Evaluation of mRNA Localization Using Double Barrel Scanning Ion Conductance Microscopy Yuji Nashimoto,†,⊥ Yasufumi Takahashi,*,†,‡,§,# Yuanshu Zhou,‡,# Hidenori Ito,† Hiroki Ida,† Kosuke Ino,† Tomokazu Matsue,*,†,‡ and Hitoshi Shiku*,∥ †
Graduate School of Environmental Studies, ‡WPI-Advanced Institute for Materials Research, and ∥Graduate School of Engineering, Tohoku University, Miyagi 980-8577, Japan § Precursory Research for Embryonic Science and Technology (PRESTO), Japan Science and Technology Agency (JST), Saitama 332-0012, Japan S Supporting Information *
ABSTRACT: Information regarding spatial mRNA localization in single cells is necessary for a better understanding of cellular functions in tissues. Here, we report a method for evaluating localization of mRNA in single cells using doublebarrel scanning ion conductance microscopy (SICM). Two barrels in a nanopipette were filled with aqueous and organic electrolyte solutions and used for SICM and as an electrochemical syringe, respectively. We confirmed that the organic phase barrel could be used to collect cytosol from living cells, which is a minute but sufficient amount to assess cellular status using qPCR analysis. The water phase barrel could be used for SICM to image topography with subcellular resolution, which could be used to determine positions for analyzing mRNA expression. This system was able to evaluate mRNA localization in single cells. After puncturing the cellular membrane in a minimally invasive manner, using SICM imaging as a guide, we collected a small amount cytosol from different positions within a single cell and showed that mRNA expression depends on cellular position. In this study, we show that SICM imaging can be utilized for the analysis of mRNA localization in single cells. In addition, we fully automated the pipet movement in the XYZ-directions during the puncturing processes, making it applicable as a high-throughput system for collecting cytosol and analyzing mRNA localization. KEYWORDS: scanning ion conductance microscopy, mRNA localization, electrochemical syringe, cytosol collection, single cell analysis, transcriptome distribution. 14,15 Although these methods are able to successfully evaluate mRNA localizations, they require fixation of specimens and, therefore, only provide a snapshot of information about mRNA distribution. The localization of mRNA changes dynamically during embryonic developments, fibroblast cell migration, and maturation of neurons.8 To analyze these dynamic changes in mRNA localization in a single cell, it is essential to be able to analyze the distribution of mRNA in live cells. Scanning probe microscopy (SPM) is an effective tool for collecting mRNA from live cells in a minimally invasive manner. Atomic force microscopy (AFM) was adapted for the analysis mRNA localization over 10 years ago.16−18 Although improvements in collection methods have been reported,19,20 the quantity of mRNA collected relies on physical adsorption on the AFM-probe, making it difficult to obtain consistent
T
ranscriptome analysis at the single cell level is critical for understanding the heterogeneity of biological systems. Microfluidic technologies are well-established solutions for single cell RNA analysis, because of the small reaction volume and the capacity for automation and parallelization.1−4 However, dissociation of tissues for single cell sorting causes loss of spatial information on individual molecules with regard to an individual cell. The asymmetric intracellular distribution of cellular components (called “cellular polarity”) is crucial for homeostasis, and in some instances cellular polarity is maintained by mRNA distribution and onsite translation.5−8 Further development of the single cell transcriptome field requires the assembly of information about functional mRNA distribution in a single cell. To analyze transcript localization with subcellular resolution, single molecule fluorescence in situ hybridization (sm-FISH) has been widely used.9,10 Significant progress has been made toward increasing the number of detectable RNA target species.11−13 In addition, in situ RNA-seq at the single cell level could be available for the analysis of intracellular mRNA © 2016 American Chemical Society
Received: April 25, 2016 Accepted: July 11, 2016 Published: July 11, 2016 6915
DOI: 10.1021/acsnano.6b02753 ACS Nano 2016, 10, 6915−6922
Article
www.acsnano.org
Article
ACS Nano
Figure 1. Nanoscale imaging and collecting a minute amount of cytosol using a double-barrel nanopipette. To evaluate intracellular mRNA distribution, a double-barrel nanopipette was used as a probe for SPM. (a) SEM image of the double-barrel nanopipette tip. (b) Schematic representation of sequential nanoscale topography imaging and cytosol collection. A blue barrel indicates a water phase barrel, and a red barrel indicates an organic phase barrel. The water phase barrel is used for SICM. The organic phase barrel is used as an electrochemical syringe for cytosol collection. After nanoscale topographic imaging, the tip was moved to the target position in a semiautomatic manner. mRNA-containing cytosol collected in the pipet was evaluated using qPCR.
PBS and 10 mM tetrahexylammonium tetrakis (4-chlorophenyl) borate (THATPBCl) in 1,2-dichloroethane (1,2-DCE). Hereafter, barrels containing PBS and THATPBCl will be referred to as water phase barrels and organic phase barrels, respectively. Water phase barrels were used for SICM to obtain a nanoscale map of the target cell, and organic phase barrels were used as an electrochemical syringe to collect the cytosol (Figure 1b). First, we evaluated whether an organic phase barrel could be used to collect cytosol from a living cell. Influx of cytosol into the organic phase barrel was evaluated using a video microscope. To regulate the position of the interface in the nanopipette, we optimized the electrochemical potentials based on a previous report.29 Prior to collection, the potential of the organic phase barrel was maintained at +1.0 V vs Ag/AgCl to prevent extracellular solution from entering the nanopipette. After the tip of the nanopipette had punctured the cellular membrane and reached the cytoplasm, the potential of the organic phase barrel was switched to the negative direction (−0.5 V vs Ag/AgCl) and held for 10 s. This stimulated the collection of cytosol from the target region. Subsequently, the potential of the organic phase barrel was switched to +0.3 V vs Ag/AgCl, to hold the collected cytosol in the nanopipette as it was quickly raised and removed from the holder. The volume of the solution was calculated using the following equation, modified from the previous paper:29
amounts of cytosol from the cell. Scanning ion conductance microscopy (SICM)21−27 is expected to be a more consistent cytosol collection tool, because it uses a (nano)pipet as a probe. However, to date there have been no reports of cytosol collection using SICM imaging, because it is difficult to precisely control the amount of cytoplasm collected at the fL− pL level. An organic solution-containing nanopipette has been used for analytical science in electrochemistry for 40 years.28 Laforge et al. reported the use of an electrochemical syringe, which has the ability to manipulate aL−fL volumes of liquids in pipettes. The electrochemical syringe is based on the regulation of an interface between the organic solution in the pipet and a water solution outside the pipet, using electrochemical potential.29 Recently, Actis et al. reported a method for cytosol collection for analyzing mRNA in the cytoplasm using an electrochemical syringe.30 While this is an effective method, linking mRNA expression with spatial information based on SPM imaging has not yet been achieved. Our group has reported the use of SPM systems to analyze cellular functions in relation to spatial information.31−33 In the current report, we developed a SPM system that links mRNA expression with spatial information to analyze mRNA localization in a single cell. This was realized by integrating two functions into a single nanopipette. Thus, a technique that combined nanoscale topographical images and collection of minute amounts of cytoplasm was established. Because nanopipettes could be adapted to other fields with relative ease, this report will contribute to advances within a number scientific disciplines, such as proteomics or metabolomics.34,35
V=
⎫ ⎞3 πa3 ⎧⎛ h ⎨⎜1 + tan α⎟ − 1⎬ ⎠ 6tan α ⎩⎝ a ⎭
where a is the pipet aperture radius, α is the angle between the cone element and its axis, and h is the height of interface from the orifice. Figure 2a shows sequential photographs during cytosol collection from MCF-7 cells. After the potential was switched to −0.5 V vs Ag/AgCl, cytosol was drawn into the organic phase barrel. We also confirmed that the target cells were morphologically changed during the collection of cytosol (Supporting Movie S1). These results clearly indicated that
RESULTS AND DISCUSSION The goal of this study was to develop a SICM system for the evaluation of mRNA localization in single cells. To achieve this, we investigated possibility of integrating two functions into a single nanopipette: acquiring a nanoscale map of the target cell and collecting a minute amount of cytosol from a living cell. We utilized double-barrel glass nanopipettes (Figure 1a) filled with 6916
DOI: 10.1021/acsnano.6b02753 ACS Nano 2016, 10, 6915−6922
Article
ACS Nano
Figure 2. Minute amount of cytosol collected using an electrochemical syringe. (a) Sequential micrographs acquired during cytosol collection. Potential of the organic phase barrel was −0.5 V vs Ag/AgCl. Arrows indicate the interface between the organic and water solutions. (b) Collection speed of the outer PBS solution is indicated by black circles, and cytosol collection speed is in red. Data are shown as mean ± standard deviation; n = 6 for PBSexperiment and n = 5 for cytoplasm-experiment. (c) Results of melting peak analysis after PCR, using GAPDH as the target gene.
Figure 3. Evaluation of cytosol collected from the nanopipette. (a) Relationship between transcripts collected in the pipet and transcripts quantified using qPCR. (b) Comparison of qPCR success rate between differentiated and undifferentiated EBs. Black bar: Gapdh, Red bar: Pou5f1; n = 10 for differentiated and undifferentiated EBs.
our method successfully collected a minute amount of cytosol from the single cells. The volume collected into the pipet increased at sub-nL levels as the time period increased. This indicates that an organic phase barrel in a double-barrel nanopipette is capable of collecting a minute amount of cytosol from living cells, which is consistent with previous reports using single-barrel pipettes.30 Interestingly, the collection speed for cytosol was about one-third that of 0.1× PBS containing PCRamplicons of glyceraldehyde-3-phosphate dehydrogenase (GAPDH (human), 2.09 × 108 copies/μL) (Figure 2b, cytoplasm, 30.6 fL/s; 0.1× PBS, 102.8 fL/s). This difference could be due to the viscosity of the cytoplasm or to resistance from the cellular membrane. mRNA in the collected cytosol was reverse transcribed, and cDNA was amplified using PCR. The melting peak of the collected samples was very close to that for the positive control (GAPDH plasmid) (Figure 2c), indicating that the cytosol was suitable for gene expression analysis. Next, we evaluated the correlation between oligonucleotides collected into the organic phase barrel and those quantified by quantitative PCR (qPCR). Oligonucleotides collected into the pipet were calculated using the volume of standard solution in the pipet and were compared with the qPCR results. As shown in Figure 3a, the quantified oligonucleotides were proportional to the collected volume, indicating that oligonucleotides collected into the pipet could be quantified by qPCR. The slope of the least-squares analysis was 103 mES cells). The percentages of mRNA detectable samples of Gapdh (murine) and Pou5f1 (murine) were compared between undifferentiated and differentiated mouse embryoid bodies (mEBs) using the cytosol collected by the electrochemical syringes (Figure 3b). Although the percentages of mRNA detectable samples for Gapdh were comparable in the undifferentiated and differentiated EBs, those for Pou5f1 were greater in the undifferentiated than in the differentiated EBs, which is consistent with the previous result.36 These results indicate that our system can be used to evaluate gene expression levels. We then evaluated whether SICM imaging could be performed using the ion current of a water phase barrel. A schematic of the double-barrel SICM is shown in Supporting Figure S1. We first tried to acquire SICM images under conventional conditions where the potential of the water phase barrel was +0.2 V vs Ag/AgCl. However, periodic noise hampered acquisition of the topography images (Supporting 6917
DOI: 10.1021/acsnano.6b02753 ACS Nano 2016, 10, 6915−6922
Article
ACS Nano
This could be due to the low concentration of ion species in the organic phase relative to the water phase barrel. The electrolyte concentration was set to
