Quantitation of Glucose-phosphate in Single Cells ... - ACS Publications

Apr 10, 2019 - among similar cells.4 Many single-cell metabolomics methods have been developed, and mass spectrometry (MS) is one of the widest applie...
0 downloads 0 Views 616KB Size
Subscriber access provided by UNIV OF LOUISIANA

Article

Quantitation of Glucose-phosphate in Single Cells by MicrowellBased Nanoliter Droplet Microextraction and Mass Spectrometry Jiaxin Feng, Xiaochao Zhang, Liang Huang, Huan Yao, Chengdui Yang, Xiaoxiao Ma, Sichun Zhang, and Xinrong Zhang Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b05226 • Publication Date (Web): 10 Apr 2019 Downloaded from http://pubs.acs.org on April 10, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 7 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Quantitation of Glucose-phosphate in Single Cells by MicrowellBased Nanoliter Droplet Microextraction and Mass Spectrometry Jiaxin Feng, a Xiaochao Zhang, a Liang Huang, b Huan Yao, a Chengdui Yang, a Xiaoxiao Ma, b Sichun Zhang, a,* Xinrong Zhang, a a Department b

of Chemistry, Tsinghua University, Beijing 100084, China. Department of Precision Instrument, Tsinghua University, Beijing 100084, China.

ABSTRACT: Changes of metabolite concentrations in single cells are significant for exploring the dynamic regulation of important biological processes, such as cell development and differentiation. Accurate quantitation of metabolites is essential for single cell analysis. In this work, we proposed a quantitative method for single-cell metabolites by combining microwell array with droplet microextraction-mass spectrometry. The microwell can confine both single cells and extraction solvent in defined space, avoiding the irregular spread of trace internal standard solution during microextraction, which was the key to improve the precision and accuracy of quantification in extremely small-volume single-cell samples. Glucose-phosphate as a crucial metabolite in glycolysis was detected and quantified in single cells at this work. The calibration curve of glucose-phosphate was obtained with a linear range from amol (10-18 mol) to fmol (10-15 mol), providing the foundation of metabolite quantitation of single cells. We applied this method to investigate the changes of metabolites including glucose-phosphate, 2-deoxy-d-glucose-phosphate and ribose-phosphate in single K562 cells stimulated by 2-deoxy-d-glucose. With the robust quantitative capabilities, the developed method holds great potentials for studying drugs’ mechanism of action and resistance at single cell level.

INTRODUCTION Quantitation of metabolites is essential for single-cell metabolomics since the metabolite concentrations provide direct reflection of the state of metabolism and the regulation of complex metabolic networks1. Metabolic flux regulation and cell biological function can be studied at single cell level via changes in concentrations of metabolites under conditions such as drug stimulation and gene (in)activation2,3. Measuring the concentration of metabolites in individual cells also aids in unravelling the heterogeneity among similar cells4. Many singlecell metabolomics methods have been developed and mass spectrometry (MS) is one of the widest applied technologies due to its capability of qualitative and quantitative detection, high sensitivity, and simultaneous analysis of multiple analytes5-11. Based on the accurate mass measurement and structural elucidation by tandem MS, hundreds of metabolites in single cells have been identified12-14. However, quantitation of metabolites in single cell is still in the early stage of development and many challenges exist due to the extremely small volume of a mammalian cell (1-10 pL) and low absolute amount of metabolites (amol - fmol)15. Currently, quantitation of metabolites in single cells by MS is mainly achieved by either external or internal standard methods. The single-cell capillary electrophoresis mass spectrometry (CEMS) method developed by Sweedler and coworkers performed quantitation of metabolites via external calibration and had been widely applied for various single cells12,16. Huang et al. also proposed quantitation of metabolites extracted form single neuro cells by external standards calibration curves8. Compared with external calibration, internal calibration can effectively reduce matrix interference and ion suppression, achieving more accurate metabolite quantitation17. The key to reliable single cell

quantitation by internal calibration is accurate introduction of internal standards into extremely small-volume single cell. Different strategies for introducing internal standard were developed depending on sampling methods of single cells. Strategies for introducing internal standard in single-cell matrixassisted laser desorption ionization (MALDI) analysis was mixing it with matrix and then deposited onto the sample surface18,19. For sampling methods that used nanopipette or capillary to extract cell cytoplasm, Laskin et al precisely controlled the volume of internal standard by electroosmotic extraction20. The nano-DESI method allowed quantitation of phospholipids in single cheek cells via adding internal standard into the continuously supplied solvent21. However, these methods of introducing internal standards still had shortcomings like uneven distribution of internal standard22, or dilution of metabolites. Hence, the accurate and simple introduction of small-volume internal standard solution is still a challenge for quantitative analysis of metabolites in single cells. Microfluidics is a wide used technology to precisely manipulate and control trace liquid. If we utilized microfluidics to simultaneously capture single cell and introduce internal standard, we could achieve high throughput single-cell dispersion and accurate quantification of metabolites. Many microfluidic devices have been widely developed and applied for single-cell capture and isolation, such as microwell array, microvalves and droplet method 23,24. For example, Hongxia Liu et al had combined microwells with MALDI for Study of Phospholipids in Single Cells25. Among these devices, microwell is a three-dimensional space that can confine the addition of internal standard solution and limit the irregular spread of the solvent, which would improve the repeatability of internal standard recovery and lead to the acquisition of reliable quantitative data26.

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Herein, we developed a quantitative method for single-cell metabolites by coupling microwell-array with droplet microextraction-MS. Individual cells were captured in the microwells and subsequently extracted by solvents casted into the microwells (Figure 1). Due to the confinement effect of the microwell, the extraction solvent established a stable liquid microjunction between the single cell and the nanopipette rather than spread irregularly on the surface, leading to improved repeatability and accuracy of solvent recovery. Besides, crosscontamination among cells were effectively eliminated by the physical boundary of microwells. Glucose-phosphate, as a crucial metabolite in glycolysis, was chosen as the model molecule and detected by the developed quantitative method. By adding the internal standard (13C6-glucose-phosphate) of known concentration in the small volume of extraction solvent, we successfully achieved quantification of glucose-phosphate in single K562 cells. We applied this method to quantitative study of metabolites change in single K562 cells stimulated by 2deoxy-d-glucose, revealing the cellular heterogeneity.

Page 2 of 7

developer to remove the uncrosslinked photoresist. After cleaning, the wafer was hard baked at 120°C for 30 minutes. The PDMS preploymer mixed with the curing agent at a ratio of 10:1 was poured onto the photoresist mold and placed in a conventional oven at 80°C for 2 hours. Peel off PDMS from the mold gently after it was cured.

Cell Culture and Capture K562 cells were cultured in RPMI-1640 medium with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin at 37 °C and 5% CO2 in a humidified incubator and passaged every 2 or 3 days. The concentration of K562 cells is determined by Hemocytometer with trypan blue staining to distinguish dead cells from living cells. For study of drug metabolism, the K562 cells were incubated with the anticancer compounds Adriamycin (20 μM) for 6 h. For study of drug absorption, the K562 cells were incubated with the 2-deoxyglucose (0, 1, 10 mM) for 4 hours respectively. To format the cell array on the microfluidic device, the chip was placed in oxygen plasma cleaner for 40s to improve the hydrophilicity. 10 μL of K562 cell suspension (1 × 107cells/mL) was added to the chip, and the cells settled into the microwells under the effect of gravity. The cells were loaded in the microwells after 10 minutes. The cell array was rinsed with PBS solution for three times to remove excess cells onto chip surface between microwells. Then the cell array was washed quickly with 0.9% ammonium formate solution to replace the PBS solution. Quench the cells with methanol by spraying it on the cell array using airbrush. The chip was then placed in a vacuum oven for 10 minutes to rupture and dry the cells.

Quantification by Droplet Microextraction in Microwell Figure 1. Workflow of droplet microextraction of single cell in microwells.

EXPERIMENTAL SECTION Reagents. Ribulose-5-phosphate, 6-Glucose-phosphate (Glu-6-P), nicotinamide adenine dinucleotide (NAD), nicotinamide adenine dinucleotide phosphate (NADP), reduced nicotinamide adenine dinucleotide (NADH), reduced nicotinamide adenine dinucleotide phosphate (NADPH) and HPLC-grade methanol were purchased from Sigma Aldrich. 13C6-Glucose-phosphate (13C6-Glu-P) was purchased from Cambridge Isotope Laboratories, Inc. 2-deoxy-glucose (2DG) was purchased from J&K Scientific Ltd. Adriamycin hydrochloride (ADM) was purchased from Solarbio Ltd. Ultrapure water was made from a Milli-Q water purification system (Millipore, resistance≥18MΩ cm−1). RPMI-1640 medium, Dulbecco’s phosphate buffered saline (DPBS), fetal bovine serum (FBS) and other materials used for cell culture were purchased from Corning (NY, USA).

Fabrication of Microfluidic Device The microfluidic devices were fabricated by soft photolithography technique. The mask was designed by AutoCAD software and fabricated by MEMS techniques. The SU-8 negative photoresist was spin-coated on a silicon wafer to form a 30 μm thick layer. After baked at 65 °C for 3 min and 95 °C for 7 min, the coated silicon wafer was exposed to UV light (365 nm) to create the pattern. The wafer was baked at 65 °C for 1 minute and 95 °C for 5 minutes and then washed by SU-8

Borosilicate glass capillaries with the inner diameter of 1.1mm and the outer diameter of 1.5 mm was chosen and pulled by the micropipette puller (P-1000, Sutter Instrument, Novato, CA). The instrument parameters were as follows: Heat = 485, Pull = 0, Velocity = 35, Time = 250 and Pressure = 500. The diameter of the tip is measured by the microscope (YX20L20, Dayueweijia Science and Technology Co. Ltd., Beijing) and is approximately 2 μm. Aspirated volume of solvent in nanopipette increases with immersion time and the solvent volume after 10 s of immersion was 1.95 nL±0.07 nL with good repeatability (n=6). Detailed information about the control of aspirated solution were in supporting information (Figure S1, S2, Table S1). Prepare Glu-6P 20%-methanol aqueous solution (0.025-2.5 mM) and 13C6-GluP 20%-methanol aqueous solution (0.20 μM). Insert capillary tip into Glu-6-P for 10 seconds and add the solution into microwells. Another capillary tip was inserted into 13C6-Glu-P solution for 10s and use it to extract Glu-6-P added in microwells for 20s. Subsequent pulsed-dc-ESI-MS analysis was conducted for the detection of the extracted Glu-6-P.

Single-cell Sampling Placed the cell array after pretreatment on an inverted microscope (DX30, Dayueweijia Science and Technology Co. Ltd., Beijing). Insert capillary tip into 20%-methanol aqueous solution containing 13C6-Glu-P (0.20 μM) for 10 seconds for solvent aspiration by capillary action. Hold the pulled capillary on the Capillary holder and connect it with a syringe through a well-tight tube. The capillary was slowly approaching to the microwells controlled by the three-dimensional manipulator (MP-225, Sutter Instrument). Due to the positive pressure provided by the syringe, the solvent was slowly pushed out from

ACS Paragon Plus Environment

Page 3 of 7 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

the tip and filled the microwells to extract cellular metabolites. After 20 seconds, suck the extraction back to the tip.

MS Detection and Data Analysis Home-made pulsed direct current electrospray ionization (pulsed-dc-ESI) sources was used for mass spectrometry detection. This pulsed-dc-ESI can determine components in small volume sample. Inserted the needle electrode into the capillary and a negative voltage of −1200 V was applied on the electrode for electrospray ionization. Detailed information about the setup of the pulsed-dc-ESI could be found in our previous work14. All mass spectrometry detection was performed by Thermo QE - Orbitrap mass spectrometer (Thermo Scientific, San Jose CA). The parameters were listed as follows: capillary temperature = 320 °C, resolution 35000 (AIF-MS/MS mode) and 70,000 (Full scan Mode), maximum inject time = 100ms, AGC = 3 × 106. The mass scan range was 100-1000. For confirming the structure, the metabolites extraction from single cell were detected by MS2 by collision induced dissociation at 25 − 40 eV collision energies. Subtract background signal from the mass spectra acquired from single cells using Thermo Xcalibur software and export peak list with exact mass from raw data. Search for possible metabolites in the human metabolome database (HMDB) (http://www.hmdb.ca/) and METLIN metabolite database (https://metlin.scripps.edu/) based on the accurate mass and isotopic profiles. MS2 data were used for further confirming relevant metabolites by comparing the major fragments with MS2 data from standards and those in HMDB.

RESULTS AND DISCUSSION Quantitation by Droplet Microextraction in Microwell The MS quantitation of analyte was performed through internal standard calibration. The internal standard was added into the extraction solvent and mixed with analytes during the microextraction in microwell. For reliable quantitative results, the repeatability of the analytical method is crucial. We performed a set of experiments to evaluate the precision of the nanoliter-droplet microextraction after introducing microwells. The known amounts of Glu-6-P were added in microwells and on plate separately, and then were extracted by nanoliter volume of solvent containing known amounts of 13C6-Glu-P. Subsequent pulsed-dc-ESI-MS analysis was conducted for the detection of the extracted Glu-6-P. The corresponding selected ion chronograms of Glu-6-P and 13C6-Glu-P were shown in Figure S3. The MS signal of Glu-6-P and 13C6-Glu-P were simultaneously appeared, demonstrating that the analyte and internal standard were well mixed. Figure 2C showed the ratios of MS intensity of Glu-6-P and 13C6-Glu-P that extracted from plate and microwells. The RSD was about 5% when extracted from the microwells while extracted from plate was about 18%, indicating a better reproducibility. When extracted on the plate, the nanoliter volume of liquid was pushed out from the tip under the pressure given by the syringe and spread irregularly on the plate due to the effect of surface tension between parallel samples (Figure 2A). It was hard to ensure the recovery of extraction solvent due to the non-uniform spread. By contrast, when extracted in microwells, the liquid was confined in the microwells and established a stable liquid microjunction with the tip, improving the recovery of trace solvent between parallel samples (Figure 2B). The confinement effect of microwell

Figure 2. Comparison of the microextraction on plate and in microwell. Optical images of droplet microextraction on plate (A) and in microwell (B). (C) Glu-6-P was extracted by aqueous extraction solution with 13C6-Glu-P as internal standard. The horizontal lines on the scatter diagram of the figure represented the mean with standard deviation (SD) between replicate samples, indicating the degree of dispersion between parallel samples under two extraction conditions (n=10).

Figure 3. Quantitation of Glu-6-P by droplet microextraction in microwell. (A) Mass spectrum obtained by extracting Glu-6-P at different concentrations in the range of 0.025-2.5 mM using 0.2 μM 13C -Glu-P as internal standard. (B) Calibration curve for Glu-6-P. 6

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. (A) Workflow of cell capture and cell array formation. The microwell size was 30 μm, 50 μm, and 70 μm respectively. (B) Fluorescence image of K562 cell array formed with the 30 μm-microwell. Most of the microwells had only one cell. (C) The percentage of 0~3 cells occupancies with the 30 μm microwell array.

Figure 5. Mass spectra of single K562. Glu-6-P and corresponding internal standard (13C6-Glu-P) were both detected.

avoided the loss of internal standard caused by irregular spread and ensured the consistency of internal standard recovery. Therefore, the extraction confined in the microwells and introduction of the internal standard can greatly reduce the RSD among the parallel sample and improve the microextraciton repeatability, which provided opportunities for quantitative analysis by droplet microextraction in microwells. The proof-of-concept experiment was conducted to verify the quantitative ability of the method. Different concentration of Glu-6-P solutions were added to the microwells and extracted by solution containing 0.20 μM 13C6-Glu-P, followed by MS detection. As shown in Figure 3, the ratio of MS intensity of Glu-6-P and 13C6-Glu-P were linearly dependent on Glu-6-P amounts in the range from 48.8 amol to 4.88 fmol with squared correlation coefficient (R2) of 0.989. The absolute amounts of many metabolites in one cell are in the range of amol to fmol15, so the linear range of our method can meet the demand for quantitative detection of Glu-6-P in single cell. Furthermore, Glu-6-P was added into cellular extract of cell population to test the recovery. Good recoveries were obtained at low, medium and high concentrations (Table

S2). Hence, the reliable quantification result can be obtained with the influence of biological matrices.

Single-cell Capture with Microwell Array High-density microwell array was obtained for single-cell capture and following single cell analysis. Microwells with 30, 40, 50 μm in diameter and 30 μm in depth were tested respectively for higher loading efficiency of single K562 cells. The process of K562 cells capture and cell array formation was shown in Figure 4A. The diameters of K562 cells were about 15 μm. Single cells were captured in the 30 μm microwells while 2-3 cells were captured in the 40 μm microwells and 3-5 cells were captured in the 50 μm microwells. Figure 4B is the fluorescent microscope image of K562 cells captured by 30 μm microwell. Among this cell array, the percent of microwells that filled with zero cell, one cell, two cells and three cells were 2.5%, 65%, 27.5% and 5% respectively. Therefore, the microwell with the diameter of 30 μm was the most effective for capture single K562 cells and was selected for the following experiment. The loading efficiency could also be adjusted by rounds of cell loading and concentrations of cell suspensions. We observed that the more rounds of cell loading and the greater

ACS Paragon Plus Environment

Page 4 of 7

Page 5 of 7 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry cell suspension concentration, the higher cell capture efficiency. In order to shorten the pre-treatment time of cells, one round of cell loading and the cell concentration of 1×107 cells/mL were chosen. And the distance between the two microwells was designed to be 300 μm to avoid interference and contamination between two adjacent microwells during cell lysis and extraction. It should be mentioned that two cells were captured in one microwell can be exclude by observing through microscopy, ensuring the accuracy of single cell analysis. The device was a universal platform and could be applied for various cells through adjusting the diameter of the microwells. For example, HeLa cells (Figure S4) and hematopoietic stem cells (Figure S5) were both captured with high loading efficiency by microwells.

MS Analysis and Quantitation of Glucose-phosphate in Single Cell The metabolite in single cells was analyzed and quantified by droplet microextraction-MS in microwell. K562 cells were isolated and captured by 30 μm-microwells and then extracted by 20% methanol-aqueous solvent containing 0.2 μM 13C6Glu-P. Figure 5 presented the mass spectrum of metabolites detected in single K562 cells. Thousands of signal peaks were profiled in Figure 5. Based on the accurate mass measurements, 67 metabolites were assigned (Table S3). MS2 spectrum of some metabolites acquired form single cell were compared with spectrum acquired from standard solutions for further metabolites identification (Figure S6). As shown in Table 1, these metabolites involved dozens of metabolic pathways, including tricarboxylic acids, glucose catabolism, amino acid pathways etc. The ability of this method to detect multiple metabolites simultaneously made it possible to be applied in single cell research widely. Table 1. Types of Metabolites Detected in Single K562 Cell.

Glu-6-P and corresponding internal standard (13C6-Glu-P) were both detected. According to the ratio of the mass intensity of Glu-6-P and 13C6-Glu-P, the absolute amounts of Glu-6-P in single K562 cell could be calculated from calibration curves. The amounts of Glu-6-P in the K562 cells

was estimated to be in the range of 111 ± 28 amol, which were obtaining from ten individual extraction experiments. Similar to previous studies27, this method can be extended to quantitation of multiple metabolites by adding appropriate internal standards or generating calibration curves for a classes of molecules.

Investigation of Metabolites Changes in Single Cell under Drug Stimulation Taking advantage of the ability of multi-component detection and the quantitation of the developed method, we further studied the metabolic interactions at single cell level. We studied the changes in metabolites quantitatively in single K562 cell stimulated by anticancer drug 2DG. K562 cells are human immortalized myelogenous leukemia cells. 2DG inhibits K562 cell growth by blocking glycolysis. It is utilized by cells as a glucose analogue and converted to 2DGphosphate (via hexokinase). However, 2DG-phosphate is unable to participate in glycolysis to generate energy for cancer cell growth. Accordingly, endogenous metabolites in the glycolysis pathway and the cellular energy state changed. To this end, a group of important metabolites were quantified, such as hex-phosphate, hexose bisphosphate, ribose-5phosphate, ADP and AMP. We incubated K562 cells with 2DG at 10 mM and used normal K562 cells as control. Both 2DG and 2DG-phosphate were detected in the experiment group (Figure S7). The liner relation between 13C6-Glu-P and hexose bisphosphate, ribose-5-phosphate and 2DG-phosphate were illustrated in Figure S8. Hence, 13C6-Glu-P could be used as the internal standard for ribose-5-phosphate, hex-phosphate and 2DG-phosphate for relative quantitation. As shown in Figure 6A~F, the key metabolites in glycolysis and energy state were significant changed after drug stimulation (n=10). 2DG-phosphate was only detected in K562 cells incubated with 2DG (Figure 6A). The amounts of hex-phosphate, hexose bisphosphate and ribose-5-phosphate (Ribose-5-P) were significantly reduced in 2DG-treated cells compared with the control group (p