Synthesis, characterization and activity of new phosphonate dipeptides as potential inhibitors of VanX

Article abstract of DOI:10.1016/j.bmcl.2011.10.094

VanX, a Zn(II)-dependent D-ala-D-ala dipeptidase, is essential for vancomycin resistance in Enterococcus faecium. The enzymatic activity of VanX was previously found to be inhibited competitively by 2-{[(1-aminoethyl) (hydroxy) phosphoryl]oxy} propanoic acid (1B). Here we report the synthesis and characterization of seven phosphonate dipeptide analogs of D-ala-D-ala with various substituent, the activity evaluation indicated that six of these phosphonate analogs inhibit VanX with IC₅₀ of 0.48-8.21 mM. These data revealed a structure-activity relationship which is that the large substituent group on β-carbon resulted in low binding affinity of the phonphonate analog to VanX. This information will be helpful to guide the design and synthesis of the tightly-binding inhibitors for VanX.

Full text of DOI:10.1016/j.bmcl.2011.10.094

Bioorganic & Medicinal Chemistry Letters 22 (2012) 482–484
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis, characterization and activity of new phosphonate dipeptides
as potential inhibitors of VanX
⇑
Chao Jia, Ke-Wu Yang , Cheng-Cheng Liu, Lei Feng, Jian-Min Xiao, Li-Sheng Zhou, Yi-Lin Zhang
Key Laboratory of Synthetic and Natural Functional Molecule Chemistry of Ministry of Education, College of Chemistry and Material Science, Northwest University, Xi’an 710069,
PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
VanX, a Zn(II)-dependent D-ala-D-ala dipeptidase, is essential for vancomycin resistance in Enterococcus
faecium. The enzymatic activity of VanX was previously found to be inhibited competitively by
2-{[(1-aminoethyl) (hydroxy) phosphoryl]oxy} propanoic acid (1B). Here we report the synthesis and
characterization of seven phosphonate dipeptide analogs of D-ala-D-ala with various substituent, the
activity evaluation indicated that six of these phosphonate analogs inhibit VanX with IC50 of
0.48–8.21 mM. These data revealed a structure–activity relationship which is that the large substituent
group on b-carbon resulted in low binding affinity of the phonphonate analog to VanX. This information
will be helpful to guide the design and synthesis of the tightly-binding inhibitors for VanX.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 14 August 2011
Revised 15 October 2011
Accepted 28 October 2011
Available online 6 November 2011
Keywords:
Vancomycin resistance
VanX
Phosphonate dipeptide
Inhibitor
Vancomycin, called the ‘antibiotic of last resort’, is used for
treatment of the infection of antibiotic resistant bacteria.1 As
glycopeptides antibiotic, vancomycin inhibits the biosynthesis of
Gram-positive bacteria cell wall by specifically binding to cell wall
peptidoglycans that terminate in D-ala-D-ala.2,3 This blockade of
the transpeptidation of uncrosslinked cell wall components by
vancomycin ultimately leads to osmotic lysis of bacteria. However,
during the last two decades, as vancomycin has become a frontline
drug for the treatment of clinical infections, there has been a
pronounced increase in the occurrence of vancomycin-resistant
microbial pathogens such as vancomycin-resistant Staphylococcus
aureus (VRSA) and vancomycin-resistant Enterococci (VRE).4,5 It is
hazardous that the Enterococci are frequent causes of nosocomial
infections and inherently resistant to most antibiotics, the only
effective treatments being synergistic penicillin/aminoglycoside
therapy or treatment with vancomycin.6 There has been an
increasing resistance and mortality from VRE and an infectious
catastrophe for VRE is looming.7
The high-level resistance to vancomycin in VRE is conferred by
at least five genes: vanS, vanR, vanH, vanA, and vanX.8,9 vanS and
vanR act as a two-component regulatory system to mediate antibi-
otic-induced transcription of the genes vanH, vanA, and vanX,
whose products (VanH and VanA) act sequentially to synthesize
the depsipeptide D-ala-D-lactate, instead of the normal
D-ala-D-ala dipeptide. VanX is a Zn(II)-dependent D-ala-D-ala
dipeptidase that hydrolyzes D-ala-D-ala but not D-ala-D-lactate,
allowing the D,D-depsipeptide to accumulate and become
incorporated into the growing peptidoglycan termini. The modified
peptidoglycan termini binds much more weakly to vancomycin
compared with D-ala-D-ala peptidoglycan, this resulting in an
overall 1000-fold reduction in affinity with vancomycin, and
results in unimpeded peptide chain cross-linking, yielding a
mechanically strong cell wall and resistance to lysis and cell death
in the presence of vancomycin.10–14 Clearly, VanX is an attractive
drug target in inhibitor design.
Walsh and co-workers examined the inhibition properties of a
number of phosphonate and phosphinate analogs of D-ala-D-ala
as well as of several dithiol compounds. These analogs were shown
to inhibit VanX with Ki values ranging from 90 nM to 300 mM and
exhibited competitive, mixed, and slow-binding modes of inhibi-
tion.13,15,16 Bussiere and co-workers have successfully determined
the crystal structure of VanX bounding with D-ala-D-ala, a phos-
phonate analog, and a phosphinate analog, which provides evi-
dence that VanX can combine with these inhibitors effectively.6,7
Subsequently, we synthesized and evaluated phosphonate, phos-
phonamidate, and phosphinate analogs of D-ala-D-ala dipeptides
as potential inhibitors for VanX17,18, the phosphonate was shown
to be a competitive inhibitor of VanX with a Ki of 400 mM, at pH
7.0, which is in excellent agreement with Walsh and co-workers.
In order to investigate the structure–activity relationship in
improving the binding affinity of the phosphonate dipeptide ana-
logs to VanX, our strategy was to introduce various substituent
onto the b-carbon in C-terminal of the molecules, because our pre-
vious studies revealed that the different substituent at this position
made the phosphinate analogs have various binding affinities to
⇑
Corresponding author. Tel./fax: +86 29 8830 2429.
E-mail address: kwyang@nwu.edu.cn (K.-W. Yang).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.10.094

C. Jia et al. / Bioorg. Med. Chem. Lett. 22 (2012) 482–484
483
VanX. Followed this strategy, a series of phosphonates analogs of
D-ala-D-ala with diversified aliphatic and aromatic substituent
were synthesized according to Scheme 1. The inhibition properties
of these synthesized compounds were evaluated with the target
enzyme VanX by determination of IC50 values.
16.8–21.2% (see Table 1). In the synthesis,
D
-
a
-hydroxy acid was
protected by benzyl ester. Compared with the original methyl ester
protection, this approach had advantage which is that the reaction
progress not only be easily monitored with silica gel plate, but also
the protecting groups including carbobenzyloxy (Cbz) were re-
moved easily by hydrogenolysis Pd/C-catalyzed. Similar synthetic
schemes are thoroughly precedent in the literatures.20–28
Towards this goal, phosphonate analogs 1(A–G) were synthe-
sized according to Scheme 1. Diphenyl 1-(phenylmethoxycarbon-
ylamino) ethyl phosphonate
5
was prepared from starting
According to the reliable protocol developed by Walsh and co-
workers, VanX was overexpressed and purified as previously de-
scribed.29 Briefly, an overnight culture of BL21(DE3) Escherichia coli
cells containing the MBP-VanX expression plasmid was used to
material acetaldehyde, triphenyl phosphate and benzyl carbamate
as previously described19 and converted by ester exchange in
methanol to the dimethyl ester 6. The monomethyl ester 7 was
obtained by partial base-catalyzed hydrolysis of 6 and was subse-
quently converted by treatment with thionyl chloride in chloro-
inoculate 4 Â 1 liter of LB containing 25
lg/ml kanamycin. The cells
were shaken at 30 °C until OD600 reaching 1.6–1.8, and protein pro-
duction was induced by making the cultures 1 mM IPTG. The cul-
tures were shaken for 75 min at 30 °C, and the E. coli cells were
collected by centrifugation and ruptured by ultrasonic. The cell lysis
was loaded onto an amylose resin column and eluted using 160 ml
of 20 mM Tris, pH 7.6, containing 10 mM maltose and 200 mM
NaCl. The fractions containing MBP-VanX were pooled, concen-
trated and cleaved using thrombin, the resulting sample was loaded
onto a Q-Sepharose column and eluted with a plateau elution pro-
file.30 VanX purity was assessed using SDS–PAGE.
form to chlorophosphonate 8.
converted by oxidation in aqueous solution of sodium nitrite to
-hydroxy acid 10(A–G), which were esterified in presence of
p-toluenesulfonic acid as catalyst to offer -hydroxy acid benzyl
ester 11(A–G). The key intermediate chlorophosphonate 8 was
coupled with -hydroxy acid benzyl ester 11(A–G) to afford com-
D-amino acids 9(A–G) were
D
-a
D-a
D
-a
pounds 12(A–G). The intermediates 13(A–G) were obtained by Pd/
C-catalyzed hydrogenolysis of 12(A–G), and were subsequently
converted by treatment with LiOH in methanol to the lithium
phosphonate, which were treated with cation exchange resin to
afford the phosphonate analogs 1(A–G) with overall yield
The activity of the newly synthesized phosphonates 1(A–G) as
inhibitors against VanX was assayed, the data are presented in
O
CH3
O
CH3
O
OC6H5
OC6H5
OCH3
OCH3
a
b
O
N
H
P
O
N
H
P
CH3CHO
P(OC6H5)3
O
NH2
O
O
2
3
4
6
5
O
CH3
O
CH3
Cl
OH
d
c
O
P
N
H
O
N
H
P
O
OCH3
O
OCH3
g
8
7
O
C
HO
COOH
NH2
COOH
HO
f
e
O
R
9(A-G)
R
R
O
10(A-G)
11(A-G)
O
C
O
CH3
CH3
CH3
O
O
O
O
h
i
P
O
O
N
H
OH
P
OH
H2N
P
H2N
O
O
R
OCH3
O
R
R
OH
OCH3
13(A-G)
12(A-G)
1(A-G)
(R = -H; -CH3; -CH(CH3)2; -CH2CH(CH3)2; -CH(CH3)CH2CH3; -CH2C6H5; -CH2CH2COOH)
Scheme 1. Synthesis of phosphonate dipeptide analogs as VanX inhibitors. Reagents and conditions: (a) CH3COOH, rt, À85 °C, 3 h (68%); (b) CH3OH, CH3ONa, 1,4-dioxane, rt,
12 h (88%); (c) NaOH ap., CH3OH, reflux, 8 h (76%); (d) CHCl3, SOCl2, rt, 8 h; (e) NaNO2, H2SO4, 24 h (82–96%); (f) benzene, benzyl alcohol, p-toluenesulfonic acid, reflux, 3 h
(77–88%); (g) CHCl3, Et3N, a-hydroxy acid ester, rt, 3d (48–61%); (h) 10% Pd/C, H2, CH3OH, rt, 3 h; (i) LiOH, CH3OH, rt, 2d.
Table 1
Inhibitory activity (IC50) of the phosphonate dipeptide analogs against VanX
CH3
O
O
Structure (R)
Compound
Properties
Yield %
IC50 (mM)
H2N
OH
P
O
R
OH
1A
1B
1C
1D
1E
1F
1G
H
CH3
White solid
White solid
White solid
White solid
White solid
White solid
White solid
18.4
21.2
20.6
18.5
20.5
18.2
16.8
0.48 0.02
0.76 0.01
2.82 0.03
1.08 0.05
1.31 0.02
8.21 0.04
No activity
CH(CH3)2
CH2CH(CH3)2
CH(CH3)CH2CH3
CH2C6H5
CH2CH2CO2H

484
C. Jia et al. / Bioorg. Med. Chem. Lett. 22 (2012) 482–484
Table 1. The determination of IC50 was performed using reported
method.31 Briefly, the inhibitor concentration that caused a 50%
reduction in enzyme activity (IC50) was determined graphically
from activity data generated. After 30 min of preincubation of
0.05–5 mM of the inhibitors and 50 nM of VanX at 25 °C, 0.5 mM
of D-ala-D-ala as substrate was added to initiate the enzymatic
reaction. The reaction was processed at 37 °C for 150 s in 50 mM
HEPES (pH 8.0) as buffer. The released free amino acid was mea-
sured with the modified Cd-ninhydrin method reported by Doi.31
0.1 ml of diluted hydrolyzed product was mixed with 0.75 ml of
Cd-ninhydrin stock solution. The color was developed by incubat-
ing the mixture at 85 °C for 5 min. The optical density was mea-
sured at 505 nm and quantified with free amino acid as standard.
The results of activity assay indicated that, except 1G, the phospho-
nates 1(A–F) were shown to be inhibitors of VanX with IC50 values
of 0.48 0.02, 0.76 0.01, 2.82 0.03, 1.08 0.05, 1.31 0.02, and
8.21 0.04 mM, respectively. The low inhibition activity may be
due to the raceme of phosphonates. Upon analysis of these data,
we found that the inhibition activity of the phosphonate analog
was depended on the substituent on b-carbon of it (see Scheme
1), with the best inhibitor having the smallest substituent (R@H),
the phosphonate with aromatic group was shown the lowest
activity.
Previously, we reported the synthesis and inhibition activity of
phosphonate, phosphonamidate and phosphinate analogs of D-ala-
D-ala dipeptides as potential inhibitors for VanX17,18, and the phos-
phonate 2-{[(1-aminoethyl) (hydroxy) phosphoryl]oxy} propanoic
acid (1B) was shown to be a competitive inhibitor. Based on this
information, in order to investigate the structure–activity relation-
ship in further developing tightly-binding inhibitor, the phospho-
nate analogs of D-ala-D-ala dipeptide 1(A–G) were synthesized,
and the target enzyme VanX was overexpressed and purified to
be used for evaluation of the phosphonates by IC50 determination.
The IC50 data revealed that the synthesized compounds 1(A–F)
were shown to inhibit VanX, and the inhibition property depended
on the substituent on b-carbon of the molecule. The best inhibitor
having the smallest R-group (R@H), the inhibitor with aromatic
group having lowest activity. This structure–activity relationship
implicated that the large R-groups may result in binding affinity
decrease of the molecules to VanX. Simultaneously, our studies
indicated that the phosphonate 1G with carboxyl did not obviously
inhibit VanX at its concentration up to 5 mM, suggesting a specific
role of the aliphatic group in the inhibitor.
analogs inhibited L1, CcrA, and ImiS at concentrations up to
1 mM at pH 7.0 using penicillin G as the substrate.
Based on the Walsh’s and Crowder’s studies that 2-{[(1-amino-
ethyl) (hydroxy) phosphoryl]oxy} propanoic acid (1B) was shown
to inhibit VanX,14,17,18 we designed, synthesized and characterized
7 phosphonate dipeptide analogs of D-ala-D-ala. In the synthetic
process, 11 step reactions in total were employed for preparation
of the phosphonates, the overall yields were 16.8–21.2%. The activ-
ity evaluation of the synthesized compounds with VanX, which
was overexpressed and purified in our lab, was shown that six of
these phosphonate analogs inhibit VanX with IC50 of 0.48–
8.21 mM. Analysis of the IC50 values, a structure–activity relation-
ship was revealed, which is that the large substituent on b-carbon
resulted in low binding affinity of the phonphonate analog to
VanX. This information will be helpful to guide the design and syn-
thesis of the tightly-binding inhibitors for VanX.
Acknowledgments
We gratefully thank Professor Michael Crowder at Miami Uni-
versity for the gift of pIADL14 plasmid used in this work. We also
acknowledge the financial support from the National Natural Sci-
ence Foundation of China (20972127), Doctoral Foundation of Chi-
na (200806970005), National Science Foundation of Shaanxi
Province (2009JM2002) and Key Fund for International Coopera-
tion of Shaanxi Province (2010KW-16).
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.10.094.
References and notes
1. Walsh, C. T. Science 1999, 284, 442.
2. Williams, D. H.; Bardsley, B. Chem. Int. Ed. 1999, 38, 1172.
3. Gao, Y. Nat. Prod. Rep. 2002, 19, 100.
4. Weigel, Linda M.; Clewell, Don B., et al Science 2003, 302, 1569.
5. Zaraket, Hassan; Otsuka, Taketo, et al Microbiol. Immunol. 2007, 51, 171.
6. Bussiere, D. E.; Pratt, S. D.; Katz, L.; Severin, J. M. Mol. Cell. 1998, 2, 75.
7. Lessard, I. A. D.; Pratt, S. D.; Bussiere, D. E.; Walsh, C. T. Chem. Biol. 1998, 5,
489.
8. Arthur, M.; Molinas, C.; Depardieu, F.; Courvalin, P. J. Bacteriol. 1993, 175, 117.
9. Arthur, M.; Reynolds, P. E.; Depardieu, F.; Evers, S. J. Infect. 1996, 32, 11.
10. Wright, G. D.; Holman, T. R.; Walsh, C. T. Biochemistry 1993, 32, 5057.
11. Holman, T. R.; Wu, Z.; Wanner, B. L.; Walsh, C. T. Biochemistry 1994, 33, 4625.
12. Reynolds, P. E.; Depardieu, F.; Dutka-Malen, S.; Arthur, M.; Courvalin, P.
Microbiology 1994, 13, 1065.
13. Wu, Z.; Wright, G. D.; Walsh, C. T. Biochemistry 1995, 34, 2455.
14. Walsh, C. T.; Fisher, S. L.; Park, I. S.; Prahalad, M.; Wu, Z. Chem. Biol. 1996, 3, 21.
15. Wu, Z.; Walsh, C. T. Proc. Nat. Acad. Sci. USA 1995, 92, 11603.
16. Wu, Z.; Walsh, C. T. J. Am. Chem. Soc. 1996, 118, 1785.
17. Yang, K. W.; Brandt, J. J.; Chatwood, L. L.; Crowder, M. W. Bioorg. Med. Chem.
Lett. 2000, 10, 1085.
In our initial report on phosphonate, phosphinate, and phos-
phonamidate as inhibitor for VanX, we noted that the inhibition
activity increased gradually with the P–O to P–N to P–C bond in
the inhibitor changes, the increased binding affinity of the
phosphorus-containing dipeptide analogs is most likely due to
the additional H-bond from the inhibitor to the enzyme.17 Here,
focusing on the substituent on b-carbon, the structure–activity
relationship of phosphonate dipeptide analogs was investigated.
The phosphonate 1(A–G) were also tested with metallo-b-lacta-
mase L1, CcrA, and ImiS because these compounds have a similar
structure as a part of the putative tetrahedral intermediate formed
during b-lactam ring hydrolysis (Fig. 1). However, none of the
18. Yang, K. W.; Golich, F. C.; Sigdel, T. K.; Crowder, M. W. Bioorg. Med. Chem. Lett.
2005, 15, 5150.
19. Oleksyszyn, J.; Subotkowska, L.; Mastalerg, P. Synthesis 1979, 985.
20. Kaplan, A. P.; Bartlett, P. A. Biochemistry 1991, 30, 8165.
21. Bartlett, P. A.; Marlowe, C. K. Biochemistry 1983, 22, 4618.
22. Hanson, J. E.; Kaplan, A. P.; Bartlett, P. A. Biochemistry 1989, 28, 6294.
23. Jacobsen, N. E.; Bartlett, P. A. J. Am. Chem. Soc. 1981, 103, 654.
24. Ellsworth, B. A.; Tom, N. J.; Bartlett, P. A. Chem. Biol. 1996, 3, 37.
25. Donkor, I. O.; Zheng, X. Z.; Han, J.; Duane, D. Miller Chirality 2000, 12, 551.
26. Kolasa, T.; Miller, M. J. J. Org. Chem. 1987, 52, 4918.
27. Jan, P.; Milos, B.; Ivona, H.; Miloslav, S.; Pavel, F. Tetrahedron 2009, 65, 6090.
28. Chang, Y. P.; Tseng, M. J. Sci. Direct. 2006, 359, 63.
29. McCafferty, D. G.; Lessard, I. A. D.; Walsh, C. T. Biochemistry 1997, 36, 10498.
30. Breece, R. M.; Costello, A.; Bennett, B.; Crowder, M. W., et al J. Biol. Chem. 2005,
280, 11074.
H
N
H
N
H
R
R
S
O
O
N
P
O-
O
-O
-O
R1
31. Doi, E.; Shibata, D.; Matoba, T. Anal. Biochem. 1981, 118, 173.
O-
-
-
CO2
CO2
A
B
Figure 1. (A) Proposed structure of tetrahedral intermediate of cephalosporin
hydrolysis by metallo-b-lactamases. (B) Structure of phosphonate 1(A–G).