Chiral separations using a polypeptide and polymeric dipeptide surfactant polyelectrolyte multilayer coating in open-tubular capillary electrochromatography

Article abstract of DOI:10.1021/ac049313t

A polyelectrolyte multilayer (PEM) coating consisting of the polypeptide, poly(L-lysine) hydrobromide, poly(L-lysine) and the polymeric dipeptide surfactant, poly(sodium undecanoyl-L-leutcyl-alaninate), poly(L-SULA), is investigated as a new medium for the separation of chiral analytes in open-tubular capillary electrochromatography (OT-CEC). In this approach, a stable PEM is constructed in situ by alternative rinses of the cationic polymer poly(L-lysine) and the anionic polymer poly(L-SULA). In previous studies, the PEM coating has been constructed by use of the cationic polyelectrolyte poly (diallydimethylammonium chloride), PDADMAC. In this study, we investigate the use of a biopolymer as the cationic polyelectrolyte. The results reported here indicate an increase in selectivity and resolution when poly(L-lysine) is used as the cationic polymer in place of PDADMAC. To evaluate the chromatographic performance of the PEM coating as a chiral stationary phase, the separation of the β-blockers, labetalol and sotalol, and the binaphthyl derivatives, 1,1′-bi-2-naphthyl-2,2′-dihydrogen phosphate, 1,1′-bi-2- naphthol, and 1,1-binaphthyl-2,2′-diamine, are investigated. In addition, the effect of varying the amino acid order of the polymeric dipeptide surfactant on resolution is investigated. The number of bilayers also significantly influences the separation efficiency and resolution of enantiomers. The run-to-run and capillary-to-capillary reproducibilities are evaluated by calculating the relative standard deviations (RSDs) of the electroosmotic flow. These RSD values were found to be less than 1%. The coating is also stable and allows more than 290 runs to be performed in the same capillary. In addition, coupling of this chiral OT-CEC column with mass spectrometry is investigated.

Full text of DOI:10.1021/ac049313t

Anal. Chem. 2004, 76, 6681-6692
Chiral Separations Using a Polypeptide and
Polymeric Dipeptide Surfactant Polyelectrolyte
Multilayer Coating in Open-Tubular Capillary
Electrochromatography
Mary W. Kamande, Xiaofeng Zhu, Constantina Kapnissi-Christodoulou, and Isiah M. Warner*
Department of Chemistry, Louisiana State University, 434 Choppin Hall, Baton Rouge, Louisiana 70803
1
A polyelectrolyte multilayer (PEM) coating consisting of
the polypeptide, poly( -lysine) hydrobromide, poly( -ly-
sine) and the polymeric dipeptide surfactant, poly(sodium
undecanoyl- -leucyl-alaninate), poly( -SULA), is investi-
gated as a new medium for the separation of chiral ana-
lytes in open-tubular capillary electrochromatography
and the other enantiomeric form may be toxic. For this reason,
the Food and Drug Administration requires the separation of a
drug into its individual enantiomers and the examination of its
biological and toxicological effects before the drug can be
L
L
L
L
2
commercialized. Thus, the development of suitable methods for
the separation of the pure enantiomers of a drug compound is
important.
A number of separation techniques have been employed for
the resolution of chiral drug compounds. These include high-
performance liquid chromatography, gas chromatography, su-
percritical fluid chromatography, and capillary electrophoresis
CE). CE has emerged as a popular technique for the separation
of enantiomers due to its high efficiency, as well as versatile
applications in micellar electrokinetic chromatography (MEKC)
(
OT-CEC). In this approach, a stable PEM is constructed
in situ by alternative rinses of the cationic polymer poly-
-lysine) and the anionic polymer poly( -SULA). In previ-
(L
L
3
4
ous studies, the PEM coating has been constructed by
use of the cationic polyelectrolyte poly (diallydimethylam-
monium chloride), PDADMAC. In this study, we inves-
tigate the use of a biopolymer as the cationic polyelectro-
lyte. The results reported here indicate an increase in
5,6
7
(
8
and capillary electrochromatography (CEC).9
selectivity and resolution when poly(L-lysine) is used as
Chiral separations may be performed using a pseudostationary
the cationic polymer in place of PDADMAC. To evaluate
the chromatographic performance of the PEM coating as
a chiral stationary phase, the separation of the â-blockers,
labetalol and sotalol, and the binaphthyl derivatives, 1,1′-
bi-2-naphthyl-2,2′-dihydrogen phosphate, 1,1′-bi-2-naph-
thol, and 1,1-binaphthyl-2,2′-diamine, are investigated.
In addition, the effect of varying the amino acid order of
the polymeric dipeptide surfactant on resolution is inves-
tigated. The number of bilayers also significantly influen-
ces the separation efficiency and resolution of enantio-
mers. The run-to-run and capillary-to-capillary reproducibil-
ities are evaluated by calculating the relative standard
deviations (RSDs) of the electroosmotic flow. These RSD
values were found to be less than 1%. The coating is also
stable and allows more than 290 runs to be performed
in the same capillary. In addition, coupling of this chiral
OT-CEC column with mass spectrometry is investigated.
1
0,11
phase that consists of chiral selectors such as cyclodextrins,
antibiotics,12 crown ethers, linear polymers, and micelles.
13
14
15
Polymeric surfactants have also been used for the chiral separation
of charged and neutral enantiomers in MEKC.1
separations, polymeric surfactants are preferred over conventional
micelles due to their stability. Conventional micelles exhibit a
limited stability due to the presence of a dynamic equilibrium
6-19
In analytical
(
(
(
(
1) Caldwell, J. J. Chromatogr., A 1996, 719, 3-13.
2) http://www.fda.gov/cder/guidance/stereo.htm.
3) Taylor, D. R.; Maher, K. J. Chromatogr. Sci. 1992, 30, 67-85.
4) Schurig, V. J. Chromatogr., A 2001, 906, 275-299.
(5) Williams, K. L.; Sander, L. C. J. Chromatogr., A 1997, 785, 149-158.
(6) Terfloth, G. J. Chromatogr., A 2001, 906, 301-307.
(7) Ward, T. J. Anal. Chem. 2002, 74, 2863-2872.
(8) Palmer, C. P. Electrophoresis 2000, 21, 4054-4072.
(9) Wistuba, D.; Schurig, V. Electrophoresis 2000, 21, 4136-4158.
10) Wang, J.; Warner, I. M. J. Chromatogr., A 1995, 711, 297-304.
11) Ozaki, H.; Ichihara, A.; Terabe, S. J. Chromatogr., A 1995, 709, 3-10.
12) Fanali, S.; Desiderio, C. HRC-J High Resolut. Chromatogr. 1996, 19, 322-
(
(
(
326.
(13) Kuhn, R. Electrophoresis 1999, 20, 2605-2613.
(
14) Maichel, B.; Potocek, B.; Gas, B.; Chiari, M.; Kenndler, E. Electrophoresis
The enantioselectivity of chiral compounds is of great impor-
tance to the pharmaceutical industries. This is because a large
number of pharmaceutical drugs exhibit chirality, and as a result,
one enantiomeric form may exhibit a desired physiological effect
1
998, 19, 2124-2128.
(15) Pascoe, R. J.; Peterson, A. G.; Foley, J. P. Electrophoresis 2000, 21, 2033-
2042.
(
(
16) Palmer, C. P. J. Chromatogr., A 1997, 780, 75-92.
17) Muijselaar, P. G.; Otsuka, K.; Terabe, S. J. Chromatogr., A 1997, 780, 41-
61.
*
Corresponding author. Tel.: (225) 578-3945. Fax: (225) 578-3971. E-mail:
(18) Otsuka, K.; Terabe, S. J. Chromatogr., A 2000, 875, 163-178.
(19) Wang, J.; Warner, I. M. Anal. Chem. 1994, 66, 3773-3776.
iwarner@lsu.edu.
1
0.1021/ac049313t CCC: $27.50 © 2004 American Chemical Society
Analytical Chemistry, Vol. 76, No. 22, November 15, 2004 6681
Published on Web 10/19/2004

SULA), demonstrated a higher selectivity and a higher resolution
for the separation of binaphthyl derivatives.29 One of the major
drawbacks associated with the use of polymeric surfactants in
MEKC is the tendency to foul the ionization source in mass
3
0
spectrometry (MS). To solve this problem, Zhu et al. have
recently studied the use of polyelectrolyte multilayer (PEM)-
coated capillaries in open-tubular capillary electrochromatogra-
phy/mass spectrometry (OT-CEC/MS).
An alternative approach to MEKC is the use of OT-CEC, which
was first introduced by Mayer and Schurig.31 They achieved chiral
separation of 1,1′-bi-2-naphthyl-2,2′-dihydrogen phosphate and
1
-phenylethanol by use of capillaries coated with immobilized
32,33
Chiral-Dex. Katayama et al.
introduced a simple coating
procedure that utilized the PEM coating approach for the
performance of achiral separations in OT-CEC. In general, a PEM
coating is formed by alternately exposing a cationic and an anionic
polyelectrolyte on a hydrophilic surface.3
4,35
The mechanism of
formation of PEMs occurs via the ion exchange process that
results in a stable coating.3
6-38
Recently, our laboratory has
investigated the use of PEM coatings for the separation of a
number of chiral and achiral analytes. These studies illustrated
Figure 1. Structures of (a) poly(L-lysine), (b) PDADMAC, (c) poly-
L-SULA), and (d) poly(L-SUAL).
(
3
9-41
an excellent reproducibility and a remarkable stability.
Al-
though this technique has a low phase ratio, selectivity can be
enhanced by increasing the number of bilayers and the salt
between individual surfactant molecules and the micelle. Thus,
organic solvents cannot be used in the background electrolyte
4
1
concentration added to the polymer deposition solutions. Thus,
in separations, the most commonly used cationic polyelectrolytes
for the construction of PEMs have been poly(diallydimethylam-
(BGE) as they interfere with micelle formation. Unlike polymeric
surfactants, conventional micelles require high surfactant con-
centrations above the critical micelle concentration in order to
provide efficient separations.
monium chloride) (PDADMAC)3
9-43
or Polybrene.
32,33,44,45
The use
of polypeptides in OT-CEC for the construction of PEMs is new.
However, it is an important step for the chiral recognition of
enantiomers as these systems mimic the biomembranes in the
human body. Rmaile and Schlenoff46 have recently reported the
Polymeric dipeptide surfactants have been traditionally em-
2
0-25
ployed for chiral separations in MEKC.
selectivity than the single amino acid polymeric surfactants due
They provide a higher
2
0
to the possession of two chiral centers. Shamsi et al. were the
first to report the use of a polymeric dipeptide surfactant. The
authors observed a higher peak efficiency, resolution, and faster
elution, when the polymeric dipeptide surfactant poly(sodium
use of optically active polypeptides consisting of poly(L-lysine) and
poly(glutamic acid) for chiral recognition of ascorbic acid in OT-
CEC. Their work illustrated remarkable permeability and chiral
selectivity.
undecanoyl-L-leucylvalinate), poly(L-SULV), was used. This was
compared to the chromatographic performance of alprenolol,
(
(
(
(
(
(
29) Billiot, F. H.; Billiot, E. J.; Warner, I. M. J. Chromatogr., A 2001, 922, 329-
338.
propanolol, and 1,1′-bi-2-naphthyl-2,2′-dihydrogen phosphate, when
30) Zhu, X.; Kamande, M. W.; Thiam, S.; Kapnissi, C. P.; Mwongela, S. M.;
Warner, I. M. Electrophoresis 2004, 25, 562-568.
31) Mayer, S.; Schurig, V. HRC-J. High Resolut. Chromatogr. 1992, 15, 129-
poly(sodium undecanoyl-L
-valinate) was used. Billiot et al.2
2,26-28
examined several factors that govern enantiomeric recognition
in polymeric dipeptide surfactants. In their studies, Shamsi et al.24
131.
32) Katayama, H.; Ishihama, Y.; Kajima, T.; Asakawa, N. Chromatography 1998,
19, 244-245.
33) Katayama, H.; Ishihama, Y.; Asakawa, N. Anal. Chem. 1998, 70, 2254-
demonstrated that poly(
was capable of resolving various molecular classes of compounds.
Although poly( -SULV) was shown to be a versatile chiral
discriminator, poly(sodium undecanoyl- -leucylalinate), poly(
L-SULV) was a broad chiral selector that
2260.
L
34) Decher, G. Science 1997, 277, 1232-1237.
L
L
-
(35) Decher, G.; Hong, J. D.; Schmitt, J. Thin Solid Films 1992, 210, 831-835.
36) Dubas, S. T.; Schlenoff, J. B. Langmuir 2001, 17, 7725-7727.
(
(
20) Shamsi, S. A.; Macossay, J.; Warner, I. M. Anal. Chem. 1997, 69, 2980-
987.
21) Haddadian, F.; Billiot, E. J.; Shamsi, S. A.; Warner, I. M. J. Chromatogr., A
999, 858, 219-227.
(37) Lowack, K.; Helm, C. A. Macromolecules 1998, 31, 823-833.
(38) Dubas, S. T.; Schlenoff, J. B. Macromolecules 1999, 32, 8153-8160.
(39) Kamande, M. W.; Kapnissi, C. P.; Zhu, X. F.; Akbay, C.; Warner, I. M.
Electrophoresis 2003, 24, 945-951.
2
(
1
(
22) Billiot, E.; Warner, I. M. Anal. Chem. 2000, 72, 1740-1748.
(40) Kapnissi, C. P.; Akbay, C.; Schlenoff, J. B.; Warner, I. M. Anal. Chem. 2002,
74, 2328-2335.
(23) Haddadian, F.; Shamsi, S. A.; Warner, I. M. Electrophoresis 1999, 20, 3011-
3
026.
24) Shamsi, S. A.; Valle, B. C.; Billiot, F.; Warner, I. M. Anal. Chem. 2003, 75,
79-387.
(41) Kapnissi, C. P.; Valle, B. C.; Warner, I. M. Anal. Chem. 2003, 75, 6097-
6104.
(42) Graul, T. W.; Schlenoff, J. B. Anal. Chem. 1999, 71, 4007-4013.
(43) Liu, Y.; Fanguy, J. C.; Bledsoe, J. M.; Henry, C. S. Anal. Chem. 2000, 72,
5939-5944.
(
3
(
(
25) Haynes, J. L.; Warner, I. M. Rev. Anal. Chem. 1999, 18, 317-382.
26) Billiot, E.; Macossay, J.; Thibodeaux, S.; Shamsi, S. A.; Warner, I. M. Anal.
Chem. 1998, 70, 1375-1381.
(44) Katayama, H.; Ishihama, Y.; Asakawa, N. Anal. Chem. 1998, 70, 5272-
5277.
(
27) Billiot, E.; Agbaria, R. A.; Thibodeaux, S.; Shamsi, S.; Warner, I. M. Anal.
Chem. 1999, 71, 1252-1256.
28) Billiot, F. H.; Thibodeaux, S.; Shamsi, S.; Warner, I. M. Anal. Chem. 1999,
(45) Bendahl, L.; Hansen, S. H.; Gammelgaard, B. Electrophoresis 2001, 22,
2565-2573.
(
7
1, 4044-4049.
(46) Rmaile, H. H.; Schlenoff, J. B. J. Am. Chem. Soc. 2003, 125, 6602-6603.
6682 Analytical Chemistry, Vol. 76, No. 22, November 15, 2004

Figure 2. Chiral separation of three binaphthyl derivatives: (a) BNP; (b) BOH; (c) BNA. Conditions: 4 bilayers; 0.02% (w/v) poly(L-lysine) with
.5 M NaCl and 0.25% (w/v) poly(L-SULA); BGE, 100 mM Tris and 10 mM Na2B4O7 (pH 10.2), 10% methanol was added to the BGE for the
separation of BNA; pressure injection, 30 mbar for 5 s; applied voltage, 30 kV; current, varied from 20 to 27 µA; temperature, 15 °C; capillary,
0
57 cm (50-cm effective length) × 50 µm i.d.; detection, 220 nm.
In this study, we report the use of a PEM coating that consists
EXPERIMENTAL SECTION
of the polypeptide poly(
L
-lysine) and the polymeric dipeptide
-SUAL) for the chiral separation
Reagents and Chemicals. Poly(L-lysine) hydrobromide (MW
) 150 000-300 000) and PDADMAC (MW )150 000-300 000)
surfactant poly( -SULA) or poly(
L
L
of three binaphthyl derivatives and two â-blockers. Several
experimental parameters are varied in order to optimize the
separation conditions. The coating exhibits remarkable reproduc-
ibility and stability, and it endures over 290 runs. In addition, the
coupling of chiral OT-CEC to MS by the use of the PEM-coated
capillary is reported for the first time.
were obtained from Sigma (St. Louis, MO). The structures of poly-
(L-lysine) and PDADMAC are illustrated in Figure 1a and b,
respectively. The pure enantiomers and the racemic mixtures of
binaphthyl derivatives 1,1′-bi-2-naphthyl-2,2′-dihydrogen phosphate
(BNP), 1,1′-bi-2-naphthol (BOH), and 1,1′-binaphthyl-2,2′-diamine
(BNA), as well as labetalol and sotalol, were purchased from Sigma
Analytical Chemistry, Vol. 76, No. 22, November 15, 2004 6683

Figure 3. Comparison of separation of BNP using (a) 0.02% (w/v) PDADMAC PEM coating or (b) 0.02% (w/v) poly(L-lysine) Conditions: 1
bilayer; 0.25% (w/v) poly(L-SULA); BGE, 100 mM Tris and 10 mM Na2B4O7 (pH 10.2); pressure injection, 30 mbar for 5 s; applied voltage, 30
kV; current 32 µA; temperature, 15 °C; capillary, 57 cm (50-cm effective length) × 50 µm i.d.; detection, 220 nm.
(St. Louis, MO). Sodium borate, tris(hydroxymethyl)aminomethane
(50-µm i.d., 320-µm o.d.) were purchased from Polymicro Tech-
(Tris), hexylamine, methanol (MeOH), cyclohexylaminopropane-
nologies (Phoenix, AZ). The effective length of the capillaries was
sulfonate (CAPS), and sodium chloride were obtained from Fisher
Scientific (Fair Lawn, NJ). Ultrapure grade ammonium acetate
was purchased from Amresco Inc. (Solon, OH). Chemicals used
50 cm, while the total length was 57 cm.
OT-CEC/ESI-MS Instrumentation. A Hewlett-Packard3D CE
instrument (Palo Alto, CA) with an electrospray ionization time-
of-flight mass spectrometer (ESI-TOFMS) Mariner Biospectrom-
etry Workstation from Applied Biosystem (Framingham, MA) was
employed for the chiral OT-CEC/ESI-MS experiments. All experi-
ments were performed at room temperature. The sheath flow OT-
CEC/ESI-MS interface was constructed by use of a CE-ESI sprayer
from Agilent Technologies (Palo Alto, CA). The sprayer is capable
of providing both a coaxial sheath liquid and a nebulization gas
to assist the electrospray. The capillary tip was set at an angle of
for the synthesis of the polymeric dipeptide surfactant poly(
SULA) and poly( -SUAL), N,N-dicyclohexylcarbodiimide, unde-
cylenic acid, and N-hydroxysuccinimide were purchased from
Sigma. The dipeptides -leucylalaninate and -alanylleucinate were
L-
L
L
L
purchased from BaChem Bioscience Inc (King of Prussia, PA).
All chemicals were used as received.
Synthesis of Poly(
L
-SULA) and Poly(L-SUAL). The struc-
tural representations of poly(L-SULA) and poly(L-SUAL) are shown
in Figure 1c and d, respectively. The surfactant monomers of
sodium undecanoyl leucine alaninate (mono-SULA) and sodium
undecanoyl alanine leucinate (mono-SUAL) were synthesized from
the N-hydroxysuccinimide ester of undecylenic acid with the
respective dipeptide according to the procedure reported by Wang
4
5° relative to the direction of the ESI-TOFMS nozzle in order to
obtain an optimum signal. The capillary tip was extended 0.5 mm
from the sprayer tip. For all OT-CEC/ESI-MS experiments, the
length of the poly(L-SULA)-coated capillary was 61 cm. The sheath
liquid was delivered at a flow rate of 2 µL/min. The ESI voltage
applied on the sprayer was set at 2.85 kV in the positive mode.
Data acquisition was performed in the range of m/z 100-1000 at
a scan rate of 3 s/spectrum. The nozzle voltage was set at 150 V,
and the skimmer voltage was set at 12 V. The nebulizer and
curtain gases were both nitrogen, and the flow rates were
optimized at 0.5 and 0.7 L/min, respectively.
1
9
and Warner. A 100 mM sodium salt solution of the monomer
6
0
was then polymerized by Co-γ radiation.
OT-CEC Instrumentation. OT-CEC experiments were per-
3
D
formed using an Hewlett-Packard CE system (Walbronn, Ger-
many). The instrument was equipped with a diode array detector
(DAD) for UV detection. In this study, the DAD was set at 220
nm for the detection of all analytes. The instrument also consisted
of a 0-30-kV high-voltage built-in power supply, and Hewlett-
Packard CE Chemstation software was used for control and data
acquisition. Fused-silica capillaries employed in all experiments
Preparation of Background Electrolyte and Analyte Solu-
tions. The BGE for the OT-CEC experiments consisted of 100
mM Tris and 10 mM Na B O at pH 10.2 or 300 mM CAPS, 50
2 4 7
6684 Analytical Chemistry, Vol. 76, No. 22, November 15, 2004

Figure 4. Effect of column temperature on the OT-CEC separation of BNP enantiomers. Conditions: 1 bilayer; 0.02% (w/v) poly(L-lysine) and
.25% (w/v) poly(L-SULA); BGE, 100 mM Tris and 10 mM Na2B4O7 (pH 10.2); pressure injection, 30 mbar for 5 s; applied voltage, 30 kV;
0
capillary, 57 cm (50-cm effective length) × 50 µm i.d.; detection, 220 nm; (a) temperature, 45 °C; current, 56 µA; (b) temperature, 35 °C;
current, 43.7 µA; (c) temperature, 25 °C; current, 38.1 µA; (d) temperature, 15 °C; current 32 µA.
mM Na
2
B
4
O
7
, and 0.15% (v/v) hexylamine at pH 8.5. The pH of
filtered through 0.45-µm polypropylene nylon syringe filters and
then sonicated for 15 min to ensure proper degassing before use.
The BGE was used to rinse the capillary for 2 min between runs.
Analyte solutions were dissolved in either 50:50 or 80:20 methanol/
water in order to give a final concentration of 0.1 mg/mL.
both BGEs was adjusted using 1 M sodium hydroxide or 1 M
phosphoric acid. The BGE for the OT-CEC/ESI-MS experiments
consisted of 25 mM ammonium acetate, and the pH was adjusted
with ammonium hydroxide solution. All BGE solutions were
Analytical Chemistry, Vol. 76, No. 22, November 15, 2004 6685

Figure 5. Effect of voltage on the migration time and resolution of BOH. Conditions: 4 bilayers; 0.02% (w/v) poly(L-lysine) and 0.25% (w/v)
poly(L-SULA); 100 mM Tris and 10 mM Na2B4O7 (pH 10.2); pressure injection, 30 mbar for 5 s; BGE, temperature, 15 °C; capillary, 57 cm
(
50-cm effective length) × 50 µm i.d.; detection, 220 nm. (a) applied voltage, 30 kV; current, 27.4 µA (b) applied voltage, 20 kV; current, 18.3
µA (c) applied voltage, 15 kV; current, 13.7 µA.
Calculations. Resolution (R
ficiency (N) were calculated using the following equations,
s
), selectivity (R), and peak ef-
where t
0
and t
r
are the respective migration times of the neutral
and w are the peak
marker (MeOH) and the enantiomer, w
1
2
widths of each enantiomer at the baseline, and w1/2 is the peak
width at half-height.
Procedure for Polyelectrolyte Multilayer Coating. All coat-
ing procedures were performed using the rinse function on the
HP3D CE system. Polymer deposition solutions consisted of
2
(t - t )
r2 r1
R )
(1)
s
w + w
1
2
t2 - t
0
R )
(2)
(3)
t1 - t
0
0
.02% (w/v) poly(
L
-lysine) with NaCl concentration ranging from
-SULA) or 0.25% (w/
2
N ) 554[t /w1/2]
r
0 to 0.5 M and either 0.25-0.5% (w/v) poly(
L
6686 Analytical Chemistry, Vol. 76, No. 22, November 15, 2004

Figure 6. (a) Effect of bilayer number on the chiral separation of BNP. (b) Effect of increasing the number of bilayers on the resolution of BNP,
labetalol, BOH, and BNA. (c) Effect of increasing the number of bilayers on efficiency and selectivity of BNP. Conditions: 4 bilayers; 0.02%
(w/v) poly(L-lysine) with 0.5 M NaCl and 0.25% (w/v) poly(L-SULA)); BGE, 100 mM Tris and 10 mM Na2B4O7 (pH 10.2); pressure injection, 30
mbar for 5 s; applied voltage, 30 kV; current, 27.4 µA temperature, 15 °C; capillary, 57 cm (50-cm effective length) × 50 µm i.d.; detection, 220
nm.
v) poly(
was conditioned with 1 M NaOH for 30 min, followed by deionized
water for 15 min. Next, the capillary was flushed with poly( -lysine)
for the deposition of the first cationic layer. A 5-min rinse with
deionized water was followed. The anionic polymeric dipeptide
L
-SUAL) with 0-0.5 M NaCl added. First, the capillary
RESULTS AND DISCUSSION
In OT-CEC, the use of poly( -SULA) required the optimization
L
L
of several parameters in order to achieve baseline chiral separa-
tions. In this study, experimental parameters such as voltage,
temperature, number of bilayers, salt concentration in the polymer
deposition solutions, and type of chiral selector were investigated.
Enantiomeric Separation of Binaphthyl Derivatives. The
three binaphthyl derivatives, BNP, BOH, and BNA, are referred
to as atropisomers because they possess a chiral plane of
symmetry rather than an asymmetric carbon. The separation of
these analytes using the PEM coating, as well as chemical
structures of each, is illustrated in Figure 2. BNP, which is anionic
at the given experimental conditions, eluted at a shorter time and
gave the highest resolution of the three binaphthyl derivatives
surfactant (poly-(
min followed by a 5-min rinse with deionized water. The cationic
and anionic layers constitute one bilayer. Consecutive bilayers
were constructed by alternate 5-min rinses of poly( -lysine) and
L-SULA) or poly(L-SULA)) was then flushed for
5
L
polymeric dipeptide surfactant in order to obtain the desired
number of bilayers. After each polyelectrolyte deposition, the
capillary was rinsed for 5 min with deionized water.
For the OT-CEC/ESI-MS experiments the coating protocol was
similar to the one described above. A 61-cm-long capillary was
prepared using a four-bilayer coating that consisted of 0.02% (w/
(Figure 2a). A possible explanation for this observation is that
BNP is the least hydrophobic, and as such, it does not penetrate
deeply into the hydrophobic core of the polymeric dipeptide
v) poly(L-lysine) and 0.25% (w/v) poly(L-SULA).
Analytical Chemistry, Vol. 76, No. 22, November 15, 2004 6687

Figure 7. Effect of variation of amino acid order on the polymeric dipeptide surfactant on separation of BNP enantiomers. Conditions: 4
bilayers; 0.02% (w/v) poly(L-lysine) with 0.5 M NaCl and 0.25% (w/v) poly(L-SUAL); pressure injection, 30 mbar for 5 s; BGE, 100 mM Tris and
1
0 mM Na2B4O7 (pH 10.2); applied voltage, 30 kV; current, 24 µA; temperature, 15 °C; capillary, 57 cm (50-cm effective length) × 50 µm i.d.;
detection, 220 nm.
surfactant.28 Figure 2b illustrates the separation of BOH under
while the peak efficiency increased. On the other hand, when a
15-kV voltage was applied, the resolution increased. It is apparent
that at lower voltages the analytes interact more with the PEM
coating, and this results in an increase in resolution. Similar
observations were made when BNA and labetalol were separated
under similar experimental conditions (data not shown).
conditions similar to those for Figure 2a. BOH, which has a pK
a
value of 9.5, is partially anionic under the given experimental
conditions. The reduction in peak efficiency, as compared to that
of BNP, may be attributed to the higher hydrophobicity of BOH.
The separation of BNA (Figure 2c) required the use of the organic
modifier methanol because it is the most hydrophobic and is
neutral under the experimental conditions. In this case, the organic
modifier solvates the solute, thus reducing hydrophobic interac-
tions with the polymeric surfactant.
Bilayer Studies. A linear relationship between the thickness
of the coating and the number of layers has been illustrated by
Dubas and Schlenoff.38 In our study, the effect of the number of
bilayers on the resolution and selectivity of BNP, BOH, BNA, and
labetalol was investigated. The coatings, which were constructed
for these experiments, consisted of one, two, three, or four
bilayers. As mentioned, a bilayer consists of a layer of a cationic
and an anionic polymer. The results are shown in Figure 6a.
Figure 6b illustrates the increase in the enantiomeric resolution
of the above analytes as the number of bilayers increases. As the
number of bilayers increased, both resolution and selectivity
increased, while the peak efficiency decreased. Figure 6c il-
lustrates the decrease in the theoretical plates of BNP and the
increase in selectivity as the number of bilayers increases. In this
study, the one-bilayer coating gave the highest number of
theoretical plates for all four analytes. Thus, these analytes
interacted less with the one-bilayer coating and caused less tailing.
The peak efficiencies of the BNP enantiomers were 312 679 and
318 706, while those of labetalol diastereoisomers were 256 248
and 262 394 in the one-bilayer coating.
A comparative study was performed to investigate the separa-
tion of BNP using PDADMAC (Figure 3a) or poly(
Figure 3b) as the cationic polymers employed for the construction
of the PEM coating. As illustrated in Figure 3, a one-bilayer
coating, which was constructed by using poly( -lysine) as the
L-lysine)
(
L
cationic polymer, was sufficient to give a baseline separation of
BNP in less than 10 min. On the other hand, no resolution was
obtained when PDADMAC was used, and the elution time of BNP
was more than doubled. It is possible that poly(L-lysine) enhances
chiral selectivity because it is chiral while PDADMAC is achiral.
In addition, the elution of the hydrophobic analyte BNA was not
achieved when PDADMAC was used as the cationic polyelectro-
lyte. Based on the results, PDADMAC forms a more hydrophobic
PEM coating, which in turn, increases the elution time of the
analytes.
Effect of Temperature. The influence of temperature on the
separation of BNP enantiomers was investigated in Figure 4. In
this study, the temperature was varied from 15 to 45 °C. As
expected, a decrease in temperature resulted in an increase in
retention time. This is probably due to a decrease in electrolyte
viscosity upon increasing the temperature. The resolution of BNP
gradually increased with a decrease in temperature. The optimal
temperature for this separation was shown to be 15 °C since it
yielded the best resolution and the minimal Joule heating.
Effect of Voltage. The influence of the applied voltage on
resolution, migration time, and efficiency was also investigated
for the separation of BOH. Figure 5 illustrates the effect of
increasing the applied voltage from 15 to 30 kV on the separation
of BOH. As expected, higher voltages decreased the elution times.
At 30 kV, the elution time of BOH enantiomers decreased and
Variation of Amino Acid Order on Polymeric Dipeptide
Surfactant on Chiral Recognition. The influence of the amino
acid order in the polymeric surfactant on enantiomeric separations
was also investigated. For the purpose of this study, the polymeric
surfactants poly(
anionic polyelectrolytes in the construction of the PEM coating.
In poly( -SULA), alanine is the outside (C-terminal) amino acid,
while in poly( -SUAL), it is the inside (N-terminal) amino acid.
Figure 7 illustrates the separation of BNP using poly( -SUAL).
The resolution obtained with poly( -SUAL) was less than 1.5, while
the resolution with poly( -SULA) was 2.5 (Figure 2a). A similar
observation was made when BOH was separated with poly(
SUAL). When the chiral selector poly( -SULA) was used, both
resolution and selectivity increased. A possible explanation for this
L-SULA) and poly(L-SUAL) were used as the
L
L
L
L
L
L-
L
6688 Analytical Chemistry, Vol. 76, No. 22, November 15, 2004

Figure 8. Effect of NaCl concentration on the chiral separation of BNP. Conditions: 4 bilayers; pressure injection, 30 mbar for 5 s; BGE, 100
mM Tris and 10 mM Na2B4O7 (pH 10.2); applied voltage, 30 kV; temperature, 15 °C; capillary, 57 cm (50-cm effective length) × 50 µm i.d.;
detection, 220 nm; (a) 0.02% (w/v) poly(L-lysine) with 0 M NaCl and 0.25% (w/v) poly(L-SULA) with 0 M NaCL; current, 24.8 µA; (b) 0.02% (w/v)
poly(L-lysine) with 0.0 M NaCl and 0.25% (w/v) poly(L-SULA) with 0.5 M NaCL; current, 26.4 µA; (c) 0.02% (w/v) poly(L-lysine) with 0.5 M NaCl
and 0.25% (w/v) poly(L-SULA) with 0.0 M NaCl; current, 27.4 µA.
observation is that BNP and BOH preferentially interact with the
inside (N-terminal) amino acid in each of the polymeric dipeptide
gated (Figure 8). Figure 8a illustrates the chiral separation of BNP
enantiomers obtained when no NaCl was used in the polymer
solutions. Baseline resolution at a shorter elution time was
obtained. A slight improvement in resolution and an increase in
electroosmotic flow (EOF) were observed when 0.5 M NaCl was
added to the anionic polymer solution (Figure 8b). Figure 8c was
obtained when 0.0 and 0.5 M NaCl were added to the cationic
and anionic polymer solutions, respectively. A significant increase
in resolution and a decrease in EOF were observed. Based on
the above results, it is possible that the addition of NaCl to the
2
7
surfactants. Thus, the two analytes interact preferentially with
the chiral center of leucine in poly( -SULA) and the chiral center
of alanine in poly( -SUAL). Therefore, the decrease in resolution
with poly( -SUAL) may be due to the steric hindrance of the butyl
L
L
L
group in leucine.
Effect of Sodium Chloride on Selectivity. There are a
number of parameters that influence the amount of polymer
deposited on a layer during the construction of the PEM coating.
These parameters include the molecular weight of the polymer,
the deposition time, and the concentration of the salt in the
polymer deposition solution. Among these factors, salt has been
found to have the greatest influence on the thickness of the
poly(
than when NaCl is added to the poly(
variation in polymer properties between poly(
-SULA) results in a variation in the degree of thickness when
L
-lysine) solution increases the thickness of the coating more
-SULA) solution. The
-lysine) and poly-
L
L
(L
3
8
coating. For the purpose of this study, the effect of the amount
of NaCl in either the cationic or the anionic polyelectrolyte on
the separation of different enantiomeric compounds was investi-
NaCl is added to each polymer deposition solution. In addition,
when NaCl was added to both cationic and anionic polymer
solutions, the stability of the current was difficult to maintain. This
Analytical Chemistry, Vol. 76, No. 22, November 15, 2004 6689

Figure 9. Illustration of a run-to-run reproducibility for the chiral separation of labetalol. Conditions: 4 bilayers; 0.02% (w/v) poly(L-lysine) and
.25% (w/v) poly(L-SULA); BGE, 100 mM Tris and 10 mM Na2B4O7 (pH 10.2); pressure injection, 30 mbar for 5 s; applied voltage, 30 kV;
current 26.5 µA; temperature, 15 °C; capillary, 57 cm (50-cm effective length) × 50 µm i.d.; detection, 220 nm.
0
may have been due to a change in the structure of the PEM
coating or an increase in film thickness that can clog the capillary.
Based on these observations, higher separation resolution of
binaphthyl derivatives can be achieved when NaCl is only added
to the cationic polyelectrolyte solution. Studies are ongoing in our
laboratory to investigate the thickness and the structure of the
PEM coating when NaCl concentration in the PEM coating is
varied.
Column Reproducibility and Stability Studies. Both repro-
ducibility and stability are very important factors for evaluating
the performance of the coating. Figure 9 is an illustration of the
remarkable run-to-run reproducibility of the 5th, 10th, and 15th
6690 Analytical Chemistry, Vol. 76, No. 22, November 15, 2004

Figure 10. Chiral separation of sotalol. Conditions: 4 bilayers; 0.5% (w/v) poly(L-lysine)) and 0.5% (w/v) poly(L-SULA) with 0.1 M NaCl;
pressure injection, 30 mbar for 3 s; BGE, 300 mM CAPs and 50 mM Na2B4O7 (pH 8.5), 0.15% hexylamine; applied voltage, 20 kV; current 50.6
µA; temperature, 15 °C; capillary, 57 cm (50-cm effective length) × 50 µm i.d.; detection, 220 nm.
Separation of â-Blockers. Labetalol is a â-blocker drug that
is used for the treatment of hypertension.47 This compound has
two chiral centers and therefore four diastereoisomers (four
peaks). However, only two diastereoisomers were resolved in this
work (Figure 9). The separation of sotalol is illustrated in Figure
1
0. Sotalol elutes before the EOF since it is cationic under the
experimental conditions used in this study. The second peak,
which is of low efficiency, is typical for the chiral separation of
â-blockers. This may be due to the higher affinity of one of the
enantiomers to the chiral PEM coating. The use of hexylamine
in the mobile phase has been shown to improve the chiral
selectivity of â-blockers.48 Therefore in this study, 0.15% (v/v)
hexylamine was added to the BGE in order to improve the
resolution between the enantiomers.
Coupling of MS to OT-CEC. The use of the PEM coating
with MS is beneficial for the coupling of MEKC to MS. In this
study, we investigated the use of the PEM-coated capillary for
the chiral separation of labetalol. A number of parameters were
varied in order to make the PEM-coated capillary amenable to
coupling with MS. The separations performed for this study
required the use of a purely aqueous electrophoresis medium.
Thus, ammonium acetate was used as the BGE. In addition, the
total capillary length was increased to 61 cm in order to facilitate
the coupling of the Hewlett-Packard3 CE instrument to the MS.
The total ion chromatogram for the separation of labetalol
diastereoisomers is illustrated in Figure 11. Although a baseline
separation was not obtained, it was possible to identify the parent
ions of the diastereoisomers in the mass spectrum.
Figure 11. Total ion chromatogram for separation of labetalol
diastereoisomers. Conditions: 4 bilayers; 0.02% (w/v) poly(L-lysine)
and 0.25% (w/v) poly(L-SULA); BGE, 10 mM NH4OH (pH 9.0);
pressure injection, 50 mbar for 5 s applied voltage, 20 kV; capillary,
61 cm × 50 µm i.d.;
runs for the separation of labetalol. In this study, a BGE of 100
mM Tris and 10 mM Na (pH 10.2) was used, and no NaCl
2
4 7
B O
was added to the polymer deposition solutions. These results were
reproducible in five different capillaries with run-to-run RSD values
D
(n ) 5) of 0.92, 0.43, 0.93, 0.70, and 0.85%. To assess the capillary-
to-capillary reproducibility, five capillaries were prepared under
similar conditions. The relative standard deviation (RSD) values
of the EOF were computed from five consecutive runs for each
coated capillary. All values were found to be less than 1%. The
endurance of the coating was tested by setting the instrument to
perform a consecutive number of runs in the same capillary over
a period of 5 days. The capillary was conditioned with a fresh
BGE after every 15 runs. The PEM-coated capillary endured over
CONCLUSIONS
A novel PEM coating has been applied for chiral separations
in OT-CEC. Several factors have been shown to affect the chiral
(47) Wong, S. H. Y. Therapeutic Drug Monitoring and Toxicology by Liquid
Chromatography; Chromatographic Science Series 32; Dekker: New York,
2
90 runs. It was also observed that the capillaries coated with no
1985.
NaCl added to the polymer solutions gave the best reproducibility
of both EOF and analyte migration times.
(48) Mangelings, D.; Hardies, N.; Maftouh, M.; Suteu, C.; Massart, D. L.; Vander
Heyden, Y. Electrophoresis 2003, 24, 2567-2576.
Analytical Chemistry, Vol. 76, No. 22, November 15, 2004 6691

selectivity. The presence or absence of NaCl in each of the
polyelectrolytes has been shown to play a significant role in the
selectivity of the chiral PEM coating. A four-bilayer coating was
found to be optimal for the separation of the three binaphthyl
derivatives BNP, BOH, and BNA. The separation of BNP enan-
tiomers and labetalol diastereoisomers resulted in high peak
efficiencies with theoretical plates of over than 250 000. Run-to-
run and capillary-to-capillary reproducibility were very good, and
the RSD values were less than 1%. The coating endured over 290
runs. In addition, the mass spectrometric detection of labetalol is
accomplished by coupling OT-CEC to MS. Future studies will
investigate the application of the PEM-coated capillary on the
separation and mass spectrometric detection of more chiral
analytes.
ACKNOWLEDGMENT
I.M.W. acknowledges the National Institutes of Health and the
Philip W. West Endowment for their support of these studies.
Received for review May 10, 2004. Accepted September 1,
2004.
AC049313T
6692 Analytical Chemistry, Vol. 76, No. 22, November 15, 2004