SPME-LC Fibers for a Variety
of Applications
Robert Shirey1, Craig Aurand1, Dajana Vuckovic2, Katherine Stenerson1,
Yong Chen1, Leonard Sidisky1, and Janusz Pawliszyn2
1Supelco,
Div. of Sigma-Aldrich, Bellefonte, PA 16823 USA
2Department of Chemistry, University of Waterloo, Waterloo, ON Canada N2L 3G1
www.sigma-aldrich.com
T410041
Abstract
An SPME fiber has been developed that is specifically designed for solvent
desorption. The fiber coating contains bonded HPLC particles embedded in a
proprietary binder that has biocompatible properties. This presentation
highlights the extraction of drugs such as carbamazepine and propranolol
from biological fluids at therapeutic levels. Low cost devices have been
developed that will allow single-use extractions for both in-vivo and in-vitro
applications.
2
Introduction
All of the commercially available SPME fibers contain coatings primarily
designed for thermal desorption and GC analysis. A new line of SPME fibers
has been developed specifically for solvent desorption with biocompatible
properties. Biocompatible fibers reduce surface adhesion of proteins and
other large macromolecules while allowing the extraction of small analytes.
Unlike the current multi-use fiber assemblies, these fibers are contained in
devices that will make single-use extractions cost effective. This presentation
highlights important parameters in the development of the coated fibers and
devices. A variety of applications showing the extraction and analysis of
small molecules from biological fluids will be presented.
3
Goals of Development of SPME Fibers
for Solvent Desorption
1. Fiber coating must be durable and reproducible
2. Fiber coating must not swell in water or organic solvents
3. Must be able to coat HPLC particles on fiber
4. Binder should not affect uptake of analytes
5. Binder should be biocompatible
a. Resists large macromolecules
b. Can be used with in-vivo type experiments without harming
organism
6. Device needs to be low cost for single use analysis
4
Experimental and Results
Fiber Coating - Process and Properties
• Silica particles (3 µm or 5 µm) are embedded in a biocompatible
proprietary binder
• Particles are coated on a durable, flexible 200 µm metal fiber using an
automated coating process (45 µm coating thickness variability 1-2%).
Findings
• Binder is inert and does not swell in water or organic solvents
• Binder does not impede extraction of small molecules
• Binder repels proteins and large macromolecules
5
Swelling of Fiber Coating upon
Solvent Exposure
Fibers soaked for 15 min. in each solvent
Bonded Silica Fiber
Coating thickness µm
Carbowax®-TPR
Coating thickness µm
15 min.
15 min.
in
in
Solvent
No Solvent Solvent Difference No Solvent Solvent Difference
Water
44
44
0
50
60
10
Acetonitrile
44
44
0
50
51
1
Methanol
44
44
0
50
61
11
Dichloromethane
44
44
0
50
52
2
Hexane
44
44
0
50
50
0
Acetone
44
44
0
50
50
0
Water:ACN
44
44
0
50
70
20
Water:MeOH
44
44
0
50
68
18
6
If swelling occurs, the coating can be more easily damaged and removed
from the fiber core, especially when a fiber is retracted into a needle.
The results show that SPME fibers developed with GC type phases are
subject to swelling in some solvents, but the newly developed fibers for
solvent desorption do not swell in most commonly used extraction solvents.
7
Extraction of Propranolol and
4-Hydroxypropranolol (4-HP) Metabolite
8
Procedure for Extraction and Analysis
Fiber Type
RPA Fiber, 45 µm coating thickness
Sample
500 µL and 100 µL, spiked phosphate buffer and rat plasma adjusted
to pH 4.0 with 25% H3PO4
Fiber Conditioning
15 min. in methanol, followed by 15 min. in water
Extraction
10 min., static
Desorption
60 min. in 100 µL 13 mM NH4OAc in 90:10 ACN:H2O
Instrument
Applied Biosystems 3200QT
Column
Discovery HS F5; 5 cm x 2.1 mm, 3 µm
Mobile Phase
2 mM ammonium formate in 90:10 acetonitrile:water
Flow
200 µL/min.
Temperature
35 °C
Injection Volume
5.0 µL
Source Conditions
Turbo ion spray ESI +, MRM
Q1 Mass (amu)
Propranolol: 260.21
4-hydroxypropranolol: 276.21
Q3 Mass (amu)
Propranolol: 183.00
4-hydroxypropranolol: 173.10
Dwell time
150 msec
®
9
HILIC Mode Separation of Propranolol
and 4-HP
XIC of +MRM (3 pairs): 260.2/183.0 amu from Sample 11 (042508011) of 042508.wiff (Turbo Spray)
Max. 3046.7 cps.
6.84
3000
Propranolol
2800
2600
2400
2200
2000
1800
1600
6.66
1400
1200
1000
800
4HP-metabolite
600
400
200
6.50
7.32
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Time, min
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
10
Linearity of Extractions of Propranolol
and 4-HP from 100 µL Samples
1.80E+05
1.60E+05
R2 = 0.998
4-HP: plasma
Propranolol: plasma
4-HP: buffer
Propranolol: buffer
1.40E+05
Response
1.20E+05
1.00E+05
R2 = 0.9588
8.00E+04
6.00E+04
R2 = 0.9847
4.00E+04
2.00E+04
R2 = 0.9087
0.00E+00
0
20
40
60
Conc. (ng/mL)
80
100
120
11
The results show that the analytes can be extracted out of a small volume of
either buffer or plasma with good linearity at low concentration levels. The
more polar metabolite is not extracted as efficiently as the parent drug. Also,
the recovery of both analytes is less out of plasma compared to buffer. This
is due to binding of the drugs to proteins in the plasma. It was shown that it
can take multiple hours for the binding equilibrium between the drugs and
protein to be met. If the extraction occurs immediately after spiking the
drugs into the plasma, the difference in recovery between buffer and plasma
is much smaller. Better linearity is obtained if full protein binding equilibrium
is obtained. An advantage of SPME is that it only extracts free (unbound)
drug that is therapeutically active.
12
LC-MS Analysis of Drugs in Plasma:
Comparison of SPME Extraction to Direct Injection
on the Matrix Background and Detection of the Drugs
Direct Injection after protein precipitation
T I C o f + M R M ( 1 p a i r ): E x p 1 , fr o m S a m p l e 4 ( 0 4 2 5 08 0 05 ) o f 0 4 2 5 0 8 . w if f ( T u r b o S p r a y )
M a x . 2 4 7 2 .0 c p s .
7 . 38
20 0 0
19 0 0
18 0 0
17 0 0
16 0 0
23 0 0
7 .6 2
MRM 184/104 for
Phospholipids
22 0 0
21 0 0
7 .2 3
20 0 0
19 0 0
7 .8 6
18 0 0
17 0 0
7 .1 4
16 0 0
8. 9 0
7 .0 8
14 0 0
9. 3 8
8 .7 6
14 0 0
13 0 0
8. 12
13 0 0
9 .5 6
12 0 0
12 0 0
11 0 0
11 0 0
5. 8 5
10 0 0
9 .6 9
10 0 0
4. 5 1
90 0
90 0
6. 1 7
80 0
9 .9 1
80 0
70 0
70 0
60 0
1 0 .1 2
6. 3 4
5 .5 7
50 0
4. 26
40 0
60 0
1 0 . 38
5 .4 6
50 0
40 0
1 0 .7 6
4. 11
30 0
30 0
20 0
1 1 .7 1
20 0
0
1 .1 1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
T im e , m i n
X IC o f + M R M
MRM 184/104 for
Phospholipids
15 0 0
15 0 0
10 0
M a x . 1 1 6. 0 c p s .
24 0 0
23 0 0
21 0 0
T I C o f + M R M ( 1 p a i r ): E x p 1 , fr o m S a m p l e 1 ( 04 2 5 0 8 0 0 1 ) o f 0 4 2 5 0 8 . w if f ( T ur b o S p r a y )
7. 52
24 7 2
24 0 0
22 0 0
SPME Extraction
( 3 p a i r s ) : E x p 2 , 2 6 0 . 2 / 1 8 3 . 0 a m u f ro m
S a m p l e 4 ( 0 4 2 5 0 8 0 0 5 ) o f 0 4 2 5 0 8 .w i ff ( T u rb o S p r a y )
M a x. 7 1 0. 0 cp s.
4 .0 5
5 . 13
6. 29 6 . 5 2 6 .7 1
10 0
7 .8 2 7. 9 2
5 .0 1
3 .4 6 3 . 8 5
1 .7 4
2 .1 0
0 .4 6 0 .7 1 1 .2 1
0
1
2
3
4
5
6
7
8
X IC o f + M R M ( 3 p a i r s ) : E x p 2 , 2 6 0 .2 /1 8 3 . 0 a m u f ro m S a m p l e 1 ( 0 4 2 5 0 8 0 0 1 ) o f 0 4 2 5 0 8 .w i ff ( T u Trbi m
o eS, pm
r ai yn )
8 .4 0 8. 76
9 .1 3
9
10 . 19
10
1 0 .6 5 1 1 .2 1
11
1 2 .1 6
12
12 .7 3
13
1 3 .4 4
1 3 .6 7 1 4 .8 6
14
M a x . 8 3 0. 0 c p s .
5 .8 9
83 0
80 0
6 .0 0
70 0
75 0
65 0
70 0
MRM 260./183.0 for
Propranolol,
276.21/173.1 for 4-HP
6 . 08
55 0
50 0
45 0
40 0
35 0
5 . 88
30 0
25 0
60 0
55 0
50 0
45 0
40 0
35 0
30 0
25 0
20 0
5. 84
20 0
15 0
15 0
10 0
10 0
5 .7 0
3 .6 4
50
6 .6 7
2. 52
0
1
2
3
4
5
6
6 .1 5
5 . 69
6 . 27
50
0
MRM 260./183.0 for
Propranolol,
276.21/173.1 for 4-HP
65 0
60 0
7
8
T im e , m in
9
10
11
12
13
14
1
2
3 .7 4
3
4 .3 5
4
6 .2 7
4 . 95
5
6
7
8
9
10
11
12
13
14
T im e , m in
13
Phospholipids (PL) are large molecules that can interfere with the analysis of
drugs by supressing ions, especially if the PL elute with the analytes of
interest. It is common to analyze drugs in plasma by precipitating proteins
followed by direct injection of the supernatant. Often not all of the PL
precipitate when acetonitrile is added to the plasma which can result in
supression specific for the drugs detected by the LC-MS system.
The above figures compare the analysis of the drugs and PL by precipitation
of proteins in plasma followed by direct injection, to extraction of untreated
plasma using SPME followed by desorption and analysis. The ion
chromatograms are at the same intensity levels. The results show that
SPME does not extract PL due to the biocompatible polymer, but the drugs
of interest are extracted at similar levels to direct injection that contained a
high concentration of PL.
14
Fiber Pipette Design and Use
15
A low cost device has been designed that contains a coated SPME fiber in a
disposable pipette tip. The tip can be inserted into low volume vials for
extraction and desorbed in 50 µL of solvent contained in a 100 µL conical
vial. Multiple samples can be extracted at one time and the desorption can
be accomplished in one step with multiple vials and fibers. The vials
containing the desorbed analytes can be placed in an autosampler for
analysis. It is possible to put tips in a plate format to simplify the extraction
process.
Reproducibility
Reproducibility of the area response of drugs extracted with 10 fiber tips per
lot, typically ranges between 3% and 9% relative standard deviation. Various
drugs used to evaluate the tips include cocaine, oxycodone, propranolol and
benzodiazepines along with their common metabolites. The variation was
calculated without correction with an internal standard.
16
Single Use Biocompatible Fiber Probes
for in-vivo Analysis
17
Probes have been designed so that blood could be sampled in-vivo from
animals with the use of a shunt device. Also, probes could be directly
inserted into plant and animal tissues for direct analysis of small
molecules.
18
Comparison of SPME in-vivo
Pharmokinetics (PK) Study of Carbamazepine (CBZ)
from Mice Whole Blood to Extracts of Plasma
Removed from Mice
CBZ Concentration
(ng/mL)
10000
SPME
1 mouse for all time
periods (triplicate runs )
Terminal blood draw
Plasma from 18 mice
1000
100
10
Slide Courtesy of
Ines de Lannoy-NoAb
BioDiscoveries.
1
0
60
120
180
Time (min)
240
300
19
The above slide shows the PK of CBZ in mice. A 2 mg/Kg dose of CBZ
was given to mice and the level of the drug in the blood stream is monitored
over time. The traditional way of monitoring the drug is to remove blood at
various time intervals usually 6-8 times over a 24 hour period followed by
extraction using SPE. Because mice have a very limited volume of blood,
the doped mouse is sacrificed between 2 and 3 extractions. In this study
triplicate samples are obtained at each time point; therefore, 18 mice were
sacrificed to obtain plasma. This is costly and the data are slightly skewed
because the metabolic rates can differ between mice.
The staff at NoAb BioDiscoveries developed a shunt device that passes the
blood of the mouse through the shunt containing a port that allows the
SPME fiber to be inserted for a 2 min. extraction of the drugs. Extraction of
drugs with the fibers can be done multiple times without harming the
mouse. For the SPME portion of this study 3 mice were used, with each
mouse surviving the entire timed study.
20
Conclusions
• SPME fiber coatings have been specifically designed for HPLC use and
bio-applications
• Fiber does not swell in water and/or solvents
• Fiber coating is biocompatible
• Fiber coating is durable and reproducible
• Fiber can be inserted into a pipette tip for easy handling and automation.
• Fiber probe is suitable for in-vivo and in-vitro applications
• Different coatings are being evaluated
21
Acknowledgements
Ines de Lannoy – In-vivo applications, NoAb Biodiscoveries
22