Project Summary
The proteins secreted from a cell constitute a complex subset of molecules referred to as the secretome.
They are key factors mediating cell-cell communication. So, eavesdropping on the secretome informs
molecular diagnostics, drug discovery and tissue engineering. The challenge then is to detect the proteins as
they are secreted only in minute amounts, and diluted and/or contaminated in culture. Moreover, since tissue is
heterogeneous, it is necessary to detect secretions from single cells, which is confounded by bulk-culture
analysis. So, sensitivity is paramount.
In response to the Focus Technology Research and Development solicitation, this proposal furnishes a
plan to develop a tool that uses a nanopore to interrogate the secretome of single cells with extreme, single
molecule sensitivity and high throughput. The blockades that develop in the ionic current through a nanopore,
when a secreted, charged molecule is impelled through it by an electric field, measure the molecular volumes
occluding the pore. A catalog of the blockades can be used to discriminate between different cellular
phenotypes non-destructively, quickly, in real-time, and to interrogate the secretome for specific biomarkers.
AIM #1: Single cell secretomics. As it reflects the different molecular constituencies comprising the
secretome, the blockade current distributions should reveal distinctive aspects unique to the cell-type. To prove
out this hypothesis, three categories of cells will be scrutinized: breast cancer cells; human induced pluripotent
stem cells and their derivatives; and mouse embryonic stem cells and their derivatives. Single cells will be
positioned with optical tweezers over a pore embedded in a microfluidic device and the resulting blockades will
be classified by the Cramér’s distance, ?, and the expression of specific biomarkers will be tracked in real-time.
AIM #2: Discriminant analysis of the single cell secretomes. To improve on ? for discriminating cell-types,
a Gaussian-mixture-model (GMM) will be implemented that captures the profile of proteins in a secretome.
This model will be fitted to the data with the number of components determined by a Bayesian Information
Criterion and classifier will be developed to discriminate cell-types in real-time. The GMM will infer which
proteins are up-/down-regulated in a cell, compared to the control, in an unbiased way.
AIM #3: Micro/Nanofluidic integrated circuits for improved throughput. To boost throughput, arrays of
eight nanopores will be fabricated and tested for concurrent single cell analysis. These arrays will be
embedded in a microfluidic device incorporating integrated pneumatic valves to be used to convey cells to
each pore, and each pore will be independently addressed by integrated electrodes for detecting blockades
and producing di-electrophoretic forces for positioning the cell (instead of optical tweezers). To slash the down-
time required to purge the microfluidic between measurements, fouling-resistant surfaces that relieve non-
specific binding of protein and prevent cell adhesion to either glass and/or PDMS microfluidics will be tested.