Luminol Electrochemiluminescence for the Analysis of Active

Mar 25, 2013 - A luminol electrochemiluminescence assay was reported to analyze active cholesterol at the plasma membrane in single mammalian cells...
11 downloads 0 Views 629KB Size
Download PDF
Article pubs.acs.org/ac

Luminol Electrochemiluminescence for the Analysis of Active Cholesterol at the Plasma Membrane in Single Mammalian Cells Guangzhong Ma,† Junyu Zhou,‡ Chunxiu Tian,† Dechen Jiang,*,† Danjun Fang,*,‡ and Hongyuan Chen† †

Key State Labortorary of Analytical Chemistry for Life Science and School of Chemistry and Chemical Engineering, Nanjing University, Jiangsu, 210093, China ‡ School of Pharmacy, Nanjing Medical University, Jiangsu, 210000, China S Supporting Information *

ABSTRACT: A luminol electrochemiluminescence assay was reported to analyze active cholesterol at the plasma membrane in single mammalian cells. The cellular membrane cholesterol was activated by the exposure of the cells to low ionic strength buffer or the inhibition of intracellular acyl-coA/cholesterol acyltransferase (ACAT). The active membrane cholesterol was reacted with cholesterol oxidase in the solution to generate a peak concentration of hydrogen peroxide on the electrode surface, which induced a measurable luminol electrochemiluminescence. Further treatment of the active cells with mevastatin decreased the active membrane cholesterol resulting in a drop in luminance. No change in the intracellular calcium was observed in the presence of luminol and voltage, which indicated that our analysis process might not interrupt the intracellular cholesterol trafficking. Single cell analysis was performed by placing a pinhole below the electrode so that only one cell was exposed to the photomultiplier tube (PMT). Twelve single cells were analyzed individually, and a large deviation on luminance ratio observed exhibited the cell heterogeneity on the active membrane cholesterol. The smaller deviation on ACAT/HMGCoA inhibited cells than ACAT inhibited cells suggested different inhibition efficiency for sandoz 58035 and mevastatin. The new information obtained from single cell analysis might provide a new insight on the study of intracellular cholesterol trafficking.

T

vided the information about the active membrane cholesterol.12,13 This method required large amount of the cells, and thus, it was difficult to apply the assay in single cell cholesterol analysis for the study of cell heterogeneity. Cholesterol oxidase was an enzyme that catalyzed the reaction of cholesterol and oxygen into cholestenone and hydrogen peroxide, which was detected using amplex red or quinoneimine assays.9,14 Since cholesterol oxidase was reported to react with only active cholesterol at the plasma membrane in cells, this method was applied to study the active membrane cholesterol at cells and the models of the lipid membrane.14−18 Although the commercial multiplate can detect 50 nM hydrogen peroxide using the amplex red assay, the sensitivity of the commercial multiplate for resorufin (the product of amplex red−hydrogen peroxide oxidation) was typically 10 pmol. Thus, 10 pmol of hydrogen peroxide or cholesterol was needed for the fluorescence assay. Considering the average molecular area of active cholesterol as 40 Å219 and the size of a single RAW264.7 cell was 10 μm, approximate 0.33 fmol of cholesterol was estimated at the outlet leaflet of the plasma membrane in a single cell. At least hundreds of cells were

he maintenance of cholesterol homeostasis is important for cell functions, which are carried out by cholesterol biosynthesis, uptake, release, and storage.1−3 Although cholesterol is mainly distributed on the entire plasma membrane of the cells, it also concentrates in specialized sphingolipid-rich domains called rafts and caveolae.4,5 The cholesterol in these domains exceeds the complexing capacity of the polar bilayer lipids leading to an elevated chemical activity (escape tendency), which is termed as active cholesterol.6−8 Recent evidence suggested that active cholesterol at the plasma membrane in the cells initialed the cholesterol trafficking in the cells, including the downregulation of cholesterol biosynthesis, less ingestion, and enhanced cholesterol efflux.9 Therefore, the full investigation of active membrane cholesterol in cells, especially at the single cell level, is significant for the study of intracellular cholesterol trafficking. Cyclodextrin and cholesterol oxidase assays have been developed to analyze active membrane cholesterol. In the cyclodextrin assay, cyclodextrin was applied to extract the cholesterol from the cellular plasma membrane.10 The rate and extent of cellular membrane cholesterol to cyclodextrin increased sharply when the active membrane cholesterol was formed.11 The determination of cholesterol amount in cyclodextrin solution by radiolabeling technique or highperformance liquid chromatography−mass spectrometry pro© 2013 American Chemical Society

Received: November 14, 2012 Accepted: March 25, 2013 Published: March 25, 2013 3912

dx.doi.org/10.1021/ac303304r | Anal. Chem. 2013, 85, 3912−3917

Analytical Chemistry

Article

required for the fluorescence analysis. The fine alignment of the optical setup can improve the detection sensitivity to minimize the cell amounts; however, the complexity of the instrumental was introduced. Recently, microelectrodes modified with cholesterol oxidase were developed to measure membrane cholesterol in single cells.20,21 In this strategy, the microelectrode was contacted with the membrane of single cells resulting in the enzymatic oxidation of cholesterol at the cell surface. Hydrogen peroxide produced was electrochemically oxidized on the electrode to give the current for the measurement of membrane cholesterol. This method was successfully applied to study single cells at different disease states;22 however, the low analysis throughput (5−10 cells/day) and the special electrochemical setup, including the cell manipulator and the electrochemical noise filter, were required limiting the application in biochemistry. Thus, developing a simple assay with low-cost setup to detect membrane active cholesterol in single cells was necessary for the biological study. In our group, lumiol electrochemiluminescence was utilized to develop the assay for active membrane cholesterol in single cells. Electrochemiluminescence was the emission of the light by the relaxes of an electronically excited product to the ground state in a chemical reaction, which was preceded by an electrochemical reaction.23−25 The luminol electrochemiluminescence was initiated by applying a certain positive voltage to the working electrode in the presence of hydrogen peroxide.26 The physical location of the light producing was near the electrode surface. The electrochemiluminescence intensity was directly proportional to the amount of luminol or hydrogen peroxide. Compared with the fluorescent assay, luminol electrochemiluminescence did not need any excitation laser light. The absence of excitation light avoided the alignment of emission pathway and improved the collection efficiency of emission light. Also, luminol electrochemiluminescence had high sensitivity, reproducibility, and relative easiness to be automatically controlled for the determination of hydrogen peroxide. For the analysis of membrane active cholesterol in adherent cells, cholesterol oxidase reacted with active membrane cholesterol in 1 min.16,17 The relatively fast generation rate of hydrogen peroxide from the cellular membrane and the relatively slow diffusion rate of hydrogen peroxide to the aqueous solution should generate a peak concentration of hydrogen peroxide near the cells on the indium tin oxide (ITO) electrode surface. Compared with luminol chemiluminescence that consumed hydrogen peroxide immediately, luminol electrochemiluminescence could give a higher signal-to-noise (S/N) ratio at the peak concentration of peroxide and was preferred for the cholesterol analysis of single adherent cells. In this paper, the details of luminol electrochemiluminescence for the single cell analysis were reported. The membrane active cholesterol at single cells was regulated by physical stimulation or biological treatment. The luminance change correlated with the alternation of the active membrane cholesterol in the cells validated our assay. The single cell analysis was conducted by the exposure of a single cell to a photomultiplier tube (PMT) through a pinhole, as schemed in Figure 1. A large deviation of active membrane cholesterol in single cells was observed, which might give new information for the study of intracellular cholesterol trafficking.

Figure 1. Setup of luminol electrochemiluminescence for the analysis of active cholesterol at the plasma membrane in single cell analysis.

■

EXPERIMENTAL SECTION Chemical. Fluo-3 a.m. was purchased from the Beyotime Institute of Biotechnology (Nantong, China). Raw264.7 cells were obtained from the Institute of Biochemistry and Cell Biology, Shanghai Institute for Biological Sciences of Chinese Academy of Science (Shanghai, China). All chemicals were from Sigma−Aldrich, unless indicated otherwise. Ultrapure water with a resistivity of 18.2 MΩ/cm was used throughout. Buffer solutions were sterilized. Cell Culture. Raw264.7 cells were seeded in DMEM/high glucose medium supplemented with 10% fetal bovine serum (FBS) and 1% antibiotics (penicillin/streptomycin). Cultures were maintained at 37 °C under a humidified atmosphere containing 5% CO2. Active Membrane Cholesterol. Approximately 20 000 cells were cultured in medium on the ITO surface (1 cm in diameter) overnight. To activate the membrane cholesterol, the cells were cultured in either 0.5 mM phosphate-buffered saline (PBS, pH 7.4) with 310 mM sucrose at 37 °C for 1 h or the medium with 2 μg/mL sandoz 58035 at 37 °C for 24 h. The cells inhibited by sandoz 58035 were not pretreated with low ion strength medium prior to the inhibition. Before the luminance detection, the cells were washed and cultured in 100 mM PBS (pH 7.4). Intracellular Calcium Measurement. The cells on the ITO electrodes were cultured in the extracellular buffer (ECB: 135 mM NaCl, 5 mM KCl, 1 mM MgCl2, 1 mM CaCl2, 10 mM HEPES, 10 mM glucose, pH 7.4) containing 2 μM Fluo-3 a.m. and 0.02% pluronic F-127 at room temperature for 30 min. Then, the cells were washed and recovered in ECB at 37 °C for 30 min. Fluorescence imaging of the cells was recorded on an inverted fluorescence microscope (Nikon Ti, Japan) with cooled CCD camera (Nikon, digital sight DS-Qi1MC, Japan). The fluorescent intensities of the cells were analyzed using Image J software. Luminance Detection. The ITO electrode cultured with the cells was used as the working electrode for the luminance detection. The transparency of the ITO electrode was over 85% at 550 nm. Ag/AgCl and Pt electrodes were connected as a reference and counter electrode, respectively. The solution volume in the cell chamber was 20 μL. Initially, the potential from −1.0 to 1.0 V with a scan rate of 1 V/s was applied on the ITO electrode to record the background luminance. Then, 1 U/mL cholesterol oxidase was introduced into the buffer for a certain time, and the luminance was measured. The ratio of luminances read at 1.0 V after and before the introduction of cholesterol oxidase was determined as the signal, which was 3913

dx.doi.org/10.1021/ac303304r | Anal. Chem. 2013, 85, 3912−3917

Analytical Chemistry

Article

termed as “luminance ratio” in this paper. The PMT voltage for the analysis of cell population was set at 600 V. For single cell analysis, a pinhole with 100 μm diameter was prepared using a 100 μm microdrill and placed between the ITO electrode and the PMT detection window. The cell density was adjusted to ∼4000 cells/mL of medium. A 0.5 mL amount of cell medium was loaded into the cell chamber, and ∼2000 cells were cultured on the ITO electrode. The distances between the adjacent cells ranged from 80 to 1000 μm. The pinhole was positioned so that a single cell was exposed to the PMT. The PMT voltage for single cell analysis was set at 1000 V. Since the noise in electrochemiluminescence increased with high PMT voltage, the luminances recorded in the voltage range 0.9−1.0 V were averaged as the final luminance to decrease the measurement error. In our work, luminescence meant the emission of light by the luminol reaction, and luminance meant the intensity of luminescence.

linear regression analysis equation was as follows: y (luminance) = 717 + 6.4 · c (conc) with a correlation coefficient of 0.997, as exhibited in Figure S2B (Supporting Information). The sensitivity was measured to be 50 nM (S/N = 6). Since the sensitivity depended on lots of parameters, such as the buffer pH, composition, PMT sensitivity, and experimental setup, a range of the sensitivity from picomolar to micromolar had been reported in the previous studies.26,28 The intercept in Figure S2B (Supporting Information) was the background luminance, which was attributed to the electrooxidation of luminol in the absence of hydrogen peroxide. Five ITO electrodes were tested, and the background luminances were varied with the electrodes, which might be attributed to the different ITO microstructure on the electrodes. To improve the batch-to-batch reproducibility, the ratio of luminance after and before the introduction of hydrogen peroxide was calculated. By using the luminol ratio, the relative standard deviation was improved to be only 5.0% for five electrodes in response of 500 nM hydrogen peroxide. Effects of Luminol and Voltage on Intracellular Calcium Concentration. For the application of luminol electrochemiluminescence on the analysis of active membrane cholesterol, luminol and the voltage applied on the cells should give the minor interruption on intracellular cholesterol trafficking. The initial experiment was performed to investigate the cell viability using THE trypan blue assay. Two groups of cells in the absence and presence of 100 μM luminol were stained with 0.06% trypan blue for 5 min. Compared with the control group, no blue cells were observed after the addition of luminol exhibiting good cell viability in the presence of 100 μM luminol. Since intracellular cholesterol was associated with calcium homeostasis in the cells,29−31 intracellular calcium was also imaged to investigate the effects of luminol and voltage on intracellular cholesterol. The first experiment included the culture of the cells in 100 μM luminol for 5 min, which was the typical time period for cell analysis. The imagings of intracellular calcium before and after the introduction of luminol were performed, and the fluorescent intensities were shown in Figure S3A (Supporting Information). No significant change in fluorescent intensity observed indicated that the intracellular calcium was not interrupted in the presence of low concentration of luminol. The second experiment was the determination of intracellular calcium change after applying the voltages of −0.5 to 0.5 V and −1.0 and 1.0 V on the electrodes. Figure S3B (Supporting Information) showed that the fluorescent intensities on the cells were similar in the absence and presence of voltage. These entire results suggested that the introduction of luminol and voltage did not change the intracellular calcium and thus might not affect the cholesterol trafficking. Detection of Active Membrane Cholesterol in the Cell Population. To create the active membrane cholesterol, the typical protocol included a pretreatment of cells in low ionic strength buffer. This pretreatment was believed to rearrange the location of cholesterol at the cellular membrane, which enhanced the membrane cholesterol activity.16 In our experiment, Raw264.7 cells were pretreated in low ionic strength medium for 1 h at 37 °C. Then, the medium was changed into 100 mM PBS (pH 7.4) with 100 μM luminol to record the background luminance. After cholesterol oxidase was added into the solution for 1 min, a significant increase in luminance was observed, as shown in Figure 2A. This luminance increase

■

RESULTS AND DISCUSSION Optimization of Luminol Electrochemiluminescence for Analysis of Aqueous Hydrogen Peroxide. Luminol electrochemiluminescence has been applied for the analysis of aqueous hydrogen peroxide on various electrode materials. In our system, the luminance from the ITO electrode was observed to increase with the voltage in PBS (pH 7.4). When the voltage was above 1.1 V, ITO turned black suggesting that the maximum voltage for the ITO electrode was less than 1.1 V. Therefore, the potential mode was set from −1.0 to 1.0 V, and the maximum luminance was read at the voltage of 1.0 V. The stability of ITO was characterized using five consecutive detections of 2.5 μM H2O2 in the presence of 100 μM luminol on the same ITO electrode. The relative standard deviation of luminance collected was 3.0%. No increasing trend in background luminescence and the decreasing trend of luminescence in the presence of H2O2 were observed in the whole experiment detection. These results supported that ITO was stable in our experiment. The effect of scan rate on the detection was investigated by varying the scan rate from 0.001 to 10 V/s. As shown in Figure S1A (Supporting Information), a decrease in the luminance of 100 μM luminol only was observed when the scan rate increased, which should be caused by the less luminol oxidized at fast scan rate. After the addition of 2.5 μM H2O2, the luminance increased when the scan rate increased from 0.001 to 0.1 V/s and started to decrease when the scan rate was over 0.1 V/s. The existence of peak luminance suggested that both the electron transfer rate and diffusion rate of hydrogen peroxide controlled the luminance. The ratio of luminance after and before the addition of hydrogen peroxide has a maximum value at a scan rate of 0.1 V/s, as shown in Figure S1B (Supporting Information). Since the luminance ratio collected at 1 V/s did not drop too much, we chosen 1 V/s as the scan rate in our work to increase the analysis rate. To find the lowest concentration of luminol with minimal cytotoxicity, different concentrations of luminol were reacted with 2.5 μM hydrogen peroxide. Figure S2A (Supporting Information) showed that the luminance reached the steady state when the luminol concentration was above 100 μM. Thus, 100 μM luminol was determined as the lowest concentration for the cell experiments. A similar result about the steady-state luminance after a certain concentration of luminol was reported previously.27 Under this condition, a linear response for hydrogen peroxide was obtained from 50 to 600 nM. The 3914

dx.doi.org/10.1021/ac303304r | Anal. Chem. 2013, 85, 3912−3917

Analytical Chemistry

Article

the analysis of aqueous hydrogen peroxide, which might be mainly caused by the error of cell numbers on the electrode. Although the cell counting was performed before the culture of cells on the electrode, an error on the real cell number on the electrode was unavoidable. Also, the difference in the proliferation rate of cells on each electrode might introduce additional error. The membrane cholesterol amount that was oxidized was estimated based on the luminescence observed after 5 min of holding time. At this point, hydrogen peroxide generated diffused into the bulk solution so that the bulk concentration was equal to the concentration near the surface for luminescence detection. The hydrogen peroxide concentration was determined by luminol electrochemiluminescence as 290 ± 53 nM, which suggested that 5.8 pmol of cholesterol reacted from 20 000 cells. Thus, the number of cholesterol molecules reacted from a single cell were determined to be 0.29 fmol, since approximately 0.33 fmol of cholesterol was estimated at the outlet leaflet of the plasma membrane in a single cell. Thus, ∼87.9% of the cholesterol at the outlet leaflet was reacted with cholesterol oxidase, which was consistent with the literature report.17 To validate our assay, 20 000 cells in the cell chamber were treated with the cholesterol oxidase for 5 min, and hydrogen peroxide generated was determined using the amplex red fluorescent assay as the gold standard. The concentration of hydrogen peroxide was determined to be 270 ± 27 nM, which was close to our luminescence analysis. Although low ionic strength buffer activated membrane cholesterol, it was more important to investigate the active membrane cholesterol in the cells with abnormal intracellular cholesterol trafficking. Acyl-coA/cholesterol acyltransferase (ACAT) is an enzyme that coverts free cholesterol in the cells into cholesterol ester. The inhibition of ACAT in the cells increased the intracellular cholesterol and membrane cholesterol,20,32 which created active membrane cholesterol. Figure 3

Figure 2. (A) Electrochemiluminescence traces of raw 264.7 cells pretreated with low ion strength buffer at 37 °C for 1 h before and after the addition of 1 U/mL cholesterol oxidase for 1 min; (B) luminances after the introduction of cholesterol oxidase for different times. The PMT voltage was set at 600 V.

exhibited that cholesterol oxidase can react with active membrane cholesterol to give hydrogen peroxide, which induced the luminol electrochemiluminescence. Different waiting times were applied after the introduction of cholesterol oxidase, and the luminance ratios were recorded in Figure 2B. As expected, a maximum in the luminance ratio was observed at 1 min. Since active membrane cholesterol was reacted with the aqueous cholesterol oxidase in 1 min,16,17 the maximum luminance observed at 1 min suggested a peak concentration of hydrogen peroxide near the cells on the electrode surface immediately after the reaction. Afterward, the concentration of hydrogen peroxide on the electrode started to decrease because only the diffusion of hydrogen peroxide into the bulk solution occurred. To achieve the highest signal-tonoise ratio, 1 min waiting time was fixed for the following experiments. Two control experiments were performed to validate our measurement. The first control experiment recorded the luminances before and after the introduction of cholesterol oxidase in the absence of cells. In the second experiment, the cells were cultured on the electrode, and PBS was added into the system in place of cholesterol oxidase. The luminance ratios close to 1 in both of the control experiments indicated that our luminance increase on the cell assay was attributed to the reaction of active membrane cholesterol with cholesterol oxidase. Eight independent cell experiments on different ITO electrodes were performed, and the relative standard deviation on the luminance ratio for cell assay was determined to be 17.4%. The relative standard deviation was larger than that on

Figure 3. Electrochemiluminescence traces of ACAT inhibited cells and ACAT/HMGCoA inhibited cells before and after the introduction of 1 U/mL cholesterol oxdiase. The PMT voltage was set at 600 V.

showed that significant increase in the luminance after the introduction of cholesterol oxidase on ACAT inhibited cells, which exhibited the existence of active membrane cholesterol in ACAT inhibited cells. The ACAT inhibited cells were further treated with 100 μM mevastatin for 24 h, which inhibited hydroxy methylglutaryl coenzyme A reductase (HMGCoA) to synthesize cholesterol in the cells.33,34 No pretreatment with low ion strength medium prior to the inhibition was needed for mevastatin inhibited cells. The decrease in the intracellular cholesterol biosynthesis was reported to decrease the cholesterol content in the membrane cholesterol.22,35 As shown in Figure 3, a decrease in the luminance on ACAT/ 3915

dx.doi.org/10.1021/ac303304r | Anal. Chem. 2013, 85, 3912−3917

Analytical Chemistry

Article

should not be caused by the error of cell number because only one cell was analyzed. Thus, the phenomenon revealed the heterogeneity of active membrane cholesterol in single cells precultured in low ion strength buffer clearly. The heterogeneity suggested that the arrangement of cholesterol in the plasma membrane to form active cholesterol in single cells might be different. It was noted that the average luminance ratio from single cell analysis cannot be compared with the result from the cell population. The reason was that the cell occupied more area in the single cell analysis than that in the cell population, which increased the background luminance and decreased the luminance ratio. Compared with the relative standard deviation (10−20%) obtained using the electrochemical method,20,22 a larger relative standard deviation was observed in our work. The electrochemical method detected the cholesterol efflux from a single cell, which was related with the chemical activity of membrane cholesterol. The difference in the relative standard deviations might suggest the large cell heterogeneity in the amount of active membrane cholesterol and the small heterogeneity in the activity of membrane cholesterol. To investigate the effect of cross-talk between adjacent cells on the single cell analysis, the distances between the target cell and the nearest cell were recorded before each analysis. As shown in Figure S4 (Supporting Information), no obvious correlation of the distance on the detection signals was observed on five single cell analyses, which suggested that the adjacent cell might not give any contribution on the signal of the target cell. Twelve ACAT inhibited cells were analyzed, and the luminance ratios were shown in Figure 5A. The deviation on

HMGCoA inhibited cells was observed. The correlation of the active membrane cholesterol and luminance supported that our assay can monitor active membrane cholesterol in the cells. Analysis of Membrane Active Cholesterol in Single Cells. The aim of our study was to develop a simple assay for single cell analysis, which was challenging for the existing technologies. Since the luminance ratio was independent of the electrode area, the luminance ratio should be the same when a small region with only one cell was measured. Thus, if the luminance on the small region was measurable, single cell analysis can be performed. Here, a 100 μm pinhole was placed between the electrode and the PMT to create a small region exposed to the PMT. By adjusting the cell density on the ITO electrode, one cell was placed above the pinhole for the analysis. Following the previous protocol, the cells were cultured in the low ionic strength solution for 1 h at 37 °C to activate the membrane cholesterol. Figure 4A showed the

Figure 4. (A) Electrochemiluminescence traces of a single cell pretreated with low ion strength buffer before and after the introduction of 1 U/mL cholesterol oxidase. The PMT voltage was set at 1000 V. The peaks before 0.5 V were the electrical noises of the PMT at high voltage. (B) Luminance ratios of 12 single cells. The luminance ratios of the cells were shown in the order of analysis.

background luminance and the luminance after the introduction of cholesterol oxidase for 1 min. An increase in the luminance indicated that membrane active cholesterol in a single cell was detectable. The value of S/N to measure a single cell with 1.75 × 108 molecules of cholesterol was averaged to be 6 suggesting that the minimum number of cholesterol molecules to generate the measurable signal was 8.7 × 107 (S/N = 3). Twelve single cells were analyzed individually, and the luminance ratios were shown in Figure 4B. The standard deviation was calculated to be 45.1%, which was much larger than the result on the cell population. This increase in deviation

Figure 5. Luminance ratios of 12 (A) ACAT inhibited single cells and (B) ACAT/HMGCoA inhibited cells, respectively. The luminance ratios of the cells were shown in the order of analysis. 3916

dx.doi.org/10.1021/ac303304r | Anal. Chem. 2013, 85, 3912−3917

Analytical Chemistry

Article

(2) Chang, T. Y.; Chang, C. C. Y.; Ohgami, N.; Yamauchi, Y. Annual Review of Cell and Developmental Biology; Annual Reviews: Palo Alto, CA, 2006; Vol. 22, pp 129−157. (3) Liscum, L.; Munn, N. J. Biochim. Biophys. Acta, Mol. Cell Biol. Lipids 1999, 1438, 19−37. (4) Oram, J. F.; Heinecke, J. W. Physiol. Rev. 2005, 85, 1343−1372. (5) Anderson, R. G. W. Annu. Rev. Biochem. 1998, 67, 199−225. (6) Ikonen, E. Nat. Rev. Mol. Cell Biol. 2008, 9, 125−138. (7) Huang, J. Y.; Feigenson, G. W. Biophys. J. 1999, 76, 2142−2157. (8) McConnell, H. M.; Radhakrishnan, A. Biochim. Biophys. Acta, Biomembr. 2003, 1610, 159−173. (9) Lange, Y.; Ye, J.; Steck, T. L. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 11664−11667. (10) Kilsdonk, E. P. C.; Yancey, P. G.; Stoudt, G. W.; Bangerter, F. W.; Johnson, W. J.; Phillips, M. C.; Rothblat, G. H. J. Biol. Chem. 1995, 270, 17250−17256. (11) Radhakrishnan, A.; McConnell, H. M. Biochemistry 2000, 39, 8119−8124. (12) Lange, Y.; Echevarria, F.; Steck, T. L. J. Biol. Chem. 1991, 266, 21439−21443. (13) Adorni, M. P.; Zimetti, F.; Billheimer, J. T.; Wang, N.; Rader, D. J.; Phillips, M. C.; Rothblat, G. H. J. Lipid Res. 2007, 48, 2453−2462. (14) Ali, M. R.; Cheng, K. H.; Huang, J. Biochemistry 2006, 45, 12629−12638. (15) Lange, Y.; Cutler, H. B.; Steck, T. L. J. Biol. Chem. 1980, 255, 9331−9337. (16) Lange, Y.; Matthies, H.; Steck, T. L. Biochim. Biophys. Acta 1984, 769, 551−562. (17) Lange, Y. J. Lipid Res. 1992, 33, 315−319. (18) Ali, M. R.; Cheng, K. H.; Huang, J. Y. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 5372−5377. (19) Weiner, N. D.; Felmeist, A. J. Lipid Res. 1970, 11, 220−222. (20) Jiang, D. C.; Devadoss, A.; Palencsar, M. S.; Fang, D. J.; White, N. M.; Kelley, T. J.; Smith, J. D.; Burgess, J. D. J. Am. Chem. Soc. 2007, 129, 11352. (21) Fang, D.; Jiang, D.; Lu, H.; Chiel, H. J.; Kelley, T. J.; Burgess, J. D. J. Am. Chem. Soc. 2009, 131, 12038. (22) Fang, D. J.; West, R. H.; Manson, M. E.; Ruddy, J.; Jiang, D. C.; Previs, S. F.; Sonawane, N. D.; Burgess, J. D.; Kelley, T. J. Respir. Res. 2010, 11, 12. (23) Marquette, C. A.; Blum, L. J. Anal. Chim. Acta 1999, 381, 1−10. (24) Marquette, C. A.; Ravaud, S.; Blum, L. J. Anal. Lett. 2000, 33, 1779−1796. (25) Forster, R. J.; Bertoncello, P.; Keyes, T. E. Annual Review of Analytical Chemistry; Annual Reviews: Palo Alto, CA, 2009; Vol. 2, pp 359−385. (26) Jirka, G. P.; Martin, A. F.; Nieman, T. A. Anal. Chim. Acta 1993, 284, 345−349. (27) Zhu, L.; Li, Y. X.; Zhu, G. Y. Sens. Actuators, B 2002, 86, 209− 214. (28) Wilson, R.; Akhavan-Tafti, H.; DeSilva, R.; Schaap, A. P. Electroanalysis 2001, 13, 1083−1092. (29) Feng, B.; Yao, P. M.; Li, Y. K.; Devlin, C. M.; Zhang, D. J.; Harding, H. P.; Sweeney, M.; Rong, J. X.; Kuriakose, G.; Fisher, E. A.; Marks, A. R.; Ron, D.; Tabas, I. Nat. Cell Biol. 2003, 5, 781−792. (30) Ahtiainen, L.; Kolikova, J.; Mutka, A. L.; Luiro, K.; Gentile, M.; Ikonen, E.; Khiroug, L.; Jalanko, A.; Kopra, O. Neurobiol. Dis. 2007, 28, 52−64. (31) Takahashi, Y.; Smith, J. D. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 11358−11363. (32) Warner, G. J.; Stoudt, G.; Bamberger, M.; Johnson, W. J.; Rothblat, G. H. J. Biol. Chem. 1995, 270, 5772−5778. (33) Roitelman, J.; Olender, E. H.; Barnun, S.; Dunn, W. A.; Simoni, R. D. J. Cell Biol. 1992, 117, 959−973. (34) Farmer, J. A. Prog. Cardiovasc. Dis. 1998, 41, 71−94. (35) Mason, R. P.; Walter, M. F.; Jacob, R. F. Circulation 2004, 109, 34−41.

active membrane cholesterol was calculated to be 40.5%. Most of the cells gave a luminance ratio between 2.8 and 4; however, four ratios were below 2.1, and one ratio reached 6. It was not yet known whether this distribution of luminance ratio was attributed to the different activity of ACAT in single cells after the inhibition or any other mechanisms. Nevertheless, this new information obtained from single cell analysis might give a new insight on the ACAT associated study in cells, which was significant to investigate intracellular cholesterol trafficking. The ACAT inhibited cells were further treated with mevastatin, and the results were shown in Figure 5B. The average luminance ratio on 12 cells was smaller than that on the ACAT inhibited cells, which was consistent with the result on the cell population. Interestingly, the relative standard deviation decreased to 18.5%. Due to the limited data, it was difficult to tell whether this decrease in the deviation has any biological significance. However, the treatment of ACAT inhibited cells by mevastain did not further increase the heterogeneity of active membrane cholesterol indicating the different inhibition efficiency for mevastatin and sandoz 58035. The effects of other statin family members on the active membrane cholesterol in single cells are being investigated to evaluate the drug efficiency.

■

CONCLUSIONS In this paper, a simple luminol electroluminescence was developed to assay active cholesterol at the plasma membrane in single mammalian cells. The luminance change was consistent with the alternation of active membrane cholesterol in the cells, which supported that our assay can monitor membrane active cholesterol. A large deviation of membrane active cholesterol in single cells was observed, which might provide new information on the cholesterol activation at the plasma membrane and the intracellular cholesterol trafficking. Further automation of this simple and low-cost assay by the arrangement of single cells on a microelectrode array can achieve fast single cell cholesterol analysis. The large amounts of data obtained can help the understanding of the cell heterogeneity and might be important for the study of intracellular cholesterol trafficking.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional information as noted in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: 086-25-83594846 (D.J.); 086-25-86868477 (D.F.). Fax: 086-25-83594846 (D.J.); 086-25-86868477 (D.F.). E-mail: [email protected] (D.J.); [email protected] (D.F.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (nos. 21135003, 21105045, and 21105049) and the 973 Program (2013 CB933800).

■

REFERENCES

(1) Maxfield, F. R.; Wustner, D. J. Clin. Invest. 2002, 110, 891−898. 3917

dx.doi.org/10.1021/ac303304r | Anal. Chem. 2013, 85, 3912−3917