Angewandte
Chemie
ultimate release of C in the self-immolation processes.
Further evidence supporting our kinetic interpretation was
found from analysis of the brightness of the intermediates
involved in S2CD self-immolation (see Supporting Informa-
tion).
The preceding kinetic analysis makes it possible to
evaluate the present strategy for uncaging with fluorescent
reporting. The substrate and its reporter are released in a one-
to-one molar ratio about ten seconds after removal of the
phenol group. In fact, this time is among the shortest times
ever reported for disassembly of a self-immolative linker.[2h,13]
Furthermore, using the values of the uncaging cross-section
ef(365 nm), where e is the molar absorption coefficient, and
the rate constants k21, k22, and k31, we simulated the change in
the concentrations of C and D upon illumination of a cS2CD
solution (see Figure 1d). C is released about ten seconds after
D. Although this delay would be detrimental to quantitative
reporting of uncaging in an open system, it is sufficiently short
to analyze most biological processes occurring in a closed
system (e.g., a cell). Indeed the delay between the release of
the substrate and its reporter remains moderate with respect
to characteristic times of cellular events, many of which
exceed the minute range (e.g., phosphorylation, protein
translation, protein degradation).
Next, we reproduced the preceding uncaging experiments
in human embryonic kidney (HEK) 293 cells. We incubated
cells for 30 minutes in a solution of 5.3 mm cS2CD in 100 mm
PBS buffer pH 7.4. We observed identical cellular fluores-
cence following uncaging with 365 nm light when cells were
incubated with 5 mm cS2CD for 30 minutes or 2 hours (data
not shown), suggesting that an incubation time of 30 min is
sufficient to equilibrate the distribution of cS2CD between the
extracellular solution and the cytoplasm. After rinsing with
PBS, cells were continuously illuminated at 365 nm and
imaged in real time with epifluorescence microscopy using
two independent channels, collecting the emission of C and D
(see the Movie in the Supporting Information).[14] We
analyzed the kinetics of photoinduced release of C and D
within HEK 293 cells (Supporting Information, Figure S8).
We did not notice any difference in the uncaging behavior
among cells that were similarly illuminated. Proceeding as
above, we calculated f(365 nm) = 3.5 ꢀ 0.2 ꢀ 10ꢁ4, k21 ꢂ
0.2 sꢁ1, and k31 ꢂ 0.1 sꢁ1 in good agreement with the values
found in vitro, showing no significant alteration of the rate
constants associated with cS2CD photoactivation in living
cells.
^
Figure 1. a) Photoreleased concentrations in C ( ) and D (*) after
illumination for a duration t of a 5.1 mm cS2CD solution at
84ꢂ10ꢁ9 einsteinsꢁ1 light intensity. Solid line: fits of the data from
Equations (34),(37) in the Supporting Information. b) Fluorescence
emission of C ICF ðtÞ (lem =500 nm) over time during illumination at
lexc =365 nm of a 5.1 mm cS2CD solution at various light intensities
(40, 80, 120, and 160ꢂ10ꢁ9 einsteinsꢁ1). Diamonds: experimental
data; solid lines: fits from Equation (49) in the Supporting Informa-
tion. c) Fluorescence emission of D IDF ðtÞ (lem =660 nm) over time
during illumination at lexc,1 =365 nm and lexc,2 =645 nm of a 5.1 mm
cS2CD solution at varying light intensities of lexc,1 (40, 80, 120, and
160ꢂ10ꢁ9 einsteinsꢁ1). Circles: experimental data; solid lines: fits
from Equation (50) in Supporting Information. d) Simulation of the
change in concentration of C (solid line) and D (dashed line) over time
upon continuous illumination of a 5 mm cS2CD solution at
10ꢁ7 einsteinsꢁ1 light intensity, as obtained from Equations (46)–
(48),(51),(52) using the values of ef(365 nm), k21, k22, and k31
extracted from the fits. Solvent: 1:1 (v/v) CH3CN/0.1m Britton–
Robinson buffer pH 8. T=298 K.
Scheme 2, the rate constant k1 is associated with uncaging,
whereas the rate constants k21, k22, k31, and k32 all refer to the
steps of self-immolation (see Supporting Information).
We obtained satisfactory fits of the kinetic data using
f(365 nm) = 8.8 ꢀ 0.9 ꢀ 10ꢁ4 for the cS2CD uncaging quantum
yield, k21 = 0.10 ꢀ 0.02 sꢁ1, k22 < 0.02 sꢁ1, and k31 = 0.11 ꢀ
0.06 sꢁ1. To support our kinetic analysis, we first noted that
the uncaging quantum yield f(365 nm) was in line with the
result calculated from the LC/MS analysis. The values of k21
and k31 were then compared with the values calculated from
the analysis of self-immolation of the model cS1C and cS2D2
spacers (see Supporting Information, Section 2.2.1 and 2.2.2).
A value for k21 of approximately 0.1 sꢁ1 compares fairly well
with the rate constant associated with the first S2D2 self-
immolation step (also approximately 0.1 sꢁ1). In fact, this
result was as expected, because the same moiety D was first
released from both S2CD and S2D2 phenols. We also saw an
agreement between k31 (approximately 0.1 sꢁ1) and the rate
constant for cS1C self-immolation (approximately 0.3 sꢁ1).
Indeed, these rate constants both corresponded to the
Finally, we used confocal microscopy to analyze the final
concentrations of C and D within the cells after uncaging with
365 nm light. Excitation at 405 nm and 633 nm light were
subsequently used to image C and D under optimized
conditions. Figures 2a–c show confocal micrographs of non-
illuminated cS2CD-incubated control cells. Under these
imaging conditions, some auto-fluorescence was observed in
the green emission channel upon exciting at 405 nm (Fig-
ure 2b). In contrast, no significant auto-fluorescence was
observed in the far-red emission channel with 633 nm light
excitation (Figure 2c). Figures 2d,e show confocal micro-
graphs of cS2CD-incubated cells that were exposed to 365 nm
light until a steady state of fluorescence was achieved. We
Angew. Chem. Int. Ed. 2012, 51, 1 – 5
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3
These are not the final page numbers!