Bisdeoxycoelenterazine derivatives for improvement of bioluminescence resonance energy transfer assays

Article abstract of DOI:10.1021/ja073936h

We report here syntheses of bisdeoxycoelenterazine derivatives that show significantly improved kinetics of the bioluminescent reaction. All derivatives contain a protecting group at the carbonyl group of the imidazopyrazinone moiety that ensures slower light signal decay. The longer lasting light signal offers the possibility of real-time imaging of protein-protein interactions. Copyright

Full text of DOI:10.1021/ja073936h

Published on Web 09/12/2007
Bisdeoxycoelenterazine Derivatives for Improvement of Bioluminescence
Resonance Energy Transfer Assays
Jelena Levi, Abhijit De, Zhen Cheng, and Sanjiv Sam Gambhir*
Molecular Imaging Program at Stanford (MIPS), Departments of Radiology and Bioengineering, Bio-X Program,
Stanford UniVersity, 318 Campus DriVe, E 150, Stanford, California 94305-5427
Received June 14, 2007; E-mail: sgambhir@stanford.edu
Protein-protein interactions are the basis of many important
cellular functions. The ability to noninvasively image these interac-
tions in living organisms provides a unique possibility of studying
important biological processes in intact, minimally perturbed
Scheme 1. Mechanism of Coelenterazine Oxidation and Light
Production
1
systems. Molecular imaging assays based on bioluminescence
resonance energy transfer (BRET) are one of the key strategies for
investigating protein-protein and other interactions in live cells
as well as in living subjects.2 BRET is a phenomenon involving
a transfer of energy, obtained from the oxidation of substrate by
the energy donor, an oxygenase protein, and an energy acceptor, a
fluorescent protein. Typically, BRET systems utilize Renilla
luciferase (RLuc) as an energy donor and green fluorescent protein
-4
(GFP) and its mutants as an energy acceptor. Based on the specific
Renilla substrate utilized, BRET systems are currently divided into
1
2
5
BRET , BRET , and recently introduced eBRET. Of the three, the
the peroxide can be formed and oxidation can take place. The
deprotection of the reaction site can be accomplished by the action
of the cellular enzymes, such as esterases, which makes the
derivatives suitable as live cell substrates.
2
BRET system has the best spectral separation between the donor
and the acceptor and thus higher signal to background ratio than
all possible BRET and eBRET combinations.
superior spectral resolution, the BRET system has not been applied
as broadly and as frequently as BRET , due to the poor quantum
yield and very fast kinetics of oxidation of the substrate utilized,
bisdeoxycoelenterazine (BDC, 1). The fast oxidation of BDC results
in a light signal that decays rapidly, limiting stable signal detection
1
6-8
Despite the
2
We have synthesized four BDC derivatives (Figure 1) with
varying size and type of the groups at the carbonyl of the
imidazopyrazinone moiety. BDC (1) was synthesized following a
1
10
published procedure by coupling 2-amino-3-benzyl-5-phenylpyra-
zine and 1,1-diethoxy-3-phenylacetone. Subsequent reaction with
acetic anhydride, di-tert-butyl dicarbonate, bromomethyl acetate,
and chloromethyl pivalate afforded acetyl bisdeoxycoelenterazine
5
to only a few seconds. We report here BDC derivatives that
improve kinetics of BDC oxidation and consequently offer signifi-
cantly longer lasting light signal. Considering that the half-times
for protein-protein interactions vary from brief to long, this
improvement in signal sustainability greatly expands the utility of
(7),O-Boc-bisdeoxycoelenterazine(8),acetoxymethylbisdeoxycoelentera-
zine (9), and pivaloyloxymethyl bisdeoxycoelenterazine (10).
The kinetics of the bioluminescent reactions were evaluated by
exposing the derivatives to HT1080 cells stably expressing the
2
BRET in real-time protein-protein interaction imaging.
Oxidation of coelenterazine (2), the natural RLuc substrate, can
generate light as a result of both enzymatic and chemical reaction,
phenomena termed bioluminescence and chemiluminescence, re-
spectively. Although the mechanism of coelenterazine oxidation is
thought to be the same (Scheme 1) for both of these reactions,9-11
the kinetics of light production differ quite remarkably. Biolumi-
nescent reactions have very fast kinetics, leading to the production
of intense, exponentially decaying light signal.12,13 On the other
hand, chemiluminescence caused by the reaction of coelenterazine
with aprotic organic solvents in the presence of oxygen and base
results in longer lasting light signal of lower intensity.10 The rate-
limiting step in chemiluminescence reactions seems to be the
formation of a peroxide intermediate (4). Coelenterazine peroxide
has been synthesized and detected at -80 °C by NMR.14 At
temperatures higher than -50 °C, the hydroperoxide intermediate
spontaneously decomposes with emission of light. The decrease in
the rate of the formation of the peroxide intermediate should
therefore result in the decrease of the rate of the bioluminescent
reaction and sustained production of light. We achieve this by
introducing protecting groups at the reaction site, the carbonyl group
of the imidazopyrazinone moiety, that have to be removed before
GFP
2-RLuc8 fusion protein15 and detecting generated light over
the course of 1 h using a cooled charge-coupled device camera
(Figure 2). The luciferase used in this system is the mutant RLuc8
developed in our lab that in the reaction with BDC gives 59-fold
16
higher light output than the native RLuc. Of the four derivatives,
only acetyl-BDC showed fast light signal decay, similar to BDC.
Not surprisingly, the acetyl ester seems to be easily cleaved by
cellular esterases, and the bioluminescent reaction is only slightly
altered. The bulkier tert-butyloxycarbamyl group at the carbonyl
of the imidazopyrazinone moiety results in the derivative that shows
considerably slower kinetics. The peak light emission for O-Boc-
BDC was observed 15 min after exposure to the RLuc8 expressing
cells and remained fairly stable over 1 h. In comparison, at the 15
min time point, the parent BDC lost nearly 75% of its initial light
emission. Bulkiness of the tert-butyl group appears to considerably
slow the enzymatic ester hydrolysis and thus the bioluminescent
reaction, providing a longer lasting light signal. The size of the
protecting group had the same effect on the rate of the biolumi-
nescent reaction in the case of the ether derivatives, acetoxymethyl-
BDC and pivaloyloxymethyl-BDC. Acetoxymethyl group delayed
the emergence of the maximum light signal by 5 min compared to
11900
9
J. AM. CHEM. SOC. 2007, 129, 11900-11901
10.1021/ja073936h CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Figure 1. Synthesized bisdeoxycoelenterazine derivatives: acetyl bis-
deoxycoelenterazine (7), O-Boc-bisdeoxycoelenterazine (8), acetoxymethyl
bisdeoxycoelenterazine (9), pivaloyloxymethyl bisdeoxycoelenterazine (10).
Figure 3. (a) The change in luminescence over time for acetyl bisdeoxy-
coelenterazine (7), O-Boc-bisdeoxycoelenterazine (8), acetoxymethyl bis-
deoxycoelenterazine (9), and pivaloyloxymethyl bisdeoxycoelenterazine
(
10). Error bars represent standard deviation from the average value. (b)
2
Bioluminescence imaging of HT1080 cells expressing GFP -RLuc fusion
protein after exposure to bisdeoxycoelenterazine and its derivatives. Color
bar units are p/s/cm /sr, where p is for photon and sr is for steradian.
2
improved kinetics of the bioluminescent reaction. Our results
indicate that the extent of the effect the protecting group has on
the rate of oxidation depends on the size and type of the protecting
group. Of the four synthesized BDC derivatives, three (8, 9, and
Figure 2. The rate of the bioluminescent reaction for BDC ((), acetyl-
BDC (9), O-Boc-BDC (b), acetoxymethyl-BDC (2), and pivaloyloxy-
methyl-BDC (0) evaluated as the change of the maximum light signal over
time.
2
10) show great promise for improving the existing BRET assays
in terms of light signal sustainability. The longer lasting signal offers
a possibility of real-time imaging of protein-protein interaction in
live cells in combination with unparalleled signal to background
ratio. Additional testing of these in various applications including
in small living subjects should help further characterize their utility.
the parent BDC, after which time the signal slowly decayed with
time. Retardation of the enzymatic ether cleavage caused by the
bulky tert-butyl group in the case of pivaloyloxymethyl-BDC
resulted in the slowest bioluminescent reaction. In terms of the
signal half-life, defined as the time required for the initial light
flux to fall to its half-value, compared to BDC, derivatives 8, 9,
and 10 have much improved characteristics. Compared to only 5
min half-life of the parent BDC, derivative 8 had a half-life of ∼50
min, while half-lives of derivatives 9 and 10 were even longer than
Acknowledgment. We would like to thank the NCI ICMIC P50
CA114747 (S.S.G.) and NCI 5RO1 CA082214 (S.S.G.) for funding
support. J.L. thanks Dr. H. Yao for helpful discussions.
Supporting Information Available: Synthetic procedures and
characterization of all four derivatives. This material is available free
of charge via the Internet at http://pubs.acs.org.
1
h.
The rate of the bioluminescent reaction also depended on the
type of the protecting group. Derivatives carrying ether protecting
groups at the carbonyl of the imidazopyrazinone ring, 9 and 10,
showed slower bioluminescent reactions compared to their ester
counterparts, derivatives 7 and 8.
The carbonyl group protection inevitably affects the intensity of
the light signal of the derivatives. Of the four derivatives, the highest
light signal was observed with acetyl-BDC and the lowest with
pivaloyloxymethyl-BDC (Figure 3a). Although the signal intensity
of pivaloyloxymethyl-BDC never reached the intensity of the signal
of any of the other three BDC derivatives, the only observable signal
References
(
1) Massoud, T. F.; Gambhir, S. S. Genes DeV. 2003, 17, 545-580.
(2) Xu, Y.; Piston, D. W.; Johnson, C. H. Proc. Natl. Acad. Sci. U.S.A. 1999,
96, 151-156.
(
(
3) De, A.; Gambhir, S. S. FASEB J. 2005, 19, 2017-2019.
4) Issad, T.; Blanquart, C.; Gonzales-Yanes, C. Exp. Opin. Ther. Targets
2007, 11, 541-556.
(
5) Pfleger, K. D.; Seeber, R. M.; Eidne, K. A. Nat. Protoc. 2006, 1, 337-
345.
(6) Jensen, A. A.; Hansen, J. L.; Sheikh, S. P.; Brauner-Osborne, H. Eur. J.
Biochem. 2002, 269, 5076-5087.
(
7) Ramsay, D.; Kellett, E.; McVey, M.; Rees, S.; Miligan, G. Biochem. J.
2002, 365, 429-440.
(
(
8) Pfleger, K. D.; Eidne, K. A. Pituitary 2003, 6, 141-151.
9) Ohmiya, Y.; Hirano, T. Chem. Biol. 1996, 3, 337-347.
24 h after exposure to the enzyme came from it (Figure 3b). Clearly,
the decrease in the rate of the bioluminescent reaction is closely
related to the decrease in signal intensity. It is important to point
out here that the lower light signal does not render a BDC derivative
(
10) Adamczyk, M.; Akireddy, S. R.; Johnson, D. D.; Mattingly, P. G.; Pan,
Y.; Reddy, R. E. Tetrahedron 2003, 59, 8129-8142.
(
(
11) Teranishi, K. Bioorg. Chem. 2007, 35, 82-111.
12) Bhaumik, S.; Gambhir, S. S. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 377-
2
inadequate for application in BRET assays. Just as in the case of
382.
(
13) Tricoire, L.; Tsuzuki, K.; Courjean, O.; Gibelin, N.; Bourout, G.; Rossier,
J.; Lambolez, B. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 9500-9505.
14) Teranishi, K.; Ueda, K.; Nakao, H.; Hisamatsu, M.; Yamada, T.
Tetrahedron Lett. 1994, 35, 8181-8184.
Enduren, the coelenterazine h derivative with prolonged lifetime,
signal intensity can be increased by simply using higher concentra-
(
5
tions of the substrate in BRET assays. What makes a derivative
2
(15) De, A.; Loening, A. M.; Gambhir, S. S. Cancer Res. 2007, 67, 7175-
have high utility in BRET applications is the decelerated kinetics
7183.
of its oxidation resulting in sustained emission of light.
In summary, protection of the carbonyl group of the imidazopy-
razinone moiety in bisdeoxycoelenterazine led to derivatives with
(16) Loening, A. M.; Fenn, T. D.; Wu, A. M.; Gambhir, S. S. Protein Eng.
Des. Sel. 2006, 19, 391-400.
JA073936H
J. AM. CHEM. SOC.
9
VOL. 129, NO. 39, 2007 11901