Synthesis and biological evaluation of phorbol-resiniferatoxin (RTX) hybrids

Article abstract of DOI:10.1002/ejoc.200400122

Phorboid 20-homovanillates modified on ring C to mimic the constitution and conformation of the ultrapotent vanilloid resiniferatoxin (RTX) and on the AB ring system to suppress the tumour-promoting potential of the terpenoid core were prepared. Several unexpected reactions were discovered, and the structure-activity relationships for this class of vanilloid ligands were expanded. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejoc.200400122

FULL PAPER
Synthesis and Biological Evaluation of Phorbol-Resiniferatoxin (RTX) Hybrids
Giovanni Appendino,*[a] Andrea Bertolino,[b] Alberto Minassi,[a] Rita Annunziata,[c]
Arpad Szallasi,[d] Luciano de Petrocellis,[e] and Vincenzo Di Marzo*[f]
Keywords: Natural products / Phorbol / Resiniferatoxin / Terpenoids / Vanilloids
Phorboid 20-homovanillates modified on ring C to mimic the
constitution and conformation of the ultrapotent vanilloid
resiniferatoxin (RTX) and on the AB ring system to suppress
the tumour-promoting potential of the terpenoid core were
prepared. Several unexpected reactions were discovered,
and the structure-activity relationships for this class of vanil-
loid ligands were expanded.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
Introduction
tagonist by simple phenolic iodination[5] lends support to
the view that this compound represents an especially advan-
tageous structure with which to investigate vanilloid recep-
tors. Since neither RTX nor its terpenoid core (resini-
feronol, 1b) are commercially available in amounts suf-
ficient to sustain a significant structure-activity effort,[6,7]
we have investigated the potential of phorbol (3a) as a resin-
iferonol surrogate for the synthesis of vanilloids. Phorbol
can be obtained in multigram amounts from croton oil, a
commercial commodity.[8] The structural differences be-
tween phorbol and resiniferonol are localised on ring C,
and are limited to a shift of the hydroxy group from the
secondary C-12 to the secondary C-14 and to the opening
of the adjacent cyclopropane moiety. Our first efforts in this
area resulted in the discovery of a series of phorbol-RTX
hybrids endowed with vanilloid activity.[9] These com-
pounds, exemplified by PPAHV (3b), raised interest because
of their unique behaviour in binding and functional assays
of vanilloid activity.[9,10] Since the vanilloid profile of phor-
boid homovanillates is distinct from those of both capsaici-
noids and resiniferonoids,[10] it was interesting to expand
their structure-activity relationships. Thus, taking PPAHV
as a lead, we have focused on modifications of ring C to
mimic the constitution and conformation of resiniferatoxin,
and on changes of the AB ring system to provide phorboid
platforms devoid of tumour-promoting potential. A signifi-
cant drawback of resiniferatoxin and phorbol homovan-
illates is indeed the generation of tumour promoters
through the loss of the 20-ester group, a reaction that takes
place under mild acidic conditions[8] and the relevance of
which in vivo is unclear.
The discovery that the phorboid resiniferatoxin (RTX,
1a) acts as an ultrapotent vanilloid, and the discovery of a
specific receptor (TRPV1), have spurred an intense research
activity aimed at the discovery of new biological analogues
of capsaicin (2), the pungent principle of hot pepper and
the archetypal vanilloid.[1] As a result of these studies, the
structural database of vanilloids has been considerably ex-
panded, and now encompasses seven distinct chemotypes
from the natural product pool,[2] two of which
(anandamide[2f] and N-acyldopamides[2g]) also occur en-
dogenously in mammal tissues. While undoubtedly provid-
ing new and exciting opportunities for discovery, none of
these new ligands approaches RTX in terms of potency and
pharmaceutical potential. RTX is currently in advanced
clinical trials for treatment of diabetic neuropathy[3] and de-
trusor hyperreflexia,[4] and it is therefore surprising that its
structure-activity relationships are still poorly understood.
Furthermore, the recent discovery that the ultrapotent van-
illoid activity of RTX can be reversed from agonist to an-
[a]
Dipartimento di Scienze Chimiche, Alimentari, Farmaceutiche
e Farmacologiche,
V.le Ferrucci 33, 28100 Novara, Italy
E-mail: appendino@pharm.unipmn.it
[b]
Dipartimento di Scienza e Tecnologia del Farmaco,
Via Giuria 9, 10125 Torino, Italy
[c]
Dipartimento di Chimica Organica e Industriale,
Via Venezian 21, 20133 Milano, Italy
[d]
Department of Pathology and Immunology, Washington
University School of Medicine,
St. Louis, MO 63110, USA
[e]
Endocannabinoid Research Group, Istituto di Cibernetica,
CNR,
While the chemistry of resiniferol is essentially uncharted
territory, phorbol has long been fascinating organic chem-
ists, and its reactivity has been extensively investigated.[8]
However, a series of surprising reactions were discovered in
Via Campi Flegrei 34, 80078 Pozzuoli, Italy
[f]
Endocannabinoid Research Group, Istituto di Chimica
Biomolecolare, CNR,
Via Campi Flegrei 34, 80078 Pozzuoli, Italy
E-mail: vdimarzo@icmib.na.cnr.it
Eur. J. Org. Chem. 2004, 3413Ϫ3421
DOI: 10.1002/ejoc.200400122
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3413

G. Appendino, V. Di Marzo et al.
FULL PAPER
the attempt to steer the reactivity of this polyol to the syn- pathways a and b).[12] However, when the reaction was car-
thesis of our targets.
ried out with sodium methoxide in methanol, the phor-
bobutanone derivative 5b was formed as the only reaction
product, the result of migration to the carbonyl of the exter-
nal cyclopropane bond. Conversely, when the reaction was
carried out in ethanol with sodium ethoxide, a mixture of
two products was obtained in a reproducible way. The
Results and Discussion
Ring C Modifications
The cyclopropane ring of phorbol can be opened under minor one (10%) was the same phorbobutanone as ob-
both acidic and basic conditions, providing the opportunity tained from the rearrangement in methanol (5b), while the
to change the constitution and conformation of ring C. This major one (78%) was the phorboisobutyl ketone derivative
ring adopts a half-chair conformation in phorbol, but is 6b. The two compounds could be distinguished very easily
constrained into a boat geometry by the orthoester group from the chemical shift of C-13, which is a ketone in 5b
of RTX.[11] Constraint into a boat conformation can also (δ ϭ 214.7 ppm) and a semiacetal in 6b (δ ϭ 108.3 ppm).
be achieved by a shift of the methano bridge (C-15), a type Similar behaviour was observed with 4c, the 20-tert-butyldi-
of reaction documented for neophorbol (ϭ 12-dehydro- methylsilyl (TBDMS) analogue of 4b. The regiochemistry
phorbol).[12] Upon mild basic treatment, 12-dehydrophor- of cyclopropane ring-opening in expansion reactions is, in
bol 13,20-diacetate (4a) undergoes an α-ketol rearrange- general, poorly understood,[13] but the steric and electronic
ment, affording a mixture of two cyclobutane derivatives similarity between methanol and ethanol makes it difficult
[hydroxyphorbobutanone (5a) and hydroxyphorboisobutyl to propose an explanation for the observed differences in
ketone (6a)].[12] The new location of the bridge forces ring product distribution in the rearrangement of 4b. The abeo-
C into a boat-like conformation, and the closure of a semia- phobols 5b and 6b were next esterified with excess phe-
cetal ring with the 9-hydroxy group of 6a further contrib- nylacetic acid, affording 5c and 6c and endowing ring C
utes to the overall mimicry of the conformation of RTX. A with the acyl appendage typical of PPAHV and RTX. The
precursor for the α-ketol rearrangement (4b) was prepared regiochemistry of esterification was verified by NOE experi-
from 13-acetylphorbol 20-trityl ether (3c)[9] by PCC oxi- ments carried out on the final 20-homovanillates 5d and 6d,
dation of the secondary C-12 hydroxy group. Surprisingly, obtained by a previously established procedure (removal of
the rearrangement of 4b took different courses in methanol the trityl group with HClO4, esterification with Mem-pro-
and in ethanol. Both the ‘‘external’’ (C-13/C-15) and the tected homovanillic acid, and SnCl4-mediated deprotection
‘‘internal’’ (C-13/C-14) cyclopropane bonds can migrate to of the phenolic hydroxy group).[9] The NOE experiments
the C-12 carbonyl, giving a mixture of two different abeo- showed that in both 5d and 6d the phenylacetyl group was
phorbols, as is indeed reported in the literature (Figure 1; bound to the hydroxy group at C-12 [NOEs between the
3414
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 3413Ϫ3421

Phorbol-Resiniferatoxin (RTX) Hybrids
FULL PAPER
phenyl protons and the α-oriented methyl groups (C-18 and the functionalisation of advanced intermediates such as 3c.
C-16)]. Alternative locations for the phenylacetyl residue, After considerable experimentation, however, we discovered
namely at the C-9 hydroxy group for the phorbobutanone that heating at reflux of a THF solution of the 13-phe-
homovanillate (5d) or at the semiacetal oxygen for the phor- nylacetyl-12-thioimidazolide 7 Ϫ formed in situ from 3d by
boisobutyl ketone homovanillate (6d), would have involved treatment with thiocarbonyldiimidazole Ϫ could trigger the
lack of correlation with the methyl group (C-16) (esterifi- cyclopropyl version of a β-elimination, affording the
cation of the 10-hydroxy group of 5d) or with the methyl enolcrotophorbolone 8a in almost quantitative yield and
group (C-18) (esterification of the semiacetalic hydroxy with preservation of the trityl protection of the primary C-
group of 6d).
20 hydroxy group and the enol ester moiety. (Scheme 1)[15]
Removal of the trityl group and introduction of the homo-
vanillyl moiety[9] eventually afforded the -seco-phorbol
(rhamnofolane) 8b.
Scheme 1. Fragmentation of the thioimidazolide 7
Figure 1. The α-ketol rearrangement of 12-dehydrophorbol deriva-
tives
2. Rings A/B Modifications
The PKC binding and tumour-promoting properties of
phorboids are critically dependent on the configuration of
the AB ring junction, and are abolished by the vinylogous
retro-aldol epimerisation of the tertiary C-4 hydroxy
group.[16] In order to suppress the tumour-promoting po-
tential of the phorboid core, the synthesis of the 4α epimer
of PPAHV and a series of derivatives was attempted. To
this end, the primary and the tertiary C-13 hydroxy groups
of 4α-phorbol (9a)[17] were acetylated, and the secondary
C-12 hydroxy group was esterified with phenylacetic acid,
affording the triester 9b. The 20-hydroxy group was re-
moved by transesterification, and the 20-alcohol 9c was
then treated with MEM-homovanillic acid and then depro-
tected to give a mixture of the homovanillate 9d and the bis-
homovanillate 9e (ca. 1:0.8).[18] The critical 20-ester bond
formation could be achieved in a selective way by switching
from an acyl to an alkyl nucleophilic substitution, and capi-
talising on the Mitsunobu reaction.[19] Thus, treatment of
9c with the triphenylphosphane/di-tert-butyl azodicarboxyl-
ate couple and homovanillic acid cleanly provided 4α-
PPHV (9d) in a gratifying 91% yield. The high reactivity of
the tertiary C-4 hydroxy group of 4α-phorbol in esterifi-
cation reaction is reminiscent of the reactivity of the tertiary
Opening of the cyclopropane ring was next investigated. C-13 hydroxy group in both phorbol and 4α-phorbol.[8]
This reaction requires the harsh acidic conditions of the so- Stabilisation of the tertiary ester by formation of strong in
called Flaschenträger reaction,[14] and is incompatible with tramolecular hydrogen bonding with the 9-hydroxy group
Eur. J. Org. Chem. 2004, 3413Ϫ3421
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3415

G. Appendino, V. Di Marzo et al.
FULL PAPER
presumably underlies these remarkable effects, which were dent from the detection of a strong NOE effect between one
also detected in lumiphorbol (vide infra). It remains un- of the homovanillate methylene groups (δ ϭ 3.61 ppm) and
clear, however, why the parent polyol 9a could be converted the two protons at C-5 (δ ϭ 1.99 and 3.27 ppm), as well as
into the triester 9b without involvement of the 4-hydroxy with 10-H (δ ϭ 2.61 ppm). On the other hand, transesterifi-
group, the reactivity of which seems to be magnified by the cation of the 13,20-diacetate-12-phenyl acetate 10f afforded
presence of a free 20-hydroxy group.
a complex mixture, as did the Mitsunobu esterification of
the 12,13-diester 10d. These results make it clear that the
rigid scaffold of lumiphorbol engenders the formation of a
web of hydrogen bonds, which makes the tertiary C-4
hydroxy group as reactive in acylation reactions as the sec-
ondary C-12 hydroxy group and the primary C-20
hydroxy group.
Interestingly, when the irradiation of 4α-PPHAV was car-
ried out under oxygen, the 2,3-seco-lactone 11 was obtained
as the major reaction product (51%). The same compound
could be obtained in similar yield (48%) by treatment of
lumi-PPAHV (10a) with peracids in buffered medium. The
photo-cleavage of 10a is presumably initiated by a Norrish
I α-cleavage of the cyclopentane ring, but the further steps
of the reaction are unclear.[21]
We lastly investigated the reduction of the 3-keto group,
stereoselectively preparing the epimeric 3-dihydro deriva-
tives of PPAHV. Reduction with NaBH4 in methanol is
dominated by steric effects, and affords the 3βΗ-hydro de-
rivative 12a as the only reaction product.[9] Conversely, re-
duction with tetrabutylammonium triacetoxyborohydride
followed a chelation-controlled pathway,[9,22] affording the
3βΗ-dihydro derivative 12b.
4α-Phorbol is cup-shaped, and easily undergoes a photo-
chemical [2πs ϩ2πs] cycloaddition to afford lumiphorbol, a
cage-like compound.[20] In the event, irradiation of PPAHV
in degassed ethanol afforded a quantitative yield of the lumi
derivative 10a. Attempts to use lumiphorobol itself as a
platform for the synthesis of specific esters were frustrated
by inability to manipulate its various hydroxy groups selec-
tively, an unfortunate event, given the potential of 4α-phor-
bol to act as a polyfunctionalized, cubane-like scaffold for
iterative, orthogonal acylations. A set of lumiphorbol esters
was therefore obtained by irradiation of the corresponding
4α-phorbol esters,[17] and then subjected to transesterifi-
cation or acylation to change their acylation patterns. Treat-
ment of lumiphorbol 13,20-diacetate (10b) with phe-
nylacetic acid with EDC promotion {EDC ϭ 1-[3-(dimeth-
ylamino)propyl]-3-ethylcarbodiimide hydrochloride} af-
forded the 4,12-bis(phenylacetate) 10c as the only reaction
product, while treatment of the 12,13-diester 10d with
MEM-homovanillic acid gave (after deprotection) the 4,20-
bis(homovanillate) 10e even with stoichiometric (1 molar
equivalent) amounts of acylating reagent. The location of
the tertiary homovanillate at the 4-oxygen function was evi-
3416
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 3413Ϫ3421

Phorbol-Resiniferatoxin (RTX) Hybrids
Table 1. Biological evaluation of the vanilloid activities of phorboid homovanillates (n.d. ϭ not detectable; n.a. ϭ not assayed)
FULL PAPER
Compound
Rat TRPV1
Maximal effect at 10 µ
Human TRPV1
Maximal effect at 10 µ
EC50 (n)
EC50 (n)
(% ionomycin)
(% ionomycin)
Capsaicin (2)
40.0 Ϯ 3.0
85.2 Ϯ 4.1
n.d.
34.6 Ϯ 1.9
43.2 Ϯ 3.1
n.a.
40.2 Ϯ 3.1
3100 Ϯ 230
n.a.
70.1 Ϯ 3.7
65.2 Ϯ 4.1
n.a.
PPAHV (3b)
5d
6d
8b
n.d.
n.a.
n.a.
25.6 Ϯ 2.7
34.8 Ϯ 6.2
34.4 Ϯ 6.0
79.8 Ϯ 5.3
n.a.
n.a.
n.a.
4180 Ϯ 360
3090 Ϯ 510
N.D.
3050 Ϯ 290
n.a.
n.a.
n.a.
11.7 Ϯ 2.9
16.1 Ϯ 3.3
5.0 Ϯ 3.0
65.1 Ϯ 3.9
n.a.
91.0 Ϯ 5.2
653 Ϯ 61
848 Ϯ 72
648 Ϯ 55
2480 Ϯ 280
n.d.
4α-PPAHV (9d)
Lumi-PPAHV (10a)
Oxylumi-PPAHV (11)
3αH-DihydroPPAHV (12a)
3βH-DihydroPPAHV (12b)
3. Biological Evaluation
Conclusions
The activities of the phorboid homovanillates 3b, 5d, 6d,
8b, 9d, 10a 11, 12a and 12b were evaluated by measuring
the entry of Ca2ϩ into embryonic kidney cells transfected
with the rat vanilloid TRPV1 receptor, with evaluation of
potency (EC50) and efficiency (maximal effect at 10 µ con-
centration).[23] Potency and efficiency reflect the affinity for
the CPS binding site on TRPV1, and the capacity to induce
a change eventually translating into a functional response
(calcium efflux), respectively. Despite their increased struc-
tural similarity to RTX relative to PPAHV, the ring C-con-
strained phorboids 5d and 6d did not show any measurable
vanilloid activity. Conversely, the crotophorbolone enol es-
ter 8b retained most of the vanilloid activity of PPAHV.
These observations show that modifications on ring C have
a profound effect on the vanilloid activity of phorbol homo-
vanillates, and highlight the key role of this moiety for bind-
ing to TRPV1.
Changes on the AB ring system that suppress the tu-
mour-promoting potential of phorboid esters also de-
creased their vanilloid potency, while their effect on efficacy
was only minimal. While the 3α-dihydro derivative 12a
showed only marginal vanilloid potency, its epimer 12b was
devoid of detectable vanilloid activity. Surprisingly, 12a
showed an increased efficacy relative to capsaicin.
Ever since its isolation, interest in phorbol within the or-
ganic community has always been high, because of its chal-
lenging structure, remarkable bioactivity, and puzzling reac-
tivity.[8] While preparing a series of phorbol-RTX hybrids
useful as vanilloid probes, we made several unexpected ob-
servations regarding the reactivity of phorboids in terms of
modification of functional groups and constitution. Since
phorboids are remarkably versatile polyhydroxylic scaffolds
for the induction of bioactivity, clarification of the mechan-
istic principles underlying our findings should be important
to make phorboids amenable to predictable chemistry. The
biological evaluation of the final homovanillates in cells
transfected with the murine and human TRPV1 ortholog-
ues showed the existence of strict structure-activity relation-
ships, but also evidenced remarkable differences within the
two species, with a marked (100-fold) decrease in potency
with use of human TRPV1 compared to rat TRPV1. These
differences had already been noticed with PPAHV,[23] and
have recently been interpreted in terms of the primary struc-
tures of the two orthologues.[24] This, and the discovery that
a 4α-phorbol derivative can activate TRPV4, a thermo-, os-
motic and mechanosensor insensitive to capsaicin and
RTX,[25] highlight the relevance of phorboids as selective
probes for neurosciences.
Remarkably, when the measurements of vanilloid activity
were done on cells transfected with the human orthologue
of the receptor, general decreases in both potency and effi-
cacy were observed, and only two compounds (PPAHV and
12a) retained efficacy similar to that of capsaicin. These fin-
dings confirm a previous report on the attenuation of the
vanilloid activity of PPAHV when assayed on human
TRPV1,[23] and show that, despite the close similarity be-
tween human and rat TRPV1, ligand discrimination is still
possible. The remarkable attenuation of potency observed
for phorbol homovanillates in the assays on human TRPV1
caution against the exclusive use of TRPV1-expressing sys-
tems from one only species in screening programs of vanil-
loid activity.
Experimental Section
Caution: 12,13-Diesters of phorbol are severe irritant to skin and
mucous membranes, and several of them display tumour-promot-
ing properties. Their handling should be carried out with latex
gloves and face protection, and avoiding contact with the skin.
General: Column chromatography: Merck silica gel. IR: Shimadzu
DR 8001 spectrophotometer. NMR: Bruker AM 500 (500 MHz
1
1
and 125 MHz for H and 13C, respectively). For H NMR, CDCl3
as solvent, CHCl3 at δ ϭ 7.26 ppm as reference. For 13C NMR,
CDCl3 as solvent, CDCl3 at δ ϭ 77.0 ppm as reference. The 1H
and 13C NMR spectra were fully assigned by a combination of 1D
Eur. J. Org. Chem. 2004, 3413Ϫ3421
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3417

G. Appendino, V. Di Marzo et al.
and 2D (COSY, HMBC, HMQC) techniques. CH2Cl2 was dried by 13) ppm. HRMS (70 eV): m/z ϭ 604.2836 (calcd. for C39H40O6,
FULL PAPER
distillation from CaH2, and THF by distillation from Na/benzo-
phenone. Na2SO4 was used to dry solutions before the evaporation.
Reactions were monitored by TLC on Merck 60 F254 (0.25 mm)
plates, and spots were viewed by UV inspection and/or by staining
with 5% H2SO4 in ethanol and heating. Phorbol was obtained from
commercial Croton oil (Alexis Biochemicals).[8] To avoid undue
manipulation of potentially tumour-promoting compounds, many
intermediates with a free 20-hydroxy group were not fully charac-
terised but directly converted into their corresponding and non-
tumour-promoting 20-esters.
604.2825).
20-Tritylphorboisobutyl Ketone (6b): M.p. 98Ϫ99 °C. IR (KBr): ν˜ ϭ
3389, 3047, 1697, 1628, 1491, 1448, 1248, 1047, 704 cmϪ1 1H
.
NMR (CDCl3): δ ϭ 0.92 (d, J ϭ 6.5 Hz, 18-H/3), 0.98 (s, 16-H/3),
1.30 (s, 17-H/3), 1.79 (br. s, 19-H/3), 1.94 (br. s, 14-H), 2.23 (d.,
J ϭ 19 Hz, 5a-H), 2.57 (q, J ϭ 6.8 Hz, 11-H), 2.60 (d, J ϭ 19.0 Hz,
5b-H), 3.42 (br. s, 10-H), 3.44 (br. s, 20-H/2), 3.40 (m, 8-H), 4.85
(br. d, J ϭ 6.1 Hz, 7-H), ca. 7.30 (m, Trit-H/15), 7.41 (br. s, 1-H)
ppm. 13C NMR (CDCl3/[D6]DMSO): δ ϭ 10.2 (C-19), 11.7 (C-18),
18.3 (C-16), 22.1 (C-17), 39.0 (C-5), 40.2 (C-15), 44.9 (C-11), 45.8
(C-8), 50.6 (C-10), 53.2 (C-14), 69.3 (C-20), 73.2 (C-4), 81.5 (C-12),
84.4 (C-9), 87.0 (Trit), 108.3 (C-13), 121.2 (C-7), 127.0 (Trit), 127.8
(Trit), 128.6 (Trit), 134.8 (C-6), 135.9 (C-2), 159.1 (C-1), 208.5 (C-
3) ppm. HRMS (70 eV): m/z ϭ 604.2843 (calcd. for C39H40O6,
604.2825).
13-Acetyl-20-tritylneophorbol (4c): PCC (664 mg, 3.08 mmol, 2 mol
˚
equiv.) and activated powdered 4-A molecular sieves (1.33 g) were
added to a stirred solution of 13-acetyl-20-tritylphorbol (3c, 1.0 g,
1.54 mmol)[9] in dry CH2Cl2 (35 mL). After stirring overnight at
room temperature, the reaction mixture was worked up by dilution
with diethyl ether (20 mL) and filtration through Celite. After evap-
oration, the residue was purified by column chromatography on
silica gel (10 g, petroleum ether/ EtOAc, 8:2 as eluent) to afford a
white powder (738 mg, 74%). M.p. 125Ϫ127 °C. IR (KBr): ν˜ ϭ
20-Homovanillyl-12-(phenylacetyl)phorbobutanone
(5d):
Phe-
nylacetic acid (excess, 873 mg, 6.4 mmol, 7 mol equiv.), EDC
(1.23 g, 6.4 mmol, 7 mol equiv.) and DMAP (280 mg, 2.3 mmol,
2.5 mol equiv.) were added to a solution of 5b (554 mg, 0.92 mmol)
in dry CH2Cl2 (30 mL). After stirring at room temperature for 72 h,
the reaction mixture was worked up by dilution with CH2Cl2
(20 mL), washing with Na2CO3 (5%, 3 ϫ 10 mL) to remove excess
phenylacetic acid, and drying (Na2SO4). Removal of the solvent
left a yellowish residue of crude 5c (570 mg), which was directly
dissolved in methanolic HClO4 (0.34%, 15 mL). After stirring at
room temperature for 15 min, the reaction mixture was worked up
by dilution with brine and extraction with EtOAc. The organic
phase was washed with water and dried (Na2SO4), and the solvents
were evaporated, to give a foam (270 mg). This was dissolved in
dry CH2Cl2 and treated with MEM-homovanillic acid (244 mg,
0.90 mmol), EDC (174 mg, 0.90 mmol) and DMAP (55 mg, cat.).
After stirring at room temperature for 45 min, the reaction mixture
was worked up by washing with brine, drying (Na2SO4) and evap-
oration. The residue was dissolved in THF (10 mL) and treated
with SnCl4 (100 µL). After 4 h, the reaction mixture was worked
up by dilution with CH2Cl2 and washing with brine. The organic
phase was evaporated, and the residue was purified by column
chromatography (10 g silica gel, petroleum ether/EtOAc, 6:4) to af-
ford 5d (196 mg, 33% from 5b) as a foam. IR (KBr): ν˜ ϭ 3400,
1761, 1720, 1710, 1685, 1439, 1356, 1260, 809, 706 cmϪ1. 1H NMR
(CDCl3): δ ϭ 0.98 (s, 16-H/3), 0.88 (d, J ϭ 6.5 Hz, 18-H/3), 1.24
(s, 17-H/3), 1.78 (br. s, 19-H/3), 2.27 (d, J ϭ 19.0 Hz, 5a-H), 2.43
(d., J ϭ 19 Hz, 5b-H), 2.65 (q, J ϭ 6.8 Hz, 11-H), 2.88 (br. s, 14-
H), 2.88 (br. s, 10-H), 3.51 (br. s, PhAc-H/2), 3.71 (br. s, HMV-H/
2), 3.77 (m, 8-H), 3.86 (s, OMe), 4.45 (br. s, 20-H/2), 5.63 (br. d,
J ϭ 6.1 Hz, 7-H), 6.73 (br. d, J ϭ 8.0 Hz, HMV-H) 6.80 (br. s,
HMV-H), 6.82 (br. d, J ϭ 8.0 Hz, HMV-H), ca. 7.33 (m, PhAc-H/
5), 7.41 (br. s, 1-H) ppm. 13C NMR (CDCl3): δ ϭ 8.9 (C-18), 10.3
(C-19), 19.4 (C-16), 25.4 (C-17), 38.2 (C-5), 41.1 (HMV), 41.6
(PhAc), 44.3 (C-8), 46.2 (C-11), 51.0 (C-14), 53.7 (C-10), 55.9
(OMe), 58.3 (C-15), 69.0 (C-20), 74.3 (C-4), 87.0 (C-9), 100.9 (C-
12), 111.7 (HMV), 114.3 (HMV), 122.1 (HMV), 125.7 (HOMV),
127.6 (PhAc), 127.8 (C-7), 128.8 (PhAc), 129.3 (PhAc), 132.7 (C-
6), 134.1 (PhAc), 136.7 (C-2), 144.8 (HMV), 146.5 (HMV), 154.9
(C-1), 171.3 (HMV), 171.8 (PhAc), 207.0 (C-3), 209.4 (C-13) ppm.
HRMS (70 eV): m/z ϭ 644.2640 (calcd. for C37H40O10, 644.2622).
3420, 1730, 1700, 1445, 1375, 1250, 760, 700 cmϪ1 1H NMR
.
(CDCl3): δ ϭ 1.19 (d, J ϭ 6.5 Hz, 18-H/3), 1.23 (s, 17-H/3), 1.37
(s, 16-H/3), 1.49 (d, J ϭ 4.9 Hz, 14-H), 1.78 (br. s, 19-H/3), 2.17 (s,
OAc), 2.43 (d, J ϭ 19.0 Hz, 5a-H), 2.61 (d, J ϭ 19.0 Hz, 5b-H),
2.94 (m. 11-H), 3.24 (br. s, 8-H and 10-H), 3.55 (br. s, 20-H/2), 5.40
(br. s, -OH), 5.76 (br. s, 7-H), ca. 7.30 (m, Trit-H/15), 7.56 (br. s,
1-H) ppm. HRMS (70 eV): m/z ϭ 646.2917 (calcd. for C41H42O7,
646.2931).
α-Ketol Rearrangement of 13-Acetyl-20-tritylneophorbol (4b):
A. With NaOMe/MeOH: Sodium methoxide (1 , 1.08 mL,
1.08 mmol, 1 mol equiv.) was added dropwise, under nitrogen, to a
solution of 4b (697 mg, 1.08 mmol) in dry MeOH (12 mL), re-
sulting in the development of a violet solution, which rapidly
turned into a thick orange suspension. After 5 minutes, the reaction
mixture was worked up by neutralisation with satd. NH4Cl and
extraction with EtOAc. Washing with brine and drying (Na2SO4)
1
afforded a yellowish powder (660 mg), that, when analysed by H
NMR (300 MHz, CDCl3) showed the presence only of 5b. An ana-
lytical sample was obtained by washing the solid residue with di-
ethyl ether, which afforded a white powder (565 mg, 88%).
B. With EtONa/EtOH: The reaction was carried out in the same
way as described for the NaOMe/MeOH couple, but the violet col-
our observed upon addition of the bases developed into a brown
limpid solution and not an orange paste. From 2.5 g 4b, 2.4 g of a
yellow powder was obtained after workup. When analysed by 1H
NMR (300 MHz, CDCl3), the presence of a ca. 1:9 mixture of 6b
and 5b was detected. The mixture was purified by column chroma-
tography (150 g silica gel, petroleum ether/EtOAc, 5:5) to afford 6b
(1.87 g, 78%) and 5b (267 mg, 10%).
20-Tritylphorbobutanone (5b): M.p. 76Ϫ78 °C. IR (KBr): ν˜ ϭ 3453,
3410, 1761, 1686, 1447, 1223, 1051, 754, 706 cmϪ1 1H NMR
.
(CDCl3): δ ϭ 1.08 (s, 16-H/3), 1.10 (d, J ϭ 6.5 Hz, 18-H/3), 1.37
(s, 17-H/3), 1.81 (br. s, 19-H/3), 2.35 (br. s, 14-H), 2.42 (q, J ϭ
6.8 Hz, 11-H), 2.44 (d., J ϭ 19 Hz, 5a-H), 2.53 (d, J ϭ 19.0 Hz,
5b-H), 2.75 (br. s, 10-H), 3.61 (br. s, 20-H/2), 3.91 (m, 8-H), 5.67
(br. d, J ϭ 6.2 Hz, 7-H), ca. 7.30 (m, Trit-H/15), 7.41 (br. s, 1-H)
ppm. 13C NMR (CDCl3/[D6]DMSO): δ ϭ 8.7 (C-18), 10.4 (C-19), 20-Homovanillyl-12-phenylacetylphorboisobutyl Ketone (6d): Start-
25.7 (C-16), 32.6 (C-17), 38.2 (C-5), 41.6 (C-8), 45.8 (C-11), 55.1 ing from 6b (820 mg), 6d was obtained as described for the syn-
(C-15), 55.3 (C-14), 55.8 (C-10), 69.0 (C-20), 74.3 (C-4), 87.2 (Trit), thesis of 5d from 5b, in an overall 39% yield. The compound was
88.5 (C-9), 97.2 (C-12), 124.3 (C-7), 127.0 (Trit), 127.8 (Trit), 128.6 obtained as a white powder. M.p. 65Ϫ68 °C. IR (KBr): ν˜ ϭ 3452,
(Trit), 136.7 (C-6), 138.3 (C-2), 154.1 (C-1), 207.0 (C-3), 214.7 (C- 1736, 1717, 1516, 1271, 1236, 1132, 1046 cmϪ1. 1H NMR (CDCl3):
3418
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 3413Ϫ3421

Phorbol-Resiniferatoxin (RTX) Hybrids
FULL PAPER
δ ϭ 0.87 (d, J ϭ 6.5 Hz, 18-H/3), 1.04 (s, 17-H/3), 1.18 (s, 16-H/
5.64 (br. s, 7-H), 6.68 (br. d, J ϭ 7.8 Hz, HMV-H), 6.75 (br. s,
3), 1.78 (br. s, 19-H/3), 1.94 (br. s, 14-H), 2.17 (d., J ϭ 19 Hz, 5a- HMV-H), 6.80 (br. d, J ϭ 7.8 Hz, HMV-H), 7.30 (m, PhAc-H/5),
H), 2.47 (d, J ϭ 19.0 Hz, 5b-H), 2.80 (q, J ϭ 6.8 Hz, 11-H), 3.35
7.52 (br. s, 1-H) ppm. 13C NMR (CDCl3): δ ϭ 10.1 (C-19), 16.8
(m, 8-H), 3.40 (br. s, 10-H), 3.49 (br. s, HMV-H/2), 3.65 (br. s, (C-18), 17.7 (C-17), 37.1 (C-5), 39.1 (C-8), 41.0 (HMV and PhAc),
PhAc-H/2), 3.87 (OMe), 4.41 (br. s, 20-H/2), 5.37 (br. d, J ϭ 42.8 (C-11), 48.7 (C-14), 54.9 (C-10), 56.0 (OMe), 69.9 (C-20), 73.3
6.0 Hz, 7-H), 6.70 (br. d, J ϭ 8.0 Hz, HMV-H), 6.77 (s, HMV-H), (C-4), 77.0 (C-9), 111.7 (HMV), 114.4 (HMV), 117.0 (C-16), 121.4
6.80 (br. d, J ϭ 8.0 Hz, HMV-H), ca. 7.3 (m, PhAc-H/5), 7.41 (br.
s, 1-H) ppm. 13C NMR (CDCl3): δ ϭ 10.2 (C-19), 12.1 (C-18), 22.2 129.4 (C-7), 129.6 (PhAc), 133.0 (C-6), 134.4 (C-2), 134.6 (PhAc),
(C-16 and C-17), 38.6 (C-5), 41.0 (C-15), 45.7 (C-8), 41.0 (PhAc), 141.7 (C-15), 144.8 (HMV), 146.5 (HMV), 147.3 (C-13), 159.5 (C-
(C-12), 122.1 (HMV), 125.7 (HMV), 127.2 (PhAc), 128.5 (PhAc),
41.6 (HMV), 45.7 (C-11), 50.5 (C-10), 56.0 (OMe), 69.6 (C-20), 1), 170.3 (PhAc), 171.4 (HMV), 208.5 (C-3) ppm. HRMS (70 eV):
73.0 (C-4), 82.9 (C-12), 84.6 (C-9), 108 (C-13), 111.9 (HMV), 114.8 m/z ϭ 628.2687 (calcd. for C37H40O9, 628.2672).
(HMV), 122.6 (HMV), 125.9 (HMV), 127.8 (PhAc), 127.9 (C-7),
13,20-Diacetyl-12-phenylacetyl-4α-phorbol (9b): TEA (30 mL,
20.9 g, 206 mmol, 16 mol equiv.) and Ac2O (20 mL, 21.1 g,
206.4 mmol, 16 mol equiv.) were added to a solution of 4α-phorbol
(4.7 g, 12.9 mmol) in dry THF/CH2Cl2 (1:1, overall 50 mL). After
stirring for 1 h at room temperature, the reaction mixture was
worked up by dilution with CH2Cl2 and washing with 2  H2SO4
128.6 (PhAc), 129.5 (PhAc), 131.7 (C-7), 133.3 (C-6), 135.0 (C-2),
158.4 (C-1), 207.6 (C-3) ppm. HRMS (70 eV): m/z ϭ 644.2639
(calcd. for C37H40O10, 644.2622).
Fragmentation of 13-Phenylacetyl-20-tritylphorbol (3d): A solution
of 13-phenylacetyl-20-tritylphorbol (3d, 180 mg, 0.25 mmol)[9] and
thiocarbonyldiimidazole (88 mg, 0.50 mmol, 2 mol equiv.) in dry and satd. NaHCO3. After drying (Na2SO4) and evaporation, the
THF (2.0. mL), was heated at reflux for 8 h, and then worked up
by dilution with EtOAc and washing with brine. After drying
residue was purified by column chromatography on silica gel (50 g,
petroleum ether/EtOAc, 4:6 as eluent) to afford 13,20-diacetyl-4α-
(Na2SO4), the organic phase was evaporated, and the residue was phorbol (2.3 g, 49%).[17] A portion of this (700 mg, 1.56 mmol) was
purified by column chromatography on silica gel (10 g, petroleum
ether/EtOAc, 9:1 as eluent) to afford 8a (174 mg, 97%) as a white
dissolved in dry CH2Cl2/THF (1:1, overall 80 mL) and treated with
phenylacetic acid (638 mg, 4.69 mmol, 3 mol equiv.), EDC (894 mg,
powder. M.p. 72Ϫ74 °C. IR (KBr): ν˜ ϭ 3453, 1748, 1491, 1449, 4.69 mmol, 3 mol equiv.) and 286 mg DMAP (2.34 mmol, 1.5 mol
1221, 1136, 700 cmϪ1. 1H NMR (CDCl3): δ ϭ 1.03 (d, J ϭ 6.5 Hz, equiv.). After stirring for 30 min at room temperature, the reaction
18-H/3), 1.58 (br. s, 17-H/3), 1.80 (br. s, 19-H/3), 2.13 (d, J ϭ
mixture was worked up by dilution with CH2Cl2 and washing with
19.0 Hz, 5a-H), 2.43 (d, J ϭ 19.0 Hz, 5b-H), 3.17 (br. s, 10-H), 3.23 brine. After drying (Na2SO4) and evaporation, the residue was
(m, 11-H), 3.40 (m, 14-H), 3.40 (br. s, 20-H/2), 3.41 (m, 8-H), 3.64 purified by column chromatography on silica gel (15 g, eluent
(s, PhAc-H/2), 4.85 (br. s, 16-H/2), 5.04 (br. s, 12-H), 5.63 (br. s, 7-
H), ca. 7.30 (m, PhAc-H/5), ca. 7.30 (m, Trit-H/15), 7.53 (br. s, 1-
H) ppm. 13C NMR (CDCl3): δ ϭ 10.1 (C-19), 16.7 (C-18), 17.8 (C-
petroleum ether/EtOAc, 6:4) to afford 9b (796 mg, 90%) as a white
powder. M.p. 161 °C. IR (KBr): ν˜ ϭ 3414, 1736, 1717, 1637, 1455,
1381, 1246, 1120, 1040, 984 cmϪ1. 1H NMR (CDCl3): δ ϭ 0.81 (d,
17), 37.1 (C-11), 39.4 (C-5), 42.8 (C-8), 41.0 (PhAc), 48.7 (C-14), J ϭ 5.4 Hz, 14-H), 0.93 (s, 16-H/3), 0.97 (s, 17-H/3), 1.02 (d, J ϭ
55.0 (C-10), 68.8 (C-20), 73.5 (C-4), 77.0 (C-9), 116.8 (C-12), 121.3 6.5 Hz, 18-H/3), 1.66 (m, 8-H), 1.83 (br. s, 19-H/3), 2.07 (s, Ac),
(C-16), 126.9 (Trit), 127.2 (Trit), 127.2 (PhAc), 127.7 (Trit) 128.5 2.09 (s, Ac), 2.28 (d, J ϭ 13.9 Hz, 5a-H), 2.88 (m, 11-H), 3.25 (s,
(PhAc), 124.9 (C-7), 129.6 (PhAc), 134.0 (C-6), 134.6 (PhAc), 136.8 10-H), 3.69 (d, J ϭ 13.9 Hz, 5b-H), 3.72 (br. s, PhAc-H/2), 4.26 (d,
(C-2), 142.0 (C-15), 147.4 (C-13), 159.6 (C-1), 170.3 (PhAc), 208.9
J ϭ 11.8 Hz, 20a-H), 4.37 (d, J ϭ 11.8 Hz, 20b-H), 5.25 (br. s, 7-
H), 5.44 (d, J ϭ 10.3 Hz, 12-H), 6.97 (s, 1-H), ca. 7.31 (m, PhAc-
H/5) ppm. HRMS (70 eV): m/z ϭ 566.2504 (calcd. for C32H38O9,
566.2516).
(C-3) ppm. HRMS (70 eV): m/z
ϭ 706.3301. (calcd. for
C47H46O6, 706.3294).
Trityl to Homovanillate Exchange in 8a: A sample of 8a (178 mg,
0.25 mmol) was dissolved in methanolic HClO4 (0.34%). After stir-
13-Acetyl-20-homovanillyl-12-phenylacetyl-4α-phorbol (4α-PPAHV,
ring for 2 h at room temperature, the reaction mixture was diluted 9d): A sample of 9b (687 mg, 1.21 mmol) was dissolved in meth-
with satd. NaHCO3 and extracted with EtOAc. After drying anolic HClO4 (0.35%, 15 mL), and the solution was stirred at room
(Na2SO4) and removal of the solvent, the residue was dissolved in
temperature for 24 h. After neutralisation with solid sodium acetate
dry CH2Cl2 (4 mL) and treated with O-MEM homovanillic acid[9] and concentration, brine and EtOAc were added. The organic
(68 mg, 0.25 mmol, 1 mol equiv.), EDC (48 mg, 0.25 mmol, 1 mol
equiv.) and DMAP (5 mg, cat.). After stirring for 1 h at room
temp., the reaction mixture was worked up by dilution with CH2Cl2
phase was dried (Na2SO4), the solvents were evaporated, and the
residue was washed with diethyl ether to afford 9c (509 mg, 80%
yield). This was dissolved in dry THF (25 mL) and, after the mix-
(5 mL), washing with brine, and evaporation. The crude residue ture had been cooled to 0 °C, triphenylphosphane (498 mg,
was dissolved in dry THF (5 mL) and treated with SnCl4 (15 µL).
After stirring overnight under a nitrogen atmosphere, the reaction
mixture was diluted with EtOAc and washed with satd. NaHCO3.
The organic phase was dried (Na2SO4), and the solvents were eva-
porated. The residue was purified by column chromatography on
silica gel (5 g; petroleum ether/EtOAc, 7:3 as eluent) to afford 8b
(83 mg, 52% from 8a) as a white powder. M.p. 49Ϫ51 °C. IR (KBr):
1.9 mmol, 2 mol equiv.), di-tert-butyl azodicarboxylate (446 mg,
1.94 mmol, mol equiv.) and homovanillic acid (353 mg,
2
1.94 mmol, 2 mol equiv.) were added. After stirring at room tem-
perature for 1 h, the reaction mixture was worked up by evapor-
ation and the residue was taken up in toluene (10 mL). After the
mixture had been kept at 4 °C overnight and filtered, the filtrate
was evaporated, and the residue was purified by column chroma-
tography on silica gel (15 g, petroleum ether-EtOAc, 6:4 as eluent)
ν˜ ϭ 3453, 1736, 1713, 1516, 1497, 1454, 1273, 1236, 1136 cmϪ1
.
1H NMR (CDCl3): δ ϭ 1.02 (d, J ϭ 6.5 Hz, 18-H/3), 1.48 (br. s, to afford a white powder (603 mg, 91%). M.p. 78 °C. IR (KBr):
17-H/3), 1.81 (br. s, 19-H/3), 2.17 (d, J ϭ 19.0 Hz, 5a-H), 2.40 (d,
J ϭ 19.0 Hz, 5b-H), 3.07 (br. s, 10-H), 3.20 (m, 11-H), 3.27 (m, 14-
H), 3.37 (m, 8-H), 3.48 (br. s, HMV-H/2), 3.63 (s, PhAc-H/2), 3.86
(s, HMV-H/3), 4.38 (d, J ϭ 13.0 Hz, 20a-H), 4.46 (d, J ϭ 13.0 Hz,
ν˜ ϭ 3414, 1719, 1605, 1516, 1433, 1375, 1271, 1250, 1148, 984
1
cmϪ1. H NMR (CDCl3): δ ϭ 0.78 (d, J ϭ 5.4 Hz, 14-H), 0.99 (s,
16-H/3), 1.01 (s, 17-H/3), 1.05 (d, J ϭ 6.4 Hz, 18-H/3), 1.54 (m, 8-
H), 1.76 (br. s, 19-H/3), 2.06 (Ac), 2.30 (d, J ϭ 14.0 Hz, 5a-H), 3.26
20b-H), 4.75 (br. s, 16a-H), 4.78 (br. s, 16b-H), 5.02 (br. s, 12-H), (br. s, 10-H), 3.60 (br. s, HMV-H/2), 3.61 (d, J ϭ 14.0 Hz, 5b-H),
Eur. J. Org. Chem. 2004, 3413Ϫ3421
www.eurjoc.org  2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 3419

G. Appendino, V. Di Marzo et al.
FULL PAPER
3.69 (br. s, PhAc-H/2), 3.89 (s, OMe), 4.33 (br. s, 20-H/2), 5.22 (s,
40%) and recovered starting 10d (16 mg). Compound 10e was ob-
7-H), 5.46 (d, J ϭ 10.0 Hz, 12-H), 6.83 (m, HMV-H), 6.85 (d. J ϭ tained as a colourless foam. IR (KBr): ν˜ ϭ 3397, 1800, 1731, 1725,
8.0 Hz, HMV-H), 6.87 (d, J ϭ 8.0 Hz, HMV-H), 6.98 (s, 1-H), ca.
1609, 1524, 1412, 1275, 1251, 1264, 1100, 1096, 982 cmϪ1 1H
7.32 (m, PhA-H/5) ppm. 13C NMR (CDCl3/[D6]DMSO): δ ϭ 10.6 NMR (CDCl3): δ ϭ 0.98, 1.02, 1.09 (s, 16-H/3, 17-H/3, 19-H/3),
(C-19), 11.9 (C-18), 16.9 (C-17), 24.0 (C-16), 24.9 (C-15), 35.0 (C- 1.10 (d, J ϭ 6.4 Hz, 18-H/3), 1.26 (d, J ϭ 6.2 Hz, 14-H), 1.42 (m,
.
5), 36.9 (C-14), 41.0 (HMV), 41.7 (PhAc), 41.9 (C-8), 43.4 (C-11), 11-H), 1.59 (m, 8-H), 1.99 (d, J ϭ 11.8 Hz, 5a-H), 2.04 (s, OAc),
55.9 (C-10), 56.7 (OMe), 65.2 (C-13), 70.7 (C-20), 75.9 (C-12), 77.0 2.39 (t, J ϭ 6.7 Hz, 1-H), 2.61 (d, J ϭ 7.1 Hz, 10-H), 2.79 (t, J ϭ
(C-4), 77.3 (C-9), 111.2 (HMV), 114.8 (HMV), 122.0 (HMV), 124.9 5.2 Hz, 7-H), 3.27 (d, J ϭ 11.8 Hz, 5b-H), 3.51, 3.61 (br. s, 2 ϫ
(C-7), 125.2 (HMV), 127.6 (PhAc), 128.5 (PhAc), 129.6 (PhAc),
HMV-H/2), 3.69 (br. s, PhAc-H/2), 3.89, 3.92 (s, 2 ϫ OMe), 4.07
132.2 (C-6), 134.0 (PhAc), 140.9 (C-2), 144.3 (HMV), 146.2 (HMV, (br. s, 20-H/2), 5.41 (d, J ϭ 7.8 Hz, 12-H), ca 6.70 (m, 2 ϫ HMV-
155.6 (C-1), 170.5 (PhAc), 171.8 (HMV),208.0 (C-3) ppm. HRMS
H/3), ca. 7.30 (m, PhAc-H/5) ppm. HRMS (70 eV): m/z ϭ 852.3369
(70 eV): m/z ϭ 688.2866 (calcd. for C39H44O11, 688.2884).
(calcd. for C48H52O14, 862.3357).
13-Acetyl-20-homovanillyl-12-(phenylacetyl)lumiphorbol
(Lumi-
Synthesis of the 2,3-seco-Lumiphorboid 11
PPAHV, 10a): A degassed ethanolic solution of 4α-PPAHV (9d;
100 mg, 0.14 mmol in 50 mL) was irradiated at 254 nm in a quartz
tube in a Rayonet apparatus. After two hours, the reaction mixture
was worked up by evaporation to afford 10a in quantitative (1H
NMR analysis) yield. Compound 10a was obtained as a white pow-
der, m.p. 62 °C. IR (KBr): ν˜ ϭ 3386, 1773, 1736, 1605, 1516, 1456,
A. By Irradiation of 4α-PPAHV under Oxygen: An ethanolic solu-
tion of 4α-PPAHV (9d; 74 mg, 0.11 mmol in 50 mL) in a quartz
tube was flushed with oxygen for 5 min and then irradiated at
254 nm in a Rayonet apparatus. After two hours, the reaction mix-
ture was worked up by flushing with nitrogen and evaporation, and
the residue was purified by column chromatography to provide 11
(39 mg, 51%) as a foam. IR (KBr): ν˜ ϭ 3400, 1740, 1605, 1516,
1375, 1271, 1244, 1150, 1032, 982 cmϪ1 1H NMR (CDCl3): δ ϭ
.
0.98, 1.02, 1.04 (s, 16-H/3, 17-H/3, 19-H/3), 1.06 (d, J ϭ 6.4 Hz,
18-H/3), 1.24 (d, J ϭ 6.2 Hz, 14-H), 1.43 (m, 11-H), 1.59 (m, 8-H),
1.92 (d, J ϭ 11.8 Hz, 5a-H), 2.09 (s, OAc), 2.34 (dd, J ϭ 7.1,
5.2 Hz, 1-H), 2.64 (d, J ϭ 7.1 Hz, 10-H), 2.75 (t, J ϭ 5.2 Hz, 7-H),
3.17 (d, J ϭ 11.8 Hz, 5b-H), 3.56 (br. s, HMV-H/2), 3.65 (br. s,
PhAc-H/2), 3.89 (s, OMe), 4.00 (br. s, 20-H/2), 5.47 (d, J ϭ 7.8 Hz,
12-H), 6.76 (m HMV-H/3), ca. 7.30 (m, PhAc-H/5) ppm. HRMS
(70 eV): m/z ϭ 688.2886 (calcd. for C39H44O11, 688.2884).
1
1456, 1371, 1273, 1151, 1066, 1032, 982 cmϪ1. H NMR (CDCl3):
δ ϭ 0.95, 1.03, 1.14 (s, 16-H/3, 17-H/3, 19-H/3), 1.16 (d, J ϭ 6.4 Hz,
18-H/3), 1.24 (m, 11-H and 14-H), 1.55 (m, 8-H), 1.76 (d, J ϭ
12.8 Hz, 5a-H), 2.11 (s, Ac), 2.31 (t, J ϭ 6.7 Hz, 1-H), 2.61 (m, 7-
H and 10-H), 3.01 (d, J ϭ 12.8 Hz, 5b-H), 3.49 (br. s, HMV-H/2),
3.65 (br. s, PhAc-H/2), 3.92 (s, OMe), 3.97 (br. s, 20-H/2), 5.41 (d,
J ϭ 7.8 Hz, 12-H), ca. 6.77 (m, HMV-H/3), ca. 7.30 (m, PhAc-H/
5) ppm. HRMS (70 eV): m/z ϭ 704.2822 (calcd. for C39H44O12,
704.2833).
Phenylacetylation of Lumiphorbol 13,20-Diacetate (10b): EDC
(23.2 mg, 0.12 mmol, 1 mol equiv.), phenylacetic acid (16.5 mg,
0.12 mmol, 1 mol equiv.) and DMAP (5 mg, catalytic) were added
to a solution of 10b (54 mg, 0.12 mmol, prepared from 13,20-di-
B. By Treatment of LumiPPAHV (10a) with Peracids: MCPBA
(80%, 18 mg, 0.108 mmol, 1 mol equiv.) and NaOAc (20 mg) were
added to a solution of lumiPPAHV (10a; 75 mg, 0.107 mmol) in
dry CH2Cl2 (10 mL). After stirring for 90 min at room temperature,
the reaction mixture was worked up by dilution with CH2Cl2, fil-
tration and washing with NaOH (2 ). The organic phase was dried
(Na2SO4), the solvents were evaporated, and the residue was puri-
fied by column chromatography on silica gel (petroleum ether/
EtOAc, 5:5) to afford 11 (39 mg, 48%), identical (1H NMR) to the
product obtained from the irradiation of 10a under oxygen.
[8]
acetyl-4α-phorbol as described for 10a from 9d) in dry CH2Cl2
(2 mL). After stirring for 5 min at room temperature, the reaction
mixture was worked up by dilution with CH2Cl2 (5 mL) and
washed with brine. After drying (Na2SO4) and evaporation, the
residue was purified by column chromatography on silica gel (2.5 g,
petroleum ether/EtOAc, 8:2 as eluent) to give 10c (39.2 mg, 47%)
and recovered 10b (15 mg). Compound 10c was obtained as a white
foam. IR (KBr): ν˜ ϭ 3397, 1782, 1740, 1456, 1377, 1240, 1156,
1
1032, 982 cmϪ1. H NMR (CDCl3): δ ϭ 0.73 (d, J ϭ 6.4 Hz, 18-
3βΗ-Dihydro-13-acetyl-20-homovanillyl-12-(phenylacetyl)phorbol
(12b): Tetramethylammonium triacetoxyborohydride (152 mg,
0.56 mmol, 4 mol equiv.) was added to a stirred solution of PPAHV
(100 mg, 0.142 mmol) in acetonitrile/HOAc (1:1, overall 2 mL).
After stirring at room temperature for 4 h, the reaction mixture
was worked up by dilution with EtOAc and washing with satd.
NaHCO3. After drying, the organic phase was evaporated and
purified by column chromatography (5 g silica gel, petroleum ether/
EtOAc, 6:4 as eluent) to afford 12b (48 mg, 48%) as a white powder.
H/3), 1.05, 1.14, 1.19 (s, 16-H/3, 17-H/3, 19-H/3), 1.35 (m, 11-H,
14-H), 1.60 (m, 8-H), 2.04, 2.09 (s, OAc), 2.28 (d, J ϭ 11.8 Hz, 5a-
H), 2.42 (t, J ϭ 6.7 Hz, 1-H), 2.85 (t, J ϭ 5.2 Hz, 7-H), 3.44 (d,
J ϭ 7.2 Hz, 10-H), 3.55 (d, J ϭ 11.8 Hz, 5b-H), 3.65 (br. s,PhAc-
H/2), 3.71 (br. s, PhAc-H/2), 3.87 (br. d, J ϭ 11.6 Hz, 20a-H), 4.06
(br. d, J ϭ 11.6 Hz, 20b-H), 5.45 (d, J ϭ 7.5 Hz, 12-H), ca. 7.30
(m, 2 ϫ PhAc-H/5). HRMS (70 eV): m/z ϭ 684.2926 (calcd. for
C40H44O10, 684.2935).
Treatment of 13-Acetyl-12-(phenylacetyl)lumiphorbol (10d) with
M.p. 75Ϫ77 °C. IR (KBr): ν˜ ϭ 3453, 1734, 1516, 1252, 1375, 1150,
1
MEM-homovanillic Acid: MEM-homovanillic acid (25 mg, 1456, 1021 cmϪ1. H NMR (CDCl3): δ ϭ 0.83 (d, J ϭ 6.5 Hz, 18-
0.097 mmol, 1 mol equiv.) and EDC (11.9 mg, 0.098 mmol, 1 mol H/3), 0.86 (d, J ϭ 5.2 Hz, 14-H), 1.04 (s, 16-H/3), 1.06 (s, 17-H/3),
equiv.) were added to a solution of 10d (59 mg, 0.097 mmol) in dry
1.74 (br. s, 19-H/3), 1.96 (m, 11-H), 2.06 (s, Ac), 2.26 (d, J ϭ
CH2Cl2 (2 mL). After stirring at room temperature for 24 h, the 19.0 Hz, 5a-H), 2.48 (s, 8-H), 2.83 (d, J ϭ 19.0 Hz, 5b-H), 2.96 (s,
reaction mixture was worked up by washing with brine. After dry-
ing (Na2SO4) and evaporation, the residue was dissolved in dry
THF (2 mL) and treated with SnCl4 (10 µL). After stirring for 4 h
10-H), 3.53 (s, HMV H/2), 3.62 (s, PhAc H/2), 3.87 (s, OMe), 4.47
(s, 20-H/2), 5.20 (br. s, OH), 5.38 (d, J ϭ 10.2 Hz, 12-H), 5.53 (s,
OMe), 5.53 (br. s, 3-H), 5.82 (br. s, 1-H), 6.73 (d, J ϭ 8.0 Hz, HMV-
at room temp., the reaction mixture was worked up by dilution H), 6.77 (s, HMV-H), 6.78 (d, J ϭ 8.0 Hz, HMV-H, 7.24 (m, PhAc
with diethyl ether and washing with satd. NaHCO3. After washing
with brine, the organic phase was evaporated, and the residue was
purified by column chromatography (3 g silica gel, petroleum ether/
H/5) ppm. 13C NMR (CDCl3): δ ϭ 14.5 (C-19), 14.6 (C-18), 16.6
(C-16), 21.1 (OAc), 23.7 (C-17), 25.2 (C-15), 36.0 (C-14), 40.4 (C-
8), 41.2 (HMV), 42.5 (PhAc), 43.5 (C-11), 43.4 (C-5), 58.0 (C-10),
EtOAc, 8:2 as eluent) to afford the bis(homovanillate) 10e (34 mg, 65.4 (C-13), 67.7 (C-20), 75.4 (C-4), 76.6 (C-12), 81.3 (C-9), 86.5
3420  2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjoc.org
Eur. J. Org. Chem. 2004, 3413Ϫ3421

Phorbol-Resiniferatoxin (RTX) Hybrids
FULL PAPER
P. Wahl, C. Foged, S. Tullin, C. Thomsen, Mol. Pharmacol.
2001, 59, 9Ϫ15.
C. S. J. Walpole, S. Bevan, G. Bloomfield, R. Breckenridge, I.
F. James, T. Ritchie, A. Szallasi, J. Winter, R. Wrigglesworth,
J. Med. Chem. 1996, 39, 2939Ϫ2952.
For a simplified procedure to obtain RTX from Euphorbia resi-
nifera Berg, see: E. Fattorusso, V. Lanzotti, O. Taglialatela-
Scafati, G. C. Tron, G. Appendino, Eur. J. Org. Chem. 2002,
71Ϫ78.
[5]
[6]
(C-3), 111.7 (HMV), 114.0 (HMV), 122.1 (HMV), 125.7 (HMV),
126.1 (PhAc), 126.2 (PhAc), 131.0 (C-1), 132.2 (C-7), 132.7 (PhAc),
135.2 (PhAc), 134.7 (C-6), 137.7 (C-2), 144.8 (HMV), 146.5
(HMV), 171.5 (HMV), 172.3 (PhAc), 173.6 (OAc) ppm. HRMS
(70 eV): m/z ϭ 6903045 (calcd. for C39H46O11, 690.3040).
[7]
Evaluation of Vanilloid Activity: The activities of compounds at
TRPV1 receptors were evaluated by assessing their effects on cyto-
solic calcium concentrations ([Ca2ϩ]i) in human embryonic kidney
(HEK) 293 cells over-expressing either the human or rat TRPV1,
a kind gift from John Davis (SmithKline Beecham, UK). Cells were
grown as monolayers in minimum essential medium supplemented
with non-essential amino acids, 10% fetal calf serum and 0.2 m
glutamine, and maintained under 95%/5% O2/CO2 at 37 °C. The
effect of the substances on [Ca2ϩ]i was determined by use of Fluo-
3, a selective intracellular fluorescent probe for Ca2ϩ. One day
prior to experiments, cells were transferred into six-well dishes co-
ated with poly--lysine (Sigma) and grown in the culture medium
described above. On the day of the experiment, the cells (50Ϫ60,000
per well) were loaded for 2 h at 25 °C with 4 µ Fluo-3 methyl
ester (Molecular Probes) in DMSO containing 0.04% Pluoronic.
After the loading, cells were washed with Tyrode’s pH ϭ 7.4, tryp-
sinized, resuspended in Tyrode’s and transferred to the cuvette of
the fluorescence detector (PerkinϪElmer LS50B) with continuous
stirring. Experiments were carried out by measurement of cell flu-
orescence at 25 °C (λEX ϭ 488 nm, λEM ϭ 540 nm) before and
after the addition of the test compounds at various concentrations.
Potency is expressed as the concentration exerting a half-maximal
effect (EC50). The efficacy of the effect was determined by compar-
ing it to the analogous effect observed with 4 µ ionomycin.
[8]
[9]
E. Hecker, R. Schmidt, Phorbol esters, the irritant and co-car-
cinogens of Croton tiglium. In Progress in the Chemistry of Or-
ganic Natural Products, 1971, vol. 31, 377Ϫ467.
G. Appendino, G. Cravotto, G. Palmisano, R. Annunziata, A.
Szallasi, J. Med. Chem. 1996, 39, 3123Ϫ3131.
[10] [10a]
L. Liu, A. Szallasi, S. A. Simon, Neuroscience 1998, 84,
[10b]
569Ϫ581.
A. Szallasi, T. Szabo, T. Biro, S. Modarres, P.
M. Blumberg, J. E. Krause, DC. N. Cortright, G. Appendino,
Brit. J. Pharmacol. 1999, 128, 428Ϫ434.
S. F. Victory, G. Appendino, D. G. Vander Velde, Bioorg. Med.
Chem. 1998, 6, 223Ϫ229.
H. Bartsch, E. Hecker, Justus Liebigs Ann. Chem. 1969, 725,
142Ϫ153.
[11]
[12]
[13] [13a]
C. H. DePuy, H. A. Andrist, P. C. Fuenfschilling, J. Am.
[13b]
Chem. Soc. 1974, 96, 948Ϫ950.
J. Org. Chem. 2000, 2777Ϫ2786.
H. Brunner, F. Stöhr, Eur.
[14]
[15]
H. W. Thielmann, E. Hecker, Justus Liebigs Ann. Chem. 1969,
728, 158Ϫ183.
Treatment of 20-acetyl-13-(phenylacetyl)phorbol with mesyl
chloride and pyridine gave a mixture of 8a and its correspond-
ing C-13 ketone, in accordance with what had been observed
with phorbol 13,20-diacetate: H. Bartsch, E. Hecker, Z. Natur-
forsch., Teil B 1969, 24, 99Ϫ106.
G. Krauter, C. W. Von der Lieth, R. Schmidt, E. Hecker, Eu.
J. Biochem. 1996, 242, 417Ϫ427.
P. Jacobi, E. Härle, H. U. Schairer, E. Hecker, Justus Liebigs
Ann. Chem. 1970, 741, 13Ϫ32.
Compounds 9d and 9e could not be completely separated. The
esterification of the 4-hydroxy group was suggested by NOE
experiments, which showed lack of correlation between the aro-
matic homovanillyl protons and 14-H and the methyl (C-18),
which are both close to the 9-oxygen atom.
G. Appendino, A. Minassi, N. Daddario, F. Bianchi, G. C.
Tron, Org. Lett. 2002, 4, 3839Ϫ3841.
E. Hecker, E. Härle, H. U. Schairer, P. Jacobi, W. Hoppe, J.
Gassmann, M. Röhrl, H. Abel, Angew. Chem. Int. Ed. Engl.
1968, 7, 890Ϫ891.
[16]
[17]
[18]
Acknowledgments
We are grateful to Dr. Enrico Mercalli for his help in obtaining
phorbol from croton oil. We thank M. I. U. R. (Rome, Italy) for
financial support within the PRIN2002 framework.
[19]
[20]
[1]
[1a]
For recent reviews, see:
A. Szallasi, Am. J. Clin. Pathol.
2002, 118, 110Ϫ121. [1b] V. Di Marzo, P. M. Blumberg, A. Szal-
lasi, Curr. Opin. Neurobiol. 2002, 12, 372Ϫ379.
[2] [2a]
[21]
Capsaicin: A. Szallasi, P. M. Blumberg, Pharmacol. Rev.
For a different cleavage of the 3,4 bond of phorbol, see: G.
Appendino, G. P. Carello, R. Enriu, J. Jakupovic, J. Nat. Prod.
1999, 51, 159Ϫ211. [2b] Resiniferatoxin: G. Appendino, A. Szal-
`
[2c]
lasi, Life Sci. 1997, 60, 681Ϫ696.
Isoprenoid dialdehydes:
1995, 58, 284Ϫ287.
[22]
[23]
O. Sterner, A. Szallasi, Trend in Pharmacol. Sci. 1999, 20,
459Ϫ464. [2d] Scutigeral: A. Szallasi, T. Biro, T. Szabo, S. Mod-
arres, M. Petersen, A. Klusch, P. M. Blumberg, J. E. Krause,
O. Sterner, Br. J. Pharmacol. 1999, 126, 1351Ϫ1358. [2e] Evodia-
mine: Y. Kobayashi, Y. Nakano, M. Kizaki, K. Hoshikuma, Y.
Yokoo, T. Kamiya, Planta Med. 2001, 67, 628Ϫ633. [2f] Anand-
P.A. Wender, K. Irie, B. L. Miller, J. Org. Chem. 1993, 58,
4179Ϫ4181.
P. McIntyre, L. M. McLatchie, A. Chambers, E. Phillips, M.
Clarke, J. Savidge, C. Toms, M. Peacock, K. Shah, J. Winter,
N. Weerasakera, M. Webb, H. P. Rang, S. Bevan, I. F. James,
Brit. J. Pharmacol 2001, 132, 1084Ϫ1094.
E. Phillips, A. Reeve, S. Bevan, P. McIntyre. J. Biol. Chem.
2004, 279, 17165Ϫ17172.
H. Watanabe, J. B. Davis, S. Darren, G. C. Jerman, G. D.
Smith, P. Hayes, J. Vriens, W. Cairns, U. Wissenbach, J. Prenen,
V. Flockerzi, G. Droogmans, C. D. Benham, B. Nilius, J. Biol.
Chem. 2002, 277, 13569Ϫ13577.
[24]
[25]
amide: D. Smart, J. C. Jerman, Trends in Pharm. Sci. 2000,
[2g]
21, 134.
N-Acyldopamides: C. J. Chu, S. M. Huang, L. De
Petrocellis, T. Bisogno, S. A. Ewing, J. D. Miller, R. E. Zipkin,
N. Daddario, G. Appendino, V. Di Marzo, J. M. Walker, J.
Biol. Chem. 2003, 278, 13633Ϫ13639.
V. Brower, Nature Biotecnol. 2000, 18, 387Ϫ391.
A. Giannantoni, S. M. Di Stasi, M. Savino, R. L. Stephen, P.
Navarra, G. Scivoletto, E. Mearini, M. Porena, J. Urol. 2002,
167, 1710Ϫ1714.
[3]
[4]
Received February 23, 2004
Early View Article
Published Online July 9, 2004
Eur. J. Org. Chem. 2004, 3413Ϫ3421
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3421