The key entity of a DCAR agonist, phosphatidylinositol mannoside Ac1PIM1: Its synthesis and immunomodulatory function

Article abstract of DOI:10.1039/c9ob02724f

Ac₁PIM₁is a potential biosynthetic intermediate for phosphatidylinositol mannosides (PIMs) fromMycobacterium tuberculosis. We achieved the first synthesis of Ac₁PIM₁by utilizing an allyl-type protecting group strategy and regioselective phosphorylation of inositol. A very potent agonist of an innate immune receptor DCAR, which is better than previously known agonists, is demonstrated.

Full text of DOI:10.1039/c9ob02724f

Organic &
Biomolecular Chemistry
View Article Online
View Journal
COMMUNICATION
The key entity of a DCAR agonist,
phosphatidylinositol mannoside Ac1PIM1: its syn-
Cite this: DOI: 10.1039/c9ob02724f
thesis and immunomodulatory function†
Received 24th December 2019,
Accepted 21st April 2020
Yohei Arai,a Shota Torigoe,b Takanori Matsumaru,a Sho Yamasakib,c,d,e and
a
DOI: 10.1039/c9ob02724f
Yukari Fujimoto
*
rsc.li/obc
Ac1PIM1 is a potential biosynthetic intermediate for phosphatidyl- we report the first chemical synthesis of Ac1PIM1 (1). The
inositol mannosides (PIMs) from Mycobacterium tuberculosis. We partial structures, PIs, were also synthesized with or without
achieved the first synthesis of Ac1PIM1 by utilizing an allyl-type polar functional groups targeting hydrophilic amino acid resi-
protecting group strategy and regioselective phosphorylation of dues in the receptor. The effects of targeting the polar func-
inositol. A very potent agonist of an innate immune receptor tional group in lipid moieties have been previously reported
DCAR, which is better than previously known agonists, is for TLR2 14 and CD1d.15 We also analyzed the ability to activate
demonstrated.
innate immune receptors of these relatively smaller syn-
thesized compounds as the basis to understand complex
immunomodulatory functions.
Mycobacterium,
particularly
Mycobacterium
tuberculosis
complex, is a causative agent of tuberculosis (TB).1 Its unique
cell envelope structure and composition with thick lipophilic
layers protect it from the environment.2 One of its cellular
components is phosphatidylinositol mannosides (PIMs), and
the molecules contain phosphatidylinositol (PI) possessing
mono- to hexa-mannosides (PIM1–6) acylated to different
degrees (Fig. 1).3–6 PIMs are recognized by various immune
receptors such as innate immune receptor Toll-like receptor
(TLR) 2 7 and C-type lectin receptors (CLRs) like DCAR (dendri-
tic cell immunoactivating receptor)8 and DC-SIGN (dendritic
cell-specific ICAM-3 grabbing non-integrin).9 It has also been
reported that non-polymorphic antigen-presenting glyco-
protein CD1d recognizes PIMs to activate NKT cells.10,11
Although Ac1PIM1 has not been synthesized, several reports
have been published on the synthesis of PIMs. However, many
of these syntheses were more focused on the glycan part rather
than the phosphatidyl inositol moiety. The first synthesis of
PIM1 and PIM2 has been reported by Van Boom et al. in 1989
and 1992, respectively.16,17 Liu et al. reported the synthesis for
PIM6 using glycosylation between the protected PIM2 moiety
and tetramannosides.18 Patil et al. recently developed regio-
selective mannosylation of the protected inositol and reported
the synthesis of PIM6 via a shorter synthetic route.19 However,
the synthesis of 1 has yet to be reported.
In this research, we first developed synthetic methods for
PIMs, namely 1 and PIs (2a–c), by modifying the synthesis for
protozoan lyso-phosphatidyl inositol compounds, EhLPPG.20
PI derivatives 2b and 2c, which contain a polar group in the
lipid moiety, are designed to allow the polar residue of the
receptor and the ligand to interact. In this case, the polar func-
Although multiple functions of PIM molecules have been
reported, the molecular basis of the receptor recognition is not
well understood. Our research focuses on a relatively small
PIM molecule with tri-acyl groups, Ac1PIM1, which are poten-
tial biosynthetic intermediates for complex PIMs.12,13 Herein
aFaculty of Science and Technology, Keio University. Hiyoshi 3-14-1, Yokohama,
Kanagawa 223-8522, Japan. E-mail: fujimotoy@chem.keio.ac.jp
bDepartment of Molecular Immunology, Research Institute for Microbial Diseases,
Osaka University, Suita 565-0871, Japan
cLaboratory of Molecular Immunology, Immunology Frontier Research Center
(WPI-IFReC), Osaka University, Suita 565-0871, Japan
dDivision of Molecular Design, Medical Institute of Bioregulation, Kyushu University,
Fukuoka 812-8582, Japan
eDivision of Molecular Immunology, Medical Mycology Research Center, Chiba
University, Chiba 260-8673, Japan
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c9ob02724f
Fig. 1 Structures of the phosphatidyl inositol mannosides (PIMs)
derived from Mycobacterium tuberculosis and phosphatidyl inositol (PI).
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Communication
Organic & Biomolecular Chemistry
Fig. 2 Retrosynthesis analysis of Ac1PIM1 1 and PI 2a and derivatives 2b
and 2c.
Scheme 1 Regioselective phosphorylation of inositol derivative 8.
tional groups in 2b and 2c should link to the polar amino
acids in mTLR2 (Ser346)14 and mCD1d (Ser28),15 respectively.
Then the structure–activity relationships are investigated using
the synthesized compounds.
Fig. 2 shows the synthetic strategy for 1 and 2a–c. An allyl-
type protecting group with either an allyl- or Alloc-group is
used as the permanent protecting group of the hydroxyl
groups, because the allyl-type protecting group can be cleaved
with a Ru complex21 under mild and near-neutral conditions
in polar solvents at the end of the synthesis.22 Compounds 3
and 4 are delivered from mannose donor 5, inositol phosphate
moiety 6, and glycerol moiety 7. To synthesize inositol phos-
phate moiety 6, we investigated the regioselective phosphoryl-
Scheme 2 Synthesis of mannose moiety 5.
ation utilizing BINOL-derived selenophosphoryl chlorides23 tection of the hydroxyl groups, the selective removal of the
with modified BINOL structures. PMP group was attempted with CAN, but the TBS group was
Firstly, the regioselective phosphorylation of the inositol also cleaved to produce 13. We then performed the silylation
was examined (Scheme 1). That is, we investigated the regio- of the 6-hydroxyl group of 13 with TBSCl and imidazole, and
selective phosphorylation of compound 8 prepared from myo- N-phenyl-2,2,2-trifluoroacetoimidate was introduced to an
inositol24 utilizing BINOL-derived phosphoryl reagents 9a–d. anomeric position to give mannosyl donor 5.
Compared to entry 1, which is the previously developed con-
Scheme 3 shows the synthesis of 1. The Alloc group of com-
dition, a lower reaction temperature (entry 2) did not improve pound 11a was converted to the allyl group using a Ru catalyst
the regioselectivity of the 1-position vs. 3-position. To investi- to give 14.26 Transesterification of the phosphate in 14, from
gate the phosphoryl reaction and to improve the regio- BINOL to allyl ester with an allyl alcohol and NaH, gave inosi-
selectivity, the BINOL moiety was modified to prepare 9b–d. tol derivative 15. The exchange of the protecting group from
Although reagent 9d gave a very high selectivity, the yield was PMB to the Alloc group yielded compound 16. After the
unsatisfactory because of its low reactivity (entry 5), even removal of the MOM group by TFA, mannosylation of 16 was
though the conditions of higher temperature and a higher achieved using mannosyl donor 5 and TMSOTf to give manno-
reagent amount were examined. Other reagents 9b–c gave syl inositol phosphate 17 in an α-selective manner. Although
lower yields and selectivities. In order to separate diastereo- we examined the equivalent of the Lewis acid (TMSOTf) for
mers and to analyze the regioselectivity in silica gel column optimizing the reaction, at a lower equivalent such as 0.1 eq.,
chromatography, phosphorylated compounds 10a–d were con- the reaction did not proceed, and at a higher equivalent (1.0
verted to 11a–d.
eq.), the TBS group was cleaved and dimannose compounds
The mannose moiety was also prepared as shown in were obtained as byproducts. So, we used 0.5 eq. as the
Scheme 2. Mannose derivative 12 was initially prepared from optimal condition. Introducing glycerol moiety 18 to the man-
D-mannose via previously reported methods.25 After Alloc pro- nosylated inositol phosphate by the Mitsunobu reaction gave
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Communication
Scheme 5 Synthesis of PI and derivatives 2a–c.
Mitsunobu reaction to give compounds 4a–c. The global de-
protection of the allyl-type protecting groups of 4a–c using Pd
(PPh3)4 afforded 2a–c.
We next evaluated the immunomodulatory functions of the
synthesized PIM and PI compounds. Initially, we investigated
the DCAR activities of PIM compounds (1 and 2a–c) using 2B4
NFAT-GFP reporter cells co-expressing DCAR and FcRγ
(Fig. 3).28 Interestingly, 1 induced a potent reporter activity. In
fact, it was higher than that of the known agonist, Ac1PIM2. In
contrast, 2a–c showed no or weak activity. The results
suggested that Ac1PIM1, which was the mannose-truncated
structure from Ac1PIM2, was the active entity for DCAR
recognition.
We also evaluated the compounds for TLR2 and CD1d
binding affinities and the activation of the synthesized ligands
(Fig. S1–S3†). Not all compounds induced definite activity to
both TLR2 and CD1d (Fig. S1 and S2), although the com-
pounds interacted with the receptors (Fig. S3†). That is, we
investigated the binding potential of the synthesized ligands
to TLR2 and CD1d. All the ligands displayed lower binding
affinities compared with the known TLR2 ligand Pam2CSK4
and CD1d ligand α-galactosyl ceramide (α-GalCer). However,
they interacted weakly with the receptor proteins.
Scheme 3 Synthesis of Ac1PIM1 (1).
compound 19. After deprotection of the TBS group and the iso-
propylidene group, palmitic acid was introduced to afford
compound 3. Cleaving the allyl ester of phosphate in 3 realized
the global deprotection of allyl/Alloc groups using a Ru cata-
lyst, and the desired Ac1PIM1 1 was synthesized. The isolated
yield of the target product 1 was relatively low, because the
purification was difficult due to the low solubility (mixture-
solvent of CHCl3, CH3OH and H2O can be only used) and also
the additional usage of a metal scavenger to remove the Ru
catalyst was necessary.
To synthesize PI and its derivatives containing an amide
group in the lipid chain, 2a–c, we used common intermediate
14 (Scheme 5). After converting BINOL to the allyl group and
deprotection of the allyl group on the phosphate, phosphoino-
sitol compound 26 was condensed with glycerolipids 25a–c
synthesized from glycerol derivative 20 27 (Scheme 4) by the
Fig. 3 DCAR recognizes synthetic Ac1PIM1 1. NFAT-GFP reporter cells
expressing FcRγ alone and FcRγ + DCAR were stimulated with phospha-
tidyl-inositol lipids (0.8 nmol per well), anti-flag tag antibody (Anti-Flag
Ab) and trehalose dibehenate (TDB) for 18 h. GFP expression was ana-
lyzed by flow cytometry. Ac1PIM2 was purified from M. bovis BCG as
previously described.8
Scheme 4 Synthesis of glycerolipid moieties 25a–c.
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Communication
Organic & Biomolecular Chemistry
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
We thank Prof. Kazuya Iwabuchi (Kitasato University) and
Dr Kazuya Kabayama (Osaka University) for providing
2E10 hybridoma, and also thank Prof. Koichi Kato and
Dr Saeko Yanaka (Institute for Molecular Science) for high
resolution NMR measurements, supported by Nanotechnology
Platform Program (Molecule and Material Synthesis) of the
Ministry of Education, Culture, Sports, Science and
Technology (MEXT), Japan.
Fig. 4 Proinflammatory cytokine induction by the synthesized com-
pounds in RAW264.7 cells. (A) MCP-1 induction. (B) TNF-α induction.
RAW264.7 cells were seeded into 96-well plates (5.0 × 104 cells per
well), and ligands were added with the indicated concentrations before
incubation at 37 °C for 24 h. Cytokine release was measured using an
ELISA kit (Affymetrix).
This research was also financially supported by JSPS KAKENHI
(No. JP17H02207, JP18H04426, JP19H04815, JP19J13192, and
JP17H05800), by AMED under Grant Numbers JP19ae0101052,
JP19gm0910010, JP19ak0101070, JP19fk0108075 and by Takeda
Science Foundation.
We also analyzed the proinflammatory cytokines of our syn-
thesized ligands. The dependence of MCP-1 and TNF-α induc-
tion on the synthesized compounds was determined in mouse
macrophage cell line RAW 264.7 cells (Fig. 4) and mouse bone
marrow-derived dendritic cells (BMDCs) (Fig. S4†), which
expressed the innate immune receptors, DCAR, TLR2, and
lipid-antigen presenting CD1d. As a result of MCP-1 and TNF-α
induction in RAW 264.7 cells and BMDCs, 1 showed weak but
definite activities. 2a–c did not exhibit MCP-1 induction
activity in both cells. However, these compounds showed a
slight activity by measuring TNF-α induction in BMDCs. The
proinflammatory cytokine production by the synthesized com-
pounds was consistent with the results of DCAR activation.
Presumably, the other TLR2 and CD1d did not contribute to
the cytokine induction activities in these cases (Fig. 3 and
S1†). The resultant immunomodulatory functions of the syn-
thesized compounds 1 and 2a–c should be fundamental to
understanding the molecular recognition of DCAR and other
lipid-recognizing innate immune receptors.
Notes and references
1 WHO, Global tuberculosis report 2019, https://www.who.
int/tb/publications/global_report/en/.
2 P. J. Brennan and H. Nikaido, Annu. Rev. Biochem., 1995,
64, 29–63.
3 M. E. Guerin, J. Kordulakova, P. M. Alzari, P. J. Brennan
and M. Jackson, J. Biol. Chem., 2010, 285, 33577–33583.
4 M. Gilleron, C. Ronet, M. Mempel, B. Monsarrat,
G. Gachelin and G. Puzo, J. Biol. Chem., 2001, 276, 34896–
34904.
5 M. Gilleron, V. F. Quesniaux and G. Puzo, J. Biol. Chem.,
2003, 278, 29880–29889.
6 M. Gilleron, B. Lindner and G. Puzo, Anal. Chem., 2006, 78,
8543–8548.
7 M. Gilleron, J. Nigou, D. Nicolle, V. Quesniaux and G. Puzo,
Chem. Biol., 2006, 13, 39–47.
Conclusions
Herein the first synthesis of 1 was achieved by utilizing an
allyl-type protecting group strategy and a regioselective phos-
phorylation reaction with BINOL-derived selenophosphoryl
chlorides. 2a–c were also synthesized with the newly developed
synthetic strategy. The immunomodulatory functions of the
synthesized compounds were then investigated. 1 exhibited a
very potent activity as a DCAR agonist, which is higher than
that of the known agonist, Ac1PIM2, while 2a–c showed no or
weak activity. Our results suggested that DCAR most strongly
recognized the Ac1PIM1 moiety in PIMs. We also showed that 1
8 K. Toyonaga, S. Torigoe, Y. Motomura, T. Kamichi,
J. M. Hayashi, Y. S. Morita, N. Noguchi, Y. Chuma,
H. Kiyohara, K. Matsuo, H. Tanaka, Y. Nakagawa,
T. Sakuma, M. Ohmuraya, T. Yamamoto, M. Umemura,
G. Matsuzaki, Y. Yoshikai, I. Yano, T. Miyamoto and
S. Yamasaki, Immunity, 2016, 45, 1245–1257.
9 N. N. Driessen, R. Ummels, J. J. Maaskant, S. S. Gurcha,
G. S. Besra, G. D. Ainge, D. S. Larsen, G. F. Painter,
C.
M.
Vandenbroucke-Grauls,
J.
Geurtsen
and
B. J. Appelmelk, Infect. Immun., 2009, 77, 4538–4547.
and 2a–c weakly interacted with other innate immune recep- 10 K. Fischer, E. Scotet, M. Niemeyer, H. Koebernick,
tors such as TLR2 and CD1d, although they did not exhibit
obvious activities. The knowledge of these molecular inter-
actions should serve as a basis for further understanding of
J. Zerrahn, S. Maillet, R. Hurwitz, M. Kursar, M. Bonneville,
S. H. Kaufmann and U. E. Schaible, Proc. Natl. Acad.
Sci. U. S. A., 2004, 101, 10685–10690.
the immunomodulatory function of lipid molecules, especially 11 D. M. Zajonc, G. D. Ainge, G. F. Painter, W. B. Severn and
on antigen-presenting cells.
I. A. Wilson, J. Immunol., 2006, 177, 4577–4583.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Communication
12 M. E. Guerin, D. Kaur, B. S. Somashekar, S. Gibbs, P. Gest, 21 E. Rüba, W. Simanko, K. Mauthner, K. M. Soldouzi,
D. Chatterjee, P. J. Brennan and M. Jackson, J. Biol. Chem.,
2009, 284, 25687–25696.
C. Slugovc, K. Mereiter, R. Schmid and K. Kirchner,
Organometallics, 1999, 18, 3843–3850.
13 Z. Svetlíková, P. Baráth, M. Jackson, J. Korduláková and 22 S. Tanaka, H. Saburi, Y. Ishibashi and M. Kitamura, Org.
K. Mikušová, Protein Expression Purif., 2014, 100, 33–39. Lett., 2004, 6, 1873–1875.
14 Y. Arai, S. Inuki and Y. Fujimoto, Bioorg. Med. Chem. Lett., 23 T. Murai, D. Matsuoka and K. Morishita, J. Am. Chem. Soc.,
2018, 28, 1638–1641. 2006, 128, 4584–4585.
15 S. Inuki, T. Aiba, N. Hirata, O. Ichihara, D. Yoshidome, 24 T. Aiba, M. Sato, D. Umegaki, T. Iwasaki, N. Kambe,
S. Kita, K. Maenaka, K. Fukase and Y. Fujimoto, ACS Chem.
Biol., 2016, 11, 3132–3139.
K. Fukase and Y. Fujimoto, Org. Biomol. Chem., 2016, 14,
6672–6675.
16 C. J. J. Elie, C. E. Dreef, R. Verduyn, G. A. Van Der Marel 25 J. Sardinha, S. Guieu, A. Deleuze, M. C. Fernandez-Alonso,
and J. H. Van Boom, Tetrahedron, 1989, 45, 3477–3486.
17 C. J. J. Elie, R. Verduyn, C. E. Dreef, G. A. van der Marel and
J. H. van Boom, J. Carbohydr. Chem., 1992, 11, 715–739.
18 X. Liu, B. L. Stocker and P. H. Seeberger, J. Am. Chem. Soc.,
2006, 128, 3638–3648.
19 P. S. Patil, T. J. Cheng, M. M. Zulueta, S. T. Yang, L. S. Lico
and S. C. Hung, Nat. Commun., 2015, 6, 7239.
20 T. Aiba, S. Suehara, S.-L. Choy, Y. Maekawa, H. Lotter,
T. Murai, S. Inuki, K. Fukase and Y. Fujimoto, Chem. – Eur.
J., 2017, 23, 8304–8308.
A. P. Rauter, P. Sinay, J. Marrot, J. Jimenez-Barbero and
M. Sollogoub, Carbohydr. Res., 2007, 342, 1689–1703.
26 H. Saburi, S. Tanaka and M. Kitamura, Angew. Chem., Int.
Ed., 2005, 44, 1730–1732.
27 W. Srisiri, H. G. Lamparski and D. F. O’Brien, J. Org. Chem.,
1996, 61, 5911–5915.
28 S. Yamasaki, M. Matsumoto, O. Takeuchi, T. Matsuzawa,
E. Ishikawa, M. Sakuma, H. Tateno, J. Uno, J. Hirabayashi,
Y. Mikami, K. Takeda, S. Akira and T. Saito, Proc. Natl.
Acad. Sci. U. S. A., 2009, 106, 1897–1902.
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.