The effects of 5-OP-RU stereochemistry on its stability and MAIT-MR1 axis

Article abstract of DOI:10.1002/cbic.202000466

Mucosal-associated invariant T (MAIT) cells are an abundant subset of innate-like T lymphocytes. MAIT cells are activated by microbial riboflavin-derived antigens, such as 5-(2-oxopropylideneamino)-6-d-ribitylaminouracil (5-OP-RU), when presented by the major histocompatibility complex (MHC) class I-related protein (MR1). We have synthesized all stereoisomers of 5-OP-RU to investigate the effects of its stereochemistry on the MR1-dependent MAIT cell activation and MR1 upregulation. The analysis of MAIT cell activation by these 5-OP-RU isomers revealed that the stereocenters at the 2’- and 3’-OH groups in the ribityl tail are crucial for the recognition of MAIT-TCR, whereas that of 4’-OH group does not significantly affect the regulation of MAIT cell activity. Furthermore, kinetic analysis of complex formation between the ligands and MR1 suggested that 5-OP-RU forms a covalent bond to MR1 in cells within 1 hour. These findings provide guidelines for designing ligands that regulate MAIT cell functions.

Full text of DOI:10.1002/cbic.202000466

Combining Chemistry and Biology
Accepted Article
Title: The effects of 5-OP-RU stereochemistry on its stability and MAIT-
MR1 axis
Authors: Takuro Matsuoka, Chihiro Motozono, Akira Hattori, Hideaki
Kakeya, Sho Yamasaki, Shinya Oishi, Hiroaki Ohno, and
Shinsuke Inuki
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemBioChem 10.1002/cbic.202000466
Link to VoR: https://doi.org/10.1002/cbic.202000466
01/2020

10.1002/cbic.202000466
ChemBioChem
FULL PAPER
The effects of 5-OP-RU stereochemistry on its stability and MAIT-
MR1 axis
Takuro Matsuoka,[a] Chihiro Motozono,[b, c] Akira Hattori,[a] Hideaki Kakeya,[a] Sho Yamasaki,[b, c] Shinya
Oishi,[a] Hiroaki Ohno,*[a] Shinsuke Inuki*[a]
[a]
T. Matsuoka, A. Hattori, H. Kakeya, S. Oishi, H. Ohno, S. Inuki
Graduate School of Pharmaceutical Sciences
Kyoto University
Sakyo-ku, Kyoto 606-8501 (Japan)
E-mail: sinuki@pharm.kyoto-u.ac.jp, hohno@pharm.kyoto-u.ac.jp
C. Motozono, S. Yamasaki
Research Institute for Microbial Diseases
Osaka University
[b]
[c]
Suita 565-0871 (Japan)
C. Motozono, S. Yamasaki
Immunology Frontier Research Center (IFReC)
Osaka University
Suita 565-0871 (Japan)
Abstract: Mucosal-associated invariant T (MAIT) cells are an
abundant subset of innate-like T lymphocytes. MAIT cells are
activated by microbial riboflavin-derived antigens, such as 5-(2-
consists of TRAV1-2 (Vα7.2) joined to TRAJ33 (Jα33) and the β-
chain repertoire is also restricted.6,7 MAIT cells are activated by
the infection of riboflavin-producing bacteria, such as Escherichia
coli or Mycobacterium tuberculosis, to produce inflammatory
cytokines (e.g., INF-γ, TNF, and GM-CSF).8–11 These cytokines
play an important role in biological defense against bacterial
infection.3,8 Namely, MAIT cells are profoundly involved in
immune responses during the initial stage of the infection. The
involvement of MAIT cells in various disorders such as cancer and
autoimmune diseases has also been reported.3,12,13 For instance,
in patients with colon adenocarcinomas and colorectal cancers,
MAIT cells are accumulated in tumor tissues.12 Therefore, the
modulation of MAIT cell activation has recently garnered
significant attention as a therapeutic target for infection, cancer
and autoimmune disorder, etc.
oxopropylideneamino)-6-D-ribitylaminouracil
(5-OP-RU),
when
presented by the major histocompatibility complex (MHC) class I-
related protein (MR1). Herein we have synthesized all stereoisomers
of 5-OP-RU to investigate the effects of its stereochemistry on the
MR1-dependent MAIT cell activation and MR1 upregulation. The
analysis of MAIT cell activation by these 5-OP-RU isomers revealed
that the stereocenters at the 2’- and 3’-OH groups in the ribityl tail are
crucial for the recognition of MAIT-TCR, while that of 4’-OH group
does not significantly affect the regulation of MAIT cell activity.
Furthermore, the kinetic analysis of complex formation between the
ligands and MR1 suggested that 5-OP-RU forms a covalent bond to
MR1 in cells within 1 hour. These findings provide guidelines for
designing ligands that regulate MAIT cell functions.
Introduction
Scheme 1. Formation of 5-OP-RU and RL-7-Me from 5-A-RU and Methylglyoxal.
Figure 1. Activation of MAIT Cells by Microbial Antigen Bound to MR1.
Recently, it has been reported that MAIT cells are strongly
activated in response to 5-(2-oxopropylideneamino)-6-D-
ribitylaminouracil (5-OP-RU) 1a presented by the major
histocompatibility complex (MHC) class I-related protein (MR1)
(Figure 1).14–16 5-OP-RU is a relatively short-lived intermediate
from the non-enzymatic Schiff base formation between 5-amino-
6-D-ribitylaminouracil (5-A-RU) 2a and methylglyoxal (Scheme 1).
Mucosal-associated invariant T (MAIT) cells are a subset of
innate-like T lymphocytes, which are found abundantly in human
tissue or blood such as gut, liver (45% of all α/β T cells) and
peripheral blood (10% of all α/β T cells).1–3 Analogous to natural
killer T (NKT) cell,4,5 MAIT cells have highly conserved T cell
receptor (TCR) repertoires. In humans, the α-chain mainly
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000466
ChemBioChem
FULL PAPER
In a previous report, decomposition of 5-OP-RU was observed in Results and Discussion
water (t1/2 = 1.5 h, 37 ˚C)17 because of the facile hydrolysis of the
All stereoisomers of 5-A-RU 2a–h23 were synthesized in
accordance with the reported method (Scheme 2).24 6-Chloro-5-
nitrouracil 4 was modified with the known amino alcohols 525 to
provide nitrouracils 6. Reduction of the nitro group provided the
expected 5-A-RU isomers 2. Each 5-OP-RU stereoisomer 126 was
prepared from their precursor 2 and methylglyoxal in aqueous
buffer immediately before testing,27 owing to their instability. We
also synthesized the RL-7-Me stereoisomers 3,28 which were
formed from 1 under aqueous conditions.17
As described above, 5-OP-RU is the potent ligand to activate
MAIT cells, but is unstable under physiological conditions.17
Considering its instability, the different stereochemistries among
the 5-OP-RU isomers might affect their chemical stability, thereby
preventing the accurate evaluation of their biological activities.
Indeed, Awad et al. reported that subtle modifications of 5-OP-RU,
including deoxy analogues, had a profound impact on their
chemical stability.18 Thus, prior to evaluating the T cell activation,
we first investigated the chemical stability of four diastereomers
[5-OP-RU (1a) and its stereoisomers 1c, 1e and 1g] prepared
from the corresponding 5-A-RU stereoisomers. Of note, the
corresponding enantiomers (1b, 1d, 1f and 1h, respectively)
should have identical physicochemical properties. After the 5-A-
RU isomers 2 (18 mM) were incubated in methylglyoxal (450 mM)
in D2O, the formation of 5-OP-RU isomers 1 and RL-7-Me isomers
3 were assessed by monitoring their characteristic signals using
1H-NMR spectroscopy (Figure 2). All four diastereomers 1
reached a maximum amount at 10 min, and then gradually
degraded within 2 hours. Four RL-7-Me stereoisomers 3
accumulated over time with similar conversion rates. These
results indicated that the stereochemistry of the ribityl groups did
not affect the formation and degradation kinetics of 1 and 3.
iminocarbonyl
moiety.
Alternatively,
an
intramolecular
condensation between the amino and carbonyl groups in 5-OP-
RU provides 7-methyl-8-D-ribityllumazine (RL-7-Me) 3a, as
shown in Scheme 1.17 5-OP-RU binds tightly to MR1 through a
covalent bond formation between the amino group of Lys43 and
iminoglyoxal group of 5-OP-RU. The ribityl group of 5-OP-RU is
mainly recognized by TCR on MAIT cells.14 For investigating the
molecular recognition by MR1 and MAIT-TCR, and the chemical
stability, several structure-activity relationship (SAR) studies of 5-
OP-RU have been carried out by various approaches.17–22 For
instance, Mak et al. designed and synthesized more stable
ligands having an enone moiety instead of the iminocarbonyl
group.17 Lange et al. developed a stable prodrug that releases an
active MAIT cell agonist precursor via enzymatic cleavage of
valine-citrulline-p-aminobenzyl carbamate linker.20
Based on these findings, we focused on the stereochemistries
of the ribityl group in 5-OP-RU as the key moiety involved in the
MR1 binding and MAIT-TCR recognition. In the present study, we
designed, synthesized and evaluated all stereoisomers of the
ribityl group in 5-OP-RU to examine the underlying molecular
mechanism for the ligand recognition by human MR1 and human
MAIT-TCR. We also investigated the bioactivities of 5-A-RU
stereoisomers 2 and bicyclic RL-7-Me stereoisomers 3 for the
MAIT-MR1 axis. Furthermore, medium-exchange experiments
were conducted to investigate the formation kinetics of the ligand-
MR1 complex. During the course of our study, Braganza et al.
reported the synthesis of some stereoisomers of 5-OP-RU to
evaluate the binding to mouse MR1 for activation of mouse MAIT-
TCR.19 More recently, Awad et al. designed deoxy analogues of
5-OP-RU to evaluate the contributions of each hydroxy group in
the ribityl moiety to MAIT cell activation.21
Scheme 2. Syntheses of 5-A-RU and RL-7-Me Stereoisomers.
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000466
ChemBioChem
FULL PAPER
(90%) between mice and humans,30 MAIT cell activation is highly
dependent on the class of TCR.31,32
Figure 2. Formation Kinetics of the 5-OP-RU and RL-7-Me Stereoisomers. (a)
Formation and degradation of 5-OP-RU 1a and its stereoisomers 1c, 1e and 1g.
(b) Formation of RL-7-Me 3a and its stereoisomers 3c, 3e and 3g. The product
formation was monitored by 1H-NMR analysis every 10 minutes after
methylglyoxal (40% aqueous solution; 58 μL, 0.33 mmol) was added to a
solution of each precursor 2a, 2c, 2e and 2g (5.0 mg, 0.013 mmol) in D2O (730
μL).
Figure 3. MAIT Cell Activation by the 5-OP-RU Stereoisomers. (a) CD69
surface expression on TG40.MAIT-TCR cells using flow cytometry after 24 h
incubation in co-culture with HeLa.MR1 cells in the presence of each ligand. (b)
IL-2 production in the supernatant measured by ELISA. The graphs show the
mean
± SD of triplicate measurements, and the results shown are
representative of at least three independent experiments. ND = not detected.
We then evaluated the MAIT cell activation of the resulting
stereoisomers by
a co-culture assay using human MR1-
expressing HeLa (HeLa.MR1) cells and TG40 cells transfected
with MAIT-TCR (TG40.MAIT-TCR). TG40.MAIT-TCR comprises
the TRAV1-2-TRAJ33 invariant α-chain paired with a TRBV6-1 β-
chain. MR1 ligands such as 5-OP-RU bind to MR1 in the antigen-
presenting cells (APC), and the resulting complexes are
recognized by TCR on MAIT cells, leading to the stimulation of
MAIT cells.13 The HeLa.MR1 and TG40.MAIT-TCR cells were co-
incubated in the presence of the ligands. After 24 h culturing, the
surface expression of CD69,29 a T cell activation marker, was
detected using flow cytometry (Figure 3a and S1). As an indicator
of MAIT cell activation, IL-2 cytokine release was also measured
by ELISA (Figure 3b). In both assays, 1b–1f and 1h displayed
significantly reduced MAIT cell activation compared with 5-OP-
RU (1a), whereas the 4’-OH epimer (1g) exhibited comparable
potent activity to 1a in a dose-dependent manner. Awad et al. re-
ported that, among the deoxy analogues of the ribityl group, the
des-4’-OH analogue displayed an agonistic activity that was as
strong as 1a, whereas the analogues without the 2’- or 3’-OH
groups were less potent.21 These results indicated that the
modification at the 4’ position in the ribityl group was tolerated with
MAIT cell activation. In contrast, unlike our findings, Braganza et
al. reported that the isomers 1b, 1c, 1e and 1g displayed
comparable activity to 5-OP-RU 1a in the assay using mouse
MR1 and MAIT-TCR.19 This discrepancy might be mainly due to
the difference in TCR used for the assay. Although MR1-α1 and -
α2 domains (ligand binding domains) have high sequence identity
Figure 4. Upregulation of Cell-Surface MR1 on HeLa.MR1 Cells in Response
to 5-OP-RU Stereoisomers. The MR1 expression on HeLa.MR1 was assessed
using flow cytometry after 8 h incubation with ligands. The graphs show the
mean
± SD of triplicate measurements, and the results shown are
representative of at least three independent experiments.
MAIT cell activation by ligands is associated with the ability to
increase cell-surface MR1 on the APC as well as its affinity with
MAIT-TCR.9 Previous studies also revealed that the surface
expression level of MR1 depends on the ligand structure.21,22 We
then investigated whether the chirality of the ribityl group
influenced the surface expression of MR1. The assay was
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000466
ChemBioChem
FULL PAPER
conducted by incubation of HeLa.MR1 cells with each ligand. The
surface expression of MR1 was measured by flow cytometry. As
shown in Figures 4 and S2, all stereoisomers of 5-OP-RU 1a had
less MR1 expression activity compared with the non-stimulatory
pteridin ligand, Ac-6-FP, which is known to increase MR1 cell-
surface expression.31 This result is in accordance with the report
by Awad et al. that indicated the polar ribityl moiety itself might
have adverse effects on the ability to upregulate MR1 cell-surface
expression.21 Among them, 1a, 1b and 1g worked as moderate
upregulators of MR1, which might be mainly associated with high-
affinity molecular recognition by MR1, although the involvement
of other cellular processes could not be ruled out. The isomer 1b
displayed no activity to MAIT cells, but increased the surface
expression of MR1. These findings suggest that the activation of
MAIT cells by 1a and 1g is partly attributable to the increase in
MR1 surface expression. The detailed studies on the correlation
between the complex stability, surface expression and MAIT cell
activation are now underway.
form 2a and (2’S,3’S,4’S)-form 2g increased CD69 expression
and IL-2 production, while the other stereoisomers showed no
activity for MAIT cell activation. The lumazine stereoisomers 3
provided results that were similar to those of the 5-A-RU
stereoisomers 2 and the 5-OP-RU stereoisomers 1. The
(2’S,3’S,4’R)-form 3a and (2’S,3’S,4’S)-form 3g exhibited
moderate agonistic activity for MAIT cell activation. In contrast, no
effect on stimulating MAIT cells was observed with the other
stereoisomers 3b, 3c and 3e. Taken together, all forms of 1–3
having the (2’S,3’S)-stereochemistries functioned as agonists for
MAIT cell activation. These results indicated that the structure of
the ribityl moiety is essential for MAIT cell activation. The
conversion or modification of the uracil and lumazine scaffolds
appears to be a promising approach for the development of novel
MAIT cell ligands.
The ligands 1a and 1g bearing an iminoglyoxal group were
more potent for MAIT cell activation compared with their precursor
(2) and lumazine (3) forms, although these are labile in water to
afford the lumazine forms (3a and 3g, respectively) within 1 h. The
previous X-ray crystallographic analysis of the complex between
5-OP-RU and MR1 revealed that 5-OP-RU are covalently bound
to the amino group of Lys43 on MR1, leading to the formation of
a stable complex.14 Given these, the formation process of stable
MR1-ligand complex in cells may be significantly associated with
the T cell activation. To estimate the formation kinetics of the
ligand-MR1 complex in cells, we carried out medium-exchange
experiments (Figure 6). HeLa.MR1 cells were incubated with 1a
or 3a in the initial 1 h. After removal of the medium and washing
with PBS, the cells were co-cultured with TG40.MAIT-TCR cells
in the medium without ligands for 23 h (Figure 6a). The control
Owing to the instability of the iminocarbonyl moiety, all isomers
1
gradually converted into their thermodynamically stable
lumazine forms 3 (Figure 2). These are also known to partially
return to the precursors 2 via hydrolysis. Namely, in these assays,
several components (including ribitylaminouracil-derived 1–3 and
methylglyoxal) would coexist during the experiment to have some
influence on the MAIT cell activation.17 Therefore, we also
evaluated the MAIT cell activities of these compounds (Figure 5
and Figure S3–S5). The methylglyoxal had no activity in the co-
culture assay using HeLa.MR1 and TG40.MAIT-TCR cells (Figure
S5). The SARs of the 5-A-RU stereoisomers 2 were similar to
those of the 5-OP-RU stereoisomers 1: only the (2’S,3’S,4’R)-
Figure 5. MAIT-cell Activation by 5-A-RU and RL-7-Me Stereoisomers. (a,b) TG40.MAIT-TCR cell activation was detected as CD69 surface expression using flow
cytometry. (c,d) IL-2 production was detected by ELISA after 24 h incubation in co-culture with HeLa.MR1 cells. The graphs show the mean ± SD of triplicate
measurements, and the results shown are representative of at least three independent experiments. ND = not detected.
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000466
ChemBioChem
FULL PAPER
experiment without washing was also conducted. There were no
differences observed between the two conditions in terms of the
activity for MAIT-TCR activation (Figure 6b), suggesting that 5-
OP-RU would immediately form a covalent bond with MR1 to
produce a stable complex during the initial 1 h incubation. Once
the complex is formed, 5-OP-RU remains tightly bound, likely due
to the Schiff base formation between the iminoglyoxal moiety and
MR1 Lys43. In contrast, the activity of 3a, lacking the iminoglyoxal
group for the covalent bond formation, remarkably decreased
under the medium-exchanged condition (Figure 6b). Taken
together, 5-OP-RU (1a) would form a covalent bond to MR1 in
cells within 1 h, before its disappearance as shown in Figure 2.
from those of 1a and 1g, causing the deterioration of the hydrogen
bond network in this region. The 4’-OH group of 1a had a
hydrogen bond formation with MR1 Arg94, while the
corresponding 4’-OH group of epimer 1g could interact with TCR
Gly98 and MR1 Tyr152. The 5’-OH group of 1a and 1g formed a
hydrogen bond with MR1 Tyr152 and MR1 Arg94, respectively.
These results highlight the important interactions of the ligands
with TCR Tyr95, MR1 Arg94 and MR1 Tyr152 to activate MAIT
cells. The potent activity of 1g would be associated with a well-
ordered hydrogen bond network within the ternary complex that is
similar to 1a.
Conclusion
In summary, we designed and synthesized all stereoisomers of 5-
OP-RU (1a) to investigate the effects of their chirality on the
activation of MAIT cells. Among the eight stereoisomers, only the
4’-OH epimer 1g exhibited potent T cell activity, revealing that the
configuration of the 2’-OH and 3’-OH group could be crucial for
agonistic activity. Further modification of the 4’-OH moiety would
be possible to improve the recognition by human MAIT-TCR. In
addition, the docking study suggested that the interaction
between the ligands and Tyr95 of TCR and/or Arg94 and Tyr152
of MR1 would contribute to the formation of the ternary complexes.
The previous results using deoxy analogues also demonstrate
that these interactions play a key role in the formation of ternary
complex.21 On the other hand, our docking study using the
stereoisomer suggested that the direct interaction between TCR
Gly98-ligand is a potential target to modulate their complex
formation. The comparable bioactivities of the lumazine forms
having the same stereochemistries suggested that modification of
the uracil core structure would be acceptable for developing
potent
ligands.
The
medium-exchange
experiments
demonstrated that the rapid formation of a covalent bond between
MR1 and 5-OP-RU could result in its potent activity, although 5-
OP-RU is a relatively short-lived intermediate under physiological
conditions. Accordingly, we obtained information for the design of
novel MAIT cell modulators. Based on these findings, the SAR
studies for the development of the drug-like agonist for MAIT cells
are underway in our laboratory.
Figure 6. Medium-Exchange Experiments. (a) Experimental scheme. (b) T cell
activation by ligand treatment. CD69 expression was monitored by flow
cytometry after 24 h incubation with or without medium-exchange. The graphs
show the mean ± SD of triplicate measurements, and the results shown are
representative of at least three independent experiments.
Experimental Section
Materials
5-A-RU, RL-7-Me and their derivatives were prepared by the reported
procedures. 5-OP-RU and its stereoisomers were freshly prepared before
each experiment by combining 10 mM of 5-A-RU or its stereoisomers with
a 40% aqueous solution of methylglyoxal (see: T Cell Activation Assay).
Acetyl-6-formylpterin (Ac-6-FP) was purchased from Cayman Chemical
Company.
Finally, we analyzed the binding mode of the 5-OP-RU
stereoisomers to hMR1 protein and MAIT-TCR with the Molecular
Operating Environment (MOE) (Figure 7). The reported X-ray
crystal structure of MR1-ligand-TCR ternary complex (PDB:
4L4V) was used as a template for the preparation of binding
models. The uracil ring and iminoglyoxal moieties of all
stereoisomers were accommodated in the MR1 cleft with similar
conformations, which are in agreement with the study by
Braganza et al.19 In terms of the ribityl moiety of the agonistic
isomers 1a and 1g having the common stereochemistries at the
2’-OH and 3’-OH groups, TCR Tyr95 and MR1 Arg94 interacted
with these 2’-OH and 3’-OH groups via hydrogen bonds,
respectively. In contrast, either 2’-OH or 3’-OH group of the non-
activating isomers 1b–1f and 1h could have a different orientation
Cell Line and Cell Culture
A human MR1-overexpressing HeLa cell line was established by retroviral
gene transduction of human MR1 and human CD8α as described
previously.33
A MAIT cell line was generated by retroviral gene
transduction of MAIT-αβTCRs into the murine thymoma TG40 (negative
for TCR), as described previously.34 HeLa cells were cultured in DMEM
medium (SIGMA) supplemented with 10% FBS (SIGMA), 1%
penicillin/streptomycin (Wako). TG40 cells were cultured in RPMI-1640
medium (nacalai) supplemented with 10% FBS, 1% penicillin/streptomycin.
During the T cell activation assay and MR1 upregulation assay, both cells
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000466
ChemBioChem
FULL PAPER
Figure 7. Binding Mode Analysis of 5-OP-RU 1a and Its Stereoisomer 1g Using the MR1-Ligand-TCR Ternary Complexes (PDB: 4L4V) as a Template.
Interactions between MAIT-TCR (yellow), MR1 (pale green) and (a) 1a, or (b) 1g with the distance (green).
were cultured in RPMI-1640 medium without folic acid, and with 10% FBS,
1% penicillin/streptomycin. All cell lines were grown at 37 °C in a
humidified 5 % CO2 atmosphere.
measured by a procedure identical with the T cell activation assay
described above.
As a control experiment, HeLa.MR1 cells (4.5×104 cells) were incubated
in 150 μL medium (RPMI 1640 without folic acid) with 1a or 3a for 1 h,
followed by the addition of 50 μL medium (RPMI 1640 without folic acid)
including TG40.MAIT-TCR cells (9.0×104 cells). Then, the mixture was co-
incubated with the derivatives for 23 h. Subsequently, the activation of
TG40 cells was measured by a procedure identical with the T cell
activation assay described above.
Stability Analysis
The stability analysis was carried out by modifications to a literature
procedure.17 To a solution of each precursor 2a, 2c, 2e and 2g (5.0 mg,
0.013 mmol) in D2O (730 μL) was added methylglyoxal (40% aqueous
solution; 58 μL, 0.33 mmol). Then, the amount of 5-OP-RU derivatives 1
1
and the RL-7-Me derivatives 3 were monitored using H-NMR with their
characteristic signals every 10 minutes until 120 min.
Binding Mode Study
The binding mode of the MR1-ligand (5-OP-RU derivative 1)-TCR ternary
complexes was analyzed using the Molecular Operating Environment
(MOE) with the X-ray crystal structure of MR1-ligand (7-hydroxy-6-methyl-
8-D-ribityllumazine)-TCR ternary complex (PDB: 4L4V) as a template. The
protein structure was preprocessed and optimized using the MOE
QuickPrep command with the default settings. After modifying the ligand
structure to 5-OP-RU or its stereoisomers, energy minimization
calculations of the complex structures were performed with the
Amber10:EHT force field. The minimization process of the residues within
8 Å around the ligand was carried out to predict the binding mode. In the
optimized ternary complexes, the distances between the ligands and the
close residues in MR1 and TCR were measured to elucidate the significant
interactions underpinning the MAIT cell activation.
T Cell Activation Assay
Briefly, HeLa.hMR1 cells (4.5×104 cells) were co-incubated with
TG40.MAIT-TCR cells (9.0×104 cells) for 24 h in 200 μL medium (RPMI
1640 without folic acid) with various ligands. Of the ligands, 5-OP-RU and
its stereoisomers (final concentration: 10 μM) were freshly prepared before
each experiment by the addition of 5-A-RU or its stereoisomers (10 mM
solution in DMSO; 4 μL), and a methylglyoxal (1 mM solution in the
medium; 1 mL) to the medium (1 mL). Then, each ligand solution (100 μL)
was added to the cell suspension (100 μL), followed by incubation. The
cells were subsequently stained with PE-conjugated anti-mouse CD3
(Biolegend), APC-conjugated anti-mouse CD69 (Biolegend) and
propidium iodide (PI), followed by Gallios flow cytometric analysis
(Beckman Coulter). Data were analyzed with Kaluza software (Beckman
Coulter, v2.1). The activation of TG40 cells was measured by an increase
in the cell-surface CD69 expression.
Acknowledgements
The supernatant was collected from the abovementioned incubation, and
the level of IL-2 was measured using an ELISA kit (BioLegend), in
accordance with the recommended protocol. The absorbance at 450 nm
was read using an iMark microplate reader (Bio-Rad).
This work was supported by the JSPS KAKENHI (Grant Numbers
JP18H04615, JP20K06938 and JP20H04773), and AMED under
Grant Number JP19gm1010007, JP20ak0101070 and
JP19ak0101072, and by the Uehara Memorial Foundation.
MR1 upregulation assay
Briefly, HeLa.hMR1 cells (1.5×105 cells) were incubated for 8 h in 200 μL
medium (RPMI 1640 without folic acid) with various ligands. The cells were
subsequently stained with PE-conjugated anti-human MR1 (Biolegend)
and PI, followed by Gallios flow cytometric analysis (Beckman Coulter).
Data were analyzed with Kaluza software (Beckman Coulter, v2.1).
Keywords: Immunochemistry • MAIT cell • MR1• 5-OP-RU•
Structure-activity relationships
[1]
[2]
L. Le Bourhis, E. Martin, I. Péguillet, A. Guihot, N. Froux, M. Coré, E.
Lévy, M. Dusseaux, V. Meyssonnier, V. Premel, C. Ngo, B. Riteau, L.
Duban, D. Robert, S. Huang, M. Rottman, C. Soudais, O. Lantz, Nat.
Immunol. 2010, 11, 701-708.
Medium-exchange experiment
HeLa.MR1 cells (4.5×104 cells) were incubated in 150 μL medium (RPMI
1640 without folic acid) with 1a or 3a for 1 h. After washing with PBS-(−)
and the addition of 200 μL medium (RPMI 1640 without folic acid) including
TG40.MAIT-TCR cells (9.0×104 cells), the mixture was co-incubated in the
absence of the derivatives for 23 h. The activation of TG40 cells was
M. Dusseaux, E. Martin, N. Serriari, I. Péguillet, V. Premel, D. Louis, M.
Milder, L. Le Bourhis, C. Soudais, E. Treiner, Blood 2011, 117, 1250-
1259.
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000466
ChemBioChem
FULL PAPER
[3]
[4]
[5]
[6]
D. I. Godfrey, H. -F. Koay, J. McCluskey, N. A. Gherardin, Nat. Immunol.
[26] The analogues of 5-OP-RU (1b, 1c, 1e and 1g) have been synthesized
previously and evaluated using mouse MR1 and mouse MAIT-TCR.19
[27] Mak et al. reported the other preparation method where, after mixing the
precursor (5-A-RU) and methylglyoxal in DMSO, the resulting product (5-
OP-RU) was purified by chromatography.17 Although we have confirmed
that 5-OP-RU isomers were produced in acceptable yields (Figure 2a),
the differences in the preparation methods could have influence on the
activity values.
2019, 20, 1110-1128.
M. Salio, J. D. Silk, E. Y. Jones, V. Cerundolo, Annu. Rev. Immunol. 2014,
32, 323-366.
D. I. Godfrey, A. P. Uldrich, J. McCluskey, J. Rossjohn, D. B. Moody, Nat.
Immunol. 2015, 16, 1114-1123.
R. Reantragoon, A. J. Corbett, I. G. Sakala, N. A. Gherardin, J. B.
Furness, Z. Chen, S. B. G. Eckle, A. P. Uldrich, R. W. Birkinshaw, O.
Patel, L. Kostenko, B. Meehan, K. Kedzierska, L. Liu, D. P. Fairlie, T. H.
Hansen, D. I. Godfrey, J. Rossjohn, J. McCluskey, L. Kjer-Nielsen, J. Exp.
Med. 2013, 210, 2305-2320.
[28] The 7-methyl-8-D-ribityllumazine (3a) has been synthesized and
evaluated with human MR1 and human MAIT-TCR.21
[29] Several cell surface markers (e.g. CD25, CD40L, CD69, CD137, OX40
etc.) can be commonly used for evaluation of T cell activation, which are
not considered to have essential differences. The measurements of IL-2
secretion are also used to investigate T cell activation.
[7]
E. Treiner, L. Duban, S. Bahram, M. Radosavljevic, V. Wanner, F. Tilloy,
P. Affaticati, S. Gilfillan, O. Lantz, Nature 2003, 422, 164-169.
A. I. Meierovics, S. C. Cowley, J. Exp. Med. 2016, 213, 2793-2809.
S. B. G. Eckle, A. J. Corbett, A. N. Keller, Z. Chen, D. I. Godfrey, L. Liu,
J. Y. W. Mak, D. P. Fairlie, J. Rossjohn, J. McCluskey, J. Biol. Chem.
2015, 290, 30204-30211
[8]
[9]
[30] H. Yamaguchi, M. Hirai, Y. Kurosawa, K. Hashimoto, Biochem. Biophys.
Res. Commun. 1997, 238, 697-702.
[31] S. B. G. Eckle, R. W. Birkinshaw, L. Kostenko, A. J. Corbett, H. E. G.
McWilliam, R. Reantragoon, Z. Chen, N. A. Gherardin, T. Beddoe, L. Liu,
O. Patel, B. Meehan, D. P. Fairlie, J. A. Villadangos, D. I. Godfrey, L.
Kjer-Nielsen, J. McCluskey, J. Rossjohn, J. Exp. Med. 2014, 211, 1585-
1600.
[10] H. E. G. McWilliam, S. B. G. Eckle, A. Theodossis, L. Liu, Z. Chen, J. M.
Wubben, D. P. Fairlie, R. A. Strugnell, J. D. Mintern, J. McCluskey, J.
Rossjohn, J. A. Villadangos, Nat. Immunol. 2016, 17, 531-537.
[11] L. Kjer-Nielsen, O. Patel, A. J. Corbett, J. Le Nours, B. Meehan, L. Liu,
M. Bhati, Z. Chen, L. Kostenko, R. Reantragoon, N. A. Williamson, A. W.
Purcell, N. L. Dudek, M. J. McConville, R. A. J. O’Hair, G. N. Khairallah,
D. I. Godfrey, D. P. Fairlie, J. Rossjohn, J. McCluskey, Nature 2012, 491,
717-723.
[32] H. -F. Koay, N. A. Gherardin, C. Xu, R. Seneviratna, Z. Zhao, Z. Chen,
D. P. Fairlie, J. McCluskey, D. G. Pellicci, A. P. Uldrich, D. I. Godfrey,
Nat. Commun. 2019, 10, 2243.
[33] T. Kitamura, Int. J. Hematol. 1998, 67, 351-359.
[12] A. Toubal, I. Nel, S. Lotersztajn, A. Lehuen, Nat. Rev. Immunol. 2019,
19, 643-657.
[34] T. Yokosuka, K. Takase, M. Suzuki, Y. Nakagawa, S. Taki, H. Takahashi,
T. Fujisawa, H. Arase, T. Saito, J. Exp. Med. 2002, 195, 991-1001.
[13] A. N. Keller, S. B. G. Eckle, W. Xu, L. Liu, V. A. Hughes, J. Y. W. Mak,
B. S. Meehan, T. Pediongco, R. W. Birkinshaw, Z. Chen, H. Wang, C.
D’Souza, L. Kjer-Nielsen, N. A. Gherardin, D. I. Godfrey, L. Kostenko, A.
J. Corbett, A. W. Purcell, D. P. Fairlie, J. McClusky, J. Rossjohn, Nat.
Immunol. 2017, 18, 402-411.
[14] A. J. Corbett, S. B. G. Eckle, R. W. Birkinshaw, L. Liu, O. Patel, J.
Mahony, Z. Chen, R. Reantragoon, B. Meehan, H. Cao, N. A. Williamson,
R. A. Strugnell, D. V. Sinderen, J. Y. W. Mak, D. P. Fairlie, L. Kjer-Nielsen,
J. Rossjohn, J. McCluskey, Nature 2014, 509, 361-365.
[15] J. Le Nours, N. A. Gherardin, S. H. Ramarathinam, W. Awad, F. Wiede,
B. S. Gully, Y. Khandokar, T. Praveena, J. M. Wubben, J. J. Sandow, A.
I. Webb, A. von Borstel, M. T. Rice, S. J. Redmond, R. Seneviratna, M.
L. Sandoval-Romero, S. Li, M. N. T. Souter, S. B. G. Eckle, A. J. Corbett,
H. H. Reid, L. Liu, D. P. Fairlie, E. M. Giles, G. P. Westall, R. W. Tothill,
M. S. Davey, R. Berry, T. Tiganis, J. McCluskey, D. G. Pellicci, A. W.
Purcell, A. P. Uldrich, D. I. Godfrey, J. Rossjohn, Science 2019, 366,
1522-1527.
[16] W. Awad, J. Le Nours, L. Kjer-Nielsen, J. McCluskey, J. Rossjohn,
Immunol. Cell Biol. 2018, 96, 588-597.
[17] J. Y. W. Mak, W. Xu, R. C. Reid, A. J. Corbett, B. S. Meehan, H. Wang,
Z. Chen, J. Rossjohn, J. McCluskey, L. Liu, D. P. Fairlie, Nat. commun.
2017, 8, 14599.
[18] G. J. M. Ler, W. Xu, J. Y. W. Mak, L. Liu, P. V. Bernhardt, D. P. Fairlie,
Chem. Eur. J. 2019, 25, 15594-15608.
[19] C. D. Braganza, K. Shibata, A. Fujiwara, C. Motozono, K. Sonoda, S.
Yamasaki, B. L. Stocker, M. S. M. Timmer, Org. biomol. Chem. 2019, 17,
8992-9000.
[20] J. Lange, R. J. Anderson, A. J. Marshall, S. T. S. Chan, T. S. Bilbrough,
O. Gasser, C. Gonzalez-Lopez, M. Salio, V. Cerundolo, I. F. Hermans,
G. F. Painter, ACS Chem. Biol. 2020, 15, 437-445.
[21] W. Awad, G. J. M. Ler, W. Xu, A. N. Keller, J. Y. W. Mak, X. Y. Lim, L.
Liu, S. B. G. Eckle, J. Le Nours, J. McCluskey, A. J. Corbett, D. P. Fairlie,
J. Rossjohn, Nat. Immunol. 2020, 21, 400-411.
[22] C. D. Braganza, C. Motozono, K. Sonoda, S. Yamasaki, K. Shibata, M.
S. M. Timmer, B. L. Stocker, Chem. Commun. 2020, 56, 5291-5294.
[23] Although the analogues of 5-A-RU (2b, 2c, 2e and 2g) have been
synthesized previously,19 the MAIT cell activation assays for these
analogues have not been conducted.
[24] K. Li, C. K. Vorkas, A. Chaudhry, D. L. Bell, R. A. Willis, A. Rudensky, J.
D. Altman, M. S. Glickman, J. Aubé, PloS ONE 2018, 13, e0191837.
[25] K. E. Peterson, M. A. Cinelli, A. E. Morrell, A. Mehta, T. S. Dexheimer, K.
Agama, S. Antony, Y. Pommier, M. Cushman, J. Med. Chem. 2011, 54,
4937-4953.
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000466
ChemBioChem
FULL PAPER
Entry for the Table of Contents
All stereoisomers of 5-OP-RU were synthesized to elucidate the effects of its stereochemistry on the MR1-dependent MAIT cell
activation. Of the stereoisomers, only the 4’-OH epimer demonstrated potent MAIT cell activity, indicating that the configuration of the
2’-OH and 3’-OH group could be important for agonistic activity.
This article is protected by copyright. All rights reserved.