Facile and Divergent Synthesis of Lamellarins and Lactam-Containing Derivatives with Improved Drug Likeness and Biological Activities

DOI: 10.1002/asia.201500611

Full Paper

Natural Products

Facile and Divergent Synthesis of Lamellarins and Lactam-

Containing Derivatives with Improved Drug Likeness and

Biological Activities

Atiruj Theppawong,^{[a, b]}Poonsakdi Ploypradith,^{[a, b]}Pitak Chuawong,^[c]

Somsak Ruchirawat,^{[a, b]}and Montakarn Chittchang*^{[a, b]}

Abstract: With the goal to improve the aqueous solubility of

lamellarins, the lactone ring in their skeleton was replaced

with a lactam moiety in azalamellarins. However, the report-

ed synthetic route produced such derivatives in very low

yields. Hence, this study focused on developing an efficient

simplified total synthetic scheme that could furnish both

azalamellarins and the parent lamellarins from the same pyr-

role ester intermediates. Subsequent comparative profiling

revealed that the introduced lactone-to-lactam replacement

rendered these molecules less lipophilic, whereas their

cancer cytotoxicity remained equipotent to that of the

parent compounds. Interestingly, their inhibitory activity was

significantly enhanced towards the multifaceted GSK-3b

enzyme. Our results clearly demonstrate the therapeutic po-

tential of this promising class of marine-derived natural

products and justify their further development, especially

into anticancer agents.

Introduction

compounds also exhibit multidrug resistance reversal activity,

HIV integrase inhibition, and antioxidant properties.^[2,3a]

Lamellarins are polyaromatic alkaloids originally isolated from

marine invertebrates, including mollusks, ascidians, and spong-

es.^[1]Since 1985, approximately 50 lamellarins and derivatives

have been discovered,^[2]most of which contain the same pen-

tacyclic skeleton and only differ in the number and position of

the oxygen-containing substituents around the core structure,

as shown in Figure 1a. Interestingly, potent cytotoxic effects of

these compounds have been demonstrated against a wide va-

riety of tumor cells from different origins.^[2,3]In addition, these

Among all the members, lamellarin D (2) appears to possess

the most promising therapeutic potential as an anticancer

agent with multiple biological targets. This compound was

originally reported to induce potent and dose-dependent stim-

ulation of DNA cleavage by trapping the complex between

topoisomerase I (Topo I) and DNA.^[4]Lamellarin D at micromolar

levels has also been demonstrated to induce caspase-depen-

dent apoptosis through mitochondria-mediated signaling

pathways.^[5]Additionally, the inhibitory activities of lamellarins

and their derivatives towards several protein kinases, including

glycogen synthase kinase-3a/b (GSK-3a/b), have been docu-

mented.^[6]Furthermore, induction of cellular senescence

through both the inhibition of Topo I and the generation of re-

active oxygen species has been proposed as another mecha-

nism for the cytotoxicity of lamellarin D.^[7]

[a] A. Theppawong, Dr. P. Ploypradith, Prof. Dr. S. Ruchirawat,

Dr. M. Chittchang

Laboratory of Medicinal Chemistry

Chulabhorn Research Institute

Laksi, Bangkok 10210 (Thailand)

E-mail: montakarn@cri.or.th

Recently, the structural isomer of lamellarin D, namely, lamel-

larin N (3), has also attracted attention owing to its interesting

cytotoxicity against certain cancer cell lines.^[3b,6]More impor-

tantly, superior potency against GSK-3a/b was reported for la-

mellarin N with a half-maximal inhibitory concentration (IC₅₀) of

5 nm, compared with 300 nm for lamellarin D.^[6]Overall, we se-

lected lamellarin D and lamellarin N as lead compounds for fur-

ther development into anticancer agents. Although not the

most lipophilic compounds in the series, these two lamellarins

still suffer from poor aqueous solubility and, thus, usually pre-

cipitate in biological media. Therefore, we aimed to develop la-

mellarin derivatives that possess improved aqueous solubility

while maintaining their biological activities.

[b] A. Theppawong, Dr. P. Ploypradith, Prof. Dr. S. Ruchirawat,

Dr. M. Chittchang

Chemical Biology Program

Chulabhorn Graduate Institute and Center of Excellence on Environmental

Health and Toxicology

Laksi, Bangkok 10210 (Thailand)

[c] Prof. Dr. P. Chuawong

Department of Chemistry and Center of Excellence for Innovation in

Chemistry

Faculty of Science and Special Research Unit for Advanced Magnetic

Resonance (AMR)

Kasetsart University

Bangkok 10900 (Thailand)

Supporting Information for this article is available on the WWW under

htt p ://dx.doi.org/10.1002/asia.201500611.

Chem. Asian J. 2015, 10, 2631 – 2650

2631

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

lamellarins and azalamellarins, as shown in Scheme 1.

The synthetic route could then branch out with

a few more steps to give the desired products. Start-

ing from the pyrrole esters, sequential reactions com-

prising lactonization, C5ÀC6 oxidation, and global de-

protection could furnish lamellarins, whereas azala-

mellarin synthesis would require an additional step

to install a nitrogen atom prior to lactam ring closure,

followed by oxidation and global deprotection.

Following our established methodology,^[9]benzyl-

dihydroisoquinoline derivatives were successfully syn-

thesized from arylacetic acid 8 or 9 and aryl ethyla-

mine 10, both of which were generated by using ap-

propriately substituted benzaldehyde derivatives as

the building blocks (see the Experimental Section for

Figure 1. a) General structure of lamellarins and b) structures of the compounds used in

this study.

As a means to simultaneously enhance the stability and sol-

ubility of lamellarins, a simple lactone-to-lactam replacement

was performed by Thasana and co-workers to generate some

aza derivatives of these compounds.^[8]The synthesis of azala-

mellarins was accomplished in eight steps (starting from pyr-

role esters). Owing to some low-yielding steps and the long se-

quence of their approach, the overall yield of azalamellarin D

(4) was only 11 %. For the other azalamellarins, only the final

products with either an N-allyl or N-propyl group were report-

ed.

Herein, we report a new synthetic route that involves pyrrole

esters as common intermediates for the preparation of both la-

mellarins and azalamellarins. The effects of structure modifica-

tion on lipophilicity, aqueous solubility, cytotoxicity, as well as

GSK-3b inhibitory activity were evaluated. Molecular modeling

was also used to probe the interactions of lamellarins and aza-

lamellarins with the GSK-3b target.

Scheme 1. Retrosynthesis for lamellarins and azalamellarins. For com-

pounds 2 and 4, R³=Me; R⁴=H. For compounds 3 and 5, R³=H; R⁴=Me.

Results and Discussion

Chemistry

details). For 2-bromo-a-nitrocinnamate (11), on the other hand,

the conditions previously used to prepare 2-bromo-O-benzyl-

vanillin (13) gave relatively low yield (25–45%) and, thus, were

optimized in this study.

The synthetic approaches for lamellarins and azalamellarins

previously reported by our laboratory^[9]and Thasana’s group,^[8]

respectively, share a common feature of pyrrole formation

through efficient Michael addition/ring closure (Mi-RC) Grob-

type condensation between benzyldihydroisoquinoline and a-

nitrocinnamate derivatives. Nevertheless, the two routes are

synthetically distinguishable from the very beginning by the

different a-nitrocinnamate derivatives required for subsequent

reactions. Whereas the 2-hydroxy group is involved in NaH-

mediated lactonization to furnish lamellarins, the presence of

a bromine atom at this particular position (see com-

pound 11 in Scheme 1) is necessary for intramolecu-

lar CÀN bond formation to generate the lactam

moiety in azalamellarins.

After some experimentation, it was found that bromination

of 12 by using either soluble or polymer-bound pyridinium tri-

bromide (PyBr₃) in MeOH gave the desired product in 65–85%

yield (Scheme 2). In contrast, the use of Br₂or N-bromosuccini-

mide (NBS) under other conditions gave lower yield. Subse-

quently, 13 could be converted into 11 in 63% yield (based on

recovered starting aldehyde 13).

To develop an efficient and divergent method, the

retrosynthetic analysis suggested the possibility of

using three different benzaldehyde derivatives as

simple building blocks to generate the pyrrole esters

(i.e., 6 or 7) as key common intermediates for both Scheme 2. Synthesis of a-nitrocinnamate 11.

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org 2632

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Synthetic development of azalamellarins

Tandem lactamization

Upon obtaining sufficient amounts of the required starting ma-

terials (see the Experimental Section for details), pyrrole ester 6

was synthesized by using our reported method.^[9]Next, the

subsequent steps en route to azalamellarin D were carefully in-

vestigated, starting with tandem lactamization both to incor-

porate a nitrogen atom into the pyrrole ester and then to form

a lactam moiety in situ. After several attempts, however, it ap-

peared that the two-step reaction of pyrrole amide formation

and lactam ring closure would be required for the azalamella-

rin synthesis.

Pd^II- and Cu^I-mediated intramolecular CÀN bond formation to

effect lactam ring closure in some biaryl systems is well estab-

lished.^[10]Therefore, a number of literature-based conditions

were applied in this study to incorporate a lactam moiety into

the lamellarin core structure, preferably in one step by tandem

lactamization. In the presence of various nitrogen nucleophiles

under different sets of reaction conditions (Scheme 4), only the

starting material was recovered in most cases. Surprisingly, if

sodium amide was used, pyrrole carboxylic acid 24 (Figure 2)

was obtained.

As summarized in Scheme 3, we then optimized the amida-

tion step by using different nitrogen sources, with or without

microwave (MW) irradiation to assist the conversion of pyrrole

ester 6 into the corresponding pyrrole amide. Desired prod-

uct 17 or 18 was successfully obtained with either a benzyl or

p-methoxybenzyl (PMB) protecting group, respectively. After-

wards, copper(I)-mediated lactam ring closure was performed

under the optimized conditions, followed by the oxidation by

using 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) to

yield N,O-protected azalamellarin D with an unsaturated

D ring. In the final step, acidolytic reaction was anticipated to

simultaneously remove all O- and N-protecting groups in the

presence of thioanisole as a benzyl cation scavenger to pro-

duce azalamellarin D. By avoiding repetitive protection and de-

protection steps, as required for 20,^[8]our current synthetic

scheme is expectedly shorter than that previously reported.

Details of our investigations focusing on the lactamization

and global deprotection steps are described below. Subse-

quently, the optimized conditions were used to synthesize the

azalamellarin N (5) structural isomer (Figure 1b). As proof of

concept, our newly developed route was also applied to syn-

thesize both lamellarin D and lamellarin N more efficiently.

Direct amidation

Given that our attempts at tandem lactamization were unsuc-

cessful, a two-step approach was employed by first converting

pyrrole ester 6 into corresponding pyrrole amide 19 without

an N-protecting group, if possible, to avoid the complications

generally experienced during its removal. Thus, model studies

were initially undertaken for direct amidation of methyl ben-

zoate (see the Supporting Information for details). As shown in

Scheme 5, the unique combination of either sodium amide or

trimethylsilyl (TMS) azide as the nitrogen nucleophile and tri-

methylaluminum as the Lewis acid^[11]unexpectedly gave prod-

uct 26 in good yield (80–89%). In contrast, the reactions per-

formed with the use of other nitrogen sources (NaN₃, NH₄Cl, or

NH₂OH·HCl) under different conditions (tetrahydrofuran (THF),

1,4-dioxane, or toluene as solvent at 60–2008C with 100–

300 W of MW power for 30–45 min) either did not proceed or

gave lower yields of the corresponding products.

These amidation conditions were next applied to our com-

pounds. Pyrrole ester 6 was thus treated with TMS azide,

sodium amide, or saturated ammonia under varying MW-assist-

ed conditions. Unfortunately, all conditions failed to give de-

sired product 19, and either the starting material or a mixture

Scheme 3. Synthetic route of azalamellarin D. [a] See Thasana et al.^[8]for details.

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2633

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Table 1. Pyrrole amide formation with N-protecting groups and MW irradia-

tion.^[a]

Scheme 4. Attempted tandem lactamization with or without N-protecting

groups. Conditions: AcNH₂, BocNH₂, BnNH₂, CbzNH₂, PMBNH₂, or NaNH₂as

nitrogen source; PdCl₂, Pd(OAc)₂, or CuI as catalyst; Cs₂CO₃or K₃PO₄as base;

Xantphos or 1,10-phenanthroline as ligand; 1,4-dioxane or toluene as sol-

vent; performed at 90–1008C for 8–24 h.

Entry Nitrogen Solvent

source

MW^[b]

[W]

T

[8C]

t

[h]

Product,

yield [%]

1

2

3

4

5

BnNH₂

PMBNH₂THF

PMBNH₂anh. 1,4-dioxane 150

PMBNH₂anh. 1,4-dioxane 200

THF

300

200

100

150

0.75

1.7

2.5

1

17, 86–96

6, 84

18, 65

18, 80

18, 99

of unidentifiable products was ob-

tained. Using conventional heating

and/or a prolonged reaction time

gave similar results. Interestingly,

the use of sodium amide in reflux-

ing 1,4-dioxane provided pyrrole

carboxylic acid 24 in excellent

yield of 99% through an unknown

[a] anh.=anhydrous, Bn=benzyl, MW=microwave, PMB=p-methoxybenzyl,

THF=tetrahydrofuran. [b] 200 psi.

conditions gave only recovered starting material 6 in 70–90%

yield.

Figure 2. Unexpected product

24 from attempted tandem

lactamization and pyrrole

amide formation in the pres-

mechanism (see below). Notably,

For benzylamine, the best yield (up to 96%) of 17 was ob-

tained if THF was used as the solvent (Table 1, entry 1). Howev-

er, under similar conditions but with the use of PMB amine in-

stead, the reaction did not proceed to give 18 (Table 1,

entry 2). After some optimization, the best conditions were

found to furnish 18 in 99% yield (Table 1, entry 5). It is note-

worthy that under MW-assisted conditions, the temperature

for the reactions performed with the use of PMB amine must

be kept under 2008C to avoid undesired side reactions, where-

as starting material 6 remained virtually unreacted.

the same pyrrole carboxylic acid

was also obtained from the at-

tempted tandem lactamization of

ence of sodium amide.

6.

Lactam ring closure and oxidation

Scheme 5. Model studies for direct amidation by using MW irradiation.

[a] 150 W, 1508C, 200 psi for NaNH₂; 300 W, 2008C, 200 psi for TMSN₃. Com-

pound 27 was further reduced by using NaBH₄to give 26 in 80% yield over

two steps.

From pyrrole amide 17 or 18, intramolecular CÀN bond forma-

tion was then performed by using copper(I) thiophene-2-car-

boxylate (CuTC) under MW-assisted conditions to generate the

lactam ring, as previously reported,^[8]except that Cs₂CO₃was

also added and the reaction time was extended to 45 min to

drive the reaction to completion (Scheme 6). The resulting

N,O-protected dihydroazalamellarin D (21 or 22) then under-

went DDQ oxidation smoothly to install the C5=C6 olefin. In

the previous synthetic routes for lamellarin D and azalamella-

rin D, this olefin installation was performed at a much later

stage and required the acetate as the O-protecting group,

which made it necessary to perform the reactions in refluxing

1,2-dichloroethane (DCE). For our current methodology, con-

ducting the DDQ-mediated oxidation reaction of 21 or 22 with

benzyl as the O-protecting group allowed such reactions to

occur under much milder conditions (CH₂Cl₂, RT, 3 h). This fur-

ther negated the possibility of premature deprotection of the

O- and/or N-benzyl, as well as N-PMB groups by DDQ at higher

temperatures.

Amidation using N-protected nitrogen sources

In a report by the Thasana’s group,^[8]a number of N-protected

amines were successfully used for the amidation of pyrrole

esters. However, only the N-allyl group could be effectively re-

moved to generate the final azalamellarin D product by a two-

step process involving the use of ClRh(PPh₃)₃as the catalyst,

followed by OsO₄with NaIO₄as reagents. Therefore, we tried

to develop a new approach focusing on the use of benzyl-type

protecting groups for all the hydroxy moieties and the lactam

nitrogen atom so that they could be later removed simultane-

ously in the final deprotection step. In addition to different ni-

trogen sources, the reaction conditions were also optimized in

terms of solvent, temperature, and reaction time. The desired

pyrrole amide was obtained only if either benzylamine or PMB

amine was employed (Table 1). The use of N-tert-butoxycarbon-

yl (Boc) amine or benzyl carbamate (CbzNH₂) under similar

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2634

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Scheme 6. Lactam ring closure and oxidation en route to azalamellarin D. [a] 200 W, 1508C, 100 psi.

Global deprotection

Table 2. Global deprotection conditions.

The final step in the synthesis of azalamellarins was removal of

all the protecting groups. A major advantage of our current

approach is that all the O- and N-protecting groups remained

intact throughout the synthetic route, which thus avoided the

additional acetylation and deacetylation steps right before and

after DDQ oxidation, respectively, as required in the previously

reported route.^[8]More importantly, these benzyl-type protect-

ing groups for both oxygen and nitrogen atoms were expect-

ed to be removed altogether in a single step.

Entry R’

Solvent

T [8C]

t [h]

Product, yield [%]

Initially, the Pd/C-catalyzed hydrogenolysis was found to

remove only the O-benzyl groups, whereas the N-benzyl

moiety was left intact. Moreover, under such reaction condi-

tions, the C5=C6 olefin was also partially reduced. As an alter-

native method, trifluoroacetic acid (TFA)-mediated acidolysis

was reported for cleavage of the N-benzyl group in some

biaryl ring systems.^[10c]The use of thioanisole as a benzyl

cation scavenger was also described to effectively reduce the

amount of undesired C-benzylated side products.^[12]Therefore,

these approaches were explored in this study.

1

2

3^[b]

4

5

6

7

Bn

TFA

RT

60

22–24

4

16–20

24

4–6

24

30^[a]

Bn

30^[a]

TFA+thioanisole RT

TFA+thioanisole 60

30, trace^[c]

30, 77

4 and 31^[a]

4, 85

PMB TFA

PMB TFA +thioanisole 60

PMB TFA+thioanisole RT, 4–6 h,

RT

4, 79

then 608C, 16 h

[a] A mixture of C-benzylated products was also obtained together with

30 or 31. [b] Upon performing this reaction under MW irradiation (150 W,

1008C, 100 psi) for 1 h, only a trace amount of 30 was obtained. [c] Deter-

mined by ¹H NMR spectroscopy. Bn=benzyl, MW=microwave, PMB=p-

methoxybenzyl, RT=room temperature, TFA=trifluoroacetic acid.

As shown in Table 2, the use of TFA alone failed to remove

the N-benzyl group and partially cleaved off the N-PMB group

(Table 2, entries 1–2 and 5). In addition, some C-benzylated

side products were also found as a result of benzyl cation

attack on the electron-rich polyphenolic core structure of the

lamellarin skeleton. In the presence of thioanisole, the N-

benzyl group still remained unaffected under various condi-

tions (Table 2, entries 3 and 4), whereas the N-PMB group was

successfully removed to furnish azalamellarin D in 79–85%

yield (Table 2, entries 6 and 7).^[13]As described in the Experi-

mental Section, all the optimized conditions were then em-

ployed to synthesize azalamellarin N, of which only the N-pro-

tected forms were obtained as the final products in the previ-

ous study.^[8]

furnished O-benzylated dihydrolamellarin D (34) in 66% yield

over two steps. Subsequent DDQ oxidation and acid-mediated

O-benzyl cleavage successfully gave lamellarin D in 81% yield

(over two steps). The developed approach was also applied to

synthesize lamellarin N from pyrrole ester 7 via the intermedia-

cy of pyrrole carboxylic acid 32,^[14]as shown in Scheme 7 and

described in the Experimental Section.

In summary, starting from the same pyrrole ester intermedi-

ates, similar sets of reactions could be effectively employed to

prepare lamellarins and azalamellarins. Functional group inter-

conversion of pyrrole esters (6 or 7) to the corresponding

amides (18 or 33) or to acids (24 or 32), followed by Cu^I-medi-

ated, MW-assisted intramolecular CÀO or CÀN bond formation,

DDQ-mediated C5ÀC6 oxidation, and acid-mediated global de-

protection of all benzyl-type protecting groups furnished la-

mellarin D (53%), lamellarin N (35%), azalamellarin D (71%),

and azalamellarin N (66%) over four steps. Notably, given that

TFA was employed for the final deprotection, the potential risk

of lactone ring opening under basic conditions otherwise re-

quired for the deacetylation step could be avoided. Therefore,

our established methodology for lamellarin synthesis was sig-

nificantly simplified.

Application to the synthesis of lamellarins

After successful improvement of the synthesis of azalamella-

rins, the possibility of diversifying our methodology to synthe-

size lamellarins was explored. As mentioned previously, treat-

ment of pyrrole ester 6 with sodium amide in refluxing 1,4-di-

oxane resulted in its unexpected conversion into pyrrole car-

boxylic acid 24 (Figure 2) in quantitative yield.^[14]Upon subject-

ing this intermediate to the Cu^I-mediated, MW-assisted condi-

tions (Scheme 7), intramolecular CÀO bond formation

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2635

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Scheme 7. Our optimized conditions for the synthesis of lamellarins 2 and 3 and azalamellarins 4 and 5. [a] 200 W, 150–1708C, 200 psi; [b] 200 W, 1508C,

100 psi. [c] For 22 or 36, the mixture was subsequently heated at 608C for an additional 16 h. For compounds 6, 18, 22, 24, and 34, R³=Me; R⁴=Bn. For com-

pounds 7, 32, 33, 35, and 36, R³=Bn; R⁴=Me. For compounds 2 and 4, R³=Me; R⁴=H. For compounds 3 and 5, R³=H; R⁴=Me.

Physicochemical studies

et al. focusing on a subtle lactone-to-lactam replacement, the

Because of their diverse biological activities, lamellarins have

been subjected to extensive investigations focusing on both

anticancer and anti-HIV effects. Since the 1990s, multiple pat-

ents have been in effect for their uses in multidrug-resistant

tumors,^[15]which further justifies their therapeutic potential.

Nevertheless, an additional requisite of equal importance is

drug likeness, the lack of which has caused a large percentage

of designed compounds to fail along the development pro-

cess.

resulting azalamellarin D molecules were observed to form

a smaller amount of precipitate in cell culture media, while still

exhibiting appreciable cytotoxicity towards selected cancer cell

lines.^[8]However, their obtained azalamellarin materials were in-

sufficient for in-depth physicochemical characterizations, which

were conducted in this study to quantitatively evaluate their

lipophilicity along with both kinetic and thermodynamic solu-

bility. All the parameters obtained for azalamellarin D (4) and

azalamellarin N (5) were then compared with those of their

parent lamellarin D (2) and lamellarin N (3), as shown in

Table 3.

With regard to drug action, sufficient lipophilicity is required

for both membrane permeation and drug-target interactions,

whereas solubility of bioactive compounds in aqueous-based

media allows the dissolved molecules to exert pharmacological

activities under physiological conditions. A balance of these

two inversely correlated properties is essential for successful

development into drug products. In the pharmaceutical arena,

two types of solubility measurements have been established

for different purposes. Generally, “thermodynamic solubility”

resulting from the dissolution of solid materials can be used as

an indicator to identify the need for bioavailability-promoting

strategies. For in vitro bioactivity screening processes, potential

precipitation issues can be predicted by measuring the “kinetic

solubility” upon adding an aliquot of DMSO stock solutions

into the desired medium.

Table 3. Physicochemical parameters of lamellarins 2 and 3 and azala-

mellarins 4 and 5.

Parameter

2

3

4

5

Exp. log P^[a]

3.55Æ0.02 3.49Æ0.01 2.65Æ0.01 2.63Æ0.01

Thermodynamic (3 days)^[b]ND

ND

Thermodynamic (7 days)^[b]0.21Æ0.23 0.12Æ0.13 0.11Æ0.07 0.05Æ0.04

Kinetic (1.5 h)^[b]

Kinetic (24 h)^[b]

2.55Æ2.53 1.71Æ1.34 5.91Æ1.57 5.58Æ1.80

2.57Æ2.58 3.78Æ2.46 5.43Æ2.16 3.07Æ0.77

[a] HPLC method. [b] Solubility in mm determined by using the shake-flask

method at pH 7.4 and 378C. Exp.=experimental, ND=not detectable.

Relative to most commercially available oral drugs, lamellarin

molecules have high molecular weights and are extremely lipo-

philic;^[16]thus, they severely suffer from precipitation in biologi-

cal media. As one of the commonly used solubilizing tech-

niques, lamellarin D has been conjugated with a number of

aqueous solubility-enhancing moieties^[17]to give rise to even

bulkier molecules. Interestingly, in a recent study by Thasana

Lipophilicity determinations

In general, for compounds without ionizable functional

groups, log P is commonly used to indicate their preferences

for a lipophilic environment to aqueous media, that is, log P is

the logarithmic ratio between the concentration of an union-

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2636

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

ized solute in octanol and its concentration in water at equilib-

rium. Using the HPLC-based method previously established in

our laboratory,^[16]the log P values of lamellarin D and lamella-

rin N were determined to be approximately 3.5 (Table 3). Such

a highly lipophilic nature of lamellarins has been attributed to

the extensive stacking interactions among these perfectly

planar molecules^[16]that predominate over the H-bonding in-

teractions between their phenolic hydroxy groups and water.

Interestingly, simple replacement of the lactone oxygen atom

in lamellarins 2 and 3 by a nitrogen atom in azalamellarins 4

and 5 led to a dramatic decrease in the log P values by almost

1 log unit (Table 3). On the basis of our preliminary molecular-

modeling studies, such structural modification did not affect

their molecular planarity (data not shown). However, the car-

bonyl group of an amide offers stronger H-bond acceptor ca-

pacity than an ester,^[18]whereas the free NÀH group can also

act as an additional H-bond donor, which would lead to en-

hanced interactions between the azalamellarins and water

molecules. This might explain why both azalamellarins ap-

peared less lipophilic than their respective parent compounds.

For each pair of structural isomers, that is, 2 vs. 3 and 4 vs. 5,

the switched OH and OMe groups at C13 and C14 on the or-

thogonal ring did not affect the lipophilicity of these molecules

as expected.

than their thermodynamic solubility, possibly as a result of the

fact that the predissolved molecules could readily interact with

the surrounding water molecules without the need for a disso-

lution process. Another driving force could be the enhanced

solvent power of the aqueous medium in the presence of 5%

DMSO from the original stock solutions. Between the two

classes of compounds, our results also demonstrated that the

azalamellarin molecules were more effectively accommodated

in the aqueous environment. Notably, the kinetic solubility of

azalamellarins and their respective lamellarins might not

appear significantly different, especially if taking into account

the expectedly large variations associated with spontaneous

precipitation of poorly soluble compounds. However, a much

smaller amount of precipitate was visually observed with azala-

mellarins even at the 12.5 mm concentration used in the cyto-

toxicity assay. As previously discussed for the lipophilicity de-

terminations, the lactam moiety can act as both an H-bond

donor and an H-bond acceptor, which may possibly lead to in-

teractions with water that are better than those of lactone-

containing lamellarins.

Biological evaluations

After demonstrating that the drug likeness of lamellarin natural

products was partly improved by lactone-to-lactam replace-

ment, we then proceeded to further evaluate the effects of

this structure modification on their biological activities. Where-

as the precipitation issue could be successfully alleviated, it

was also necessary to confirm that the reduced lipophilicity

did not impair their membrane permeability such that the

overall cytotoxicity against selected cancer cells would be se-

verely abolished. In addition, their GSK-3b inhibitory activity

was also evaluated, as both parent compounds, lamellarin D

and lamellarin N, have been previously reported to potently in-

hibit this enzyme in vitro.^[6]

Aqueous solubility measurements

In this study, both types of solubility data were determined for

the synthesized lamellarins and azalamellarins by using the tra-

ditional shake-flask method that was specifically optimized for

these compounds in our laboratory (unpublished results).

Under physiological conditions, that is, pH 7.4 and 378C, both

lamellarin D and lamellarin N exhibited extremely low thermo-

dynamic solubility in the range of 100 to 200 nm, even after

7 days of incubation (Table 3). Rather unexpectedly, this param-

eter was not much improved in the presence of the lactam

moiety, which effectively rendered azalamellarins 4 and 5 less

lipophilic. However, upon considering the dissolution process

of any solid material, the interactions among the solute mole-

cules must be overcome prior to the solvation step. The

amount of energy required to disrupt the crystal lattice can be

realized by the melting points, which were equally high for

both lamellarins and azalamellarins, and this makes it consider-

ably difficult for these compounds to go into the solution

form.

Cytotoxic activity

Through various mechanisms as described in the Introduction,

both lamellarin D and lamellarin N exhibit potent cytotoxic ac-

tivity against a number of cancer cell lines according to previ-

ous reports.^[2,3]Without focusing on any particular underlying

mechanisms, the overall cytotoxic effects of the parent lamel-

larins and their structurally related analogues with altered mo-

lecular properties were evaluated by using three adherent

cancer cell lines (i.e., A549, HepG2, and HuCCA-1) and one sus-

pended cancer cell line (i.e., MOLT-3) for primary screening.^[19]

As shown in Table 4, the IC₅₀values of azalamellarin D

against three adherent cell lines were roughly comparable to

those of lamellarin D within experimental errors, whereas the

suspended MOLT-3 appeared more sensitive to the parent la-

mellarin D. Nonetheless, the activity of azalamellarin D was still

comparable to that of the positive control, etoposide, against

these leukemic cells. For lamellarin N and azalamellarin N, on

the other hand, the analogue was apparently more active than

the parent compound towards A549 and HepG2 cells, whereas

equipotent effects of both compounds were observed with

As a result of poor aqueous solubility commonly observed

with most organic compounds, the use of cosolvents, mostly

dimethyl sulfoxide (DMSO), is normally required to enable

compound screening in preclinical studies. Unfortunately, upon

dilution into aqueous media, precipitation can still occur and

lead to erroneous results. Therefore, a second type of solubility

measurements was established to mimic this situation by

adding DMSO stock solutions of the test compounds into an

aqueous medium, followed by determination of the concentra-

tion remaining in solution after a short incubation period.

As shown in Table 3, the kinetic solubility of both lamellarins

2 and 3 and azalamellarins 4 and 5 was dramatically higher

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2637

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

ed,^[6]in spite of some differences in the experimental

design. Our results also confirmed that the activity of

lamellarin N is superior to that of lamellarin D, al-

though there is a significant difference between the

IC₅₀values obtained in the two studies. Interestingly,

replacement of the lactone moiety with a lactam

functionality further potentiated the GSK-3b inhibito-

ry activity of these compounds.

Table 4. Cytotoxicity against cancer cells and GSK-3b inhibitory activity of lamellarins

2 and 3 and azalamellarins 4 and 5.

Target

IC₅₀^[a][mm]

2

3

4

5

Doxorubicin Etoposide SB 415286

A549^[b]

1.51

0.30

0.29

0.006 0.01

56.94

0.32

I

0.82

0.23

0.13

0.04

0.30

0.38

0.09

0.12

0.01

8.29

0.59

0.26

0.69

ND

3.09

ND

0.05

ND

HepG2^[b]

0.40

0.18

HuCCA-1^[b]

MOLT-3^[c]

MRC-5^[b]

GSK-3b^[d]

(10 mm ATP)

GSK-3b^[d]

(100 mm ATP)

Upon increasing the ATP concentration to 100 mm,

only a 1.5-fold increase in the IC₅₀values was ob-

served with lamellarins, whereas those of both ana-

logues remained virtually unaffected. In contrast, the

inhibitory activity of SB 415286, an ATP-competitive

inhibitor of GSK-3, was significantly diminished with

increasing ATP concentrations. However, caution

should be exercised upon interpreting the results for

lamellarins. As illustrated with some representative

dose–response curves in Figure 3, over 90% inhibi-

tion was always achieved with SB 415286 and azala-

I

0.036 0.018 0.008 ND

0.24

0.52

0.053 0.022 0.009 ND

ND

0.78

[a] Standard deviations are omitted for visual clarity. [b] MTT assay. [c] XTT assay.

[d] ADP-Glo assay. A549=human nonsmall cell lung carcinoma, HepG2=human hep-

atocellular carcinoma, HuCCA-1=human cholangiocarcinoma, MOLT-3 =human acute

T-lymphoblastic leukemia, MRC-5=human fetal/embryonic lung fibroblast, I=inactive

(IC₅₀>100 mm), ND=not determined.

HuCCA-1 and MOLT-3 cells. Our results for the D series com-

pounds were mostly in agreement with those reported by the

Thasana’s group,^[8]and we further confirmed by using the pre-

viously unobtained azalamellarin N that the cytotoxicity of la-

mellarins was not adversely affected by lactone-to-lactam re-

placement. Interestingly, although both analogues were signifi-

cantly more toxic than the parent lamellarins against the MRC-

5 cells originally derived from normal lung tissue,^[20]azalamel-

larin N exhibited better selectivity in killing cancer over normal

cells than azalamellarin D.

mellarins at their maximum concentrations used in our experi-

ments. In contrast, 20–40% GSK-3b activity still remained if ex-

posed to lamellarins at micromolar levels, which may potential-

ly result in lower apparent IC₅₀values and overestimated po-

tency of these compounds.

The incomplete GSK-3b inhibition by lamellarins at micromo-

lar concentrations, possibly as a result of their limited aqueous

solubility (Table 3), precluded our IC₅₀determinations at higher

ATP levels. Nevertheless, we attempted to expose the GSK-3b

enzyme to five different equimolar concentrations of various

inhibitors in the presence of 1 mm ATP, which not only ap-

proaches its physiological concentrations^[23]but is also sug-

gested as the maximum limit to be used in the ADP-Glo assay.

The results in Table 5 indicate that at any equimolar concentra-

tion of the inhibitors, the enzyme was inhibited much more ef-

fectively by both azalamellarins than either their parent com-

pounds or the positive control. On the basis of these results, it

is unlikely that azalamellarins would inhibit GSK-3b enzyme by

GSK-3b inhibitory activity

Among several protein kinases that are reportedly sensitive to

lamellarins and derivatives,^[6]GSK-3 was selected as the molec-

ular target in this study owing to its implications in the pathol-

ogy of several diseases, not only diabetes, but also Alzheimer’s

and other neurodegenerative diseases, bipolar affective disor-

der, as well as cancer.^[21]Within the lamellarin family, lamella-

rin N has been demonstrated as the most potent inhibitor

(IC₅₀=5 nm) against GSK-3a/b from porcine brain in the pres-

ence of 15 mm adenosine triphosphate (ATP).^[6]Although this

enzyme could also be inhibited by lamellarin D, a higher IC₅₀

value of 300 nm was observed. Recently, two HPLC-resolved

atropisomers of 16-methyllamellarin N were developed to in-

vestigate the role of axial chirality towards their kinase selectiv-

ity.^[22]However, both compounds were not only undistinguish-

ed by GSK-3b but were also substantially less active than la-

mellarin N.

Generally, kinase activity can be measured in terms of either

product formation [i.e., phosphorylated peptide or adenosine

diphosphate (ADP)] or substrate disappearance (i.e., peptide

substrate or ATP). From both safety and sensitivity standpoints,

the commercially available, luminescence-based ADP-Glo assay

was used in this study to quantify ATP-to-ADP conversion by

recombinant GSK-3b, either with or without the test com-

pounds. As shown in Table 4, in the presence of 10 mm ATP, the

IC₅₀value of lamellarin D coincided with that previously report-

Figure 3. Representative dose–response curves of lamellarin D (2), azalamel-

larin D (4), and SB 415286 in the presence of 10 or 100 mm ATP.

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2638

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

competing with ATP molecules, as commonly observed with

kinase inhibitors.

pair of structural isomers (i.e., 2 vs. 3 or 4 vs. 5), as well as

those observed between each lamellarin and its respective

lactam-containing derivative (i.e., 2 vs. 4 or 3 vs. 5). Notably,

atropisomerism of lamellarins is generally expected to exist

owing to the theoretically restricted C1ÀC11 rotation. However,

our synthetic methods are nonstereoselective and are likely to

generate a racemic mixture of these compounds, similarly to

those isolated from natural sources (except for lamellarin S^[1f]).

Previously, our attempts to determine the optical activity of se-

lected lamellarins, including lamellarin D, yielded fluctuating

values of their specific rotation ([a]_D) within the range of Æ108

(c=0.50, THF), compared with +66.58 (c=0.84, MeOH) report-

ed for optically active lamellarin S.^[1f]Nevertheless, both aR and

aS forms of compounds 2–5 were virtually constructed and op-

timized prior to use in our docking studies to confirm whether

the GSK-3b enzyme is selective towards a particular atropiso-

mer.

To preliminarily determine their mode of inhibition, an addi-

tional IC₅₀value of the less-active lamellarin D and SB 415286

was determined in the presence of 25 mm ATP, whereas that of

lamellarin N and both azalamellarins was evaluated at 1 mm

ATP. As depicted in Figure 4, an ATP-dependent shift in the po-

tency^[24]of lamellarins, especially lamellarin D, suggested the

competition with ATP as a possible mode of GSK-3b inhibition,

similar to that of SB 415286. Distinctively, the IC₅₀values of

both azalamellarins remained at a maximum of 20 nm regard-

less of the ATP concentration up to 1 mm, which is indicative

of ATP-noncompetitive inhibition. Such findings are of high im-

portance for the development of kinase inhibitors into drug

candidates, as ATP-competitive inhibitors usually suffer from

limited selectivity among various kinases possessing conserved

ATP-binding pockets, as well as the loss of potency if exposed

to ATP at physiological levels.^[24]

Whereas the actual binding site of lamellarins and deriva-

tives on the GSK-3b enzyme is still unknown, the ATP-binding

pockets are commonly used for molecular-docking studies of

kinase inhibitors owing to the availability of the co-crystallized

enzyme–inhibitor complexes in the Protein Data Bank (PDB).

On the basis of a recent study by Yoshida et al.,^[22]the GSK-3b/

staurosporine co-crystal structure (PDB ID: 1Q3D^[26]) was first

used to simulate the interactions formed between the active

site residues and our inhibitor molecules. To verify the em-

ployed docking method, the original ligand (staurosporine)

was first removed and then redocked back onto GSK-3b. The

best-docked conformation was found to be very similar to the

original complex, which confirmed the suitability of the

method utilized herein for GSK-3b. Subsequently, the aR and

aS forms of compounds 2–5 were individually docked onto the

ATP-binding pocket of GSK-3b, from which the original ligand

had been removed. The best-docked conformation of each la-

mellarin was identified by assessing the binding energy be-

tween the ligand and the active site. However, the binding

energy values obtained by using 1Q3D (data not shown) did

not correlate with either the observed interactions or the ex-

perimentally determined inhibitory activities of our com-

pounds, which might also interact with some other residues

that are not involved in the current model.

Molecular-modeling studies

Following the GSK-3b assay, molecular-docking studies were

undertaken. Possible interactions underlying the affinity of our

inhibitor molecules for the GSK-3b enzyme were explored to

rationalize the different inhibitory activities exerted by each

Table 5. GSK-3b inhibition in the presence of 1 mm ATP.

Concentration [nm]

Inhibition^[a][%]

2

3

4

5

SB 415286

1

ND

5.4

13.1

ND

11.4

68.4

10

ND

4.8

28.0

73.2

96.5

ND

41.6

88.1

96.1

ND

100

1000

10000

ND

15.3

49.6

56.9

À3.0

29.4

[a] Average values from duplicate wells with <10% standard deviations.

ND=not determined.

Hence, another co-crystal structure between GSK-3b and 3-

aryl-4-(arylhydrazono)-1H-pyrazol-5-ones (PDB ID: 3L1S^[27]) was

then selected, and the docking experiments were performed

by the same protocols utilized earlier for 1Q3D. The docking

results are illustrated in Figure 5 by using the aS form of each

compound (see Figure S1 in the Supporting Information for

the docking results of the aR form). We focused on hydrogen

bonds, which usually play an important role in enzyme–ligand

interactions. The comparable calculated binding energies and

similar interactions observed for both atropisomeric forms of

each lamellarin indicated that these isomers were not distin-

guished by the relatively large ATP-binding pocket of GSK-3b,

as previously reported for the two HPLC-resolved atropisomers

of 16-methyllamellarin N.^[22]

Figure 4. Dependence of the IC₅₀values on the ATP concentrations. A litera-

ture-based Michaelis–Menten constant (K_m) value of 10 mm for ATP and GSK-

3b enzyme^[25]was used for data analysis.

As expected, the lactone carbonyl group as well as the hy-

droxy groups at both C8 and C20 of lamellarin D are all within

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2639

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

the H-bond distances from Val135, Glu97, and Pro136, respec-

tively (Figure 5a,c). Additionally, the oxygen atoms of the hy-

droxy group at C8 and the methoxy group at C9 serve as H-

bond acceptors that interact with Lys85. Unexpectedly, the

C14 hydroxy moiety, which is also considered beneficial for the

cytotoxicity of this compound,^[17a]does not appear to interact

with any residues in the ATP-binding pocket. Instead, if this

phenolic hydroxy group was switched to C13 of lamellarin N,

an additional hydrogen bond with Asn64 was observed (Fig-

ure 5e,g). These observations, in conjunction with the C9 me-

thoxy oxygen atom serving as a potential H-bond acceptor for

the side-chain amino group of Lys85, might explain the signifi-

cantly different inhibitory activities of these two isomeric la-

mellarins (Table 4). In addition, replacing the lactone of lamel-

larin D with a lactam in azalamellarin D resulted in simultane-

ous H-bond formation between the amide NÀH group and

both the carbonyl and NÀH moieties of Val135 (Figure 5a–d).

Additive stabilizing effects were observed with azalamellarin N

possessing both the C13 hydroxy group and the lactam

moiety (Figure 5 f,h). Although the binding energy differences

among 2, 3, 4, and 5 are not drastic, our docking results are in

line with the inhibitory activities towards the GSK-3b enzyme.

On the basis of these results combined with those from the

above-mentioned GSK-3b assays, it is possible that multiple H-

bonding interactions between azalamellarins and their binding

site could lead to strong binding to the enzyme, which might

hamper its catalytic activity without affecting its binding to

ATP owing to their structural nonsimilarities.

molecules could better interact with water molecules, which

thereby alleviated their precipitation issues in aqueous media.

Interestingly, both azalamellarin D and azalamellarin N could

still effectively kill cancer cells, albeit with significant toxicity

against normal cells. More importantly, such subtle structural

modification led to substantially enhanced inhibitory activity

towards the multifunctional GSK-3b enzyme with a distinct

mode of inhibition from that of the parent lamellarins. Current-

ly, additional analogues are being developed in our laboratory

to further improve the aqueous solubility and cancer selectivity

of these compounds.

Experimental Section

Synthesis: general

Acetonitrile and toluene were purified by a solvent purification

system, whereas tetrahydrofuran (THF) was either distilled from

sodium benzophenone ketyl or purified by the solvent purification

system. All the other chemicals were used as received from suppli-

ers. Unless stated otherwise, reactions were performed in oven-

dried, round-bottomed flasks and were monitored by analytical

thin-layer chromatography (TLC) on silica gel 60 F₂₅₄aluminum

sheets. Crude reaction mixtures were concentrated under reduced

pressure. Column chromatography and preparative TLC (P-TLC)

were performed by using silica gel 60 (particle size 0.06–0.2 mm;

70–230 mesh ASTM).

The final lamellarin and azalamellarin products were purified by

crystallization from appropriate solvent systems and, if necessary,

were further subjected to semipreparative HPLC purification

through

a SymmetryPrep C18 column (7 mm, 19 mm i.d.

Conclusions

300 mm). Stepwise elution (10 mLmin^À1) started with MeOH/

water (70:30 v/v) for 20 min, followed by a linear gradient over

5 min to 100% MeOH, which was then maintained for 10 min

before the column was equilibrated for the next injection. The

compound purity was determined to be >95% on the basis of an-

alytical HPLC through a ZORBAX Eclipse Plus C18 column (5 mm,

4.6 mm i.d. 250 mm) by using two mobile phase systems at

1 mLmin^À1and UV detection at l=276 nm, as described in the

Supporting Information (Table S2). The purified lamellarins and aza-

lamellarins were stored refrigerated at 48C until use.

Herein, we reported significant advancements in the synthesis

of marine-derived lamellarin natural products and their lactam-

containing derivatives called azalamellarins. In this study, a gen-

eral divergent method featuring the use of only a common 2-

bromo-a-nitrocinnamate to prepare a late common pyrrole

ester intermediate en route to both classes of compounds was

successfully devised. Subsequently, either the novel NaNH₂-

mediated or the AlMe₃-mediated conversion of the pyrrole

ester was accomplished to furnish the corresponding pyrrole

carboxylic acid or pyrrole amide, respectively. The desired lac-

tonization or lactamization was effectively performed through

a single and similar reaction type of the CuTC-mediated, micro-

wave-assisted intramolecular CÀO or CÀN bond formation. Our

designed approach also allowed the O-benzyl protecting

groups to remain throughout the DDQ-mediated oxidation,

which thereby avoided the additional acetylation and deacety-

lation steps that put lamellarin molecules at risk of lactone

ring opening. Simple TFA-mediated acidolysis in the presence

of thioanisole effectively removed all O- and N-benzyl-type

protecting groups in one step, which furnished the desired la-

mellarins and azalamellarins in good overall yields over four

steps starting from the pyrrole esters.

Depending on the compound solubility, proton (¹H) and carbon

(¹³C) nuclear magnetic resonance (NMR) spectra of the synthesized

materials were recorded with a 300 MHz spectrometer (300 MHz

1

for H; 75 MHz for ¹³C) by using either CDCl₃or [D₆]DMSO as sol-

vent. The chemical shifts are reported in parts per million (ppm, d)

downfield from tetramethylsilane (SiMe₄), whereas the splitting

patterns of the H NMR spectra are listed as singlet (s), doublet (d),

1

triplet (t), quartet (q), multiplet (m), broad (br), and doublet of dou-

blet (dd), along with the coupling constants (Hz) and integration.

Melting points (m.p.) are uncorrected, and infrared (IR) resonances

are reported in wavenumber (cm^À1). Mass spectrometry (MS) was

performed by using electron impact (EI) mode, whereas high-reso-

lution mass spectrometry (HRMS) was conducted by using the

time of flight (TOF) method with atmospheric-pressure chemical

ionization (APCI) or electrospray ionization (ESI). Compound char-

acterization data are as described below.

From detailed comparative physicochemical and biological

profiling, our results clearly highlight the promising benefits of

replacing the lactone oxygen atom in the lamellarin skeleton

with a nitrogen atom. With lower lipophilicity, azalamellarin

4-(Benzyloxy)-3-methoxybenzaldehyde (12): A mixture of vanillin

(16.0 g, 105 mmol), ethanol (150 mL), potassium carbonate (22.8 g,

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2640

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Figure 5. Lamellarins and azalamellarins in the best-docked conformations within the ATP-binding pocket of GSK-3b (PDB ID: 3L1S). Numbers indicate the cal-

culated binding energies.

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2641

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

164 mmol), and benzyl bromide (13.7 mL, 116 mmol) was stirred at

reflux for 6 h. Then, the mixture was concentrated, and the residue

was diluted with water and extracted with EtOAc (330 mL). The

combined organic extract was washed with water (330 mL) and

brine (30 mL), dried with anhydrous sodium sulfate, and concen-

trated under reduced pressure. The crude product was subse-

quently purified by recrystallization in 25–45% EtOAc/hexane to

give white crystals of 12 (24.2 g, 95%). Physical and spectroscopic

data are in agreement with those previously reported.^[28]

ical and spectroscopic data are in agreement with those previously

reported.^[31]

1-(Benzyloxy)-4-(chloromethyl)-2-methoxybenzene (39): Thionyl

chloride (1.96 mL, 16.4 mmol) was slowly added to a solution of 38

(2.68 g, 11.0 mmol) in CH₂Cl₂(22 mL) at 08C. Then, the mixture was

warmed up to room temperature. After 30 min, the reaction was

quenched with water, and this was followed by the extraction with

CH₂Cl₂(320 mL) and washing with water (320 mL). The com-

bined organic layer was then washed with 10% w/v sodium hy-

droxide (620 mL) until basic pH, as indicated by litmus paper, fol-

lowed by washing with brine (30 mL) and drying with anhydrous

sodium sulfate. The crude product was concentrated under re-

duced pressure, dried in vacuo, and used directly in the next step

without further purification.

3-(Benzyloxy)-4-methoxybenzaldehyde (37): A mixture of isovanillin

(16.7 g, 110 mmol), ethanol (150 mL), potassium carbonate (22.8 g,

164 mmol), and benzyl bromide (14.4 mL, 121 mmol) was stirred at

reflux. After 3 h, ethanol was evaporated, and this was followed by

the addition of water and extraction with EtOAc (330 mL). Then,

the combined organic layer was washed with water (330 mL)

and brine (30 mL), dried with anhydrous sodium sulfate, and con-

centrated under reduced pressure. Next, the crude product was

purified by recrystallization in 25–45% EtOAc/hexane to give pale

yellow crystals of 37 (24.0 g, 91%). Physical and spectroscopic data

are in agreement with those previously reported.^[29]

2-[4-(Benzyloxy)-3-methoxyphenyl]acetonitrile (40): Crude 39 was

dissolved in a biphasic mixture of CH₂Cl₂/water (1:1, 40 mL), and

this was followed by the addition of potassium cyanide (1.07 g,

16.4 mmol) and tetra-n-butylammonium bromide (3.53 g,

11.0 mmol). The mixture was stirred at room temperature over-

night. Then, the crude product was extracted with CH₂Cl₂(3

50 mL). The combined organic layer was washed with water (3

50 mL) and brine (60 mL), dried with anhydrous sodium sulfate,

and concentrated under reduced pressure. Finally, the crude prod-

uct was purified by column chromatography (10–20% EtOAc/

hexane) to give pale yellow crystals of 40 (2.14 g, 76%). M.p. 67.2–

68.58C (lit.^[32]68.0–69.08C); ¹H NMR (300 MHz, CDCl₃, SiMe₄): d=

3.70 (s, 2H), 3.92 (s, 3H), 5.17 (s, 2H), 6.79–6.89 (m, 3H), 7.28–

7.46 ppm (m, 5H); ¹³C NMR (75 MHz, CDCl₃, SiMe₄): d=23.2, 56.1,

71.1, 111.5, 114.4, 120.2, 122.7, 127.2, 127.9, 128.6, 136.8, 147.9,

150.2 ppm; IR (UATR): n˜ =2927, 2249, 1512, 1224 cm^À1; MS (EI): m/z

(%): 253 (10) [M⁺], 91 (100), 65 (12).

4-(Benzyloxy)-2-bromo-5-methoxybenzaldehyde (13): Pyridinium

tribromide (4.12 g, 12.9 mmol) was added to a stirred solution of

12 (2.00 g, 8.34 mmol) in MeOH (80 mL) at room temperature.

After 3 h, MeOH was evaporated, and the mixture was diluted with

water before extraction with EtOAc (330 mL). The combined or-

ganic layer was washed with water (330 mL) and brine (30 mL),

dried with anhydrous sodium sulfate, and concentrated under re-

duced pressure. The crude material was purified by recrystallization

in 10–15% EtOAc/hexane to give pale orange crystals of 13

(2.30 g, 85%). Physical and spectroscopic data are in agreement

with those previously reported.^[30]

Ethyl

3-[4-(benzyloxy)-2-bromo-5-methoxyphenyl]-2-nitroacrylate

2-[4-(Benzyloxy)-3-methoxyphenyl]acetic acid (8): Potassium hy-

droxide (10.1 g, 181 mmol) was added to a mixture of 40 (5.72 g,

22.6 mmol) in ethanol (44 mL) and water (22 mL), and the mixture

was stirred at reflux overnight. Afterwards, the mixture was cooled

to room temperature, concentrated under reduced pressure, acidi-

fied with 6 n HCl (pH~2–3), and extracted with EtOAc (360 mL).

The combined organic layer was washed with water (360 mL)

and brine (70 mL), dried with anhydrous sodium sulfate, and con-

centrated under reduced pressure. Finally, the crude product was

purified by recrystallization in 50–80% EtOAc/hexane to yield 8 as

a white solid (5.97 g, 97%). Physical and spectroscopic data are in

agreement with those previously reported.^[33]

(11): Diethylamine hydrochloride (4.11 g, 37.4 mmol) and ethyl ni-

troacetate (3.42 mL, 31.1 mmol) were added to a solution of 13

(8.00 g, 24.9 mmol) in toluene (300 mL). The reaction vessel was

equipped with a Dean–Stark apparatus, and the mixture was

stirred at reflux under an argon atmosphere for 3 days. Subse-

quently, toluene was removed under reduced pressure, and this

was followed by the addition of water and extraction with EtOAc

(2100 mL). The combined organic layer was washed with water

(3100 mL) and brine (30 mL), dried with anhydrous sodium sul-

fate, and concentrated under reduced pressure. The crude product

was purified by column chromatography (20–30% EtOAc/hexane)

to give a 1.5:1 mixture of 11/13 as a brown sticky product (3.76 g)

along with the starting material (3.87 g). The spectroscopic data

are in agreement with those previously reported.^[8]

2-[4-(Benzyloxy)-3-methoxyphenyl]acetyl chloride (41): Oxalyl chlo-

ride (2.80 mL, 32.9 mmol) and N,N-dimethylformamide (DMF;

2 drops) were sequentially added to a solution of acid 8 (5.97 g,

22.0 mmol) in CH₂Cl₂(44 mL). The reaction vessel was equipped

with a drying tube, and the mixture was stirred at room tempera-

ture. After 2 h, the crude product was evaporated to remove all or-

ganic solvents. The product was immediately used in the next step

of the amide synthesis without further purification. Owing to its

unstable nature, complete spectroscopic characterization was not

performed.

Synthesis: azalamellarin D (4) and lamellarin D (2)

[4-(Benzyloxy)-3-methoxyphenyl]methanol (38): Sodium borohy-

dride (1.14 g, 30.2 mmol) was added to a solution of 12 (2.93 g,

12.1 mmol) in ethanol (120 mL) and THF (60 mL). The mixture was

stirred at room temperature for 2 h and then concentrated under

reduced pressure; this was followed by the addition of water and

extraction with EtOAc (330 mL). Then, the combined organic

layer was washed with water (330 mL) and brine (30 mL), dried

with anhydrous sodium sulfate, and concentrated under reduced

pressure. The crude product was purified by recrystallization in 40–

60% EtOAc/hexane to give white crystals of 38 (2.68 g, 91%). Phys-

(E)-2-(Benzyloxy)-1-methoxy-4-(2-nitrovinyl)benzene (14): Ammoni-

um acetate (28.9 g, 395 mmol) and nitromethane (31.9 mL,

592 mmol) were sequentially added to a solution of 37 in glacial

acetic acid (38 mL) at room temperature. Then, this mixture was

stirred at reflux for 30 min. Afterwards, the mixture was cooled to

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2642

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

room temperature, during which time crystallization of the crude

product spontaneously occurred. The yellow crystals were filtered

and washed with an excess amount of petroleum ether or hexane,

followed by an excess amount of water to wash away impurities.

After redissolving in heated absolute ethanol, the product was al-

lowed to recrystallize at room temperature. Finally, the product

was filtered and then dried in vacuo to obtain bright yellow crys-

tals of 14 (25.0 g, 89%). Physical and spectroscopic data are in

agreement with those previously reported.^[34]

Pyrrole ester 6: Sodium bicarbonate (0.375 g, 4.47 mmol) and a-ni-

trocinnamate 11 (2.20 g, 3.44 mmol) were added to crude 15 dis-

solved in anhydrous acetonitrile (50 mL) at room temperature. The

mixture was then stirred at reflux overnight before the solvent was

removed. Next, the reaction was quenched with water, and the

mixture was extracted with EtOAc (340 mL). The combined or-

ganic layer was washed with water (340 mL) and brine (30 mL),

dried with anhydrous sodium sulfate, and concentrated under re-

duced pressure. Finally, the crude product was purified by column

chromatography (30–50% EtOAc/hexane) to give pyrrole ester 6

(1.60 g, 51% over two steps from amide 42). M.p. 76.4–77.58C;

¹H NMR (300 MHz, CDCl₃, SiMe₄): d=0.85 (t, J=6.9 Hz, 3H), 2.90–

3.10 (m, 2H), 3.23 (s, 3H), 3.63 (s, 3H), 3.65 (s, 3H), 3.94–4.10 (m,

2H), 4.31–4.36 (m, 1H), 4.87–4.92 (m, 1H), 5.11 (s, 2H), 5.13 (s, 4H),

6.54 (s, 1H), 6.71–6.74 (m, 5H), 7.03 (s, 1H), 7.28–7.44 ppm (m,

15H); ¹³C NMR (75 MHz, CDCl₃, SiMe₄): d=13.6, 28.9, 42.7, 55.2,

55.8, 56.0, 59.7, 70.7, 71.0, 71.1, 108.9, 113.2, 113.5, 114.3, 115.0,

115.6, 117.0, 118.6, 121.3, 121.4, 122.8, 125.5, 127.0, 127.2, 127.4,

127.8, 127.9, 128.0, 128.3, 128.5, 128.6, 128.7, 130.4, 130.8, 131.6,

136.5, 136.9, 137.2, 146.5, 147.1, 147.2, 147.8, 148.1, 149.2,

161.5 ppm; IR (UATR): n˜ =2935, 1686, 1495, 1454, 1422, 1402, 1383,

1322, 1248, 1210, 1176, 1134, 1022 cm^À1; MS (EI): m/z (%): 881 (44)

[M+2H⁺], 879 (50) [M⁺], 618 (12), 368 (20), 91 (87); HRMS (ESI-

TOF): m/z: calcd for C₅₁H₄₇⁸¹BrNO₈: 882.2472 [M+H⁺]; found:

882.2469; C₅₁H₄₇⁷⁹BrNO₈: 880.2479 [M+H⁺]; found: 880.2452. These

spectroscopic data are similar to those previously reported.^[8]

2-[3-(Benzyloxy)-4-methoxyphenyl]ethanamine (10): A solution of

14 (2.66 g, 70.1 mmol) in anhydrous THF (45 mL) was added to

a stirred slurry of lithium aluminum hydride (5.00 g, 17.5 mmol) in

anhydrous THF (45 mL) at 08C under an argon atmosphere. The

mixture was then brought to room temperature and stirred over-

night. Next, the reaction was quenched by adding a cold mixture

of EtOAc/water (1:1) until the generation of gas ceased. The result-

ing mixture was filtered to remove all gray solid particles. The

crude product was then concentrated under reduced pressure. Fi-

nally, amine 10 was immediately used for the synthesis of amide

without further purification.

2-[4-(Benzyloxy)-3-methoxyphenyl]-N-[3-(benzyloxy)-4-methoxyphe-

nethyl]acetamide (42): Acid chloride 41 (21.9 mmol) dissolved in

CH₂Cl₂(50 mL) was added to a solution of crude 10 (17.5 mmol)

in CH₂Cl₂(40 mL), and this was followed by the addition of a

10% w/v aqueous solution (46 mL) of sodium carbonate (4.64 g,

43.8 mmol). After 3 h at room temperature, the reaction was

quenched with water and extracted with CH₂Cl₂(330 mL). The

combined organic layer was washed with water (330 mL) and

brine (30 mL), dried with anhydrous sodium sulfate, and concen-

trated under reduced pressure. Finally, the crude product was puri-

fied by column chromatography (50–100% EtOAc/hexane) to pro-

vide pale yellow crystals of 42 (4.58 g, 52% over two steps from

Pyrrole amide 17: Trimethylaluminum (1.0m in toluene, 0.11 mL,

0.11 mmol) was added to a mixture of pyrrole ester 6 (0.0401 g,

0.0456 mmol) and benzylamine (0.024 mL, 0.23 mmol) in THF

(1 mL). Then, microwave irradiation (300 W, 2008C, 45 min) was

used to convert the pyrrole ester into the corresponding pyrrole

amide. Afterwards, the mixture was cooled to room temperature

before the reaction was quenched with 2 n HCl and then extracted

with EtOAc (220 mL). The combined organic layer was washed

with water (320 mL) and brine (30 mL), dried with anhydrous

sodium sulfate, and concentrated under reduced pressure. Finally,

the crude product was purified by using P-TLC (30% EtOAc/

hexane) to give pyrrole amide 17 (0.0381 g, 89%). M.p. 63.4–

65.88C; ¹H NMR (300 MHz, CDCl₃, SiMe₄): d=2.90–3.08 (m, 2H),

3.23 (s, 3H), 3.59 (s, 3H), 3.64 (s, 3H), 4.26 (dd, J=6.0, 5.0 Hz, 1H),

4.37–4.41 (m, 1H), 4.44 (dd, J=6.1, 5.0 Hz, 1H), 4.90–4.96 (m, 1H),

5.00 (s, 2H), 5.11 (s, 2H), 5.12 (s, 2H), 5.82 (t, J=5.4 Hz, 1H), 6.60 (s,

1H), 6.68 (s, 1H), 6.73 (s, 4H), 6.93–6.97 (m, 3H), 7.18–7.43 ppm (m,

18H); ¹³C NMR (75 MHz, CDCl₃, SiMe₄): d=29.0, 29.7, 43.0, 43.5,

55.2, 55.8, 56.0, 70.7, 71.0, 71.1, 108.8, 113.4, 113.6, 114.2, 115.4,

115.8, 117.2, 120.8, 121.5, 121.7, 122.8, 125.6, 126.7, 127.1, 127.2,

127.3, 127.4, 127.5, 127.8, 127.9, 128.0, 128.2, 128.4, 128.5, 128.56,

128.62, 128.7, 129.3, 136.2, 137.0, 137.2, 137.9, 146.6, 147.0, 147.9,

148.4, 148.9, 149.2, 161.6 ppm; IR (UATR): n˜ =3416, 2926, 1643,

1

14). M.p. 108.7–112.58C; H NMR (300 MHz, CDCl₃, SiMe₄): d=2.63

(br t, J=6.6 Hz, 2H), 3.39 (q, J=6.0 Hz, 2H), 3.44 (s, 2H), 3.84 (s,

3H), 3.86 (s, 3H), 5.11 (s, 2H), 5.14 (s, 2H), 5.35 (br s, 1H), 6.54 (d,

J=8.1 Hz, 1H), 6.62 (d, J=7.8 Hz, 1H), 6.68 (s, 2H), 6.75 (d, J=

8.4 Hz, 1H), 6.82 (d, J=8.1 Hz, 1H), 7.28–7.46 ppm (m, 10H);

¹³C NMR (75 MHz, CDCl₃, SiMe₄): d=34.9, 40.7, 43.4, 55.9, 56.0, 71.1,

112.0, 113.0, 114.3, 114.7, 121.4, 121.5, 127.2, 127.4, 127.9, 128.51,

128.54, 137.09, 137.12, 147.4, 148.2, 148.6, 149.9, 171.2 ppm; IR

(UATR): n˜ =3300, 2933, 1509, 1258 cm^À1; MS (EI): m/z (%): 511 (4)

[M⁺], 240 (20), 137 (22), 91 (100), 69 (42); HRMS (APCI-TOF): m/z:

calcd for C₃₂H₃₃NO₅+Na⁺: 534.2251 [M+Na⁺]; found: 534.2238.

6-(Benzyloxy)-1-[4-(benzyloxy)-3-methoxybenzyl]-7-methoxy-3,4-di-

hydroisoquinoline (15): Amide 42 (2.28 g, 4.47 mmol) was dissolved

in anhydrous acetonitrile (20 mL) in a two-necked, round-bot-

tomed flask, to which phosphorus oxychloride (1.23 mL,

13.4 mmol) was added. The mixture was heated at reflux under an

argon atmosphere for 2 h. After that, the mixture was cooled to

room temperature before acetonitrile was removed. The crude

mixture was then basified with a 10% w/v aqueous solution

(21 mL) of sodium carbonate (2.13 g, 20.1 mmol), which was fol-

lowed by the extraction with EtOAc (340 mL). The combined or-

ganic layer was washed with water (340 mL) and brine (40 mL),

dried with anhydrous sodium sulfate, concentrated under reduced

pressure, and then dried in vacuo to yield key intermediate dihy-

droisoquinoline 15, which was used in the next step without fur-

ther purification.

1512, 1494, 1454, 1384, 1330, 1248, 1211, 1177, 1141, 1024 cm^À1

;

MS (EI): m/z (%): 941 (1) [M⁺], 149 (12), 148 (13), 104 (100), 76 (65),

73 (13), 57 (12); HRMS (ESI-TOF): m/z: calcd for C₅₆H₅₀⁸¹BrN₂O₇:

943.2790 [M+H⁺]; found: 943.2653; C₅₆H₅₀⁷⁹BrN₂O₇: 941.2795

[M+H⁺]; found: 941.2801.

Pyrrole amide 18: Trimethylaluminum (1.0m in toluene, 0.11 mL,

0.11 mmol) was added to a mixture of pyrrole ester 6 (0.0401 g,

0.0456 mmol) and p-methoxybenzylamine (0.032 mL, 0.23 mmol) in

anhydrous 1,4-dioxane (1 mL). Then, microwave irradiation (200 W,

1658C, 65 min) was used to convert the pyrrole ester into the cor-

responding pyrrole amide. Subsequent steps were performed as

described for compound 17 to furnish compound 18 (0.0434 g,

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2643

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

1

98%). M.p. 74.9–76.18C; H NMR (300 MHz, CDCl₃, SiMe₄): d=2.88–

2,3-dichloro-5,6-dicyano-1,4-benzoquinone

(DDQ;

0.0151 g,

3.09 (m, 2H), 3.23 (s, 3H), 3.59 (s, 3H), 3.64 (s, 3H), 3.70 (s, 3H),

4.20 (dd, J=9.4, 4.9 Hz, 1H), 4.35 (dd, J=8.5, 5.8 Hz, 1H), 4.42 (m,

1H), 4.89–4.91 (m, 1H), 4.93 (s, 2H), 5.10 (s, 2H), 5.12 (s, 2H), 5.73

(br t, J=6.0 Hz, 1H), 6.58 (s, 1H), 6.68–6.76 (m, 7H), 6.88–6.91 (m,

3H), 7.25–7.44 ppm (m, 15H); ¹³C NMR (75 MHz, CDCl₃; SiMe₄): d=

29.0, 31.7, 43.0, 55.19, 55.24, 55.8, 56.0, 70.8, 71.1, 108.8, 113.5,

113.7, 113.8, 114.3, 115.5, 115.8, 117.3, 120.8, 121.6, 121.8, 122.9,

124.3, 125.6, 126.7, 127.1, 127.3, 127.4, 127.8, 127.9, 128.2, 128.5,

128.55, 128.64, 128.9, 129.3, 130.0, 136.2, 137.1, 137.2, 146.6, 147.0,

147.9, 148.4, 149.0, 149.2, 158.8, 161.5 ppm; IR (UATR): n˜ =3418,

2932, 2835, 1685, 1645, 1602, 1511, 1494, 1463, 1384, 1330, 1247,

1211, 1175, 1137, 1026 cm^À1; MS (EI): m/z (%): 972 (1) [M+2H⁺],

281 (24), 208 (18), 207 (82), 135 (33), 129 (47), 99 (33), 97 (53), 85

(59), 83 (53), 73 (100); HRMS (ESI-TOF): m/z: calcd for

C₅₇H₅₂⁸¹BrN₂O₃: 973.2896 [M+H⁺]; found: 973.2915; C₅₇H₅₂⁷⁹BrN₂O₃:

971.2902 [M+H⁺]; found: 971.2888.

0.0667 mmol) was added, and the mixture was stirred at room

temperature for 3 h. Subsequently, water (2 mL) was added to

quench the reaction, followed by the extraction with CH₂Cl₂(3

10 mL). The combined organic layer was washed with water (3

20 mL) and brine (30 mL), dried with anhydrous sodium sulfate,

and concentrated under reduced pressure. Finally, the crude prod-

uct was purified by P-TLC (40% EtOAc/hexane) or column chroma-

tography (35–60% EtOAc/hexane) to produce compound 28

(0.0220 g, 95%). M.p. 116.5–118.18C; ¹H NMR (300 MHz, CDCl₃,

SiMe₄): d=3.34 (s, 3H), 3.36 (s, 3H), 3.89 (s, 3H), 4.97 (s, 2H), 5.24

(s, 2H), 5.31 (s, 2H), 5.54 (br s, 2H), 6.79 (s, 1H), 6.88 (d, J=7.5 Hz,

1H), 7.00 (s, 1H), 7.08 (s, 1H), 7.14–7.22 (m, 5H), 7.26–7.51 (m,

19H), 9.62 ppm (d, J=7.5 Hz, 1H); ¹³C NMR (75 MHz, CDCl₃, SiMe₄):

d=45.6, 55.1, 55.2, 56.2, 70.7, 70.9, 71.1, 102.5, 105.7, 106.5, 109.6,

110.2, 110.9, 111.9, 114.7, 115.1, 119.5, 123.8, 124.1, 126.4, 127.0,

127.1, 127.2, 128.0, 128.05, 128.11, 128.6, 128.67, 128.72, 128.8,

130.1, 131.4, 132.5, 136.5, 137.0, 137.2, 144.9, 147.6, 148.5, 149.2,

150.6, 156.0 ppm; IR (UATR): n˜ =3336, 3027, 2925, 1706, 1626,

1512, 1493, 1454, 1430, 1259, 1217, 1182, 1027 cm^À1; MS (EI): m/z

(%): 859 (1) [M+H⁺], 149 (24), 104 (100), 91 (16), 76 (60); HRMS

(ESI-TOF): m/z: calcd for C₅₆H₄₇N₂O₇: 859.3383 [M+H⁺]; found:

859.3347.

N,O-Arylmethylenated dihydroazalamellarin D (21 and 22): Pyrrole

amide 17 (0.0250 g, 0.0266 mmol) was dissolved in DMF (2 mL).

Copper(I) thiophene-2-carboxylate (CuTC; 0.0122 g, 0.0638 mmol)

and cesium carbonate (0.0087 g, 0.027 mmol) were added. The

mixture was heated under microwave irradiation (200 W, 1508C,

45 min). Afterwards, the reaction was quenched with a saturated

ammonium chloride solution (3 mL), and the mixture was extracted

with EtOAc (315 mL). The combined organic layer was then

washed with water (315 mL) and brine (20 mL), dried with anhy-

drous sodium sulfate, and concentrated under reduced pressure.

The crude product was purified by using P-TLC (30% EtOAc/

hexane) to give compound 21 (0.0227 g, 96%). M.p. 64.7–66.88C;

¹H NMR (300 MHz, CDCl₃, SiMe₄): d=3.04 (br t, J=6.3 Hz, 2H), 3.28

(s, 3H), 3.31 (s, 3H), 3.87 (s, 3H), 4.89 (s, 2H), 4.84–4.91 (m, 1H),

5.00–5.08 (m, 1H), 5.15 (s, 2H), 5.27 (s, 2H), 5.46 (br s, 2H), 6.72–

6.75 (m, 3H), 6.88 (s, 1H), 7.06–7.21 (m, 5H), 7.26–7.48 ppm (m,

18H); ¹³C NMR (75 MHz, CDCl₃, SiMe₄): d=29.0, 42.4, 45.7, 55.15,

55.18, 56.2, 70.8, 70.9, 71.1, 102.6, 105.8, 108.9, 112.5, 113.3, 114.2,

114.6, 114.7, 121.2, 123.6, 125.6, 126.3, 126.5, 126.96, 127.02, 127.1,

127.2, 127.89, 127.93, 128.04, 128.5, 128.6, 128.7, 128.75, 128.83,

130.5, 133.8, 136.6, 136.8, 137.0, 137.2, 144.9, 146.8, 147.4, 147.5,

147.9, 150.4, 155.9 ppm; IR (UATR): n˜ =3027, 2931, 1637, 1515,

1486, 1454, 1418, 1262, 1213, 1177 cm^À1; MS (EI): m/z (%): 861 (2)

[M+H⁺], 860 (3) [M⁺], 129 (34), 104 (100), 91 (21); HRMS (ESI-TOF):

m/z: calcd for C₅₆H₄₉N₂O₇: 861.3534 [M+H⁺]; found: 861.3571.

Using a similar procedure, compound 22 (0.0490 g, 0.0549 mmol)

was employed to furnish corresponding product 29 (0.0454 g,

93%). M.p. 224.8–226.18C; ¹H NMR (300 MHz, CDCl₃, SiMe₄): d=

3.34 (s, 3H), 3.36 (s, 3H), 3.76 (s, 3H), 3.89 (s, 3H), 5.02 (s, 2H), 5.24

(s, 2H), 5.30 (s, 2H), 5.47 (br s, 2H), 6.77–6.83 (m, 4H) 6.87 (d, J=

7.5 Hz, 1H), 7.01 (s, 1H), 7.07–7.16 (m, 7H), 7.29–7.51 (m, 14H),

9.61 ppm (d, J=7.4 Hz, 1H); ¹³C NMR (75 MHz, CDCl₃, SiMe₄): d=

45.1, 55.1, 55.22, 55.25, 56.2, 70.8, 70.9, 71.1, 102.5, 105.8, 106.6,

109.7, 110.2, 110.9, 112.0, 114.2, 114.8, 115.2, 119.6, 123.8, 124.2,

127.02, 127.05, 127.19, 127.25, 127.6, 127.9, 128.0, 128.1, 128.6,

128.66, 128.72, 129.2, 130.1, 131.4, 132.5, 136.6, 137.0, 145.0, 147.7,

148.6, 149.3, 150.7, 156.0, 158.6 ppm; IR (UATR): n˜ =3749, 2933,

2308, 1611, 1513, 1455, 1431, 1260, 1220, 1182 cm^À1; MS (EI): m/z

(%): 889 (1) [M+H⁺], 207 (36), 129 (55), 104 (100), 76 (58), 73 (86);

HRMS (ESI-TOF): m/z: calcd for C₅₇H₄₉N₂O₈: 889.3483 [M+H⁺];

found: 889.3486.

N-Benzylated azalamellarin D (30): Compound 28 (0.015 g,

0.0175 mmol) was treated with trifluoroacetic acid (TFA; 0.4 mL) in

the presence of thioanisole (0.04 mL) at 608C. After 24 h, the mix-

ture was cooled to room temperature and extracted with EtOAc

(310 mL). The combined organic layer was washed with water

(310 mL) and brine (20 mL), dried with anhydrous sodium sulfate,

and concentrated under reduced pressure. The crude product was

purified by crystallization (MeOH/Et₂O/hexane) to give product 30

(0.008 g, 77%). M.p. 278.6–281.18C (MeOH/Et₂O/hexane); ¹H NMR

(300 MHz, [D₆]DMSO): d=3.27 (s, 3H), 3.32 (s, 3H), 3.71 (s, 3H),

5.52 (br s, 2H), 6.78 (s, 1H), 6.92–7.28 (m, 12H), 9.23 (br s, 1H), 9.41

(d, J=7.5 Hz, 2H), 9.68 ppm (br s, 1H); ¹³C NMR (75 MHz,

[D₆]DMSO): d=44.9, 55.0, 55.2, 56.5, 103.6, 106.0, 107.1, 110.1,

110.4, 111.1, 111.97, 112.00, 115.8, 117.0, 118.3, 123.1, 124.5, 124.6,

126.9, 127.3, 127.4, 129.1, 132.0, 132.5, 138.1, 143.7, 147.1, 147.4,

148.0, 148.5, 149.3, 155.6 ppm; IR (UATR): n˜ =3429, 2926, 1617,

1548, 1490, 1456, 1432, 1257, 1207, 1177 cm^À1; MS (EI): m/z (%):

589 (28) [M+H⁺], 588 (61) [M⁺], 207 (25), 149 (33), 104 (100), 97

(44), 84 (51); HRMS (ESI-TOF): m/z: calcd for C₃₅H₂₉N₂O₇: 589.1969

[M+H⁺]; found: 589.1969.

Similarly, compound 22 (0.0207 g, 90%) was also synthesized from

pyrrole amide 18 (0.0250 g, 0.0257 mmol). M.p. 114.9–116.18C;

¹H NMR (300 MHz, CDCl₃, SiMe₄): d=3.03 (br t, J=6.3 Hz, 2H), 3.28

(s, 3H), 3.32 (s, 3H), 3.76 (s, 3H), 3.87 (s, 3H), 4.89–4.95 (m, 1H),

4.99 (s, 2H), 4.99–5.06 (m, 1H), 5.14 (s, 2H), 5.27 (s, 2H), 5.38 (br s,

2H), 6.71 (s, 1H), 6.75–6.80 (m, 4H), 6.88 (s, 1H), 7.05–7.10 (m, 5H),

7.27–7.48 ppm (m, 15H); ¹³C NMR (75 MHz, CDCl₃, SiMe₄): d=29.0,

42.4, 45.2, 55.18, 55.22, 55.3, 56.2, 70.2, 71.0, 102.7, 105.9, 109.0,

112.5, 113.5, 114.2, 114.75, 114.81, 118.9, 121.3, 123.6, 125.6, 126.5,

127.0, 127.1, 127.2, 127.6, 127.87, 127.94, 128.0, 128.5, 128.6, 128.7,

129.2, 129.9, 130.6, 133.8, 136.7, 136.9, 137.0, 144.9, 146.9, 147.5,

147.6, 148.0, 150.5, 156.0, 158.6 ppm; IR (UATR): n˜ =3032, 2931,

1636, 1512, 1486, 1455, 1441, 1418, 1252, 1213, 1176 cm^À1; MS (EI):

m/z (%): 890 (3) [M⁺], 551 (26), 129 (70), 111 (51), 104 (80), 98 (86),

97 (85), 91 (80), 73 (100); HRMS (ESI-TOF): m/z: calcd for

C₅₇H₅₁N₂O₇: 891.3645 [M+H⁺]; found: 891.3673.

N,O-Arylmethylenated azalamellarin D (28 and 29): Compound 21

(0.0230 g, 0.0267 mmol) was first dissolved in CH₂Cl₂(2 mL). Then,

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2644

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Azalamellarin D (4): Using a procedure similar to that described for

N-benzylated azalamellarin (30), compound 29 (0.250 g,

(m, 2H), 7.10–7.14 (m, 4H), 7.30–7.50 (m, 16H), 9.19 ppm (d, J=

7.3 Hz, 1H); ¹³C NMR (75 MHz, CDCl₃, SiMe₄): d=55.2, 55.6, 56.3,

70.8, 70.90, 70.94, 102.7, 105.4, 105.6, 107.9, 109.5, 110.3, 111.0,

112.4, 114.6, 114.9, 119.3, 123.2, 123.9, 124.6, 125.0, 127.0, 127.2,

127.9, 128.1, 128.7, 129.3, 134.3, 136.2, 136.3, 136.9, 146.0, 146.4,

147.9, 148.5, 149.2, 149.7, 150.6, 155.5 ppm; IR (UATR): n˜ =2928,

D

0.281 mmol) underwent TFA/thioanisole-mediated acidolysis, which

was performed at room temperature for 4–6 h to remove the O-

benzyl groups. Then, the mixture was heated to 608C for 16 h, and

subsequent steps were performed as described for compound 30

to give azalamellarin D (0.120 g, 86%). M.p.>2958C (MeOH/Et₂O/

1697, 1510, 1486, 1453, 1419, 1263, 1215, 1166, 1034, 1010 cm^À1

;

MS (EI): m/z (%): 770 (1) [M+H⁺], 281 (12), 207 (41), 129 (42), 104

(100), 84 (69), 73 (81), 60 (57); HRMS (ESI-TOF): m/z: calcd for

C₄₉H₄₀NO₈: 770.2748 [M+H⁺]; found: 770.2730.

1

hexane, dec.); H NMR (300 MHz, [D₆]DMSO): d=3.34 (s, 3H), 3.37

(s, 3H), 3.76 (s, 3H), 6.80 (s, 1H), 6.89 (s, 1H), 7.00–7.03 (m, 2H),

7.08–7.13 (m, 4H), 9.27 (s, 1H), 9.38 (d, J=7.5 Hz, 1H), 9.49 (s, 1H),

9.72 (s, 1H), 11.27 ppm (s, 1H); ¹³C NMR (75 MHz, [D₆]DMSO): d=

55.0, 55.3, 56.4, 102.7, 106.0, 106.4, 109.0, 110.5, 110.8, 112.0, 112.5,

115.8, 116.9, 118.3, 123.1, 124.4, 127.3, 128.4, 131.8, 143.8, 147.0,

147.5, 147.8, 148.4, 149.1, 155.9 ppm; IR (UATR): n˜ =3296, 2924,

1644, 1625, 1455, 1428, 1274, 1211 cm^À1; HRMS (ESI-TOF): m/z:

calcd for C₂₈H₂₃N₂O₇: 499.1505 [M+H⁺]; found: 499.1480. These

spectroscopic data are similar to those previously reported.^[8]

Lamellarin D (2): Using a procedure similar to that described for

azalamellarin D (4), compound 43 (0.0105 g, 0.0136 mmol) under-

went TFA/thioanisole-mediated O-debenzylation at room tempera-

ture for 4–6 h to furnish lamellarin D (0.0056 g, 82%). M.p.>2908C

1

(MeOH/Et₂O, dec.); H NMR (300 MHz, [D₆]DMSO): d=3.386 (s, 3H),

3.390 (s, 3H), 3.78 (s, 3H), 6.72 (s, 1H), 6.87 (s, 1H), 7.02–7.08 (m,

1H), 7.11–7.21 (m, 5H), 9.00 (d, J=7.5 Hz, 1H), 9.32 (s, 1H), 9.81 (s,

1H), 9.91 ppm (s, 1H); ¹³C NMR (75 MHz, [D₆]DMSO): d=54.5, 55.0,

56.0, 103.7, 105.3, 105.8, 106.4, 108.3, 110.8, 111.5, 112.3, 115.0,

116.4, 117.5, 122.0, 123.8, 124.6, 125.4, 128.9, 134.0, 144.6, 146.3,

146.8, 147.8, 148.3, 148.5, 148.7, 154.3 ppm. Physical and spectro-

scopic data are in agreement with those previously reported.^[34]

Benzylated dihydrolamellarin D (34): Pyrrole ester 6 (0.0305 g,

0.0346 mmol) underwent reaction with sodium amide (0.0067 g,

0.17 mmol) in anhydrous 1,4-dioxane (5 mL). The mixture was

stirred at 1008C for 18 h, and the reaction was then quenched

with water. The mixture was extracted with EtOAc (230 mL). The

combined organic extract was then washed with water (220 mL)

and brine (20 mL), dried with anhydrous sodium sulfate, and con-

centrated under reduced pressure. Crude product 24 was dried in

Synthesis: azalamellarin N (5) and lamellarin N (3)

1

vacuo. H NMR (300 MHz, CDCl₃, SiMe₄): d=2.91–3.03 (m, 2H), 3.22

2-[3-(Benzyloxy)-4-methoxyphenyl]acetic acid (9): Using a procedure

similar to that described for arylacetic acid 8, benzaldehyde 37

(2.50 g, 10.3 mmol) was used to prepare arylacetic acid 9 (2.32 g,

83% over four steps). The physical and spectroscopic data are in

agreement with those previously reported.^[35]

(s, 3H), 3.64 (s, 3H), 3.65 (s, 3H), 4.42 (br s, 1H), 4.83–4.87 (m, 1H),

5.07 (s, 2H), 5.13 (s, 4H), 6.60 (s, 1H), 6.70–6.74 (m, 5H), 7.28–

7.44 ppm (m, 16H); HRMS (ESI-TOF): m/z: calcd for C₄₉H₄₃⁸¹BrNO₈:

854.2158 [M+H⁺]; found: 854.2132; C₄₉H₄₃⁷⁹BrNO₈: 852.2167

[M+H⁺]; found: 852.2159.

2-[3-(Benzyloxy)-4-methoxyphenyl]acetyl chloride (44): Oxalyl chlo-

ride (1.09 mL, 12.7 mmol) and DMF (2 drops) were sequentially

added to a solution of acid 9 (2.32 g, 8.45 mmol) in CH₂Cl₂(20 mL).

Subsequent steps were then performed as described for com-

pound 41, and the product was immediately used in the next step

of amide synthesis without further purification. Owing to its unsta-

ble nature, complete spectroscopic characterization was not per-

formed.

Subsequently, the lactone ring closure of crude product 24 was

performed by using CuTC (0.0158 g, 0.0830 mmol) in DMF under

microwave irradiation (200 W, 1508C, 45 min). Afterwards, a saturat-

ed ammonium chloride solution (3 mL) was added to stop the re-

action, and the mixture was then extracted with EtOAc (320 mL).

The combined organic layer was washed with water (220 mL)

and brine (40 mL), dried with anhydrous sodium sulfate, and con-

centrated under reduced pressure. Finally, the crude product was

purified by P-TLC (30% EtOAc/hexane) to yield product 34

(0.0180 g, 66% over two steps). M.p. 181.0–182.48C; ¹H NMR

(300 MHz, CDCl₃, SiMe₄): d=3.03 (br t, J=6.6 Hz, 2H), 3.28 (s, 3H),

3.37 (s, 3H), 3.39 (s, 3H), 4.62–4.71 (m, 1H), 4.76–4.85 (m, 1H), 5.15

(s, 4H), 5.27 (s, 2H), 6.63 (s, 1H), 6.70 (s, 1H), 6.75 (s, 1H), 6.88 (s,

1H), 7.00–7.10 (m, 3H), 7.29–7.48 ppm (m, 15H); ¹³C NMR (75 MHz,

CDCl₃, SiMe₄): d=28.6, 42.4, 55.2, 55.6, 56.3, 70.87, 70.93, 71.0,

102.7, 104.8, 109.1, 110.7, 113.3, 113.8, 114.4, 114.5, 114.9, 120.5,

123.4, 126.4, 127.0, 127.2, 127.3, 128.0, 128.1, 128.4, 128.68, 128.72,

135.9, 136.3, 136.7, 136.9, 145.8, 146.0, 147.6, 147.7, 148.0, 148.1,

150.4, 155.6 ppm; IR (UATR): n˜ =2924, 2853, 1701, 1606, 1542,

N-[3-(Benzyloxy)-4-methoxyphenethyl]-2-[3-(benzyloxy)-4-methoxy-

phenyl]acetamide (45): A solution of acid chloride 44 (8.45 mmol)

in CH₂Cl₂(30 mL) was added to a solution of amine 10 (6.76 mmol)

in CH₂Cl₂(20 mL); this was followed by the addition of a 10% w/v

aqueous solution (18 mL) of sodium carbonate (1.79 g, 16.9 mmol).

Subsequent steps were then performed as described for com-

pound 42 to provide the desired amide (3.01 g, 87%). M.p. 113.1–

1

114.88C (lit.^[36]113.5–115.08C); H NMR (300 MHz, CDCl₃, SiMe₄): d=

2.58 (t, J=6.9 Hz, 2H), 3.35 (q, J=6.6 Hz, 2H), 3.41 (s, 2H), 3.85 (s,

3H), 3.88 (s, 3H), 5.11 (s, 4H), 5.31 (br s, 1H), 6.55 (d, J=8.1 Hz,

1H), 6.62–6.84 (m, 5H), 7.28–7.46 ppm (m, 10H); ¹³C NMR (75 MHz,

CDCl₃, SiMe₄): d=34.8, 40.5, 43.3, 59.9, 70.8, 70.9, 111.8, 112.0,

114.5, 115.0, 121.2, 122.1, 127.0, 127.3, 127.8, 127.9, 128.48, 128.50,

130.9, 136.8, 137.0, 148.1, 148.2, 148.3, 148.9, 171.1 ppm. HRMS

(ESI-TOF): m/z: calcd for C₃₂H₃₄NO₅: 512.2431 [M+H⁺]; found:

512.2431.

1511, 1484, 1463, 1417, 1269, 1238, 1211, 1166, 1039, 1012 cm^À1

;

MS (EI): m/z (%): 771 (1) [M⁺], 149 (18), 148 (18), 104 (100), 99 (16),

76 (63); HRMS (ESI-TOF): m/z: calcd for C₄₉H₄₂NO₈: 772.2910

[M+H⁺]; found: 772.2899.

Benzylated lamellarin D (43): Compound 34 (0.0120 g,

0.0156 mmol) was oxidized by DDQ (0.0088 g, 0.039 mmol) by

using a procedure similar to that described for compound 28 to

furnish product 43 (0.0117 g, 99%) as a solid. M.p. 213.6–215.88C;

¹H NMR (300 MHz, CDCl₃, SiMe₄): d=3.36 (s, 3H), 3.39 (s, 3H), 3.90

(s, 3H), 5.19 (s, 2H), 5.25 (s, 2H), 5.31 (s, 2H), 6.72 (s, 1H), 6.95–6.96

6-(Benzyloxy)-1-[3-(benzyloxy)-4-methoxybenzyl]-7-methoxy-3,4-di-

hydroisoquinoline (46): Amide 45 (3.01, 5.89 mmol) was dissolved

in anhydrous acetonitrile (30 mL) in a two-necked, round-bot-

tomed flask, to which phosphorus oxychloride (1.61 mL,

17.6 mmol) was added. Subsequent steps were then performed as

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2645

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

described for dihydroisoquinoline 15 to yield key intermediate di-

hydroisoquinoline 46, which was used in the next step without fur-

ther purification.

1212, 1176 cm^À1

;

HRMS (ESI-TOF): m/z: calcd for C₅₇H₅₁N₂O₇:

891.3640 [M+H⁺]; found: 891.3621.

N,O-Arylmethylenated azalamellarin N (47): Compound 36 (0.100 g,

0.112 mmol) was dissolved in CH₂Cl₂(6 mL). Then, DDQ (0.051 g,

0.224 mmol) was added, and the mixture was stirred at room tem-

perature for 4 h. Subsequent steps were then performed as de-

scribed for compound 28 to yield product 47 (0.099 g, 100%).

M.p.>2908C (dec.); ¹H NMR (300 MHz, CDCl₃, SiMe₄): d=3.41 (s,

3H), 3.42 (s, 3H), 3.79 (s, 3H), 4.00 (s, 3H), 5.05 (s, 2H), 5.16 (s, 2H),

5.27 (s, 2H), 5.50 (br s, 2H), 6.81–6.86 (m, 3H), 6.89 (d, J=7.5 Hz,

1H), 7.00 (s, 1H), 7.10–7.18 (m, 4H), 7.21–7.43 (m, 16H), 7.42–7.49

(m, 2H), 9.64 ppm (d, J=7.5 Hz, 1H); ¹³C NMR (75 MHz, CDCl₃,

SiMe₄): d=45.1, 55.1, 55.2, 55.3, 56.5, 70.8, 71.0, 71.1, 102.5, 105.7,

106.6, 109.7, 110.1, 110.9, 112.0, 112.9, 114.2, 117.3, 119.6, 123.8,

124.1, 124.8, 127.0, 127.17, 127.24, 127.5, 127.7, 127.9, 128.0, 128.1,

128.5, 128.6, 128.7, 129.3, 129.6, 131.4, 132.5, 136.3, 136.56, 136.62,

144.9, 147.8, 148.6, 149.1, 149.2, 149.6, 156.0, 158.6 ppm; IR (UATR):

n˜ =2927, 1727, 1638, 1627, 1452, 1431, 1247, 1220, 1179,

1019 cm^À1; MS (EI): m/z (%): 889 (1) [M+H⁺], 888 (2) [M⁺], 368

(23), 208 (71), 195 (100), 129 (42), 97 (43), 73 (70); HRMS (ESI-TOF):

m/z: calcd for C₅₇H₄₉N₂O₈: 889.3483 [M+H⁺]; found: 889.3504.

Pyrrole ester 7: Sodium bicarbonate (0.500 g, 5.89 mmol) and a-ni-

trocinnamate 11 (1.98 g, 4.53 mmol) were added to crude 46 dis-

solved in anhydrous acetonitrile (50 mL) at room temperature. Sub-

sequent steps were then performed as described for compound 6

to give pyrrole ester 7 (2.90 g, 64% over two steps from amide

1

45). M.p. 69.8–70.48C; H NMR (300 MHz, CDCl₃, SiMe₄): d=0.86 (t,

J=6.9 Hz, 3H), 2.89–3.10 (m, 2H), 3.32 (s, 3H), 3.62 (s, 3H), 3.82 (s,

3H), 3.94–4.09 (m, 2H), 4.34–4.42 (m, 1H), 4.84–4.86 (m, 1H), 4.90

(s, 2H), 5.08 (s, 2H), 5.14 (s, 2H), 6.55 (s, 1H), 6.65 (s, 1H), 6.73 (s,

1H), 6.78–6.81 (m, 3H), 7.03 (s, 1H), 7.25–7.44 ppm (m, 15H);

¹³C NMR (75 MHz, CDCl₃, SiMe₄): d=13.7, 28.9, 42.8, 55.2, 56.01,

56.03, 59.7, 70.9, 71.0, 71.2, 109.0, 111.6, 113.3, 115.1, 115.6, 116.4,

117.1, 118.6, 121.36, 121.39, 123.6, 125.6, 127.2, 127.3, 127.4, 127.8,

127.9, 128.0, 128.4, 128.56, 128.59, 130.4, 131.0, 131.6, 136.5,

136.98, 137.01, 147.2, 147.3, 147.9, 148.1, 148.4, 161.6 ppm; IR

(UATR): n˜ =2925, 1687, 1455, 1402, 1382, 1248, 1176, 1024 cm^À1

;

MS (EI): m/z (%): 880 (1) [M+H⁺], 368 (51), 236 (52), 149 (41), 121

(86), 97 (91), 83 (83), 71 (86), 57 (100); HRMS (ESI-TOF): m/z: calcd

for C₅₁H₄₇⁸¹BrNO₈: 882.2472 [M+H⁺]; found: 882.2441;

C₅₁H₄₇⁷⁹BrNO₈: 880.2479 [M+H⁺]; found: 880.2445. These spectro-

scopic data are similar to those previously reported.^[8]

Azalamellarin N (5): Using a procedure similar to that described for

azalamellarin D (4), TFA/thioanisole-mediated acidolysis of com-

pound 47 (0.090 g, 0.101 mmol) gave azalamellarin N (0.041 g,

83%). M.p.>2908C (MeOH/Et₂O/hexane, dec.); ¹H NMR (300 MHz,

[D₆]DMSO): d=3.36 (s, 3H), 3.39 (s, 3H), 3.87 (s, 3H), 6.84 (s, 1H),

6.89 (s, 1H), 6.99–7.03 (m, 3H), 7.14 (d, J=4.8 Hz, 2H), 7.24 (d, J=

8.7 Hz, 1H), 9.37 (d, J=7.5 Hz, 2H), 9.53 (br s, 1H), 9.76 (br s, 1H),

11.29 ppm (s, 1H); ¹³C NMR (75 MHz, [D₆]DMSO): d=55.0, 55.3,

56.6, 102.8, 105.9, 106.3, 108.9, 110.2, 110.9, 112.1, 112.5, 114.2,

118.2, 119.0, 122.7, 123.1, 124.4, 128.1, 129.3, 131.6, 131.8, 143.8,

147.5, 147.9, 148.2, 148.5, 155.9 ppm; IR (UATR): n˜ =3124, 2924,

1643, 1625, 1491, 1427, 1273, 1212 cm^À1; MS (EI): m/z (%): 498 (24)

[M⁺], 207 (36), 129 (46), 97 (63), 84 (100), 69 (72), 57 (90); HRMS

(ESI-TOF): m/z: calcd for C₂₈H₂₃N₂O₇: 499.1505 [M+H⁺]; found:

499.1480.

Pyrrole amide 33: Trimethylaluminum (1.0m in toluene, 0.19 mL,

0.19 mmol) was added to a mixture of pyrrole ester 7 (0.0705 g,

0.0800 mmol) and p-methoxybenzylamine (0.052 mL, 0.40 mmol) in

anhydrous 1,4-dioxane (1 mL). Subsequently, microwave irradiation

(200 W, 1658C, 65 min) was used to convert the pyrrole ester into

the corresponding pyrrole amide. Subsequent steps were then per-

formed as described for compound 17 to give pyrrole amide 33

1

(0.077 g, 99%). M.p. 88.5–89.48C; H NMR (300 MHz, CDCl₃, SiMe₄):

d=2.91–3.04 (m, 2H), 3.31 (s, 3H), 3.58 (s, 3H), 3.70 (s, 3H), 3.81 (s,

3H), 4.16–4.22 (m, 1H), 4.32–4.44 (m, 2H), 4.85–4.95 (m, 1H), 4.89

(s, 2H), 4.95 (s, 2H), 5.13 (s, 2H), 5.71 (br t, J=5.4 Hz, 1H), 6.57 (s,

1H), 6.62 (s, 1H), 6.72–6.80 (m, 6H), 6.87 (s, 1H), 6.90 (s, 2H), 7.25–

7.45 ppm (m, 15H); ¹³C NMR (75 MHz, CDCl₃, SiMe₄): d=29.0, 31.7,

42.9, 55.1, 55.2, 55.92, 55.94, 70.8, 70.99, 71.03, 108.8, 111.5, 113.3,

113.7, 115.3, 115.7, 116.2, 117.1, 120.7, 121.5, 121.7, 123.5, 124.2,

125.5, 126.6, 127.2, 127.4, 127.7, 127.8, 127.9, 128.1, 128.4, 128.5,

128.6, 128.8, 129.3, 129.9, 136.1, 136.9, 137.0, 146.9, 147.8, 148.29,

148.34, 148.8, 158.7, 161.4 ppm; IR (UATR): n˜ =3417, 2931, 1643,

1510, 1494, 1246, 1210, 1174, 1024 cm^À1; MS (EI): m/z (%): 971 (1)

[M+H⁺], 970 (2) [M⁺], 163 (21), 121 (100), 91 (92); HRMS (ESI-TOF):

m/z: calcd for C₅₇H₅₂⁸¹BrN₂O₃: 973.2896 [M+H⁺]; found: 973.2818;

C₅₇H₅₂⁷⁹BrN₂O₃: 971.2902 [M+H⁺]; found: 971.2859.

Benzylated dihydrolamellarin N (35): Pyrrole ester 7 (0.0380 g,

0.0431 mmol) underwent reaction with sodium amide (0.017 g,

0.431 mmol) in anhydrous 1,4-dioxane (10 mL) at 1008C for 24 h.

Subsequent steps were then performed as described for com-

pound 34 to afford compound 32 as an intermediate. ¹H NMR

(300 MHz, CDCl₃, SiMe₄): d=2.91–3.02 (m, 2H), 3.31 (s, 3H), 3.64 (s,

3H), 3.82 (s, 3H), 4.44 (br s, 1H), 4.79–4.89 (m, 1H), 4.84 (s, 2H),

5.03 (s, 2H), 5.14 (s, 2H), 6.61 (s, 1H), 6.64 (s, 1H), 6.77–6.79 (m,

4H), 7.05 (s, 1H), 7.26–7.44 ppm (m, 16H); ¹³C NMR (75 MHz, CDCl₃,

SiMe₄): d=28.9, 43.0, 55.2, 56.0, 70.9, 71.0, 71.2, 109.2, 111.5, 113.1,

115.1, 115.4, 116.3, 117.0, 121.1, 122.0, 123.6, 125.9, 127.29, 127.32,

127.5, 127.6, 127.8, 128.0, 128.1, 128.4, 128.6, 129.6, 132.3, 133.2,

136.5, 136.9, 147.5, 147.7, 147.9, 148.2, 148.4, 164.9, 165.5 ppm;

HRMS (ESI-TOF): m/z: calcd for C₄₉H₄₂⁸¹BrNO₈+Na⁺: 876.1977

[M+Na⁺]; found: 876.1992; C₄₉H₄₂⁷⁹BrNO₈+Na⁺: 874.1986 [M+Na⁺

]; found: 874.1994.

N,O-Arylmethylenated dihydroazalamellarin N (36): Pyrrole amide

33 (0.170 g, 0.175 mmol) was dissolved in DMF (2 mL). CuTC

(0.067 g, 0.350 mmol) and cesium carbonate (0.119 g, 0.350 mmol)

were then added. Subsequent steps were performed as described

for compound 21 to give product 36 (0.137 g, 80%). M.p. 134.6–

1

136.18C; H NMR (300 MHz, CDCl₃, SiMe₄): d=3.02 (br t, J=6.5 Hz,

2H), 3.31 (s, 3H), 3.37 (s, 3H), 3.75 (s, 3H), 3.93 (s, 3H), 4.90–5.04

(m, 4H), 5.12 (s, 2H), 5.14 (s, 2H), 5.16 (s, 2H), 5.40 (br s, 2H), 6.66

(s, 1H), 6.75–6.80 (m, 4H), 6.89 (s, 1H), 7.06–7.41 ppm (m, 20H);

¹³C NMR (75 MHz, CDCl₃, SiMe₄): d=29.0, 42.5, 45.2, 55.1, 55.3, 56.5,

70.0, 71.1, 102.7, 105.9, 109.0, 112.6, 112.8, 113.4, 114.16, 114.20,

116.9, 118.8, 121.2, 124.3, 125.7, 126.5, 127.1, 127.2, 127.4, 127.6,

127.9, 128.0, 128.5, 128.57, 128.64, 129.3, 129.4, 130.6, 133.9, 136.5,

136.7, 137.0, 145.0, 146.9, 147.6, 148.0, 149.0, 149.3, 156.0,

158.6 ppm; IR (UATR): n˜ =2931, 1635, 1512, 1485, 1454, 1418, 1250,

Subsequently, the lactone ring closure of 32 was performed by

using a procedure similar to that described for 34 to yield product

35 (0.017 g, 51% over two steps). M.p. 87.9–90.18C; ¹H NMR

(300 MHz, CDCl₃, SiMe₄): d=3.01 (br t, J=6.6 Hz, 2H), 3.34 (s, 3H),

3.48 (s, 3H), 3.98 (s, 3H), 4.71 (br t, 2H), 5.00 (s, 2H), 5.13 (s, 2H),

5.25 (s, 2H), 6.65 (d, J=5.1 Hz, 2H), 6.74 (d, J=9.3 Hz, 2H), 7.11 (s,

2H), 7.20–7.45 ppm (m, 16H); ¹³C NMR (75 MHz, CDCl₃, SiMe₄): d=

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2646

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

28.5, 42.2, 55.2, 55.5, 56.4, 70.6, 70.8, 102.3, 104.8, 109.0, 110.6,

112.6, 113.2, 113.6, 114.7, 116.6, 120.4, 124.0, 126.4, 127.1, 127.2,

127.4, 127.9, 127.96, 128.02, 128.5, 128.59, 128.65, 135.8, 136.3,

136.6, 136.7, 145.7, 145.9, 147.5, 148.0, 148.1, 148.8, 149.5,

155.5 ppm; IR (UATR): n˜ =2925, 1698, 1509, 1454, 1259, 1214,

1168 cm^À1; MS (EI): m/z (%): 772 (1) [M+H⁺], 149 (37), 148 (22), 104

(100), 98 (70), 73 (94); HRMS (ESI-TOF): m/z: calcd for C₄₉H₄₂NO₈:

772.2905 [M+H⁺]; found: 772.2896.

corrected and then compared with the column dead time (t₀, de-

termined from the retention time of unretained formamide) to

give the retention factor (k), the logarithm of which was then plot-

ted against the known log P value to construct a calibration curve.

Subsequently, the 10 mm stock solutions of lamellarins and azala-

mellarins were freshly diluted with 3 volumes of MeOH and

1 volume of deionized water. Their retention factors were then de-

termined as described for the reference compounds. Finally, their

log P values were assessed from the calibration curve constructed

on the same day.

Benzylated lamellarin N (48): Compound 35 (0.076 g, 0.099 mmol)

was oxidized by DDQ (0.045 g, 0.198 mmol) by using a procedure

similar to that described for compound 28 to furnish product 48

as a solid (0.074 g, 97%). M.p. 203.4–206.38C; ¹H NMR (300 MHz,

CDCl₃, SiMe₄): d=3.42 (s, 3H), 3.46 (s, 3H), 4.01 (s, 3H), 5.13 (s, 2H),

5.20 (s, 2H), 5.24 (s, 2H), 6.69 (s, 1H), 6.89 (s, 1H), 6.96 (d, J=

7.5 Hz, 1H), 7.09 (d, J=9.0 Hz, 2H), 7.19–7.48 (m, 18H), 9.16 ppm

(d, J=7.2 Hz, 1H); ¹³C NMR (75 MHz, CDCl₃, SiMe₄): d=55.2, 55.5,

56.5, 70.7, 70.9, 71.0, 102.6, 105.4, 105.5, 109.5, 110.2, 111.0, 112.3,

112.7, 117.1, 119.3, 123.1, 124.5, 124.6, 127.18, 127.22, 127.5,

128.07, 128.12, 128.2, 128.5, 128.67, 128.70, 129.3, 134.3, 136.28,

136.32, 146.0, 146.4, 148.5, 149.0, 149.2, 149.6, 149.8, 155.4 ppm; IR

(UATR): n˜ =2924, 1698, 1430, 1421, 1264, 1223, 1167 cm^À1; MS (EI):

m/z (%): 770 (1) [M+H⁺], 207 (95), 111 (39), 105 (45), 104 (56), 98

(99), 91 (100), 83 (79), 69 (77); HRMS (ESI-TOF): m/z: calcd for

C₄₉H₄₀NO₈: 770.2748 [M+H⁺]; found: 770.2730.

Solubility measurements

Freshly prepared HEPES buffer (20 mm, pH 7.4) served as the

medium for determining the aqueous solubility of lamellarins and

azalamellarins by using the traditional shake-flask method coupled

with HPLC analysis, as previously optimized for these compounds

(unpublished results). Briefly, each sample for thermodynamic solu-

bility measurements was prepared by adding the solid material

(1 mg) into buffer (40 mL) in a 100 mL Erlenmeyer flask, which was

then loaded into a shaking incubator operated at 378C and

250 rpm for up to 7 days. In contrast, kinetic solubility samples

were prepared by adding 10 mm stock solutions (100 mL) of our

test compounds into vials containing HEPES buffer (1900 mL) and

were then incubated under the above-mentioned conditions for

up to 24 h. All measurements were performed at least in triplicate.

Lamellarin N (3): Using a procedure similar to that described for

azalamellarin D (4), TFA/thioanisole-mediated O-debenzylation of

compound 48 (0.0656 g, 0.0853 mmol) at room temperature for 4–

6 h gave lamellarin N (0.030 g, 70%). M.p. 312.4–314.68C (MeOH/

At predetermined time points during the specified incubation

period, an aliquot (500 mL) was taken from each sample without

buffer replacement and was transferred into a microcentrifuge

tube for centrifugation at 258C and 18000 rpm for 5 min. Subse-

quently, clear supernatant (198 mL) was carefully withdrawn from

each sample into an HPLC insert containing DMSO (2 mL) to pre-

vent precipitation. Lastly, HPLC analysis was then conducted on

both the samples and the serially diluted standard solutions to de-

termine the concentrations of the test compounds that either dis-

solved or remained in HEPES buffer, which represented their ther-

modynamic or kinetic solubility, respectively.

1

Et₂O); H NMR (300 MHz, [D₆]DMSO): d=3.39 (s, 3H), 3.40 (s, 3H),

3.87 (s, 3H), 6.76 (s, 1H), 6.87 (s, 1H), 7.01–7.04 (m, 2H), 7.18–7.26

(m, 4H), 9.01 (d, J=7.2 Hz, 1H), 9.38 (br s, 1H), 9.83 (br s, 1H),

9.94 ppm (br s, 1H); ¹³C NMR (75 MHz, [D₆]DMSO): d=55.0, 55.5,

56.1, 104.2, 105.8, 106.2, 107.0, 108.7, 111.0, 112.0, 112.9, 114.2,

117.9, 118.7, 122.5, 112.6, 125.1, 127.9, 129.2, 134.3, 145.0, 146.8,

148.2, 148.3, 148.4, 148.8, 149.0, 154.8 ppm; HRMS (ESI-TOF): m/z:

calcd for C₂₈H₂₂NO₈: 500.1340 [M+H⁺]; found: 500.1349. Physical

and spectroscopic data are similar to those previously repor-

ted.^[1j,37]

Biological evaluations: general

Each test compound or positive control was dissolved in DMSO at

20 mm for cytotoxicity assay and 10 mm for GSK-3b inhibition

assay. The stock solutions for cytotoxicity assay were filter-sterilized

before use, whereas the remaining solutions were aliquoted and

stored at À208C for replicate experiments within the same week.

The sources of cell lines and the culture media used for their prop-

agation are described in the Supporting Information (Table S3). All

reagents for cell culture, as well as those for 3-(4,5-dimethylthiazol-

2-yl)-2,5-diphenyltetrazolium bromide (MTT) and 2,3-bis-(2-me-

thoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide (XTT)

assays, were obtained from commercial sources and were used as

received. The ADP-Glo Kinase Assay Kits, including the GSK-3b

Kinase Enzyme Systems, were supplied by Promega Corporation

(Madison, WI, USA).

Physicochemical characterizations: general

All chemicals were obtained from commercial sources with at least

98% purity. Unless otherwise mentioned, the test compounds

were prepared as 10 mm stock solutions in DMSO and were further

diluted as described for each experiment. The HPLC analyses men-

tioned below were performed by using isocratic elution with

MeOH/water (75:25 v/v) for 15 min through a ZORBAX Eclipse Plus

C18 column (5 mm, 4.6 mm i.d. 250 mm). UV detection was per-

formed at l=210, 254, and 276 nm for formamide, reference com-

pounds, and lamellarins, respectively.

Lipophilicity determinations

Experimental log P values of the lamellarins and azalamellarins

were determined by using reverse-phase HPLC, as previously es-

tablished in our laboratory.^[16]Seven chemicals with known log P

values, including benzyl alcohol, benzonitrile, methyl benzoate,

benzophenone, naphthalene, benzyl benzoate, and diphenyl ether,

were selected off the list of recommended reference substances^[38]

Cytotoxicity assays

Throughout the assay, all cell lines were maintained by using stan-

dard procedures at 378C with 95% humidity and 5% CO₂. For ad-

herent cells (i.e., A549, HuCCA-1, HepG2, and MRC-5), the MTT

assay was performed to determine the number of cells that sur-

vived the cytotoxicity of each test compound, as previously repor-

ted.^[3c]Briefly, 5000–10000 cells suspended in the corresponding

and were prepared as

a reference mixture in MeOH/water

(75:25 v/v) for daily system calibration. By using a standard equa-

tion, the retention time (t_R) of each reference compound was first

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2647

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

media (100 mL) were first inoculated into each well of 96-well mi-

crotiter plates and were then incubated for 24 h. Afterwards, the

attached cells were treated with an equal volume (100 mL per well)

of the cell culture medium containing either each test compound

or the positive control (doxorubicin) at various concentrations,

except for the negative control cells, which were treated with

0.25% v/v DMSO. After 48 h of exposure, the cell viability was de-

termined by adding the MTT reagent [0.5 mgmL^À1in fetal bovine

serum (FBS)-free medium, 100 mL per well] into each well, followed

by 2.5–4 h of incubation. Then, the medium was replaced with an

equal volume of DMSO to dissolve the insoluble purple formazan.

Finally, the absorbance at l=550 nm was measured with a refer-

ence wavelength of l=650 nm by using a microplate reader.

ATP, which was then used in the luciferase/luciferin reaction. After

30 min, the luminescence generated was measured by using

a SpectraMax L luminescence microplate reader (Molecular Devices,

Sunnyvale, CA, USA) and was compared with that obtained in the

absence of inhibitor. The percent inhibition averaged from dupli-

cate wells was then plotted against the concentration of the test

compound to determine the IC₅₀value by using nonlinear regres-

sion analysis. At least three independent experiments were con-

ducted to get the mean IC₅₀value of each test compound, as

shown in Table 4.

Molecular-docking studies

Three-dimensional structures of lamellarin D (2), lamellarin N (3),

azalamellarin D (4), and azalamellarin N (5) were constructed and

optimized by using the AM1 method implemented in the Gaussian

03 package.^[39]The crystal structures of human GSK-3b were ob-

tained from the Protein Data Bank (PDB ID: 1Q3D^[26]and 3L1S^[27]).

Docking analysis was performed by using the AutoDock 4.0 pro-

gram.^[40]All ligands (i.e., compounds 2–5) and the protein were

prepared by using AutoDockTools 1.5.6.^[41]Briefly, the Gastiger

charges on all ligands were calculated, and all nonpolar hydrogen

atoms were merged. For GSK-3b, hydrogen atoms were added to

the structure, and the Gastiger charges were computed. All water

molecules were removed from the crystal structure of GSK-3b. The

nonpolar hydrogen atoms were merged, and the grid box was as-

signed to sufficiently include the ATP-binding site of the GSK-3b

with spacing of 0.375 , whereas the number of points was set to

be 60 in the x, y, and z axes. The search parameter used was Ge-

netic Algorithm (GA) with 100 runs, and the population size was

set to 150. The best-docked conformation for each compound was

visualized and analyzed by using the Discovery Studio Visualizer

(Accelrys, San Diego, CA, USA).

On the other hand, the suspended MOLT-3 cell line was subjected

to the XTT assay as follows. MOLT-3 cells in RPMI 1640 medium

(50 mL) were seeded at a density of 20000 cells per well and al-

lowed to acclimatize for 30 min. Subsequently, the cells were treat-

ed with an equal volume (50 mL per well) of the cell culture

medium containing serial dilutions of each test compound or the

positive control (etoposide), whereas the negative control cells

were exposed to 0.05% v/v DMSO. Following an additional 44 h in-

cubation, the XTT reagent (50 mL, prepared by mixing 5 mL of

1 mgmL^À1XTT sodium in sterile water with 100 mL of

0.383 mgmL^À1phenazine methosulfate in water) was added to

each well, and the cells were then incubated for 4 h. As the orange

formazan formed is water soluble, the final solubilization step

could be avoided, which thus allowed the absorbance at l=

492 nm to be readily measured with a reference wavelength of l=

690 nm.

Data analysis was then performed by determining the percentage

of surviving cells after 48 h exposure to each concentration of the

test compounds or the positive controls. The background-subtract-

ed absorbances of treated wells were averaged and then com-

pared with those of the negative control wells. The dose–response

curve was then constructed and regressed by using the SigmaPlot

software (Systat Software, Inc., San Jose, CA, USA) to obtain the

IC₅₀value. The results shown in Table 4 are the mean values deter-

mined from at least three independent experiments for each com-

pound.

Acknowledgements

This work was financially supported by the Thailand Research

Fund (TRF; DBG5480011 to M.C., BRG5580010 to P.P.) and the

Scientist Development Project Commemorating His Majesty

the King’s 84^thBirthday Anniversary (to A.T.). Insightful informa-

tion on the previous synthetic route of azalamellarins from Dr.

Nopporn Thasana (Laboratory of Medicinal Chemistry, Chula-

bhorn Research Institute) is gratefully acknowledged. The au-

thors are thankful to Dr. Vilailak Prachyawarakorn (Laboratory

of Natural Products, Chulabhorn Research Institute) for her

kind suggestions for semipreparative HPLC purification. Special

thanks go to Mr. Nitirat Chimnoi (Laboratory of Natural Prod-

ucts, Chulabhorn Research Institute) and Dr. Amornmart Jarat-

rungtawee and Dr. Jaran Jainhuknan (Bruker Biospin AG, Bang-

kok, Thailand) for LC–MS analysis. The authors are also grateful

for generous advice from Dr. Kriengsak Lirdprapamongkol and

Dr. Lukana Ngiwsara (Laboratory of Biochemistry, Chulabhorn

Research Institute) during GSK-3b assay optimization. In addi-

tion, technical guidance on cell culture protocols and cytotox-

icity assays from Ms. Pakamas Intachote (Laboratory of Immu-

nology, Chulabhorn Research Institute), together with Ms. Su-

chada Sengsai and Ms. Busakorn Saimanee (Integrated Re-

search Unit, Chulabhorn Research Institute) is greatly appreci-

ated. Finally, the authors would like to thank Ms. Wandee

Sirapat and Ms. Suchonnama Puttakan (Office of Academic Af-

GSK-3b inhibition assays

According to the ADP-Glo Kinase Assay Protocol, the assay was

performed at room temperature in 1X buffer containing 40 mm

Tris (pH 7.50), 20 mm MgCl₂, and 0.1 mgmL^À1bovine serum albu-

min (BSA). The optimized assay conditions were as follows. The

GSK-3b enzyme (20 ng in 10 mL buffer) was first exposed for

10 min to each test compound or the positive control (i.e.,

SB 415286), all of which had been prepared at predetermined con-

centrations in buffer (5 mL). Likewise, the enzyme exposed to the

DMSO-containing buffer (1% v/v, 5 mL) served as the negative con-

trol. Subsequently, the ATP/substrate mixture (10 mL) was added

into each well to initiate the kinase-mediated phosphorylation of

a primed peptide substrate derived from glycogen synthase [i.e.,

YRRAAVPPSPSLSRHSSPHQ(pS)EDEEE, in which pS represents phos-

phorylated serine]. Following 60 min incubation, the ADP-Glo Re-

agent (25 mL per well) was added to the mixture and was allowed

to react for an additional 40 min to simultaneously terminate the

kinase reaction and to remove all the remaining ATP. Finally, addi-

tion of the Kinase Detection Reagent (50 mL per well) mediated the

conversion of the ADP product from the kinase reaction back to

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2648

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

dergo Cu^I-mediated, MW-assisted intramolecular CÀO bond formation

rather than decarboxylation. For detailed discussion, see the Supporting

Information.

fairs, Chulabhorn Research Institute) for their kind assistance

with the graphic design.

[15] a) J. L. F. Puentes, D. Garcia Gravalos, A. Rodriguez Quesada (PharmaMar,

S. A.), World Patent WO/1997/001336, 1997; b) A. Rodriguez Quesada,

D. Garcia Gravalos, J. L. F. Puentes (PharmaMar, S. A.), Canadian Patent

CA 2225807, 1997; c) J. L. F. Puentes, D. Garcia Gravalos, A. Rodriguez

Quesada (PharmaMar, S. A.), U. S. Patent 5,852,033, 1998; d) J. L. F.

Puentes, D. Garcia Gravalos, A. Rodriguez Quesada (PharmaMar, S. A.),

U. S. Patent 6,087,370, 2000.

Keywords: cytotoxicity

lamellarins · total synthesis

·

drug likeness

·

inhibitors

·

[1] a) R. J. Andersen, D. J. Faulkner, C.-H. He, G. D. Van Duyne, J. Clardy, J.

Am. Chem. Soc. 1985 , 107 , 549 2 –5 495; b) N. Lindquist, W. Fenical, G. D.

Van Duyne, J. Clardy, J. Org. Chem. 1988 , 53 , 457 0 –4 574; c) A. R. Carroll,

B. F. Bowden, J. C. Coll, Aust. J. Chem. 1993 , 46 , 48 9 –5 01; d) S. Urban,

M. S. Butler, R. J. Capon, Aust. J. Chem. 1994 , 47 , 191 9 –1 924; e) S.

Urban, L. Hobbs, J. N. A. Hooper, R. J. Capon, Aust. J. Chem. 1995 , 48 ,

149 1 –1 494; f) S. Urban, R. J. Capon, Aust. J. Chem. 1996, 49, 711 –713;

g) M. V. R. Reddy, D. J. Faulkner, Y. Venkateswarlu, M. R. Rao, Tetrahedron

1997 , 53 , 345 7 –3 466; h) R. A. Davis, A. R. Carroll, G. K. Pierens, R. J.

Quinn, J. Nat. Prod. 1999 , 62 , 41 9 –4 24; i) J. Ham, H. Kang, Bull. Korean

Chem. Soc. 2002, 23, 163–166; j) P. Krishnaiah, V. L. N. Reddy, G. Venka-

taramana, K. Ravinder, M. Srinivasulu, T. V. Raju, K. Ravikumar, D. Chan-

drasekar, S. Ramakrishna, Y. Venkateswarlu, J. Nat. Prod. 2004 , 67 , 116 8 –

1171; k) S. Malla Reddy, M. Srinivasulu, N. Satyanarayana, A. K. Kondapi,

Y. Venkateswarlu, Tetrahedron 2005 , 61 , 924 2 –9 247.

[16] M. Chittchang, M. P. Gleeson, P. Ploypradith, S. Ruchirawat, Eur. J. Med.

Chem. 2010 , 45 , 216 5 –2 172.

[17] a) C. Tardy, M. FacomprØ, W. Laine, B. Baldeyrou, D. García-Gravalos, A.

Francesch, C. Mateo, A. Pastor, J. A. JimØnez, I. Manzanares, C. Cuevas,

C. Bailly, Bioorg. Med. Chem. 2004 , 12 , 169 7 –1 71 2; b) D. Pla, A. Fran-

cesch, P. Calvo, C. Cuevas, R. AliguØ, F. Albericio, M. lvarez, Bioconju-

gate Chem. 2009 , 20 , 110 0 –1 11 1 ; c) D. Pla, M. Martí, J. Farrera-Sinfreu, D.

Pulido, A. Francesch, P. Calvo, C. Cuevas, M. Royo, R. AliguØ, F. Albericio,

M. lvarez, Bioconjugate Chem. 2009 , 20 , 111 2 –1 121.

[18] G. L. Patrick in An Introduction to Medicinal Chemistry, 3rd ed., Oxford

University Press, Oxford, 2005, pp. 8–23.

[19] Following 48 h of exposure, the number of cancer cells that survived

the toxicity of our test compounds was determined indirectly from re-

duction of the tetrazolium-based MTT and XTT reagents by the mito-

chondrial dehydrogenase enzyme in the adherent and suspended cells,

respectively, that remained metabolically active. On the basis of nonlin-

ear regression analysis of the dose–response curves, the half-maximal

inhibitory concentration (IC₅₀) of each test compound was then calcu-

lated as the concentration that inhibited cell growth by 50% of the

maximum response relative to untreated cells.

[2] D. Pla, F. Albericio, M. lvarez, Anticancer Agents Med. Chem. 2008, 8,

746–760 and the references cited therein.

[3] a) H. Fan, J. Peng, M. T. Hamann, J.-F. Hu, Chem. Rev. 2008 , 108 , 26 4 –

287 and the references cited therein; b) M. Chittchang, P. Batsomboon,

S. Ruchirawat, P. Ploypradith, ChemMedChem 2009 , 4 , 45 7 –4 65; c) K.

Tangdenpaisal, R. Worayuthakarn, S. Karnkla, P. Ploypradith, P. Intachote,

S. Sengsai, B. Saimanee, S. Ruchirawat, M. Chittchang, Chem. Asian J.

2015 , 10 , 92 5 –9 37.

[20] a) J. P. Jacobs, Nature 1966 , 210 , 10 0 –1 01; b) J. P. Jacobs, C. M. Jones,

J. P. Baille, Nature 1970 , 227 , 16 8 –1 70.

[4] M. FacomprØ, C. Tardy, C. Bal-Mahieu, P. Colson, C. Perez, I. Manzanares,

C. Cuevas, C. Bailly, Cancer Res. 2003, 63, 7392–7399.

[21] a) L. Meijer, M. Flajolet, P. Greengard, Trends Pharmacol. Sci. 2004 , 25 ,

47 1 –4 80; b) J. Luo, Cancer Lett. 2009 , 273 , 19 4 –2 00; c) H. Eldar-Finkel-

man, A. Martinez, Front. Mol. Neurosci. 2011, 4, 32; d) T. Kramer, B.

Schmidt, F. Lo Monte, Int. J. Alzheimer’s Dis. 2012, 2012, 381029.

[22] K. Yoshida, R. Itoyama, M. Yamahira, J. Tanaka, N. Loac, O. Lozach, E.

Durieu, T. Fukuda, F. Ishibashi, L. Meijer, M. Iwao, J. Med. Chem. 2013 ,

56 , 728 9 –7 301.

[5] J. Kluza, M.-A. Gallego, A. Loyens, J.-C. Beauvillain, J.-M. F. Sousa-Faro, C.

Cuevas, P. Marchetti, C. Bailly, Cancer Res. 2006 , 66 , 317 7 –3 187.

[6] D. Baunbæk, N. Trinkler, Y. Ferandin, O. Lozach, P. Ploypradith, S. Ruchi-

rawat, F. Ishibashi, M. Iwao, L. Meijer, Mar. Drugs 2008 , 6 , 51 4 –5 27.

[7] C. Ballot, A. Martoriati, M. Jendoubi, S. Buche, P. Formstecher, L. Mortier,

J. Kluza, P. Marchetti, Mar. Drugs 2014 , 12 , 77 9 –7 98.

[8] S. Boonya-udtayan, N. Yotapan, C. Woo, C. J. Bruns, S. Ruchirawat, N.

Thasana, Chem. Asian J. 2010 , 5 , 211 3 –2 123.

[9] P. Ploypradith, T. Petchmanee, P. Sahakitpichan, N. D. Litvinas, S. Ruchira-

wat, J. Org. Chem. 2006 , 71 , 944 0 –9 448.

[10] a) H. Zhang, Q. Cai, D. Ma, J. Org. Chem. 2005 , 70 , 516 4 –5 173; b) J.

Yuen, Y.-Q. Fang, M. Lautens, Org. Lett. 2006 , 8 , 65 3 –6 56; c) R. Bernini,

S. Cacchi, G. Fabrizi, A. Sferrazza, Synthesis 2008, 729–738; d) S. Kumar,

H. Ila, H. Junjappa, J. Org. Chem. 2009 , 74 , 704 6 –7 051.

[11] a) A. Basha, M. Lipton, S. M. Weinreb, Tetrahedron Lett. 1977 , 18 , 417 1 –

417 2; b) M. F. Lipton, A. Basha, S. M. Weinreb, Org. Synth. 1979, 59, 49–

53; c) J. I. Levin, E. Turos, S. M. Weinreb, Synth. Commun. 1982 , 12 , 98 9 –

993; d) F.-Z. Zradni, J. Hamelin, A. Derdour, Synth. Commun. 2002 , 32 ,

352 5 –3 531.

[12] a) R. Dçlling, P. Jeschke, A. Otto, J. Eichler, Collect. Czech. Chem.

Commun. 1990 , 55 , 179 2 –1 800; b) G. N. Jubilut, E. M. Cilli, M. Tominaga,

A. Miranda, Y. Okada, C. R. Nakaie, Chem. Pharm. Bull. 2001 , 49 , 108 9 –

1092.

[23] D. Kitagawa, K. Yokota, M. Gouda, Y. Narumi, H. Ohmoto, E. Nishiwaki, K.

Akita, Y. Kirii, Genes Cells 2013 , 18 , 11 0 –1 22.

[24] a) “Assay Development for Protein Kinase Enzymes”, J. F. Glickman in

Assay Guidance Manual [Internet] (Eds.: G. S. Sittampalam, N. Gal-Edd,

M. Arkin, D. Auld, C. Austin, B. Bejcek, M. Glicksman, J. Inglese, V.

Lemmon, Z. Li, J. McGee, O. McManus, L. Minor, A. Napper, T. Riss, O. J.

Trask, J. Weidner), Bethesda (MD), 2012. Available from: htt p ://

www.ncbi.nlm.nih.gov/pubmed/2255386 3 ; b) ”Mechanism of Action

Assays for Enzymes”, J. Strelow, W. Dewe, P. W. Iversen, H. B. Brooks,

J. A. Radding, J. McGee, J. Weidner in Assay Guidance Manual [Internet]

(Eds.: G. S. Sittampalam, N. Gal-Edd, M. Arkin, D. Auld, C. Austin, B.

Bejcek, M. Glicksman, J. Inglese, V. Lemmon, Z. Li, J. McGee, O. McMa-

nus, L. Minor, A. Napper, T. Riss, O. J. Trask, J. Weidner), Bethesda (MD),

2012. Available from: http://www.ncbi.nlm.nih.gov/pubmed/22553872.

[25] A. Baki, A. Bielik, L. Molnµr, G. Szendrei, G. M. Keserü, Assay Drug Dev.

Technol. 2007 , 5 , 7 5 –8 3.

[26] J. A. Bertrand, S. Thieffine, A. Vulpetti, C. Cristiani, B. Valsasina, S. Knapp,

H. M. Kalisz, M. Flocco, J. Mol. Biol. 2003 , 333 , 39 3 –4 07.

[27] M. Arnost, A. Pierce, E. ter Haar, D. Lauffer, J. Madden, K. Tanner, J.

Green, Bioorg. Med. Chem. Lett. 2010 , 20 , 166 1 –1 664.

[28] P. Ploypradith, P. Cheryklin, N. Niyomtham, D. R. Bertoni, S. Ruchirawat,

Org. Lett. 2007 , 9 , 263 7 –2 640.

[13] Despite the use of thioanisole as a benzyl cation scavenger, the ob-

tained crystallized azalamellarin products were contaminated with

some C-monobenzylated side products (ꢀ5–15%) as evidenced by the

peak at m/z=589.1962–589.1995 [M+H⁺] from LC–MS analysis. There-

fore, it was still necessary to use semipreparative HPLC for their final

purification.

[29] V. Lisowski, S. LØonce, L. Kraus-Berthier, J. Sopkovµ-de Oliveira Santos, A.

PierrØ, G. Atassi, D.-H. Caignard, P. Renard, S. Rault, J. Med. Chem. 2004 ,

47 , 144 8 –1 464.

[30] a) S. Kobayashi, K. Okimoto, Y. Imakura, Chem. Pharm. Bull. 1982 , 30 ,

156 7 –1 573; b) Y. Fukuyama, H. Yaso, T. Mori, H. Takahashi, H. Minami,

M. Kodama, Heterocycles 2001 , 54 , 25 9 –2 74.

[31] R. Singh, G. C. Singh, S. K. Ghosh, Eur. J. Org. Chem. 2007 , 537 6 –5 385.

[32] A. R. Battersby, R. Binks, R. J. Francis, D. J. McCaldin, H. Ramuz, J. Chem.

Soc. 1964 , 360 0 –3 610.

[14] In contrast to the previously reported labile nature towards decarboxy-

lation of the 2-pyrrole carboxylic acids,^[9]the structures of which are

similar to those of 24 or 32 except for the proton in place of the bro-

mine atom, these two intermediates could be isolated, chromatograph-

ically purified, and spectroscopically characterized. Structural assign-

ments of 24 and 32 were based on their spectroscopic data (Support-

ing Information). Whereas the exact mechanism of their formation

under the reaction conditions is not fully understood, the presence of

the bromine atom appeared to direct 24 and 32 to preferentially un-

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2649

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

[33] a) R. L. Douglas, J. M. Gulland, J. Chem. Soc. 1931 , 289 3 –2 903; b) T. Ka-

metani, S. Takano, K. Masuko, F. Sasaki, Chem. Pharm. Bull. 1966 , 14 , 6 7 –

72.

[34] F. Ishibashi, Y. Miyazaki, M. Iwao, Tetrahedron 1997 , 53 , 595 1 –5 962.

[35] a) T. Fujii, M. Ohba, J. Sakaguchi, Chem. Pharm. Bull. 1987 , 35 , 362 8 –

3640; b) J. H. Schrittwieser, V. Resch, S. Wallner, W.-D. Lienhart, J. H. Sat-

tler, J. Resch, P. Macheroux, W. Kroutil, J. Org. Chem. 2011 , 76 , 670 3 –

6714.

[36] T. Kametani, T. Terui, H. Agui, K. Fukumoto, J. Heterocycl. Chem. 1968 , 5 ,

75 3 –7 55.

[37] N. Fujikawa, T. Ohta, T. Yamaguchi, T. Fukuda, F. Ishibashi, M. Iwao, Tetra-

hedron 2006 , 62 , 59 4 –6 04.

kajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P.

Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,

R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Och-

terski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,

V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K.

Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui,

A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko,

P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,

C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.

Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wallingford,

CT, USA, 2004.

[40] G. M. Morris, D. S. Goodsell, R. S. Halliday, R. Huey, W. E. Hart, R. K.

Below, A. J. Olson, J. Comput. Chem. 1998 , 19 , 163 9 –1 662.

[41] M. F. Sanner, J. Mol. Graphics Mod. 1999, 17, 57–61.

[38] OECD Guideline for Testing of Chemicals #117 [Internet], Organization

for Economic Co-operation and Development (OECD) Council, 2004.

Available from: htt p ://www.oecd.org/document/40/0,3343,en_2649_

34377_37051368_1_1_1_1,00.html.

[39] Gaussian 03, Revision C.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.

Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr. , T. Vreven,

K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone,

B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsu-

ji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Shida, T. Na-

Manuscript received: June 12, 2015

Accepted Article published: July 16, 2015

Final Article published: September 4, 2015

Chem. Asian J. 2015, 10, 2631 – 2650

www.chemasianj.org

2650

ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Article Doi

Facile and Divergent Synthesis of Lamellarins and Lactam-Containing Derivatives with Improved Drug Likeness and Biological Activities

DOI: 10.1002/asia.201500611

Source and publish data:

Authors:

Article abstract of DOI:10.1002/asia.201500611

Full text of DOI:10.1002/asia.201500611

Products guided by the article

R&D Labs maybe for 1253736-31-0

Relevant to this article

Hot Product