Studies on the biodegradation of fosfomycin: Growth of Rhizobium huakuii PMY1 on possible intermediates synthesised chemically

Article abstract of DOI:10.1039/b821829c

The first step of the mineralisation of fosfomycin by R. huakuii PMY1 is hydrolytic ring opening with the formation of (1R,2R)-1,2- dihydroxypropylphosphonic acid. This phosphonic acid and its three stereoisomers were synthesised by chemical means and tes

Full text of DOI:10.1039/b821829c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Studies on the biodegradation of fosfomycin: Growth of Rhizobium huakuii
PMY1 on possible intermediates synthesised chemically
John W. McGrath,a Friedrich Hammerschmidt,*b Werner Preusser,b John P. Quinn*a and Anna Schweiferb
Received 5th December 2008, Accepted 23rd February 2009
First published as an Advance Article on the web 23rd March 2009
DOI: 10.1039/b821829c
The first step of the mineralisation of fosfomycin by R. huakuii PMY1 is hydrolytic ring opening with
the formation of (1R,2R)-1,2-dihydroxypropylphosphonic acid. This phosphonic acid and its three
stereoisomers were synthesised by chemical means and tested as their ammonium salts for
mineralisation as evidenced by release of Pi. Only the (1R,2R)-isomer was degraded. A number of salts
of phosphonic acids such as ( )-1,2-epoxybutyl-, ( )-1,2-dihydroxyethyl-, 2-oxopropyl-,
(S)-2-hydroxypropyl-, ( )-1-hydroxypropyl- and ( )-1-hydroxy-2-oxopropylphosphonic acid were
synthesised chemically, but none supported growth. In vitro C–P bond cleavage activity was however
detected with the last phosphonic acid. A mechanism involving phosphite had to be discarded as it
could not be used as a phosphorus source. R. huakuii PMY1 grew well on (R)- and (S)-lactic acid and
hydroxyacetone, but less well on propionic acid and not on acetone or (R)- and ( )-1,2-propanediol.
The Pi released from (1R,2R)-1,2-dihydroxypropylphosphonic acid labelled with one oxygen-18 in the
PO3H2 group did not stay long enough in the cells to allow complete exchange of 18O for 16O by enzymic
turnover.
Although the biosynthesis of natural phosphonates and their
Introduction
biomineralisation (including that of man-made phosphonates)
Esters of orthophosphate (Pi), characterised by four P–O bonds,
has been actively pursued, our knowledge is still fragmentary,
especially with regard to their catabolism.7 Unlike the labile P–O
bond, the P–C bond is highly stable under a variety of conditions,
even under forcing acidic and basic conditions.
are ubiquitous in the biosphere. However, the number of known
natural products containing a P–C bond is limited, but steadily
growing.1 Four biologically interesting examples are given (Fig. 1).
2-Aminoethylphosphonic acid (2-AEP, 1) was isolated first from
rumen protozoa and is the most simple and widely distributed
phosphonic acid.2 Fosfomycin (2) containing additionally an
oxirane ring is a cell-wall active antibiotic used clinically.3 The
commercial herbicide phosphinothricin (3) containing two P–C
bonds inhibits glutamine synthetase.4 Fosmidomycin (4), a potent
antimalarial agent blocking the Rhomer pathway of terpenoid
biosynthesis, is currently undergoing field trials.5 Glyphosate (5)
is a synthetic herbicide used extensively throughout the world.6
Acquisition of adequate supplies of Pi as the preferred P source
is vital for all living cells. Under conditions of Pi starvation gene
systems are induced, whose products are involved in acquisition
and assimilation of phosphonates. Earlier studies suggested that
their breakdown is under the strict control of the pho regulon
which is expressed only under conditions of phosphate limitation.8
It was assumed that P–C compounds were therefore utilised only
as P source, but failed to serve as a carbon (or nitrogen) source, as
excess Pi released during catabolism would repress any further
breakdown. The majority of these studies involved synthetic
alkyl- and arylphosphonates and glyphosate, whose P–C bond is
cleaved by microbial multi-enzyme systems with a broad substrate
specificity (‘P–C lyases’) with release of Pi and a hydrocarbon.8
Later, P–C hydrolase enzymes and genes were found, which for
the most part were not subject to control by Pi. The supplied
phosphonates are used as the sole carbon and energy source for
growth of the host bacterium. Phosphonoacetaldehyde hydrolase
(phosphonatase), a well characterised pho-regulated enzyme, is
part of the degradation pathway of AEP.9 It is first transam-
inated (2-AEP:pyruvate aminotransferase) and then converted
to acetaldehyde and Pi by phosphonatase. A pho independent
AEP degradative operon has been identified in Pseudomonas
putida.7 Similarily, phosphonoacetate10 and phosphonopyruvate11
hydrolases cleave the respective phosphonates to give acetate and
pyruvate, respectively, and Pi.
Fig. 1 Compounds containing a P–C bond.
aSchool of Biology and Biochemistry, The Queen’s University of Belfast,
Medical Biology Centre, 97 Lisburn Rd., Belfast, BT9 7BL, Northern
Ireland. E-mail: j.quinn@qub.ac.uk; Fax: +44 28 90975877; Tel: +44 28
90972287
bUniversity of Vienna, Institute of Organic Chemistry, Wa¨hringerstrasse
38, 1090, Vienna, Austria. E-mail: friedrich.hammerschmidt@univie.ac.at;
Fax: +43 (1) 4277 9521; Tel: +43 (1) 4277 52105
The biosynthesis of fosfomycin [cis-(1R,2S)-1,2-epoxy-
propylphosphonic acid], produced by strains of Streptomyces
1944 | Org. Biomol. Chem., 2009, 7, 1944–1953
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
fradiae, S. wedmorensis, S. viridochromogenes and Pseudomonas
syringae, has been studied extensively and features five, in-part
unique steps, starting from phosphoenol pyruvate.12 First reports
showed that the cleavage of the P–C bond of fosfomycin is in-
ducible only under conditions of phosphate limitation. Some years
ago, we however reported the isolation of the bacterium R. huakuii
PMY1 using fosfomycin as a carbon source or as a carbon
and phosphorus source.13 The mineralisation was independent
of the phosphate status of the cell and resulted in essentially
quantitative extracellular accumulation of organophosphonate-
derived Pi. None of a series of phosphonates could replace
fosfomycin. Cell extracts of R. huakuii PMY1 grown on fosfomycin
(5 mM) as carbon, energy and phosphorus source, contained no
in vitro P–C bond-cleaving activity either on fosfomycin or other
organophosphonate substrates tested. These findings induced us
to synthesise putative intermediates of the biodegradative pathway
of fosfomyin and study their utilisation by R. huakuii PMY1.
Results and discussion
Chemical stability of oxirane ring in fosfomycin
From a chemical point of view, an oxirane ring is fairly labile under
acidic and basic conditions. A nucleophile can attack fosfomycin
regioselectively on either C-1 or C-2. Vidal et al. found that a
solution of the disodium salt of fosfomycin in 0.1 N H2SO4,
when heated at 50 ◦C for 24 h, furnished a diol homogeneous
Scheme 1 Acid and base-catalysed opening of oxirane ring in fosfomycin.
opening of fosfomycin is less regioselective than the acid-catalysed
one.
1
by paper chromatography and H NMR spectroscopy, assuming
Compared to H2O and OH-, ammonia17 attacks the epoxide ring
less regioselectively (C-2 : C-1 = 2.5 : 1), CF3CO2H, HCl and HBr
attack18 at C-2. The mode of action of fosfomycin and the mode
of resistance to it are based on the alkylation of the sulfur atoms
of cysteins of specific proteins by fosfomycin, whereby inversion
of configuration is caused at C-219 on oxirane ring opening in the
former case and at C-120 in the latter.
nucleophilic attack of water at C-2.14 Later, it was proven by
the synthesis of reference samples, that the diol had indeed
(R)-configuration at carbon atoms 1 and 2.15 In both studies a
small amount of the (1S,2S)-diol formed by attack of water at
C-1 of fosfomycin might have gone undetected. To exclude that
possibility, we reinvestigated the acid-catalysed ring opening of
fosfomycin, and studied the base-catalysed one as well (Scheme 1).
We found that its free acid, when left as an aqueous solution
without added H2SO4 at RT for 2–3 days, was converted to diol 6
quantitatively as evidenced by NMR spectroscopy. To determine
whether it was only stereoisomer (1R,2R)-6 or in admixture
with (1S,2S)-6, the crude diol obtained was esterified with dia-
zomethane to give dimethyl phosphonates 7, which were converted
with (S)-MTPA-Cl16 to Mosher esters (1R,2R)-7·[MTPA-(R)]2 and
(1S,2S)-7·[MTPA-(R)]2. Similarly, an authentic sample of diol
(1S,2S)-6 (for preparation see later) was transformed into the
Mosher ester (1S,2S)-6·[MTPA-(R)]2. Careful inspection of their
NMR spectra (1H, 31P) revealed that they were diastereomerically
homogeneous. Therefore, acid-catalysed opening of the epoxide
ring of fosfomycin furnished exclusively (>99%) diol (1R,2R)-6,
implicating attack of water at C-2.
Chemical synthesis of possible intermediates and their use in
degradative studies
With the aim of unravelling the stepwise mineralisation of
fosfomycin by R. huakuii PMY1, we decided to prepare putative
intermediates of the catabolic pathway and test them as phospho-
rus, carbon and energy sources for this bacterium. Additionally
for those organophosphonate substrates under test, to negate the
possibility of their lack of transport into the cell, studies were
also performed on cell-free extracts prepared from fosfomycin
grown cells of R. huakuii PMY1. As virtually nothing was known
about the biodegradation of natural organophosphonates when
we started this work, we tested even unlikely intermediates as
outlined in Scheme 2.
Direct enzymatic cleavage of the P–C bond of fosfomycin by
a specific P–C lyase could yield Pi and epoxide (S)-8, which
might be hydrolysed by an epoxide hydrolase to give diol
(S)- or (R)-9, depending on which carbon atom inversion was
effected (Route A). Deoxygenation of fosfomycin could give cis-
1-propenylphosphonic acid (10) (Route B) and reductive ring
opening either hydroxypropylphosphonic acid (R)-11 or (S)-12
(Routes C and D).
When a solution of fosfomycin in 2 M NaOH was left for 14 days
at RT, the ratio of diol/fosfomycin was 11 : 89, indicating very
slow hydrolysis of the epoxide ring. The reaction rate increased
by heating the mixture at 90 ◦C, at which it was finished after
1 day. Part of the reaction mixture was converted to the free
acid using Dowex 50, H+, followed by esterification to give
Mosher esters of dimethyl phosphonates 7 as before. This time, the
NMR spectra revealed that it was a mixture, derived from diols
(1R,2R)- and (1S,2S)-6 in a ratio of 10 : 1. Base-catalysed ring
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1944–1953 | 1945
©

View Article Online
Scheme 4 Conversion of fosfomycin by an epoxide hydrolase to diol
(1R,2R)- or (1S,2S)-6.
groups were removed with TMSBr/allylsilane and the benzyl
group by catalytic hydrogenation (Scheme 5).15 The second and
more convenient method was to keep an aqueous solution of the
free acid of fosfomycin at RT (see Scheme 1), neutralise it with
NH3 and finally lyophilise it. When this salt was added to growth
medium (5 mM) of R. huakuii PMY1, it was used as a carbon
and phosphorus source with concomitant release of Pi (4.1 mM)
into the growth medium in virtually quantitiative yield. To test
the degradation of the other stereoisomers of (1R,2R)-6 as well,
their ammonium salts were prepared similarly. Surprisingly, none
supported growth as evidenced by the lack of release of Pi. These
findings in combination with the lack of growth of R. huakuii
on (R)-1,2-propanediol [(R)-9] show that (1) the first step of the
biodegradation of fosfomycin is the stereospecific epoxide ring
opening at C-2 by a putative epoxide hydrolase and (2) that the
following step is not cleavage of the P–C bond. If that were to
occur, Pi and (R)-9 would result. The metabolism of the latter
would be a prerequisite for its use as an energy and carbon
source. However, it does not take place. Moreover no in vitro
cleavage of the P–C bond of (1R,2R)-6 could be detected. Addi-
tionally, we fed the ammonium salts of phosphonic acids ( )-17
and ( )-19, homologues of fosfomycin and 1,2-dihydroxypropyl-
phosphonic acid, respectively, to R. huakuii. The former was
prepared by epoxidation of the triethylammonium salt of cis-
1-butenylphosphonic acid (16) with H2O2/Na2WO4·H2O and
isolation as ammonium salt by ion exchange chromatography,12d
the latter from hydroxyphosphonate ( )-1824 (Scheme 6). None
supported growth or were degraded in cell-extract preparations.
Scheme 2 Putative intermediates of biodegradation of fosfomycin.
Therefore, we had to get the putative intermediates 9, of
which we wanted to test the racemic and (R)-configured form,
10, and the racemates of 11 and 12. However, epoxide (S)-8
is very volatile (bp 35 ◦C) and was not considered a suitable
compound for testing. ( )-1,2-Propanediol (9) is commercially
available, the (R)-enantiomer was obtained by deblocking the
corresponding benzyl ether21 (R)-13 (Scheme 3). Phosphonic acids
1022 and ( )-1212c left as their cyclohexylammonium salts from
previous studies on the biosynthesis of fosfomycin were converted
to the ammonium salts (Dowex 50,H+, then neutralisation with
ammonia). The ammonium salt of ( )-11 was easily accessible by
deblocking hydroxyphosphonate ( )-1423 using mild conditions15
(Me3SiBr/allylSiMe3), followed by neutralisation of the free acid
with ammonia. This ammonium salt and others obtained by
lyophilisation as fluffy powders were homogeneous by NMR
and contained varying amounts of nitrogen and water as found
by microanalysis. They were normally used directly for the
degradative studies, assuming a 95% yield for the preparation
of the ammonium salts. They could be purified by ion exchange
-
chromatography (Dowex 1, HCO3 form), if necessary. Unfortu-
nately, none of these putative intermediates was used by R. huakuii
PMY1 as carbon and phosphorus sources. Additionally no in vitro
C–P bond cleavage activity was detected. We therefore envisoned
formal opening of the epoxide ring by water as an alternative first
step in the mineralisation of fosfomycin (Scheme 4).
Scheme 3 Synthesis of putative intermediates 9 and 11 of biodegradation
of fosfomycin.
Scheme 5 Conversion of protected stereoisomeric hydroxyphosphonates
15 to the ammonium salts of the corresponding acids 6.
Depending on the regioselectivity of the enzyme, diol 6 could
have either (1R,2R)- or (1S,2S)-configuration, if water attacked
at C-2 or C-1 of fosfomycin, respectively. The former isomer was
prepared by two methods. The first one was based on the protected
diol (1R,2R)-15 prepared by base-catalysed addition of diiso-
propyl phosphite to (R)-O-benzyllactaldehyde.15 The isopropyl
The next step on the biodegradative pathway of 1,2-
dihydroxypropylphosphonic acid (1R,2R)-6 cannot be the cleav-
age of the P–C bond by a P–C lyase as the resulting propane-
diol [(R)-9] was not metabolised as already found. Evidently,
(1R,2R)-6 is modified in the side chain before the P–C bond
1946 | Org. Biomol. Chem., 2009, 7, 1944–1953
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
carboxylic acids.27 However, the corresponding salts show in-
creased stability compared to the free acids. Possibly, an en-
zyme with a new mode of action, could hydrolyse a-hydroxy-
phosphonate (1R,2R)-6 to lactaldehyde [(R)-25] and phosphite
(22, Route D) or a-oxophosphonates 20 and (R)-28 to propionic
acid [21, Route B] and lactic acid [(R)-29, Route F], respectively,
and phosphite. Naturally, cleavage of the P–C bonds in these
latter substrates as well as in 23 and (R)-26 by a novel P–C lyase,
whereby Pi and aldehydes and acetone (24) and hydroxyacetone
(27), respectively, are formed, is also possible. Metcalf and Wolfe28
found that phosphite-oxidising organisms may in fact be quite
common and Vrtis et al.29 studied the mechanism of a phosphite
dehydrogenase.
However phosphite could not be used as a phosphorus source
by R. huakuii PMY1 and no phosphite oxidase activity (using
both phosphorous acid and sodium phosphite as substrates) could
be detected in PMY1 cell-free-extracts. Some of the mechanisms
outlined in Scheme 7 must therefore be dismissed (Routes B, D,
and F). Furthermore, R. huakuii grew very well on hydroxyacetone
(27), (R)- and (S)-lactic acid [(R)- and (S)-29], but less well
on propionic acid and not on acetone. Thus route E was the
only one left for the degradation of (1R,2R)-6. We decided to
synthesise 26, in the first place in its racemic form, and its ana-
logue 23 without the hydroxyl group (Scheme 8). Base-catalysed
addition of dimethyl phosphite (31) to metacrolein (30) furnished
a-hydroxyphosphonate ( )-32 in 87% yield. It was deblocked as
usual and converted to the sodium salt, which was ozonolysed
to give the desired 2-oxopropylphosphonic acid ( )-23 as sodium
salt. The analogue without the hydroxyl group was produced by
deblocking 2-oxopropylphosphonate 34 and neutralising the acid
with NaOH. Ammonia was replaced in these two cases here by
NaOH, as ammonia might react with the carbonyl groups and
give side products.
Scheme 6 Preparation of ammonium salts of epoxy- and dihydroxy-
phosphonic acids ( )-17·and ( )-19.
is split (Scheme 7). Enzymatic elimination of water from the
diol could give either 1-oxopropylphosphonic acid (20) or
2-oxopropylphosphonic acid (23), depending on the hydroxyl
group which is removed. Dehydrogenation however, could yield
isomeric hydroxy-oxopropylphosphonic acids (R)-26 or (R)-28.
Whereas the P–C bond in alkyl- and arylphosphonic acid esters
are extremely stable chemically, those in a-hydroxyalkyl- and
a-oxoalkylphosphonic acid esters are fairly labile. The former are
base labile and are split to give dialkyl phosphite and aldehydes
or ketones, from which they are formed easily (Pudovik25 and
retro-Abramov reaction26). a-Oxoalkylphosphonic acid esters are
even more labile and are hydrolysed to dialkyl phosphites and
Scheme 8 Synthesis of 2-oxopropylphosphonic acids 23 and ( )-26 (as
sodium salts).
The sodium salt of 23 was not used as carbon and phosphorus
source by R. huakuii PMY1 as anticipated. Surprisingly, the
putative intermediate ( )-26 was not used either. To account
for this unexpected finding in combination with the fact that
hydroxyacetone is degraded, we have to assume either that
phosphonic acid ( )-26 does not enter the cells or that (S)-26
Scheme 7 Putative intermediates for degradation of fosfomycin beyond
(1R,2R)-1,2-dihydroxypropylphosphonic acid.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1944–1953 | 1947
©

View Article Online
acts as a growth inhibitor [since it is most likely to be (R)-26
formed in vivo from the (1R,2R)-6 precursor]. Yet in vitro C–P
bond cleavage activity using ( )-26 as a substrate was detectable.
Control assays containing no cell-extract or no substrate were
performed to confirm this.
Conclusions
In summary we have shown that fosfomycin is stereospecifi-
cally opened under acidic conditions to give the (1R,2R)-1,2-
dihydroxypropylphosphonic acid exclusively and under basic
conditions in admixture with 9% of the (1S,2S)-enantiomer. The
first step of the mineralisation of fosfomycin by R. huakuii PMY1
is the hydrolytic ring opening with the formation of (1R,2R)-1,2-
dihydroxypropylphosphonic acid. Of the four stereoisomeric
1,2-dihydroxypropylphosphonic acids prepared chemically as
ammonium salts only the (1R,2R)-isomer was mineralised as
evidenced by release of Pi and growth. A number of the phosphonic
acids considered as putative intermediates on the biodegradative
pathway of fosfomycin were synthesised chemically and tested
for growth and release of Pi. None supported growth; in vitro
C–P bond cleavage activity was however detected with racemic
1-hydroxy-2-oxopropylphosphonic acid. A mechanism involving
phosphite had to be dismissed as it could not be used as
phosphorus source. Hydroxyacetone, a putative intermediate of
the catabolic pathway, supported growth.
Finally, we addressed the length of stay of Pi formed by the
cleavage of the P–C bond in the the cells using (1R,2R)-6 with an
oxygen-18 labelled phosphonate group (Scheme 9). Thus, isobutyl
lactate [(R)-35] was TIPS-protected30 and reduced with DIBAH
at -78 ◦C to give lactaldehyde [(R)-37], to which 18O-labelled
diethyl trimethylsilyl phosphite was added at low temperature
diastereoselectively.31 The labelled phosphite was obtained by
silylation32 of 18O-labelled diethyl phosphite33 derived from triethyl
phosphite and H218O. The reaction mixture was treated with HCl
in EtOH to remove the trimethylsilyl group from the a-hydroxy
group to furnish a mixture of diastereomers separable by flash
chromatography [(1R,2R)-38-(1S,2R)-38, 89 : 11]. Deblocking
of the major diastereomer (84% oxygen-18 labelled) yielded the
ammonium salt of the 18O-labelled (1R,2R)-6. The 18O-labelled
phosphonic acid was used by R. huakuii PMY1 as a carbon
and phosphorus source. After release of Pi (4 mmol) from
5 mM phosphonate, cells were removed by centrifugation and the
supernatant was freeze-dried. The residue was found to contain
inorganic phosphate (0.72% w/w) and the ratio of PO43-/PO318O3-
was 95.6 : 4.4 as determined by LC-MS. Evidently, the inorganic
phosphate does not stay long enough in the cells after release from
the labelled precursor to allow complete exchange of 18O for 16O
by enzymic turnover.
Experimental
General experimental
1H, 13C and 31P NMR spectra were recorded in CDCl3 and D2O
at 300 K on a Bruker Avance DRX 400 at 400.13, 100.61 and
161.98 MHz, respectively. Chemical shifts were referenced to
residual CHCl3 (dH 7.24) or HDO (dH 4.80) and CDCl3 (dC 77.00)
and external H3PO4 (85%). Chemical shifts are given in d in ppm
and J values in Hz. The FD mass spectra were recorded on a
◦
Finnigan MAT 900S at 30 C. IR spectra were run on a Perkin-
Elmer 1600 FT-IR spectrometer; liquid samples were measured
as films on a silicon disc.34 Optical rotations were measured at
20 ◦C on a Perkin-Elmer 351 polarimeter in a 1 dm cell. TLC was
carried out on 0.25 mm thick Merck plates, silica gel 60 F254. Flash
(column) chromatography was performed with Merck silica gel
60 (230–400 mesh). Spots were visualised by UV and/or dipping
the plate into a solution of (NH4)6Mo7O24·4H2O (23.0 g) and of
Ce(SO4)2·4H2O (1.0 g) in 10% aqueous H2SO4 (500 mL), followed
by heating with a heat gun. Phosphonic acids eluted from Dowex
-
1, HCO3 were also detected by TLC (iPrOH–H2O–conc. NH3
6 : 3 : 1, then visualising). The following ion exchange resins
-
were used: Dowex 1 ¥ 8, HCO3 (100–200 mesh) Dowex 50 ¥ 8,
H+ (50–100 mesh). DMF and pyridine were dried by refluxing
over powdered CaH2, then distillation and storage over molecular
˚
sieves (4 A). Dichloromethane and 1,2-dichloroethane were dried
by passing through aluminium oxide 90 active neutral (0.063–
˚
0.200 mm, activity I) and stored over molecular sieves (3 A and
˚
4 A, respectively). Hexane was dried by storage over molecular
˚
sieves (4 A). Et2O was refluxed over LiAlH4, THF over potassium
and distilled prior to use. Melting points were determined on a
Reichert Thermovar instrument and were uncorrected.
Determination of isotopic composition of inorganic phosphate
by LC-MS: HPLC HP 1100 with UV/Vis Diode Array Detector,
equipped with a vacuum degasser and quaternary pump and a HP
1050 Autosampler (all Hewlett Packard, Palo Alto, CA, USA);
Mass spectrometry: quadrupole system HP 5989 B (Hewlett
Packard) with Atmospheric Pressure Ionisation Interface HP
Scheme 9 Preparation of ammonium salt of (1R,2R)-6 18O-labelled in the
phosphonate group.
1948 | Org. Biomol. Chem., 2009, 7, 1944–1953
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
59987 A (Hewlett Packard) equipped with an RF-only hexapole
(Analytica of Branford, Branford, CT, USA).
1H NMR spectra were identical to those of the corresponding
racemates.
It was assumed that the conversion of phosphonates to the
respective ammonium or sodium salts was effected in 95% yield.
Ammonium salts of hydroxyphosphonic acids could be purified
by ion exchange chromatography by analogy to the conversion of
1,2-dihydroxypropylphosphonic acid ¥ 1.5 cyclohexylamine to the
ammonium salt given below, if necessary.
(1R,2R)-15. dC (100.6 MHz; CDCl3) 138.0, 128.4 (2 C), 128.0
(2 C), 127.7, 74.3 (d, J 3.8), 72.4 (d, J 162.9), 71.4 (d, J 6.9), 71.4,
70.8 (d, J 6.9), 24.2 (d, J 3.1), 24.04 (d, J 4.6), 24.0 (d, J 4.6), 23.8
(d, J 5.4), 16.7 (d, J 7.7).
(1S,2R)-15. dC (100.6 MHz; CDCl3) 138.2, 128.3 (2 C), 127.8
(2 C), 127.5, 74.9 (d, J 6.1), 71.2 (d, J 6.9), 71.1 (d, J 6.9), 70.7,
70.6 (d, J 159.9), 24.1 (d, J 3.8), 24.0 (d, J 3.8), 23.8 (d, J 4.6),
23.8 (d, J 5.4), 16.7 (d, J 3.8).
Conversion of 1,2-dihydroxypropylphosphonic acid ¥ 1.5 cyclo-
hexylamine (prepared from the disodium salt of fosfomycin) to
ammonium salt. The salt (0.300 g, 0.984 mmol) was dissolved
in water and applied to a column filled with Dowex 1 ¥ 8,
Deblocking of phosphonates 15. The four diisopropyl phos-
phonates were deblocked according to a literature procedure15
except that CCl4 was replaced by dry 1,2-dichloroethane (molar
ratio of 15-allylTMS-TMSBr 1 : 1.5 : 5; 17 h at 50 ◦C). After
hydrogenolytic removal (Pd/C/H2) of the benzyl group in dry
EtOH and removal of the catalyst, ammonia (25%) was added
and the solution was concentrated. The residue was dissolved in
water and lyophilisation gave fluffy or gum-like ammonium salts.
If there was still EtOH present in the salt (by 1H NMR in D2O), it
was dissolved again in water and lyophilised. It was assumed that
the conversion of the phosphonates to the respective ammonium
salts was effected in 95% yields. These salts were used directly for
the degradation studies.
The spectroscopic data (1H, 13C and 31P NMR; D2O) of the
ammonium salts of (1R,2R)- and (1S,2S)-6 were identical to the
one derived from fosfomycin. The spectroscopic data of (1R,2S)-
and (1S,2R)-6 were identical; dH (400.1 MHz; D2O) 4.06 (1 H, m,
PCCH), 3.62 (1 H, dd, J 10.1, 5.3, PCH), 1.28 (3 H, d, J 6.6, CH3);
dC (100.6 MHz; D2O) 73.7 (d, J 146.9), 68.5 (d, J 6.9), 17.6 (d, J
5.4); dP (162 MHz; D2O) 17.6.
-
HCO3 (75 mL, i.d. 2 cm ¥ 23 cm) and eluted with aqueous
NH4HCO3 (0.1 M, 350 mL, then 0.25 M; fractions of 25 mL). The
phosphonic acid was eluted in fractions 19–25 (TLC, Rf 0.42).
Fractions containing product were pooled, concentrated under
reduced pressure to a small volume and lyophilised to give the
ammonium salt as a white, fluffy product (0.176 g).
(1R,2R)-Dihydroxypropylphosphonic acid, ammonium salt
[(1R,2R)-6·(NH3)x], prepared from fosfomycin. Fosfomycin
(1.32 g of the disodium salt, 7.25 mmol, dissolved in 5 mL of water)
was passed down a column of Dowex 50 ¥ 8, H+ (30 mL). It was
eluted with water until neutral. The solution (about 50 mL) was
left at RT for 48 h and then basified with ammonia (25%, 2 mL).
Lyophilisation gave the ammonium salt (1.470 g) as a hygroscopic
fluffy powder. Lyophilisation of the aqueous solution without
prior addition of ammonia gave the oily free acid. Ammonium
salt: dH (400.1 MHz; D2O) 4.06 (1 H, m, PCCH), 3.44 (1H, dd, J
9.6, 4.3, PCH), 1.26 (3 H, d, J 6.6, CH3); dC (100.6 MHz, D2O)
73.2 (PCH, d, J 145.3), 68.3 (CH, d, J 3.1), 18.9 (CH3, d, J 7.7);
dP (162 MHz, D2O) 17.4.
Base-catalysed ring opening of fosfomycin. A solution of
fosfomycin (Na2 salt, 0.500 g) in 2 M NaOH (10 mL) was
heated at 90 ◦C for 24 h. After 17 h a sample (0.25 mL) was
withdrawn, neutralised with solid CO2 and lyophilised. The residue
Synthesis of all four stereoisomeric 1,2-dihydroxypropyl-
phosphonic acids, ammonium salts [(1R,2R)-, (1S,2S)-, (1R,2S)-
and (1S,2R)-6·(NH3)x], from protected precursors 15. They
were prepared by deblocking the corresponding diisopropyl
2-benzyloxy-1-hydroxypropylphosphonates 15, which were ob-
tained by analogy to the compounds deuterated at C-1, but
starting from nondeuterated (R)- and (S)-2-benzyloxypropanal.15
Diisopropyl phosphite was added to the aldehyde in THF,
catalysed by DBU (-78 ◦C to room temperature). The ratio of
the diastereomeric phosphonates was 1 : 1 (1H NMR, C6D6).
Homogeneous samples were obtained by flash chromatography
(CH2Cl2-EtOAc 5 : 1, TLC for 3 : 1, (1R,2R)-15: Rf 0.38; (1S,2R)-
15: 0.31) and the combined yields were 60–70%.
1
was dissolved in D2O to record a H and 31P NMR spectrum
(besides diol there was still 3% of fosfomycin present). The cold
solution was diluted with water and applied to a column of Dowex
50 ¥ 8, H+. Washing with water gave an eluate, which was basified
with ammonia (25%) and lyophilised to give the ammonium salt
containing a small amount of an impurity.
Preparation of Mosher ester {7·[MTPA-(R)]2}. Dimethyl
1,2-dihydroxypropylphosphonate (7) was derived from fosfomycin
(by acid- or base-catalysed ring opening) and (1S,2S)-6. A sample
(up to 50 mg) of 1,2-dihydroxypropylphosphonic acid derived
from fosfomycin (by acid- or base-catalysed ring opening) or
(1S,2S)-6 was esterified in dry MeOH with a distilled etheral
solution of diazomethane at 0 ◦C. Concentration of the solution
yielded a mixture of dimethyl 1,2-dihydroxypropylphosphonate
◦
(1R,2R)-15. Colourless needles (hexanes), mp 83 C (lit.15
for C-1 deuterated compound: 84–86 ◦C); [a]D -24.2 (c 2.0 in
20
CH2Cl2) {lit.15 for C-1 deuterated compound: [a]D -22.8 (c 1.63
20
20
in CH2Cl2)} and (1S,2R)-15, colourless oil, [a]D -6.3 (c 2.0 in
1
and its isomeric monomethyl ethers (10 : 1 : 1, by H and 31P
CH2Cl2) {lit.15 for C-1 deuterated compound: [a]D -5.8 (c 2.08
20
NMR). A portion of the mixture (15 mg) was esterified at RT
in dry CH2Cl2 (1 mL) containing dry pyridine (5 drops) and
(S)-MTPACl (0.5 mL of a 0.5 M solution in dry CH2Cl2).
After 18 h the solution was concentrated under reduced pressure
and water and CH2Cl2 were added to the residue. The organic
phase was washed with 2 M HCl, a saturated aqueous solution
of NaHCO3, dried (Na2SO4) and concentrated under reduced
in CH2Cl2)} were obtained from (R)-2-benzyloxypropanal.
(1S,2S)-15. Colourless needles (hexanes), mp 83 ◦C, [a]D
20
+23.5 (c 2.18 in CH2Cl2) and (1S,2R)-15, colourless oil, [a]D20 +6.5
(c 2.0 in CH2Cl2), were obtained from (S)-2-benzyloxypropanal.
The 1H and 13C NMR spectra of (1R,2R)- and (1S,2S)-15 and
of (1S,2R)- and (1R,2S)-15, respectively, were identical. Their
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1944–1953 | 1949
©

View Article Online
pressure to give an oil. Part of it was investigated directly by
1H and 31P NMR spectroscopy, the remaining portion after flash
chromatography (hexanes-EtOAc 1 : 1, Rf 0.13 for 1 : 2).
(25%, 0.5 mL). The solution was concentrated under reduced
pressure. The residue was dissolved twice in water (2 ¥ 20 mL)
and lyophilised to give the ammonium salt of ( )-19 as a viscous
oil (0.600 g); dH (400.1 MHz; D2O) 3.91 (1 H, ddd, J 11.6, 4.6, 2.4,
PCH), 3.83–3.65 (2 H, m, CH2); dc (100.6 MHz; D2O) 71.4 (d, J
147.6), 63.5 (d, J 9.2); dP (162 MHz; D2O) 16.8.
(1R,2R)-7·[MTPA-(R)]2. dH (400.1 MHz; CDCl3) 7.56–7.32
(10 H, m), 5.58–5.60 (1 H, m), 5.52 (1 H, dd, J 12.9, 3.3, PCH),
3.63 (3 H, d, J 10.9, OCH3), 3.48 (3 H, d, J 10.9, OCH3), 3.47 (3
H, q, J ~1, OCH3), 3.44 (3 H, q, J ~1, OCH3), 1.34 (3 H, dd, J 6.6,
1.3, CH3); dP (162 MHz; CDCl3) 17.9.
2-Oxopropylphosphonic acid (23), sodium salt. A solution
of dimethyl 2-oxopropylphosphonate (0.347 g, 2.09 mmol), al-
lyltrimethylsilane (0.358 g, 0.50 mL, 3.14 mmol) and TMSBr
(1.60 g, 1.38 mL,◦10.45 mmol) in dry 1,2-dichloroethane (8 mL)
was heated at 50 C for 2 h under argon. After cooling, volatile
components were removed under reduced pressure (0.5 mbar). The
residue was dissolved in dry 1,2-dichloroethane (5 mL) and again
concentrated under reduced pressure. The residue was dissolved
in a mixture of ethanol and water (10 mL, 1 : 1) and concentrated
after 15 min at reduced pressure and dissolved in water (5 mL).
After bringing the pH to 7.0 (1 M NaOH) the product was
lyophilised to give the sodium salt of 23 as a white powder (0.353 g);
dH (400.1 MHz; D2O) 3.017 (2 H, d, J 21.2, PCH2), 2.35 (3 H, s,
CH3); dC (100.6 MHz; D2O) 212.016 (d, J 4.6), 48.136 (d, J
107.1), 30.9; dP (162 MHz; D2O) 12.1. The hydrogen atoms of
the PCH2CO group were exchanged for deuterium. After leaving
the NMR probe for 5.5 h at room temperature, the ratio of PCH2-
PCHD-PCD2 was 0.36 : 0.48 : 0.16; dH (400.1 MHz; D2O) 3.000
(br. d, J 21.2, PCHD); dC (100.6 MHz; D2O) 212.061 (d, J 4.6,
CHDCO), 47.840 (td, J 107.1, 19.1, PCHD).
(1S,2S)-7·[MTPA-(R)]2. dH (400.1 MHz; CDCl3) 7.60–7.31 10
H, m), 5.60–5.49 (2 H, m), 5.55 (3 H, d, J 10.9, CH3), 3.48 (3 H,
q, J ~1, OCH3), 3.44 (3 H, q, J 10.9, OCH3), 3.43 (3 H, q, J ~1,
CH3), 1.32 (3 H, d, J 6.6, CH3); dP (162 MHz; CDCl3) 18.2.
(R)-(-)-1,2-Propanediol [(R)-9]. (R)-(-)-2-Benzyloxy-propa-
nol21 [3.00 g, 18.05 mmol; aD -28.0 (neat), obtained by re-
20
duction of (R)-(+)-isobutyl O-benzyllactate with LiAlH4] was
hydrogenated over Pd/C (0.20 g, 10%) in dry ethanol (60 mL)
containing 2 drops of conc. HCl in a Parr apparatus at 3.4 bar
for 3 h at RT. The catalyst was removed and the filtrate was
concentrated under reduced pressure. The crude product was bulb
to bulb distilled (80 ◦C/10 mm) (lit.35 92 ◦C/14 mm) to give diol
20
(R)-9 (0.79 g, 58%) as a colourless oil; [a]D -24.8 (c 1.02 in CHCl3)
24.4
{lit.35 [(a)]D -28.6 (in CHCl3)}; dH (400.1 MHz; CDCl3) 3.86
(1 H, qdd, J 7.9, 6.4, 3.0, CHO), 3.57 (1 H, dd, J 11.3, 3.0, one
of CH2O), 3.35 (1 H, dd, J 11.3, 7.9, one of CH2O), 3.13 (2 H, s,
2 ¥ OH), 1.13 (3 H, d, J 6.4, CH3).
( )-1-Hydroxypropylphosphonic acid, ammonium salt [( )-11·
( )-Dimethyl 1-hydroxy-2-methyl-2-propenylphosphonate [( )-
32]. A saturated solution of MeONa in MeOH (0.1 mL) was
added to a stirred solution of 2-methyl-2-propenal (0.70 g,
10 mmol) and dimethyl phosphite (1.10 g, 10 mmol) in dry
Et2O (20 mL) at -35 ◦C under argon. After stirring for 25 min
at -35 ◦C, conc. H2SO4 (7 drops) was added and the solvent
was removed (-35 ◦C/0.5 mm). The residue was purified by
flash column chromatography (EtOAc-CHCl3 9 : 1, Rf 0.28) to
yield hydroxyphosphonate ( )-32 (1.57 g, 87%) as a colourless
oil (Found: C, 39.9, H, 7.5. C6H13O4P requires C, 40.0, H,
7.3%); nmax/NaCl, film/cm-1 3299, 2958, 1649, 1236, 1033; dH
(400.1 MHz; CDCl3) 5.17 (1 H, dq, J 4.9, 1.0, CH=), 5.05 (1 H,
dq, J 2.5, 1.0, CH=), 4.43 (1 H, dd, J 12.8, 5.9, PCH), 4.01 (1 H,
dd, J 9.4, 5.9, OH), 3.788 and 3.785 (2 ¥ 3 H, 2 d, J 10.8, 2 ¥
OCH3), 1.87 (3 H, dt, J 3.0, 1.0, CH3); dC (100.6 MHz; CDCl3)
140.8 (d, J 3.8), 113.9 (d, J 11.1), 71.8 (d, J 157.1), 53.7 (d, J 6.9),
53.6 (d, J 7.1), 19.5 (d, J 2.2).
(NH3)x].
A solution of ( )-dimethyl 1-hydroxypropylphos-
phonate (0.325 g, 1.93 mmol), allyltrimethylsilane (0.332 g,
0.46 mL, 2.90 mmol) and TMSBr (1.48 g, 1.27 mL, 9.65 mmol)
in dry 1,2-dichloroethane (8 mL) was heated at 50 ◦C under
argon for 2 h. After cooling, volatile components were removed
under reduced pressure (0.5 mm). The residue was dissolved in dry
1,2-dichloroethane (5 mL) and again concentrated under reduced
pressure. The residue was then dissolved in a mixture of ethanol
and water (10 mL, 1 : 1) and after 15 min concentrated at reduced
pressure and dissolved in water (5 mL)/conc. ammonia (0.5 mL).
Lyophilisation gave the ammonium salt of ( )-11 as a white powder
(0.312 g); dH (400.1 MHz; D2O) 3.61 (1 H, ddd, J 10.1, 6.8, 3.3,
PCH), 1.83 (1 H, m), 1.60 (1 H, m), 1.04 (3 H, t, J 7.4, CH3);
dC (100.6 MHz, D2O) 71.0 (d, J 156.8), 25.1 (d, J 1.5), 10.7 (d, J
13.7); dP (162 MHz; D2O) 21.7.
cis-( )-1,2-Epoxybutylphosphonic acid, ammonium salt [( )-17·
(NH3)x]. The crude product obtained by epoxidation12d of cis-1-
butenylphosphonic acid (as triethylammonium salt) was purified
( )-1-Hydroxy-2-oxopropylphosphonic acid [( )-26], sodium salt.
A solution of hydroxyphosphonate ( )-31 (0.62 g, 3.4 mmol),
allyltrimethylsilane (0.777 g, 1.08 mL) and TMSBr (2.603 g,
2.24 mL) in dry 1,2-dichloroethane (5 mL) was left for 3 h at
RT under argon. Then, volatile components were removed under
reduced pressure (0.5 mm/RT). The residue was dissolved in dry
MeOH (20 mL) and the pH was brought to 7–8 by addition of
aqueous 1 M NaOH. After cooling to -78 ◦C, ozone-containing
dry air was bubbled through the stirred solution until a faint
blue color persisted. The cooling bath was removed to allow the
reaction mixture to warm up to RT. Ph3P (30 mg, dissolved in 1 mL
of CH2Cl2) was added, followed by water (40 mL) 5 min later. The
mixture was extracted with CH2Cl2 (3 ¥ 10 mL). The aqueous
layer was neutralised (pH 7, 1 M NaOH) and lyophilised to yield
-
by ion exchange chromatography (Dowex 1 ¥ 8, HCO3 ) by eluting
first with 0.1 M NH4HCO3, then 0.25 M. The fractions containing
the epoxyphosphonic acid (TLC, Rf 0.62) were pooled and
lyophilised to give ( )-17·(NH3)x as a white foam; dH (400.1 MHz;
D2O) 3.00 (1 H, m, PCCH), 2.73 (1 H, dd, J 18.8, 4.8, PCH), 1.69
(2 H, m), 0.90 (3 H, t, J 7.5, CH3); dP (162 MHz; D2O) 11.5.
( )-1,2-Dihydroxyethylphosphonic acid, ammonium salt [( )-19·
(NH3)x]. ( )-Dimethyl 2-benzyloxy-1-hydroxethyl-phosphonate
1824 (0.781 g, 3 mmol) was deblocked according to the proce-
dure used for phosphonates 15, except that demethylation was
performed at RT for 2 h. The free phosphonic acid in EtOH
after removal of the benzyl group was neutralised with ammonia
1950 | Org. Biomol. Chem., 2009, 7, 1944–1953
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
the solid sodium salt of ( )-26; nmax/nujol mull/cm-1 3300–2500,
1703, 1170, 1077; dH (400.1 MHz; D2O) 4.63 (1 H, d, J 19.7, PCH),
2.31 (3 H, s, CH3), contained 14 mol% of HCO2Na (s at 8.42); dC
(100.6 MHz; D2O) d 211.0, 79.0 (d, J 135.0), 27.7, formate: 120.8;
dP (162 MHz; D2O) 10.8, Pi at 0.9 (ratio 100 : 7).
16 h, while the bath was allowed to warm slowly to +10 ◦C. Volatile
components were removed (finally at 0.50 mm/1 h) to yield a
mixture of the crude silylated phosphonates as an oil. It was
dissolved in ethanol (40 mL) and stirred at RT for 1 h after the
addition of 4 drops of conc. HCl [TLC: hexanes–acetone 3 : 1, Rf
0.45 for (1R,2R)-38 and Rf 0.36 for (1S,2R)-38].
The solution was concentrated under reduced pressure. Water
(10 mL) was added to the residue and the mixture was extracted
with EtOAc (3 ¥ 15 mL). The combined organic layers were
dried (MgSO4) and concentrated in vacuo. The residue was
purified by flash chromatography (hexanes–acetone 10 : 1) to give
phosphonates (1R,2R)-38 (1.28 g, 50%) and (1S,2R)-38 (0.15 g,
6%) as colourless oils.
Preparation of diethyl trimethylsilyl [18O1]phosphite. A solu-
tion of freshly distilled triethyl phosphite (8.30 g, 50 mmol), dry
THF (5 ml), and H218O (1.0 g, 90% 18O) was kept for 16 h at room
temperature, then concentrated under reduced pressure and bulb
to bulb distilled (70 ◦C/8 mm) to give diethyl [P(18O)]phosphite33
20
20
(6.34 g, 91%); nD 1.4071 (unlabelled compound: nD 1.4065).
A mixture of the labelled phosphite (6.34 g, 45.3 mmol), hexam-
ethyldisilazane (3.66 g, 4.79 mL, 22.7 mmol) and TMSCl (2.47 g,
2.90 mL, 22.7 mmol) and dry hexane (45 mL) was refluxed for
1 h. The precipitated NH4Cl was removed by filtration through
Celite. The solution was concentrated under reduced pressure and
the residue was bulb to bulb distilled (75–80 ◦C/18 mm, lit.32 76–
77 ◦C/20 mm for unlabelled compound) to give oxygen-18 labelled
silyl phosphite (8.36 g, 87%) as a colourless liquid.
(1R,2R)-38. [a]D -2.1 (c 2.0 in acetone) {lit.31a [a]D -7.3
20
20
(c 1.2 in CHCl3}. The spectroscopic data agree with that of the
literature, except for the 31P NMR spectrum (162 MHz, CDCl3),
31a
=
which showed a resonance at 23.37 (P O, 18%) (lit. 23.1) with
a shoulder at 23.42 (P 18O, 82%; isotope induced satellite signal);
=
n
max/Si, film/cm-1 3384, 3081, 2941, 1464, 1382, 1216, 1031; dH
(400.1 MHz; CDCl3) 4.27 (1 H, m, PCCH), 4.15 (4 H, sext, J 7.0,
2 ¥ OCH2), 3.57 (1 H, t, J 6.0, CHP), 2.50 (1 H, bs, OH), 1.30 (3
H, d, J 6.0, CH3), 1.28 and 1.27 (each 3 H, t, J 7.0, OCCH3), 1.10–
0.99 (21 H, m, TIPS); dC (100.6 MHz; CDCl3) 72.9 (d, J 162.9),
68.3 (d, J 5.4), 62.9 (d, J 6.9), 62.4 (d, J 6.9), 21.4 (d, J 6.1), 18.1
(R)-(+)-Isobutyl 2-(triisopropylsilyloxy)propionate [(R)-36].
iPr3SiCl (4.71 g 24.0 mmol, 5.32 mL) was added dropwise
to a stirred solution of (R)-(+)-isobutyl 2-hydroxypropionate
(2.92 g, 20.0 mmol) and imidazole (3.00 g, 44.0 mmol) in dry
DMF (15 mL) at 0 ◦C under argon. The reaction mixture was
stirred for 78 h at ambient and was then poured into a mixture
of Et2O (50 mL) and water (50 mL). The organic phase was
separated. The aqueous layer was extracted with Et2O (3 ¥
25 mL). The combined organic layers were washed with 1 N HCl
(50 mL) and a saturated NaHCO3 solution, dried (MgSO4) and
concentrated under reduced pressure. The residue was purified
by flash chromatography (hexanes–acetone 4 : 1, Rf 0.20) to give
silylated ester (R)-36 (5.70 g, 74%) as a colourless oil (Found: C,
64.0, H, 11.0. C23H32O3Si requires C, 63.5, H, 11.3%); [a]D20 +24.4
(c 1.80 in acetone); nmax/NaCl, film/cm-1 2944, 2868, 1759, 1734,
1460, 1305, 1277, 1147; dH (400.1 MHz; CDCl3) 4.41 (1 H, q, J
6.9, CHO), 3.88 (AB part of an ABX system, JAB = JBX = 6.9,
JAB 10.5, CO2CH2), 1.93 (1 H, nonett, J 6.9, CH), 1.41 (d, J 6.9,
CH3), 1.15–0.98 (21 H, m, TIPS), 0.92 (6 H, d, J 6.9, C(CH3)2);
dC (100.6 MHz; CDCl3) 174.3, 70.8, 68.5, 27.7, 21.9, 19.0, 17.9
and 17.8 (6 C), 12.1 (3 C).
and 18.0 (2 s, each 3 C), 16.4 (d, J 5.4), 12.7 (3 C); dP (162 MHz;
18
CDCl3) 23.37 (P O) and 23.42 (P O ); 84% 18O [FD-MS: m/z:
=
=
369 (19%), 371 (100%)].
(1S,2R)-38.
n
max/Si, film/cm-1 3420, 2940, 2865, 1463, 1382,
1221, 1092, 1078, 1026; dH (400.1 MHz; CDCl3) 4.27 (1 H, m,
PCCH), 4.13 (4 H, m, 2 ¥ OCH2), 3.97 (1 H, dd, J 10.5, 3.0,
PCH), 2.45 (1 H, bs, OH), 1.29 (3 H, d, J 7.0, CH3), 1.28 (6 H, t, J
6.5, 2 ¥ OCCH3), 1.00 (21 H, m, iPr3Si); dC (100.6 MHz; CDCl3)
72.8 (d, J 161.4), 68.5 (d, J 8.4), 62.6 (d, J 6.9), 62.4 (d, J 6.9),
19.0 and 18.5 (6 C), 17.7 (d, J 2.3), 16.5 (d, J 5.4), 16.4 (d, J 5.4),
12.3 (3 C).
(1R,2R)-Diethyl 1,2-dihydroxypropyl-[P 18O]phosphonate [(1R,
=
2R)-39]. A mixture of (1R,2R)-2-(triisopropylsilyloxy)-propyl-
[P 18O]phosphonate (1.11 g, 3.01 mmol), glacial acetic acid
=
(16 mL) and water (4 mL) was stirred for 16 h at 90 ◦C (TLC:
EtOAc-MeOH 20 : 1, Rf 0.36). Volatile components were re-
moved under reduced pressure and the residue was purified
by flash chromatography (EtOAc-MeOH 20 : 1) to give 1,2-
dihydroxyphosphonate (1R,2R)-39 (0.39 g, 62%) as a colourless
(1R,2R)
and
=
(1S,2R)-Diethyl
2-triisopropylsilyloxy-1-
hydroxypropyl[P 18O]phosphonate [(1R,2R)- and (1S,2R)-38].
DIBAH (7.8 mL, 7.8 mmol, 1 M solution in n-heptane) was
added dropwise to a stirred solution of silylated ester (R)-36
20
oil; [a]D -12.7 (c 2.60 in acetone); dH (400.1 MHz; CDCl3) 4.13
(5 H, m, 2 ¥ OCH2 and CHCP), 3.60 (1 H, dd, J 7.0, 3.0, PCH),
3.31 (2 H, bs, 2 ¥ OH), 1.29 (3 H, t, J 7.0, CH3), 1.28 (3 H, t, J
7.0, CH3), 1.24 (3 H, dd, J 6.5, 1.5, CH3); dC (100.6 MHz; CDCl3)
71.6 (d, J 159.1), 66.4 (d, J 2.3), 63.4 (d, J 6.9), 62.6 (d, J 7.7),
19.1 (d, J 10.7), 16.5 and 16.4 (2 d, J 5.3); dP (162 MHz; CDCl3)
◦
(2.69 g, 7. 0 mmol) in dry Et2O (25 mL) at -78 C under argon.
The reaction mixture was stirred for 2 h at -78 ◦C. Water (1.5 mL)
was added and the mixture was stirred for 0.5 h at 0 ◦C. The
aluminium hydroxide was removed by filtration through Celite
and washed with Et2O (50 mL). The combined solutions were
dried (MgSO4) and concentrated under reduced pressure. The
residue dissolved in dry toluene, concentrated under reduced
pressure and dried (0.50 mm/1 h) to yield the crude aldehyde
(R)-37 (1.61 g, quantitatively) as an oil.31
16
18
24.66 (P= O) with shoulder at 24.70 (P= O); 81% 18O [FD-MS:
m/z:213 (23%), 215 (100%)].
(1R,2R)-(1,2)-Dihydroxypropyl-[18O]phosphonic acid, ammo-
nium salt {[P18O](1R,2R)-6}. Under an argon atmosphere
TMSBr (3.55 g, 23.30 mmol, 3.06 mL) was added to a stirred
Diethyl 18O-trimethylsilyl phosphite (1.78 g, 8.40 mmol) dis-
solved in dry toluene (5 mL) was added dropwise to the stirred
solution of the crude aldehyde (1.61 g, 7.00 mmol) in dry toluene
(20 ml) at -78 ◦C under argon. The reaction mixture was stirred for
solution of (1R,2R)-1,2-dihydroxypropyl-[P 18O]phosphonate
=
(1R,2R)-39 (0.39 g, 1.85 mmol) and allyltrimethylsilane (1.07 g,
9.40 mmol, 1.50 mL) in dry 1,2-dichloroethane (7.5 mL). The
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1944–1953 | 1951
©

View Article Online
reaction mixture was stirred at RT for 16 h. Volatile components
were removed under reduced pressure (RT to 40 ◦C/0.5 mm) and
the residue was dissolved in a mixture (10 mL) of ethanol–water
1 :1 and concentrated again. The residue was dissolved in water
(10 mL) and conc. ammonia (1.5 mL) was added. Lyophilisation
gave the phosphonic acid [P18O](1R,2R)-6 as ammonium salt
(0.23 g). Its NMR spectra recorded in D2O were identical to the
ones of (1R,2R)-6 as ammonium salt, derived from fosfomycin.
The 31P NMR spectrum did not show an isotope-induced satellite
signal.
Protein assay
Protein concentration was determined by the method38 of Brad-
ford using bovine serum albumin as standard.
Acknowledgements
This work was supported by grant no. P11929-CHE from
FWF. We are grateful to S. Felsinger for recording the NMR
spectra, Sandoz GmbH (Kundl) for Fosfomycin Sandoz.ꢀ
R
Microorganism and culture conditions. Rhizobium huakuii
PMY1 was isolated as previously described.10 Cells were grown
in batch cultures at 30 ◦C on an orbital shaker at 100 rpm in
minimal medium (pH 7.0) of the following composition: KCl,
0.20 g; MgSO4·7H2O, 0.20 g; CaCl2·2H2O, 1.0 mg; NH4Cl, 1.0 g;
ferric ammonium citrate, 1.0 mg; phosphate free yeast extract10,
0.05 g; BME essential amino acids solution, 20 mL/L (Sigma),
and 1 mL each of trace element solution10 and vitamin solution10
in 1 L. Either filter-sterilised (0.22 mm) fosfomycin (5 mM) or those
organophosphonates under test as potential pathway intermedi-
ates were added as sole carbon and phosphorus source. Microbial
growth was measured by the increase in absorbance at 650 nm
using a ATI-Unicam PU 8625 UV/VIS spectrophotometer while
phosphate release into the culture supernatant was monitored via
the method36 of Fiske and SubbaRow.
References
1 For a review see: H. Seto and T. Kuzuyama, Nat. Prod. Rep., 1999, 16,
589–596.
2 S. T. Dyhrman, P. D. Chappell, S. T. Haley, J. W. Moffett, E. D. Orchard,
J. B. Waterbury and E. A. Webb, Nature, 2006, 439, 68–71.
3 (a) D. Hendlin, E. O. Stapley, M. Jackson, H. Wallick, A. K. Miller,
F. J. Wolf, T. W. Miller, L. Chaiet, F. M. Kahan, E. L. Foltz, H. B.
Woodruff, J. M. Mata, S. Hernandez and S. Mochales, Science, 1969,
166, 122–123; (b) F. M. Kahan, J. S. Kahan, P. J. Cassidy and H. Kropp,
Ann. N. Y. Acad. Sci., 1974, 235, 364–386.
4 E. Bayer, K. H. Gugel, K. Ha¨gele, H. Hagenmaier, S. Jessipow, W. A.
Ko¨nig and H. Za¨hner, Helv. Chim. Acta, 1972, 55, 224–239.
5 (a) Y. Mine, T. Kamimura, S. Nonoyama and M. Nishida, J. Antibiot.,
1980, 33, 36–43; (b) M. A. Missinou, S. Borrmann, A. Schindler, S.
Issifou, A. A. Adegnika, P.-B. Matsiegui, R. Binder, B. Lell, J. Wiesner,
T. Baranek, H. Jomaa and P. G. Kremsner, Lancet, 2002, 360, 1941.
6 J. E. Franz, M. K. Mao, J. A. Sikorski, Glyphosate: A Unique
Global Herbicide, ACS Monograph 189, American Chemical Society,
Washington, DC, 1997.
7 (a) For reviews see: J. P. Quinn, A. N. Kulakova, N. A. Cooley and J. W.
McGrath, Environ. Microbiol., 2007, 9, 2392–2400; (b) N. G. Ternan,
J. W. McGrath, G. McMullan and J. P. Quinn, World J. Microbiol. &
Biotechnol., 1998, 14, 635–647.
8 (a) B. L. Wanner, Biodegradation, 1994, 5, 175–184; (b) M. Kertesz, A.
Elgorriaga and N. Amrhein, Biodegradation, 1991, 2, 53–59; (c) J. P.
Quinn, J. M. M. Peden and R. E. Dick, Appl. Microbiol. Biotechnol.,
1989, 31, 283–287; (d) J. P. Quinn, Lett. Appl. Microbiol., 1989, 8, 113–
116; (e) L. P. Wackett, S. L. Shames, C. P. Venditti and C. T. Walsh,
J. Bacteriol., 1987, 169, 710–717.
9 (a) S.-L. Lee, T. W. Hepburn, W. H. Swartz, H. L. Ammon, P. S. Mariano
and D. Dunaway-Mariano, J. Am. Chem. Soc., 1992, 114, 7346–7354;
(b) A.-M. Lacoste, C. Dumora and A. Cassaigne, Biochem. (Life Sci.
Adv.), 1989, 8, 97–111; (c) N. G. Ternan and J. P. Quinn, Syst. Appl.
Microbiol., 1998, 21, 346–352.
10 (a) J. W. McGrath, G. B. Wisdom, G. McMullan, M. J. Larkin and J. P.
Quinn, Eur. J. Biochem., 1995, 234, 225–230; (b) J. W. McGrath, A. N.
Kulakova and J. P. Quinn, J. Appl. Microbiol., 1999, 86, 834–840.
11 (a) A. N. Kulakova, G. B. Wisdom, L. A. Kulakov and J. P. Quinn,
J. Biol. Chem., 2003, 278, 23426–23431; (b) C. C. H. Chen, Y. Han,
W. L. Niu, A. N. Kulakova, A. Howard, J. P. Quinn, D. Dunaway-
Mariano and O. Herzberg, Biochemistry, 2006, 45, 11491–11504.
12 (a) R. D. Woodyer, G. Li, H. Zhao and W. A. van der Donk, Chem.
Commun., 2007, 359–361; (b) F. Yan, S.-J. Moon, P. Liu, Z. Zhao, J. D.
Lipscomb, A. Liu and H.-w. Liu, Biochemistry, 2007, 46, 12628–12638;
(c) F. Hammerschmidt, J. Chem. Soc., Perkin Trans. I, 1991, 1993–
1996; (d) A. Schweifer and F. Hammerschmidt, Bioorg. Med. Chem.
Lett., 2008, 18, 3056–3059.
Enzyme asssays
(i) In vitro C–P bond cleavage activity. Cell-extracts of
fosfomycin-grown cells were prepared by sonication as previ◦ously
described.10 C–P bond cleavage activity was assayed at 30 C by
measuring the amount of phosphate liberated from the supplied
organophosphate substrate. Unless otherwise stated the assay
mixture (1 mL) consisted of: 50 mM Tris-HCl buffer (pH 7.0) and
5 mM organophosphonate substrate. The reaction was initiated
by the addition of approximately 5 mg/mL of cell-free extract
from fosfomycin grown cells, incubated for 60 min and terminated
by the addition of 200 mL of 50% (w/v) trichloroacetic acid.
After centrifugation, the phosphate released was measured via
the method36 of Fiske and SubbaRow. Activity was expressed as
the mean of three replicates, in units of nanomoles of phosphate
liberated per minute per milligram of protein.
(ii) Phosphite oxidase. Phosphite oxidase activity was as-
sayed via the method37 of Malacinski in cell-extracts of fosfomycin
grown cells of R. huakuii PMY1. The assay mixture contained
the following (l mL); 450 mL of 50 mM Tris-HCl buffer (pH 7.5);
300 mL of 10 mM NAD+ and 200 mL of either 50 mM phosphorous
acid or sodium phosphite. The reaction was initiated by the
addition of approximately 5 mg/mL of crude cell-extract and the
mixture incubatedfor 60min at30 ◦C. The reaction was terminated
by the addition of 200 mL of 50% (w/v) trichloroacetic acid and
the phosphate produced measured via the method36 of Fiske and
SubbaRow.
13 J. W. McGrath, F. Hammerschmidt and J. P. Quinn, Appl. Environ.
Microbiol., 1998, 64, 356–358.
14 Y. Vidal, B. Clin, A. Cassaigne and E. Neuzil, Bull. Soc. Pharm.
Bordeaux, 1982, 121, 1–11.
15 F. Hammerschmidt, Liebigs Ann. Chem., 1991, 469–475.
16 J. M. Seco, E. Quinoa´ and R. Riguera, Chem. Rev., 2004, 104, 17–117.
17 F. Hammerschmidt, G. Bovermann and K. Bayer, Liebigs Ann. Chem.,
1990, 1055–1061.
18 N. N. Girotra and N. L. Wendler, Tetrahedron Lett., 1969, 10, 4647–
For all enzyme assays, controls containing no crude cell-extract
were prepared to evaluate the degree of background phosphate
release.
4649.
19 T. Skarzynski, A. Mistry, A. Wonacott, S. E. Hutchinson, V. A. Kelly
and K. Duncan, Structure, 1996, 4, 1465–1474.
1952 | Org. Biomol. Chem., 2009, 7, 1944–1953
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
20 B. A. Bernat, L. T. Laughlin and R. N. Armstrong, J. Org. Chem., 1998,
63, 3778–3380.
21 F. Hammerschmidt, Monatsh. Chem., 1991, 122, 389–398.
22 F. Hammerschmidt, G. Bovermann and K. Bayer, Liebigs Ann. Chem.,
1990, 1055–1061.
23 F. Hammerschmidt and Y.-F. Li, Tetrahedron, 1994, 50, 10253–10264.
24 (a) F. Hammerschmidt and H. Vo¨llenkle, Liebigs Ann. Chem., 1989,
577–583; (b) for monodeuterated species of ( )-19 see: F. Hammer-
schmidt and H. Ka¨hlig, J. Org. Chem., 1991, 56, 2364–2370.
25 A. N. Pudovik and I. V. Konovalova, Synthesis, 1979, 81–96.
26 (a) V. S. Abramov, L. P. Semenova and L. G. Semenova, Dokl. Akad.
Nauk SSR, 1952, 84, 281–284[Chem. Abstr., 1953, 47, 3227i]; (b) L.
Horner and H. Ro¨der, Chem. Ber, 1970, 103, 2984–2986.
27 K. Sasse, in Houben-Weyl, 4th edn, XII/1, Georg Thieme Verlag,
Stuttgart, 1963, pp. 453–458.
29 J. M. Vrtis, A. K. White, W. W. Metcalf and W. A. van der Donk, J. Am.
Chem. Soc., 2001, 123, 2672–2673.
30 For a review see: C. Ru¨cker, Chem. Rev., 1995, 95, 1009–1064.
31 (a) A. Bongini, R. Camerini and M. Panunzio, Tetrahedron: Asymme-
try, 1996, 7, 1467–1476; (b) G. Cainelli, M. Panunzio, E. Bandini,
G. Martelli and G. Spunta, Tetrahedron, 1996, 52, 1685–
1698.
32 X. Creary, C. C. Geiger and K. Hilton, J. Am. Chem. Soc., 1983, 105,
2851–1858.
33 W. J. Stec, N. Goddard and J. R. Van Wazer, J. Phys. Chem., 1971, 75,
3547–3549.
34 W. Mikenda, Vibrational Spectrosc., 1992, 3, 327–330.
35 N. Shieh and C. C. Price, J. Org. Chem., 1959, 24, 1169.
36 C. H. Fiske and Y. SubbaRow, J. Biol. Chem., 1925, 66, 375–400.
37 G. M. Malacinski and W. A. Konetzka, J. Bacteriol., 1967, 93, 1906–
1910.
28 (a) W. W. Metcalf and R. W. Wolfe, J. Bacteriol., 1998, 21, 5547–5558;
(b) W. Buckel, Angew. Chem. Int. Ed., 2001, 40, 1417–1418.
38 M. M. Bradford, Anal. Biochem., 1976, 72, 248–254.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1944–1953 | 1953
©