An improved procedure for the preparation of aminomethyl polystyrene resin and its use in solid phase (peptide) synthesis

Article abstract of DOI:10.1016/j.tetlet.2011.09.010

2-Aminoethanol was used to successively replace hydrazine in the preparation of aminomethyl polystyrene resin thereby facilitating purification and by-product removal. The syntheses of the polypeptides ACP (65-74) and oxytocin demonstrated that the use of aminomethyl polystyrene resin prepared in this manner was equal to or better than that prepared using the hydrazine method.

Full text of DOI:10.1016/j.tetlet.2011.09.010

Tetrahedron Letters 52 (2011) 6024–6026
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
An improved procedure for the preparation of aminomethyl polystyrene
resin and its use in solid phase (peptide) synthesis
⇑
⇑
Paul W. R. Harris , Sung Hyun Yang, Margaret A. Brimble
School of Chemical Sciences, University of Auckland, 23 Symonds St, Auckland 1142, New Zealand
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 July 2011
Revised 23 August 2011
Accepted 2 September 2011
Available online 10 September 2011
2-Aminoethanol was used to successively replace hydrazine in the preparation of aminomethyl polysty-
rene resin thereby facilitating purification and by-product removal. The syntheses of the polypeptides
ACP (65–74) and oxytocin demonstrated that the use of aminomethyl polystyrene resin prepared in this
manner was equal to or better than that prepared using the hydrazine method.
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Solid phase
Aminomethyl resin
2-Aminoethanol
Peptides
Peptide synthesis
The pioneering work of Merrifield on peptide synthesis using an
insoluble solid support has revolutionized the field of peptide
chemistry.1 Solid phase peptide synthesis (SPPS) is arguably the
first choice method for synthesizing a polypeptide of any length
as simple, robust protocols can be easily executed under standard
laboratory conditions. While SPPS has undergone many refine-
ments, such as the development of coupling reagents and micro-
wave enhancement, the critical factor to successful syntheses is
arguably the properties of the solid support.2–5
One of the most popular solid matrices used for SPPS is a co-poly-
mer of polystyrene (PS) and 1% divinylbenzene (DVB) abbreviated as
1% DVB–PS. This resin is commonly functionalized with either a suit-
able handle, such as an aminomethyl (AM) (NH2CH2–) group or pre-
loaded with the desired linker and first amino acid as required for
Fmoc- or Boc-SPPS. We and others have shown, however, that
batch-to-batch variation of commercially supplied resins is a signif-
icant problem and ultimately results in poor quality peptides.6–9 We
and others have also previously established that polypeptides pre-
pared using ‘in house synthesized’ AM-PS resin dramatically im-
proved the quality of the synthetic polypeptides.6,8
prolonged reaction times and difficulty in by-product removal. In
the case of the widely used phthalimidomethylation procedure,
the ensuing dephthalimidomethylation uses hydrazine to unmask
the aminomethyl handle, and is accompanied by formation of the
highly insoluble by-product, 2,3-dihydrophthalazine-1,4-dione
(3a), and consequently requires extensive washing with large vol-
umes of organic solvents (hot or boiling) to ensure its complete
removal.19
From an environmental and safety viewpoint this procedure is
undesirable. Furthermore, replacement of the toxic and potentially
explosive hydrazine with a more benign reagent is advantageous.
We herein describe a significant improvement in the prepara-
tion of aminomethyl polystyrene resin (AM-PS resin) whereby
the phthalimidomethylation–dephthalimidomethylation reaction
times have been reduced, by-products are easily removed with
minimal washing steps, and hydrazine has been substituted by less
toxic reagents. We also demonstrate that AM-PS synthesized using
this new procedure enables the preparation of several polypeptides
with good recovery and in comparable purity to those obtained
using AM-PS prepared by the existing methods using hydrazine.
The N-phthalimidomethyl resin 2 was prepared in multigram
(5 g) amounts using N-(hydroxymethyl)phthalimide (1) and
methanesulfonic acid (MsOH) as catalyst in dichloromethane.6
Incorporation of the phthalimide was confirmed by IR spectro-
Several methods have been previously reported for introducing
an aminomethyl handle (or masked form thereof) on PS resin includ-
ing phthalimidomethylation,10–15 chloromethylation,16 amidome-
thylation,16 acetamidomethylation,17 and co-polymerization18
although all of these methods suffer from drawbacks, such as
scopic analysis of resin 2 showing a C@O absorption at 1717 cmÀ1
.
For introduction of the phthalimide moiety use of the inexpen-
sive methanesulfonic acid is preferred as it is easier to handle than
hydroscopic trifluoromethanesulfonic acid.10 Furthermore the use
of a strong acid scavenges any undesired pre-existing functionality
⇑
Corresponding authors. Tel.: +64 9 3737 599; fax: +64 9 3737 422.
E-mail addresses: paul.harris@auckland.ac.nz (Paul W. R. Harris), m.brimble@-
auckland.ac.nz (M.A. Brimble).
0040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.09.010

Paul W. R. Harris et al. / Tetrahedron Letters 52 (2011) 6024–6026
6025
Table 1
Aminolysis of N-phthalimidomethyl PS resin 2
Entry Reaction conditions
Aminolysis
Resin washing
%
1a
2b
3b
4b
5b
6a
5% NH2NH2, ethanol,
reflux, 5 h
5% PhNHNH2, ethanol,
reflux, 7 h
5% BnNH2, ethanol, reflux, 50
3 h
5% BnNH2, ethanol, reflux, 60
6 h
20% BnNH2, ethanol,
reflux, 7 h
20% HOCH2CH2NH2,
ethanol, reflux, 7 h
100
Hot EtOH (400 mL), hot MeOH, (400 mL), DMF (200 mL), CH2Cl2 (200 mL), CH2Cl2–TFA (100 mL, 1:1, v/v), CH2Cl2
(200 mL), 10% DIPEA in DMF (200 mL), DMF (200 mL), CH2Cl2 (200 mL)
n/a
None
n/a
n/a
100
n/a
100
1 Â EtOH, 1 Â DMF, 1 Â CH2Cl2 (100 mL each)
a
5 g of resin 2 used.
1 g of resin 2 used.
b
on the starting polystyrene resin. The level of substitution is esti-
mated by the amount (mmol) of N-(hydroxymethyl)phthalimide
used per gram of resin and proceeds to completion in ꢀ5 h. Impor-
tantly, no decrease in swelling properties of the N-phthalimidom-
ethyl resin 2 is noticed20 indicating that no additional cross links
have been introduced which would hinder diffusion of reagents
during SPPS.
We next examined the reaction of resin 2 with various amine
nucleophiles such as phenylhydrazine, benzylamine, or 2-amino-
ethanol in refluxing ethanol. These reagents were chosen as they
are expected not only to react rapidly with the N-phthalimidom-
ethyl group, but the resultant by-products containing aromatic or
alkoxy groups were expected to be more soluble in organic sol-
vents facilitating removal by filtration. The results of this study
are listed in Table 1 and illustrated in Scheme 1.21 For a direct com-
parison to existing protocols, hydrazinolysis using hydrazine was
also undertaken.
Using 5% (v/v) hydrazine to affect hydrazinolysis was rapid, but
the resultant by-product, 2,3-dihydrophthalazine-1,4-dione (3a),
required extensive washing to completely remove it from the resin
(entry 1). Substitution of hydrazine with phenylhydrazine failed to
give any reaction as judged by analysis of the resin by IR spectros-
copy (entry 2). Replacing hydrazine with benzylamine required a
more concentrated solution (20%, v/v) to enable complete aminol-
ysis within 7 h (entries 3–5). However, benzylamine, a known
lachrymator, was deemed to be unsuitable hence it was replaced
with 2-aminoethanol resulting in complete aminolysis (entry 6)
in a similar timeframe (7 h). Moreover, following completion of
the reaction, the resin was only washed minimally with ethanol,
dimethylformamide, and finally dichloromethane effecting total
removal of the highly soluble by-product 3c. The absence of 3c
was confirmed by IR analysis of a dried resin sample which did
not exhibit a carbonyl (C@O) stretch. The loading capacities of res-
ins 3, 4, and 5 were 0.86, 0.75, and 0.87 mmol/g, respectively, as
determined by elemental analysis.
The suitability of each aminomethyl resin prepared using either
benzylamine or 2-aminoethanol was evaluated using Fmoc SPPS.
The so-called ‘difficult’ peptide sequence ACP (65–74), containing
a C-terminal acid and the peptide hormone oxytocin, containing
a C-terminal amide were chosen as representative polypeptides
for this purpose (Fig. 1). As control experiments, both ACP (65–
74) and oxytocin, were concurrently synthesized using amino-
methyl PS resin derived from the hydrazine protocol.
1% DVB-PS
O
N
MeSO3H/CH2Cl2
HO
r.t./5 h
O
1
Each resin (0.1 g), functionalized with [4-[(R,S)-a-[1-(9H-fluo-
O
ren-9-yl]methoxycarbonylamino]-2,4-dimethoxy]phenoxyacetic
acid (Rink linker)22 or Fmoc-Gly-hydroxymethylphenoxy propionic
acid (HMPP linker)23 to deliver a C-terminal amide or acid, respec-
tively, was elongated using 20% piperidine (v/v in DMF) as the
Fmoc deblocking reagent and HBTU as the coupling agent in the
presence of DIPEA. Identical batches of reagents and identical
protocols for deprotection and coupling were utilized and for
uniformity, machine-assisted synthesis was employed.24 Upon
completion of the synthesis, cleavage with trifluoroacetic acid/tri-
isopropylsilane/water (95:2.5:2.5, v/v) for ACP (65–74) or trifluoro-
acetic acid/triisopropylsilane/water/ethanedithiol (95:1:2.5:2.5, v/
v) for oxytocin was performed for 2 h. Crude products were precip-
itated using ether, isolated by centrifugation, dissolved in aqueous
~1 mmol/g
N
loading
2
O
NH2CH2CH2OH
EtOH/reflux
BnNH2 EtOH
reflux
NH2NH2/EtOH
reflux
NH2
NH2
NH2
3
5
4
+
O
+
+
O
O
HN
HN
ACP (65-74) :
NH2-Val-Gln-Ala-Ala-Ile-Asp-Tyr-Ile-Asn-Gly-OH
N
N
Bn
HO
O
O
O
3a
Oxytocin :
3c
3b
NH2-Cys-Tyr-Ile-Gln-Asn-Cys-Pro-Leu-Gly-NH2
Scheme 1. Preparation of aminomethyl resin by hydrazinolysis or aminolysis.
Figure 1. Amino acid sequences of test peptides oxytocin and ACP (65–74).

6026
Paul W. R. Harris et al. / Tetrahedron Letters 52 (2011) 6024–6026
Table 2
Preparation of ACP (65–74) or oxytocin using amino methyl resins 3, 4, or 5
Resin
Scale (mmol)
Peptide
Calculated weight of
peptide resin (mg)
Weight of peptide
resin (mg)
Calculated weight
of crude product (mg)
Weight of crude
product (mg)
Purity (%)
3a
4b
5c
3a
4b
5c
0.086
0.075
0.087
0.086
0.075
0.087
ACP (65–74)
ACP (65–74)
ACP (65–74)
Oxytocin
Oxytocin
Oxytocin
256
236
258
299
274
301
248
225
252
276
263
288
91.4
79.9
92.5
86.7
75.6
87.8
74.5
66.1
82.5
75.3
65.3
82.2
86
89
86
79
75
86
a
b
c
Loading 0.86 mmol/g.
Loading 0.75 mmol/g.
Loading 0.87 mmol/g.
and oxytocin afforded crude products in excellent purity and good
recovery which compared well to resin 3 obtained from the hydra-
zine protocol. We have used resin 5 exclusively for the preparation
of many (>50) polypeptides by Fmoc- or Boc-SPPS, the results of
which will be communicated in the subsequent publications.
Acknowledgment
The authors wish to thank The Maurice Wilkins Centre for
Molecular Biodiscovery for the financial support (P.W.R.H.).
References and notes
Figure 2. RP-HPLC of crude ACP (65–74) synthesized on aminomethyl resins 3, 4, or
5. Column C18 Gemini 2.0 Â 50, gradient 5–30% B (1% B/min), A = 0.1% TFA–H2O,
B = 0.1% TFA–MeCN.
1. Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149–2154.
2. Pugh, K. C.; York, E. J.; Stewart, J. M. Int. J. Pept. Prot. Res. 1992, 40, 208–213.
3. MacDonald, A. A.; Dewitt, S. H.; Ghosh, S.; Hogan, E. M.; Kieras, L.; Czarnik, A.
W.; Ramage, R. Mol. Diversity 1996, 1, 183–186.
4. Meldal, M. In Solid Phase Peptide Synthesis; Fields, C. G., Ed.; Academic Press:
New York, 1988; pp 83–104.
5. Yan, B. Comb. Chem. High Throughput Screening 1988, 215–229.
6. Harris, P. W. R.; Brimble, M. A. Synthesis 2009, 3460–3466.
7. Bouillon, I.; Soural, M.; Miller, M. J.; Krchnak, V. J. Comb. Chem. 2009, 11, 213–
215.
8. Deng, F. K.; Mandal, K.; Luisier, S.; Kent, S. B. H. J. Pept. Sci. 2010, 16, 545–550.
9. Alewood, D.; Hopping, G.; Brust, A.; Reid, R. C.; Alewood, P. F. J. Pept. Sci. 2010,
16, 551–557.
10. Mitchell, A. R.; Kent, S. B. H.; Engelhard, M.; Merrifield, R. B. J. Org. Chem. 1978,
43, 2845–2852.
11. Meisenbach, M.; Allmendinger, T.; Mak, C. P. Org. Process Res. Dev. 2003, 7, 553–
558.
12. Adams, J. H.; Cook, R. M.; Hudson, D.; Jammalamadaka, V.; Lyttle, M. H.;
Songster, M. F. J. Org. Chem. 1998, 63, 3706–3716.
Figure 3. RP-HPLC of crude oxytocin synthesized on aminomethyl resins 3, 4, or 5.
Column C18 Gemini 2.0 Â 50, gradient 5–30% B (1% B/min), A = 0.1% TFA–H2O,
B = 0.1% TFA–MeCN.
13. Ljungdahl, N.; Martikainen, L.; Kann, N. Tetrahedron Lett. 2008, 49, 6108–6110.
14. Zikos, C. C.; Ferderigos, N. G. Tetrahedron Lett. 1995, 36, 3741–3744.
15. Zikos, C.; Alexiou, G.; Ferderigos, N. Tetrahedron Lett. 2006, 47, 8711–8715.
16. Mitchell, A. R.; Kent, S. B. H.; Erickson, B. W.; Merrifield, R. B. Tetrahedron Lett.
1976, 3795–3798.
acetonitrile, and lyophilized. The results are given in Table 2 and
Figures 2 and 3.
17. Lee, T. K.; Ryoo, S. J.; Byun, J. W.; Lee, S. M.; Lee, Y. S. J. Comb. Chem. 2005, 7,
170–173.
18. Garibay, P.; Toy, P. H.; Hoeg-Jensen, T.; Janda, K. D. Synlett 1999, 1438–1440.
19. Ref. 9 used MeNH2 for de-phthalimidomethylation resulting in a soluble by-
product but requiring 16 h to complete.
20. Swelling (in CH2Cl2) of 1% DVB–PS is 8 mL/g, swelling of 2 was 8.1 mL/g.
21. N-Phthalimidomethyl resin 2 was prepared as outlined in Ref. 5 A 250 mL
round-bottomed flask was charged with N-phthalimidomethyl resin 2 (6.02 g),
20% (v/v) ethanolamine in absolute EtOH (AR, 100 mL) and the mixture is
refluxed by stirring for 7 h. The resulting mixture was cooled to room
temperature and washed successively with EtOH (100 mL), DMF (100 mL),
and CH2Cl2 (100 mL) and dried in vacuo to afford resin 5 (5.01 g, 94%) as a
white solid. IR analysis showed an absence of carbonyl absorptions. Elemental
analysis gave 1.22% N corresponding to a loading of 0.87 mmol/g.
22. Rink, H. Tetrahedron Lett. 1987, 28, 3787–3790.
For aminomethyl resins 3, 4, and 5 the weight of the peptide re-
sin following assembly by SPPS was comparable to that antici-
pated. The quantities recovered and the purities of the crude
products, obtained after cleavage from the resin, as estimated by
analytical HPLC (Figs. 2 and 3), were also similar although the high-
est recoveries were found for aminomethyl resin 5. Therefore,
based on the data presented in Table 1, it is evident that either
of the dephthalimidomethylation methods using benzylamine or
2-aminoethanol can be used confidently for the preparation of
AM-PS resin.
In conclusion, we have demonstrated that for the preparation of
AM-PS resin by phthalimidomethylation–dephthalimidomethyla-
tion on polystyrene resin, the common dephthalimidomethylation
reagent hydrazine can be replaced effectively by 2-aminoethanol
or benzylamine. Significantly, the use of 2-aminoethanol or benzyl-
amine enabled the resultant aminomethyl-functionalized polysty-
rene resin to be easily separated from highly soluble by-products
by filtration using minimal solvent washing. Fmoc-SPPS of amino-
methyl resins 4 or 5 for the synthesis of the peptides ACP (65–74)
23. Albericio, F.; Barany, G. Int. J. Pept. Prot. Res. 1985, 26, 92–97.
24. General procedure for peptide synthesis with aminomethyl resins: For ACP (65–
74); resin
5 (0.1 g) was swelled in CH2Cl2 and reacted with FmocGly-
OCH2PhOCH2CH2CO2H (2 equiv) and diisopropylcarbodiimide (DIC) (2 equiv)
in CH2Cl2 for 1 h. The Kaiser test was negative. For oxytocin; resin 5 (0.1 g) was
swelled in DMF and reacted with Rink linker (3 equiv), HOBt (3 equiv), and
diisopropylcarbodiimide (DIC) (3 equiv) in DMF for 1 h. The Kaiser test was
negative. All Fmoc-SPPS was performed using a Tribute 2 channel peptide
synthesizer (Protein Technologies, Tucson, Az) using a 2 Â 5 min deprotection
with 20% (v/v) piperidine/DMF, followed by 5 Â 30 s DMF washes. Coupling
was performed for 30 min with Fmoc-Aa (5 equiv), HBTU (4.6 equiv) and DIPEA
(10 equiv) followed by a 5 Â 30 s DMF wash.