Novel organophosphorus scaffolds of urease inhibitors obtained by substitution of Morita-Baylis-Hillman adducts with phosphorus nucleophiles

Article abstract of DOI:10.1016/j.ejmech.2017.03.070

The reactivity of Morita-Baylis-Hillman allyl acetates was employed to introduce phosphorus-containing functionalities to the side chain of the cinnamic acid conjugated system by nucleophilic displacement. The proximity of two acidic groups, the carboxyla

Full text of DOI:10.1016/j.ejmech.2017.03.070

Accepted Manuscript
Novel organophosphorus scaffolds of urease inhibitors obtained by substitution of
Morita-Baylis-Hillman adducts with phosphorus nucleophiles
Vassilis Ntatsopoulos, Stamatia Vassiliou, Katarzyna Macegoniuk, Łukasz Berlicki,
Artur Mucha
PII:
S0223-5234(17)30240-4
10.1016/j.ejmech.2017.03.070
EJMECH 9330
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 16 January 2017
Revised Date: 2 March 2017
Accepted Date: 28 March 2017
Please cite this article as: V. Ntatsopoulos, S. Vassiliou, K. Macegoniuk, E. Berlicki, A. Mucha, Novel
organophosphorus scaffolds of urease inhibitors obtained by substitution of Morita-Baylis-Hillman
adducts with phosphorus nucleophiles, European Journal of Medicinal Chemistry (2017), doi: 10.1016/
j.ejmech.2017.03.070.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical Abstract
Novel organophosphorus scaffolds of urease inhibitors obtained by substitution of
Morita-Baylis-Hillman adducts with phosphorus nucleophiles
Vassilis Ntatsopoulos, Stamatia Vassiliou, Katarzyna Macegoniuk, Łukasz Berlicki and Artur
Mucha

ACCEPTED MANUSCRIPT
Novel organophosphorus scaffolds of urease inhibitors obtained by substitution of
Morita-Baylis-Hillman adducts with phosphorus nucleophiles
a
a,
b
b
Vassilis Ntatsopoulos , Stamatia Vassiliou *, Katarzyna Macegoniuk , Łukasz Berlicki ,
and Artur Mucha b,*
a
Laboratory of Organic Chemistry, Department of Chemistry, University of Athens,
Panepistimiopolis, Zografou, 15701 Athens, Greece
b
Department of Bioorganic Chemistry, Faculty of Chemistry, Wrocław University of
Technology, Wybrzeże Wyspiańskiego 27, 50-370 Wrocław, Poland
* Corresponding authors
SV. E-mail: svassiliou@chem.uoa.gr
AM. E-mail: artur.mucha@pwr.edu.pl
Keywords: phosphonic and phosphinic acids, substitution of allyl acetates, enzyme inhibitors,
structure-activity relationship, binding mode.
1

ACCEPTED MANUSCRIPT
Abstract
The reactivity of Morita-Baylis-Hillman allyl acetates was employed to introduce
phosphorus-containing functionalities to the side chain of the cinnamic acid conjugated
system by nucleophilic displacement. The proximity of two acidic groups, the carboxylate and
phosphonate/phosphinate groups, was necessary to form interactions in the active site of
urease by recently described inhibitor frameworks. Several organophosphorus scaffolds were
obtained and screened for inhibition of the bacterial urease, an enzyme that is essential for
survival of urinary and gastrointestinal tract pathogens. α-Substituted phosphonomethyl- and
2-phosphonoethyl-cinnamate appeared to be the most potent and were further optimized. As a
result, one of the most potent organophosphorus inhibitors of urease, α-phosphonomethyl-p-
methylcinnamic acid, was identified, with Ki = 0.6 µM for Sporosarcina pasteurii urease.
High complementarity to the enzyme active site was achieved with this structure, as any
further modifications significantly decreased its affinity. Finally, this work describes the
challenges faced in developing ligands for urease.
2

ACCEPTED MANUSCRIPT
Introduction
Urease (EC 3.5.1.5) is a heteropolymeric amidohydrolase involved in the degradation of urea
to ammonia and carbamate; the latter compound is then spontaneously hydrolyzed to another
ammonia and carbonic acid [1,2]. Urease is characterized by an exceptional hydrolysis rate
enhancement (1014-fold) and holds a special position among milestone biochemistry
developments as the first ever enzyme crystallized and demonstrated as nickel-dependent [3-
5]. Resolution of the crystal structures of urease from different sources allowed investigators
to determine the role of the two Ni2+ ions in the active site and the subsequent mechanism of
action [5-10]. Urease is involved in global nitrogen circulation and occurs in plants, fungi and
bacteria, such as Proteus mirabilis, Klebsiella aerogenes and Helicobacter pylori [11,12].
This distribution determines the importance of urease to humans and the environment in two
fundamental ways. In agriculture, the efficacy of nitrogen fertilizers can be decreased by soil
macrobiotic ureases [13], which requires heavy use of urea and causes water eutrophication
[14]. In the context of public health, bacterial urease activity is associated with colonization of
humans with pathogenic microorganisms and causes persistent infections/diseases, in
particular those of the urinary and gastrointestinal tracts. A Proteus mirabilis-induced
increase in the environment’s pH results in the crystallization of urinary stones [11,12,15],
while Helicobacter pylori is the main etiological agent responsible for peptic ulcers. In the
stomach, the alkaline local environment facilitates H. pylori colonization of the mucosal
lining and ultimately leads to disruption of the gastric mucous membrane [11,12,16-18].
Accordingly, massive efforts have been devoted to the development of inhibitors of bacterial
urease, especially those responsible for human disorders [19-21]. Although these inhibitors
are not precisely antibacterial, they can eradicate the microorganism-induced conditions
essential for the development of pathogens. Organophosphorus compounds are one of the
most important classes of urease inhibitors. Originally, they comprised phosphoramides of
varied structures that mimicked the substrate in the transition state of the enzymatic process
[4,22-24]. Unfortunately, the P-N bond is not fully stable in water conditions, which limits
further applications of phosphoramides. Recently, compounds containing a P-C linkage
(phosphonates and phosphinates) emerged as an alternative to overcome the hydrolytic
lability. Several active di- and tri-functional α-aminomethylphosphonic and α-
aminomethylphosphinic acids were found after extensive screening and structural
optimization (Figure 1) [25-28].
3

ACCEPTED MANUSCRIPT
N
P
Ki = 0.62 µM, B. pasteurii
O OH
Ki = 0.36 µM, P. vulgaris
Ki = 0.43 µM, B. pasteurii
N
H
P
OH
O OH
Ki = 0.11 µM, S. pasteurii
Ki = 0.20 µM, P. mirabilis
N
H
P
NH2
O OH
Figure 1. Exemplified P-methyl, P-hydroxymethyl and P-aminomethyl-substituted α-
aminomethylphosphinic acids as inhibitors of bacterial ureases of submicromolar inhibition
constants [26-28].
In this work, we suggest a novel framework for organophosphorus inhibitors of ureases. Our
concept is partially based on the results presented in a study describing the crystal structure of
Sporosarcina pasteurii urease complexed with citrate [29]. The polar groups of citrate, in
particular two carboxylates, form a complex network of hydrogen bonds in the enzyme active
site (Figure 2). Central 2-carboxylate is involved in bidendate nickel ion complexation and in
an interaction with the imidazole nitrogen atom of Hiss222. Terminal 1-carboxylate forms
two close hydrogen bonds with the oxygen atom of Gly280 and α-nitrogen atom of Arg339.
The two remaining functionalities (3-carboxylate and hydroxyl) also interact with oxygen
atoms of the neighboring amino acid residues, although these contacts seem to be less crucial
for binding.
Figure 2. The mode of binding of citrate to Sporosarcina pasteurii urease (PDB id 4AC7)
[29]. Inhibitor and enzyme amino acid residues are shown as sticks, while nickel ions are
4

ACCEPTED MANUSCRIPT
shown as dark blue spheres. Hydrogen bonds are shown as thin green lines. Enzyme surface is
colored according to the interpolated charge.
Inspired by these data, we propose construction of a phosphonate(phosphinate)/carboxylate
system which can imitate the 1,2-dicarboxylate portion of citrate in its affinity to urease.
Phosphonic acid should have improved complexation abilities with catalytic metal ions.
Contrary to citrate, both acidic groups are planned to be fixed on a hydrophobic, partially
rigid core. To examine the synthetic feasibility of such an arrangement, we turn our attention
to appropriately modified cinnamic acids. These compounds can be readily obtained from
activated allylic systems of Morita-Baylis-Hillman acetates under SN2’ nucleophilic
displacement [30,31]. For phosphorus nucleophiles, such transformations were reported in the
literature mostly with the use of tervalent phosphorus compounds [32-35]. Here, we intended
to examine the reactivity of a broad range of P-nucleophiles with allyl acetates derived from
aromatic aldehydes (Scheme 1). As an alternative to allylic substitution, certain derivatives
could also be prepared by derivatization via intermediate H-phosphinic acids. We aimed to
obtain various scaffolds containing phosphonic, H-, methyl-, hydroxymethyl- and
aminomethyl-phosphinic functionalities and evaluate their affinity to S. pasteurii urease.
Additionally, modification and optimization of the hydrophobic core structure was also
planned.
O
O
COOMe
Morita-Baylis-Hillman
hydrophobic fragment
X
acetates
+
O
phosphorus or phosphorus-containing
nucleophiles,
e.g. bis(trimethylsilyl) hypophosphite
metal-binding
and hydrogen
bond-forming
functionalities
OH
OH
X
n
P
Y
O
X = H, Me, F, Cl, NHAc, NO2, OMe
n = 1, 2
Y = H, OH, Me, CH2OH, CH2NH2
COOMe
OH
X
P
O
H
Scheme 1. A general concept of the construction of novel inhibitors of urease and the
retrosynthetic analysis.
5

ACCEPTED MANUSCRIPT
Results and Discussion
The Morita-Baylis-Hillman (MBH) acetates were synthesized in a typical two-step procedure
(Scheme 2) [31,33]. First, benzaldehyde and its selected p-substituted analogues (1a-g) were
reacted with methyl acrylate. As is standard, the reaction proceeded in the presence of
DABCO as the catalyst for an extended time to form the corresponding MBH adducts, allyl
alcohols, which were used in the subsequent step without purification. Acetylation of the
hydroxyl group was readily performed with acetyl chloride and pyridine in CH2Cl2.
Compounds 2a-g were purified by column chromatography, and the yield was good to
excellent (45-90%).
2a,
X = H
O
2b
, X = Me
2c,
X = F
2d, X = Cl
O
O
a, b
COOMe
+
COOMe
2e X = NHAc
,
X
2f, X = OMe
X
1a-g
X = NO2
2g,
Scheme 2. The structure and preparation of the allylic substrates, the Morita-Baylis-Hillman
acetates 2a-g. Reagents and conditions: (a) DABCO, rt, 3 d (20 d for 1e); (b) AcCl, pyridine,
CH2Cl2, 0 °C, 1 h, rt, 16 h; 45-90%.
To establish the synthetic procedures of the allylic substitution with phosphorus and
phosphorus-containing nucleophiles, benzaldehyde-derived compound 2a was employed as
the model substrate. The primary P-C bond formation was performed with bis(trimethylsilyl)
hypophosphite (Scheme 3). The silylated starting material was obtained in situ by heating
ammonium hypophosphite with hexamethyldisilazane (HMDS). After cooling to 0 °C, the
esterification was followed by the addition of the MBH acetate 2a and dissolved in CH2Cl2.
Methanolysis of the produced silyl ester intermediate provided the final compound 3a in 89%
yield after column chromatography. The Z stereoisomer of phosphinic acid 3a was
exclusively obtained as determined by 1H NMR. A single signal at 7.91 ppm appeared, and no
resonance at a higher field (approximately 6.90-7.00 ppm) was observed. This assumption is
in accordance with Georgiadis’ data on the stereochemical preferences of MBH acetate
substitution with phosphinic acid silyl esters [36]. Alkaline hydrolysis followed by
6

ACCEPTED MANUSCRIPT
acidification provided the corresponding product 4a in the form of free carboxylic acid in
96% yield.
Scheme 3. Substitution of the Morita-Baylis-Hillman acetate 2a and with bis(trimethylsilyl)
hypophosphite to obtain H-phosphinic acid 3a, and hydrolysis of the latter to 4. Reagents and
conditions: (a) HMDS, 120 °C, 2 h; (b) 2a in CH2Cl2, 0 °C → rt, 16 h; (c) MeOH, rt, 30 min;
89%, three steps; (d) 0.75 M NaOH, MeOH, rt, then H+; 96%.
The H-phosphinic acids contain functionality that can be divergently modified. We envisaged
three types of transformations: P-H oxidation, P-alkylation and P-H addition to a carbonyl
compound (Scheme 4). The P-H oxidation of phosphinic acid 3a with I2/DMSO provided
phosphonic acid 5a in 81% yield after chromatographic purification. Apparently, the
phosphorus group transformation does not alter the diastereometric composition, and the Z
isomer was solely obtained. This result was again verified by the presence of an exclusive
1
peak at 7.75 ppm in H NMR. Acidic hydrolysis of 5a in refluxing 6 M HCl resulted in free
carboxylic acid 6a in 74% yield after purification on reversed phase column chromatography.
Typically, such α-(phosphonomethyl)cinnamic acids are obtained by substitution of
compounds 2 with a trialkyl phosphite [33] followed by exhaustive hydrolysis. Here, we
envisaged an alternative route that provides easy access to more diversified products
including alkyl cinnamates 3 and 5 and cinnamic acids 4 and 6 containing a free phosphorus
moiety.
7

ACCEPTED MANUSCRIPT
Scheme 4. Divergent modifications of H-phosphinic acid 3a leading to oxidized and P-
alkylated products. Reagents and conditions: (a) DMSO, I2, THF, reflux, 24 h; 81%; (b) 6 M
HCl, reflux, 24 h, reversed phase column chromatography; 74% for 6a, 83% for 8 and 69%
for 10; (c) CH3I, CH2Cl2, BSA, 0 °C → rt, 24 h; 68%; (d) HCHO, CH2Cl2, BSA, 0 °C → rt,
24 h, 35%.
The subsequent P-C bonds were formed by reactions involving in situ generation of tervalent
bis(trimethylsilylated) phosphorus species, in this case using BSA (N,O-
bis(trimethylsilyl)acetamide). The activation was followed by either an Arbuzov-type reaction
with iodomethane to obtain 7 or with the addition of formaldehyde to obtain 9. The
corresponding phosphinic acids were obtained after chromatographic purification in 68% and
35% yield, respectively. Both compounds were converted to the form of free carboxylates by
acidic hydrolysis to give compounds 8 and 10 in 83% and 69% yield, respectively (Scheme
4).
The Cbz-protected H-phosphinic analog of glycine 11 [25] was selected as the appropriate
nucleophile for the elaboration of aminomethylphosphinic acid 12 (Scheme 5). The
conversion was achieved via BSA-mediated allylic substitution of MBH acetate 2a with
bis(trimethylsilyl) phosphinate intermediate (41% yield after purification on column
chromatography). The Cbz-protecting group was successfully removed, in the presence of the
double bond, with 33% HBr/AcOH giving phosphinic acid 13 in 94% yield. Simultaneous N-
and C-deprotection were achieved in refluxing HCl to obtain 14 in 67% yield.
8

ACCEPTED MANUSCRIPT
COOMe
O OH
H
a
2a
+
N
P
Cbz
c
Cbz
H
P
NH
O OH
11
12
b
COOH
COOMe
P
NH2
P
NH2
O OH
14
O OH
13
Scheme 5. Substitution of MBH acetate 2a with N-benzyloxycarbonylaminomethyl-H-
phosphinate 11. Reagents and conditions: (a) CH2Cl2, BSA, 0 °C → rt, 24 h; 41%; (b) 33%
HBr/AcOH, 30 min, rt, then reversed phase column chromatography; 94%; (c) 6 M HCl,
reflux, 24 h, then reversed phase column chromatography; 67%.
Finally, allylic substitution with a dimethyl methylphosphonate anion was required to extend
the inhibitor backbone (Scheme 6). The starting material 15 was deprotonated with LiHMDS
(lithium bis(trimethylsilyl)amide) at -15 °C and reacted with acetate 2a. Phosphonate 16a of
1
predominantly E stereochemistry (7.71 ppm in H NMR) was obtained in 41% yield.
Selective demethylation of the phosphonate group in the presence of the carboxylic methyl
ester was achieved with TMSBr (bromotrimethylsilane) in CH2Cl2 giving 17a in 79% yield.
The fully deprotected compound 18a was obtained by acidic hydrolysis in a manner similar to
the cases mentioned above.
COOMe
O OMe
a
P
+ 2a
OMe
P
O
OMe
c
16a
15
MeO
b
COOH
COOMe
P
P
O
O
OH
OH
18a
17a
HO
HO
9

ACCEPTED MANUSCRIPT
Scheme 6. Substitution of MBH acetate 2a with dimethyl methylphosphonate 11. Reagents
and conditions: (a) THF, -15 °C, LiHMDS, 1 h; 41%; (b) TMSBr, CH2Cl2, 0 °C → rt, 24 h;
79%; (c) 6 M HCl, reflux, 24 h, reversed phase column chromatography; 67%.
All the synthesized structural variants, both esters and acids, together with the starting allyl
acetates, were tested for inhibition of urease derived from Sporosarcina pasteurii. The MBH
acetates 2 either did not inhibit or weakly inhibited the enzyme (Table S1), showing
irreversible mode of binding to the target protein. The most likely mechanism of action
involved formation of a covalent complex via nucleophilic displacement of the allylic system
with thiolate of Cys322. As illustrated by molecular modeling such a modification does not
produce any further specific ligand-protein contacts (Figure S1).
Several phosphorus compounds were found to be poor or completely inactive inhibitors, while
the others appeared to range from moderate to good inhibitors (Table 1). Contrary to MBH
acetates and in agreement with the designing concept they all bound to the enzyme in a
competitive and reversible manner (see below for details). C-P(O)(OH)-C phosphinic
derivatives, namely, P-methyl 7 and 8, P-hydroxymethyl 9 and 10 and P-aminomethyl acids
13 and 14 belonged to the less potent structures. The majority of them indicated no effect on
urease activity. P-H phosphinic compounds 3a and 4 acted somewhat more efficiently. Their
inhibition constants Ki were measured at approximately 100 µM and did not depend on the
status of carboxylate function, methyl ester 3a or free acid 4. Phosphonic compounds 6a, 17a
and 18a (with the exception of 5a) exhibited the most significant inhibitory influence on
urease. Evidently, the presence of three oxygen atoms in the phosphonate group was favored
for effective binding. α-(Phosphonomethyl)cinnamic acid 6a displayed a low micromolar
inhibition constant, while the corresponding carboxylate ester 5a unexpectedly appeared
inactive. Even more surprising was that their higher homologs behaved in a quite opposite
manner. The ester 17a was found to be the most active compound among all the tested
scaffolds with Ki ~ 3 µM, approximately 20-fold more potent than the free acid 18a.
Considering the data set, phosphonic compounds 6, 17 and 18 were selected for further
structural optimization—phenyl ring modifications.
10

ACCEPTED MANUSCRIPT
Table 1. Inhibition of S. pasteurii urease by methyl cinnamates and cinnamic acids α-
substituted with phosphorus-containing functionalities. Scaffold validation and optimization
of the structure of the phosphonate/phosphinate moiety. The most significant inhibition is
highlighted in bold.
Entry
Ester
Ki [µM]
Entry
Acid
Ki [µM]
COOH
COOMe
97.5 ± 14
117 ± 8.9
3a
4
H
H
P
P
O OH
COOH
O OH
COOMe
NI
NI
13.1 ± 2.3
5a
7
6a
8
OH
OH
P
O OH
P
O OH
COOMe
COOH
NI
NI
P
P
O OH
O OH
COOH
COOMe
293 ± 20
9
10
P
OH
P
OH
O OH
O OH
COOMe
COOH
NI
1 319 ± 126
66.7 ± 6.6
13
14
P
NH2
P
NH2
O OH
O OH
COOMe
COOH
17a
3.05 ± 0.16 18a
P
HO
P
HO
O
O
OH
OH
To facilitate the structure development of the inhibitor, the binding mode of the two most
potent compounds 6a (data not shown) and 17a (Figure 3), and a moderate one (4, Figure S2),
was studied in some detail by molecular modeling. The optimized inhibitor-enzyme
complexes had general similarities in their interactions at the enzyme active site. In all cases,
the phosphonate/H-phosphinate group coordinates with two nickel ions, while the aromatic
portion of the inhibitors is lipophilically stuck to the hydrophobic surface of the entrance
cavity formed by Ala170 and Ala366. The carboxylate oxygen atom of methyl ester 17a
11

ACCEPTED MANUSCRIPT
forms a hydrogen bond with Arg339. However, the distance between the ester and phosphonic
functions in 17a seems to be too long to fully employ the potential of both oxygen-rich
moieties as the ligand driving forces. For 6a, the inclusion is somewhat deeper and
conveniently involves charge-assisted hydrogen bonds with Arg339 (for details, see Figure 4
showing the binding mode of p-methyl derivative 6b with improved inhibitory activity). In
the case of H-phosphinic acid 4 the binding is not optimal. The lack of the third oxygen atom
on phosphorus prevents formation of an interaction with His222. The less polar and spatial H-
phosphinate group is moved closer to the nickel ions. This shift partially disables the potential
of hydrogen bonding between the ligand carboxylate and the guanidine of Arg339. In
conclusion, as the following optimization step, we suggested substitution of the phenyl ring
with small groups to modify the properties and size of the lipophilic fragment of phosphonic
acid-containing scaffolds.
Figure 3. Modeled binding mode of compound 17a to S. pasteurii urease. Inhibitor and
enzyme amino acid residues are shown as sticks, while nickel ions as dark blue spheres.
Hydrogen bonds and interactions with metal ions are shown as thin green lines. Enzyme
surface is colored according to interpolated charge.
Substituted α-(phosphonomethyl)cinnamic acids 6b-f were obtained via corresponding esters
5b-f starting from the MBH acetates 2b-f (Scheme 7), following the conditions of the allylic
substitution procedure outlined in Scheme 3. Subsequently, oxidation of intermediate H-
12

ACCEPTED MANUSCRIPT
phosphinic acids 3b-f was conducted analogously to the reaction shown in Scheme 4.
Similarly to the model derivatives, the substitution of 2b-f, oxidation of 3b-f and hydrolysis
of 5d-b and 5f produced carboxylates 6d-b and 6f in good to excellent yield (6e was not
obtained).
O
COOMe
COOMe
O
a
b
COOMe
H
P
OH
P
X
X
O OH
O OH
5b-f
3b-f
X
2b-f
COOH
c
b, X = Me; c, X = F; d, X = Cl;
OH
P
X
e
f
, X = NHAc; , X = OMe
O OH
6b-d,f
Scheme 7. Synthesis of methyl α-H-(phosphinomethyl)cinnamates, methyl α-
(phosphonomethyl)cinnamates and α-(phosphonomethyl)cinnamic acids comprising a p-
substituted phenyl ring. Reagents and conditions: (a) bis(trimethylsilyl) hypophosphite,
CH2Cl2, 0 °C → rt, 16 h; then MeOH, rt, 30 min; 49-98%; (b) DMSO, I2, THF, reflux, 24 h;
49-94%; (c) 6 M HCl, reflux, 24 h for 5b-d; 0.75 M NaOH, MeOH, rt, 3 d for 5f; then
reversed phase column chromatography; 50-52% for 6b-d, 89% for 6f. 6e was not obtained
(hydrolysis of the amide bond).
Different small-size groups, including an alkyl—methyl group, halogens—fluoro and chloro
groups, and heteroatom-derived—acetamido and methoxy groups, were used to modify the
properties of the phenyl portion of the molecules. The anti-urease activity was evaluated for
H-phosphinic intermediates and phosphonic products, and almost all substitutions gave rise to
activity improvement (Table 2). For H-phosphinic carboxylate esters 3b-e the potency
reached a Ki range of 13.4-23.4 µM, which is 5-7-fold lower than that measured for reference
compound 3a. Only methoxy derivative 3f did not follow the same tendency. The
improvement was also visible for phosphonic carboxylate esters 5b-f compared to the
simplest analogue 5a which did not inhibit urease. The most significant result was obtained
for phosphonic carboxylate acids. The addition of p-methyl to the aromatic ring produced 6b,
the most potent inhibitor identified in this study and one of the most potent organophosphorus
ligands of the enzyme reported so far (Ki ~ 0.6 µM). Substitution of phenyl with halogens
13

ACCEPTED MANUSCRIPT
retained the activity of 6c and 6d at a low micromolar level. Interestingly, methoxy-
substituted compound 6f was only slightly less potent than the analogous esters of H-
phosphinic 3f and phosphonic acid 5f.
Table 2. Inhibition of S. pasteurii urease by methyl α-H-(phosphinomethyl)cinnamates 3,
methyl α-(phosphonomethyl)cinnamates 5 and α-(phosphonomethyl)cinnamic acids 6.
Exploration of the structure of phenyl substituents. The most significant inhibition is
highlighted in bold.
Entry
Ki [µM]
Entry
Ki [µM]
Entry
Ki [µM]
97.5 ± 14
NI
13.1 ± 2.3
23.4 ± 1.4
14.5 ± 1.2
16.7 ± 1.3
20.7 ± 1.3
62.9 ± 3.2
81.2 ± 4.9
0.594 ± 0.028
9.79 ± 0.65
16.2 ± 0.98
13.4 ± 0.85
149 ± 8.2
386 ± 19
229 ± 13
-
34.4 ± 2.1
The modeled structure of the optimized 6b-urease complex confirmed a favorable binding
mode of the inhibitor (Figure 4). In particular, the phosphonic acid group perfectly chelates
the catalytic nickel cations. One of the phosphoryl oxygen atoms forms a bridge between both
metallic centers and is additionally involved in a hydrogen bond with the imidazole nitrogen
atom of neighboring His222. Each of the two remaining P-oxygen atoms coordinate
monodendately counterpart Ni2+ ions. In contrast to 17a (Figure 3) and 4 (Figure S2), the
carboxylate group of 6b is located opposite to guanidine of Arg339 and forms a classical two-
hydrogen bond-assisted salt bridge. Thus, the polar heads of 6b are in a suitable arrangement
14

ACCEPTED MANUSCRIPT
to be involved in strong interactions in the enzyme active site. The p-methylphenyl group is
conveniently accommodated in the hydrophobic cleft.
Figure 4. Modeled mode of binding of compounds 6b to Sporosarcina pasteurii urease.
Inhibitor and enzyme amino acid residues are shown as sticks, while nickel ions are shown as
dark blue spheres. Hydrogen bonds and interactions with metal ions are shown as thin green
lines. Enzyme surface is colored according to interpolated charge.
The α,β-unsaturated system of 6b, conjugated with free carboxylate and an aromatic ring, of a
certain size and rigidity appeared to be the most complementary compound to the sterical and
electronic demands of the enzyme active site. High specificity towards the organophosphorus
ligands is not an unusual observation for urease [26-28]. To further prove the significance of
scaffold rigidity, the potency of 6b was compared with a saturated counterpart. The latter (19)
was obtained by catalytic hydrogenation over Pd/C. Indeed, the drop in activity reached 150-
fold (Scheme 8).
COOH
COOH
a
OH
P
OH
P
6b
O OH
19
O OH
Ki = 0.594 ± 0.028 µM
Ki = 90.2 ± 5.3 µM
15

ACCEPTED MANUSCRIPT
Scheme 8. Synthesis and activity of saturated phosphonic analog of the most potent inhibitor
identified 6b. Reagents and conditions: (a) H2, 10% Pd/C, MeOH, 1 atm, rt, 2 d; 60%.
Phenyl substitution of extended phosphonic structures 17a and 18a was not as profitable as in
the cases of methylene-shorter compounds. Selected phosphonates 17b-d (carboxylate esters)
and 18b,d (carboxylic acids) were synthesized by the LiHMDS deprotonation procedure
followed by either TMSBr-mediated phosphonate demethylation or acidic hydrolysis (Scheme
9), similar to that envisaged in Scheme 6. Any structural extension or modification of non-
substituted ester 17a dramatically altered the activity of the products, at least by an order of
magnitude (Table 3). For acids (18b and 18d), the Ki values were quite similar to esters and
remained at a moderate, micromolar level.
O
2c,
X = F
O
2f
, X = OMe
COOMe
2g
, X = NO2
X
a
COOMe
COOMe
COOH
b
c
X
X
X
P
P
P
O
O
O
OH
OMe
18b,d
OH
17b-d
16b-d
HO
MeO
HO
16-18b, X = F; 16-18c, X = OMe; 16-18d, X = NO2
Scheme 9. Synthesis of methyl α-(2-phosphonoethyl)cinnamates and α-(2-
phosphonoethyl)cinnamic acids comprising a p-substituted phenyl ring. Reagents and
conditions: (a) MeP(O)(OMe)2, THF, -15 °C, LiHMDS, 1 h; 11-44%; (b) TMSBr, CH2Cl2, 0
°C → rt, 24 h; 43-60% (c) 6 M HCl, reflux, 24 h, then reversed phase column
chromatography; 37% (18b), 68% (18d).
The majority of studied organophosphorus inhibitors appeared irreversible, competitive and
fast-binding. p-Nitro derivatives 17d and 18d showed a slow-binding mechanism of action,
while a few more compounds exhibited a mixed mechanism. To achieve the stationary state
for all the ligands presented in Tables 1-3, they were preincubated with urease before data
16

ACCEPTED MANUSCRIPT
collection. To distinguish between the mechanisms, the inhibition was also measured without
incubation. The initial and stationary state inhibition constants are compared in Supporting
Information (Table S2).
Table 3. Inhibition of S. pasteurii urease by methyl α-(2-phosphonoethyl)cinnamates 17 and
α-(2-phosphonoethyl)cinnamic acids 18. Exploration of the structure of phenyl substituents.
18c was not obtained.
Entry
Ester 17
Ki [µM]
Entry
Acid 18
Ki [µM]
3.05 ± 0.16
66.7 ± 6.6
17a
18a
COOH
47.4 ± 1.3
292 ± 18
27.9 ± 1.3
17b
17c
18b
18c
F
P
O
OH
HO
COOMe
COOH
-
O
O
P
P
O
O
OH
OH
HO
HO
COOH
54.1 ± 3.6*
59.9 ± 3.2*
17d
18d
O2N
P
O
OH
HO
* stationary state (slow-binding mechanism)
Conclusions
Structure-based design and development of low molecular mass competitive inhibitors of
urease is not a trivial task. The enzyme active site is adapted to accommodate and to perform
hydrolysis of the small specific substrate [5-10,25]; thus, the potential ligands are limited to
compounds that fulfill those sterical and functional restrictions. P-C organophosphorus
compounds have been recently proven to satisfy the complementary to the enzyme demands
and emerged as one of the most important classes of inhibitors of urease. Chemically stable
17

ACCEPTED MANUSCRIPT
and resistant to hydrolysis, phosphonic and phosphinic acids offer a tetrahedral functionality
which combines analogously to that of the gem-diolate substrate in the transition state of an
enzymatic reaction with metal-binding properties. Nevertheless, to achieve a substantial level
of inhibition these generally favorable features needed to be extensively optimized. The SAR
studies performed for modifications of a lead structure typically appointed the optimization
process to a specific compound of a particular size and functional group arrangement. For
example, among collections of methyl-, hydroxymethyl- and aminomethyl(aminophosphinic)
acids, only single structures of submicromolar activity were identified. They outscored even
their closest homologs/analogs in affinity [26-28].
In this work, we found new organophosphorus ligand leads of a bacterial urease
(Sporosarcina pasteurii). The two most potent compounds (6b and 17a) were additionally
tested for inhibition of urease from pathogenic Proteus mirabilis strain. According with high
homology of these bacterial enzymes they exhibited quite a similar potency, with Ki = 0.485 ±
0.029 µM for P. mirabilis urease (0.594 ± 0.028, S. pasteurii) for 6b, and Ki = 1.52 ± 0.11 µM
for P. mirabilis urease (3.05 ± 0.16, S. pasteurii) for 17a. They were selected among several
structural variants obtained by nucleophilic displacement in the allylic system of the Morita-
Baylis-Hillman acetates with phosphorus and phosphorus-containing reactants. The
phosphonomethyl group installed as the α-substituent of the cinnamic acid conjugated system
gave rise to the most significant results. Similar to the aforementioned literature cases,
inhibition was spectacularly structure-specific (Figure 5). Only a single compound, α-
phosphonomethyl-p-methylcinnamic acid (6b), exhibited the submicromolar Ki value. The
lowest possible structural variations diminished this activity by an order of magnitude, at least
20-fold. An alteration of the aromatic portion bulkiness and hydrophobicity appeared to be
relatively less influential on the potency. A transformation of one of the functional groups,
either by esterification of carboxylate or through a phosphonic-to-phosphinic replacement
(one oxygen atom loss), caused the Ki to increase to 20-25 µM (35-40-fold). The backbone
restrictions were found to be the most important. Separation of the acidic groups from a three-
to four-bond distance was manifested by a two-orders of magnitude decrease in activity (110-
fold). Loss of compound rigidity appeared even more fatal and induced a 150-fold decrease in
Ki. The enzyme preferences were quite uniquely satisfied by inhibitor 6b, which was selected
among a variety of fundamental structures and further optimized. Convenient tools to
accomplish the syntheses, structure diversification and optimization were provided by the
allylic system of the Morita-Baylis-Hillman adducts.
18

ACCEPTED MANUSCRIPT
Ki = 90.2 µM
Ki = 20.7 µM
double bond
esterification
(with Me)
reduction
COOH
lipophilicity
decrease
(Me/H)
Ki = 13.0 µM
OH
P
OH
O OH
replacement
(with H)
linker
extension
(by CH2)
Ki = 23.4 µM
Ki = 66.7 µM
Ki = 0.594 µM
Figure 5. The specificity of inhibition of Sporosarcina pasteurii urease by compound 6b and
its close analogs/homologs.
Experimental
1. Chemistry – general
All used reagents were purchased at the highest quality from commercial suppliers, and were
used without further purification. Reactions were monitored by thin-layer chromatography
(TLC) carried out on 0.25 mm silica gel plates (silica gel 60F254) and components were
visualized by UV light absorbance. Purification of compounds by column chromatography
was carried out on silica gel (70-230 mesh) or reversed phase silica gel (Kieselgel 60, RP-18,
1
40-63 µm) with the indicated solvents. H, 13C and 31P NMR spectra were recorded on a 200
MHz Mercury spectrometer. 1H and 13C spectra are referenced according to the residual peak
of the solvent based on literature data. 31P NMR chemical shifts are reported in ppm
downfield from 85% H3PO4 (external standard). 13C and 31P NMR spectra are fully proton
decoupled. ESI mass spectral analyses were performed on a mass spectrometer, using direct
sample injection. Negative or positive ion ESI spectra were acquired by adjusting the needle
and cone voltages accordingly. HRMS spectra were registered using a Bruker Maxis Impact
QTOF mass spectrometer. All synthesized compounds gave satisfactory NMR spectra
confirming >95% purity. All tested organophosphorus compounds 3a-f, 4, 5a-f, 6a-d, 6f, 7-
10, 13, 14, 17a-d, 18a, 18b and 18d gave satisfactory HRMS analysis (found value within
0.0014 m/z unit of the calculated value of the parent ion). The spectroscopy and analytical
data were obtained in the Laboratory of Organic Chemistry of the University of Athens.
2. Allylic substitution of MBH acetates with bis(trimethylsilyl) hypophosphite, the general
procedure for the synthesis of H-phosphinic acids 3a-f
In an oven-dried flask, NH4H2PO2 (0.25 g, 3.00 mmol) and HMDS (0.74 mL, 3.60 mmol) are
19

ACCEPTED MANUSCRIPT
placed and the mixture was heated at 110 οC for 2 h under Ar. Then, the reaction mixture was
ο
cooled at 0 C and the appropriate MBH acetate 2 (1.20 mmol) dissolved in CH2Cl2 (5 mL)
was added. The mixture was stirred at room temperature for 16 h, then, MeOH (5 mL) was
added and stirring was continued for 30 min. The volatiles were evaporated under vacuum
and the residue was acidified to pH = 1 with 6 M HCl, taken up with CH2Cl2, washed with 2
M HCl and H2O. The organic phase was dried over Na2SO4, evaporated in vacuo and purified
by column chromatography using CHCl3/MeOH/AcOH, 7:1:0.5 (7:1.5:1 for 3e) as eluent.
Methyl (Z)-2-{[hydrogen(hydroxy)phosphoryl]methyl}-3-phenylpropenoate, 3a
1
Colorless viscous oil, 0.64 g, yield 89%. Rf (CHCl3/MeOH/AcOH, 7:1:0.5) = 0.27. H NMR
3
1
(200 MHz, CDCl3) δ 7.91 (d, JP-H = 4.5 Hz, 1H), 7.50-7.33 (m, 5H), 7.30 (d, JP-H = 578.5
1
Hz, 1H), 3.81 (s, 3H), 3.10 (d, JP-H = 20.5 Hz, 2H). 13C NMR (50 MHz, CDCl3) δ 167.5,
142.9 (d, 3JC-P = 10.7 Hz), 134.2, 129.1, 129.0, 128.5, 122.2 (d, 2JC-P = 9.7 Hz), 52.3, 29.9 (d,
1JC-P = 94.8 Hz). 31P NMR (81 MHz, CDCl3) δ 33.1. MS (ESI) m/z calculated for C11H12O4P-
[M-H]- 239.0, found 239.0. HRMS m/z calculated for C11H12O4P- [M-H]- 239.0479, found
239.0479.
Methyl (Z)-2-{[hydrogen(hydroxy)phosphoryl]methyl}-3-(4-methylphenyl)propenoate,
3b
Yellowish foamy solid, 0.49 g, yield 98%. Rf (CHCl3/MeOH/AcOH, 7:1:0.5) = 0.35. 1H NMR
(200 MHz, CDCl3) δ 7.66 (s, 1H), 7.35-7.05 (m, 4H), 7.09 (d, 1JP-H = 527.5 Hz, 1H), 3.65 (s,
13
3H), 3.00 (m, 2H), 2.20 (s, 3H). C NMR (50 MHz, CDCl3) δ 168.9, 140.9, 138.7, 131.9,
129.4, 129.1, 123.6, 52.3, 31.8 (d, 1JP-C = 82.4 Hz), 21.1. 31P NMR (81 MHz, CDCl3) δ 26.1.
MS (ESI) m/z calculated for C12H14O4P- [M-H]- 253.1, found 253.1. HRMS m/z calculated
for C12H14O4P- [M-H]- 253.0635, found 253.0642.
Methyl (Z)-3-(4-fluorophenyl)-2-{[hydrogen(hydroxy)phosphoryl]methyl}propenoate, 3c
1
White sticky solid, 0.35 g, yield 78%. Rf (CHCl3/MeOH/AcOH, 7:1:0.5) = 0.32. H NMR
(200 MHz, CD3OD) δ 7.56-6.77 (m, 5H), 7.09 (d, 1JP-H = 525.7 Hz, 1H), 3.74 (s, 3H), 3.06 (d,
13
1
2JP-H = 22.1 Hz, 2H). C NMR (50 MHz, CD3COOD) δ 170.4, 164.8 (d, JF-C = 250.0 Hz),
1
142.4, 134.6, 132.1, 125.8, 117.2, 53.8, 33.1 (d, JP-C = 75.6 Hz). 31P NMR (81 MHz,
CD3OD) δ 23.6. MS (ESI) m/z calculated for C11H11FO4P- [M-H]- 257.0, found 257.2. HRMS
m/z calculated C11H11FO4P- [M-H]- 257.0384, found 257.0384.
20

ACCEPTED MANUSCRIPT
Methyl (Z)-3-(4-chlorophenyl)-2-{[hydrogen(hydroxy)phosphoryl]methyl}propenoate,
3d
1
Foamy solid, 0.47 g, yield 94%. Rf (CHCl3/MeOH/AcOH, 7:1:0.5) = 0.31. H NMR (200
MHz, CDCl3) δ 7.68 (s, 1H), 7.42 (d, J = 7.9 Hz, 2H), 7.25 (d, J = 7.9 Hz, 2H), 7.12 (d, 1JP-H
2
13
= 549.2 Hz, 1H), 3.71 (s, 3H), 2.97 (d, JP-H = 18.2 Hz, 2H). C NMR (50 MHz, CDCl3) δ
3
2
168.3, 140.3 (d, JC-P = 9.7 Hz), 134.8, 133.0, 130.7, 128.7, 124.3 (d, JC-P = 9.0 Hz), 52.5,
1
31
31.3 (d, JC-P = 89.5 Hz). P NMR (81 MHz, CDCl3) δ 27.78. MS (ESI) m/z calculated for
C11H11ClO4P- [M-H]- 273.0, found 273.1. HRMS m/z calculated for C11H11ClO4P- [M-H]-
273.0089, found 273.0087.
Methyl
(Z)-3-(4-acetamidophenyl)-2-
{[hydrogen(hydroxy)phosphoryl]methyl}propenoate, 3e
1
White solid, 0.39 g, yield 49%. Rf (CHCl3/MeOH/AcOH, 7:1.5:1) = 0.35. H NMR (200
MHz, CD3COOD) δ 7.89 (d, 4JP-H = 4.5 Hz, 1H), 7.68 (d, J = 8.4 Hz, 2H), 7.54 (d, J = 8.4 Hz,
1
2
2H), 7.36 (d, JP-H = 587.7 Hz, 1H), 3.85 (s, 3H), 3.28 (d, JP-H = 20.0 Hz, 2H), 2.21 (s, 3H).
13C NMR (50 MHz, CD3COOD) δ 172.8, 169.3, 143.8, 140.2, 131.3, 122.3, 120.9, 113.1,
1
53.2, 30.9 (d, JP-C = 90.1 Hz), 23.9. 31P NMR (81 MHz, CD3OD) δ 44.6. MS (ESI) m/z
calculated for C13H15NO5P- [M-H]- 296.1, found 296.1. HRMS m/z calculated for
C13H15NO5P- [M-H]- 296.0693, found 296.0693.
Methyl (Z)-2-{[hydrogen(hydroxy)phosphoryl]methyl}-3-(4-methoxyphenyl)propenoate,
3f
1
White solid, 0.34 g, yield 68%. Rf (CHCl3/MeOH/AcOH, 7:1:0.5) = 0.29. H NMR (200
MHz, CDCl3) δ 7.60-6.79 (m, 5H), 7.07 (d, 1JP-H = 535.2 Hz, 1H), 3.65 (br s, 6H) 3.18 (d, 2JP-
= 20.5 Hz, 2H). 13C NMR (50 MHz, CDCl3) δ 169.5, 160.1, 140.8, 131.4, 127.3, 122.1,
H
113.9, 55.1, 52.4, 32.0 (d, 1JC-P = 84.5 Hz).31P NMR (81 MHz, CDCl3) δ 24.8. MS (ESI) m/z
calculated for C12H14O5P- [M-H]- 269.1, found 269.1. HRMS m/z calculated for C12H14O5P-
[M-H]- 269.0584, found 269.0586.
3. Alkaline hydrolysis of carboxylate methyl esters, the general procedure
To a stirred solution of a methyl ester (1 mmol) in MeOH (5.4 mL), 4 M NaOH (1.25 mL, 5
mmol) was added and the reaction mixture was stirred at room temperature till total
consumption of the starting material (typically 4-5 days, the reaction progress was monitored
by TLC). Then, the volatiles were evaporated in vacuo and the residue was acidified with 6 M
21

ACCEPTED MANUSCRIPT
HCl to pH = 1. The aqueous phase was extracted with AcOEt, dried over Na2SO4 and
concentrated in vacuo. The residue was purified by normal phase or reversed phase column
chromatography using CHCl3/MeOH/AcOH, 7:2:0.5 (4) or CH3CN/H2O, 1:9 to 9:1 (6f) as
eluent.
(Z)-2-{[Hydrogen(hydroxy)phosphoryl]methyl}-4-phenylpropenoic acid, 4
Phosphinic acid 4 was obtained from 3a (70.8 mg, 0.29 mmol) as colorless viscous oil after
column chromatography using CHCl3/MeOH/AcOH, 7:2:0.5 as eluent in 96% yield (64 mg).
Rf (CHCl3/MeOH/AcOH, 7:2:0.5) = 0.19. 1H NMR (200 MHz, CD3OD) δ 7.94 (d, 3JP-H = 3.8
Hz, 1H), 7.57-7.40 (m, 5H), 7.09 (d, 1JP-H = 643.0 Hz, 1H), 3.08 (d, 1JP-H = 20.2 Hz, 2H). 13C
NMR (50 MHz, DMSO) δ 168.8, 140.6 (d, 3JC-P = 9.8 Hz), 134.9, 129.5, 129.2, 128.7, 124.8
2
1
31
(d, JC-P = 9.4 Hz), 31.2 (d, JC-P = 88.5 Hz). P NMR (81 MHz, DMSO) δ 26.5. MS (ESI)
m/z calculated for C10H10O4P- [M-H]- 225.0, found 225.0. HRMS m/z calculated for
C10H10O4P- [M-H]- 225.0322, found 225.0323.
4. Oxidation of H-phosphinic acids 3a-f to phosphonic acids 5a-f, the general procedure
The appropriate H-phosphinic acid 3 (1.00 mmol) was dissolved in THF (7 mL), then DMSO
(148 µL, 2.00 mmol) and Ι2 (2 mg) were added, and the reaction mixture was refluxed for 24
h. The volatiles were removed in vacuo and the residue was purified by column
chromatography using CHCl3/MeOH/AcOH, 7:1:0.5 (7:1.5:1 for 5e) as eluent.
Methyl (Z)-3-phenyl-2-(phosphonomethyl)propenoate, 5a
Phοsphonic acid 5a was obtained from H-phosphinic acid 3a (0.25 g, 1.04 mmol) as highly
1
viscous oil in 81% yield (0.21 g). Rf (CHCl3/MeOH/AcOH, 7:1:0.5) = 0.36. H NMR (200
1
MHz, DMSO) δ 8.33 (br s, 2H), 7.76-7.38 (m, 5H), 3.71 (s, 3H), 2.96 (d, JP-H = 22.1 Hz,
13
3
2H). C NMR (50 MHz, DMSO) δ 168.2, 139.1 (d, JC-P = 10.4 Hz), 134.9, 129.8, 129.1,
128.7, 126.2 (d, 2JC-P = 10.9 Hz), 52.3, 28.5 (d, 1JC-P = 134.6 Hz). 31P NMR (81 MHz, DMSO)
δ 21.0. MS (ESI) m/z calculated for C11H12O5P- [M-H]- 255.0, found 255.1. HRMS m/z
calculated for C11H12O5P- [M-H]- 255.0428, found 255.0414.
Methyl (Z)-3-(4-methylphenyl)-2-(phosphonomethyl)propenoate, 5b
1
White solid, 0.36 g, yield 89%. Rf (CHCl3/MeOH/AcOH, 7:1:1) = 0.35. H NMR (200 MHz,
CD3COOD) δ 7.74 (s, 1H), 7.51 (d, J = 7.2 Hz, 2H), 7.19 (d, J = 7.2 Hz, 2H), 3.79 (s, 3H),
3.16 (d, 2JP-H = 22.2 Hz, 2H), 2.33 (s, 3H). 13C NMR (50 MHz, CD3COOD) δ 170.8, 142.3 (d,
22

ACCEPTED MANUSCRIPT
2
1
3JP-C = 12.1 Hz), 140.0, 132.9, 130.5, 129.9, 125.3 (d, JP-C = 11.1 Hz), 53.0, 29.4 (d, JP-C
=
133.8 Hz), 20.7. 31P NMR (81 MHz, CD3COOD) δ 33.2. MS (ESI) m/z calculated for
C12H14O5P- [M-H]- 269.1, found 269.1. HRMS m/z calculated for C12H14O5P- [M-H]-
269.0584, found 269.0584.
Methyl (Z)-3-(4-fluorophenyl)-2-(phosphonomethyl)propenoate, 5c
1
White solid, 0.14 g, yield 90%. Rf (CHCl3/MeOH/AcOH, 7:1:1) = 0.39 H NMR (200 MHz,
CD3COOD) δ 7.77-7.63 (m, 3H), 7.12 (t, J = 8.7 Hz, 2H), 3.81 (s, 3H), 3.17 (d, 2JP-H = 21.8
1
Hz, 2H). 13C NMR (50 MHz, CD3COOD) δ 170.5, 164.0 (d, JF-C = 248.2 Hz), 140.9, 132.3,
2
2
1
126.3 (d, JP-C = 10.4 Hz), 116.3 (d, JF-C = 21.9 Hz), 114.8, 53.1, 29.4 (d, JP-C = 134.0 Hz).
31P NMR (81 MHz, CD3COOD) δ 33.5. MS (ESI) m/z calculated for C11H11FO5P- [M-H]-
273.0, found 273.1. HRMS m/z calculated for C11H11FO5P- [M-H]- 273.0334, found
273.0331.
Methyl (Z)-3-(4-chlorophenyl)-2-(phosphonomethyl)propenoate, 5d
1
White solid, 0.28 mg, yield 94%. Rf (CHCl3/MeOH/AcOH, 7:1:1) = 0.32. H NMR (200
MHz, CD3COOD) δ 7.71-7.38 (m, 5H), 3.81 (s, 3H), 3.12 (m, 2H). 13C NMR (50 MHz,
CD3COOD) δ 168.7, 140.5, 135.0, 134.2, 131.7, 129.0, 126.9, 52.9, 30.4 (d, 1JP-C = 36.8 Hz).
31P NMR (81 MHz, CD3COOD) δ 31.8. MS (ESI) m/z calculated for C11H11ClO5P- [M-H]-
289.0, found 289.1. HRMS m/z calculated for C11H11ClO5P- [M-H]- 289.0038, found
289.0041.
Methyl (Z)-3-(4-acetamidophenyl)-2-(phosphonomethyl)propenoate, 5e
1
White solid, 0.11 g, yield 49%. Rf (CHCl3/MeOH/AcOH, 7:1.5:1) = 0.23. H NMR (200
2
MHz, CD3COOD) δ 7.73-7.62 (m, 5H), 3.79 (s, 3H), 3.17 (d, JP-H = 21.6 Hz, 2H), 2.19 (s,
3H). 13C NMR (50 MHz, CD3COOD) δ 172.6, 170.8, 141.3, 139.8, 131.8, 131.4, 125.3,
1
120.5, 53.0, 29.4 (d, JP-C = 112.6 Hz), 23.9. 31P NMR (81 MHz, CD3COOD) δ 33.2. MS
(ESI) m/z calculated for C13H15NO6P- [M-H]- 312.1, found 312.1. HRMS m/z calculated for
C13H15NO6P- [M-H]- 312.0642, found 312.0636.
Methyl (Z)-3-(4-methoxyphenyl)-2-(phosphonomethyl)propenoate, 5f
1
White solid, 0.27 g, yield 91%. Rf (CHCl3/MeOH/AcOH, 7:1:1) = 0.27. H NMR (200 MHz,
CD3COOD) δ 7.73 (d, 4JP-H = 5.0 Hz, 1H), 7.60 (d, J = 8.6 Hz, 2H), 6.93 (d, J = 8.6 Hz, 2H),
2
13
3.80 (s, 3H), 3.79 (s, 3H), 3.17 (d, JP-H = 22.3 Hz, 2H). C NMR (50 MHz, CD3COOD) δ
23

ACCEPTED MANUSCRIPT
171.2, 161.4, 142.2, 132.7 (d, 3JP-C = 18.4 Hz), 128.3, 123.7 (d, 2JP-C = 11.4 Hz), 114.7, 55.6,
1
53.1, 29.4 (d, JP-C = 133.8 Hz). 31P NMR (81 MHz, CD3COOD) δ 33.2. MS (ESI) m/z
calculated for C12H16O6P- [M-H]- 287.1, found 287.1. HRMS m/z calculated for C12H14O6P-
[M-H]- 285.0533, found 285.0534.
5. Acidic hydrolysis of carboxylate methyl esters, the general procedure
The corresponding methyl ester (2 mmol) was subjected to reflux in 6 M HCl (10 mL) for 24
h. The reaction mixture was concentrated under reduced pressure and the residue was purified
using reversed phase column chromatography with CH3CN/H2O, 1:9 to 9:1 as eluent.
(Z)-3-Phenyl-2-(phosphonomethyl)propenoic acid, 6a
Phosphonic acid 6a was obtained from 5a (0.21 g, 0.80 mmol) as white solid in 74% yield
(0.14 g). Rf (n-BuOH/AcOH/H2O, 4:1:1) = 0.51. 1H NMR (200 MHz, DMSO) δ 7.75-7.09 (m,
6H), 2.88 (d, 1JP-H = 22.0 Hz, 2H). 13C NMR (50 MHz, DMSO) δ 169.5, 138.5 (d, 3JC-P = 10.2
2
1
31
Hz), 135.2, 129.8, 129.0, 128.8, 127.3 (d, JC-P = 10.6 Hz), 28.6 (d, JC-P = 135.1 Hz). P
NMR (81 MHz, DMSO) δ 20.8. MS (ESI) m/z calculated for C10H10O5P- [M-H]- 241.0, found
241.1. HRMS m/z calculated for C10H10O5P- [M-H]- 241.0271, found 241.0259.
(Z)-3-(4-Methylphenyl)-2-(phosphonomethyl)propenoic acid, 6b
1
White solid, 0.18 g, yield 52%. Rf (n-BuOH/AcOH/H2O, 4:1:1) = 0.47. H NMR (200 MHz,
CD3COOD) δ 7.90 (s, 1H), 7.56 (d, J = 7.2 Hz, 2H), 7.24 (d, J = 7.2 Hz, 2H), 3.30 (d, 2JP-H
=
17.7 Hz, 2H), 2.36 (s, 3H). 13C NMR (50 MHz, CD3COOD) δ 170.7, 140.8, 132.8, 131.1,
130.8, 130.2, 123.6, 28.5 (d, 1JP-C = 132.8 Hz), 21.4. 31P NMR (81 MHz, CD3COOD) δ 37.9.
MS (ESI) m/z calculated for C11H12O5P- [M-H]- 255.0, found 255.0. HRMS m/z calculated
for C11H12O5P- [M-H]- 255.0428, found 255.0428.
(Z)-3-(4-Fluorophenyl)-2-(phosphonomethyl)propenoic acid, 6c
1
White solid 77 mg, yield 50%. Rf (n-BuOH/AcOH/H2O, 4:1:1) = 0.50. H NMR (200 MHz,
4
CD3COOD) δ 7.93 (d, JP-H = 5.1 Hz, 1H), 7.73-7.66 (m, 2H), 7.17 (t, J = 8.8 Hz, 2H), 3.31
(d, 2JP-H = 22.8 Hz, 2H). 13C NMR (50 MHz, CD3COOD) δ 173.3, 164.3 (d, 1JF-C = 249.0 Hz),
2
2
1
133.1, 131.9, 124.1 (d, JP-C = 11.2 Hz), 119.9, 116.6 (d, JF-C = 18.1 Hz), 27.9 (d, JP-C
=
31
139.7 Hz). P NMR (81 MHz, CD3COOD) δ 39.4. MS (ESI) m/z calculated for C10H9FO5P-
[M-H]- 259.0, found 259.2. HRMS m/z calculated for C10H9FO5P- [M-H]- 259.0177, found
259.0178.
24

ACCEPTED MANUSCRIPT
(Z)-3-(4-Chlorophenyl)-2-(phosphonomethyl)propenoic acid, 6d
1
White solid, 76 mg, yield 52%. Rf (n-BuOH/AcOH/H2O, 4:1:1) = 0.54. H NMR (200 MHz,
CD3COOD) δ 7.90 (d, 4JP-H = 5.7 Hz, 1H), 7.63 (d, J = 8.5 Hz, 2H), 7.43 (d, J = 8.5 Hz, 2H),
2
13
3.30 (d, JP-H = 22.8 Hz, 2H). C NMR (50 MHz, CD3COOD) δ 173.0, 136.1, 134.2, 132.1
3
2
1
31
(d, JP-C = 14.9 Hz), 129.9, 129.6, 125.1 (d, JP-C = 11.4 Hz), 28.0 (d, JP-C = 137.3 Hz). P
NMR (81 MHz, CD3COOD) δ 38.7. MS (ESI) m/z calculated for C10H9ClO5P- [M-H]- 275.0,
found 275.0. HRMS m/z calculated for C10H9ClO5P- [M-H]- 274.9882, found 274.9883.
(Z)-3-(4-Methoxyphenyl)-2-(phosphonomethyl)propenoic acid, 6f
Following the general alkaline hydrolysis (chapter 3), phosphonic acid 6f was obtained from
5f as white solid in 89% yield (76 mg). Rf (n-BuOH/AcOH/H2O, 4:1:1) = 0.50. 1H NMR (200
MHz, CD3COOD) δ 7.93 (d, 4JP-H = 5.1 Hz, 1H), 7.65 (d, J = 8.5 Hz, 2H), 6.99 (d, J = 8.5 Hz,
2
13
2H), 3.84 (s, 3H), 3.38 (d, JP-H = 23.0 Hz, 2H). C NMR (50 MHz, CD3COOD) δ 173.7,
162.0, 144.7, 132.9 (d, 3JP-C = 22.0 Hz), 127.9, 121.0 (d, 2JP-C = 11.5 Hz), 115.1, 55.7, 27.7 (d,
1JP-C = 140.0 Hz). 31P NMR (81 MHz, CD3COOD) δ 40.6. MS (ESI) m/z calculated for
C11H12O6P- [M-H]- 271.0, found 271.2. HRMS m/z calculated for C11H12O6P- [M-H]-
271.0377, found 271.0377.
6. P-Methylphosphinic compounds 7 and 8
Methyl (Z)-2-{[methyl(hydroxy)phosphoryl]methyl}-3-phenylpropenoate, 7
Methyl iodide (0.23 mL, 3.66 mmol) was added to an oven-dried flask containing 3a (0.24 g,
o
1.00 mmol) in dry CH2Cl2 (2.5 mL). After cooling to 0 C, BSA (1.80 mL, 7.36 mmol) was
added dropwise under argon, and the reaction mixture was stirred at room temperature for 24
h. Then, 1.5 M HCl (10 mL) was added and the organic phase was taken up with AcOEt (30
mL), washed with H2O, dried over Na2SO4 and concentrated under reduced pressure. The
crude product was purified with column chromatography using CHCl3/MeOH/AcOH, 7:1:0.5
1
giving 7 as viscous oil in 68% yield (0.17 g). Rf (CHCl3/MeOH/AcOH, 7:1:0.5) = 0.62. H
3
NMR (200 MHz, CDCl3) δ 10.18 (br s, 1H), 7.82 (d, JP-H = 3.8 Hz, 1H), 7.52-7.20 (m, 5H),
3.78 (s, 3H), 3.16 (d, 1JP-H = 18.4 Hz, 2H), 1.42 (d, 1JP-H = 14.3 Hz, 3H). 13C NMR (50 MHz,
3
2
CDCl3) δ 168.0, 141.9 (d, JC-P = 9.8 Hz), 134.6, 129.3, 128.9, 128.5, 123.9 (d, JC-P = 9.6
1
1
31
Hz), 52.3, 30.6 (d, JC-P = 92.7 Hz), 15.9 (d, JC-P = 94.7 Hz). P NMR (81 MHz, CDCl3) δ
52.2. MS (ESI) m/z calculated for C12H14O4P- [M-H]- 253.06, found 253.16. HRMS m/z
25

ACCEPTED MANUSCRIPT
calculated for C12H14O4P- [M-H]- 253.0635, found 253.0637.
(Z)-2-{[Methyl(hydroxy)phosphoryl]methyl}-3-phenylpropenoic acid, 8
Following the general acidic hydrolysis procedure (chapter 5), phosphinic acid 8 was obtained
from 7 (99.6 mg, 0.39 mmol) as white solid after chromatographic purification with
CHCl3/MeOH/AcOH, 7:2:0.5 in 83% yield (78 mg). Rf (CHCl3/MeOH/AcOH, 7:2:1) = 0.32.
1H NMR (200 MHz, DMSO) δ 7.70-7.39 (m, 6H), 3.00 (d, 1JP-H = 18.2 Hz, 2H), 1.32 (d, 1JP-H
= 14.2 Hz, 3H). 13C NMR (50 MHz, DMSO) δ 170.4, 141.6 (d, 3JC-P = 9.7 Hz), 135.5, 130.7,
130.4, 129.8, 125.9 (d, 2JC-P = 9.6 Hz), 29.7 (d, 1JC-P = 87.9 Hz), 16.9 (d, 1JC-P = 91.5 Hz). 31P
NMR (81 MHz, DMSO) δ 45.2. MS (ESI) m/z calculated for C11H12O4P- [M-H]- 239.1, found
239.2. HRMS m/z calculated for C11H12O4P- [M-H]- 239.0479, found 239.0480.
7. P-Hydroxymethylphosphinic compounds 9 and 10
Methyl (Z)-2-{[hydroxymethyl(hydroxy)phosphoryl]methyl}-3-phenylpropenoate, 9
Paraformaldehyde (0.11 g, 3.66 mmol) was added to an oven-dried flask containing 3a (0.24
g, 1.00 mmol) in dry CH2Cl2 (2.5 mL). After cooling to 0 oC, BSA (1.80 mL, 7.36 mmol) was
added dropwise under argon and the reaction mixture was stirred at room temperature for 24
h. Then, 1.5 M HCl (10 mL) was added and the organic phase was taken up with AcOEt (30
mL), washed with H2O, dried over Na2SO4 and concentrated under reduced pressure. The
crude product was purified with column chromatography using CHCl3/MeOH/AcOH, 7:1:0.5
1
giving 9 as viscous oil in 35% yield (95 mg). Rf (CHCl3/MeOH/AcOH, 7:1:0.5) = 0.40. H
1
NMR (200 MHz, CDCl3) δ 7.82 (s, 1H), 7.51-7.25 (m, 5H), 3.83 (d, JP-H = 22.5 Hz, 2H),
3.63 (s, 3H), 3.16 (d, 1JP-H = 17.2 Hz, 2H). 13C NMR (50 MHz, CDCl3) δ 168.4, 142.7 (d, 3JC-
2
1
= 8.9 Hz), 134.4, 129.3, 129.1, 128.6, 122.9 (d, JC-P = 8.8 Hz), 60.0 (d, JC-P = 105.5 Hz),
P
1
31
52.5, 28.7 (d, JC-P = 93.7 Hz). P NMR (81 MHz, CDCl3) δ 38.1. MS (ESI) m/z calculated
for C12H14O5P- [M-H]- 269.1, found 269.1. HRMS m/z calculated for C12H14O5P- [M-H]-
269.0584, found 269.0585.
(Z)-2-{[Hydroxymethyl(hydroxy)phosphoryl]methyl}-3-phenylpropenoic acid, 10
Following the general acidic hydrolysis procedure (chapter 5), phosphinic acid 10 was
obtained from 9 (0.14 g, 0.52 mmol) as white solid after reversed phase column
chromatography in 69% yield (92 mg). Rf (n-BuOH/AcOH/H2O, 4:1:1) = 0.42. 1H NMR (200
MHz, DMSO) δ 7.75 (d, 3JP-H = 4.3 Hz, 1H), 7.32-7.28 (m, 5H), 3.41 (s, 2H), 2.72 (d, 1JP-H
=
26

ACCEPTED MANUSCRIPT
13
3
13.9 Hz, 2H). C NMR (50 MHz, DMSO) δ 171.2, 136.6 (d, JC-P = 7.9 Hz), 135.6, 130.0,
129.4, 128.4, 128.1, 60.9 (d, 1JC-P = 112.2 Hz), 29.1 (d, 1JC-P = 74.6 Hz). 31P NMR (81 MHz,
DMSO) δ 34.1. MS (ESI) m/z calculated for C11H12O5P- [M-H]- 255.0, found 255.1. HRMS
m/z calculated for C11H12O5P- [M-H]- 255.0428, found 255.0425.
8. P-Aminomethylphosphinic compounds 13 and 14
Methyl
(Z)-2-{[N-benzyloxycarbonylaminomethyl(hydroxy)phosphoryl]methyl}-3-
phenylpropenoate, 12
MBH acetate 2a (1.52 g, 6.49 mmol) was added to an oven-dried flask containing N-
benzyloxycarbonylaminomethyl-H-phosphinic acid 11 (0.40 g, 1.75 mmol) in dry CH2Cl2 (4.5
mL). After cooling to 0 oC, BSA (3.20 mL, 13.09 mmol) was added dropwise under argon and
the reaction mixture was stirred at room temperature for 24 h. Then, 1.5 M HCl (10 mL) was
added and the organic phase was taken up with AcOEt (30 mL), washed with H2O, dried over
Na2SO4 and concentrated under reduced pressure. The crude product was purified with
column chromatography using CHCl3/MeOH/AcOH, 7:0.5:0.5 giving 12 as sticky solid in
41% yield (0.29 g). Rf (CHCl3/MeOH/AcOH, 7:0.5:0.5) = 0.56. 1H NMR (200 MHz, CDCl3)
δ 7.77 (s, 1H), 7.54-7.02 (m, 10H), 6.13 (br s, 1H), 5.00 (s, 2H), 3.69 (d, 1JP-H = 12.8 Hz, 2H),
3.50 (s, 3H), 3.08 (d, 1JP-H = 17.2 Hz, 2H). 13C NMR (50 MHz, CDCl3) δ 168.5, 156.6, 142.0
(d, 3JC-P = 9.4 Hz), 136.1, 134.4, 129.2, 128.8, 128.7, 128.2 (d, 2JC-P = 10.5 Hz), 128.0, 127.8,
1
1
31
127.7, 66.7, 52.3, 40.5 (d, JC-P = 107.8 Hz), 28.0 (d, JC-P = 95.7 Hz). P NMR (81 MHz,
CDCl3) δ 43.3. MS (ESI) m/z calculated for C20H21NO6P- [M-H]- 402.1, found 402.0. HRMS
m/z calculated for C20H21NO6P- [M-H]- 402.1112, found 402.1107.
Methyl (Z)-2-{[aminomethyl(hydroxy)phosphoryl]methyl}-3-phenylpropenoate, 13
33% HBr/CH3COOH (2.0 mL) was added to 12 (0.27 g, 0.67 mmol) and the mixture was
stirred at room temperature for 30 min. Then, toluene was added and the mixture was
concentrated in vacuo (3 times). Phosphinic acid 13 was obtained as white solid after reversed
phase column chromatography purification using CH3CN/H2O, 1:9 to 9:1 in 94% yield (0.17
g). Rf (n-BuOH/AcOH/H2O, 4:1:1) = 0.50. 1H NMR (200 MHz, CD3COOD) δ 7.89 (d, 3JP-H
4.7 Hz, 1H), 7.68-7.44 (m, 5H), 3.86 (s, 3H), 3.37 (d, 1JP-H = 8.8 Hz, 2H), 3.20 (d, 1JP-H = 18.6
=
13
3
Hz, 2H). C NMR (50 MHz, CD3COOD) δ 169.7, 135.3 (d, JC-P = 3.0 Hz), 131.3, 130.3,
2
1
1
129.3, 129.1, 125.2 (d, JC-P = 10.3 Hz), 52.9 (d, JC-P = 4.0 Hz), 40.7, 31.1 (d, JC-P = 93.4
31
Hz). P NMR (81 MHz, CD3COOD) δ 40.6. MS (ESI) m/z calculated for C12H15NO4P- [M-
27

ACCEPTED MANUSCRIPT
H]- 268.1, found 268.2. HRMS m/z calculated for C12H15NO4P- [M-H]- 268.0744, found
268.0740.
(Z)-2-{[Aminomethyl(hydroxy)phosphoryl]methyl}-3-phenylpropenoic acid, 14
Following the general acidic hydrolysis procedure (chapter 5), phosphinic acid 14 was
obtained from 12 (93.5 mg, 0.35 mmol), as white solid after reversed phase column
1
chromatography using in 67% yield (60 mg). Rf (n-BuOH/AcOH/H2O, 4:1:1) = 0.40. H
NMR (200 MHz, CD3COOD) δ 7.96 (s, 1H), 7.66-7.17 (m, 5H), 3.42 (d, 1JP-H = 8.8 Hz, 2H),
1
13
3
3.21 (d, JP-H = 18.8 Hz, 2H). C NMR (50 MHz, CD3COOD) δ 172.9, 144.9 (d, JC-P = 9.0
Hz), 135.5, 131.6, 130.5, 129.4, 124.7 (d, 2JC-P = 10.4 Hz), 40.0 (d, 1JC-P = 91.6 Hz), 31.0 (d,
1JC-P = 94.6 Hz). 31P NMR (81 MHz, CD3COOD) δ 41.8. MS (ESI) m/z calculated for
C11H13NO4P- [M-H]- 254.1, found 254.0. HRMS m/z calculated for C11H13NO4P- [M-H]-
254.0588, found 254.0579.
9. Allylic substitution of MBH acetates with dimethyl methylphosphonate, the general
procedure for the synthesis of phoshonates 16a-d
Dry THF (2.9 mL) was added to an oven-dried flask containing the appropriate MBH acetate
2 (1.75 mmol) and dimethyl methylphosphonate 15 (0.20 mL, 1.88 mmol). After cooling to -
15 oC, LiHMDS (1 Μ in THF, 1.88 mL, 1.88 mmol) was added dropwise under argon and the
reaction mixture was stirred at -15 oC for 1h. Then, 1 M HCl was added and the organic phase
was taken up in AcOEt, washed with H2O, dried over Na2SO4 and concentrated under reduced
pressure. The crude product was purified with column chromatography using AcOEt followed
by CHCl3/MeOH, 9.5:0.5 as eluent.
Trimethyl (E)-3-phenyl-2-(2-phosphonoethyl)propenoate, 16a
Phosphonate 16a was obtained from 2a (0.40 g, 1.75 mmol) as viscous oil in 41% yield (0.20
1
g). Rf (CHCl3/MeOH, 9.5:0.5) = 0.73. H NMR (200 MHz, CDCl3) δ 7.71 (s, 1H), 7.43-7.25
(m, 5H), 3.81-3.69 (m, 9H), 2.85-2.72 (m, 2H), 2.10-1.93 (m, 2H). 13C NMR (50 MHz,
CDCl3) δ 167.9, 140.3, 134.8, 131.1 (d, 3JC-P = 18.3 Hz), 129.1, 128.7, 128.5, 52.3, 52.2, 52.0,
1
2
31
23.8 (d, JC-P = 138.5 Hz), 20.6 (d, JC-P = 4.0 Hz). P NMR (81 MHz, CDCl3) δ 34.2. MS
(ESI) m/z calculated for C14H20O5P+ [M+H]+ 299.1, found 299.0. HRMS m/z calculated for
C14H20O5P+ [M+H]+ 299.1043, found 299.1042.
Trimethyl (E)-3-(4-flourophenyl)-2-(2-phosphonoethyl)propenoate, 16b
28