Stereochemistry, lipid length and branching influences Mincle agonist activity of monoacylglycerides

Article abstract of DOI:10.1039/c9ob02302j

Herein, we report on the synthesis of a series of enantiomerically pure linear, iso-branched, and α-branched monoacyl glycerides (MAGs) in 63-72% overall yield. The ability of the MAGs to signal through human macrophage inducible C-type lectin (hMincle) using NFAT-GFP reporter cells was explored, as was the ability of the compounds to activate human monocytes. From these studies, MAGs with an acyl chain length ≥C22 were required for Mincle activation and the production of interleukin-8 (IL-8) by human monocytes. Moreover, the iso-branched MAGs led to a more pronounced immune response compared to linear MAGs, while an α-branched MAG containing a C-32 acyl chain activated cells to a higher degree than trehalose dibehenate (TDB), the prototypical Mincle agonist. Across the compound classes, the activity of the sn-1 substituted isomers was greater than the sn-3 counterparts. None of the representative compounds were cytotoxic, thus mitigating cytotoxicity as a potential mediator of cellular activity. Taken together, 6h (sn-1, iC26+1), 8a (sn-1, C32) and 8b (sn-3, C32) exhibited the best immunostimulatory properties and thus, have potential as vaccine adjuvants.

Full text of DOI:10.1039/c9ob02302j

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Stereochemistry, lipid length and branching
influences Mincle agonist activity of
monoacylglycerides†
Cite this: DOI: 10.1039/c9ob02302j
a,b
Ayesha Khan,a,b Chriselle D. Braganza,a,b Kristel Kodar,a,b Mattie S. M. Timmer
*
and Bridget L. Stocker*a,b
Herein, we report on the synthesis of a series of enantiomerically pure linear, iso-branched, and
α-branched monoacyl glycerides (MAGs) in 63–72% overall yield. The ability of the MAGs to signal
through human macrophage inducible C-type lectin (hMincle) using NFAT-GFP reporter cells was
explored, as was the ability of the compounds to activate human monocytes. From these studies, MAGs
with an acyl chain length ≥C22 were required for Mincle activation and the production of interleukin-8
(IL-8) by human monocytes. Moreover, the iso-branched MAGs led to a more pronounced immune
response compared to linear MAGs, while an α-branched MAG containing a C-32 acyl chain activated
cells to a higher degree than trehalose dibehenate (TDB), the prototypical Mincle agonist. Across the
compound classes, the activity of the sn-1 substituted isomers was greater than the sn-3 counterparts.
None of the representative compounds were cytotoxic, thus mitigating cytotoxicity as a potential
mediator of cellular activity. Taken together, 6h (sn-1, iC26+1), 8a (sn-1, C32) and 8b (sn-3, C32) exhibited
the best immunostimulatory properties and thus, have potential as vaccine adjuvants.
Received 24th October 2019,
Accepted 19th November 2019
DOI: 10.1039/c9ob02302j
rsc.li/obc
the host cell.8,9 Liposomal formulations of synthetically pre-
pared C32 monoacylglycerol (MAG-C32, 3a) also induce potent
Introduction
Glycerides are membrane constituents found in most organ- Th-1 activity, and thus show promise as vaccine adjuvants,10
isms.1 These compounds, which are fatty acid esters of gly- while analogues of momocorynomycoloylglycerol MCMG
cerol, have many fundamental physiological functions includ- (4a and 4b) activate dendritic cells (DCs), with lipid length and
ing acting as anhydrous reservoirs for the efficient storage of stereochemistry, but not the configuration of the glycerol
caloric reserves and as intracellular membrane messengers in moiety, affecting the immune response.11 In this latter work,
signal transduction and molecular recognition processes.2,3 the inability of short chain MCMGs to activate DCs was attribu-
Several glycerides also have antifungal, antibacterial and/or ted to their cytotoxicity.
immunomodulatory activities. For example, iso-branched
To explain how MMGs exert their immunomodulatory
monoacylglycerides (iMAGs, 1) (Fig. 1), which have been iso- profile, Yamasaki and co-workers recently determined that
lated from Streptomyces sp. and the marine sponge Stelletta MMG 2 was a ligand for the human ortholog of the macro-
sp.,4,5 possess antifungal and antibacterial properties along phage inducible C-type lectin (hMincle), with hMincle signal-
with platelet aggregation inhibitory effects,4,6,7 while monomy- ling leading to the production of inflammatory cytokines.12
coloylglycerol (MMG, 2), which was first isolated from Monobehenoyl-rac-glycerol (MBG, 3b), a synthetic analogue of
Mycobacterium bovis Bacillus Calmette-Guérin, both as sn-1 MMG, failed to signal through mouse Mincle (mMincle), while
and sn-3 isomers, downregulates the T-helper (Th)-1 immune strong hMincle activity was observed, thus demonstrating the
response when formulated in liposomes and is thought to con- species-specific activity of select classes of Mincle ligands.12
tribute to the long-term survival of dormant mycobacteria in These findings sparked much interest in the Mincle agonist
activity of other glycerides, with Van der Peet et al. preparing
the (R,R)-meromycolate isomers of both sn-1 (4a) and sn-3 (4b)
aSchool of Chemical and Physical Sciences, Victoria University of Wellington, PO Box
substituted MCMG and, using a nuclear factor of activated T
cell-green fluorescent protein (NFAT-GFP) assay,13,14 demon-
600, Wellington, New Zealand
bCentre for Biodiscovery, Victoria University of Wellington, PO Box 600, Wellington,
strating that Mincle agonist activity predominantly resides
New Zealand. E-mail: bridget.stocker@vuw.ac.nz, mattie.timmer@vuw.ac.nz
with sn-1-isomer 4a.15 In 2019, it was also determined that
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
linear racemic MAGs 3b and 5 containing C22–C30 acyl chains
c9ob02302j
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 1 Structure of representative glycerides.
activate human, but not murine, Mincle.16 Here, the activity of
the MAGs was dependent on the length of the acyl chains.16
With a long-standing interest in determining how the struc-
ture of Mincle ligands influences their biological
response,17–22 we sought to better understand how glyceride
structure affects the ability of the compound to activate
murine and human Mincle. Previous investigations into the
immunostimulatory profile of Mincle ligands,23 including tre-
halose diesters,20 6′-acylated mannose and glucose,24
β-gentiobiosyl diacylglycerides,25 and lipidated brartemicin
analogues,18 demonstrated that an increase in lipid length
generally enhances the Mincle-dependent immune response.
The inclusion of iso-branching on the lipid backbone has also
been correlated to an enhanced immune response, both for
trehalose diesters,22 and for C17–C20 iMAGs,4,6 whereby the
latter exhibited better antimicrobial activity than their shorter
chain or linear counterparts. In view of these findings, as well
as the observation that the stereochemistry of the glycerol
moiety can influence Mincle activation,15 we sought to prepare
a series of monoacylglycerides with different stereochemistry
in the glycerol moiety and different iso-branched 6a–i, linear
7a–g, and α-branched 8a,b lipid chains (Fig. 2). These deriva-
tives would then be assessed for their ability to activate
hMincle reporter cells, as well as primary human cells.
Fig. 2 sn-1 and sn-3 MAGs to be synthesised.
modest, acids 13 were the only products observed by TLC [Rf =
0.8 (PE)], and were isolated as a mixture of E- and Z-isomers in
15 : 1 ratio. The instability of the ylide synthesised in situ could
be a contributing factor to the low yield of the Wittig olefina-
tion. Separation of the E- and Z-isomers was not undertaken,
Results and discussion
To synthesise the iso-branched MAGs, commercially available as hydrogenation of the alkene moiety was to be undertaken in
iso-branched fatty acids were used except for iso-methyl-cerotic a subsequent step.
acid (C26+1), which was prepared from cyclopentadecanolide 9
With the lipids in hand, iso-branched sn-1 and sn-3 MAGs
(Scheme 1). Here, Grignard reaction of cyclopentadecanolide 9 6a–i were prepared via esterification of commercially available
with methyl magnesiumiodide, followed by deoxygenation of (S)- or (R)-1,2-O-isopropylidene glycerol 14 with the appropriate
diol 10 at the tertiary position using boron trifluoride etherate fatty acid (15a–e or 13) in the presence of coupling reagents
and triethylsilane,26 gave primary alcohol 11 in excellent 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDCI) and
overall yield. Oxidation of 11 using PCC then gave aldehyde 12 4-dimethylaminopyridine (DMAP) (Scheme 2). Here, the reac-
in 83% yield.27 Finally, Wittig olefination was carried out tion mixture was stirred for 48 h at 70 °C to yield the corres-
using the triphenylphosphonium salt derived from 10-bromo- ponding isopropylidene protected iso-branched glycerides
decanoic acid and PPh3, and aldehyde 12, to yield iso- 16a–i in good yields (65–78%). For each product,
a
branched lipids 13.28 Although the yield of this reaction was Heteronuclear Multiple Bond Correlation (HMBC) between the
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 1 Synthesis of iso-branched fatty acids 13.
Scheme 2 Synthesis of iMAGs 6a–i.
methylene protons of the glycerol moiety and the carbonyl expressing hMincle and FcRγ or FcRγ-only.13,14 In accordance
carbon of the lipid confirmed the successful installation of the with recently reported findings,16 MAGs incorporating short
lipid moiety. Deprotection of 16a–g was then achieved under acyl chains (iMAGs 6a–e and MAGs 7a–c) did not activate
the agency of trifluoroacetic acid in tetrahydrofuran/water hMincle expressing reporter cells (Fig. 3). The iMAGs 6f (sn-1,
(TFA/THF/H2O, 3 : 8 : 1),15 with stirring of the reaction mixture iC22+1) and 6g (sn-3, iC22+1) showed better activity than their
at r.t. for 30 min. This resulted in the desired products 6a–g linear counterparts 7d (sn-1, C22) and 7e (sn-3, C22), and a
being isolated in excellent yields (85–93%) following purifi- similar trend was observed for iMAG 6h (sn-1, iC26+1) and 6i
cation by silica gel flash column chromatography. To prepare (sn-1, iC26+1) when compared to 7f (sn-1, C26) and 7g (sn-3,
glycerides 6h and 6i, alkene 16h and 16i were subjected to iC26). Moreover, iMAG 6h (sn-3, iC26 + 1) showed hMincle
hydrogenation in the presence of Pearlman’s catalyst which agonist activity equivalent to the positive control, trehalose
led to both double bond reduction and isopropylidene de- dibehenate (TDB), which is the prototypical Mincle agonist.23
protection to yield the target compounds in excellent yields In addition, the αMAGs 8a (sn-1, C32) and 8b (sn-3, C32),
(87–89%).
showed the highest degree of reporter cell activation and were
The sn-1 or sn-3 linear chain MAGs 7a–g and α-branched found to induce more GFP production than TDB. These results
MAGs 8a,b were then prepared in an analogous manner to the clearly demonstrate the importance of lipid length and
iMAGs (Scheme 3). Isopropylidine protected glycerol 14 was branching on the ability of MAGs to signal through Mincle.
esterified using carboxylic acids 17a–e or 18,24 to yield the iso-
The stereochemistry of the glycerol moiety also affected
propylidene protected linear and α-branched glycerides 19a–g Mincle signalling, as demonstrated by all three subsets of
and 20a,b, respectively. Deprotection of 19a–g and 20a,b gave MAGs with the sn-1 (S) isomers leading to greater reporter cell
the desired products 7a–g and 8a,b in excellent yields. The activation as compared to their sn-3 (R) isomeric counterparts
characterisation data of all MAGs was in accordance with pre- at both concentrations of glycolipid tested. For example, 6f
viously reported racemic iso-branched29 and short chain12,30,31 (sn-1, iC22+1) led to greater activity compared to 6g (sn-3,
MAGs.
iC22+1), as did 7d (sn-1, C22) compared to 7e (sn-3, C22), 7f
With the synthesised MAGs in hand, their ability to signal (sn-1, C26) compared to 7g (sn-3, C26), 6h (sn-1, iC26+1) com-
through Mincle was first assessed using NFAT-GFP cell lines pared to 6i (sn-3, iC26+1) and 8a (sn-1, C32) compared to 8b
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 3 Synthesis of linear MAG 7a–g and αMAG 8a,b.
Fig. 3 NFAT-GFP 2B4 reporter cells expressing hMincle + FcRγ, or
FcRγ-only were stimulated using MAG-coated plates (0.1 or 1 nmol per
well) for 18 h. The cells were then harvested and examined for
NFAT-GFP expression. Data reported is representative of two indepen-
dent experiments performed in duplicate (mean SEM).
Fig. 4 IL-8 production of human monocytes by treatment with plate-
coated MAGs or TDB (0.1 or 1 nmol per well) or solubilised LPS (100 ng
mL−1). Cytokine production was measured by ELISA from the super-
natant collected after 24 hours. Mean SEM of triplicate samples from
representative experiment of three performed are shown.
(sn-1, C32). This result corroborates the earlier findings of Van
der Peet et al. and further demonstrates that the Mincle agonist (i.e. compare 6f to 7d, 6g to 7e, 6h to 7f and 6i to 7g). Relative
activity predominantly resides with the sn-1-isomer.15 Overall, cytokine production by the monocytes following stimulation
8a (α-branched C32, sn-1) was the most potent agonist of those with the MAGs also followed the trends observed when using
tested. Moreover, iMAGs 6a–e and MAGs 7a–e were also tested the NFAT-GFP reporter assay, whereby monocytes stimulated
for their ability to activate murine Mincle using the NFAT-GFP with 1 nmol per well of iMAG 6h (sn-1, iC-26+1) led to IL-8 pro-
mMincle reporter cell assay.13,14 However, as anticipated,12,15 duction that was comparable to TDB, while αMAG 8a (sn-1,
these compounds did not signal through Mince (ESI Fig. 1†).
C32) and 8b (sn-3, C32) led to significantly higher levels of IL-8
In addition to the reporter cell assays, the proficiency of compared to TDB. The stereochemistry of the glycerol back-
MAGs to activate human monocytes, which have previously bone also affected IL-8 production by the monocytes, with the
been shown to express Mincle,32–34 was explored. After iso- sn-1 analogues primarily leading to higher levels of IL-8 cyto-
lation of the monocytes from peripheral blood mononuclear kine production than the corresponding sn-3 isomer. While we
cells (PMBCs), the production of the proinflammatory cytokine cannot conclusively say that the production of IL-8 by human
IL-835,36 by the monocytes was assessed following stimulation monocytes in response to the monoacylglycerides is solely
with the C22 (6f, 6g, 7d, 7e), C26 (6h, 6i, 7f, 7g) and C32 (8a, dependent on Mincle signalling, our results from the reporter
8b) MAGs at 0.1 or 1 nmol per well (Fig. 4). Once again, lipid cell assay where Mincle-dependence was observed indicates
length played an important role in the immunostimulatory that IL-8 production, at least in part, is due to the engagement
activity of the compounds, with increasing lipid length corres- of the monoacylglycerides with hMincle.
ponding to increased IL-8 production. Moreover, the pro-
To determine whether there was a correlation between the
duction of IL-8 in response to the iso-branched MAGs was cytotoxicity of the glycerides and their ability to activate
greater than that elicited by the corresponding linear MAGs immune cells, as was previously demonstrated by Bertelsen
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
were equivalent to, if not better than, TDB. Accordingly, the
ease of their syntheses and their immunostimulatory pro-
perties makes MAGs particularly promising vaccine adjuvants.
Conflicts of interest
There are no conflicts of interest to declare.
Acknowledgements
We would like to thank the Marsden Fund (VUW1401) and the
Health Research Council of New Zealand (Hercus Fellowship,
BLS, 2013/33) for funding. We would also like to thank
Professor Sho Yamasaki (RIMD Research Institute for
Microbial Disease) for kindly providing the Mincle reporter
cell lines and Amy T. Lynch (Immunoglycomics Lab, Victoria
University of Wellington) for drawing blood from donors.
Fig. 5 HL-60 were treated with 0.001, 0.01, 0.1, 1, 10, or 100 nmol per
well of plate-coated 6b, or 7d, or 7f, or 8a. Mean
SEM of triplicate
MTT assay samples from representative experiment of three performed
are shown. The toxin cyclohexamide was used as a positive control.
et al. when assessing the ability of MCMGs to active DCs,11 we
determined the cytotoxic effects of representative MAGs on
HL-60 cells using the MTT assay.37 To this end, HL-60 cells
were treated with titrated amounts of 6b (sn-1, iC14+1), 7d
(sn-1, C22), 7f (sn-1, C26) and 8a (sn-1, C32), and, after 18 h,
the percentage of live/dead cells was analysed (Fig. 5). No cyto-
toxicity was observed for the MAGs with an acyl chain length
≥C22 at any concentration tested, while the short acyl chain
length iMAG 6b (iC14+1) exhibited cytotoxicity only at a very
high concentration, i.e. 100 nmol per well, which is a concen-
tration that greatly surpasses that used in the NFAT-GFP repor-
ter assay. Accordingly, the inability of 6b to activate reporter
cells does not appear to be due to the cytotoxicity of the
compound but is likely due to ineffective binding of this
ligand in the hydrophobic groove of Mincle.17–22,38 Moreover,
the pro-inflammatory effects of the longer chain glycerides
(7d, 7f and 8a) were not due to cytotoxicity.
References
1 B. Lerique, J. Lepetit-Thévenin, A. Vérine, C. Delpéro and
J. Boyer, Life Sci., 1994, 54, 831–840.
2 R. Bittman, in Encyclopedia of Biophysics, ed. G. C. K.
Roberts, Springer Berlin Heidelberg, Berlin, Heidelberg,
2013, vol. 3, 907–914.
3 G. van Meer, D. R. Voelker and G. W. Feigenson, Nat. Rev.
Mol. Cell Biol., 2008, 9, 112.
4 S. Omura, A. Nakagawa, N. Fukamachi, K. Otoguro and
B. Kobayashi, J. Antibiot., 1986, 39, 1180–1181.
5 Q. Zhao, T. A. Mansoor, J. Hong, C.-O. Lee, K. S. Im,
D. S. Lee and J. H. Jung, J. Nat. Prod., 2003, 66, 725–728.
6 X. Ping, Y. Akeda, K.-i. Fujita, T. Tanaka and M. Taniguchi,
J. Antibiot., 1998, 51, 852–856.
7 Y. Akeda, K. Shibata, X. Ping, T. Tanaka and M. Taniguchi,
J. Antibiot., 1995, 48, 363–368.
8 Y. Hattori, I. Matsunaga, T. Komori, T. Urakawa,
T. Nakamura, N. Fujiwara, K. Hiromatsu, H. Harashima
and M. Sugita, Biochem. Biophys. Res. Commun., 2011, 409,
304–307.
9 E. Layre, A. Collmann, M. Bastian, S. Mariotti, J. Czaplicki,
J. Prandi, L. Mori, S. Stenger, G. De Libero, G. Puzo and
M. Gilleron, Chem. Biol., 2009, 16, 82–92.
Conclusion
In this study a series of linear, iso-branched, and α-branched
MAGs were synthesised with defined stereochemistry and in
good overall yields (63–72%). The effect of acyl chain length,
iso-branching and stereochemistry on Mincle activation was
explored and it was determined that the incorporation of
longer lipid chains and branching on the lipid backbone 10 V. Bhowruth, D. E. Minnikin, E. M. Agger, P. Andersen,
enhanced the immunmodulatory activity of the compounds,
with an acyl chain length ≥C22 being required for signalling
V. W. Bramwell, Y. Perrie and G. S. Besra, Bioorg. Med.
Chem. Lett., 2009, 19, 2029–2032.
through hMincle and the production of IL-8 by human mono- 11 B. Martin-Bertelsen, K. S. Korsholm, F. Rose, P. Nordly,
cytes. The stereochemistry of the glycerol moiety also influ-
H. Franzyk, P. Andersen, E. M. Agger, D. Christensen,
enced cellular activation with the sn-1 isomers leading to more
A. Yaghmur and C. Foged, RSC Adv., 2013, 3, 20673–20683.
pronounced immune responses compared to the sn-3 isomers. 12 Y. Hattori, D. Morita, N. Fujiwara, D. Mori, T. Nakamura,
The immune response to the MAGs was not due to cytotoxicity.
Thus, in summary, key structure–activity relationships with
H. Harashima, S. Yamasaki and M. Sugita, J. Biol. Chem.,
2014, 289, 15405–15412.
regard to the Mincle-agonist activity of MAGs have been deter- 13 E. Ishikawa, T. Ishikawa, Y. S. Morita, K. Toyonaga,
mined, with three MAGs [6h (sn-1, iC26+1), 8a (sn-1, C32) and
8b (sn-3, C32)] exhibiting immunostimulating properties that
H. Yamada, O. Takeuchi, T. Kinoshita, S. Akira, Y. Yoshikai
and S. Yamasaki, J. Exp. Med., 2009, 206, 2879–2888.
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
14 S. Yamasaki, E. Ishikawa, M. Sakuma, H. Hara, K. Ogata 25 M. B. Richardson, S. Torigoe, S. Yamasaki and
and T. Saito, Nat. Immunol., 2008, 9, 1179–1188. S. J. Williams, Chem. Commun., 2015, 51, 15027–15030.
15 P. L. van der Peet, C. Gunawan, S. Torigoe, S. Yamasaki and 26 M. B. Richardson and S. J. Williams, Beilstein J. Org. Chem.,
S. J. Williams, Chem. Commun., 2015, 51, 5100–5103. 2013, 9, 1807–1812.
16 T. Matsumaru, R. Ikeno, Y. Shuchi, T. Iwamatsu, 27 E. Corey and G. Schmidt, Tetrahedron Lett., 1979, 20, 399–
T. Tadokoro, S. Yamasaki, Y. Fujimoto, A. Furukawa and
K. Maenaka, Chem. Commun., 2019, 55, 711–714.
402.
28 S. Zlatanos, K. Laskaridis, E. Koliokota and A. Sagredos,
17 J. H. Bird, A. A. Khan, N. Nishimura, S. Yamasaki,
Grasas Aceites, 2011, 62, 462–466.
M. S. M. Timmer and B. L. Stocker, J. Org. Chem., 2018, 83, 29 X. Ping, Y. Usuki, Y. Akeda and M. Taniguchi, J. Antibiot.,
7593–7605. 1999, 52, 345–347.
18 A. J. Foster, M. Nagata, X. Lu, A. T. Lynch, Z. Omahdi, 30 R. J. Craven and R. W. Lencki, Cryst. Growth Des., 2011, 11,
E. Ishikawa, S. Yamasaki, M. S. M. Timmer and
1723–1732.
B. L. Stocker, J. Med. Chem., 2018, 61, 1045–1060.
31 D. R. Kodali, J. Lipid Res., 1987, 28, 464–469.
19 A. A. Khan, S. H. Chee, R. J. McLaughlin, J. L. Harper, 32 J. Ostrop, K. Jozefowski, S. Zimmermann, K. Hofmann,
F. Kamena, M. S. M. Timmer and B. L. Stocker,
ChemBioChem, 2011, 12, 2572–2576.
E. Strasser, B. Lepenies and R. Lang, J. Immunol., 2015,
195, 2417–2428.
20 A. A. Khan, B. L. Stocker and M. S. M. Timmer, Carbohydr. 33 T. Hupfer, J. Schick, K. Jozefowski, D. Voehringer, J. Ostrop
Res., 2012, 356, 25–36. and R. Lang, Front. Immunol., 2016, 7, 423.
21 B. L. Stocker, A. A. Khan, S. H. Chee, F. Kamena and 34 D. Vijayan, K. J. Radford, A. G. Beckhouse, R. B. Ashman
M. S. M. Timmer, ChemBioChem, 2014, 15, 382–388. and C. A. Wells, Immunol. Cell Biol., 2012, 90, 889–895.
22 A. Khan, K. Kodar, M. S. M. Timmer and B. L. Stocker, 35 W. Holmes, J. Lee, W. Kuang, G. Rice and W. Wood,
Tetrahedron, 2018, 74, 1269–1277. Science, 1991, 253, 1278–1280.
23 C. D. Braganza, T. Teunissen, M. S. M. Timmer and 36 P. Murphy and H. Tiffany, Science, 1991, 253, 1280–1283.
B. L. Stocker, Front. Immunol., 2018, 8, 1940.
24 A. Decout, S. Silva-Gomes, D. Drocourt, S. Barbe, I. André,
37 M. V. Berridge, P. M. Herst and A. S. Tan, in Biotechnol.
Ann. Rev., Elsevier, 2005, vol. 11, pp. 127–152.
F. J. Cueto, T. Lioux, D. Sancho, E. Pérouzel, A. Vercellone, 38 K. Momoi, Y. Sugita, M. Ishihara, K. Satoh, H. Kikuchi,
J. Prandi, M. Gilleron, G. Tiraby and J. Nigou, Proc. Natl.
Acad. Sci. U. S. A., 2017, 114, 2675–2680.
K. Hashimoto, I. Yokoe, H. Nishikawa, S. Fujisawa and
H. Sakagami, In vivo, 2005, 19, 157–163.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2019