General Fmoc-Based Solid-Phase Synthesis of Complex Depsipeptides Circumventing Problematic Fmoc Removal

DOI: 10.1002/ejoc.201901459

Source and publish data:

European Journal of Organic Chemistry p. 183 - 192 (2020)

Update date:2022-08-10

Topics:

Authors:

Lobo-Ruiz, Ariadna

Tulla-Puche, Judit

Read Full Text PDF DownLoad Join now for total 90,000,000 free articles

Article abstract of DOI:10.1002/ejoc.201901459

Development of an Fmoc-based solid-phase depsipeptide methodology has been hampered by base-promoted fragmentation and diketoperazine formation upon Fmoc group elimination. Such a strategy would be a useful tool given the number of commercially available Fmoc-protected residues. Herein we report that the addition of small percentages of organic acids to the Fmoc-removal cocktail proves effective to circumvent these drawbacks and most importantly, allowed the development of an exclusively solid-phase stepwise methodology to prepare a highly complex depsipeptide with multiple and consecutive esters bonds. Alongside, the optimal protecting group scheme for residue incorporation, which is not as straightforward as it is for traditional peptide synthesis, was explored. The developed stepwise strategy proved effective for the synthesis of a highly complex cyclodepsipeptide, being comparable to the yields obtained when using traditional combined chemistry approaches.

Full text of DOI:10.1002/ejoc.201901459

A Journal of

Accepted Article

Title: Development of a general Fmoc-based solid-phase methodology

for the synthesis of complex depsipeptides by circumventing

problematic Fmoc removal

Authors: Ariadna Lobo-Ruiz and Judit Tulla-Puche

This manuscript has been accepted after peer review and appears as an

Accepted Article online prior to editing, proofing, and formal publication

of the final Version of Record (VoR). This work is currently citable by

using the Digital Object Identifier (DOI) given below. The VoR will be

published online in Early View as soon as possible and may be different

to this Accepted Article as a result of editing. Readers should obtain

the VoR from the journal website shown below when it is published

to ensure accuracy of information. The authors are responsible for the

content of this Accepted Article.

To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201901459

Link to VoR: http://dx.doi.org/10.1002/ejoc.201901459

Supported by

10.1002/ejoc.201901459

European Journal of Organic Chemistry

FULL PAPER

Development of a general Fmoc-based solid phase methodology

for the synthesis of complex depsipeptides by circumventing

problematic Fmoc removal

Ariadna Lobo-Ruiz,^[a]and Judit Tulla-Puche^*[a],[b]

[a]

Dr. A. Lobo-Ruiz, Dr. J. Tulla-Puche

Department of Inorganic and Organic Chemistry – Organic Chemistry Section

University of Barcelona

Martí i Franquès 1-11, 08028 – Barcelona, Catalonia, Spain

E-mail: judit.tulla@ub.edu

https://judittullapuche.wordpress.com/

[b]

Dr. J. Tulla-Puche

Institut de Biomedicina de la Universitat de Barcelona (IBUB)

Martí i Franquès 1-11, 08028 – Barcelona, Catalonia, Spain

Supporting information for this article is given via a link at the end of the document.

Abstract: Development of an Fmoc-based solid phase depsipeptide

methodology has been hampered by base-promoted fragmentation

and diketoperazine formation upon Fmoc group elimination. Such an

strategy would be a useful tool given the number of commercially

available Fmoc-protected residues. Herein we report that the addition

of small percentages of organic acids to the Fmoc-removal cocktail

proves effective to circumvent these drawbacks and most importantly,

allowed the development of an exclusively solid phase stepwise

methodology to prepare a highly complex depsipeptide with multiple

and consecutive esters bonds. Alongside, the optimal protecting

group scheme for residue incorporation, which is not as

straightforward as it is for traditional peptide synthesis, was explored.

The developed stepwise strategy proved effective for the synthesis of

a highly complex cyclodepsipeptide, being comparable to the yields

obtained when using traditional combined chemistry approaches.

peptide APIs are being produced using hybrid strategies after

observing the benefits in the overall yield obtained after

purification.^[21-22]

However these strategies also present some disadvantages. For

instance, the synthetic route must be designed and optimized for

each particular case, and therefore a general synthetic method

cannot be outlined. This can be a limitation in screening

processes, where only small amounts of products are required

and fast-delivery of multiple analogues is more important than

high yields. Furthermore, the synthesis of rather short peptides

and depsipeptides (less than 20 amino acids) might not benefit

from the advantages of hybrid strategies, since the drop in yields

is still more than acceptable even for typical low purity ranges.^[20]

An exclusive solid phase strategy for the preparation of complex

depsipeptides would become a valuable tool to rapidly generate

numerous synthetic analogues for structure-activity relationship

studies (SARs), screening drug candidates and in general

working in projects where time limits the possibility of process

optimization. Among the advantages of solid phase synthesis are

included the convenient elimination of excess reagents after each

step by simple washing and filtration processes, and stepwise

incorporation of orthogonally protected amino and hydroxy acid

monomer units. In this context, it would be extremely useful the

use of commercially available derivatives to avoid building block

preparation in solution. However, ester bond instability upon

removal of certain protecting groups has been reported. Whereas

the strong acidic conditions required to eliminate the tert-

Butyloxycarbonyl (Boc) group might lead to fragmentation,^[23]the

Introduction

Depsipeptides are biomolecules commonly found in nature that

are characterized by the presence of at least one ester bond

1-10

within the peptide backbone.^

Despite growing interest in the

exploitation of naturally-occurring depsipeptides as therapeutic

agents,^^6,10–15the difficulties encountered during the isolation and

purification of large quantities from natural sources, as well as

their challenging chemical synthesis, have hampered their growth

in the drug market. Up to date, the most general and effective

strategy for the preparation of complex depsipeptides combines

solid phase synthesis with solution chemistry approaches, where

basic

conditions

needed

remove

the

Fluorenylmethyloxycarbonyl (Fmoc) group can trigger the

following side-reactions: (i) racemization; (ii) depsipeptide

fragmentation; and (iii) formation of undesired α,β-elimination

side-products. Concomitantly, diketopiperazine (DKP) formation,

which ultimately leads to lower yields due to dipeptide loss, is a

prevalent drawback encountered during Fmoc removal of the

second residue.^24-29By contrast, Allyloxy carbonyl (Alloc)

removal under neutral conditions is likely to prevent fragmentation.

Unfortunately, there are not many commercially available Alloc-

derivatives. Given the number of Fmoc-protected residues

compared to Boc-derivatives commercially available in the market,

we put great efforts into developing a convenient Fmoc-based

building blocks containing the ester moieties are prepared in

16-19

solution and later incorporated onto the polymeric support.^

In these hybrid strategies the chemical and chiral purities of the

different intermediates can be rigorously controlled, and crude

purities tend to be higher when compared to fully stepwise

strategies, usually leading to higher post-purification yields. This

is especially true for high molecular weight peptides. For this

reason, combined strategies are regarded as the best alternative

for the large scale synthesis of peptide-based Active Principle

Ingredients (APIs) in the pharmaceutical industry, where the

economic viability of the products is paramount.^[20]Several

10.1002/ejoc.201901459

European Journal of Organic Chemistry

FULL PAPER

stepwise strategy for the synthesis of complex depsipeptides as

well as circumventing the problems associated with the use of

piperidine when using an Fmoc/tBu strategy. With that purpose,

cyclization in solution at the latest stage of the synthesis (Scheme

1). The peptide chain elongation consisted of coupling and

deprotection repetitive cycles following the well-known Fmoc/tBu

strategy. Couplings were carried out using as coupling system

cyclodepsipeptide YM-254890 (1, Figure 1) as a model peptide to

address our research.^30YM-254890 is

synthetically

used

analogue

naturally-occurring

Fmoc-protected

amino

acid

(DIC)

(Fmoc-AA),N,N′-

Diisopropylcarbodiimide

and

challenging depsipeptide that presents a head-to-side-chain

cyclic arrangement, an overall of three ester bonds (two of them

being consecutive) and a highly N-methylated structure. In 2005,

228 applicants participated in a worldwide contest where a

$100.000 reward was offered for the preparation of at least 1 mg

of this compound, however, none of the candidates succeeded in

this task.^31It was not until 2016 that the total synthesis of YM-

254890 was achieved for the first time.^29A combined solid phase

and solution chemistry approach was applied to prepare the so-

called molecule. Development of an exclusive Fmoc/tBu strategy

was hampered by the presence of the two consecutive ester

bonds and undesired -elimination side reactions arising from

Fmoc removal of the N,O-Me₂Thr-OH residue. In fact, another

group reported complete α,β-elimination at the same point of the

synthesis when attempting the synthesis.^28

Ethyl(hydroxyimino)cyanoacetate (usually known as OxymaPure),

in which the residue was incorporated upon mild and neutral

conditions, hence preventing epimerization. Strong coupling

conditions were required for residue incorporation onto secondary

amines, which present lower reactivity compared to primary

amines. Couplings onto secondary amines such as Pro or N-

methyl amino acids were performed using as coupling system the

Fmoc-AA,

1-[Bis(dimethylamino)methylene]-1H-1,2,3-

triazolo[4,5-b]pyridinium 3-oxid hexafluorophosphate (HATU), 1-

Hydroxy-7-azabenzotriazole

diisopropylethylamine (DIEA).

(HOAt)

and

N,N-

Fmoc removal studies to minimize DKP formation

A good approach to determine the extent of this secondary

reaction is to estimate the resin loading after the third residue

incorporation and compare it to the substitution level after the first

amino acid incorporation. A decrease in this value implies

dipeptide loss due to DKP formation. The tested conditions for

Fmoc elimination from 2 are summarized in Table 1. In all cases,

a quick treatment (2 x 1 min) shorter than the conventional

procedure was performed.

Figure 1. Structure of our proposed model cyclodepsipeptide (1) and naturally-

occurring depsipeptide YM-254890.

Table 1. Study of the optimal Fmoc removal conditions to minimize DKP

formation.

Fmoc removal conditions

% DKP

formationa

Piperidine–DMF (1:4 v/v) (2  1 min)

0.1 M DBU in DMF (2  1 min)

0.1 M HOBt in piperidine–DMF (1:4

v/v) (2  1 min)

The work presented herein describes the development of a fully

Fmoc-based solid phase methodology for the preparation of

complex depsipeptides. The synthetic complexity conferred by

the presence of two consecutive ester bonds and the overall

chemical diversity provided by depsipeptide 1, allowed Fmoc

removal studies after ester bond formation in diverse chemical

environments as well as evaluation of the optimal protecting group

scheme for each residue incorporation via ester bond formation.

Additionally, Fmoc elimination conditions to fully prevent or

minimize DKP formation were evaluated. Note that the proposed

sequence modifications (1) preserve the synthetic challenges

whilst making the methodological studies more viable, since (i)

higher coupling rates were expected due to the less hindered

environment observed when replacing hydroxyleucine (-

HyLeu) residues by Thr, and (ii) the constrained dehydroalanine

moiety, which is prone to undergo Michael addition secondary

reactions and therefore must be selectively protected, was

substituted by a likewise residue, Pro.

0.1 M OxymaPure in piperidine–DMF

(1:4 v/v) (2  1 min)

0.1 M HOBt, 0.13 M DBU in DMF (2

 1 min)

0.1 M OxymaPure, 0.13 M in DMF (2

 1 min)

^aThe DKP formation percentage was determined by the difference between the

loading level after the first and third residue incorporation. The loading level was

determined by Fmoc-UV quantification at 290 nm.

An improvement in the Fmoc removal outcome was observed

when replacing the traditional piperidine system by the 1,8-

Diazabicyclo[5.4.0]undec-7-ene (DBU) cocktail (entries #1 and #2,

Table 1). Addition of small percentages of slightly acidic

molecules such as Hydroxybenzotriazole (HOBt) or OxymaPure

to the Fmoc removal cocktail has proven effective to minimize

aspartimide formation during Fmoc removal.^32Inspired by this

principle, we assessed whether addition of HOBt and OxymaPure

could further reduce DKP formation. Surprisingly, OxymaPure

increased the secondary reaction when added to either cocktail

Results and Discussion

The starting point of the synthesis was incorporation of N-MeAla-

OH onto the 2-Chlorotrityl chloride (2-CTC) resin, which renders

the C-terminus as a free carboxylic acid function and allows

10.1002/ejoc.201901459

European Journal of Organic Chemistry

FULL PAPER

(entries #4 and #6, Table 1). Contrary, HOBt slightly reduced DKP

formation when added to the traditional piperidine mixture and

halved undesired dipeptide loss when added to the DBU solution

(entries #3 and #5, Table 1). Remarkably, Fmoc removal with a

0.1 M HOBt, 0.13 M DBU in dimethylformamide (DMF, 2 x 1 min)

solution seemed to be the most effective treatment, resulting in a

small DKP formation compared to the DKP formed when using

the traditional piperidine–DMF (1:4 v/v) or the alternative DBU–

DMF (2:98 v/v) treatments. However, these optimized conditions

(0.1 M HOBt, 0.13 M DBU in DMF (2 x 1 min)) become a good

alternative to considerably reduce the undesired side-reaction.

Scheme 1. Reaction conditions: a) Fmoc-NMeAla-OH, DIEA, DCM, 50 min; b) MeOH, 10 min; c) piperidine–DMF (1:4 v/v) (1  1 min + 2  10 min); d) Fmoc-Ala-

OH, HATU, HOAt, DIEA, DMF, 1 h; e) 0.1 M HOBt, 0.13 M DBU in DMF (2  1 min); f) Fmoc-Pro-OH, OxymaPure, DIC, DMF, 40 min; g) piperidine–DMF (1:4 v/v)

(1  1 min + 2  5 min); h) D-Pla, HATU, HOAt, DIEA, DMF, 1 h; i) Fmoc-Thr(tBu)-OH, DIC, DMAP, DCM–DMF (9:1 v/v), 35 °C, 2 h 30 min; j) Fmoc-Thr(TBDMS)-

OH (9, see SI for details), DIC, DMAP, DCM–DMF (9:1 v/v), 35 °C, 2 h 30 min; k) 0.1 M OxymaPure, 0.13 M DMF in DMF (1  1 min); l) 1.0 M TBAF in THF, 10

min, N₂atm., this step was repeated twice; m) AcOH, OxymaPure, DIC, DCM, 40 min; n) Fmoc-N,O-Me₂Thr-OH (14, see SI for details), DIC, DMAP, DCM–DMF

(9:1 v/v), 35 °C, 2 h 30 min; o) DBU–DMF (2:98 v/v) (1  1 min); p) Boc-Thr-OH, HATU, HOAt, DIEA, rt, 1 h, this step was repeated twice; q) 0.1 M HOBt, 0.13 M

DBU in DMF (2:98 v/v) solution (1  1 min); r) TFA–TIS–DCM (90:5:5 v/v), 25 °C, 30 min, anhydrous conditions; s) HATU, 2,4,6-collidine, DMF, 2 h.

Selection of the protecting group scheme for the formation of the

first ester linkage and subsequent Fmoc removal studies

Ester bond formation between the α-hydroxyl group of D-

phenyllactic acid (Pla, 4) and the carboxylic acid of the

subsequent residue, Ac-Thr(OH)-OH, requires protection of the

-amino and

β-hydroxyl moiety of the Thr derivative. Selection of the alcohol

protecting group entails significant importance, since the Barlos

resin does not allow deprotection under acidic conditions.

Moreover, the formed ester linkage must be stable to hydroxyl

effect of the protecting group scheme for both the amine and the

hydroxyl functions on the esterification extent (see Table 2). As a

first approach to considerably simplify the screening process, tBu

was selected as the hydroxyl protecting group (note that Fmoc-

Thr(tBu)-OH is readily used in peptide synthesis and can be

purchased at low prices) and the α-amino function was protected

with two different orthogonal protecting groups (Fmoc and Ac).

Unfortunately, ester bond formation with the conveniently

acetylated Thr derivative (Ac-Thr(tBu)-OH, 5) was not

accomplished (entry #1, Table 2), complying with previous

deprotecting conditions. On the basis of the above considerations, reported experiments.^{[ ]}Therefore, the amine function of the Thr

incorporation of the Thr derivative as Ac-Thr(Silyl)-OH turns into

the smartest choice. Note that silyl protecting groups are

orthogonal to the Fmoc/tBu strategy and can be removed by

treatment with tetrabutylammonium fluoride (TBAF).^33However,

Kaur et al. reported the low reactivity of the Ac-Thr(Silyl)-OH

residue upon esterification conditions. It was suggested that the

acetyl moiety might deactivate the carbonyl function resulting in

no residue incorporation.^28In order to develop a fully stepwise

strategy, we explored this residue incorporation by evaluating the

derivative must be acetylated after residue incorporation.

Surprisingly, the esterification product, 7, with commercially

available Fmoc-Thr(tBu)-OH (6) was obtained with a quantitative

conversion

according

High

Performance

Liquid

Chromatography – Mass Spectrometry (HPLC-MS) analysis

(entry #2, Table 2), confirming that the amine protecting group

plays an important role in the reaction outcome. With this

promising result in hands, optimal Fmoc removal conditions of the

model depsipeptide (7) to afford 8 were extensively studied and

10.1002/ejoc.201901459

European Journal of Organic Chemistry

FULL PAPER

are summarized in Table 3. In all cases, a quick treatment of 1

min was performed and the deprotection efficiency was evaluated

by HPLC-MS analysis. Treatment with the traditional piperidine

cocktail resulted in high epimerization rates (54%, entry #1, Table

3). Unfavorable elimination of the Fmoc group was also observed

with the milder DBU cocktail. At this stage, we evaluated whether

addition of HOBt and OxymaPure to the Fmoc removal cocktail

could reduce racemization of 8. Whilst HOBt had a negative effect

when used as additive, incorporation of OxymaPure to the DBU

cocktail proved effective to fully prevent epimerization and afford

the unprotected model depsipeptidyl-resin 8 (entry #5, Table 3).

additional treatment and subsequent acetylation resulted in the

formation of the desired product 13 with quantitative yields and no

evidence of fragmentation according to HPLC analysis.

Table 3. Study of the optimal Fmoc removal conditions to deprotect

depsipeptidyl-resin 7 to afford 8.

Fmoc removal conditions

8:7

ratio

Epimerization

of 5^a

Piperidine–DMF (1:4 v/v) (2 

1 min)

95:5

54%

60%

0.1 M HOBt in Piperidine–DMF

(1:4 v/v) (1  1 min)

Table 2. Study of the optimal protecting group scheme of the Thr derivative for

the formation of the first ester linkage.

DBU–DMF (2:98 v/v) (2 x 1

min)

100:0

Broad peak

56%

0.1 M HOBt, 0.13 M DBU in

DMF (2  1 min)

0.1

OxymaPure, 0.13

DBU in DMF (2  1 min)

a

Epimerization yields were determined by HPLC-MS analysis. HPLC data

processed at 220 nm. See all the chromatograms in the SI.

Formation of the second ester linkage and subsequent Fmoc

removal studies

Amino acid

HPLC conversion (%)^[a]

Ac-Thr(tBu)-OH (5)

With pentadepsipeptidyl-resin 13 in hand, esterification of the

Fmoc-N,O-Me₂Thr-OH (14, see SI for details) residue was carried

out, and the incorporation yield was quantitative according to

HPLC-MS analysis. However, 22% of epimerization was

observed, probably due to the presence of an N-alkylated residue,

which is prone to undergo epimerization at the -carbon.^[34]

Further efforts to minimize epimerization resulted in failure. It has

been reported that Fmoc removal after incorporation of the Fmoc-

N,O-Me₂Thr-OH residue (for the synthesis of the natural analogue,

YM-254890) completely failed to afford the desired product, with

the α,β-elimination side-product formed instead.^28,29We

hypothesized that the presence of two consecutive ester bonds

might enhance the α,β-elimination side-reaction. In order to

establish the best conditions for this particular deprotection step,

the previously tested Fmoc removal conditions were applied to

the present system. In all cases, a quick treatment with the

corresponding deprotection cocktail was performed. The HPLC

yields as well as the structures of the desired product (16) and the

two main side-products corresponding to the N,O-Me₂Thr-OH

residue loss (13) and the α,β-elimination product (17) are shown

in Table 4. As expected, Fmoc removal with the traditional

piperidine–cocktail mainly led to the formation of the α,β-

elimination product (17), but also to N,O-Me₂Thr-OH residue loss

(13) and only 37 % of the desired product 16 (entry #1, Table 4).

Remarkably, the best conditions were obtained with the DBU–

DMF (2:98 v/v) system, where the desired product (16) was

obtained with a 57% HPLC yield, and we were able to significantly

decrease side-products 13 and 17 formation (entry #3, Table 4).

Although small percentages of HOBt and OxymaPure were added

to the Fmoc removal cocktail, the deprotection outcome could not

be further improved. To the best of our knowledge, a big step

forward compared to the results reported in the literature (where

only traces of the desired product 16 were detected and 17 was

obtained as a major product) was made, since formation of the

α,β-elimination side-product (17) was considerably minimized.

Fmoc-Thr(tBu)-OH (6)

Fmoc-Thr(TBDPS)-OH (9)

Fmoc-Thr(TBDMS)-OH (10)

100

a

Reaction monitoring was carried out by HPLC-MS analysis. HPLC data

processed at 220 nm. See all the chromatograms in the SI.

Once Fmoc was selected as the optimal protecting group for the

amine moiety, assessment of the most suitable protecting group

for the -hydroxyl function was carried out with tert-

butyldimethylsilyl (TBDMS) and tert-butyldiphenylsilyl (TBDPS,

which allowed evaluation of the bulkiness effect on the

esterification rates (Table 2). Whereas incorporation of bulkier

Fmoc-Thr(TBDPS)-OH (9) resulted in failed attempts to introduce

the Thr derivative (entry #3, Table2), incorporation of Fmoc-

Thr(TBDMS)-OH (10) was successfully achieved with

quantitative incorporation yield (entry #4, Table2). Although

incorporation of Fmoc-Thr(TBDMS)-OH (10) was a good starting

point, further studies on TBDMS and Fmoc removal of 11 needed

to be carried out. Initial attempts consisted of Fmoc elimination

with the already optimized conditions (treatment with a 0.1 M

OxymaPure, 0.13 M DBU in DMF solution for 1 min), which fully

prevented epimerization, and subsequent acetylation with the

AcOH–OxymaPure–DIC system. Next, stability of the ester

linkage upon treatment with

a 1.0 M TBAF solution in

tetrahydrofuran (THF) under anhydrous conditions was studied.

Unfortunately, treatment of the peptidyl-resin 12 with TBAF failed

to afford the desired product, and depsipeptide fragmentation was

observed instead yielding 4 with a quantitative HPLC conversion.

In order to avoid depsipeptide fragmentation, simultaneous Fmoc

and TBDMS removal was explored by treating depsipeptidyl-resin

11 with a 1.0 M TBAF solution in THF. This strategy led to partial

elimination of the Fmoc and TBDMS protecting groups. An

10.1002/ejoc.201901459

European Journal of Organic Chemistry

FULL PAPER

We were able to prevent complete α,β-elimination, and obtain the

desired Fmoc-unprotected depsipeptide in a good enough HPLC

yield (57%) to carry on with the synthesis.

conditions. The linear precursor 21 was subjected to HPLC

purification and obtained with a yield of 25% (over 18 steps).

Considering the synthetic complexity of the target molecule and

the numerous steps required for its preparation, 21 was prepared

with an outstanding overall yield. Obtention of the final cyclic

analogue (1) was accomplished by treatment of the pure linear

unprotected depsipeptide chain with the HATU–collidine coupling

system.

Table 4. Study of the optimal Fmoc removal conditions to deprotect

depsipeptidyl-resin 15, which afforded the desired product 16 and side-products

13 and 17.

Side products derived from the Fmoc removal step

Table 5. Study of the optimal protecting group scheme of the Thr derivative for

the formation of the third ester linkage.

Fmoc removal conditions

(%)

(%)^a

Piperidine–DMF (1:4 v/v) (2  1

min)

0.1 M HOBt in Piperidine–DMF

(1:4 v/v) (1  1 min)

DBU–DMF (2:98 v/v) (2  1 min)

0.1 M HOBt, 0.13 M DBU in DMF

(2  1 min)

0.1 M OxymaPure, 0.13 M DBU in

DMF (2  1 min)

Amino acid

HPLC conversion (%)^[a]

a

Ac-Thr(tBu)-OH (5)

Fmoc-Thr(tBu)-OH (6)

Product percentages were determined by HPLC-MS analysis. HPLC data

processed at 220 nm. See all the chromatograms in the SI

100

a

Selection of the optimal N^-protecting group for the formation of

the third ester linkage and subsequent Fmoc removal studies

Peptide coupling of Boc-Thr-OH was followed by assembly of the

Ac-Thr(tBu)-OH residue via an ester bond formation between its

carboxylic acid and the β-hydroxyl group of 18. Two N^-protecting

groups, Ac and Fmoc (see Table 5), were explored for this

esterification reaction. Complying with previously obtained results

for the first esterification, formation of the ester bond was only

accomplished with the Fmoc derivative (6) (entry #2, Table 5). As

a part of the Fmoc removal studies after formation of all three

ester linkages, efficient Fmoc removal from 19 was extensively

studied. Again, a quick treatment was performed and the

deprotection efficiency was evaluated by HPLC-MS analysis. The

same conditions as the ones used for the first and second ester

deprotection were tested and are summarized in Table 6. The α,β-

elimination side-product (17) was formed in high percentages

(90%) when piperidine was used as base (entry #1, Table 6).

Depsipeptide fragmentation was also observed when using 0.1 M

HOBt in piperidine–DMF (2:98 v/v), DBU–DMF (2:98 v/v) and 0.1

M OxymaPure, 0.13 M in DMF (entries #2-4, Table 6). Optimal

Fmoc removal conditions after formation of the third ester linkage

were accomplished by treatment of 19 with a 0.1 M HOBt in DBU–

DMF (2:98 v/v) solution, in which formation of the α,β-elimination

product was fully suppressed (entry #5, Table 6). Again, we were

able to improve the Fmoc removal outcome when adding small

percentages of organic acids.

Reaction monitoring was carried out by HPLC-MS analysis. HPLC data

processed at 220 nm. See all the chromatograms in the SI.

Table 6. Study of the optimal Fmoc removal conditions to deprotect

depsipeptdiyl-resin 19, which rendered the desired product 20 and side-product

17.

Side product derived from the Fmoc removal step

Fmoc removal conditions

(%)

(%)^a

Piperidine–DMF (1:4 v/v) (2  1 min)

0.1 M HOBt in Piperidine–DMF (1:4

v/v) (1  1 min)

DBU–DMF (2:98 v/v) (2  1 min)

0.1 M OxymaPure in DBU–DMF (2:98

v/v) (2  1 min)

0.1 M HOBt, 0.13 M DBU in DMF (2 

100

1 min)

a

Product percentages were determined by HPLC-MS analysis. HPLC data

processed at 220 nm. See all the chromatograms in the SI.

Lastly, acetylation of the free amine was carried out prior to

depsipeptide cleavage and global deprotection with

trifluoroacetic acid (TFA) triisopropyl silane (TIS)

–

dichloromethane (DCM) (90:5:5 v/v) cocktail under anhydrous

10.1002/ejoc.201901459

European Journal of Organic Chemistry

FULL PAPER

Evaluation of the stepwise strategy efficiency by comparison with

traditional combined chemistry approaches

condensation (Scheme 2, see SI for details) and (ii) a strategy

containing two depsipeptide segment condensations (Scheme 3,

see SI for details). All three strategies led to comparable yields

(between 24-26%) and therefore prove effective for the

preparation of the target molecule. However, the fully stepwise

strategy presents some advantages over segment condensation

approaches.

In order to evaluate whether it is more efficient to incorporate the

residues in an exclusive stepwise manner or following traditional

segment condensation approaches, two synthetic strategies

parallel to the synthesis of natural YM-254890 were developed

and optimized for the preparation of the target linear depsipeptide

(21), including (i) a strategy containing one depsipeptide segment

Scheme 2. Synthetic strategy containing one-segment condensation. Yield: 24% over 14 steps.

Scheme 3. Synthetic strategy containing two-segment condensations. Yield: 26% over 11 steps.

Additionally, the numerous commercially available Fmoc-

protected amino acid derivatives facilitate the synthetic process.

A major advantage of this strategy is that the excesses of

reagents can be washed away by simple suction, and therefore

only one purification step is required throughout the synthesis,

thus the synthesis is less time-consuming and laborious.

10.1002/ejoc.201901459

European Journal of Organic Chemistry

FULL PAPER

The peptidyl-resin was washed with dry DCM (3 x 1 min), dried under

vacuum for 20 min and transferred to a pyrex® culture tube provided with

a magnetic stirrer. Esterification with Fmoc-N,O-Me₂Thr-OH (14, 52 mg,

0.14 mmol, 8.0 eq), DIC (22 µL, 0.14 mmol, 8.0 eq) and DMAP (1 mg, 0.01

mmol, 0.5 eq) was performed as before. Fmoc removal was achieved by

treatment with a DBU–DMF (2:98 v/v) (1 x 1 min) solution. The peptidyl-

resin was washed with DMF (3 x 1 min), DCM (3 x 1 min) -and DMF (3 x 1

min).

Conclusion

In conclusion, a stepwise Fmoc-based solid phase methodology

for the synthesis of a highly complex depsipeptide was developed

by selecting the most suitable protecting group scheme for each

residue incorporation as well as circumventing problematic Fmoc

group elimination of the second residue or after ester bond

formation. In the latter, replacement of the traditional piperidine-

based Fmoc removal cocktail by a DBU solution led in all cases

to better deprotection outcomes. Noteworthy, the addition of small

percentages of organic acids (HOBt or OxymaPure) to the

deprotection cocktail resulted in most cases in an improvement in

this step, being secondary reactions fully prevented or

significantly minimized. Nevertheless, the best deprotection

conditions (HOBt, OxymaPure or no additive) are highly chemical

environment-dependent and therefore should be evaluated for

each problematic Fmoc elimination step. This newly developed

methodological study is a valuable tool for the preparation of

synthetically challenging depsipeptides and the rapid generation

of numerous analogues.

A mixture of Boc-Thr-OH (12 mg, 0.05 mmol, 3.0 eq), HATU (20 mg, 0.05

mmol, 3.0 eq), HOAt (7 mg, 0.05 mmol, 3.0 eq) and DIEA (18 µL, 0.11

mmol, 6.0 eq) in DMF (500 µL) was added to the resin through an

amidation reaction and the peptidyl-resin was shaken for 1 h. In this case,

re-coupling was required to accomplish full incorporation of the Thr

derivative. Incorporation of the last amino acid was achieved again via an

esterification reaction using Fmoc-Thr(tBu)-OH (56 mg, 0.14 mmol, 8.0 eq),

DIC (22 µL, 0.14 mmol, 8.0 eq) and DMAP (1 mg, 0.01 mmol, 0.5 eq).

Incorporation of the last amino acid was achieved via a Steglich

esterification reaction. Fmoc-Thr(tBu)-OH (56 mg, 0.14 mmol, 8.0 eq), DIC

(22 µL, 0.14 mmol, 8.0 eq) and DMAP (1 mg, 0.01 mmol, 0.5 eq) were

added to the peptidyl-resin following the conditions reported above. Fmoc

removal was achieved by treatment with a solution of 0.1 M HOBt in DBU–

DMF (2:98 v/v) (1 x 1 min). The peptide resin was washed with DMF (3 x

1 min), DCM (3 x 1 min) and DMF (3 x 1 min). The peptidyl-resin was

washed with DMF (3 x 1 min), DCM (3 x 1 min) and DMF (3 x 1 min), and

the corresponding free amine was acetylated by treatment with a solution

of AcOH (2 µL, 0.05 mmol, 3.0 eq), OxymaPure (8 mg, 0.05 mmol, 3.0 eq)

and DIC (8 µL, 0.05 mmol, 3.0 eq) in DMF (500 µL) for 40 min. The peptide

resin was washed with DMF (3 x 1 min), DCM (3 x 1 min) and DMF (3 x 1

min). Treatment of the peptidyl-resin with a TFA–TIS–DCM (90:5:5 v/v)

over 30 min furnished the crude linear depsipeptide. The lyophilized crude

linear peptide was purified by HPLC using a XBridgeTM C18 reversed-

phase column (3.5 µm x 4.6 mm x 42 mm) (purification gradient: C18

G0100t20T25) to afford pure 21 (4.0 mg, 25% over 16 steps) as white solid.

XBridge^TMC18 reversed-phase analytical column (3.5 m x 4.6 mm x 42

mm) linear gradients (0% to 100%) of ACN over 9 min, with a flow rate of

1.0 mL/min, t_R= 10.07 min, purity (λ = 220 nm) = 96%.HRMS-ESI(+)

characterization: m/z calculated for C₄₃H₆₅N₇O₁₆935.4566, found [M + H]⁺

936.4558

Experimental Section

Depsipeptide assembly on solid-phase using

strategy

a fully stepwise

2-CTC resin (25 mg, 1.6 mmol/g resin) was placed in a 2 mL-polypropylene

syringe fitted with two polyethylene filter discs. The conditioning of the

resin and incorporation of the first amino acid, Fmoc-NMeAla-OH (6 mg,

0.02 mmol, 1.0 eq), was carried out by standard means. The Fmoc group

was removed by treatment with a piperidine–DMF solution (1:4 v/v) (1 x 1

min + 2 x 10 min). The peptidyl-resin was washed with DMF (3 x 1 min),

DCM (3 x 1 min) and DMF (3 x 1 min), and a solution of Fmoc-Ala-OH*H₂O

(17 mg, 0.05 mmol, 3.0 eq), HATU (20 mg, 0.05 mmol, 3.0 eq), HOAt (7

mg, 0.05 mmol, 3.0 eq) and DIEA (18 µL, 0.11 mmol, 6.0 eq) in DMF (500

µL) was added to the peptidyl-resin and the coupling was shaken for 1 h.

In order to minimize diketopiperazine formation, the Fmoc group of the

second residue was eliminated by treatment with a 0.1 M HOBt in DBU–

DMF (2:98 v/v) solution (2 x 1 min). A mixture of Fmoc-Pro-OH (18 mg,

0.05 mmol, 3.0 eq), OxymaPure (8 mg, 0.05 mmol, 3.0 eq) and DIC (8 µL,

0.05 mmol, 3.0 eq) in DMF (500 µL) was added to the peptide-bound and

the resin was shaken for 40 min, and then the Fmoc group was

subsequently removed by treatment with a piperidine–DMF (1:4 v/v) (1 x 1

min + 2 x 5 min) solution. A mixture of D-(+)-phenyllactic (9 mg, 0.05 mmol,

3.0 eq), HATU (20 mg, 0.05 mmol, 3.0 eq), HOAt (7 mg, 0.05 mmol, 3.0

eq) and DIEA (18 µL, 0.11 mmol, 6.0 eq) in DMF (500 µL) was added to

the resin and the peptidyl-resin was shaken for 1 h.

Next, the peptidyl-resin was washed with dry DCM (3 x 1 min) and it was

dried under vacuum for 20 min. The peptide-resin was transferred to a

pyrex® culture tube provided with a magnetic stirrer. A solution containing

Fmoc-Thr(TBDMS)-OH, (10, 64 mg, 0.14 mmol, 8.0 eq), DIC (22 µL, 0.14

mmol, 8.0 eq) and 4-dimethytlaminopyridine (DMAP) in dry DCM was pre–

activated for 5 min before it was added to the tube. A solution of DMAP (1

mg, 0.01 mmol, 0.5 eq) in dry DMF was also added to the tube and the

reaction was shaken for 2 h and 30 min at 35 C. The reaction mixture was

cooled to 25 C, transferred to a polypropylene syringe fitted with two

polyethylene discs and washed with dry DCM (3 x 1 min), DMF (3 x 1 min)

and DCM (3 x 1 min). Esterification completion was monitored by HPLC-

MS. Next, the TBDMS and Fmoc groups were simultaneously removed by

treatment with a 1.0 M TBAF (36 µL, 0.02 mmol, 1.0 eq) (2 x 10 min)

solution in THF (500 µL) under N₂atmosphere. The resin was washed with

dry THF (3 x 1 min), DCM (3 x 1 min) and DMF (3 x 1 min), and the

corresponding free amine was acetylated by treatment with a solution of

AcOH (2 µL, 0.05 mmol, 3.0 eq), OxymaPure (8 mg, 0.05 mmol, 3.0 eq)

and DIC (8 µL, 0.05 mmol, 3.0 eq) in DMF (500 µL) for 40 min.

Depsipeptide assembly on solid-phase containing one depsipeptide

building block segment condensation