Letter pubs.acs.org/ac
Analysis of Intracellular Glucose at Single Cells Using Electrochemiluminescence Imaging Jingjing Xu, Peiyuan Huang, Yu Qin, Dechen Jiang,* and Hong-yuan Chen State Key Laboratory of Analytical Chemistry for Life Science, School of Chemistry and Chemical Engineering, Nanjing University, Nanjing, Jiangsu 210093, China S Supporting Information *
ABSTRACT: Here, luminol electrochemiluminescence was first applied to analyze intracellular molecules, such as glucose, at single cells. The individual cells were retained in cell-sized microwells on a gold coated indium tin oxide (ITO) slide, which were treated with luminol, triton X-100, and glucose oxidase simultaneously. The broken cellular membrane in the presence of triton X-100 released intracellular glucose into the microwell and reacted with glucose oxidase to generate hydrogen peroxide, which induced luminol luminescence under positive potential. To achieve fast analysis, the luminescences from 64 individual cells on one ITO slide were imaged in 60 s using a charge-coupled device (CCD). More luminescence was observed at all the microwells after the introduction of triton X-100 and glucose oxidase suggested that intracellular glucose was detected at single cells. The starvation of cells to decrease intracellular glucose produced less luminescence, which confirmed that our luminescence intensity was correlated with the concentration of intracellular glucose. Large deviations in glucose concentration at observed single cells revealed high cellular heterogeneity in intracellular glucose for the first time. This developed electrochemiluminescence assay will be potentially applied for fast analysis of more intracellular molecules in single cells to elucidate cellular heterogeneity.
E
electron multiplying charge-coupled device (EM-CCD). Although our work has proved luminol electrochemiluminescence as an effective method for single cell analysis, the application is restricted in the analysis of surface molecules. The further development of luminol electrochemiluminescence for the analysis of intracellular molecules will be significant because the fluctuation of the intracellular molecule amount associated with the activity is important for the biological study. In this Letter, the analysis of cytosol in single cells was attempted using luminol electrochemiluminescence for the first time, and intracellular glucose was chosen as the model molecule. As demonstrated in Figure 1, the individual cells were retained in cell-sized microwells on an indium tin oxide (ITO) slide. The introduction of the detergent, such as triton X-100, could break the cellular membrane so that intracellular glucose was released to react with aqueous glucose oxidase. Hydrogen peroxide generated induced luminol luminescence under positive potential to realize the analysis of intracellular glucose. The design of the microwell structure around the cell could not only retain the single cell but also slow down the diffusion of hydrogen peroxide away from the ITO region in the microwell. As a result, the luminescence could be continuously collected in a certain time period using a luminescence imaging system for fast single cell analysis.
lectrochemiluminescence (ECL) is an optical radiation process induced from energy relaxation of excited species after certain electrochemical reactions.1−3 As a marriage between electrochemical and spectroscopic methods, ECL has many distinctive advantages over other detection systems, such as chemiluminescence and fluorescence. Specifically, compared with chemiluminescence, ECL is more controllable and selective for the separation between applied potential and light emission.4 Also, in comparison with fluorescence, the absence of excitation light in ECL minimizes the interruption of background light on the measurement so that higher sensitivity is achievable.5,6 Therefore, as a powerful analytical tool, this technique has attracted considerable attention in bioanalysis for sensing trace amounts of target molecules.7−13 The luminol−hydrogen peroxide system is well-known to produce strong luminescence under positive potential.14,15 During this process, luminol and hydrogen peroxide go through electro-oxidation to diazaquinone and oxygen containing species and further undergo a high-energy electron-transfer reaction to produce 3-aminophthalate for light emission.6,15 Considering corresponding oxidase reacts with the target molecule to produce hydrogen peroxide, our group utilized luminol electrochemiluminescence to analyze active cholesterol in plasma membrane at single cells through the introduction of cholesterol oxidase.16,17 To increase the analysis rate, parallel determination of single cells using electrochemiluminescence imaging was realized.18 In this strategy, the luminescence associated with hydrogen peroxide or active cholesterol at the surface of multiple cells was recorded in one image using an © 2016 American Chemical Society
Received: March 18, 2016 Accepted: April 20, 2016 Published: April 20, 2016 4609
DOI: 10.1021/acs.analchem.6b01073 Anal. Chem. 2016, 88, 4609−4612
Letter
Analytical Chemistry
electrochemical reactions associated with luminol electrochemiluminescence and generate more luminescence,19 our ITO regions in the microwells were electrodeposited with Au NPs to improve the detection limit.20 As shown in Figure S1, scanning electron microscopy (SEM) imaging showed that Au NPs were electrodeposited at ITO regions in the microwells. With the optimized deposition time of 400 s, Au NP coated ITO regions generated a 10-fold increase in the luminescence compared to uncoated ITO regions, as shown in Figure S2. The enhancement in the luminescence permits electrochemiluminescence imaging of hydrogen peroxide at ITO regions under the microwells with good detection limit. Figure 2A showed a bright-field image of a Au NP coated ITO slide with 64 microwells (30 μm in diameter and height) in 200 μM L012 (luminol analog with higher luminescence) using a 10× objective. When the concentration of hydrogen peroxide was elevated from 5 to 500 μM on the slide, the luminescence images were collected. The luminescence difference in the absence and presence of hydrogen peroxide were calculated and exhibited in Figure 2B−H. More luminescence observed at the microwells only exhibited the visualization of hydrogen peroxide at ITO regions. The luminescence increase was correlated with the concentration of hydrogen peroxide, as presented in Figure 2H, which suggested that our platform could offer the quantitative measurement of single cells. The detection limit of hydrogen peroxide was 5 μM, which was better than our previous limit (10 μM). The relative standard deviation of luminescence observed at these 64 microwells was 17.6%, which was close to our previous deviation of 15.6% reported on eight microelectrode arrays. This similarity on the
Figure 1. Schematic electrochemiluminescence imaging setup and the detection strategy for fast analysis of intracellular glucose at single cells. The green balls presented the individual cells in the microwells.
For the validation of the intracellular glucose analysis, the imaging of aqueous hydrogen peroxide in the microwells is critical. Although our previous work exhibited that 10 μM hydrogen peroxide was visible on the ITO slide using electrochemiluminescence imaging, ITO slides through multiple microfabrication processes for the preparation of microwells were observed to offer weak luminescence intensity resulting in a higher detection limit. The loss of the luminescence efficiency might be attributed to the photoresist residue remaining on the ITO regions in the microwells after the microfabrication. Since gold nanoparticles (Au NPs) were reported to accelerate
Figure 2. Electrochemiluminescence imaging of aqueous hydrogen peroxide at ITO/microwell regions in 10 mM PBS with 200 μM L012. (A) bright-field image; (B−H) the luminescence difference images of 5, 10, 20, 50, 100, 200, 500 μM hydrogen peroxide; (I) the relationship of luminescence difference with the concentrations of hydrogen peroxide. The error bar presented the standard deviation measured from 64 microwells on the ITO slide. 4610
DOI: 10.1021/acs.analchem.6b01073 Anal. Chem. 2016, 88, 4609−4612
Letter
Analytical Chemistry
glucose oxidase to exclude the possible contribution of intracellular reactive oxygen species on the luminescence. As shown in Figure S4, no obvious luminescence enhancement was observed at the microwells after the introduction of triton X-100 only, which confirmed that the luminescence increase observed at the microwells was attributed to the reaction of intracellular glucose and glucose oxidase. These results suggested the visualization of intracellular glucose at single cells using our electrochemiluminescence imaging. For the calibration of the luminescence from these cells, the ITO slide with the cells was washed to remove any hydrogen peroxide remaining in the microwells. The fresh buffer containing 50 μM hydrogen peroxide, 200 μM L012, and 0.1% titron X-100 was introduced at the ITO slide for the reimaging of the luminescence. The luminescence collected from 128 cells and 50 μM hydrogen peroxide was rationed as the signal in Figure 4A. The average luminescence ratio was
deviation supports the relative uniformity of Au NP coated ITO regions in the microwells. To achieve fast analysis of intracellular glucose at single cells, electrochemiluminescence imaging was applied for the recording of luminescence from 64 cells/microwells simultaneously. Figure 3A showed a bright field image of 64 microwells loaded
Figure 3. (A) The bright-field image of single cells located in the microwells; (B) the luminescence image collected in the first 60 s immediately after the introduction of glucose oxidase/triton X-100; (C) the luminescence image collected in the second 60 s; (D) the luminescence difference image between those two images.
with individual cells. The typical efficiency for the loading of a single cell in the microwells was over 80% under the microscopic observation. The microwells without the cell or with more than one cell were excluded in the following luminescence analysis. After the retaining of single cells in the microwells, the cellular viability was investigated by staining the cells with 10 μM fluorescein esters for 5 min and then recovering them in 10 mM PBS for 30 min at 37 °C. As shown in Figure S3, strong fluorescence emitted from the cells demonstrated that the individual cells in the microwells converted fluorescein ester into fluorescein actively. The different fluorescence intensity collected at the cells was attributed to cellular heterogeneity in the endocytosis of fluorescein ester and intracellular hydrolase activity in the conversion of fluorescein ester into fluorescein, which had been reported before.21 As a result, the maintenance of cellular function through the loading process into the microwells was confirmed. Then, glucose oxidase/titron X-100 was introduced into the buffer. The video recording exhibited that the cells in the microwells started to break in 45−60 s. Therefore, two consecutive imagings with the exposure time of 60 s were performed, as shown in Figure 3B,C. In the first image, the intensity presented background luminescence from L012 only, and in the second image, the luminescence was contributed to L012 and hydrogen peroxide from the reaction between intracellular glucose released and aqueous glucose oxidase. The luminescence difference between these two images, as shown in Figure 3D, exhibited more luminescence at each microwell. The control experiment was performed at the cells in the absence of
Figure 4. Luminescence ratio of (A) 128 single normal cells and (B) 128 single cells after the starvation for 4 h. Each diamond represents one cell.
calculated to be 1.06 ± 0.46, indicating that the concentration of hydrogen peroxide in the microwells generated from cells was near 50 μM. Considering the volumes of one cell and microwell were ∼1 and 21 pL, the intracellular glucose was estimated to be ∼1 mM, which was close to the literature value.22 The relative standard deviation from 128 cells was 43.4%, which was higher than the error in the measurement of aqueous hydrogen peroxide (17.6%) exhibiting high cellular heterogeneity of intracellular glucose. To the best of our knowledge, this was the first report to show the large difference in intracellular glucose. Since glucose is important for nutrition and is a metabolite in the cells, this difference will facilitate the future understanding of cell heterogeneity. 4611
DOI: 10.1021/acs.analchem.6b01073 Anal. Chem. 2016, 88, 4609−4612
Analytical Chemistry
■
To further validate our assay of intracellular glucose, a negative control experiment was conducted on the starved cells, which were cultured in the medium without glucose for 4 h. Since extracellular glucose as the energy source was removed from the medium, intracellular glucose was exhausted in this period resulting in lower concentration inside the cells. The bright-field, two luminescence, and the luminescence difference images of the starved cells are shown in Figure S5. Weak luminescence was observed at the microwells confirming the measurement of intracellular glucose after the starvation. Similar to the assay of normal cells, the luminescences from 128 cells and 50 μM hydrogen peroxide were rationed and shown in Figure 4B. As compared to the result on normal cells, a smaller luminescence ratio of 0.36 ± 0.19 was calculated on the starved cells, which was attributed to glucose consumption during the starvation. According to the calibration curve in Figure 2I, the intracellular glucose was estimated to be 0.36 mM. This observation of glucose change in the cells at different states confirmed that our luminescence intensity was correlated with the concentration of intracellular glucose. The relative standard deviation was 52.8% revealing the existence of cellular heterogeneity after the starvation. In conclusion, the luminol electrochemiluminescence was first applied to analyze the intracellular glucose at single cells. Coupled with the electrochemiluminescence imaging platform, 64 individual cells were analyzed in 60 s that speeded up the single cell analysis rate greatly. A high heterogeneity in intracellular glucose was observed on normal and starved cells, which exhibited the cellular difference that might be associated with multiple cellular behaviors. The future biological application of this technology will focus on the study of glucose uptake at single cells to elucidate the connection between blood glucose and intracellular glucose in various cell types. Also, the first success in the electrochemiluminescence analysis of intracellular molecules at single cells will remove the previous restriction of this technique on the measurement of cellular surface only. Therefore, this strategy could be applied for the analysis of more intracellular metabolites to investigate more biological phenomena.
■
Letter
REFERENCES
(1) Knight, A. W.; Greenway, G. M. Analyst 1994, 119, 879−890. (2) Mitschke, U.; Bauerle, P. J. Mater. Chem. 2000, 10, 1471−1507. (3) Richter, M. M. Chem. Rev. 2004, 104, 3003−3036. (4) Hu, L.; Xu, G. Chem. Soc. Rev. 2010, 39, 3275−3304. (5) Fähnrich, K. A.; Pravda, M.; Guilbault, G. G. Talanta 2001, 54, 531−559. (6) Miao, W. Chem. Rev. 2008, 108, 2506−2553. (7) Dennany, L.; Forster, R. J.; White, B.; Smyth, M.; Rusling, J. F. J. Am. Chem. Soc. 2004, 126, 8835−8841. (8) Jiang, H.; Wang, X. Anal. Chem. 2012, 84, 6986−6993. (9) Jie, G.; Zhang, J.; Wang, D.; Cheng, C.; Chen, H.-Y.; Zhu, J.-J. Anal. Chem. 2008, 80, 4033−4039. (10) Nie, G.; Bai, Z.; Yu, W.; Chen, J. Biomacromolecules 2013, 14, 834−840. (11) Wu, M.-S.; Liu, Z.; Shi, H.-W.; Chen, H.-Y.; Xu, J.-J. Anal. Chem. 2015, 87, 530−537. (12) Zhang, L. H.; Guo, Z. H. Huaxue Xuebao 2013, 71, 644−648. (13) Zhuo, Y.; Liao, N.; Chai, Y.-Q.; Gui, G.-F.; Zhao, M.; Han, J.; Xiang, Y.; Yuan, R. Anal. Chem. 2014, 86, 1053−1060. (14) Knight, A. W. TrAC, Trends Anal. Chem. 1999, 18, 47−62. (15) Marquette, C. A.; Blum, L. J. Biosens. Bioelectron. 2004, 20, 197− 203. (16) Ma, G.; Zhou, J.; Tian, C.; Jiang, D.; Fang, D.; Chen, H. Anal. Chem. 2013, 85, 3912−3917. (17) Tian, C.; Zhou, J.; Wu, Z.-Q.; Fang, D.; Jiang, D. Anal. Chem. 2014, 86, 678−684. (18) Zhou, J.; Ma, G.; Chen, Y.; Fang, D.; Jiang, D.; Chen, H. Y. Anal. Chem. 2015, 87, 8138−8143. (19) Park, J. H.; Lim, Y. T.; Park, O. O.; Kim, J. K.; Yu, J.-W.; Kim, Y. C. Chem. Mater. 2004, 16, 688−692. (20) Wu, M. S.; Yuan, D. J.; Xu, J. J.; Chen, H. Y. Anal. Chem. 2013, 85, 11960−11965. (21) Jiang, D. C.; Sims, C. E.; Allbritton, N. L. Electrophoresis 2010, 31, 2558−2565. (22) Teusink, B.; Diderich, J. A.; Westerhoff, H. V.; van Dam, K.; Walsh, M. C. J. Bacteriol. 1998, 180, 556−562.
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.6b01073. Experimental section; SEM and luminescence analysis of Au NPs coated ITO regions in microwells; cellular function; more luminescence images. (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*Phone: 086-25-89684846. Fax: 086-25-89684846. E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by the 973 Program (2013 CB933800) and the National Natural Science Foundation of China (Nos. 21327902, 21135003, and 21575060). 4612
DOI: 10.1021/acs.analchem.6b01073 Anal. Chem. 2016, 88, 4609−4612
