background fluorescence even when closed, due to imperfect
quenching of the fluorophore by the quencher. Thus, when
delivered into cells, background fluorescence, nonspecific
interactions and cell-to-cell variations in MB delivery can all
lead to misinterpretation of high uptake efficiency for MB
hybridization signals. Ratiometric imaging, a technique that
enables correction for unequal probe delivery, has been a pop-
ular technique for normalizing delivery efficiency in studies
involving the use of fluorescent probes [20]. In the context of
MBs, the technique generally involves comparing the fluores-
cence of the MBs of interest with that of an optically distinct,
nonreactive reference probe both in solution and codelivered
into cells. Comparing the fluorescence ratio of the MBs to the
reference probe in cells with that obtained under well-
controlled conditions provides a means for determining
whether the MB signal in cells reflects MB hybridization or
nonspecific opening. Using this approach, we identified that
MBs with a fully PS-modified loop domain and a phosphodie-
ster stem (2Me/PSLOOPMB) exhibit the lowest false-positive
signal among the six different MB architectures tested in living
cells.- 1PBS can also be used as microinjection buffer.
- ImageJ can also be used for image analysis. Fiji can be down-
loaded at: http://fiji.sc/. ImageJ can be downloaded at:
https://imagej.nih.gov/ij/. - Dissolve the appropriate quantity of IRDye®800 NHS ester in
1–5μL anhydrous DMSO or DMF before reacting with the
aminodextran. Make sure the volume of DMSO or DMF is less
than 10% of the total reaction mixture. - To identify MB backbone architectures that are stable in living
cells, it is critical to use an appropriate control that enables
normalization of cell-to-cell variations in delivery. As the back-
bone stability is unknown, it might not be appropriate to use an
unquenched beacon as a normalization reference, as both
quenched and unquenched beacons may be degraded. Alterna-
tively, we have previously used fluorescently labeled dextran as a
means for normalizing cellular delivery of molecular beacons
[18, 19]. The IRDye®800-labeled dextran used in this study
has a molecular weight (~10 kDa) similar to the MBs, therefore
the two molecules were expected to exhibit similar transfection
efficiencies when codelivered into cells by microporation, as
shown in our previous results [18, 19]. - Microemulsion bubbles form a range of sizes, and increasing
the stirring speed can reduce bubble sizes. - Experimenters should try to image bubbles that have sizes
similar to cells.
254 Mingming Chen et al.