New Tools for Molecular Imaging of Redox Metabolism: Development of a Fluorogenic Probe for 3α-Hydroxysteroid Dehydrogenases

Article abstract of DOI:10.1021/ja039799f

A new fluorogenic substrate was developed for 3α-hydroxysteroid dehydrogenases (3α-HSD), including the human enzymes implicated in important physiological functions (androgen deactivation, neurosteroid activation). While ketone 5 is nonfluorescent, the co

Full text of DOI:10.1021/ja039799f

Published on Web 02/06/2004
New Tools for Molecular Imaging of Redox Metabolism: Development of a
Fluorogenic Probe for 3r-Hydroxysteroid Dehydrogenases
Dominic J. Yee, Vojtech Balsanek, and Dalibor Sames*
Department of Chemistry, Columbia UniVersity, 3000 Broadway, New York, New York 10027
Received November 25, 2003; E-mail: sames@chem.columbia.edu
The development of noninvasive methods for the visualization
and quantification of molecular processes in vivo is of central
interest to both basic science and medicine. Recent advances in
photon detection technologies enable new opportunities in optical
imaging of living matter (i.e., higher resolution and sensitivity).1
However, molecular optical imaging requires the generation of
Figure 1. Design of an optical switch based on carbonyl-alcohol redox
“smart” probes or functional dyes2 which would provide an optical
chemistry. EDG ) electron-donating group, EWG ) electron-withdrawing
signal for a particular molecular event (cf., enzymatic activity).3
group.
We herein communicate the key findings of a study that began
with a chemical concept and resulted in the development of a
selective fluorogenic probe for 3R-hydroxysteroid dehydrogenases
(3R-HSD), an important class of enzymes in human physiology.
Oxidoreductases, including alcohol dehydrogenases, play es-
sential roles in maintaining the balance of metabolic energy and
regulating the concentration of critical metabolites, hormones, and
xenobiotics. Redox optical probes must have a built-in mechanism
for coupling the chemical redox event to a change in emission
properties. However, two mechanisms frequently used for construc-
Figure 2. Synthesis of compound arrays based on three fluorophore cores
tion of fluorogenic substrates (e.g., probes for hydrolases), fluo-
(the corresponding alcohols are not shown).
rescence energy transfer (FRET)4 and phenol- or aniline-releasing
reactions,5 are generally not suitable for alcohol dehydrogenase
0.1); and (3) photochemical stability and chemical stability (includ-
probes.6 Consequently, we have formulated a simplified electronic
ing stability to intracellular reductants). Following these strict
argument that guided the design of a ketone-alcohol redox switch.
Many organic fluorophores are based on the “push-pull” structural
feature wherein the electron-donating and electron-withdrawing
groups are electronically connected via an extended π-conjugated
system.7 This class of fluorophores seemed particularly suitable for
the design of redox probes wherein the ketone carbonyl would be
a part of the “push-pull” system. Reduction of the carbonyl group
to an alcohol converts an electron-withdrawing group (and often a
quenching group) to an electron-donating group, resulting in a
profound electronic change of the system, which in turn may lead
to a change in the emission profile (Figure 1).8
The accurate prediction of light emission properties of organic
compounds remains an underdeveloped art, and thus we undertook
a semiempirical approach. An array of compounds was synthesized
according to the design shown in Figure 1, founded on three
aromatic core structures (Figure 2). The ketone moiety was attached
to the core at two positions, either directly or via a linker. The
linker (benzene, alkene, and alkyne) was introduced to explore the
consequences of spatial separation of the ketone and the fluorophore
while maintaining the conjugation between these two components.
Specifically, we were interested in the effect of the length and nature
of the π-conjugation system on both emission properties and the
enzyme activity and selectivity (accessibility of the carbonyl group
to the enzyme active site).
criteria, we found seven fluorogenic probes possessing a suitable
profile; in all cases (with the exception of probe 1), the alcohols
were highly fluorescent while the corresponding ketones showed
only a background level of emission, thus constituting an optical
redox switch (Figure 3A, for spectroscopic data see Supporting
Information).8 We were pleased that the selected candidates
contained three different cores and a variety of linkers, thereby
increasing the structural diversity of the set.
Probes 1-7 were subsequently tested against a collection of
dehydrogenases in the presence of NAD(P)H; the extent of
reduction was assessed by the measurement of fluorescence intensity
at the emission maximum of each probe (Figure 3B). This assay
included enzymes from two major oxidoreductase superfamilies,
the short chain alcohol dehydrogenases (SDR) and the aldo-keto
reductases (AKR), ranging from bacterial to mammalian as well
as human enzymes.
Probe 5 (λem ) 510 nm for the corresponding alcohol) clearly
stood out as it was converted rapidly and selectively by 3R-
hydroxysteroid dehydrogenases (3R-HSD), the bacterial (Pseudomo-
nas) and the rat liver enzymes (Figure 3B).9
No other enzymes examined in this assay catalyzed the reduction
of probe 5. Similarly, both 3R-HSD enzymes demonstrated high
selectivity for 5 among the tested probes. Compound 6 showed
good conversion, however, at a significantly slower rate in compari-
son to that of probe 5 (Figure 3B). Surprisingly, horse liver alcohol
dehydrogenase (HLAD) and Thermoanaerobium brockii alcohol
dehydrogenase (TBAD), both well known for their substrate promi-
scuity,10 were not tolerant of probe 5. In contrast, these two latter
enzymes catalyzed reduction of alkynyl-ketone probes 4 and 7.
Approximately 50 compounds were synthesized (see Supporting
Information) and evaluated in terms of the following physical and
chemical properties in aqueous solution: (1) emission switching
between the oxidized (ketone) and reduced forms (alcohol); (2)
emission wavelength (λem > 430 nm) and quantum yield (Φ >
9
2282
J. AM. CHEM. SOC. 2004, 126, 2282-2283
10.1021/ja039799f CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Figure 4. Kinetic parameters for probe 5 and the physiological substrate
for human 3R-HSD (type 2, AKR1C3).
5 was in fact a far better substrate for this enzyme. This point may
prove significant in future explorations of this probe in living matter.
In summary, this study addressed two key points: (1) ketone-
alcohol redox optical switches with suitable emission profiles
(emission wavelength, quantum yield) may be generated, and (2)
these synthetic compounds may function as substrates for oxi-
doreductase enzymes. Specifically, a new fluorogenic probe was
developed for 3R-HSD enzymes, including a human isozyme that
has been implicated in important physiological functions. On the
basis of its good optical properties and excellent enzyme selectivity,
structure 5 represents an exciting lead for the development of a
redox imaging probe. Future work will focus on determining the
selectivity of probe 5 within the HSD family in vitro, followed in
due course by studies in vivo.
Acknowledgment. We thank Professor Trevor M. Penning and
Vladi Heredia (University of Pennsylvania School of Medicine)
for assistance with enzymology studies and for supplying rat and
human 3R-HSD enzymes. We also thank Dr. J. B. Schwarz
(editorial assistance). This work was generously supported by The
G. Harold & Leila Y. Mathers Charitable Foundation.
Supporting Information Available: Synthetic procedures, spectral
data, excitation and emission profiles for compounds 1-7, and
enzymology assay protocol (PDF). This material is available free of
charge via the Internet at http://pubs.acs.org.
Figure 3. Screening of selected probes (A) against a panel of oxidoreduc-
tases (B). Percentage fluorescence increase after 12 h incubation of 30-50
µM substrate, 100 mM phosphate buffer (pH 7), 250 µM NAD(P)H, and
100 nM enzyme. Substrates 1, 2, 3, and 4 were monitored at λexc ) 340
nm, λem ) 440 nm. Substrates 5, 6, and 7 were monitored at λexc ) 440
nm, λem ) 510 nm. 3HSD, 3R-hydroxysteroid dehydrogenase (PT,
Pseudomonas testosteroni); HLAD, horse liver alcohol dehydrogenase;
TBAD, Thermoanaerobium brockii alcohol dehydrogenase; BS 12HSD,
Bacillus sp. 12R-hydroxysteroid dehydrogenase; ABAD, amyloid-â binding
alcohol dehydrogenase (human); GDH, glycerol dehydrogenase; YADH,
yeast alcohol dehydrogenase; LDH, lactate dehydrogenase.
References
(1) Weissleder, R.; Ntziachristos, V. Nat. Med. 2003, 9, 123-128.
(2) (a) Moreira, R.; Havranek, M.; Sames, D. J. Am. Chem. Soc. 2001, 123,
3927-3931. (b) Chen, C.-A.; Yeh, R.-H.; Lawrence, D. S. J. Am. Chem.
Soc. 2002, 124, 3840-3841.
(3) (a) Boonacker, E.; Van Noorden, C. J. F. J. Histochem. Cytochem. 2001,
49, 1473-1486. (b) Handbook of Fluorescent Probes and Research
Chemicals, Molecular Probes, 9th ed.; Haugland, R. P., Ed.; Molecular
Probes: Eugene, OR, 2002.
(4) Wang, G. T.; Matayoshi, E.; Huffaker, H. J.; Krafft, G. A. Tetrahedron
Lett. 1990, 31, 6493-6496.
(5) (a) Rotman, B.; Zderic, J. A.; Edelstein, M. Proc. Natl. Acad. Sci. U.S.A.
1963, 50, 1-6. (b) Zimmerman, M.; Ashe, B.; Yurewicz, E.; Patel, G.
Anal. Biochem. 1977, 78, 47-51.
The 3R-HSD enzymes are members of the aldo-keto reductase
(AKR) superfamily and play numerous physiological functions. For
example, they are involved in the activation/deactivation of steroid
hormones in various tissues, including prostate (androgen deactiva-
tion) and brain (neurosteroid activation).11
Consequently, we addressed the next key question of whether
the activity of human enzymes could be imaged by probe 5. The
type 2 isozyme of 3R-HSD (AKR 1C3) was selected for this study
because of its important physiological role. To our delight, probe
5 was rapidly converted by this enzyme, and subsequent quantitative
measurements afforded the kinetic parameters (Km ) 2.5 µM, kcat
) 8.2 min-1). Remarkably, when compared to 5R-dihydrotestoster-
one (5R-DHT, Km ) 26 µM, kcat ) 0.25 min-1, Figure 4),11 a likely
physiological substrate in prostate, we found that synthetic probe
(6) Alcohol dehydrogenase probes which require two catalytic steps (oxidation
and â-elimination): Klein, G.; Reymond, J.-L. Bioorg. Med. Chem. Lett.
1998, 8, 1113-1116.
(7) Rettig, W. Angew. Chem., Int. Ed. Engl. 1986, 25, 971-988.
(8) Previous examples of carbonyl-alcohol fluorogenic probes suffered from
short excitation/emission wavelengths in the near UV region. (a)
Wierzchowski, J.; Dafeldecker, W. P.; Holmquist, B.; Vallee, B. L. Anal.
Biochem. 1989, 178, 57-62. (b) List, B.; Barbas, C. F., III; Lerner, R. A.
Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 15351-15355.
(9) Jez, J. M.; Schlegel, B. P.; Penning, T. M. J. Biol. Chem. 1996, 271,
30190-30198.
(10) Fang, J.-M.; Lin, C.-H.; Bradshaw, C. W.; Wong, C.-H. J. Chem. Soc.,
Perkin Trans. 1 1995, 967-978.
(11) Penning, T. M.; Burczynski, M. E.; Jez, J. M.; Hung, C.-F.; Lin, H.-K.;
Ma, H.; Moore, M.; Palackal, N.; Ratnam, K. Biochem. J. 2000, 351,
67-77.
JA039799F
9
J. AM. CHEM. SOC. VOL. 126, NO. 8, 2004 2283