Acid-Stable Ester Linkers for the Solid-Phase Synthesis of Immobilized Peptides

Article abstract of DOI:10.1002/cplu.202000246

A series of N-terminally Fmoc-protected linkers of the general formula Fmoc-X?CO?O?Y?COOH have been prepared, where X is ?NH?CH₂?CH₂- or -p-(aminomethyl)phenyl- and Y is ?(CH₂)_n? (n is 1 or 4) or -p-(methyl)phenyl-. These linkers can easily be covalently attached via their C-terminal carboxyl group to a resin bearing a free amino group. After cleavage of the N-terminal Fmoc group, the linkers can be extended by standard solid-phase peptide synthesis techniques. These ester linkers are acid-stable and resistant to the base-mediated diketopiperazine formation that often occurs during the synthesis of ester-bound peptides; they are stable at neutral pH in aqueous buffers for days but can be effectively cleaved with 0.1 m NaOH or aq. ammonia within minutes or hours, respectively. These properties make these ester handles well suited for use as linkers for the solid-phase peptide synthesis of immobilized peptides when the stable on-resin immobilization of the peptides and the testing of their biological properties in aqueous buffers at neutral pH are necessary.

Full text of DOI:10.1002/cplu.202000246

Full Papers
doi.org/10.1002/cplu.202000246
ChemPlusChem
1
2
3
4
5
Acid-Stable Ester Linkers for the Solid-Phase Synthesis of
Immobilized Peptides
Jan Pícha, Miloš Buděšínský, Katarína Mitrová, and Jiří Jiráček*[a]
6
7
8
9
A series of N-terminally Fmoc-protected linkers of the general
azine formation that often occurs during the synthesis of ester-
bound peptides; they are stable at neutral pH in aqueous
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
formula Fmoc-XÀ COÀ OÀ YÀ COOH have been prepared, where X
is À NHÀ CH2À CH2- or -p-(aminomethyl)phenyl- and Y is À (CH2)nÀ buffers for days but can be effectively cleaved with 0.1 m NaOH
(n is 1 or 4) or -p-(methyl)phenyl-. These linkers can easily be
covalently attached via their C-terminal carboxyl group to a
resin bearing a free amino group. After cleavage of the N-
terminal Fmoc group, the linkers can be extended by standard
solid-phase peptide synthesis techniques. These ester linkers
are acid-stable and resistant to the base-mediated diketopiper-
or aq. ammonia within minutes or hours, respectively. These
properties make these ester handles well suited for use as
linkers for the solid-phase peptide synthesis of immobilized
peptides when the stable on-resin immobilization of the
peptides and the testing of their biological properties in
aqueous buffers at neutral pH are necessary.
Introduction
support can be useful in certain situations, such as monitoring
the quality of the synthesis of immobilized peptides or
especially in combinatorial approaches for identifying the
peptides on “positive” resin beads by mass spectrometry after
biological screening libraries of compounds immobilized on
beads.[3a,c–g]
Compounds immobilized on solid supports are broadly appli-
cable, e.g., in the affinity purification of proteins,[1] the
identification of new protein targets for immobilized ligands[2]
or in combinatorial approaches for the identification of new
protein ligands and drug leads.[3]
A variety of linkers and cleavage strategies have been
developed for solid-phase peptide chemistry.[6] Some of these
strategies, such as photochemical approaches,[7] metal-assisted
strategies[8] or so-called “safety catch linkers”,[9] offer alternatives
to acid-mediated cleavage of peptides from solid supports.
However, these approaches often involve the use of less-
straightforward chemistries or reagents that can cause side
reactions (Ref.[6b] and references therein). In this respect,
connecting the peptide carboxy terminus to the resin-bound 4-
hydroxymethyl benzoic acid (HMBA) linker via an ester bond
seems to be the simplest and best solution.[10] However,
esterification of the hydroxyl group with an Nα-Fmoc-protected
amino acid can be accompanied by incomplete couplings[11] or
side reactions such as base-mediated Fmoc cleavage[12] and
racemization.[13] Importantly, resin-bound C-terminal amino acid
esters are also prone to diketopiperazine formation during
peptide synthesis.[5,14] Diketopiperazine formation and subse-
quent cleavage of the peptides from the resin can be a serious
complication in combinatorial synthesis of peptide mixtures
where an equimolar distribution of peptides is expected.
The aim of this work was to develop a linker that (i) can be
easily attached to a resin bearing free amino groups, (ii) is
stable during piperidine-mediated Fmoc deprotection, (iii) is
resistant to the racemization and diketopiperazine formation
often accompanying peptide synthesis, (iv) is resistant to the
TFA used for amino acid side chains deprotection, (v) is stable
in aqueous buffers of neutral pH, which are common media for
many biological assays and, finally, (vi) can be cleaved under
conditions that will not damage the synthesized peptide and
will allow mass spectrometric characterization. Such a linker
should have properties (summarized in Figure 1) suitable for
Peptides are frequently used as ligands for targeting
proteins of pharmacological interest. Solid-phase peptide syn-
thesis is a common method for the generation of new ligands
in the form of resin-bound combinatorial libraries.[4] The most
frequently employed variant of solid-phase peptide synthesis
uses Nα-Fmoc and side-chain tBu-protected amino acids[5] and
involves elongation of a peptide sequence from the C- to the N-
terminus and a final acid-mediated cleavage of protecting
groups of the amino acid side-chains together with the
simultaneous cleavage of the linker moiety between the
peptide C-terminus and the solid support (resin). On the other
hand, stable peptide immobilization on a solid-phase support
after completion of a synthesis, i.e., after cleavage of the acid-
labile side chain protecting groups, requires that the linkage
between the peptide and the resin is resistant to concentrated
trifluoroacetic acid (TFA), which is usually used for deprotection.
This can be achieved simply by coupling the first C-terminal
amino acid to the resin-bound primary amino group, forming a
TFA-stable amide bond. However, this approach does not allow
nondestructive cleavage of the final peptide from the resin. A
nondestructive cleavage/release of compounds from the solid
[a] Dr. J. Pícha, Dr. M. Buděšínský, K. Mitrová, Dr. J. Jiráček
Institute of Organic Chemistry and Biochemistry
The Czech Academy of Sciences
Flemingovo nám. 2, 166 10 Prague 6 (Czech Republic)
E-mail: jiracek@uochb.cas.cz
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/cplu.202000246
This article is part of a Special Collection on “Chemistry in the Czech Re-
public”.
ChemPlusChem 2020, 85, 1297–1306
1297
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
doi.org/10.1002/cplu.202000246
ChemPlusChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Scheme 1. Linkers for the solid-phase peptide synthesis prepared in this
work. The color scheme is the same as in Figure 1. Linker III has already been
prepared by Valerio et al.[15]
byproducts during the peptide synthesis. Linker III is identical
to the Fmoc-NHÀ CH2À CH2À COÀ O-p-(methyl)phenyl-COOH linker
prepared by Valerio et al.[15] We prepared a series of model
peptides on these linkers, and we investigated in detail their
stabilities in aqueous buffers of different pH values, which is a
crucial property for biological applications of immobilized
peptides. We found that the ester linkers prepared here can be
useful as acid-stable and base-labile linkers for the solid-phase
synthesis of resin-immobilized peptides and their applications.
Results and Discussion
Solution-phase synthesis of compounds
Figure 1. General principles and results of this work. The blue C-terminal
carboxyl group enables a simple and stable attachment to the amino-
substituted resin, and cleavage of the blue Fmoc-protected amino group
enable elongation with amino acids by classic Fmoc/t-Bu peptide synthesis
techniques. The chemical structure of the green X part of the linkers
provides resistance to diketopiperazine formation during peptide synthesis,
the black Y part of the linker modulates the stability at neutral pH and,
finally, the central orange ester bond enables the cleavage of the peptide
from the resin under mild alkaline conditions.
Scheme 2 shows our strategy for the synthesis of target
compounds
5 (I)–7 (III) and 12 (IV)–14 (VI), namely, a
straightforward three-step synthetic sequence starting from
Boc-protected acids 1 and 8. Recently, we directly alkylated
propargylamine with tert-butyl 5-bromovalerate in acetonitrile
in the presence of potash.[16] Here, we applied these reaction
conditions for the esterification of 1 or 8 with the correspond-
ing brominated derivatives. Using this reaction, we prepared a
series of intermediates, 2–4 and 9–11, in high yields (approx-
imately 90%). Another positive feature of this straightforward
protocol is the fact that after filtering off the solid potassium
carbonate from the reaction mixture, the filtrate is concentrated
under reduced pressure, and the oily residue is loaded directly
on a column without any further aqueous work-up. The starting
tert-butyl 2-bromoacetate is commercially available, and tert-
butyl 5-bromovalerate was prepared by a previously published
method.[16] In the case of the synthesis of tert-butyl 4-
(bromomethyl)benzoate,(16) we tested different esterification
methods using a series of condensation agents, such as DIC and
PyBroP, but we ultimately found that the highest yield was
obtained by heating isourea 15 with 4-(bromomethyl)benzoic
acid in DCM (Scheme 2, f). In the next step, both the Boc and
tBu acid-labile groups of intermediates 2–4 and 9–11 were
cleaved by treatment with TFA in DCM with rapid liberation of
isobutene and carbon dioxide. The fully deprotected intermedi-
ates were not isolated but dissolved in aqueous sodium
hydrogen carbonate solution, and then the free amino group
both the synthesis and screening of resin-bound combinatorial
libraries of compounds and could be an alternative to linkers
cleavable with UV light,[3d,7] cleavable with acids stronger than
TFA (e.g. PAM ester linker[3e]
) or methionine-containing
linkers[3f,g] cleavable with CNBr, and could help to avoid side
reactions connected with these reactions and reagents.
In 1994, Bray et al.[10c] published a series of ester handles of
general formulas Fmoc-AAxÀ O-p-(methyl)phenyl-COOH or
Fmoc-AAxÀ OÀ CH2-COOH (where AAx is an α-amino acid) and
explored their properties in conjunction with multipin peptide
synthesis. In 1991, Valerio et al.[15] prepared an ester linker,
Fmoc-NHÀ CH2À CH2À COÀ O-p-(methyl)phenyl-COOH, bearing a
β-alanine at the N-terminus that provides stability to diketopi-
perazine formation after cleavage of the Fmoc group from the
linker. Inspired by these studies, we prepared and characterized
a series of six ester handles, I–VI, containing Nα-Fmoc-protected
β-alanine or p-(aminomethyl)benzoic acid moieties (Scheme 1),
and the presence of these groups at the N-terminus of the
linkers prohibits the formation of undesirable diketopiperazine
ChemPlusChem 2020, 85, 1297–1306
www.chempluschem.org
1298
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
doi.org/10.1002/cplu.202000246
ChemPlusChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
°
Scheme 2. Reagents, conditions and yields: (a) Boc2O, Na2CO3, water and dioxane, 1 h at 0 C then at rt overnight (87% for 1, 60% for 8); (b) tert-butyl 2-
°
bromoacetate for 2 and 9, tert-butyl 5-bromovalerate for 3 and 10, 16 for 4 and 11, K2CO3, AcCN, 60 C overnight (92% for 2, 88% for 3, 87% for 4, 92% for 9,
°
°
87% for 10, 91% for 11); (c) TFA, DCM, 1 h at 0 C then 2 h at rt; (d) Fmoc-Osu, NaHCO3 water, dioxane, 1 h at 0 C then at rt overnight (95% for 5, 94% for 6,
°
92% for 7, 94% for 12, 92% for 13, 90% for 14, each yield is over two steps); (e) tert-butanol, CuCl, 6 d at rt (88%), (f) 15, DCM, at 40 C for 24 h (88%).
was acylated with Fmoc-Osu. After classical aqueous work-up
and trituration of the solids, pure products 5 (I)–7 (III) and 12
(IV)–14 (VI) were isolated in excellent yields (above 90% for two
steps).
As mentioned in the Introduction, dipeptides attached by
ester bonds to HMBA immobilized on resin can undergo
intramolecular aminolysis.[17] This process is accelerated by
piperidine, which is used for cleavage of the Nα-Fmoc protect-
ing group and leads to diketopiperazine side-products. The
structures of linkers 5 (I)–7 (III), derived from β-alanine, and
linkers 12 (IV)–14 (VI), derived from 4-(aminomethyl)benzoic
acid, preclude the formation of diketopiperazines, as they do
not allow the creation of a stable six-membered ring after
attachment of an additional α-amino acid and cleavage of its
Fmoc group. In this respect, we sought to determine if ester
linkers having a standard α-amino acid moiety at their N-
terminal part, e.g., the handles prepared by Bray et al.[10c]
(Fmoc-AAxÀ O-p-(methyl)phenyl-COOH or Fmoc-AAxÀ OÀ CH2-
COOH), which are mentioned in the Introduction, are prone or
resistant to diketopiperazine formation after cleavage of their
ChemPlusChem 2020, 85, 1297–1306
www.chempluschem.org
1299
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
doi.org/10.1002/cplu.202000246
ChemPlusChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Scheme 3. Reagents, conditions and yields: (a) tert-butyl 5-bromo valerate (for 17), tert-butyl 2-bromo acetate (for 18 or 19) or 16 (for 20), K2CO3, AcCN, at
°
°
60 C overnight (36% for 17, 91% for 18, 86% for 19, 71% for 20); (b) TFA, DCM, at 0 C for 1 h then for 2 h at rt; (c) Fmoc-Osu, NaHCO3 water, dioxane, for 1 h
°
at 0 C then at rt overnight (91% for 21, 83% for 22, 89% for 23, 90% for 24).
N-terminal Fmoc group (see Investigation of diketopiperazine
formation below). Therefore, linkers 21–24 derived from glycine
and alanine were prepared (Scheme 3). The synthetic strategy
was the same as that used for compounds 5 (I)–7 (III) and 12
(IV)–14 (VI). Boc-Gly and Boc-Ala were reacted with protected
bromo esters to give corresponding intermediates 17–20.
Surprisingly, unlike the other intermediates, product 17 was
isolated in a lower yield (36%), and no modifications of the
reactions conditions lead to better results. The last two
synthetic steps, acid deprotection and a reaction with Fmoc-
Osu, proceeded smoothly with high cumulative yields (83–91%)
for compounds 21–24.
group of O-methyl-L-tyrosine was Cbz protected. Then, the
PyBroP-mediated reaction of intermediate 25 with Gly-OMe·HCl
or Ala-OMe·HCl led to dipeptides 26 or 27, respectively.
Catalytic hydrogenolysis of 26 or 27 afforded the free amino
dipeptides, which quickly underwent spontaneous cyclization
and afforded products 28 or 29, respectively.
The purities of compounds 5 (I)–7 (III), 12 (IV)–14 (VI), 21–
24, and 28–30 (Fmoc-piperidine adducts) were determined by
RP-HPLC. The chromatograms are shown in Figures S1–S13 (see
the Supporting Information).
Next, we prepared model cyclic compounds 28 and 29
(Scheme 4). They were designed to represent “standards” for
the easy detection of potential diketopiperazine products in
reaction mixtures after Fmoc deprotection of the amino acids
attached to resin-bound linkers 21–24. Thus, the free amino
Solid-phase peptide synthesis
For the synthesis of model peptide sequences, we chose a
standard Fmoc strategy that is fully compatible with automated
synthesis in peptide synthesizers and can be used to test the
stability of the linker ester bonds in the presence of piperidine
and TFA. Linker 14 (VI) could not be tested due to its negligible
solubility in DMF and other commonly used solvents, and thus
it was excluded from further experiments.
A schematic of the synthesis of model peptides 31 and 32 is
shown in Scheme 5. Briefly, DMF-swollen beads of resin bearing
free amino groups in plastic syringes equipped with Teflon frits
were treated with linkers 5 (I)–7 (III) or 12 (IV) and 13 (V) using
either HBTU/DIPEA twice per hour or PyBroP/DIPEA for 8 and
17 h as condensation agents. Any remaining free amino groups
were capped with acetic anhydride, and the Fmoc groups were
cleaved, which enabled spectrophotometric quantitative esti-
mation of the linker loading on the resin. Then, syringes with
the resins were installed in an automatic peptide synthesizer,
and a Tyr-Ala-Lys-Gly-Glu-Ala peptide chain was built using a
Scheme 4. Reagents, conditions and yields: (a) Cbz-Cl, Na2CO3, water and
°
dioxane, at 0 C for 1 h then at rt overnight (77%); (b) Gly-OMe·HCl or Ala-
OMe·HCl, PyBroP, DIPEA, DMF, overnight at rt (70% for 26, 69% for 27); (c)
10% Pd/C, H2, 15 psi, methanol at rt (63% for 28, 65% for 29).
ChemPlusChem 2020, 85, 1297–1306
www.chempluschem.org
1300
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
doi.org/10.1002/cplu.202000246
ChemPlusChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Scheme 5. Reagents and conditions: (a) HBTU, DIPEA, 5–7 or 12–13, DMF, 2× for 1 h; (b) PyBroP, DIPEA, 5–7 or 12 and 13, DMF, for 18 and 7 h; (c) Ac2O,
DIPEA, DMF, 2× for 15 min; (d) 20% piperidine/DMF for 5 and 20 min; (e) Fmoc solid-phase peptide synthesis automated cycles with individual amino acids
(see Methods); (f) 5% TIPS/TFA 2× for 1 h; (g) 100 mM NaOH, 45 min for 5–7, 24 h for 12 and 13. NH2 denotes the N-terminus of the peptides.
standard Fmoc solid-phase peptide synthesis protocol. Acid-
labile protecting groups (Boc and tBu) on the side chains of Tyr,
Lys and Glu were removed by treatment with an acidic cocktail.
The cleavage solutions were filtered off from the resin, and the
filtrates were concentrated and analyzed by HPLC. No acid-
catalyzed linker hydrolysis products were found. Finally, the
linker ester bond was saponified with an aqueous solution of
0.1 M NaOH. The linkers derived from β-alanine were cleaved in
45 min, and the linkers 12 (V) and 13 (VI) derived from 4-
(aminomethyl)benzoic acid were cleaved in 24 h because we
expected the carboxyphenyl ester moieties to be more stable to
hydrolysis. Later (see below, Stability of linkers), the hydrolytic
stabilities of the linkers were investigated, and the results are
discussed below in detail. Briefly, these experiments confirmed
that a longer cleavage time was suitable for 13 (V), but for 12
(V), 45 min in 0.1 M NaOH would be generally sufficient.
lower than the loading of linkers 5 (I)–7 (III), probably due to a
lower reactivity of the benzene-bound carboxylic group.
RP-HPLC analyses (Figures S14–S19 in the Supporting
Information) of crude peptides 31 (entries 1–6) revealed high
purities for the products synthesized with linkers 5 (I)–7 (III); all
over 90%. In contrast, crude peptides 32 (entries 7–10),
synthesized on linkers 12 (IV) or 13 (V) (Figures S20–S23 in the
Supporting Information), were contaminated with several
impurities, and their purities were approximately 75%, which is
still satisfactory. These lower yields could be explained by the
longer treatment with aqueous sodium hydroxide, which can
damage peptides. Shorter cleavage times would likely provide
purities similar to those of linkers 5 (I)–7 (III). The RP-HPLC
profiles of HPLC-purified peptides 31 and 32 (Entries 1–10) are
shown in Figures S24–S33 (see the Supporting Information).
The alkaline cleavage of the peptides synthesized on resin-
bound linkers I–V resulted in compounds with C-terminal β-
alanine (e.g., 31 in Scheme 5) or C-terminal 4-(aminomethyl)
benzoic acid groups (e.g., 32 in Scheme 5). This is consistent
with the purpose of linkers I–V, which are designed for the
stable immobilization of peptides for affinity purification of
The loadings on the linkers and the yields after cleavage of
the peptides from the resin are summarized in Table 1. Similar
loadings of linkers 5 (I)–7 (III) and 12 (IV) and 13 (V) were
achieved with both condensation agents. It seems that the
loadings of linkers 12 (IV) and 13 (V) were generally slightly
Table 1. Summary of the syntheses of compounds 31 and 32 on TentaGel resin using linkers I–V.
Compounds
31
Entry/Linker
Loading [%][c]
Purity [%][d]
Yield [%][e]
HRMS[f]
Entry (1)/5 (I)[a]
83
91
91
87
93
93
84
81
87
86
91
94
98
93
93
97
73
77
75
75
64
60
67
61
64
69
54
59
55
51
709.3527/731.3323
709.3528/731.3308
709.3525/731.3324
709.3502/731.3314
709.3531/731.3307
709.3539/731.3311
771.3664
771.3693/793.3454
771.3667
771.3662
Entry (2)/6 (II) [a]
Entry (3)/7 (III)[a]
Entry (4)/5 (I)[b]
Entry (5)/6 (II)[b]
Entry (6)/7 (III)[b]
Entry (7)/12 (IV)[a]
Entry (8)/13 (V)[a]
Entry (9)/12 (IV)[b]
Entry (10)/13 (V)[b]
31
32
32
[a] Attachment of the linker to the resin with HBTU and DIPEA or [b] PyBroP and DIPEA. [c] Yield (relative to the amount of unsubstituted resin) of the loading
of the linkers on the resin determined by spectrophotometric measurement of the adduct of dibenzofulvene and piperidine 30. [d] Relative purities of the
crude peptides determined by integration of their HPLC chromatograms. [e] Final yields of HPLC-purified peptides relative to the initial loading of resin. [f]
HRMS-MALDI, [M+H]+/[M+Na]+.
ChemPlusChem 2020, 85, 1297–1306
www.chempluschem.org
1301
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
doi.org/10.1002/cplu.202000246
ChemPlusChem
proteins or similar biological applications and not primarily for
the synthesis of peptides with a native C-terminus. It also
means that the linkers can serve as integral parts of “spacers”
separating the target ligand (e.g., a peptide) from the resin.
Separating immobilized ligands from the solid support by
spacers is a common strategy that can enhance the probability
of affinity capture, especially for larger proteins.
1
2
3
4
5
6
7
8
9
Investigation of diketopiperazine formation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
As previously mentioned (see Solution synthesis of com-
pounds), dipeptides with an ester bond at the C-terminus can
undergo spontaneous cyclization to form 6-membered rings
under basic conditions. We expected that such side reactions
would occur in the case of linkers 21–24, which are derived
from glycine and alanine. Hence, we performed model experi-
ments and tested whether the formation of diketopiperazines
28 and 29 can occur after piperidine deprotection of the first
Fmoc-amino acid attached to the linker. The mechanism for
diketopiperazine formation is depicted in Scheme 6. Thus, 21–
24 were attached to the resin via their carboxylic groups, and
this was followed by the capping of the remaining free amino
groups on the resin, Fmoc group removal with piperidine, and
elongation of the immobilized linker with FmocTyr(OMe).
Finally, the resins were treated with piperidine, and cleavage
mixtures were subjected to the work-up described in detail in
the Experimental section.
Figure 2. Superposition of the HPLC profiles (measured using Method 1, see
Experimental) of piperidine solutions collected after treatment with resin-
bound Fmoc-Tyr(OMe)-linker 24 (lower profile in red), 1-(9H-fluoren-9-
ylmethyl)piperidine 30 (middle profile in blue) and previously prepared
model diketopiperazine 29 (upper profile in green).
perazine 29 and a compound formed in the piperidine solution
after the reaction with Fmoc-Tyr(OMe)-linker 24 attached to the
resin indicate the formation of the expected byproduct. Similar
results were also obtained for linkers 21–23, and their
chromatograms are shown in Figures S34–S36, respectively (see
the Supporting Information).
Compounds with HPLC retention times identical to those of
model diketopiperazines 28 and 29 were isolated, and their
structures were confirmed by spectroscopic methods. In the
case of linker 21, the amount of 28 isolated was low, albeit
detectable, but the yields of diketopiperazine byproducts
formed from linkers 22, 23 and 24 were 43%, 54% and 44%,
respectively.
Figure 2 shows the HPLC chromatogram of a piperidine/
DMF solution that was used for deprotection of the Fmoc group
from Fmoc-Tyr(OMe) bound to linker 24 attached to the resin.
This chromatogram was superimposed on HPLC chromato-
grams of model diketopiperazine 29 and fluorenyl-piperidine
adduct 30. The identical retention times (8.8 min) of diketopi-
Finally, by analogy with the experiments described in
Scheme 6, we also attached Fmoc-Tyr(OMe) to linkers I–V, and
Scheme 6. Diketopiperazine formation on resin-bound linkers 21–24. Reagents and conditions: (a) PyBroP, DIPEA, 21–24, DMF, for 18 and 7 h; (b) Ac2O, DIPEA,
DMF, 2×15 min; (c) 20% piperidine/DMF 5 and 20 min; (d) Fmoc-Tyr(OMe)-OH, DIC, DMF, 2×1.5 h.
ChemPlusChem 2020, 85, 1297–1306
www.chempluschem.org
1302
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
doi.org/10.1002/cplu.202000246
ChemPlusChem
piperidine solution after the cleavage of the Fmoc group was
analyzed. Chromatograms shown in Figure S37 did not reveal
any traces of hypothetical diketopiperazine products. The
results clearly show that the designed linkers bearing β-alanine
(5 (I)–7 (III)) or 4-(aminomethyl)benzoic acid (12 (IV) and 13 (V))
species in their structures prohibit the formation of diketopiper-
azines.
stability is optimal for the purposes described in the Introduc-
tion; sufficient lability in 0.1 M NaOH enabling mild cleavage of
compounds but high stability up to pH 8.0, enabling biological
applications. Therefore, we tested the stability of linker 6 (II) in
aqueous ammonia (Figure 3F). The advantage of cleavage with
aqueous ammonia is that it can be easily evaporated without
any contamination by salts, as is the case for aqueous NaOH.
The experiment showed that aqueous ammonia can cleave the
peptide from linker 6 (II) in a time-dependent manner after 24 h
at room temperature. MS analysis after RP-HPLC isolation of two
peaks (Figure 4) revealed two products: a C-terminal carbox-
amide (peptide 31, entry 5, carboxamide, approximately 75%,
HRMS-ESI calculated for C31H49O10N9Na [M+Na]+730.3495,
found 730.3502) and C-terminal carboxylic acid (peptide 31,
entry 5, carboxylic acid, approximately 25%, HRMS-ESI calcu-
lated for C31H49O11N8 [M+H]+709.3515, found 709.3511). The
lower yield of peptides cleaved with ammonia compared to
that cleaved with 0.1 M NaOH (Figure 3F) could be simply
caused by a difference in the amounts of wet resin used for the
stability experiments.
1
2
3
4
5
6
7
8
9
Stability of linkers
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Linkers 5 (I), 6 (II), 7 (III), 12 (IV) or 13 (V) were attached to
TentaGel® S NH2 Resin, and the model peptide, Tyr-Ala-Lys-Gly-
Glu-Ala, was synthesized on each of the linkers. The stability of
each of the ester bonds in the linkers was investigated in
aqueous solutions with different pH values: 50 mM citrate/
NaOH at pH 6.1; 50 mM potassium phosphate at pH 6.5, 7.0, 7.5
and 8.0; 200 mM sodium bicarbonate/sodium carbonate at
pH 8.7; and 50 mM glycine/NaOH at pH 10.5. The resin was
incubated in each of these buffers, and the release of the
cleaved products (peptide 31 for linkers 5–7 or peptide 32 for
linkers 12 and 13) into the supernatant was monitored by HPLC
(see Methods). The results were compared to the effects of Conclusion
aqueous 100 mM NaOH under the same conditions (Figure 3).
According to the results shown in Figures 3A-3E, the pH stability
of linkers 5 (I), 6 (II), 7 (III), 12 (IV) and 13 (V) in aqueous buffers
can be ranked approximately as follows (from the most stable
to the least stable): 13 (V)>12 (IV) ~7 (III)>6 (II)>5 (I). This
stability order can be explained by considering the chemical
reactivities of the respective compounds. Alkyl benzoate 13 (V)
is the most stable under aqueous basic conditions due to the
electron donating effect of its benzene ring. The stability of
12(IV) is ranked second because its sensitivity to nucleophilic
attack is increased more by the “leaving group ability” of the
methyl carboxamide group. Linker 7 (III), which has a stability
similar to that of 12, is composed of more reactive alkanoic
acid, but its effect is counterbalanced by the electron-donating
4-(carboxamide)phenylmethyl group. Alkyl alkanoate 6 (II)
showed the penultimate reactivity, and finally, the most reactive
linker is 5 (I) because this compound, in addition to being an
alkanoic acid, also contains an “easily leaving” methyl carbox-
amide group.
Linker 13 (V) is completely stable up to pH 8.0; only
approximately 10% of the peptide was released at pH 10.5 after
48 h and even 0.1 M NaOH did not quantitatively cleave the
peptide from the resin during the first 30 min (Figure 3E).
Linkers 7 (III) (Figure 3C) and 12 (IV) (Figure 3D) share similar
pH stabilities and show negligible cleavage up to pH 8 and
approximately 15% cleavage at pH 8.7. Linker 6 (II) is stable up
to pH 8.0, only marginally cleaved at pH 8.7 and cleaved to only
approximately 25% at pH 10.5 (Figure 3B). The most pH-labile
linker, 5 (I) (Figure 3A), was fully cleaved at pH 10.5 after 10 h
and cleaved to approximately 70% at pH 8.7 after 48 h.
However, 5 (I) was still relatively stable up to pH 8 (with
approximately 10% cleavage after 24 h).
In conclusion, we have presented a systematic study in which
we prepared and characterized the properties of five new acid-
stable and diketopiperazine formation-resistant ester linkers, 5
(I), 6 (IV), 12 (IV) and 13 (V), and a previously prepared linker, 7
(III). The limited solubility of new linker 14 (IV) precluded its
further evaluation. The main features of these linkers can be
summarized in the following points. (i) Linkers 5 (I), 6 (IV), 7 (III),
12 (IV) and 13 (V) can be easily prepared by a straightforward
3-step procedure on a multigram scale and in high yields. (ii) All
the linkers can be easily and effectively attached to TentaGel S
NH2 resin and are fully compatible with Fmoc/tBu protocol
solid-phase peptide synthesis protocols, and the ester bonds of
all the linkers are stable in 20% piperidine and in concentrated
TFA. (iv) Model peptides synthesized on all investigated linkers
provided crude products with high purities and high yields. (v)
The relative pH stabilities of the linkers in aqueous buffers was
as follows: 13 (V)>12 (IV)=7 (III)>6 (II)>5 (I), with linker 13
(V) being the most stable. This stability order is consistent with
the chemical reactivities of the respective ester groups. (vi) All
the linkers can be effectively cleaved with 0.1 M NaOH. (vi) All
the linkers, except 5 (I), which is the most base-labile, are stable
up to pH 7.5 in water, and linker 6 (II) is stable up to pH 8. This
allows their use in the testing of immobilized peptides under
standard biological conditions. (vii) We showed that the best
linker, 6 (II), can also be cleaved with aqueous ammonia, which
allows the mild release of a peptide from a resin without any
contaminating salts. These properties of the linkers prepared in
this study, and especially of linker 6 (II), make them useful for
solid-phase peptide synthesis when the stable on-resin immobi-
lization of peptides, the testing of their biological properties in
aqueous buffers at neutral pH and finally, their release from the
solid support under non-destructive conditions are necessary.
Based on the above described stabilities of the linkers, we
selected linker 6 (II) for further characterization because its
ChemPlusChem 2020, 85, 1297–1306
www.chempluschem.org
1303
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
doi.org/10.1002/cplu.202000246
ChemPlusChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 3. Time-, pH- and solvent-dependent release of Tyr-Ala-Lys-Gly-Glu-Ala-β-Ala peptide attached to TentaGel S NH2 resins through linkers 5 (I) (in A), 6 (II)
(in B and F), 7 (III) (in C), 12 (IV) (in D) or 13 (V) (in E). The following solvents were used: 50 mM citrate/NaOH at pH 6.1 (red); 50 mM potassium phosphate at
pH 6.5 (orange), 7.0 (light green), 7.5 (green) and 8.0 (dark green); 200 mM sodium bicarbonate/sodium carbonate at pH 8.7 (light blue); 50 mM glycine/NaOH
at pH 10.5 (dark blue); 0.1 M aqueous NaOH (black) and aqueous ammonia (blue circle), based on the integration of two peaks, see text. The chemical
structures of linkers with cleavable ester bonds highlighted in red are also shown.
compounds were visualized by exposure to UV light at 254 nm, by
ninhydrin spraying (stains amines dark blue), and by spraying with
Experimental Section
a 1% (v/v) ethanolic solution of 4-(4-nitrobenzyl)pyridine followed
by heating and treating with gaseous ammonia (stains esters or
General Chemistry
alkylating compounds blue). Flash chromatography purifications
were carried out on silica gel (40–63 μm, Fluka).
Unless otherwise stated, the reagents and solvents used in this
study were obtained from commercial suppliers (Sigma-Aldrich,
Fluka, and Merck) and were used without purification. If needed,
Analytical RP-HPLC was carried out using Method 1 at a flow rate of
1 ml/min on a C8 column (Macherey-Nagel 120–5 C8; 250×4 mm,
5 μm) using the following gradient: t=0 min (20% B), t=30 min
(100% B), t=31 min (20% B), or using Method 2 on a C18 column
°
some solvents were distilled at 50 C and 2 kPa, and the products
were dried over phosphorus pentoxide at rt and 13 Pa. TLC was
performed on silica gel-coated aluminum plates (Fluka). The
ChemPlusChem 2020, 85, 1297–1306
www.chempluschem.org
1304
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
doi.org/10.1002/cplu.202000246
ChemPlusChem
supernatant were taken after various time periods and diluted with
1
2
3
4
5
185 μl of water. The samples (100 μl) were the analyzed by RP HPLC
on a C18 analytical column (Watrex Nucleosil 120-5C18, 250×
4 mm) using a linear gradient (v/v) of acetonitrile in water with
0.1% TFA (v/v) (0 min–1.6% CH3CN, 30 min–16% CH3CN at 1 ml/
min). The chromatograms were evaluated at 218 nm, and the peaks
of the products were integrated.
6
7
8
9
Acknowledgements
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
This work was supported by the European Regional Development
Fund; OP RDE; Project: “Chemical Biology for drugging undrug-
gable targets (ChemBioDrug)” (No. CZ.02.1.01/0.0/0.0/16_019/
0000729) and by the Academy of Sciences of the Czech Republic
(Research Project RVO:6138963, support to the Institute of Organic
Chemistry and Biochemistry).
Figure 4. HPLC profile of aqueous ammonia cleavage solutions combined
from Tyr-Ala-Lys-Gly-Glu-Ala-β-Ala-linker 6 (II)-resin. Chromatographic sepa-
ration was performed on a preparative C18 column (Watrex Nucleosil 120-
5C18, 250×21 mm) using a linear gradient (v/v) of acetonitrile in water with
0.1% TFA (v/v) (0 min–1.6% CH3CN, 30 min–16% CH3CN at 9 ml/min). The
chromatograms were obtained at 276 nm. The compound that eluted in
peak 1 was identified as peptide 31 (entry 5) carboxamide, and the
compound that eluted in peak 2 was identified as peptide 31 (entry 5) with
a free C-terminal carboxylic group.
Conflict of Interest
The authors declare no conflict of interest.
(Nucleosil 120 C18; 250×4 mm) using the following gradient: t=
0 min (0% B), t=30 min (60% B), t=31 min (0% B). Preparative RP-
HPLC was carried out using Method 3 on a C4 column (Grace Vydac
C4 214TP; 250×22 mm, 10–15 μm) at a flow rate of 9 ml/min. The
following gradient was used: t=0 min (10% B), t=30 min (100%
B), and t=31 min (10% B). In all cases, solvent A was 0.1% TFA and
solvent B: 80% CH3CN, 0.1% TFA. The eluted compounds were
detected at 218 nm.
Keywords: diketopiperazines · esters · peptides · protecting
groups · solid-phase synthesis
[1] a) M. Wilchek, T. Miron, J. Kohn, Methods Enzymol. 1984, 104, 3–55; b) M.
Urh, D. Simpson, K. Zhao, Methods Enzymol. 2009, 463, 417–438.
[2] a) J. Taunton, C. A. Hassig, S. L. Schreiber, Science 1996, 272, 408–411;
b) M. Collinsova, C. Castro, T. A. Garrow, A. Yiotakis, V. Dive, J. Jiracek,
Chem. Biol. 2003, 10, 113–122; c) S. E. Ong, X. Y. Li, M. Schenone, S. L.
Schreiber, S. A. Carr, Methods Mol. Biol. 2012, 803, 129–140.
[3] a) K. S. Lam, M. Lebl, V. Krchnak, Chem. Rev. 1997, 97, 411–448; b) R. W.
Liu, X. C. Li, K. S. Lam, Curr. Opin. Chem. Biol. 2017, 38, 117–126; c) T.
Kodadek, P. J. McEnaney, Chem. Commun. 2016, 52, 6038–6059; d) T.
Kodadek, N. G. Paciaroni, M. Balzarini, P. Dickson, Chem. Commun. 2019,
55, 13330–13341; e) Z. P. Gates, A. A. Vinogradov, A. J. Quartararo, A.
Bandyopadhyay, Z.-N. Choo, E D. Evans, K. H. Halloran, A. J. Mijalis, S. K.
Mong, M. D. Simon, E. A. Standley, E. D. Styduhar, S. Z. Tasker, F. Touti,
J. M. Weber, J. L. Wilson, T. F. Jamison, B. L. Pentelute, Proc. Natl. Acad.
Sci. USA 2018, 115, E5298–E5306; f) P. Dickson, T. Kodadek, Org. Biomol.
Chem. 2019, 17, 4676–4688; g) T. M. Doran, P. Dickson, J. M. Ndungu, P.
Ge, I. Suponitsky-Kroyter, H. An, T. Kodadek, Methods Enzymol. 2019,
622, 91–127.
Melting points were determined on a Boetius block and are
uncorrected. 1H and 13 C NMR spectra were acquired on a Bruker
AVANCE-600 spectrometer (1H at 600.13 MHz, 13 C at 150.9 MHz) in
CDCl3 or DMSO-d6 at 300 K. The 2D HÀ H COSY, 2D HÀ C HSQC and
2D HÀ C HMBC spectra were recorded and used for the structural
assignment of proton and carbon signals. IR spectra were recorded
on a Bruker IFS 55 Equinox apparatus. HRMS data were obtained on
an FTMS mass spectrometer LTQ-orbitrap XL (Thermo Fisher,
Bremen, Germany) in electrospray ionization mode, or in the case
of HRMS (EI), data were acquired on a GCT Premier (Waters). The
purity of each target compound was confirmed by elemental
analysis (C, H, N), and the results agreed with the calculated values
within 0.3%. The determination of C, H, and N in solid samples was
performed using an automatic PE 24000 series II CHNS/O analyzer.
[4] M. Lebl, V. Krchnak, Methods Enzymol. 1997, 289, 336–392.
[5] G. B. Fields, R. L. Noble, Int. J. Pept. Protein Res. 1990, 35, 161–214.
[6] a) I. W. James, Tetrahedron 1999, 55, 4855–4946; b) F. Guillier, D. Orain,
M. Bradley, Chem. Rev. 2000, 100, 2091–2157.
[7] a) R. J. T. Mikkelsen, K. E. Grier, K. T. Mortensen, T. E. Nielsen, K. Qvortrup,
ACS Comb. Sci. 2018, 20, 377–399; b) M. Rinnova, M. Novakova, V.
Kasicka, J. Jiracek, J. Pept. Sci. 2000, 6, 355–365.
Synthetic details and analytical data for compounds 1–30 and
methodology for the solid-phase synthesis of model peptides are
provided in the Supporting Information.
Determination of the stability of the linkers
[8] J. U. Peters, S. Blechert, Synlett 1997, 348–350.
Linkers 5 (I), 6 (II), 7 (III), 12 (IV) or 13 (V) were separately attached
to TentaGel S NH2 resin (Rapp Polymere Gmbh, 130 μm, 0.27 mmol/
g), and the same peptide sequence (Tyr-Ala-Lys-Gly-Glu-Ala) were
synthesized on each of the linkers as described above. After
completion of the synthesis, the resins were thoroughly washed
with deionized water until the rinsate was neutral with pH indicator
paper. The excess water was removed. Approximately 100 mg of
the wet resin was weighed to test tubes. Then, 1.5 ml of the
selected buffer (see above in the Results section), aqueous 0.1 M
NaOH or aqueous ammonia (24% w/w) was added to the resin. The
test tubes were shaken at room temperature. Aliquots (15 μl) of the
[9] a) M. Gongora-Benitez, J. Tulla-Puche, F. Albericio, Acs Comb Sci 2013,
15, 217–228; b) N. Jung, M. Wiehn, S. Brase, Top. Curr. Chem. 2007, 278,
1–88.
[10] a) E. Atherton, M. J. Gait, R. C. Sheppard, B. J. Williams, Bioorg. Chem.
1979, 8, 351–370; b) A. M. Bray, N. J. Maeji, A. G. Jhingran, R. M. Valerio,
Tetrahedron Lett. 1991, 32, 6163–6166; c) A. M. Bray, A. G. Jhingran, R. M.
Valerio, N. J. Maeji, J. Org. Chem. 1994, 59, 2197–2203; d) S. C. Story, J. V.
Aldrich, Int. J. Pept. Protein Res. 1992, 39, 87–92.
[11] a) S. M. Dankwardt, Synlett 1998, 761–761; b) B. Blankemeyermenge, M.
Nimtz, R. Frank, Tetrahedron Lett. 1990, 31, 1701–1704.
[12] J. W. van Nispen, J. P. Polderdijk, H. M. Greven, Recl. Trav. Chim. Pays-Bas
1985, 104, 99–100.
ChemPlusChem 2020, 85, 1297–1306
www.chempluschem.org
1305
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
doi.org/10.1002/cplu.202000246
ChemPlusChem
[13] a) E. Atherton, N. L. Benoiton, E. Brown, R. C. Sheppard, B. J. Williams, J.
Chem. Soc. Chem. Commun. 1981, 336–337; b) P. Sieber, Tetrahedron
Lett. 1987, 28, 6147–6150.
[14] B. F. Gisin, R. B. Merrifield, J. Am. Chem. Soc. 1972, 94, 3102–3106.
[15] R. M. Valerio, M. Benstead, A. M. Bray, R. A. Campbell, N. J. Maeji, Anal.
Biochem. 1991, 197, 168–177.
A. Grandas, M. A. Saralegui, E. Giralt, C. Granier, J. Vanrietschoten,
Tetrahedron 1982, 38, 1183–1192; d) E. Pedroso, A. Grandas, X.
Delasheras, R. Eritja, E. Giralt, Tetrahedron Lett. 1986, 27, 743–746.
1
2
3
4
[16] J. Picha, B. Fabre, M. Budesinsky, J. Hajduch, M. Abdellaoui, J. Jiracek,
Eur. J. Org. Chem. 2018, 5180–5192.
[17] a) M. Rothe, J. Mazánek, Liebigs Ann. Chem. 1974, 439–459; b) E. Giralt,
R. Eritja, E. Pedroso, Tetrahedron Lett. 1981, 22, 3779–3782; c) E. Pedroso,
5
6
7
Manuscript received: March 27, 2020
Revised manuscript received: May 28, 2020
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
ChemPlusChem 2020, 85, 1297–1306
www.chempluschem.org
1306
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim