Light-mediated spatial control via photolabile fluorescently quenched peptide cassettes

Article abstract of DOI:10.1021/ja907427p

(Figure Presented) Light-regulatable compounds are finding increasing utility as spatial and temporal probes of biological behavior. An independent measure of successful light-induced structural change is possible when alteration (e.g., activation, deactivation, etc.) of the bioprobe can be directly linked to a fluorescent readout. We have identified a series of photolabile fluorescently quenched cassettes that display extraordinarily large fluorescence enhancements upon photolysis. A pair of cassettes has been inserted into mitochondrial localization sequences to assess an organelle-targeted light-mediated release strategy for controlling biological activity. The peptide constructs are readily absorbed by mitochondria and subsequently can be cleaved in a light-dependent fashion as assessed by the predicted changes in absorbance and fluorescence.

Full text of DOI:10.1021/ja907427p

Published on Web 01/14/2010
Light-Mediated Spatial Control via Photolabile Fluorescently Quenched
Peptide Cassettes
Hsien-Ming Lee, Melanie A. Priestman, and David S. Lawrence*
Department of Chemistry, DiVision of Medicinal Chemistry and Natural Products, and Department of
Pharmacology, UniVersity of North Carolina, Chapel Hill, North Carolina 27599
Received September 2, 2009; E-mail: lawrencd@email.unc.edu
Chart 1. Quenched Fluorescent Cassette Library (1 and 2) Derived
from Fluorophores 3 and 4, Photolabile Linkers 5 and 6, and
Quenchers 7-10
Light-regulatable compounds are finding increasing utility as
spatial and temporal probes of biological behavior.1 Although these
compounds are presumed to undergo photolysis (or photoisomer-
ization) in response to illumination, the observed biological response
is typically taken as the only evidence that a photoinitiated event
has transpired. An independent measure of successful light-induced
structural change is possible when alteration (e.g., activation,
deactivation, etc.) of the bioprobe can be directly linked to a
fluorescent readout. Furthermore, such a design element not only
would allow one to quantify the amount of photoaltered bioprobe
that had been generated but also could be used to visualize the
intracellular localization of the bioprobe after illumination. To the
best of our knowledge, Muir and co-workers2 were the first to put
this concept into practice. These investigators prepared a protein
construct (Smad2) via expressed protein ligation that generates a
26-fold fluorescence enhancement upon photolysis. Subsequently,
Kutateladze and co-workers3 described a thioxanthone-based system
that furnishes an up to 17-fold enhancement in response to
photocleavage.
Unlike enzyme-catalyzed reactions, in which readouts (e.g.,
fluorescence) are continuously amplified as a function of time,
photolysis produces a fixed amount of product. A large fluorescence
change in response to illumination reduces the amount of bioprobe
required for visualization, which in turn reduces the likelihood of
undesired “observer effect”-induced alterations in cellular biochem-
istry.4 A case in point is the mitochondrion, the so-called energy
factory of the cell, which contains several suborganelle compart-
ments that can be targeted using specific amino acid sequences.
However, because of the small size of these organelles, targeting-
sequence oversaturation of these compartments is possible if large
quantities are required for visualization. With this concern in mind,
we initiated a program to evaluate an array of structural motifs in
order to identify quenched fluorescent cassettes that furnish a large
fluorescent response upon photolysis.
A library of 32 modularly designed tripeptides having the general
structures 1 and 2 was prepared (Chart 1). Fluorescein (3) and
tetramethylrhodamine (4; TAMRA) analogues were evaluated as
the fluorophore component since they are commonly employed in
cell-based studies. The other variables constituting the library
included two photolinkers (5 and 6), four different quenchers (7-9),
and two sequences [with the fluorophore at the C-terminus (1) or
the N-terminus (2)]. The library was prepared via solid-phase
synthesis followed by stepwise modification of the Lys and Cys
side chains with the fluorophore and quencher, respectively. All of
the library members were HPLC-purified (Figure S-1 in the
Supporting Information) and subsequently characterized. Fluores-
cence readings were obtained prior to and following photolysis
(Figures S-2-S-6). The fluorescein- and TAMRA-derivatized
library members are listed in Table 1. Our leads [Ac-Lys(Fl)-photo-
linker-Cys(Q)-amide, where Lys(Fl) ) 4, photolinker ) 5,
Cys(Q) ) 7 (4-5-7), 8 (4-5-8)] display a greater than 300-fold
fluorescent enhancement upon photolysis.
The most impressive light-induced fluorescence enhancements
appear to be a consequence of two structural features: First, 7 and
8 quench the fluorescence of fluorescein and TAMRA more deeply
than do 9 or 10 [which may be a consequence of efficient
Table 1. Light-Induced Fluorescence Changes (in italics) of the
Quenched Fluorescent Cassette Library (see Chart 1 for
Structures); The Fluorescein (Upper Half of Table) and TAMRA
(Lower Half of Table) Cassettes Are Segregated (See Figures S-4
and S-5)
7-5-3
25 ( 1
7-6-3
11 ( 1
3-5-7
75 ( 1
3-6-7
75 ( 7
8-5-3
6.5 ( 1
8-6-3
22 ( 1
3-5-8
48 ( 15
3-6-8
57 ( 13
9-5-3
3.0 ( 0.2
9-6-3
10-5-3
4.0 ( 0.3
10-6-3
14 ( 2
3-5-10
35 ( 1
2.8 ( 0.1
3-5-9
17 ( 3
3-6-9
3-6-10
12 ( 1
6.0 ( 0.1
7-5-4
160 ( 4
7-6-4
55 ( 10
4-5-7
340 ( 10
4-6-7
100 ( 20
8-5-4
290 ( 20
8-6-4
56 ( 2
4-5-8
360 ( 10
4-6-8
120 ( 10
9-5-4
32 ( 2
9-6-4
31 ( 1
4-5-9
71 ( 7
4-6-9
63 ( 2
10-5-4
8.0 ( 0.1
10-6-4
28 ( 1
4-5-10
86 ( 2
4-6-10
30 ( 3
9
1446 J. AM. CHEM. SOC. 2010, 132, 1446–1447
10.1021/ja907427p  2010 American Chemical Society

C O M M U N I C A T I O N S
Figure 1. Mitochondrial-targeted light-mediated release strategy. Photolysis
releases the quencher (and any associated biochemically active agent “X”)
from the mitochondrial surface into the cytoplasm.
fluorescence resonance energy transfer (FRET) and collisional
quenching] and thus deliver a larger fluorescent response. Second,
nitrobenzyl derivatives are modest quenchers of fluorescence as
well.5 Cassettes in which photolysis detaches the nitrobenzyl-based
photolinkers from the fluorophore-appended segment (e.g., 4-5-8)
produce larger fluorescence changes than the corresponding cas-
settes in which the nitrobenzyl linker remains associated with the
fluorophore (e.g., 8-5-4). These large fluorescence changes can
be easily observed using a handheld UV-vis lamp (see the movie
in the Supporting Information), which both photolyzes the linker
and excites the fluorophore.
Figure 2. HeLa cells exposed to CPMLS-4-5-7 and mitotracker FarRed:
row A (unphotolyzed), row B (photolyzed); column 1 (Cy 3 window, green,
TAMRA-labeled peptide), column 2 (Cy 5.5 window, red, mitotracker
FarRed), column 3 (merged columns 1 and 2, orange signifies regions of
overlap, Pearson coefficient 0.88 ( 0.03); green circle (laser focus).
S-12 and S-13) with little remaining in the supernatant, as assessed
by visual inspection as well as absorbance and fluorescence spectros-
copy. Photolysis produces a sharp increase in supernatant absorbance,
consistent with the release of the colored fluorescence quencher 8 from
the mitochondria. Mitochondrial fluorescence quantitation was assessed
via flow cytometry,9 which revealed a light-mediated 13.4-fold
fluorescence enhancement (Figure S-13).
Unfortunately, the Trp-induced dim fluorescence of photolyzed
MLS-PEG-4-5-8 precluded single-cell imaging experiments.
Consequently, we synthesized alternative constructs that utilized the
cell-permeable MLS (CPMLS)10 Ac-Fx-r-Fx-K-Fx-r-Fx-K, where Fx
) cyclohexyl-Ala and r ) D-Arg); CPMLS-4-5-7 furnished the
largest light-induced fluorescence change (51-fold; Figure S-8).
Exposure of HeLa cells to CPMLS-4-5-7 and subsequent photolysis
produced an (11.8 ( 0.7)-fold fluorescence increase and subcellular
mitochondrial localization as assessed by overlap with mitotracker
FarRed (Figure 2). In contrast, photolyzed 4-5-8 gave a diffuse
pattern similar to TAMRA itself (Figure S-16).
We have identified a series of photolabile fluorescently quenched
cassettes that display large fluorescence enhancements upon pho-
tolysis. Cassette-containing MLS peptides are mitochondrially
absorbed and photolyzed in the expected fashion. These reagents
are under evaluation as the basis for a new strategy for spatial
manipulation of intracellular biochemical activity.
The construction of “caged” compounds commonly relies upon
transformation of a biologically active species into an inert
derivative via covalent modification of an essential functional group
with a light-sensitive moiety.1 However, direct modification of a
single key site for complete biological caging purposes is not always
feasible. It occurred to us that an alternative approach for
manipulating activity would be light-driven spatial control of the
cellular distribution of the biological entity. For example, mito-
chondrial localization sequences (MLSs), as well as related species,
could be used to deliver activators (or inhibitors) to mitochondria,
thereby triggering (or inhibiting) an organelle-specific biochemical
cascade.6 If the bioactive species were linked to the MLS via a
photocleavable moiety, then photolysis could potentially be used
to switch off mitochondrial-specific activity via release of the active
reagent from the MLS, leading to its subsequent dilution in the
cytoplasm (Figure 1). We used cassettes from Table 1 to examine
this “organelle targeted/light-mediated release strategy”.
Acknowledgment. We thank the NIH (CA79954) for financial
support.
Supporting Information Available: Details of the quenched
fluorophore cassette and MLS peptide syntheses and the photolysis,
mitochondrial, and cellular studies; movie (QT) showing fluorescence
changes. This material is available free of charge via the Internet at
http://pubs.acs.org.
References
The peptide MLALLGWWWFFSRKK-PEG-4-5-8-amide
(“MLS-PEG-4-5-8”), where the italicized component is based
on an MLS that targets the mitochondrial outer membrane7 and
PEG ) -NHCH2(CH2CH2O)3(CH2)3NHCO-CH2OCH2CO-, was
prepared via solid-phase peptide synthesis. Photolysis of
MLS-PEG-4-5-8 splits the quencher (8) from the TAMRA (4)-
appended peptide, which furnishes an 11-fold fluorescence enhance-
ment (Figure S-7). The latter is dramatically reduced from the
fluorescence yield observed for the 4-5-8 cassette itself, which
we ascribe to the presence of the MLS Trp (a known fluorescence
quencher8) triplet. The mauve-colored MLS-PEG-4-5-8 is
rapidly taken up by isolated bovine heart mitochondria (Figures
(1) (a) Lee, H. M.; Larson, D. R.; Lawrence, D. S. ACS Chem. Biol. 2009, 4,
409. (b) Mayer, G.; Heckel, A. Angew. Chem., Int. Ed. 2006, 45, 4900.
(2) Pellois, J. P.; Hahn, M. E.; Muir, T. W. J. Am. Chem. Soc. 2004, 126,
7170.
(3) Majjigapu, J. R.; Kurchan, A. N.; Kottani, R.; Gustafson, T. P.; Kutateladze,
A. G. J. Am. Chem. Soc. 2005, 127, 12458.
(4) Sharma, V.; Lawrence, D. S. Angew. Chem., Int. Ed. 2009, 48, 7290.
(5) Carmel, A.; Keller, E.; Yaron, A. Eur. J. Biochem. 1977, 73, 617.
(6) Yousif, L. M.; Stewart, K. M.; Kelley, S. O. ChemBioChem 2009, 10, 1939.
(7) Huang, L. J.; Wang, L.; Ma, Y.; Durick, K.; Perkins, G.; Deerinck, T. J.;
Ellisman, M. H.; Taylor, S. S. J. Cell Biol. 1999, 145, 951.
(8) Doose, S.; Neuweiler, H.; Sauer, M. ChemPhysChem 2005, 6, 2277.
(9) Fuller, K. M.; Arriaga, E. A. Curr. Opin. Biotechnol. 2003, 14, 35.
(10) Horton, K. L.; Stewart, K. M.; Fonseca, S. B.; Guo, Q.; Kelley, S. O. Chem.
Biol. 2008, 15, 315.
JA907427P
9
J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010 1447