Diazotrifluoropropionamido-containing prenylcysteines: Syntheses and applications for studying isoprenoid-protein interactions

Article abstract of DOI:10.1021/ol026752a

Photoaffinity-labeled prenylcysteines (1 and 2) incorporating a diazotrifluoropropionamide-based photophore have been prepared. Photolyses of 2 in the presence of RhoGDI, a protein that interacts with prenylated proteins, and prenylcysteine-containing competitors demonstrate the effectiveness of this photoaffinity-labeled analogue as a tool for studying isoprenoid binding sites.

Full text of DOI:10.1021/ol026752a

ORGANIC
LETTERS
2003
Vol. 5, No. 5
609-612
Diazotrifluoropropionamido-Containing
Prenylcysteines: Syntheses and
Applications for Studying
Isoprenoid−Protein Interactions
Tamara A. Kale and Mark D. Distefano*
Department of Chemistry, UniVersity of Minnesota, Minneapolis, Minnesota 55455
distefan@chem.umn.edu
Received August 19, 2002
ABSTRACT
Photoaffinity-labeled prenylcysteines (1 and 2) incorporating a diazotrifluoropropionamide-based photophore have been prepared. Photolyses
of 2 in the presence of RhoGDI, a protein that interacts with prenylated proteins, and prenylcysteine-containing competitors demonstrate the
effectiveness of this photoaffinity-labeled analogue as a tool for studying isoprenoid binding sites.
Protein prenylation, whereby certain cysteine residues are
appended with one or more isoprenoid units, has been
extensively studied in the past 15 years due to the role
prenylated proteins play in the development of cancer.
Considerable progress has been made in developing prenyl-
ation inhibitors for chemotherapeutic purposes.1 However,
an alternative to this mode of attack could lie in the disruption
of prenylated protein-protein interactions. Isoprenoid bind-
ing sites in several proteins that interact with prenylated
proteins have been either detected or suggested. Interruption
of this type of isoprenoid-protein interaction may provide
an alternative, and possibly more favorable, target of clinical
interest.2,3 Prenylated proteins have numerous opportunities
to interact with other proteins, both cytosolic and membrane
bound.4,5 Various methodologies have been employed to
study the way in which prenylated proteins interact with other
proteins at the molecular level, including NMR, X-ray
crystallography, and photoaffinity labeling. Photoaffinity
labeled probes can lead to the identification of specific amino
acid residues involved in recognition and/or binding events.
Along with others,6-8 our laboratory has exploited this
technique, incorporating photoactivatable benzophenone and
diazotrifluoropropanoyl moieties into either prenyl diphos-
phates9-13 or prenylcysteines14-16 and then photochemically
cross-linking these compounds to proteins of interest (see
Figure 1).
The benzophenone unit has proved to be a desirable photo-
affinity labeling reagent in the field of protein prenylation,
in part due to the structural similarities between the photo-
(6) Marecak, D. M.; Horiuchi, Y.; Arai, H.; Shimonaga, M.; Maki, Y.;
Koyama, T.; Ogura, K.; Prestwich, G. D. Bioorg. Med. Chem. Lett. 1997,
7, 1973-1978.
(7) Yokoyama, K.; McGeady, P.; Gelb, M. H. Biochemistry 1995, 34,
1344-1354.
(8) Omer, C. A.; Kral, A. M.; Diehl, R. E.; Prendergast, G. C.; Powers,
S.; Allen, C. M.; Gibbs, J. B.; Kohl, N. E. Biochemistry 1993, 32, 5167-
5176.
(9) Turek, T. C.; Gaon, I.; Distefano, M. D.; Strickland, C. L. J. Org.
Chem. 2001, 66, 3253-3264.
(10) Turek, T. C.; Gaon, I.; Gamache, D.; Distefano, M. D. Bioorg. Med.
Chem. Lett. 1997, 7, 2125-2130.
* To whom correspondence should be addressed. Phone: (612) 624-
0544. Fax: (612) 626-7541.
(11) Gaon, I.; Turek, T. C.; Distefano, M. D. Tetrahedron Lett. 1996,
37, 8833-8836.
(1) Sebti, S. M.; Hamilton, A. D. Oncogene 2000, 19, 6584-6593.
(2) Cox, A. D.; Der, C. J. Curr. Opin. Cell. Biol. 1992, 4, 1008-1016.
(3) Marshall, C. J. Science 1993, 259, 1865-1866.
(4) Zhang, F. L.; Casey, P. J. Annu. ReV. Biochem. 1996, 65, 241-269.
(5) Geyer, M.; Wittinghofer, A. Curr. Opin. Struct. Biol. 1997, 7, 786-
792.
(12) Edelstein, R. L. Ph.D. Thesis, University of Minnesota, Minneapolis,
Minnesota, 1997.
(13) Edelstein, R. L.; Distefano, M. D. Biochem. Biophys. Res. Commun.
1997, 235, 377-382.
(14) Kale, T. A.; Raab, C.; Yu, N.; Dean, D. C.; Distefano, M. D. J.
Am. Chem. Soc. 2001, 123, 4373-4381.
10.1021/ol026752a CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/14/2003

that would proceed via intermediate 4.21 Unfortunately, either
cleavage of the phthalimide protecting group or removal of
the cysteinyl methyl ester resulted in the formation of several
degradation products.
As an alternative source for the amine nitrogen, the
azidation method employed by Thompson et al.22 was found
to be effective (83% yield) whereby the hydroxyl group of
5 (Scheme 1), a previously reported compound,10 was directly
converted to the azide (6). However, regioisomers resulted
in each attempt to introduce an azide at the C-8 position (6a
and 6b).23,24
Azidations of geraniol and farnesol, using a variety of
conditions, including palladium-catalyzed chemistry and
Mitsunobu-type substitution employing zinc azide,22,23,25-27
also resulted in two sigmatropically rearranged products in
an invariably equimolar ratio. Furthermore, isomers persisted
during the syntheses of our prenylcysteine analogues, with
each new compound being formed as an inseparable mixture
(see Supporting Information). This is not altogether surprising
since allyl azides are known to undergo 1,3 thermal re-
arrangements at room temperature.28 However, this compli-
cation was not seen as an impediment: in principle, both
isomers of 1 and 2 could be photoaffinity labeling reagents
for isoprenoid binding sites, although the targeted isomers,
derived from the primary azides, would most likely be the
more effective cross-linkers since the other isomers (obtained
from the secondary azides) are less structurally similar to
isoprenoid units.
Figure 1. Examples of benzophenone- and diazotrifluoropro-
panoyl-containing mimics of prenylcysteines.
phore and an isoprene unit. Despite its successes, there are
some potential drawbacks to using this reactive group: the
lack of precise steric complementarity and the relatively long
irradiation times that are often needed for optimal cross-
linking emphasize the benefits of having a different photo-
active structure.17 For these reasons, the diazotrifluoro-
propanoyl (DATFP) group may prove to be a useful alterna-
tive. Recent work by Quellhorst et al., who showed that a
DATFP-containing isoprenoid could be enzymatically in-
corporated into Rab5 and used to photolabel GDI-1, validates
the use of this photophore as a probe for isoprenoid binding
sites.18 Prenylcysteine analogues such as N-acetyl farnesyl-
cysteine (AFC) and N-acetyl geranylgeranylcysteine (AGGC)
have been used to examine prenylated protein-protein
behavior because they mimic the prenylated C-terminal
cysteine residue of prenylated proteins.19 Here, we describe
the synthesis and cross-linking properties of a new class of
small molecule probes that incorporate a DATFP photophore
into prenylated cysteine methyl esters 1a and 2a.20 To confer
greater stability on the probe, we elected to incorporate an
amide linkage between the DATFP group and the prenyl
moiety instead of the ester linkage that has been used in most
DATFP-containing isoprenoid analogues; this change sig-
nificantly complicated the syntheses of the target compounds.
To commence with the synthesis of 1a, we initially
pursued the use of a phthalimide for introduction of the
DATFP amide nitrogen due to the success of this protecting
group in the preparation of amide-linked benzophenone- and
DATFP-containing prenyl diphosphates.10,12 Compound 3
was prepared on the basis of previous work12 in hopes that
halogenation of the C-1 alcohol would lead to a species
primed to alkylate a cysteinyl thiol. Unfortunately, prepara-
tion of a brominated or chlorinated intermediate was unsuc-
cessful. The presence of the secondary amide linkage
(although not related tertiary amides) appears to cause several
side reactions to occur.10 Thus, a new route was conceived
Thus, the THP protecting group of compounds 6a and 6b,
as a mixture, was removed, and the resulting C-1 position
was brominated (7). Using acidic conditions,29 the thiol of
(15) Kale, T. A.; Raab, C.; Yu, N.; Aquino, E.; Dean, D. C. J. Labelled
Compds. Radiopharm. 2003, 46, 29-54.
(16) Kale, T. A.; Turek, T. C.; Chang, V.; Gautam, N.; Distefano, M.
D. Methods Enzymol. 2002, 344, 245-258.
(17) Dorma´n, G.; Prestwich, G. D. Biochemistry 1994, 33, 5661-5673.
(18) Quellhorst, G. J., Jr.; Allen, C. M.; Wessling-Resnick, M. J. Biol.
Chem. 2001, 276, 40727-40733.
(19) Mondal, M. S.; Wang, Z.; Seeds, A. M.; Rando, R. R. Biochemistry
2000, 39, 406-412.
(20) Syntheses of 1a and 2a produced inseparable mixtures containing
regioisomers 1b and 2b. References to 1 and 2 (and intermediates in the
syntheses) in the text refer to mixtures of these compounds. The presence
1
and ratio of these isomers was relatively easy to determine via H NMR.
(21) Both cysteine methyl esters and their free acid counterparts are
desirable prenylcysteine analogues, as not all proteins are methylesterified.
(22) Thompson, A. S.; Humphrey, G. R.; DeMarco, A. M.; Mathre, D.
J.; Grabowski, E. J. J. J. Org. Chem. 1993, 58, 5886-5888.
(23) Murahashi, S.-I.; Taniguchi, Y.; Imada, Y.; Tanigawa, Y. J. Org.
Chem. 1989, 54, 3292-3303.
(24) Compound 6b and subsequent products therefrom are, in reality,
each mixtures of two diastereomers due to the chirality at C-6.
(25) Viaud, M. C.; Rollin, P. Synthesis 1990, 130-133.
(26) Safi, M.; Fahrang, R.; Sinou, D. Tetrahedron Lett. 1990, 31, 527-
530.
(27) Trost, B. M.; Pulley, S. R. Tetrahedron Lett. 1995, 36, 8737-8740.
(28) Gagneux, A.; Winstein, S.; Young, W. G. J. Am. Chem. Soc. 1969,
82, 5956-5957.
610
Org. Lett., Vol. 5, No. 5, 2003

Scheme 1a
a Reagents and conditions: (a) (PhO)2PON3, DBU, toluene; (b) PPTS, EtOH; (c) PPh3 (polymer supported), CBr4, CH2Cl2; (d) N-acetyl
cysteine methyl ester, Zn(OAc)2, 2:1:1 DMF/butanol/0.025% TFA; (e) PPh3 (polymer supported), THF/H2O. Isomers are present for 7 and
8 as well as intermediates leading up to 1; the undesired isomers were omitted for clarity.
N-acetyl cysteine methyl ester was alkylated by 7, followed
by reduction of the azide using polymer-supported PPh3.30
This resin-bound reagent was easily removed by filtration,
and the free amine was then acylated using diazotrifluoro-
propionyl chloride (11) to produce compound 1a and its
isomer 1b. Fortuitously, the ratio of isomers ranged from
2.5 to 4:1 in favor of 1a as PPh3 appeared to preferentially
reduce the primary azide in 8.
focus was to incorporate a tag to trace any protein products
cross-linked by the DATFP group following irradiation, as
well as provide the means to eventually identify specific
cross-linked amino acids. Biotin was chosen because sub-
sequent blotting to a nitrocellulose membrane and treatment
with HRP-conjugated streptavidin would allow the visualiza-
tion of a chemiluminescent signal signifying cross-linked
protein.33,34 Scheme 2 shows the synthesis of biotinylated
prenylcysteine 2, starting from 8.
Triazolone formation is sometimes a competing problem
with the preparation of DATFP-containing compounds.31,32
Thus, 1H NMR, IR, and UV techniques were used to verify
Alkylation of cysteine methyl ester using bromide 8
proceeded efficiently using DIEA in DMF to produce 12,35
which was then biotinylated. Reduction of the azide with
PPh3 and incorporation of the DATFP moiety proceeded in
a fashion similar to the generation of 1 to give 2.36 1H NMR
analysis indicated that the primary DATFP isomer (2a) was
formed in a 7:1 ratio over the secondary isomer (2b).
Triazolone formation was again avoided in the synthesis of
2 as evidenced by 1H and COSY NMR, IR, and UV analyses,
all of which support the assigned structure of 2.
1
the production of 1a and its isomer 1b. H COSY NMR of
1 showed coupling between key protons of 1a (at C-8 and
N-9, Scheme 1) as well as its isomer 1b (at C-6 and its
neighboring amide proton), suggesting the presence of the
intact DATFP moiety. This was corroborated by IR and UV
analyses that further indicated the presence of the DATFP
group.
Once a route had been established for the synthesis of
DATFP-containing, N-acetylated prenylcysteine 1, our next
Scheme 2a
a Reagents and conditions: (a) cysteine methyl ester hydrochloride, DIEA, DMF; (b) biotin-XX, DIEA, DMF; (c) PPh3 (polymer
supported), THF/H2O. Isomers are present at each stage of this synthesis; undesired isomers were omitted for clarity, excluding 2.
Org. Lett., Vol. 5, No. 5, 2003
611

within the isoprenoid binding site of RhoGDI: Ac-KKSRRC
(S-geranylgeranyl) (termed CR6-gg). The cross-linking signal
is decreased by roughly 75% in the presence of 5 µM CR6-
gg, and nearly 80% in the presence of 50 µM prenylated
peptide; notable competition was also observed using AFC
as a competitor. These results imply that photoprobe 2 and
CR6-gg are competing for the same isoprenoid binding site,
thus suggesting that 2 is an effective prenylcysteine mimic.
Direct comparison between DATFP-based 2 reported here
and earlier benzophenone-based probes is difficult because
the methods used for detection were different (biotin/
streptavidin versus radioactivity, respectively). However,
using a compound analogous to 2 that retained the biotin
reporter but replaced the DATFP moiety with a benzo-
phenone group made a comparison possible. Irradiation of
2 and its benzophenone-containing counterpart with RhoGDI
at comparable concentrations suggests that 2 may be a
somewhat more efficient cross-linking agent. Moreover, 2
appears to be the more specific reagent since its photolabeling
of RhoGDI can be significantly attenuated by the inclusion
of competitors such as CR6-gg and AFC. In contrast, cross-
linking by the benzophenone homologue of 2 is relatively
unaffected by the addition of these competitors at comparable
concentrations.
Figure 2. Autoradiographic analysis of reaction products detected
by SDS-PAGE, blotting, and treatment with HRP-conjugated
streptavidin following photolabeling of RhoGDI by 2 alone and in
the presence of various potential competitors. Reactions, irradiated
for 2 min, were performed using 4.0 µM RhoGDI. Lane assign-
ments: (1) 500 nM 2; (2) 200 nM 2; (3) 200 nM 2, 5 µM CR6-gg;
(4) 200 nM 2, 50 µM CR6-gg.
To examine the utility of photoprobe 2, we chose RhoGDI
for photolysis experiments because this cytosolic protein is
known to interact with geranylgeranylated Rho proteins, and
has been shown to possess an isoprenoid binding site.37
Irradiating RhoGDI under various conditions established that
2 min of irradiation in our system resulted in no detectable
protein degradation, as judged by SDS-PAGE. RhoGDI (4.0
µM) was then irradiated in the presence of 2 at multiple
concentrations, and the reactions were first subjected to SDS-
PAGE, blotted to nitrocellulose and treated with HRP-
conjugated streptavidin. Cross-linking by 2 was detected in
concentrations ranging from 200 nM to 2.0 µM, and a
histogram representing a portion of the resultant chemi-
luminescent signals associated with the blots, as detected by
autoradiography, is presented in Figure 2. Competition
experiments were performed using a prenylated peptide that
not only mimics the C-terminal residues of Rho proteins but
whose farnesylated counterpart is known to affect residues
In addition to protein prenylation processing enzymes,
there are numerous effector, activator, and inhibitory proteins
that contribute to the regulation of prenylated Ras and Ras-
like proteins, virtually all of which have the potential of
possessing isoprenoid binding sites. Discovering the molec-
ular basis for prenylated protein recognition is a necessary
step in understanding how these proteins, and their possible
oncoprotein isoforms, operate. Small molecules such as 1
and 2, acting as prenylated protein mimics, offer a means to
access these details. Since many putative prenyl-binding
proteins may be membrane bound, photoaffinity labeling
reagents may prove to be superior to other methods such as
X-ray crystallography and NMR for elucidating such struc-
tural information. Greater specificity of photoprobe-protein
interactions may likely occur by using photoactivatable
peptides, and efforts are underway to develop such agents.
The simple compounds presented here, however, appear to
be well-suited for providing initial guidance in studying
proteins that interact with prenylated proteins.
(29) Xue, C.-B.; Becker, J. M.; Naider, F. Tetrahedron Lett. 1992, 33,
1435-1438.
(30) Bernard, M.; Ford, W. T. J. Org. Chem. 1983, 48, 326-332.
(31) Hahn, K. M.; Hastie, S. B.; Sundberg, R. J. Photochem. Photobiol.
1992, 55, 17-27.
(32) Theodore, L. J.; Nelson, W. L.; Dave, B.; Giacomini, J. C. J. Med.
Chem. 1990, 33, 873-877.
(33) Willchek, M., Bayer, E. A., Eds. AVidin-Biotin Technology;
Acknowledgment. The authors thank Dr. George
O’Doherty for invaluable advice during the preparations of
these compounds. This work was supported by a grant from
the National Institutes of Health (GM58442).
Academic Press: San Diego, 1990; Vol. 184.
(34) Biotinylated prenylated peptides have been prepared via solid-phase
synthesis, where biotin-avidin chemistry was utilized in subsequent
enzymatic assays (Dolence, E. K.; Dolence, J. M.; Poulter, C. D.
Bioconjugate Chem. 2001, 12, 35-43. Liu, L.; Jang, G.-F.; Farnsworth,
C.; Yokoyama, K.; Glomset, J. A.; Gelb, M. H. Methods Enzymol. 1995,
250, 189-206).
Supporting Information Available: Product syntheses
and characterization as well as a description of the prepara-
tion of acyl chloride 11, experimental procedures for
photolysis experiments, and details of the comparison
between 2 and its benzophenone-containing homologue. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(35) Yang, C.-C.; Marlowe, C. K.; Kania, R. J. Am. Chem. Soc. 1991,
113, 3177-3178.
(36) It should be noted that instead of biotinylation of the amine-
containing prenylcysteine compound 12 in Scheme 2, attempts were made
to synthesize N-biotinylated cysteine methyl ester and then alkylate its free
thiol with bromide 8. This method did not give reproducible results. The
potential for oxidation of the cysteinyl sulfhydryl group as well as disulfide
formation complicated this reaction. The method shown in Scheme 2 is
considerably more reliable.
(37) Hoffman, G. R.; Nassar, N.; Cerione, R. A. Cell 2000, 100, 345-
356.
OL026752A
612
Org. Lett., Vol. 5, No. 5, 2003