A fluorescent substrate for carbon-phosphorus lyase: Towards the pathway for organophosphonate metabolism in bacteria

Article abstract of DOI:10.1016/j.bmcl.2009.08.035

Many species of bacteria can use naturally occurring organophosphonates as a source of metabolic phosphate by cleaving the carbon-phosphorus bond with a multi-enzyme pathway collectively called carbon-phosphorus lyase (CP-lyase). Very little is known about the fate of organophosphonates entering this pathway. In order to detect metabolic intermediates we have synthesized a fluorescently labelled organophosphonate and show that this is a viable substrate for the CP-lyase pathway in Escherichia coli and that the expected product of CP-bond cleavage is formed. The in vivo competence of one potential metabolic intermediate, 1-ethylphosphonate-α-d-ribofuranose, is also demonstrated.

Full text of DOI:10.1016/j.bmcl.2009.08.035

Bioorganic & Medicinal Chemistry Letters 19 (2009) 5954–5957
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
A fluorescent substrate for carbon–phosphorus lyase: Towards the pathway
for organophosphonate metabolism in bacteria
Shu-Mei He a, Yan Luo a, Bjarne Hove-Jensen b, David L. Zechel a,
*
a Department of Chemistry, Queen’s University, Kingston, Ontario K7L 3N6, Canada
b Department of Biology, University of Copenhagen, DK-2200 Copenhagen N, Denmark
a r t i c l e i n f o
a b s t r a c t
Article history:
Many species of bacteria can use naturally occurring organophosphonates as a source of metabolic phos-
phate by cleaving the carbon–phosphorus bond with a multi-enzyme pathway collectively called carbon–
phosphorus lyase (CP-lyase). Very little is known about the fate of organophosphonates entering this
pathway. In order to detect metabolic intermediates we have synthesized a fluorescently labelled organ-
ophosphonate and show that this is a viable substrate for the CP-lyase pathway in Escherichia coli and that
the expected product of CP-bond cleavage is formed. The in vivo competence of one potential metabolic
Received 23 June 2009
Revised 7 August 2009
Accepted 7 August 2009
Available online 13 August 2009
Keywords:
Carbon–phosphorous lyase
phn operon
intermediate, 1-ethylphosphonate-
a
-
D-ribofuranose, is also demonstrated.
Ó 2009 Elsevier Ltd. All rights reserved.
Phosphonate metabolism
Fluorescent label
Substrate
Escherichia coli
Appropriately called the ‘staff of life’,1 phosphate is an essential
component of biological energy (ATP), genetic information (DNA,
RNA) and cell signalling (protein phosphorylation, cAMP). Phos-
phate is also frequently a life-limiting nutrient for microorganisms,
particularly in aquatic environments where concentrations can fall
to picomolar levels.2 To make up for this shortfall, many species of
bacteria will utilize organophosphonates, which can comprise a
significant fraction of total soluble phosphorus.3 For example, the
most abundant naturally occurring organophosphonate, 2-amin-
oethylphosphonate, is used by phytoplankton as a surrogate for
phosphocholine in the biosynthesis of phospholipids.4 However,
for phosphate to be liberated from organophosphonates for other
cellular roles, bacteria must cleave the highly stable carbon–phos-
phorus bond (BDE = 70 kcal molÀ1). A multi-enzyme pathway, col-
lectively called carbon phosphorus lyase (CP-lyase), is commonly
used by bacteria (primarily gram-negative) to cleave this bond,
producing inorganic phosphate and a hydrocarbon.3,5 Commensu-
rate with its scavenging role, CP-lyase is notably relaxed in its
substrate specificity (which varies according to the bacterial
species6), and is capable of cleaving CP bonds in alkyl-, phenyl-,
vinyl-, and alkynylphosphonic acids, as well as doubly cleaving
alkylphosphinic acids.6,7 Such substrate promiscuity makes CP-
lyase an excellent candidate for bioremediation applications.
Despite considerable knowledge of the remarkably complex
genetic structure of the phn operon encoding CP-lyase,8–10 little is
known about the enzymes encoded by the individual phn genes
(phnCDEFGHIJKLMNOP),11–14 particularly the individual reactions
mediated by each and how these reactions might comprise an over-
all pathway for organophosphonate metabolism. Initial evidence for
a metabolic pathway came from the detection of 1-ethylphospho-
nate-a-D-ribofuranose 1 (Scheme 1) in the medium of an Escherichia
coli culture subsisting on ethylphosphonate as the sole phosphorus
source.15 This compound bore a striking resemblance to the
substrate for PhnN (one of the few CP-lyase enzymes with a
demonstrated function), which phosphorylates 5-phospho-
ribofuranosyl phosphate 2 to give the glycosyl donor 5-phospho-
-ribofuranosyl diphosphate 3 (Scheme 1).11 To identify other
a-D-
a-D
potential metabolic organophosphonate intermediates it would be
useful to directly track the fate of an alkylphosphonate in the cell.
Although it is possible to synthesize 32P labelled alkylphospho-
nates,15 we opted to avoid radiolabelled substrates.
Inspired by the tolerance of CP-lyase for a diverse range of sizes
and structures in the organic moiety of organophosphonates, we
hypothesized that a fluorescently labelled alkylphosphonate would
also be a viable substrate. To this end we synthesized dansyl
labelled 3-aminopropylphosphonic acid 7 in a simple three step
procedure (Scheme 2). The commercially available nitrile 4 was
reduced with sodium borohydride to obtain 3-aminopropyl phos-
phonate 5. Reaction of the amine with dansyl chloride afforded
fluorescently labelled 6, which was deprotected with bromotri-
methylsilane to afford 7. We also synthesized the predicted
product 8 of the CP-lyase mediated cleavage of 7 through a direct
reaction of dansyl chloride with n-propylamine. Details of the
* Corresponding author. Tel.: +1 613 533 3259.
E-mail address: dlzechel@chem.queensu.ca (D.L. Zechel).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.08.035

S.-M. He et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5954–5957
5955
HO
O
O
O P OH
OH OH
1
O
O
PhnN
HO
P
O
HO
P
O
ATP ADP
O
O
OH
OH
O
O
O
O P OH
OH
O
P
O P OH
OH
OH OH
OH OH
OH
3
2
Scheme 1. The putative metabolic intermediate 1 identified in E. coli cultures subsisting on ethylphosphonic acid. The reaction for PhnN, one of 14 enzymes and proteins in
the CP-lyase pathway, is shown below. 1 is a potential precursor to the substrate 2 for PhnN.
O
P
O
P
O
S
O
P
O
S
a
b
OR
O
N
H
O
H2N
N
H
OEt
OEt
N
OEt
OEt
OR
4
5
6
(R = Et)
8
c
N
N
7 (R = H)
Scheme 2. Reagents and conditions: (a) NaBH4, CoCl2, THF/H2O (2:1), 0 °C, 3 h (40%); (b) dansyl chloride, Et3N, CH2Cl2, rt, 32 h (53%); (c) TMSBr, ACN, rt, 24 h then MeOH/H2O
(9:1), rt, 2 h (77%).
synthesis and characterization of 7 and 8 are given in the Supple-
mentary data following this Letter.
methylphosphonic acid, a known substrate for E. coli CP-lyase,7
and media containing no source of phosphorus. The plates were
simultaneously incubated at 37 °C for 3 days (growth of E. coli is
typically slow under these nutrient deprived conditions). Gratify-
ingly, E. coli HO1429 grew as well on minimal media containing
7 as media containing an equivalent concentration of methylphos-
phonic acid (Fig. 1). Plates containing no source of phosphorus did
We then tested the ability of the E. coli phn+ strain HO1429 to
grow on MOPS minimal media agar plates16 where 7 (0.5 mM)
was the sole source of phosphorus (see Supplementary data for
strain preparation and growth conditions). For comparison strain
HO1429 was also grown on MOPS-agar media containing 0.5 mM
Figure 1. Growth of the E. coli phn+ strain HO1429 on MOPS minimal media agar containing (l to r) methyl phosphonate (0.5 mM), 7 (0.5 mM), and no phosphorus source. The
plates were illuminated with ambient light (top row) or 360 nm light (bottom row).

5956
S.-M. He et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5954–5957
CH2Cl2/MeOH indicated the presence of
Rf = 0.44 (Fig. 2A, lane 1) which exhibited the characteristic yel-
low–green fluorescence of dansyl fluorophore (k emis-
a product with an
a
sion = 525 nm) when viewed under long UV light (ꢀ360 nm).
This compound matched the Rf of synthetically derived 8, the
expected product of the CP-lyase reaction with 7 (Fig. 2A, lane 2).
The substrate 7 remained on the baseline (Fig. 2A, lane 3). Incuba-
tion of 7 in minimal media in the absence of HO1429 did not result
in the formation of 8 (Fig. 2B, lane 1). Likewise, disruption of the
phn operon in E. coli resulted in a strain that was unable to cleave
7 to form 8 (unpublished data). Purification of this primary product
by preparative TLC (Fig. 2A, lane 4) and analysis by EI-MS (positive
ion mode) revealed a molecular ion peaks at m/z = 292.12 and m/
z = 171.11, which correspond to the expected values for the molec-
ular ion of 8 and a dimethylamino naphthalene fragment ion,
respectively (Fig. 3). It is clear from these analyses that E. coli CP-
lyase accepts 7 as a substrate and cleaves the CP bond to form inor-
ganic phosphate (which sustains bacterial growth) and the
expected labelled product 8.
Encouraged by these results, we subsequently tested syntheti-
cally derived 1,17 a putative metabolic intermediate in the E. coli
CP-lyase pathway. The E. coli strain HO1429 was once again grown
on MOPS-agar minimal media supplemented with 0.5 mM methyl-
phosphonic acid, 0.5 mM 1, or media lacking any phosphorus
source. Likewise, glassware was washed with dilute nitric acid to
eliminate contaminating phosphate. After 2–3 days incubation at
37 °C, equivalent growth of HO1429 colonies were observed on
plates containing methylphosphonic acid or 1, but not on the med-
ium lacking phosphorus (Fig. 4). This is an intriguing result as 1 is
formally a phosphonate ester, and it was previously shown that
ethyl ethylphosphonic acid could not sustain growth of E. coli, sug-
gesting that this was not a substrate for CP-lyase.7 Although this is
good preliminary evidence that 1 is a competent intermediate in
the CP-lyase pathway, it is possible that 1 is hydrolyzed in vivo
to release ethylphosphonic acid, which is a viable substrate for
E. coli CP-lyase.7 Having demonstrated that fluorophore labelled
alkyl phosphonates are acceptable substrates for CP-lyase, this
Figure 2. (A) TLC analysis of the metabolism of 7 by E. coli in MOPS minimal media.
The plate was developed with 96:4 CH2Cl2/MeOH and illuminated with 366 nm
light. The major product 8 is indicated with an arrow. Lane 1, CH2Cl2 extract of the
used growth medium of HO1429 grown in liquid culture with 7; lane 2, synthetic
standard 8, the predicted CP-lyase product; lane 3, substrate 7, lane 4; major
product of E. coli culture purified by preparative TLC for EI-MS analysis; lane 5,
CH2Cl2 extract of minimal media without 7. (B) Lane 1, CH2Cl2 extract of MOPS
minimal media incubated with 7 in the absence of HO1429. Lane 2, synthetic
standard 8.
not show bacterial growth. Growth of HO1429 on 7 is not likely to
have arisen from contaminating phosphate as no phosphate was
present during synthesis and purification of 7, and glassware was
made phosphate-free by liberal washing with dilute nitric acid
before use.
The growth of HO1429 on 7 in the absence of contaminating
phosphate should lead to the production of the CP-bond cleavage
product 8 (Scheme 2). To perform a product analysis HO1429
was cultured in liquid MOPS minimal media containing 0.5 mM 7
until the bacterial culture reached an optical density at 600 nm
of ꢀ1 (37 °C, 220 rpm in an air shaker, ꢀ2 days). After removing
the cells by centrifugation, the remaining medium was extracted
with CH2Cl2 and the solvent removed in vacuo. TLC analysis of
the concentrated extract on silica gel plates developed with 96:4
Figure 3. EI-MS spectrum (positive ion mode) of the major product isolated from HO1429 subsisting on 7. Molecular (m/z = 292.1) and fragment ions (m/z = 170.1) are
indicated.

S.-M. He et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5954–5957
5957
Figure 4. Growth of E. coli phn+ strain HO1429 on MOPS-agar minimal media containing (l to r) methyl phosphonate (0.5 mM), 1 (0.5 mM), and no phosphorus source. The
plates were illuminated with ambient light.
possibility can be addressed by synthesizing the corresponding
dansyl substituted 1 and tracking its fate in vivo by TLC or HPLC
analysis of cell extracts.
References and notes
1. Karl, D. M. Nature 2000, 406(31), 33.
2. Hudson, J. J.; Taylor, W. D.; Schindler, D. W. Nature 2000, 406, 54.
3. Quinn, J. P.; Kulakova, A. N.; Cooley, N. A.; McGrath, J. W. Environ. Microbiol.
2007, 9, 2392.
4. Van Mooy, B. A.; Fredricks, H. F.; Pedler, B. E.; Dyhrman, S. T.; Karl, D. M.;
Koblizek, M.; Lomas, M. W.; Mincer, T. J.; Moore, L. R.; Moutin, T.; Rappe, M. S.;
Webb, E. A. Nature 2009, 458, 69.
5. White, A. K.; Metcalf, W. W. Annu. Rev. Microbiol. 2007, 61, 379.
6. Wackett, L. P.; Shames, S. L.; Venditti, C. P.; Walsh, C. T. J. Bacteriol. 1987, 169,
710.
7. Frost, J.; Loo, S.; Cordeiro, M.; Li, D. J. Am. Chem. Soc. 1987, 109, 2166.
8. Chen, C. M.; Ye, Q. Z.; Zhu, Z. M.; Wanner, B. L.; Walsh, C. T. J. Biol. Chem. 1990,
265, 4461.
9. Metcalf, W. W.; Wanner, B. L. J. Bacteriol. 1993, 175, 3430.
10. Huang, J.; Su, Z.; Xu, Y. J. Mol. Evol. 2005, 61, 682.
11. Hove-Jensen, B.; Rosenkrantz, T. J.; Haldimann, A.; Wanner, B. L. J. Bacteriol.
2003, 185, 2793.
12. Rizk, S. S.; Cuneo, M. J.; Hellinga, H. W. Protein Sci. 2006, 15, 1745.
13. Gebhard, S.; Cook, G. M. J. Bacteriol. 2008, 190, 1335.
14. Podzelinska, K.; He, S. M.; Wathier, M.; Yakunin, A.; Proudfoot, M.; Hove-
Jensen, B.; Zechel, D. L.; Jia, Z. J. Biol. Chem. 2009, 284, 17216.
15. Avila, L. Z.; Draths, K. M.; Frost, J. W. Biorg. Med. Chem. Lett. 1991, 1, 51.
16. Neidhardt, F. C.; Bloch, P. L.; Smith, D. F. J. Bacteriol. 1974, 119, 736.
17. Luo, Y.; Zechel, D. L. Can. J. Chem. 2006, 84, 743.
The viability of 7 as a substrate for E. coli CP-lyase provides a
powerful tool for the dissection of the metabolic pathway of organ-
ophosphonate degradation. Strains of E. coli can be created where
the phn operon is constitutively expressed so that the strains will
degrade organophosphonates even in the presence of low concen-
trations of inorganic phosphate.18 Subsequently, individual phn
genes can be ‘knocked out’ in this strain and the resulting mutants
cultured with 7. When 7 is incorporated into metabolic intermedi-
ates such as 1, the absence of a downstream CP-lyase enzyme
should result in accumulation of the intermediate, which can be
easily detected due to the attached fluorophore. Purification of
such intermediates for identification by MS and NMR will also be
facilitated by the fluorophore. Ultimately the use of 7 in this strat-
egy will allow the assignment of specific reactions and substrates
to individual CP-lyase enzymes.
Supplementary data
18. Wanner, B. L.; Latterell, P. Genetics 1980, 96, 353.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.08.035.