Single-Cell Mobility Analysis of Metastatic Breast Cancer Cells.

Efforts have been taken to enhance the study of single-cells, however, the task remains challenging because most previous investigations cannot exclude the interactions between single cells or separately retrieved cells with specificity for further analyses. Here, a single-cell mobility analysis platform (SCM-Chip) is developed that can not only real-time monitor single-cell migration in independent niches but can also selectively recover target cells one by one. The design of each channel with a single-cell capture unit and an outlet enables the system to place single cells in different isolated niches with fluidic capture and to respectively collect target cells based on mobilities. SCM-Chip characterization of breast cancer cells reveals the presence of high- and low-migratory populations. Whole-cell transcriptome analysis establishes that monocyte chemotactic protein induced protein 1 (MCPIP1) is related with cell mobility; cells with a high expression of MCPIP1 exhibit low mobility in vitro and metastasis in vivo. The SCM platform provides a generic tool for accurate single-cell isolation and differentiation that can be readily adapted for the study of cancer and drug development.

Polydimethylsiloxane (PDMS RTV615) was obtained from Momentive Performance Materials (Waterford, NY). Next, the structure on the silicon wafer was used to fabricate the PDMS layer.
The mold and PDMS layers were baked at 80°C for 2 h, and the cured PDMS was cut and removed from the mold. The holes for the inlets and outlets were punched using needle sizes that were compatible with the size of the fluid input/output pins. The PDMS layer was then cleaned by briefly rinsing with isopropyl alcohol and deionized water and dried with nitrogen gas. After treatment with oxygen plasma, the PDMS layer was bonded immediately to a glass slide. Finally, the bonded device was baked for 2 h at 80°C.

Cell loading
Cells were harvested by treatment with 0.25% trypsin-EDTA. After centrifugation, cells were resuspended in serum-free medium at a final concentration of 5×10 6 cells/ml. First, cells were loaded in the device from the main inlet by the application of pressure via a Laboratorial Syringe Pump (LongerPump, TS-2A). The flow is directed from the inlet to the outlets through 3 branches of main channels. The cell flow rate was maintained as 2.5 L/min and about 12.5 L or less of the cell suspension was gently introduced into the chip to seed cells in close proximity to the capture units within different microchannels. Second, after most of single cells were captured by the hooks, low serum culture medium (1%) were introduced into the inlet at a flow rate of 20 L/min for 1 min in order to wash the uncaptured cells within the channels.
Meanwhile, the residual cells in the main inlet were cleared up with micro pipette tips and washed with cell-free medium for three times.

Single cell mobility array
After cell loading, the chip will be settled in the microscope incubator (Tokai Hit). During the experiment, the temperature of the cell culture chamber was maintained at 37° C in a humidified atmosphere of 5% CO 2 . Under this condition, single cell with different mobility will start migration after accommodation to microenvironment change. After 18 hours of cell migration, cultivation observations of the whole chip were collected for the profile of single cell migration.

Whole Chip Imaging
Imaging was performed using a high-resolution camera connected to an inverted optical microscope (Olympus, X81) equipped with objectives of 10× and 20×magnification. It cost about 5 minutes for taking the image of the whole chip (10×, bright field and FITC). Time-lapse images of the cell migration process were capture during the whole experiment. The migrated cells within the microchannel were identified at once and the cell migration paths were plotted ( Figure S3). Migration distance of individual cells were quantified based on the on the distance form star line to the final cell frontier of each migration channels after 18 hours of incubation without medium replacement. After analysis of the raw data of the cell migrating distance, we can access the two key factors associated with cell mobility, that is, the migration distance of individual cells and a percentage of the high-migratory cells.
Cell Retrieval, single RNA-sequencing analysis and bulk RNA-sequencing analysis First 20 L of PBS was introduced into the inlet for 2 mintues for wash the whole microfluidic channels, followed by flowing 50 L of trypsin in the inlet for 5 mintues in the stage top incubator. Then all single cells will flow toward different outlets and cell recovery was processed with microinjector. Single cells with different mobility were collected from individual outlets based on the migration profiles. According to cell mobility, cells were divided into two populations, high mobility cells and low mobility cells. Consequently, both of them were counted as three repicates of an amount of about 20 cells for single-cell RNA-sequencing. The single-cell suspensions were directly lysed for RNA purification, followed by the generation of cDNA through Smart-seq2 method for transcriptome sequencing. Briefly, these cells were transefered into lyasate buffer. Terminal deoxynucleotidyl transferase was used to add a poly(A) tail to the 3 end of the first-strand cDNA, and then perform 21 PCR cycles to pre-amplify the single-cell cDNA. After cDNA purification by DNA clean beads, cDNA was resuspended in elution buffer (EB) (Qiagen). The fragmented cDNA was end-repaired, dA-tailed, adaptor ligated and then subjected to 10 cycles of PCR amplification. After cDNA amplification, libraries were quantified using Qubit 3.0 (Life Technologies) and quality of libraries was checked using Bioanalyser (Agilent 2100 Bioanalyzer). Using the Illumina HiSeq2000 sequencer, we generated 48 Gb of sequencing data from the 120 single cells, with, on average, 0.4 million reads per cell with read length of 100 bp.
For bulk RNA sequencing analysis, total RNA samples were extracted from cells between MDA-MB-231/Vector and MDA-MB-231/MCPIP1 cells, followed by the generation of cDNA through Smart-seq2 method for transcriptome sequencing. Briefly, after PBS washes, these cells were transefer into lyasate buffer. Terminal deoxynucleotidyl transferase was used to add a poly(A) tail to the 3 end of the first-strand cDNA, and then perform 21 PCR cycles to pre-amplify the single-cell cDNA. After cDNA purification by DNA clean beads, cDNA was resuspended in elution buffer (EB) (Qiagen). The fragmented cDNA was end-repaired, dA-tailed, adaptor ligated and then subjected to 10 cycles of PCR amplification. After cDNA amplification, libraries were quantified using Qubit 3.0 (Life Technologies) and quality of libraries was checked using Bioanalyser (Agilent 2100 Bioanalyzer). Using the Illumina HiSeq2000 sequencer, we generated 98 Gb of sequencing data from 6 samples with read length of 150 bp.
Quality control was conducted: 1. The reads that aligned to adaptors with no more than two mismatches. 2. The reads with more than 10% unknown bases (N bases). 3. The reads with more than 50% of low-quality bases (quality value ≤ 5) in one read. We subsequently only used samples with Q20 > 75% and a rate of clean data (percentage of obtained final data with respect to raw data) of >45% for further analysis.
After the RNA library preparation, reads were aligning to GRCh38 (hg19 RefSeq), which as downloaded from the UCSC Genome Browser (http://genome.ucsc.edu), with HISAT2. Based on the above results, gene expression levels were estimated by StringTie. FPKM (expected number of Fragments Per Kilobase of transcript sequence per Millions base pairs sequenced) (calculated using Cufflinks (v 2.2.1)) was used to access the gene expression levels.
Differential expression between the putative groups was conducted using the R package Wound healing assays Cells were plated in 96-well plates in triplicate. After 24 hours, the cells were treated overnight with mitomycin C (5 g/mL, Santa Cruz Biotechnology). Then, a wound was made in the monolayer with a pipette tip, the medium was replaced by serum-free medium. Pictures of the wounds were taken 0 and 18 hours after treatment initiation, and wound closure was measured using ImageJ software.

Transwell Migration assay
Transwell migration assays were performed using a 24-well Transwell chamber system                   Table   Table S1. The calculated relative migration speed of three breast cancer cell lines based on the raw data from SCM-Chip.