Synthesis of functional ras lipoproteins and fluorescent derivatives

Article abstract of DOI:10.1021/ja002723o

For the study of biological signal transduction, access to correctly lipidated proteins is of utmost importance. Furthermore, access to bioconjugates that embody the correct structure of the protein but that may additionally carry different lipid groups or labels (i.e., fluorescent tags) by which the protein can be traced in biological systems, could provide invaluable reagents. We report here of the development of techniques for the synthesis of a series of modified Ras proteins. These modified Ras proteins carry a number of different, natural and non-natural lipid residues, and the process was extended to also provide access to a number of fluorescently labeled derivatives. The maleimide group provided the key to link chemically synthesized lipopeptide molecules in a specific and efficient manner to a truncated form of the H-Ras protein. Furthermore, a preliminary study on the biological activity of the natural Ras protein derivative (containing the normal farnesyl and palmitoyl lipid residues) has shown full biological activity. This result highlights the usefulness of these compounds as invaluable tools for the study of Ras signal transduction processes and the plasma membrane localization of the Ras proteins.

Full text of DOI:10.1021/ja002723o

J. Am. Chem. Soc. 2001, 123, 1023-1035
1023
Synthesis of Functional Ras Lipoproteins and Fluorescent Derivatives
Karsten Kuhn,‡ David J. Owen,† Benjamin Bader,§ Alfred Wittinghofer,§
Ju1rgen Kuhlmann,§ and Herbert Waldmann*,‡
Contribution from the Max-Planck-Institut fu¨r Molekulare Physiologie, Department of Chemical Biology
and Department of Structural Biology, Otto-Hahn-Strasse 11, 44227 Dortmund, Germany, and
Department of Medicinal Chemistry, Victorian College of Pharmacy, 381 Royal Parade, ParkVille,
Victoria 3052, Australia
ReceiVed July 24, 2000
Abstract: For the study of biological signal transduction, access to correctly lipidated proteins is of utmost
importance. Furthermore, access to bioconjugates that embody the correct structure of the protein but that
may additionally carry different lipid groups or labels (i.e., fluorescent tags) by which the protein can be
traced in biological systems, could provide invaluable reagents. We report here of the development of techniques
for the synthesis of a series of modified Ras proteins. These modified Ras proteins carry a number of different,
natural and non-natural lipid residues, and the process was extended to also provide access to a number of
fluorescently labeled derivatives. The maleimide group provided the key to link chemically synthesized
lipopeptide molecules in a specific and efficient manner to a truncated form of the H-Ras protein. Furthermore,
a preliminary study on the biological activity of the natural Ras protein derivative (containing the normal
farnesyl and palmitoyl lipid residues) has shown full biological activity. This result highlights the usefulness
of these compounds as invaluable tools for the study of Ras signal transduction processes and the plasma
membrane localization of the Ras proteins.
Introduction
The biological relevance of lipid-modified proteins is high-
lighted by the crucial role of the so-called Ras proteins in
maintaining the regular life cycle of cells. The Ras proteins are
a class of plasma membrane-bound lipoproteins which serve
as central molecular switches in cellular signaling cascades.7
In response to receptor protein tyrosine kinases or other growth
factor-dependent stimuli, these proteins become activated,
exchanging the normally bound guanosine diphosphate (GDP)
for guanosine triphosphate (GTP). In the GTP-bound active
state, the Ras proteins can then interact with and activate
downstream targets, resulting in the activation of transcription
factors.7 The Ras cellular signaling pathway is of central
importance for the control of cell growth and proliferation and
if disturbed uncontrolled proliferation may occur, resulting in
transformation of the cell. The importance of the correct
functioning of the Ras signaling pathway is clearly illustrated
by the fact that a point mutation in the ras oncogenes (coding
for the Ras protein) is found in approximately 30% of human
cancers. This figure can increase to a dramatic 80% for some
major malignancies such as cancer of the lung, colon, and
pancreas.8 To perform their normal and oncogenic functions,
Ras proteins must be membrane-associated, and the covalent
attachment of lipid residues to their peptide chains is critical
to this localization.9 For the posttranslational processing of
Ras, a conserved CaaX amino acid sequence (C is cysteine, a
is generally an aliphatic amino acid, and X is a serine or
methionine residue) is recognized and farnesylated at the
The transduction of stimuli from the extracellular space across
the plasma membrane and ultimately into the cell nucleus is
one of the most important processes by which cells regulate
growth, differentiation, and proliferation. Recent discoveries
have shown that the posttranslational modification of certain
proteins with lipid units is an essential feature of several critical
signal transduction systems in eukaryotic cells.1,2 Lipidation of
protein molecules can arise in three different ways: (1) the co-
or posttranslational attachment of myristic acid to a N-terminal
glycine, (2) the attachment of palmitic acid to the thiol group
of cysteine, and (3) the formation of farnesyl- or geranylgeranyl
thioethers of cysteine.3 It is believed that the attached lipid
groups not only serve to influence the cellular distribution and
anchor the modified protein to cytoplasmic or vesicular mem-
branes but may also be involved in the transduction of signals,
for instance by facilitating protein-protein and protein-lipid
interactions.4-6
* To whom correspondence should be addressed.
‡ Max-Planck-Institute, Department of Chemical Biology.
§ Max-Planck-Institute, Department of Structural Biology.
† Victorian College of Pharmacy.
(1) Lodish, H.; Baltimore, D.; Berk, A.; Zipursky, S. L.; Matsudeira, P.;
Darnell, J. Molecular Cell Biology, 3rd ed.; W. H. Freeman & Co.: New
York, 1995; Chapter 20.
(2) (a) Krauss, G. Biochemie der Regulation und Signaltransduktion,
VCH: New York, 1997. (b) Hidaka, H., Nairin A. C., Eds. Intracellular
Signal Transduction, AdVances in Pharmacology; Academic Press: San
Diego, 1996; Vol. 36.
(3) (a) Casey, P. J. Science 1995, 268, 221. (b) Milligan, G.; Pateri, M.;
Magee, A. L. TIBS 1995, 181.
(5) (a) Moffet, S.; Mouillac, B.; Bonin, H.; Bouvier, M. EMBO J. 1993,
12, 349. (b) Weiss, E. R.; Osawa, S.; Shi, W.; Dickerson, C. D. Biochemistry
1994, 33, 7587.
(6) Wedegaertner, P. B.; Bourne, H. R. Cell 1994, 77, 1063.
(7) (a) Boguski, M. S.; McCormick, F. Nature 1993, 366, 643. (b) Egan,
S. E.; Weinberg, R. A. Nature, 1993, 365, 781.
(4) For the importance of farnesyl groups, see: (a) Kisselev, O. G.;
Ermolaeva, M. V.; Gautam, N. J. J. Biol. Chem. 1994, 269, 21399. (b)
Scheer, A.; Gierschik, P. Biochemistry 1995, 34, 4952. (c) Matsuda, T.;
Takao, T.; Shimonishi, Y.; Murata, A.; Asano, T.; Yoshizawa, T.; Fukada,
Y. J. Biol. Chem. 1995, 269, 30358. (d) For a review, see: Rando, R. R.
Biochim. Biophys. Acta 1996, 1300, 5.
(8) Levitzki, A. Eur. J. Biochem. 1994, 226, 1.
(9) Hall, A. Science 1994, 264, 1413 and references therein.
10.1021/ja002723o CCC: $20.00 © 2001 American Chemical Society
Published on Web 01/18/2001

1024 J. Am. Chem. Soc., Vol. 123, No. 6, 2001
Kuhn et al.
cysteine residue by the enzyme protein farnesyl transferase
(PFTase). After subsequent proteolytic removal of the three
C-terminal amino acids and methylation of the resulting
farnesylated C-terminal cysteine residue, the peptide chain can
then be further modified by introduction of palmitic acid
thioesters. The H-, N- and K-Ras proteins have been cloned
and can be overexpressed and purified from bacterial systems.10
However, as the protein is not naturally found in prokaryotic
cells, such as Escherichia coli, the protein is expressed in the
unmodified state. Farnesylation, proteolysis and methylation of
the isolated protein can be achieved in vitro either via enzymatic
reactions, or with cell free extracts.11 These procedures, however,
are inefficient and can only provide very small quantities of
partially processed protein. Thus, until this time the only access
to fully lipidated Ras proteins required lengthy and tedious
purification from eukarotic systems.12
Previous work from our groups has focused on the synthesis
and biological evaluation of correctly lipidated Ras peptide
fragments.13 However, access to large quantities of correctly
lipidated protein would be even more valuable. Furthermore,
access to bioconjugates that embody the correct structure of
the protein but may additionally carry different lipid groups or
labels by which the protein can be traced in biological systems
(i.e. fluorescent tags) could provide invaluable reagents. Re-
cently we described a method to couple synthetically derived
lipopeptide fragments to recombinant Ras proteins to form
modified Ras protein derivatives.14 Building on our preliminary
data, in this report we outline the development of the general
method for the engineering of protein molecules with the
subsequent production of a small library of modified functional
Ras proteins.
Figure 1. Coupling of the maleimide group 1 to a sulfhydryl group in
a protein to form a stable thioether bond 2.
Maleic acid imides (maleimides 1) are popular constituents
of many heterofunctional protein cross-linking agents, and
readily undergo alkylation reactions with sulfhydryl groups to
form stable thioether bonds 2 (Figure 1).15 It was realized that
this group could provide the key by which chemically synthe-
sized lipopeptide molecules could be linked in a specific manner
to thiol residues such as those found in the amino acid cysteine.
In particular maleimidocaproic acid (MIC-OH 3) enabled the
specific coupling of chemically synthesized lipopeptides to a
modified form of the Ras protein.
For the synthesis of the desired lipopeptide molecules, a
modular strategy was adopted, whereby a tetrapeptide formed
the key intermediate (Figure 2). This tetrapeptide intermediate
would allow further elongation at the C-terminus with lipidated
or nonlipidated amino acids, as well as the addition of various
N-terminal units. For each compound synthesized the MIC-OH
3 group would be positioned at the N-terminus to allow for
simple subsequent coupling to the Ras proteins. Utilizing this
pathway a number of Ras derivatives containing natural and
Figure 2. Retrosynthetic scheme for the synthesis of the lipopeptide
molecules.
(10) Bourne, H. R.; Sanders, D. A.; McCormick, F. Nature 1991, 117,
349.
(11) Porfiri, E.; Evans, T.; Bollag, G.; Clark, R.; Hancock, J. F. Methods
Enzymol. 1995, 255, 13.
(12) Cox, A. D.; Solski, P. A.; Jordan, J. D.; Der C. J. Methods Enzymol.
1995, 255, 195.
(13) (a) Waldmann, H.; Schelhaas, M.; Na¨gele, E.; Kuhlmann, J.;
Wittinghofer, A.; Schroeder, H.; Silvius, J. R. Angew. Chem. 1997, 109,
2334 and Angew. Chem., Int. Ed. Engl. 1997, 36, 2238. (b) Schroeder, H.;
Leventis, R.; Rex, S.; Schelhaas, M.; Na¨gele, E.; Waldmann, H.; Silvius,
J. R. Biochemistry 1997, 36, 13102. (c) Na¨gele, E.; Schelhaas, M.; Kuder,
N.; Waldmann, H. J. Am. Chem. Soc. 1998, 120, 6889. (d) Cotte´ A.; Bader,
B.; Kuhlmann, J.; Wittinghofer, A.; Waldmann, H. Chem. Eur. J. 1999, 5,
922.
nonnatural lipid residues were produced and the technique was
extended to also include a number of fluorescent derivatives.
For the synthesis of the fluorescent derivatives the N-methyl-
anthranilate group was chosen as it is one of the smallest known
fluorescent groups that produces relatively intense fluorescence
upon irradiation with UV light.16
Preliminary biological assays on the bioconjugates have
shown that at least for the natural lipid derivative 43c, full
biological activity is observed. This fact underlines the ef-
fectiveness of this general strategy in providing fully active,
engineered protein derivatives. It is expected that these proteins
(14) Bader, B.; Kuhn, K.; Owen, D. J.; Waldmann, H.; Wittinghofer,
A.; Kuhlmann, J. Nature 2000, 403, 223.
(15) Hermanson, G. T. Bioconjugate Techniques; Academic Press:
London, 1996; Chapters 2.2 and 5.
(16) Hiratsuka, T. J. Biol. Chem. 1982, 257, 13354.

Synthesis of Functional Ras Lipoproteins
J. Am. Chem. Soc., Vol. 123, No. 6, 2001 1025
Scheme 1. Synthesis of the Fluorescently Labeled Lipidated
Cysteine Molecules 9a, 9b
Scheme 2. Synthesis of the Key Tetrapeptide Intermediate
19
The Prenylated Pentapeptides. Synthesis of the tetrapeptide
intermediate 19 followed standard peptide coupling reactions,13c,22
and furnished the desired compound in 44% overall yield
(Scheme 2). With this compound in hand, condensation with
the lipidated cysteine residues 20, 9a and 9b, followed by
standard Fmoc deprotection provided the three central penta-
peptide intermediates 22a, 22b, and 22c in 60-69% yield
(Scheme 3). Access to the first three target molecules was then
easily achieved by coupling of the pentapeptides with maleimido
caproic acid (MIC-OH 3), to furnish compounds 41a, 41b, and
41c in excellent overall yields (Scheme 6).
The Prenylated Heptapeptides. For the production of the
heptapeptide derivatives a modular strategy was adopted,
whereby the pentapeptides 22a, 22b, and 22c would provide
the central intermediates to which would be coupled the
dipeptides 29, 30, and 31 (Scheme 6). Access to the protected
disulfides 23a and 23b was available via literature known
procedures.13c,22 Cleavage of the disulfide bridge by reduction
with excess DTT formed the free thiols, which were immediately
reacted with either palmitoyl chloride or hexadecyl bromide to
provide the desired lipidated dipeptides 24a and 25a,b in high
yields (Scheme 4). Reaction of the disulfides 23a,b with tert-
butanethiol and air also proceeded smoothly to furnish the
desired S-tert-butyl disulfide derivatives 26a,b.23
will provide useful tools for further dissecting the biological
signal transduction process and the study of selective plasma
membrane localization of the Ras proteins.
Results
Fluorescent Probes. The synthesis of the novel fluorescently
labeled analogues, shown in Scheme 1, starts with the known
alcohols 4a and 4b,17,18 both of which are obtained in a two-
step sequence starting from commercially available geraniol and
farnesol. Attachment of the fluorescent N-methylanthranilate
group19 was achieved by acylation of the allylic alcohols with
N-methylisatoic anhydride 5, in good yields. Removal of the
tetrahydropyran-protecting group from the esters 6a and 6b,
under acidic conditions furnished the desired fluorescent alco-
hols 7a and 7b in excellent yields. For attachment of the labeled
lipids to the cysteine a number of methods were tried. However,
by far the best result was obtained utilizing the Corey-Kim
chlorination20 followed by reaction of the allyl chlorides 8a and
8b with cysteine methyl ester in THF and liquid ammonia. This
procedure, based on a known literature procedure,21 provided
the desired compounds 9a and 9b in excellent yields without
the need for chromatography.
As an initial approach to the protected dipeptides 29, 30, and
31 the N-protected Boc/allyl ester protecting group strategy,24
found in compound 25a, was initially chosen. Deprotection of
(17) (a) Turek, T. C.; Gaon, I.; Gamache, D.; Distefano, M. D. Bioorg.
Med. Chem. Lett. 1997, 7, 2125. (b) Gaon, I.; Turek, T. C.; Weller, V. A.;
Edelstein, R. L.; Singh, S. K.; Distefano, M. D. J. Org. Chem. 1996, 61, 7738.
(18) Umbriet, M. A.; Sharpless, K. B. J. Am. Chem. Soc. 1977, 99, 5526.
(19) (a) Venuti, M. C. Synthesis 1982, 266. (b) For reviews on isatoic
anhydrides, see: Coppola, G. M. Synthesis 1980, 505.
(21) Brown, M. J.; Milano, P. D.; Lever, D. C.; Epstein, W. W.; Poulter,
C. D. J. Am. Chem. Soc. 1991, 113, 3176.
(22) (a) Sto¨ber, P.; Schelhaas, M.; Na¨gele, E.; Hagenbuch, P.; Re´tey, J.;
Waldmann, H. Bioorg. Med. Chem. 1997, 5, 75. (b) Joaquina, M.; Amaral,
S. A.; Macedo, M. A.; Oliveira, M. I. J. Chem. Soc., Perkin Trans. 2 1977,
205.
(20) Corey, E. J.; Kim, C. V.; Taheeda, M. Tetrahedron Lett. 1972, 4339.

1026 J. Am. Chem. Soc., Vol. 123, No. 6, 2001
Kuhn et al.
Scheme 4. Formation of the Dipeptide Intermediates 24a,
Scheme 3. Coupling of the Lipidated Cysteine Moieties 9a,
9b, and 20 to the Tetrapeptide Intermediate 19, Followed by
Deprotection of the Fmoc Group to Produce Pentapeptides
22a,b,c
25a,b, and 26a,b
followed by coupling to the MIC-OH 3, afforded the protected
intermediates 39 and 40. These two compounds could then be
smoothly deprotected by treatment with trifluoroacetic acid in
dichloromethane at room temperature for 2 h to provide 30 and
31 in quantitative yields (Scheme 5). This later method, although
providing a slightly lower overall yield for the desired com-
pounds, was decided to be superior due to the ease of handling
and purification of the intermediates.
With ready access to all three lipidated maleimidocaproyl
dipeptides 29, 30, and 31, as well as a sufficient supply of the
three major pentapeptide intermediates 22a, 22b, and 22c, the
coupling reactions to the target heptapeptides could then be
undertaken (Scheme 6). A total of seven heptapeptide derivatives
were produced (see Table 1). Each condensation was performed
essentially using the same conditions (utilizing EDC and HOBt
as coupling reagent), and all reactions provided the desired
targets in moderate to good yields. Confirmation of structure
for each compound was obtained by 1H and 13C NMR, as well
as by FAB and/or MALDI mass spectrometry.
Other Heptapeptide Derivatives. Four more lipid derivatives
were also required to undertake a full study of the biological
importance of the lipid groups present in the natural protein. In
these compounds the effect of the C-terminal cysteine moiety
would be examined. Thus, the target compounds for this study
would require the C-terminal amino acid to either be lipidated
as a hexadecyl thioether, as in compounds 56 and 59 (Scheme
8), or replaced by serine methyl ester as in compounds 62 and
the N-terminal Boc group, followed by coupling to MIC-OH 3
proceeded smoothly to furnish 28 (Scheme 5). Unfortunately,
deprotection of the allyl ester in compound 28 with tetrakis-
triphenylphosphine palladium(0) utilizing a number of different
allyl scavengers was unsuccessful. In this reaction, the allyl ester
was removed cleanly; however, attack also occurred on the
olefin moiety within the maleimide group, resulting in the loss
of the double bond. Two alternative approaches were then
devised to circumvent this problem. The first approach focused
on the complete deprotection of the dipeptide to produce
compounds 35 and 36. These compounds were then subse-
quently coupled to MIC-OH 3 to produce the desired dipeptides
29 and 30, which could be used for the subsequent coupling
reaction. Later studies showed that the maleimide group was
quite stable under the conditions of Boc deprotection, with
trifluoroacetic acid. Thus, in the second approach to the desired
dipeptides, the protecting group arrangement of compound 23a
was reversed. Thereby, removal of the Aloc group from the
dipeptides 25b or 26b, utilizing tetrakistriphenylphosphine
palladium(0) with dimethylbarbituric acid as a scavenger,
(23) (a) Wu¨nsch, E. In Methoden der Organischen Chemie (Houben-
Weyl); Thieme, G., Ed.; Verlag: Stuttgart, 1974; Vol.15/1. (b) Jones, J.
The Chemical Synthesis of Peptides; Clarendon: Oxford, 1992.
(24) (a) Schmittberger, T.; Cotte´, A.; Waldmann, H. J. Chem. Soc., Chem.
Commun. 1998, 937. (b) Schmittberger, T.; Waldmann, H. Bioorg. Med.
Chem. 1999, 7, 749-762.

Synthesis of Functional Ras Lipoproteins
J. Am. Chem. Soc., Vol. 123, No. 6, 2001 1027
Scheme 5. Three Approaches to the Desired Dipeptides 29,
30, and 31.
The serine-substituted derivative 53 was elaborated in a
manner similar to that used for the cysteine hexadecyl com-
pounds. Thus, condensation with the dipeptides 32 and 33,
produced the lipidated heptapeptide derivatives 60 and 63.
Deprotection of the Boc-protecting group with trifluoroacetic
acid/thioanisole, followed by coupling of the heptapeptides with
MIC-OH 3 under standard conditions furnished the final two
target compounds 62 and 65 in excellent overall yields (Scheme
9).
Coupling to Ras. With a large number of lipopeptide
derivatives in place (See Table 1), the compounds could then
be coupled to mutated forms of the Ras protein. Two truncated
versions of the H-Ras protein were utilized for this study. The
main mutant utilized in this study had the last eight amino acids
from the C-terminus removed (including the CaaX box sequence
CVLS), leaving a cysteine residue at the C-terminus possibly
well accessible for coupling. Truncation of full length H-Ras
cDNA was achieved with standard PCR-methods. Briefly, a stop
codon was introduced into position 182 of the H-Ras cDNA,
and the resulting PCR product was cloned into an E. coli
expression vector (ptac expression vector). Protein expression
in E. coli strain CK600K and purification by ion-exchange
chromatography and gel filtration was performed as described.25
The second mutant utilized in this study was similar to the first
but also contained a mutation of a glycine residue at position
12 to a valine residue.26a This mutation removes the ability for
the Ras protein to hydrolyze bound GTP, thus making the
protein oncogenic.26b Coupling of the peptides to the protein
was readily achieved by gently mixing the peptide and protein
together in a 1:1 mixture of 11% Triton X-114 detergent and
Tris buffer overnight at 4 °C. Purification of the resulting labeled
lipoprotein was easily provided by an extraction procedure with
Triton X-114 detergent.27 This detergent has the added property
that it undergoes a phase separation from aqueous solutions at
37 °C.28 Thus, the lipoprotein is extracted into the Triton phase
leaving unlabeled Ras protein in the aqueous phase, and as can
be seen in Figure 3a, yields virtually 95% pure labeled protein.
The protein was then separated from the Triton X-114 detergent
by simple DEAE ion exchange chromatography. The entire
labeling procedure was found to work extremely well, and as
can be seen in Table 2, excellent yields for the coupling reaction
were obtained. Initial attempts at the coupling reaction utilized
a 10-fold excess of the lipopeptide coupling partner. The
coupling clearly worked very well; however, it was found in
subsequent reactions that the reaction was virtually quantitative
and only a 1:1 equivalent of peptide/protein was actually
required, with no reduction in the isolated yield.
Table 1. Table of All the Target Lipopeptides Synthesized.
type peptide
peptide
MIC-Met-Gly-Leu-Pro-Cys(Far)-OMe
MIC-Met-Gly-Leu-Pro-Cys(Ger/Mant)-OMe
MIC-Met-Gly-Leu-Pro-Cys(Far/Mant)-OMe
41c
41a
41b
43c
44c
42c
44a
42a
44b
42b
56
MIC-Gly-Cys(Pal)-Met-Gly-Leu-Pro-Cys(Far)-OMe
MIC-Gly-Cys(HD)-Met-Gly-Leu-Pro-Cys(Far)-OMe
MIC-Gly-Cys(StBu)-Met-Gly-Leu-Pro-Cys(Far)-OMe
MIC-Gly-Cys(HD)-Met-Gly-Leu-Pro-Cys(Ger/Mant)-OMe
MIC-Gly-Cys(StBu)-Met-Gly-Leu-Pro-Cys(Ger/Mant)-OMe
MIC-Gly-Cys(HD)-Met-Gly-Leu-Pro-Cys(Far/Mant)-OMe
MIC-Gly-Cys(StBu)-Met-Gly-Leu-Pro-Cys(Far/Mant)-OMe
MIC-Gly-Cys(Pal)-Met-Gly-Leu-Pro-Cys(HD)-OMe
MIC-Gly-Cys(StBu)-Met-Gly-Leu-Pro-Cys(HD)-OMe
MIC-Gly-Cys(Pal)-Met-Gly-Leu-Pro-Ser-OMe
59
62
65
Characterization of the labeled proteins was obtained by SDS-
PAGE gel electrophoresis and electrospray mass spectrometry
of the intact labeled proteins (see Figure 3). Full proof for the
specific incorporation of the label into the C-terminal cysteine
residue was obtained for three representative proteins (proteins
coupled with the peptides 41c, 43c, and 44c). In these experi-
ments the labeled proteins were digested with trypsin or
chymotrypsin proteases, and the resulting digested samples were
then analyzed by electrospray, as well as MALDI mass
MIC-Gly-Cys(HD)-Met-Gly-Leu-Pro-Ser-OMe
65 (Scheme 9). For the synthesis of these compounds, the Fmoc
group was thought to be unsatisfactory thus, the Boc-protected
tetrapeptide 48 formed the key intermediate (Scheme 7).
Synthesis of the tetrapeptide was relatively straightforward based
on standard protocols. Condensation of the tetrapeptide with
either hexadecylated cysteine methyl ester or serine methyl ester,
followed by deprotection of the Boc group provided the desired
pentapeptide intermediates 51 and 53 (Scheme 7).
The cysteine hexadecyl pentapeptide was then condensed with
both the Boc protected dipeptides 32 and 34 to produce the
lipidated heptapeptides 54 and 57. Deprotection of the Boc
groups in both compounds with trifluoroacetic acid/thioanisole,
followed by coupling with MIC-OH 3 under standard conditions,
provided the targets 56 and 59 (Scheme 8).
(25) Tucker, J.; Sczakiel, G.; Feuerstein, J.; John, J.; Goody, R. S.;
Wittinghofer, A. EMBO J. 1986, 5, 1351.
(26) (a) The point mutation in codon 12 was introduced by PCR
mutagenesis according to: Picard, V.; Ersdal-Badju, E.; Lu, A.; Bock, S.
C Nucleic Acids Res. 1994, 22, 2587. (b) For reviews see: Bourne, H., R.;
Sanders, D. A.; McCormick, F. Nature 1990, 348, 125-132. Bourne, H.
R.; Sanders, D. A.; McCormick, F. Nature 1991, 349, 117.
(27) Gutierrez, L.; Magee, A. I.; Marshall, C. J.; Hancock, J. F. EMBO
J. 1989, 8, 1093.
(28) Bordier, C. J. Biol. Chem. 1981, 256, 1604.

1028 J. Am. Chem. Soc., Vol. 123, No. 6, 2001
Kuhn et al.
Scheme 6. Formation of the Ten Prenylated Target Molecules, Compounds 41a,b,c; 42a,b,c; 43c; 44a,b,c
Table 2. Table of Results from Coupling Reactions of
Lipopeptides to the Truncated Ras Protein
Scheme 7. Synthesis of the N-terminally Deprotected
Nonprenylated Pentapeptides 51 and 53
recovered
protein conjugate
peptide
(mg)
protein
(mg)
molar
ratio
peptide
(mg)
yield (%)
41c
41a
41b
43c
44c
44a
44b
42c
42a
42b
56
5.03
1.50
0.64
6.50
6.60
0.87
0.61
0.62
o.52
0.79
0.33
0.60
0.58
0.73
0.55
9.1a
13.0a
11.5a
11.6a
11.0a
10.9a
10.3a
10.8b
8.5a
12:1
2.5:1
1:1
8.5:1
9:1
1:1
1:1
1:1
1:1
1:1
1:1
6.24
7.3
6.2
5.7
5.9
6.2
3.2
5.3
4.3
6.8
2.2
6.4
3.5
5.5
5.6
66
54
51
46
50
53
29
49
50
56
40
62
35
50
59
12.2a
5.5b
59
62
65
10.2b
9.9b
1:1
1:1
1.25:1
1:1
10.9a
9.5b
a Normal truncated H-Ras. b Mutated (G12V) truncated H-Ras.
of the lipopeptides 41c, 43c, and 44c to the C-terminal cysteine,
Cys181 of the truncated Ras protein, thereby, proving the
specificity of the reaction.
Biological Assays. As a preliminary investigation into the
biological activity, the oncogenic protein coupled with the
natural lipid derivative 43c was chosen for microinjection studies
in PC12 cells. It had been previously shown that these cells
can be induced to differentiate, producing long outgrowths from
the cell, after microinjection with functional oncogenic Ras
protein.29 In this experiment, two additional protein samples
were also microinjected into the PC12 cells as controls, the first
sample was the truncated oncogenic H-Ras protein (1-181),
spectrometry. Through this technique, the peptide fragments
from the protease digestion are separated and the molecular
weights of individual fragments can be determined. The results
from these experiments clearly showed the covalent attachment
(29) Schmidt, G.; Lenzen, C.; Simon, I.; Deuter, R.; Cool, R. H.; Goody,
R. S.; Wittinghofer, A. Oncogene 1996, 12, 87.

Synthesis of Functional Ras Lipoproteins
J. Am. Chem. Soc., Vol. 123, No. 6, 2001 1029
Scheme 8. Synthesis of the C-terminal Hexadecylated Heptapeptides 56 and 59
Scheme 9. Synthesis of the C-terminal Serine Heptapeptides 62 and 65
while the second was a full length bacterially expressed and
therefore unmodified sample of H-Ras, that is capable of being
modified by the cellular enzymatic machinery. As can be seen
in Figure 4, the uncoupled truncated Ras, as expected had no
effect on the growing cells (Figure 4a), while the full length
Ras protein sample (Figure 4b) is fully active and produces the
expected outgrowths from the cells. Much to our delight, upon
microinjection of the coupled oncogenic protein, long out-
growths from the PC12 cells began to form (Figure 4c), thereby
proving that this coupled protein derivative was biologically
active.
teins.13,30 Efficient methods for the synthesis of these sensitive
compounds have been developed and problems associated with
the acid and base sensitivity of these compounds can be
alleviated by the correct protecting group strategy.22a,24,31 As
useful as the lipopeptides have been fully lipid modified proteins
would provide the ultimate tools. Such molecules would
naturally contain the correct protein and lipid structures or could
alternatively carry different lipid groups or labels by which the
proteins could be traced in biological systems (i.e., fluorescent
tags which can be detected by fluorescence microscopy and
(30) For reviews see (a) Hinterding, K.; Alonso-D´ıaz, D.; Waldmann,
H. Angew. Chem. 1998, 110, 716 and Angew. Chem., Int. Ed. 1998, 37,
688. (b) Kappes, T.; Waldmann, H. Liebigs Ann./ Recl. 1997, 803. (c) Eisele,
F.; Owen, D. J.; Waldmann, H. Bioorg. Med. Chem. 1999, 7, 193.
(31) Schelhaas, M.; Waldmann, H. Angew. Chem. 1996, 108, 2192 and
Angew. Chem., Int. Ed. Engl. 1996, 35, 2056.
Discussion
Previous studies from our group have focused on the
production of lipopeptides as tools for studying cellular signaling
cascades and selective membrane targeting of lipidated pro-

1030 J. Am. Chem. Soc., Vol. 123, No. 6, 2001
Kuhn et al.
a
b
Figure 3. An example of characterization of the product from the coupling reaction of H-Ras 1-181 with lipopeptide 41b. (a) Picture of SDS-
PAGE gels for coupling reaction. The differences in electrophoretic mobility between labeled and unlabeled H-Ras 1-181 are very small, unlabeled
H-Ras migrated slightly faster. Lanes 6 and 8 of both gels show the good separation of labeled and unlabeled H-Ras employing the Triton X-114
method. Gel A shows the slight difference in electophoretic mobility, gel B shows under UV-excitation virtually no fluorescent protein in lane 6.
A small amount of labeled protein is lost during washing the detergent phases with fresh buffer (lane 7). Lanes 8 and 9 show the high purity of
isolated labeled H-Ras. (b) Electrospray mass spectra of H-Ras 1-181 (black) and of the protein coupled with a lipopeptide 42a (gray). Nanoelectrospray
mass spectrometry of starting material and product in a LCQ ion-trap instrument from Finnigan shows the expected shift from Ras∆C (calculated
MW: 20446 g‚mol-1, black trace) to Ras-7merGer/Mant (21617 g‚mol-1, gray trace). Starting material and product show minor peaks due to small
amounts of protein without the first N-terminal cycteine (20312/21486 g‚mol-1). Precision of the instrument (4 Da.
Figure 4. Pictures from PC 12 cells after microinjection of three different oncogenic protein samples of H-Ras. Picture (a) shows cells in which
the truncated form of the oncogenic H-Ras (1-181) was injected. Because of the absence of modification sites, this protein can’t be modified by
the cellular enzymatic machinery. As expected, there is no effect on the cell growth. As shown in the picture (b), when the full-length oncogenic
H-Ras, which can be modified, is injected into the cells, long outgrowths are formed. In picture (c) cells are shown which formed such outgrowths,
too, after injection of the truncated oncogenic H-Ras (1-181) coupled with the lipid derivative 43c. This result shows the biological activity of
those protein conjugates.
fluorescence spectroscopy). We recently reported a general
method by which synthetically derived lipopeptide chains could
be selectively coupled to a recombinant Ras protein.14 Here we
describe the production of a small library of engineered Ras
protein derivatives. These engineered proteins were easily
synthesized, purified and characterized and may provide the
foundation for a wide arsenal of derivatives which would be of
use for further biological studies. It is expected that this coupling
procedure should also be completely general and thus could
be further utilized for the production of other important bio-
conjugates.
The maleimide group has been widely used as a popular
constituent of many heterobifunctional cross-linking agents.15
The double bond of the maleimide group 1 can undergo
alkylation reactions with sulfhydryl groups to form stable
thioether bonds 2 (Figure 1). Although maleimides can also react
with amines and other nucleophiles, the reaction is specific with
sulfhydryl groups in the pH range of 6.5-7.5. Thus, at pH 7
the maleimide group will react a 1000 times faster with the
sulfhydryl of cysteine than with the ꢀ amine of lysine.32
Furthermore, the derivative maleimidocaproic acid (MIC-OH
3) is a stable commercially available compound, which can
easily be incorporated into peptide molecules via standard
peptide coupling procedures. It was for these reasons that the
maleimide group was thought to provide the key to linking the
lipopeptide molecules specifically to the Ras protein.
For the synthesis of the desired target molecules we initially
saw that the tetrapeptide 19 would provide the major intermedi-
ate to all targets (Figure 2). This compound would allow further
elongation at the C-terminus with differently labeled cysteine
amino acids or other amino acids, as well as allow the addition
of various N-terminal pieces. For each compound the maleimido
caproic acid group would be placed at the N-terminus for
subsequent attachment onto the mutated Ras protein. As noted
above, synthesis of the desired lipopeptides followed established
protocols developed within our group,13c,22a,24 and access to
eight mono- and dilipidated peptide derivatives, as well as six
mono- and dilipidated fluorescent derivatives was achieved (see
Table 1).
(32) (a) Smyth, D. G.; Nagamatsu, A.; Fruton, J. S. J. Am. Chem. Soc.
1960, 82, 4600. (b) Smyth, D. G.; Blumenfeld, O. O.; Konigsberg, W.
Biochem. J. 1964, 91, 589.

Synthesis of Functional Ras Lipoproteins
J. Am. Chem. Soc., Vol. 123, No. 6, 2001 1031
The synthesis of a range of lipid derivatives was thought to
be useful to probe the biological activity of the newly
synthesized Ras proteins, in particular, the influence of the
different membrane anchors on distribution and signaling ability
of the new biopolymers. A number of biological assays for
monitoring the in vivo activity of Ras already exist,29,33 and so
by producing a series of differently lipidated Ras proteins, the
specific effect of the different lipid groups could be studied.
Thus, in one set of compounds the S-palmitoyl thioester, which
occurs in the natural protein, was replaced by either; (1) no
lipidation site at all (compounds 41a,b,c), (2) by a metabolically
stable thioether of similar lipophilicity (compounds 44a,b,c),
or (3) by a free thiol group (compounds 42a,b,c). The free thiol
group can be generated in situ from the S-tert-butyl disulfide
by treatment of the coupled protein with dithiothreitol.13b In
one case (compound 43c), the S-palmitoyl thioester was
introduced into the peptide to form a Ras derivative with both
natural lipid residues. It has previously been shown that
S-palmitoylation/deacylation of both Ras proteins and peptide
fragments is a rapidly reversible process within a cell.13b For
this reason we believe that for microinjection studies, proteins
coupled with either the palmitoyl derivative 43c or the free thiol
derivatives 42a,b,c are essentially equivalent.
To undertake a full study of the biological importance of the
lipid groups present in the natural protein, four more compounds
were synthesized. In these compounds the effect of the C-
terminal cysteine moiety would be examined. Thus, the second
set of compounds contained the normal native palmitoylation
site (either as the palmitoylated thioester derivative or as the
free thiol derivative) combined with changes in the S-farnesyl
thioether moiety. In these molecules, replacement of the
farnesylated cysteine methyl ester with a hexadecylated cysteine
methyl ester (compounds 56 and 59) is expected to show if the
farnesyl group itself is important or if the presence of any lipid
group is sufficient for biological activity of the Ras proteins.
Finally with the removal of a C-terminal lipidation site, by
replacement with a nonlipidizable serine residue (compounds
62 and 65), the monolipidated derivatives were completed.
The Fluorescent Derivatives. It was believed at the onset
of the project that fluorescent groups attached to the peptide
could provide invaluable compounds for further in vivo biologi-
cal studies utilizing the technique of microinjection followed
by fluorescence microscopy. Earlier studies from our groups
had shown that problems could be encountered with loss of
fluorescence from fluorescently labeled lipopeptides.34 It was
believed that by placing the fluorescent marker within the lipid
group, no interference with the attachment of the peptide to the
protein, should occur. Furthermore, with the fluorescent group
in the lipid moiety, once the labeled protein is injected into cells,
the lipid group would insert into intracellular membranes and
thus, would (1) protect the fluorescent group from hydrolysis,
and (2) is expected to increase the groups quantum yield. There
were some initial reservations that the presence of a bulky
fluorescent group within the lipid portion of the protein
molecules may effect the binding of the lipid molecule to the
biological lipid bilayer. It was for this reason that the fluorescent
N-methylanthranilate group was chosen, as it is one of the
smallest known compounds that produces relatively intense
fluorescence upon irradiation with UV light.16 It was decided
that two different lipid chain lengths would also be advantageous
as the effect of the chain length on protein binding could also
be observed.
Coupling to H-Ras. With a large number of lipopeptide
derivatives in place the compounds could then be coupled to
the mutated forms of the Ras protein (see Table 2). The main
mutant used in this study, H-Ras 1-181, was a mutant of the
H-Ras protein in which the last eight amino acid residues had
been removed from its sequence leaving a single surface
available cysteine residue at the C-terminus. This protein is
expressed in high level from an E. coli expression vector, and
is easily purified.25 The coupling reaction between the peptide
and the protein was extremely straightforward, resulting in high
yields of coupled protein after a relatively simple extraction
procedure followed by anion exchange chromatography. Further-
more, it was found that the coupling reaction required only one
equivalent of each of the coupling partners, thus highlighting
the efficiency of the maleimide-sulfhydryl reaction. The H-Ras
protein used in this study does in fact have three more cysteine
residues within its sequence. It could thus not be excluded, that
the maleimide group would react with one of these cysteine
residues, although the crystal structure shows that they are not
easily accessible.35 Proof for the specific incorporation of the
lipopeptide tail to the C-terminal cysteine residue (and not to
other amino acid residues) was obtained for three representative
proteins. Thus, proteolysis of the labeled proteins, followed by
mass spetrometry of the fragments clearly showed the presence
of a single lipopeptide covalently attached to the C-terminal
cysteine 181 residue.
A second truncated form of the Ras protein was also used in
the coupling reactions. This protein carries basically the same
truncation of the tail as the other “natural” protein. However,
this protein in addition carries a mutation at the twelve position
whereby, the natural glycine is replaced by a valine residue.
This mutation renders the protein unable to hydrolyze bound
GTP, thereby, keeping the protein in a continuously active
state.26 This particular oncogenic protein, once derivatized with
the various lipopeptides, provides a useful tool for assessing
the biological activity of the protein-peptide bioconjugates, as
the continuous “ON” signal enables the quick detection of active
and membrane-bound Ras proteins. As expected there was no
detectable difference between the handling and coupling of the
two different Ras proteins with the various peptide fragments.
Biological Activity. As a preliminary investigation into the
biological activity of these derivatives, the product from the
coupling reaction of the oncogenic H-Ras protein with the
natural lipopeptide N-Ras fragment 43c was tested by micro-
injection into PC12 cells. Microinjection into these cells forms
a typical standard assay to assess the biological activity of Ras
proteins, as once injected with oncogenic Ras proteins they
quickly begin to differentiate, producing long outgrowths. In
our experiment after microinjection of the protein sample into
the PC12 cells a formation of long outgrowths was observed.
This result proves the biological activity of the engineered
protein, and also proves that the maleimide group has little, if
any effect on protein activity.
Conclusions
The study of Ras signal transduction is an important field
especially when it is considered that many useful drugs to
combat diseases such as cancer, may develop from full
understanding of this important pathway.8 The development of
(33) Biological assays were adopted for microinjection by Schmidt, G.
in her Ph.D. thesis, based on different cell lines: NIH3T3 cells: Stacey,
D. W.; Kung, H. F. Nature 1984, 310, 508. Fos-LacZ cells: Lloyd, A. C.;
Paterson, H. F.; Morris, J. D.; Hall, A, Marshall; C. J. EMBO J. 1989, 8,
1099. PC 12 cells: Greene, L. A.; Tischler, A. S. Proc. Nat. Acad. Sci.
U.S.A. 1976, 73, 2424.
(35) Pai, E. F.; Kabsch, W.; Krengel, U.; Holmes, K. C.; Wittinghofer,
(34) Schelhaas, M.; Na¨gele, E.; Waldmann, H. Unpublished results.
A. Nature 1989, 341, 209.

1032 J. Am. Chem. Soc., Vol. 123, No. 6, 2001
Kuhn et al.
triazole (HOBt) (1.5 equiv), followed by ethyl-(dimethylamino)-
propylcarbodiimide (EDC) (1.2 equiv). The reaction was warmed to
room temperature and stirred for 18 h. The reaction mixture was then
diluted with ethyl acetate (50-100 mL) and was extracted with 0.5 M
HCl (2 × 10 mL), 1 M NaHCO3 (2 × 10 mL), and finally with brine
solution (2 × 10 mL). The organic layers were dried over MgSO4,
filtered, and concentrated under reduced pressure. The product could
then be isolated from the remaining residue by flash chromatography
on silica gel using the solvent systems noted for each particular
compound.
tools to study such processes is an integral part of this goal.
Here through the successful combination of organic chemistry,
biochemistry and cell biology we hope to better understand the
molecular details of complex processes such as those involved
in signal transduction and membrane targeting. In this study,
access to the rather sensitive lipopeptide molecules was ef-
ficiently achieved utilizing a range of powerful synthetic
protocols, previously developed within our group,13,22a,24 while
the maleimide group provided the key reagent for the efficient
and specific coupling of the synthetic lipopeptide derivatives
to truncated forms of the H-Ras protein. The natural and
nonnatural engineered protein derivatives produced in this work
are expected to provide important tools for further studying the
Ras signaling pathway. Microinjection experiments on one of
the lipopeptide-protein conjugates has shown the bioconjugate
to possess full biological activity.
It is important to note that the methodology developed here
forms the basis for a general biological readout system,
comprised of a protein head and a lipopeptide tail. It is expected
that the methodology should be completely general and will
allow for the specific engineering of many different protein
bioconjugates. As shown in this study, by using an oncogenic
Ras protein molecule, coupled to different lipopeptide tails, we
can obtain a quantifiable biological readout of the ability of
different lipopeptide tails to localize the protein to the plasma
membrane after microinjection into PC12 cells. Conversely,
taking a lipopeptide tail that can efficiently anchor a protein to
the plasma membrane (such as compounds 42, 43, or 44) and
coupling it to various protein molecules we may be able to
determine the possible interaction of those particular protein
molecules with or at the plasma membrane. Such a system may
also prove useful for studying the interactions of these various
protein molecules with other membrane bound proteins. The
development of general biological readout systems, such as
the one outlined here, should provide invaluable tools for
determining membrane-protein interactions and may assist in
determining the molecular details of important biological
processes such as cellular signaling cascades.
Standard Fmoc-Protecting Group Removal. To a solution of the
protected peptide in CH2Cl2 (5-10 mL) at room temperature was added
piperidine (1-2 mL). The reaction was warmed to room temperature
and stirred for 18 h. After removing the solvent by azeotropic distillation
with toluene (2 × 30 mL) and chloroform (30 mL), the product was
isolated from the remaining residues by flash chromatography on silica
gel using the solvent systems noted for each particular compound.
Standard Procedure for the Removal of Aloc Groups or Allyl
Esters. To a solution of the protected peptide in dry THF (5-10 mL)
was added dimethylbarbituric acid (0.55 equiv), followed by a catalytic
amount of tetrakis(triphenylphosphine)palladium(0). The reaction was
monitored by TLC, and was judged complete with the disappearance
of starting material (generally 2-3 h). The solvent was then removed
under reduced pressure. The product could be isolated from the
remaining residue by the methods noted for each particular compound.
Standard Procedure of the Removal of Boc Groups or tert-Butyl
Esters. To a solution of the protected peptide in CH2Cl2 (5 mL) and
thioanisole (0.5 mL) was added trifluoroacetic acid (2.5-5 mL). The
reaction was then left to stir for 1-2 h at room temperature. The solvent
was removed by azeotropic distillation with toluene (2 × 50 mL) and
chloroform (50 mL). The desired product was isolated from the
remaining residues by the methods noted for each particular compound.
(E,E)-8-O-(2-N-Methyl-aminobenzoyl)-1-(tetrahydropyran)-3,7-
dimethyl-2,6-octandiene (6a). A mixture of the anhydride 5 (440 mg,
2.4 mmol) and DMAP (26 mg, 0.24 mmol), dissolved in dry DMF (2
mL), was heated to 65 °C with stirring under an atmosphere of dry
argon. A mixture of the alcohol 4a (620 mg, 2 mmol) and dry
triethylamine (340 µL, 2.4 mmol), dissolved in dry DMF (2 mL), was
then slowly added to the anhydride solution over a period of 40 min.
The reaction was left stirring at this temperature for 3.5 h, after which
time TLC indicated the reaction was virtually complete. The solution
was poured into a separatory funnel containing ethyl acetate (100 mL)
and was washed once with 1:1 water/brine solution (10 mL) followed
by brine (10 mL). The combined aqueous phases were back-extracted
three times with ethyl acetate (10 mL). The combined organic phases
were dried over MgSO4 and filtered, and the solvent was removed under
reduced pressure. Purification of the resulting oily residue by flash
chromatography on silica gel with hexane/ethyl acetate (8:1), yielded
711 mg (75%) of the desired product 6a as a colorless oil, followed by
60 mg (10%) of starting material: Rf ) 0.65 (hexane/ethyl acetate 3:1).
UV (MeOH) λmax 223 nm (ꢀ 29 958), 256 nm (ꢀ 9960), 355 nm (ꢀ
7130). IR (KBr) νmax 3378, 2940 br s, 1682, 1607, 1581, 1520, 1239,
Experimental Section
General Methods. Proton and carbon NMR spectra were recorded
on Bruker AC-250, Bruker AM-400 and Bruker DRX-500 spectrom-
eters. NMR spectra were obtained in either deuteriochloroform, deuterio
methanol, or deuteriochloroform/deuterio methanol mixture. Proton and
carbon spectra are reported in parts per million downfield from an
internal standard of Me4Si. EI-MS and FAB-MS were recorded on a
Finnigan MAT 90 machine, MALDI-MS were recorded on a Voyager
machine by Applied Biosystems, applied matrices are given for each
particular compound. ESI-MS were recorded on a LCQ Ion Trap
instrument by Finnigan. Infrared Spectra were recorded on a Bruker
IFS88 FTIR spectrometer as a KBr disk. UV spectra were recorded on
a Perkin-Elmer UV/vis spectrometer in methanol. Combustion analysis
was carried out on a Heraeus CHN-Rapid analyzer. Flash chroma-
tography was carried out on columns packaged with Baker silica
gel (30-60 µm). TLC was carried out on Kieselgel 60F254 aluminum
sheets (Merck Darmstadt, Germany). All reagents were obtained from
Fluka (Buchs, Switzerland), Aldrich (Steinheim, Germany) or Sigma
(Deisenhofen, Germany). All solvents were dried and distilled using
standard procedures.36 All peptide couplings were performed utilizing
standard synthetic methods (see Supporting Information).
1
1128, 1085, 1023, 751 cm-1. H NMR (500 MHz, CDCl3): δ 7.90
(dd, J ) 1.6, 8.0 Hz, 1H, arom CH); 7.67 (br s, 1H, NH); 7.34 (ddd,
J ) 1.6, 7.1, 8.4 Hz, 1H, arom CH); 6.62 (dd, J ) 0.8, 8.4 Hz, 1H,
arom CH); 6.56 (ddd, J ) 0.8, 7.1, 8.0 Hz, 1H, arom CH); 5.51 (dt, J
) 1.0, 6.4 Hz, 1H, CH Ger); 5.38 (dt, J ) 1.1, 7.4 Hz, 1H, CH Ger);
4.62 (s, 2H, CO2CH2); 4.61 (t, J ) 3.6 Hz, 1H, OCHO); 4.23 (dd, J )
6.4, 11.7 Hz, 1H, CH2aO Ger); 4.01 (dd, J ) 7.4, 11.7 Hz, 1H, CH2bO
Ger); 3.87 (m, 1H, CH2aO THP); 3.49 (m, 1H, CH2bO THP); 2.87 (s,
3H, NHCH3); 2.17-2.22 (m, 2H, CH2 Ger); 2.07-2.10 (m, 2H, CH2
Ger); 1.79-1.85 (m, 1H, CH2a THP); 1.71 (s, 3H, CH3 Ger); 1.69 (m,
1H partially obscured, CH2b THP); 1.68 (s, 3H, CH3 Ger); 1.47-1.60
(m, 4H, 2 * CH2 THP). 13C NMR (125.6 MHz, CDCl3): δ 168.3 (CO2);
Standard Peptide Coupling Conditions. Unless otherwise stated,
all peptide coupling reactions were performed based on the following
standard procedure: To a 1:1 (molar) solution of each coupling partner
dissolved in CH2Cl2 (5-10 mL) at 0 °C was added 1-hydroxybenzo-
(37) Waldmann, H.; Kunz, H. Liebigs Ann. Chem. 1983, 11, 1724.
(38) Schelhaas, M.; Na¨gele, E.; Kuder, N.; Bader, B.; Kuhlmann, J.;
Wittinghofer, A.; Waldmann, H. Chem. Eur. J. 1999, 5, 1239.
(39) Amaral, M. J. S. A.; Adelina Macedo, M.; Olivera, M. I. A. J. Chem.
Soc., Perkin Trans. 1 1977, 205.
(36) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals, 3rd ed.; Pergamon: Oxford, 1988.
(40) Fukase, K.; Kitazawa, M.; Wakamiya, T.; Shiba T. Bull. Chem. Soc.
Jpn. 1990, 1838.

Synthesis of Functional Ras Lipoproteins
J. Am. Chem. Soc., Vol. 123, No. 6, 2001 1033
152.0 (arom quart. CNH); 139.4 (quart. C Ger); 134.5; 131.5 (2 * arom
CH); 130.6 (quart. C Ger); 128.6; 121.1 (2 * CH Ger); 114.3; 110.6 (2
* arom CH); 110.0 (arom quart. CCO2); 97.7 (CH THP); 69.6
(CO2CH2); 63.5 (CH2O THP); 62.1 (CH2O Ger); 39.0 (CH2 Ger); 30.7
(CH2 THP); 29.4 (NHCH3); 26.0; 25.5; 19.6 (3 * CH2 THP); 16.4;
14.0 (2 * CH3 Ger). MS (EI) m/z (rel intensity) 387 (M+, 8.2), 302
(26), 151 (100), 134 (24), 105 (14), 85 (88). HRMS (EI) m/z: calcd
for (M+) C23H33NO4 387.2409, found 387.2397. Anal. Calcd: C, 71.29;
H, 8.58; N, 3.61. Found: C, 71.54; H, 8.47; N, 3.90.
S-[(E,E)-8-O-(2-N-Methyl-aminobenzoyl)-3,7-dimethyl-2,6-octan-
diene]-L-cysteine Methyl Ester (H-Cys(Ger/Mant)-OMe) (9a). To
a solution L-cysteine methyl ester hydrochloride (73.0 mg, 0.42 mmol)
in dry liquid ammonia (10 mL), stirring vigorously at -40 °C under
an atmosphere of argon, was added a solution of the chloride 8a
dissolved in dry THF (10 mL). The resulting clear solution was then
left to slowly warm overnight. After evaporation of the remaining
solution the resulting white solid was dissolved in water (50 mL) and
extracted with Et2O (5 × 20 mL). The combined organic phase was
dried over MgSO4, filtered and the solvent removed under reduced
pressure to yield compound 9a as a pale yellow oil (142 mg, 89%). Rf
) 0.53 (hexane/ethyl acetate 1:2). [R]22D ) +11.2 (c ) 1.0, CH3OH).
UV (MeOH) λmax 199 nm (ꢀ 29 653), 220 nm (ꢀ 31 033), 253 nm (ꢀ
9106), 354 nm (ꢀ 6220). IR (KBr) νmax 3379, 2925 br s, 1742, 1681,
(E,E)-8-O-(2-N-Methyl-aminobenzoyl)-3,7-dimethyl-2,6-octan-
dien-1-ol (7a). Compound 6a (500 mg, 1.3 mmol) was dissolved in
ethanol (5 mL). Solid PPTS (36 mg, 0.14 mmol) was added and the
reaction was heated at 60 °C for 3 h. After this time, the solvent was
removed under reduced pressure, and the oily residue was purified by
flash chromatography on silica gel using hexane/ethyl acetate (3:1) as
eluent, and yielded 362 mg (92%) of compound 7a as a colorless oil:
Rf ) 0.34 (hexane/ethyl acetate 3:1). UV (MeOH) λmax 223 nm (ꢀ
36 325), 256 nm (ꢀ 11 857), 355 nm (ꢀ 8497). IR (KBr) νmax 3379 br
s, 2914 br s, 1680, 1607, 1581, 1520, 1242, 1128, 1087, 751 cm-1. 1H
NMR (500 MHz, CDCl3): δ 7.92 (dd, J ) 1.6, 8.0 Hz, 1H, arom CH);
7.60 (br s, 1H, NH); 7.37 (ddd, J ) 1.6, 7.1, 8.5 Hz, 1H, arom CH);
6.65 (dd, J ) 1.0, 8.5 Hz, 1H, arom CH); 6.58 (ddd, J ) 1.0, 7.1, 8.5
Hz, 1H, arom CH); 5.50 (dt, J ) 1.3, 7.0 Hz, 1H, CH Ger); 5.40 (dt,
J ) 1.3, 6.8 Hz, 1H, CH Ger); 4.63 (s, 2H, CO2CH2); 4.12 (d, J ) 6.8
Hz, 2H, CH2O Ger); 2.89 (s, 3H, NHCH3); 2.12-2.21 (m, 2H, CH2
Ger); 2.06-2.09 (m, 2H, CH2 Ger); 1.71 (s, 3H, CH3 Ger); 1.67 (s,
3H, CH3 Ger). 13C NMR (125.6 MHz, CDCl3): δ 168.4 (CO2); 152.0
(arom quart. CNH); 138.8 (quart. C Ger); 134.6; 131.5 (2 * arom CH);
130.6 (quart. C Ger); 128.4; 123.9 (2 * CH Ger); 114.3; 110.7 (2 *
arom CH); 110.0 (arom quart. CCO2); 69.6 (CO2CH2); 59.2 (CH2O
Ger); 38.9 (CH2 Ger); 29.5 (NHCH3); 25.9 (CH2 Ger); 16.2; 14.0 (2 *
CH3 Ger). MS (EI) m/z (rel intensity) 303 (M+, 12), 151 (100), 134
(21), 105 (23). HRMS (EI) m/z: calcd for (M+) C18H25NO3 303.1834,
found 303.1823. Anal. Calcd: C, 71.24; H, 8.31; N, 4.62. Found: C,
71.09; H, 8.31; N, 4.91.
1
1607, 1580, 1520, 1240 cm-1. H NMR (500 MHz, CDCl3): δ 7.89
(dd, J ) 1.6, 8.0 Hz, 1H, arom CH); 7.62 (br d, J ) 4.7 Hz, 1H, NH);
7.34 (ddd, J ) 1.6, 7.0, 8.4 Hz, 1H, arom CH); 6.62 (dd, J ) 1.0, 8.4
Hz, 1H, arom CH); 6.55 (ddd, J ) 1.0, 7.0, 8.0 Hz, 1H, arom CH);
5.47 (t, J ) 7.0 Hz, 1H, CH Ger); 5.21 (t, J ) 7.8 Hz, 1H, CH Ger);
4.60 (s, 2H, CO2CH2); 3.70 (s, 3H, OCH3); 3.59 (dd, J ) 4.7, 7.7 Hz,
1H, R-CH Cys); 3.09-3.20 (m, 2H, R-CH2 Ger); 2.86 (d, J ) 4.7 Hz,
3H, NHCH3); 2.83 (dd, J ) 4.7, 13.5 Hz, 1H, â-CH2a Cys); 2.64 (dd,
J ) 7.7, 13.5 Hz, 1H, â-CH2b Cys); 2.00-2.20 (m, 4H, 2 * CH2 Ger);
1.76 (s, 2H, NH2); 1.69 (s, 3H, CH3 Ger); 1.64 (s, 3H, CH3 Ger). 13C
NMR (125.6 MHz, CDCl3): δ 174.4 (CO2Me); 168.3 (CO2); 151.9
(arom quart. CNH); 138.8 (quart. C Ger); 134.5; 131.4 (2 * arom CH);
130.6 (quart. C Ger); 128.3; 120.3 (2 * CH Ger); 114.2; 110.6 (2 *
arom CH); 110.0 (arom quart. CCO2); 69.5 (CO2CH2); 54.1 (R-CH
Cys); 52.1 (OCH3); 38.9 (R-CH2 Ger); 36.3 (â-CH2 Cys); 29.6 (CH2
Ger); 29.4 (NHCH3); 26.0 (CH2 Ger); 16.0; 14.0 (2 * CH3 Ger). MS
(EI) m/z (rel intensity) 420 (M+, 27), 151 (100), 134 (68), 105 (20), 88
(22). HRMS (EI) m/z: calcd for (M+) C22H32N2O4S 420.2070, found
420.2083.
N-(9-Fluorenylmethyloxycarbonyl)-L-methionylglycyl-L-leucyl-L-
prolyl-S-[(E,E)-8-O-(2-N-methyl-aminobenzoyl)-3,7-dimethyl-2,6-
octandiene]-L-cysteine Methyl Ester (Fmoc-Met-Gly-Leu-Pro-Cys-
(Ger/Mant)-OMe) (21a). This compound was prepared utilizing the
standard peptide coupling procedure with the tetrapeptide 19 (250 mg,
0.39 mmol) and the lipidated cysteine 9a (164 mg, 0.39 mmol).
Purification by flash chromatography on silica gel starting with ethyl
acetate, followed by ethyl acetate/methanol (95:5) as eluent, yielded
272 mg (68%) of the desired product 21a as a colorless oil. Rf ) 0.6
(ethyl acetate). [R]22D ) -52.9 (c ) 1.0, CH3OH). 1H NMR (500 MHz,
CDCl3): δ 7.91 (dd, J ) 1.6, 8.0 Hz, 1H, arom CH Mant); 7.75 (d, J
) 7.5 Hz, 2H, arom CH Fmoc); 7.58 (d, J ) 7.4 Hz, 2H, arom CH
Fmoc); 7.35-7.47 (m, 4H, 2 * arom CH Fmoc, arom CH Mant,
CONH); 7.30 (m, 2H, arom CH Fmoc); 7.03 (br s, 1H, CONH); 6.92
(br s, 1H, CONH); 6.67 (d, J ) 8.4 Hz, 1H, arom CH Mant); 6.59 (t,
J ) 8.0 Hz, 1H, arom CH Mant); 5.62 (d, J ) 7.7 Hz, 1H, OCONH);
5.48 (t, J ) 6.9 Hz, 1H, CH Ger); 5.15 (t, J ) 7.9 Hz, 1H, CH Ger);
4.81-4.84 (m, 1H, R-CH); 4.60-4.68 (m, 2H, 2 * R-CH), 4.63 (s,
2H, CO2CH2); 4.30-4.47 (m, 3H, OCH2 Fmoc, R-CH); 4.21 (t, J )
6.6 Hz, 1H, CH Fmoc); 3.90-4.13 (m, 2H, R-CH2 Gly); 3.75-3.79
(m, 1H, δ-CH2a Pro); 3.72 (s, 3H, OCH3); 3.61-3.64 (m, 1H, δ-CH2b
Pro); 3.16 (dd, J ) 8.1, 13.3 Hz, 1H, R-CH2a Ger); 3.06 (dd, J ) 7.4,
13.3 Hz, 1H, R-CH2b Ger); 2.92 (dd, J ) 4.7, 13.8 Hz, 1H, â-CH2a
Cys); 2.90 (s, 3H, NHCH3); 2.72 (dd, J ) 6.9, 13.8 Hz, 1H, â-CH2b
Cys); 2.54 (br s, 2H, γ-CH2 Met); 2.09 (s, 3H, SCH3); 1.90-2.30 (m,
8H, 2 * CH2 Ger, â-CH2 Met, â-CH2 Pro); 1.71 (s, 3H, CH3 Ger);
1.65 (s, 3H, CH3 Ger); 1.50-1.71 (m, 5H, γ-CH2 Pro, â-CH2 Leu,
γ-CH Leu); 0.94 (d, J ) 6.5 Hz, 3H, 1 * ω-CH3 Leu); 0.91 (d, J ) 6.5
Hz, 3H, 1 * ω-CH3 Leu). 13C NMR (125.6 MHz, CDCl3): δ 171.8;
171.6; 171.1; 168.3; 168.2 (5 * CdO); 156.1 (OCONH); 151.9 (arom
quart. CNH); 143.7; 143.5; 141.1 (3 * quart. C Fmoc); 139.1 (quart. C
Ger); 134.5; 131.4 (2 * arom CH); 130.5 (quart. C Ger); 128.3 (CH
Ger); 127.6; 126.9; 125.0 (3 * arom CH, Fmoc); 120.0 (CH Ger); 119.8
(arom CH, Fmoc), 114.1; 110.6 (2 * arom CH); 109.8 (arom Cq-CO2);
69.5 (CO2-CH2); 66.9 (OCH2 Fmoc); 59.6 (R-CH Pro); 53.9 (R-CH
Met); 52.2 (OCH3); 52.1 (R-CH Cys); 48.6 (R-CH Leu); 47.4 (δ-CH2
Pro); 47.0 (CH Fmoc); 42.9 (R-CH2 Gly); 41.6 (â-CH2 Leu); 38.9 (R-
CH2 Ger); 32.5 (â-CH2 Cys); 32.0 (CH2 Met); 29.9 (CH2 Ger); 29.4
(E,E)-8-O-(2-N-Methyl-aminobenzoyl)-3,7-dimethyl-2,6-octandi-
ene-1-chloride (8a). To a solution of N-chlorosuccinimide (30.5 mg,
0.23 mmol) in CH2Cl2 (1 mL), stirring at -40 °C under an atmosphere
of argon, was added dimethyl sulfide (18.5 uL, 0.28 mmol) dropwise.
The reaction was then warmed to 0 °C with an ice bath and kept at
this temperature for 5 min. The reaction was then cooled back to -40
°C, and a solution of the alcohol 7a (63 mg, 0.21 mmol) in CH2Cl2 (2
mL), was added dropwise. The resulting milky white solution was then
left stirring to warm to 0 °C over 1 h. The reaction was maintained at
0 °C for another hour, after which time it was then warmed to room
temperature for a further 15 min. The resulting clear yellow solution
was poured into a separating funnel containing pentane (100 mL) and
ice cold brine solution (20 mL). The layers were separated, and the
organic phase was washed once more with ice cold brine solution. The
combined aqueous phases were back-extracted with pentane (2 × 20
mL). The combined organic phase was dried over Na2SO4 and filtered,
and the solvent was removed under reduced pressure, to yield the allylic
chloride 8a as a pale yellow oil (62 mg, 92%): Rf ) 0.63 (hexane/
ethyl acetate 6:1). UV (MeOH) λmax 219 nm (ꢀ 44 908), 253 nm (ꢀ
11 954), 353 nm (ꢀ 8432). IR (KBr) νmax 3379, 2933 br s, 1682, 1607,
1
1581, 1520, 1241, 1174, 1128, 1087, 844, 751 cm-1. H NMR (500
MHz, CDCl3): δ 7.92 (dd, J ) 1.6, 8.0 Hz, 1H, arom CH); 7.66 (br s,
1H, NH); 7.38 (ddd, J ) 1.6, 7.1, 8.4 Hz, 1H, arom CH); 6.66 (dd, J
) 1.0, 8.4 Hz, 1H, arom CH); 6.58 (ddd, J ) 1.0, 7.1, 8.0 Hz, 1H,
arom CH); 5.43-5.49 (m, 2H, 2 * CH Ger); 4.63 (s, 2H, CO2CH2);
4.08 (d, J ) 8.0 Hz, 2H, CH2Cl Ger); 2.90 (s, 3H, NHCH3); 2.18-
2.21 (m, 2H, CH2 Ger); 2.09-2.13 (m, 2H, CH2 Ger); 1.73 (s, 3H,
CH3 Ger); 1.72 (s, 3H, CH3 Ger). 13C NMR (125.6 MHz, CDCl3): δ
168.4 (CO2); 152.1 (arom quart. CNH); 142.1 (quart. C Ger); 134.6;
131.5 (2 * arom CH); 130.9 (quart. C Ger); 128.0; 120.7 (2 * CH Ger);
114.3; 110.7 (2 * arom CH); 110.0 (arom quart. CCO2); 69.6 (CO2-
CH2); 41.0 (CH2Cl Ger); 38.8 (CH2 Ger); 29.6 (NHCH3); 25.8 (CH2
Ger); 16.1; 14.1 (2 * CH3 Ger). MS (EI) m/z (rel intensity) 321 (M+,
19), 151 (100), 134 (30), 133 (17), 105 (32). HRMS (EI) m/z: calcd
for (M+) C18H24ClNO2 321.1495, found 321.1511. Anal. Calcd: C,
67.26; H, 7.53; N, 4.36. Found: C, 67.07; H, 7.61; N, 4.30.

1034 J. Am. Chem. Soc., Vol. 123, No. 6, 2001
Kuhn et al.
(NHCH3); 29.3 (CH2 Met); 28.6 (â-CH2 Pro); 26.7; 26.0 (2 * CH2 Ger);
24.7 (γ-CH2 Pro); 24.4 (γ-CH Leu); 23.1; 21.9 (2 * ω-CH3 Leu); 15.9
(CH3 Ger); 15.1 (SCH3); 13.9 (CH3 Ger). HRMS (FAB; 3-NBA) m/z:
calcd for (M + H)+ C55H73N6O10S2 1041.452, found 1041.483. Anal.
Calcd: C, 63.44; H, 6.97; N, 8.07. Found: C, 63.05; H, 6.82; N, 7.96.
131.5 (arom CH); 130.7 (quart. C Ger); 128.4; 120.0 (2 * CH Ger);
114.3; 110.7 (2 * arom CH); 110.0 (arom quart. CCO2); 69.7 (CO2-
CH2); 59.8 (R-CH Pro); 52.5; 52.2 (R-CH Met, R-CH Cys); 52.0
(OCH3); 48.9 (R-CH Leu); 47.5 (δ-CH2 Pro); 42.9 (R-CH2 Gly); 41.7
(â-CH2 Leu); 39.1 (R-CH2 Ger); 37.6 (NCH2 MIC); 36.1 (R-CH2 MIC);
32.9 (â-CH2 Cys); 31.5; 30.1 (3 signals 2 * CH2 Met, CH2 Ger); 29.5
(NHCH3); 28.2 (2 signals overlapped, CH2 MIC, â-CH2 Pro); 26.3 (CH2
Ger); 26.2 (CH2 MIC); 24.9; 24.8 (γ-CH2 Pro, CH2 MIC); 24.6 (γ-CH
Leu); 23.3; 21.9 (2 *ω-CH3 Leu); 16.1 (CH3 Ger); 15.3 (SCH3); 14.0
(CH3 Ger). HRMS (FAB; 3-NBA) m/z: calcd for (M + H)+
C50H74N7O11S2 1012.489, found 1012.465.
L-Methionylglycyl-L-leucyl-L-prolyl-S-[(E,E)-8-O-(2-N-methyl-
aminobenzoyl)-3,7-dimethyl-2,6-octandiene]-L-cysteine Methyl Ester
(H-Met-Gly-Leu-Pro-Cys(Ger/Mant)-OMe) (22a). This compound
was prepared utilizing the standard Fmoc deprotection procedure using
the pentapeptide 21a (130 mg, 0.125 mmol). Purification by flash
chromatography on silica gel starting with ethyl acetate, followed by
ethyl acetate/methanol (80:20), yielded 95 mg (93%) of the desired
product 22a as a colorless oil. Rf ) 0.2 (ethyl acetate/methanol 95:5).
[R]22D ) -43.1 (c ) 1.0, CH3OH). 1H NMR (500 MHz, CDCl3/CD3-
OD 5:1): δ 7.83 (dd, J ) 1.5, 8.0 Hz, 1H, arom CH); 7.70 (d, J ) 7.7
Hz, 1H, CONH); 7.56 (d, J ) 7.5 Hz, 1H, CONH); 7.29-7.32 (m,
2H, arom CH, CONH); 6.60 (d, J ) 8.4 Hz, 1H, arom CH); 6.51 (t, J
) 7.5 Hz, 1H, arom CH); 5.43 (t, J ) 6.6 Hz, 1H, CH Ger); 5.13 (t,
J ) 7.2 Hz, 1H, CH Ger); 4.61 (m, 1H, R-CH); 4.55 (s, 3H, R-CH,
CO2CH2); 4.46-4.49 (m, 1H, R-CH); 4.04 (obscured by MeOH, 2H,
R-CH2 Gly); 3.72-3.76 (m, 1H, δ-CH2a Pro); 3.68 (s, 3H, OCH3);
3.51-3.55 (m, 1H, δ-CH2b Pro); 3.28 (br, 1H, R-CH Met); 3.12 (dd,
J ) 8.3, 13.6 Hz, 1H, R-CH2a Ger); 3.04 (m, 1H, R-CH2b Ger); 2.88
(dd, J ) 5.0, 13.7 Hz, 1H, â-CH2a Cys); 2.83 (s, 3H, NHCH3); 2.72
(dd, J ) 7.0, 13.8 Hz, 1H, â-CH2b Cys); 2.55 (t, J ) 7.4 Hz, 2H,
γ-CH2 Met); 1.91-2.12 (m, 10H, 2 * CH2 Ger, â-CH2 Met, â-CH2
Pro, γ-CH2 Pro); 2.06 (s, 3H, SCH3); 1.76 (m, 1H, γ-CH Leu); 1.65
(s, 3H, CH3 Ger); 1.60 (s, 3H, CH3 Ger); 1.40-1.65 (m, 2H, â-CH2
Leu); 0.88 (d, J ) 5.2 Hz, 6H, 2 *ω-CH3 Leu). 13C NMR (125.6 MHz,
CDCl3):δ 172.9; 171.9; 171.5; 169.5; 169.4; 168.7 (6 * CdO); 152.2
(arom quart. CNH); 139.7 (quart. C Ger); 134.9; 131.7 (2 * arom CH);
130.8 (quart. C Ger); 128.6; 120.1 (2 * CH Ger); 114.6; 111.0 (2 *
arom CH); 110.1 (arom quart. CCO2); 69.9 (CO2CH2); 60.2 (R-CH
Pro); 52.8 (R-CH Met); 52.7 (OCH3); 52.3 (R-CH Cys); 49.8 (R-CH
Leu); 47.5 (δ-CH2 Pro); 42.7 (R-CH2 Gly); 40.6 (â-CH2 Leu); 39.2
(R-CH2 Ger); 32.9 (â-CH2 Cys); 30.4 (CH2 Met); 29.8 (CH2 Ger); 29.5
(NHCH3); 29.2 (CH2 Met); 28.4 (â-CH2 Pro); 26.3 (CH2 Ger); 25.0
(γ-CH2 Pro); 24.8 (γ-CH Leu); 23.3; 21.5 (2 *ω-CH3 Leu); 16.1 (CH3
Ger); 15.0 (SCH3); 14.1 (CH3 Ger). HRMS (FAB; 3-NBA) m/z: calcd
for (M + H)+ C40H62N6O8S2 819.4218, found 819.4149.
N-(6-Maleimidocaproyl)-glycyl-tert-butylthio-L-cysteyl-L-methio-
nylglycyl-L-leucyl-L-prolyl-S-[(E,E)-8-O-(2-N-methyl-aminobenzoyl)-
3,7-dimethyl-2,6-octandiene]-L-cysteine Methyl Ester (MIC-Gly-
Cys(StBu)-Met-Gly-Leu-Pro-Cys(Ger/Mant)-OMe) (42a). This
compound was prepared following the standard coupling procedure
using the pentapeptide 22a (29.5 mg, 0.036 mmol) and MIC-Gly-Cys-
(StBu)-OH 31 (18.2 mg, 0.039 mmol, 1.1 eq.). The compound was
purified by flash chromatography on silica gel using gradient elution
starting with ethyl acetate, followed by ethyl acetate/methanol (98:2)
and finally with ethyl acetate/methanol (95:5). After purification 20.0
mg (44%) of the desired target molecule 42a was obtained as a colorless
oil. Rf ) 0.3 (ethyl acetate/methanol 95:5). [R]22 ) -51.8 (c ) 1.0,
D
1
CH2Cl2). H NMR (500 MHz, CDCl3/CD3OD, 3:1): δ 8.30 (d, J )
7.2 Hz, 1H, CONH); 8.22 (d, J ) 6.9 Hz, 1H, CONH); 7.87 (dd, J )
1.6, 8.0 Hz, 1H, arom CH); 7.76 (br s, 1H, NH); 7.67 (d, J ) 7.6 Hz,
1H, CONH); 7.50 (d, J ) 8.0 Hz, 1H, CONH); 7.44 (d, J ) 5.3 Hz,
1H, CONH); 7.34-7.36 (m, 1H, arom CH); 6.96 (s, 2H, CHdCH
MIC); 6.65 (d, J ) 8.5 Hz, 1H, arom CH); 6.56 (dt, J ) 1.0, 8.0 Hz,
1H, arom CH); 5.47 (t, J ) 6.6 Hz, 1H, CH Ger); 5.16 (t, J ) 7.1 Hz,
1H, CH Ger); 4.52-4.70 (m, 4H, 4 *R-CH, R-CH CysGer, R-CH CysHD
,
R-CH Pro, R-CH Leu); 4.59 (s, 2H, CO2CH2); 4.13-4.22 (obscured
by MeOH, 2H, R-CH2a Gly, R-CH Met); 3.85-3.90 (m, 2H, R-CH2
Gly′); 3.81-3.85 (m, 1H, δ-CH2a Pro); 3.71 (s, 3H, OCH3); 3.56-
3.71 (m, 2H, δ-CH2b Pro, R-CH2b Gly); 3.47 (t, J ) 7.2 Hz, 2H, NCH2
MIC); 3.15 (dd, J ) 8.3, 13.2 Hz, 1H, R-CH2a Ger); 3.01-3.07 (m,
2H, R-CH2b Ger, â-CH2a CysStBu); 2.94 (dd, J ) 6.5, 13.5 Hz, 1H,
â-CH2b CysStBu); 2.90 (dd, J ) 4.9, 13.8 Hz, 1H, â-CH2a CysGer);
2.87 (s, 3H, NHCH3); 2.70 (dd, J ) 7.2, 13.8 Hz, 1H, â-CH2b CysGer);
2.58-2.62 (m, 1H, γ-CH2a Met); 2.50-2.54 (m, 1H, γ-CH2b Met);
2.00-2.29 (m, 12H, 2 * CH2 Ger, R-CH2 MIC, â-CH2 Met, â-CH2
Pro, γ-CH2 Pro); 2.06 (s, 3H, SCH3); 1.69 (s, 3H, CH3 Ger); 1.64 (s,
3H, CH3 Ger); 1.44-1.78 (m, 7H, 2 * CH2 MIC, γ-CH Leu, â-CH2
Leu); 1.30 (br s, 11H, CH2 MIC, SC(CH3)3); 0.92 (d, J ) 6.7 Hz, 3H,
1 *ω-CH3 Leu); 0.90 (d, J ) 6.7 Hz, 3H, 1 *ω-CH3 Leu). MS (FAB;
3-NBA) m/z: found 1260 (M + H)+.
N-(6-Maleimidocaproyl)-L-methionylglycyl-L-leucyl-L-prolyl-S-
[(E,E)-8-O-(2-N-methyl-aminobenzoyl)-3,7-dimethyl-2,6-octandiene]-
L-cysteine Methyl Ester (MIC-Met-Gly-Leu-Pro-Cys(Ger/Mant)-
OMe) (41a). This compound was prepared following the standard
coupling procedure using the pentapeptide 22a (64 mg, 0.08 mmol)
and maleimidocaproic acid 3 (18 mg, 0.09 mmol, 1.1 eq.). The
compound was purified by flash chromatography on silica gel starting
with ethyl acetate, followed by ethyl acetate/methanol (98:2), and
furnished 44 mg (55%) of the desired target molecule 41a as a colorless
oil. Rf ) 0.33 (ethyl acetate/methanol 98:2). [R]22D ) -41.6 (c ) 1.0,
N-(6-Maleimidocaproyl)-glycyl-S-hexadecyl-L-cysteyl-L-methio-
nylglycyl-L-leucyl-L-prolyl-S-[(E,E)-8-O-(2-N-methyl-aminobenzoyl)-
3,7-dimethyl-2,6-octandiene]-L-cysteine Methyl Ester (MIC-Gly-Cys-
(HD)-Met-Gly-Leu-Pro-Cys(Ger/Mant)-OMe) (44a). This compound
was prepared utilizing the standard coupling procedure with the
pentapeptide 22a (27 mg, 0.033 mmol) and MIC-Gly-Cys(HD)-OH
30 (20 mg, 0.033 mmol). The compound was purified by flash
chromatography on silica gel using gradient elution starting with ethyl
acetate, followed by ethyl acetate/methanol (98:2) and furnished 25.5
mg (52%) of the desired target molecule 44a as a colorless oil. Rf )
0.63 (ethyl acetate/methanol 98:2). [R]22D ) -56.3 (c ) 1.0, CH2Cl2).
1H NMR (500 MHz, CDCl3): δ 7.91 (dd, J ) 1.5, 7.9 Hz, 2H, arom
CH, CONH); 7.63 (br s, 1H, NH); 7.55 (d, J ) 7.5 Hz, 1H, CONH);
7.37 (t, J ) 6.9 Hz, 2H, arom CH, CONH); 7.20 (d, J ) 7.9 Hz, 1H,
CONH); 6.68 (s, 2H, CHdCH MIC); 6.66-6.68 (m, 1H, arom CH);
6.58 (t, J ) 7.9 Hz, 1H, arom CH); 5.49 (t, J ) 6.5 Hz, 1H, CH Ger);
5.17 (t, J ) 7.7 Hz, 1H, CH Ger); 4.74-4.78 (m, 1H, R-CH Leu);
4.69-4.72 (m, 1H, R-CH CysGer); 4.62 (s, 2H, CO2CH2); 4.51-4.59
(m, 2H, R-CH CysHD R-CH Pro); 4.37 (dd, J ) 7.7, 17.0 Hz, 1H,
R-CH2a Gly); 4.27 (br, 1H, R-CH Met); 3.91-3.96 (m, 2H, R-CH2
Gly′); 3.85-3.88 (m, 1H, δ-CH2a Pro); 3.74 (s, 3H, OCH3); 3.54-
3.73 (m, 2H, δ-CH2b Pro, R-CH2b Gly); 3.50 (t, J ) 7.1 Hz, 2H, NCH2
MIC); 3.16 (dd, J ) 8.2, 13.3 Hz, 1H, R-CH2a Ger); 3.05 (dd, J ) 4.5,
13.3 Hz, 1H, R-CH2b Ger); 2.90 (d, J ) 4.9 Hz, 3H, NHCH3); 2.81-
2.96 (m, 3H, â-CH2a CysGer, â-CH2 CysHD); 2.71 (dd, J ) 6.6, 13.7
1
CH2Cl2). H NMR (500 MHz, CDCl3): δ 7.90 (dd, J ) 1.6, 8.0 Hz,
1H, arom CH); 7.58-7.62 (m, 2H, CONH, NH); 7.57 (d, J ) 7.6 Hz,
1H, arom CH); 7.36 (m, 1H, arom CH); 7.24 (br s, 1H, CONH); 6.67
(s, 2H, CHdCH MIC); 6.64 (d, J ) 8.5 Hz, 1H, arom CH); 6.55-
6.57 (m, 2H, arom CH, CONH); 5.46-5.49 (m, 1H, CH Ger); 5.17 (t,
J ) 8.2 Hz, 1H, CH Ger); 4.83 (dt, J ) 4.4, 9.1 Hz, 1H, R-CH); 4.59-
4.66 (m, 3H, 3 *R-CH); 4.61 (s, 2H, CO2CH2); 4.00-4.06 (m, 2H,
R-CH2 Gly); 3.80-3.84 (m, 1H, δ-CH2a Pro); 3.71 (s, 3H, OCH3);
3.62-3.66 (m, 1H, δ-CH2b Pro); 3.48 (t, J ) 7.2 Hz, 2H, NCH2 MIC);
3.16 (dd, J ) 8.3, 13.2 Hz, 1H, R-CH2a Ger); 3.04 (dd, J ) 7.2, 13.2
Hz, 1H, R-CH2b Ger); 2.90 (partially obscured, 1H, â-CH2a Cys); 2.89
(d, J ) 4.8 Hz, 3H, NHCH3); 2.70 (dd, J ) 7.3, 13.8 Hz, 1H, â-CH2b
Cys); 2.50-2.55 (m, 2H, γ-CH2 Met); 2.12-2.25 (m, 3H, â-CH2a Met,
R-CH2 MIC); 2.08 (s, 3H, SCH3); 1.72-2.11 (m, 9H, 2 * CH2 Ger,
â-CH2b Met, â-CH2 Pro, γ-CH2 Pro); 1.70 (s, 3H, CH3 Ger); 1.65 (s,
3H, CH3 Ger); 1.50-1.70 (m, 6H, 2 * CH2 MIC, γ-CH Leu, â-CH2a
Leu); 1.23-1.31 (m, 3H, CH2 MIC, â-CH2b Leu); 0.93 (d, J ) 6.5 Hz,
3H, 1 *ω-CH3 Leu); 0.91 (d, J ) 6.5 Hz, 3H, 1 *ω-CH3 Leu). 13C
NMR (125.6 MHz, CDCl3):δ 173.1; 172.3; 171.7; 171.3; 171.2; 170.9
(2 signals overlapped); 168.4; 168.4 (9 * CdO); 152.1 (arom quart.
CNH); 139.4 (quart. C Ger); 134.6 (arom CH); 134.1 (CHdCH MIC);

Synthesis of Functional Ras Lipoproteins
J. Am. Chem. Soc., Vol. 123, No. 6, 2001 1035
Hz, 1H, â-CH2b CysGer); 2.55-2.67 (m, 2H, γ-CH2 Met); 2.52 (t, J )
7.4 Hz, 2H, R-CH2 HD); 1.98-2.41 (m, 12H, 2 * CH2 Ger, R-CH2
MIC, â-CH2 Met, â-CH2 Pro, γ-CH2 Pro); 2.08 (s, 3H, SCH3); 1.72 (s,
3H, CH3 Ger); 1.66 (s, 3H, CH3 Ger); 1.42-1.80 (m, 9H, 2 * CH2
MIC, â-CH2 HD, γ-CH Leu, â-CH2 Leu); 1.24 (br s, 28H, CH2 MIC,
(CH2)13); 0.93 (t, J ) 6.4 Hz, 6H, 2 *ω-CH3 Leu); 0.87 (t, J ) 6.9 Hz,
3H, ω-CH3 HD). 13C NMR (125.6 MHz, CDCl3):δ 174.9; 173.4; 171.4;
171.2; 171.1; 170.9 (2 signals overlapped); 170.8; 170.6; 169.7; 168.5
(11 * CdO); 152.1 (arom quart. CNH); 139.5 (quart. C Ger); 134.7
(arom CH); 134.1 (CHdCH MIC); 131.5 (arom CH); 130.8 (quart. C
Ger); 128.4; 120.0 (2 * CH Ger); 114.3; 110.7 (2 * arom CH); 110.0
(arom quart. CCO2); 69.7 (CO2CH2); 60.1 (R-CH Pro); 53.4; 53.1 (R-
CH Met, R-CH Cys); 52.6 (OCH3); 51.9 (R-CH Cys); 49.1(R-CH Leu);
47.4 (δ-CH2 Pro); 44.0; 43.0 (2 *R-CH2 Gly); 40.6 (â-CH2 Leu); 39.1
(R-CH2 Ger); 37.6 (NCH2 MIC); 35.9 (R-CH2 MIC); 33.2; 32.7; 32.7;
31.9; 30.6; 29.7 (2 *â-CH2 Cys, 2 * CH2 Met, CH2 Ger, 9 * CH2 HD);
29.6 (NHCH3); 29.6; 29.4; 29.3; 28.9; 28.5; 28.2; 27.8; 26.3; 26.2;
25.0; 24.9 (3 * CH2 MIC, 5 * CH2 HD, CH2 Ger, â-CH2 Pro, γ-CH2
Pro); 24.6 (γ-CH Leu); 23.3 (1 *ω-CH3 Leu); 22.7 (CH2 HD); 21.7
(1 *ω-CH3 Leu); 16.1 (CH3 Ger); 15.0 (SCH3); 14.2 (CH3 Ger); 14.1
(ω-CH3 HD). MS (FAB; 3-NBA) m/z: found 1396.0 (M + H)+.
S-Hexadecyl-L-cysteine Methyl Ester (H-Cys(HD)-OMe) (49). To
a solution of 1.0 g (5.81 mmol) HCl * H-Cys-OMe (1.0 g, 5.81 mmol),
dissolved in dry DMF (10 mL), was added at 0 °C diisopropylethyl-
amine (2.5 mL, 14.50 mmol) followed by hexadecylbromide (3.6 mL,
11.62 mmol). After stirring for 3 h at room temperature, the solvent
was removed under reduced pressure and the residue was taken up
in 1 M HCl (80 mL). The product was extracted by washing with
CH2Cl2 (3 × 60 mL). The organic layer was washed with 1 M NaHCO3
(100 mL) and water (100 mL) and dried over MgSO4. After removing
the solvent, the product was purified by flash chromatography on silica
gel using hexane/ethyl acetate (1:1) as eluent and furnished 541 mg
containing solutions). The reaction tube was sealed under argon and
placed on a rotary mixer and left to react at 4 °C. After 15 h, the solution
was then centrifuged to remove any denatured protein. The supernatant
was transferred into a fresh 15 mL centrifuge tube, diluted with 3 mL
buffer (containing 2mM DTE), and heated to 37 °C upon which the
solution turned milky white. Centrifugation at room-temperature resulted
in a phase separation of an 11% Triton X-114 detergent phase from
the aqueous phase. The aqueous phase was removed and extracted two
more times with a 11% Triton X-114 detergent solution (2 × 1 mL).
The detergent phases were combined and washed three times with fresh
buffer (3 × 7 mL). The resulting extracted solution was diluted to 30
mL with fresh buffer, and the sample was purified by DEAE ion
exchange chromatography following a standard protocol: After removal
of the Triton X-114 by purging the column with buffer, the labeled
protein was eluted in 2 mL portions using a NaCl-gradient (0 M to 1
M NaCl within 25 min with a flow rate of of 1 mL/min). Fractions
containing the desired protein were combined, concentrated, and
desalted using a Vivaspin protein concentrator with a 10,000 Da
exclusion membrane. The protein was then snap frozen and stored at
-80 °C. Yields for each coupling reaction are given in Table 2.
Proof for the specific coupling of the lipopeptides to the proteins
was obtained by SDS-PAGE gel electrophoresis, as well as by
electrospray mass spectrometry of the intact labeled protein. Further
proof for the specific incorporation of the lipopeptide onto cysteine
residue 181 was obtained by digestion of three of the protein samples
(H-Ras (1-181) + peptide 41c, H-Ras (1-181) + peptide 43c, H-Ras
(1-181) + peptide 44c) with trypsin and chymotrypsin proteases
followed by mass spectrometry of the individual fragments. This final
result clearly showed incorporation of the lipopeptide tail to the last
cysteine Cys181, thereby proving the specific incorporation of the
peptide group.
Microinjection Experiments. PC12 cell culture and microinjections
were performed as described.29 Briefly, PC12 cells were prestimulated
with NGF (100 ng/mL) for 3 days and kept in NGF-free medium for
another 2 days. Ras proteins were diluted to 150 µM in PBS containing
10 µM fluorescein dextran as a marker for identification. 40 h after
injection the transformed cells showed a typical differentiated phenotype
with neurite outgrowth.
(26%) of the desired material as a white solid. Rf ) 0.3 (hexane/ethyl
1
acetate 2:3). [R]22 ) -3.1 (c ) 1.0, CHCl3). H NMR (250 MHz,
D
CDCl3): δ 3.74 (s, 3H, OCH3); 3.65 (dd, J ) 4.7, 7.3 Hz, 1H, R-CH
Cys); 2.91 (dd, J ) 4.6, 13.5 Hz, 1H, â-CH2a Cys); 2.76 (dd, J ) 6.3,
13.5 Hz, 1H, â-CH2b Cys); 2.51 (t, J ) 7.4 Hz, 2H, R-CH2 HD); 2.08
(s, br, 2H, NH2); 1.50-1.59 (m, 2H, â-CH2 HD); 1.24 (s, br, 26H,
(CH2)13 HD); 0.86 (t, J ) 7.0 Hz, 3H, ω-CH3 HD). HRMS (EI) m/z:
calcd for (M+) C20H41NO2S 359.2858, found 359.2842. Anal. Calcd:
C, 66.80; H, 11.49; N, 3.89. Found: C, 67.20; H, 11.25; N, 3.41.
Coupling to Ras. Prior to the coupling reaction, a sample of the
mutant Ras protein (1-181) was passed through a desalting column to
remove any excess salts and DTE required for storage of the protein.
At all stages of the coupling reaction, all samples containing protein,
unless otherwise stated, were kept at or below 4 °C. To a solution of
the desired peptide dissolved in methanol (20 µL) in a 2 mL Eppendorf
tube was added a solution of 11% Triton X-114 in water (1 mL). The
solution was sonicated for approximately 10 min until a slightly cloudy,
homogeneous solution was obtained. The detergent solution was cooled
to 0 °C, and an aqueous solution (buffer: 20 mM Tris/HCl, 5 mM
MgCl2, pH 7.4) containing the Ras protein was added to the peptide
solution (approximately a 1:1 solution of the peptide- and protein-
Acknowledgment. We are grateful to D. Vogt for cloning
of the H-Ras mutants, C. Nowak for help in protein purification
and Dr. H. Prinz for mass spectrometric analyses of the proteins.
This research was supported by Deutsche Forschungsgemein-
schaft and the Fonds der Chemischen Industrie. D.J.O. thanks
the Alexander von Humboldt Foundation for a Postdoctoral
Fellowship.
Supporting Information Available: Experimental proce-
dures, spectroscopic and analytical data for all compounds
(PDF). This material is available free of charge via the Internet
at http://pubs.acs.org.
JA002723O