High-Throughput Multiparameter Analysis of Individual Mitochondria

Chem. , 2012, 84 (15), pp 6421–6428 .... Figure 1. (A) Schematic design for the high-throughput multiparameter ... A program written in LabVIEW was ...
0 downloads 0 Views 3MB Size
Article pubs.acs.org/ac

High-Throughput Multiparameter Analysis of Individual Mitochondria Shuyue Zhang,†,§ Shaobin Zhu,†,§ Lingling Yang,† Yan Zheng,† Min Gao,† Shuo Wang,† Jin-zhang Zeng,‡ and Xiaomei Yan*,† †

The Key Laboratory of Analytical Science, The Key Laboratory for Chemical Biology of Fujian Province, Department of Chemical Biology, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen, Fujian 361005, People’s Republic of China ‡ School of Pharmaceutical Sciences and Institute for Biomedical Research, Xiamen University, People’s Republic of China S Supporting Information *

ABSTRACT: Mitochondria are one of the most important organelles responsible for cellular energy metabolism and apoptosis regulation. However, single-mitochondrion analysis is challenging, because of their small sizes and the low content of organelle constituents. Here, we report the development of a sensitive and versatile platform for high-throughput multiparameter analysis of individual mitochondria. Employing specific fluorescent staining with a laboratory-built high-sensitivity flow cytometer (HSFCM), we demonstrate the simultaneous detection of side scatter, cardiolipin, and mitochondria DNA (mtDNA) of a single mitochondrion. Simultaneous measurements of side scatter, porin, and cytochrome c of individual mitochondria are reported for the first time. Correlation analysis among multiple attributes on an organelle-by-organelle basis could provide a more definitive assessment of the purity, structure integrity, and apoptosis-related proteins of isolated mitochondria than bulk measurement. This work represents a significant advancement in single-mitochondrion analysis. We believe that the HSFCM holds great potential for studying apoptotic signal transduction pathways at the single-mitochondrion level.

M

electrophoresis (CE) coupled with post-column laser-induced fluorescence (LIF).8−14 Recently, a capillary flow cytometer with on-column LIF detection was applied to determine heteroplasmy of individual mitochondrial particles containing one or more nucleoids.15 Information about the distribution of one or two properties within the population of mitochondria can be obtained by single-mitochondrion CE studies.16 Compared with CE, flow cytometry (FCM) can measure multiple parameters of individual mitochondria with high throughput. However, because of the small sizes of isolated mitochondrial particles (0.5−1.0 μm in diameter) and the low content of specific organelle constituents, flow cytometric analysis of individual mitochondria has been limited to attributes that can be brightly stained, such as cardiolipin, ROS, and membrane potential.17−19 Because of the pivotal role of mitochondrial proteins in apoptotic signal transduction,20 it is important to develop advanced techniques for the simultaneous detection of multiple specific mitochondrial proteins at the single-organelle level. Employing the techniques for single-molecule fluorescence detection in sheathed flow,21−25 we have developed a high-sensitivity flow cytometer (HSFCM) for the analysis of individual nanoparticles and

itochondria are membrane-enclosed organelles responsible for cellular energy metabolism, calcium homeostasis, free-radical production, and apoptosis regulation.1,2 Insights into the biological functions of mitochondria have been mainly acquired via ensemble measurements of isolated mitochondria, such as spectrophotometry, chromatography, enzyme-linked immunosorbent assay, and Western blot. The number of mitochondria within a cell can vary from tens to several thousands, depending on cell type and energy activity. Studies have shown that mitochondria within an individual cell are morphologically heterogeneous, and this heterogeneity has been implicated as important for the diverse functionality exhibited by mitochondria.3−6 Therefore, probing the properties of a single mitochondrion is of the utmost importance to uncover individual differences between mitochondria, resolve mitochondrial diversity, and possibly identify cellular processes associated with a specific mitochondrial subpopulation.4 In addition, analyzing isolated mitochondria at the single-organelle level can simplify the complexity of biological sample by eliminating the potential interference of other subcellular structures. The Arriaga group has pioneered studies in characterizing mitochondria heterogeneity.4,7 Properties of individual mitochondria including protein content, electrophoretic mobility, cardiolipin content, reactive oxygen species (ROS), mitochondrial DNA (mtDNA), isoelectric point, and cytoskeleton binding have been measured using capillary © 2012 American Chemical Society

Received: January 23, 2012 Accepted: July 9, 2012 Published: July 9, 2012 6421

dx.doi.org/10.1021/ac301464x | Anal. Chem. 2012, 84, 6421−6428

Analytical Chemistry

Article

Figure 1. (A) Schematic design for the high-throughput multiparameter analysis of individual mitochondria. Upon fluorescence staining, the isolated mitochondria are passed individually through a laboratory-built high-sensitivity flow cytometer (HSFCM). (B) Schematic diagram of the optical path for the laboratory-built three-channel HSFCM. [Legend: P, half-wave plate; S, polarizing beam splitter; M, mirror; L, achromatic-doublet lens; C, flow cell; AL, aspheric lens; DicF, dichroic filter; EF, edge filter; LP, long-pass filter; BP, band-pass filter; PMT, photomultiplier tube.]

1013 and 4.5 × 1012 particles/mL for the 110-nm and 200-nm FluoSpheres, respectively. Sphero rainbow calibration particles RCP-30−5A (8 peaks, 3.0−3.4 μm) were purchased from Spherotech (Libertyville, IL, USA). Anti-cytochrome c monoclonal antibody (mAb, purchased from eBiosciences, San Diego, CA, USA) was labeled with Alexa Fluor 488 following the instruction of Alexa Fluor 488 Monoclonal Antibody Labeling Kit (Molecular Probes). Anti-porin mAb (purchased from MitoSciences, Eugene, OR, USA) was biotinylated, following the instruction of a biotin-XX microscale protein labeling kit (Molecular Probes). Trypsin, paraformaldehyde, digitonin, tetraethyl orthosilicate (TEOS), fluoresceinisothiocyanate (FITC), and 3-aminopropyltriethoxysilane (APTES) were purchased from Sigma (St. Louis, MO, USA). Staurosporine was obtained from Enzo Life Sciences (New York, NY, USA). Unless otherwise stated, all the reagents used in present study were of analytical grade and obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, PRC). Mitochondrial buffer (MT buffer) containing 250 mM sucrose, 10 mM HEPES, 1 mM EGTA, 4.2 mM sodium succinate dibasic hexahydrate, 1 mM potassium dihydrogen phosphate, 0.01% Tween 20, and adjusted to pH 7.0 with 1 M potassium hydroxide was used for mitochondria washing and staining. Distilled, deionized water supplied by a Milli-Q RG unit (Millipore, Bedford, MA, USA) was filtered through a 0.22-μm filter and used for buffer preparation and served as the sheath fluid. All the buffers were filtered through a 0.22-μm filter and used within three weeks. Instrumentation. A laboratory-built three-channel highsensitivity flow cytometer (HSFCM) was used for mitochondria analysis. A solid-state 488-nm continuous-wave laser (Newport Corp., Irvine, CA) was used as the excitation source. The laser excitation power used in the present study (measured after mirror reflection) was 5.0 mW. The 0.7-mm laser output beam was focused to an ∼8.9-μm-diameter spot (1/e2) by an achromatic doublet lens onto the hydrodynamically focused sample stream inside a 250 μm × 250 μm quartz flow channel (NSG Precision Cells, Farmingdale, NY). The light emitted from individual mitochondria was collected by an aspheric lens and then directed by the first dichroic beam splitter (FF500Di02, Semrock, Inc., Rochester, NY) into two light paths. The reflected side-scattering light was directly detected by a photomultiplier tube (PMT-1, R3788, Hamamatsu, Japan).

single phycoerythrin molecules.26−30 Sub-picoliter probe volume and millisecond transit times are used for background reduction and photon emission enhancement, respectively. By upgrading the laboratory-built HSFCM to facilitate simultaneous detection of side scatter, green and red fluorescence signals, we here report the development of a sensitive and versatile platform for the rapid and multiparameter analysis of individual mitochondria. Correlation analysis among the attributes on an organelle-by-organelle basis should provide a more definitive assessment of the purity, structure integrity, and apoptosis-related proteins of isolated mitochondria. Figure 1 depicts the experimental design and the schematic diagram of the optical path for the laboratory-built HSFCM. After fluorescent labeling with specific probes or antibodies, the mitochondria were forced to transverse individually through the full width of the tightly focused laser beam at the same rate via hydrodynamic focusing. The side scatter signal reflects the mitochondrial internal structure and refractive index,31 and the intensities of fluorescence signals give quantitative measurements of specific mitochondrial components. Using mitochondria isolated from HeLa cells, two types of correlation analysis were demonstrated for the first time at the single-mitochondrion level by simultaneous detection of (i) side scatter, cardiolipin in mitochondrial inner membrane, and mtDNA in the matrix; and (ii) side scatter, cytochrome c in the intermembrane space, and porin protein in the outer membrane. With an analysis rate of up to 100 events per second, statistically meaningful distribution profiles of mitochondrial properties can be acquired in 1 min after analyzing thousands of mitochondria.



EXPERIMENTAL SECTION Reagents and Chemicals. Alexa Fluor 647-R-phycoerythrin streptavidin conjugate, 10-N-nonyl-acridine orange (NAO), SYTO 62 nucleic acid stain, and Alexa Fluor 488 Annexin V/Dead cell Apoptosis Kit were purchased from Molecular Probes (Eugene, OR, USA). Yellow-green FluoSpheres with sizes of 110 ± 5 nm and 200 ± 9 nm, with excitation/emission maxima of 505/515 nm, were also obtained from Molecular Probes. The manufacture-reported FITC equivalents are 7400 and 110000 molecules per sphere, and the manufacturer-reported particle concentrations are 2.7 × 6422

dx.doi.org/10.1021/ac301464x | Anal. Chem. 2012, 84, 6421−6428

Analytical Chemistry

Article

The transmitted light was spectrally filtered by a Raman edge filter (LP03-488RS, Semrock) and separated by the second dichroic beam splitter (FF593-Di02, Semrock) into two light paths for green and red fluorescence detection, respectively. The reflected fluorescence passed through a band-pass filter (FF01-520/35, Semrock) and was detected by PMT-2 (R928, Hamamatsu) for green fluorescence detection. The transmitted fluorescence passed through a long-pass filter (565ALP, Omega Optical, Inc. Brattleboro, VT), a band-pass filter (FB700-40, Thorlabs, for SYTO 62 measurement or FF01-670/30, Semrock, for Alexa Fluor 647 measurement), and then was detected by PMT-3 (R928) for red fluorescence detection. The output signals from the PMT detectors were routed to a National Instruments DAQ card (PCI-6221, Austin, TX) and recorded at a sampling rate of 10 kHz. A program written in LabVIEW was used to perform the data acquisition and processing. The criteria used for peak identification were the threshold levels in both the peak height (a digital discriminator level set to 3−10 times of the standard deviation of the background) and the peak width (0.5 ms). Burst height and burst area distribution histograms for all three channels can be generated dynamically. Two-parameter and three-parameter correlation analyses were conducted directly with the data processing program. Sample fluid was delivered pneumatically via a precise pressure regulator to a fused-silica capillary (40-μm ID, 240 μm OD, ∼45 cm long, Polymicro Technology, Inc., Phoenix, AZ) inserted into the 250 μm × 250 μm square bore flow channel of the flow cuvette. The inserted tip of the capillary, positioned ∼500 μm below the laser beam, was ground to a ∼12° taper (New Objective, Inc., Woburn, MA) to facilitate smooth laminar flow of the sheath fluid around the capillary tip. Ultrapure water served as a sheath fluid via gravity feeding, and the flow rate can be regulated by adjusting the relative height between the sheath supply bottle and the waste container. For each mitochondrial sample, 60 s of data acquisition time was used. The mean transit time of intact mitochondria passing through the focused laser beam individually was ∼1.5 ± 0.3 ms (see Figure S1 in the Supporting Information), corresponding to a linear sample flow rate of 5.9 mm/s in the flow cell. The sheath flow rate was measured to be ∼22 μL/min by an increment method. The sample volumetric flow rate was measured to be ∼5 nL/min by calibrating with the 200-nm FluoSpheres with a known concentration. Defined by the overlap of the focused laser spot (∼8.9 μm) and the sample stream (∼4.2 μm in diameter), the detection volume was calculated to be ∼0.13 pL. Based on Poisson statistics, when the concentration of mitochondria is ∼1 × 109/mL, the probability that two mitochondria will happen to pass through the probe volume simultaneously is 0.7%. Synthesis and Calibration of Fluorescent Silica Nanoparticles. To enable the comparison of fluorescence sensitivity between the HSFCM and the conventional flow cytometer, a set of particle standards in the nanometer scale and in the weak fluorescence range were prepared. Fluorescent silica nanoparticles doped with varying amount of FITC were prepared through a typical Stöber-based synthesis method.32,33 Briefly, FITC molecules were first covalently conjugated to an aminecontaining silane agent (APTES) at a molar ratio of 1:1 by stirring the mixture in ethanol at room temperature for 24 h in darkness. Second, 0.7 mL of TEOS was added to a clean glass reaction vessel containing 16.75 mL of ethanol, 1.2 mL of ammonium hydroxide (28%), and 0.9 mL of deionized (DI)

water. After the mixture was stirred at room temperature for 1 h, the APTES-FITC conjugates with a desired volume were added and stirred for another 24 h. After the reaction, the samples were centrifuged at 9000 rpm for 10 min to collect the silica nanoparticles. The nanoparticles were washed with ethanol and DI water three times via centrifugation to remove the unreacted chemicals. The particle size was measured to be 298 ± 8 nm by scanning electron microscopy (SEM) and nanoparticles in four different intensities were prepared (the blank beads, F1, F2, and F3). The fluorescence intensity of each bead population was calibrated in the unit of molecules of equivalent soluble fluorochrome (MESF) of FITC (MESFFITC) or FITC equivalents per nanoparticle by analyzing their mixture with 110-nm FluoSpheres (7400 FITC equivalents) on the HSFCM. With a 1-point calibration factor, the calibrated intensities are 298, 987, and 2062 FITC equivalents for populations of F1, F2, and F3, respectively. Cell Culturing and Mitochondria Isolation. HeLa cells were cultured in Dulbecco’s modified Eagles medium (Hyclone) supplemented with 10% (v/v) FBS (Sigma), 1% penicillin (Gibco), and 1% streptomycin. Cells were seeded into 150-mm Falcon culture dishes and maintained by the addition of new media every 2−3 days. Before mitochondria isolation or staurosporine inducement of normal cells, cell viability was confirmed to be >95% using the trypan blue exclusion assay. Apoptosis was induced by culturing the cells in medium containing 500 nM staurosporine for 10 h. Mitochondria extraction was performed using the Pierce Mitochondria Isolation Kit (Pierce, Rockford, IL, USA) in conjunction with the Dounce homogenization method, following the manufacturet’s instructions. HeLa (2 × 107) cells were used for each mitochondria isolation experiment. The mitochondria were resuspended in the storage buffer provided in the isolation kit, maintained on ice, and used for subsequent staining within 3 h. Fluorescent Staining of Mitochondria. For NAO and SYTO 62 staining, the isolated mitochondria were centrifuged at 12 000 g for 5 min at 4 °C and resuspended in 600 μL of MT buffer. Every 100 μL of the mitochondria suspension was pipetted into an Eppendorf tube. For NAO staining, the mitochondrial sample was pelleted and resuspended in 200 μL of an NAO working solution freshly prepared with MT buffer. The NAO and mitochondria were allowed to react at 37 °C for 15 min. After washing once with 500 μL MT buffer, the NAOstained mitochondria were resuspended in 200 μL MT buffer for HSFCM analysis. For SYTO 62 staining, the mitochondrial sample was pelleted and resuspended in 200 μL SYTO 62 working solution freshly prepared with MT buffer. After 15 min incubation on ice, the sample was analyzed on the HSFCM. For NAO and SYTO 62 double staining, after washing the NAOstained mitochondria once with 500 μL MT buffer, the pellet was resuspended in 100 μL SYTO 62 working solution prepared freshly with MT buffer and incubated on ice for 15 min. For immunostaining, the isolated mitochondria were centrifuged and resuspended in 600 μL fixing buffer (1% paraformaldehyde, 250 μg/mL digitonin in MT buffer) at room temperature for 10 min. Every 100 μL of the mitochondria suspension was pipetted into an Eppendorf tube and centrifuged. The pellet was resuspended with 50 μL of incubation buffer (250 μg/mL digitonin in MT buffer) containing 20 μg/mL of Alexa Fluor 488-labeled anticytochrome c mAb, or 20 μg/mL of biotinylated anti-porin 6423

dx.doi.org/10.1021/ac301464x | Anal. Chem. 2012, 84, 6421−6428

Analytical Chemistry

Article

Figure 2. Flow cytometric analysis of the unstained, 200-nM NAO-stained, and 500-nM SYTO 62-stained mitochondria on the laboratory-built HSFCM. Panels A-1, A-2, and A-3 respectively represent side scatter, green fluorescence, and red fluorescence burst traces for the unstained samples; panels B-1, B-2, and B-3 respectively represent the same for the 200-nM NAO-stained samples, and panels C-1, C-2, and C-3 respectively represent the same for the 500-nM SYTO 62-stained mitochondria on the laboratory-built HSFCM. The saturated signal intensity is 10 V, and the concentration of mitochondria was ∼1 × 109/mL.

When the mitochondria were stained with 200 nM NAO, welldefined green fluorescence bursts were detected concurrently with their corresponding side scatter bursts (see panel B-2 in Figure 2). No crosstalk of the green fluorescence signals to the red fluorescence channel was observed (see panel B-3 in Figure 2). In order to estimate the sensitivity required for side-scatter detection of a single mitochondrion, a mixture of 110-nm and 200-nm FluoSpheres was analyzed in parallel. As disclosed by the bivariate frequency distribution for side scatter versus green fluorescence burst area (see Figure S3 in the Supporting Information), the side scatter of mitochondria was broadly distributed and mainly fell in the range of 110-nm and 200-nm polystyrene nanospheres. Therefore, despite of their relatively large particle sizes, mitochondria scatter less incident light than polystyrene nanoparticles. This can be attributed to the small refractive index of mitochondria. Because it is difficult for the conventional flow cytometer to discriminate nanoparticles