Using the EWOD actuation principle, a droplet with suspended
protoplasts was merged with a hypo-osmotic droplet (water) or
hyper-osmotic droplet (buffer) (Fig. 1(d)) which resulted in the
transportation of water from the external solution toward the protoplast
interior or from the protoplast interior toward the external
solution. This water transport over the membrane resulted in an
increase or decrease of the protoplast volume and consequently
a change in volume. In Fig. 3(a), typical volumetric changes are
shown for five single protoplasts that were simultaneously treated
at a of −0.99 MPa and monitored over a period of 360 s (see
supplementary information for other tested conditions).
For the on-chip monitored protoplasts, the volume increased
in a non-linear way with time before reaching a final constant
value. At the most extreme hypo-osmotic condition tested
( = −0.99 MPa), 55% of the tested protoplasts did burst after
an initial volume increase (Fig. 3(a)), whereas for the moderate
hypo-osmotic test condition ( = −0.70 MPa), this phenomenon
occurred in 10% of the tested protoplast (supplementary information).
This reveals a low tolerance of certain protoplasts toward
the abrupt change in the external osmotic condition. One of the
possible reasons could be that due to the rapid influx of water in
the protoplasts, the plasma membrane gets excessively stretched,
finally leading to the lysis of the protoplasts [36]. Protoplasts
when treated with hyper-osmotic conditions ( = 1.21 MPa
and = 2.44 MPa) started to loose water molecules that lead
to shrinkage of protoplasts. We observed that the shrinking
phenomenon was highly non-uniform in all the directions, (supplementary
information) which lead to change in protoplast shape
over time and the model could no longer be applied on the protoplasts.
Also extreme hyper-osmotic treatments could lead to
implosion and hence death of protoplasts which could not be
detected clearly during the experiments. Hence we decided to evaluate
only hypo-osmotic conditions because the analysis was more
suitable for the developed modeled. After the droplet merging step,
the volume changes started with a short delay indicating a finite
delay in reaching a uniform concentration in the droplet. This delay
was considered in the model for calculating P values (see model,
Eq. (10). The volume changes were calculated from the measured
axis changes (D1 and D2). As depicted in Fig. 3(b), the two axis
lengths showed a non-linear trend which stabilized once the external
and internal concentration were equal. The presented results
illustrate the suitability of DMF chips for the treatment of multiple
non-adhering cells with different stimuli and simultaneously
monitoring their response at a single cell resolution. The data also
showed that on-chip merging of droplets was effective in changing
the osmolarity of the final droplet as the kinetics of the volumetric
changes observed are in agreement with those obtained by traditional
methods and reported in literature [36]. Next, the rates of
major and minor axis change were used for calculating the water
permeability coefficients of every single protoplasts.