Master thesis 45 hp

March 2014

Analysis of most common endogenous steroids in

plasma

Wenjun Zhao

Degree project in analytical chemistry, Master of Science (2 years)

Supervisor: Dr. Kumari Ubhayasekera

Analytical Chemistry, Department of Chemistry BMC and Science for Life Laboratory (SciLife),

Uppsala University

Box 599, 75124 Uppsala, Sweden.

ISRN UU MASTERUTB-EX--(M2014: ) Examiner: Christer Elvingson

1

Contents Abstract ......................................................................................................................................................... 3

List of Abbreviations .................................................................................................................................... 4

1. Introduction ............................................................................................................................................... 6

1.1 Endogenous steroid hormones and their metabolites .......................................................................... 8

1.2 Analytical methodology ...................................................................................................................... 9

1.2.1 Sample preparation ...................................................................................................................... 9

1.2.2 Analytical techniques for steroid analysis .................................................................................. 10

1.3 Scope of this study ............................................................................................................................ 13

2. Materials and methods ............................................................................................................................ 14

2.1 Chemicals and reagents ..................................................................................................................... 14

2.2 Standard solutions ............................................................................................................................. 14

2.3 Optimization of sample preparation .................................................................................................. 15

2.4 TLC analysis ..................................................................................................................................... 16

2.5 Sample- free plasma preparation ....................................................................................................... 16

2.5.1 Using ISOLUTE SI 6 ml SPE column ....................................................................................... 16

2.5.2 Using charcoal method............................................................................................................... 16

2.6 Steroids extraction using ISOLUTE SLE + 96-well plates .............................................................. 17

2.7 Method validation ............................................................................................................................. 17

2.8 Instrumentation ................................................................................................................................. 18

3. Results and discussion ............................................................................................................................ 19

3.1 Identification of each steroid............................................................................................................. 20

3.2 Optimized derivatization conditions ................................................................................................. 27

3.3 Retention times and specificity ......................................................................................................... 31

3.4 Steroid-free experiments ................................................................................................................... 32

3.5 Method validation ............................................................................................................................. 33

3.5.1 Calibration curve for each steroid .............................................................................................. 33

3.5.2 Limit of detection (LOD), limit of quantification (LOQ) and precision. ................................... 35

3.5.3 Normal sample analysis ............................................................................................................. 36

3.5.4 Recovery test .............................................................................................................................. 36

4. Conclusion .............................................................................................................................................. 38

5. Future work ............................................................................................................................................. 38

2

6. Acknowledgments ................................................................................................................................... 39

7. Popular scientific summary ..................................................................................................................... 40

8. References ............................................................................................................................................... 42

3

Abstract

Steroid hormones are biologically active low molecular weight lipid compounds which are synthesized

from cholesterol. These steroids are important in controlling human body functions and regulating the

nervous and immune systems. Therefore, a sensitive, stable and rapid method for the analysis and

quantification of steroids in the human plasma is necessary in pharmacokinetic studies. This work is

establishing a novel GC-MS method for the simultaneous determination of as many as 9 analytes covering

3 classes of endogenous steroids.

The methoxylation and silylation conditions for steroids were optimized in this work. The linearity of the

GC-MS method was evaluated at the steroids concentration range of 2 – 50 µg/ml. Correlation coefficient

(R2) of cholesterol and estradiol ≥ 0.99; both dehydroepiandrosterone and cortisone’s parameters ≥ 0.93.

The recoveries for all the steroids determined at low, medium and high concentration levels varied from

52.70 – 162.78 %. The concentration of cholesterol in normal plasma is 34.82 ± 1.35µg/ml.

Keywords

Steroids, methoxylation, silylation, GC-MS, quantification, human plasma

4

List of Abbreviations

AS Androsterone

BSA N,O-Bis(trimethylsilyl)acetamide

BSTFA N,O-Bis(trimethylsilyl)trifluoroacetamide

BSTFA+TMCS N,O-Bis(trimethylsilyl)trifluoroacetamide and trimethylchlorosilane

C2Cl3F3 1,2,2-trichloro-1,2,2-trifluoroethane

CAN Concentration of the analyte

CIS Concentration of the internal standard

CHOL Cholesterol

DHEA Dehydroepiandrosterone

DTE Dithioerythritol

E Cortisone

E1 Estrone

E2 Estradiol

E3 Estriol

ECN Etiocholanolone

EI Electron impact

GC Gas chromatography

GC–MS/MS Gas chromatography mass spectrometry

HDL High-density lipoprotein

HFBA Heptafluorobutyric Acid

HPLC High performance liquid chromatography

IS Internal standard

LC–MS/MS Liquid chromatography tandem mass spectrometry

LOD Limit of detection

LOQ Limit of quantification

MS Mass spectrometry

MSTFA N-Methyl-N-(trimethylsilyl)trifluoroacetamide

m/z Mass-to-charge ratio

Rt Retention time

SD Standard deviation

S/N Signal-to-noise ratio

5

SIM Selected ion monitoring

SPE Solid-phase extraction

T Testosterone

TBDMCI tert-Butyldimethylsilylchlorosilane

TEA Triethylaluminium

TMIS Trimethyliodosilane

TMS Trimethylsilyl

TMSI N-trimethylsilylimidazole

TMCS Trimethylchlorosilane

6

1. Introduction

Steroid hormones are biologically active low molecular lipid compounds which are synthesized from

cholesterol (figure 1.1). All the steroid hormones have a tetracyclic base structure which consists of four

fused rings: three cyclohexane rings and one cyclopentane [1]; the chemical structure and IUPAC names of

the steroids of interest are listed in table 1.1.

These steroids are important in controlling human body functions and regulating the nervous and immune

systems [2, 3]. In recent years, steroids have became useful biomarkers for the diagnosis of several

pathological conditions [4]. Therefore it is important to improve and continue developing new analysis

methods. The most important requirements for the techniques of steroid measurement are reliability,

specificity and sensitivity. The tandem mass spectrometry (MS/MS) technique can satisfy these

requirements, hence mass spectrometry (MS) coupled with gas chromatography (GC) or high performance

liquid chromatography (HPLC) is commonly used in steroid hormone detection during modern times [34].

Figure 1.1: Biosynthesis of steroid hormones in the cholesterol pathway

(http://en.wikipedia.org/wiki/File:Steroidogenesis.svg)

7

Table 1.1. List of the names, abbreviations and chemical structures of the investigated steroids.

IUPAC Common name Abbreviation Chemical structure

(3β)-Cholest-5-en-3-ol Cholesterol CHOL

3-Hydroxyestra-1,3,5(10)-trien-

17-one

Estrone E1

(17β)-Estra-1,3,5(10)-triene-

3,17-diol

Estradiol E2

(16α,17β)-Estra-1,3,5(10)-

triene-3,16,17-triol

Estriol E3

(3β)-3-Hydroxyandrost-5-en-

17-one

Dehydroepiandroster

one

DHEA

(3α,5α)-3-Hydroxyandrostan-

17-one

Androsterone AS

(3α,5β)-3-Hydroxyandrostan-

17-one

Etiocholanolone ECN

(17β)-17-Hydroxyandrost-4-en-

3-one

Testosterone T

11,17,21-Trihydroxypregn-4-

ene-3,20-dione

Cortisone E

The aim of this work is to develop and optimize a GC–MS/MS analytical method to serve as a diagnostic

test for steroids/biomarkers of endocrine and metabolic diseases.

8

1.1 Endogenous steroid hormones and their metabolites

The steroids that will be measured in this study are divided into the following catogories: estrogen; androgen;

and glucocorticoid, with their respective metabolites. Steroid hormones are synthesized from cholesterol

(CHOL) which can be transformed into other steroids (figure1.1).

Steroids with their metabolites show great clinical effectiveness as biomarkers for disease detection. Fouad

(et al) found that the concentration levels of steroids could indicate some disorders such as defective

steroidogenesis and other related clinical diseases [5];

Other impacts of steroids on the human body are list in table 1.2.

Table 1.2. List of the steroids and their metabolites and some effects in the human body.

Steroid and Metabolite Impact

Cholesterol It can control membrane fluidity over the range of physiological

temperatures and functions in intracellular transport, cell signaling

and nerve conduction [6].

Estrone It is a known carcinogen for human females [7].

Estradiol It could protect women from heart disease by raising the level of

HDL and protect bone density in both men and women [8].

Estriol In pregnant women with multiple sclerosis, it reduces the disease's

symptoms [9].

Dehydroepiandrosterone It works as a metabolic intermediate in the biosynthesis of the

androgen and estrogen sex steroids [10].

Androsterone It would influence human behavior and low levels could be the major

correlate of decreased sexual desire [11].

Etiocholanolone It would cause fever, immune stimulation and leukocytosis [12].

Testosterone Development of male reproductive tissues such as the testis and

prostate as well as promoting secondary sexual characteristics [13].

Cortisone It can elevate blood pressure and prepare the body for a fight or flight

response [14].

Synthesis of steroids occurs in the mitochondria and the smooth endoplasmic reticulum of cells in the

adrenal cortex, the gonads and the placenta [15]. The adrenal cortex is mainly responsible for the production

of corticosteroid and androgen hormones which has three zones. Aldosterone and mineralocorticoids are

mainly produced in the outmost zone and zona glomerulosa [16]. The middle zone, zona fasciculate, releases

glucocorticoids such as cortisol and cortisone into the body. Zona reticularis and the innermost zone is

responsible for producing androgens like dehydroepiandrosterone (DHEA) and androstenedione (the

precursor to testosterone (T)) [16]. Estrogens (E1, E2 and E3) are produced mainly by the ovaries and

placenta with small amounts being produced by the adrenal glands.

9

1.2 Analytical methodology

1.2.1 Sample preparation

Extraction/Purification

Common biological samples used for analysis of steroids includes serum, plasma, urine and saliva. Sample

sizes typically ranges from a few microliters to 1mL. Sample preparation for steroid analysis generally

involves extraction and enrichment. Extraction techniques are usually compound specific and depend on

the sample type, requiring sensitivity and high throughput. Routine sample preparation usually consist of

liquid-liquid extraction (LLE), protein precipitation and solid phase extraction (SPE) [17].

Derivatization

i. Silylation

Most steroids are required to be derivatized before injection into GC-MS. Silylation is one of the most

common techniques employed in steroid derivatization. The reaction occurs by substituting the proton in

hydroxyl (-OH), amino (-NH2) and sulfide (-SH) groups with a silyl group (R3Si) [18].

Some of the most common reagents used for derivatization, include MSTFA[19-26], TMIS[23,24], DTE[24,27],

BSTFA[24], trimethylsilylimidazole(TMSI)[24,27-29], trimethylchlorosilane (TMCS)[24] , HFBA[25,30] and

TEA[25]. To enhance the derivatization efficiency, these reagents are sometimes mixed in different ratios.

For example, Cangialosia et al. [21], mixed MSTFA and TMIS in a ratio of 1000 to 2 (v/v) for derivatization

at 60°C for 15 min, and shown a good sensitivity. Other derivatization mixtures have also been used, namely

MSTFA, TMIS and DTE (1000:2:5, v/v/w); BSTFA and 1% TMCS; MSTFA and 1% TMCS; BSTFA and

1% TMSI; MSTFA and 1% TMSI; a good linearity of response with a coefficient R2 > 0.994 was obtained;

Limits of detection (LODs) ranging from 0.01 to 1.0 ng/mL for all steroid hormones were also obtained;

All the derivatives were stable for quite some time (24-48 hours) [24]. The conditions for derivatization were

as follows: MSTFA with 2.0% NH4I and 1.5% dithioerythritol (v/w/w) was heated for 10 min at 90°C and

the lowest quantification limits were reported to be from 0.002 to 0.6 μg/L for 1.0 mL of saliva [26].

From the discussion above, it is clear that the optimization of derivatization conditions still remains a big

challenge in the determination of steroids.

10

ii. Methoxylation

Steroids containing ketone groups such as cortisone or cortisol sometimes cannot be detected by GC–MS

after silylation. Normally, the ketone group of these compounds should be derivatized by methoxylamine

or ethoxylamine to form a stable oxime. The methoxylation reagent can be used separately or coupled with

a silylation step [31].

1.2.2 Analytical techniques for steroid analysis

Currently, the main techniques employed for the analysis of steroids are immunoassays [32, 33], LC–MS [34,

35-37] and GC–MS/MS [30, 34, 38-44]. However, immunoassays are mainly applied in the analysis of clinical

disorders influenced by cross-reactions, thus negatively affecting the results. The latest techniques are

mainly GC–MS and LC–MS based. Both of these techniques are superior to immunoassays in terms of

increased specificity, quantification and linearity.

GC–MS/MS method

The principle of GC–MS is that gas-phase ions are separated according to mass-to-charge (m/z) ratio and

then sequentially detected in single and multiple modes. A GC–MS instrument is divided into four major

parts, i.e. sample introduction, ion source, mass analyzer and detector. A typical schematic diagram of a

GC–MS/MS system is shown in figure 1.2.

11

Figure 1.2: Schematic diagram of Scion GC-MS/MS and q0 is ion guide of this instrument

GC–MS is normally used as the predominant method in the determination of steroids in human plasma,

because it ensures high specificity and sensitivity as well as a low limit of detection (LOD) [45]. According

to Thenot and Horning all possible steroids with their metabolites can be detected by GC–MS [46]. In 1960,

the analysis of steroids by GC was described for the first time by Van den Heuvel [19] who used cholestane

as internal standard (IS), and reported that the compounds with a hydroxyl group eluted before the

compounds with a ketone group. Four years later, in 1964, Ryhage reported on coupling GC with MS for

the first time [47]. This invention paved the way for the determination of the structures of known and

12

unknown steroids [47]. The mass analyzer of most GC–MS systems can be operated in full scan mode and/or

selected ion monitoring mode. This provides the operator with the opportunity to detect most of the

compounds present in a sample or to detect only a selected few. A review of the literature on steroid analysis

revealed reports on the simultaneous determination of some of the steroids of interest as well as methods

for individual detection [27, 30].

It has been reported by Jun Chen, et al (2009) that DHEA, cortisol (E), estrone (E1), 17β-estradiol (E2), T

and androsterone (AS) were analyzed simultaneously and the resulting detection limit (LOD) and limit of

quantification (LOQ) range show that the sensitivity is quite good and stable[30]. Martin et al.[27] also found

the concentration levels of estrogen and androgen groups in the plasma and serum, down to 0.08–0.16

ng/mL and 0.20 – 0.36 ng/mL, respectively; the accuracy of this method was 50–112% in the range 0.10 to

2.00 ng/mL.

LC–MS/MS method

GC–MS methods are well suited for the analysis of almost all steroids and their metabolites, however, some

steroids and their metabolites are preferably analyzed using LC–MS/MS methods; e.g. corticosterone[48],

tetrahydrocortisol[49,50], and tetrahydrocorticosterone[51], to name a few. All the results show good sensitivity

and reproducibility.

Comparison of GC–MS and LC–MS/MS techniques for steroid analysis

GC–MS offers higher resolution than LC–MS/MS due to the longer analysis running time. For some

diastereomeric compounds, GC–MS has better separation than LC–MS/MS, which renders it more useful

in the diagnosis of some metabolic disorders like apparent cortisone reductase deficiency (ACRD) [54]. GC–

MS is also a good tool for analyzing non-target steroids, especially when used in full scan mode which

enables storage of the fragmentation information for evaluation at a later stage. This is very useful in, for

example, the field of sports for the randomized anti-doping analysis of athletes’ body fluid samples which

can be re-checked for previously unidentified drugs [34]. GC-MS needs less sample volume than LC-MS

however GC-MS is more sensitive than LC-MS.

13

Table 1.3: Advantages and disadvantages of the current methods for analysis of steroids

Method Advantage(s) Disadvantage(s)

GC–MS Good resolution and can be used for

volatile compounds

High sensitivity/ throughput

Small sample volume

High cost and time consuming sample

preparation procedures

Cannot be used for non-volatile and

thermal unstable molecules

LC–MS/MS Shorter analysis time and can be used for

non-volatile and high molecular weight

compounds

Not good for volatile compounds

Immunoassay Simple, rapid, inexpensive Overestimate steroid concentrations

One of the biggest drawbacks of GC–MS is that it is marked by a time consuming sample preparation step

which is often less prominent in LC–MS/MS. In many cases, derivatization of steroids are needed, which

usually include methoxylation of ketone groups and silylation of hydroxyl groups, to improve their

volatility when using GC–MS. After derivatization, compounds can give a mass fragmentation of the

molecular ion, so it is easier for determination the functional groups on the steroid [18].

After a comparison of these two techniques, it is evident that GC–MS is a powerful tool that can be

employed for the analysis of steroids, especially unknown steroids, despite being prone to time-consuming

sample preparation steps and longer analysis times [54, 55]. The advantages and disadvantages of these

methods for the analysis of steroids are summarized in Table 1.3.

1.3 Scope of this study

The main objective of this study was to develop a GC–MS method for the complete separation and

quantification of several steroids to use as a diagnostic test for steroid/biomarkers of endocrine and

metabolic disease. Optimizations of sample preparation and clean-up steps as well as derivatization

conditions are needed. After method development and optimization, the method should be evaluated and

applied in analyzing biological tissues and fluids.

14

2. Materials and methods

2.1 Chemicals and reagents

Cholesterol (CHOL), estrone (E1), estradiol (E2), estriol (E3), dehydroepiandrosterone (DHEA),

androsterone (AS), etiocholanolone (ECN), testosterone (T) and cortisone (E) were purchased from

Steraloids (Newport, RI, USA). 5α-Cholestane(C27H48) was used as internal standard (IS); N,O-

bis(trimethylsilyl)acetamide (BSA); N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA), BSTFA with 1%

trimethylchlorosilane (BSTFA+TMCS), tert-butyldimethylsilylchloride (TBDMCL); 1,1,1,3,3,3-

hexamethyldisilane + trimethylchlorosilane + pyridine (HMDS+TMCS+Pyridine (3:1:9)); methoxyamine

hydrochloride and anhydrous pyridine (99.8%) were obtained from Sigma-Aldrich (Stockholm, Sweden).

All chemicals and chromatographic reagents used were of analytical grade. All other chemicals and reagents

were purchased from Merck, Eurolab (Stockholm, Sweden), unless otherwise stated. Ultrapure water was

deionized and osmosed with a MilliQ purification system (Millipore, Bedford, MA, USA) was used.

2.2 Standard solutions

Stock solutions (1 mg/mL) of all steroids were prepared by weighing off, with analytical precision, each

steroid and dissolving it in ethanol. A stock solution of 100 ng/µL in ethanol was prepared for the internal

standard. The stock solutions were kept in a freezer at -20 °C within tightly capped glass tubes. In order to

obtain the working solutions and calibration standards at different concentrations, the stock solutions were

diluted with methanol (MeOH). The concentration of methoxyamine hydrochloride was 20 mg/mL in

anhydrous pyridine.

Working solutions and calibration standards were prepared by a general protocol. After pipetting 20 µL

from stock solution (1mg/mL) and 100µl from internal standard solution (100ng/µL) into a glass tube, the

solvent was evaporated by a gentle flow of nitrogen gas. The methoxylation reagent (30 µL) was added.

The glass tube was kept in an oven at 60°C for 45min. After incubation, the solvent was evaporated and

100 µL of silylation reagent was added to the sample. The samples were incubated again at 120°C for 30

min. Then the solution was cooled to the room temperature and the solvent was evaporated under nitrogen

gas. Finally 100 µL of C2Cl3F3 was added. The solution was centrifuged at room temperature at 3,000 × g

for 5 min. The solution was transferred to a small GC vial and injected into the GC–MS instrument for

analysis.

15

2.3 Optimization of sample preparation

Methoxylation

Different methoxylation conditions were investigated. Therefore, one sample from each set was paired and

subjected to the same set of incubation conditions for methoxylation as shown in Table 2.1. Methoxyamine

hydrochloride in pyridine was used as methoxylation reagent. After methoxylation of the compounds, the

silylation reagent was added to each set and incubated.

Table 2.1: Different incubation conditions for methoxylation of standards.

Set Name of silylation reagent

(100µL)

Incubation conditions

for methoxylation (30µL)

Incubation conditions

for silylation (100µL)

A

BSA

60°C for 30 min

120°C for 30min 60°C for 45min

70°C for 30min

B

BSTFA

60°C for 30 min

120°C for 30min 60°C for 45min

70°C for 30min

Silylation

Different silylation conditions were investigated. Five different silylation reagents or mixtures, namely

BSA, BSTFA, HMDS+TMSC, TBDMCI and BSTFA+TMCS, were investigated by incubating at different

temperatures and inclubation times and the conditions are shown in Table 2.2.

Table 2.1: Different incubation conditions for silylation of standards.

Set Name of silylation

reagent (100µL)

Incubate condition for the

methoxylated derivatization(30µL)

Incubate condition

for the silylation

A BSA 60°C for 45min 70°C for 1 hour

120°C for 30 min

B BSTFA 60°C for 45min 70°C for 1 hour

120°C for 30 min

C HMDS+TMCS 60°C for 45min 70°C for 1 hour

120°C for 30 min

D TBDMCL 60°C for 45min 70°C for 1 hour

120°C for 30 min

16

E BSTFA+TMCS 60°C for 45min 70°C for 1 hour

120°C for 30 min

The working flow diagram of experiment with standards

Pipette 20µl from each steroid stock solution and 100µl from internal standard solution evaporate

solvent add 30µl methoxylation reagent keep the tube in the oven at different programs

evaporate solvent add silylation reagent 100 µl into the mixture solution under the N2 gas flow

keep in the oven at different programs evaporate solvent the derivatized standards re-dissolved

in 100µl C2Cl3F3 centrifuge for 5min GC-MS/MS analysis

2.4 TLC analysis

To check the completion of derivatization, underivatized and derivatized steroids were analyzed by

analytical pre-modified silica with aliphatic hydrocarbons. The TLC plates silica gel 60, 20 × 20 cm, 0.25

mm thickness were used (Merck, Eurolab AB, Stockholm, Sweden) to check visually the completeness of

derivatization. The derivatized samples along with the standard steroids (underivatized) as TLC-references

(Steraloids, Newport, RI, USA), were spotted onto a TLC plate and developed in the solvent system diethyl

ether: cyclohexane (90:10, v/v). The TLC plate was dried briefly in air and then sprayed with 10 %

phosphomolybdic acid in diethyl ether: ethanol (50:50, v/v). The plate was placed in an oven at 120 ºC for

15 min for color development.

2.5 Sample- free plasma preparation 2.5.1 Using ISOLUTE SI 6 ml SPE column

Normal plasma (1mL) was diluted with 1mL hexane: diethyl ether (9:1 v/v) and vortexed at room

temperature. The diluted sample was loaded onto an ISOLUTE SI 6 ml SPE cartridge (pre-conditioned with

5ml hexane). After distribution of the aqueous phase in the column, the steroids of interest were eluted with

15ml hexane: diethyl ether (9:1 v/v). After that process, washing with 10ml hexane: diethyl ether (1:1 v/v)

again. Finally, elution of pure steroid free plasma with 10ml acetone into silanized glass tubes. The collected

elution was dried under the N2 gas after the methoxylation and silylation process. The sample solution was

compared with the standard cholesterol solution using a TLC plate to check if the steroids were removed.

2.5.2 Using charcoal method

The steroid-free plasma was prepared by activated charcoal as described by Aburuz et al [56]. Two sets of

normal plasma (200µl) were prepared, one set without charcoal and one set with 8mg charcoal then vortexed

17

and centrifuged. After the filtering, Methanol: CHCl3 (2:1, v/v) was added twice into these two sets of

plasma samples. The sample solution dried under the N2 gas after the methoxylation and silylation process.

It was then compared with the standard cholesterol solution using a TLC plate to check if the steroids were

removed.

2.6 Steroids extraction using ISOLUTE SLE + 96-well plates

Normal plasma (100µl) with 2µg internal standard was diluted with MQ water (100µl). The pre-treated

sample (200µl) was loaded into a plate followed by a pulse of vacuum to initiate flow and left for 5 minutes.

Then eluted with dichloromethane (1ml) dropped directly into a deep well collection plate which was left

for 5 minutes. Then a short pulse of vacuum was applied. The eluent was collected followed by

methoxylation and derivatization and finally injected into GC-MS.

2.7 Method validation

The proposed method was validated using spiked plasma samples as described in the guidelines by the FDA

(http:\\www.fda.gov\cder\guidance\4252fnl.pdf.2001). The method was validated using four different

steroids (CHOL, E2, DHEA and E). Stock solution of steroids and IS (Cholestane) were prepared

independently at 1mg/mL in ethanol and stored at -20˚C. Prior to the measurements, these solutions were

serially diluted with the same solvent and further used for the calibration and quality control (QC).

Linearity

In order to accurately quantify the targeted steroids, the measured response must be correlated with the

concentration of the steroid hormones. Therefore, calibration curves need to be constructed. Calibration

standards of CHOL, E2, DHEA and E were prepared from 1 mg/mL stock solutions according to the general

protocol described in section 2.2. The standards for CHOL, E2 and DHEA were prepared to a final

concentration of 2, 4, 10, 15, 25 and 50 µg/mL and E was 50, 75, 100, 150 and 200 µg/mL in triplicate,

respectively. Each sample contained 15 µg/mL of the internal standard. The calibration range reflected the

steroid concentration levels in the human body and the number of replicates was sufficient to determine

certain parameters and perform various statistical analyses (linearity, precision, accuracy, sensitivity,

stability, limit of detection, etc.). Samples were run in triplicates. Calibration curves were separately

prepared by plotting the peak area ratio of each steroid area of IS against the concentration of each

compound. The linearity was determined by linear regression analysis.

18

The ratio of the integrated peak areas (AAN)/(AIS) and the concentration ratio (CAN/CIS) of the analyte (AN)

and internal standard (IS) were calculated. Calibration curves of (AAN)/(AIS) versus (CAN/CIS) were

constructed for each compound. A linear relationship over the analyte concentration range was displayed

by all four calibration curves.

Precision and accuracy

Quality control samples were used for precision and accuracy determination on intra-day and inter-day

basis. The intra-day (n=3) and inter-day precisions (n=6) were determined for the spiked samples.

Limit of detection and quantification

The detection limit (LOD) and the quantification limit (LOQ) were calculated using calibration curves for

samples spiked with steroids. LOD was calculated using the formula 3.3 δ/S and LOQ was obtained by the

formula 10 δ/S, here δ is the residual standard deviation of the regression line and S is the slope of the

regression line.

Recovery

Samples were spiked in triplicates with known amounts of four different steroids at five different

concentration levels (CHOL, E2, and DHEA was 500ng, 1000ng, 2000ng, 3000ng and 5000ng; E was

5000ng, 7500ng, 10000ng, 12500ng and 15000ng respectively) and analyzed to determine the recovery of

each steroid.

2.8 Instrumentation

Analysis of steroids was performed using a triple quad mass spectrometer coupled to the gas chromatograph

(SCION TQTM GC-MS/MS System, Billerica, MA, USA). Separation of steroids was performed on a non-

polar capillary column (DB-5MS, J&W Scientific, Folsom, CA, USA), 30 m 0.18 mm 0.18 m film

thickness. Helium was the carrier gas at a linear velocity (1 mL/min). The injector temperature was 250°C

and the samples were injected using the split injection mode (20:1). The oven temperature was 120°C

followed by an increase of 50°C /min to 240°C and an isothermal hold of 1 min, then ramping at 1.5°C

/min to 250 °C, after that increasing of 10°C /min to 260°C, finally by an increase of 50°C /min to 320°C

with a hold of 5 min, total running time was 17.27 min. The ionization mode was electron impact (EI). The

mass spectra were recorded at an electron energy of 70 eV and the ion source temperature was 260°C. The

transfer line temperature was 300°C. The spectra were scanned in the range of 50-600 m/z. Identification

19

of the steroids was done by comparing the mass spectra of standard samples of the steroids and their

retention times. Chromatograms were collected in full scan mode (TIC). Quantification of the steroids was

accomplished using cholestane as an internal standard due to its physical and chemical behavior similar to

that of target analyte as well as its absence in biological samples. Quantification of steroids (area response)

was accomplished with the help of the internal standard.

3. Results and discussion

A simple analytical procedure for the simultaneous determination of 9 steroids by a GC-MS/MS method

has been developed. Many target analytes are not intrinsically volatile and will decompose during the

transition between the liquid and gas phase inside the injection port or during elution from the

column at high temperatures. Consequently, several steps, including pre-analytical derivatization, are

required for steroid analysis by GC/MS. The derivatization is one of the main steps in the sample

preparation. This step is required to improve the volatility and thermal stability of steroids. Derivatization

of steroids is generally required before analyzing by GC-MS/MS. This process often time consuming.

In the present study, several strategies have been employed to enhance this process, like incubation

temperature, time of incubation and different silylating reagents.

Mass spectrometry is the most recent method that has wide acceptance in the steroid field. GC for

separation prior to MS (GC/MS) is used, due to the relative ease of coupling these orthogonal

techniques with respect to the vacuum considerations. Given the high resolving capability of a GC

equipped with a modern capillary column and the superb precision of MS, GC/MS systems have been

a mainstay of analytical chemistry. To facilitate the interpretation and discussion of the results, mass

spectrometry identification of steroids is briefly discussed below (Figure 3.1).

20

c

Figure 3.1: a Schematic diagram of a simple mass spectrometer with source, quadrupole, and

detector, sorting ions by m/z. b Generation of a mass spectrum for a pure compound. The

molecular ions fragment, and the charged fragments are separated by the quadrupole and

quantitated by the detector, yielding a standard mass spectrum. c Mass spectrum of

testosterone.

During the methoxylation reaction a methyl group (–CH3) from the methoxylation reagent, with a

molecular mass of 15 Da, binds to the oxygen of a ketone group of the steroid. A trimethylsilyl

group (–Si(CH3)3) from the silylation reagent, with a molecular mass of 73 Da, will attach to the

oxygen of a hydroxyl group of the steroid.

3.1 Identification of each steroid

The MS detection of every steroid showed abundant molecular ions [(M-n×H) +n×73+m×15]+, or [(M-

n×H) +n×73]+ (M represents the molecular weight of each steroid, n represents the number of hydroxyl

groups and m represents the number of ketone groups) providing evidence of the steroid's molecular weight

(Table 1.1).

21

Figures 3.2-3.12 illustrate the chromatogram and mass spectrum using full scan mode for each steroid.

1) The molecular mass of cholesterol is 386.65 Da (Table 1.1) and the structure has one hydroxyl

group. The MS detection of cholesterol showed an abundance singly charged molecular ions [(M-

H) +73]+ at m/z of 458 providing evidence of cholesterol. Also fragments of m/z 129, 329 and 368

could be detected. The mass to charge ratio m/z of 129 is a basic fragment formation for identifying

steroids; also [M-H2O] gave the formation of m/z at 368. As we can see from the figure, the intensity

of the molecular ion with m/z of 458 was relatively low comparing with the other fragments due to

that the EI source is a hard ionization technique with high voltage so that most of analyte would be

fragmented and leaving a weak molecular ion peak.

Figure 3.2: Representative cholesterol chromatograms and mass spectrum; cholestane was the internal

standard. The molecular ion m/z is 458.

2) The molecular mass of androsterone is 290.44 Da (table 1.1); the structure has one ketone group

and one hydroxyl group, respectively. During the methoxylation process, the ketone group cannot

be completely converted to a methoxy group, so androsterone would show two peaks after GC

separation. The MS detection of androsterone showed one charged molecular ions [(M-H)+73]+ at

m/z of 362.3 and another was [(M-H) +73+15]+ at 377.3 providing evidence of androsterone. Figure

3.2 shows androsterone with one TMS group (m/z 362.3) and figure 3.3 shows androsterone with

one TMS group and one methoxy group (m/z 377.3).

22

Figure 3.3: Representative Androsterone-TMS chromatograms and mass spectrum; cholestane was the

internal standard. The molecular ion m/z is 362.3.

Figure 3.4: Representative Androsterone-TMS-CH3 chromatograms and mass spectrum; cholestane was

the internal standard. The molecular ion m/z is 377.3.

3) The molecular mass of etiocholanolone is 290.44 Da (table 1.1), the structure of etiocholanolone

has one ketone group and one hydroxyl group, respectively. The ketone group can be completely

converted to a methoxy group. The MS detection of etiocholanolone showed an abundance singly

charged molecular ions [(M-H) +73+15]+ at m/z of 377.4 providing evidence of etiocholanolone.

23

Figure 3.5: Representative etiocholanolone chromatograms and mass spectrum; cholestane was the

internal standard. The molecular ion m/z is 377.4.

4) The molecular mass of DHEA is 288.42 Da (table 1.1). The structure of DHEA has one ketone

group and one hydroxyl group, respectively. The ketone group can be converted to methoxy group

completely. The MS detection of DHEA showed an abundance singly charged molecular ions [(M-

H) +73+15]+ at m/z of 375.4 providing evidence of DHEA.

Figure 3.6: Representative DHEA chromatograms and mass spectrum; cholestane was the internal

standard. The molecular ion m/z is 375.4.

5) The molecular mass of estrone is 270.37 Da (table 1.1). The structure of estrone has one ketone

group and one hydroxyl group, respectively. During the methoxylation process, the ketone group

cannot be completely converted to a methoxy group, so the estrone still show two peaks in the

chromatogram. The MS detection of estrone showed one charged molecular ion [(M-H) +73]+ at

24

m/z of 342.3 and another was [(M-H) +73+15]+ at 357.4 providing evidence of estrone. Figure 3.7

shows estrone with one TMS group (m/z 342.3) and figure 3.8 shows estrone with one TMS group

and one methoxy group (m/z 357.4).

Figure 3.7: Representative estrone-TMS chromatograms and mass spectrum; cholestane was the internal

standard. The molecular ion m/z is 342.3.

Figure 3.8: Representative estrone-TMS-CH3 chromatograms and mass spectrum; cholestane was the

internal standard. The molecular ion m/z is 357.4.

6) The molecular mass of estradiol is 272.38 Da (table 1.1). The structure of estradiol has two

hydroxyl groups. The MS detection of estradiol showed one charged molecular ions [(M-2H)

+2×73]+ at m/z of 416.3 providing evidence of estradiol. Figure 3.9 shows the results for the

molecular ion of estradiol.

25

Figure 3.9: Representative estradiol-2TMS chromatograms and mass spectrum; cholestane was the

internal standard. The molecular ion m/z is 416.3.

7) The molecular mass of testosterone is 288.42 Da (table 1.1). The structure of testosterone has one

ketone group and one hydroxyl group, respectively. The ketone group can be completely converted

to a methoxy group. Testosterone has enantiomers, so after the derivatization, there were two split

peaks with exactly the same fragment information. The MS detection of testosterone showed an

abundance singly charged molecular ions [(M-H) +73+15]+ at m/z of 375.4 providing evidence of

testosterone. This is shown in figure 3.10.

Figure 3.10: Representative testosterone chromatograms and mass spectrum

26

8) The molecular mass of estriol is 288.38 Da (table 1.1). The structure has three hydroxyl groups and

can be silylated completely. The MS detection of estriol showed an abundance singly charged

molecular ions [(M-3H)+3×73]+ at m/z of 504.3 providing evidence of estriol. Figure 3.11 shows

the information about estriol.

Figure 3.11: Representative estriol-3TMS chromatograms and mass spectrum; cholestane was the internal

standard. The molecular ion m/z is 504.3

9) The molecular mass of cortisone is 360.44 Da (table 1.1). The structure of cortisone has three

ketone groups and two hydroxyl groups, respectively. The ketone groups can be completely

converted to methoxy groups. The MS detection of cortisone showed an abundance singly charged

molecular ions [(M-2H) +2×73+3×15]+ at m/z of 549.6 providing evidence of cortisone. The figure

3.12 shows the fragment information about cortisone.

27

Figure 3.12: Representative cortisone-2TMS-3CH3 chromatograms and mass spectrum; cholestane was

the internal standard. The molecular ion m/z is 549.6.

3.2 Optimized derivatization conditions

Silylation conditions such as different silylation reagents (BSA, BSTFA, BSTFA+TMCS, TBDMCI and

HMDS+TMCS+Pyridine (3:1:9)) with different reaction times and reaction temperatures. I also tried to test

different reaction times and reaction temperatures for the methoxylation reaction. To measure the efficiency

of derivatization, we should compare the peak area for each steroid. Table 3.1 shows the results from the

silylation reaction.

Table 3.1. Peak area of every steroid using different silylation reaction condition (methoxylated reaction

condition was the same at 60°C for 45min)

Steroid

BSA BSTFA BSTFA+TMCS HMDS

+TMCS

TBDM

CL

70°C

60min

120°C

30min

70°C

60min

120°C

30min

70°C

60min

120°C

30min

70°C

60min

120°C

30min

AN-TMS 1.115e8 3.837e8 1.745e8 4.420e7 1.868e8 3.072e8 1.112e8 1.868e8

ETIO-

TMS

2.320e8

9.171e8

4.371e8

1.069e8

4.084e8

6.937e8

2.393e8

4.084e8

AN-TMS-

CH3

1.679e8

8.144e8

2.870e8

2.580e8

2.145e8

6.390e8

1.845e8

2.145e8

ETIO-

TMS-CH3

6.147e8

2.611e9

1.050e9

7.680e8

8.279e8

2.051e9

6.571e8

8.279e8

28

DHEA 3.230e8 1.556e9 5.676e8 5.093e8 4.120e8 1.229e9 3.608e8 4.120e8

E1-TMS 1.221e8 5.315e8 2.506e8 7.414e7 2.414e8 4.012e8 1.567e8 2.414e8

E2 4.389e8 1.891e9 8.655e8 5.444e8 6.373e8 1.579e9 5.080e8 6.373e8

E1-TMS-

CH3

1.429e8

9.242e8

3.190e8

2.809e8

2.291e8

7.431e8

2.045e8

2.291e8

T 2.860e8 1.226e9 4.815e8 2.683e8 3.507e8 9.713e8 2.504e8 3.507e8

E3 4.824e8 2.088e9 9.235e8 6.457e8 6.202e8 1.692e9 5.425e8 6.202e8

CHOL 4.384e8 1.517e9 7.778e8 4.763e8 6.181e8 1.475e9 4.736e8 6.181e8

E 2.368e8 9.507e8 5.575e8 3.197e8 4.394e8 1.088e9 5.575e8 4.394e8

*For some unknown errors, I did not collect the data information about HMDS+TMCS at 120°C, 30min

and TBDMCL at 70°C, 60min; *e represents the base 10.

Comparing all the derivatization yield efficiency information, the reactions of BSA and BSTFA+TMCS

(99:1) at 120°C, 30 min seem approximately the same, and better than the other reagents. BSTFA shows

the best result among all the reagents at 70°C, 60min. During 120°C, the efficiency of BSA is much better

than at 70°C; BSTFA shows the opposite result. Finally, BSTFA+TMCS (99:1) and BSA were chosen to

do the experiment for optimization of the methoxylation reaction. The figure 3.13 illustrates that the peak

areas are not much different between BSTFA-TMCS (99:1) and BSA as silylation reagents.

Figure 3.13: the chromatograms (overlaid) for 9 steroids when using BSTFA+TMCS (99:1) and BSA as

silylation reagent. The red color is for BSA and blue color is for BSTFA-TMCS (99:1).

29

After finishing the optimization of the silylation reaction, the silylation condition was set to 120°C, 30min

to continue the experiment for the methoxylation reaction. Table 3.2 shows the result for optimization of

the methoxylation reaction.

Table 3.2 Peak area of every steroid using different methoxylation reaction conditions (silylation reaction

condition was everywhere the same at 120°C for 30min).

Steroid BSA BSTFA + TMCS (99:1)

60°C

30min

60°C

45min

70°C

30min

60°C

30min

60°C

45min

70°C

30min

AN 7.902e8 5.982e8 1.175e9 3.449e8 3.971e8 3.749e8

ETIO 1.932e9 1.430e9 2.897e9 1.073e9 1.160e9 1.058e9

DHEA 1.409e9 1.005e9 2.151e9 7.054e8 8.127e8 7.573e8

E1 9.175e8 7.332e8 1.556e9 4.801e8 6.018e8 5.110e8

E2 9.695e8 6.978e8 1.822e9 4.375e8 5.594e8 5.588e8

E1 4.701e8 2.541e8 6.159e8 1.992e8 2.064e8 1.740e8

T 3.809e7 2.373e7 11.887e7 2.495e7 2.738e7 2.443e7

E3 1.071e9 7.698e8 1.860e9 5.335e8 5.830e8 5.303e8

CHOL 7.986e8 5.225e8 1.329e9 3.616e8 4.232e8 3.338e8

*For some unknown errors, I did not collect the data information about cortisone. *e represents the base 10.

According the results above, BSA shows the best yield efficiency for the methoxylation reaction at 60°C,

30min (figure 3.14); all the conditions for BSTFA + TMCS (99:1) gave the approximately the same yield

efficiency for the methoxylation reaction (figure 3.15). Comparing all the results, using BSA as silylation

reagent at 120°C, 30min and methoxylation condition at 60°C, 30min gives the best derivatization

efficiency. We still, however, wanted to extend the time for the methoxylation reaction to keep all the

analytes stable, and a methoxylation reaction of 60°C, 45min was chosen as the final methoxylation

condition.

30

Figure 3.14: The chromatograms (overlaid) for 9 steroids when using BSA as silylation reagent with

different methoxylated conditions.

Figure 3.15: The chromatograms (overlaid) for 9 steroids when using BSTFA+TMCS (99:1) as silylation

reagent with different methoxylated conditions

31

In this study, that samples reacting with 30µl methoxyamine hydrochloride at 60°C for 45min, then with

100 µL BSA at 120°C for 30 min was the optimized condition.

3.3 Retention times and specificity

The retention times of 9 steroids are shown in Table 3.3, and the gas chromatogram of 9 steroids is shown

in figure 3.16.

Figure 3.16: Gas chromatography of 9 steroids and internal standard. The 9 Steroids were detected via

mass spectrometry after being derivatization. (1) Androsterone-TMS; (2) Androsterone-TMS-CH3; (3)

ETIO-TMS- CH3; (4) DHEA; (5) Estrsdiol; (6) Estrone-TMS-CH3; (7) and (8) Testosterone; (9) Cholestane;

(10) Estriol; (11) Cholestrol; (12) Cortisone.

Table 3.3 Analysis parameters for 9 steroids

Steroids Retention time (min) Fragments (amu)

AN-TMS 6.606 129.0;272.3;347.3;362.3a

AN-TMS-CH3 7.176 129.1;270.3;360.3;377.3a

ETIO-TMS-CH3 7.313 129.1;270.3;360.3;377.3a

DHEA 8.061 129.0; 358.3;360.4a

E2 8.748 129.0;232.3;285.3;416.3a

32

E1-TMS-CH3 8.847 129.1;231.2;340.3;357.4a

T 9.091 and 9.207 129.1;358.3;375.4a

E3 11.576 129.1;147.0;270.2;504.3a

CHOL 12.726 129.0;329.4;368.4;458.5a

E 14.027 129.1;369.3;459.3;549.6a

a represents the qualifier and quantifier ion of each compound.

3.4 Steroid-free experiments

Thin layer chromatography experiments were used for testing the methods of removed the steroids.

Among all the steroids in the human body, cholesterol shows the highest concentration level, so the

standard of cholesterol was chosen as the reference steroid. Figure 3.17 shows the extraction result of

ISOLUTE SI 6 ml SPE column and Figure 3.18 is the result of the charcoal method.

Figure 3.17: SPE extreaction method Figure 3.18: Activated charcoal method

As we can see, cholesterol was not found in the plasma when we used ISOLUTE SI 6 ml SPE column,

and the method with activated charcoal shows that this method cannot remove the steroids.

33

3.5 Method validation

3.5.1 Calibration curve for each steroid

During the experimentation process, various errors occurred at different steps of the assay, so the peak area

ratio between the steroids and internal standard instead of the absolute peak area of steroids, was utilized

to calculate the concentration of steroids in plasma. The peak area ratios were plotted against the

concentration of the steroids (Figure 3.19-3.22). The quality of the linear fit was scrutinized by means

of several statistical analysis methods as presented in the following section.

Figure 3.19: Calibration curve for cholesterol, showing average (n=6) cholesterol to internal standard

peak area ratio versus concentration

y = 0.045x - 0.0792R² = 0.9984

-5.00E-01

0.00E+00

5.00E-01

1.00E+00

1.50E+00

2.00E+00

2.50E+00

3.00E+00

0.00E+00 1.00E+01 2.00E+01 3.00E+01 4.00E+01 5.00E+01 6.00E+01Pe

ak a

rea

of

Ch

ole

ste

rol/

Pe

ak a

rea

of

IS

Cholesterol concentration (µg/mL)

34

Figure 3.20: Calibration curve for estradiol, showing average (n=6) estradiol to internal standard peak

area ratio versus concentration

Figure 3.21: Calibration curve for DHEA, showing average (n=6) DHEA to internal standard peak area

ratio versus concentration

y = 0.0617x - 0.0743R² = 0.9900

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0.0 10.0 20.0 30.0 40.0 50.0 60.0

Pe

ak a

rea

of

Estr

adio

l/P

eak

are

a o

f IS

Estradiol concentration (g/mL)

y = 0.0322x - 0.1663R² = 0.9833

-0.20

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

0.0 10.0 20.0 30.0 40.0 50.0 60.0

Pe

ak a

rea

of

DH

EA/P

eak

are

a o

f IS

DHEA Concentration (g/mL)

35

Figure 3.22: Calibration curve for cortisone, showing average (n=6) cortisone to internal standard peak

area ratio versus concentration

3.5.2 Limit of detection (LOD), limit of quantification (LOQ) and precision.

LOD and LOQ were determined as the concentrations producing signal to noise ratio of 3 and 10

respectively. The instrument precision was determined by multiple injections of the samples. All the results

summarized in Table 3.4.

Table 3.4 Relative standard deviation (RSD) values of standard steroid/internal standard peak area ratio at

each concentration and each steroid LOD and LOQ

Concentration

(µg/ml)

RSD (%)

n=6

LOD(µg/ml) LOQ(µg/ml)

Cholesterol

4

10

15

50

25.9

17.0

8.9

8.5

2.6

8.0

Cortisone

50

65

75

100

15.0

31.7

11.0

21.3

62.4

189

4 9.4

y = 0.0032x - 0.1275R² = 0.9312

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

40.0 90.0 140.0 190.0

Pe

ak a

rea

of

Co

rtis

on

e/P

eak

are

a o

f IS

Cortisone concentration (g/mL)

36

DHEA 10

15

50

22.0

26.6

11.2

9.0 27.2

Estradiol

4

10

15

50

10.4

3.6

9.9

4.8

6.6

20.0

3.5.3 Normal sample analysis

The purpose of the steroid quantification method development is eventually to measure concentration of

steroids in normal and patient plasma samples. In this study, we used normal plasma to check what kind of

steroids we can find using this method. After extraction by ISOLUTE SLE + 96-well plates, only

cholesterol can be identified. The chromatogram is shown below (figure 3.23).

Figure 3.23: normal plasma chromatogram

According the peak area ratio and the fitted equation in figure 3.19, concentration of cholesterol is 34.82 ±

1.35µg/ml

3.5.4 Recovery test

The recovery of each steroid (cholesterol, cortisone, DHEA, and cortisone) was estimated as the ratio

between the areas under the peaks of the extracted samples. The spiking level for DHEA, Estradiol, and

Cholesterol was 500ng, 1000ng, 2000ng, 3000ng and 5000ng; recovery level for Cortisone was 5000ng,

7500ng, 10000ng, 12500ng and 15000ng respectively. All the results are illustrated below (table 3.5).

37

Table 3.5 Recovery of four steroids

Recovery (%) Spiking level (ng)

500 1000 2000 3000 5000

Cholesterol 119.49 - 122.35 126.69 125.27

Cortisone 128.88 125.25 127.26 - -

DHEA 99.27 52.70 83.56 - -

Estradiol 149.75 115.20 105.89 - 162.78

A dash represents that I did not collect the data information because some unknown errors.

38

4. Conclusion

The aim of my study was to develop a sensitive, stable and rapid method for the analysis of steroids in

biological tissues by GC-MS, a more reliable technique than the usual immunoassay technique, which

allows also a wide range of steroids to be identified in a single sample. The steroids I tried belonged to the

androgens, estrogens, and corticoids sub-groups. The utilization of different derivatization methods during

the sample preparation part is also important. The developed GC-MS method is robust and reached the level

of sensitivity and reproducibility which is required by clinical testing. Hence, the method can be

incorporated into the daily routine testing of steroids.

5. Future work Future work can possibly explore the derivatization method, looking for effective ways to convert a ketone

group to a methoxy group. In this project my focus was to explore only 5 silylation reagents but in the

future one can also test more reagents. Examples would be looking at MSTFA, TMIS, DTE, (TMSI), HFBA

and TEA or mixing different reagents using different ratio to compatible more steroids which did not test

in this project.

39

6. Acknowledgments

My gratitude to all of you is beyond any words. First of all I would like to Professor Jonas Bergquist for

accepting me to work on this project. You gave me a lot of trusts and helps since the first day we met.

Thank you for trust and giving me the opportunity to switch my master program from another department

2 years ago. Your guidance and encouragement from the start to the end of this project is highly appreciated.

My sincere thanks is to my supervisor Dr. Kumari Ubhayasekara for the continuous guidance valuable

suggestions. You are really like my father as I mentioned before and I’m very lucky to be with you during

these months. You are really cared so much on my work, patience to my questions and explained everything

to me. My sincere thanks is also to Neil De Kock for your valuable helps, good suggestions and enjoyable

discussions during the lab experiments. Also I thank Warunika Aluthgedara, for her kindness, sweet and

valuable discussion. I would also like to thank all the staff at the analytical chemistry department for their

support and making my time fun here. Finally, I would like to give my deepest thanks to my father, for your

many years of enormous support to my study in Sweden. Last but not least, I would like to say thanks to

my boyfriend, Eugene, thank you for your love and support, for your understanding everything of me during

my study.

40

7. Popular scientific summary

Steroids are lipid compounds which can be synthesized from cholesterol. The steroids are mainly divided

into the following catogories: estrogen; androgen; glucocorticoid; progesterone; and mineralocorticoid. The

synthesis of these steroids occurs in the mitochondria and the smooth endoplasmic reticulum of cells in the

adrenal cortex, the gonads and the placenta. The adrenal cortex is mainly responsible for the production of

corticosteroid and androgen hormones which has three zones. Aldosterone and mineralocorticoids are

mainly produced in the outmost zone and zona glomerulosa. The middle zone, zona fasciculate, releases

glucocorticoids such as cortisol and cortisone into the body. Zona reticularis and the innermost zone is

responsible for producing androgens androstenedione (the precursor to testosterone (T)). Estrogens are

produced mainly by the ovaries and placenta with small amounts being produced by the adrenal glands.

These steroids with their metabolites are important in controlling human body functions and regulating the

nervous and immune systems. For example, estrogens could develop and maintain the female secondary

sex characteristics; androgens could develop and maintain the male secondary sex characteristics;

glucocorticoids could be acted as an immunosuppressant; progesterones could maintain the pregnancy and

regulate the cyclical changes of menstruation; and mineralocorticoids could regulate the blood pressure.

In recent years, steroids have became useful biomarkers for the diagnosis of several pathological conditions

and the concentration levels of steroids in the human body could indicate some disorders. Therefore it is

important to develop rapid and accurate analysis methods for steroids.

Currently, the main techniques for the analysis of steroids are immunoassays, liquid chromatography or gas

chromatography coupled to mass spectrometry. The most important requirements for the techniques of

steroid measurement are reliability, specificity and sensitivity. The technique of immunoassays might give

negative results due to cross-reactions, so this technique is not reliable enough. The tandem mass

spectrometry (MS/MS) technique can give reliable, specific and sensitive results, hence mass spectrometry

(MS) coupled with gas chromatography (GC) or high performance liquid chromatography (HPLC) is now

being commonly used in steroid hormone detection.

GC–MS/MS offers a higher resolution compared to LC–MS due to the longer analysis running time and

can analyse volatile compounds, so GC–MS/MS is a powerful tool that can be employed for the analysis

of steroids and is more sensitive than LC–MS. In this study, GC–MS/MS was used for the analysis of

steroids.

41

Normally, biological samples need extraction and an enrichment process for purification before injection

into the instrument. Routine sample preparation usually consists of liquid-liquid extraction (LLE), protein

precipitation and solid phase extraction (SPE). After comparing different techniques, the method of

ISOLUTE SLE + 96-well plates were shown to give high throughput, so this method was used for extraction

of steroids.

Most steroids are required to be derivatized for improving their volatility before injection into GC-MS.

Optimization of derivatizated conditions was investigated for enhancing the derivatization efficiency. In

this study, that samples reacting with 30µl methoxyamine hydrochloride at 60°C for 45min, then with 100

µL BSA at 120°C for 30 min was the optimized condition.

In order to accurately quantify the targeted steroids, the measured response must be correlated with the

concentration of the steroid hormones. Therefore, calibration curves need to be constructed. The calibration

range reflected the steroid concentration levels in the human body and the number of replicates was

sufficient to determine certain parameters and perform various statistical analyses (linearity, precision,

accuracy, sensitivity, stability, limit of detection, etc.). The linearity of the GC-MS method was evaluated

at a concentration range of 2 – 50 µg/ml. The correlation coefficient (R2) of cholesterol and estradiol ≥

0.99; both DHEA and cortisone's parameters ≥ 0.93. The recoveries for all the steroids determined at low,

medium and high concentration levels varied from 52.70 – 162.78 %. The concentration of cholesterol in

normal plasma is 34.82 ± 1.35µg/ml.

The aim of this study was to develop a sensitive, stable and rapid method for the analysis of steroids in

biological tissues by GC-MS, which allows also a wide range of steroids to be identified in a single sample.

The utilization of different derivatization methods during the sample preparation part is also important. The

developed GC-MS method is robust and reached the level of sensitivity and reproducibility which is

required by clinical testing. Hence, the method can be incorporated into the daily routine testing of steroids.

Future work can be done on the derivatization method, to find more effective derivatization methods

compatible with steroids which were not tested in this project.

42

8. References

1. Steroid Chemistry and Steroid Hormone Action. Chapter 1.

2. Synthia H. Mellon and Lisa D. Griffin. Neurosteroids: biochemistry and clinical significance.

Trends in Endocrinology & Metabolism, Volume 13, Issue 1, 1 January 2002, Pages 35-43.

3. M. Terqui a**, J.H.M. Wrathall b, M.A. Driancourt a, P.G. Knight b, Modulation of ovarian

function by steroid and inhibin immunization; Livestock Production Science 42 ( 1995) 181-192

4. Ziad J. Sahab, Suzan M. Semaan and Qing-Xiang Amy Sang, Methodology and Applications of Dis.

Paul J. Taylor,1,2*Donald P. Cooper,3Richard D. Gordon,1 and Michael Stowasser.

Measurement of Aldosterone in Human Plasma by Semiautomated HPLC–Tandem Mass

Spectrometry.ease Biomarker Identification in Human Seru. Biomarker Insights 2007: 2; 21-43.

5. Fouad R. Kandeel, Vivien K. T. Koussa and Ronald S. Swerdloff ; Male Sexual Function and Its

Disorders: Physiology, Pathophysiology, Clinical Investigation, and Treatment; Endocr Rev. 2001

Jun;22(3):342-88

6. Sadava D, Hillis DM, Heller HC, Berenbaum MR (2011). Life: The Science of Biology 9th Edition.

San Francisco: Freeman. pp. 105–114. ISBN 1-4292-4646-4.

7. "Estrone". OSHA. Retrieved 21 February 2013

8. Douma SL, Husband C, O'Donnell ME, Barwin BN, Woodend AK (2005). "Estrogen-related mood

disorders: reproductive life cycle factors". ANS Adv Nurs Sci 28 (4): 364–75.

9. Sicotte NL, Liva SM, Klutch R, Pfeiffer P, Bouvier S, Odesa S, Wu TC, Voskuhl RR (October 2002).

"Treatment of multiple sclerosis with the pregnancy hormone estriol". Ann. Neurol. 52 (4): 421–8.

10. Mo Q, Lu SF, Simon NG (April 2006). "Dehydroepiandrosterone and its metabolites: differential

effects on androgen receptor trafficking and transcriptional activity" . J. Steroid Biochem. Mol.

Biol. 99 (1): 50–8.

11. Bloch M1, Rubinow DR, Berlin K, Kevala KR, Kim HY, Schmidt PJ; Monoamines and

neurosteroids in sexual function during induced hypogonadism in healthy men; Arch Gen

Psychiatry. 2006 Apr;63(4):450-6.

12. Julien F. Blebuyck, M.B., D.Phil; The metabolic response to stress: an overview and update;

Anesthesiology 73: 308-327, 1990.

13. Mooradian AD, Morley JE, Korenman SG (February 1987). "Biological actions of androgens".

Endocr. Rev. 8 (1): 1–28.

14. "Prednisone and other corticosteroids: Balance the risks and benefits". MayoClinic.com. 2010-06-

05.

15. Steven J. Soldin1,2*and Offie P. Soldin1,3,4; Steroid Hormone Analysis by Tandem Mass

Spectrometry; Clin Chem. 2009 Jun;55(6):1061-6. doi: 10.1373/clinchem.2007.100008. Epub

2009 Mar 26.

16. Curnow KM, Tusie-Luna MT, Pascoe L, Natarajan R, Gu JL, Nadler JL, White PC (October 1991).

"The product of the CYP11B2 gene is required for aldosterone biosynthesis in the human adrenal

cortex.". Mol. Endocrinol. 5 (10): 1513–1522.

17. Ronald E. Majors, Wilmington,De; sample preparationfundamentals for chromatography

18. C.H. Shackleton, Genetic disorders of steroid metabolism diagnosed by mass spectrometry, in: N.

Blau, M. Duren, K.M. Gibson (Eds.), Laboratory guide to the methods in biochemical genetics, 1st

ed., Springer, Berlin Heidelberg, 2008, pp.549–605.

19. W.J.A. Vanden Heuvel, C.C. Sweeley, E.C. Horning, J. Am; sepration of steroids by gas

chromatography; Chem. Soc. 82 (1960) 3481.

20. Blau, J. Halket, Handbook of Derivatives for Chromatography, second edition, 1993, John Wiley

& Sons Ltd, England.

43

21. Maria Vittoria Cangialosia, Egidio Pucciaa, Antonio Mazzol b, Valentina Mansuet a, Augustine

Arukwec; Screening of ovarian steroidogenic pathway in Ciona intestinalis and its modulation

after tributyltin exposure; Toxicol Appl Pharmacol. 2010 May 15;245(1):124-33. doi:

10.1016/j.taap.2010.03.001. Epub 2010 Mar 7.

22. Sarah Combalbert & Marie-Laure Pype &Nicolas Bernet & Guillermina Hernandez-Raquet;

Enhanced methods for conditioning, storage, and extraction of liquid and solid samples of manure

for determination of steroid hormones by solid-phase extraction and gas chromatography–mass

spectrometry; Anal Bioanal Chem. 2010 Sep;398(2):973-84. doi: 10.1007/s00216-010-3954-7.

Epub 2010 Jul 28.

23. Homma K, Hasegawa T, Masumoto M, Takeshita E, Watanabe K, Chiba H, Kurosawa T, Takahashi

T, Matsuo N; Reference values for urinary steroids in Japanese newborn infants: gas

chromatography/mass spectrometry in selected ion monitoring; Endocr J. 2003 Dec;50(6):783-92.

24. Kai Fang, Xuejun Pan&, Bin Huang, Jingliang Liu, Yu Wang, Jianpei Gao; Simultaneous

Derivatization of Hydroxyl and Ketone Groups for the Analysis of Steroid Hormones by GC–MS;

Chromatographia, November 2010, Volume 72, Issue 9-10, pp 949-956.

25. Philippe Liere, 1,*Antoine Pianos,*Bernard Eychenne,* Annie Cambourg,* Karl Bodin,†William

Griffiths,†,2Michael Schumacher, * Etienne-Emile Baulieu,* and Jan Sjövall†; Analysis of

pregnenolone and dehydroepiandrosterone in rodent brain: cholesterol autoxidation is the key; J

Lipid Res. 2009 Dec;50(12):2430-44. doi: 10.1194/jlr.M900162-JLR200. Epub 2009 Jun 8.

26. Jin-Aa Oh1and Ho-Sang Shin2; Rapid determination of natural steroidal hormones in saliva for the

clinical diagnoses; Chemistry Central Journal (Impact Factor: 1.31). 03/2012; 6(1):22.

DOI:10.1186/1752-153X-6-22.

27. Martin Hansen. Naja W. Jacobsen. Frederik K. Nielsen. Erla nd Björklund; Determinatio of steroid

hormones in blood by GC–MS/MS; Anal Bioanal Chem. 2011 Jul;400(10):3409-17. doi:

10.1007/s00216-011-5038-8. Epub 2011 May 1.

28. Man Ho Choi and Bong Chul Chung*; GC-MS determination of steroids related to androgen

biosynthesis in human hair with pentafluorophenyldimethylsilyl–trimethylsilyl derivatisation;

Analyst. 1999 Sep;124(9):1297-300.

29. P. A. Voogt,* J. G. D. Lambert, J. C, M. Grannemantt and M, Jansen*; Confirmation of the

presence of oestradiol-17β in sea star Asterias rubens by GC-MS; Comp. Biochem. Physiol. Vol.

101B, No. 1/2, pp. 13-16, 1992.

30. Jun Chen, Qionglin Lianga,∗ , Haiyin Huaa, Yiming Wanga, Guo’an Luoa,∗ ∗ , Mingqiu Hub,

Yanqun Nab; Simultaneous determination of 15 steroids in rat blood via gas chromatography–

mass spectrometry to evaluate the impact of emasculation on adrenal; Talanta. 2009 Dec

15;80(2):826-32. doi: 10.1016/j.talanta.2009.07.066. Epub 2009 Aug 8.

31. Wilfried M.A. Niessen ; Current Practice of Gas Chromatography-Mass Spectrometry; volume 86;

32. Dorgan JF, Fears TR, McMahon RP, Aronson Friedman L, Patterson BH, Greenhut SF.

Measurement of steroid sex hormones in serum: a comparison of radioimmunoassay and mass

spectrometry. Steroids 2002;67:151– 8.

33. Poulson K, Sancho J, Haber E. A simplified radio-immunoassay for plasma aldosterone employing

an antibody of unique specificity. Clin Immunol 1974;2:373– 80.

34. Jonas Abdel-Khalik∗ , Erland Björklund1, Martin Hansen∗ , Simultaneous determination of

endogenous steroid hormones in human and animal plasma and serum by liquid or gas

chromatography coupled to tandem mass spectrometry; J Chromatogr B Analyt Technol Biomed

Life Sci. 2013 Jun 1;928:58-77. doi: 10.1016/j.jchromb.2013.03.013. Epub 2013 Mar 21.

35. Guo T, Chan M, Soldin SJ. Steroid profiles using liquid chromatography tandem mass spectrometry

with atmospheric pressure photoionization source. Arch Pathol Lab Med 2004;128:469 –75.

44

36. Nelson RE, Grebe SK, Okane DJ, Singh RJ. Liquid chromatography-tandem mass spectrometry as-

say for simultaneous measurement of estradiol and estrone in human plasma. Clin Chem 2004;

50:373– 84.

37. Paul J. Taylor,1,2*Donald P. Cooper,3Richard D. Gordon,1 and Michael Stowasser. Measurement

of Aldosterone in Human Plasma by Semiautomated HPLC–Tandem Mass Spectrometry; Clin

Chem. 2009 Jun;55(6):1155-62. doi: 10.1373/clinchem.2008.116004. Epub 2009 Apr 9.

38. B.G. Wolthers, G.P.B. Kraanb,1, Clinical applications of gas chromatography and gas

chromatography–mass spectrometry of steroids; Journal of Chromatography A, 843 (1999) 247–

274.

39. P. Magnisalia, M. Dracopouloub, M. Mataragasc, A. Dacou-Voutetakisb, P. Moutsatsoua,∗ .

Routine method for the simultaneous quantification of 17β-hydroxyprogesterone, testosterone,

dehydroepiandrosterone, androstenedione, cortisol, and pregnenolone in human serum of neonates

using gas chromatography–mass spectrometry; J Chromatogr A. 2008 Oct 10;1206(2):166-77. doi:

10.1016/j.chroma.2008.07.057. Epub 2008 Jul 25.

40. Stefan A. Wudya,∗ , Michaela Hartmanna, Janos Homokib; Determination of 11-deoxycortisol

(Reichstein’s compound S) in human plasma by clinical isotope dilution mass spectrometry using

benchtop gas chromatography–mass selective detection; Steroids Volume 67, Issue 10, September

2002, Pages 851–857.

41. Hans Jorg Leis , Gunter Fauler, Gerald N. Rechberger, Werner Windischhofer. Quantitative trace

analysis of estriol in human plasma by negative ion chemical ionization gas chromatography–mass

spectrometry using a deuterated internal standard; Journal of Chromatography B, 794 (2003)

205–213.

42. K.M. Trippa,1,*, M. Duboisb, P. Delahautb, J.P. Verstegena; Detection and identification of

plasma progesterone metabolites in the female Florida manatee ( Trichechus manatuslatiros tris)

using GC-MS/MS.Theriogenology,2009Aug;72(3):365-71.doi:10.1016/j.theriogenology.

43. Sati C. Chattoraj and Antonio Scommegna with the technical assistance of Leonard Rambow; A

gas chromatographic technique for the simultaneous determination of urinary pregnanediol and

pregnanetriol after ammonium sulfate precipitation; Steroids; Volume 9, Issue 3, March 1967,

Pages 327–346.

44. Liere,P., Y. Akwa , S. Weill-Engerer, B. Eychenne, A. Pianos,P. Robel,J. Sjövall, M. Schumacher ,

and E. E. Baulieu. 2000. Validation of an analytical procedure to measure trace amounts of

neurosteroids in brain tissue by gaschromatography-mass spectrometry; Journal of

Chromatography B, 739 (2000) 301–312.

45. Xu X, Duncan AM, Merz BE, Kurzer MS. Effects of soy isoflavones on estrogen and phytoestrogen

metabolism in premenopausal women; Cancer Epidemiol Biomarkers Prev. 1998; 7:1101–1108.

46. Jeffrey M. McGuire* and Phillip J. Powis. Gas Chromatographic Analysis of Tall Oil

Fractionation Products After Methylation with Ν,Ν-Dimethylformamide Dimethylacetal; Journal

of Chromatographic Science, Vol. 36, February 1998.

47. RAGNAR RYHAGE, Use of a Mass Spectrometer as a Detector and Analyzer for Effluents

Emerging from High Temperature Gas Liquid Chromatography Columns; Anal. Chem., 1964, 36

(4), pp 759–764, DOI: 10.1021/ac60210a019,Publication Date: April 1964.

48. Ivan Miksık , Martina Vylitova , Jirı Pacha , Zdenek Deyla; Separation and identification of

corticosterone metabolites by liquid chromatography–electrospray ionization mass spectrometry;

Journal of Chromatography B, 726 (1999) 59–69.

49. Turpeinen U, Markkanen H, Sane T, Hämäläinen E. Determination of free tetrahydrocortisol and

tetrahydrocortisone ratio in urine by liquid chromatography-tandem mass spectrometry; Scand J

Clin Lab Invest. 2006;66(2):147-59.

45

50. Raffaelli A, Saba A, Vignali E, Marcocci C, Salvadori P. Direct determination of the ratio of

tetrahydrocortisol+allo-tetrahydrocortisol to tetrahydrocortisone in urine by LC-MS-MS; J

Chromatogr B Analyt Technol Biomed Life Sci. 2006 Jan 18;830(2):278-85. Epub 2005 Nov 23.

51. Tatsuya Higashi∗ , Hiroyuki Yokoi, Hitoe Maekubo, Ayako Honda, Kazutake Shimada; Studies on

neurosteroids XXIII. Analysis of tetrahydrocorticosterone isomers in the brain of rats exposed to

immobilization using LC–MS; Steroids, Volume 72, Issue 13, November 2007, Pages 865–874.

52. N. Andrási,A. Helenkár,A. Vasanits-Zsigrai,Gy. Záray,I. Molnár-Perl∗; The role of the acquisition

methods in the analysis of natural and synthetic steroids and cholic acids by gas chromatography–

massspectrometry;J.ChromatogrA.2011Nov11;1218(45):8264-72.doi: 10.1016/j.chroma Epub

2011 Sep 14.

53. Jin Wei, Huang Bin, Wang Bin, Wang Dao-Wei, Zhao Shi-Min, Pan Xue-Jun*, Simultaneous

Determination of Androgens and Progestogen in Surface Water and Sediment by Gas

Chromatography-Mass Spectrometr; Chinese Journal of Analytical Chemistry 2013-2-14, 205-209

54. Nils Krone,[low asterisk] Beverly A. Hughes, Gareth G. Lavery, Paul M. Stewart, Wiebke Arlt, and

Cedric H.L. Shackleton; Gas chromatography/mass spectrometry (GC/MS) remains a pre-eminent

discovery tool in clinical steroid investigations even in the era of fast liquid chromatography

tandem mass spectrometry (LC/MS/MS); J Steroid Biochem Mol Biol. Aug 2010; 121(3-5): 496–

504.

55. Eunjung Han,Okkyoung Yim,Sun-young Beak; Determination of Adrenosterone and its Metabolites

in Human Urine by LC/APCI/MS and GC/MS; Bull. Korean Chem. Soc. 2009 , Vol. 30, No. 7,1489

56. S. AbuRuz, J. Millership, L. Heaney, J. McElnay, J. Chromatogr. B 798 (2003) 193.