Exploring the influence of indololactone structure on selectivity for binding to the C1 domains of PKCα, PKCε, and RasGRP

Article abstract of DOI:10.1016/j.bmc.2017.03.022

C1 domain-containing proteins, such as protein kinase C (PKC), have a central role in cellular signal transduction. Their involvement in many diseases, including cancer, cardiovascular disease, and immunological and neurological disorders has been extensi

Full text of DOI:10.1016/j.bmc.2017.03.022

Accepted Manuscript
Exploring the influence of indololactone structure on selectivity for binding to
the C1 domains of PKCα, PKCε, and RasGRP
Eleonora Elhalem, Lucía Gandolfi Donadío, Xiaoling Zhou, Nancy E. Lewin,
Lia C. Garcia, Christopher C. Lai, James A. Kelley, Megan L. Peach, Peter M.
Blumberg, María J. Comin
PII:
S0968-0896(16)31461-4
DOI:
http://dx.doi.org/10.1016/j.bmc.2017.03.022
BMC 13619
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
23 December 2016
6 March 2017
9 March 2017
Please cite this article as: Elhalem, E., Donadío, L.G., Zhou, X., Lewin, N.E., Garcia, L.C., Lai, C.C., Kelley, J.A.,
Peach, M.L., Blumberg, P.M., Comin, M.J., Exploring the influence of indololactone structure on selectivity for
binding to the C1 domains of PKCα, PKCε, and RasGRP, Bioorganic & Medicinal Chemistry (2017), doi: http://
dx.doi.org/10.1016/j.bmc.2017.03.022
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Exploring the influence of indololactone structure on selectivity for binding to the C1 domains
of PKCα, PKCε, and RasGRP
Eleonora Elhalema, Lucía Gandolfi Donadíoa, Xiaoling Zhoub, Nancy E. Lewinb, Lia C. Garciaa,
Christopher C. Laic, James A. Kelleyc, Megan L. Peachd, Peter M. Blumbergb and María J. Comina*
a Laboratory of Organic Synthesis, Center of Research and Development in Chemistry, National
Institute of Industrial Technology, Buenos Aires, Argentina; bLaboratory of Cancer Biology and
Genetics, Center for Cancer Research, National Cancer Institute, Bethesda, MD 20892, USA;
cChemical Biology Laboratory, Center for Cancer Research, National Cancer Institute at
Frederick, National Institutes of Health, Frederick, MD 21702, U.S.A.
dBasic Science Program, Leidos Biomedical Research Inc., Chemical Biology Laboratory,
Frederick National Laboratory for Cancer Research, National Institutes of Health, Frederick, MD
21702, U.S.A.
* Corresponding author. Tel.: +54-11-4724-6289; fax: +54-11-4724-6289; email:
jcomin@inti.gob.ar.
Abbreviations: BCl3, boron trichloride; CAN, ceric ammonium nitrate; DAG,
diacylglycerol; DBU, 1,8-diazabicyclo[5.4.0]undec-7-ene; DMAP, 4-
dimethylaminopyridine; GEFs, guanine nucleotide exchange factors; LiHMDS, lithium
hexamethyldisilazide; MRCKs, myotonic dystrophy kinase-related Cdc42-binding kinases
(MRCKs); PDBu, [20-3H]phorbol 12,13-dibutyrate; PKC, protein kinase C; PKD, protein
kinase D; PMP, p-methoxyphenyl; RasGRP, Ras guanine nucleotide-releasing protein;
1

TBAF, tetra-n-butylammonium fluoride; TBDMS, tert-butyldimethylsilyl; TBDPS, tert-
butyldiphenylsilyl.
Abstract
C1 domain-containing proteins, such as protein kinase C (PKC), have a central role in
cellular signal transduction. Their involvement in many diseases, including cancer,
cardiovascular disease, and immunological and neurological disorders has been extensively
demonstrated and has prompted a search for small molecules to modulate their activity. By
employing a diacylglycerol (DAG)-lactone template, we have been able to develop ultra
potent analogs of diacylglycerol with nanomolar binding affinities approaching those of
complex natural products such as phorbol esters and bryostatins. One current challenge is
the development of selective ligands capable of discriminating between different protein
family members. Recently, structure-activity relationship studies have shown that the
introduction of an indole ring as a DAG-lactone substituent yielded selective Ras guanine
nucleotide-releasing protein (RasGRP1) activators when compared to PKCα and PKCε. In
the present work, we examine the effects of ligand selectivity relative to the orientation of
the indole ring and the nature of the DAG-lactone template itself. Our results show that the
indole ring must be attached to the lactone moiety through the sn-2 position in order to
achieve RasGRP1 selectivity.
1. Introduction
The second messenger DAG plays a central role in cell signaling from both receptor
tyrosine kinases and G-protein-coupled receptors.1 Upon receptor stimulation, the
2

activation of phospholipase C catalyzes phosphatidylinositol 4,5-bisphosphate hydrolysis,
liberating DAG and inositol 1,4,5-triphosphate. The rise in DAG concentration is
transduced into downstream signals via activation of seven families of signaling proteins
through binding of DAG to their C1 domains, which function as the DAG recognition
motif. The most extensively studied proteins containing these domains are members of the
PKC family, which belongs to the superfamily of serine/threonine kinases and is known to
regulate numerous cellular functions such as cell growth, cell differentiation, metabolism,
and apoptosis.2 After the identification of the PKCs, six other families of proteins with
homologous, DAG-responsive C1 domains have been identified. The protein kinase D
family is involved in Golgi function, proliferation, metastasis and apoptosis.3 The
chimaerins act as inhibitors (GAPs, GTPase activating proteins) for the small GTPase Rac
and are candidate tumor suppressors.4 The Unc-13 family members are responsible for
promoting vesicle priming.5 The DAG kinases terminate DAG signaling by
phosphorylating DAG.6 MRCK is a downstream effector of cdc42 involved in filopodia
formation, contributing to tumor invasion.7 Finally, the RasGRP family members function
as activators (GEFs, GTP Exchange factors) for Ras.8
PKC was first identified as a cellular receptor for the phorbol ester tumor promoters
more than 30 years ago.9,10 Today, it is recognized that the involvement of the PKC
isoforms in cellular biology extends far beyond its role in tumor promotion.11 Their central
role in cellular signal transduction underlies their extensive involvement in many diseases,
including cancer, cardiovascular disease, and immunological and neurological disorders.
Thus far, the clinical trial results of PKC modulators have been disappointing, largely due
to inadequate therapeutic effect and/or unanticipated adverse reactions. Given PKC’s
critical roles in both normal physiology and disease, this family of kinases remains an
3

enticing target for drug development. In this context, the development of selective
modulators will contribute to the unraveling of the biology of PKC and to the future
success in developing drugs for PKC-mediated diseases.11
RasGRP family members function as activators (GEFs, guanine nucleotide
exchange factors) for the Ras family of small GTPases and are prominently expressed in
blood cells. RasGRP malfunction likely contributes to autoimmunity and may contribute to
blood malignancies.12 In addition, a role for RasGRP3 in prostate cancer 13 and melanoma
14 and of RasGRP1 in skin cancer15 has been demonstrated. Furthermore, it has been shown
that RasGRPs are targets of the anticancer drug ingenol-3-angelate.16 Consequently, given
the important biological roles of RasGRP family members, the discovery of selective
agents capable of specifically interacting with their C1 domains could provide exciting lead
structures for drug development and biochemical studies.
C1 domains are zinc finger structures of ~50 amino acids. Those that bind phorbol
ester / diacylglycerol are termed “typical” whereas those that do not do so are referred to as
“atypical”.17 DAG-lactones have provided a powerful platform for probing C1 domain
structure-activity relations. They combine impressive structural simplification compared to
the structurally complex phorbol esters while still achieving comparable nanomolar
potencies.18 A current challenge is the development of ligands possessing selectivity,
whether between the different families of C1 domain containing proteins or between
isoforms within a single family, while retaining high potency for their selective target. A
key previous finding using a combinatorial chemical approach was that DAG-lactones
substituted with sn-1 and sn-2 aromatic rings displayed much higher binding affinities for
RasGRP1/3 than for PKCα and other PKC isozymes.14
4

Building on these findings, exploration of heteroaryl moieties at the sn-2 position
yielded the DAG-indololactone 1 as the most selective and potent compound in the series
for the activation of RasGRP3.20 This compound showed a 22-fold more potent binding
affinity for RasGRP3 as compared to PKCα and exhibited subnanomolar affinities. Starting
with this lead DAG-indololactone 1, we synthesized a family of regioisomers that probed
the influence on potency and selectivity of the position of the linkage between the indole
ring and the lactone moiety.16 All compounds were potent and selective activators of
RasGRP1 and RasGRP3 when compared to PKCα and revealed that the orientation of the
1-methyl-1H-indole ring on the DAG-lactone plays a critical role in selectivity, with the
most selective compound having the linkage at position 3 of the indole ring (1).
In the present study, we have synthesized and evaluated a new family of DAG-
indololactones in which the 1-methyl-1H-indole ring is attached through the sn-1 ester
position rather than in the sn-2 position and we have then varied the nature of the
substituent in the sn-2 position (Figure 1). As in previous studies, DAG-indololactones
were synthesized as racemates but bearing in mind that the R isomer would be twice as
potent.22 The results presented here show that both binding affinity and selectivity are
decreased when the heterocyclic ring is present at the sn-1 position.
5

O
O
O
O
N
O
HO
O
O
N
O
R1
HO
R2
sn-1 indololactones
; R = R = CH
2
3
1
2
3
; R = H, R = C H19; E
1
2
9
4; R1 = H, R2 = CH((CH2)2CH3)2; E
5; R1 = H, R2 = CH2CH(CH2(i-Pr))2; E
1
sn-2 indololactone
; R = CH CH(CH (i-Pr))2, R2 = H; Z
6
7
8
1
2
2
; R = H, R =
; E
1
2
; R = H, R =
; E/Z
1
2
Figure 1. Structures of parent sn-2 indololactone 1 and target sn-1 indololactones.
2. Results
2.1. Chemistry
The synthesis of the target compounds was accomplished through a well established
procedure utilizing a sequential alkylation-elimination and acylation steps.17 Previously, we
presented the synthesis of a family of regioisomers of lead compound 1 that differ in the
position of the linkage between the 1-methyl-1H-indole ring and the DAG-lactone moiety
at the sn-2 position.16 During the course of that work it became evident that the stability of
indole ring towards deprotection conditions depended on its substitution pattern. From
previous work we knew that 3-substituted indole rings were stable to the deprotection
chemistry used for silyl ethers, so, we decided to synthesize the simple sn-1 indololactone
derivative 2 starting with the asymmetrically protected DAG-lactone 9.24
Aldol condensation with acetone followed by mesylation and elimination of the
aldol product gave 10 in 62% yield (Scheme 1). The benzyl ether group was selectively
removed with BCl3 in 82% yield giving pure compound 11 after column chromatography
purification. Acylation employing 1-methyl-1H-indole-3-carbonyl chloride and subsequent
6

treatment with triethylamine trihydrofluoride to remove the tert-butyldiphenylsilyl
ether afforded the expected lactone 2.
Scheme 1. Synthesis of indololactone 2.
Reagents and conditions: (a) 1. (CH3)2CO, LiHMDS, THF, -78 ºC, 4 h; 2. (i) CH3SO2Cl, Et3N, 0
°C, 2 h; (ii) DBU, CH2Cl2, room temp, 24 h, 62%; (b) BCl3, CH2Cl2, -78 °C, 2 h, 82%; (c) 1-
methyl-1H-indole-3-carbonyl chloride, Et3N, DMAP, room temp, 24 h, 85%; (d) HF.Et3N, THF,
reflux, 24 h, 78%.
During scale up of the synthesis of lead compound 1, we learned that 3-substituted indole
rings were also stable during Lewis Acid treatment required for benzyl removal
(unpublished results). So, for the preparation of target compounds 3, 5, 6 and 8 we
employed lactone 13 that was available in higher amounts (Scheme 2). In most cases, the
alkylation step generated mixtures of the E/Z-isomers; the only exception was the aromatic
aldehyde that afforded exclusively E-16. As with previously synthesized DAG-lactones, the
7

geometry of the exocyclic double bond was assigned by 1H NMR. The vinyl proton
corresponding to the E-isomer appears as a characteristic multiplet that is further downfield
compared to that of the corresponding Z-isomer.23 After separation of the E/Z-isomers by
column chromatography, 14 (E-isomer) and 15 (individual regioisomers E and Z) were
converted to the corresponding DAG-lactones 3, 5 and 6, respectively. Selective
deprotection of the TBDMS group with TBAF, acylation of the free primary hydroxyl
group, and final deprotection of the second primary hydroxyl group were carried out with
very good yields. In the case of the aromatic derivative 16, the treatment with TBAF led to
the desired monoalcohol 20 but with a high degree of isomerization, presumably due to the
basic nature of the reagent. The isomerization was confirmed by HPLC analysis and two-
dimensional NMR experiments (see Supplementary Data). As the isomers of 20 were not
chromatographically separable, we continued the DAG-lactone synthesis with the mixture.
Thus, after acylation and benzyl ether removal compound 8 was obtained as an E/Z
mixture.
Scheme 2. Synthesis of indololactones 3-8.
8

Reagents and conditions: (a) 1. RCHO, LiHMDS, THF, -78 ºC, 2-4 h; 2. (i) CH3SO2Cl, Et3N, 0
°C, 5 h; (ii) DBU, CH2Cl2, room temp, 24 h; (b) TBAF,THF, room temp, 30 min; (c) CAN,
CH3CN/H2O, 0 ºC, 20 min; (d) 1-methyl-1H-indole-3-carbonyl chloride, Et3N, DMAP, room temp,
24 h; (e) BCl3, CH2Cl2, -78 °C, 1 h.
At this point we had several examples of compatibility between lactones bearing the
3-substituted indole moiety and typical conditions employed for benzyl group cleavage
(BCl3,-78 °C, CH2Cl2). Consequently, in order to avoid double bond isomerization we
prepared compound 7 starting with known lactone 2525 protected with benzyl and p-
methoxyphenyl (PMP) groups and obtained excellent results. Compound 4 was synthesized
following the same strategy.
2.2. Binding of ligands to PKCα, PKCε and RasGRP1
9

Table 1. Binding selectivity of DAG-lactones 1- 8 for PKCα and PKCε versus RasGRP1
[3H]PDBub
1
12.9 1.1
31.0 1.6
0.58 0.02
22
53
0.42
5.0
2
3
106 16
59.6 9.5
63.3 8.3
1.7
0.94
1.8
6.4
4
504 49
295 41
133 12
3.8
2.2
1.7
5.2
5
73 12
27.3 7.9
31.3 4.9
2.3
0.87
2.7
6.6
6
83.1 6.3
55 10
22.6 3.1
3.7
2.4
1.5
6.6
7
257 38
119 14
40 11
6.5
3.0
2.2
7.0
8
116 18
131 38
22.6 1.8
5.1
5.8
0.89
7.0
0.28 0.02
698 57
301 31
666 13
1.08
0.45
2.3
PKCα
PKCε
RasGRP1
PKCα/RasGRP1
PKCε/RasGRP1
PKCα/PKCε
ClogPc
0.22 0.05
0.40 0.03
---
---
---
---
2.6
a Values (nM) represent mean SEM of the Kd or Ki from at least triplicate independent experiments.
b [Kd(nM)].
c The Clog P values were calculated using ChemBioDraw Ultra, version 12.0.2.
The binding affinities of compounds 1-8 for PKCα, PKCε and RasGRP1 were determined
in vitro by competition with bound [20-3H]phorbol 12,13-dibutyrate (PDBU) in the
presence of 100 µg/ml phosphatidylserine as previously described.26 For characterization,
we examined PKCα, PKCε, and RasGRP1. We selected PKCα and PKCε because these
isoforms had behaved differently from one another in a previous study, in which our lead
DAG-indololactone showed marked selectivity for RasGRP1/3, compared to PKCα, with
intermediate selectivity relative to PKCε.27 We limited our comparison to RasGRP1, since
both RasGRP1 and RasGRP3 closely resembled one another in binding affinity for aryl-
DAG lactones27 and for the DAG-indololactones in our previous study.21 In general, none
of the compounds were appreciably more potent than parent compound 1 for any of the
studied proteins (Table 1). Although all compounds showed equal or stronger affinity for
RasGRP1 than for PKCα, the obtained values of selectivity, ranging from 1 to 6.5-fold,
indicated that none of the synthesized DAG-indololactones improved the selectivity of the
parent compound 1 (22-fold in parallel measurements, Table 1). Likewise, compound 1
remained the most selective for PKCε with a selectivity of 53, compared to values of 0.45
to 5.8 for compounds 2-8. In terms of potencies, compounds 6 and 8 were the most potent
for RasGRP1 but 56-fold and 39-fold less potent than the phorbol ester PDBu and
compound 1, respectively.
10

We have observed in previous studies18 that the binding affinity or Ki for the DAG-lactones
is correlated with their lipophilicity (logP). This is because binding to the C1 domain
requires ligand partitioning into the phosphatidylserine lipid bilayer (or the cell membrane
in vivo) and the formation of a ternary complex between C1 domain, ligand, and the
membrane. As expected from its lower lipophilicity, compound 2 was the least potent of the
sn-1 indololactones studied here, and there is a general trend toward increasing affinity with
increased logP (Table 1). The direction and orientation of the aliphatic chains relative to the
lactone ring is also important.18 This can be seen here in the striking difference between
compounds 1 and 4, where the 1-methyl-1H-indole ring and the branched alkyl group are
reciprocally substituted. Similarly, compound 8, bearing a mixture of 1:1 E/Z isomers at the
sn-2 aromatic group, presented approximately a 2-fold difference in Ki values for PKCα
and RasGRP1 compared to the pure E isomer (7).
2.3. Modeling
The parent compound 1 and the reciprocally substituted compound 4 were docked into the
C1 domain of RasGRP1. The available crystal structure of RasGRP128 is relatively low
resolution (3 Å) and the C1 domain has no ligand bound. We compared this structure to
that of the phorbol-bound PKCδ C1b domain29 to determine whether the binding site would
be suitable for docking, and found that while the backbone trace of the binding site loops
could be aligned quite closely, residue Q568 at the bottom of the binding site was in a
conformation that would clash with a bound ligand. The side chain amide orientation in this
glutamine residue also appeared to be mis-assigned, since when flipped it could form three
hydrogen bonds to surrounding residues (Figure 2). Before docking, we therefore adjusted
the conformation of Q568 to an orientation similar to that seen in the equivalent residue
11

(Q257) in the phorbol-bound C1b domain of PKCδ.29 Both compounds adopted the
expected binding mode for DAG-lactones, in which hydrogen bonds are formed between
the C1 domain backbone and the lactone carbonyl oxygen as well as the hydroxymethyl
group. The overall docking scores and the calculated hydrogen bonding energy were
virtually identical between the two compounds due to this similarity in binding mode. The
indole group does not appear to form any strong intermolecular interactions with binding
site residues in either docked structure (Figure 3). To explain the difference in binding
affinity between compounds 1 and 4, therefore, we consider a model of the ternary complex
between C1 domain, ligand, and the cell membrane, which as discussed above is the
relevant structure for understanding DAG-lactone activities. Although the exact depth of
penetration and orientation of the RasGRP1 C1 domain relative to the membrane bilayer is
not known experimentally, we can develop a reasonable estimate based on the location of
hydrophobic residues in the binding site loops. The positioning of the indole group relative
to the bilayer differs significantly between compounds 1 and 4. In compound 1 (Figure 3A),
the indole is predicted to be located in the interfacial region of the bilayer whereas when
compound 4 is bound (Figure 3B), the indole ring is directed more deeply into the bilayer
to interact with the hydrophobic lipid acyl chains.
12

Q568
Figure 2. Comparison of the C1 domain binding sites of PKCδ C1b with bound phorbol
(1PTR),29 shown in light blue, and RasGRP1 (4L9M),28 shown in black. The original
position of residue Gln 568 in the RasGRP1 structure is colored white, and its adjusted
position, as used for docking, is in black. Hydrogen bonds formed by this residue to the
backbone are shown as dashed green lines.
A
B
V565
V565
L550
K551
L550
K551
G564
W563
G564
W563
P552
P552
L562
L562
T553
T553
Figure 3. Model of the RasGRP1 C1 domain relative to the lipid bilayer. The colored
background represents the location of bilayer regions: blue – polar region of charged lipid
headgroups and water; green – interfacial region with lipid glycerol backbone and ester
groups; yellow – hydrophobic regions with lipid acyl chains. A) Docked complex of
13

compound 1. B) Docked complex of compound 4. Hydrogen bonds to the C1 domain
backbone are shown as dashed yellow lines.
3. Discussion
As a starting point, compound 2 was selected as the simplest indololactone bearing
the 1-methyl-1H-indole-3-carbonyl group at the sn-1 position of the DAG-lactone template.
Initially, because of our concern with the stability of the indole ring to the reaction
conditions required for final hydroxyl group deprotection, we decided to use a starting
lactone scaffold containing at least one silyl protecting group, e.g., the TBDPS or TBDMS
ether. However, for aromatic derivative 7 this approach generated an E/Z mixture of the
deprotected compound upon removal of the silyl ether from the intermediate 16 with
TBAF. As an alternative, we employed the available paramethoxyphenyl- and benzyl-
protected lactone 25 as starting material (Scheme 2). From a chemical point of view, this
approach seemed feasible since the most reactive 3-position of the indole ring was blocked,
and its derivatives should have sufficient stability in the presence of the Lewis acids
required for benzyl group removal.20 Indeed, this proved to be the case and the single E-
isomer 7 was obtained in very good yield after final deprotection by BCl3 treatment. Thus
there is no need to use silylated lactones such as 9 and 13 as the starting material. We
therefore proceeded to use lactone 25 as the appropriate starting material for preparation of
our family of target indololactones.
A clear finding of our study was that the exchange of the heterocyclic indole from
the sn-2 position, as in the parent compound 1, to the sn-1 position decreased the affinity of
the lactones to the proteins. Comparing compounds 1 and 4, which possess the identical
indole group and alkyl chain arrayed in opposite orientations on the DAG-lactone, we
14

observe 40-fold and 229-fold weaker affinities for PKCα and RasGRP1, respectively, when
the indole group is in the sn-1 position. This result is consistent with a proposed binding
model in which the DAG-lactone binds to the C1 domain with its acyl chain oriented
toward the inner lipid core of the membrane and its R-alkylidene chain projected along the
membrane surface in the interfacial region (Figure 2).29 This binding model is also
supported by observed differences in affinity for other pairs of compounds with
functionalized phenyl groups alternatively in either the sn-1 or sn-2 positions of the DAG-
lactone template where again, more polar substituents such as phenol rings showed stronger
C1 domain binding when located in the sn-2 position.19,29
It is well known that indole-bearing tryptophan residues are preferentially located in
membrane protein structures where they can partition into the membrane-water interface.30
Furthermore, in C1 domains themselves, the conserved tryptophan residue in the binding
site (located at position 563 in RasGRP1), has been shown to be important for ligand
binding and membrane translocation in PKC θ and PKCα,32,33 and is predicted in our model
to be located in the interfacial region upon membrane insertion of the C1 domain. It has
been demonstrated computationally that isolated indole moieties also retain an energetic
preference for the interfacial region of the lipid bilayer over the hydrophobic center.34 In
the case of the 1-methyl-1H-indole substituent analyzed here, we believe that the
contrasting behavior displayed by structural isomers 1 and 4 points to differences in the
membrane interactions the two compounds can form once bound to the C1 domain in the
ternary complex.
4. Conclusions
15

Indololactones bearing an indole ring at the sn-2 position were previously shown to
possess potent affinity and selectivity for RasGRP1 versus PKCα. Here, we sought to
compare their behavior when the indole ring was shifted to the sn-1 position. We
developed a synthetic strategy for preparation of indololactones bearing the indole ring at
the sn-1 position and assessed their binding affinities for PKCα, PKCε, and RasGRP1. We
conclude that locating the heteroaromatic ring at the sn-2 position affords superior binding
affinity as well as enhanced selectivity for RasGRP1. This preference probably arises from
more favorable interactions of the indole ring in the interfacial region of cellular
membranes. In summary, at the current time, compound 1 still remains the most selective
RasGRP activator and a valuable tool for activating this class of C1 domain-containing
proteins.
Experimental section
General procedures
All chemical reagents were commercially available. [20-3H]phorbol 12,13-dibutyrate
(PDBu) was obtained from Perkin-Elmer, Waltham, MA. Melting points were determined
on an Electrothermal IA9000series digital melting point apparatus and are uncorrected.
Column chromatography was performed on a Teledyne Isco CombiFlash Companion
instrument under gradient elution conditions with RediSep disposable flash columns.
Analytical TLC was performed on Merck silica gel 254F plates. 1H and 13C NMR spectra
were recorded on a Bruker Avance DPX 400 instrument at 400 and 100 MHz, respectively.
Spectra are referenced to the solvent in which they were run (7.26 ppm for CDCl3). High
resolution, positive ion, electrospray ionization (ESI) mass spectra were obtained for all
compounds. These analyses were carried out on a Thermo-Fisher LTQ-XL Orbitrap hybrid
16

LC/MS/MS system equipped with an in-line diode-array UV detector and operated at a
resolution of 30,000 (FWHM). Sample introduction into the mass spectrometer was
accomplished by either flow-injection analysis (FIA) or by gradient LC/MS analysis; the
choice of sample introduction mode was based on compound structure and estimated
sample hydrophobicity. For FIA mode a sample solution was injected directly into the mass
spectrometer using 1:1 CH3CN/H2O containing 0.1% HCOOH at a flow rate of 200 µl/min.
Selected compounds, including indololactone target compounds 2-8, were analyzed by
LC/MS to confirm identity and assess homogeneity. Separations were carried out on a
narrow-bore (100 X 2.1 mm), small-particle (3.5-µm), Zorbax Rapid-Resolution reversed-
phase C18 column coupled with a C18 guard column (12.5 X 2.1 mm) that was eluted with
a 20-min combination of linear gradient and isocratic elution using mobile phases varying
from 2 - 90% of CH3CN/H2O containing 0.1% HCOOH at a flow rate of 250 µl/min. Both
the total-ion chromatogram (TIC) and the UV-chromatogram generated by this LC/MS
analysis were used to confirm compound purity. For LC/MS mode, the chromatographic
conditions were capable of separating E/Z isomers. For both FIA and LC/MS high
resolution MS analyses, the resulting accurate mass measurement of a molecular species
([M+H]+, [M+Na]+ or M+NH4]+) was then used to determine a unique elemental
composition for each particular compound. When appropriate, 1H and 13C NMR data were
used to set elemental constraints for this calculation.
Chemistry
General Procedure A. Aldol condensation followed by olefination.
Standard Alkylation Procedure.
17

A solution of lactone (1 equiv) in THF (5 mL/mmol) at -78 °C was treated dropwise with
LiHMDS (1.5 equiv, 1 M in THF). After the mixture was stirred at -78 °C for 1 h, a
solution of R2CHO (1.5 equiv) in THF (1 mL/mmol) was added and stirring continued for
2-3 h more at -78 °C. The reaction was quenched by slow addition of a saturated aqueous
solution of NH4Cl, and allowed to warm to room temperature. The layers were separated,
and the aqueous layer was extracted with Et2O (3×). The combined organic phases were
washed with H2O (1×) and brine (1×), dried over MgSO4, and concentrated in vacuo.
Purification by silica gel flash column chromatography (gradient 0-20% EtOAc/hexanes)
gave a mixture of diastereomers, which were used directly in the next step.
Standard Mesylation-Olefination Procedure.
A solution of the alkylation product in dichloromethane (10 mL/mmol) at 0 °C was treated
with methanesulfonyl chloride (2 eq) and triethylamine (4 eq) and then stirred at room
temperature for 2-5 h. DBU (5 equiv) was added at 0 °C, and the resulting solution was
stirred overnight at ambient temperature. The reaction mixture was treated with a saturated
solution of NH4Cl (10 mL/mmol) and extracted with CH2Cl2 (3 ×). The combined organics
were washed with H2O (2 ×) and brine (1 ×), dried (Na2SO4), and concentrated in vacuo.
Purification by silica gel flash column chromatography (gradient 0-20% EtOAc/hexanes)
gave 10, 14, 15, 16 and 27.
5-(benzyloxymethyl)-5-((tert-butyldiphenylsilyloxy)methyl)-3-(propan-2-
ylidene)dihydrofuran-2(3H)-one (10)
Starting from 9 (1.38 g, 2.90 mmol) and following general procedure A, 10 was isolated as
a colorless oil (926 mg, 62% yield): 1H NMR (400 MHz, CDCl3) δ 7.63 (d, J = 7.1 Hz, 4 H,
18

Ph), 7.44-7.26 (m, 11 H, Ph), 4.54 (s, 2 H, OCH2Ph), 3.72 (AB q, J = 10.7 Hz, 2 H,
CH2OSi), 3.55 (AB q, J = 10.1 Hz, 2 H, CH2OBn), 2.76 (m, 2 H, H-4a,b), 2.25 (s, 3 H,
13
CH3), 1.83 (s, 3 H, CH3), 1.01 (s, 9 H, (CH3)3CSi); C NMR (50 MHz, CDCl3): δ 169.6,
149.2, 137.7, 135.6, 135.6, 132.9, 132.7, 129.8, 128.3, 127.7, 127.6, 127.6, 120.3, 82.5,
73.6, 72.0, 66.5, 32.6, 26.6, 24.4, 19.8, 19.2. IR (neat) 1748 cm-1; HRMS (ESI) m/z calcd.
for C32H42O4NSi (M+NH4)+: 532.2878, found: 532.2873. Elem. Anal. Calcd. for
C32H38O4Si: C, 74.67; H, 7.44. Found: C, 74.57; H, 7.23.
(E) and (Z)-5-(benzyloxymethyl)-5-((tert-butyl(methoxy)(phenyl)silyloxy)methyl)-3-
decylidenedihydrofuran-2(3H)-one (14)
Starting from 13 (489 mg, 1.40 mmol) and following general procedure A, 14 were isolated
as a colorless oil, 7:3 mixture of E- and Z-isomers respectively (341 mg, 50% yield).
1
14E: H NMR (400 MHz, CDCl3) δ 7.36-7.26 (m, 5 H, Ph), 6.70-6.64 (m, 1 H, C=CH),
4.56 (s, 2 H, OCH2Ph), 3.69 (s, 2 H, CH2OSi), 3.55 (AB q, J = 10.3 Hz, 2 H, CH2OBn),
2.78-2.64 (AB q, J = 16.8 Hz, 2 H, H-4a,b), 2.18-2.10 (m, 2 H, CH2CH2(CH2)6CH3), 1.50-
1.40 (m, 2 H, CH2CH2(CH2)6CH3), 1.26 (m, 12 H, CH2CH2(CH2)6CH3), 0.88 (t, J = 6.8 Hz,
3 H, CH2CH2(CH2)6CH3), 0.85 (s, 9 H, (CH3)3CSi,), 0.04 (s, 6 H, Si(CH3)2); 13C NMR (50
MHz, CDCl3) δ 170.5, 140.1, 137.8, 128.4, 127.7, 127.6, 127.1, 84.4, 73.7, 72.0, 65.6, 31.8,
30.1, 29.9, 29.5, 29.4, 29.3, 29.3, 28.1, 25.7, 22.6, 18.1, 14.1, -5.6; IR (neat) 2926, 1760,
1682 cm-1; HRMS (ESI) m/z calcd. for C29H49O4Si (M+H)+: 489.3395, found 489.3383.
1
14Z: H NMR (400 MHz, CDCl3) δ 7.36-7.27 (m, 5 H, Ph), 6.16-6.10 (m, 1 H, C=CH),
4.56 (s, 2 H, OCH2Ph), 3.67 (s, 2 H, CH2OSi), 3.53 (AB q, J = 10.3 Hz, 2 H, CH2OBn),
2.72-2.84 (m, 2 H, H-4a,b), 2.72-2.64 (m, 2 H, CH2(CH2)7CH3), 1.45-1.22 (m, 14 H,
19

CH2(CH2)7CH3), 0.90-0.84 (m, 12 H, (CH3)3CSi, CH2(CH2)7CH3), 0.04 (s, 6 H, Si(CH3)2);
13C NMR (50 MHz, CDCl3) δ 169.3, 143.7, 137.8, 128.4, 127.7, 127.6, 125.0, 83.6, 73.7,
72.0, 65.5, 33.2, 31.9, 29.5, 29.5, 29.3, 29.1, 27.6, 25.7, 22.7, 18.1, 14.1, -5.6, -5.6; IR
(neat) 2925, 1756, 1671 cm-1; HRMS (ESI) m/z calcd. for C29H49O4Si (M+H)+: 489.3395,
found 489.3376.
(E) and (Z)-5-(benzyloxymethyl)-5-((tert-butyldimethylsilyloxy)methyl)-3-(3-isobutyl-
5-methylhexylidene)dihydrofuran-2(3H)-one (15)
Starting from 13 (950 mg, 2.7 mmol) and following general procedure A, 15E and 15Z
were isolated as a colorless oil, 6:4 mixture of E- and Z-isomers respectively (298 mg, 22%
yield).
1
15E: H NMR (400 MHz, CDCl3): δ 7.37-7.25 (m, 5 H, Ph), 6.75-6.67 (m, 1 H, C=CH),
4.56 (s, 2 H, OCH2Ph), 3.69 (s, 2 H, CH2OSi), 3.55 (AB q, J = 10.2 Hz, 2 H, CH2OBn),
2.71 (AB q, J = 16.9 Hz, 2 H, H-4a,b), 2.10 (t, J = 6.4 Hz, 2 H, C=CHCH2), 1.73-1.55 (m, 3
H, CH(CH2(CH(CH3)2)2), 1.09 (m, 4 H, CH(CH2(CH(CH3)2)2), 0.86 (m, 21 H, (CH3)3CSi,
CH(CH2(CH(CH3)2)2), 0.04 (s, 6 H, Si(CH3)2); 13C NMR (50 MHz, CDCl3) δ 170.3, 139.0,
137.7, 128.4, 128.0, 127.7, 127.6, 84.3, 73.7, 72.0, 65.5, 43.8, 34.6, 32.8, 30.0, 25.7, 25.2,
22.9, 22.6, 18.1, -5.6; IR (neat): 2953, 1760, 1681 cm-1; HRMS (ESI) m/z calcd. for
C30H51O4Si (M+H)+: 503.3551, found 503.3541.
1
15Z: H NMR (400 MHz, CDCl3): δ 7.36-7.26 (m, 5 H, Ph), 6.14 (t, J = 7.3 Hz, 1 H,
C=CH), 4.56 (s, 2 H, OCH2Ph), 3.70-3.63 (m, 2 H, CH2OSi), 3.56-3.50 (m, 2 H, CH2OBn),
2.80 (AB q, J = 16.2 Hz, 2 H, H-4a,b), 2.68-2.62 (m, 2 H, C=CHCH2), 1.69-1.56 (m, 3 H,
CH(CH2(CH(CH3)2)2), 1.08 (m, 4 H, CH(CH2(CH(CH3)2)2), 0.86 (m, 21 H, (CH3)3CSi,
20

CH(CH2(CH(CH3)2)2), 0.04 (s, 6 H, Si(CH3)2); 13C NMR (50 MHz, CDCl3) δ 169.2, 142.6,
137.8, 128.3, 127.6, 127.5, 125.8, 83.5, 73.6, 71.8, 65.3, 43.90, 43.9, 33.2, 33.2, 31.9, 25.7,
25.1, 25.1, 23.0, 22.7, 18.1, -5.6, -5.6; IR (neat): 2953, 1756, 1671 cm-1; HRMS (ESI) m/z
calcd. for C30H51O4Si (M+H)+: 503.3551, found 503.3537.
(E)-5-(benzyloxymethyl)-5-((tert-butyldimethylsilyloxy)methyl)-3-(4-(p-
tolylethynyl)benzylidene)dihydrofuran-2(3H)-one (16)
Starting from 13 (501 mg, 1.43 mmol) and following general procedure A, 16 was obtained
as a white solid (196 mg, 25% yield): 1H NMR (400 MHz, CDCl3) δ 7.53 (d, J = 8.1 Hz, 2
H, Ph), 7.46-7.39 (m, 5 H, Ph, C=CH), 7.32-7.23 (m, 5 H, Ph), 7.14 (d, J = 7.8 Hz, 2 H,
Ph), 4.55 (s, 2 H, OCH2Ph), 3.75-3.69 (m, 2 H, CH2OSi), 3.61-3.56 (m, 1 H, CH2OBn),
3.07 (AB q, J = 17.7 Hz, 2 H, H-4a,b), 2.35 (s, 3 H, CH3Ar), 0.79 (s, 9 H, (CH3)3Si), 0.00 (s,
6 H, CH3Si); 13C NMR (50 MHz, CDCl3) δ 171.5, 138.8, 137.6, 134.8, 134.4, 131.8, 131.6,
129.8, 129.2, 128.4, 127.8, 127.7, 126.4, 124.7, 119.8, 91.9, 88.4, 84.7, 73.7, 71.8, 65.7,
32.5, 25.6, 21.5, 18.1, -5.6; IR (neat) 3031, 2927, 2855, 1742, 1652 cm-1; HRMS (ESI) m/z
calcd. for C35H41O4Si (M+H)+: 553.2769, found 553.2753; Mp: 128-131 °C.
(E)-5-(benzyloxymethyl)-5-((4-methoxyphenoxy)methyl)-3-(2-propylpentylidene)
dihydrofuran-2(3H)-one (26)
A solution of lactone 25 (398 mg, 1.16 mmol) was treated following general procedure A,
using potassium tert-butoxide (3 equiv) in place of DBU and stirring solution at room
temperature for 20 min in the final step, to give 26 as a colorless oil, one single
1
diastereomer (163 mg, 31% yield): H NMR (400 MHz, CDCl3) δ 7.33-7.26 (m, 5 H, Ph),
21

6.81 (s, 4 H, Ph), 6.51 (d, J = 10.8 Hz, 1 H, C=CH), 4.58 (s, 2 H, OCH2Ph), 4.03 (AB q, J =
9.7 Hz, 2 H, CH2OPhOCH3), 3.76 (s, 3 H, OCH3), 3.67 (AB q, J = 10.0 Hz, 2 H, CH2OBn),
2.85 (AB q, J = 16.2 Hz, 2 H, H-4a,b), 2.28-2.17 (m, 1 H, CH((CH2)2CH3)2), 1.48-1.15 (m,
8 H, CH((CH2)2CH3)2), 0.88-0.82 (m, 6 H, CH((CH2)2CH3)2); 13C NMR (50 MHz, CDCl3)
δ 170.2, 154.3, 152.5, 145.9, 137.6, 128.4, 127.8, 127.6, 126.1, 115.6, 114.6, 82.8, 73.7,
72.1, 70.7, 55.7, 40.8, 37.1, 37.1, 30.8, 20.6, 20.6, 14.2, 14.2; IR (neat) 2926, 1760, 1682
cm-1; HRMS (ESI) m/z calcd. for C28H37O5 (M+H)+: 453.2636, found 453.2628.
(E)-5-(benzyloxymethyl)-5-((4-methoxyphenoxy)methyl)-3-(4-(p-
tolylethynyl)benzylidene)dihydrofuran-2(3H)-one (27)
Starting from 25 (440 mg, 1.28 mmol) and following general procedure A, 27 was obtained
as a white solid, one single diastereomer (328 mg, 47% yield): 1H NMR (400 MHz, CDCl3)
δ 7.57 (d, J = 8.4 Hz, 2 H, Ph), 7.55 (s, 1 H, C=CH), 7.48-7.42 (m, 4 H, Ph), 7.32-7.26 (m,
5 H, Ph), 7.16 (d, J = 7.9 Hz, 2 H, Ph), 6.80 (s, 4 H, Ph), 4.59 (s, 2 H, OCH2Ph), 4.08 (AB
q, J = 9.7 Hz, 2 H, CH2OPhOCH3), 3.74 (s, 3 H, CH3OAr), 3.71 (AB q, J = 10 Hz, 2 H,
CH2OBn), 3.23 (AB q, J = 17.8 Hz, 2 H, H-4a,b), 2.37 (s, 3 H, CH3Ar); 13C NMR (50 MHz,
CDCl3) δ 171.1, 154.4, 152.4, 138.9, 137.4, 135.8, 134.1, 131.9, 131.6, 129.9, 129.2, 128.4,
127.8, 127.6, 125.2, 125.0, 119.7, 115.7, 114.6, 92.1, 88.3, 83.1, 73.7, 71.9, 70.7, 55.7,
32.9, 21.5; IR (neat) 3031, 2918, 2866, 1747, 1659 cm-1; HRMS (ESI) m/z calcd. for
C36H33O5 (M+H)+: 545.2323, found 545.2307; Mp: 88°C.
General Procedure B. TBDMS deprotection.
22

To a stirred solution of 15E, 15Z or 16 (1 equiv) in anhydrous THF (30 mL/mmol)
tetrabutylammonium fluoride (TBAF, 1.0 M solution in THF, 1.5 equiv) was added. The
reaction mixture was stirred at room temperature for 15 min. The crude solution was
concentrated in vacuo and purified by silica gel flash column chromatography (gradient 0-
40% EtOAc/hexanes) to obtain 18, 19, or 20.
(E)-5-(benzyloxymethyl)-5-(hydroxymethyl)-3-(3-isobutyl-5-
methylhexylidene)dihydrofuran-2(3H)-one (18)
Starting from 15E (130 mg, 0.26 mmol) and following general procedure B, 18 was
1
obtained as a colorless oil (79 mg, 79% yield): H NMR (400 MHz, CDCl3) δ 7.37- 7.26
(m, 5 H, Ph), 6.79-6.73 (m, 1 H, C=CH), 4.56 (s, 2 H, OCH2Ph), 3.76 (dd, J = 11.7, 5.6 Hz,
1H, CHaHOH), 3.67 (dd, J = 11.6, 4.9 Hz, 1H, CHHbOH), 3.57 (AB q, J = 10.0 Hz, 2 H,
CH2OBn), 2.73 (AB q, 2 H, H-4a,b), 2.11 (m, 3 H, C=CHCH2, OH), 1.74-1.57 (m, 3 H,
CH(CH2(CH(CH3)2)2), 1.09 (m, 4 H, CH(CH2(CH(CH3)2)2), 0.86 (m, 12 H,
CH(CH2(CH(CH3)2)2); 13C NMR (50 MHz, CDCl3) δ 170.1, 140.4, 137.5, 128.5, 127.9,
127.6, 127.2, 84.0, 73.7, 71.9, 65.6, 43.8, 34.7, 32.8, 30.1, 25.2, 23.0, 22.9, 22.6; IR (neat)
3432, 2953, 2912, 1755, 1677 cm-1; HRMS (ESI) m/z calcd. for C24H37O4 (M+H)+:
389.2686, found 389.2678.
(Z)-5-(benzyloxymethyl)-5-(hydroxymethyl)-3-(3-isobutyl-5-
methylhexylidene)dihydrofuran-2(3H)-one (19)
Starting from 15Z (120 mg, 0.24 mmol) and following general procedure B, 19 was
obtained as a colorless oil (68 mg, 73% yield): 1H NMR (400 MHz, CDCl3) δ 7.38-7.26 (m,
23

5 H, Ph), 6.20 (t, J = 7.4 Hz, 1 H, C=CH), 4.56 (s, 2 H, OCH2Ph), 3.70 (AB q, J = 11.6 Hz,
2 H, CH2OH), 3.55 (AB q, J = 10.0 Hz, 2 H, CH2OBn), 2.82 (s, 2 H, H-4a,b), 2.66 (m, 2 H,
C=CHCH2), 2.20 (s, 1 H, OH), 1.69-1.56 (m, 3 H, CH(CH2(CH(CH3)2)2), 1.08 (m, 4 H,
13
CH(CH2(CH(CH3)2)2), 0.85 (m, 12 H, CH(CH2(CH(CH3)2)2); C NMR (50 MHz, CDCl3)
δ 169.1, 144.0, 137.5, 128.4, 127.8, 127.6, 125.1, 83.3, 73.7, 71.8, 65.4, 43.9, 33.4, 33.2,
32.0, 25.1, 23.0, 22.7, 22.7; IR (neat) 3432, 2953, 2908, 1752, 1668 cm-1; HRMS (ESI) m/z
calcd. for C24H37O4 (M+H)+: 389.2686, found 389.2673.
5-(benzyloxymethyl)-5-(hydroxymethyl)-3-(4-(p-tolylethynyl)benzylidene)
dihydrofuran-2(3H)-one (20)
Starting from 16 (170 mg, 0.30 mmol) and following general procedure B, 20 was obtained
1
as a colorless oil, inseparable 1:1 mixture of E- and Z-isomers (84 mg, 64% yield): H
NMR (400 MHz, CDCl3) δ 7.58-7.12 (m, 27 H, Ar isomer E, Ar isomer Z, C=CH isomer
E), 6.92 (s, 1 H, C=CH isomer Z), 4.55 (s, 2 H, OCH2Ph isomer E), 4.51 (AB q, 2 H, J =
11.3 Hz, OCH2Ph isomer Z), 3.54-3.84 (m, 10 H, CH2OBn, CH2OH, H-4a,b isomer Z), 3.13
(s, 2 H, H-4a,b isomer E), 2.36 (s, 6 H, CH3Ar); 13C NMR (50 MHz, CDCl3) δ 172.4, 171.5,
148.9, 138.9, 138.4, 137.3, 137.2, 137.1, 136.0, 135.4, 134.0, 131.9, 131.6, 131.4, 131.1,
130.7, 130.0, 129.5, 129.2, 129.1, 128.9, 128.5, 128.0, 127.9, 127.7, 125.3, 125.0, 122.0,
120.1, 119.7, 92.1, 89.6, 88.4, 88.4, 84.5, 73.8, 73.7, 71.8, 70.4, 65.3, 63.9, 32.4, 31.7, 22.6,
21.5; IR (neat) 3365, 2920, 2865, 1729, 1652 cm-1; HRMS (ESI) m/z calcd. for C29H27O4
(M+H)+: 439.1904, found 439.1891.
General Procedure C. PMP deprotection.
24

Ceric ammonium nitrate (3 equiv) was added to a solution of 26, 27 (1 equiv) in 80%
CH3CN/H2O (v/v, 10 mL/mmol) at 0 °C. The reaction mixture was stirred for 15 min. A
solution of 5% Na2S2O3 was added at room temperature and the aqueous phase was
extracted with CH2Cl2 (3 ×). The organic phase was dried (Na2SO4) and concentrated in
vacuo. Purification by silica gel flash column chromatography (gradient 0-30%
EtOAc/hexanes) gave 28, 29.
(E)-5-(benzyloxymethyl)-5-(hydroxymethyl)-3-(2-propylpentylidene)dihydrofuran-
2(3H)-one (28)
Starting from 26 (149 mg, 0.33 mmol) and following general procedure C, 28 was obtained
1
as a colorless oil (101 mg, 89% yield): H NMR (400 MHz, CDCl3) δ 7.37-7.26 (m, 5 H,
Ph), 6.49 (dt, J = 10.7, 2.7 Hz, 1 H, C=CH), 4.55 (d, J = 2.3 Hz, 2 H, OCH2Ph), 3.76 (dd, J
= 12.1, 7.1 Hz, 1 H, CHaHOH), 3.68 (dd, J = 12.1, 6.6 Hz, 1 H, CHHbOH), 3.57 (AB q, J =
10.0 Hz, 2 H, CH2OBn), 2.79 (dd, J = 16.9, 2.6 Hz, 1 H, H-4a), 2.71 (dd, J = 16.8, 2.9 Hz,
1H, H-4b), 2.26-2.16 (m, 1 H, CH((CH2)2CH3)2), 2.09 (t, J = 6.8 Hz, 1H, CH2OH), 1.48-
1.13 (m, 8 H, CH((CH2)2CH3)2), 0.87 (t, J = 7.2 Hz, 3 H, CH((CH2)2CH3)2), 0.84 (t, J = 7.2
Hz, 3 H, CH((CH2)2CH3)2); 13C NMR (50 MHz, CDCl3) δ 171.0, 147.0, 138.2, 129.1,
128.6, 128.3, 126.8, 84.7, 74.5, 72.7, 66.3, 41.5, 37.8, 30.9, 21.3, 21.2, 14.8; IR (neat)
3437 , 2955 , 1755 , 1737 , 1678 cm-1; HRMS (ESI) m/z calcd. for C21H31O4 (M+H)+:
347.2217, found 347.2208.
(E)-5-(benzyloxymethyl)-5-(hydroxymethyl)-3-(4-(p-tolylethynyl)benzylidene)
dihydrofuran-2(3H)-one (29)
25

Starting from 27 (308 mg, 0.56 mmol) and following general procedure C, 29 was obtained
1
as a yellow solid (181 mg, 74% yield): H NMR (400 MHz, CDCl3) δ 7.58-7.50 (m, 3 H,
Ar, C=CH isomer E), 7.47-7.40 (m, 3 H, Ar), 7.35-7.25 (m, 6 H, Ar ), 7.19-7.12 (m, 2 H,
Ar), 4.55 (s, 2 H, OCH2Ph), 3.80 (m, 2 H, CH2OH), 3.63 (AB q, J = 10.0 Hz, 2 H,
CH2OBn), 3.12 (s, 2 H, H-4a,b), 2.36 (s, 3 H, CH3Ar), 1,92 (s, 1H, OH).
General Procedure D. BnO Deprotection.
A solution of 10, 21, 22, 23, 30, 31 (1 equiv) in anhydrous dichloromethane (20 mL/mmol)
at -78°C was treated dropwise with boron trichloride (3 equiv). The reaction was monitored
by TLC and quenched upon completion (0.5-2 h) by the slow addition of NaHCO3 and the
aqueous phase was extracted with dichloromethane (2 ×). The organic phase was dried
(Na2SO4) and concentrated in vacuo. Purification by silica gel flash column
chromatography (gradient 0-40% EtOAc/hexanes) gave 11, 3, 4, 5, 6, 7 and 8.
General Procedure E. Acylation.
A solution of 11, 17, 18, 19, 20, 28 or 29 (1 equiv) in dichloromethane (12 ml/mmol) was
treated with Et3N (3 equiv), 1-methyl-1H-indole-3-carbonyl chloride (1.3 equiv) and a
catalytic amount of DMAP (0.1 equiv). The reaction was stirred at room temperature and
monitored by TLC, and upon completion (20-27 h) it was concentrated in vacuo.
Purification by silica gel flash column chromatography (gradient 0-30% EtOAc/hexanes)
gave 12, 21, 22, 23, 30, 31.
26

5-((tert-butyldiphenylsilyloxy)methyl)-5-(hydroxymethyl)-3-(propan-2-
ylidene)dihydrofuran-2(3H)-one (11)
Starting from 10 (229 mg, 0.44 mmol) and following general procedure D, 11 was obtained
as an amber oil (154 mg, 82%yield): 1H NMR (400 MHz, CDCl3) δ 7.66-7.61 (m, 4 H, Ph),
7.46-7.36 (m, 6 H, Ph), 3.76-3.60 (m, 4 H, CH2OSi, CH2OH), 2.76 (AB q, J = 16.2 Hz, 2
H, H-4a,b), 2.24 (s, 3 H, CH3), 1.94 (t, J = 6.8 Hz, 1 H, OH), 1.86 (s, 3 H, CH3), 1.03 (s, 9
H, (CH3)3CSi); 13C NMR (50 MHz, CDCl3) δ 169.5, 150.1, 135.6, 135.6, 132.8, 132.6,
129.9, 127.8, 120.1, 83.2, 66.4, 65.4, 31.9, 26.6, 24.5, 19.8, 19.2; IR (neat): 3417, 1746 cm-
1; HRMS (ESI) m/z calcd. for C25H33O4Si (M+H)+: 425.2143, found 425.2124.
(2-((tert-butyldiphenylsilyloxy)methyl)-5-oxo-4-(propan-2-ylidene)tetrahydrofuran-2-
yl)methyl 1-methyl-1H-indole-3-carboxylate (12)
Starting from 11 (434 mg, 1.02 mmol) and following general procedure E, 12 was obtained
as a colorless oil (508 mg, 85% yield): 1H NMR (400 MHz, CDCl3) δ 8.06 (d, J = 7.7 Hz, 1
H, H-4’), 7.70-7.63 (m, 5 H, Ph, H-2’), 7.44-7.21 (m, 9 H, Ph, H-5’, H-6’, H-7’), 4.47 (AB
q, J = 11.7 Hz, 2 H, CH2CO2), 3.86-3.81 (m, 2 H, CH2OSi), 3.80 (m, 3 H, CH3N), 2.84 (AB
q, J = 16.3 Hz, 2 H, H-3a,b), 2.23 (s, 3 H, CH3), 1.81 (s, 3 H, CH3), 1.04 (s, 9 H,
(CH3)3CSi); 13C NMR (100 MHz, CDCl3): δ 169.3, 164.2, 149.8, 137.2, 135.6, 135.6,
132.7, 132.6, 129.8, 129.8, 127.8, 127.8, 126.5, 122.8, 122.1, 121.5, 119.9, 109.7, 106.1,
81.6, 66.5, 65.0, 33.4, 32.7, 26.6, 24.4, 19.8, 19.2; IR (neat): 1746, 1699 cm-1; HRMS (ESI)
m/z calcd. for C35H40NO5Si (M+H)+: 582.2670, found 582.2662.
27

(E)-(2-(benzyloxymethyl)-4-decylidene-5-oxotetrahydrofuran-2-yl)methyl-1-methyl-
1H-indole-3-carboxylate (21)
Starting from 14E (191 mg, 0.37 mmol) and following general procedures B and E, 21 was
1
obtained as a white solid (138 mg, 70% yield): H NMR (400 MHz, CDCl3) δ 8.02 (d, J =
7.2 Hz, 1 H, H-4’), 7.69 (s, 1 H, H2’), 7.37-7.24 (m, 8 H, Ph, H-5’, H-6’, H-7’), 6.77-6.70
(m, 1 H, C=CH), 4.60 (s, 2 H, OCH2Ph), 4.57 (d, J = 11.9 Hz, 1 H, CO2CHaH), 4.44 (d, J =
11.8 Hz, 1 H, CO2CHHb), 3.83 (s, 3 H, CH3N), 3.68 (AB q, J = 10.1 Hz, 2 H, BnOCH2),
2.84 (AB q, J = 17.4 Hz, 2 H, H-3a,b), 2.15-2.06 (m, 2 H, CH2(CH2)7CH3), 1.39-1.14 (m, 14
13
H, CH2(CH2)7CH3), 0.87 (t, J = 6.9 Hz, 3 H, CH3); C NMR (50 MHz, CDCl3): δ 170.2,
164.1, 141.2, 137.4, 137.2, 135.6, 128.4, 127.8, 127.7, 126.5, 126.3, 122.9, 122.1, 121.5,
109.8, 105.9, 82.8, 73.7, 72.0, 65.2, 33.5, 31.8, 30.6, 30.2, 29.4, 29.3, 29.3, 28.0, 22.6, 14.1;
IR (neat): 2926, 2854, 1750, 1688, 1538 cm-1; HRMS (ESI) m/z calcd. for C33H42NO5
(M+H)+: 532.3057, found 532.3045; Mp: 92°C.
(E)-(2-(benzyloxymethyl)-4-(3-isobutyl-5-methylhexylidene)-5-oxotetrahydrofuran-2-
yl)methyl 1-methyl-1H-indole-3-carboxylate (22)
Starting from 18 (74 mg, 0.18 mmol) and following general procedure E, 22 was obtained
1
as a colorless oil (92 mg, 93% yield): H NMR (400 MHz, CDCl3) δ 8.08 (d, 1 H, J = 7.3
Hz, H-4’), 7.69 (s, 1 H, H2’), 7.36-7.24 (m, 8 H, Ph, H-5’, H-6’, H-7’), 6.77 (m, 1 H,
C=CH), 4.59 (s, 2 H, OCH2Ph), 4.56 (d, 1 H, J = 11.8 Hz, CO2CHaH), 4.44 (d, 1 H, J =
11.8 Hz, CO2CHHb), 3.82 (s, 3 H, CH3N), 3.67 (AB q, J = 10.1 Hz, 2 H, BnOCH2), 2.83
(AB q, J = 17.1 Hz, 2 H, H-3a,b), 2.14-2.01 (m, 2 H, C=CHCH2), 1.68-1.50 (m, 3 H,
CH(CH2(CH(CH3)2)2), 1.07-0.98 (m, 4 H, CH(CH2(CH(CH3)2)2), 0.83-0.75 (m, 12 H,
28

CH(CH2(CH(CH3)2)2); 13C NMR (50 MHz, CDCl3) δ 170.0, 164.1, 140.1, 137.4, 137.2,
135.6, 128.4, 127.8, 127.6, 127.1, 126.6, 122.9, 122.1, 121.5, 109.8, 105.9, 82.7, 73.7, 71.9,
65.1, 43.7, 34.6, 33.5, 32.7, 30.8, 25.1, 22.9, 22.8, 22.6, 22.5. IR (neat): 2953, 2912, 2867,
1757, 1701, 1534, 748 cm-1; HRMS (ESI) m/z calcd. for C34H44NO5 (M+H)+: 546.3214,
found 546.3199.
(Z)-(2-(benzyloxymethyl)-4-(3-isobutyl-5-methylhexylidene)-5-oxotetrahydrofuran-2-
yl)methyl 1-methyl-1H-indole-3-carboxylate (23)
Starting from 19 (64 mg, 0.16 mmol) and following general procedure E, 23 was obtained
as a colorless oil (74 mg, 85% yield): 1H NMR (400 MHz, CDCl3) δ 8.09 (d, J = 6.8 Hz, 1
H, H-4’), 7.71 (s, 1 H, H-2’), 7.36-7.23 (m, 8 H, Ph, H-5’, H-6’, H-7’), 6.17 (m, 1 H,
C=CH), 4.59 (s, 2 H, OCH2Ph), 4.55 (d, J = 11.8 Hz, 1 H, CO2CHaH), 4.43 (d, J = 11.8 Hz,
1 H, CO2CHHb), 3.82 (s, 3 H, CH3N), 3.66 (AB q, J = 10.1 Hz, 2 H, BnOCH2), 2.97 (dd, J
= 16.4, 1.9 Hz, 1 H, H-3a), 2.86 (dd, J = 16.4, 1.9 Hz, 1 H, H-3b), 2.81-2.73 (m, 1 H,
C=CHCHaH), 2.52-2.43 (m, 1 H, C=CHCHHb), 1.65-1.53 (m, 3 H, CH(CH2(CH(CH3)2)2),
13
1.07-1.02 (m, 4 H, CH(CH2(CH(CH3)2)2), 0.84-0.80 (m, 12 H, CH(CH2(CH(CH3)2)2); C
NMR (50 MHz, CDCl3 δ 169.0, 164.2, 143.7, 137.5, 137.2, 135.6, 128.4, 127.8, 127.6,
126.5, 124.9, 122.9, 122.2, 121.6, 109.8, 106.1, 82.0, 73.7, 71.8, 64.9, 43.9, 43.9, 34.1,
33.5, 33.2, 32.0, 25.1, 23.0, 23.0, 22.7, 22.6; IR (neat): 2952, 2909, 2866, 1755, 1702, 1534,
748 cm-1; HRMS (ESI) m/z calcd. for C34H44NO5 (M+H)+: 546.3214, found 546.3197.
(E)-(2-(benzyloxymethyl)-5-oxo-4-(2-propylpentylidene)tetrahydrofuran-2-yl)methyl
1-methyl-1H-indole-3-carboxylate (30)
29