Asymmetric synthesis of isobenzofuranone derivatives and their unique character as protein kinase Cα (PKCα) activators

Article abstract of DOI:10.1016/j.tetlet.2009.03.069

Efficient enantio-selective synthesis of conformationally constrained diacylglycerol analogues, 7-substituted isobenzofuranone derivatives, originally developed by us as PKCα ligands, was achieved by asymmetric dihydroxylation and γ-lactone formation via

Full text of DOI:10.1016/j.tetlet.2009.03.069

Tetrahedron Letters 50 (2009) 3609–3612
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Asymmetric synthesis of isobenzofuranone derivatives and their unique
character as protein kinase Ca (PKCa) activators
Go Hirai a, Yosuke Ogoshi b, Megumi Ohkubo a,b, Yuki Tamura a, Toru Watanabe a,b, Tadashi Shimizu a,
Mikiko Sodeoka a,
*
a Synthetic Organic Chemistry Laboratory, Advanced Science Institute, RIKEN, 2-1, Hirosawa, Wako, Saitama 351-0198, Japan
b Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, 2-1-1, Katahira, Sendai, Miyagi 980-8577, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Efficient enantio-selective synthesis of conformationally constrained diacylglycerol analogues, 7-substi-
Received 15 January 2009
Revised 9 March 2009
Accepted 10 March 2009
Available online 16 March 2009
tuted isobenzofuranone derivatives, originally developed by us as PKC
metric dihydroxylation and -lactone formation via ortho-lithiation and carboxylation. A series of
derivatives having straight and/or branched side chains were synthesized and evaluated, and low-nano-
molar-concentration affinity ligands and highly potent PKC activators were found among them. These
a ligands, was achieved by asym-
c
a
potent ligands induced phenotypic change of K562 cells, which is characteristic of PKC activators.
Keywords:
Protein kinase C
Ó 2009 Elsevier Ltd. All rights reserved.
a
Asymmetric dihydroxylation
Isobenzofuranone
Bleb-forming assay
1,2-Diacylglycerol (DAG), which is generated via enzymatic
hydrolysis of various phospholipids, is one of the most important
second messengers. It is an activator of protein kinase C (PKC).
PKC isozymes are known to play important roles in intracellular
signal transduction for cellular events, such as proliferation, differ-
entiation, and apoptosis.1 They are divided into three classes, con-
ity. In contrast, bryostatin is an anticancer agent.4 Therefore, much
effort has been made to create analogues of these natural products;
indeed, several of the analogues obtained so far have interesting bio-
logical activities.5 On the other hand, the nontumor promotingphys-
iological ligand DAG is also of interest. However, DAG is difficult to
utilize as a template for further structural modification because of
its structural flexibility and lability under biological conditions. Con-
formationally constrained analogues of DAG are therefore promis-
ing candidates as DAG surrogates for uncovering the roles of C1
domain-containing proteins.6 Among such DAG analogues, monocy-
ventional PKCs (cPKCs:
a, bI/bII, c), novel PKCs (nPKCs: d, e, g, h),
and atypical PKCs (f, /k). PKCs have a kinase domain that phos-
s
phorylates their substrate proteins and regulatory domains that
control the kinase activity. Regulatory domains of cPKCs are com-
posed of the membrane targeting domains, two C1 domains (C1A
and C1B) and a C2 domain. C1 domain is the binding site of DAG.
The activation mechanism of cPKCs was proposed to be as follows.
The cPKCs initially bind to the membrane via Ca2+-dependent bind-
ing of the C2 domain, and then binding of DAG on the lipid mem-
brane to C1 domain allows release of a pseudo-substrate region
from the active site in the kinase domain and stabilizes the active
form of PKCs. However, the molecular mechanism of PKC activa-
tion is not yet understood in detail.
clic
c-lactones (DAG-lactone derivatives) have been extensively
studied by Marquez and co-workers.7 Independently, we have also
developed bicyclic isobenzofuranone derivatives as C1 domain li-
gands of PKCa 8
. Our initial structure–activity relationship (SAR)
studies showed that chirality is important, and (R)-isobenzofura-
none derivatives are better ligands than the (S)-isomers. It was also
shown that bulkiness of the acyl side chain is important for PKC
affinity, and (R)-pivaloyl derivative 1 was found to be a potent ligand
for PKCa
(Fig. 1B, Ki = 122 nM).8a Here we report a novel enantiose-
Several complex natural products, such as phorbol esters (PMA
and PDBu, Fig. 1A), bryostatin, aplysiatoxins, and tereocidins, are
also known to be C1 domain ligands and strong agonists of cPKCs
and nPKCs.2 These natural products are valuable tools in cancer re-
search and in the development of anticancer therapeutics,3 because
with the exception of bryostatin, they show tumor-promoting activ-
lective synthetic route to isobenzofuranone derivatives having
straight and/or branched side chains. The activities of these com-
pounds were evaluated at the enzyme and cellular levels.
As reported previously,8c a binding model of 1 with PKC
aC1B
domain suggested that an acyl side chain carbonyl group would
form a hydrogen bond with Gly124 in a similar manner to the
3-carbonyl group of phorbol ester. The side chain of the acyl group
would be located between Leu121 and Leu125, and the alkyl chain
on the benzene ring would be buried in the phospholipid bilayer,
* Corresponding author. Fax: +81 48 462 4666.
E-mail address: sodeoka@riken.jp (M. Sodeoka).
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.03.069

3610
G. Hirai et al. / Tetrahedron Letters 50 (2009) 3609–3612
Figure 1. (A) Structure of phorbol esters. (B) Structures of 7-substituted isobenzof-
uranone derivatives 1. (C) Binding model of the complex of PKC C1B domain with 1
and representative interaction between them. (D) Structures of designed isobenzof-
Scheme 1. Asymmetric synthesis of (R)-(+)-8. Reagents and conditions: (a) MOMCl,
iPr2NEt, CH2Cl2, rt (98%); (b) sBuLi, THF, À78 °C, then I2, rt (85%); (c) 6, PdCl2(PPh3)2,
LiCl, DMF, 70 °C (65%); (d) AD-mix-b, tBuOH–H2O, CH3SO2NH2, 0 °C; (e) TBSCl,
Imidazole, DMF, rt (54%, 2 steps, 13% ee); (f) MOMCl, iPr2NEt, CH2Cl2, rt (95%); (g) 6,
a
uranones 2a–e.
PdCl2(PPh3)2, LiCl, DMF, 90 °C (93%); (h) AD-mix-a
, tBuOH–H2O, 0 °C (95%, 78% ee);
(i) TBSCl, imidazole, DMF, rt, (75%); (j) nBuLi (10 equiv), hexane, rt, then CO2 (dry
ice), rt (80%, 78% ee); (k) separation by HPLC (CHIRALPAK AD-H, hex-
ane/iPrOH = 9:1) ((R)-(+)-8, 78%; (S)-(À)-8, 4%); (l) TBAF, THF, rt, (95%).
like the acyl side chain at C12 of phorbol ester (Fig. 1C). Previous
SAR studies8a indicated that shape of the hydrophobic acyl group
is important. Cyclohexylcarbonyl derivative showed comparable
affinity to the pivaloyl derivative 1, but too bulky 3,5-di(tert-
butyl)benzoyl derivative showed negligible binding with PKC.
Based on our binding model, we designed derivatives 2c and 2d
having a 2,2-di(tert-butyl)acetyl group, which is expected to fit
into the space between the two Leu side chains (Fig. 1D).9 Recently,
we also found that the direction of the hydrophobic long alkyl
chain on the benzene ring of isobenzofuranone derivatives was
and the desired c-lactone was formed simultaneously. Protection
of the newly generated primary alcohol as TBS ether gave 8. But,
unfortunately and unexpectedly, the enantioselectivity was disap-
pointing. The optical purity of 8 was only 13% ee even when the
dihydroxylation was conducted in the presence of MeSO2NH2.11
The reason for this low enantioselectivity is unclear, but might
be related to the participation of the bulky diethylaminocarbonyl
group at the ortho-position.12 To solve this problem, we next tried
asymmetric dihydroxylation of 10. After conversion of commer-
cially available m-iodophenol (9) to its MOM-ether, an alkenyl
group was introduced by means of coupling reaction with vinylst-
important for PKC
isobenzofuranone derivatives activate PKC
a
activation. Interestingly, 7- or 6-substituted
strongly, but 4-substi-
a
tuted derivatives do not activate it.8c These facts suggest that a
long alkyl chain would be important for stabilization of the mem-
brane-bound active conformer. Furthermore, in this model, the
lactone carbonyl group would not participate in the interaction
annane 6 to give the
a-benzyloxymethylstyrene derivative 10 in
good yield. Asymmetric dihydroxylation of 10 using AD-mix-
a
pro-
with C1 domain of PKC
interaction with phospholipids and, consequently, to the binding
and activation of PKC . Thus, to examine the roles of the straight
a (Fig. 1C), but might contribute to the
vided the desired diol 11 (78% ee), and the resulting primary alco-
hol was converted into the corresponding TBS ether. Regioselective
ortho-lithiation and carboxylation were conducted with excess n-
a
alkyl side chain and lactone carbonyl group, we also planned to
synthesize 2a, 2b, and 2e (Fig. 1D).
Since an achiral isobenzofuranone diol 3 was the key intermedi-
ate of the previous synthesis,8a we first tried an enzyme-catalyzed
enantioselective mono-acylation/hydrolysis approach, but this was
not successful. Therefore we needed to develop a novel synthetic
butyllithium and CO2 to give the
tio-purity.13 Optically pure (R)-(+)-8 was obtained by HPLC purifi-
cation using chiral phase column, and the absolute
c-lactone 8 without loss of enan-
a
stereochemistry was determined after conversion to 1. The com-
mon intermediate 12 for the preparation of 2a–d was obtained
by removal of the TBS group. Further transformation to 2a–d was
conducted by conventional methods (Scheme 2). Acylation of pri-
mary alcohol with lauroyl chloride or acid chloride 14,14 followed
by treatment with TMSBr, afforded phenol 13 or 15 in good yield.
After alkylation of the phenolic hydroxyl group via Mitsunobu
reaction using dodecanol or alcohol 16,14 synthesis of 2a–d was
completed by hydrogenolysis to remove benzyl protection. To re-
route. We planned to use asymmetric dihydroxylation of a-arylsty-
rene derivative 7 as a key step for creating the chiral tertiary
alcohol.10 Intermediate 7 was prepared from salicylic acid diethyl-
amide (4) by protection with MOM group, selective ortho-lithiation
followed by reaction with iodine, and Stille coupling with 6
(Scheme 1). Dihydroxylation of 7 using AD-mix-b was carried out

G. Hirai et al. / Tetrahedron Letters 50 (2009) 3609–3612
3611
Scheme 2. Synthesis of 2a–e. Reagents and conditions: (a) CH3(CH2)10COCl,
pyridine, DMAP, rt (91%); (b) TMSBr, CH2Cl2, À30 °C (for 13, 98%; for 15, 95%);
(c) 14, pyridine, DMAP, 90 °C (60%); (d) 1-dodecanol, PPh3, DEAD, THF, rt; (e) 16,
PPh3, DEAD, THF, rt; (f) Pd(OH)2/C, H2, MeOH, rt (for 2a, quant.; for 2b, 76%; for 2c,
40%; for 2d, 60%, 2 steps); (g) DIBAL-H, CH2Cl2, À78 °C (88%); (h) CSA, CH(OMe)3/
CH2Cl2 = 1:1, rt (88%); (i) Et3SiH, BF3ÁOEt2, CH2Cl2, À78 °C; (j) TBAF, THF, rt (86%, 2
steps); (k) 14, pyridine, DMAP, 90 °C (70%); (l) TMSBr, CH2Cl2, À30 °C (72%); (m) 16,
PPh3, DEAD, THF, rt; (n) Pd(OH)2/C, H2, MeOH, rt (13%, 2 steps).
Figure 2. Evaluation of the activity of the phorbol ester PMA and isobenzofura-
nones (2a–d). (A) Dose-dependent inhibition of [3H]PDBu binding to PKC
a. (B)
Relative activity of PKCa (100% = activity with 10 lM PMA).
chains at the C7 position participated in strong activation of PKC
and 2c was the most potent activator of PKC . As mentioned above,
we recently showed the importance of the direction of the alkyl
chain on the benzene ring for PKC
activation.8c In this case, all
four compounds have the hydrophobic alkyl group at C7, and are
expected to interact with fatty acid chains in PS lipid (Fig. 1C).
But, it is likely that a straight alkyl chain would have a strong
a,
move the lactone carbonyl group of the key intermediate, (R)-(+)-8
was first reduced with DIBAL, and the resulting hemiacetal was
converted into methyl acetal 17. Further reduction was conducted
with the Et3SiH–BF3ÁOEt2 system. After removal of the TBS group,
isobenzofuran 18 was obtained in good yield. Acylation with 14
and deprotection of the MOM group afforded phenol 19. Although
the yield was unsatisfactory because of the absence of the lactone
carbonyl group, synthesis of the desired 2e was achieved by the
introduction of the bulky alkyl group onto phenol and hydrogenol-
ysis of the benzyl group.15
a
a
anchoring effect, fixing the PKC
resulting in the stabilization of active PKC
fect of the branched alkyl group seemed to be weaker than that of
the straight chain, the direction of the interaction of PKC -2d com-
plex with PS lipid would be less effectively fixed. Therefore, the
population of active PKC would be smaller, and the activation
a-2c complex to the lipid bilayer,
a. Since the anchoring ef-
a
The binding affinity toward PKC
sessed by competitive inhibition of the binding of [3H]PDBu in
the presence of phosphatidyl-
-serine (PS) vesicles (Fig. 2A).16,17
a of these compounds was as-
a
L
ability of 2d would be lower than that of 2c.
Compound 2d (ClogP = 7.43) with two bulky alkyl chains was
found to show the highest binding affinity (Ki = 2.1 nM), and the
affinity decreased dramatically in the order of 2c, 2b, and 2a. The
Ki values were 43 nM (2c, ClogP = 9.14), 755 nM (2b, ClogP = 9.23),
and >3000 nM (2a, ClogP = 10.9), respectively. These results indi-
cated that the branched hydrophobic group greatly improved bind-
ing to PKCa, as we had expected, but its contribution is much
larger when it is introduced into the acyl group (2c) than at the
7-position (2b).
Comparison of the results of the binding (Fig. 2A) and activation
assays (Fig. 2B) indicated that, in the case of the phorbol ester PMA,
both binding and activation of PKC
centration of 1 M, and no further increase in the activation level
was observed. In contrast, relative activities induced by the iso-
benzofuranone derivative 2a–d at 100 M were higher than the
maximum activity induced by PMA. Although binding of 2d and
2c seems to reach the maximum level at 1 and 10 M, respectively,
activation levels continued to increase at concentrations
M. Moreover, significant activation was observed with
a were saturated over the con-
l
l
l
the PKC
a
Next, PKC
a activator potency was measured by comparison of
over 10
l
phosphorylation levels in the absence and presence of the iso-
benzofuranone derivatives 2a–d using a fluorescence-labeled pep-
tide substrate (Fig. 2B).16 As expected, all four compounds showed
the weaker binders 2a and 2b at concentrations at which negligible
inhibition of [3H]PDBu binding by these compounds was observed.
These results could be explained by different relative affinities of
the ligand molecules for the C1A and C1B domains. Cho et al. re-
ported that phorbol esters such as PDBu and PMA have strong pref-
strong, dose-dependent activation of PKC
levels of PKC induced by these compounds were not correlated
with the binding affinity. Relative activity of 2d, which is the stron-
gest binder of PKC , was slightly lower than that of 2c. These re-
sults indicated that branched alkyl chains in 2d were less
effective for the induction of PKC activation, though they contrib-
uted to strong binding of 2d to PKC . In contrast, long straight alkyl
a. However, activation
a
erence for the C1B domain of PKCa 18
Based on their results, it is
.
considered that [3H]PDBu would bind only to the C1B domain un-
der our binding assay conditions, suggesting that only the binding
a
a
affinity to the C1B domain of PKCa can be estimated by this assay.
a
Therefore, we believe that 2a and 2b may bind preferentially to the

3612
G. Hirai et al. / Tetrahedron Letters 50 (2009) 3609–3612
References and notes
1. (a) Nishizuka, Y. Science 1992, 258, 607–614; (b) Yang, C.; Kazanietz, M. G.
Trends Pharmacol. Sci. 2003, 24, 602–608; (c) Newton, A. C. Chem. Rev. 2001,
101, 2353–2364.
2. (a) Kishi, Y.; Rando, R. R. Acc. Chem. Res. 1998, 31, 163–172; (b) Castagna, M.;
Takai, Y.; Kaibuchi, K.; Sano, K.; Kikkawa, U.; Nishizuka, Y. J. Biol. Chem. 1982,
257, 7847–7851; (c) Fujiki, H.; Tanaka, Y.; Miyake, R.; Kikkawa, U.; Nishizuka,
Y.; Sugimura, T. Biochem. Biophys. Res. Commun. 1984, 120, 339–343; (d) Smith,
J. B.; Smith, L.; Pettit, G. R. Biochem. Biophys. Res. Commun. 1985, 132, 939–945;
(e) Berkow, R. L.; Kraft, A. S. Biochem. Biophys. Res. Commun. 1985, 131, 1109–
1116.
3. Teicher, B. A. Clin. Cancer Res. 2006, 12, 5336–5345.
4. Newman, D. J.; Cragg, G. M. J. Nat. Prod. 2004, 67, 1216–1238.
5. Selected recent reports: (a) Yanagita, R. C.; Nakagawa, Y.; Yamanaka, N.;
Kashiwagi, K.; Saito, N.; Irie, K. J. Med. Chem. 2008, 51, 46–56; (b) Wender, P. A.;
Verma, V. A. Org. Lett. 2008, 10, 3331–3334; (c) Keck, G. E.; Kraft, M. B.; Truong,
A. P.; Li, W.; Sanchez, C. C.; Kedei, N.; Lewin, N. E.; Blumberg, P. M. J. Am. Chem.
Soc. 2008, 130, 6660–6661; (d) Mach, U. R.; Lewin, N. E.; Blumberg, P. M.;
Kozikowski, A. P. ChemMedChem 2006, 1, 307–314; (e) Bertolini, T. M.;
Giorgione, J.; Harvey, D. F.; Newton, A. C. J. Org. Chem. 2003, 68, 5028–5036.
6. Marquez, V. E.; Blumberg, P. M. Acc. Chem. Res. 2003, 36, 434–443.
7. Selected recent reports: (a) El Kazzouli, S.; Lewin, N. E.; Blumberg, P. M.;
Marquez, V. E. J. Med. Chem. 2008, 51, 5371–5386. and references therein; (b)
Lee, J.; Lee, J.-H.; Kim, S. Y.; Perry, N. A.; Lewin, N. E.; Ayres, J. A.; Blumberg, P. M.
Bioorg. Med. Chem. 2006, 14, 2022–2031.
8. (a) Baba, Y.; Ogoshi, Y.; Hirai, G.; Yanagisawa, T.; Nagamatsu, K.; Mayumi, S.;
Hashimoto, Y.; Sodeoka, M. Bioorg. Med. Chem. Lett. 2004, 14, 2963–2967; (b)
Baba, Y.; Mayumi, S.; Hirai, G.; Kawasaki, H.; Ogoshi, Y.; Yanagisawa, T.;
Hashimoto, Y.; Sodeoka, M. Bioorg. Med. Chem. Lett. 2004, 14, 2969–2972; (c)
Hirai, G.; Shimizu, T.; Watanabe, T.; Ogoshi, Y.; Ohkubo, M.; Sodeoka, M.
ChemMedChem 2007, 2, 1006–1009.
Figure 3. Bleb-forming of K562 cells induced by PKCa activators.
C1A domain and thus can activate PKCa even at concentrations at
which negligible binding was observed in the binding assay.
Although 2d having the two branched side chains was the
strongest binder to PKC
to PKC was dramatically lower than that of 2d, and the Ki value
is 704 nM.19 Moreover, induction of phosphorylation activity of
PKC by 2e was also decreased, and the relative activity was about
60% even at 100 M. These facts indicate that the lactone carbonyl
group is also essential for strong binding and activation of
PKCa 20,21
a, the binding affinity of 2e (ClogP = 8.67)
a
a
9. Marquez and co-workers reported that DAG and DAG–lactone derivatives with
bulky hydrophobic substituents were potent ligands with low nanomolar
binding affinity. See: (a) Nacro, K.; Sigano, D. M.; Yan, S.; Nicklaus, M. C.; Pearce,
L. L.; Lewin, N. E.; Garfield, S. H.; Blumberg, P. M.; Marquez, V. E. J. Med. Chem.
2001, 44, 1892–1904; (b) Nacro, K.; Bienfait, B.; Lee, J.; Han, K.-C.; Kang, J.-H.;
Benzaria, S.; Lewin, N. E.; Bhattacharyya, D. K.; Blumberg, P. M.; Marquez, V. E.
J. Med. Chem. 2000, 43, 921–944.
l
.
Finally, the activity of these isobenzofuranone derivatives was
evaluated at the cellular level by using the bleb-forming assay of
K562 cells.22 As reported, typical bleb formation was observed on
10. Wang, Z.-M.; Sharpless, K. B. Synlett 1993, 603–604.
11. Bennani, Y. L.; Sharpless, K. B. Tetrahedron Lett. 1993, 34, 2079–2082.
12.
A carboxymethyl group at the ortho-position did not disturb asymmetric
treatment of K562 cells with PDBu (1
the strong ligands 2c and 2d caused similar phenotypic change
at 3 M, whereas 2a, 2b, and 2e were inactive even at 10 M. In
case of 2a and 2b, strong activation of PKC was observed at this
lM, Fig. 3). As expected,
dihydroxylation: Ohzeki, T.; Mori, K. Biosci. Biotechnol. Biochem. 2003, 67,
2240–2244; Moderate ee was observed in the case of a highly substituted
styrene derivative: Ohzeki, T.; Mori, K. Biosci. Biotechnol. Biochem. 2003, 67,
2584–2590.
l
l
a
13. (a) Uemura, M.; Tokuyama, S.; Sakan, T. Chem. Lett. 1975, 1195–1198; (b) Trost,
B. M.; Rivers, G. T.; Gold, J. M. J. Org. Chem. 1980, 45, 1835–1838.
14. Olah, G. A.; Wu, A.; Farooq, O. Synthesis 1989, 566–567.
15. The complete characterization for all new compounds has been done using
NMR and MALDI-TOFMS.
concentration, but low membrane permeability or instability of
the linear ester group under biological conditions might be a rea-
son for the ineffectiveness of these compounds at cell level.
In conclusion, we have developed an asymmetric synthetic route
to isobenzofuranone derivatives. Among them, the derivatives 2c
and 2d, having a branched acyl side chain, were found to be affinity
16. Assay protocol is described in Ref. 8c.
17. Tanaka, Y.; Miyake, R.; Kikkawa, U.; Nishizuka, Y. J. Biochem. 1986, 99, 257–261.
18. Ananthanarayanan, B.; Stahelin, R. V.; Digman, M. A.; Cho, W. J. Biol. Chem.
2003, 278, 46886–46894.
ligands of PKC
PKC induced by these derivatives was also evaluated, and the re-
sults indicated the importance of a straight alkyl chain for strong
activation of PKC . The lactone carbonyl group was also important.
a at low nanomolar concentrations. Activation of
19. This Ki value was determined in
a different assay system: Lewin, N. E.;
a
Blumberg, P. M. Methods Mol. Biol. 2003, 233, 129–156. The Ki value of 2d
determined with this system was 11.2 nM.
20. There are two possible reasons for the observed low affinity of 2e to PKC
Based on our proposed binding model (Fig. 1C), the lactone carbonyl group
would not contribute to the direct interaction with PKC C1B domain;
a.
a
Moreover, 2c and 2d showed remarkable bleb-forming activity in
K562 cells, suggesting that these derivatives may be useful as cell-
permeable DAG analogues for research at the cellular level. Further
study on the biological activities of these derivatives is under way.
a
however, this carbonyl group may participate in the interaction with a head
group of the phospholipids (PS) and contributes to the binding through
stabilization of the ligand–PKC–PS complex.21 Another possibility is that 2d
has an alternative binding mode, in which the lactone carbonyl group instead
of the side chain carbonyl group makes critical hydrogen bond with Gly124. If
this is the case, the hydrogen bonding with Gly124 would be missing in the
Acknowledgments
compound 2e causing drastic decrease in affinity to PKCa.
21. The importance of the both carbonyl groups in DAG–lactone was discussed in
the following literature: Kang, J.-H.; Peach, M. L.; Pu, Y.; Lewin, N. E.; Nicklaus,
M. C.; Blumberg, P. M.; Marquez, V. E. J. Med. Chem. 2005, 48, 5738–5748.
22. Osada, H.; Magae, J.; Watanabe, C.; Isono, K. J. Antibiot. 1988, 41, 925–931.
We thank Ms. Rie Onose, Dr. Hideaki Kakeya, and Professor Hiro-
yuki Osada (RIKEN) for technical advice about bleb-forming assay.