Photoactivatable synthetic Ras proteins: "Baits" for the identification of plasma-membrane-bound binding partners of Ras

Article abstract of DOI:10.1002/1521-3773(20020715)41:14<2546::AID-ANIE2546>3.0.CO;2-E

The combination of organic synthesis with cell biology provides access to a chimera of the signal-transducing Ras protein, which contains a photoactivatable benzophenone group in its isoprenoid membrane anchor, yet retains biological activity. The semisyn

Full text of DOI:10.1002/1521-3773(20020715)41:14<2546::AID-ANIE2546>3.0.CO;2-E

COMMUNICATIONS
Polymerization reactions: A 1-L glass autoclave was charged with toluene
(200 mL) and triisobutylaluminum (0.5 mL). The mixture was stirred
(700 rpm), thermostated at 258C, and saturated with ethylene (2 bar). The
polymerization reaction was started by the injection of a solution of the
respective zwitterionic complex in toluene, generated in situ by treatment
of 8a or 8b (ca. 35 mmol) with an equimolar amount of B(C6F5)3 in toluene
(8 mL). After 1 h the reaction was quenched by adding aqueous HCl in
methanol (15 mL). The polymer was precipitated with additional methanol
(100 mL), collected by filtration, washed with 6n aqueous HCl (100 mL),
water (200 mL), acetone (50 mL), and then dried in vacuo.
2005 2007; B. L. Small, M. Brookhart, J. Am. Chem. Soc. 1998, 120,
7143 7144; S. A. Svejda, L. K. Johnson, M. Brookhart, J. Am. Chem.
Soc. 1999, 121, 10634 10635; NMR characterization: G. B. Galland,
R. F. de Souza, R. S. Mauler, F. F. Nunes, Macromolecules 1999, 32,
1620 1625.
[17] Programs used: data collection COLLECT (B. V. Nonius, 1998), data
reduction Denzo-SMN (Z. Otwinowski, W. Minor, Methods in
Enzymol. 1997, 276, 307 326), absorption correction SORTAV
(R. H. Blessing, Acta Crystallogr. Sect. A 1995, 51, 33 37 ; R. H.
Blessing, J. Appl. Crystallogr. 1997, 30, 421 426), structure solution
SHELXS-97(G. M. Sheldrick, Acta Crystallogr. Sect. A 1990, 46, 467
473), structure refinement SHELXL-97 (G. M. Sheldrick, Universit‰t
Gˆttingen, 1997), graphics SCHAKAL (E. Keller, Universit‰t Frei-
burg, 1997).
Received: March 20, 2002 [Z18937]
[1] H.-H. Brintzinger, D. Fischer, R. M¸lhaupt, B. Rieger, R. M. Way-
mouth, Angew. Chem. 1995, 107, 1255 1283; Angew. Chem. Int. Ed.
Engl. 1995, 34, 1143 1170.
[2] G. J. P. Britovsek, V. C. Gibson, D. F. Wass, Angew. Chem. 1999, 111,
448 468; Angew. Chem. Int. Ed. 1999, 38, 428 447; S. D. Ittel, L. K.
Johnson, M. Brookhart, Chem. Rev. 2000, 100, 1169 1203.
[3] X. Yang, C. L. Stern, T. J. Marks, J. Am. Chem. Soc. 1994, 116, 10015
10031.
[4] H. Sinn, W. Kaminsky, Adv. Organomet. Chem. 1980, 18, 99 149; X.
Yang, C. L. Stern, T. J. Marks, J. Am. Chem. Soc. 1991, 113, 3623
3625; J. C. W. Chien, W.-M. Tsai, M. D. Rausch, J. Am. Chem. Soc.
1991, 113, 8570 8571; M. Bochmann, A. J. Jaggar, J. C. Nicholls,
Angew. Chem. 1990, 102, 830 832; Angew. Chem. Int. Ed. Engl. 1990,
29, 780 782; review: E. Y.-X. Chen, T. J. Marks, Chem.Rev. 2000, 100,
1391 1434.
[5] B. Temme, G. Erker, J. Karl, H. Luftmann, R. Frˆhlich, S. Kotila,
Angew. Chem. 1995, 107, 1867 1869; Angew. Chem. Int. Ed. Engl.
1995, 34, 1755 1757; review: G. Erker, Acc. Chem. Res. 2001, 34, 309
317.
[6] M. Dahlmann, G. Erker, M. Nissinen, R. Frˆhlich, J. Am. Chem. Soc.
1999, 121, 2820 2828; M. Dahlmann, G. Erker, K. Bergander, J. Am.
Chem. Soc. 2000, 122, 7986 7998.
[7] T. M. Kooistra, Q. Knijnenburg, J. M. M. Smits, A. D. Horton, P. H. M.
Budzelaar, A. W. Gal, Angew. Chem. 2001, 113, 4855 4858; Angew.
Chem. Int. Ed. 2001, 40, 4719 4722.
[8] V. C. Gibson, M. J. Humphries, K. P. Tellmann, D. F. Wass, A. J. P.
White, D. J. Williams, Chem. Commun. 2001, 2252 2253.
[9] Electron transfer had previously been observed or used in activation
processes of metallocene Ziegler Natta catalysts: R. F. Jordan, Adv.
Organomet. Chem. 1991, 32, 325 387.
[10] For recent evidence of electron-transfer processes occurring in
B(C6F5)3 activation chemistry see, for example: C. J. Harlan, T.
Hascall, E. Fujita, J. R. Norton, J. Am. Chem. Soc. 1999, 121, 7 27 4
7275; C. J. Beddows, A. D. Burrows, N. G. Connelly, M. Green, J. M.
Lynam, T. J. Paget, Organometallics 2001, 20, 231 233; R. J. Kwaan,
C. J. Harlan, J. R. Norton, Organometallics 2001, 20, 3818 3820.
Photoactivatable Synthetic Ras Proteins:
™Baits∫ for the Identification of Plasma-
Membrane-Bound Binding Partners of Ras**
J¸rgen Kuhlmann,* Andreas Tebbe, Martin Vˆlkert,
Melanie Wagner, Koji Uwai, and Herbert Waldmann*
The regulation of cell growth and differentiation by
proteins of the Ras superfamily[1] requires the correct
subcellular distribution of the small GTP-binding proteins
(GTP ˆ guanosine triphosphate).[2] The biological function of
Ras is strictly dependent on its correct translocation to the
plasma membrane, and this localization is directly linked to
posttranslational S-farnesylation and S-palmitoylation of Ras.[3]
The real mechanism of translocation is still the subject of
debate. On the one hand for K-RasB, which embodies a
polycationic hexalysine stretch and a farnesyl thioether at the
C terminus, a model was proposed in which unspecific but
highly anionic ™sites∫ (formed at least in part by the lipid
bilayer) at the plasma membrane instead of a classical specific
proteinaceous receptor are responsible for association of this
Ras isoform with the plasma membrane.[4] On the other hand
for H- and N-Ras, which have S-farnesyl and S-palmitoyl
substituents at the C terminus, a membrane-trapping model[5]
was postulated in which a prenyl protein specific palmitoyl-
[11] For
a formally related homogeneous alkyl-free Group 4 metal
[*] Dr. J. Kuhlmann, Dipl.-Biochem. A. Tebbe,
Dipl.-Biochem. M. Wagner
Ziegler Natta catalyst see: J. Cano, P. Royo, M. Lanfranchi, M. A.
Pellinghelli, A. Tiripicchio, Angew. Chem. 2001, 113, 2563 2565;
Angew. Chem. Int. Ed. 2001, 40, 2495 2497, J. Jin, D. R. Wilson,
E. Y.-x. Chen, Chem. Commun. 2002, 708 709, and references
therein.
Max-Planck-Institut f¸r molekulare Physiologie
Abteilung Strukturelle Biologie
Otto-Hahn-Strasse 11, 44227Dortmund (Germany)
Fax : (‡49)231-133-1435
[12] H. Yasuda, Y. Kajihara, K. Mashima, K. Nagasuna, K. Lee, A.
Nakamura, Organometallics, 1982, 1, 388 396, and references therein.
[13] J. C. M. Sinnema, G. H. B. Fendesak, H. tom Dieck, J. Organomet.
Chem. 1990, 390, 237 250.
[14] A. G. Massey, A. J. Park, J. Organomet. Chem. 1964, 2, 245 250;
A. G. Massey, A. J. Park in Organometallic Synthesis, Vol. 3, (Eds.:
R. B. King, J. J. Eisch), Elsevier, New York, 1986, p. 461.
[15] For structurally related metallocene complexes with metals of Group
4 see: J. Karl, G. Erker, R. Frˆhlich, J. Organomet. Chem. 1997, 535,
59 62; M. Dahlmann, G. Erker, R. Frˆhlich, O. Meyer, Organo-
metallics 2000, 19, 2956 2967.
E-mail: juergen.kuhlmann@mpi-dortmund.mpg.de
Prof. Dr. H. Waldmann, Dipl.-Chem. M. Vˆlkert, Dr. K. Uwai
Max-Planck-Institut f¸r molekulare Physiologie
Abteilung Chemische Biologie
Otto-Hahn-Strasse 11, 44227Dortmund (Germany)
and
Fb. 3, Organische Chemie, Universit‰t Dortmund
Otto-Hahn-Strasse 6, 44227Dortmund (Germany)
Fax : (‡49)231-133-2499
E-mail: herbert.waldmann@mpi-dortmund.mpg.de
[**] We thank Christine Nowak for excellent technical assistance, Lilianna
Wielitzek for help with the UV exposure system, and Fred Wit-
tinghofer for continuous encouragement. The work was supported by
the Fonds der Chemischen Industrie and the Alexander von Hum-
boldt Foundation.
[16] The resulting polyethylene contains mostly methyl branches, as
expected: L. K. Johnson, C. M. Killian, M. Brookhart, J. Am. Chem.
Soc. 1995, 117, 6414 6415; C. M. Killian, D. J. Tempel, L. K. Johnson,
M. Brookhart, J. Am. Chem. Soc. 1996, 118, 11664 11665; C. M.
Killian, L. K. Johnson, M. Brookhart, Organometallics 1997, 16,
2546
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4114-2546 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 14

COMMUNICATIONS
transferase (PTase) is located in the plasma membrane.
However, these proposals have been challenged by recent
findings that suggest that K-Ras and H-/N-Ras traffic to the
cell surface through different routes and that palmitoyl
substitution of the H- and N-isoform occurs in the endoplas-
matic reticulum.[6]
lipopeptide to a Ras mutant with a C-terminal cysteine
residue.[15, 16] Such peptides were required to demonstrate
that the conjugate is selectively directed to the plasma
membrane and that the photophore does not significantly
influence the binding to a putative Ras binding protein, if
at all.
In addition, the correlation of subcompartmentation of the
plasma membrane[7] with signal transduction events has
received increasing attention. Thus, H-Ras is enriched in
membrane domains with high levels of sphingolipids and
cholesterol (rafts),[8] whereas K-Ras shows a random distri-
bution in the plasma membrane.[6] To unravel the mechanisms
that direct Ras localization and to identify putative binding
partners of Ras in the plasma membrane or other intracellular
compartments, new tools that go beyond the techniques
available from regular biological methodology alone are
urgently needed.
These criteria are met by the peptides 4a and 4b, which
were synthesized as shown in Scheme 1. The N-terminal
hexapeptide of the Ras sequence was synthesized efficiently
Pro
SPPS
H-Gly-Cys-Met-Gly-Leu-Pro
S
1
S
In addressing this challenge, we have synthesized Ras
lipopeptides and a Ras protein that incorporate a photo-
activatable isoprenoid thioether. Our evaluations demon-
strated that the lipopeptides will be suitable ™baits∫ for
covalent coupling of proteins, whereas the protein construct
demonstrated biological activity required to trap Ras binding
partners.
As suitable chemical tools for covalent trapping and
identification of possible membrane-embedded binding part-
ners of the Ras-proteins, lipopeptides were considered that
meet the following requirements:
O
a) or b)
c)
OH
O
5
R-Gly-Cys-Met-Gly-Leu-Pro-OH
SStBu
e) f)
2a,b
O
H2N
OMe
d)
SGer-O-pBP
3
1. They should embody the characteristic C-terminal amino
acid sequence of the Ras protein that is in contact with the
membrane and most likely also with a putative binding
partner. To fulfill this criterion, the C-terminal heptapep-
tide sequence of N-Ras was chosen which terminates in a
S-farnesyl-substituted cysteine methyl ester.[9]
2. The peptides should be selectively localized to and stably
inserted into the plasma membrane of cells, in our case by
insertion of a palmitoylatable cysteine residue that is
located at the C terminus of N-Ras. This cysteine was
masked as a tert-butyl disulfide, which can be cleaved
readily by treatment with dithiothreitol. The cysteine thiol
functional group liberated is readily derivatized with a
palmitoyl group, and the resulting S-palmitoyl- and S-far-
nesyl-substituted Ras peptide is selectively localized at the
plasma membrane.[10, 11]
3. The conjugate should contain a group that can be photo-
activated,[12] preferably incorporated into the membrane-
embedded isoprenoid residue. For this purpose the benzo-
phenone group (BP) was chosen as an established photo-
phore.[13, 14]
4. They should carry a group that allows the isolation and
detection of the covalently crosslinked products by estab-
lished methods. For this purpose the biotin label was
chosen because it can serve both as a tag for product
enrichment and as a detection label.
Biot-Aca-OH:
R-Gly-Cys-Met-Gly-Leu-Pro-Cys-OMe
O
NH
SStBu
S
O
O
O
HN
S
N
OH
OH
4
H
O
MIC-OH:
O
4a,b
N
2a, 4a: R = Biot-Aca
2b, 4b: R = MIC
O
O
Scheme 1. Solid-phase synthesis of the hexapeptides 2a/b and fragment
coupling with the BP-labeled isoprenyl-derivatized cysteine 3. a) Biot-Aca-
OH, HBTU, HOBt, DIPEA, DMF; b) MIC-OH, HBTU, HOBt, DIPEA,
DMF; c) TFA in CH2Cl2 (10%), 2a: 62%, 2b: 33%; d) 3, EDCI, HOBt,
CH2Cl2, 4a: 43%, 4b: 55%; e) NCS, DMS, CH2Cl2; f) CysOMe ¥ HCl, NH3,
MeOH, 66% over two steps. BP ˆ benzophenone, SPPS ˆ solid-phase
peptide synthesis, DEAD ˆ diethylazodicarboxylate, HBTU ˆ 2-(1H-ben-
zotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate, HOBtˆ
1-hydroxybenzotriazole, DIPEA ˆ diisopropylethylamine, DMF ˆ N,N-di-
methylformamide, DMS ˆ dimethyl sulfide, EDCIˆ 1-ethyl-3-(3-dimethyl-
aminopropyl)carbodiimide hydrochloride, NCS ˆ N-chlorosuccinimide.
on a solid support. The peptide was linked to the polymeric
carrier through the chlorotrityl (chlorotriphenylmethyl) link-
er, and Fmoc-protected amino acids were employed for chain
elongation. After assembly of the hexapeptide and N-termi-
nal deprotection, either biotinylaminocaproic acid (Biot-Aca-
OH) or maleimidocaproic acid (MIC-OH) were introduced
followed by release of the Biot-Aca- and MIC-labeled
peptides 2a and 2b from the solid support. Both compounds
were isolated in high yields and then coupled with the
S-alkylated cysteine methyl ester 3 to yield the target peptides
4a and 4b. In 3 the terminal isoprene unit of the farnesyl
5. The synthesis strategy should give access to photoactivat-
able lipopeptides for the labeling experiments and to
protein lipopeptide conjugates, which can be used to
determine the activity of the lipopeptides. The maleimi-
docaproic acid (MIC) group was chosen to couple the
Angew. Chem. Int. Ed. 2002, 41, No. 14
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4114-2547 $ 20.00+.50/0
2547

COMMUNICATIONS
group has been replaced by the benzophenone photophore,
which approximates it in size and lipophilicity. The lipid-
modified and benzophenone-labeled cysteine methyl ester 3
was synthesized by conversion of allyl alcohol 5[17] into the
corresponding allyl chloride and subsequent nucleophilic
substitution with cysteine methyl ester (Scheme 1).
residue at position 181 of the N-Ras
mutant proceeded almost quantitatively.
N-RasG12VD181 terminates in a
free cysteine residue, which selectively
reacts with the MIC group. The cou-
pling product was extracted by treat-
ment with Triton-X114 and purified by
ion-exchange chromatography. Gel
electrophoresis gave a single band for
the hydrophobic protein (Figure 2).
Mass spectrometric analysis (Figure 3)
gave rise to signals for the coupling
product without the S-tert-butyl pro-
tecting group at the cysteine group,
which can be substituted with a palmi-
The synthesis strategy is both efficient and flexible. The
chosen combination of solid-phase and solution synthesis and
the assembly of the target peptides from building blocks that
can be varied readily opens up the opportunity to generate a
variety of further analogues rapidly if required. Furthermore,
variation of the photophore is straightforward. Thus, a variety
of suitable tools for further biological investigations can be
generated efficiently by means of this approach. To inves-
tigate if the benzophenone isoprenoid groups could be
applied to the photoaffinity labeling of Ras interaction
partners, we performed exposure experiments with solutions
of lipopeptide 4a (1.6 mm) and bovine serum albumin (BSA;
8 mm). The reaction mixture was exposed to UV light and
samples were taken before exposure to UV light, and after 5,
15, 30, and 60 min of exposure. The time course of the reaction
was analyzed by SDS-PAGE and Western Blot. Biotinylated
BSA was detected by incubation with a streptavidine alka-
line phosphatase conjugate and subsequent enzymatic reac-
tion. The benzophenone moiety was stable in the absence of
UV irradiation, whereas exposure to 300–400-nm light led to
rapid photoactivation and coupling of the lipopeptide to BSA
(Figure 1). Five minutes of exposure to UV light were
sufficient for maximum biotinylation of the BSA sample.
Figure 2. Characteri-
zation of the coup-
ling
product
of
N-RasG12VD181 with
lipopeptide 4a by SDS-
PAGE. SDS-PAGE of
the coupling product
shows a protein band
with an apparent mass
of 22 kDa and a purity
>90%.
toyl
group
(N-RasG12VCysGerBP
,
21659 Da), and a smaller signal for
the coupling product with the protect-
ing group (21747 Da). These data
verify a 1:1 stoichiometry for the
protein lipopeptide coupling product. Specific reaction of
MIC lipopeptide 4b with the C-terminal cysteine (Cys181) of
N-RasG12VD181 was validated by tryptic digestion of the
chimeric protein with trypsin and chymotrypsin. ESI-MS and
MALDI-TOF analysis of the proteolytic products excluded
significant formation of lipopeptide thioethers with Cys51,
Cys80, or Cys118 of N-RasG12VD181 (data not shown).
Figure 1. Coupling of biotinylated lipopeptide with BSA. A solution of
lipopeptide 4a (10 mm) and a fivefold excess of BSA were exposed to UV
light, and samples were taken before (1, 2) and 5 min (3, 4), 15 min (5, 6),
30 min (7, 8), and 60 min (9, 10) after start of irradiation. Aliquots
corresponding to 3 mg (1, 3, 5, 7, 9) and 5mg (2, 4, 6, 8, 10) of BSA were
applied in SDS-PAGE (A) and Western Blot (B).
Figure 3. Characterization of the coupling product of N-RasG12VD181
with lipopeptide 4a by ESI MS. Mass spectra of N-RasG12VD181 (black)
and the coupling product of N-RasG12VD181 with lipopeptide 4a (gray).
The base peak in the spectra corresponds to the theoretical mass of the
N-RasG12VD181 that has been truncated at the C terminus (20440 Da).
The minor peak at 20311 Da can be assigned to N-RasG12VD181 without
the N-terminal Met1 (theoretical mass: 20309 Da). The MS spectrum of
the coupling product shows a major peak at 21659 Da and a minor peak at
21747 Da. These signals match the masses of the coupling product
N-RasG12VD181 4a after and before removal of the S-tert-butyl protect-
ing group, respectively.
Utilization of lipopeptides with a benzophenone isoprenoid
group for the affinity labeling of Ras-interaction proteins
requires that the replacement of the farnesyl group by a
geranyl benzophenone moiety does not prevent the recog-
nition of the lipopeptides by receptors or modifying enzymes
for Ras. We generated a lipoprotein by coupling oncogenic
N-Ras mutant N-RasG12VD181 (which has a truncated
C terminus) with lipopeptide 4b (which corresponds to the
native C-terminus of N-Ras with a free palmitoylation site and
a geranyl benzophenone instead of the farnesyl thioether at
the last amino acid residue) as shown in Scheme 1. The reaction
of the MIC group of the lipopeptide with the free cysteine
The rat pheochromocytoma cell line PC12 can be induced
to differentiate by oncogenic Ras mutants.[18] This effect can
be correlated to the transforming potential of Ras mutants in
a microinjection assay.[19] To ensure that benzophenone
lipopeptides are applicable as photoaffinity probes for the
identification of Ras receptors and modifiers, we analyzed the
2548
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4114-2548 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 14

COMMUNICATIONS
differentiating potential of the oncogenic coupling product of
N-RasG12VD181 and the benzophenone lipopeptide 4b and
compared it with the efficiency of N-RasG12V full-length
protein and the ™natural∫ chimera with a C-terminal farnesyl
thioether (N-RasG12VCysFar) to induce the corresponding
morphological pattern in PC12 cells. Gratifyingly, the lip-
oprotein with the benzophenone isoprenoid moiety induced
neurite-like outgrowths in PC12 cells (Figure 4), thus indicat-
ing biological activity of the protein. This result implies that
the construct with a geranyl benzophenone thioether and a
carboxymethylation at the C terminus is sufficient to address
the oncogenic chimera to its functional location, which
includes palmitoylation of the free cysteine of the lipopeptide
moiety.
Figure 5. Differentiation efficiency of oncogenic N-Ras constructs. PC12
cells were microinjected with
a protein solution that contained
CysFar
N-RasG12VCysGerBP
(
~
), N-RasG12V
^
( ), and full-length N-RasG12V
&
( ) in the given concentrations.
Studies with green fluorescent protein constructs of Ras
proteins have promoted the idea that it is the hydrophobically
modified C terminus of the Ras isoforms that determines
targeting and localization,[6, 20] whereas the G domain is less
important for targeting the protein. The Ras C terminus is
flexible and was not resolved in the crystal structures of the
protein[21, 22] and its complexes[23] nor in the NMR spectro-
scopic analysis of H-Ras.[24] Therefore, it is most probable that
affinity tags presenting the Ras C terminus are sufficient for
recognition by and binding to putative receptors and modi-
fiers.
By combining the power of organic synthesis with cell
biology numerous synthetic lipopeptide and lipoprotein
™baits∫ are accessible. For instance different natural or non-
natural hydrophic modifications, alternative affinity tags, or
radiolabels may be introduced or the entire peptide backbone
may be replaced by a nonpeptide scaffold, thereby opening up
a multitude of possible applications. The combination of
chemical synthesis, cell biology, and elaborated biochemical
and mass spectrometric approaches delineated herein should
provide a new powerful approach to solve some of the most
prevailing problems and answer open questions concerning
signal transduction through Ras proteins.
Figure 4. Differentiation of PC12 cells by N-RasG12VCysGerBP. Overlay of a
phase contrast and fluorescence microscopy image of PC12 cells after co-
injection of 100 mm oncogenic Ras construct and FITC-dextran. Only cells
with a fluorescent marker (dark cell body, white arrowheads) have formed
neurite-like outgrowths (black-bordered white arrows). FITC ˆ fluorescein
isothiocyanate.
The data show that the potential of the benzophenone
derivative to induce neurite-like outgrowths is approximately
one third of the values recorded for N-RasG12V (full length)
or N-RasG12VCysFar (Figure 5). This activity is sufficient to
allow the successful execution of biological experiments,
including photoactivation studies.
Received: January 7, 2002 [Z18487]
[1] L. Van Aelst, M. Barr, S. Marcus, Polverino, A. M. Wigler, Proc. Natl.
Acad. Sci. USA 1993, 90, 6213 6217.
[2] A. Wittinghofer, Biol. Chem. 1998, 379, 933 937.
[3] A. Wittinghofer, H. Waldmann, Angew. Chem. 2000, 112, 4360 4383;
Angew. Chem. Int. Ed. 2000, 39, 4193 4214.
[4] M. O. Roy, R. Leventis, J. R. Silvius, Biochemistry 2000, 39, 8298
8307.
[5] S. Shahinian, J. R. Silvius, Biochemistry 1995, 34, 3813 3822.
[6] A. Apolloni, I. A. Prior, M. Lindsay, R. G. Parton, J. F. Hancock, Mol.
Cell. Biol. 2000, 20, 2475 2487.
[7] D. A. Brown, E. London, J. Biol. Chem. 2000, 275, 17221 17224.
[8] S. Roy, R. Luetterforst, A. Harding, A. Apolloni, Etheridge, M, E.
Stang, B. Rolls, J. F. Hancock, R. G. Parton, Nat. Cell Biol. 1999, 1,
98 105.
[9] J. F. Hancock, A. I. Magee, J. E. Childs, C. J. Marshall, Cell 1989, 57,
1167 1177.
[10] H. Waldmann, M. Schelhaas, E. Nagele, J. Kuhlmann, A. Wittinghof-
er, Schroeder, J. R. Silvius, Angew. Chem. 1997, 109, 2334 2337;
Angew. Chem. Int. Ed. Engl. 1997, 36, 2238 2241.
Our results demonstrate that introduction of a benzophe-
none group as a photoactivatable ligand in the isoprenoid
moiety of Ras-lipopeptides does not obstruct the ability of a
corresponding oncogenic Ras-lipoprotein chimera to differ-
entiate PC12 cells. Furthermore, photochemical activation of
the benzophenone can be initiated by exposure to near-UV
light (350 360 nm), thus limiting photodamage of the protein
targets. Therefore, probes such as peptide 4a with the
C-terminal peptide sequence of Ras-proteins, a benzophe-
none isoprenoid substitution of the farnesyl group, and a
suitable affinity tag (e.g. biotin) are promising tools for the
isolation of Ras-modifying enzymes and Ras-interaction
partners, in particular of plasma-membrane-localized pro-
teins.
Angew. Chem. Int. Ed. 2002, 41, No. 14
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4114-2549 $ 20.00+.50/0
2549

COMMUNICATIONS
study,[5] all experimental information on BiH3 (synthesis,
isolation, and vapor pressure) originates from the report by
Amberger in 1961.[2] To the best of our knowledge nobody has
to date been able to repeat Amberger×s (admittedly pre-
sumptuous) synthesis of BiH3 by decomposition of CH3BiH2
at À55 to À458C, or to develop an alternative route to this
compound. Hence, unlike the case of the short-lived mono-
hydride BiH,[6] neither the structure nor the vibrational
spectrum of BiH3 have been determined experimentally.
We report herein the successful repetition of Amberger×s
synthesis of BiH3 and its unambiguous characterization by
independent, modern spectroscopic methods, which are
supported by ab initio caclulations. Our particular interest
in BiH3 concerns those structural and spectroscopic properties
that are expected to be unique for BiH3. First, BiH3 should
have the smallest H-X-H bond
[11] H. Schroeder, R. Leventis, S. Rex, M. Schelhaas, E. Nagele, H.
Waldmann, J. R. Silvius, Biochemistry 1997, 36, 13102 13109.
[12] F. Kotzyba-Hibert, I. Kapfer, M. Goeldner, Angew. Chem. 1995, 107,
1391 1408; Angew. Chem. Int. Ed. Engl. 1995, 34, 1296 1312.
[13] T. A. Kale, C. Raab, N. Yu, D. E. Dean, M. D. Distefano, J. Am. Chem.
Soc. 2001, 123, 4373 4381.
[14] G. Dorman, G. D. Prestwich, Trends Biotechnol. 2000, 18, 64 7 7 .
[15] B. Bader, K. Kuhn, D. J. Owen, H. Waldmann, A. Wittinghofer, J.
Kuhlmann, Nature 2000, 403, 223 226.
[16] K. Kuhn, D. J. Owen, B. Bader, A. Wittinghofer, J. Kuhlmann, H.
Waldmann, J. Am. Chem. Soc. 2001, 123, 1023 1035.
[17] D. M. Marecak, Y. Horiuchi, H. Arai, M. Shimonaga, Y. Maki, T.
Koyama, K. Ogura, G. D. Prestwich, Bioorg. Med. Chem. Lett. 1997, 7,
1973 1978.
[18] D. Bar-Sagi, J. R. Feramisco, Cell 1985, 42, 841 848.
[19] R. H. Cool, G. Schmidt, C. U. Lenzen, H. Prinz, D. Vogt, A.
Wittinghofer, Mol. Cell. Biol. 1999, 19, 6297 6305.
[20] E. Choy, V. K. Chiu, J. Silletti, M. Feoktistov, T. Morimoto, D.
Michaelson, I. E. Ivanov, M. R. Philips, Cell 1999, 98, 69 80.
[21] A. M. de Vos, L. Tong, M. V. Milburn, P. M. Matias, J. Jancarik, S.
Noguchi, S. Nishimura, K. Miura, E. Ohtsuka, S. H. Kim, Science 1988,
239, 888 893.
angle (ca. 90.08) of any of the
hydrides of the composition
[22] E. F. Pai, W. Kabsch, U. Krengel, K. C. Holmes, J. John, A.
Wittinghofer, Nature 1989, 341, 209 214.
XH2 and XH3 (Figure 1). In
the vibrational ground state,
BiH3 should thus be an oblate
[23] K. Scheffzek, M. R. Ahmadian, W. Kabsch, L. Wiesmuller, A.
Lautwein, F. Schmitz, A. Wittinghofer, Science 1997, 277, 333 338.
[24] K. Yamasaki, G. Kawai, Y. Ito, Y. Muto, J. Fujita, T. Miyazawa, S.
Nishimura, S. Yokoyama, Biochem. Biophys. Res. Commun. 1989, 162,
1054 1062.
symmetric top (Ia ˆ Ib < Ic)
extremely close to the spheri-
cal top (Ia ˆ Ib ˆ Ic) limit, even
more so than SbH3.[7] In vibra-
tionally excited states it may
switch to a prolate symmetric
Figure 1. Molecular structure
of BiH3 with principal axes of
top with Ia < Ib ˆ Ic. Therefore
inertia indicated; ae, re ˆ bond
it should reveal, with unprece-
dented distinction, particular
ground and excited state rota-
angle and length in the equili-
brium structure.
Bismuthine BiH3: Fact or Fiction?
High-Resolution Infrared, Millimeter-Wave,
and Ab Initio Studies**
tional energy patterns that require appropriate reductions of
the rotational vibrational Hamiltonian.[8] Owing to the near-
rectangular H-Bi-H angle, the heavy central atom, the
expected small HBi/BiH' coupling, and the large anharmo-
nicity of the BiH stretching motion, BiH3 would also be a
prototype molecule for local mode behavior.[9]
Wolfgang Jerzembeck, Hans B¸rger,*
¡
Lucian Constantin, Laurent Margules, Jean Demaison,
J¸rgen Breidung, and Walter Thiel
After numerous failures we were able to repeat the
reported synthesis[2] and eventually obtained apparently pure
BiH3 in quantities sufficient to enable us to carry out gas-
phase infrared (IR) and millimeter-wave (MMW) measure-
ments over periods of minutes to hours. In brief, we first
prepared Bi(CH3)3 from BiCl3 and CH3MgI, and then reacted
this with BiCl3 to give CH3BiCl2 by using standard proce-
dures.[10] CH3BiCl2 was then reduced with LiAlH4 in di-n-
butyl ether at À788C to give CH3BiH2 whose disproportio-
nation at À55 to À458 C yielded BiH3 along with methyl-
bismuthanes. Some hydrogen, which formed by decomposi-
tion of the hydrides, was pumped off while the reaction
mixture was cooled by using liquid nitrogen. Thereafter this
mixture was allowed to warm to about À508 C and volatile
material expanded into cooled absorption cells until a total
vapor pressure of between 10 and 100 Pa was reached.
IR spectra were recorded in the region for stretching
fundamentals n1(A1)/n3(E) at about 1700 cmÀ1 with a reso-
lution of 4.4 Â 10À3 cmÀ1 (Figure 2), and in the region for
bending modes n2(A1)/n4(E) at about 750 cmÀ1 with a
resolution of 6.6 Â 10À3 cmÀ1 (Figure 3).[11] An external dou-
Standard Inorganic Chemistry textbooks[1] report that
bismuthine, BiH3, is a common although unstable Group 15
hydride. Typically only the boiling point[2] of ‡16.88C is
mentioned for characterization. Apart from the early obser-
vation of a ™volatile bismuth hydride∫ by Paneth in 1918,[3] its
relevance in analytical chemistry,[4] and a mass spectrometric
[*] Prof. Dr. H. B¸rger, Dr. W. Jerzembeck
Anorganische Chemie
Bergische Universit‰t Wuppertal
42097Wuppertal (Germany)
Fax : (‡49)202-439-2901
E-mail: buerger1@uni-wuppertal.de
¡
Dr. L. Constantin, Dr. L. Margules, Dr. J. Demaison
¬
Laboratoire de Physique des Lasers, Atomes et Molecules
¬
Universite de Lille 1
59655 Villeneuve d×Ascq (France)
Dr. J. Breidung, Prof. Dr. W. Thiel
Max-Planck-Institut f¸r Kohlenforschung
45470 M¸lheim (Germany)
[**] We thank the German French PROCOPE program and the PICS
599 project for support. We also thank Prof. H. Stoll, Stuttgart, for
helpful discussions.
2550
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4114-2550 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 14