- Email: [email protected]
Development and application of a fluorescent glucose uptake assay for the high-throughput screening of non-glycoside SGLT2 inhibitors
PHASCI 3213
No. of Pages 5, Model 5G
31 March 2015 European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx 1
Contents lists available at ScienceDirect
European Journal of Pharmaceutical Sciences journal homepage: www.elsevier.com/locate/ejps 5 6 3 4
Development and application of a fluorescent glucose uptake assay for the high-throughput screening of non-glycoside SGLT2 inhibitors
8
Szu-Huei Wu 1, Chun-Hsu Yao 1, Chieh-Jui Hsieh 2, Yu-Wei Liu, Yu-Sheng Chao 3, Jen-Shin Song ⇑, Jinq-Chyi Lee ⇑
9
Institute of Biotechnology and Pharmaceutical Research, National Health Research Institutes, 35 Keyan Road, Zhunan Town, Miaoli County 35053, Taiwan, ROC
7
10 11 1 2 3 6 14 15 16 17 18 19 20 21 22 23 24 25
a r t i c l e
i n f o
Article history: Received 12 December 2014 Received in revised form 5 March 2015 Accepted 20 March 2015 Available online xxxx Keywords: Sodium-dependent glucose co-transporter Type 2 diabetes mellitus Non-glycoside High-throughput screening Click chemistry
a b s t r a c t Sodium-dependent glucose co-transporter 2 (SGLT2) inhibitors are of current interest as a treatment for type 2 diabetes. Efforts have been made to discover phlorizin-related glycosides with good SGLT2 inhibitory activity. To increase structural diversity and better understand the role of non-glycoside SGLT2 inhibitors on glycemic control, we initiated a research program to identify non-glycoside hits from high-throughput screening. Here, we report the development of a novel, fluorogenic probe-based glucose uptake system based on a Cu(I)-catalyzed [3 + 2] cycloaddition. The safer processes and cheaper substances made the developed assay our first priority for large-scale primary screening as compared to the well-known [14C]-labeled a-methyl-D-glucopyranoside ([14C]-AMG) radioactive assay. This effort culminated in the identification of a benzimidazole, non-glycoside SGLT2 hit with an EC50 value of 0.62 lM by high-throughput screening of 41,000 compounds. Ó 2015 Elsevier B.V. All rights reserved.
27 28 29 30 31 32 33 34 35 36 37 38 39
40 41
1. Introduction
42
Sodium-dependent glucose co-transporter 2 (SGLT2) is a highcapacity, low-affinity transporter located on the S1 segment of the proximal tubule in the kidney, and belongs to the 12 member solute carrier family 5A (SLC5A). It reabsorbs the majority (90%) of renal glucose filtered by kidney glomerulus back into the blood stream (Wright, 2001; Wright and Turk, 2004). Inhibition of SGLT2 has been demonstrated to suppress renal glucose reabsorption, increase urine glucose excretion, and consequentially reduce glycemic levels. This mode of action constitutes an insulinindependent pathway to control hyperglycemia, without the risk of hypoglycemia (Abdul-Ghani et al., 2011). SGLT2 inhibitors were also found to result in modest weight loss and blood pressure reduction, both additional advantages which make SGLT2 inhibitors more attractive in the treatment of type 2 diabetes (Nauck, 2014).
43 44 45 46 47 48 49 50 51 52 53 54 55 56
⇑ Corresponding authors. Tel.: +886 37 246166x35739 (J.-S. Song), +886 37 246166x35781 (J.-C. Lee). E-mail addresses: [email protected] (J.-S. Song), [email protected] (J.-C. Lee). 1 S.-H. Wu and C.-H. Yao equally contributed to this work. 2 Present address: TPV Technology Group, Top Victory Electronics (Taiwan) Co., Ltd., 12F-1, No. 186, Jian 1st Rd., Zhonghe Dist., New Taipei City 23511, Taiwan, ROC. 3 Present address: Diamond BioFund INC., 5F., No. 3, Sec. 1, Dunhua S. Rd., Taipei City 105, Taiwan, ROC.
Several SGLT2 inhibitors have been developed and disclosed. Some are in the advanced stages of clinical trials (Ho et al., 2011), and all are C-linked b-glycosides derived from the structural modification of phlorizin, the natural product isolated from the root bark of the apple tree and subsequently identified as the first SGLT2 inhibitor (Ehrenkranz et al., 2005). Of these, dapagliflozin, canagliflozin, ipragliflozin, tofogliflozin and empagliflozin were recently approved for the treatment of type 2 diabetes mellitus (T2DM) (Traynor, 2014; Elkinson and Scott, 2013; Poole and Dungo, 2014; Poole and Prossler, 2014; Neumiller, 2014). However, their use requires long-term follow-up due to the potential for cancer, liver injury, cardiovascular diseases, and/or other unexpected side-effects (Fujita and Inagaki, 2014). The overwhelming majority of known SGLT2 inhibitors are phlorizin-related glycosides. Identification of non-glycoside inhibitors could be advantageous, and the use of high-throughput screening is an important strategy, having been utilized in drug discovery for many years to identify active hits from large libraries of compounds (Hertzberg and Pope, 2000; Liu et al., 2004). The most widely used in vitro assay for evaluating inhibitory activities against SGLT2 is a radioactive functional assay, which is based on the suppression of uptake of [14C]-labeled a-methyl-Dglucopyranoside ([14C]-AMG) into cells stably expressing SGLT2 (Castaneda and Kinne, 2005). Unfortunately, this system is severely limited by the high cost of the substrate and the difficulties associated with the handling and disposal of radioactive substances, and
http://dx.doi.org/10.1016/j.ejps.2015.03.011 0928-0987/Ó 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: Wu, S.-H., et al. Development and application of a fluorescent glucose uptake assay for the high-throughput screening of non-glycoside SGLT2 inhibitors. Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.03.011
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
PHASCI 3213
No. of Pages 5, Model 5G
31 March 2015 2
S.-H. Wu et al. / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx
97
we declined to use it in the large-scale primary screening for nonglycoside SGLT2 inhibitors even though it has been previously applied in drug discovery and in an automated, 96-well format (Li et al., 2011; Castaneda and Kinne, 2005). Instead we sought to develop a new assay system with safer processes and a cheaper substrate. Herein, we outline a fluorescence system for probing the glucose transported by hSGLT2 into cells, which should obviate the aforementioned shortcomings of the radioassay. In this approach, the uptake azido-modified glucose analogues are labeled with a fluorogenic probe, 4-ethynyl-N-ethyl-1,8-naphthalimide, using a Cu(I)-catalyzed [3 + 2] cycloaddition (azide–alkyne ‘‘click’’ chemistry) (Sawa et al., 2006; Rostovtsev et al., 2002). We also exemplify the utility of this newly developed fluorescence assay in high-throughput screening by the identification of non-glycoside SGLT2 inhibitor hits.
98
2. Materials and methods
83 84 85 86 87 88 89 90 91 92 93 94 95 96
99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114
2.1. Establishment of human SGLT-2 stable cell line The hSGLT-2 cDNA was subcloned into pIRES2-EGFP vector (Clontech Laboratories, Inc., Mountain View, CA, USA). Before transfection, cells were seeded in 24-well plates at a density of 2 105 cells/well. Cells were cultured until they reached 70–80% confluence and then transfected with 1 lg of vector containing the hSGLT-2 coding region using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) per well. Transfected CHO-K1 cell stably expressed hSGLT-2 (hSGLT-2/CHO-K1) was selected by EGFP and 1 mg/ml G418 (Calbiochem, Merck-Millipore, Darmstadt, Germany) for 10 to 14 days. The selected clone was maintained in Ham’s F12 medium (HyClone, Logan, UT, USA) supplemented with 10% fetal bovine serum (HyClone, Logan, UT, USA), 1% of penicillin–streptomycin (Biological Industries, Beit-Haemek, Israel), and 0.5 mg/ml G418 with 5% CO2 at 37 °C in the humidified incubator.
115
2.2. Radioactive sodium dependent glucose co-transport 2 assay
116
The transport assays were performed according to method from Castaneda and Kinne with some modification (Castaneda and Kinne, 2005). Stably transfected CHO-K1 cells (hSGLT-2/CHO-K1) were used for transporter studies. Sodium-dependent glucose cotransporter was determined by means of [14C]-a-methyl-Dglucopyranoside ([14C]-AMG, specific radioactivity 300 mCi/mmol) purchased from Perkin Elmer (Boston, USA). For the purpose of this study, Krebs–Ringer–Henseleit (KRH) solution containing 120 mM NaCl, 4.7 mM KCl, 1.2 mM MgCl2, 2.2 mM CaCl2, 10 mM Hepes (pH 7.4 with Tris) was used as rinse buffer. All chemicals were purchased from Sigma (Deisenhofen, Germany). Briefly, hSGLT2/ CHO-K1 cells were seeded into a white-walled 96-well culture plate (Corning, NY, USA) at a density of 3 104 cells/well and incubated for 48 h at 37 °C with 5% CO2 atmosphere in growth medium. After 48 h, the culture medium of the 96-wells was taken off and wells were rinsed twice with 200 lL of KRH-Na and incubated in KRH-Na containing 3 lM [14C]-AMG in the absence or presence of inhibitors for up to 120 min at 37 °C. At the end of the uptake period, the transport buffer was removed and the uptake of [14C]-AMG was stopped by adding ice-cold KRH-Na containing 0.5 mM phlorizin. The wells were rinsed three times with 100 lL stop buffer using the microplate washer (Tecan, Männedorf, Switzerland). After the third rinse, the stop buffer was completely removed from the wells and the cells were solubilized by adding 1% sodium dodecyl sulfate (Sigma). After 24 h, the microtiter plate was taken for scintillation counting of radioactive [14C]-AMG using a TopCount (Perkin Elmer). The percent of inhibition was
117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142
calculated by comparing counts per minute (cpm) in the inhibitor-containing well with cpm in the well containing only the DMSO vehicle. Phlorizin and dapagliflozin were evaluated in parallel in every assay. A dose–response curve was fitted to a sigmoidal dose–response model using GraphPad software to determine the inhibitor concentration at half-maximal response (EC50).
143
2.3. Fluorescent sodium dependent glucose co-transport 2 assay
149
The stably transfected CHO-K1 clone hSGLT-2/CHO-K1 was incubated with indicated concentration of azidoglucose in the absence or presence of 10–50 lM phlorizin in KRH buffer (120 mM NaCl, 4.7 mM KCl, 1.2 mM MgCl2, 2.2 mM CaCl2, 10 mM HEPES with Tris, pH = 7.4) for 3 h at 37 °C. At the end of the uptake period, the medium was removed and washed thrice with KRH buffer to remove the un-transported azidoglucose. Cells were then treated with click-activated alkyne probe, 4-ethynyl-N-ethyl-1,8naphthalimide 8, in SDS lysis buffer (100 lM 8, 250 lM triazole ligand (Zhang and Zhang, 2013), 5 mM sodium ascorbate, 2.5 mM CuSO45H2O, 0.1% SDS-PBS) and incubated at 37 °C for 3 h (for high-throughput screening, the duration of the incubation is 18 h) to afford the desired fluorescence adduct. Next, the fluorescence intensity of the generated fluorescently labeled triazolyl glycan was measured by a Victor2-V fluorescence plate reader with an excitation/emission wavelength of 355 nm/460 nm. The percent of inhibition of inhibitor was calculated by comparing fluorescence intensity (relative fluorescence unit, RFU) in inhibitor-containing well with RFU in well containing only DMSO vehicle. A dose– response curve was calculated by nonlinear regression (GraphPad software, San Diego, CA) to determine the functional inhibitory concentration at half-maximal response (EC50).
150
2.4. MTS cell viability assay
172
Cells were seeded in 96-well clear plates at a density of 1 104 cells per well overnight. Then, cells were treated with 10 lM compounds for 72 h. At the end of incubation, for a 96-well microtiter plate, 12 ml of MTS Mix reagent contained culture medium, MTS (Promega, Madison, WI), and PMS (Sigma, St. Louis, MO) in a ratio of 8:2:0.1, respectively. The medium in well was removed and the MTS Mix reagent was then added to cells (100 lL/well). The plates were incubated for 2.5 h at 37 °C in a humidified, 5% CO2 atmosphere and the absorbance was then recorded at 490 nm.
173
3. Results and discussion
182
3.1. Assay development
183
3.1.1. Strategy for specific fluorescent SGLT2 assay The bioorthogonal reactions of click chemistry have been well studied in biomolecule labeling and detection (Best, 2009; Zhang and Zhang, 2013). The remarkable results obtained encouraged us to incorporate it into a non-radioactive assay strategy, which is depicted in Fig. 1. The function of SGLT2 is to transport glucose and sodium into cells using the sodium gradient created by sodium/potassium ATPase pumps in the basolateral cell membranes. In the developed assay, we substituted glucose for azidoglucoses, which were hypothesized to be transported into cells via SGLT2. Firstly, hSGLT2/CHO-K1 cells were incubated with azido-glucoses in the presence of screened molecules derived from our in-house compound library (Step 1). Cells were then subjected to Cu(I)-catalyzed [3 + 2] cycloaddition to couple the transported glucosyl azide with alkyne probe in lysis buffer (Step 2). Subsequently, the fluorescence intensity of the generated fluorescently labeled triazolyl glycans were detected by fluorescence
184
Please cite this article in press as: Wu, S.-H., et al. Development and application of a fluorescent glucose uptake assay for the high-throughput screening of non-glycoside SGLT2 inhibitors. Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.03.011
144 145 146 147 148
151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171
174 175 176 177 178 179 180 181
185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
PHASCI 3213
No. of Pages 5, Model 5G
31 March 2015 S.-H. Wu et al. / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx
Screened Molecules
N3
Probe
O
Step 2
SGLT2 Na+
Fluorescent Adduct
O
N3
3
N N
N
O
Step 3
Na+
Inhibition Fluorescence Fluorescence Reader Step 1 Fig. 1. Strategy for specific fluorescent labeling of azidoglucose transported into cells. Step 1: the uptake of azidoglucose via sodium-dependent glucose co-transporter 2 in the presence of screened molecules. Step 2: cells were subjected to Cu(I)-catalyzed [3 + 2] cycloaddition with alkyne-probe in lysis buffer. Step3: the fluorescently labeled triazolyl glycans were detected by fluorescence reader.
201 202 203 204
205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230
reader to find compounds with inhibitory activity against SGLT2 (Step 3). This is a competition assay, i.e. if the screened molecules exhibit good SGLT2 inhibition, less azidoglucose will be delivered into cells and therefore a weak fluorescence signal will result. 3.1.2. Identification of the substrate and its most suitable concentration The modified sugar analogues and fluorogenic alkyne probe used in this study are presented in Fig. 2; a detailed description of these substrates is published as Supporting Information. With the desired materials in hand, we tested whether azidoglucoses were indeed SGLT2 substrates. Stably transfected CHO-K1 cells expressing hSGLT2 were incubated with 25 mM azidoglucose (1– 7) in the absence or presence of 50 lM phlorizin in KRH buffer for 3 h at 37 °C. At the end of the uptake period, the medium was removed and washed thrice with KRH buffer to remove the untransported azidosugars. Cells were then treated with click-activated alkyne probe, 4-ethynyl-N-ethyl-1,8-naphthalimide 8, in SDS lysis buffer and incubated at 37 °C for another 3 h to afford the desired fluorescence adduct. Next, the fluorescence intensity of the generated fluorescently labeled triazolyl glycan was measured in a Victor2-V fluorescence plate reader with an excitation/ emission wavelength of 355 nm/460 nm. The C1N3-glucose-treated cells were found to exhibit a distinct increase in fluorescence signal as compared to other azidoglucoses (Fig. 3A). The resulting triazolyl product, obtained from Cu(I)-catalyzed [3 + 2] cycloaddition of alkyne probe 8 with C1N3-glucose 1, has been proven to show strong emission at 460 nm previously (data presented in Supplementary Fig. 1). Interestingly, only 70.5% increased azidoglucose uptake was inhibited by phlorizin, which may be because some other glucose transporters presented in CHO-K1 OH
OH HO HO
O
N3
OH 1: C1N3
HO HO
OH N3 HO
O HO OH 4: C4N3
O N3 OH 2: C2N3
OH HO N3
N3 HO HO
O
O
O HO OH 3: C3N3 Et N
O
HO OR
5: C6N3; OR = α,β-OH 6: α-OMeC6N3; OR = α-OMe 7: β-OMeC6N3; OR = β-OMe
8: Probe
Fig. 2. Modified sugar analogues and probe used in this study.
cells are not suppressed by phlorizin. These results prove that C1N3-glucose is the substrate of SGLT2 in this cell-based system. The most suitable concentration of C1N3-glucose for the uptake was also evaluated. As shown in Fig. 3B, the significant difference in fluorescence intensity in the absence or presence of phlorizin was observed when the concentration of C1N3-glucose ranged from 6.25 mM to 50.0 mM. In addition, the preliminary data of robustness, or Z0 factor (Zhang et al., 1999), suggested the best concentration range of C1N3-glucose for uptake to be 12.5 mM to 25 mM (data not shown). These positive results constitute a good starting point for assay development.
231
3.1.3. Evaluation of the seeding density and the time course of Cu(I)-catalyzed [3 + 2] cycloaddition Having confirmed that C1N3-glucose is a substrate of SGLT2, we evaluated the seeding density of cells on C1N3-glucose uptake. Cells were seeded at increasing densities (from 10,000 to 40,000 cells/ well), treated with 25 mM C1N3-glucose, reacted with clickactivated alkyne probe, and the fluorescence measured. The observed data revealed no difference in fluorescence intensity when the numbers of cells are greater than 20,000 per well (Fig. 4A). In the same experiment, we also conducted exploratory analysis on the duration of the C1N3-glucose uptake. A three hour incubation time post seeding at 20,000 cells/well is considered an optimal condition. The time course of Cu(I)-catalyzed [3 + 2] cycloaddition of transported C1N3-glucose with alkyne-probe was then explored. As shown in Fig. 4B, the reaction reached equilibrium after a 3 h-incubation at 37 °C, and this was chosen as the standard reaction time.
242
3.1.4. The performance of the established fluorescence assay As the assay details were developed, two well-known SGLT2 inhibitors, phlorizin and dapagliflozin, were used to test whether the established fluorescence assay could be used for the evaluation of inhibitory activity against SGLT2. As shown in Fig. 5, the obtained EC50 values of these two SGLT2 inhibitors were 300.8 nM and 7.0 nM, respectively, consistent with the trend of the inhibitory activity found using the [14C]-AMG radioactive assay (EC50 = 85.2 nM and 1.2 nM, respectively). It is well known that non-radioactive assays are in general less sensitive than radioactive ones; however, the safer processes and cheaper substrate used, in our opinion, greatly outweigh this disadvantage.
259
3.2. High-throughput screening of non-glycoside SGLT2 inhibitors
271
To ensure the feasible operation and the data quality of highthroughput screening in 96-well plates, we optimized the
272
Please cite this article in press as: Wu, S.-H., et al. Development and application of a fluorescent glucose uptake assay for the high-throughput screening of non-glycoside SGLT2 inhibitors. Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.03.011
232 233 234 235 236 237 238 239 240 241
243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258
260 261 262 263 264 265 266 267 268 269 270
273
PHASCI 3213
No. of Pages 5, Model 5G
31 March 2015 4
S.-H. Wu et al. / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx
without PZN
(A)
% of Control
80 60 40
80 60 40
R H K
-C 6N
3
6.25
12.50 25.00 C1N3 (mM)
50.00
-O
-O
M e
M e
-C 6N
3 C 6N
3 C 4N
C 3N
3
3
0
3
20
0
C 2N
with PZN
100
20
C 1N
without PZN
120
100
3
% of Control
(B)
with PZN
120
β
α
Fig. 3. (A) Azidoglucoses uptake was measured in hSGLT2-transfected CHO-K1 cells in the presence and absence of phlorizin (PZN). (B) Uptake concentration evaluation of the substrate C1N3-glucose.
(B)
(A) % of Control
10000 cells / well 20000 cells / well 30000 cells / well 40000 cells / well
100 80 60 40
% of Control
120
120
Blank Glc-C1N3 Glc-C1N3 + PZN
80 60 40 20
20 0
100
0 0
1
2
3
4
5
0
10
20
30
40
50
Time (h)
Uptake time (h)
Fig. 4. (A) Effects of cell density and incubation time on C1N3-glucose uptake. (B) Time course evaluation of Cu(I)-catalyzed [3 + 2] cycloaddition of transported C1N3-glucose with alkyne-probe.
(A) Compound
Radioactivity Assay EC50 (nM)
Fluorescence Assay EC50 (nM)
Fold
Phlorizin (PZN)
85.2 ± 25.2
300.8 ± 43.2
3.5
Dapagliflozin (DAPA)
1.2 ± 0.2
7.0 ± 1.0
5.8
(B) [14C]-AMG Radioactive Assay
(C) Fluorescence Assay
Fig. 5. (A) The EC50 values of phlorizin and dapagliflozin in the radioactivity assay and fluorescence assay. (B) Phlorizin and dapagliflozin dose response curves obtained in the [14C]-AMG radioactive assay. (C) Phlorizin and dapagliflozin dose response curves obtained in the fluorescence assay.
274 275 276 277 278 279 280 281 282 283
concentration of C1N3-glucose and the duration of Cu(I)-catalyzed [3 + 2] cycloaddition. It was found that the fluorescence system performed best at 15 mM C1N3-glucose treatment and an 18 hincubation with alkyne probe 8. Subsequently, the Z0 factor was determined to evaluate whether the data was suitable for highthroughput screening assay under the optimized conditions: 20,000 hSGLT2/CHO-K1 cells per well were seeded in 96-well plates, 3 h-incubation with 15 mM C1N3-glucose, and 50 lM tested compounds, followed by 18 h Cu(I)-catalyzed [3 + 2] cycloaddition. The Z0 factor was found to be 0.77 (Fig. 6A), indicating that this
system is reliable and meets the criteria of an excellent assay for high-throughput screening. A schematic diagram of high-throughput screening to identify hits against hSGLT2 is shown in Fig. 6B. In the primary screening, 96-well microtiter plates (80 compounds tested per/plate) were used to analyze 41,000 compounds in duplicate for their ability to inhibit C1N3-glucose transport via SGLT2. Of these, 438 compounds which exhibited >70% inhibition of SGLT2 activity when tested at 50 lM concentration were selected for next stage evaluation. To select hit compounds with potent activity and low
Please cite this article in press as: Wu, S.-H., et al. Development and application of a fluorescent glucose uptake assay for the high-throughput screening of non-glycoside SGLT2 inhibitors. Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.03.011
284 285 286 287 288 289 290 291 292 293
PHASCI 3213
No. of Pages 5, Model 5G
31 March 2015 5
S.-H. Wu et al. / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx
(A)
(B)
NO2
H N N
O2 N
Z' factor = 0.77
HO
hSGLT2 EC 50 = 0.62 μM
1.0
120
0.6
% of Control
Z' Factor
0.8
0.4 0.2 0.0
0
20
40
60
80
100
100 80 60 40 20
Number of Plates
0 -9
-8
-7
-6
-5
-4
log[Drug] (M) Fig. 6. (A) Z0 factor determination under the optimized culture conditions, 3 h-incubation with C1N3-glucose/tested compounds followed by 18 h Cu(I)-catalyzed [3 + 2] cycloaddition. (B) Schematic representation of high-throughput screening and the resulting benzimidazole hit compound.
304
cytotoxicity, the MTS cell viability assay was performed in both CHO-K1 and Detroit 551 cells, and 156 hits were obtained with a criterion of a cell survival rate of >70%. Next, these compounds were tested at a 20 lM concentration using a SGLT2 [14C]-AMG radioactivity assay, the secondary screening assay, to confirm their inhibitory activity. Fourteen compounds exhibited >50% inhibition and were submitted for EC50 determination. The EC50 values obtained were from 17.7 to 0.62 lM, the most potent of which contained a benzimidazole scaffold (Fig. 6B). This constitutes an exciting starting hit for the future development of novel non-glycoside SGLT2 inhibitors.
305
4. Conclusion
306
317
A novel fluorescent glucose uptake assay based on a Cu(I)-catalyzed [3 + 2] cycloaddition has been successfully established. Although the developed system is not as sensitive as the radioactive assay, the consistent nature of the results obtained gave us the confidence to undertake a high-throughput screening of nonglycoside hits against hSGLT2 using this assay. A total of 41,000 molecules were screened, and 14 inhibitory hits with an EC50 of <20 lM were identified. The most potent hit, a benzimidazole derivative with an EC50 value of 0.62 lM, was chosen for further studies that will be helpful in the elucidation of the role of non-glycoside SGLT2 inhibitors, and perhaps the development of novel type 2 diabetes treatments.
318
5. Uncited reference
294 295 296 297 298 299 300 301 302 303
307 308 309 310 311 312 313 314 315 316
319
Zhou and Fahrni (2004).
320
Acknowledgements
321
324
We are grateful to the National Health Research Institutes and National Science Council of Taiwan (NSC 97-2323-B-400-002, NSC 98-2323-B-400-002, and NSC 99-2323-B-400-003) for financial support.
325
Appendix A. Supplementary material
326
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ejps.2015.03.011.
322 323
327
References
328
Abdul-Ghani, M.A., Norton, L., DeFronzo, R.A., 2011. Role of sodium–glucose cotransporter 2 (SGLT2) inhibitors in the treatment of type 2 diabetes. Endocr. Rev. 32, 515–531. Best, M.D., 2009. Click chemistry and bioorthogonal reactions: unprecedented selectivity in the labeling of biological molecules. Biochemistry 48, 6571–6584. Castaneda, F., Kinne, R.K., 2005. A 96-Well automated method to study inhibitors of human sodium-dependent D-glucose transport. Mol. Cell. Biochem. 280, 91–98. Ehrenkranz, J.R.L., Lewis, N.G., Kahn, C.R., Roth, J., 2005. Phlorizin: a review. Diabetes Metab. Res. Rev. 21, 31–38. Elkinson, S., Scott, L.J., 2013. Canagliflozin: first global approval. Drugs 73, 979–988. Fujita, Y., Inagaki, N., 2014. Renal sodium glucose cotransporter 2 inhibitors as a novel therapeutic approach to treatment of type 2 diabetes: clinical data and mechanism of action. J. Diabetes Invest. 5, 265–275. Hertzberg, R.P., Pope, A.J., 2000. High-throughput screening: new technology for the 21st century. Curr. Opin. Chem. Biol. 4, 445–451. Ho, L.T., Kulkarni, S.S., Lee, J.C., 2011. Development of sodium-dependent glucose co-transporter 2 inhibitors as potential anti-diabetic therapeutics. Curr. Top. Med. Chem. 11, 1476–1512. Li, A.R., Zhang, J., Greenberg, J., Lee, T., Liu, J., 2011. Discovery of non-glucoside SGLT2 inhibitors. Bioorg. Med. Chem. Lett. 21, 2472–2475. Liu, B., Li, S., Hu, J., 2004. Technological advances in high-throughput screening. Am. J. Pharmacogenomics 4, 263–276. Nauck, M.A., 2014. Update on developments with SGLT2 inhibitors in the management of type 2 diabetes. Drug Des. Dev. Ther. 8, 1335–1380. Neumiller, J.J., 2014. Empagliflozin: a new sodium–glucose co-transporter 2 (SGLT2) inhibitor for the treatment of type 2 diabetes. Drugs Context 3, 212262. Poole, R.M., Dungo, R.T., 2014. Ipragliflozin: first global approval. Drugs 74, 611– 617. Poole, R.M., Prossler, J.E., 2014. Tofogliflozin: first global approval. Drugs 74, 939– 944. Rostovtsev, V.V., Green, L.G., Fokin, V.V., Sharpless, K.B., 2002. A stepwise Huisgen cycloaddition process: copper(I)-catalyzed regioselective ‘‘Ligation’’ of azides and terminal alkynes. Angew. Chem. Int. Ed. 41, 2596–2599. Sawa, M., Hsu, T.L., Itoh, T., Sugiyama, M., Hanson, S.R., Vogt, P.K., Wong, C.H., 2006. Glycoproteomic probes for fluorescent imaging of fucosylated glycans in vivo. Proc. Natl. Acad. Sci. U.S.A. 103, 12371–12376. Traynor, K., 2014. Dapagliflozin approved for type 2 diabetes. Am. J. Health Syst. Pharm. 71, 263. Wright, E.M., 2001. Renal Na+–glucose cotransporters. Am. J. Physiol. Ren. Physiol. 280, F10–F18. Wright, E.M., Turk, E., 2004. The sodium/glucose cotransport family SLC5. Pflugers Arch. 447, 510–518. Zhang, X., Zhang, Y., 2013. Applications of azide-based bioorthogonal click chemistry in glycobiology. Molecules 18, 7145–7159. Zhang, J.H., Chung, T.D.Y., Oldenburg, K.R., 1999. A simple statistical parameter for use in evaluation and validation of high throughput screening assays. J. Biomol. Screen. 4, 67–73. Zhou, Z., Fahrni, C.J., 2004. A fluorogenic probe for the copper(I)-catalyzed azide– alkyne ligation reaction: modulation of the fluorescence emission via 3(n, pi)1(pi, pi) inversion. J. Am. Chem. Soc. 126, 8862–8863.
329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380
Please cite this article in press as: Wu, S.-H., et al. Development and application of a fluorescent glucose uptake assay for the high-throughput screening of non-glycoside SGLT2 inhibitors. Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.03.011
381
No. of Pages 5, Model 5G
31 March 2015 European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx 1
Contents lists available at ScienceDirect
European Journal of Pharmaceutical Sciences journal homepage: www.elsevier.com/locate/ejps 5 6 3 4
Development and application of a fluorescent glucose uptake assay for the high-throughput screening of non-glycoside SGLT2 inhibitors
8
Szu-Huei Wu 1, Chun-Hsu Yao 1, Chieh-Jui Hsieh 2, Yu-Wei Liu, Yu-Sheng Chao 3, Jen-Shin Song ⇑, Jinq-Chyi Lee ⇑
9
Institute of Biotechnology and Pharmaceutical Research, National Health Research Institutes, 35 Keyan Road, Zhunan Town, Miaoli County 35053, Taiwan, ROC
7
10 11 1 2 3 6 14 15 16 17 18 19 20 21 22 23 24 25
a r t i c l e
i n f o
Article history: Received 12 December 2014 Received in revised form 5 March 2015 Accepted 20 March 2015 Available online xxxx Keywords: Sodium-dependent glucose co-transporter Type 2 diabetes mellitus Non-glycoside High-throughput screening Click chemistry
a b s t r a c t Sodium-dependent glucose co-transporter 2 (SGLT2) inhibitors are of current interest as a treatment for type 2 diabetes. Efforts have been made to discover phlorizin-related glycosides with good SGLT2 inhibitory activity. To increase structural diversity and better understand the role of non-glycoside SGLT2 inhibitors on glycemic control, we initiated a research program to identify non-glycoside hits from high-throughput screening. Here, we report the development of a novel, fluorogenic probe-based glucose uptake system based on a Cu(I)-catalyzed [3 + 2] cycloaddition. The safer processes and cheaper substances made the developed assay our first priority for large-scale primary screening as compared to the well-known [14C]-labeled a-methyl-D-glucopyranoside ([14C]-AMG) radioactive assay. This effort culminated in the identification of a benzimidazole, non-glycoside SGLT2 hit with an EC50 value of 0.62 lM by high-throughput screening of 41,000 compounds. Ó 2015 Elsevier B.V. All rights reserved.
27 28 29 30 31 32 33 34 35 36 37 38 39
40 41
1. Introduction
42
Sodium-dependent glucose co-transporter 2 (SGLT2) is a highcapacity, low-affinity transporter located on the S1 segment of the proximal tubule in the kidney, and belongs to the 12 member solute carrier family 5A (SLC5A). It reabsorbs the majority (90%) of renal glucose filtered by kidney glomerulus back into the blood stream (Wright, 2001; Wright and Turk, 2004). Inhibition of SGLT2 has been demonstrated to suppress renal glucose reabsorption, increase urine glucose excretion, and consequentially reduce glycemic levels. This mode of action constitutes an insulinindependent pathway to control hyperglycemia, without the risk of hypoglycemia (Abdul-Ghani et al., 2011). SGLT2 inhibitors were also found to result in modest weight loss and blood pressure reduction, both additional advantages which make SGLT2 inhibitors more attractive in the treatment of type 2 diabetes (Nauck, 2014).
43 44 45 46 47 48 49 50 51 52 53 54 55 56
⇑ Corresponding authors. Tel.: +886 37 246166x35739 (J.-S. Song), +886 37 246166x35781 (J.-C. Lee). E-mail addresses: [email protected] (J.-S. Song), [email protected] (J.-C. Lee). 1 S.-H. Wu and C.-H. Yao equally contributed to this work. 2 Present address: TPV Technology Group, Top Victory Electronics (Taiwan) Co., Ltd., 12F-1, No. 186, Jian 1st Rd., Zhonghe Dist., New Taipei City 23511, Taiwan, ROC. 3 Present address: Diamond BioFund INC., 5F., No. 3, Sec. 1, Dunhua S. Rd., Taipei City 105, Taiwan, ROC.
Several SGLT2 inhibitors have been developed and disclosed. Some are in the advanced stages of clinical trials (Ho et al., 2011), and all are C-linked b-glycosides derived from the structural modification of phlorizin, the natural product isolated from the root bark of the apple tree and subsequently identified as the first SGLT2 inhibitor (Ehrenkranz et al., 2005). Of these, dapagliflozin, canagliflozin, ipragliflozin, tofogliflozin and empagliflozin were recently approved for the treatment of type 2 diabetes mellitus (T2DM) (Traynor, 2014; Elkinson and Scott, 2013; Poole and Dungo, 2014; Poole and Prossler, 2014; Neumiller, 2014). However, their use requires long-term follow-up due to the potential for cancer, liver injury, cardiovascular diseases, and/or other unexpected side-effects (Fujita and Inagaki, 2014). The overwhelming majority of known SGLT2 inhibitors are phlorizin-related glycosides. Identification of non-glycoside inhibitors could be advantageous, and the use of high-throughput screening is an important strategy, having been utilized in drug discovery for many years to identify active hits from large libraries of compounds (Hertzberg and Pope, 2000; Liu et al., 2004). The most widely used in vitro assay for evaluating inhibitory activities against SGLT2 is a radioactive functional assay, which is based on the suppression of uptake of [14C]-labeled a-methyl-Dglucopyranoside ([14C]-AMG) into cells stably expressing SGLT2 (Castaneda and Kinne, 2005). Unfortunately, this system is severely limited by the high cost of the substrate and the difficulties associated with the handling and disposal of radioactive substances, and
http://dx.doi.org/10.1016/j.ejps.2015.03.011 0928-0987/Ó 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: Wu, S.-H., et al. Development and application of a fluorescent glucose uptake assay for the high-throughput screening of non-glycoside SGLT2 inhibitors. Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.03.011
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
PHASCI 3213
No. of Pages 5, Model 5G
31 March 2015 2
S.-H. Wu et al. / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx
97
we declined to use it in the large-scale primary screening for nonglycoside SGLT2 inhibitors even though it has been previously applied in drug discovery and in an automated, 96-well format (Li et al., 2011; Castaneda and Kinne, 2005). Instead we sought to develop a new assay system with safer processes and a cheaper substrate. Herein, we outline a fluorescence system for probing the glucose transported by hSGLT2 into cells, which should obviate the aforementioned shortcomings of the radioassay. In this approach, the uptake azido-modified glucose analogues are labeled with a fluorogenic probe, 4-ethynyl-N-ethyl-1,8-naphthalimide, using a Cu(I)-catalyzed [3 + 2] cycloaddition (azide–alkyne ‘‘click’’ chemistry) (Sawa et al., 2006; Rostovtsev et al., 2002). We also exemplify the utility of this newly developed fluorescence assay in high-throughput screening by the identification of non-glycoside SGLT2 inhibitor hits.
98
2. Materials and methods
83 84 85 86 87 88 89 90 91 92 93 94 95 96
99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114
2.1. Establishment of human SGLT-2 stable cell line The hSGLT-2 cDNA was subcloned into pIRES2-EGFP vector (Clontech Laboratories, Inc., Mountain View, CA, USA). Before transfection, cells were seeded in 24-well plates at a density of 2 105 cells/well. Cells were cultured until they reached 70–80% confluence and then transfected with 1 lg of vector containing the hSGLT-2 coding region using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) per well. Transfected CHO-K1 cell stably expressed hSGLT-2 (hSGLT-2/CHO-K1) was selected by EGFP and 1 mg/ml G418 (Calbiochem, Merck-Millipore, Darmstadt, Germany) for 10 to 14 days. The selected clone was maintained in Ham’s F12 medium (HyClone, Logan, UT, USA) supplemented with 10% fetal bovine serum (HyClone, Logan, UT, USA), 1% of penicillin–streptomycin (Biological Industries, Beit-Haemek, Israel), and 0.5 mg/ml G418 with 5% CO2 at 37 °C in the humidified incubator.
115
2.2. Radioactive sodium dependent glucose co-transport 2 assay
116
The transport assays were performed according to method from Castaneda and Kinne with some modification (Castaneda and Kinne, 2005). Stably transfected CHO-K1 cells (hSGLT-2/CHO-K1) were used for transporter studies. Sodium-dependent glucose cotransporter was determined by means of [14C]-a-methyl-Dglucopyranoside ([14C]-AMG, specific radioactivity 300 mCi/mmol) purchased from Perkin Elmer (Boston, USA). For the purpose of this study, Krebs–Ringer–Henseleit (KRH) solution containing 120 mM NaCl, 4.7 mM KCl, 1.2 mM MgCl2, 2.2 mM CaCl2, 10 mM Hepes (pH 7.4 with Tris) was used as rinse buffer. All chemicals were purchased from Sigma (Deisenhofen, Germany). Briefly, hSGLT2/ CHO-K1 cells were seeded into a white-walled 96-well culture plate (Corning, NY, USA) at a density of 3 104 cells/well and incubated for 48 h at 37 °C with 5% CO2 atmosphere in growth medium. After 48 h, the culture medium of the 96-wells was taken off and wells were rinsed twice with 200 lL of KRH-Na and incubated in KRH-Na containing 3 lM [14C]-AMG in the absence or presence of inhibitors for up to 120 min at 37 °C. At the end of the uptake period, the transport buffer was removed and the uptake of [14C]-AMG was stopped by adding ice-cold KRH-Na containing 0.5 mM phlorizin. The wells were rinsed three times with 100 lL stop buffer using the microplate washer (Tecan, Männedorf, Switzerland). After the third rinse, the stop buffer was completely removed from the wells and the cells were solubilized by adding 1% sodium dodecyl sulfate (Sigma). After 24 h, the microtiter plate was taken for scintillation counting of radioactive [14C]-AMG using a TopCount (Perkin Elmer). The percent of inhibition was
117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142
calculated by comparing counts per minute (cpm) in the inhibitor-containing well with cpm in the well containing only the DMSO vehicle. Phlorizin and dapagliflozin were evaluated in parallel in every assay. A dose–response curve was fitted to a sigmoidal dose–response model using GraphPad software to determine the inhibitor concentration at half-maximal response (EC50).
143
2.3. Fluorescent sodium dependent glucose co-transport 2 assay
149
The stably transfected CHO-K1 clone hSGLT-2/CHO-K1 was incubated with indicated concentration of azidoglucose in the absence or presence of 10–50 lM phlorizin in KRH buffer (120 mM NaCl, 4.7 mM KCl, 1.2 mM MgCl2, 2.2 mM CaCl2, 10 mM HEPES with Tris, pH = 7.4) for 3 h at 37 °C. At the end of the uptake period, the medium was removed and washed thrice with KRH buffer to remove the un-transported azidoglucose. Cells were then treated with click-activated alkyne probe, 4-ethynyl-N-ethyl-1,8naphthalimide 8, in SDS lysis buffer (100 lM 8, 250 lM triazole ligand (Zhang and Zhang, 2013), 5 mM sodium ascorbate, 2.5 mM CuSO45H2O, 0.1% SDS-PBS) and incubated at 37 °C for 3 h (for high-throughput screening, the duration of the incubation is 18 h) to afford the desired fluorescence adduct. Next, the fluorescence intensity of the generated fluorescently labeled triazolyl glycan was measured by a Victor2-V fluorescence plate reader with an excitation/emission wavelength of 355 nm/460 nm. The percent of inhibition of inhibitor was calculated by comparing fluorescence intensity (relative fluorescence unit, RFU) in inhibitor-containing well with RFU in well containing only DMSO vehicle. A dose– response curve was calculated by nonlinear regression (GraphPad software, San Diego, CA) to determine the functional inhibitory concentration at half-maximal response (EC50).
150
2.4. MTS cell viability assay
172
Cells were seeded in 96-well clear plates at a density of 1 104 cells per well overnight. Then, cells were treated with 10 lM compounds for 72 h. At the end of incubation, for a 96-well microtiter plate, 12 ml of MTS Mix reagent contained culture medium, MTS (Promega, Madison, WI), and PMS (Sigma, St. Louis, MO) in a ratio of 8:2:0.1, respectively. The medium in well was removed and the MTS Mix reagent was then added to cells (100 lL/well). The plates were incubated for 2.5 h at 37 °C in a humidified, 5% CO2 atmosphere and the absorbance was then recorded at 490 nm.
173
3. Results and discussion
182
3.1. Assay development
183
3.1.1. Strategy for specific fluorescent SGLT2 assay The bioorthogonal reactions of click chemistry have been well studied in biomolecule labeling and detection (Best, 2009; Zhang and Zhang, 2013). The remarkable results obtained encouraged us to incorporate it into a non-radioactive assay strategy, which is depicted in Fig. 1. The function of SGLT2 is to transport glucose and sodium into cells using the sodium gradient created by sodium/potassium ATPase pumps in the basolateral cell membranes. In the developed assay, we substituted glucose for azidoglucoses, which were hypothesized to be transported into cells via SGLT2. Firstly, hSGLT2/CHO-K1 cells were incubated with azido-glucoses in the presence of screened molecules derived from our in-house compound library (Step 1). Cells were then subjected to Cu(I)-catalyzed [3 + 2] cycloaddition to couple the transported glucosyl azide with alkyne probe in lysis buffer (Step 2). Subsequently, the fluorescence intensity of the generated fluorescently labeled triazolyl glycans were detected by fluorescence
184
Please cite this article in press as: Wu, S.-H., et al. Development and application of a fluorescent glucose uptake assay for the high-throughput screening of non-glycoside SGLT2 inhibitors. Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.03.011
144 145 146 147 148
151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171
174 175 176 177 178 179 180 181
185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
PHASCI 3213
No. of Pages 5, Model 5G
31 March 2015 S.-H. Wu et al. / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx
Screened Molecules
N3
Probe
O
Step 2
SGLT2 Na+
Fluorescent Adduct
O
N3
3
N N
N
O
Step 3
Na+
Inhibition Fluorescence Fluorescence Reader Step 1 Fig. 1. Strategy for specific fluorescent labeling of azidoglucose transported into cells. Step 1: the uptake of azidoglucose via sodium-dependent glucose co-transporter 2 in the presence of screened molecules. Step 2: cells were subjected to Cu(I)-catalyzed [3 + 2] cycloaddition with alkyne-probe in lysis buffer. Step3: the fluorescently labeled triazolyl glycans were detected by fluorescence reader.
201 202 203 204
205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230
reader to find compounds with inhibitory activity against SGLT2 (Step 3). This is a competition assay, i.e. if the screened molecules exhibit good SGLT2 inhibition, less azidoglucose will be delivered into cells and therefore a weak fluorescence signal will result. 3.1.2. Identification of the substrate and its most suitable concentration The modified sugar analogues and fluorogenic alkyne probe used in this study are presented in Fig. 2; a detailed description of these substrates is published as Supporting Information. With the desired materials in hand, we tested whether azidoglucoses were indeed SGLT2 substrates. Stably transfected CHO-K1 cells expressing hSGLT2 were incubated with 25 mM azidoglucose (1– 7) in the absence or presence of 50 lM phlorizin in KRH buffer for 3 h at 37 °C. At the end of the uptake period, the medium was removed and washed thrice with KRH buffer to remove the untransported azidosugars. Cells were then treated with click-activated alkyne probe, 4-ethynyl-N-ethyl-1,8-naphthalimide 8, in SDS lysis buffer and incubated at 37 °C for another 3 h to afford the desired fluorescence adduct. Next, the fluorescence intensity of the generated fluorescently labeled triazolyl glycan was measured in a Victor2-V fluorescence plate reader with an excitation/ emission wavelength of 355 nm/460 nm. The C1N3-glucose-treated cells were found to exhibit a distinct increase in fluorescence signal as compared to other azidoglucoses (Fig. 3A). The resulting triazolyl product, obtained from Cu(I)-catalyzed [3 + 2] cycloaddition of alkyne probe 8 with C1N3-glucose 1, has been proven to show strong emission at 460 nm previously (data presented in Supplementary Fig. 1). Interestingly, only 70.5% increased azidoglucose uptake was inhibited by phlorizin, which may be because some other glucose transporters presented in CHO-K1 OH
OH HO HO
O
N3
OH 1: C1N3
HO HO
OH N3 HO
O HO OH 4: C4N3
O N3 OH 2: C2N3
OH HO N3
N3 HO HO
O
O
O HO OH 3: C3N3 Et N
O
HO OR
5: C6N3; OR = α,β-OH 6: α-OMeC6N3; OR = α-OMe 7: β-OMeC6N3; OR = β-OMe
8: Probe
Fig. 2. Modified sugar analogues and probe used in this study.
cells are not suppressed by phlorizin. These results prove that C1N3-glucose is the substrate of SGLT2 in this cell-based system. The most suitable concentration of C1N3-glucose for the uptake was also evaluated. As shown in Fig. 3B, the significant difference in fluorescence intensity in the absence or presence of phlorizin was observed when the concentration of C1N3-glucose ranged from 6.25 mM to 50.0 mM. In addition, the preliminary data of robustness, or Z0 factor (Zhang et al., 1999), suggested the best concentration range of C1N3-glucose for uptake to be 12.5 mM to 25 mM (data not shown). These positive results constitute a good starting point for assay development.
231
3.1.3. Evaluation of the seeding density and the time course of Cu(I)-catalyzed [3 + 2] cycloaddition Having confirmed that C1N3-glucose is a substrate of SGLT2, we evaluated the seeding density of cells on C1N3-glucose uptake. Cells were seeded at increasing densities (from 10,000 to 40,000 cells/ well), treated with 25 mM C1N3-glucose, reacted with clickactivated alkyne probe, and the fluorescence measured. The observed data revealed no difference in fluorescence intensity when the numbers of cells are greater than 20,000 per well (Fig. 4A). In the same experiment, we also conducted exploratory analysis on the duration of the C1N3-glucose uptake. A three hour incubation time post seeding at 20,000 cells/well is considered an optimal condition. The time course of Cu(I)-catalyzed [3 + 2] cycloaddition of transported C1N3-glucose with alkyne-probe was then explored. As shown in Fig. 4B, the reaction reached equilibrium after a 3 h-incubation at 37 °C, and this was chosen as the standard reaction time.
242
3.1.4. The performance of the established fluorescence assay As the assay details were developed, two well-known SGLT2 inhibitors, phlorizin and dapagliflozin, were used to test whether the established fluorescence assay could be used for the evaluation of inhibitory activity against SGLT2. As shown in Fig. 5, the obtained EC50 values of these two SGLT2 inhibitors were 300.8 nM and 7.0 nM, respectively, consistent with the trend of the inhibitory activity found using the [14C]-AMG radioactive assay (EC50 = 85.2 nM and 1.2 nM, respectively). It is well known that non-radioactive assays are in general less sensitive than radioactive ones; however, the safer processes and cheaper substrate used, in our opinion, greatly outweigh this disadvantage.
259
3.2. High-throughput screening of non-glycoside SGLT2 inhibitors
271
To ensure the feasible operation and the data quality of highthroughput screening in 96-well plates, we optimized the
272
Please cite this article in press as: Wu, S.-H., et al. Development and application of a fluorescent glucose uptake assay for the high-throughput screening of non-glycoside SGLT2 inhibitors. Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.03.011
232 233 234 235 236 237 238 239 240 241
243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258
260 261 262 263 264 265 266 267 268 269 270
273
PHASCI 3213
No. of Pages 5, Model 5G
31 March 2015 4
S.-H. Wu et al. / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx
without PZN
(A)
% of Control
80 60 40
80 60 40
R H K
-C 6N
3
6.25
12.50 25.00 C1N3 (mM)
50.00
-O
-O
M e
M e
-C 6N
3 C 6N
3 C 4N
C 3N
3
3
0
3
20
0
C 2N
with PZN
100
20
C 1N
without PZN
120
100
3
% of Control
(B)
with PZN
120
β
α
Fig. 3. (A) Azidoglucoses uptake was measured in hSGLT2-transfected CHO-K1 cells in the presence and absence of phlorizin (PZN). (B) Uptake concentration evaluation of the substrate C1N3-glucose.
(B)
(A) % of Control
10000 cells / well 20000 cells / well 30000 cells / well 40000 cells / well
100 80 60 40
% of Control
120
120
Blank Glc-C1N3 Glc-C1N3 + PZN
80 60 40 20
20 0
100
0 0
1
2
3
4
5
0
10
20
30
40
50
Time (h)
Uptake time (h)
Fig. 4. (A) Effects of cell density and incubation time on C1N3-glucose uptake. (B) Time course evaluation of Cu(I)-catalyzed [3 + 2] cycloaddition of transported C1N3-glucose with alkyne-probe.
(A) Compound
Radioactivity Assay EC50 (nM)
Fluorescence Assay EC50 (nM)
Fold
Phlorizin (PZN)
85.2 ± 25.2
300.8 ± 43.2
3.5
Dapagliflozin (DAPA)
1.2 ± 0.2
7.0 ± 1.0
5.8
(B) [14C]-AMG Radioactive Assay
(C) Fluorescence Assay
Fig. 5. (A) The EC50 values of phlorizin and dapagliflozin in the radioactivity assay and fluorescence assay. (B) Phlorizin and dapagliflozin dose response curves obtained in the [14C]-AMG radioactive assay. (C) Phlorizin and dapagliflozin dose response curves obtained in the fluorescence assay.
274 275 276 277 278 279 280 281 282 283
concentration of C1N3-glucose and the duration of Cu(I)-catalyzed [3 + 2] cycloaddition. It was found that the fluorescence system performed best at 15 mM C1N3-glucose treatment and an 18 hincubation with alkyne probe 8. Subsequently, the Z0 factor was determined to evaluate whether the data was suitable for highthroughput screening assay under the optimized conditions: 20,000 hSGLT2/CHO-K1 cells per well were seeded in 96-well plates, 3 h-incubation with 15 mM C1N3-glucose, and 50 lM tested compounds, followed by 18 h Cu(I)-catalyzed [3 + 2] cycloaddition. The Z0 factor was found to be 0.77 (Fig. 6A), indicating that this
system is reliable and meets the criteria of an excellent assay for high-throughput screening. A schematic diagram of high-throughput screening to identify hits against hSGLT2 is shown in Fig. 6B. In the primary screening, 96-well microtiter plates (80 compounds tested per/plate) were used to analyze 41,000 compounds in duplicate for their ability to inhibit C1N3-glucose transport via SGLT2. Of these, 438 compounds which exhibited >70% inhibition of SGLT2 activity when tested at 50 lM concentration were selected for next stage evaluation. To select hit compounds with potent activity and low
Please cite this article in press as: Wu, S.-H., et al. Development and application of a fluorescent glucose uptake assay for the high-throughput screening of non-glycoside SGLT2 inhibitors. Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.03.011
284 285 286 287 288 289 290 291 292 293
PHASCI 3213
No. of Pages 5, Model 5G
31 March 2015 5
S.-H. Wu et al. / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx
(A)
(B)
NO2
H N N
O2 N
Z' factor = 0.77
HO
hSGLT2 EC 50 = 0.62 μM
1.0
120
0.6
% of Control
Z' Factor
0.8
0.4 0.2 0.0
0
20
40
60
80
100
100 80 60 40 20
Number of Plates
0 -9
-8
-7
-6
-5
-4
log[Drug] (M) Fig. 6. (A) Z0 factor determination under the optimized culture conditions, 3 h-incubation with C1N3-glucose/tested compounds followed by 18 h Cu(I)-catalyzed [3 + 2] cycloaddition. (B) Schematic representation of high-throughput screening and the resulting benzimidazole hit compound.
304
cytotoxicity, the MTS cell viability assay was performed in both CHO-K1 and Detroit 551 cells, and 156 hits were obtained with a criterion of a cell survival rate of >70%. Next, these compounds were tested at a 20 lM concentration using a SGLT2 [14C]-AMG radioactivity assay, the secondary screening assay, to confirm their inhibitory activity. Fourteen compounds exhibited >50% inhibition and were submitted for EC50 determination. The EC50 values obtained were from 17.7 to 0.62 lM, the most potent of which contained a benzimidazole scaffold (Fig. 6B). This constitutes an exciting starting hit for the future development of novel non-glycoside SGLT2 inhibitors.
305
4. Conclusion
306
317
A novel fluorescent glucose uptake assay based on a Cu(I)-catalyzed [3 + 2] cycloaddition has been successfully established. Although the developed system is not as sensitive as the radioactive assay, the consistent nature of the results obtained gave us the confidence to undertake a high-throughput screening of nonglycoside hits against hSGLT2 using this assay. A total of 41,000 molecules were screened, and 14 inhibitory hits with an EC50 of <20 lM were identified. The most potent hit, a benzimidazole derivative with an EC50 value of 0.62 lM, was chosen for further studies that will be helpful in the elucidation of the role of non-glycoside SGLT2 inhibitors, and perhaps the development of novel type 2 diabetes treatments.
318
5. Uncited reference
294 295 296 297 298 299 300 301 302 303
307 308 309 310 311 312 313 314 315 316
319
Zhou and Fahrni (2004).
320
Acknowledgements
321
324
We are grateful to the National Health Research Institutes and National Science Council of Taiwan (NSC 97-2323-B-400-002, NSC 98-2323-B-400-002, and NSC 99-2323-B-400-003) for financial support.
325
Appendix A. Supplementary material
326
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ejps.2015.03.011.
322 323
327
References
328
Abdul-Ghani, M.A., Norton, L., DeFronzo, R.A., 2011. Role of sodium–glucose cotransporter 2 (SGLT2) inhibitors in the treatment of type 2 diabetes. Endocr. Rev. 32, 515–531. Best, M.D., 2009. Click chemistry and bioorthogonal reactions: unprecedented selectivity in the labeling of biological molecules. Biochemistry 48, 6571–6584. Castaneda, F., Kinne, R.K., 2005. A 96-Well automated method to study inhibitors of human sodium-dependent D-glucose transport. Mol. Cell. Biochem. 280, 91–98. Ehrenkranz, J.R.L., Lewis, N.G., Kahn, C.R., Roth, J., 2005. Phlorizin: a review. Diabetes Metab. Res. Rev. 21, 31–38. Elkinson, S., Scott, L.J., 2013. Canagliflozin: first global approval. Drugs 73, 979–988. Fujita, Y., Inagaki, N., 2014. Renal sodium glucose cotransporter 2 inhibitors as a novel therapeutic approach to treatment of type 2 diabetes: clinical data and mechanism of action. J. Diabetes Invest. 5, 265–275. Hertzberg, R.P., Pope, A.J., 2000. High-throughput screening: new technology for the 21st century. Curr. Opin. Chem. Biol. 4, 445–451. Ho, L.T., Kulkarni, S.S., Lee, J.C., 2011. Development of sodium-dependent glucose co-transporter 2 inhibitors as potential anti-diabetic therapeutics. Curr. Top. Med. Chem. 11, 1476–1512. Li, A.R., Zhang, J., Greenberg, J., Lee, T., Liu, J., 2011. Discovery of non-glucoside SGLT2 inhibitors. Bioorg. Med. Chem. Lett. 21, 2472–2475. Liu, B., Li, S., Hu, J., 2004. Technological advances in high-throughput screening. Am. J. Pharmacogenomics 4, 263–276. Nauck, M.A., 2014. Update on developments with SGLT2 inhibitors in the management of type 2 diabetes. Drug Des. Dev. Ther. 8, 1335–1380. Neumiller, J.J., 2014. Empagliflozin: a new sodium–glucose co-transporter 2 (SGLT2) inhibitor for the treatment of type 2 diabetes. Drugs Context 3, 212262. Poole, R.M., Dungo, R.T., 2014. Ipragliflozin: first global approval. Drugs 74, 611– 617. Poole, R.M., Prossler, J.E., 2014. Tofogliflozin: first global approval. Drugs 74, 939– 944. Rostovtsev, V.V., Green, L.G., Fokin, V.V., Sharpless, K.B., 2002. A stepwise Huisgen cycloaddition process: copper(I)-catalyzed regioselective ‘‘Ligation’’ of azides and terminal alkynes. Angew. Chem. Int. Ed. 41, 2596–2599. Sawa, M., Hsu, T.L., Itoh, T., Sugiyama, M., Hanson, S.R., Vogt, P.K., Wong, C.H., 2006. Glycoproteomic probes for fluorescent imaging of fucosylated glycans in vivo. Proc. Natl. Acad. Sci. U.S.A. 103, 12371–12376. Traynor, K., 2014. Dapagliflozin approved for type 2 diabetes. Am. J. Health Syst. Pharm. 71, 263. Wright, E.M., 2001. Renal Na+–glucose cotransporters. Am. J. Physiol. Ren. Physiol. 280, F10–F18. Wright, E.M., Turk, E., 2004. The sodium/glucose cotransport family SLC5. Pflugers Arch. 447, 510–518. Zhang, X., Zhang, Y., 2013. Applications of azide-based bioorthogonal click chemistry in glycobiology. Molecules 18, 7145–7159. Zhang, J.H., Chung, T.D.Y., Oldenburg, K.R., 1999. A simple statistical parameter for use in evaluation and validation of high throughput screening assays. J. Biomol. Screen. 4, 67–73. Zhou, Z., Fahrni, C.J., 2004. A fluorogenic probe for the copper(I)-catalyzed azide– alkyne ligation reaction: modulation of the fluorescence emission via 3(n, pi)1(pi, pi) inversion. J. Am. Chem. Soc. 126, 8862–8863.
329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380
Please cite this article in press as: Wu, S.-H., et al. Development and application of a fluorescent glucose uptake assay for the high-throughput screening of non-glycoside SGLT2 inhibitors. Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.03.011
381