Fluorescence probe enhanced spectrofluorimetric method for the determination of gatifloxacin in pharmaceutical formulations and biological fluids

Article abstract of DOI:10.1016/j.saa.2007.04.026

A spectrofluorimetry for the determination of gatifloxacin (GFLX) was developed based on the strong fluorescence of gatifloxacin after adding fluorescence probe yttrium in buffer solution (pH 7.0) and various factors of influencing fluorescence have been researched. Under the optimum conditions, the liner range was 4.00 × 10^-8 to 1.00 × 10^-6 g mL^-1 and the detection limit is 3.36 × 10^-9 g mL^-1 (correlation coefficient r = 0.9997), respectively. The relative standard deviation was 1.1% for 11 measurements of 5.6 × 10^-7 g mL^-1 gatifloxacin standard solution. The mechanism of sensitizing effect of probe was discussed. The proposed method has been successfully applied to determine real samples and the obtained results are in good agreement with the results of HPLC.

Full text of DOI:10.1016/j.saa.2007.04.026

Available online at www.sciencedirect.com
Spectrochimica Acta Part A 69 (2008) 478–482
Fluorescence probe enhanced spectrofluorimetric method
for the determination of gatifloxacin in pharmaceutical
formulations and biological fluids
Xiashi Zhu∗, Aiqin Gong, Suhai Yu
College of Chemistry & Chemical Engineering, Yangzhou University, Yangzhou 225002, China
Received 16 July 2006; accepted 25 April 2007
Abstract
A spectrofluorimetry for the determination of gatifloxacin (GFLX) was developed based on the strong fluorescence of gatifloxacin after adding
fluorescence probe yttrium in buffer solution (pH 7.0) and various factors of influencing fluorescence have been researched. Under the optimum
conditions, the liner range was 4.00 × 10−8 to 1.00 × 10−6 g mL−1 and the detection limit is 3.36 × 10−9 g mL−1 (correlation coefficient r = 0.9997),
respectively. The relative standard deviation was 1.1% for 11 measurements of 5.6 × 10−7 g mL−1 gatifloxacin standard solution. The mechanism
of sensitizing effect of probe was discussed. The proposed method has been successfully applied to determine real samples and the obtained results
are in good agreement with the results of HPLC.
© 2007 Published by Elsevier B.V.
Keywords: Gatifloxacin; Spectrofluorimetry; Probe
1. Introduction
etc. These methods were limited to widely use because of expen-
sive instruments cost or low sensitivity.
Gatifloxacin { -1-cyclopropyl-6-fluoro-7-(3-methyl-1-pi-
peraziny)-8-methoxy-1,4-dihydro-4-oxo-3-quinolinecarboxy-
lic acid sesquihydrate} is a new broad-spectrum 8-methoxy
fluoroquinolone antibacterial agent. Its molecular formula is
C19H22FN3O4·1.5H2O and its molecular weight is 402.23. The
structure of GFLX is shown in Fig. 1.
The new fluoroquinolone, which has shown activity against
both Gram (+), Gram (−) and anaerobic species, behaves
excellent pharmacokinetic and pharmacodynamic characters. Its
negative action is very small and even no photoactive reaction
[1–3]. Thus, a rapid, low cost, and accessible procedure for
the quantification of gatifloxacin in dosage forms is desirable.
Various techniques have been utilized for the determination
of the fluoroquino-lone, among them HPLC is the most fre-
quently used technique [4,5]. Other analytical methods are
spectrophotometry (UV) [6], capillary electrophoresis [7] and
fluorospectrophotometry [1,3,8], FIA-chemiluminescence [9],
Probes are some small organics, inorganic ions, metal com-
plex ions and quantum points. The spectrophotometric or
electrochemical characters of the analytes would change when
they form extra-molecular complexes with probes, therefore it
could provide concentrative or structural information of the ana-
lytes or enhance sensitivity of determination [10]. Nowadays
there are many cases to use probes in assaying pharmaceuti-
cals. Literature [11] studied the reaction of oxyfloxacin with
DNA using Tb as probe. There was also a paper [12] that used
methylene blue as probe to research the reaction of p-(N,Nꢀ-
dimethyl aminomethyl) calix [8] arene with DNA. Some papers
have reported to determine gatifloxacin by spectrofluorimetry
without probes [1,8], but the sensitivity was low. The sensiti-
vity of assay gatifloxacin could be enhanced in the medium of
SDS micelle solution [3]. Fluorescence probe enhanced spectro-
fluorimetric method for the determination of GFLX seems to be
lacking. In this paper, Y3+ as probe, spectrofluorimetric method
for the determination of GFLX is investigated. To the best of
our knowledge, this is the first attempt to determine GFLX
in pharmaceutical formulations by fluorescence probe enhan-
ced spectrofluorimetry. The main advantages of the proposed
∗
Corresponding author. Tel.: +86 514 7975244; fax: +86 514 7975244.
E-mail addresses: xszhu@yzu.edu.cn, zhuxiashi@sina.com (X. Zhu).
1386-1425/$ – see front matter © 2007 Published by Elsevier B.V.
doi:10.1016/j.saa.2007.04.026

X. Zhu et al. / Spectrochimica Acta Part A 69 (2008) 478–482
479
treated serum and urine samples were spiked with convenient
amounts of GFLX stock solutions. Human serum and urine were
kindly provided by healthy volunteers.
2.4. Determination of gatifloxacin
Aliquots of working solutions or sample solutions of gati-
floxacin, 3.0 mL of acetic acid–ammonium acetate buffer (pH
7.0), 1.0 mL of Y2O3 were pipetted in to 25 mL calibrated flasks
and diluted to the mark with water. The obtained solutions were
thermostated at 20.0 0.5 ◦C and the fluorescence intensity was
measured, using an excitation wavelength of 292 nm, against a
blank solution.
Fig. 1. The structure of gatifloxacin.
methods compared with previously reported are their simplicity
and lower cost. The detection limits are comparable or better
than the detection limits of the other methods [1,3,8].
2.5. Determination of relative fluorescence quantum yields
Fluorescence quantum yields of GFLX and GFLX-Y3+ were
measured using 1.0 × 10−6 g mL−1 quinine sulfate as reference
material. Under the same apparatus conditions, according to the
equation ϕ2 = (ϕ1A1F2)/(F1A2) [13], the quantum yields of the
analyte was calculated. Briefly, ϕ1 and ϕ2 are corresponding the
standard and unknown fluorescence quantum yield and F1 and
F2 the integral areas of two calibration fluorescence emission
curves, A1 and A2 the absorbance (λabsorbance = λemission) of the
standard and unknown and ϕ1 = 0.55 (25 ◦C) is known.
2. Experimental
2.1. Chemicals
Gatifloxacin was kindly provided by Jinghua institution.
Gatifloxacin capsule and injection were from Biocause (Hunai,
China) and Suzhong (Jiangsu, China) pharmaceutical industries,
respectively. Acetic acid, ammonium acetate, hydrochloric acid
and Y2O3 were purchased from reagent company of Guoyao
(Shanghai, China). All chemicals were of analytical reagent
grade. Water was distilled, deionised.
3. Results and discussion
Gatifloxacin stock standard solution of 2.0 × 10−4 g mL−1
was prepared by dissolving 0.0100 g of gatifloxacin in 50 mL
of 0.1 mol L−1 HCl and kept in the dark. Working standard of
2.0 × 10−5g mL−1 was prepared daily by dilution of stock stan-
dard solution with high-purity water. Buffer solution (pH 7.0)
was prepared by dissolving 77.0 g ammonium acetate in 500 mL
high purity water and was adjusted by the addition of acetate
acid. The solution of Y3+ was prepared by dissolving 0.1270 g
Y2O3 in 1 mL of HCl and diluted to 100 mL with water.
3.1. The fluorescence spectra of gatifloxacin
The fluorescence spectra of gatifloxacin in pH 7.0 buffer solu-
tion are shown in Fig. 2. Comparing the curves, it was found the
fluorescence intensity was enhanced about one time when using
4.0 × 10−5 g mL−1 Y3+ as probe.
3.2. The selection of probes
The selection of probes is very important in spectrofluorime-
try. In this work, probes such as aluminum, terbium, europium,
2.2. Apparatus
An Hitachi F-4500 spectrofluorimeter (Japan) was used for
all the measurement, with excitation and emission slits at 2.5 nm,
λexc = 292 nm and 1 cm quartz cell. The pH was measured on a
pH S-25 pH-meter (Shanghai).
2.3. Sample preparation
Thesolutionofgatifloxacincapsuleandgatifloxacininjection
were prepared by dissolving suitable amount of the commercial
samples in 0.1 mol L−1 HCl and diluted the resulting solution
to adjust the concentration with water to that required by the
experimental conditions adopted.
A 1.0 mL serum sample was deproteinized by adding 4.0 mL
10% trichloroacetic acid (CCl3COOH) in a centrifuge tube,
whichwasthencentrifugedfor15 minat4000 rpm. Thesuperna-
tant was diluted with deionized water to suitable concentration.
No further pre-treatment was required for urine samples. The
Fig. 2. Fluorescence spectra of GFLX (GFLX: 4.0 × 10−7 g mL−1).

480
X. Zhu et al. / Spectrochimica Acta Part A 69 (2008) 478–482
Fig. 3. Effect of different probe on F (GFLX: 4.0 × 10−7 g mL−1).
Fig. 5. Effect of Vbuffer solution on ration of fluorescence intensity (GFLX:
4.0 × 10−7 g mL−1).
lanthanum, erbium, yttrium were studied (Fig. 3). The influence
of probes on fluorescence intensity (F) could be divided two
kinds. One was to enhance gatifloxacin fluorescence inten-
sity such as aluminum, lanthanum or yttrium. The other was
to quench fluorescence such as terbium, europium or erbium.
Maximum fluorescence intensity was observed in the present of
yttrium. The F was approximately one time higher in the pre-
sence of probe Y3+ than in the absence of Y3+. Hence, yttrium
was the probe of choice.
3.4. Effect of probe Y3+ concentration
The influence of Y3+ concentration on the fluorescence inten-
sity was studied by changing the concentration of Y3+. It was
found that 4.0 × 10−5 g mL−1 Y3+ was appropriate for maxi-
mum fluorescence intensity (Fig. 6).
3.5. Effect of temperature
3.3. The effect of pH
Temperature is an important factor in spectrofluorimetry. As
expected, when the temperature is increased the fluorescence
intensity is lowered slowly. In this work the fluorescence inten-
sity varied very small about 20 ◦C, and this temperature was
easily kept constant, the temperature of 20 ◦C (Fig. 7) was selec-
ted.
The influence of pH on the fluorescence intensity of gati-
floxacin was investigated. As could be seen in Fig. 4, the ration
of fluorescence intensity with or without Y3+ (F1/F2) reaches a
maximum value at pH 7.0. Moreover, yttrium would be hydro-
lysised after pH > 8.0. Thus, an acetic acid–ammonium acetate
buffer of pH 7.0 was chosen for the determination.
3.6. Effect of ionic strength
The effect of the value of buffer solution was also studied.
Fig. 5 shows that the F1/F2 remains stable for a buffer value of
2.0–4.0 mL. So a 3.0 mL buffer solution was selected as suitable
for the optimized method (Fig. 5).
The fluorescence intensity remained constant with increasing
KCl concentrations up to 0.48 mol L−1; after that, the fluores-
Fig. 4. Effect of pH on the ration of fluorescence intensity (GFLX:
4.0 × 10−7 g mL−1).
Fig. 6. Effect of Y3+ concentration on F (GFLX: 4.0 × 10−7 g mL−1).

X. Zhu et al. / Spectrochimica Acta Part A 69 (2008) 478–482
481
ability of translating absorption energy to fluorescence. In this
paper, the fluorescence quantum yields of GFLX and GFLX-Y3+
were determined, respectively. The results were listed in Table 1.
The quantum yield Φf was approximately 33% times higher in
the presence of probe Y3+ than in the absence of Y3+. The Φf
value was tightly related to chemical structure and microenvi-
ronment of the system. The Φf value is higher, the ability of
translating absorption energy to fluorescence is stronger, it is
also the reason of enhanced determination sensitivity by Y3+.
3.8. Effect of potential interferences
The effects of different foreign substrates were discussed in
the determination of 4.0 × 10−7 g mL−1 gatifloxacin. The level
of tolerated concentrations of foreign ions was considered as
maximum concentration found to cause a change in signal, less
than 5%, compared with the signal for gatifloxacin alone. The
tolerance limits are shown in Table 2.
Fig. 7. Effect of temperature on F (GFLX: 4.0 × 10−7 g mL−1).
3.9. Analytical performance
The calibration graph for the determination of gatifloxacin
was obtained under the experimental conditions above descri-
bed. The results showed that a good linear relationship was
observed over the range 4.00 × 10−8 to 1.00 × 10−6 g mL−1
.The
calibration
curve
was
F = −1.268 + 129.765C
(10−6 g mL−1), r = 0.9997. The detection limit estimated
(S/N = 3) was 3.36 × 10−9 g mL−1
.
3.10. Analysis of pharmaceutical formulations
The proposed method was applied for the determination of
GFLXinrealsamples. Thestandardadditionmethodwasusedto
avoid matrix effects and the recoveries are presented in Table 3.
The average recoveries obtained by the proposed method were
between 97.1 and 103.8% (Table 3).
Fig. 8. Effect of ionic strength on F (GFLX: 4.0 × 10−7 g mL−1).
cence intensity decreased. That is, lower ionic strength has no
effect on F, while higher ionic strength F decreased, this is
because the effect of the electrostatic shielding of charges would
reduce the binding of GFLX and Y3+ and result in a decreased
signal (Fig. 8).
Table 2
Tolerance limits of interfering ions (mg mL−1
)
Tested ions
Tolerance limit
3.7. Discussion of mechanism of sensitizing effect
Ca2+
0.08
0.08
0.24
0.24
0.12
0.16
0.2
0.004
0.0032
0.0032
0.2
Mg2+
Ba2+
The fluorescence quantum yields of (Φf) in various media
were determined, respectively, in order to discuss the influence
of the microenvironment on the fluorescence intensity of GFLX.
The results are listed in Table 1. The fluorescence quantum yield
was one of the mostly basic and significant parameters in all
the characters of fluorescence substance [14]. It represented the
Zn2+
Cd2+
Hg2+
Al3+
Eu3+
Er3+
Tb3+
La3+
Table 1
Lactose
Cane sugar
Glucose
Urea
>8
>8
>8
3.2
The results of relative fluorescence quantum yield of GFLX
Sample
Quantum yield Φf
GFLX
GFLX + Y3+
0.12
0.16
Amylin
Amylum
3.2
3.2

482
X. Zhu et al. / Spectrochimica Acta Part A 69 (2008) 478–482
Table 3
Recoveries of GFLX from pharmaceutical products
a
Sample
Content (g mL−1
)
Added (g mL−1
)
Found (×10−6 g mL−1
)
Recovery (%)
R.S.D. (%)
8.0 × 10−8
2.4 × 10−7
4.0 × 10−7
4.83 × 10−7
6.33 × 10−7
7.99 × 10−7
103.8
97.1
99.8
2.5
1.3
1.1
GFLX
cap-
sule
4.0 × 10−7
a
Average value of three determination.
Table 4
Determination of GFLX in pharmaceutical products
Samples
Batch number
Labeled values
Proposed methoda
R.S.D. (%, n = 3)
HPLC methoda
Relative errorb (%)
Capsule
Injection
050102
051104031
0.10 g/particle
0.20 g/100 mL
0.099 g/particle
0.202 g/100 mL
0.5
2.5
0.0985 g/particle
0.199 g/100 mL
0.51
1.5
a
After suitable dilution.
Between results determined by proposed method and HPLC.
b
4. Conclusions
Table 5
Results of determination of recovery from biological fluids (n = 3)
A fluorescence probe enhanced fluorospectrophotometry for
the determination of GFLX has been developed and success-
fully applied to the determination of GFLX in pharmaceutical
formulations and biological fluids. It shows that proposed
yttrium-enhanced fluorimetric method is a simple, fast, accurate
and precise procedure.
Sample Added (g mL−1
)
Found (g mL−1
)
Recovery (%) R.S.D. (%)
Urine
8.0 × 10−8
2.4 × 10−7
4.0 × 10−7
8.0 × 10−8
2.4 × 10−7
4.0 × 10−7
8.10 × 10−8
2.32 × 10−7
3.77 × 10−7
7.70 × 10−8
2.34 × 10−7
3.88 × 10−7
101.2
96.7
94.3
1.8
1.1
1.2
Serum
96.2
97.5
97.0
2.1
1.7
1.3
References
[1] L. Ning, S. Chunyan, Z. Huichun, J. Instrum. Anal. (Chinese) 21 (1) (2000)
79.
[2] L. Jie, S. Chengchun, Y. Yujuan, Prac. J. Med. Pharma. 21 (5) (2004) 470.
[3] A.O. Juan, J.B. Francisco, C. Manuel, J. Pharm. Biomed. Anal. 37 (2005)
327.
[4] H.A. Nguyen, J.G. Boubakar, B. Ba, C. Quentin, M.C. Saux, J. Chromatogr.
B810 (2004) 77.
In all cases, the results obtained by the proposed method
were in good agreement with the labeled amounts (Table 4). In
order to further testing the accuracy of the proposed method,
the proposed method was compared with HPLC method [15].
The statistical t-test was used to compare the results from both
methods, which showed that there were not significantly dif-
ferent between them.
[5] R.P. Jayachandra, R.Y. Padmanabha, Int. J. Chem. Sci. 3 (2) (2005) 374.
[6] K. Venugopal, R.N. Saha, New Farmaco. 60 (2005) 906.
[7] R.F. Sane, S. Menon, Chromatogr. A 61 (5/6) (2005) 303.
[8] L. Colunga-Gonzalez, M. Guadalupe de Lerma, Anal. Lett. 38 (14) (2005)
2355.
[9] S. Hanwen, L. Liqing, C. Xueyan, Anal. Bioanal. Chem. 384 (2005) 1314.
[10] M. Shannigrahi, S. Bagchi, Spectrochim. Acta Part A 61 (2005) 2131.
[11] L. Li, J. Chongqiu, L. Lei, Anal. Chem. 31 (12) (2003) 1504.
[12] L. Laisheng, H. Zhibing, L. Xu, Appl. Chem. 21 (10) (2004) 1011.
[13] C. Guozhen, H. Xianzhi, X. Jingou, Spectrofluorimetric Analytical Method
[M], second ed., Science Publishing House, Beijing, 1990, p. 16.
[14] Z. Xiashi, B. Li, G. Rong, W. Jun, Anal. Chim. Acta 523 (2004) 43.
[15] C. Yulong, L. Liwei, Ch. J. Pharm. 32 (12) (2001) 553.
3.11. Analysis of spiked serum and urine samples
Inordertostudytheinfluenceofbiologicalfluidsonthedeter-
mination of GFLX, the proposed method was used to determine
GFLX in spiked human urine and serum samples. The results
are given in Table 5. The recovery was 94.3–101.2% for urine
and 96.2–97.5% for serum.