Angewandte
Research Articles
Chemie
annealing showed peaks at similar positions around 0.40 and
0.45 nmꢁ1, but the peak intensity was largely increased by the
annealing process (Figure 5c). It indicates the improved
packing order of micelles resulting from annealing, which
would reduce the oxygen diffusivity and consequently en-
hance the air stability of the UC emission. Since the TTA-UC
in nonionic surfactant micelles dispersed in aqueous solution
usually requires deaerated conditions,[13] the observed air-
stability improvement of the Pluronic hydrogels is noteworthy
for biological applications.
2+
The TET efficiency from Os(peptpy)2 to TTBP was
estimated based on the phosphorescence lifetime in Pluronic
hydrogel. Without TTBP, the perylene-conjugated Os-
2+
(peptpy)2 showed a phosphorescence lifetime of 23 ms
Figure 6. a) Fluorescence microscopy images of GFP only (white),
mCherry only (white), and an overlay of GFP (green) and mCherry
(magenta). Arrowheads indicate GFP and mCherry double-positive
cells. Arrows indicate mCherry single-positive cells. Grid represents the
counting frame (200ꢀ200 mm) for stereological analysis. b) Cell scor-
ing of GFP-positive cells among mCherry-positive cells. * p<0.05
(compared to values for control).
(Figure 5d). The addition of TTBP shortened the lifetime of
Os(peptpy)22+ to 6.7 ms (Supporting Information, Figure S8c),
which gave a FTET (= 1ꢁtp/tp,0) of 71%. The occurrence of
triplet back energy transfer from TTBP to Os(peptpy)22+ was
also suggested by the long phosphorescence decay component
of 128 ms. The fact that the UC emission disappeared at
ꢁ1968C indicates that the TET process operates via molec-
ular diffusion and collision rather than the energy migration
in dye aggregates (Supporting Information, Figure S8d).
The TTA-based UC mechanism was supported by the
excitation intensity dependence of the UC emission. A double
logarithmic plot for the UC emission intensity of the Os-
(peptpy)22+-TTBP-Pluronic hydrogel showed a quadratic-to-
linear transition with threshold excitation intensity Ith of
13 Wcmꢁ2 (Supporting Information, Figure S8e). The abso-
lute quantum efficiency of this hydrogel was relatively low
(< 0.1%). The UC emission intensity was not improved by
deaeration of hydrogels by freeze-pump-thaw cycles, suggest-
ing the minor effect of dissolved oxygen on the UC efficiency.
While the reduction of excitation intensity and the improve-
ment of UC efficiency by enhanced energy migration in
controlled molecular assemblies remain as important future
works,[14] the current TTA-UC properties were found to be
enough to demonstrate the proof-of-concept of NIR opto-
genetics as shown below.
To apply this upconverting hydrogel for optogenetic
genome engineering, we used a PA-Cre system (Supporting
Information, Figure S9a).[8] We used Cre-reporter GFP
(Supporting Information, Figure S9b) and transfection-re-
porter mCherry (Supporting Information, Figure S9c) to
visualize optogenetic manipulation. Cre-reporter GFP starts
to be expressed in the presence of activated PA-Cre, while it is
not expressed in the absence of activated PA-Cre because of
a polyA signal sequence between two loxP sites. The cerebral
cortex was dissected from a day (E) 13 embryo and pCAG-
PA-Cre, pCALNL-GFP, and pCAG-mCherry plasmids were
transfected into cortical cells by electroporation[15] (Support-
ing Information, Figure S9d). After 48 hours in culture,
transfected cortical neurons were illuminated as described
in the Supporting Information. The total number of cortical
cells, counted after removing dead cells, was not significantly
increased the ratio of GFP-positive cells, while NIR-light
stimulation with the TTBP hydrogel did not (Figure 6a,b),
indicating that the UC hydrogel works as a tool for NIR UC
optogenetics.
Using the UC hydrogel, we examined whether NIR light
stimulation regulates hippocampal dendritic spines involved
in learning and long-term memory by receiving excitatory
input from axons.[16] Rac1(Q61L), the constitutive active form
of Rac1,[17] is known to promote dendritic spine formation.[18]
We used Cre-reporter Rac1(Q61L) to promote hippocampal
dendritic spine formation (Supporting Information, Fig-
ure S11a). The hippocampus was dissected from E15 embryo
and pCAG-PA-Cre, pCALNL-Rac1(Q61L), pCALNL-GFP,
and pCAG-mCherry plasmids were transfected into hippo-
campal cells (Supporting Information, Figure S11b). At
10 hours after NIR-light stimulation with the UC hydrogel,
both GFP expression and the formation of dendritic-spine-
like structures were observed (Figure 7e–h). On the other
hand, the non-light stimulation condition did not show GFP
expression or the formation of dendritic-spine-like structures
(Figure 7a–d). To examine the dynamics of spine-like for-
mation, we performed time-lapse imaging analysis (Support-
ing Information, Figure S11b). Newly generated dendritic
protrusions were observed by NIR-light stimulation with the
UC hydrogel from 6 hours after stimulation but not with the
control (Figure 7i,j and Supporting Information, Movie S1),
suggesting that the NIR-light stimulation promoted the
formation of dendritic-spine-like structures. From these
results, the NIR-light stimulation with the TTA-UC hydrogel
regulates neuronal morphology, which is important for
learning and long-term memory.
affected by the NIR-light illumination in the presence of the Conclusion
UC hydrogel (Supporting Information, Figure S10), suggest-
ing that the UC hydrogel did not affect cell viability in this
experiment. NIR-light stimulation with the UC hydrogel
We demonstrate NIR optogenetics by NIR-to-blue TTA-
UC for the first time. The stable TTA-UC emission in the
&&&&
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2019, 58, 2 – 9
These are not the final page numbers!
Sasaki, Yoichi
