Acid-mediated prevention of aspartimide formation in solid phase peptide synthesis

Article abstract of DOI:10.1021/ol3007925

Aspartimide formation is one of the major obstacles that impedes the solid phase synthesis of large peptides and proteins. Until now, no cost-effective strategy to suppress this side reaction has been developed. Here it is demonstrated that addition of sm

Full text of DOI:10.1021/ol3007925

ORGANIC
LETTERS
2012
Vol. 14, No. 20
5218–5221
Acid-Mediated Prevention of Aspartimide
Formation in Solid Phase Peptide Synthesis
†
Tillmann Michels, Rudolf Dolling, Uwe Haberkorn, and Walter Mier*
‡
†
,†
€
Department of Nuclear Medicine, University Hospital Heidelberg, INF 400,
€
69120 Heidelberg, Germany, and BIOSYNTAN GmbH, Robert-Rossle-Strasse 10,
13125 Berlin, Germany
walter.mier@med.uni-heidelberg.de
Received August 22, 2012
ABSTRACT
Aspartimide formation is one of the major obstacles that impedes the solid phase synthesis of large peptides and proteins. Until now, no cost-
effective strategy to suppress this side reaction has been developed. Here it is demonstrated that addition of small amounts of organic acids to
the standard Fmoc cleavage agent piperidine efficiently prevents formation of aspartimide side products. This effect is shown to be virtually
independent of the acid strength.
Advances in molecular biology have paved the way for
patient-specific therapies. Many of these novel drugs are
proteins, which unfortunately show unfavorable pharma-
cokinetic behavior as a consequence of their high molec-
ular weight. The downsizing of such proteins to their
effective chemophore, such as an antibody to its epitope
recognition site,1 would provide an ideal strategy for
closing the gap between drugs of a molecular weight below
500 Da, corresponding to the traditional understanding of
druglikeness, and today’s proteinaceous drug candidates.
Furthermore, this strategy would be applicable for chemi-
cal synthesis techniques, thereby avoiding complications
that arise with drug approval of recombinant proteins.
Solid phase peptide synthesis (SPPS), originally devel-
oped by Merrifield,2 is a key technology for drug synthesis
within the downsizing development strategy. The synthesis
of large peptides by SPPS can be achieved by chemical
ligation; however, side reactions can hamper the overall
synthesis leading to decreased yields and/or quality.
Aspartimide is a major pitfall in Fmoc-based SPPS.3 Its
formation, which can either be acid or base catalyzed,
occurs while the piperidine-catalyzed Fmoc cleavage of
peptides containing aspartic acid.4 The propensity of
aspartimide formation mainly depends on the aspartate
carboxyl neighboring residue.5 Aspartimide is, therefore,
the result of an attack of an amidate species at the carbonyl
carbon of the OtBu protected side chain carboxylate of
aspartic acid (Scheme 1).
Many approaches have been developed to suppress the
formation of aspartimides. Mergler et al. achieved signifi-
cant improvements by using sterically demanding Asp
side chain protection groups such as 3-methylpent-3-yl
(OMpe)6 and β-2,3,4-trimethyl-pent-3-yl.7 Replacement
€
(3) Yang, Y.; Sweeney, W. V.; Schneider, K.; Thronqvist, S.; Chait,
B. T.; Tam, J. P. Tetrahedron Lett. 1994, 35, 9689–9692.
€
(4) Dolling, R.; Beyermann, M.; Haenel, J.; Kernchen, F.; Krause,
E.; Franke, P.; Brudel, M.; Bienert, M. J. Chem. Soc., Chem. Commun.
1994, 38, 853–854.
(5) (a) Lauer, J. L.; Fields, C. G.; Fields, G. B. Lett. Pept. Sci. 1995, 1,
197–205. (b) Mergler, M.; Dick, F.; Sax, B.; Stahelin, C.; Vorherr, T.
† University Hospital Heidelberg.
‡ BIOSYNTAN GmbH.
€
J. Pept. Sci. 2003, 9, 518–526.
(6) Mergler, M.; Dick, F.; Sax, B.; Weiler, P.; Vorherr, T. J. Pept. Sci.
2003, 9, 36–46.
(1) (a) Nedwidek, M. N.; Hecht, M. H. Proc. Natl. Acad. Sci. U.S.A.
1997, 94, 10010–10011. (b) Bray, B. L. Nat. Rev. Drug Discovery 2003, 2,
587–593.
(2) Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149–2154.
(7) Mergler, M.; Dick, F. J. Pept. Sci. 2005, 11, 650–657.
r
10.1021/ol3007925
Published on Web 10/01/2012
2012 American Chemical Society

for the absence of piperidide formation due to steric
hindrance preventing the attack of piperidine.14
To determine the influence of acids on the formation
of aspartimide in the deprotection step, PreS9-33-y was
synthesized either with or without 5% formic acid (v/v in
piperidine). The synthesis under standard conditions (no
formic acid) led to a massive occurrence of aspartimide
(Figure 1).
Scheme 1. Proposed Mechanism of Aspartimide Formation and
Its Piperidinium Ion Mediated Suppression
ofpiperidine (pKa = 11.12) withthe milderbase piperazine
(pKa = 9.73) can also reduce aspartimide formation,
however, at the cost of the reaction rate.8 This can be
overcome by microwave heating;9 however this must be
carefully controlled to avoid other side reactions.10 Com-
plete prevention of aspartimide formation can be achieved
using N-(2-hydroxy-4-methoxybenzyl) (Hmb) as a back-
bone protecting group.11 However, Hmb protected amino
acids show low coupling efficiencies and so must be used as
dipeptide building blocks. Furthermore, Hmb-protected
building blocks are difficult to synthesize and only the
dipeptide containing glycine (Fmoc-Asp(tBu)-(Hmb)Gly)
is commercially available. The addition of hydroxybenzo-
triazole (HOBt) to the piperidine deprotection agent has been
shown to slightly reduce the formation of aspartimide in
Fmoc-based SPPS. However, commercial HOBt hydrate
contains ∼12% water, and its explosive potential together
with the restricted availibility and light sensitivity would cause
problems for its routine application in SPPS. Interestingly,
2,4-dinitrophenol12 and ethyl 2-cyano-2-(hydroxyimino)-
acetate (Oxyma)13 have also been described as additives for
reducing aspartimide formation. This effect might be attrib-
uted to their acidic character. Consequently, suppression of
aspartimide formation by adding small amounts of organic
acids to the deprotection agent piperidine was studied.
PreS9-33-y, a 26-mer peptide derived from the HBV sur-
face antigen (NPLGFFPDHQLDPAFRANTANPDWDy-
NH2), was used to analyze aspartimide formation. This
peptide contains three sites that are prone to aspartimide
formation (Asp-D-Tyr, Asp-Trp, and Asp-His). The fourth
Asp containing motif (Asp-Pro) is hindered due to the
steric nature of proline. The motif Asp-Trp is not suscep-
tible to aspartimide formation. As Asp-X motifs contain-
ing a D-amino acid are highly prone to aspartimide forma-
tion, this is the most probable site of formation. The close
proximity of this motif to the resin may be an explanation
Figure 1. Analysis of aspartimide formation during solid phase
synthesis. Reversed-phase HPLC of crude PreS9-33-y synthe-
sized under (A) standard conditions, (B) using Fmoc-Asp-
(OMpe)-OH and (C) with 5% formic acid.
In a second synthesis, the peptide was synthesized using
Fmoc-Asp(OMpe)-OH. Under these conditions HPLC
analysis revealed a decrease in aspartimide formation,
although high levels of aspartimide remained (50% as
compared to 81%). Finally, addition of 5% formic acid
to piperidine significantly reduced aspartimide formation
(13% aspartimide). Thus, addition of formic acid was
superior to the usage of the sterical hindered side chain
protecting group OMpe.
(8) Wade, J.; Mathieu, M.; Macris, M.; Tregear, G. Lett. Pept. Sci.
2000, 7, 107–112.
(9) Palasek, S. A.; Cox, Z. J.; Collins, J. M. J. Pept. Sci. 2007, 13, 143–148.
(10) Nissen, F.; Kraft, T. E.; Ruppert, T.; Eisenhut, M.; Haberkorn,
U.; Mier, W. Tetrahedron Lett. 2010, 51, 6216–6219.
(11) (a) Quibell, M.; Owen, D.; Packman, L. C.; Johnson, T. J. Chem.
Soc., Chem.Commun. 1994, 2317–2421. (b) Sampson, W. R.; Patsiouras,
H.; Ede, N. J. J. Pept. Sci. 1999, 5, 403–409. (c) Offer, J.; Quibell, M.;
Johnson, T. J. Chem. Soc., Perkin Trans. 1 1996, 175–182.
(12) (a) Lauer, J. L.; Fields, C. G.; Fields, G. B. Lett. Pept. Sci. 1994,
1, 197–205. (b) Martinez, J.; Bodanszky, M. Int. J. Pept. Protein Res.
1978, 12, 277–283.
Figure 2. Analysis of aspartimide formation by piperidine treat-
ment of resin bound PreS9-33-y. RP-HPLC analyses of a sample
incubated with 20% piperidine (A), 20% piperidine with
0.1 equiv formic acid (B), and the untreated control (C).
ꢀ
(13) Subiros-Funosas, R.; El-Faham, A.; Albericio, F. Biopolymers
2012, 98, 89–97.
Org. Lett., Vol. 14, No. 20, 2012
5219

To simulate deprotection conditions of prolonged syn-
thesis, the resin bound peptide, synthesized in the presence
of formic acid, was incubated with piperidine (20% v/v in
DMF) alone or with piperidine containing 0.1 equiv of
formic acid at 50 °C for 60 min. Under these conditions,
treatment with piperidine alone caused the formation of
59% aspartimide (Figure 2). Addition of 0.1 equiv of for-
mic acid, on the other hand, completely prevented any
further aspartimide formation (13% aspartimide).
A series of 18 different acids covering the range of pKa
values of common organic acids (0.1 equiv each) were
added to 20% piperidine in DMF to establish whether
prevention of aspartimide formation depends on the acid
used. Analytic HPLC analysis did not show a correlation
between the pKa and the efficiency of aspartimide suppres-
sion (Table 1). With apparent independence from their
pKa values, all acids were shown to suppress aspartimide
Table 2. Concentration Dependent Suppression of Piperidine
Mediated Aspartimide Formation of Resin Bound PreS9-33-y
by Different Acids
relative yields [%]
mol equiv
acid
relative to piperidine producta aspartimidea
formic acid
1
93.6
91.2
90.6
91.6
87.8
87.8
90.6
90.1
88.8
86.2
85.6
80.8
92.7
92.3
90.3
88.7
87.1
82.2
92.5
88.6
74
6.4
0.5
0.2
0.1
0.05
0.01
1
8.8
9.4
8.4
12.2
12.2
9.4
HOBt
0.5
0.2
0.1
0.05
0.01
1
9.9
11.2
13.8
14.4
19.2
7.3
4-nitrophenol
Table 1. Piperidine-Catalyzed Aspartimide Formation of Resin
Bound PreS9-33-y under the Influence of Different Organic
Acids
0.5
0.2
0.1
0.05
0.01
1
7.7
9.7
11.3
12.9
17.8
7.5
relative yields [%]
producta
aspartimidea
HFIP
acid
TFMSAb
pKa1, pKa2
0.5
0.2
0.1
0.05
0.01
0
11.4
26
À13
0.30
0.77
0.7
85.1
86.8
85.3
88.9
90.1
84.1
88.4
87.3
87.3
88.2
88.2
88.9
87.9
87.2
67.9
88.1
88.2
41.3
88.2
8.1
TFA
13.2
14.7
11.1
9.9
64.4
54.3
60
35.6
45.7
40
trichloroacetic acid
PTSA
dichloroacetic acid
taurine
1.25
1.5, 8.74
2.87
3.75
4.09
4.17, 11.6
4.2
no acid
41
59
15.9
11.6
12.7
12.7
11.8
11.8
11.1
12.2
12.8
32.1
11.9
11.8
58.7
11.8
a Determined by RP-HPLC (UV detection at 214 nm).
chloroacetic acid
formic acid
2,4-dinitrophenol
ascorbic acid
benzoic acid
HOBt
identify optimal concentions for solid phase protein synth-
esis (Table 2). While low concentrations of formic acid,
hydroxybenzotriazole (HOBt), and 4-nitrophenol drama-
tically suppressed aspartimide formation, HFIP did not
show a significant effect, indicating that effects other
than the pKa (values determined from aqueous media)
are responsible for this phenomenon. A strong base, 1,8-
diazabicyclo[5.4.0]undec-7-en (DBU), was tested to simu-
late the even harsher Fmoc cleavage conditions (2% DBU
and 20% piperidine in DMF) used to deprotect cleavage-
resistant Fmoc groups.15 This led to 67% aspartimide
(compared to 17% with formic acid). Using DBU, the
piperidide, which was not observed in the syntheses with
pure piperidine, was formed in ∼1.6% yield as determined
by integration of the respective LC/MS signals. Using
0.1 equiv of formic acid, only 0.4% of the piperidine
adduct could be detected. Furthermore, a second asparti-
mide was formed after treatment with DBU (26%), which
was not visible in the sample treated with additional formic
acid (0.6%). Thus, small amounts of acid prevented as-
partimide formation even under very harsh deprotection
conditions. Small amounts of formic acid (5% v/v in
4.6
acetic acid
4.75
7.2
4-nitrophenol
HFIP
9.6
phenol
10.0
10.1
benzenesulfonamide
20% piperidine
untreated
a Determined by RP-HPLC (UV detection at 214 nm). b Trifluor-
omethanesulfonic acid (TFMSA) caused the formation of side products
besides aspartimide.
formation compared to the untreated sample. Citric acid
and oxalic acid formed precipitates in the cleavage solu-
tion. Attempts to use inorganic acids such as HCl and
sulfuric acid failed due to the low solubility of their
piperidinium salts in piperidine. Interestingly, the amino
sulfonic acid taurine gave results similar to those obtained
with the common organic acids. The concentration depen-
dency was determined using different equivalents of formic
acid, HOBt, 4-nitrophenol, and hexafluoroisopropanol to
(15) (a) Tickler, A. K.; Barrow, C. J.; Wade, J. D. J. Pept. Sci. 2001, 7,
488–494. (b) Srivastava, K.; Davis, M. Adv. Exp. Med. Biol. 2009, 611,
585–591.
ꢀ
(14) Subiros-Funosas, R.; El-Faham, A.; Albericio, F. Tetrahedron
2011, 67, 8595–8606.
5220
Org. Lett., Vol. 14, No. 20, 2012

Even low amounts of acid (e.g., 0.05 equiv) prevented
aspartimide formation.
The peptide parathyroid hormone (PTH) is known to be
difficult to synthesize by SPPS.16 To test the efficiency of
aspartimide suppression strategy described above, a com-
perative synthesis of a PTH fragment (46À84 = AGSQ-
RPRKKEDNVLVESHEKSLGEADKADVNVLTKA-
KSQ-NH2) was performed using standard conditions or
additionof5% formicacidtothe piperidine. Massspectro-
metry showed the formation of aspartimide at all three
possible sites in the peptide synthesized under standard
conditions (Figure 3). As expected, the addition of 5%
formic acid reduced aspartimide formation by ∼90%, as
determinedby integrationof the respectiveLC/MS signals.
Today, extensive synthesis strategies are required to
avoid aspartic acid in drug developmental candidates.
The described method therefore provides an efficient
means of preventing aspartimide formation and therefore
presents an interesting alternative tothe usage of backbone
protected dipeptides, the gold standard for prevention of
aspartimide side products. In contrast to the commercialy
available dipeptides that are limited to specific motifs, this
method can be used for all Asp containing peptides.
In conclusion, a straightforward, efficient, and cost-
effective method for the prevention of aspartimide during
deprotection in Fmoc-based solid phase peptide synthesis is
described. This study clearly showed that the addition of
organic acids to the piperidine cleavage solution does not have
any limitations. Since aspartic acid is found in most oligopep-
tides, the authors recommend that, in general, 5% (v/v)
formic acid (fresh baches to avoid eventual side reactions)
should be added to piperidine based Fmoc cleavage mixtures.
Figure 3. Product region of the UPLC-MS analysis of the crude
reaction product of the SPPS of parathyroid (46À84) amide.
Total ion current obtained with (A) and without the addition of
acid (B). As the aspartimides eltue near the product, the positive
ion mode ESI-MS mass spectrum obtained at the front of the
product peak shows strong aspartimide signals. While the
reaction with acid (C) leads to a significant aspartimide reduc-
tion in favor of the product, the aspartimides dominate after
reaction without acid (D).
piperidine) almost completely prevented aspartimide for-
mation in the PreS9-33-y peptide. Under normal condi-
tions, this peptide was prone to significant amounts of
aspartimide. Simulation of prolonged incubation under
harsh conditions confirmed those results. Furthermore,
it was shown that this was not formic acid specific, but a
general effect of acids during piperidine driven deprotection.
With the exception of TFMSA and HFIP, all acids
(0.1 equiv) reduced the level of aspartimide as comparred
to the untreated control. It can therefore be concluded that
this is due to a decrease of the amount of nucleophiles
of adjacent amides by protonation of the amidate form.
Acknowledgment. This work was supported by the
€
Bundesministerium fur Bildung und Forschung (Grant
No. 13N10269) and the Deutsche Forschungsgemeinschaft
(Grant No. HA 2901/6-1).
Supporting Information Available. Materials and meth-
ods; reversed-phase HPLC and HPLC-MS data. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(16) Goud, N. A.; McKee, R. L.; Sardana, M. K.; DeHaven, P. A.;
Huelar, E.; Syed, M. M.; Goud, R. A.; Gibbons, S. W.; Fisher, J. E.;
Levy, J. L.; Rodkey, J. A.; Bennett, C.; Ramjit, H. G.; Caporale, L. H.;
Claudfield, M. P.; Rosenblatt, M. J. Bone Miner. Res. 1991, 6, 781–891.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 20, 2012
5221