Determination of levofloxacin in human urine with capillary electrophoresis and fluorescence detector

Article abstract of DOI:10.1002/jccs.200700142

In this study, we developed an analytical method for the enantioseparation of ofloxacin, using capillary electrophoresis with fluorescence detection. The optimum background electrolyte was obtained to be 60 mM hydroxylpropyl-β- cyclodextrin (HP-β-CD) in 50 mM phosphate buffer at pH 2.30. Under these conditions, the (+) and (-) ofloxacin were completely separated, with the detection limit of 10 nM when the sample was prepared in deionized water. The linear ranges of levofloxacin in deionized water and untreated urine were 10^-7 to 5 × 10^-3 M with R² = 0.9989 and 5 × 10^-6 to 5 × 10^-3 M with R² = 0.9943, respectively. We also applied this method to investigate the purity of a commercial drug. The results revealed that the ratio between (+)-ofloxacin and (-)-ofloxacin (levofloxacin) was 99.9:0.1, and there is about 93 mg levofloxacin per tablet (200 mg). The concentration of levofloxacin in patient's urine was founded to be 7.9 × 10^-4 M, and the ratio between the two optical isomers was 99.3:0.7.

Full text of DOI:10.1002/jccs.200700142

Journal of the Chinese Chemical Society, 2007, 54, 991-995
991
Determination of Levofloxacin in Human Urine with Capillary
Electrophoresis and Fluorescence Detector
a
Yu-Hsien Tsai (
b
), Ming-Jong Bair (
a,c
) and Cho-Chun Hu * (
)
a
Institute of Life Science, National Taitung University, 684 Chunghua Rd. Sec. 1,
Taitung, Taiwan 950, R.O.C.
b
Gastroenterology Division, Department of Internal Medicine, Mackay Memorial Hospital,
Taitung Branch, Taiwan, R.O.C.
c
Department of Natural Science Education, National Taitung University, 684 Chunghua Rd. Sec. 1,
Taitung, Taiwan 950, R.O.C.
In this study, we developed an analytical method for the enantioseparation of ofloxacin, using capil-
lary electrophoresis with fluorescence detection. The optimum background electrolyte was obtained to be
60 mM hydroxylpropyl-b-cyclodextrin (HP-b-CD) in 50 mM phosphate buffer at pH 2.30. Under these
conditions, the (+) and (-) ofloxacin were completely separated, with the detection limit of 10 nM when
the sample was prepared in deionized water. The linear ranges of levofloxacin in deionized water and un-
-
7
-3
2
-6
-3
2
treated urine were 10 to 5 ´ 10 M with R = 0.9989 and 5 ´ 10 to 5 ´ 10 M with R = 0.9943, respec-
tively. We also applied this method to investigate the purity of a commercial drug. The results revealed that
the ratio between (+)-ofloxacin and (-)-ofloxacin (levofloxacin) was 99.9:0.1, and there is about 93 mg
levofloxacin per tablet (200 mg). The concentration of levofloxacin in patient’s urine was founded to be
-
4
7
.9 ´ 10 M, and the ratio between the two optical isomers was 99.3:0.7.
Keywords: Levofloxacin; Capillary electrophoresis; Fluorescence.
INTRODUCTION
Levofloxacin is the levo-enantiomer of ofloxacin, a
than 5% of an administered dose was recovered in the urine
as the desmethyl and N-oxide metabolites, the only metab-
2
olites identified so far in humans. These metabolites have
member of the fluoroquinolones, and a class of synthetic
antimicrobial agents. The chemical name of ofloxacin is 9-
fluoro-2,3-dihydro-3-methyl-10-(4-methyl-1-piperazinyl)-
little relevant pharmacological activity.
The methods available for the analysis of levofloxacin
in biological fluids are based on high-performance liquid
3
chromatography, spectrofluorimetry, high-perform-
-9
10
7
-oxo-7H-pyrido[1,2,3-de]-1,4-benzoxazine-6-carboxylic
1
1
acid. Levofloxacin has found to be involved in inhibition of
bacterial topoisomerase IV and DNA gyrase which are im-
portant enzymes required for DNA replication, transcrip-
tion, repair and recombination. Levofloxacin is often used
for the treatment of infections of the respiratory and urinary
tracts, skin, and soft tissues.
ance thin layer chromatography and the electroanalytical
2
method. These methods have been used for the analysis of
levofloxacin in tablets, human urine, and serum. So far,
only a few chiral separations of ofloxacin by capillary elec-
trophoresis (CE) have been mentioned. One study de-
scribed the use of a g-CD-Zn(II)-D-phenylalanine complex
1
2
Pharmacological research has shown that the anti-
bacterial activity of the levo-enantiomer of ofloxacin ((-)-
ofloxacin or levofloxacin) is 8 ~ 128 times higher than that
to separate several new quinolone drugs. Alternatively,
neutral derivatized cyclodextrins (CDs) as chiral selectors
in micelle electrokinetic chromatography (MEKC) sys-
tems have demonstrated the chiral separation of oflox-
1
of (+)-ofloxacin. Levofloxacin is rapidly absorbed after
1
3-18
oral administration and undergoes limited metabolism in
humans. Approximately 87% of an administered dose was
recovered as unchanged drug in urine within 48 h, and less
acin.
Charged CDs were also used for the chiral separa-
tion of ofloxacin, with the advantages of greater speed at
low concentrations and the potential for the separation of
*
Corresponding author. Tel and Fax: +886-89-318855; E-mail: cchu@nttu.edu.tw

9
92 J. Chin. Chem. Soc., Vol. 54, No. 4, 2007
Tsai et al.
2
4,25
neutral racemates.
Recently, Ouyang’s group also used
prior to use. The capillary was regenerated by rinsing it
with running buffer for 3 min between every run. Samples
were injected with a pressure of 100 psi for 0.1 min. The
electrolyte was varied according to the experimental sec-
tions.
hydroxylpropyl-b-cyclodextrin (HP-b-CD) as chiral selec-
tor to measure the fluoroquinolone compounds spiked in
2
drug free urine by CE with UV-Vis absorption detection.
6
Capillary electrophoresis provides the advantages of
ease of automation, simplicity of operation, high speed,
high separation efficiency, and low consumption of sample
and reagents. The fluorescence detection has high sensitiv-
ity, with little matrix effect. In this study, we developed an
analytical method for the enantioseparation of ofloxacin,
using capillary electrophoresis with a fluorescence detec-
tor. We investigated the effects of various pH values, con-
centration of running buffer, and different concentrations
of HP-b-CD on the resolution. We also applied this method
to confirm the purity of a commercial drug and the concen-
tration of levofloxacin in patients’ urine.
Preparation of Samples
The urine was obtained from the patients in Mackay
Memorial Hospital (Taitung Branch, Taiwan). The patient
had been treated with 500 mg levofloxacin by oral adminis-
tration for three days. The urine samples were collected 4
hours after the last administration, stored at 4 °C and ana-
lyzed within 48 hours without any pretreatment. The com-
mercial drug at about 0.2 g per tablet was triturated, weighed
and dissolved in 10 mL of methanol. After vibration for 15
min and centrifugation for 2 min, the filtrate was collected
by passing through a 0.22 mm nylon filter and diluted by
methanol to 25 mL as a stock solution.
EXPERIMENT
Reagents and Materials
RESULTS AND DISCUSSION
Hydroxypropyl b-cyclodextrin, levofloxacin and
ofloxacin were from Aldrich (Madrid, Spain). All constitu-
ents of the buffer and chemical reagents were from Sigma
Effects of the concentration and pH value of phos-
phate buffer
(
St. Louis, MO) and were analytical grade.
The ionic strength is an important parameter for the
efficiency and resolution of the separation system. The ef-
fects of the phosphate buffer concentration from 20 to 100
mM on the migration time and resolution of the ofloxacin
were examined. We found that the resolution improved and
the migration time increased with increasing buffer con-
centration. Higher electrical currents (higher Joule heat)
were observed in the higher buffer concentration. Hence, a
phosphate buffer concentration of 50 mM was selected for
the following studies.
Urine was collected from patients of Mackay Memo-
rial Hospital (Taitung Branch, Taiwan). All solutions were
prepared with water purified by a Barnstead Ultrapure Wa-
ter System. Stock solutions of ofloxacin and levofloxacin
were prepared in methanol at the respective concentrations
of 1 mM and 10 mM. The solutions were then kept in a re-
frigerator for about two weeks.
Apparatus
The fluorescence spectra were taken with a Hitachi
F-4500 (Tokyo, Japan) spectrometer. Separation experi-
ments were carried out on a PRINCE 4-tray CE System.
The system consisted of a +30 to -30 kV high-voltage built-
in power supply, equipped with a fluorescence detector
The effect of different pH values on the fluorescence
intensity is shown in Fig. 1 when the excitation wavelength
was fixed at 292 nm. As the pH increased, the fluorescence
intensity increased and the maximum fluorescence shifted
from a long to a short wavelength (495-455 nm). However,
we also found that the migration time became longer and
the resolution improved with decreasing pH. Considering
the sensitivity, selectivity and analytical time, 50 mM phos-
phate buffer (pH 2.30) was chosen as optimal.
(
Jasco FP-1520, Tokyo, Japan), WPrince 32 bit software
for system controlling and SISC32 2.0 software (SIS Crop.,
Taiwan) for data collection. The separation capillary has an
internal diameter of 75 mm and outer diameter of 375 mm,
with 82 cm of total length, and 50 cm of effective length.
The effect of HP-b-CD
Electrophoresis Conditions
We tested a CE method using HP-b-CD as a chiral se-
lector for the separation of the two isomers of ofloxacin in
Capillaries were treated with 0.5N NaOH overnight

Determination of Levofloxacin in Human Urine
J. Chin. Chem. Soc., Vol. 54, No. 4, 2007 993
5
0 mM phosphate buffer (pH 2.30). HP-b-CD and two opti-
tion, the resolution improved and the migration times be-
came longer. Although 40 mM of HP-b-CD achieved a res-
olution value of about 2, we used 60 mM HP-b-CD in the
running buffer for the sake of accurate quantitative deter-
mination of ofloxacins. We note that commercial drugs
contain relatively small amounts of ofloxacins in (+) form
and large amounts in (-) from. The optimum background
electrolyte (pH 2.30) consists of 60 mM HP-b-CD and 50
mM phosphate. Under these conditions, the two isomers
were completely separated, with the detection limit of
levofloxacin in deionized water to be 10 nM. This analyti-
cal method using a fluorescence detector is 100-fold more
sensitive than using a UV detector. The linear range of
cal isomers of ofloxacin from host-guest complexes or in-
clusion complexes, leading to alternation in the mobility in
CE. The effects of the HP-b-CD concentrations from 20 to
8
0 mM on the resolution and migration times of ofloxacins
are shown in Fig. 2. With increasing HP-b-CD concentra-
-
7
-3
2
levofloxacin was from 10 to 5 ´ 10 M with R = 0.9989.
We also spiked levofloxacin into untreated urine from
healthy human patients. Even in this matrix, the linear
-
6
-3
range of levofloxacin was from 5 ´ 10 to 5 ´ 10 M with
2
R = 0.9943. The data are shown in Table 1. The signals of
levofloxacin in urine matrix were about 1 order less than
those in deionized water. We speculated that salts or other
chemical compounds in the urea caused the fluorescence
quenching of levofloxacin. As a result, the detection limit
in urine was about 10 fold higher than that in deionized wa-
ter.
Fig. 1. Fluorescence spectra of ofloxacin at different
pH values of the phosphate buffer. The concen-
-
4
tration of ofloxacin was 10 M and the excita-
tion wavelength was 292 nm.
Analysis of the commercial drug and patients’ urine
The results of using the optimized CE separation con-
ditions to analyze the commercial drug are shown in Fig. 3.
Because the concentration of the standard solution was
over the linear range, we diluted the sample with deionized
water. By comparing Fig. 3 (a) and (b), we found a small
-
5
peak on the right side of levofloxacin. After spiking 5 ´ 10
M ofloxacin into the standard solution, we identified the
small peak to be (+)-ofloxacin, an impurity of levofloxacin.
From the data of the calibration curve, we inferred that the
ratio between (+)-ofloxacin and S(-)-ofloxacin in the com-
mercial drug was 99.9:0.1. We also ascertained the exact
composition of levofloxacin in a tablet (200 mg) to be 93
mg.
This method was also used to investigate the quantity
of levofloxacin in patients’ urine. In Fig. 4(a), an electro-
pherogram from a healthy person’s urine is presented. Due
to an excellent selectivity of fluorescence detection, we ac-
quired a very clear and stable baseline. This verifies that
our analytical method can be employed in a real urine sam-
ple without matrix effect. The urine samples require no pre-
Fig. 2. Enantioseparation of ofloxacin in various HP-
b-CD and 50 mM phosphate buffer at pH 2.30.
(
a) 80 mM; (b) 60 mM; (c) 40 mM; (d) 20 mM.
Peak 1: R(+)-ofloxacin; peak 2: S(-)-ofloxacin
levofloxacin).
(

9
94 J. Chin. Chem. Soc., Vol. 54, No. 4, 2007
Tsai et al.
Table 1. Analytical parameters of the proposed method
Levofloxacin in deionized water
Levofloxacin in urine
-
7
-3
10 ~ 5 ´ 10 M
-6
5 ´ 10 ~ 5 ´ 10 M
-3
Linear range
LOD
-
10 M
8
-7
10 M
-
8
4 ´ 10 M
-7
4 ´ 10 M
LOQ
2
R
0.9985
0.9941
LOD, limit of detection (S/N = 3); LOQ, limit of quantification (S/N = 10)
Fig. 3. Electropherograms of the commercial drug. (a)
The stock solution was diluted with deionized
water ten times; (b) Drugs diluted with deionized
water and spiked with 50 mM ofloxacin. Run-
ning buffer: 50 mM phosphate buffer (pH 2.30)
and 60 mM HP-b-CD. Sample injection: 0.1
min at 100 psi; separation at 15 kV; fluores-
cence detection at lex = 292 nm and lem = 495
nm.
Fig. 4. Electropherograms of (a) urine from a healthy
person; (b) urine from a patient diluted 10 times
with deionized water; (c) urine from a patient
diluted 10 times with deionized water and spiked
with 50 mM ofloxacin. The other conditions were
the same as those given in Fig. 3.
CONCLUSION
treatment in this method. Fig. 4(b) and (c) show a patient’s
urine diluted 10 times with deionized water and the diluted
urine sample spiked with 50 mM ofloxacin, respectively.
By comparing Fig. 4(b) and (c), one can find two peaks in
patient’s urine, with the ratio between the two optical iso-
mers of about 99.3:0.7, which is a little higher than that in
the original drug. We speculated that these two optical iso-
mers have different anabolic rates in the human body or the
In this paper, we developed an analytical method for
the enantioseperation of ofloxacin, using CE with fluores-
cence detection. This method tolerates higher concentra-
tion of chiral selector in running buffer and discriminates
these two optical isomers that have large concentration dif-
ferences. The optimum background electrolyte (pH 2.30)
consists of 60 mM hydroxylpropyl-b-cyclodextrin and 50
mM phosphate. This analytical method using fluorescence
detection has less matrix effect and is 100-fold more sensi-
tive than that using UV detection. We believe that this method
is very convenient and accurate for the study of pharmaco-
kinetics as well as for routing examination.
(
-) isomer became racemic mixtures after being subjected
to human metabolism. After the quality was measured, the
concentration of levofloxacin in a patient’s urine collected
after 4 h on the third day of administration was found to be
-
4
about 7.9 ´ 10 M.

Determination of Levofloxacin in Human Urine
Received December 15, 2006.
J. Chin. Chem. Soc., Vol. 54, No. 4, 2007 995
Separat. Sci. 2003, 26, 1698.
2. Horimai, T.; Ohara, M.; Ichinose, M. J. Chromatogr. A 1997,
60, 235.
1
1
1
7
3. Fu, X.-Y.; Sun, C.-R.; Lu, J.-D.; Chen, Y.-Z. Chem. J. Chin.
Univ. Chin. 1997, 18, 1957.
REFERENCES
4. Chankvetadze, B.; Endresz, G.; Blaschke, G. Chem. Soc.
Rev. 1996, 25, 141.
1
. De Boer, T.; Mol, R.; De Zeeuw, R. A.; De Jong, G. J.;
Ensing, K. Electrophoresis 2001, 22, 1413.
1
5. Lelievre, F.; Gareil, P.; Jardy, A. Anal. Chem. 1997, 69, 385.
6. Luong, J. H. T.; Nguyen, A. L. J. Chromatogr. A 1997, 792,
2
3
. Radi, A.; El-Sherif, Z. Talanta 2002, 58, 319.
. Wright, D. H.; Herman, V. K.; Kinstantunides, F. N.;
Rotschafer, J. C. J. Chromatogr. B 1998, 709, 97.
. Wong, F.-A.; Juzwin, S.-J.; Flor, S.-C. J. Pharm. Biomed.
Anal. 1997, 15, 765.
1
4
31.
1
1
1
7. Lurie, I. S. J. Chromatogr. A. 1997, 792, 297.
4
5
6
7
8. Verleysen, K.; Sandra, P. Electrophoresis 1998, 19, 2798.
9. Fillet, M.; Hubert, P.; Crommen, J. Electrophoresis 1997,
. Liang, H.; Kays, M. B.; Sowinski, K. M. J. Chromatogr. B
1
8, 1013.
2
002, 772, 53.
2
2
0. Galaverna, G.; Corradini, R.; Dossena, A.; Marchelli, R.
Electrophoresis 1999, 20, 2619.
. Neckel, U.; Joukhadar, C.; Frossard, M; Jäger, W.; Müller,
M.; Mayer, B. X. Anal. Chim. Acta 2002, 463, 199.
. Djabarouti, S.; Boselli, E.; Allaouchiche, B.; Ba, B.; Nguyen,
A. T.; Gordien, J. B.; Bernadou, J. M.; Saux, M. C.; Breilh,
D. J. Chromatogr. B 2004, 799, 165.
1. Lurie, I. S.; Klein, R. F. X.; Dal Cason, G. A.; LeBelle, M. J.;
Brenneisen, R.; Weinberger, R. E. Anal. Chem. 1994, 66,
4
019.
2
2
2
2
2. Terabe, S. TrAC, Trends Anal. Chem. 1989, 8, 129.
8
. Nguyen, H. A.; Grellet, J.; Ba, B. B.; Quentin, C.; Saux, M.
C. J. Chromatogr. B 2004, 810, 77.
3. Wang, F.; Khaledi, M. G. Electrophoresis 1998, 19, 2095.
4. Tanaka, Y.; Terabe, S. J. Chromatogr. A 1997, 781, 151.
5. Fillet, M.; Hubert, P.; Crommen, J. J. Chromatogr. A 2000,
9
. Santoro, M. I. R. M.; Kassab, N. M.; Singh, A. K.; Kedor-
Hackmam, E. R. M. J. Pharm. Biomed. Anal. 2006, 40, 179.
8
75, 123.
1
0. Ocaña González, J. A.; Callejón Mochón, M.; Barragán de la
26. Zhou, S.; Ouyang, J.; Baeyens, W. R. G.; Zhao, H.; Yang, Y.
Rosa, F. J. Talanta 2000, 52, 1149.
J. Chromatogr. A 2006, 1130, 296.
11. Meyyanathan, S. N.; Ramasarma, G. V. S.; Suresh, B. J.