Synthesis of C-glycosyl phosphonate derivatives of 4-amino-4-deoxy-α-L-arabinose

Article abstract of DOI:10.3762/bjoc.16.2

The incorporation of basic substituents into the structurally conserved domains of cell wall lipopolysaccharides has been identified as a major mechanism contributing to antimicrobial resistance of Gram-negative pathogenic bacteria. Inhibition of the corr

Full text of DOI:10.3762/bjoc.16.2

Synthesis of C-glycosyl phosphonate derivatives of
4-amino-4-deoxy-α-ʟ-arabinose
Lukáš Kerner and Paul Kosma*
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2020, 16, 9–14.
University of Natural Resources and Life Sciences, Vienna
Department of Chemistry, Muthgasse 18, A-1190 Vienna, Austria
doi:10.3762/bjoc.16.2
Received: 23 October 2019
Accepted: 06 December 2019
Published: 02 January 2020
Email:
Paul Kosma* - paul.kosma@boku.ac.at
* Corresponding author
Associate Editor: D. Spring
Keywords:
© 2020 Kerner and Kosma; licensee Beilstein-Institut.
License and terms: see end of document.
antibiotic resistance; glycosyl phosphonate; glycosyl transferase;
lipid A; lipopolysaccharide
Abstract
The incorporation of basic substituents into the structurally conserved domains of cell wall lipopolysaccharides has been identified
as a major mechanism contributing to antimicrobial resistance of Gram-negative pathogenic bacteria. Inhibition of the correspond-
ing enzymatic steps, specifically the transfer of 4-amino-4-deoxy-ʟ-arabinose, would thus restore the activity of cationic antimicro-
bial peptides and several antimicrobial drugs. C-glycosidically-linked phospholipid derivatives of 4-amino-4-deoxy-ʟ-arabinose
have been prepared as hydrolytically stable and chain-shortened analogues of the native undecaprenyl donor. The C-phosphonate
unit was installed via a Wittig reaction of benzyl-protected 1,5-arabinonic acid lactone with the lithium salt of dimethyl
methylphosphonate followed by an elimination step of the resulting hemiketal, leading to the corresponding exo- and endo-glycal
derivatives. The ensuing selective monodemethylation and hydrogenolysis of the benzyl groups and reduction of the 4-azido group
gave the α-ʟ-anomeric arabino- and ribo-configured methyl phosphonate esters. In addition, the monomethyl phosphonate glycal
intermediates were converted into n-octyl derivatives followed by subsequent selective removal of the methyl phosphonate ester
group and hydrogenation to give the octylphosphono derivatives. These intermediates will be of value for their future conversion
into transition state analogues as well as for the introduction of various lipid extensions at the anomeric phosphonate moiety.
Introduction
Glycosyltransferases are important enzymes that accomplish the dolichol in mammalian systems or to undecaprenol in prokary-
transfer of activated sugar phosphates onto their respective otic donor substrates for bacterial glycosyltransferases [2].
acceptor molecules [1]. In most cases, nucleotide diphosphate 4-Amino-4-deoxy-ʟ-arabinose (Ara4N) is an important micro-
sugars serve as the reactive species, but lipid-linked diphos- bial carbohydrate in bacterial lipopolysaccharides (LPS) and
phate derivatives are equally important, e.g., when connected to has been implicated in resistance mechanisms of pathogenic
9

Beilstein J. Org. Chem. 2020, 16, 9–14.
Gram-negative bacteria against antibiotics, such as We have recently set out to study the substrate specificity of
polymyxin B and colistin [3]. The main effect of Ara4N incor- ArnT enzymes in more detail using phosphodiester-linked de-
poration into the lipid A part – and, less frequently, into the rivatives in both anomeric configurations and containing short
inner core region of LPS [4] – is thought to originate from lipid appendages in order to define the minimum structural
blocking the electrostatic interaction of cationic antimicrobial requirements for Ara4N glycosyltransferase substrates [9]. In
peptides with the negatively charged phosphate and carboxyl- parallel, we have started to develop C-phosphonate analogues
ate groups in LPS domains. Suitable inhibitors intercepting the as these have frequently been exploited as potential inhibitors
attachment of Ara4N units to the lipid A and inner core region for glycosyl transferases since the carbon–phosphorus bond is
might restore sensitivity towards the cationic antimicrobial not hydrolyzed in the active site of glycosyl transferases [10-
drugs as a novel approach to combat the looming antibiotic 13]. Herein, we report on the synthesis of α-anomeric C-arabi-
crisis [5]. 4-Amino-4-deoxy-ʟ-arabinose units are activated as nosyl methylphosphonate ester derivatives as model com-
the phosphodiester-linked undecaprenyl derivative [6], which is pounds to allow for future incorporation of different lipid chains
then transferred by the action of several Ara4N transferases and options towards glycal analogues as potential transition
(ArnT, Figure 1) [7].
state analogues to inhibit 4-amino-4-deoxy-ʟ-arabinose transfer
to bacterial LPS.
The synthesis of potential inhibitors of the biosynthesis of
Ara4N and the glycosyl transfer have not been fully explored Results and Discussion
yet. Previously, Kline and co-workers reported on the synthesis The previously synthesized [14] methyl 4-azido-4-deoxy-α-ʟ-
of acetylated 4-azido-arabinose phosphate and uridine diphos- arabinopyranoside (1), available in multigram amounts, was
phate (UDP) derivatives. In addition, a 4-aminophosphoami- benzylated in 79% yield using benzyl bromide and sodium
date UDP derivative was also obtained [8]. Whereas these com- hydride in DMF, followed by hydrolysis of the anomeric
pounds were inactive towards enzyme upstream of the biosyn- aglycon of 2 with 5 M HCl and SrCl2 in acetic acid, which
thetic pathway to undecaprenyl Ara4N, the peracetylated afforded the hemiacetal 3 as a 2:3 α/β anomeric mixture in 60%
4-azido derivative showed modest reduction of Ara4N incorpo- yield (Scheme 1). Conversion of benzylated reducing sugars,
ration into the lipid A part of Salmonella typhimurium [8].
such as glucose [15], N-acetyl-ᴅ-glucosamine [16], and galac-
Figure 1: Modification of lipid A by ArnT.
10

Beilstein J. Org. Chem. 2020, 16, 9–14.
Scheme 1: Phosphonate and glycal synthesis.
tose [17], into the corresponding C-glycosyl phosponates using which gave 8 in 57% yield [27]. The Z-configuration of the
a Wittig reaction with methylenetriphenylphosphorane hexenitol unit was derived from the 3JC,P coupling constant
furnished the respective glycoenitols, which were then subject- (14.1 Hz), which was in good agreement with reported data
ed to mercuriocyclization, followed by iodination and phos- [22,28]. Next, selective O-demethylation of the phosphonate
phonate introduction by an Arbusov reaction. Alternative ap- diester was elaborated. Using sodium iodide in acetone afforded
proaches were elaborated from olefinic C-glycosides, which the mono-O-demethylated derivative 9 in 98% yield [29]. At
were converted into the corresponding C-linked hydroxymethyl this stage, the Z-configuration of the enol ether was unambigu-
derivatives and processed to give the glycosyl methylphos- ously assigned on the basis of NOESY experiments. Specifi-
phonic acid derivatives [18,19]. As the Ara4N transferase cally, NOE correlations were seen between the olefinic proton
reacted with the equatorial phosphodiester lipid, we opted to and the methylene protons of a benzyl group. The position of
use lactone 4 as a suitable precursor for C-glycosyl phos- the 2-O-benzyl group was assigned on the basis of an HMBC
phonates, based on literature precedents [10,20,21]. Thus, a correlation to position H-2 of the pyranose ring. Additional
modified Swern oxidation of lactol 3 with acetic anhydride in interactions were also found from the olefinic proton to hydro-
DMSO afforded a near-quantitative yield of lactone 4, which, gen atom H-2 and the O-methyl group, respectively. The
however, was unstable towards chromatographic purification preferred formation of Z-configured Wittig products fully
and used as crude material after lyophilization to remove the agreed with similar results in the literature for 2-O-benzyl-pro-
solvent. Coupling of 4 with the lithium salt of dimethyl tected gluco and galacto derivatives [20,27,30,31].
methylphosphonate (5) in THF afforded the β-anomeric ketol
phosphonate derivative 6 in 57% yield. The axial orientation of In addition to exocyclic glycals mimicking putative planar tran-
the anomeric hydroxy group was proven by NOESY correla- sition states of substrates involved in enzymatic reactions, such
tions between the H-2 proton and both geminal protons of the as glycosyl transfer, mutase, and epimerization, endo-glycals
CH2P unit. A strong NOE was also observed for the upfield- are also of interest [21,22,32-35].
shifted doublet of doublets at 1.71 ppm of the methylene group,
which had an additional NOE correlation with the broad signal Under basic conditions, such as by treatment of 9 with 1 M
of the anomeric OH group at 5.79 ppm. Next, the ensuing elimi- NaOH in MeOH, the exo double bond of compound 9 could be
nation step was carried out to explore the access to transition shifted to produce the corresponding endo-glycal 10 in 80%
state analogues [22] potentially mimicking the sp2 character of yield. The m/z value of the high-resolution mass spectrum indi-
the oxocarbenium intermediate in the enzymatic transfer reac- cated that 10 was an isomer of 9. In the 1H NMR (600 MHz)
tion. In addition, exo-glycals are versatile precursors for the spectrum of 10, the olefinic proton was absent, whereas the
introduction of fluoromethyl phosphonates, which are the better 13C NMR spectrum showed downfield shifts for the anomeric
bioisosters of phosphates [23-25]. Exo-glycal 8 was obtained in carbon atom (146.48 ppm) and the adjacent ring carbon atom
74% yield when using methyl oxalyl chloride [26], which (133.95 ppm). Evidence from an HMBC experiment additional-
proved to be superior to the use of trifluoroacetic anhydride, ly indicated a correlation between the benzylic protons and the
11

Beilstein J. Org. Chem. 2020, 16, 9–14.
latter carbon atom as well as signals of two upfield-shifted The HPLC-based purification step even allowed for further sep-
deoxy protons at 2.72 and 2.55 ppm, respectively, to the aration of the diastereomers on phosphorus of both compounds.
anomeric carbon atom. In addition, for compound 10, lacking
the conjugation to the phosphorus atom, a significant down- Selective methyl ester cleavage of 12 and 14 – again using NaI
field shift of the 31P NMR signal was observed (17.97 ppm in in a minimum volume of acetone – was straightforward and
10 versus 10.95 ppm in compound 9).
furnished the octyl phosphonate ester derivatives 13 and 15 in
good yields. For the future synthesis of deprotected glycal ester
Full deprotection, including the conversion of the 4-azido group derivatives containing prenyl units, benzyl groups will have to
into an amino function, was accomplished by hydrogenation of be exchanged for protecting groups that can be removed under
9 in the presence of palladium hydroxide in a 1:1 mixture of nonhydrogenolytic conditions, e.g., silyl ether protecting
methanol/acetic acid to deliver the target derivative 11 in 30% groups. In preliminary experiments, Lewis acid cleavage using
yield after final HILIC purification (Scheme 2). The equatorial BBr3, BCl3, and TiCl4, respectively, was not successful, al-
arrangement of the C-glycosyl linkage was supported from the though the latter reagent had previously been used to deprotect
large value of the coupling constant J1,2 (9.5 Hz), indicating a two benzyl groups from an allyl glycoside containing a
1,2-trans orientation of the respective protons.
4-amino-4-deoxy-ʟ-arabinose residue [36]. Cleavage of the
benzyl protecting groups with concomitant reduction of the
The monomethyl phosphonate ester 9 was then subjected to azido group afforded the α-anomeric octyl ((4-amino-4-deoxy-
alkylation reactions in order to allow for the selective introduc- ʟ-arabinopyranosyl)methyl)phosphonate 16 in 38% yield.
tion of longer-chain alkyl ester groups as better mimics of the Deprotection of the endo-glycal ester 15 was also investigated.
native prenyl-activated donor substrate. Since the reaction of 9 An intermediate enol resulting from hydrogenolytic cleavage of
with 1-bromooctane as model system involved basic conditions, the benzyl ether was envisaged to be prone to the facile forma-
mixtures of endo- and exo-glycal derivatives 12 and 14 were tion of tautomers, which would then lead to different reduction
obtained. The highest yields were obtained for reactions per- products.
formed in DMF, whereas DMSO and MeCN as solvents were
less effective. Reaction times of 2–5 h at 80 °C were sufficient, Notably, however, the reduction proved to be highly selective,
the ratio of isomers obtained, however, critically depended on and hence might have involved a fast hydrogen addition from
the work-up procedure. Addition of MeOH and silica gel in the bottom face of the pyranose ring [37]. After full hydrogena-
combination with dry loading of the column mainly produced tion and azide-to-amine conversion, compound 17 was isolated
the endo-glycal 14 as the major product (55%) as well as the as the 4-amino-4-deoxy-ʟ-ribopyranosyl derivative. The config-
exo-isomer (31%), since Cs2CO3 was still present to induce uration was determined from the NMR data. Position H-2 at
isomerization. In contrast, direct separation of a concentrated 4.02 ppm appeared as a broad doublet with small homonuclear
reaction mixture by HPLC afforded 12 as the major product coupling constants, as would be expected for a manno spin
(61%), with isolation of 14 as the minor isomer in 17% yield. system. In addition, NOESY correlations were observed be-
Scheme 2: Synthesis of methyl phosphonate 11 and octyl phosphonates 16 and 17.
12

Beilstein J. Org. Chem. 2020, 16, 9–14.
4. Needham, B. D.; Trent, M. S. Nat. Rev. Microbiol. 2013, 11, 467–481.
doi:10.1038/nrmicro3047
5. Karam, G.; Chastre, J.; Wilcox, M. H.; Vincent, J.-L. Crit. Care 2016,
20, No. 136. doi:10.1186/s13054-016-1320-7
tween atom H-3 and H-5ax as well as the anomeric proton,
which was consistent with the equatorial position of the C-phos-
phonate entity.
6. Trent, M. S.; Ribeiro, A. A.; Doerrler, W. T.; Lin, S.; Cotter, R. J.;
Raetz, C. R. H. J. Biol. Chem. 2001, 276, 43132–43144.
doi:10.1074/jbc.m106962200
7. Petrou, V. I.; Herrera, C. M.; Schultz, K. M.; Clarke, O. B.;
Vendome, J.; Tomasek, D.; Banerjee, S.; Rajashankar, K. R.;
Belcher Dufrisne, M.; Kloss, B.; Kloppmann, E.; Rost, B.; Klug, C. S.;
Trent, M. S.; Shapiro, L.; Mancia, F. Science 2016, 351, 608–612.
doi:10.1126/science.aad1172
8. Kline, T.; Trent, M. S.; Stead, C. M.; Lee, M. S.; Sousa, M. C.;
Felise, H. B.; Nguyen, H. V.; Miller, S. I. Bioorg. Med. Chem. Lett.
2008, 18, 1507–1510. doi:10.1016/j.bmcl.2007.12.061
9. Olagnon, C.; Monjaras Feria, J.; Grünwald‐Gruber, C.; Blaukopf, M.;
Valvano, M. A.; Kosma, P. ChemBioChem 2019, 20, 2936–2948.
doi:10.1002/cbic.201900349
10.Forget, S. M.; Bhattasali, D.; Hart, V. C.; Cameron, T. S.;
Syvitski, R. T.; Jakeman, D. L. Chem. Sci. 2012, 3, 1866–1878.
doi:10.1039/c2sc01077a
Conclusion
In conclusion, Wittig reactions of the suitably protected arabi-
nonic lactone allowed for the straightforward implementation of
a phosphonate dimethyl ester that could be readily converted
into exo- and endo-glycal ester derivatives. The selective
cleavage of one of the methyl groups opened various options
for the introduction of a separate alkyl chain since the
remaining methyl group was amenable to selective ester
cleavage at a later stage. The glycal ester derivatives them-
selves offered additional options for the preparation of transi-
tion state analogues but can also be fully deprotected to provide
hydrolytically stable substrate derivatives in the correct
anomeric configuration.
11.Hajduch, J.; Nam, G.; Kim, E. J.; Fröhlich, R.; Hanover, J. A.; Kirk, K. L.
Carbohydr. Res. 2008, 343, 189–195.
Supporting Information
doi:10.1016/j.carres.2007.10.027
12.Beaton, S. A.; Huestis, M. P.; Sadeghi-Khomami, A.; Thomas, N. R.;
Jakeman, D. L. Chem. Commun. 2009, 238–240.
Supporting Information File 1
Experimental section, spectral data, and copies of 1H, 13C,
and 31P NMR spectra of compounds 4, 6, and 8–17.
[https://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-16-2-S1.pdf]
doi:10.1039/b808078j
13.Gordon, R. D.; Sivarajah, P.; Satkunarajah, M.; Ma, D.; Tarling, C. A.;
Vizitiu, D.; Withers, S. G.; Rini, J. M. J. Mol. Biol. 2006, 360, 67–79.
doi:10.1016/j.jmb.2006.04.058
14.Müller, B.; Blaukopf, M.; Hofinger, A.; Zamyatina, A.; Brade, H.;
Kosma, P. Synthesis 2010, 3143–3151. doi:10.1055/s-0030-1258174
15.Nicotra, F.; Ronchetti, F.; Russo, G. J. Org. Chem. 1982, 47,
4459–4462. doi:10.1021/jo00144a011
16.Casero, F.; Cipolla, L.; Lay, L.; Nicotra, F.; Panza, L.; Russo, G.
J. Org. Chem. 1996, 61, 3428–3432. doi:10.1021/jo9519468
17.Heissigerová, H.; Kočalka, P.; Hlaváčková, M.; Imberty, A.; Breton, C.;
Chazalet, V.; Moravcová, J. Collect. Czech. Chem. Commun. 2006, 71,
1659–1672. doi:10.1135/cccc20061659
Acknowledgements
The authors thank Dr. Markus Blaukopf for help with HILIC
purification steps and NMR spectroscopic measurements and
Simon Krauter and Thomas Grasi for HRMS data.
Funding
18.Torres-Sanchez, M. I.; Zaccaria, C.; Buzzi, B.; Miglio, G.; Lombardi, G.;
Polito, L.; Russo, G.; Lay, L. Chem. – Eur. J. 2007, 13, 6623–6635.
doi:10.1002/chem.200601743
Financial support by the Austrian Science Fund FWF (Project
no. P-28826-N28) is gratefully acknowledged.
19.Chang, R.; Vo, T.-T.; Finney, N. S. Carbohydr. Res. 2006, 341,
1998–2004. doi:10.1016/j.carres.2006.05.008
20.Dondoni, A.; Marra, A.; Pasti, C. Tetrahedron: Asymmetry 2000, 11,
305–317. doi:10.1016/s0957-4166(99)00477-2
ORCID® iDs
Lukáš Kerner - https://orcid.org/0000-0001-7886-750X
Paul Kosma - https://orcid.org/0000-0001-5342-7161
21.Caravano, A.; Vincent, S. P. Eur. J. Org. Chem. 2009, 1771–1780.
doi:10.1002/ejoc.200801249
22.Stolz, F.; Reiner, M.; Blume, A.; Reutter, W.; Schmidt, R. R.
J. Org. Chem. 2004, 69, 665–679. doi:10.1021/jo0353029
23.Romanenko, V. D.; Kukhar, V. P. Chem. Rev. 2006, 106, 3868–3935.
doi:10.1021/cr051000q
Preprint
A non-peer-reviewed version of this article has been previously published
as a preprint doi:10.3762/bxiv.2019.129.v1
24.Berkowitz, D. B.; Bose, M.; Pfannenstiel, T. J.; Doukov, T.
J. Org. Chem. 2000, 65, 4498–4508. doi:10.1021/jo000220v
25.Loranger, M. W.; Forget, S. M.; McCormick, N. E.; Syvitski, R. T.;
Jakeman, D. L. J. Org. Chem. 2013, 78, 9822–9833.
doi:10.1021/jo401542s
26.Yang, W.-B.; Wu, C.-Y.; Chang, C.-C.; Wang, S.-H.; Teo, C.-F.;
Lin, C.-H. Tetrahedron Lett. 2001, 42, 6907–6910.
doi:10.1016/s0040-4039(01)01427-7
References
1. Liang, D.-M.; Liu, J.-H.; Wu, H.; Wang, B.-B.; Zhu, H.-J.; Qiao, J.-J.
Chem. Soc. Rev. 2015, 44, 8350–8374. doi:10.1039/c5cs00600g
2. Bugg, T. D. H.; Brandish, P. E. FEMS Microbiol. Lett. 1994, 119,
255–262. doi:10.1111/j.1574-6968.1994.tb06898.x
3. Baron, S.; Leulmi, Z.; Villard, C.; Olaitan, A. O.; Telke, A. A.;
Rolain, J.-M. Int. J. Antimicrob. Agents 2018, 51, 450–457.
doi:10.1016/j.ijantimicag.2017.11.017
13

Beilstein J. Org. Chem. 2020, 16, 9–14.
27.Yang, W.-B.; Yang, Y.-Y.; Gu, Y.-F.; Wang, S.-H.; Chang, C.-C.;
Lin, C.-H. J. Org. Chem. 2002, 67, 3773–3782. doi:10.1021/jo0255227
28.Eppe, G.; Dumitrescu, L.; Pierrot, O.; Li, T.; Pan, W.; Vincent, S. P.
Chem. – Eur. J. 2013, 19, 11547–11552. doi:10.1002/chem.201300969
29.Sigal, I.; Loew, L. J. Am. Chem. Soc. 1978, 100, 6394–6398.
doi:10.1021/ja00488a019
30.Molina, A.; Czernecki, S.; Xie, J. Tetrahedron Lett. 1998, 39,
7507–7510. doi:10.1016/s0040-4039(98)01627-x
31.Xie, J.; Molina, A.; Czernecki, S. J. Carbohydr. Chem. 1999, 18,
481–498. doi:10.1080/07328309908544013
32.Qian, X.; Palcic, M. M. Glycosyltransferase Inhibitors. In Carbohydrates
in Chemistry and Biology; Ernst, B.; Hart, G. W.; Sinaý, P., Eds.;
Wiley-VCH: Weinheim, 2000; pp 293–312.
doi:10.1002/9783527618255.ch56
33.Compain, P.; Martin, O. R. Bioorg. Med. Chem. 2001, 9, 3077–3092.
doi:10.1016/s0968-0896(01)00176-6
34.Schwörer, R.; Schmidt, R. R. J. Am. Chem. Soc. 2002, 124,
1632–1637. doi:10.1021/ja017370n
35.Frische, K.; Schmidt, R. R. Liebigs Ann. Chem. 1994, 297–303.
doi:10.1002/jlac.199419940312
36.Blaukopf, M.; Müller, B.; Hofinger, A.; Kosma, P. Eur. J. Org. Chem.
2012, 119–131. doi:10.1002/ejoc.201101171
37.Koester, D. C.; Kriemen, E.; Werz, D. B. Angew. Chem., Int. Ed. 2013,
52, 2985–2989. doi:10.1002/anie.201209697
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(https://creativecommons.org/licenses/by/4.0). Please note
that the reuse, redistribution and reproduction in particular
requires that the authors and source are credited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(https://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.16.2
14