Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2023 May 19;25(21):3927–3931. doi: 10.1021/acs.orglett.3c01293

Solid-Phase Synthesis of Glycosyl Phosphate Repeating Units via Glycosyl Boranophosphates as Stable Intermediates

Kazuki Sato †,*, Kazumasa Muramoto †, Tomoya Hagio †, Rintaro I Hara †,‡, Takeshi Wada †,*
PMCID: PMC10243113  PMID: 37204168

Abstract

graphic file with name ol3c01293_0007.jpg

Solid-phase synthesis of glycosyl phosphate repeating units was investigated using glycosyl boranophosphates as stable precursors. The stable nature of glycosyl boranophosphate enables the elongation of a saccharide chain without remarkable decomposition. After deprotection of the boranophosphotriester linkages to boranophosphodiesters, the intersugar linkages were converted to the phosphate counterparts quantitatively using an oxaziridine derivative. This method significantly improves the synthesis of oligosaccharides containing glycosyl phosphate units.

Leishmania is a tropical protozoan parasite that causes Leishmaniasis.1 Leishmaniasis is one of the neglected tropical diseases, and although there is a few therapeutics for Leishmaniasis,1 a fundamental means to eradicate it is highly required. Developing a vaccine against Leishmania is promising, and lipophosphoglycans (LPGs) on the Leishmania surface2 are reported to play a vital role in its survival and infectious processes.3 Notably, purified LPGs have been used as a vaccine against Leishmania major.4 Synthetic phosphoglycans conjugated with tetanus toxin fragment C protein have also shown significant protection against Leishmania infection in mice.5 LPGs have a characteristic repeating unit of [→6)β-d-Gal-(1 → 4)-α-d-Man-(1-P)]2 (Figure 1). The length of the repeating units varies from 15 to 30 in accordance with the life stage of Leishmania.6 Therefore, there is a need for a method to synthesize phosphoglycans with structural homogeneity to progress in this field.

Figure 1.

Figure 1

Open in a new tab

Structure of the glycosyl phosphate repeating units found in the LPGs of Leishmania donovani.

The traditional method of synthesizing glycosyl phosphates involves using the H-phosphonate method, where a glycosyl H-phosphonate monoester monomer unit is condensed with a hydroxy group of another molecule in the presence of a condensing reagent such as pivaloyl chloride (PivCl) and the resultant H-phosphonate diester is further oxidized to form its phosphodiester counterpart.7 However, it has been reported that the oxidation of H-phosphonate diester is prone to side reactions such as hydrolysis of the intersugar linkages.8,9 In addition, the phosphodiester can be a potential reaction site and may react with a condensing reagent, which causes undesired reactions. Although some reports have attempted to synthesize repeating units using a polycondensation approach,10,11 the length of the repeating units prepared stepwise is limited to four.12 Recently, Zhang et al. reported the α-selective synthesis of glycosyl phosphate using glycosyl o-alkynylbenzoate and phosphate as the glycosyl donor and acceptor, respectively, in the presence of a gold(I) catalyst. They effectively constructed intersugar phosphate linkages with good stereoselectivity.13 Another approach is to use a glycosyl phosphoramidite monomer, which can be condensed with a hydroxy group to form a phosphite triester and then oxidized to a phosphotriester.7 In nucleic acid chemistry, this method is the most widely used due to the high coupling efficiency.14 However, as one can see from the fact that glycosyl phosphotriesters and phosphite triesters are used as potent glycosyl donors under acidic conditions,15,16 the liability of these moieties should be taken into consideration for the synthesis of glycosyl phosphate derivatives. To address this issue, our group has shown that the substitution of a 2-position hydroxy group with a fluorine atom prevents the degradation of glycosyl phosphates and phosphites, probably due to its electron-withdrawing nature, which prohibits the formation of oxocarbenium cations.17 In this study, two strategies were developed to synthesize glycosyl phosphate repeating units in LPGs of Leishmania: (1) introduction of the electron-withdrawing group as a hydroxy protecting group and (2) utilization of glycosyl boranophosphates as stable precursors of glycosyl phosphates. As for (1), the preliminary investigation revealed that the use of the benzoyl (Bz) protecting group did not prevent the decomposition of a glycosyl phosphite and/or phosphate (data not shown). Thus, the o-chlorobenzoyl group was chosen as a strong electron-withdrawing group. Regarding (2), Prosperi et al.18 and our group19 have shown that a glycosyl boranophosphate, in which one of nonbridging oxygen atoms of a phosphate is replaced with a borano group, has substantial stability under acidic conditions. Additionally, glycosyl boranophosphodiesters were converted to phosphodiesters via either H-phosphonate19−21 or acyl phosphite intermediates.22 Therefore, we anticipated an efficient synthesis of oligo-(glycosyl phosphate) would be possible by creating boranophosphotriester intersugar linkages and converting the linkages to the corresponding phosphodiester linkages in the final stage of the synthesis.

At first, we started to synthesize a monomer for the construction of the repeating units. The hydroxy groups of the mannose residue were protected by o-chlorobenzoyl groups, while those of the galactose residue were protected by Bz groups except for the 6-position, which had an o-(4-methoxytrityl) group. The phosphoramidite monomer 2 was obtained by the phosphitylation of the hemiacetal 1 with the preference for an α-isomer (α/β = 89:11–94:6, Scheme 1). The configuration of the anomeric position was confirmed by the 1JC1–H1 value (172 Hz) of 2.23 The detailed experimental procedures are described in the Supporting Information.

Scheme 1. Synthesis of Disaccharide 1-Phosphoramidite 2.

Scheme 1

Open in a new tab

Next, the solid-phase synthesis of disaccharide 1-phosphate was initiated (Scheme 2). Controlled pore glass (CPG) was chosen as a solid support, and the 1,4-bis(2-hydroxyethyl) hydroquinone spacer, which enables the detection of products with its UV absorption, was introduced via a succinyl linker to CPG.17 The hydroxy group of 3 was condensed with phosphoramidite 2 in MeCN in the presence of 4,5-dicyanoimidazole (DCI) as an acidic activator for 10 min. The resulting phosphite was oxidized to its phosphotriester counterpart using (+)-(8,8-dichlorocamphorylsulfonyl)oxaziridine (DCSO). After the MMTr group removal at the 6-position of the galactose residue and the deprotection of the phosphate moiety with Et3N, the removal of hydroxy protecting groups and cleavage of the linker were simultaneously performed via treatment with a 40% MeNH2 aqueous solution to afford glycosyl phosphate 4. The crude mixture was analyzed by reverse-phase high-performance liquid chromatography (RP-HPLC). The profile is shown in Figure S1 (Supporting Information). Although it was indicated that the disaccharide 1-phosphate 4 was obtained as a main product, we detected the formation of the byproduct 6. Additionally, the crude HPLC profile indicated that the disaccharide 1-phosphate 4 was formed as an α/β mixture, with two peaks having similar retention times, NMR analysis suggested that the former eluted peak was the α-isomer (isolated yield of 71%; the chemical shift of Man-H-1 was δ 5.43 ppm, in good agreement with α disaccharide 1-phosphate derivatives24), while the latter was the β-isomer (isolated yield of 21%; the chemical shift of Man-H-1 was δ 5.11 ppm). In addition, they had similar m/z values (calcd for [M – H]−, 601.1539; the former was 601.1530, and the latter was 601.1528).

Scheme 2. Solid-Phase Synthesis of Disaccharide 1-Phosphophate.

Scheme 2

Open in a new tab

The formation of 6 could be attributed to eliminating the phosphate moiety from the anomeric position, probably during the detritylation reaction. Thus, it was suggested that introducing an o-chlorobenzoyl protecting group could not prevent the formation of an oxocarbenium ion under acidic conditions. Subsequently, we attempted to synthesize a glycosyl boranophosphate using a similar procedure as that for glycosyl phosphate except for the oxidation step, which was replaced by a boronation step, and the removal of the cyanoethyl group was conducted by DBU treatment (Scheme S2). The crude RP-HPLC profile indicated the predominant formation of the disaccharide 1-boranophosphate, indicating that the glycosyl boranophosphate was tolerant of the detritylation conditions (Figure S2). Furthermore, treatment with a DCSO solution in MeCN resulted in an almost quantitative conversion of the glycosyl boranophosphodiester to its phosphate counterpart. We serendipitously found the reaction. Actually, we investigated the conversion of a boranophosphodiester to a phosphodiester via an acyl phosphite intermediate. As a control experiment, we examined the reaction between a boranophosphodiester and DCSO and found that the boranophosphate was converted to the phosphodiester. The mechanism and the scope of the reaction is under investigation. The procedure in Scheme 3 gave the glycosyl phosphate 4 as a major product (Table 1, entry 1, Figure S3). This result indicated that the degradation of the reaction intermediates was significantly suppressed via boranophosphate intermediates. In our previous reports, a boranophosphodiester linkage was converted to a phosphodiester linkage by multiple steps.20−22 In contrast, treatment with DCSO afforded a phosphodiester derivative in a single step, significantly facilitating the synthesis.

Scheme 3. Solid-Phase Synthesis of Disaccharide 1-Phosphophate via a Boranophosphate Intermediate.

Scheme 3

Open in a new tab

Table 1. Investigation of Preactivation Conditions.

entry pre-activation condensation time (min) HPLC yield (%)a α/β ratioa
1   10 97 82:18
2 + 10 77 97:3
3 + 60 90 97:3
Open in a new tab
a

Determined by RP-HPLC.

As for the low diastereoselectivity of the condensation reaction, a similar phenomenon was also observed in our precedent study, which used an α/β mixture of a glycosyl phosphoramidite as a monomer.17 Since the β-isomer of the phosphoramidite seemed to be more reactive than the α-counterpart, a preactivation strategy was effective to suppress the formation of β-glycosyl phosphate linkages. In the preactivation strategy, the α/β mixture of glycosyl phosphoramidite was treated with an alcohol to consume the reactive β-isomer before a condensation reaction on a solid support.17 Thus, this strategy was applied to the synthesis (Scheme 4, Table 1). The monomer 2 was treated with an activator and an alcohol in MeCN solvent in a round-bottom flask to consume the β-isomer, and after 3 min the mixture was added to the reactor for the manual solid-phase synthesis by a syringe. It was found that 3-phenyl-1-propanol effectively prevented the formation of the β-isomer. However, the condensation efficiency was not satisfactory with a 10 min reaction time (Table 1, entry 2, Figure S4). Extending the condensation reaction time to 1 h improved the HPLC yield (Table 1, entry 3, Figure S4). Thus, the condensation reactions were conducted for 1 h in the following experiments.

Scheme 4. Solid-Phase Synthesis of Glycosyl Phosphate Repeating Units (n = 1, 2, 4, 5).

Scheme 4

Open in a new tab

Capping steps were conducted except for after the final condensation step.

We proceeded to synthesize glycosyl phosphates with multiple phosphate moieties. The synthesis of tetrasaccharide having two phosphate linkages was attempted following Scheme 4. A capping step was added after the condensation reaction using acetic anhydride. The coupling yields were estimated from RP-HPLC profiles of the reaction mixtures. The results are shown in Table 2. The second coupling efficiency was low compared with the first step when the condensation reactions were performed in the presence of DCI, regardless of DCI concentration (Table 2, entries 1 and 2, Figure S5). In contrast, when N-phenylimidazolium triflate (PhIMT) was used, high coupling efficiencies were achieved, and tetrarasaccharide was obtained with a 59% isolated yield (Table 2, entry 3, Figure S5). We proceeded to synthesize octasaccharide and decasaccharide, which contained four and five repeating units, respectively, by repeating the cycles of condensation in the presence of PhIMT, boronation, and detritylation, followed by the removal of cyanoethyl groups at boranophosphate linkages, oxidation with DCSO, and deprotection of the hydroxy groups and linker cleavage. It is worth noting that the solid support was washed with MeOH following the boronation steps to remove the boronation reagent and/or its residue, which would interfere with the next condensation reaction.25 The HPLC profiles of the reaction mixtures shown in Figures 2 and S6 indicates the predominant formation of the desired products. The octasaccharide and decasaccharide were isolated in 46% and 36% yields, respectively (Table 2, entries 4 and 5, Figures 2 and S6). However, in the 1H NMR spectrum of the isolated decasaccharide derivative, a signal around 5.2 ppm was observed (SI page S61), indicating the presence of small amounts of the β-isomer. The improvement in stereoselectivity still needs to be solved.

Table 2. Synthesis of Oligo(glycosyl phosphate).

      coupling efficiencya (%)
  
entry n activator (concentration) n = 1 n = 2 isolated yield (%)
1 2 DCI (0.5 M) 95 75  
2 2 DCI (1.0 M) 91 82  
3 2 PhIMT (1.0 M) 95 92 59
4b 4 PhIMT (1.0 M)     46
5b 5 PhIMT (1.0 M)     36
Open in a new tab
a

Determined by RP-HPLC.

b

Washing steps with MeOH were conducted after each condensation step.

Figure 2.

Figure 2

Open in a new tab

RP-HPLC profiles of the crude mixture of 9 (ODS, H2O (containing 0.1 M 1,1,1,3,3,3-hexafluoro-2-propanol and 8 mM Et3N)/MeOH = 95:5–77:23 over 30 min with a flow rate of 0.5 mL/min, detection at 286 nm and 60 °C, and tR = 6.4 min (compound 9).

In conclusion, we successfully synthesized glycosyl phosphate repeating units found in LPGs of Leishmania via solid-phase synthesis using boranophosphates as key reaction intermediates. Using glycosyl boranophosphates provides stability under acidic conditions, enabling chain elongation without notable side reactions. The conversion of boranophosphodiester into phosphodiester was almost quantitative using the DCSO treatment. The repeating units up to decasaccharide were successfully obtained following this strategy. Notably, this is the first report of synthesizing a homogeneous oligo-(glycosyl phosphate) structure bearing more than four phosphate groups without chemical modifications on pyranose moieties. Our method would significantly facilitate the synthesis and property evaluation of biomolecules bearing glycosyl phosphate moieties.

Acknowledgments

We thank Ms. Noriko Sawabe and Mr. Motoo Iida (Tokyo University of Science) for their technical assistance with the NMR measurements. We also thank Ms. Fukiko Hasegawa and Dr. Yayoi Yoshimura (Tokyo University of Science) for the mass spectrometry measurements. This research was partly conducted at the Joint Usage/Research Center on Tropical Disease, Institute of Tropical Medicine, Nagasaki University (2020-Ippan-3, 2021-Ippan-22) under a SUNBOR GRANT from the Suntory Institute for Bioorganic Research and MEXT KAKENHI Grant JP23K06032. We would like to thank Enago (www.enago.jp) for the English language review.

Data Availability Statement

The data underlying this study are available in the published article and its online Supporting Information.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.orglett.3c01293.

  • Experimental procedures; HPLC profiles; and copies of 1H, 13C, 31P NMR, COSY, HMQC, HSQC, and HMBC spectra of new compounds (PDF)

Author Contributions

K.S., R.I.H., and T.W. conceived the research and designed the experiment plans. K.S. and K.M. conducted the experiments. All authors analyzed the data. K.S. wrote the manuscript.

The authors declare no competing financial interest.

Supplementary Material

ol3c01293_si_001.pdf (2.7MB, pdf)

References

  1. Nagle A. S.; Khare S.; Kumar A. B.; Supek F.; Buchynskyy A.; Mathison C. J. N.; Chennamaneni N. K.; Pendem N.; Buckner F. S.; Gelb M. H.; Molteni V. Recent Developments in Drug Discovery for Leishmaniasis and Human African Trypanosomiasis. Chem. Rev. 2014, 114, 11305–11347. 10.1021/cr500365f. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Forestier C. L.; Gao Q.; Boons G. J. Leishmania Lipophosphoglycan: How to Establish Structure-Activity Relationships for This Highly Complex and Multifunctional Glycoconjugate?. Front Cell Infect. Microbiol. 2014, 4, 193. 10.3389/fcimb.2014.00193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Novozhilova N. M.; Bovin N. V. Structure, Functions, and Biosynthesis of Glycoconjugates of Leishmania Spp. Cell Surface. Biochemistry (Moscow) 2010, 75, 686–694. 10.1134/S0006297910060027. [DOI] [PubMed] [Google Scholar]
  4. McConville M. J.; Bacic A.; Mitchell G. F.; Handman E. Lipophosphoglycan of Leishmania Major That Vaccinates against Cutaneous Leishmaniasis Contains an Alkylglycerophosphoinositol Lipid Anchor. Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 8941–8945. 10.1073/pnas.84.24.8941. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Rogers M. E.; Sizova O. V.; Ferguson M. A. J.; Nikolaev A. V.; Bates P. A. Synthetic Glycovaccine Protects against the Bite of Leishmania- Infected Sand Flies. J. Infect. Dis. 2006, 194, 512–518. 10.1086/505584. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cabezas Y.; Legentil L.; Robert-Gangneux F.; Daligault F.; Belaz S.; Nugier-Chauvin C.; Tranchimand S.; Tellier C.; Gangneux J.-P.; Ferrières V. Leishmania Cell Wall as a Potent Target for Antiparasitic Drugs. A Focus on the Glycoconjugates. Org. Biomol. Chem. 2015, 13, 8393–8404. 10.1039/C5OB00563A. [DOI] [PubMed] [Google Scholar]
  7. Nikolaev A. V.; Botvinko I. V.; Ross A. J. Natural Phosphoglycans Containing Glycosyl Phosphate Units: Structural Diversity and Chemical Synthesis. Carbohydr. Res. 2007, 342, 297–344. 10.1016/j.carres.2006.10.006. [DOI] [PubMed] [Google Scholar]
  8. Hansson J.; Garegg P. J.; Oscarson S. Syntheses of Anomerically Phosphodiester-Linked Oligomers of the Repeating Units of the Haemophilus Influenzae Types c and f Capsular Polysaccharides. J. Org. Chem. 2001, 66, 6234–6243. 10.1021/jo001302m. [DOI] [PubMed] [Google Scholar]
  9. Berkin A.; Coxon B.; Pozsgay V. Towards a Synthetic Glycoconjugate Vaccine Against Neisseria Meningitidis A. Chem. Eur. J. 2002, 8, 4424–4433. . [DOI] [PubMed] [Google Scholar]
  10. Nikolaev A. V.; Chudek J. A.; Ferguson M. A. J. The Chemical Synthesis of Leishmania Donovani Phosphoglycan via Polycondensation of a Glycobiosyl Hydrogenphosphonate Monomer. Carbohydr. Res. 1995, 272, 179–189. 10.1016/0008-6215(95)00067-4. [DOI] [PubMed] [Google Scholar]
  11. Ruhela D.; Vishwakarma R. A. Iterative Synthesis of Leishmania Phosphoglycans by Solution, Solid-Phase, and Polycondensation Approaches without Involving Any Glycosylation. J. Org. Chem. 2003, 68, 4446–4456. 10.1021/jo0341867. [DOI] [PubMed] [Google Scholar]
  12. Nikolaev A. V.; Ivanova I. A.; Shibaev V. N. The Stepwise Synthesis of Oligo(Glycosyl Phosphates) via Glycosyl Hydrogenphosphonates. The Chemical Synthesis of Oligomeric Fragments from Hansenula Capsulata Y-1842 Exophosphomannan and from Escherichia Coli K51 Capsular Antigen. Carbohydr. Res. 1993, 242, 91–107. 10.1016/0008-6215(93)80024-9. [DOI] [PubMed] [Google Scholar]
  13. Zhang X.; Yang Y.; Ding J.; Zhao Y.; Zhang H.; Zhu Y. Stereoselective Gold(I)-Catalyzed Approach to the Synthesis of Complex α-Glycosyl Phosphosaccharides. Nat. Commun. 2022, 13, 421. 10.1038/s41467-022-28025-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Beaucage S. L.; Iyer R. P. Advances in the Synthesis of Oligonucleotides by the Phosphoramidite Approach. Tetrahedron 1992, 48, 2223–2311. 10.1016/S0040-4020(01)88752-4. [DOI] [Google Scholar]
  15. Martin T. J.; Schmidt(rk) R. R. Efficient Sialylation with Phosphite as Leaving Group. Tetrahedron Lett. 1992, 33, 6123–6126. 10.1016/S0040-4039(00)60022-9. [DOI] [Google Scholar]
  16. Palmacci E. R.; Plante O. J.; Seeberger P. H. Oligosaccharide Synthesis in Solution and on Solid Support with Glycosyl Phosphates. Eur. J. Org. Chem. 2002, 2002, 595–606. . [DOI] [Google Scholar]
  17. Hara R. I.; Yaoita A.; Takeda K.; Ueki H.; Ishii A.; Imoto H.; Kobayashi S.; Sano M.; Noro M.; Sato K.; Wada T. Solid-Phase Synthesis of Fluorinated Analogues of Glycosyl 1-Phosphate Repeating Structures from Leishmania Using the Phosphoramidite Method. ChemistryOpen 2018, 7, 439–446. 10.1002/open.201800030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Prosperi D.; Panza L.; Poletti L.; Lay L. Boranophosphate Diesters as Stable Synthetic Analogues of 1-O- Glycosylphosphates. Tetrahedron 2000, 56, 4811–4815. 10.1016/S0040-4020(00)00386-0. [DOI] [Google Scholar]
  19. Matsumura F.; Oka N.; Wada T. Synthesis of Glycosyl Boranophosphates and Their Applications as Precursors of Glycosyl Phosphate Analogues. Org. Lett. 2008, 10, 1557–1560. 10.1021/ol800205e. [DOI] [PubMed] [Google Scholar]
  20. Sato K.; Oka N.; Fujita S.; Matsumura F.; Wada T. 1,2-Trans-Selective Synthesis of Glycosyl Boranophosphates and Their Utility as Building Blocks for the Synthesis of Phosphodiester-Linked Disaccharides. J. Org. Chem. 2010, 75, 2147–2156. 10.1021/jo902725g. [DOI] [PubMed] [Google Scholar]
  21. Fujita S.; Oka N.; Matsumura F.; Wada T. Synthesis of Oligo(α-D-Glycosyl Phosphate) Derivatives by a Phosphoramidite Method via Boranophosphate Intermediates. J. Org. Chem. 2011, 76, 2648–2659. 10.1021/jo102584g. [DOI] [PubMed] [Google Scholar]
  22. Sato K.; Wada T. One-Pot Conversion Reactions of Glycosyl Boranophosphates into Glycosyl Phosphate Derivatives: Via Acyl Phosphite Intermediates. Org. Biomol. Chem. 2016, 14, 11092–11095. 10.1039/C6OB02309F. [DOI] [PubMed] [Google Scholar]
  23. Bock K.; Pedersen C. A Study of 13CH Coupling Constants in Hexopyranoses. J. Chem. Soc. Perkin Trans. 2 1974, 293, 293–297. 10.1039/p29740000293. [DOI] [Google Scholar]
  24. Nikolaev A. V.; Rutherford T. J.; Ferguson M. A. J.; Brimacombe J. S. Parasite Glycoconjugates. Part 4. Chemical Synthesis of Disaccharide and Phosphorylated Oligosaccharide Fragments of Leishmania Donovani Antigenic Lipophosphoglycan. J. Chem. Soc. Perkin Trans. 1 1995, 1977. 10.1039/p19950001977. [DOI] [Google Scholar]
  25. Sato K.; Hagio T.; Sano M.; Muramoto K.; Yaoita A.; Noro M.; Hara R. I.; Wada T. Solid-Phase Stereocontrolled Synthesis of Oligomeric P-Modified Glycosyl Phosphate Derivatives Using the Oxazaphospholidine Method. ACS Omega 2021, 6, 20026–20041. 10.1021/acsomega.1c03058. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ol3c01293_si_001.pdf (2.7MB, pdf)

Data Availability Statement

The data underlying this study are available in the published article and its online Supporting Information.

Articles from Organic Letters are provided here courtesy of American Chemical Society

RESOURCES