Synthesis of aromatic glycoconjugates. building blocks for the construction of combinatorial glycopeptide libraries

Article abstract of DOI:10.3762/bjoc.10.256

New aromatic glycoconjugate building blocks based on the trifunctional 3-aminomethyl-5-aminobenzoic acid backbone and sugars linked to the backbone by a malonyl moiety were prepared via peptide coupling. The orthogonally protected glycoconjugates, bearing an acetyl-protected glycoside, were converted into their corresponding acids which are suitable building blocks for combinatorial glycopeptide synthesis.

Full text of DOI:10.3762/bjoc.10.256

Synthesis of aromatic glycoconjugates. Building blocks for
the construction of combinatorial glycopeptide libraries
Markus Nörrlinger and Thomas Ziegler*
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2014, 10, 2453–2460.
Institute of Organic Chemistry, University of Tuebingen, Auf der
Morgenstelle 18, 72076 Tuebingen, Germany
doi:10.3762/bjoc.10.256
Received: 13 August 2014
Accepted: 10 October 2014
Published: 22 October 2014
Email:
Thomas Ziegler* - thomas.ziegler@uni-tuebingen.de
* Corresponding author
Associate Editor: S. Flitsch
Keywords:
© 2014 Nörrlinger and Ziegler; licensee Beilstein-Institut.
License and terms: see end of document.
amino acids; aniline; carbohydrates; glycoconjugates; glycopeptides
Abstract
New aromatic glycoconjugate building blocks based on the trifunctional 3-aminomethyl-5-aminobenzoic acid backbone and sugars
linked to the backbone by a malonyl moiety were prepared via peptide coupling. The orthogonally protected glycoconjugates,
bearing an acetyl-protected glycoside, were converted into their corresponding acids which are suitable building blocks for combi-
natorial glycopeptide synthesis.
Introduction
Glycans or other complex oligosaccharide structures, present on appears to be a useful tool to investigate, for instance, specific
the surface of every prokaryotic and eukaryotic cell, are impor- carbohydrate–protein or carbohydrate–carbohydrate interac-
tant for a large number of biological recognition processes like, tions. Recently our group has prepared a series of trifunctional
for example, intercellular communication, signal transduction, glycopeptide building blocks with aliphatic backbones, which
pathogen recognition or immunological responses [1-4]. In allow for the automated construction of combinatorial libraries
order to investigate these processes it is essential that a large of highly divers glycopeptides suitable for studying carbohy-
amount of the respective polysaccharide structure is available. drate–protein interactions [5-8]. Hitherto, our focus was on
Unfortunately, isolation of pure oligosaccharides from natural glycosylated amino acid building blocks derived from aspartic
sources is difficult due to the micro heterogenity of naturally acid and from the PNA-like N-(2-aminoethyl)glycine (AEG)
occurring saccharides. For this reason chemical oligosaccharide backbone to which the sugar moieties were attached through
synthesis is the only alternative for providing sufficient amounts either simple alkyl chains [5,6], amino alcohols [7,8] or 1,2,3-
of pure material for detailed biological studies. However, the triazoles [9-11]. These building blocks were used for combina-
synthetic preparation of complex oligosaccharides is still diffi- torial solid phase or automated spot synthesis of libraries of
cult despite the great achievements in this field during the past highly glycosylated peptides and shown to specifically bind to
decades. Therefore, the application of oligosaccharide mimetics lectins [5,8,10]. Here, we now describe the preparation of a
which may be synthesized more easily in larger amounts series of glycopeptide building blocks which allow for the
2453

Beilstein J. Org. Chem. 2014, 10, 2453–2460.
construction of glycopeptide libraries with an aromatic back- methyl ester in 3 with aqueous LiOH solution in THF afforded
bone based on 3-aminomethyl-5-aminobenzoic acid to which the corresponding carboxylic acid 4 in 96% yield. Next, acid 4
the sugar moieties are attached through a malonyl linker. For was converted into tert-butyl ester 5 in 89% overall yield by a
that purpose we designed the two building blocks 1 and 2 two-step procedure via the corresponding intermediate acid
(Figure 1) which both can be converted into the respective chloride (Scheme 1). Selective reduction of the azido group in 5
glycopeptides using standard Fmoc strategies [10].
without affecting the nitro group was achieved with a
Staudinger reaction [14]. Thus, treatment of 5 with triphenyl-
phosphine in aqueous THF gave tert-butyl 3-aminomethyl-5-
nitrobenzoate which was not isolated but immediately
converted into the corresponding Fmoc-protected derivative 6
in 78% overall yield. Final hydrogenation of the latter with
Lindlar catalyst afforded 7 almost quantitatively. Likewise,
copper(I)-catalyzed 1,3-dipolar cycloaddition (Click reaction)
[15-18] of 5 with Fmoc-protected propargylamine afforded first
t-butyl benzoate 8 in 87% yield. Hydrogenation of the latter
with Pd on charcoal then gave 9 in 88% yield. It should be
noted that hydrogenation of 6 and 8 had to be carefully opti-
mized with respect to the reaction conditions in order to
completely suppress the hydrogenation of the Fmoc group
(Scheme 1).
Figure 1: Malonyl-linked aromatic glycoconjugate building blocks for
spot synthesis of combinatorial glycopeptides libraries.
For the construction of the two desired glycopetide building
blocks 1 and 2 we needed a series of 1-malonylamidoglycopy-
ranoses 12 for condensation with the aromatic backbone
building blocks 7 and 9. For that purpose we chose four sugar
Results and Discussion
Our synthesis of building blocks 1 and 2 started from known ligands in the gluco, galacto, N-acetylglucosamine and galac-
3-azidomethyl-5-nitrobenzoic acid methyl ester 3 which was tosamine series which were prepared as outlined in Table 1.
prepared from commercially available dimethyl 5-nitroiso- Glycosylamines 10a–d were prepared from the corresponding
phthalate in 64% overall yield [12,13]. Saponification of the glycosylazides by hydrogenation according to known pro-
Scheme 1: Synthesis of the aromatic backbone building blocks 7 and 9.
2454

Beilstein J. Org. Chem. 2014, 10, 2453–2460.
cedures [19-23]. Next, glycosylamines 10 were condensed with Finally, glycopeptide building blocks 1 and 2 were prepared as
tert-butyl malonate [24], N,N,N′,N′-tetramethyl-O-(1H-benzotri- follows (Table 2). Both Fmoc-protected benzoic acid deriva-
azol-1-yl)uronium hexafluorophosphate (HBTU) and triethyl- tives 7 and 9 (Scheme 1) were each condensed with each of the
amine in THF to give monosaccharides 11 in 53–76% yield. four malonylamidoglycosides 12a–d (Table 1) to afford eight
The medium yields in case of 11b–d (Table 1, entries 2–4) were intermediate tert-butyl esters 13 and 14 in 45–67% yield. For
due to some decomposition during chromatographic purifica- the condensation step between 7 and 9 with 12, respectively, we
tion. Nevertheless, in our hands, HBTU was superior to other chose 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDCI)
peptide coupling reagents because it resulted in the highest instead of HBTU as the coupling reagent because the side prod-
yields of compounds 11. HBTU was also previously used for uct (1-(3-(dimethylamino)propyl)-3-ethylurea) released from
the preparation of a similar t-butyl succinate of glucosamine EDCI was easily removed from the crude reaction mixture by
[25]. Upon treatment with trifluoroacetic acid in dichloro- washing with aqueous citric acid. The medium yields for these
methane, esters 11 were converted into the corresponding free condensation steps were due to some difficulties to completely
acids 12 in 91–98% yield (Table 1).
separate the products from concomitant minor unidentified
Table 1: Synthesis of glycosides 11 and 12.
Entry
Starting Material [ref]
Products
Yield 11 (%) Yield 12 (%)
1
11a 76
11b 56
11c 56
11d 53
12a 91
12b 96
12c 98
12d 98
10a [19,20]
R = t-Bu 11a
R = H 12a
2
3
4
10b [20]
R = t-Bu 11b
R = H 12b
10c [21]
R = t-Bu 11c
R = H 12c
10d [22,23]
R = t-Bu 11d
R = H 12d
2455

Beilstein J. Org. Chem. 2014, 10, 2453–2460.
Table 2: Synthesis of glycopeptide building blocks 1 and 2.
Entry
Products 13, 14
Yield (%)
Products 1, 2
Yield (%)
1
54
95
13a
13b
13c
1a
1b
1c
2
57
90
3
48
68
2456

Beilstein J. Org. Chem. 2014, 10, 2453–2460.
Table 2: Synthesis of glycopeptide building blocks 1 and 2. (continued)
4
5
6
7
8
61
94
13d
14a
14b
14c
14d
1d
62
45
67
65
92
84
91
94
2a
2b
2c
2d
2457

Beilstein J. Org. Chem. 2014, 10, 2453–2460.
byproducts which showed similar mobilities during chromato- libraries we chose glucose-containing derivatives 1a and 2a for
graphic purification. Next, the tert-butyl ester groups of com- an exemplified preparation of the corresponding fully protected
pounds 13 and 14 were hydrolysed with a 2:1 mixture of formic dimers. For comparison reasons and the possibility to later
acid and dichloromethane at room temperature to give the construct glycopeptides containing non-glycosylated chain
corresponding free acids, i.e., building blocks 1a–d and 2a–d in links, we also prepared two dipeptides from nitro-benzoates 6
68–95% yield (Table 2).
and 8 in the following way (Scheme 2). Treatment of 6 and 8
with a 2:1 mixture of formic acid and dichloromethane at room
In order to demonstrate that building blocks 1 and 2 are indeed temperature for 38 h, as described for the preparation of
suitable for the construction of combinatorial glycopeptide building blocks 1 and 2 (see Table 2), afforded the benzoic acid
Scheme 2: Synthesis of dimers 17, 20, 22 and 24.
2458

Beilstein J. Org. Chem. 2014, 10, 2453–2460.
derivatives 15 and 18 in 93% yield for each. Likewise, treat- Acknowledgements
ment of 5 and 8 with 20% piperidine in DMF at room tempera- We thank Dr. Dorothee Wistuba for the measurement of the
ture for 3.5 h gave crude aminomethyl compounds 16 and 19 high resolution mass spectra, Dr. Markus Kramer and the
which were used for the next step without further purification. members of the NMR-division of the Institute of Organic
Final coupling of 15 with 16 and 18 with 19 using HBTU, Chemistry for recording the NMR spectra and Petra Krüger for
1-hydroxybenzotriazole (HOBt) and diisopropylethylamine performing the elemental analyses. We also thank Dr. Gregor
(DIPEA) in DMF as the condenation agent gave non-glyco- Lemanski for numerous discussions on the topic.
sylated fully protected dipeptides 17 and 20 in 73% and 88%
References
1. Benett, H. S. J. Histochem. Cytochem. 1963, 11, 14–23.
yield, respectively (Scheme 2). Likewise, the Fmoc protecting
groups in glucosylated building blocks 13a and 14a were first
doi:10.1177/11.1.14
removed with piperidine in DMF to give crude aminomethyl
2. Lee, Y. C.; Lee, R. T. Acc. Chem. Res. 1995, 28, 321–327.
derivates 21 and 23. Next, the latter were coupled with 1a (for
doi:10.1021/ar00056a001
21) and 2a (for 23) under similar conditions as described for the
3. Dwek, R. A. Chem. Rev. 1996, 96, 683–720. doi:10.1021/cr940283b
non-glycosylated counterparts above to afford glycol-dipep-
4. Solis, D.; Romero, A.; Menendez, M.; Jimenez-Barbero, J. In The
tides 22 and 24 in 51% and 55% yield, respectively. As was
observed for EDCI-promoted preparation of glycosylated
derivatives 13 and 14 (see Table 2), the yields of the two gluco-
dipeptides were only in the medium range due to traces of
unidentified byproducts which could not easily removed during
chromatographic purification. Nevertheless, the only medium
yields in this case can be circumvented in solid phase
syntheses of such glycopeptides where an excess of one
building block can be applied and no chromatographic
purification is necessary. Such combinatorial solid phase
syntheses with the building blocks described here are now
underway.
Sugar Code; Gabius, H.-J., Ed.; Wiley-Blackwell: Weinheim, 2009;
p 233.
5. Ziegler, T.; Röseling, D.; Subramanian, L. R. Tetrahedron: Asymmetry
2002, 13, 911–914. doi:10.1016/S0957-4166(02)00212-4
6. Daiber, R.; Ziegler, T. ARKIVOC 2013, No. ii, 408–420.
7. Schips, C.; Ziegler, T. J. Carbohydr. Chem. 2005, 24, 773–788.
doi:10.1080/07328300500326859
8. Ziegler, T.; Schips, C. Nat. Protoc. 2006, 1, 1987–1994.
doi:10.1038/nprot.2006.307
9. Pietrzik, N.; Schips, C.; Ziegler, T. Synthesis 2008, 519–526.
doi:10.1055/s-2008-1032150
10.Günther, K.; Schips, C.; Ziegler, T. J. Carbohydr. Chem. 2008, 27,
446–463. doi:10.1080/07328300802419873
11.Günther, K. U.; Ziegler, T. Synthesis 2014, 46, 2362–2370.
doi:10.1055/s-0033-1339137
12.Harris, T. D.; Rajopadhye, M.; Damphousse, P. J.; Glowacka, D.;
Yu, K.; Bourque, J. P.; Barrett, J. A.; Damphousse, D. J.;
Heminway, S. J.; Lazewatsky, J.; Mazaika, T.; Carroll, T. R.
Bioorg. Med. Chem. Lett. 1996, 6, 1741–1746.
Conclusion
We have described the preparation of a series of new aromatic
glycopeptoids and have demonstrated their usefulness for the
preparation of corresponding glycosylated or non-glycosylated
dipeptides. The benzoic acid derived building blocks described
here will be used in combination with previously described
similar glycopeptoids based on asparaginic acid or PNA-like
backbones for automated SPOT syntheses of combinatorial
glycopeptide libraries which will be screened for their capa-
bility to bind to specific proteins [5-11]. Those results will be
published elsewhere.
doi:10.1016/0960-894X(96)00282-X
13.Watzke, A.; Gutierrez-Rodriguez, M.; Köhn, M.; Wacker, R.;
Schroeder, H.; Breinbauer, R.; Kuhlmann, J.; Alexandrov, K.;
Niemeyer, C. M.; Goody, R. S.; Waldmann, H. Bioorg. Med. Chem.
2006, 14, 6288–6306. doi:10.1016/j.bmc.2006.05.006
14.Staudinger, H.; Meyer, J. Helv. Chim. Acta 1919, 2, 635–646.
doi:10.1002/hlca.19190020164
15.Rostovtsev, V. V.; Green, L.; Fokin, V. V.; Sharpless, K. B.
Angew. Chem., Int. Ed. 2002, 41, 2596–2599.
doi:10.1002/1521-3773(20020715)41:14<2596::AID-ANIE2596>3.0.CO
;2-4
16.Tornoe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002, 67,
3057–3064. doi:10.1021/jo011148j
Supporting Information
17.Bock, V. D.; Hiemstra, H.; van Maarseveen, J. H. Eur. J. Org. Chem.
2006, 51–68. doi:10.1002/ejoc.200500483
Supporting Information File 1
Experimental data.
18.Dedola, S.; Nepogodiev, S. A.; Field, R. A. Org. Biomol. Chem. 2007,
5, 1006–1017. doi:10.1039/b618048p
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-10-256-S1.pdf]
19.Helferich, B.; Mitrowsky, A. Chem. Ber. 1952, 85, 1–8.
doi:10.1002/cber.19520850103
Supporting Information File 2
NMR Spectra.
20.Badia, C.; Souard, F.; Vicent, C. J. Org. Chem. 2012, 77,
10870–10881. doi:10.1021/jo302238u
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-10-256-S2.pdf]
21.Neumann, J.; Thiem, J. Eur. J. Org. Chem. 2010, 900–908.
doi:10.1002/ejoc.200901106
2459

Beilstein J. Org. Chem. 2014, 10, 2453–2460.
22.Dunstan, D.; Hough, L. Carbohydr. Res. 1972, 25, 246–248.
doi:10.1016/S0008-6215(00)82766-4
23.Tanaka, M.; Yamashina, I. Carbohydr. Res. 1973, 27, 175–183.
doi:10.1016/S0008-6215(00)82437-4
24.Tararov, V. I.; Korosteylev, A.; König, G.; Börner, A. Synth. Commun.
2006, 36, 187–191. doi:10.1080/00397910500334405
25.Fujita, Y.; Abdel-Aal, A.-B. M.; Wimmer, N.; Batzloff, M. R.;
Good, M. F.; Toth, I. Bioorg. Med. Chem. 2008, 16, 8907–8913.
doi:10.1016/j.bmc.2008.08.064
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/2.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.10.256
2460