Synthesis of desthio prenylcysteine analogs: Sulfur is important for biological activity

Article abstract of DOI:10.1016/j.bmcl.2005.07.075

N-Acetyl-S-farnesyl cysteine (AFC) is the minimal synthetic substrate for the enzyme Icmt, which methylates prenylated proteins. The desthio-AFC isostere 2 has been synthesized in racemic form. This analog was not an Icmt substrate, but instead a weak inh

Full text of DOI:10.1016/j.bmcl.2005.07.075

Bioorganic & Medicinal Chemistry Letters 15 (2005) 5080–5083
Synthesis of desthio prenylcysteine analogs: Sulfur is important
for biological activity
Brian S. Henriksen,a Jessica L. Anderson,b Christine A. Hrycynab and Richard A. Gibbsa,*
aDepartment of Medicinal Chemistry and Molecular Pharmacology and the Purdue Cancer Center, Purdue University,
West Lafayette, IN 47907, USA
bDepartment of Chemistry and the Purdue Cancer Center, Purdue University, West Lafayette, IN 47907, USA
Received 27 May 2005; revised 23 July 2005; accepted 25 July 2005
Abstract—N-Acetyl-S-farnesyl cysteine (AFC) is the minimal synthetic substrate for the enzyme Icmt, which methylates prenylated
proteins. The desthio-AFC isostere 2 has been synthesized in racemic form. This analog was not an Icmt substrate, but instead a
weak inhibitor with an IC50 of ꢀ325 lM.
Ó 2005 Elsevier Ltd. All rights reserved.
Biological and chemical interest in protein lipidation1
CO2CH3
has lagged behind that in protein phosphorylation and
glycosylation. As we learn more about the proteome, it
is becoming clear that protein prenylation is a critically
important post-translational modification.1,2 It has
recently been estimated that 120 human proteins are
farnesylated or geranylgeranylated, as shown in Figure
1.3,4 Protein prenylation is now an issue of significant sci-
entific and medical importance, due in large part to the
development and potential use of farnesyltransferase
inhibitors (FTIs) as anti-cancer agents.5,6 This medical
relevance has led to increased interest in other aspects
of the chemical biology of protein prenylation, including
the synthesis of prenylated peptides as probes of the
function of protein prenylation,7,8 and the investigation
of the enzymes Rce1 and Icmt,4 which are responsible
for the proteolytic processing and C-terminal methyla-
tion of prenylated proteins. Inhibitors of Icmt are of
particular interest, due to the recent report by Casey
and co-workers that a small molecule Icmt inhibitor
(cysmethylnil) exhibits significant in vitro antitumor
activity.9 This report prompts us to report some of our
own recent work in this area.10
S
S
NH
farnesylated
protein
O
O
CO2CH3
NH
geranylgeranylated
protein
-
CO2
NH
S
1
AFC
H3C
O
O
-
CO2
H3C
NH
2
"all-carbon AFC"
-
CO2
H3C
NH
3
A key characteristic of prenylated peptides and proteins
is the biologically rare allyl sulfide moiety. In this paper,
N-acetyl-farnesylglycine
O
Figure 1.
Keywords: Isoprenoids; Methyltransferase; Enzyme inhibitors; Amino
acid analogs.
we report the synthesis of a racemic Ôall-carbonÕ analog
of the minimal Icmt substrate N-acetyl-S-farnesylcys-
teine (1, AFC),11 where the sulfur is replaced with a
*
Corresponding author. Tel.: +1 765-494-1456; fax: +1 765-494-1414;
e-mail: rag@pharmacy.purdue.edu
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.07.075

B. S. Henriksen et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5080–5083
5081
methylene (2). It is hypothesized that this replacement
would lead to a compound that would have a signifi-
cantly longer half-life in cell-based assays than AFC.
This conclusion was drawn based on the chemically12
and enzymatically13,14 labile nature of the allylic thio-
ether moiety. The analog would have the additional
advantage of being stable in the harshly acidic condi-
tions employed in standard Fmoc peptide coupling
schemes, facilitating the synthesis of prenylated pep-
tides, which are quite useful for chemical biology studies
on protein prenylation.7,8 Note that Waldmann and co-
workers have developed elegant alternative routes for
the synthesis of prenylated peptides. In the course of
the synthetic studies on the preparation of 2, we also
synthesized N-acetylfarnesylglycine (3) as a model sys-
tem. Herein, we report the synthesis of 2 and 3 in race-
mic form and their biochemical evaluation versus yeast
Icmt. Note that previously Rando and co-workers had
prepared and evaluated AFC analogs bearing oxygen,
selenium, sulfoxide, and amino substitutions for the
allylic sulfur in 1.15
6 by deprotonating 4 with potassium bis(trimethylsi-
lyl)amide. Surprisingly, under these conditions the
bisfarnesylated product was preferred over the desired
monoalkyated product 6 at room temperature, despite
the use of an excess of 4. A mixture of the two products
was obtained at 0 °C. Subsequent acidic deprotection of
the benzophenone imine 6 afforded ester 7. The free
amine was then acylated to give 8, and saponification
of the farnesyl glycine gave a mixture of products, from
which the desired farnesyl glycine analog 3 was obtained
in modest yield. This unoptimized route provided a prec-
edent for the synthesis of 2 and also provided sufficient
amounts of 3 for its evaluation versus Icmt.
The initial challenge in the synthesis of 2 was the conver-
sion of farnesyl bromide (5, Scheme 2) into bis-homo-
farnesol (10). Attempts were made to employ a
previously published homologation protocol using the
Grignard reagent ClMgCH2CH2OMgCl,18 but this
approach was not successful in our hands. The primary
product formed was squalene, the farnesyl bromide
dimer. Efforts to directly alkylate the farnesyl bromide
with the potassium enolate of ethyl acetate to give 9
failed, leading to the O-alkylated product. This difficult
first step was accomplished using the procedure
We examined the synthesis of the model system N-acetyl-
farnesylglycine to optimize the amino acid coupling
chemistry originally developed by OÕDonnell and co-
workers for use with an isoprenoid side chain.16,17 Earli-
er, unsuccessful attempts were made to prepare 2 using
other amino acid synthesis procedures. We chose to uti-
lize the direct stoichiometric deprotonation of 4 rather
than the standard phase transfer protocol for operation-
al simplicity. The protected glycine variant 4 (Scheme 1)
was deprotonated using NaH as the base, followed by
coupling with farnesyl bromide (5), to afford 6 in modest
yield. Attempts were made to optimize the production of
Br
5
EtOAc, LDA, CuI, THF
84%
-108 to -30 ˚C; 5
CO2Et
9
DIBAL-H,
70%
PhMe, -78 ˚C
Ph
N
CO2Et
4
Ph
OH
10
NaH,THF; 5
0 ˚C to rt
39%
a) CBr4, PPh3, rt
57% (for both steps)
b) 4, NaH, THF, 0 ˚C
CO2Et
CO2Et
CO2Et
6
7
Ph
N
Ph
N
11
1 M HCl/Et2O,
0 ˚C to rt
Ph
69%
94%
52%
1 M HCl/Et2O,
0 ˚C to rt
Ph
26%
CO2Et
NH2
NH2
12
Ac2O, Et3N, rt
Ac2O, Et3N, rt
85%
CO2Et
NH
CO2Et
NH
8
3
13
NaOH, EtOH,
0 ˚C to rt
O
NaOH, EtOH,
0 ˚C to rt
O
83%
2
Scheme 1.
Scheme 2.

5082
B. S. Henriksen et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5080–5083
developed by Kuwajima and Doi,19 and later utilized by
Coates et al.20 The lithium–copper enolate of ethyl ace-
tate was formed by deprotonation with LDA, followed
by addition to anhydrous CuI. Farnesyl bromide addi-
tion then afforded the desired ester 9, as shown below
in Scheme 1. DIBAL-H reduction of the ester produced
bis-homofarnesol (10). The alcohol was then converted
to the bromide, which was not purified but instead
directly coupled to the anion of the glycine benzophen-
imine ethyl ester, to give 11 in 57% overall yield. Subse-
quently, the imine-protecting group was removed under
acidic conditions, and the free amine 12 was acetylated
with acetic anhydride to give 13. Saponification of the
ethyl ester to the carboxylate gave the desired AFC ana-
log 2.21
110
100
90
80
70
60
50
40
30
20
10
0
0
100
200
300
400
500
600
[2], µM
Figure 2. Inhibition of Ste14p activity by 2. In vitro vapor diffusion
methyltransferase activity was performed, as previously described,22
with 33 lM N-acetyl-S-farnesylcysteine, 20 lM S-adenosyl-[14C-meth-
yl]methionine, 5 lg of crude membrane preparation from His-Ste14p
overexpressing cells, and increasing concentrations of compound 2, all
in 100 mM Tris–HCl, pH 7.5. The reaction was allowed to proceed for
30 min at 30 °C. The IC50 value calculated from the data above using
GraphPad Prism 4 was 325 lM (95% confidence interval = 298–
354 lM).
The desthio-AFC analog 2 and the model system 3
were biochemically evaluated versus the well-character-
ized Saccharomyces cerevisiae Icmt variant (Ste14p).22
The previously reported vapor diffusion assay, with
AFC (or 2/3) as the prenylcysteine co-substrate, was
utilized to measure Ste14p activity.23 The biological
activity of N-acetylfarnesylglycine (3) was examined
first. This compound is not recognized by Icmt and
has essentially no ability to act either as a substrate
(maximal specific activity of ꢀ4.5, compared to 700
for AFC) or an inhibitor (IC50 > 1000 lM). This find-
ing emphasized the importance of the proper length of
the isoprenoid moiety in its recognition by Icmt. The
geranyl analog of AFC, with a C10 rather than C15
isoprenoid, also binds very poorly to Icmt due to
the shortened length of the backbone and is not an
Icmt substrate (data not shown). We expected poor
activity of 3 based on the modest biological activity
of the geranyl variant,24 however, a complete lack of
activity of the longer Ôall-carbonÕ farnesylglycine is
somewhat surprising.
O-
-
CO2
S+
14
H3C
NH
AFC sulfoxide
Ki = 13.2 µM
O
-
S
CO2
15
FTA
Ki = 4.6 µM
The Ôall-carbonÕ AFC analog 2 was next examined as
both a substrate for and inhibitor of Ste14p. It also
has essentially no ability to act as a substrate, with a
maximal specific activity of ꢀ3 versus ꢀ700 for AFC.
Compound 2 is a weak inhibitor of Ste14p, with an
IC50 value of 325 lM (Fig. 2), although this level of inhi-
bition may in part be due to non-specific detergent ef-
fects on the activity of the membrane-bound Icmt.
This is in sharp contrast to the behavior of the oxygen,
selenium, and sulfoxide analogs of AFC, which all bind
effectively to mammalian Icmt.15 Note that the racemic
DL-AFC exhibits significant substrate activity, in sharp
contrast to 2, with the unnatural D-enantiomer exhibit-
ing modest inhibitory potency (Ki = 73 lM).25 To pro-
vide a comparison with 2, the structures and inhibitory
potencies of three representative Icmt inhibitors are giv-
en in Figure 3.
-
CO2
S
16
H3C
NH
3-isobutenylAFC
IC50 = 7-14 µM;a 44 µMb
O
17
cysmethylnil
IC50 = 0.2-2.4 µM
N
CONH2
In summary, the poor substrate activity of the previous-
ly reported oxygen, selenium, and amine AFC ana-
logs,15 coupled with the data from the Ôall-carbonÕ
AFC isostere 2, highlights the importance of the sulfur
in the interaction of prenylcysteine moieties with Icmt.
It is interesting to contrast this finding with recent model
system studies, which have suggested that the thioether
Figure 3. Selected Icmt inhibitors. The Ki or IC50 values for the AFC
sulfoxide,25 FTA,24 3-isobutenylAFC,10 and cysmethylnil9 are from
the indicated literature references. (a) IC50 value for Icmt by 16 under
the conditions utilized by Casey and co-workers. (b) IC50 value for
Icmt by 16 under the conditions utilized in this study (see Fig. 2
caption). The Ki for 16 for ste14p is 17.1 lM.10

B. S. Henriksen et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5080–5083
5083
Toone, E. J.; Casey, P. J. Proc. Natl. Acad. Sci. U.S.A.
2005, 102, 4336.
10. Anderson, J. L.; Henriksen, B. S. Gibbs, R. A.; Hrycyna,
C. A. J. Biol. Chem. 2005, 280, 29454.
11. Hrycyna, C. A.; Sapperstein, S. K.; Clarke, S.; Michaelis,
S. EMBO J. 1991, 10, 1699.
12. Gordon, E. M.; Pluscec, J. J. Am. Chem. Soc. 1992, 114,
1521.
in methionine can be replaced with a methylene group
with few negative consequences.26 Enantiomerically en-
riched variants of 2 could be used for the synthesis of
Ôall-carbonÕ prenylcysteine peptide analogs. Such
analogs, which should possess enhanced synthetic and
biological stability, may prove valuable biological
probes for other prenyl binding proteins, such as
RhoGDI27 and RabGDI.8,28
13. Park, S. B.; Howald, W. N.; Cashman, J. R. Chem. Res.
Toxicol. 1994, 7, 191.
14. Tschantz, W. R.; Zhang, L.; Casey, P. J. J. Biol. Chem.
1999, 274, 35802.
Acknowledgments
15. Tan, E. W.; Perez-Sala, D.; Rando, R. R. J. Am. Chem.
Soc. 1991, 113, 6299.
16. OÕDonnell, M.; Eckrich, T. M. Tetrahedron Lett. 1978, 47,
4625.
17. OÕDonnell, M.; Barney, C. L. Tetrahedron Lett. 1985, 26,
3067.
18. Overhand, M.; Stuivenberg, H. R.; Pieterman, E.; Cohen,
L. H.; vanLeeuwen, R. E. W.; Valentijn, A. R. P. M.;
Overkleeft, H. S.; VanDerMarel, G. A.; vanBoom, J. H.
Bioorg. Chem. 1998, 26, 269.
We thank Ying Shao (Department of Pharmaceutical
Sciences, Wayne State University) for performing
exploratory experiments directed toward the synthesis
of 2. This work was supported in part by the Purdue
Cancer Center (Indiana Elks Award to R.A.G./
C.A.H.), the Army CDMRP (NF020054; subcontract
to R.A.G.), and the National Pancreas Foundation (to
C.A.H.).
19. Kuwajima, I.; Doi, Y. Tetrahedron Lett. 1972, 13, 1163.
20. Coates, R. M.; Ley, D. A.; Cavender, P. L. J. Org. Chem.
1978, 43, 4915.
References and notes
21. NMR and MS data for 2: 1H NMR (CDCl3): 1.8 (three s,
12H), 1.9 (m, 6H), 2.2 (t, 2H), 2.3 (s, 3H), 4.8 (t, 1H), 5.2
(t, 3H), 6.5 (q, 1H) and 10.4 (br s, 1H) 13C NMR (CDCl3):
16.49, 18.11, 23.35, 23.68, 25.84, 26.12, 27.06, 27.83, 32.86,
40.13, 45.84, 52.81, 123.91, 124.2, 125.02, 134.5, 134.5,
137.4, 171.61, 176.11. MS-ESI (M-H) = 348.
22. Anderson, J. L.; Frase, H.; Michaelis, S.; Hrycyna, C. A.
J. Biol. Chem. 2005, 280, 7336.
23. Hrycyna, C. A.; Clarke, S. Mol. Cell Biol. 1990, 10, 5071.
24. Tan, E. W.; Perez-Sala, D.; Canada, F. J.; Rando, R. R.
J. Biol. Chem. 1991, 266, 10719.
25. Gilbert, B. A.; Tan, E. W.; Perez-Sala, D.; Rando, R. R.
J. Am. Chem. Soc. 1992, 114, 3966.
26. Tatko, C. D.; Waters, M. L. Protein Sci. 2004, 13, 2515.
27. Mondal, M. S.; Wang, Z.; Seeds, A. M.; Rando, R. R.
Biochemistry 2000, 39, 406.
28. An, Y.; Shao, Y.; Alory, C.; Matteson, J.; Sakisaka, T.;
Chen, W.; Gibbs, R. A.; Wilson, I. A.; Balch, W. E.
Structure 2003, 11, 347.
1. Magee, T.; Seabra, M. C. Curr. Opin. Cell Biol. 2005, 17,
190.
2. Gibbs, R. A.; Zahn, T. J.; Sebolt-Leopold, J. S. Curr.
Med. Chem. 2001, 8, 1437.
3. Reid, T. S.; Terry, K. L.; Casey, P. J.; Beese, L. S. J. Mol.
Biol. 2004, 343, 417.
4. Winter-Vann, A. M.; Casey, P. J. Nat. Rev. Cancer 2005,
5, 405.
5. Brunner, T. B.; Hahn, S. M.; Gupta, A. K.; Muschel, R.
J.; McKenna, G.; Bernhard, E. J. Cancer Res. 2003, 63,
5656.
6. Bell, I. M. J. Med. Chem. 2004, 47, 1869.
7. Naider, F. R.; Becker, J. M. Biopolymers 1997, 43, 3.
8. Watzke, A.; Brunsveld, L.; Durek, T.; Alexandrov, K.;
Rak, A.; Goody, R. S.; Waldmann, H. Org. Biomol.
Chem. 2005, 3, 1157.
9. Winter-Vann, A. M.; Baron, R.; Wong, W.; de la Cuze, J.;
York, J. D.; Gooden, D. M.; Bergo, M.; Young, S. G.;