Generation of Stable Isopentenyl Monophosphate Aryloxy Triester Phosphoramidates as Activators of Vγ9Vδ2 T Cells

Article abstract of DOI:10.1002/cmdc.202100198

Aryloxy triester phosphoramidate prodrugs of the monophosphate derivatives of isopentenyl pyrophosphate (IPP) and dimethylallyl pyrophosphate (DMAPP) were synthesized as lipophilic derivatives that can improve cell uptake. Despite the structural similarity of IPP and DMAPP, it was noted that their phosphoramidate prodrugs exhibited distinct stability profiles in aqueous environments, which we show is due to the position of the allyl bond in the backbones of the IPP and DMAPP monophosphates. As the IPP monophosphate aryloxy triester phosphoramidates showed favorable stability, they were subsequently investigated for their ability to activate Vγ9/Vδ2 T cells and they showed promising activation of this subset of T cells. Together, these findings represent the first report of IPP and DMAPP monophosphate prodrugs and the ability of IPP aryloxy triester phosphoramidate prodrugs to activate Vγ9/Vδ2 T cells highlighting their potential as possible immunotherapeutics.

Full text of DOI:10.1002/cmdc.202100198

Accepted Article
Title: Generation of Stable Isopentenyl Monophosphate Aryloxy
Triester Phosphoramidates as Activators of Vγ9Vδ2 T Cells
Authors: Qin Xu, Taher E Taher, Elizabeth Ashby, Maria Sharif,
Benjamin E Willcox, and Youcef Mehellou
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: ChemMedChem 10.1002/cmdc.202100198
Link to VoR: https://doi.org/10.1002/cmdc.202100198
01/2020

10.1002/cmdc.202100198
ChemMedChem
FULL PAPER
Generation of Stable Isopentenyl Monophosphate Aryloxy
Triester Phosphoramidates as Activators of Vγ9Vδ2 T Cells
Qin Xu,[a] Taher E. Taher,[b,c] Elizabeth Ashby,[b,c] Maria Sharif,[b,c] Benjamin E. Willcox,* [b,c] Youcef
Mehellou*[a]
[a]
Qin Xu, and Dr. Youcef Mehellou
School of Pharmacy and Pharmaceutical Sciences
Cardiff University
King Edward VII Avenue
Cardiff CF10 3NB, U.K.
E-mail: MehellouY1@cardiff.ac.uk
[b]
[c]
Dr. Taher E. Taher, Elizabeth Ashby, Maria Sharif and Prof. Benjamin E. Willcox
Cancer Immunology and Immunotherapy Centre
University of Birmingham
Edgbaston, Birmingham B15 2TT, U.K.
Dr. Taher E. Taher, Elizabeth Ashby, Maria Sharif and Prof. Benjamin E. Willcox
Institute of Immunology and Immunotherapy
University of Birmingham
Edgbaston, Birmingham B15 2TT, U.K.
E-mail: B.Willcox@bham.ac.uk
Supporting information for this article is given via a link at the end of the document.((Please delete this text if not appropriate))
intracellular B30.2 domain of Butyrophilin 3A1 (BTN3A1)[5]
,
Abstract: Aryloxy triester phosphoramidate prodrugs of the
monophosphate derivatives of isopentenyl pyrophosphate (IPP) and
dimethylallyl pyrophosphate (DMAPP) were synthesized as lipophilic
derivatives that can improve cell uptake. Despite the structural
similarity of IPP and DMAPP, it was noted that their phosphoramidate
prodrugs exhibited distinct stability profiles in aqueous environments,
which we show is due to the position of the allyl bond in the backbones
of the IPP and DMAPP monophosphates. As the IPP monophosphate
aryloxy triester phosphoramidates showed favorable stability, they
were subsequently investigated for their ability to activate Vγ9/Vδ2 T
cells and they showed promising activation of this subset of T cells.
Together, these findings represent the first report of IPP and DMAPP
monophosphate prodrugs and the ability of IPP aryloxy triester
phosphoramidate prodrugs to activate Vγ9/Vδ2 T cells highlighting
their potential as possible immunotherapeutics.
in a process that is dependent upon co-expression of
Butyrophlin 2A1, a direct ligand for the Vγ9Vδ2 T cell
receptor (TCR).[6]
IPP and DMAPP exhibit limited activation of Vγ9/Vδ2 T cells
(EC50 = 3 μM and 10 μM, respectively)[4] relative to (E)-4-
Hydroxy-3-methyl-but-2-enyl pyrophosphate (HMBPP, EC50
= 60-500 pM).[7] Although this limited activation may be due
to the structural differences between these compounds that
8]
affect their affinity to the BTN3A1 B30.2 domain,[5, we
hypothesised that inefficient cellular uptake resulting from
their polar nature that arises from the negatively charged
pyrophosphate groups at physiological pH (<7.4) may also
contribute to their limited activation of Vγ9/Vδ2 T cells.
Additionally, chemical or enzymatic dephosphorylation of
these compounds when used exogenously may be another
reason.
Introduction
To address these drawbacks, we decided to first reduce the
polarity of IPP and DMAPP and simplify the chemistry by
Vγ9/Vδ2 T cells, the predominant subtype of γδ T cells in
peripheral blood, have received substantial attention in
recent years due to their role in infectious and autoimmune
diseases as well as cancer.[1] An increasing number of
studies indicate that Vγ9/Vδ2 T cells are one of the more
potentially exploitable subsets of γδ T cells, with effector
mechanisms that can underpin powerful anti-microbial[2] and
anti-tumour[3] potential. Among the key small molecule
applying
biocleavable
masking
groups
to
the
monophosphate derivatives of IPP and DMAPP, namely 3-
activators of Vγ9/Vδ2 T cells is Zoledronate (1, Figure 1A),
which inhibits farnesyl diphosphate synthase, and this
ultimately leads to intracellular accumulation of isopentenyl
pyrophosphate (IPP
2
,
Figure 1A
) and dimethylallyl
pyrophosphate (DMAPP
3
,
Figure 1A).[4] The intracellular
Figure 1. A. Chemical structures of Zoledronate (1), IPP (2) and DMAPP (3). B.
Chemical structures and IPP and DMAPP monophosphate derivatives 4 and 5,
which are used in this study.
accumulation of IPP and DMAPP has been shown to lead to
the activation of Vγ9/Vδ2 T cells via their binding to the
1
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202100198
ChemMedChem
FULL PAPER
methylbut-3-en-1-yl dihydrogen phosphate (
4
,
Figure 1B
)
reported procedures.[11] These phosphorochloridates were
subsequently coupled to the commercially available 3-
methylbut-3-en-1-ol or 3-methylbut-2-en-1-ol in DCM at room
temperature for 16 h and using TEA as a base. The desired
and 3-methylbut-2-en-1-yl dihydrogen phosphate (5,
Figure
1B). This strategy of switching pyrophosphates to
monophosphates in the design of small molecule Vγ9/Vδ2 T
cell activators was previously used with success.[9]
phosphoramidates 8a
-
d
and 9a
-
d
were generated in low
mixture of
yields ranging from
diastereoisomers.
5
to 25% as
a
Results and Discussion
Notably, during the purification of phosphoramidates 9a-d,
we noticed that they were partially decomposed when
passing through the silica column, which is weakly acidic.
In terms of the choice of the monophosphate biocleavable
masking groups, we selected an aryl group and an amino
acid ester moieties to generate aryloxy triester
phosphoramidate prodrugs.[10] This approach has been
widely employed for similar purposes and has delivered
prodrugs with improved stability and cellular uptake
profiles.[10] Once inside the cell, the monophosphate masking
groups undergo enzymatic processing to release the native
monophosphate species, through a now well-established
process that involves two enzymes; carboxypeptidase Y and
the phosphoramidase enzyme Hint-1.[10]
Thus, upon the synthesis of 9a-d, we decided to compare the
stability of phosphoramidates 8d and 9d in acidic
environments. These two phosphoramidates were incubated
in aqueous acid (pH = 1) and monitored by 31P NMR over 12
h. The results showed that the IPP monophosphate
phosphoramidate, 8d, was stable in acid (pH =1) for the 12
h of the study, as the two 31P NMR singlets corresponding to
8d remained intact during the timeframe of the study and no
new metabolites were detected (Figure 3A). The DMAPP
monophosphate phosphoramidate 9d, however, was quickly
broken down in acid (pH = 1) and was consumed completely
With these in mind, we designed four different aryloxy triester
phosphoramidates of compounds
4 and 5 (Figure 2, 8a-d
A.
and 9a ). For the masking groups, we used phenol as the
-
d
8d + acidic buffer (12 h)
aryl motif and L-alanine as the amino acid since these two
moieties are the most widely in aryloxy triester
phosphoramidates and generate compounds with excellent
stability and cellular uptake.[10] In terms of the ester groups,
methyl (Me), isopropyl (iPr), tert-butyl (tBu), and benzyl (Bn)
were used. The choice of these esters is influenced by their
effect on the lipophilicity of the generated compounds as they
give a good spread of cLogP values (Supporting Table S1),
which often translates into noticeable differences in the
pharmacological activity.[10]
8d + acidic buffer (10 h)
8d + acidic buffer (8 h)
8d + acidic buffer (4 h)
8d + acidic buffer (2 h)
8d + acidic buffer (1 h)
8d + acidic buffer (0.5 h)
8d alone
The synthesis of aryloxy triester phosphoramidates 8a
9a started by synthesising phosphorochloridates 7a
which was achieved by reacting phenyl dichlorophosphate
-d and
-d
-d,
6
15
10
5
0
-5
-10
-15
-20
with the appropriate amino acid ester in DCM and in the
presence of triethylamine (TEA) at -78 ºC for 3 h using
ppm
B.
9d + acidic buffer (12 h)
9d + acidic buffer (8 h)
9d + acidic buffer (5 h)
9d + acidic buffer (3 h)
9d + acidic buffer (2 h)
9d + acidic buffer (1.5 h)
9d + acidic buffer (1 h)
9d + acidic buffer (0.5 h)
9d alone
15
10
5
0
-5
-10
-15
-20
ppm
Figure 3. Stability of IPP and DMAPP monophosphate derivatives
aryloxytriester phosphoramidates in acid. A. 31P NMR of compound 8d in acidic
buffer (pH = 1) at 37 ºC for 12 h. B. 31P NMR of compound 9d in acidic buffer
(pH = 1) at 37 ºC for 12 h.
2
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202100198
ChemMedChem
FULL PAPER
within 1 h as judged by the 31P NMR (Figure 3B). Since the
only difference between 8d and 9d is the position of the
double bond, we hypothesised that in aqueous acid
environments where there is an abundance of water, water
molecules act as nucleophiles and perform a nucleophilic
attack on the double bond of phosphoramidate 9d where the
electrons are passed to phosphoramidate moiety through the
allylic carbons resulting in the breakdown of 9d as illustrated
S5) whereas phosphoramidate 9d reacted with O,O-diethyl
thiophosphate and formed the metabolite O,O-diethyl S-(2-
methylbut-3-en-2-yl)
Collectively, this stability data supports our initial hypothesis
regarding the instability of phosphoramidates 9a being a
phosphorothioate
(
Figure
5).
-
d
result of the location of the double bond and the presence of
a good leaving group, the phosphoramidate moiety (11) in
this case. It is worth noting that we previously[9] observed
such instability with the phosphoramidate prodrugs of
in Figure 4A
.
compounds similar to 9a-d.
compound
13
compound 9d
Figure 4. A. Proposed mechanism of 9d breakdown in aqueous acidic buffer.
B. Scheme showing the possible mechanism by which strong nucleophiles
mediate the breakdown of 9d.
To test this hypothesis, we first replaced the masking groups
of the phosphoramidate moiety of 9d with diethyl groups so
that a diethyl monophosphate derivative, compound 12, is
generated (see Supporting Figure S2). In this case, we
hypothesised that because the diethyl monophosphate is a
poor leaving group compared to the phosphoramidate moiety
Figure 5. 31P NMR of the reaction involving compound 9d and the nucleophile
O,O-diethyl thiophosphate after 1 h.
Given that phosphoramidates 9a
aqueous environments, we subsequently focused our
investigation on phosphoramidates 8a and studied their
-d were unstable in
11
(Figure 4A), the electrons will not be pushed and
delocalised in the diethyl monophosphate group. Indeed,
conducting the stability of compound 12 in acidic buffer (pH
= 1) and monitoring the sample with 31P NMR showed that it
is completely stable for 12 h as expected (see Supporting
Figure S3).
-
d
ability to activate Vγ9/Vδ2 T cells. For this, peripheral blood
mononuclear cells (PBMCs) containing Vγ9/Vδ2 T cells
derived from healthy donors were incubated with known
Vγ9/Vδ2 T cell activators Zoledronate (30 μM and 100 μM)
To probe this stability hypothesis further, we subsequently
reacted phosphoramidate 9d with O,O-diethyl thiophosphate
or HMBPP (30 μM and 100 μM), or phosphoramidates 8a-d
at two concentrations (30 μM and 100 μM). Peripheral blood
γδ T cells lack appreciable levels of surface CD69 or CD25
under steady state conditions, but TCR stimulation
upregulates both T cell activation markers.[12] Vγ9/Vδ2 T cells
responsive to stimulation by Zoledronate, HMBPP or
(Figure 4B), a stronger nucleophile than water. The choice
of this reagent was also driven by the fact that we would be
able to purify the products of the reaction and reveal what
takes place. Following the reaction by TLC indicated the
rapid breakdown of 9d and the formation of two new
products. Upon column chromatography, only one product
was isolated. Analytical chemistry techniques (NMR and
mass spectrometry) confirmed this product to be O,O-diethyl
phosphoramidates 8a-d were then distinguished by TCR Vγ9
and Vδ2 expression and assessed for the upregulation of
CD69 and CD25.
As shown in Figure 6 and Supporting Figure S6
,
S-(2-methylbut-3-en-2-yl) phosphorothioate 13 (Figure 4B)
Zoledronate induced 34.3±4.4% and 47±4.4% activation of
as expected (Supporting Figure S4). Repeated attempts
Vγ9/Vδ2 T-cells at 30 μM and 100 μM, respectively. In
aimed to isolate the second reaction product, which we
comparison, phosphoramidates 8a
of activation of Vγ9/Vδ2
-
d
showed varying levels
believed it to be compound 11
(Figure 4B) were not
T
cells. Out of these
successful. This is likely due to the instability of this
compound.
phosphoramidates, compound 8a exhibited the highest
activation of Vγ9/Vδ2 T cells and it was able to induce
30.7±4.3% and 23.3±4.6% activation of Vγ9/Vδ2 T cells at
30 μM and 100 μM, respectively. Notably, the level of
activation of Vγ9/Vδ2 T cells by phosphoramidate 8a at 30
μM was comparable to that of Zoledronate at the same
concentration (30 μM). However, such activation did not
increase at the higher concentration studied (100 μM). This
may suggest that phosphoramidate 8a is a potent activator
of Vγ9/Vδ2 T cells, and at the lowest concentration we
To confirm the role of the position of the double bond and the
nature of the phosphate masking groups being critical for the
stability of phosphoramidates 8d and 9d, we run in parallel
two reactions where phosphoramidates 8d and 9d were
reacted with O,O-diethyl thiophosphate for 1 h, and studied
the reaction by 31P NMR. The results showed that
phosphoramidate 8d was stable in the presence of the
nucleophile O,O-diethyl thiophosphate (Supporting Figure
3
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202100198
ChemMedChem
FULL PAPER
B.
A.
80
80
60
40
20
0
did not induce any activation of CD8+ αβ T cells
(Supporting Figure S7), which are activated by peptides
60
40
20
0
and are not responsive to zoledronate. This confirms the
specificity of phosphoramidates 8a-d toward Vγ9/Vδ2 T
cells.
Conclusion
Figure 6. IPP monophosphate aryloxytriester phosphoramidates-mediated
activation of Vγ9/Vδ2 T-cells. Levels of activation are measured as the % of
Vγ9/Vδ2 T-cells expressing both CD69 and CD25 following overnight
incubation with 100 µM (A) or 30 µM (B) of IPP monophosphate aryloxytriester
phosphoramidates, Zoledronate and HMBPP. Data is shown as mean ± SEM
(n = 3).
Aryloxy triester phosphoramidates of the monophosphate
derivative of DMAPP (9a-d) were unstable in acidic aqueous
environments compared to those of the IPP monophosphate
derivative (8a-d). In terms of Vγ9/Vδ2 T cells activation, the stable
IPP monophosphate phosphoramidate prodrugs (8a-d) exhibited
varied activation of Vγ9/Vδ2 T cells with phosphoramidate 8a
being the most potent activator. Given the interest in small
molecule activators of Vγ9/Vδ2 T cells in immunotherapeutic
strategies targeting cancer and microbial infections, IPP
monophosphate phosphoramidates disclosed in this work are
thus possible candidates to be further investigated for the
treatments of these diseases.
studied (30 μM), it had already passed the maximum level of
Vγ9/Vδ2 T cells activation. This notion is supported by the
fact that at 100 μM the level of activation was almost the
same (or a little bit less) than that observed at 30 μM. Beyond
phosphoramidate 8a, compounds 8b and 8c showed some
activation of Vγ9/Vδ2 T cells, with phosphoramidate 8b
exhibiting a slightly better activation than 8c at both
concentrations studied (30 μM and 100 μM). This activation
pattern is in line with the established structure-activity
relationship (SAR) of aryloxy triester phosphoramidate
prodrugs where phosphoramidates bearing methyl (8a) and
isopropyl (8b) often show better pharmacological activity
than those bearing a tert-butyl ester (8c). This is due to the
fact that the tert-butyl ester of aryloxy triester
phosphoramidates is not as efficiently metabolised by
esterases compared to phosphoramidates bearing methyl
Experimental Section
General information. All reagents and solvents were of general purpose
or analytical grade and were purchased from Sigma-Aldrich Ltd., Fisher
Scientific, Fluorochem, or Acros. 31P, 1H, and 13
C NMR data were
recorded on a Bruker AVANCE DPX500 spectrometer operating at 202,
500, and 125 MHz, respectively. Chemical shifts (δ) are quoted in ppm,
and J values are quoted in Hz. In reporting spectral data, the following
abbreviations were used: s (singlet), d (doublet), t (triplet), q (quartet), dd
(doublet of doublets), td (triplet of doublets), and m (multiplet). All of the
reactions were carried out under a nitrogen atmosphere and were
monitored using analytical thin layer chromatography on precoated silica
plates (kiesel gel 60 F254, BDH). Compounds were visualized by
illumination under UV light (254 nm) or by the use of KMnO4 stain followed
by heating. Flash column chromatography was performed with silica gel
60 (230–400 mesh) (Merck).
General method for the synthesis of phosphorochloridates. L-alanine
ester hydrochloride (1 eq., 4.74 mmol) was dissolved in 20 mL anhydrous
DCM under nitrogen atmosphere and phenyl dichlorophosphate (6) (1 eq.,
0.71 mL, 4.74 mmol) was added before cooling the mixture down to -78
ºC. Triethylamine (2 eq., 1.3mL, 9.48 mmol) was then slowly added to the
mixture in half an hour. The solution was then allowed to warm up to room
temperature and stirred for 3 hrs. After checking by TLC (EtOAc: hexane
7:3, Rf = 0.9), the solvents were removed under reduced pressure and the
resulting mixture was dissolved by 10 mL diethyl ether and filtered. The
undissolved residue was washed by diethyl ether (2 × 10 mL) and all the
ether parts were combined, concentrated and used for the next step
without further purification.
General method for the synthesis of IPP and DMAPP
phosphoramidates. In oven-dried round bottom flask was charged with
either 3-methyl-3-buten-1-ol for 8a-d (1 eq., 0.46 mL, 4.74 mmol) or 3-
methyl-2-buten-1-ol for 9a-d (1 eq., 0.46 mL, 4.74 mmol) in 10 mL
anhydrous DCM under nitrogen atmosphere. The solution was cooled to 0
ºC and 1 eq. of triethylamine (0.66mL, 4.74 mmol) was added, followed by
the appropriate phosphorochloridate (4.74 mmol) in 1 mL dichloromethane
dropwise. The mixture was allowed to warm up slowly to room temperature
and stirred overnight. Then, the solvent was removed under reduced
pressure and the crude was purified by column chromatography (5%
methanol: 1% triethylamine: 94% chloroform) to give the desired pure
product.
and
isopropyl
esters.[10]
Surprisingly,
however,
phosphoramidate 8d, which bears a benzyl ester did not
show any activation of Vγ9/Vδ2 T cells. In the SAR of aryloxy
triester phosphoramidates, those bearing a benzyl ester
often show better pharmacological activities compared to
their phosphoramidate counterparts, which carry a methyl,
isopropyl and tert-butyl esters. This is because
phosphoramidates with a benzyl ester have comparatively
higher lipophilicity, which allows them to enter cells (via
passive diffusion) more efficiently, and this ultimately
translates into better pharmacological activity. Critically, the
established SAR of aryloxy triester was drawn from applying
this technology to highly polar compounds (mostly
nucleosides) and thus notable increases of lipophilicity by
applying the aryloxy triester phosphoramidates, especially
the ones bearing a benzyl ester, were needed to bring the
lipophilicity to 2-4 (logP) so that significant cellular uptake is
achieved. However, in this study, the parent backbone 3-
methylbut-3-en-1-ol is not very polar and thus by making its
aryloxy triester phosphoramidates that bear benzyl ester, the
lipophilicity of the generated 9d was most probably too high
(cLogP = 4.99, Supporting Table S1). This compromised
the aqueous solubility of 8d and thus it was not significantly
soluble in the cell culture media for it to induce a detectable
Vγ9/Vδ2 T cells activation.
Beyond γδ T cell activation, phosphoramidates 8a
-d and
Methyl
(((3-methylbut-3-en-1-yl)oxy)(phenoxy)phosphoryl)-L-
1
alaninate (8a). Yield: 373 mg (24% over two steps). H NMR (500 MHz,
Zoledronate tested at concentrations of 30 μM and 100 μM
4
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202100198
ChemMedChem
FULL PAPER
CDCl3, δ): 7.22-7.26 (m, 2H, Ph), 7.12-7.15 (m, 2H, Ph), 7.06-7.09 (m, 1H,
Ph), 4.74 (d, J = 4.9 Hz, 1H, =CH), 4.67(d, J = 8.3 Hz, 1H, =CH), 4.09-4.16
(m, 2H, OCH2), 3.96 (m, 1H, NH), 3.62, 3.65(2s, 3H, OCH3, from R and S
isomer), 3.39-3.47 (m, 1H, CH), 2.29-2.34 (m, 2H, CH2), 1.67 (d, J = 5.65
Hz , 3H, CH3), 1.30 (t, J = 7.3 Hz, 3H, CH3). 13C NMR (125 MHz, CDCl3,
δ): 174.0, 150.82, 141.0, 129.60, 124.78, 120.24, 112.70, 65.32, 52.44,
50.13, 38.19, 22.42, 21.06. 31P NMR (202 MHz, CDCl3, δ): 2.19, 2.11.
HRMS (ES+,m/z) calcd. for (M+H)+ C15H23NO5P+: 328.1308; found:
328.1306. HRMS (ES+,m/z) calcd. for (M+Na)+ C15H22NNaO5P+:
350.1128; found: 350.1127.
2H, CH2), 3.90-3.93 (m, 1H, CH), 3.51 (s, 1H, NH), 1.77 (s, 3H, CH3), 1.71
(s, 3H, CH3), 1.45 (d, J = 10.20, 9H, 3×CH3), 1.33-1.39 (m, 3H, CH3). 13
C
NMR (125 MHz, CDCl3, δ): 172.70, 151.02, 139.36, 129.54, 124.60,
120.25, 119.21, 63.83, 50.72, 45.78, 27.89, 25.73, 21.17, 18.01. 31P NMR
(202MHz, CDCl3, δ): 2.76, 2.82. HRMS (ES+,m/z) calcd. for (M+H)+
C18H29NO5P+: 370.1778; found: 370.1780. HRMS (ES+,m/z) calcd. for
(M+Na)+ C18H28NNaO5P+: 392.1597; found: 392.1593.
Benzyl
(((3-methylbut-2-en-1-yl)oxy)(phenoxy)phosphoryl)-L-
alaninate (9d). Yield: 125.39 mg (6.56% over two steps). 1H NMR (500
MHz, CDCl3, δ): 7.04-7.25 (m, 10H, Ph), 5.25-5.29 (m, 1H, =CH), 5.04,
5.06 (2s, 2H, CH2, from R and S isomer), 4.48-4.55 (m, 2H, CH2), 3.95-
4.04 (m, 1H, CH), 3.40-3.47 (m, 1H, NH), 1.66 (d, J = 8.50 Hz, 3H, CH3),
1.60 (d, J = 2.75 Hz, 3H, CH3), 1.30 (dd, J = 9.10 Hz, 7.15 Hz, 3H, CH3).
13C NMR (125 MHz, CDCl3, δ): 173.53, 151.38, 139.71, 135.47, 129.73,
128.82, 128.64, 128.37, 124.84, 120.40, 119.30, 67.33, 63.93, 50.45,
25.91, 21.28, 18.22. 31P NMR (202MHz, CDCl3, δ): 2.42, 2.51. HRMS
(ES+,m/z) calcd. for (M+H)+ C21H27NO5P+: 404.1621; found: 404.1616.
HRMS (ES+,m/z) calcd. for (M+Na)+ C21H26NNaO5P+: 426.1441; found:
426.1440.
Isopropyl
(((3-methylbut-3-en-1-yl)oxy)(phenoxy)phosphoryl)-L-
alaninate (8b). Yield: 357 mg (21.2% over two steps). 1H NMR (500 MHz,
CDCl3, δ): 7.3 (td, J = 7.8 Hz, 2.9 Hz, 2H, Ph), 7.2 (t, J = 8.82 Hz, 2H, Ph),
7.13 (t, J = 7.35 Hz, 1H, Ph), 5.0 (m, 1H, CH), 4.81 (m, 1H, =CH), 4.74 (d,
J = 7.1 Hz, 1H, =CH), 4.15-4.23 (m, 2H, CH2), 3.92-3.99 (m, 1H, NH), 3.52
(q, J = 10.43 Hz, 1H, CH), 2.36-2.41 (m, 2H, CH2), 1.73 (d, J = 5.25 Hz,
3H, CH3), 1.35 (t, J = 6.97 Hz, 4.67 Hz, 3H, CH3), 1.22 (m, 6H, 2×CH3).
13C NMR (125 MHz, CDCl3, δ): 173.0, 150.86, 141.08, 129.58, 124.75,
120.26, 112.69, 69.15, 65.31, 50.32, 38.24, 22.43, 21.69, 21.61, 21.07. 31
P
NMR (202 MHz, CDCl3, δ): 2.36, 2.28. HRMS (ES+,m/z) calcd. for (M+H)+
C17H27NO5P+: 356.1621; found: 356.1617. HRMS (ES+,m/z) calcd. for
(M+Na)+ C17H26NNaO5P+: 378.1441; found: 378.1439.
Diethyl
(3-methylbut-2-en-1-yl)
phosphate
(12).
Diethyl
phosphorochloridate (1 eq., 2.9 mmol, 0.42 ml) and 3-methyl-2-buten-1-ol
(1 eq, 2.9 mmol, 0.29 ml) were added in 5 ml DCM. The mixture was cooled
to -78ºC and triethylamine (1 eq., 2.9 mmol, 0.4 ml) was added dropwise.
After stirring for 30 min, the reaction was allowed to warm up to room
temperature and stirred overnight. The solvents were then removed, and
the crude was purified by flash column chromatography (EtOAc: Hexane
6:4). Yield: 204 mg (32%). 1H NMR (500 MHz, CDCl3, δ): 5.40 (m, 1H,
(CH3)2C=CHCH2), 4.55 (m, 2H, (CH3)2C=CHCH2), 4.10 (m, 4H,
2×CH2CH3), 1.77 (s, 3H, (CH3)2C=CHCH2), 1.71 (s, 3H, (CH3)2C=CHCH2),
1.33 (t, J = 7.1 Hz, 6H, 2×CH2CH3). 13C NMR (125 MHz, CDCl3, δ): 139.40,
119.22, 64.06, 63.61, 25.75, 18.00, 16.11. 31P NMR (202MHz, CDCl3, δ):
-0.59.
In Vitro Stability Assays Protocols. 4.0 mg of the prodrug was dissolved
in 0.2 mL of methanol-d, and 0.2 ml of acidic buffer (1:1 0.2 M HCl: 0.2 M
KCl) was added. The experiment was run on NMR phosphorus mode and
scanned every half an hour for 12 hrs. The incubation temperature was
37°C.
Activation of Vγ9/Vδ2 T-cells. Peripheral blood mononuclear cells
(PBMCs) were isolated from heparinised blood obtained from consented
healthy donors (Ethical approval was obtained from the NRES Committee
West Midlands ethical board; REC reference 14/WM/1254) using
Lymphoprep (Stem Cell Technologies). PBMCs were seeded out into
round-bottom 96-well plates at 5 × 105 cells in a total volume of 200 μL of
RPMI-1640 media supplemented with 2 mM L-glutamine, 25 mM HEPES,
1% sodium pyruvate, 50 μg/ml penicillin/streptomycin (Invitrogen) and
10% foetal calf serum (Sigma) per well, and cultured for 20 hours in the
presence of medium alone or the indicated concentration of zoledronate
(Sigma), HMBPP (Echelon Biosciences Inc) and IPP monophosphate
phosphoramidate prodrugs. To assess Vγ9Vδ2 T-cell activation, cells
were stained with Zombie Aqua viability dye (1:400), CD3 (UCHT1; 1:100),
CD8 (SK1; 1:100), CD25 (2A3; 1:100); all Biolegend, CD69 (TP1.55.3;
1:25) and TCR Vγ9 (IMMU360; 1:400); Beckman Coulter, and TCR Vδ2
(123R3; 1:200); Miltenyi. All samples were acquired using LSRFortessa
X20 (BD Biosciences), and all data were analysed with FlowJo v10 and
GraphPad Prism software.
Tert-butyl
(((3-methylbut-3-en-1-yl)oxy)(phenoxy)phosphoryl)-L-
alaninate (8c). Yield: 301.1 mg (17.2% over two steps). 1H NMR (500 MHz,
CDCl3, δ): 7.23 (td, J = 7.95 Hz, 1.95 Hz, 2H, Ph), 7.14 (t, J = 7.72 Hz, 2H,
Ph), 7.06 (t, J = 7.35 Hz, 1H, Ph), 4.74 (s, 1H, =CH), 4.67 (d, J = 2.35, 1H,
=CH), 4.09-4.18 (m, 2H, CH2), 3.82 (s, 1H, NH), 3.44 (m, 1H, CH), 2.32 (q,
J = 7.05, 2H, CH2), 1.66 (d, J = 3.35 Hz, 3H, CH3), 1.36 (d, J = 11.65 Hz,
9H, 3×CH3), 1.26 (dd, J = 6.97 Hz, 3.42 Hz, 3H, CH3).13C NMR (125 MHz,
CDCl3, δ): 172.64, 150.91, 141.12, 129.57, 124.71, 120.27, 112.67, 81.92,
65.28, 50.74, 38.25, 27.92, 22.47, 21.19. 31P NMR (202MHz, CDCl3, δ):
2.49, 2.44. HRMS (ES+,m/z) calcd. for (M+H)+ C18H29NO5P+: 370.1778;
found: 370.1774. HRMS (ES+,m/z) calcd. for (M+Na)+ C18H28NNaO5P+:
392.1597; found: 392.1594.
Benzyl
(((3-methylbut-3-en-1-yl)oxy)(phenoxy)phosphoryl)-L-
alaninate (8d). Yield: 301.3 mg (15.76% over two steps). 1H NMR (500
MHz, CDCl3, δ): 7.14-7.37 (m, 10H, Ph), 5.15 (d, J = 14 Hz, 2H, CH2), 4.85
(d, J = 9.35 Hz, 1H, =CH), 4.75 (d, J = 14 Hz, 1H, =CH), 4.14-4.28 (m, 2H,
CH2), 4.05-4.13 (m, 1H, CH), 3.57 (brs, 1H, NH), 2.38 (dt, J = 21.25, 6.82,
2H, CH2), 1.74 (d, J = 10.20, 3H, CH3), 1.41 (t, J = 5.72, 3H, CH3). 13C
NMR (125 MHz, CDCl3, δ): 173.38, 150.84, 141.08, 135.27, 129.58,
128.64, 128.50, 128.22, 124.77, 120.26, 112.71, 67.19, 65.32, 50.29,
38.23, 22.44, 21.05. 31P NMR (202MHz, CDCl3, δ): 2.19, 2.10. HRMS
(ES+,m/z) calcd. for (M+H)+ C21H27NO5P+: 404.1621; found: 404.1612.
HRMS (ES+,m/z) calcd. for (M+Na)+ C21H26NNaO5P+: 426.1441; found:
426.1435.
Methyl
(((3-methylbut-2-en-1-yl)oxy)(phenoxy)phosphoryl)-L-
alaninate (9a). Yield: 167.7 mg (10.8% over two steps). 1H NMR (500 MHz,
CDCl3, δ): 7.23 (t, J = 7.17 Hz, 2H, Ph), 7.14 (t, J = 8.32 Hz, 2H, Ph), 7.06
(t, J = 7.30, 1H, Ph), 5.28-5.34 (m, 1H, =CH), 4.57-4.51 (m, 2H, CH2), 3.92-
3.99 (m, 1H, NH), 3.62 (d, J = 12.10, 3H, OCH3), 3.46 (q, J = 11.13, 1H,
CH), 1.68 (d, J = 4.23, 3H, CH3), 1.62 (d, J = 4.53, 3H, CH3), 1.29 (dd, J =
11.47, 7.07, 3H, CH3). 13C NMR (125 MHz, CDCl3, δ): 174.09, 151.01,
139.56, 129.63, 124.72, 120.28, 119.19, 63.81, 52.45, 50.21, 25.80, 21.12,
18.09. 31P NMR (202MHz, CDCl3, δ): 2.44, 2.53. HRMS (ES+,m/z) calcd.
for (M+H)+ C15H23NO5P+: 328.1308; found: 328.1326. HRMS (ES+,m/z)
calcd. for (M+Na)+ C15H22NNaO5P+: 350.1128; found: 350.1126.
Isopropyl
(((3-methylbut-2-en-1-yl)oxy)(phenoxy)phosphoryl)-L-
Acknowledgements
alaninate (9b). Yield: 100 mg (5.94% over two steps). 1H NMR (500 MHz,
CDCl3, δ): 7.23 (td, J = 7.88 Hz, 2.72 Hz, 2H, Ph), 7.14 (t, J = 8.52, 2H,
Ph), 7.06 (t, J = 7.26 Hz, 1H, Ph), 5.32-5.35 (m, 1H, =CH), 4.89-4.98 (m,
1H, CH), 4.51-4.58 (m, 2H, CH2), 3.85-3.92 (m, 1H, CH), 3.43-3.50 (m, 1H,
NH), 1.67 (s, 3H, CH3), 1.61(s, 3H, CH3), 1.27 (t, 3H, CH3), 1.12-1.17 (m,
6H, 2×CH3).13C NMR (125 MHz, CDCl3, δ): 173.2, 151.11, 136.45, 129.69,
124.78, 123.83, 120.40, 120.34, 119.30, 69.17, 59.48, 50.46, 25.89, 21.77,
21.23, 17.97. 31P NMR (202MHz, CDCl3, δ): 2.67, 2.59. MS not found.
The work was supported by a Wellcome Trust ISSF grant (Grant
code: 514079) and a MRC Confidence in Concept (Grant code:
519638) awarded to Y.M.; Wellcome Trust Investigator award
funding to B.E.W. (Grant code: 099266/Z/12/Z); and Rosetrees
Trust PhD funding to M.S. (Grant code: M924).
Tert-butyl
(((3-methylbut-2-en-1-yl)oxy)(phenoxy)phosphoryl)-L-
alaninate (9c). Yield: 105 mg (6% over two steps). 1H NMR (500 MHz,
CDCl3, δ): 7.14-7.34 (m, 5H, Ph), 5.41-5.43 (m, 1H, =CH), 4.62-4.65 (m,
Conflict of Interest
5
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202100198
ChemMedChem
FULL PAPER
K. Muqit, Y. Mehellou, J. Med. Chem. 2017, 60, 3518-
3524.
The authors declare no conflict of interests.
[12] M. S. Davey, C. Y. Lin, G. W. Roberts, S. Heuston, A. C.
Brown, J. A. Chess, M. A. Toleman, C. G. Gahan, C. Hill, T.
Parish, J. D. Williams, S. J. Davies, D. W. Johnson, N.
Topley, B. Moser, M. Eberl, PLoS Pathog. 2011, 7,
e1002040.
Keywords: IPP • DMAPP • Prodrug • Phosphoramidate • T-cell
[1] a) J. Howard, I. Zaidi, S. Loizon, O. Mercereau-Puijalon, J.
Dechanet-Merville, M. Mamani-Matsuda, Front.
Immunol. 2018, 9, 2760; b) C. Junqueira, R. B. Polidoro,
G. Castro, S. Absalon, Z. Liang, S. Sen Santara, Â. Crespo,
D. B. Pereira, R. T. Gazzinelli, J. D. Dvorin, J. Lieberman,
Nat. Immunol. 2021, 22, 347-357; c) C. M. Shiromizu, C.
C. Jancic, Front. Immunol. 2018, 9, 2389; d) B. Silva-
Santos, S. Mensurado, S. B. Coffelt, Nat. Rev. Cancer
2019, 19, 392-404.
[2] K. S. Singh, R. Sharma, P. A. N. Reddy, P. Vonteddu, M.
Good, A. Sundarrajan, H. Choi, K. Muthumani, A.
Kossenkov, A. R. Goldman, H. Y. Tang, M. Totrov, J.
Cassel, M. E. Murphy, R. Somasundaram, M. Herlyn, J. M.
Salvino, F. Dotiwala, Nature 2021, 589, 597-602.
[3] K. K. Payne, J. A. Mine, S. Biswas, R. A. Chaurio, A.
Perales-Puchalt, C. M. Anadon, T. L. Costich, C. M. Harro,
J. Walrath, Q. Ming, E. Tcyganov, A. L. Buras, K. E.
Rigolizzo, G. Mandal, J. Lajoie, M. Ophir, J. Tchou, D.
Marchion, V. C. Luca, P. Bobrowicz, B. McLaughlin, U.
Eskiocak, M. Schmidt, J. R. Cubillos-Ruiz, P. C. Rodriguez,
D. I. Gabrilovich, J. R. Conejo-Garcia, Science 2020, 369,
942-949.
[4] Y. Tanaka, C. T. Morita, Y. Tanaka, E. Nieves, M. B.
Brenner, B. R. Bloom, Nature 1995, 375, 155-158.
[5] M. Salim, T. J. Knowles, A. T. Baker, M. S. Davey, M.
Jeeves, P. Sridhar, J. Wilkie, C. R. Willcox, H. Kadri, T. E.
Taher, P. Vantourout, A. Hayday, Y. Mehellou, F.
Mohammed, B. E. Willcox, ACS Chem. Biol. 2017, 12,
2631-2643.
[6] a) M. M. Karunakaran, C. R. Willcox, M. Salim, D. Paletta,
A. S. Fichtner, A. Noll, L. Starick, A. Nöhren, C. R. Begley,
K. A. Berwick, R. A. G. Chaleil, V. Pitard, J. Déchanet-
Merville, P. A. Bates, B. Kimmel, T. J. Knowles, V.
Kunzmann, L. Walter, M. Jeeves, F. Mohammed, B. E.
Willcox, T. Herrmann, Immunity 2020, 52, 487-498.e486;
b) M. Rigau, S. Ostrouska, T. S. Fulford, D. N. Johnson, K.
Woods, Z. Ruan, H. E. G. McWilliam, C. Hudson, C.
Tutuka, A. K. Wheatley, S. J. Kent, J. A. Villadangos, B. Pal,
C. Kurts, J. Simmonds, M. Pelzing, A. D. Nash, A. Hammet,
A. M. Verhagen, G. Vairo, E. Maraskovsky, C. Panousis, N.
A. Gherardin, J. Cebon, D. I. Godfrey, A. Behren, A. P.
Uldrich, Science 2020, 367, eaay5516.
[7] a) H. Kadri, T. E. Taher, Q. Xu, M. Sharif, E. Ashby, R. T.
Bryan, B. E. Willcox, Y. Mehellou, J. Med. Chem. 2020, 63,
11258-11270; b) M. Hintz, A. Reichenberg, B. Altincicek,
U. Bahr, R. M. Gschwind, A. K. Kollas, E. Beck, J. Wiesner,
M. Eberl, H. Jomaa, FEBS Lett. 2001, 509, 317-322.
[8] A. Sandstrom, C. M. Peigné, A. Léger, J. E. Crooks, F.
Konczak, M. C. Gesnel, R. Breathnach, M. Bonneville, E.
Scotet, E. J. Adams, Immunity 2014, 40, 490-500.
[9] M. S. Davey, R. Malde, R. C. Mykura, A. T. Baker, T. E.
Taher, C. S. Le Duff, B. E. Willcox, Y. Mehellou, J. Med.
Chem. 2018, 61, 2111-2117.
[10] Y. Mehellou, H. S. Rattan, J. Balzarini, J. Med. Chem. 2018
61, 2211-2226.
,
[11] L. Osgerby, Y. C. Lai, P. J. Thornton, J. Amalfitano, C. S. Le
Duff, I. Jabeen, H. Kadri, A. Miccoli, J. H. R. Tucker, M. M.
6
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202100198
ChemMedChem
FULL PAPER
Entry for the Table of Contents
Aryloxy triester phosphoramidates of the monophosphate derivatives of DMAPP and IPP were designed and synthesized. Despite their
structural similarity, the aryloxy triester phosphoramidates of the IPP monophosphate derivative were more stable in aqueous
environments than those of DMAPP and they induced specific activation of Vγ9/Vδ2 T cells.
Institute and/or researcher Twitter usernames: @Youcef_Mehellou; @LabWillcox
7
This article is protected by copyright. All rights reserved.