Page 75 - Read Online

Page 8 of 19 Shi et al. J Cancer Metastasis Treat 2018;4:47 I http://dx.doi.org/10.20517/2394-4722.2018.32
[63]
lung cancer using this CTC-chip to perform an EGFR mutational analysis . An improved microfluidic
[64]
CTC isolation platform, the herringbone (HB)-chip, is also developed by the same group . The HB-chip
uses calibrated microfluidic flow patterns to drive cells to come in contact with the antibody-coated walls
of the device, thereby reducing cell collisions and improving target cell capture efficiency. A commercial
microfluidic circuitry chip DEPArray System (Menarini Silicon Biosystems, Inc.) containing an array of in-
dividually controllable electrodes to create a dielectrophoretic (DEP) cage around each cell for single CTC
[65]
isolation is also available . Besides isolation of CTCs from blood, the microfluidic platform can also be
used for single-cell isolation from other tissues [66,67] . For example, an innovative workflow using DEPArray
system was established to examine tumor heterogeneity using FFPE samples, providing a solution for genetic
[68]
analysis using minute archival clinical samples .
Immunoaffinity-based cell isolation
The CellSearch Circulating Tumor Cell Kit (Menarini Silicon Biosystems, Inc.) is based on ferrofluid- and
fluorochrome-couple antibodies with high binding affinities for the EpCAM antigen of CTCs. After immu-
nomagnetic capture and enrichment, CTCs in peripheral blood are detected and enumerated as measured
by fluorescence intensity. ITH has been reported for PIK3CA and TP53 mutations in metastatic breast cancer
using a combination of CellSearch and DEPArray technologies [69,70] . CTCs can also be purified and enriched
[71]
using an immunomagnetic enrichment device termed MagSweeper . Using this technique, high level of
[72]
heterogeneity among individual CTCs was detected in the blood of metastatic breast cancer patients .
Isolation of single cells using Fluorescence-activated cell sorting
Flow cytometry using fluorescence-activated cell sorting is a powerful method of isolating single cells that
share the same marker from liquid suspensions. Cells passing through the lasers emit optical signals en-
abling their separation and capture from other cells that lack the signal [73,74] . Single cells can be sorted indi-
vidually onto a 96 well plate format [Figure 4B]. Alternatively, a serial dilution can be performed using the
sorted cell suspensions into a 96 well plate such that each well contains a single cell. Downstream sequenc-
ing can be performed using a 96 well plate format.

Isolated single cells can be interrogated by a variety of genomic technologies for deeper genotype-phenotype
characterization. Significant technological advancement summarized in the next section is producing novel
insights into the biology of the disease and applications in the clinic.

Downstream analysis of single cells
Single-cell genomics
The work-flow of single-cell sequencing involves amplification of genomic DNA or RNA transcripts to pro-
duce enough material for library construction. The earliest method of sequencing DNA from single-cells
combined flow-sorting cells by DNA ploidy followed by single-nucleus sequencing by degenerative-oligonu-
cleotide-PCR technique [74,75] . However, this method failed to generate genome-wide single nucleotide variants
due to low coverage of ~6% [74,75] . A non-PCR-based multiple-displacement DNA amplification method using
Phi29 enzyme and random hexamers [Table 1] produced good genome coverage with high sequence fidelity
in multiple single-cell studies [58,76-79] . Another amplification method - multiple annealing and looping-based
amplification cycles (MALBAC) reduced whole-genome amplification bias and improved genome coverage
[Table 1]. In the MALBAC method, limited isothermal amplification using degenerate primers, followed by
PCR amplification produced 93% genome coverage for a single cell and both copy-number variations and
[80]
single nucleotide variations were detected . Amplification bias is a serious limitation in single-cell sequenc-
[81]
ing, which can reduce the accuracy of genomic information from single-cell genomes . Statistical models
have been developed to calibrate allelic bias in single-cell whole-genome amplification to reduce the sequenc-
[81]
ing artifacts .