Syntheses, biological evaluation and SAR of ingenol mebutate analogues for treatment of actinic keratosis and non-melanoma skin cancer

Article abstract of DOI:10.1016/j.bmcl.2013.08.038

Ingenol mebutate is the active ingredient in Picato a new drug for the treatment of actinic keratosis. A number of derivatives related to ingenol mebutate were prepared by chemical synthesis from ingenol with the purpose of investigating the SAR and potency in assays relating to pro-inflammatory effects (induction of PMN oxidative burst and keratinocyte cytokine release), the potential of cell death induction, as well as the chemical stability. By modifications of the ingenol scaffold several prerequisites for activity were identified. The chemical stability of the compounds could be linked to an acyl migration mechanism. We were able to find analogues of ingenol mebutate with comparable in vitro properties. Some key features for potent and more stable ingenol derivatives have been identified.

Full text of DOI:10.1016/j.bmcl.2013.08.038

Bioorganic & Medicinal Chemistry Letters 23 (2013) 5624–5629
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Syntheses, biological evaluation and SAR of ingenol mebutate
analogues for treatment of actinic keratosis and non-melanoma skin
cancer
Xifu Liang a, Gunnar Grue-Sørensen a, Kristoffer Månsson a, Per Vedsø a, Anke Soor a, Martin Stahlhut b,
Malene Bertelsen b, Karen Margrethe Engell c, Thomas Högberg a,
⇑
a Chemical Research, LEO Pharma A/S, 55 Industripaken, DK-2750 Ballerup, Denmark
b Biological Research, LEO Pharma A/S, 55 Industriparken, DK-2750 Ballerup, Denmark
c Product Reprofiling, LEO Pharma A/S, 55 Industriparken, DK-2750 Ballerup, Denmark
a r t i c l e i n f o
a b s t r a c t
Ingenol mebutate is the active ingredient in PicatoÒ a new drug for the treatment of actinic keratosis. A
number of derivatives related to ingenol mebutate were prepared by chemical synthesis from ingenol
with the purpose of investigating the SAR and potency in assays relating to pro-inflammatory effects
(induction of PMN oxidative burst and keratinocyte cytokine release), the potential of cell death induc-
tion, as well as the chemical stability. By modifications of the ingenol scaffold several prerequisites for
activity were identified. The chemical stability of the compounds could be linked to an acyl migration
mechanism. We were able to find analogues of ingenol mebutate with comparable in vitro properties.
Some key features for potent and more stable ingenol derivatives have been identified.
Ó 2013 Elsevier Ltd. All rights reserved.
Article history:
Received 28 June 2013
Revised 6 August 2013
Accepted 7 August 2013
Available online 14 August 2013
Keywords:
Ingenol derivatives
Ingenol 3-acylates
Oxidative burst
Cytokine release
PKC activation
Immune system activation
Cell death induction
A search for bioactive compounds in plants used in traditional
medicine has led to the identification of ingenol mebutate
(PEP005, ingenol 3-angelate, 1, Fig. 1) from Euphorbia peplus.1,2 This
compound has been developed as the active ingredient in PicatoÒ,
a new drug for field treatment of actinic (solar) keratosis (AK)
offering a short treatment schedule (2–3 days), providing effective
and sustained clearance of AK lesions with a predictable onset and
short duration of local skin responses.1,3
Ingenol mebutate (1) is a 3-mono-ester of the diterpene ingenol
and angelic acid. More than thirty different 3-mono-esters of inge-
nol have been isolated from the plant genus Euphorbia, mainly with
non-cyclic unbranched lipophilic aliphatic acids like hexadecanoic,
decadienoic, decatrienoic and dodecatetraenoic acid.4,5 3-Mono-
esters of ingenol display high affinity towards protein kinase C
(PKC),6 a family of related serine/threonine kinases mediating a
number of important cellular signal transduction responses.7 Inge-
keratinocyte cell death and (ii) induction of a lesion-directed im-
mune response, at least partially mediated by PKC.9
In our search for novel analogues of ingenol mebutate (1) with
improved properties, such as chemical stability, potency in activa-
tion of the immune system and therapeutic efficacy, we have
investigated the consequences of minor structural changes of 1.
The ingenol scaffold is a very rigid structure with four hydroxyl
groups in the southern part of which only the 3-OH is esterified
with angelic acid in 1. We have explored the importance of the
5- and 20-OH groups as well as the geometry of the seven-mem-
bered ring. Furthermore, we have prepared transposed esters at
O-3, and investigated cyclic ether lactones connecting O-3 and
O-4 in order to understand the importance of the ester function
in relation to biological activity. The results from these studies
led us to focus on aliphatic ester analogues of 1 having intact
ingenol scaffolds. Herein we report the preparation, the chemical
and biological characterization of these novel ingenol 3-angelate
analogues and the identification of compounds having improved
properties.
nol mebutate (1) is an activator of novel (d,
e, g, h) and classical
(a, bI, bII,
c
) PKC isoenzymes,8 The therapeutic effect of ingenol
mebutate in the treatment of actinic keratosis is believed to be
caused by a dual mechanism of action: (i) induction of aberrant
With a few modifications we have adopted a previously re-
ported method10 for the preparation of the new 3-acyl derivatives
of ingenol (Scheme 1). The formation of ingenol 5,20-acetonide (3),
previously described,11 can potentially lead to several mono
⇑
Corresponding author. Tel.: +45 44 94 58 88.
E-mail address: thomas.hogberg@leo-pharma.com (T. Högberg).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.08.038

X. Liang et al. / Bioorg. Med. Chem. Lett. 23 (2013) 5624–5629
5625
H
(4)
H
a
O
H
H
H
H
O
H
O
O
O
3
H
H
H
5
H
b
c
O
HO
HO
H
O
OH
1
20
O
O
HOHO
HO
HO
O
HO
O
O
HO
O
Figure 1. Ingenol mebutate (1, PEP005).
6
5
7
acetonide products and a bis-3,4-5,20-acetonide. In our hands, the
main product 3 was isolated after crystallization in 73% yield. By
chromatographic isolation from the mother liquor, we obtained a
minor 3,4-regioisomer (4), which was used to prepare some
5-O and 20-O-methyl ether analogues of ingenol 3-angelate (cf.
Schemes 2 and 3). The angeloylation to form 1 required special
conditions to avoid isomerization induced by Michael addition by
nucleophilic catalysts.12 For other 3-acyl-ingenols not associated
with this issue, standard procedures, such as Steglich coupling con-
ditions or using acyl chlorides, could be applied in the preparation
of I. The final acid catalyzed deprotection step to produce II was a
minor modification of a previously described method.10,13
20-O-Methyl-ingenol mebutate (7) was synthesized from the
3,4-protected ingenol 4 as described in Scheme 2. Treatment of 4
with dimethyl sulfate and lithium bis(trimethylsilyl)amide as base
provided 5 in a reasonable yield. Removal of the protecting group
followed by angeloylation gave 7.
The synthesis of 5-O-methyl-ingenol mebutate (11) was also
based on 4 (Scheme 3). The primary alcohol was selectively pro-
tected as a triphenylmethyl (Tr) ether. Methylation of 8 followed
by removal of the two protecting groups under mild acidic condi-
tions delivered 5-O-methyl-ingenol (9). The primary alcohol of 9
again required protection prior to esterification which was realized
with a tert-butyldimethylsilyl ether. Finally, angeloylation of 10
followed by removal of the silyl group under mild conditions pro-
vided 11.
Scheme 2. Synthesis of 20-O-methyl-ingenol mebutate (7). Reagents and condi-
tions: (a) (CH3O)2SO2 (1.2 equiv), LiHMDS (1.05 equiv), THF, 0 °C ? rt, overnight,
35%; (b) aq HCl (4 M), THF, rt, 20 h, 51%; (c) angelic anhydride (1 equiv), Cs2CO3
(1.1 equiv), CH3CN, 0–5 °C, overnight, 47%.
H
H
O
O
H
H
a
b, c
H
H
(4)
O
HO
O
HO
HO
OTr
O
OH
8
9
H
H
O
O
H
H
e, f
d
H
H
O
O
HO
HO
HO
O
OH
O
OTBS
10
11
Scheme 3. Synthesis of 5-O-methyl-ingenol mebutate (11). Reagents and condi-
tions: trityl chloride (1.1 equiv), DMAP, Et3N, CH2Cl2, rt, 62%; (b) (CH3O)2SO2
(3 equiv), LiHMDS (2 equiv), THF, ꢀ78 °C ? rt, 1.2 h; (c) MeOH, concd HCl (cat), rt,
1 h, 7% over two steps; (d) tBuMe2SiCl (2.4 equiv), DMAP (2.9 equiv), CH2Cl2, 89%;
(e) angelic anhydride (3.7 equiv), Cs2CO3 (4 equiv), CH3CN, rt, 1 h, 70 °C, 1 h; (f)
MeOH, concd HCl (cat), rt, 0.5 h, 28% over two steps.
Scheme 4 depicts the preparations of the derivatives 13 and 14
with modified seven-membered rings. The synthesis of the fluoro
derivative 13 was inspired by Appendino et al.14 Selective acetyla-
tion of the primary alcohol in ingenol mebutate provided 12. Treat-
ment of 12 with diethylaminosulfur trifluoride followed by
removal of the acetyl group delivered 13. The two double bonds
of ingenol show different reactivities due to the steric hindrance.11
Thus, the epoxide 14 was prepared by oxidation of 1 with m-chlo-
H
H
a
(1)
O
H
O
H
b, c
H
H
F
O
H
O
d
O
HO
HO
O
O
HO
H
OAc
OH
H
13
12
O
H
O
OH
O
HO
HO
H
14
H
H
O
H
O
O
H
Scheme 4. Syntheses of fluoro analogue 13 and epoxide 14. Reagents and
conditions: (a) Ac2O (3 equiv), Et3N, CH2Cl2, 88%; (b) DAST (2 equiv), CH2Cl2,
ꢀ78 °C, 2 h; (c) Na2CO3, MeOH, rt; (d) MCPBA (4.2 equiv), saturated aq NaHCO3,
CH2Cl2, rt, 3 h.
H
a
H
H
H
+
HO
HO
O
O
HO
HO
HO
O
OH
HO
OH
O
3
Ingenol (2)
4
roperoxybenzoic acid and the stereochemistry was assigned from a
1H NOESY spectrum showing the distance between H-7 and H-8 to
be about 4 Å in three dimensional space, which is consistent with a
trans-configuration.
The 3-ether 15 and 3-ether-4-lactone 17 modifications were
made under the same conditions in two steps (Scheme 5). Alkyl-
ation of the 5,20-acetonide 3 with ethyl 2-chloroacetate led to a
lactone as main product via a two-step reaction. However, alkyl-
ation with the bulky tert-butyl ester prevents formation of lactone.
Hydrolysis under acidic conditions provided the two 3-ether deriv-
atives 15 and 17. A cyclic 3,4-carbonate 16 was prepared by react-
ing 3 with 1,10-carbonyldiimidazole followed by deprotection as
described above.
b or c
H
O
H
O
H
O
H
(II)
d or e
H
O
H
(I)
O
O
HO
R
O
HO
HO
O
R
OH
Scheme 1. General synthetic route from ingenol to ingenol 3-acylates (II). Reagents
and conditions: (a) acetone, PTSA (cat), rt, 0.5 h, 73% for 3, 4.5% for 4; (b) RCO2H,
DCC, DMAP, CH3CN, rt ? 140 °C; (c) RCOCl, DMAP, DIPEA, CH3CN, rt ? 140 °C; (d)
aq HCl/THF; (e) aq HCl/MeOH, rt.

5626
X. Liang et al. / Bioorg. Med. Chem. Lett. 23 (2013) 5624–5629
(3)
test systems. Selective epoxidation of the D
6,7-double bond of 1
b, c
a, c
H
led to compound 14 with one order of magnitude less potency in
ROS and cytokine release assays compared to 1. Thus, the modest
modifications of the ingenol scaffold leads to significantly less
active compounds (cf. Table 1).
H
d, c
H
O
H
O
H
H
H
O
O
H
H
O
O
HO
HO
Ingenol is a 1,2,3,5-tetra-ol and this particular alignment makes
acyl migration possible.10 In aqueous solution ingenol mebutate
will form an equilibrium between the 3-, 4-, 5- and 20-mono-ange-
lates in an approximate ratio of 1:0:1:18 and with no immediate
hydrolysis (Scheme 6). A pH-dependent mechanism for acyl migra-
tion in aqueous solution has been proposed for corticosteroid
esters,19 which we believe to be applicable for this system as well.
Compounds with improved stability would prolong the shelf-life
and facilitate the conditions for distribution of the drug. Ingenol
mebutate is most stable at pH 2.0–4.5. In order to quickly rank
the stability of new ingenol 3-acylates we determined the recovery
at pH 7.4 after 16 h at ambient temperature. Under these condi-
tions 60% ingenol mebutate was recovered. The degradation prod-
ucts of the 3-mono esters of ingenol were identified by LC–MS and
they were generally the acyl migration products: 5-mono esters
and 20-mono esters. It can be noted that 11 and 13 lacking the
5-OH was stable (>95%) and 7 lacking the 20-OH had a stability
(77%) between 1 and 11, which supports the acyl migration mech-
anism in the degradation depicted in Scheme 6.
The purpose of making 3-(t-butoxycarbonylmethyl) ingenol
(15) was to produce a derivative not capable of acyl migration,
but still containing an ester carbonyl in a neighboring position.
The ‘transposition’ of the ester carbonyl one C-atom away from
O-3 of ingenol did produce a stable compound (95% recovered after
16 h) but with greatly reduced ROS and cytokine release activity
(Table 1). Likewise, the ether lactone 17 was inactive. The cyclic
3,4-carbonate 16 was roughly 20 times less active than ingenol
mebutate (cf. Table 1).10 Thus the position of the ester carbonyl
group as well as the free rotation to allow the carbonyl group to
be placed in an optimal position for target interaction is of impor-
tance as previously shown in simulations for ingenol 3-benzoate in
complex with PKCd.20
O
OH
O
OH
HO
O
O
O
O
17
15
OH
HO
16
Scheme 5. Synthesis of compounds modified in the 3-position. Reagents and
conditions: (a) ethyl 2-chloroacetate (5 equiv) K2CO3 (8 equiv), CH3CN, 80 °C, 18 h;
or (b) t-butyl 2-bromoacetate (5 equiv); (c) aq HCl (4 M), THF, rt, 16 h; (d) 1,10-
carbonyldiimidazole (2 equiv), KHMDS (2 equiv), THF, rt, 28%.
To test the biological activity, potencies for the induction of
necrotic cell death were determined in HeLa cells, and immune
response-related effects were measured as oxidative burst induc-
tion, that is, release of reactive oxygen species (ROS), in polymor-
phonuclear (PMN) leukocytes as well as cytokine release (IL8 and
TNFa
) in human primary epidermal keratinocytes.15 Preincubation
with Bisindolylmaleimide I,16 an inhibitor of classical and novel
PKC isoforms, abolished ingenol mebutate-mediated cytokine
release and PMN oxidative burst, indicating a requirement for
PKC activation in these read-outs. Activation of PKCd was deter-
mined for several compounds to confirm the correlation to oxida-
tive burst and cytokine release potency (Table 2).17 As expected,
ingenol (2) itself was lacking activity in all assays (cf. Ref. 18).
The 20-O methyl ether of ingenol mebutate (7) was completely
devoid of any oxidative burst and cytokine release activity, which
confirms the importance of a free 20-OH group capable to engage
as hydrogen bond donor as previously shown for other ingenol
3-acylates with long-chain aliphatic acids.5,10,14 The 5-O methyl
ether of ingenol mebutate (11) was almost two orders of magni-
tude less potent than 1 in the ROS and cytokine release assays. Loss
of activity could point towards the need of a free 5-OH group, but
the steric influence of the 5-methoxy group on the conformation of
the 20-OH group shown to be pivotal for the activity cannot be
ruled out. The allylic fluoride, 13, also lacking a 5-OH, showed
approximately 100-fold less cellular activity, which could be a
combination of a missing 5-OH and a change of required spatial
position of the 20-OH caused by the double bond. No measurable
PKCd activation could be observed for this compound in
accordance with the putative involvement of PKC in the cellular
As minor modifications of the ingenol scaffold led to inferior
biological activity we started the next phase with exploring the
SAR features related to small modifications of the angelate acyl
group. We focused on stability as well as ROS and cytokine release
potency. Angelic acid is a weak carboxylic acid (pKa 4.30)21 and
even weaker acids should form more stable esters. However,
significantly weaker aliphatic carboxylic acids were not readily
Table 1
Chemical stability and biological activity of 3-mebutate esters of modified ingenol scaffolds and of 3-ether ingenols
No
Stabilitya
Oxidative burst (ROS)b
EC50 nM Emax (%)
TNF
a
releasec
Emax (%)
IL8 releasec
EC50 (nM)
Necrosisd
% Recovered
EC50 (nM)
Emax (%)
LC50 (lM)
1
2
7
11
13
14
15
16
17
60
>95
77
>95
>95
Nde
95
Nd
25
8.7
113
—
—
11.2
98
—
—
131
67
156
—
112
—
10.3
95
—
—
230
>400
400
Nd
Nd
Nd
311
400
Nd
10,000
10,000
281
1240
50
3300
181
10,000
10,000
10,000
748
1210
367
10,000
253
10,000
10,000
10,000
738
1130
320
10,000
171
10,000
119
—
57
119
—
109
—
114
111
124
—
a
The chemical stability over 16 h was evaluated in an aqueous buffer with less than 30% organic solvent at pH 7.4. Reported as % recovered material.25
PMN respiratory burst after 40 min incubation of test compound was quantified by measuring fluorescence expressed in relative light units.26 EC50 denotes the test
b
compound concentration producing 50% of the maximum effect given by 1. Emax indicates the maximal response in relation to 1.
c
Cytokine secretions were measured in human adult log-phase primary epidermal keratinocytes after incubation of test compound for 6 h at 37 °C.27 The EC50 and Emax
calculations according to oxidative burst protocol.
d
HeLa cells were treated with test compounds for 30 min at 37 °C and then measured the remaining metabolic activity. LC50 denotes concentration giving 50% loss of
metabolic activity.28
e
Nd: not determined

X. Liang et al. / Bioorg. Med. Chem. Lett. 23 (2013) 5624–5629
5627
Table 2
Chemical stability and biological activity of ingenol 3-acylates
H
O
H
H
O
R
HO
O
HO
OH
No
R-group
Stabilitya
Oxidative burst (ROS)b
TNF
a
releaseb
Emax (%)
IL8 releaseb
EC50 (nM) Emax (%)
Necrosisc
PKCd activationd
% Recovered
EC50 (nM)
Emax (%)
EC50 (nM)
LC50
(lM)
EC50 (nM)
1
60
8.7
113
11.2
98
10.3
95
230
4.1
18
19
20
0
9.1
17
110
112
109
802
91
—
821
93
—
Nd
Nd
Nd
Nd
5
10,000
66
10,000
86
Nd
55
4.3
82
74
283
21
22
20
30
6.1
5.2
114
102
213
29
77
246
21
74
406
129
Nd
Nd
101
100
Et
23
24
45
75
7.5
5.9
92
96
81
97
112
8.6
75
237
174
Nd
5.3
106
8.0
102
25
26
72
45
7.4
5.9
126
126
4.9
3.1
110
101
3.8
2.1
97
87
103
109
5.2
Nd
Ph
4-MePh
27
28
18
30
25
19
88
318
112
87
72
313
120
115
63
Nd
Nd
Nd
Nd
104
Et
Et
29
60
6.3
109
27
62
43
59
86
4.5
30
31
30
10
26
10
112
111
243
450
81
60
256
882
83
55
Nd
Nd
Nd
Nd
32
33
30
35
6.5
6.0
107
104
65
58
47
69
54
43
Nd
Nd
Nd
10,000
10,000
112
34
35
36
60
30
70
6.7
6.0
7.6
99
10,000
88
42
60
62
10,000
182
45
54
52
Nd
Nd
61
Nd
Nd
2.9
110
127
13
22
H
H
37
70
8.3
136
36
55
27
65
77
Nd
H
O
N
38
39
68
55
49
28
120
117
1600
52
47
147
54
45
92
10.6
Nd
Ph
N
O
10,000
10,000
Nd
Ph
40
41
CH3-(CH2)8–
Nd
Nd
10
28
137
106
11
8.4
96
115
8.0
7.3
94
109
117
400
5.3
4.7
CH3-(CH2)14
–
eNd: not determined.
a
Stability was measured according to footnote in Table 1.25
ROS, TNF
Necrosis was measured according footnote in Table 1.28
Activation potency of PKCd was measured by the use of an in vitro phosphorylation assay using a PKC-peptide substrate.29
a
and IL8 release were measured according to footnotes in Table 1.26,27
b
c
d

5628
X. Liang et al. / Bioorg. Med. Chem. Lett. 23 (2013) 5624–5629
H
H
O
(ingenol 3-hexadecanoate, 41) may also exhibit anti-tumour activ-
H
O
ity.24 Ingenol 3-decanoate and 3-hexadecanoate (40 and 41) are
potent PKCd activators and close to equipotent with the 3-angelate
1 in the ROS and cytokine release assays, but these lipophilic com-
pounds are in a special risk zone and might not supply an accept-
able safety profile as therapeutic drugs.
O
H
H
O
H
H
H
H
O
HO
HO
HO
HO
HO
O
HO
HO
O
O
OH
OH
O
The present study shows that the ingenol scaffold conveys sev-
eral unique structural features, including the 5- and 20-hydroxyl
groups, required for the biological activity of the 3-angelate ester
1 linked to induction of oxidative burst, cytokine release and
necrosis. However, these essential hydroxyl groups also impact
the stability of the ester derivatives as they provide a structural
framework for facile acyl migration. The ester moiety positioned
in a specific manner in the 3-position was also shown to be essen-
tial. We have also provided support, using positive and negative
data, for the role of PKC in stimulating the immune response by re-
Scheme 6. Angeloyl migration in aqueous solution exemplified with 1.
available, so we focused on steric hindrance as means of control-
ling acyl migration.22 A number of aliphatic esters II were prepared
according to Scheme 1 and tested.15
The series of simple
a,b-unsaturated 3-acyl ingenol analogues
showed that no -substituent gave poor stability (18 and 19) and
a
low potency in cytokine release. The latter result is probably due
to significant rearrangement by acyl migration of the test sub-
stances during the test conditions (6 h/37 °C/pH 7.4) to the inactive
5- and 20-esters, which had not proceeded during the shorter
oxidative burst assay (40 min/37 °C/pH 7.4). Surprisingly, the
b,b-disubstituted isomer showed comparable ROS release and
lease of reactive oxygen species (ROS) and cytokines IL8 and TNFa.
An extensive SAR exploration showed that a few new compounds
were associated with equal or slightly improved overall properties
compared to ingenol mebutate. Thus, the
tuted 24, the -phenyl-substituted 25, the
a
,b,b-trimethyl-substi-
a
a-methylcyclohexyl es-
stability despite the lack of
a
-substituent (20 vs 1). On the other
ter 36, and the noradmantane 37 warrant further investigations on
their usefulness in treatment of actinic keratosis and possibly non-
melanoma carcinomas (e.g., BCC, SCC and Bowen’s disease) as they
also display an interesting spread in their ability to induce kerati-
nocyte cell death.
hand, the -mono-substituted acrylates 21 and 22 had comparable
a
activity on ROS release but were still clearly inferior to the angelate
1 regarding stability. The ethyl substituted derivative 22 displayed
significantly better effect on cytokine release and necrosis than the
methyl analogue 21. The tigloyl stereoisomer 23 showed lower
cytokine release potency and was somewhat less stable than the
angelate 1, despite a more favorable pKa of 5.0221 (see above). In
this particular case, the lower stability of the isomer 23 may well
be caused by sterically more favourable conditions for acyl migra-
Acknowledgments
We acknowledge Thomas Vifian, Karin Hvidtfeldt Hansen, Svitl-
ana Tkach, Nina Ravn Borch, Ninette Winther Hansen, Nannette
Svendsen for skillful assistance.
tion. Thus, the alkyl-disubstituted (a,b or b,b) 3-acryl ingenol ana-
logues showed slightly better stability than the monosubstituted
compounds, albeit potency in cytokine release was still inferior
to the angelate 1. Of the simple acrylate derivatives only the
References and notes
a
,b,b-trimethyl-substituted 24 appeared to be slightly superior to
the angelate 1 in all assays performed. Alternatively, a phenyl sub-
stituent on the -position of the unsaturated acyl group provided
1. (a) Lebwohl, M.; Sohn, A. Expert Rev. Dermatol. 2012, 7, 121; (b) Berman, B. Clin.
Cosmet. Investig. Dermatol. 2012, 5, 111; (c) Ali, F. R.; Wlodek, C.; Lear, J. T.
Dermatol. Ther. 2012, 2, 8.
2. Vasas, A.; Rédei, D.; Csupor, D.; Molnár, J.; Hohmann, J. Eur. J. Org. Chem. 2012,
5115.
a
comparable PKCd activation but appeared to improve stability
and cytokine release potency compared to the level of 1 (25),
whereas b-substitution only improved cytokine release over 1 (26).
Saturated aliphatic acyl groups gave a similar picture. 3-Butan-
oyl ingenol10 (27) had poor stability and probably as a consequence
also poor cytokine release potency. Alpha-substituents in the acyl
group (28 and 29) increased, as expected, the stability and the lat-
ter was equipotent to the angelate 1 in the tests performed, includ-
ing PKCd activation (IC50 4.5 nM compared to 4.1 nM for 1).
A series of three to six-membered cyclic acyl carboxylates
(30–35) all had poor stability, with the exception of the cyclohex-
enyl 34, and all compounds were also inferior to the angelate 1
regarding cytokine release properties even if ROS release were
3. (a) Lebwohl, M.; Swanson, N.; Anderson, L. L.; Melgaard, A.; Xu, Z.; Berman, B.
N. Engl. J. Med. 2012, 366, 1010; (b) Martin, G.; Swanson, N. J. Am. Acad.
Dermatol. 2013, 68, S39; (c) Keating, G. M. Drugs 2012, 72, 2397; (d) Lebwohl,
M.; Shumack, S.; Gold, L. S.; Melgaard, A.; Larsson, T.; Tyring, S. K. JAMA
Dermatol. 2013, 1. http://dx.doi.org/10.1001/jamadermatol.2013.2766.
4. Opferkuch, H. J.; Hecker, E. Tetrahedron Lett. 1974, 261.
5. Gotta, H.; Adolf, W.; Opferkuch, H. J.; Hecker, H. Z. Naturforsch. 1983, 39b, 683.
6. (a) Winkler, J. D.; Hong, B.-C.; Bahador, A.; Kazanietz, M. G.; Blumberg, P. M.
Bioorg. Med. Chem. Lett. 1993, 3, 577; (b) Winkler, J. D.; Kim, S.; Harrison, S.;
Lewin, N. E.; Blumberg, P. M. J. Am. Chem. Soc. 1999, 121, 296.
7. (a) Bosco, R.; Melloni, E.; Celeghini, C.; Rimondi, E.; Vaccarezza, M.; Zauli, G.
Mini-Rev. Med. Chem. 2011, 11, 185; (b) Reyland, M. Front. Biosci. 2009, 14,
2386; (c) Steinberg, S. Physiol. Rev. 2008, 88, 1341.
8. Kedei, N.; Lundberg, D. J.; Toth, A.; Welburn, P.; Garfield, S. H.; Blumberg, P. M.
Cancer Res. 2004, 64, 3243.
9. (a) Hampson, P.; Kavanagh, D.; Smith, E.; Wang, K.; Lord, J. M.; Ed Rainger, G.
Cancer Immunol. Immunother. 2008, 57, 1241; (b) Ersvaer, E.; Kittang, A. O.;
Hampson, P.; Sand, K.; Gjertsen, B. T.; Lord, J. M.; Bruserud, Ø. Toxins 2010, 2,
174; (c) Rosen, R. H.; Gupta, A. K.; Tyring, S. K. J. Am. Acad. Dermatol. 2012, 3,
486; (d) Stahlhut, M.; Bertelsen, M.; Hoyer-Hansen, M.; Svendsen, N.; Eriksson,
A. H.; Lord, J. M.; Scheel-Toellner, D.; Young, S. P.; Zibert, J. R. J. Drugs Dermatol.
2012, 11, 1181.
10. Sorg, B.; Hecker, E. Z. Naturforsch. 1982, 37b, 748.
11. Opferkuch, H. J.; Adolf, W.; Sorg, B.; Kusumoto, S.; Hecker, E. Z. Naturforsch.
1981, 36b, 878.
12. Liang, X.; Grue-Sørensen, G.; Petersen, A. K.; Högberg, T. Synlett 2012, 23, 2647.
13. Sorg, B.; Hecker, E. Z. Naturforsch. 1982, 37b, 1640.
14. Appendino, G.; Tron, G. S.; Cravotto, G.; Palmisano, G.; Annunziata, R.; Baj, G.;
Surico, N. Eur. J. Org. Chem. 1999, 3413.
15. (a) Grue-Sørensen, G.; Liang, X.; Högberg, T. PCT Int. Appl. WO2012/085189 A1,
2012.; (b) Grue-Sørensen, G.; Liang, X.; Högberg, T.; Månsson, K.; Vedsø, P. PCT
Int. Appl. WO2012/083954 A1, 2012.
16. Martiny-Baron, G.; Kazanietz, M. G.; Mischak, H.; Blumberg, P. M.; Kochs, G.;
Hug, H.; Marmé, D.; Schächtele, C. J. Biol. Chem. 1993, 268, 9194.
17. Tauber, A. Blood 1987, 69, 711.
comparable to 1. Only the
a-trisubstituted esters 36 and 37 were
marginally more stable than ingenol mebutate, and ROS and cyto-
kine release data were comparable to those of 1, including the
PKCd activation measured for 36. In addition both 36 and 37
appeared to have a higher cell death induction potential. Acyl
groups with oxime functionalities were also investigated (38 and
39), without reaching notable potency levels. Thus, both oximes
displayed poor cytokine and ROS release and 38 showed moderate
PKCd activation.
From the diversity of structures in Table 2 it appears that
potency in the cytokine release and oxidative burst assays do not
depend on very strict structural demands on the acid of the ester
group. However, it is well known that increased lipophilicity of
linear acid side chains enhances the tumor promoting properties
of the ingenol 3-acylates,23 even though the same compound

X. Liang et al. / Bioorg. Med. Chem. Lett. 23 (2013) 5624–5629
5629
18. Hasler, C. M.; Acs, G.; Blumberg, P. M. Cancer Res. 1992, 52, 202.
19. Bundgaard, H.; Hansen, J. Int. J. Pharm. 1981, 3, 197.
20. Pak, Y.; Enyedy, I. J.; Varady, J.; Kung, J. W.; Lorenzo, P. S.; Blumberg, P. M.;
Wang, S. J. Med. Chem. 2001, 44, 1690.
21. Brown, H. C.; McDaniel, D. H.; Hafliger, O. Determ. Org. Struct. Phys. Methods
1955, 567. http://research.chem.psu.edu/brpgroup/pKa_compilation.pdf.
22. Nicholls, A. W.; Akira, K.; Lindon, J. C.; Farrant, R. D.; Wilson, I. D.; Harding, J.;
Killick, D. A.; Nicholson, J. K. Chem. Res. Toxicol. 1996, 9, 1414.
Envision plate reader. EC50 was calculated as the concentration of test
compound producing 50% of the maximum effect given by ingenol
3-angelate (1). Emax indicates the maximal response in relation to 1.
27. Human adult log-phase primary primary epidermal keratinocytes were sub-
cultured in 96-well plates at 10,000 cells/well and incubated with titrated test
compound for 6 h at 37 °C in humidified air/CO2 (95%/5%). Secreted TNFa levels
were quantified using Meso Scale Discovery 4-spot cytokine plates. Secreted
IL8 levels were based on homogeneous time-resolved fluorescence (Human IL-
8 HTRFÒ kit, CisBio). EC50 and Emax were calculated according to previous
reference.
23. Sorg, B.; Schmidt, R.; Hecker, E. Carcinogenesis 1987, 8, 1.
24. Itokawa, H.; Ichihara, Y.; Watanabe, K.; Takeya, K. Planta Med. 1989, 55, 271.
25. A DMSO stock solution containing a known amount of compound was diluted
with pre-heated (37 °C) aqueous buffer pH 7.4 to an organic content 630% v/v.
After thorough shaking, the solution was placed in the HPLC auto sampler
(37 °C) and injected within 5 min and then repeatedly injected over a period of
16 h. Based on the decrease of area of the compound signal (UV detection at
normally 270 nm) the recovery of the compound over time was assessed.
26. Polymorphonuclear leukocytes (PMN) were isolated and purified from fresh
buffy coats and incubated for 40 min at ambient temperature with titrated test
28. HeLa cells were treated with increasing concentrations of ingenol esters up to
400 lM for 30 min at 37 °C. Subsequently, metabolic activity was quantified by
the use of the resazurin-based dye formulation PrestoBlue (Invitrogen). LC50
was calculated as the concentration of test compound producing 50% loss of
metabolic activity.
29. PKCd was tested in the KinaseProfiler™ assay (Millipore) using 0.05 mg/mL
phosphatidylserine in the absence of diacylglycerol to allow for observations of
any stimulatory effects. The compounds were serially diluted, at
compound (highest test concentration 10
l
M) pre-mixed with hydroethidine.
semilogarithmic intervals, for twelve points starting from 10 lM. Assays
The PMN respiratory burst was quantified by measuring fluorescence
expressed in relative light units at 579 nm (excitation: 485 nm) using an
were started by the addition of ATP. Data points for the EC50 determinations
were performed in duplicate.