Synthesis and biological evaluation of aeroplysinin analogues: a new class of receptor tyrosine kinase inhibitors

Article abstract of DOI:10.1016/S0968-0896(98)00070-4

Receptor tyrosine kinases (RTKs), such as the epidermal growth factor receptor (EGFR) and the platelet-derived growth factor receptor (PDGFR), are critically involved in the transduction of mitogenic signals across the plasma membrane and therefore in the regulation of cell growth and proliferation. Enhanced RTK activity is associated with proliferative diseases such as cancer, psoriasis and atherosclerosis, while decreased function may be associated for instance with diabetes. EGFR and PDGFR are selectively inhibited by analogues of the marine natural product aeroplysinin. The synthetic inhibitors display IC₅₀ values in the low micromolar range and in contrast to the natural product show pronounced inhibitory activity in cultured cells in vivo. The mechanism of inhibition is likely based on a covalent modification of the target enzymes by reaction of epoxy ketone 8 with various nucleophiles. Copyright (C) 1998 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(98)00070-4

BIOORGANIC &
MEDICINAL
CHEMISTRY
Bioorganic & Medicinal Chemistry 6 (1998) 1153±1162
Synthesis and Biological Evaluation of Aeroplysinin
Analogues: a New Class of Receptor Tyrosine Kinase
Inhibitors
Klaus Hinterding,a Axel Knebel,b Peter Herrlichb and Herbert Waldmanna,*
aUniversitat Karlsruhe, Institut fur Organische Chemie, Richard-Willstatter-Allee 2, D-76128 Karlsruhe, Germany
È
È
È
bForschungszentrum Karlsruhe GmbH, Institut fur Genetik, D-76344 Eggenstein-Leopoldshafen, Germany
È
Received 15 January 1998; accepted 6 March 1998
AbstractÐReceptor tyrosine kinases (RTKs), such as the epidermal growth factor receptor (EGFR) and the platelet-
derived growth factor receptor (PDGFR), are critically involved in the transduction of mitogenic signals across the
plasma membrane and therefore in the regulation of cell growth and proliferation. Enhanced RTK activity is asso-
ciated with proliferative diseases such as cancer, psoriasis and atherosclerosis, while decreased function may be asso-
ciated for instance with diabetes. EGFR and PDGFR are selectively inhibited by analogues of the marine natural
product aeroplysinin. The synthetic inhibitors display IC50 values in the low micromolar range and in contrast to the
natural product show pronounced inhibitory activity in cultured cells in vivo. The mechanism of inhibition is likely
based on a covalent modi®cation of the target enzymes by reaction of epoxy ketone 8 with various nucleophiles. # 1998
Elsevier Science Ltd. All rights reserved.
Introduction
activates transcription factors, which then in¯uence
gene transcription in the nucleus. Dysregulation of these
well balanced pathways may contribute to the genera-
tion of various diseases. For instance, enhanced activity
of receptor tyrosine kinases has been implicated in car-
cinogenesis and in the development of proliferative dis-
eases such as atherosclerosis and psoriasis,2 whereas
decreased function of the insulin RTK is associated with
various types of diabetes.3 Therefore, receptor tyrosine
kinases are considered promising targets for the devel-
opment of new drugs. Such RTK inhibitors are, in
addition, likely useful in the dissection of signalling
pathways.2,4 To be used in biological studies, they
should preferably be active both in vitro and in vivo.
The proliferation and growth of cells are regulated by
growth factors such as the insulin-like growth factor
(IGF), the nerve growth factor (NGF), the epidermal
growth factor (EGF), and the platelet-derived growth
factor (PDGF). These growth factors are ligands for
speci®c receptors with tyrosine kinase activity (RTKs).
The receptors transmit growth-promoting signals across
the cell membrane. After binding of the ligand to the
extracellular domain of the receptors, they interphos-
phorylate tyrosine residues present in their intracellular
domains and subsequently trigger various intracellular
signalling cascades by a variety of speci®c protein-
protein interactions (Fig. 1).1 For instance, as a result of
several protein±protein interactions the Ras-protein, a
small GTP-binding protein, is activated. Ras then
passes the signal on to a kinase cascade that eventually
Natural products like lavendustin A 1,5 quercetin 2,6
and erbstatin 37 (Fig. 2), which can be applied advanta-
geously in biological investigations, have proven to be
RTK blockers. Based on these structures, synthetic
inhibitors, the so called tyrphostins 4,2a,b,10 were devel-
oped. Whereas most compounds introduced so far dis-
play IC50 values in the low micromolar range,2,4,8±10
more ecient and speci®c inhibitors such as 5 and 6
(Fig. 2) have recently been reported.2a,9 Since numerous
Key words: Tyrosine kinase; enzyme inhibition; aeroplysinin;
EGF receptor; PDGF receptor.
*Author for correspondence. Tel: +49-721-6082091; Fax:
+49-721-6084825;
E-mail: waldmann@ochhades.chemie.uni-karlsruhe.de
0968-0896/98/$Ðsee front matter # 1998 Elsevier Science Ltd. All rights reserved
PII: S0968-0896(98)00070-4

1154
K. Hinterding et al./Bioorg. Med. Chem. 6 (1998) 1153±1162
Figure 2. Structures of PTK inhibitors.
against the EGF-receptor was investigated employing
immortalized rat-1-HER cells14 (rat ®broblasts that
over-express the human EGF receptor, Fig. 3). Under
the assay conditions, however, 7 was not active at all.
We reasoned that this lack of inhibition in vivo might be
attributed to the inability of 7, due to the presence of a
polar trans-diol unit, to pass the plasma membrane and
thereby reach the kinase domain of the EGF receptor,
which is located in the intracellular domain of the
proteins. Therefore, we designed the aeroplysinin ana-
logue 8 as a possible RTK inhibitor for which a higher
membrane permeability and, therefore, an enhanced
Figure 1. The Ras/Map-kinase signal transduction cascade.
tyrosine kinases are involved in various biological
phenomena, the identi®cation of new classes of subtype-
selective receptor tyrosine kinase inhibitors is still of
great importance to bioorganic and medicinal chemistry.
Aeroplysinin 7 (Scheme 1), a highly functionalized
marine natural product, is cytotoxic against human
breast cancer cells and inhibits the EGFR tyrosine
kinase in an in vitro test system.11 Its structure di€ers
signi®cantly from the structures of the RTK inhibitors
reported so far (see Figure 2), which are almost exclu-
sively aromatic compounds. Aeroplysinin might there-
fore serve as a promising lead for the development of a
new class of these biologically active substances. How-
ever, compounds with structural similarity to 7 have yet
not been investigated and data on the eciency of 7 in
vivo are lacking. We now report on the synthesis and
biological evaluation of aeroplysinin analogues that
display pronounced RTK inhibitory activity in vivo and
receptor subtype selectivity.12
Results and Discussion
To investigate the activity of aeroplysinin13 in a whole
cell assay system (vide infra), its inhibitory activity
Scheme 1. Structure of aeroplysinin 7 and the spiro-epoxy
analogue 8; retrosynthetic analyses of 7 and 8.

K. Hinterding et al./Bioorg. Med. Chem. 6 (1998) 1153±1162
1155
inhibitory e€ect could be expected. In 8 the vicinal diol
unit and the b-cyano alcohol are replaced by an a-spiro
epoxy ketone. Thus, 8 should on the one hand be sig-
ni®cantly more lipophilic than 7 and on the other hand
it should be prone to nucleophilic attack (Scheme 1) at
the epoxide and/or the vinylogous ketone function.
Thereby an ecient inactivation of the receptor via
covalent modi®cation of nucleophiles in the binding site
might be brought about. In addition, compound 8 may
serve as a promising intermediate for the synthesis of
the natural product itself. For this purpose the epoxide
in 8 would have to be transformed to the b-cyano alco-
hol by reaction with cyanide nucleophiles and the keto
function would have to be diastereoselectively reduced
to the alcohol 7 (Scheme 1). The a-spiro epoxy ketone
present in 8 can be generated via Becker±Adler15 oxida-
tion from the ortho hydroxy benzylic alcohol 12, which
in turn should be accessible from the commercially
available aldehyde 9 (Scheme 1).
The a-spiro epoxy ketone 8 was synthesized in a four-
step sequence from aldehyde 916 (Scheme 2). To this end,
the methyl ether ortho to the aldehyde group of 9 was
regioselectively cleaved in 93% yield by treatment with
BCl3. The Lewis acid coordinates to the aldehyde and
then regioselectively cleaves the ortho methoxy group.
The resulting phenol 10 was converted to the dibromo
compound 11 (75% yield) through the agency of the
pyridinium±hydrobromide±bromine complex.17 Reduc-
tion of the aldehyde function of 11 with sodium boro-
hydride generated compound 12 in 93% yield. This
ortho-hydroxy-substituted benzyl alcohol was subjected
to oxidation with sodium periodate in an acidic med-
ium. Under these conditions the desired a-spiro-epoxy
ketone 8 was formed in 60% yield. The amount and
Scheme 2. Synthesis of the spiro-epoxy aeroplysinin analogue
.
8; (a) BCl3, CH2Cl2, rt, 93%; (b) Br2 HBrpyridine, pyridine,
65 ꢀC, 75%; (c) NaBH4, THF, rt, 93%; (d) NaIO4, HCl, H2O,
THF, rt, 60%.
Figure 3. Schematic drawing of the in vivo assay system.

1156
K. Hinterding et al./Bioorg. Med. Chem. 6 (1998) 1153±1162
concentration of hydrochloric acid present in the reac-
tion mixture of this Becker±Adler15 oxidation had to be
carefully controlled, since otherwise the aldehyde 11Ð
after formation of an intermediary benzylic periodate
esterÐwas formed predominantly.
investigated by subjecting NIH-3T3-b-PDGFR cells to
the assay conditions described above.18 Compound 8
inhibited this tyrosine kinase receptor not at all up to a
concentration of 100 mM, indicating a pronounced sub-
type-speci®ty.
To investigate whether the a-spiro-epoxy ketone 8 and
other analogues of 7 inhibit RTKs in vivo, the inhibi-
tion of the EGFR tyrosine kinase was ®rst studied. We
used the assay system mentioned above. Immortalized
rat-1-HER cells14 (rat ®broblasts that over express
human EGF receptor) were incubated with varying
concentrations of inhibitor for 5 min. After this
short time period, human EGF was added to induce
signal transduction and phosphorylation of unblocked
receptor. The cells were subsequently lysed, the proteins
resolved by electrophoresis (SDS±PAGE) and then
transferred onto an Immobilon-PDVF membrane via
electroblotting. The extent of tyrosine phosphorylation
of the EGFR was determined by phosphotyrosine
immunoblotting (see Figure 3).18 Antibody binding was
visualized with an antibody-linked peroxidase that
caused an emission of light proportional to the amount
of phosphorylated protein. This was used to darken a
photo plate. Therefore, the more active the investigated
compound inhibited the autophosphorylation of the
RTK, the less dark were the corresponding protein pat-
ches on the Immobilon-PDVF membrane. Figure 4
demonstrates that compound 8, but not compound 13,
inhibits the EGF-induced increase of EGF receptor
tyrosine phosphorylation in rat-1-HER cells.
To delineate which structural properties are responsible
for the RTK-inhibitory activity of 8, several analogues
of the spiro-epoxy ketone were synthesized. Compounds
13, 14, 15, and 17 were synthesized according to a proto-
col described by Anderson and Faulkner.19 The g, d
saturated ketone hydrate 20 was generated from the
unsaturated ketone 15 by reduction with NaBH4. In
contrast to reference 19 our experiments with this reac-
tion did not form the expected natural product itself.
Benzodioxolane 19 was synthesized from epoxy ketone 8
via a 1,3 sigmatropic shift analogous to the rearran-
gement of vinyl cyclopropanes to cyclopentenes
(Scheme 3). Catechol 18, in principle to be derived from
dioxolane 19, could be instead formed in one step from
epoxy ketone 8 by nucleophilic opening of the epoxide
(Scheme 3, see Scheme 4 for reaction mechanisms). These
compounds were investigated as possible inhibitors of
the EGF and PDGF receptors using the in vivo assay
systems described above. The data compiled in Table 1
demonstrate that neither aeroplysinin 713 nor its isomer
iso-aeroplysinin19 13 inhibit the EGF or the PDGF
RTK in vivo at concentrations up to 100 mM. The ana-
logue 14 in which the epoxide is opened and the result-
ing tertiary alcohol is masked as an ester also inhibits
the EGF receptor although it is less ecient than 8.
Likewise 14 does not block the kinase activity of the
PDGF receptor at all. Enhancing the polarity by sapo-
nifying the acetate (14!15) yielded a compound inac-
tive under the assay conditions probably due to its
reduced ability to suciently pass the plasma membrane
of the cells within the time of assay. Interestingly, also
the aromatic aldehydes 16 (which does not embody a
methoxy group) and 11 both inhibited the EGFR
(entries 6 and 7) albeit less ecient than 2. Furthermore
16 (but not the methoxyaldehyde 11) inhibited the
PDGFR with an IC50 value of 30 mM. On the other
hand, neither the aldehyde 10 (entry 8), nor the alcohols
12, 17 and 20 (entries 9, 10, and 13), the catechol 18
(entry 11) or the benzodioxolane 19 (entry 12) are inhi-
bitors. These ®ndings taken together suggest the fol-
lowing requirements for inhibitory activity: (1) the
bromine atoms are necessary, probably to e€ect lipo-
philicity (Table 1, compare entries 6, 7, and 8); (2) more
than one free hydroxy-functions render the compounds
inactive under the assay conditions (Table 1, compare
entries 6, 9, and 11); (3) an electrophilic group is
required (Table 1, compare entries 3, 4, 5, 6, and 12); (4)
the nature and position of the electrophilic group may
vary (Table 1, compare entries 3, 4, and 6). The IC50
values reported in this study are in the low micromolar
The results were very promising: 8 inhibited the kinase
activity of the EGF receptor with an IC50 value of
10 mM (Table 1). Preincubation of the cells for just a few
minutes was sucient to induce this e€ect, whereas
under the conditions of our assay an inhibition of the
receptor tyrosine kinase with tyrphostins required pre-
incubation of the cells for at least several hours. These
®ndings prove substantial in vivo activity of 8 and its
sucient lipophilicity. In a ®rst attempt to determine
whether 8 is receptor-subtype speci®c, the inhibition
of the platelet derived growth factor receptor was
Figure 4. Determination of the IC50 values for the inhibition of
the EGFR by immunoblotting of phosphotyrosine (aPY) and
EGFR) to con®rm that equal amounts of proteins were loaded.

K. Hinterding et al./Bioorg. Med. Chem. 6 (1998) 1153±1162
1157
Table 1. Inhibition of the EGF and the PDGF receptor tyrosine kinases by means of aeroplysinin analogues in an in
vivo assay system (for the description of the assay see the text)
Entry
Compd
Structure
Inhibition of the EGF
a
Inhibition of the PDGF
a
receptor IC50 (mM)
receptor IC50 (mM)
aConcentration at which 50% of the enzymatic activity are inhibited.
`inactive'=no inhibition at <100 mM inhibitor concentration.

1158
K. Hinterding et al./Bioorg. Med. Chem. 6 (1998) 1153±1162
Scheme 3. Synthesis of the aeroplysinin analogues 18 and 19;
(a) THF, 1 N HCl, 70 ꢀC, 72%; (b) toluene, 60 ꢀC, 78%.
range and therefore do not reach the best values repor-
ted so far (vide supra). We stress, however, that the
values are determined by employing a whole-cell-based
assay system, whereas many of the IC50 values reported
in the literature were determined by employing isolated
kinases. Consequently, in our assay system the inhibi-
tors have to di€use across the plasma membrane to the
interior of the cells. We do not know their intracellular
concentration which is sucient for the observed inhi-
bition. It is, however, prudent to assume that the cyto-
solic inhibitor concentration is signi®cantly lower than
the concentrations reported in Table 1.
Scheme 4. Reaction of epoxide 8 with nucleophiles at di€erent
electophilic positions. (a) NaBH4, CeCl3, MeOH, 70%; (b)
NaCN, DMSO or Et2AlCN, toluene, 19: 20%, 24: 14%; (c)
TMSCN, Bu4NCN, CH2Cl2, 54% or TMSCN, PPh3, CH2Cl2,
25%.
requires the expected cleavage of the C O bond. To test
whether these electrophilic groups may be able to cova-
lently modify the enzyme by reaction with nucleophiles
in the active site, compound 8 was treated with biologi-
cally relevant nucleophiles, which may occur in the side
chains of amino acids (Scheme 5). In the event reaction
with Z-Tyr gave oxoacetal 26 by nucleophilic attack at
the epoxide function (position 2). On the other hand
epoxy ketone 8 was reduced to benzylic alcohol 12
To prove the high susceptibility of 8 towards nucleo-
philic attack, which may be responsible for the inhibi-
tion of the tyrosine kinases the epoxy ketone was treated
with di€erent nucleophiles. By variation of the reaction
conditions a speci®c attack at three di€erent positions 1,
2 and 3 (Scheme 4) could be induced. Reducing agents
like NaBH4 attacked the ketone (position 1) and yielded
the benzyl alcohol 12 after rearrangement of the inter-
mediate 21. Cyanide nucleophiles NaCN in DMSO or
Et2AlCN attacked the epoxide followed by an unex-
pected aromatization of the system via phenolate 22,
yielding benzodioxolane 19 and cyanohydrin 24. A
nucleophilic attack at position 3 could be initiated with
Bu4NCN or PPh3 in TMSCN. Deprotonation of the
acidic position of intermediate 23 yielded the hexasub-
stituted benzene derivative 25. All attempts to synthesize
aeroplysinin from the intermediate 8 therefore failed
(Scheme 4). Reduction of the ketone 8 with di€erent
reducing agents yielded the benzyl alcohol 12, whereas
the desired alcohol 21 could not be isolated. On the
other hand, reaction with cyanide nucleophiles gave the
aromatic compounds 19 and 24 after cleavage of the
C C bond instead of the desired alcohol 15, which
Scheme 5. Reaction of epoxy ketone 8 with biological relevant
nucleophiles at di€erent electophilic positions. (a) N-Ac-Cys,
TosOH, THF, 50 ꢀC, 56%; (b) Z-Tyr, TosOH, THF, 50 ꢀC,
27%.

K. Hinterding et al./Bioorg. Med. Chem. 6 (1998) 1153±1162
1159
(attack at position 1) upon treatment with N-acetyl-
Cys. These ®ndings indicate that a covalent modi®ca-
tion of the enzyme may be responsible for the inhibitory
activity of the compounds described.
temperature for 16 h the mixture was cooled in an ice-
bath and 100 mL of 1 N HCl were added followed by
200 mL of water. The aqueous phase was extracted with
CH2Cl2 (3Â100 mL) and the combined organic layers
were washed with brine (2Â150 mL). Drying over
Na2SO4 and removal of the solvent in vacuo yielded an
oil, which slowly crystallized. The crude pink product
(9.89 g, 0.065 mol) was puri®ed by chromatography on
silica gel (ethyl acetate/hexane=1/3 (v/v), Rf=0.30).
Yield: 9.49 g, 0.062 mol, 93%, white crystals, mp 42 ꢀC
Conclusion
In conclusion, we have demonstrated that a-substituted
cyclohexadienones like 8 and 14 and structurally related
aromatic compounds such as 11 and 16 inhibit receptor
tyrosine kinases, possibly by covalent modi®cation of
the enzyme in its active site. They display a marked in
vivo activity and appreciable receptor-subtype speci®-
city. Analogues of the natural product aeroplysinin are
thus interesting lead compounds for the development of
new drugs and tools in bioorganic and medicinal chem-
istry (i.e., for the dissection of signal transduction path-
ways) and the introduction of new tyrosine kinase
inhibitors into medical use.
(lit. 21; 42 ꢀC); H NMR (CDCl3, 250 MHz) d 3.87 (s,
1
3H, OMe), 6.43 (d, J=1.5 Hz, 1H, CH), 6.54 (dd,
J=8.6 Hz, J=1.5 Hz, 1H, CH), 7.42 (d, J=8.6 Hz, 1H,
CH), 9.72 (s, 1H, OH), 11.50 (s, 1H, CHO); 13C NMR
(CDCl3, 100 MHz) d 55.75 ppm (OMe), 100.68 (CH),
108.43 (CH), 115.21 (C-CHO), 135.28 (CH), 164.57
(C-OH), 166.87 (COMe), 194.43 (CHO).
3,5-Dibromo-2-hydroxy-4-methoxy-benzaldehyde (11). A
solution of pyridinium±hydrobromide±perbromide (4.5g,
14 mmol) in pyridine (12 mL) was added dropwise at
45 ꢀC to a solution of aldehyde 10 (1.0 g, 6.6 mmol) in
pyridine (12 mL). After stirring for 120 min at 45 ꢀC the
mixture was hydrolysed with ice (100 mL) and water
(500 mL) to yield a brownish precipitate, which was ®l-
tered and dried in vacuo. Chromatography on silica gel
(ethyl acetate/hexane=1/2 (v/v), Rf=0.46) gave white
crystals: 1.71 g, 5.5 mmol, 83%. 1H NMR (CDCl3,
250 MHz) d 3.99 (s, 3H, OMe), 7.75 (s, 1H, arom. H),
9.78 (s, 1H, OH), 11.75 (s, 1H, CHO); 13C NMR
(CDCl3, 100 MHz) d 60.94 (OMe), 107.84 (CBr), 107.88
(CBr), 118.67 (CCHO), 136.29 (CH), 159.50 (COH),
161.11 (COMe), 194.10 (CHO); HRMS: calcd for
C8H6Br2O3 309.8664, found: 309.8646; Anal. calcd for
C8H6Br2O3: C 31.00% H 1.96%, found: C 31.16% H
2.00%.
Experimental
General
Melting points were determined in open capillaries using
1
a Buchi 535 apparatus and are uncorrected. H and 13C
NMR spectra were recorded on Bruker AC 250, AM
400 or DRX 500 spectrometers at room temperature. IR
spectra were recorded on a Bruker IFS 88 spectrometer.
Mass spectra and high-resolution mass spectra (HRMS)
were measured on a Finnigan MAT MS70 spectro-
meter. Elemental analyses were performed on a Heraeus
CHN-Rapid apparatus.
Materials
3,5-Dibromo-2-hydroxy-4-methoxy-benzyl alcohol (12).
NaBH4 (0.117 g, 3.1 mmol) was added at room tem-
perature to a stirred solution of aldehyde 11 (0.96 g,
3.1 mmol) in THF (30 mL). After 60 min 1 N HCl
(10 mL) was added and the aqueous phase was extracted
with ether (3Â30 mL). The organic layers were dried
with Na2SO4 and the solvent was removed in vacuo to
yield a yellow oil, which was puri®ed by chromato-
graphy on silica gel (ethyl acetate/hexane=1/1 (v/v),
Rf=0.33;). Yield: 0.9 g, 2.9 mmol, 93%, white crystals.
1H NMR (CDCl3, 250 MHz) d 2.24 (t, J=5.7 Hz, 1H,
OH), 3.88 (s, 3H, OMe), 4.76 (d, J=5.7 Hz, 2H, CH2),
6.75 (s, 1H, OH), 7.37 (s, 1H, arom. H); 13C NMR
(CDCl3, 100 MHz) d 60.73 ppm (CH2), 62.20 (OMe),
107.18 (CBr), 107.40 (CBr), 124.24 (C-CH2OH), 130.77
(CH), 151.89 (COH), 153.88 (COMe); MS (70 eV, EI):
m/z (%): 310/312/314 (16/27/11) [M+], 292/294/296 (47/
100/49) [M+ 18], 264/266/268 (6/11/6), 249/251/253 (3/
7/4), 213/215 (12/12), 91 (6), 63 (7); HRMS: calcd for
Solvents were dried by standard methods and stored
over molecular sieves. For column chromatography
silica gel (40±60 mm, Baker) was used. Commercial
reagents were used without further puri®cation.
Pyridinium±hydrobromide±perbromide was prepared
according to reference 17. Compounds 13, 14, 15, and
17 were synthesized according to reference 19. Human
EGF was purchased from Sigma (Munich), PDGF-BB
from Biomol (Hamburg), Immobilon-PDVF membrane
from Millipore (Bedford, UK), antibodies from Trans-
duction Laboratories (Exeter, UK) except for perox-
idase-linked antibodies, which were from DAKO
(Hamburg).
2-Hydroxy-4-methoxy-benzaldehyde (10).21 To a solu-
tion of the aldehyde 9 (11.07 g, 0.067 mol) in dry CH2Cl2
(100 mL) was added dropwise a 1 M solution of BCl3 in
CH2Cl2 (100 mL, 0.1 mol) at 0 ꢀC. After stirring at room

1160
K. Hinterding et al./Bioorg. Med. Chem. 6 (1998) 1153±1162
C8H8Br2O3 311.8820, found: 311.8809; Anal. calcd for
C8H8Br2O3: C 30.81% H 2.59%, found: C 30.87% H
2.59%.
(s, 2H, CH2), 6.94 (s, 1H, CH); 1H NMR (DMSO,
250 MHz) d 3.74 ppm (s, 3H, OMe), 6.16 (s, 2H, CH2),
7.25 (s, 1H, CH); 13C NMR (CDCl3, 250 MHz) d 60.92
(OMe), 98.02 (CBr), 102.35 (CH2), 107.96 (CBr), 111.10
(CH), 144.32 (CO), 146.70 (CO), 148.92 (CO); IR
(KBr): n 3090, 2943, 2908, 1606, 1495, 1466, 1404, 1245,
1201, 954, 838 cm 1; MS (70 eV, EI): m/z (%): 308/310/
312 (50/100/49) [M+], 293/295/297 (44/88/43) [M+ 15],
235/237/239 (5/9/4), 209 (7), 143 (9), 131/133 (14/13), 77
(18), 53 (14); HRMS: calcd for C8H6Br2O3: 309.8664
found: 309.8647; Anal. calcd for C8H6Br2O3: C 31.00%
H 1.96%, found: C 31.18% H 2.16%. A more e€ective
protocol for preparing compound 19 is the following: a
solution of epoxyketone 8 (50 mg, 0.16 mmol) in toluene
(3 mL) is heated for 10 h at 60 ꢀC. After evaporation of
the solvent in vacuo the residue is puri®ed by chroma-
tography on silica gel (ethyl acetate/hexane=1/4 (v/v),
Rf=0.45). Yield: 39 mg, 0.13 mmol, 78%, white crystals,
mp: 86 ꢀC.
2,4-Dibromo-3-methoxy-cyclohexa-2,4-diene-1-one-5-spiro-
oxirane (8). To a solution of the benzylic alcohol 12
(0.5 g, 1.6 mmol) in a mixture of THF (8 mL) and 1 N
HCl (1 mL) was added dropwise a solution of NaIO4
(2.74 g, 12.8 mmol, 8 equiv) in water (15 mL) and 1 N
HCl (5 mL). After stirring at room temperature for
90 min the deep-yellow solution was extracted with ether
(3Â30 mL). The combined organic layers were dried
with Na2SO4 and the solvent was removed in vacuo.
Chromatography on silica gel (ethyl acetate/hexane=1/
5 (v/v), Rf=0.27) yielded yellow crystals. Yield: 0.3 g,
1 mmol, 60%. 1H NMR (CDCl3, 250 MHz) d 3.26 (d,
J=8.6 Hz, 1H, CH2), 3.43 (d, J=8.6 Hz, 1H, CH2), 4.14
(s, 3H, OMe), 6.54 (s, 1H, CH); 13CNMR (CDCl3,
100 MHz) d 57.55 (quart. C), 59.19 (CH2), 61.84 (OMe),
110.80 (CBr), 119.82 (CBr), 138.23 (CH), 164.44
(COMe), 187.06 (C=O); IR (KBr): n3337, 3064, 2944,
2849, 1762, 1679, 1534, 1460, 1051 cm 1; MS (70 eV,
EI): m/z (%): 308/310/312 (40/75/35) [M+], 293/295/297
(53/100/50) [M+ 15], 265 (5), 235/237/239 (4/10/5), 209
(8), 131 (18), 77 (15); HRMS: calcd for C8H6Br2O3
309.8664, found: 309.8648; Anal. calcd for C8H6Br2O3:
C 31.00% H 1.96%, found: C 31.05% H 1.99%.
1
24: H NMR (CDCl3, 250 MHz) d 3.88 (s, 3H, OMe),
6.18 (s, 2H, CH2), 6.41 (s, 1H, CH); H NMR (DMSO,
1
250 MHz) d 3.82 (s, 3H, OMe), 6.29 (s, 2H, CH2), 6.88
(s, 1H, CH); 13C NMR (CDCl3, 100 MHz) d 57.06
(OMe), 90.82 (CBr), 98.47 (CBr), 103.26 (CH2), 103.69
(CH), 114.25 (CN), 144.77 (CO), 148.23 (CO), 151.91
(CO); IR (KBr): n 3115, 3083, 2915, 2235 (CN), 1625,
1612, 1472, 1453, 1438, 1366, 1226, 1100, 935 cm1; MS
(70 eV, EI): m/z (%):333/335/337 (0.42/0.80/0.36)
[M+ 2], 318/320/322 (0.33/0.73/0.33) [M+ 17], 308/
310/312 (0.10/0.18/0.09) [M+ 27], 255/257 (100/98),
240/242 (69/67), 210/212 (12/12), 182/184 (8/9), 131/133
(10/10), 90 (5), 75 (12); HRMS calcd for C9H7Br2NO3
336.8773, found: 336.8739.
Reduction of the ketone 8 with NaBH4/CeCl3. To a
solution of the ketone 8 (0.3 g, 0.97 mmol) and CeCl3
(0.89 g, c=0.4 M) in MeOH (6 mL) was added at 0 ꢀC
under stirring NaBH4 (0.036 g, 0.97 mmol). After
30 min the mixture was hydrolysed with 1 N HCl (3 mL)
and extracted with ether (5Â20 mL). The combined
organic layers were dried over Na2SO4 and the solvent
was removed in vacuo. Chromatography on silica gel
(ethyl acetate/hexane=1/2 (v/v), Rf=0.23) yielded
white crystals of alcohol 12. Yield: 0.21 g, 0.67 mmol,
70%.
3,5-Dibromo-2-hydroxy-4-methoxy-phenol (18). A solu-
tion of the epoxyketone 8 (200 mg, 0.65 mmol) in a
mixture of THF (35 mL), 1 N HCl (5 mL) and water
(20 mL) was heated under re¯ux for 5 h. After cooling
the mixture was extracted with ether (3Â50 mL), the
combined organic layers were washed with brine
(30 mL) and dried with Na2SO4. The solvent was
removed in vacuo and the residue was puri®ed by chroma-
tography (ethyl acetate/hexane=2/3 (v/v), Rf=0.23).
Yield: 140 mg, 0.47 mmol, 72%, yellowish crystals, mp
108 ꢀC; 1H NMR (CDCl3, 250 MHz) d 3.83 (s, 3H,
OMe), 5.30 (br, 1H, OH), 5.54 (br, 1H, OH), 7.11 (s,
1H, CH); 13C NMR (CDCl3, 100 MHz) d 60.86 (OMe),
105.81 (CBr), 107.59 (CBr), 118.10 (CH), 140.34 (COH),
141.19 (COH), 147.27 (COMe); IR (KBr): n 3352, 2944,
1583, 1486, 1375, 1271, 1138, 964 cm‹1; MS (70 eV,
EI): m/z (%): 296/298/300 (50/100/47) [M+], 281/283/
285 (45/89/42) [M+ 15], 253/255/257 (19/40/18), 175
(8), 131 (8), 77 (13), 53 (12); HRMS (70 eV): calcd for
C7H6Br2O3: 297.8664, found: 297.8640; Anal. calcd: C
28.22% H 2.03%, found: C 28.32% H 2.18%.
3,5-Dibromo-4-methoxy-benzo-1,2-dioxolane (19) and
3,5-dibromo-2-hydroxy-4-methoxy-phenoxyacetonitrile
(24). To a solution of the epoxyketone 8 (100 mg,
0.32 mmol) in toluene (2 mL) was added dropwise under
nitrogen
a 1 M solution of Et2AlCN (2.24 mL,
2.24 mmol, 7 equiv) at room temperature. After stirring
at room temperature for 3 h the reaction mixture was
poured into 2 N NaOH (10 mL) and ice (10 mL). The
aqueous phase was extracted with CH2Cl2 (4Â20 mL)
and the combined organic layers were dried over
Na2SO4. After evaporation of the solvent in vacuo,
chromatography on silica gel (ethyl acetate/hexane=1/4
(v/v)) yielded the benzodioxolane 19 (20 mg, 0.07 mmol,
20%, white crystals; Rf=0.45) and the cyanohydrin 24
(15 mg, 0.04 mmol, 14%, white crystals; Rf=0.35). 19:
1H NMR (CDCl3, 250 MHz) d 3.82 (s, 3H, OMe), 6.05

K. Hinterding et al./Bioorg. Med. Chem. 6 (1998) 1153±1162
1161
3,5-Dibromo-2-hydroxybenzaldehyde (16).22 To a solu-
tion of 2-hydroxy benzaldehyde (2.00 g, 16.4 mmol) in
pyridine (20 mL) was added dropwise at 45 ꢀC a solution
of pyridinium±hydrobromide±perbromide (11.25 g,
35 mmol) in pyridine (35 mL). After stirring for 2 h at
45 ꢀC the mixture was poured onto ice (200 mL). The
precipitated solid was ®ltered o€, dried and recrys-
tallized from ethyl acetate/hexane (2/1 (v/v), 20 mL).
Yield: 3.90 g, 13.9ꢀ4 mmol, 85%, yellow crystals, mp
the combined organic phases with Na2SO4 was followed
by evaporation of the solvent in vacuo. Chromato-
graphy on silica gel (ethyl acetate/hexane=1/2 (v/v),
Rf=0.20) gave yellow crystals. Yield: 110 mg,
0.33 mmol, 54%; 1H NMR (CDCl3, 250 MHz)
d
3.90 ppm (s, 3H, OMe), 5.10 (s,2H,CH2); 1H NMR
(DMSO-d6, 250 MHz) d 3.80 ppm (s, 3H, OMe), 4.70 (s,
2H, CH2); 13C NMR (DMSO-d6, 100 MHz) d 58.00 ppm
(CH2), 60.45 ppm (OMe), 110.13 (CCN), 114.46 (CBr),
114.94 (CBr), 115.79 (CN), 130.49 (CCH2OH), 153.94
(COMe), 154.55 (COH); IR (KBr): n 3439, 3107, 2947,
2240, 1557, 1400, 1173, 1021 cm 1; MS (70 eV, EI): m/z
(%): 335/337/339 (23/30/20) [M+], 317/319/321 (76/100/
74) [M+ 18], 289/291/293 (9/21/13), 246/248/250 (4/9/
4), 238/240 (16/17), 223/225 (8/8), 180/182 (3/4), 116 (7),
88 (8), 77 (3); HRMS (70 eV): calcd for C9H7Br2NO3:
336.9663, found: 336.9655.
ꢀ
1
81 C (lit.22: 81±84 C); H NMR (CDCl3, 250 MHz) d
3.94 (s, 3H, OCH3), 7.20 (d, J=8 Hz, 1H, arom. CH),
7.94 (dd, J=8 Hz, J=1 Hz, 1H, arom. CH), 8.07 (d,
J=1 Hz, 1H, arom. CH), 12.94 (br, 1H, COOH).
2,6-Dibromo-4-hydroxy-40-cyanomethyl-1-methoxy-cyclo-
hexene-3-one-hydrate (20). NaBH4 (8 mg, 0.21 mmol)
was added at 0 ꢀC under nitrogen to a solution of ketone
15 (35 mg, 0.10 mmol) in dry EtOH (10 mL). After
10 min at 0 ꢀC the reaction was stopped by adding a
saturated solution of NH4Cl in water (5 mL). Extraction
with CH2Cl2 (3Â20 mL) was followed by drying the
combined organic layers over MgSO4 and evaporation
of the solvent in vacuo. Chromatography on silica gel
(ethyl acetate/hexane=1/1 (v/v), Rf=0.15) yielded a
colourless solid. Yield: 12 mg (0.03 mmol), 34%. 1H
NMR (CD3OD, 250 MHz) d 2.50 (t, J=14 Hz, 1H,
CH2), 2.78 (d, J=12 Hz, 1H, CH2CN), 2.85 (d,
J=12 Hz, 1H, CH2CN), 3.12 (dd, J=14 Hz, J=6 Hz,
1H, CH2), 3.70 (s, 3H, OCH3), 5.02 (dd, J=14 Hz,
J=6 Hz, 1H, CHBr); 1H NMR (DMSO-d6, 250 MHz) d
2.41 (t, J=14 Hz, 1H, CH2), 2.65 (d, J=15 Hz, 1H,
CH2CN), 2.96 (d, J=15 Hz, 1H, CH2CN), 3.04 (dd,
J=14 Hz, J=6 Hz, 1H, CH2), 3.63 (s, 3H, OCH3), 5.15
(dd, J=14 Hz, J=6 Hz, 1H, CHBr), 6.25 (br, 1H, OH),
7.11 (br, 1H, OH), 7.58 (br, 1H, OH); 13C NMR
(CDCl3, 100 MHz) d 41.18 (CH2), 44.31 (CH2), 46.14
(CH), 52.90 (OCH3), 71.93 (CCH2CN), 93.60 (CN),
163.45 (quart. C), 172.03 (quart. C), 180.23 (quart. C);
Reaction of epoxy ketone 8 with Z-Tyr (oxoacetal 26).
A solution of epoxide 8 (10 mg, 0.03 mmol), Z-Tyr (20
mg, 0.07 mmol) and p-TosOH (0.5 mg, mmol) in dry
THF (2 mL) was heated in the presence of molecular
sieves (4 A) for 5 h. Ethyl acetate (30 mL) and NEt3
(0.1 mL) was added and the mixture was washed with
water (2Â20 mL). After evaporation of the solvent in
vacuo, the residue was puri®ed by chromatography
(ethyl acetate/hexane=1/2, Rf=0.15). Yield: 5 mg,
0.008 mmol, 27%, colourless oil. 1H NMR (CDCl3,
250 MHz) d 3.01 (d, J=7 Hz, 2H, Tyr b-CH2), 3.88 (s,
3H, OCH3), 4.60 (m, 1H, a-CH), 5.10 (s, 2H, Ph-CH2),
5.68 (d, J=6 Hz, 1H, O-CH2-O), 5.78 (d, J=6 Hz, 1H,
O-CH2-O), 6.21 (s, 1H, NH), 6.67 (d, J=8 Hz, 2H, Tyr
arom. CH), 6.90 (d, J=8 Hz, 2H, Tyr arom. CH), 7.35
(br, 5H, arom. CH); MS (70 eV, EI): m/z (%): 623/625/
627 (0.15/0.40/0.16) [M++H2O], 606/608/610 (0.03/
0.07/0.04) [M+], 528 (0.08) [M+ Br], 386/388/390
(0.14/0.41/0.26), 308/310/312 (12/25/12) [M+ Tyr], 298
(48), 283 (45), 107 (83), 91 (100); HRMS (70 eV): calcd
for C25H23Br2NO7: 606.9841, found: 606.9858.
IR (KBr): n 3389, 3267, 3206, 2947, 2250 (CN), 1730,
1
1621, 1547, 1442, 1343, 1260, 1232, 1000, 771, 634 cm
;
MS (70 eV, EI): m/z (%): 356/358/360 (15/29/13)
[M++1], 355/357/359 (50/100/45) [M+], 325 (10) [M+-
OCH3-H], 312 (9), 284 (16) [M+-2OH-CH2CN], 250/
252 (28/26) [M+-Br-CN], 249/251 (90/89) [M+-HBr-
CN], 217/219/221 (24/42/20), 189/191 (44/49), 170 (100),
138 (41), 55 (49); HRMS (70 eV): calcd for
C9H11Br2NO4: 356.9035, found: 356.9058.
Inhibition of RTK autophosphorylation. Immortalized
rat-1/HER cells14 (in the case of EGF-R autophos-
phorylation) or NIH 3T3-b-PDGFR cells20 (in the case
of PDGF-R autophosphorylation) were grown at 37 ꢀC
and 6% CO2 in Dulbecco's modi®ed Eagle's Medium
(DMEM) containing 10% calf serum (FCS) and anti-
biotics (100 units/mL penicillin, 100 mg/mL streptomy-
cin). Three hours before the experiments the medium
was exchanged for DMEM without phenol red and
without FCS. The cells were incubated for 5 min with
DMSO-solutions of the inhibitors (®nal concentrations:
1 mM, 10 mM and 100 mM) and then treated for 5 min.
with human EGF in the case of EGF-R autophos-
phorylation (2 ng/mL ®nal concentration in the med-
ium) or PDGF-BB in the case of PDGF-R auto-
phosphorylation (10 ng/mL ®nal concentration in the
3,5-Dibromo-6-cyano-2-hydroxy-4-methoxybenzyl alcohol
(25). A solution of epoxy ketone 8 (190 mg, 0.61 mmol)
in dry CH2Cl2 (1.5 mL) and TMSCN (1.5 mL, 1.12 g,
11.25 mmol) was stirred under nitrogen at room tem-
perature for 24 h. The mixture was cooled to 0 ꢀC and
Bu4NCN (165 mg, 0.61 mmol) was added. After 45 min
a saturated solution of NaHCO3 in water (5 mL) was
added. Extraction with ether (4Â30 mL) and drying of

1162
K. Hinterding et al./Bioorg. Med. Chem. 6 (1998) 1153±1162
medium). After removal of the medium the cells were
washed twice with PBS (2Â5 mL; bu€ered isotonic
solution: 137 mM NaCl, 2.7 mM KCl, 0.7 mM CaCl2,
0.6 mM MgCl2, 6.5 mM Na2HPO4, 1.5 mM KH2PO4)
and subsequently lysed with sample bu€er (0.5 mL;
0.16 M Tris±HCl, 4% SDS, 20% glycerol, 4% b-mer-
captoethanol, 0.05% bromophenol blue). The lysate was
sonicated to shear the DNA and aliquots were boiled
for 5 min. The proteins were resolved by SDS±PAGE
(10% acrylamide). After transferring the proteins onto
Immobilon-PDVF-membrane by electroblotting immuno-
staining was performed using the antibody dilutions
recommended by the manufacturer (anti-phosphotyro-
sine PY20-, monoclonal a-EGFR- and monoclonal
a-PDGFR-antibodies). Counterstaining was performed
with horseradish peroxidase-linked antibodies, which
were visualized using ECL detection reagent. In all
cases, unblotted parts of the SDS-gels were Coomassie
stained to con®rm that equal amounts of proteins were
loaded.
Naganawa, H.; Takeuchi, T.; Tatsuta, K.; Umezawa, K. J.
Nat. Prod. 1989, 52, 1252.
6. Glossmann, H.; Presek, P.; Ergenbrodt, E. Naunyn Schmie-
deberg's Arch. Pharmacol. 1981, 317, 100.
7. Umezawa, H.; Imoto, M.; Sawar, T.; Isshiki, K.; Matsuda,
N.; Uchida, T.; Finuma, H.; Hamada, M.; Takeuchi, T. J.
Antibiot. 1986, 39, 170.
8. For recent studies of tyrosine kinase inhibitors see for
example: (a) Buchdunger, E.; Zimmermann, J.; Mett, H.;
Meyer, T.; Muller, M.; Regenass, U.; Lydon, N. B. Proc. Natl.
Acad. Sci., U.S.A. 1995, 92, 2558. (b) Zimmermann, J.; Buch-
dunger, E.; Mett, H.; Meyer, T.; Lydon, N. B.; Traxler, P.
Bioorg. Med. Chem. Lett. 1996, 6, 1221. (c) Kobayashi, J.;
Madono, T.; Shigemoni, H. Tetrahedron 1995, 51, 10867. (d)
Alvi, K.; Diaz, M. C.; Crews, P.; Slate, D. L.; Lee, R. H.;
Moretti, R. J. Org. Chem. 1992, 57, 6604.
9. (a) Rewcastle, G. W.; Palmer, B. D.; Thompson, A. M.;
Bridges, A. J.; Cody, D. R.; Zhou, H.; Frey, D. W.; McMi-
chael, A.; Denny, W. A. J. Med. Chem. 1996, 39, 1823. (b)
Frey, W.; Kratow, A. J.; McMichael, A.; Ambroso, L. A.;
Nelson, J. M.; Leopold, W. R.; Connors, R. W.; Bridges, A. J.
Science 1994, 265, 1093. (c) Spada, A. P.; Myers, M. R. Exp.
Opin. Ther. Patents 1995, 5, 805. (d) Bridges, A. J.; Zhou, H.;
Cody, D. R.; Rewcastle, G. W.; McMichael, A.; Showalter, H.
D. H.; Fry, D. W.; Kraker, A. J.; Denny, W. A. J. Med. Chem.
1996, 39, 267. (e) Traxler, P.; Lydon, N. Drugs Future 1995, 20,
1261.
Acknowledgements
This research was supported by the Deutsche For-
schungsgemeinschaft and the Fonds der Deutschen
Chemischen Industrie.
10. Review: Levitzki, A. FASEB J. 1992, 6, 3275.
11. Kreuter, M.-H.; Leake, R. E.; Rinaldi, F.; Muller-Klieser,
W.; Maidhof, A.; Muller, W. E. G.; Schroder, H. C. Comp.
Biochem. Physiol. 1990, 978, 151.
References and Notes
12. Part of this work was published in preliminary form:
Waldmann, H.; Hinterding, K.; Herrlich, P.; Rahmsdorf, H. J.;
Knebel, A. Angew. Chem. 1997, 109, 1553; Angew. Chem. Int.
Ed. Engl. 1997, 36, 1541.
1. (a) Protein phosphorylation; Marks, F., Ed.; VCH: Wein-
heim 1996. (b) Protein Kinases; Woodgett, J. R., Ed.; Oxford,
1994. (c) Cooper, G. F. Oncogenes, 2nd ed.; Jones and Bartlett:
Boston 1995.
13. (+)-Aeroplysinin was kindly supplied by Professor
Proksch, Lehrstuhl fur Pharmazeutische Biologie, Universitat
Wurzburg.
2. (a) Levitzki, A.; Gazit, A. Science 1995, 267, 1782. (b)
Levitzki, A. Eur. J. Biochem. 1994, 226, 1. (c) Bridges, A. J.
ChemtractsÐOrganic Chemistry 1995, 8, 73. (d) Fry, D. W.
Ann. Rep. Med. Chem. 1996, 31, 151.
14. Wasilenko, W. J.; Payne, D. M.; Fitzgerald, D. L.; Weber,
M. J. Mol. Cell. Biol. 1991, 11, 309.
3. Taylor, S. I.; Cana, A.; Accili, D.; Barbetti, F.; Quon, M. J.;
Sierra, M. D.; Suzuki, Y.; Koller, E.; Levytoledano, R.; Wer-
theimer, E.; Moncada, V. Y.; Kadowaki, H.; Kadowaki, T.
Endocr. Rev. 1992, 13, 566.
15. Becker, H.-D.; Bremholt, T.; Adler, E. Tetrahedron Lett.
1972, 13, 4205. For recent applications of the Becker±Adler
oxidation in natural product synthesis see: (a) Corey, E. J.;
Dittami, J. P. J. Am. Chem. Soc. 1985, 107, 256. (b) Danish-
efsky, S.; Shair, M. D. J. Org. Chem. 1996, 61, 16.
4. For the use of tyrosine kinase inhibitors in the dissection of
signalling pathways see: (a) Daub, H.; Weiss, F. U.; Wallasch,
C.; Ullrich, A. Nature 1996, 379, 557. (b) Novogrodsky, A.;
Vanichin, A.; Patya, M.; Gazit, A.; Osherov, N.; Levitzki, A.
Science 1994, 264, 1319. (c) Meydan, N.; Grunberger, T.; Dadi,
H.; Shahar, M.; Arpaia, E.; Lapidot, Z.; Leeder, J. S.; Freed-
man, M.; Cohen, A.; Gazit, A.; Levitzki, A.; Roifman, C. M.
Nature 1996, 379, 645. (d) Dudley, D. T.; Pong, L.; Decker, S.
J.; Bridges, A. J.; Saltiel, A. R. Proc. Natl. Acad. Sci. U.S.A.
1995, 92, 7686. (e) Lepley, R. A.; Kirkpatrick, R. A. Arch.
Biochem. Biophys. 1996, 331, 141. (f) Review: Hinterding, K.;
Alonso-Dõaz, D.; Waldmann, H. Angew. Chem. 1998, 110, 716;
Angew. Chem. Int. Ed. Engl. 1998, 37, 688.
16. Aldehyde 3 is commercially available.
17. Fieser, L. F.; Fieser, M. Reagents for Organic Synthesis;
Wiley, New York 1967, Vol. 1, p 967.
18. Knebel, A.; Rahmsdorf, H. J.; Ullrich, A.; Herrlich, P.
EMBO J. 1996, 15, 5314.
19. Synthesis of aeroplysinin and iso-aeroplysinin: Andersen,
R. J.; Faulkner, D. J. J. Am. Chem. Soc. 1975, 97, 936.
20. The NIH-3T3-b-PDGFR cells were kindly supplied by
Professor Dr A. Ullrich, MPI fur Biochemie, Martinsried.
21. O€e, J.; Jatzkewitz, K. Chem. Ber. 1947, 80, 469.
5. Onoda, T.; Iinuma, H.; Saski, Y.; Hamada, M.; Isshiki, K.;
22. Lindemann, J.; Forth, P. J. Liebigs Ann. Chem. 1924, 223.