Content uploaded by Mariam M. Tadros
Author content
All content in this area was uploaded by Mariam M. Tadros on Aug 21, 2020
Content may be subject to copyright.
RESEARCH ARTICLE
Liquid chromatographic and spectrofluorimetric assays of
empagliflozin: Applied to degradation kinetic study and content
uniformity testing
Q1Q2 Maha F. Abdel‐Ghany |Miriam F. Ayad |Mariam M. Tadros
Analytical Chemistry Department, Faculty of
Pharmacy, Ain Shams University, Abbassia,
Cairo, Egypt
Correspondence
Mariam M. Tadros, Analytical Chemistry
Department, Faculty of Pharmacy, Ain Shams
University, Abbassia, Cairo 11566, Egypt.
Email: mariam.tadros@hotmail.com
Funding information
Faculty of Pharmacy Ain Shams
Q3 University
Abstract
Stability‐indicating high‐performance liquid chromatography (HPLC) and
spectrofluorimetric methods were developed for determination of empagliflozin
(EGF). EGF was subjected to oxidation, wet heat, photo‐degradation, acid hydrolysis
and alkali hydrolysis. The alkaline degradation pathway was subjected to a kinetics
study as the major product obtained after stress conditions. Arrhenius plots were
constructed and the activation energies of the degradation process were calculated.
HPLC was used for the kinetic study as it enabled simultaneous determination of
EGF and the degradation product while the spectrofluorimetric assay was applied
to content uniformity testing due to its higher sensitivity and lower limit of detection
(LOD). Isocratic chromatographic elution was attained for HPLC on a Intersil® C
18
column (150 mm × 4 mm, 5 μm), using a mobile phase of acetonitrile–potassium
dihydrogen phosphate buffer pH 4, (50:50, v/v) at a flow rate of 1 ml/min with
ultraviolet (UV) detection at 225 nm. The relative fluorescence intensity was
recorded by spectrofluorimeter applying synchronous mode using Δλ=70nmat
297.6 nm. Linearity ranges were found to be 5–50 μg/ml and 50–1000 ng/ml for
HPLC and spectrofluorimetric methods, respectively.
KEYWORDS
empagliflozin, forced degradation, HPLC, kinetics study, stability indicating assay, synchronous
spectrofluorimetry
1|INTRODUCTION
The kidney has a role in blood glucose regulation and consequently it
serves as a target for empagliflozin (EGF), (Figure
F1 1), which is an
inhibitor of sodium glucose co‐transporter‐2 (SGLT‐2) that inhibits
glucose re‐absorption into the blood.
[1,2]
Enhancements of glycemic
control and waist circumference provided further reasoning for use
in patients with type 2 diabetes.
[3]
According to the literature review
(Tables T11 and T22), some spectrophotometric and chromatographic
methods
[4–23]
were developed for EGF assay in its pharmaceutical
dosage form in addition to a review article discussing EGF assay in
its different combinations.
[24]
It is worthy to mention that this study is the first to consider the
kinetic degradation of EGF with full validation following the Interna-
tional Conference on Harmonization (ICH) guidelines.
[25]
Moreover,
it is the first study that considers the sensitive spectrofluorimetric
determination of EGF based on its native fluorescence with applica-
tion on the content uniformity testing of Jardiance® tablets following
United States Pharmacopeia (USP) guidelines.
[26]
High‐performance liquid chromatography (HPLC) was used for the
kinetic study because of its advantage of simultaneous determination
Abbreviations used: EFG, empagliflozin; ESI, electrospray ionization; HPLC,
high‐performance liquid chromatography; LOD, limit of detection; LOQ, limit
of quantification; %RSD, percent relative standard deviation; UV, ultraviolet.
Received: 8 December 2017 Revised: 6 February 2018 Accepted: 13 March 2018
DOI: 10.1002/bio.3491
Luminescence. 2018;1–14. Copyright © 2018 John Wiley & Sons, Ltd.wileyonlinelibrary.com/journal/bio 1
Journal Code Article ID Dispatch: 12.04.18 CE:
B I O 3 4 9 1 No. of Pages: 14 ME:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
of EGF and the degradation product while the spectrofluorimetric assay
was applied to content uniformity testing due to its higher sensitivity
and lower limit of detection (LOD). The major difference between the
developed methods is that HPLC enabled monitoring of two separate
peaks, one for the parent drug and the other for the degradation.
Percent of degradation was calculated at different time intervals and
temperatures using HPLC to study the factors affecting the kinetics
of EGF degradation. Furthermore, to save the time consuming method
development that is usually used to adjust the excitation and emission
wavelengths and to face the problem of broad spectra, synchronous
technique was conducted with simple runs, enhanced resolution and
better regression parameters. Synchronous technique had been used
for analysis of many pharmaceutical products because of its advantages
over the conventional mode.
[27–31]
FIGURE 1 Chemical structure of EGF
TABLE 1 Spectrophotometric methods for analysis of EGF either alone or in combination with linagliptin (LIG) or metformin (MRN)
Method Solvent Linearity (μg/ml) Detection wavelength (λ) Application
Direct UV Methanol and water 1–3 224 nm Determination of EGF in tablets
[4]
Direct UV Methanol and water 2–6 277 nm Determination of EGF and LIG in tablets
[5]
Simultaneous equation Methanol and water 6–12 277 nm Determination of EGF and LIG in tablets
[5]
First derivative Methanol 2.5–30 221 and 238 nm Determination of EGF and LIG in tablets
[6]
Simultaneous equation Methanol 2–12 225 and 237 nm Determination of EGF and MRN in tablets
[7]
PLS‐2 Methanol 2–10 200–300 nm Determination of EGF and MRN in tablets
[7]
Simultaneous equation Methanol 2–25 272 and 234 nm Determination of EGF and MRN in tablets
[8]
Absorption ratio Methanol 2–25 254 and 226 nm Determination of EGF and MRN in tablets
[8]
First derivative Methanol 2–12 223.5 and 233.5 nm Determination of EGF and MRN in tablets
[9]
Direct UV Water 2–12 247 nm Determination of EGF in tablets
[10]
Phenanthroline reaction Water 5–30 438 nm Determination of EGF in tablets
[10]
K Ferricyanide reaction Water 10–60 782 nm Determination of EGF in tablets
[10]
Derivative ratio Methanol 2–12 230 and 242 nm Determination of EGF and MRN in tablets
[11]
Ratio subtraction Methanol 2–12 237 nm Determination of EGF and MRN in tablets
[11]
Ext. ratio subtraction Methanol 2–12 225 nm Determination of EGF and MRN in tablets
[11]
Direct UV Methanol 5–25 296 nm Determination of LIG with EGF and MRN
[12]
Mean centering Methanol 2–12 222 and 249 nm Determination of EGF and MRN in tablets
[13]
First derivative Methanol 5–25 310 nm Determination of LIG with EGF and MET
[14]
Spectrum subtraction Methanol 2–12 225 nm Determination of EGF and MRN in tablets
[15]
Constant multiplication Methanol 2–12 237 nm Determination of EGF and MRN in tablets
[15]
TABLE 2 Chromatographic methods for analysis of EGF either alone or in combination with linagliptin (LIG) or metformin (MRN)
Stationary
phase Mobile phase Applications Detection
C
18
column Phosphate buffer (pH 3)–methanol
(30:70, v/v)
Determination of EGF in pharmaceutical
dosage form with MRN
UV 240 nm
[16]
C
18
column Acetonitrile–water (90:10, v/v) Pharmacokinetic study of EGF on human
volunteers
MS/MS m/z449.01
to 371.21
[17]
C
18
column Acetonitrile–water (75:25, v/v) Determination of EGF impurity MS/MS m/z785
to 475
[18]
C
18
column phosphate buffer (pH 4.8)–acetonitrile–methanol
(15:80:5, v/v/v)
Determination of EGF pharmaceutical
dosage form with MRN
UV 227 nm
[19]
C
18
column 0.1% formic acid–acetonitrile (50:50, v/v) Bio‐assay of EGF in plasma with MRN MS/MS m/z451.13
to 71.1
[20]
C
8
column 0.1 orthophosphoric acid–acetonitrile,
(70:30, v/v)
Assay of EGF in pharmaceutical dosage form UV 233 nm
[21]
C
18
column Potassium dihydrogen phosphate buffer
(pH 4–methanol (50:50, v/v)
Assay of EGF, LIG and MRN in tablets UV 225 nm
[22]
C
18
column 0.1% aqueous formic acid–acetonitrile
(75:25, v/v)
Assay of EGF and MRN in tablets MS/MS m/z451.04
to 71.07
[23]
2ABDEL‐GHANY ET AL.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
2|EXPERIMENTAL
2.1 |Instruments and underlying conditions
Schimadzu® Liquid Chromatography (JapanQ4 ), Intersil® C
18
column
(150 mm × 4 mm, 5 μm), system controller (SCL ‐10A VP),
degasser (DGU ‐12A) and ultraviolet (UV) detector (SPD‐10A VP)
were used. Acetonitrile–potassium dihydrogen phosphate buffer
pH 4 (50:50, v/v) was used as the mobile phase (adjusted with
orthophosphoric acid) with UV‐detection at 225 nm. The injection
volume was 20 μl.
Relative fluorescence intensity for EGF was measured using a
Shimadzu spectrofluorophotometer RF600 (Japan
Q5 ). Monochromator
was based on blazed holographic grating, 1300 grooves per millimeter.
The Δλof 70 nm was used in the synchronous mode measuring the
intense band at 297.6 nm.
Thermostatic multiple water bath, model BT‐15 (Spain
Q6 ) was used.
The ultra‐performance liquid chromatography with tandem mass spec-
trometry (UPLC–MS/MS) system consisted of Waters® ultra‐pressure
liquid chromatography (Milford, MA, USA) using Agilent® C
18
column
(50 mm × 2.1 mm, 5 μm) (USA
Q7 ). The system was equipped with a MS/
MS detector (Waters ACQUITY® TQD, Milford, MA, USA), using
electrospray ionization (ESI) technique.
2.2 |Samples and reagents
Pharmaceutical grade EGF (99.71%), according to a reference
method
[22]
and Jardiance® tablets (25 mg EGF) were supplied from
Boehringer Ingelheim (Germany Q8). Acetonitrile, methanol, ethanol,
deionized water, isopropyl alcohol, methylene chloride, acetone,
butanol, dimethylformamide (DMF) and ethyl acetate (HPLC) were
purchased from (Sigma‐Aldrich, St Louis, MO, USA). Glacial acetic
acid, sodium acetate trihydrate, boric acid, sodium hydroxide (NaOH),
hydrochloric acid (HCl), hydrogen peroxide (H
2
O
2
) (30% w/v),
potassium dihydrogen phosphate and orthophosphric acid (85%) were
purchased from VWR Chemicals (Poole, UK). Phosphate buffer (0.2 M,
pH 3–10), acetate buffer (0.2 M, pH 3–5) and borate buffer (0.2 M,
pH 8–10) solutions were freshly prepared. Surfactants 0.5% (w/v)
aqueous solutions were used including sodium dodecyl sulfate (SDS),
TABLE 3 System suitability tests for LC method for the determina-
tion of EGF in bulk
Item EGF ALK Limits
N 2116 1600 >2000
Ta 1.01 1.05 ≤2
Re, resolution between EGF and ALK peaks 4.3 >2
%RSD of six injections
Peak area 0.11 0.22 ≤2%
Retention time 0.15 0.13 ≤2%
Note: N, number of theoretical plates; Ta, tailing factor; Re, resolution;
RSD, relative standard deviation; ALK, alkaline degradation.
FIGURE 2 (a) HPLC chromatogram of EGF
(40 μg/ml) at 225 nm and (b) chromatogram of
Jardiance® tablet extract (10 μg/ml) using
HPLC‐UV at 225 nm
FIGURE 3 Three‐dimensional (3D) fluorescence scan spectrum of
EGF (1 μg/ml)
ABDEL‐GHANY ET AL.3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
cetrimide, β‐cyclodextrin (β‐CD), hydroxyl propyl‐β‐cyclodextrin
(HP‐β‐CD) (Darmstadt, Germany); carboxymethylcellulose (CMC) and
Tween‐80 from (El‐Nasr Chemicals, Cairo, Egypt).
2.3 |Standard stock solutions
Standard stock solution of EGF (1 mg/ml) was prepared in methanol
and then the required concentrations were prepared by serial dilutions
in acetonitrile–water (50:50, v/v) for the HPLC method (LC method)
while methanol was used as a diluent for the spectrofluorimetric
method (SF method) dilutions.
2.4 |Procedure for LC method
Accurately measured aliquots from the standard stock solution
equivalent to 50–500 μg EGF were transferred into a series of
10 ml volumetric flasks and then completed to final volume with ace-
tonitrile–water (50:50, v/v). A volume of 20 μl of each solution was
injected to HPLC.
2.5 |Procedure for SF method
Procedure for the SF method required preparation of a working
solution of EGF (10 μg/ml) in methanol. Accurately measured aliquots
FIGURE 4 (a) Excitation and emission
spectra of EGF (750 ng/mL) in methanol. (b)
Synchronous fluorescence spectra at
297.6 nm of EGF (750 ng/ml) in methanol. (c)
Effect of different Δλ10–150 nm on
fluorescence of EGF (1 μg/ml)
4ABDEL‐GHANY ET AL.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
FIGURE 5 Synchronous fluorescence
spectra at 297.6 nm of EGF (50–1000 ng/ml)
in methanol
FIGURE 6 Effect of addition of different
organized media (a), different buffers and
different pH values (b) and effect of different
diluents (c) on fluorescence of EGF (1 μg/ml)
ABDEL‐GHANY ET AL.5
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
from the working solution equivalent to 0.5–10 μg EGF were trans-
ferred into a series of 10 ml volumetric flasks and then completed to
final volume with methanol and transferred to the spectrofluorimeter.
The Δλof 70 nm was used in the synchronous mode measuring the
intense band at 297.6 nm.
2.6 |Assay of EGF in Jardiance® tablets and content
uniformity test
An accurately weighed amount of the finely powdered Jardiance®
tablets equivalent to 100 mg of EGF were extracted with 100 ml
methanol. The solutions were sonicated, filtered and serially diluted
by acetonitrile–water (50:50, v/v) for LC method while they were
diluted with methanol for SF method. The concentrations of EGF were
calculated using the constructed calibration equation and the standard
addition technique had been applied. Furthermore, for the content
uniformity test, 10 Jardiance®tablets were analyzed separately using
synchronous spectrofluorimetry, the uniformity of their contents
was tested and the nominal contents were calculated from the corre-
sponding regression equation.
FIGURE 7 Effect of time (a) and temperature (b) on fluorescence of
EGF (1 μg/ml)
TABLE 4 Results of assay validation of chromatographic (LC) and spectrofluorimetric (SF) methods for determination of EGF in bulk
Item LC method SF method
Retention time (min) 2.267 ‐
Wavelength of detection (nm) 225 Δλ= 70 at 297.6
Range of linearity 5–50 (μg/ml) 50–1000 (ng/ml)
Regression equation Area × 10
−6
= 0.4765
Cμg/ml + 0.1981
RFI = 22.584
C ng/ml + 821.1
Regression coefficient (r) 0.9997 0.9999
LOD 1.19 (μg/ml) 16.5 (ng/ml)
LOQ 3.62 (μg/ml) 50 (ng/ml)
S
b
4.42 × 10
−3
0.133
S
a
0.16 91.608
Confidence limit of the slope 0.4765 ± 2.11 × 10
−3
22.584 ± 3.01
Confidence limit of the intercept 0.1981 ± 3.25 × 10
−2
821.1 ± 75200
Standard error of the estimation 0.17 112.93
Precision
Intraday %RSD 0.11–0.19 0.24–0.17
Interday %RSD 0.31–0.25 0.15–0.28
Robustness
Flow rate (± 0.2), %RSD 0.14 —
Organic strength (% ± 2), %RSD 0.23 —
Mobile phase pH (± 0.2), %RSD 0.26 —
Note: RFI, relative fluorescence intensity; LOD, limit of detection; LOQ, limit of quantification; S
b
, standard error of slope; S
a
, standard error of intercept;
%RSD, percent relative standard deviation.
TABLE 5 Results for determination of EGF in bulk powder by the
proposed LC and SF methods
LC method SF method
Pure
(μg/ml)
Found
(μg/ml) R(%)
a
Pure
(ng/ml)
Found
(ng/ml) R(%)
a
7.5 7.48 99.73 75 74.80 99.73
15 15.20 101.33 200 201.52 100.76
25 25.09 100.36 400 397.50 99.38
35 34.95 99.86 600 595.31 99.22
45 45.07 100.16 800 804.05 100.51
Mean ± standard
deviation
100.29 Mean ± standard
deviation
99.92
0.63 0.68
a
Mean of three determinations.
6ABDEL‐GHANY ET AL.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
2.7 |Stress degradation of EGF
To a series of test tubes, each one contains 1 ml of standard stock
solution of EGF (1 mg/ml), 2 ml of 0.1, 0.5, 1 M NaOH, 2 ml of
0.1, 0.5, 1 M HCl and 2 ml of 0.3% H
2
O
2
were added separately
and each mixture was allowed to stand in a thermally controlled
water bath at 60°C for 30 min except for H
2
O
2
which was kept at
room temperature. The H
2
O
2
oxidation reaction was stopped using
sonication while both alkaline and acid degradation were stopped
by chilling the test tube in ice cold water and neutralization (with
1 M HCl or 1 M NaOH) using pH meter. The degraded samples were
quantitatively transferred to 25 ml volumetric flasks separately and
diluted to final volume with acetonitrile–water (50:50, v/v) for the
LC method. Hence, 20 μl of the degraded solutions were injected
to HPLC.
2.8 |Alkaline degradation kinetics of EGF using LC
method
Three series of six test tubes were prepared by transferring 1 mL of
EGF standard stock solution (1 mg/mL), 2 mL 0.5 M NaOH were
added to each tube and allowed to stand in a thermally controlled
water bath at 60, 70, and 80°C. The degradation was stopped at
10‐min intervals (10, 20, 30, 40, 50, and 60 min) by chilling the test
tube in ice cold water and neutralization using pH meter (to give
pH 6.5–7.5). The degraded sample was quantitatively transferred to
25 ml volumetric flask and diluted to final volume with acetonitrile–
water (50:50, v/v). Then 20 μl of the degraded solutions were
injected to HPLC.
3|RESULTS AND DISCUSSION
In spite of both the chromatographic and the spectrofluorimetric
methods being developed successfully for stability indicating assays,
the chromatographic method enables the monitoring of the degrada-
tion peak on chromatograms offering the data required for the kinetic
degradation study.
3.1 |HPLC method development
The two most simple mobile phases were tried in the preliminary inves-
tigations as 50% organic solvent (acetonitrile or methanol) and 50%
buffer at the pH value away from the pKa of the drug. Acetonitrile
showed the optimum results and decreased the noise significantly.
The retention time for EGF was 2.267 min as presented in Figure F22(a)
with no interference from the dosage form excipients as shown in
Figure 2(b). The percent relative standard deviation (%RSD) of
peak areas for six injections of EGF and the reproducibility of retention
times were checked with the other system suitability tests as shown in
Table T33.
3.2 |Fluorimetric characteristics of EGF
An essential state for a drug to fluoresce is the conjugated π‐electron
system that gives re‐emission of the absorbed energy. The
spectrofluorimetric scan for EGF showed excitation and emission
spectra at 226.5 and 299.4 nm, respectively (Figures F33 and F44a)
because of the conjugated system that appears in its structure
(Figure 1). In the underlying investigation, synchronous technique
was used applying Δλ= 70 nm with enhanced sensitivity and features
bands
[27–31]
at 297.6 nm as shown in Figure 4(b). A marked decrease
in spectra overlapping and marked increase in linearity parameters
confirmed the advantages of using synchronous technique over the
conventional spectrofluorimetric analysis.
TABLE 6 Determination of EGF in pharmaceutical dosage form and standard addition technique by LC and SF methods
Pharmaceutical dosage form
% Recovery ± standard deviation
a
Standard addition technique
EGF
Claimed concentration
(μg/ml)
Pure added
(μg/mL)
Pure
found
%Rpure
added
a
LC method
Jardiance® tablets (EGF 25 mg)
Batch no: L5727A
99.37 ± 1.15 10 μg/ml 5 μg/ml 5.06 101.20
10 μg/ml 9.96 99.60
15 μg/ml 15.11 100.73
Mean ± standard deviation 100.51 ± 0.82
SF method
Jardiance® tablets (EGF 25 mg)
Batch no: L5727A
99.97 ± 1.01 200 ng/ml 100 ng/ml 101.05 101.05
200 ng/ml 197.37 98.69
300 ng/ml 301.28 100.43
Mean ± standard deviation 100.06 ± 1.23
a
Mean of three determinations.
TABLE 7 Statistical comparison between the proposed methods and
the reference method
Statistical term Reference method
[22]
LC method SF method
Mean 99.71 100.29 99.92
SD ± 0.97 0.63 0.68
%RSD 0.97 0.63 0.68
n555
V0.94 0.40 0.46
t(2.306), p= 0.05 1.12 0.40
F(6.39), p= 0.05 2.35 2.04
Note: SD, standard deviation; %RSD, percent relative standard deviation.
ABDEL‐GHANY ET AL.7
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
3.3 |Factors affecting fluorescence of EGF
The addition of a surfactant at a concentration above its CMC may
increase fluorescence intensity but for EGF, no increase was
observed after studying the effect of adding many surfactants and
macromolecules. Furthermore, adding different buffers with different
pH values showed no significant enhancement of EGF native fluores-
cence. The native fluorescence of the drug was enough to conduct a
sensitive synchronous spectrofluorimetric assay as adding different
organized media and/or different buffers for sensitization of EGF
fluorescence resulted in no effect or slight decrease in its relative
fluorescence intensity.
3.3.1 |Effect of Δλon EGF fluorescence
Different Δλvalues were tested from 10 to 150 nm as in Figure 4(c)
and the best synchronous spectrum was obtained for Δλ =70nm
showing the best linearity results (Figure F55).
FIGURE 8Q10 HPLC alkaline degradation chromatograms of 40 μg/ml EGF (at retention time 2.267 min) and its main degradation product (at retention
time 1.43–1.45 min) withdrawn at 10 (a), 20 (b), 30 (c), 40 (d), 50 (e) and 60 (f) min using 0.5 M NaOH at 60°C
8ABDEL‐GHANY ET AL.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
3.3.2 |Effect of surfactant on EGF fluorescence
FigureF6 6 describes the effect of different surfactants including
anionic (SDS), cationic (cetrimide) and non‐ionic (Tween‐80), and
macromolecules such as β‐CD, HP‐β‐CD and CMC (1 ml of a 0.5%,
w/vfreshly prepared aqueous solution of each) on EGF (1 μg/mL)
methanolic solution fluorescence showing that no increase in relative
fluorescence was observed after using any sensitizing agent. There-
fore, no surfactant was used in the study.
3.3.3 |Effect of buffer on EGF fluorescence
The influence of pH was investigated using 1 ml of freshly prepared
solutions of different buffer types with different pH values ranging from
3 to 10, as shown in Figure 6, phosphate buffer (0.2 M, pH 3–10),
acetate buffer (0.2 M, pH 3–5) and borate buffer (0.2 M, pH 8–10),
without enhanced fluorescence of EGF (1 μg/ml) methanolic solution.
It was found that the maximum relative fluorescence intensity was
achieved in methanol without the addition of any buffer.
FIGURE 9Q11 HPLC alkaline degradation chromatograms of 40 μg/ml EGF (at retention time 2.267 min) and its main degradation product (at
retention time 1.43–1.45 min) withdrawn at 10 (a), 20 (b), 30 (c), 40 (d), 50 (e) and 60 (f) min using 0.5 M NaOH at 70°C
ABDEL‐GHANY ET AL.9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
3.3.4 |Effect of diluent on EGF fluorescence
The effect of different diluting solvent on the relative fluoresence of
EGF was investigated using water, ethanol, methanol, n‐butanol,
DMF, ethyl acetate, acetonitrile and acetone. Figure 6 confirmed that
using any solvent other than methanol as a diluent decreased EGF rela-
tive fluorescence intensity that may be attributed to high blank readings
or quenching effect of the used solvents. So, it was found that methanol
was the best solvent for dilution, as it gave the highest relative
fluoresence and the lowest blank reading with reproducible results.
3.3.5 |Effect of time and temperature on EGF
fluorescence
The effect of time on the relative fluorescence of the drug was also
studied. It was found that the relative fluorescence intensity remained
stable for more than 2 h as shown in Figure F77(a). The effect of temper-
ature was studied at room temperature (25 ± 5°C) and at higher
temperatures. It was found that increasing the temperature resulted
in a decrease in the relative fluorescence of the drug as shown in
Figure 7(b). This effect may be due to the facilitated non‐radiated
FIGURE 10Q12 HPLC alkaline degradation chromatograms of 40 μg/mL EGF (at retention time 2.267) and its main degradation product (at retention
time 1.43) withdrawn at 10 (a), 20 (b), 30 (c), 40 (d), 50 (e) and 60 (f) min using 0.5 M NaOH at 80°C
10 ABDEL‐GHANY ET AL.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
deactivation of the excited singlet state caused by high temperatures.
Therefore, all the experiments were carried out at room temperature.
3.4 |Method validation
A linear relationship was found between the response (AUPQ9 in LC
method and relative fluorescence intensity in SF method) and EGF
concentration. Limit of detection (LOD = 3.3 × residual standard
deviation of regression line/slope) and limit of quantification
(LOQ = 10 × residual standard deviation of regression line/slope) were
determined for the proposed methods. The analytical data of the
calibration curve are summarized in Table T44. To check the accuracy,
the LC and SF procedures were adopted using five concentrations
ranging from 7.5 to 45 μg/ml and 75 to 800 ng/ml of EGF, respec-
tively. The concentrations of EGF were calculated using the calibration
equations and the mean of the recovery and standard deviation are
shown in Table T55. To check the precision, three concentrations of
EGF (20, 25 and 30 μg/mL, LC method and 400, 500 and 600 ng/
FIGURE 11Q13 HPLC chromatograms of EGF (40 μg/ml) subjected to peroxide degradation (a), acid degradation after 30 min at 60°C using 0.1 M
HCl (b), 0.5 M HCl (c), 1 M HCl (d), wet heat degradation (e) and UV photodegradation (f)
ABDEL‐GHANY ET AL.11
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
mL, SF method) were analyzed three times, within the same day and
on three successive days. The %RSD was calculated and found to be
less than 1% (Table 4). Robustness of LC method was confirmed by
minor changes in the flow rate, the organic strength and pH value of
the phosphate buffer (Table 4).
3.5 |Application to tablets (specificity) and content
uniformity testing
Both methods had been applied to the pharmaceutical dosage form
and the standard addition technique was applied where the included
excipients do not interfere with the results (Table T66). The developed
synchronous spectrofluorimetric method was suited for content
uniformity testing according to USP guidelines.
[26]
The mean of the
percent recoveries of the investigated 10 tablets was found to be
100.29% ± 1.38 (standard deviation) and the acceptance value (AV)
was found to be 2.4 lower than the maximum allowed acceptance
value (L1 = 15) confirming drug uniformity.
3.6 |Statistical analysis
Statistical comparison with the reference method
[22]
was conducted
by ‘SPSS statistical package version 11’at P= 0.05 as shown in
Table T77 where the calculated Fand ttest are less than the tabulated
ones which proves our method to be accurate and precise.
3.7 |Degradation and kinetic behavior
Different molarities of NaOH showed the appearance of additional
peaks at 1.43–1.45 min using the LC method, (Figures 8 F8F10–10)
confirming that EGF is susceptible to alkaline degradation. HPLC was
used for the kinetic study because of its advantage of simultaneous
determination of EGF and the degradation product. The samples
subjected to acid induced, H
2
O
2
, wet heat and photo‐induced degra-
dation, showed no additional peaks with EGF except for H
2
O
2
peak;
as shown in Figure F1111.
Kinetic of the basic degradation of EGF was studied using the LC
method by following the concentrations of the basic degradation
product within 60 min at 10 min intervals at 60, 70, and 80°C. A
decrease in concentration of drug with increasing time was observed.
The final alkaline degradation percentage was found to be 59%, 29%
and 25% at 60, 70 and 80°C, respectively. Figure F1212(a) shows plots
of Ln (concentration) of the remaining drug against time. In compari-
son to the drug concentration, the reaction was performed in large
excess of NaOH (0.5 M). The linear relationships obtained indicate
pseudo‐first order reaction kinetics. The slopes of the straight lines
at each temperature were used to calculate the half‐life (t
1/2
) and
shelf‐life (t
90
) as shown in Table T88.
The activation energy (E
a
) of the basic degradation reaction was
determined by calculating the rate constant at 60, 70 and 80°C using
0.5 M NaOH as presented in Figure 12(a), then fitting the obtained
data in the Arrhenius equation (lnK=lnC–E
a
/RT) where Kis the
observed rate constant (in min
−1
), Athe frequency factor (in min
−1
),
E
a
the energy of activation (in Kcal mol
−1
), Rthe gas constant
(1.987 cal/K/mol) and Tis the absolute temperature (K). The
Arrhenius plot of (ln K
obs
) values versus the reciprocal temperature
(1/T), (60–80 ±
°
C) was illustrated in Figure 12(b). The degradation rate
constant, the half‐life (t
1/2
) and shelf‐life (t
90
) at room temperature
(25°C) were calculated as shown in Tables 8 and T99.
One of the alkaline degradation samples that was prepared earlier,
(0.5 M NaOH at 70°C for 0.5 h), was investigated by LC–MS either in
positive mode (Figure F1313a) or in negative mode (Figure 13b) and the
FIGURE 12 Pseudo‐first order kinetic plot of the alkaline
degradation of EGF (40 μg/ml) at 60, 70, and 80°C using 0.5 M
NaOH (a) and Arrhenius plot of the alkaline degradation of EGF
(40 μg/ml), versus different heating times (in minutes) with 0.5 M
NaOH (b)
TABLE 8 Kinetic data of EGF alkaline degradation in the presence of
0.5 M NaOH
Temperature (K) K
obs
(min
−1
)t
1/2
(min)
a
t
90
(min)
b
298.15 0.00295 234.9 35.7
333.15 0.0068 101.9 15.5
343.15 0.0086 80.6 12.3
353.15 0.0102 67.9 10.3
a
t
1/2
was calculated as 0.693/K
obs
.
b
t
90
was calculated as 0.10536/K
obs
.
TABLE 9 Kinetic data of EGF alkaline degradation at 25°C
Parameters 0.5 M NaOH
Activation energy, E
a
(Kcal mol
−1
) 4.7
Degradation rate constant K
25
(min
−1
) 0.00295
Half‐life, t
1/2
(min) 234.9
Shelf‐life, t
90
(min) 35.7
Arrhenius frequency factor, A(min
−1
) 8.9
12 ABDEL‐GHANY ET AL.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
expected degradation product is shown in Figure F1414 with explanation
of the m/zof EGF (473 in positive ESI and 485.18 in negative ESI) and
m/zof the formed alkaline degradation product (59.93 in positive ESI
and 58.92 in negative ESI). Furthermore, the degradation product
(Figure 14) was more polar than the drug so it appears early on the
chromatogram with significant lower retention time.
4|CONCLUSION
The proposed stability indicating chromatographic and
spectrofluorimetric methods proved to be accurate and precise for
quantification of EGF in the presence of its degradation products as
fast, economic and reliable analytical procedures. The methods were
applied effectively for the determination of EGF in Jardiance® tablets.
FIGURE 13Q14 Full scan mass spectrum in positive ESI (a) and negative ESI (b) ion detection mode for EGF (40 μg/ml) and its main degradation
product withdrawn after 30 min heating at 70°C
FIGURE 14 Proposed pathway for the major alkaline degradation
product of EGF
ABDEL‐GHANY ET AL.13
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
It is the first study that considered the degradation kinetics of EGF,
the corresponding Arrhenius plots and further identification of
the major degradation product with full validation study. The kinetics
of the alkaline hydrolysis was investigated to study the effect of
temperature and time on the degradation product. Moreover, it is
the first described synchronous spectrofluorimetric method for EGF
based on the native fluorescence of the drug with valuable application
to content uniformity testing following USP guidelines, the method is
ideally suited for content uniformity testing which is a time‐consuming
process when using conventional assay techniques. Finally, to
maximize the time‐saving procedures to its higher limits, synchronous
scan was necessary to achieve satisfying results.
ORCID
Mariam M. Tadros http://orcid.org/0000-0001-9928-1057
REFERENCES
[1] H. H. Hansen, J. Jelsing, C. F. Hansen, G. Hansen, N. Vrang, M. Mark,
T. Klein, E. Mayoux, J. Pharm. Exp. Ther. 2014,350, 657.
[2] S. Nair, J. P. Wilding, J. Clin. Endocrinol. Metab. 2010,95, 34.
[3] K. M. Munir, S. N. Davis, Clin. Pharmacol. 2016,8, 19.
Q15
[4] S. D. Patil, S. K. Chaure, M. A. H. Rahman, P. U. Varpe, S. Kshirsagar,
Asian J. Pharm. Ana. 2017,7, 18.
[5] N. Padmaja, G. Veerabhadram, Pharm. Lett. 2015,7, 306.
[6] F. R. Amrutiya, B. R. Patel, J. G. Patel, K. L. Vegad, A. S. Patel, V. C.
Darji, Int. J. Pharm. Drug Anal. 2017,5, 129.
[7] B. M. Ayoub, Spectrochim. Acta Mol. Biomol. Spectrosc. 2016,168, 118.
[8] N. Padmaja, M. S. Babu, G. Veerabhadram, Pharm. Lett. 2016,8, 207.
[9] B. M. Ayoub, Der Pharma Chemica. 2016,8, 12.
[10] N. Jyothirmai, B. Nagaraju, K. M. Anil, J. De Afrikana 2016,3, 177.
[11] B. M. Ayoub, AOAC. 2017,100, 985.
[12] L. Shaker, Pharm. Lett. 2016,8, 267.
[13] B. M. Ayoub, R. M. Emam, M. M. Youssef, M. N. El‐Kattan, M. A. Sayed,
A. M. Kowider, A. H. Seha, E. A. Rabea, R. M. Yakout, R. H. Faried,
Marmara Pharm. J. 2017,21, 669.
[14] L. Shaker, Pharm. Lett. 2016,8, 256.
[15] B. M. Ayoub, J. Anal. Chem. 2017,84(5), 884.
[16] N. Madana Gopal, C. Sridhar, J. Appl. Pharmacol. 2017,9, 45.
[17] B. M. Ayoub, S. Mowaka, E. S. Elzanfaly, N. Ashoush, M. M. Elmazar,
Sci. Rep. 2017,7(1), art. no., 2583.
[18] H. Zhou, F.‐H. Meng, L. Sun, S.‐Y. Qiao, G.‐G. Zhang, J. Int. Pharm. Res.
2016,43, 753.
[19] P. Geetha Swarupa, K. Lakshmana Rao, K. R. S. Prasad, K. Suresh Babu,
Asian J. Pharm. Clin. Res. 2016,9, 126.
[20] B. M. Ayoub, Der Pharma Chemica. 2016,8, 163.
[21] Shyamala, K. Nirmala, J. Mounika, B. Nandini, Pharm. Lett. 2016,8, 457.
[22] B. M. Ayoub, RSC Adv. 2015,5, 95703.
[23] B. M. Ayoub, S. Mowaka, J. Chrom. Sci. 2017,55, 742.
[24] B. M. Ayoub, Der Pharma Chemica. 2016,8, 23.
[25] ICH, Q2 (R1), Validation of Analytical procedures. International Con-
ference on Harmonization, Commision of the European Communities,
Geneva, 1996.
[26] M. Rockville, The United States Pharmacopoeia 30, the National For-
mulary 25 US Pharmacopeial Convention,Electronic version, 2007.
[27] A. D. Bani‐Yaseen, Spectrochim. Acta Mol. Biomol. Spectrosc. 2015,
148, 93.
[28] F. Ibrahim, J. J. Nasr, Luminescence 2016,31, 255.
[29] E. Vazquez‐Giao, J. López‐Hernández, M. A. Lage‐Yusty, Curr. Top.
Toxicol. 2015,11, 41.
[30] H. W. Darwish, A. Bakheit, RSC Adv. 2015,5, 54471.
[31] O. A. A. Ghonim, A. M. El‐Kosasy, S. M. El‐Sayed Okeil, J. AOAC Int.
2014,97, 921.
How to cite this article: Abdel‐Ghany MF, Ayad MF, Tadros
MM. Liquid chromatographic and spectrofluorimetric assays of
empagliflozin: Applied to degradation kinetic study and content
uniformity testing. Luminescence. 2018;1–14. https://doi.org/
10.1002/bio.3491
14 ABDEL‐GHANY ET AL.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114