Synthesis of enantiomericalty pure 1,5,5-trideuterated cis- and trans-2,4-dioxa-3-phosphadecalins. 31P-NMR evidence of covalent-bond formation and the stereochemical implications in the course of the inhibition of δ-chymotrypsin

Article abstract of DOI:10.1002/hlca.200790215

The irreversible inhibition of δ5-chymotrypsin with the enantiomerically pure, P(3)-axially and P(3)-equatorially X-substituted cis- and trans-configurated 2,4-dioxa-3-phospha(1,5,5-²H₃) bicyclo[4.4.0]-decane 3-oxides (X = F, 2,4-din

Full text of DOI:10.1002/hlca.200790215

2058
Helvetica Chimica Acta – Vol. 90 (2007)
Synthesis of Enantiomerically Pure 1,5,5-Trideuterated cis- and trans-2,4-
Dioxa-3-phosphadecalins. 31P-NMR Evidence of Covalent-Bond Formation
and the Stereochemical Implications in the Course of the Inhibition of
d-Chymotrypsin
by Markus J. Stçckli and Peter Rüedi*
Organisch-chemisches Institut der Universität Zürich, Winterthurerstrasse 190, CH-8057 Zürich
(phone: þ 41446354241)
The irreversible inhibition of d-chymotrypsin with the enantiomerically pure, P(3)-axially and P(3)-
equatorially X-substituted cis- and trans-configurated 2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]-
1
decane 3-oxides (X ¼ F, 2,4-dinitrophenoxy) was monitored by 31P-NMR spectroscopy. H-Correlated
31P{2H}-NMR spectra enabled the direct observation of the vicinal coupling (3J) between the P-atom of
the inhibitor and the CH2O moiety of Ser195 (¼ꢀSer195ꢁ(CH2O)), thus establishing the covalent nature of
the ꢀSer195ꢁ(CH2OÀP) bond in the inhibited enzyme. The stereochemical course of the phosphorylation is
dependent on the structure of the inhibitor, and neat inversion, both inversion and retention, as well as
neat retention of the configuration at the P-atom was found.
1. Introduction. – Earlier 31P-NMR spectroscopic investigations on the irreversible
inhibition of d-chymotrypsin with the optically active, P(3)-axially and P(3)-
equatorially substituted cis- and trans-configurated 3-(2,4-dinitrophenoxy)-2,4-dioxa-
3-phosphadecalin
3-oxides
(¼ 3-(2,4-dinitrophenoxy)-2,4-dioxa-3-phosphabicy-
clo[4.4.0]decane 3-oxides ¼ 2-(2,4-dinitrophenoxy)hexahydro-4H-1,3,2-benzodioxa-
phosphorin 2-oxides) have shown that the stereochemical course of the inhibition
reaction is dependent on the substrate [1][2]1). The general concept is summarized in
Scheme 1, and the key spectroscopic features for the assignment of the configuration at
the P-atom are outlined in Scheme 2 (left)2). Accordingly, the existence of covalent
OÀP bonds between Ser195 of the enzyme and the inhibitors, i.e., ꢀSer195ꢁ (CH2OÀP),
has been anticipated due to the similarity of the chemical shifts of the phosphorylated
enzyme sample with those of model compounds. This most probable but not fully
unambiguous assumption has recently been verified by a preliminary experiment based
on the concept depicted in Scheme 2 [5]3): The direct evidence of a covalent bond is the
1
)
)
For a full account on the background of the project, its context, and pertinent references, see [1][3].
The magnitude of the vicinal 3J(P,H) is the key argument in assigning the conformation of the
heterocyclic ring and, as a consequence, the configuration at the P-atom (Scheme 2) (see [1][2][4]
and refs. cit. therein). In addition, the 31P-NMR resonance of the axial P(3)-epimers is
diamagnetically shifted with respect to their equatorially substituted counterparts.
2
3
)
Although the concept has been presented [5], it is recapitulated for the detailed, concluding report
on that subject. Notice that for convenience, the implied CH2O moiety of 195Ser of the enzyme, i.e.,
ꢀSer195ꢁ(CH2O), is represented in Figures and Schemes by Ser195O, Ser195ÀOCH2, or CTÀSer195
(CT¼ d-chymotrypsin).
O
ꢂ 2007 Verlag Helvetica Chimica Acta AG, Zürich

Helvetica Chimica Acta – Vol. 90 (2007)
2059
Scheme 1
Scheme 2
existence of a vicinal coupling (3J) between the P-atom of the inhibitor and the
1
ꢀSer195ꢁ(CH2O) of the enzyme. In principle, this coupling can be observed in the H-
coupled 31P-NMR spectra, but the splitting caused a priori by HÀC(1) and CH2(5) of
the oxaphosphadecalin interferes and makes an authentic monitoring difficult. Hence,
the crucial 3J(P,H) can be detected reliably only when all couplings with the 1H-atoms
2
of the inhibitor are supressed. This is accomplished by replacing them by H-atoms.
2
31
However, since H also couples with the P-atom4), 31P{2H}-NMR spectra have to be
recorded5) to obtain unequivocal information. This prerequisite is unfavorable with
respect to the expected intensities of the resonances because coupling of 31P with H
1
and quadrupol relaxation (2H) significantly decrease the intensity of the signals.
4
)
This holds at least for the 1D spectra. Sophisticated 2D experiments would allow to avoid that
particular problem, but as the d values of HÀC(1) and CH2(5) in the 1H-isomers of 9 – 12 are close
to the ꢀSer195ꢁ(CH2O), recording 31P{2H}-spectra was considered to be the most straightforward
approach. In general, coupling with nuclei of different quantum numbers causes complex splitting
patterns which render the interpretation difficult. Although 3J(P,2H) is usually negligibly small and
would only contribute to line broadening, it becomes significant in the present case
(3J(P,HeqÀC(5)) ꢀ 25Hz and the expected 3J(P,2HeqÀC(5)) ꢀ 3.5Hz).
2
5
)
Since the H-channel is usually the lock signal, this requirement is not trivial.

2060
Helvetica Chimica Acta – Vol. 90 (2007)
The preliminary experiment was performed with the axially P-substituted (Æ)-
trans-3-(2,4-dinitrophenoxy)-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-ox-
ide ((Æ)-9a) [5]. It demonstrated the feasibility of the concept and confirmed the
stereochemical course in terms of inversion of the configuration at the P-atom as found
[1]. However, due to the application of racemic inhibitors, two diastereoisomeric
phosphorylated enzyme samples are generated a priori, a fact that further decreases the
sensitivity of the experiment and complicates its interpretation. To increase the
sensitivity and, in particular, to improve the significance of the method, we decided to
investigate enantiomerically pure 1,5,5-trideuterated inhibitors. In addition to the 3-
(2,4-dinitrophenoxy)(1,5,5-2H3) compounds 9 and 10 (cf. Scheme 4), also the corre-
sponding cyclic (1,5,5-2H3)-3-fluoridates 11 and 12 (cf. Scheme 4) were prepared and
investigated.
2. Synthesis and Characterization of the Deuterated Phosphadecalins. – 2.1. Starting
(2H3)Alcohols 4 and 5. The enantiomerically pure diols (þ)- and (À)-4, and (þ)- and
(À)-5, resp.), were obtained after reduction of ethyl 2-oxocyclohexanecarboxylate (1)
with NaBD4 and chromatographic separation of the resulting ethyl (Æ)-trans- and (Æ)-
cis-2-hydroxy(2-2H1)cyclohexanecarboxylates ((Æ)-2 and (Æ)-3, resp.; Scheme 3).
After reduction of the hydroxy esters (Æ)-2 and (Æ)-3 with LiAlD4, the trans-diol
(Æ)-4 was esterified with (À)-(1S)-camphanoyl chloride to yield the mixture (ca. 1 :1)
of the diastereoisomeric bis-camphanates 6/76), and the cis-diol (Æ)-5 was transformed
into the bis-4-bromobenzoate (Æ)-8 (¼ 8/ent-8). Prep. HPLC (Chiralcelꢀ OD) of 6/7
afforded 6 and 7 (de > 99%), and the same procedure with (Æ)-8 gave 8 and ent-8
(ee > 99%). Saponification of 6 and 7 afforded the trans-configurated (þ)-(1R,2S)- and
(À)-(1S,2R)-2-hydroxy(2-2H1)cyclohexane(a,a-2H2)methanols ((þ)-4 and (À)-4, resp.;
ee > 99%). Saponification of 8 and ent-8 gave the cis-configurated (þ)-(1S,2S)- and
(À)-(1R,2R)-2-hydroxy(2-2H1)cyclohexane(a,a-2H2)methanols ((þ)-5 and (À)-5, resp.,
ee > 99%).
The absolute configurations of the (2H3)diols were established independently with
respect to the 1H-isomers7) by X-ray crystallographic analyses. In the case of the trans-
bis-camphanates 6 and 7, the absolute configuration of the parent diols (þ)- and (À)-4,
respectively, was inferred via the (1S)-camphanoyl moiety, whereas the cis-bis-4-
6
)
It is remarkable that diastereoisomer 7 (1S,1’S,2’R) partly crystallized (de > 99%), whereas its
enantiomer ent-7 (1R,1’R,2’S), derivatized with (þ)-(1R)-camphanoyl chloride, did not crystallize
from the mixture (see Exper. Part). This unexpected fact made a simple separation of the
enantiomers impossible.
The preparation and characterization of the optically active 1H-isomeric trans- and cis-diols was
reported earlier [1][2]. The procedures and the assignments of the absolute configurations had been
performed according to [6] despite several inconsistencies, mainly concerning the magnitude of the
optical rotations. The absolute configurations were unambiguously confirmed by X-ray crystal-
lographic analyses of the corresponding phosphoramidates that had been obtained after
derivatization of the optically active cis- and trans-configurated 3-chloro-2,4-dioxa-3-phosphabicy-
clo[4.4.0]decane 3-oxides with (þ)-(R)- and (À)-(S)-(1-phenylethyl)amine [7].
7
)

Helvetica Chimica Acta – Vol. 90 (2007)
2061
Scheme 3
a) NaBD4, EtOH, reflux (cis/trans ca. 3 :2 from 1). b) CC (SiO2, hexane/Et2O). c) LiAlD4, Et2O, reflux.
d) (À)-(1S)-Camphanoyl chloride, N,N-dimethylpyridin-4-amine (DMAP), pyridine, reflux. e) Prep.
HPLC (Chiralcelꢃ OD, hexane/MeOH/EtOH). f) LiAlH4, THF, 08 ! r.t. g) CC (SiO2, CH2Cl2, AcOEt).
h) 4-BrC6H6COCl, DMAP, pyridine, reflux. i) Prep. HPLC (Chiralcelꢀ OD, hexane/EtOH).
8
)
According to Chem. Abstr., the diols are substituted methanols. The IUPAC numbering system is
different as it considers the same compounds as substituted cyclohexanols.

2062
Helvetica Chimica Acta – Vol. 90 (2007)
bromobenzoates 8 and ent-8 enabled a direct determination due to the presence of the
Br-atoms9). It is worth mentioning that the signs of the optical rotations and the
absolute configurations of the (2H3)diols and the 1H-isomers are coincident.
2.2. 3-Substituted 2,4-Dioxa-3-phosphadecalins 9 – 12, 14, and 15. The trans-2,4-
dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-oxides 9a, 9b, 11a, and 11b
(Scheme 4) were prepared from (þ)- or (À)-4 by reaction with the appropriate
phosphorus reagent (2,4-dinitrophenyl phosphorodichloridate for 9 and POFCl2 for 11)
and chromatographic separation of the resulting P(3)-epimer mixture (axial/equatorial
ca. 1:1) into the pure axial (9a and 11a) and equatorial epimers (9b and 11b). Similarly,
starting from (þ)- or (À)-5, the 3-substituted cis-2,4-dioxa-3-phospha(1,5,5-2H3)bicy-
clo[4.4.0]decane 3-oxides 10a, 10b, 12a, and 12b were obtained (Scheme 4). With
1
respect to the H-isomers [1][2][8], the sign of the optical rotations of all optically
active 3-phospha(1,5,5-2H3)decalins remained unchanged, only the absolute values
varied to some extent.
The trans- and cis-, axially and equatorially 3-substituted phosphoserine model
compounds 14a/14a’ and 14b/14b’ (trans), and 15a/15a’ and 15b/15b’ (cis) were
prepared from (Æ)-4 and (Æ)-5, respectively, by reaction with in situ generated N-
[(benzyloxy)carbonyl]-O-(dichlorophosphinyl)-l-serine methyl ester (13) and ob-
tained as a mixture of diastereoisomers (ca. 1:1) after chromatographic separation
(Scheme 5).
The NMR data of the 2,4-dioxa-3-phospha(1,5,5-2H3)decalins 9 – 12, 14, and 15 fully
1
confirmed their structures (see Exper. Part). Significant H- and 13C-NMR character-
istics are both the lack of the ABX-P-system in the heterocyclic moiety, and, due to
1J(C,2H) and 2J(C,P), the complex m of C(1) and C(5) in 9 – 12 and of C(1’) and C(5’) in
14 and 15, which hardly exceed the noise. Compared to the 1H-isomers [1][2][8], d(31P)
remained almost unchanged, the axially substituted epimers resonating at higher
field with respect to the equatorially substituted counterparts. The decisive spec-
tral feature is provided by the 31P{2H}-NMR spectra (CDCl3) that exhibit a t at
d À 6.66 (3J(P,CH2(3)) ¼ 7.1 Hz ) 10) for 14a/14a’ and two t at d À 4.11 and À 4.13
10
1
(3J(P,CH2(3)) ¼ 6.5Hz) ) for 14b/14b’. The H-correlated 31P{2H}-NMR spectra, as
exemplified for the trans-isomers (Fig. 1), clearly showed the respective cross-peaks
with HAÀC(3) and HBÀC(3) of the serine moiety at d 4.50 and 4.39 (14a/14a’) and at d
4
4.51 and 4.37 (14b/14b’), and also indicated a J(P,HÀC(2)) coupling at d 4.57 in the
case of 14b/14b’. The same features hold for the cis-configurated model compounds
15a/15a’ (d À 6.49 (m, w1/2 ꢀ 22 Hz)) and 15b/15b’ (d À 3.93 (m, w1/2 ꢀ 20 Hz))11)
4
(Scheme 5). In addition, significant J(P,CH2(10)) couplings are observed at d 1.34 –
1.43 (15a/15a’) and at d 1.52 (15b/15b’). To compare the chemical shifts with the
conditions of the enzyme experiments, the spectra of the model compounds were also
9
)
The X-ray crystallographic analyses were performed by PD Dr. A. Linden, X-ray department of our
institute. CCDC-650295 – CCDC-650298 contain the full crystallographic data for the compounds 6,
7, 8, and ent-8. These data can be obtained free of charge from the Cambridge Crystallographic Data
Centre, via http://www.ccdc.cam.ac.uk/data_request/cif.
10
)
)
The signal is accidentally a t as HAÀ and HBÀC(3) are diastereotopic.
11
Contrary to 14a/14a’ and 14b/14b’, the cis-configurated model compounds 15a/15a’ and 15b/15b’
displayed only marginally resolved 31P{2H}-NMR spectra.

Helvetica Chimica Acta – Vol. 90 (2007)
2063

2064
Helvetica Chimica Acta – Vol. 90 (2007)

Helvetica Chimica Acta – Vol. 90 (2007)
2065
Fig. 1. 1H-Correlated 31P{2H}-NMR spectra (CDCl3, 242.9 MHz, 278) of the model compounds a) 14a/
14a’ and b) 14b/14b’
recorded in CD3CN/D2O/0.2m Tris (pH 7.8) 11:44 :45. The respective d(31P) are À 5.12
(14a/14a’), À 4.10 (14b/14b’), À 5.38 (15a/15a’), and À 3.77 and À 3.95( 15b/15b’)
(Scheme 5).
3. Results of the 31P-NMR Investigations. – 3.1. General. In analogy to the
preceding setups [1][2], the experiments were designed so that d-chymotrypsin and the
inhibitors 9 – 12 reacted in stoichiometric amounts (ca. 2 mmol in Tris buffer at pH 7.8
and 278). For the reason of the expected low sensitivity, the progression of the
inhibition reaction was not spectroscopically tracked as in [1], and the NMR spectra
were recorded only after complete inhibition had taken place. As observed in the
preliminary experiment [5], the resolution of the 1D 31P-NMR spectra was too low to
3
reliably evidence the crucial J(P,H) splitting. The existence of covalent bonds was
1
decisively demonstrated by the 2D H-correlated 31P{2H}-NMR spectra that reveal
1
cross-peaks of the P-atoms of the inhibitors with the H-resonances at d ca. 4.1 – 4.5.
According to the 1H-NMR chemical shifts of the model compounds 14a/14a’, 14b/14b’,
15a/15a’, and 15b/15b’, this signal group is allocated to the ꢀSer195ꢁ(CH2O). Despite the
fact that the 31P-NMR resonances did not very well agree with those of the model
compounds, they were consistent within a series, and all the conclusions could be drawn
unambiguously. In particular, the diamagnetic and the paramagnetic displacements
could always be distinguished and clearly assigned to the corresponding axially or
equatorially substituted phosphorylated enzyme species. This is well demonstrated in
the 3-fluoro series where the inhibition did not proceed stereospecifically in most cases.
But the respective Dd values were always consistent.
3.2. 3-(2,4-Dinitrophenoxy) Inhibitors 9 and 10. The 2D 31P, 1H-NMR spectra of the
irreversible inhibition of d-chymotrypsin with the enantiomerically pure trans- and cis-
configurated, axially and equatorially 3-substituted 3-(2,4-dinitrophenoxy) derivatives
9 and 10 are compiled in the Figs. 2 – 4, and the interpretation of the stereochemical
course is depicted in Schemes 6 – 8. After reaction of the enzyme with the trans-
configurated compounds (þ)- and (À)-9a and (þ)- and (À)-9b, and with the cis-
configurated (À)-10b, distinct NMR signals were detected, but no signals could be

2066
Helvetica Chimica Acta – Vol. 90 (2007)
Fig. 2. 1H-Correlated 31P{2H}-NMR spectra (242.9 MHz, 278) of the inhibition of d-chymotrypsin with the
trans-axial 3-(2,4-dinitrophenoxy)-3-phosphadecalins a) (þ)-9a and b) (À)-9a3). Solvent: 0.2m Tris
buffer, pH 7.8, D2O (44%), CD3CN (11%).
Fig. 3. 1H-Correlated 31P{2H}-NMR spectra (242.9 MHz, 278) of the inhibition of d-chymotrypsin with the
trans-equatorial 3-(2,4-dinitrophenoxy)-3-phosphadecalins a) (À)-9b and b) (þ)-9b3). Solvent: 0.2m Tris
buffer, pH 7.8, D2O (44%), CD3CN (11%).

Helvetica Chimica Acta – Vol. 90 (2007)
2067
Fig. 4. 1H-Correlated 31P{2H}-NMR spectrum (242.9 MHz, 278) of the inhibition of d-chymotrypsin with
the cis-equatorial 3-(2,4-dinitrophenoxy)-3-phosphadecalin (À)-10b3). Solvent: 0.2m Tris buffer, pH 7.8,
D2O (44%), CD3CN (11%).
Scheme 6

2068
Helvetica Chimica Acta – Vol. 90 (2007)
Scheme 7
Scheme 8
found when the enzyme was reacted with (þ)- and (À)-10a and (þ)-10b. This finding
confirms the earlier results obtained with the 1H-isomers [1][2]: It was shown
kinetically and by 31P-NMR spectroscopy that the four trans- and the (À)-cis-equatorial
3-(2,4-dinitrophenoxy)-3-phosphadecalins inhibited d-chymotrypsin irreversibly,
whereas the other cis-compounds turned out to be only weak reversible inhibitors.
Figs. 2 – 4 clearly demonstrate the reactions to proceed stereospecifically in terms of
neat retention or inversion of the configuration at the P-atom. Reaction of the trans-
configurated 3-axially substituted inhibitors (þ)- and (À)-9a with d-chymotrypsin gave
31P, 1H-correlated signals at d À 3.88 and À 3.95( Fig. 2). Comparison of the 31P-NMR
chemical shifts with those of the trans-configurated model compounds 14a/14a’
(d À 5.12) and 14b/14b’ (d À 4.10) shows that these resonances can be attributed to the
enzyme/inhibitor adducts in which the Ser195 moiety of d-chymotrypsin occupies the
equatorial position at the P-atom. Therefore, the diastereoisomeric structures 16b and
17b were assigned to the phosphorylated enzyme species, and, as a consequence, the
inhibition reaction proceeded with inversion (Scheme 6). The same product 16b
(d À 3.87) was detected after inhibition of the enzyme with the trans-configurated 3-

Helvetica Chimica Acta – Vol. 90 (2007)
2069
equatorially substituted inhibitor (À)-9b (Fig. 3,a). On the other hand, reaction of its
enantiomer (þ)-9b yielded a high-field correlated signal at d À 4.77 (Fig. 3,b),
characteristic for an axial substitution pattern at P(3). Therefore, structure 17a was
assigned to the inhibited enzyme. These findings are interpreted in terms of retention at
P(3) with (À)-9b and inversion with (þ)-9b to yield the reaction products 16b and 17a
(Scheme 7). Inhibition of d-chymotrypsin with the cis-configurated 3-equatorially
substituted (À)-10b (the only irreversible inhibitor in the cis-series) gave rise to a low-
field signal at d À 3.71 (Fig. 4), characteristic for equatorial substitution at P(3). In
accordance with the cis-configurated model compounds 15a/15a’ (d À 5.38) and 15b/
15b’ (d À 3.77 and À 3.95), structure 18b was deduced for the reaction product, and its
formation can be explained by retention of the configuration at P(3) (Scheme 8).
Although the chemical shifts are slightly different from those reported, the results
1
confirm the stereochemical course derived from the experiments with the H-isomers
[1][2]12). In particular, the trans-axial compounds (þ)- and (À)-9a react with inversion,
the trans-equatorial congeners (þ)- and (À)-9b with retention and inversion, and the
cis-equatorial one ((À)-10b) with retention of the configuration at P(3).
3.3. 3-Fluoro Inhibitors 11 and 12. The results of the reaction of d-chymotrypsin
with the enantiomerically pure 3-phospha-3-fluoro derivatives 11 and 12 are depicted in
the Figs. 5 – 7 and in Schemes 9 – 11. After reaction of the enzyme with the trans-
configurated compounds (þ)- and (À)-11a and (þ)- and (À)-11b, and with the cis-
configurated (þ)- and (À)-12a, the 31P-NMR spectra showed these compounds to be
Scheme 9
Generally, the d(31P) are shifted upfield by ca. 0.4 – 0.5ppm with respect to those reported for the
1H-isomers [1][2], but the respective relative differences between the P(3)-axially and P(3)-
equatorially substituted species are consistent. In particular, the reliability of the interpretations is
corroborated by the fact that the slower reacting enantiomer, i.e., (À)-9a, of the trans-axial
inhibitors gives rise to the paramagnetically shifted diastereoisomer, i.e., 17b, as described in [1].
12
)

2070
Helvetica Chimica Acta – Vol. 90 (2007)
Scheme 10
irreversible inhibitors (Figs. 5 – 7), whereas no signals could be detected when d-
chymotrypsin was treated with the cis-congeners (þ)- and (À)-12b. Figs. 5 and 6 reveal
that the reactions with the four trans-configurated inhibitors give each rise to a pair of
correlated peaks at d À 5.99 and À 5.13, À 5.89 and À 5.06, À 6.01 and À 5.15, and
À 5.92 and À 5.09, respectively. As discussed above, the diamagnetically shifted signals
are indicative of axial substitution at P(3) (!diastereoisomeric phosphorylated
enzyme species 16a and 17a) and the paramagnetically shifted ones for equatorial
substitution at P(3) (!diastereoisomers 16b and 17b)13). This outcome clearly shows
that the reactions do not proceed stereospecifically as both retention and inversion of
the configuration at the P-atom take place (Schemes 9 and 10). On the other hand,
13
)
Formally, the phosphorylated diastereoisomeric enzymes 16a, 16b, 17a, and 17b originating either
from the 2,4-dinitrophenoxy or the fluoro inhibitors (þ)- and (À)-9, or (þ)- and (À)-11, respectively
(Schemes 6 and 7) are depicted as identical species. Hence, the respective pairs are expected to
exhibit equal 31P-chemical-shift displacements. However, although all the experiments were run
under identical conditions, the compounds that are derived from the fluoro inhibitors 11 display
upfield shifts (Dd(31P) ca. 1.2) with respect to those from the 2,4-dinitrophenoxy inhibitors 9, a fact
that needs a comment. Besides a non-controlled periodic change of the characteristics of the NMR
probe-head, we definitely can rule out extrinsic factors that influence the chemical shift such as
referencing, temperature, pH, and concentration. Since the 31P-NMR resonances are consistent
within an inhibitor series (9 or 11), we conclude that the P-atoms in the respective inhibited enzyme
species are situated in a different environment. With regard to the characteristics of the leaving
groups, we suppose that the properties of the F-atom are significantly different from those of the
sterically and electronically demanding 2,4-dinitrophenoxy group, i.e., the conformation of the
active site and its environment are affected differently (actually in an unpredictable manner).
Moreover, due to the structural features surrounding the catalytic triade such as the oxy-anion hole,
further cationic and p-donor-type coordination sites [9], the 2,4-dinitrophenolate might remain
embedded in the active-site pocket. However, this idea could not be corroborated by simple
experiments with the model compounds, as d(31P) is invariable after adding various concentrations
of 2,4-dinitrophenolate.

Helvetica Chimica Acta – Vol. 90 (2007)
2071
Scheme 11
Fig. 5. 1H-Correlated 31P{2H}-NMR spectra (242.9 MHz, 278) of the inhibition of d-chymotrypsin with the
trans-axial 3-fluoro-3-phosphadecalins a) (À)-11a and b) (þ)-11a3). Solvent: 0.2m Tris buffer, pH 7.8,
D2O (44%), CD3CN (11%).
inhibiton of d-chymotrypsin with the cis-configurated 3-axially substituted congeners
(þ)- and (À)-12a gave rise to single correlated high-field signals at d À 6.15and À 6.14
(Fig. 7), characteristic for an axial substitution pattern at P(3). Hence, the reactions
proceeded stereospecifically in terms of neat retention of the configuration at the P-
atom to yield the diastereoisomeric phosphorylated enzyme species 18a and 19a
(Scheme 11).
4. Interpretation of the Data and Conclusions. – 4.1. General Mechanistic
Considerations. The mechanism of nucleophilic displacement reactions at a pentacoor-

2072
Helvetica Chimica Acta – Vol. 90 (2007)
Fig. 6. 1H-Correlated 31P{2H}-NMR spectra (242.9 MHz, 278) of the inhibition of d-chymotrypsin with the
trans-equatorial 3-fluoro-3-phosphadecalins a) (À)-11b and b) (þ)-11b3). Solvent: 0.2m Tris buffer,
pH 7.8, D2O (44%), CD3CN (11%).
Fig. 7. 1H-Correlated 31P{2H}-NMR spectra (242.9 MHz, 278) of the inhibition of d-chymotrypsin with the
cis-axial 3-fluoro-3-phosphadecalins a) (þ)-12a and b) (À)-12a3). Solvent: 0.2m Tris buffer, pH 7.8, D2O
(44%), CD3CN (11%).
dinate P-center such as phosphate esters and related compounds has been the subject of
many fundamental studies from which conflicting results have emerged. In particular,
exocyclic displacements occur with a bewildering variety of stereochemical implica-
tions, dependent on the nature of the substrate, attacking nucleophile, leaving group,

Helvetica Chimica Acta – Vol. 90 (2007)
2073
solvent, and further extrinsic factors14), and many results are equivocal. As a matter of
fact, the stereochemistry of six-membered phosphorus-containing rings remains not
completely understood [11]15). Nonenzymic reactions of both five- and six-membered
systems are comparable in wealth of mechanistic pathways available, including in-line
and adjacent displacement mechanisms, pseudorotation of trigonal bipyramidal
intermediates, and competing pathways. In comparable displacement reactions,
retention at the P-atom is generally rationalized by an equatorial (adjacent) entry of
the nucleophile followed by ligand reorganization (pseudorotation) according to their
relative apicophilicities and apical departure of the leaving group [11][12]. In contrast,
it was stated by Westheimer that all enzymic reactions at the P-atom proceed with
inversion and, therefore, occur without pseudorotation [12]. In fact, there seems no
unambiguous evidence that pseudorotation or adjacent attack at the P-atom is a
significant process in any biological systems, and formal retention is rationalized by a
multistep process with an even number of inversions [11 – 13].
4.2. Rationalization of the Results. Our investigations establish that the irreversible
inhibition of d-chymotrypsin is effected by covalent-bond formation between the
nucleophilic OÀC(b) of the active-site serine195 residue in the catalytic triade (Asp102
··· His57 ··· Ser195) in the enzyme [9] and the P-atom of the inhibitors. The phosphory-
lated enzyme is a stable tetrahedral adduct and considered to be an analogue of the
tetrahedral carbonyl addition intermediate or its transition state [14].
The mechanistic rationalization in terms of SN2(P)-type displacement processes is
summarized in the Schemes 12 – 14. In-line apical attack (!A) of the activated
OÀC(b) of Ser195 followed by apical departure of the leaving group from the
pentacoordinated intermediate B [12] yields the equatorially substituted enzyme
species (e.g., 16b) as found in the trans-axial series (e.g., (þ)-9a or (À)-11a); Scheme 12,
left). In the trans-equatorial series (e.g., (À)-9b or (À)-11b) an in-line attack of the
nucleophile opposite to the leaving group is unfavored in the chair conformers, only an
adjacent entry (!C) [11][12] would be feasible (Scheme 12, middle). The arrange-
ment of the substituents in the intermediate D where the leaving group is already in the
required apical position enables its facile departure without prior pseudorotation (see
4.1). The result of this process are the equatorially substituted phosphoenzymes (e.g.,
16b) where the configuration at the P-atom is retained. According to stereoelectronic
considerations (anomeric effect) which strongly favor the axial arrangement of an
electronegative substituent, equatorially substituted cyclic phosphates preferentially
adopt distorted conformations such as a boat (e.g., (À)-9b’ or (À)-11b’) or a twist-boat
conformation (e.g., (À)-9b’’ or (À)-11b’’) [4]. Hence, the observed inversion can be
explained by a direct in-line displacement process with these conformers (via
intermediate E) during which the conformation in the trigonal bipyramid F may be
maintained until the favorable chair is re-established from intermediate G yielding the
14
)
)
We have demonstrated that also the bulkiness of otherwise very similar auxiliary bases (DBN
(¼1,5-diazabicyclo[4.3.0]non-5-ene), DBU (¼1,8-diazabicyclo[5.4.0]undec-7-ene) determines the
stereochemical outcome of nucleophilic displacements at trans-3-chloro-2,4-dioxa-3-phosphabicy-
clo[4.4.0]decane 3-oxides [10].
The review [11] is the most comprehensive, critical paper covering several decades of research on
this subject. Todate, no further significant contributions to this subject have appeared.
15

2074
Helvetica Chimica Acta – Vol. 90 (2007)
Scheme 12
axially substituted phosphoenzymes (e.g., 16a, Scheme 12, right). We have no stringent
explanation for the fact that (À)-9b reacts with retention (!16b, Fig. 3,a, and
Scheme 7), whereas its enantiomer (þ)-9b reacts with inversion of the configuration at

Helvetica Chimica Acta – Vol. 90 (2007)
2075
P(3) (!17a, Fig. 3,b and Scheme 7). However, since all these processes are strongly
dependent on complex diastereoisomeric interactions, such results must never be
excluded a priori16).
The situation in the conformationally flexible cis-series of the inhibitors is less
straightforward (Schemes 13 and 14). The neat retention of (À)-10b can be explained
by an adjacent displacement process (! H) via the intermediate I as discussed above
(Scheme 13, left). The same argumentation considering the conformational changes
associated with the anomeric effect would facilitate an in-line process with the
conformers (À)-10b’ and (À)-10b’’ and result in inversion of the configuration at the P-
atom. However, besides these distorted conformations, also the completely ring-
inverted double-chair conformer (À)-10b’’’ has to be regarded as a prominent substrate
[4][8] (Scheme 13, right)17). It would enable a facile in-line attack (! K) to
intermediate L, resulting in the equatorially phosphorylated enzyme species 18b’’’.
From the experimental fact that no trace of a high-field 31P-NMR signal corresponding
to the axially substituted 18a could be detected (see Fig. 4), we anticipate that 18b’’’
might well be formed but does not undergo a complete ring inversion to 18a, probably
due to steric factors in the active-site pocket. As a consequence, this particular reac-
tion sequence consists of two different inversion processes that result in overall
retention.
Rationalizing the enzyme inhibition with the cis-fluoridates (þ)- and (À)-12a
(Scheme 14), it has to be expected a priori that an in-line attack (! M) would be
feasible, leading to inversion of the configuation at P(3) via intermediate N. But the
experimental results clearly show the reactions to proceed stereospecifically, and no
trace of a low-field signal corresponding to an equatorial epimer (e.g., 18b) could be
detected. Therefore, the reaction products have to be assigned to axially substituted
phosphoenzymes (e.g., 18a). A possible explanation for the observed retention of the
configuration at P(3) would be that an in-line attack (! M) from the ꢀbottomꢁ
side is hindered in the enzyme due to sterical crowding (Scheme 14, left), and only
an adjacent attack (!O) via intermediate P and retention (!18a) takes place
(Scheme 14, right).
5. Remarks. – As shown above, depending on the structure of the inhibitor, neat
inversion, both inversion and retention, as well as neat retention of the configuration at
the P-atom was found (Schemes 6 – 11). In particular, the stereochemical outcome of
the inhibition depends not only on the configurational characteristics of the inhibitor as
presented earlier [1][2] but also on the nature of the leaving group. This is exemplified
16
)
Such phenomena are not exceptional. It has been demonstrated earlier that the phosphonylation of
a-lytic protease by P-epimeric hexapeptide analogues yields the same covalent adduct [15]. This
result indicates inversion of the configuration at the P-atom for one diastereoisomer and retention
for the other one. Although the authors are aware that displacement processes with pseudorotation
are unprecedented in enzymic reactions, they consider it to be more plausible than a two-step
mechanism.
Recently, the complete ring-inverted double-chair conformers of the (þ)- and (À)-cis-3-fluoro-2,4-
dioxa-3-phosphabicyclo[4.4.0]decane 3-oxides (the 1H-isomers of (þ)- and (À)-12b) have been
evidenced by X-ray crystallographic analyses [8].
17
)

2076
Helvetica Chimica Acta – Vol. 90 (2007)
Scheme 13

Helvetica Chimica Acta – Vol. 90 (2007)
2077
Scheme 14
by the nonstereospecific reactions of the axially substituted trans-fluoridates (þ)- and
(À)-11a, in particular the partial retention of their configuration (Scheme 9). The latter
fact cannot be explained by the actual argumentation. We are well aware that the real
reaction pathways may be significantly more complex than the experiments make
apparent. Hence, our approach is simplified as it does not consider postinhibitory
phenomena [16] (e.g., aging [17]) nor dynamic and kinetic processes. Thus, the
inhibition might also be crucially determined by the dynamics of the active site and pre-

2078
Helvetica Chimica Acta – Vol. 90 (2007)
or postinhibitory epimerization of the phosphoenzymes to the most compatible
stereoisomer18).
As a consequence, our interpretations remain with reservations and are based
simply on the experimental data to the best of our knowledge. The presented results
seem contradictory to the generally accepted knowledge that enzymatic mechanisms
proceed stereospecifically. This holds indeed for biological processes that had been
optimized by an evolutionary development but not necessarily for nonspecific
substrates such as the investigated inhibitors.
We thank dipl. chem. M. Wächter for additional, clarifying experiments, PD Dr. O. Zerbe, Mrs. N.
Bross-Walch, and S. Jurt from the NMR department of our institute for their continuous advice and help,
and PD Dr. A. Linden for the X-ray crystallographic analyses. The financial support of the project by the
Swiss National Science Foundation is gratefully acknowledged.
Experimental Part
1. General. See [1 – 3]. For the particular precautions in preparing and handling the organo-
phosphates, see [3]. In addition: The camphanoyl derivatives were prepared with (þ)-(1R)- and (À)-
(1S)-camphanoyl chloride (¼(þ)-(1R,4R)- and (À)-(1S,4R)-4,7,7-trimethyl-3-oxo-2-oxabicyclo[2.2.1]-
heptane-1-carbonyl chloride ¼ ChiraSelect Fluka 21286 and 21287). [a]2D5: Perkin-Elmer 241-MC
polarimeter with thermostat B. Braun Thermomix 1441; 10 cm cell; ee based on the integration of the
peak areas of the anal. HPLC separations (Rs > 1.5). Anal. HPLC: Chiralcelꢀ OD-H (Daicel Chemical
Industries, Ltd.; 5 m, 25 0ꢁ 4.6 mm) column; flow rate 1 ml/min; at r.t.; Pharmacia LKB HPLC pump
2248; Hewlett-Packard 1040M diode-array detection system; data handling with a Hewlett-Packard
Chemstation for LC, Rev. A.04.02. Prep. HPLC: Chiralcelꢃ OD (10 m, 25 0ꢁ 20 mm) column; flow rate
12 ml/min; at r.t.; Applied-Biosystems 400 solvent delivery system; Applied-Biosystems 783A program-
mable absorbance detector.
2. Ethyl (Æ)-trans- and (Æ)-cis-2-Hydroxy(2-2H1)cyclohexanecarboxylates ((Æ)-2 and (Æ)-3, resp.).
To a cooled soln. (ca. 08) of ethyl 2-oxocyclohexanecarboxylate (1) (16.6 g, 97.6 mmol) in anh. EtOH
(250 ml), NaBD4 (1.40 g, 33.4 mmol) was added in portions, and the mixture was kept for 3 h at ca. 08.
Workup and continuous extraction with Et2O yielded (Æ)-2/(Æ)-3 as a slightly greenish oil (15.62 g,
94%). CC (SiO2, hexane/Et2O 5:4) afforded pure ( Æ)-2 (trans-isomer; 5.29 g, 34%) and (Æ)-3 (cis-
isomer; 8.23 g, 53%) as colorless oils. Data: identical with those of (Æ)-2 and (Æ)-3 reported in [5].
3. (Æ)-trans and (Æ)-cis-2-Hydroxy(2-2H1)cyclohexane(a,a-2H2)methanols (¼(Æ)-trans- and (Æ)-cis-
2-Hydroxy(2,2-2H2)methyl(1-2H1)cyclohexan-1-ols; (Æ)-4 and (Æ)-5, resp.). To a cooled soln. (ca. 08) of
(Æ)-2 (5.29 g, 30.5 mmol) in anh. Et2O (10 ml), LiAlD4 (1.28 g, 30.5mmol) was added in portions. After
40 min at 08, the mixture was refluxed overnight, then usual workup and continuous extraction with Et2O
afforded the trans-diol (Æ)-4 (2.76 g, 68%) as a white amorphous solid.
Starting from (Æ)-3 (8.23g, 47.5mmol) in Et 2O (150 ml) and LiAlD4 (2.0 g, 47.5mmol), the
analogous procedure yielded the cis-diol (Æ)-5 (5.42 g, 86%) as a colorless viscous oil that solidified in the
refrigerator after distillation at 1008/1 Torr. Data: identical with those of (Æ)-4 and (Æ)-5 reported in [5].
4. (þ)-(1R,2S)- and (À)-(1S,2R)-2-Hydroxy(2-2H1)cyclohexane(a,a-2H2)methanols ((þ)- and (À)-4,
resp.). 4.1. Diastereoisomeric Bis-camphanoyl Derivatives 6 and 7. To a soln. of (Æ)-4 (2.00 g, 15.04 mmol)
in pyridine (100 ml) and DMAP (50 mg) at 608, (À)-(1S)-camphanoyl chloride (6.53 g, 30.1 mmol) was
added and the mixture refluxed for 18 h. After workup, treating the residue with charcoal/EtOH, and
crystallization, the ca. 1:1 mixture of the diastereoisomers 6 (1S,1’R,2’S) and 7 (1S,1’S,2’R) was obtained
as colorless crystals (6.89 g, 93%). Recrystallization from toluene afforded pure 7 (2.06 g, 30%; de
> 99%)6). Repeated prep. HPLC of the mother liquor (Chiralcelꢃ OD, hexane/MeOH/EtOH 20 :1:0.1;
18
)
Several observations pointing towards that direction have been mentioned in [1][2]. Although
planned, we did not investigate these phenomena thoroughly, mainly because they were partly
accidental and transient and not really reproducible.

Helvetica Chimica Acta – Vol. 90 (2007)
2079
ldet 220 nm; a ¼ 1.21; Rs ¼ 1.6) afforded from the less polar fractions (k’ ¼ 2.8) pure 6 (3.32 g, 48%;
de > 99%) and from the more polar ones (k’ ¼ 3.4) additional 7 (1.25g, 18%; de > 99%).
When (Æ)-4 was treated analogously with (þ)-(1R)-camphanoyl chloride, ent-7 (1R,1’R,2’S) did not
crystallize as expected and prevented a simple enantiomer separation.
Data of (1S,4R)-4,7,7-Trimethyl-3-oxo-2-oxabicyclo[2.2.1]heptane-1-carboxylic Acid {(1R,2S)-2-
{{[(1S,4R)-4,7,7-trimethyl-3-oxo-2-oxabicyclo[2.2.1]hept-1-yl]carbonyl}oxy}(2-1H2)cyclohexyl}(2H2)-
methyl Ester (6): Colorless needles. M.p. 1658. Rf (hexane/AcOEt 1:2) 0.45. UV/VIS (EtOH): 230. IR
(KBr): 3557w, 3431w, 2971s, 2940s, 2870s, 1789s, 1725s, 1454s, 1276s, 1167s, 1103s, 1060s, 821s, 743s, 621m,
1
2
3
585m, 5 07m. H-NMR (300 MHz, CDCl3): 2.41 (ddd, J ¼ 13.4, J ¼ 10.8, 4.2, 2 HexoÀC(6)); 2.11 – 2.01
(m, 2 HendoÀC(6), HeqÀC(3’)); 1.96 – 1.86 (m, 2 CH2(5)); 1.80 – 1.63 (m, HÀC(1’), HeqÀC(6’), CH2(4’));
1.41 – 1.24 (m, HaxÀC(6’), CH2(5’), HaxÀC(3’)); 1.11 (s, 1 MeÀC(4)); 1.06, 1.05, 0.96 (each s, 1 MeÀC(4),
4 MeÀC(7)). 13C-NMR (75.4 MHz, CDCl3): 177.9 (C(3)); 167.3, 166.7 (COÀC(1)); 91.0 (C(1)); ca. 79
(m, C(2’))19); ca. 70 (m, CD2OCamph.)19); 54.7 (C(4)); 54.0 (C(7)); 41.1 (C(1’)); 31.4 (C(3’)); 30.6
(C(6)); 28.8 (C(5)); 28.0 (C(6’)); 24.5(C(5 ’)); 24.1 (C(4’)); 16.7, 16.6, 9.6 (MeÀC(4), MeÀC(7)). CI-MS
(NH3): 510 (100, [M þ NH4]þ, [C27H35D3O8 þ NH4]þ), 494 (15, [M þ H]þ), 330 (12, [M þ NH4 À
Camph.]þ). ESI-MS (CH2Cl2/MeCN/NaI): 516 (100, [M þ Na]þ, [C27H35D3O8 þ Na]þ).
Data of (1S,4R)-4,7,7-Trimethyl-3-oxo-2-oxabicyclo[2.2.1]heptane-1-carboxylic Acid {(1S,2R)-2-
{{[(1S,4R)-4,7,7-trimethyl-3-oxo-2-oxabicyclo[2.2.1]hept-1-yl]carbonyl}oxy}(2-2H1)cyclohexyl}(2H2)-
methyl Ester (7): Colorless cubes. M.p. 1748. Rf (hexane/AcOEt 1:2) 0.45. UV/VIS (EtOH): 230. IR
(KBr): 3556w, 3435w, 2974s, 2943s, 2871s, 1785s, 1725s, 1454s, 1276s, 1167s, 1103s, 1060s, 821s, 743s, 621m,
1
2
3
585m, 5 07m. H-NMR (300 MHz, CDCl3): 2.43 (ddd, J ¼ 13.4, J ¼ 10.8, 4.2, 2 HexoÀC(6)); 2.12 – 1.80
(m, 2 HendoÀC(6), HeqÀC(3’), 2 CH2(5)); 1.75– 1.64 (m, HÀC(1’), HeqÀC(6’), CH2(4’)); 1.40 – 1.27 (m,
HaxÀC(6’), CH2(5’), HaxÀC(3’)); 1.11, 1.06, 0.97 (each s, 2 MeÀC(4), 4 MeÀC(7)). 13C-NMR
(75.4 MHz, CDCl3)20): 177.9 (C(3)); 167.3, 166.7 (COÀC(1)); 91.0 (C(1)); 54.7 (C(4)); 54.0 (C(7));
41.1 (C(1’)); 31.4 (C(3’)); 30.6 (C(6)); 28.9 (C(5)); 28.0 (C(6’)); 24.5(C(5 ’)); 24.1 (C(4’)); 16.7, 16.6, 9.6
(MeÀC(4), MeÀC(7)). CI-MS (NH3): 510 (100, [M þ NH4]þ, [C27H35D3O8 þ NH4]þ), 494 (18, [M þ
H]þ), 330 (15, [M þ NH4 À Camph.]þ). ESI-MS (CH2Cl2/MeCN/NaI): 516 (100, [M þ Na]þ,
[C27H35D3O8 þ Na]þ).
4.2. trans-Diols (þ)- and (À)-4. To a cooled soln. (ca. 08) of 6 (1.04 g, 2.11 mmol) in anh. THF
(30 ml), LiAlH4 (240 mg, 6.33 mmol) was slowly added in portions. After 30 min at 08, the mixture was
refluxed until no starting material was detected (TLC). Extraction with Et2O and workup afforded a
mixture of (þ)-5 (Rf (CH2Cl2/AcOEt 3 :4) 0.12) and the reduction products of camphanic acid (Rf
(CH2Cl2/AcOEt 3 :4) 0.09) as a colorless viscous residue (916 mg). CC (SiO2, CH2Cl2/AcOEt 3 :4)
afforded pure (þ)-4 (145mg, 52%) 21) as a colorless viscous oil that solidified in the refrigerator.
The analogous procedure, starting from 7 (1.16 g, 2.35mmol) in THF (30 ml) and LiAlH 4 (270 mg,
7.1 mmol) yielded (À)-4 (172 mg, 55%)21).
Data of (þ)-4: White amorphous solid. M.p. 10 – 158. Rf (AcOEt) 0.22. [a]2D5 ¼ þ6.3 (c ¼ 1.00,
CHCl3). IR (film): 3323s (br.), 2925s, 2853s, 2665m, 2197m, 2085m, 1650w, 1449s, 1412s, 1332s, 1147s,
1
1083s, 1002s, 961s, 906m, 85 5m, 817m, 484s. H-NMR (300 MHz, CDCl3): 3.83 (br. s, w1/2 ꢀ 12, 2 OH);
3
3
1.93 (dd, J(1,6ax) ¼ 12, J(1,6eq) ¼ 3, HÀC(1)); 1.76 – 1.45( m, HeqÀC(3), CH2(4), HeqÀC(6)); 1.40 –
2
3
3
3
1.12 (m, CH2(5), HaxÀC(3)); 0.91 (qd, J ꢀ J(6ax,1) ꢀ J(6ax,5ax) ¼ 12, J(6ax,5eq) ¼ 3.6, HaxÀC(6)).
13C-NMR: 75.6 (t-like, J(C,2H) ¼ 21.3, C(2)); 67.5( quint.-like, J(C,2H) ¼ 21.5, CD2OH); 45.8 (C(1));
35.2 (C(3)); 27.3 (C(6)); 25.1 (C(5)); 24.5 (C(4)). EI-MS: 133 (1, Mþ, [C7H11D3O2]þ), 115(22, [ M À
H2O]þ), 97 (55, [M À 2 H2O]þ), 81 (26), 70 (100), 58 (62).
1
1
Data of (À)-4: [a]2D5 ¼ À6.4 (c ¼ 1.00, CHCl3). All other data: identical with those of (þ)-4.
Due to 1J(C,2H) and 2J(C,P), the 2H-substituted C-atoms gave rise to complex, not resolved m that
19
)
hardly exceeded the noise.
The signals of the H-substituted C-atoms did not exceed the noise and could not be detected.
Because the separation of the diols from the optically active reduction products of camphanic acid
required laborious chromatography, only the very best fractions were pooled (GC control) to obtain
an ee > 99%.
2
20
21
)
)

2080
Helvetica Chimica Acta – Vol. 90 (2007)
5. (þ)-(1S,2S)- and (À)-(1R,2R)-2-Hydroxy(2-2H1)cyclohexane(a,a-2H2)methanols ((þ)-5 and (À)-
5, resp.). 5.1. Bis-4-bromobenzoate (Æ)-8 and Enantiomers 8 (1S,2S) and ent-8 (1R,2R). To a soln. of (Æ)-
5 (2.45g, 18.42 mmol) and DMAP (200 mg) in pyridine (100 ml), 4-bromobenzoyl chloride (8.55g,
39 mmol) was added and the mixture refluxed overnight. After workup, the crude residue was subjected
to CC (SiO2, hexane/CH2Cl2/Et2O 6 :4 :1): (Æ)-8 (6.16 g, 67%) as colorless crystals. Repeated prep.
HPLC (Chiralcelꢃ OD, hexane/EtOH 12 :1; ldet 240 nm; a ¼ 2.33; Rs > 6) afforded from the less polar
fractions (k’ ¼ 1.2) pure 8 (2.88 g; ee > 99%) and from the more polar ones (k’ ¼ 2.8) pure ent-8 (2.91 g;
ee > 99%).
Data of 4-Bromobenzoic Acid {[1S,2S)-2-[(4-Bromobenzoyl)oxy](2-2H1)cyclohexyl}(2H2)methyl
Ester (8): Colorless needles. M.p. 130.58. Rf (hexane/CH2Cl2/Et2O 4 :6 :1) 0.60. IR (KBr): 3407w, 2936s,
1
2799m, 1917w, 1791m, 1714vs, 1585s, 1267s, 1065s, 75 3s, 680s, 471m. H-NMR (300 MHz, CDCl3): 7.86
(dd, 3J ¼ 8.6, 4J ¼ 1.9, 2 Ho); 7.56 (dd, 3J ¼ 8.7, 4J ¼ 1.9, 2 Hm); 2.16 (dd, 3J(1,6ax) ¼ 10.3, 3J(1,6eq) ¼ 4.2,
HÀC(1)); 2.11 – 2.04 (m, HeqÀC(3)); 1.85( m, d-like, w1/2 ꢀ 20, HaxÀC(3)); 1.78 – 1.39 (m, CH2(4),
CH2(5), CH2(6)). 13C-NMR (75.4 MHz, CDCl3)20): 165.6, 164.9 (CO); 131.7, 131.6 (Co); 130.9 (Cm); 128.9
(Cp); 127.9 (Cipso); 39.4 (C(1)); 29.6 (C(3)); 24.5(C(6)); 24.2 (C(5)); 20.7 (C(4)). EI-MS: 501, 499, 497
(<1, 1, < 1, M(81Br)þ, M(81Br79Br)þ, M(79Br)þ, C21H17Br2D3O4þ ); 299, 298, 297, 296 (6, 3, 6, 2, [M À
BrC6H4COO]þ), 281 (5), 207 (10), 185 (78, 81BrC6H4COþ), 183 (80, 79BrC6H4COþ), 157 (19, [185 À
CO]þ), 155 (20, [183 À CO]þ), 114 (4, C7H8D3Oþ), 104 (8), 97 (100, C7H4Dþ3 ).
The analogous data are displayed by ent-8.
5.2. cis-Diols (þ)- and (À)-5. To a cooled soln. (ca. 08) of 8 (2.91 g, 5.83 mmol) in anh. THF (50 ml),
LiAlH4 (445mg, 11.7 mmol) was slowly added in portions and the mixture then kept at r.t. for 2 h (TLC
monitoring). Extraction with Et2O, workup, and CC (SiO2, AcOEt) gave (þ)-5 (625mg, 88%) as a
colorless viscous oil that quickly solidified.
Starting from ent-8 (2.88 g, 5.77 mmol) and LiAlH4 (440 mg, 11.6 mmol), the analagous procedure
afforded (À)-5 (680 mg, 89%).
Data of (þ)-5: White amorphous solid. M.p. 48 – 508. Rf (AcOEt) 0.25. [a]2D5 ¼ þ31.9 (c ¼ 1.00,
CHCl3). IR (film): 3365w, 3324s (br.), 2923s, 2857s, 2661m, 2038w, 1445s, 1190s, 1093s, 1021s, 808m.
1H-NMR (300 MHz, CDCl3): 2.70 (CD2OH); 2.59 (OHÀC(2)); 1.82 – 1.73 (m, dt-like, HÀC(1)); 1.70 –
1.23 (m, CH2(3), CH2(4), CH2(5), CH2(6)). 13C-NMR (75.4 MHz, CDCl3): 69.9 (t-like, 1J(C,2H) ¼ 21.3,
C(2)); 66.2 (quint.-like, 1J(C,2H) ¼ 21.7, CD2OH); 42.5(C(1)); 32.9 (C(3)); 24.9 (C(6)); 23.6 (C(5)); 20.5
(C(4)). EI-MS: 133 (2, Mþ, C7H11D3O2þ ), 115(25, [ M À H2O]þ), 97 (48, [M À 2 H2O]þ), 81 (30), 70
(100), 58 (60).
Data of (À)-5: [a]2D5 ¼ À31.7 (c ¼ 1.00, CHCl3). All other data: identical with those of (þ)-5.
6. Phosphorus Heterocycles 9 – 12. 6.1. Phosphorus Reagents. The 2,4-dinitrophenyl phosphorodi-
chloridate was prepared according to [18]: POCl3 (5ml), anh. NaCl (7 mg), and 2,4-dinitrophenol were
heated under reflux (ca. 1208) in a glove box under N2 (48 h). After filtration and distilling off excess
POCl3, (bulb-to-bulb), the viscous residue crystallized (1.87 g, 92%). The product consisted of
Cl2P(O)OC6H3(NO2)2 (85%; 31P-NMR (CDCl3): 5.0) and ClP(O)[OC6H3(NO2)2]2 (15%; 31P-
NMR(CDCl3): À 6.5) and was not purified further.
POCl2F was prepared according to [19]: POCl3 (100 g) and dry, finely crystalline NH4F (48 g; Fluka
09737, puriss. p.a.) were heated at 1108 (15h) in a flask fitted with a condenser. The volatile fluorinated
compounds passing through the condenser (POCl2F (b.p. 5 48), POClF2 (b.p. 38), and POF3 (b.p. À 408))
were trapped at À 788. The mixture from the cold trap was fractionally distilled (Vigreux, 5 0 cm) to
1
afford POCl2F as a colorless liquid (11.6 g, 13%). B.p. 528. 31P-NMR (CDCl3): 1.80 (d, J(P,F) ¼ 1191).
6.2. trans- and cis-3-(2,4-Dinitrophenoxy)-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-
Oxides ((þ)- and (À)-9a, (þ)- and (À)-9b, (þ)- and (À)-10a, and (þ) and (À)-10b, resp.). To a cooled
soln. (08) of (þ)-4 (50 mg, 0.38 mmol) and pyridine (60 ml) in anh. CHCl3 (5ml) in a glove box (N 2
atmosphere), a soln. (08) of 2,4-dinitrophenyl phosphorodichloridate (145mg, 0.48 mmol) in anh. CHCl 3
(2 ml) was added dropwise and the mixture kept at 08 for 2 h (TLC and GC/MS monitoring). The volatile
components were evaporated and the residue subjected to CC (SiO2, pH 5.6 [3], hexane/AcOEt 3 :1 !
3 :2): (þ)-9a (38 mg, 28%) and (À)-9b (36 mg, 26%).
Applying the identical procedure, the following compounds were prepared from the different
starting diols (each 50 mg): from (À)-4: (À)-9a (36 mg, 26%) and (þ)-9b (40 mg, 29%); from (þ)-5: (þ)-

Helvetica Chimica Acta – Vol. 90 (2007)
2081
10a (61 mg, 45%), and (À)-10b (42 mg, 31%); from (À)-5: (À)-10a (65mg, 47%) and ( þ)-10b (43 mg,
31%); in all cases the axial epimer was less polar.
(þ)-(1S,3S,6R)-3-(2,4-Dinitrophenoxy)-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Oxide
((þ)-9a): Slightly yellowish prisms (from Et2O/hexane). M.p. 125.58. Rf (hexane/AcOEt 1:3) 0.57.
[a]2D5 ¼ þ4.5( c ¼ 1.00, CHCl3). IR (KBr): 3271m, 3123m, 3065w, 2932s, 2864s, 2670w, 2508w, 2264w,
1916w, 1804w, 1613s, 1543s, 1453m, 1346s, 1261s, 1076s, 1022s, 934s, 899s, 835s, 782s, 661m, 607m, 485s.
4
5
3
1H-NMR (300 MHz, CDCl3): 8.83 (dd, J(3’,5’) ¼ 2.7, J(3’,6’) ¼ 1.2, HÀC(3’)); 8.46 (dd, J(5’,6’) ¼ 9.2,
4J(5’,3’) ¼ 2.7, HÀC(5’)); 8.13 (dd, 3J(6’,5’) ¼ 9.2, 5J(6’,3’) ¼ 1.2, HÀC(6’)); 2.12 (ddd, 3J ¼ 12.5,
3J(10eq,9ax) ¼ 3.4, 3J(10eq,9eq) ¼ 2.0, HeqÀC(10)); 2.05(br. d, 3J(6,7ax) ¼ 11, HÀC(6))22); 1.94 – 1.80
(m, HeqÀC(9)); 1.84 – 1.70 (m, tt-like, HeqÀC(7), HeqÀC(8)); 1.59 (br. td, 2J ꢀ J(10ax,9ax) ꢀ 11,
3
3J(10ax,9eq) ¼ 3.5, HaxÀC(10))23); 1.43 – 1.26 (m, quint.-like, HaxÀC(8), HaxÀC(9)); 1.07 (qd, 2J ꢀ
3J(7ax,6) ꢀ J(7ax,8ax) ꢀ 12, 3J(7ax,8eq) ¼ 3.5, HaxÀC(7)). 13C-NMR (75.4 MHz, CDCl3): 148.2
3
(C(1’)); 143.4 (C(4’)); 140.6 (C(2’)); 129.2 (C(5’)); 123.0 (C(6’)); 121.9 (C(3’)); ca. 85( m, C(1))19); ca.
74 (m, C(5 )) ); 40.7 (d, 3J(6,P) ¼ 6.4, C(6)); 32.3 (d, 3J(10,P) ¼ 8.8, C(10)); 25.2 (C(7)); 24.2 (C(8)); 23.9
19
(C(9)). 31P-NMR (121.5MHz, CDCl 3): À 14.3 (s, w1/2 ꢀ 10, 3J(P,2H) ꢀ 3)). EI-MS: 184 (100,
(NO2)2C6H3OHþ), 154 (45), 107 (45), 91 (55), 79 (45), 63 (80), 53 (75). ESI-MS (MeOH/CHCl /
3
NaI): 384 (100, [M þ Na]þ, [C13H12D3N2O8P þ Na]þ).
(À)-(1R,3R,6S)-3-(2,4-Dinitrophenoxy)-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Ox-
ide ((À)-9a): [a]2D5 ¼ À4.3 (c ¼ 1.00, CHCl3). All other data: identical with those of (þ)-9a.
(þ)-(1R,3S,6S)-3-(2,4-Dinitrophenoxy)-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Ox-
ide ((þ)-9b): Slightly yellowish needles (from Et2O/hexane). M.p. 1148. Rf (hexane/AcOEt 1:3) 0.35.
[a]2D5 ¼ þ46.9 (c ¼ 1.00, CHCl3). IR (KBr): 3450w, 3116m, 3063m, 2939s, 2864m, 2544w, 2258w, 2175w,
2138w, 1936w, 1824w, 1613s, 1546s, 1487s, 1449m, 1417m, 1348s, 1314s, 1171m, 1149m, 1046s, 934s, 837s,
1
4
5
741s, 663m, 5 5s1, 491s. H-NMR (300 MHz, CDCl3): 8.80 (dd, J(3’,5’) ¼ 2.7, J(3’,6’) ¼ 1.2, HÀC(3’));
3
4
3
5
8.45( dd, J(5’,6’) ¼ 9.2, J(5’,3’) ¼ 2.7, HÀC(5’)); 7.98 (dd, J(6’,5’) ¼ 9.2, J(6’,3’) ¼ 1.2, HÀC(6’)); 2.33
(br. d, J(6,7ax) ¼ 11.5, HÀC(6))22); 2.20 (dt-like, J ¼ 12, J(10eq,9ax) ꢀ J(10eq,9eq) ꢀ 2, HeqÀC(10));
3
2
3
3
1.92 – 1.88 (m, HeqÀC(9)); 1.84 – 1.76 (m, HeqÀC(7), HeqÀC(8)); 1.57 (br. td, 2J ꢀ J(10ax,9ax) ꢀ 12,
3
3J(10ax,9eq) ¼ 3.5, HaxÀC(10))23); 1.39 – 1.26 (m, quint.-like, HaxÀC(8), HaxÀC(9)); 1.03 (qd, 2J ꢀ
3J(7ax,6) ꢀ J(7ax,8ax) ꢀ 12, 3J(7ax,8eq) ¼ 3.5, HaxÀC(7)). 13C-NMR (75.4 MHz, CDCl3)20): 148.3
3
3
(C(1’)); 143.6 (C(4’)); 140.3 (C(2’)); 128.9 (C(5’)); 123.9 (C(6’)); 121.5(C(3 ’)); 38.9 (d, J(6,P) ¼ 12.4,
C(6)); 32.5( d, 3J(10,P) ¼ 6.3, C(10)); 26.3 (C(7)); 24.1 (C(8)); 23.8 (C(9)). 31P-NMR (121.5MHz,
CDCl3): À 13.6 (s, w1/2 ꢀ 8, J(P,2H) ꢀ 2)). EI-MS: 184 (100, (NO2)2C6H3OHþ), 154 (20), 107 (60), 91
3
(55), 79 (45), 63 (80), 53 (75). ESI-MS (MeOH/CHCl3/NaI): 384 (100, [M þ Na]þ, [C13H12D3N2O8P þ
Na]þ).
(À)-(1S,3R,6R)-3-(2,4-Dinitrophenoxy)-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Ox-
ide ((À)-9b): [a]D25 ¼ À46.6 (c ¼ 1.00, CHCl3). All other data: identical with those of (þ)-9a.
(þ)-(1S,3S,6S)-3-(2,4-Dinitrophenoxy)-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Oxide
((þ)-10a): Colorless viscous oil. Rf (hexane/AcOEt 1:3) 0.61. [a]2D5 ¼ þ36.8 (c ¼ 1.00, CHCl3). IR
(CHCl3): 3114w, 3026m, 2945s, 2870m, 1609s, 1541s, 1485m, 1347s, 1317s, 1267s, 1224m, 1040s, 1013s, 934s,
899s, 867w, 668s, 612m, 5 5 m3 , 45 3m. 1H-NMR (300 MHz, CDCl3): 8.81 (dd, 4J(3’,5’) ¼ 2.8, 5J(3’,6’) ¼ 1.4,
HÀC(3’)); 8.45( dd, 3J(5’,6’) ¼ 9.0, 4J(5’,3’) ¼ 2.8, HÀC(5’)); 7.98 (dd, 3J(6’,5’) ¼ 9.0, 5J(6’,3’) ¼ 1.4,
3
3
HÀC(6’)); 2.05( dd, J(6,7ax) ¼ 12.0, J(6,7eq) ¼ 4.0, HÀC(6)); 1.98 – 1.85( m, HeqÀC(10)); 1.84 – 1.74
(m, HaxÀC(10)); 1.72 – 1.51 (m, CH2(7), CH2(9)); 1.45– 1.23 ( m, CH2(8)). 13C-NMR (75.4 MHz,
CDCl3)20): 148.2 (C(1’)); 143.2 (C(4’)); 131.5(C(2 ’)); 129.0 (C(3’)); 122.8 (C(5’)); 121.5(C(6 ’)); 35.7 (d,
3
3
3J(6,P) ¼ 5.1, C(6)); 31.0 (d, J(10,P) ¼ 9.2, C(10)); 24.4 (C(7)); 22.9 (d, J(8,P) ¼ 16.0, C(8)); 18.6 (d,
3J(9,P) ¼ 9.7, C(9)). 31P-NMR (121.5MHz, CDCl 3): À 14.4 (s, w1/2 ꢀ 10, 3J(P,2H) ꢀ 3). ESI-MS (MeOH/
CHCl3/NaI): 384 (100, [M þ Na]þ, [C13H12D3N2O8P þ Na]þ).
(À)-(1R,3R,6R)-3-(2,4-Dinitrophenoxy)-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Ox-
ide ((À)-10a): [a]2D5 ¼ À35.1 (c ¼ 1.00, CHCl3). All other data: identical with those of (þ)-10a.
3
3
Line broadening due to J(6ax,2HaxÀC(5)); J(6ax,7eq) was not resolved.
22
23
)
)
3
Line broadening due to J(10ax,2HaxÀC(1)).

2082
Helvetica Chimica Acta – Vol. 90 (2007)
(þ)-(1R,3S,6R)-3-(2,4-Dinitrophenoxy)-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Ox-
ide ((þ)-10b): Tiny colorless needles (from Et2O/hexane). M.p. 1588. Rf (hexane/AcOEt 1:3) 0.41.
[a]2D5 ¼ þ19.3 (c ¼ 1.00, CHCl3). IR (KBr): 3120w, 3074w, 2936m, 2857w, 1610s, 1537s, 1483w, 1555w,
1342s, 1308s, 1255m, 1049s, 996m, 966m, 932s, 896s, 779s, 5 5m3, 5 31m, 472m. 1H-NMR (300 MHz,
CDCl3): 8.81 (dd, 4J(3’,5’) ¼ 2.8, 5J(3’,6’) ¼ 1.4, HÀC(3’)); 8.45( dd, 3J(5’,6’) ¼ 9.0, 4J(5’,3’) ¼ 2.8,
3
5
3
3
HÀC(5’)); 7.98 (dd, J(6’,5’) ¼ 9.0, J(6’,3’) ¼ 1.4, HÀC(6’)); 2.36 (t, J(6,7ax) ꢀ J(6,7eq) ꢀ 6, HÀC(6));
2.19 – 2.04 (m, HeqÀC(10)); 1.90 – 1.72 (m, CH2(7), HaxÀC(10)); 1.70 – 1.58 (m, HeqÀC(9)); 1.44 (m, w1/2
ꢀ 14, CH2(8), HaxÀC(9)). 13C-NMR (75.4 MHz, CDCl3): 148.3 (C(1’)); 143.6 (C(4’)); 131.5(C(2 ’)); 128.8
(C(3’)); 123.7 (C(5’)); 121.5(C(6 ’)); ca. 82 (m, C(1)19)24); 35.1 (d, 3J(6,P) ¼ 7.6, C(6)); 29.0 (C(10)); 24.9
(C(7)); 22.2 (C(8)); 22.0 (C(9)). 31P-NMR (121.5MHz, CDCl 3): À 13.8 (s, w1/2 ꢀ 8, 3J(P,2H) ꢀ 2). EI-MS:
184 (100, (NO2)2C6H3OHþ), 154 (50), 94 (5), 91 (55), 79 (50), 53 (80). ESI-MS (MeOH/CHCl3/NaI):
384 (100, [M þ Na]þ, [C13H12D3N2O8P þ Na]þ).
(À)-(1S,3R,6S)-3-(2,4-Dinitrophenoxy)-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Ox-
ide ((À)-10b): [a]2D5 ¼ À21.5( c ¼ 1.00, CHCl3). All other data: identical with those of (þ)-10b.
6.3. trans- and cis-3-Fluoro-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Oxides ((þ)- and
(À)-11a, (þ)- and (À)-11b, (þ)- and (À)-12a, and (þ)- and (À)-12b, resp.). To a cooled soln. (08) of (þ)-4
(80 mg, 0.60 mmol) in Et2O (1 ml) in a glove box (N2 atmosphere), anh. pyridine (97 ml (95mg),
1.2 mmol) and a soln. of POCl2F (92 ml (137 mg), 1.0 mmol) in Et2O (1 ml) were added with a syringe,
and the mixture was kept for 5min at 0 8 after which the mixture was withdrawn and quickly passed
through SiO2 (pH 5.6 [3], Et2O). The precipitated pyridinium salt in the reaction flask was thoroughly
washed with Et2O, the combined soln. evaporated, and the residue subjected to CC (SiO2 pH 5.6 [3],
hexane/Et2O 4 :1) to yield (À)-11a (34 mg, 29%) and (À)-11b (37 mg, 31%).
Applying the identical procedure, the following compounds were prepared from the different
starting diols (each 80 mg): from (À)-4: (þ)-11a (38 mg, 32%) and (þ)-11b (33 mg, 28%); from (þ)-5:
(þ)-12a (39 mg, 33%) and (À)-12b (42 mg, 36%); from (À)-5: (À)-12a (41 mg, 35%) and (þ)-12b (46 mg,
39%); in all cases the axial epimer was less polar.
(þ)-(1R,3R,6S)-3-Fluoro-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Oxide ((þ)-11a):
Colorless crystals (from Et2O/hexane). M.p. 1078. Rf (hexane/Et2O 1:3) 0.36. [a]2D5 ¼ þ14.0 (c ¼ 1.00,
CHCl3), [a]2D5 ¼ þ13.2 (c ¼ 1.00, acetone). IR (KBr): 3030w, 2946s, 2866m, 1474w, 1453w, 1331s, 1248m,
1066s, 996s, 900s, 885s, 626m. 1H-NMR (300 MHz, CDCl3): 2.14 (br. d, 3J(6,7ax) ¼ 11.6, HÀC(6))22); 1.98
(dd, 2J ¼ 20.5, 3J(10eq,9ax) ¼ 8.5, HeqÀC(10)); 1.93 – 1.83 (m, HeqÀC(9)); 1.79 – 1.69 (m, HeqÀC(7),
3
HeqÀC(8)); 1.58 (br. t, 2J ꢀ J(10ax,9ax) ꢀ 12, HaxÀC(10))25); 1.41 – 1.18 (m, quint.-like, HaxÀC(8),
HaxÀC(9)); 0.98 (qd, 2J ¼ 3J(7ax,6) ¼ 3J(7ax,8ax) ¼ 12.5, 3J(7ax,8eq) ¼ 3.5, HaxÀC(7)). 13C-NMR
(75.4 MHz, CDCl3): ca. 84 (m, C(1))19); ca. 75( m, C(5 )1)9); 40.3 (d, 3J(6,P) ¼ 6.2, C(6)); 32.1 (d,
3J(10,P) ¼ 9.2, C(10)); 25.0 (C(7)); 24.0 (C(8)); 23.7 (d, 4J(9,P) ¼ 2.4, C(9)). 31P-NMR (121.5MHz,
1
1
CDCl3): À 15.4 (d, J(P,F) ¼ 1003). 19F-NMR (282.4 MHz, CDCl3): À 86.7 (d, J(F,P) ¼ 1003). EI-MS:
197 (2, Mþ, C7H9D3FO3Pþ), 154 (7), 153 (5), 141 (2), 128 (1), 116 (3), 101 (12), 97 (90), 83 (26), 82 (82),
81 (100), 80 (89), 79 (23), 70 (23), 69 (19), 68 (25), 67 (17), 58 (18), 57 (16), 56 (22), 55 (28), 54 (23), 53
(17). CI-MS (NH3): 409 (21, [2M þ NH4]þ), 215(100, [ M þ NH4]þ, [C7H9D3FO3P þ NH4]þ).
(À)-(1S,3S,6R)-3-Fluoro-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Oxide ((À)-11a):
[a]D25 ¼ À13.8 (c ¼ 1.00, CHCl3), [a]2D5 ¼ À12.9 (c ¼ 1.00, acetone). All other data: identical with those
of (þ)-11a.
(þ)-(1R,3S,6S)-3-Fluoro-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Oxide ((þ)-11b):
Colorless tiny needles (from Et2O/hexane). M.p. 688. Rf (hexane/Et2O 1:3) 0.23. [a]2D5 ¼ þ26.5( c ¼
1.00, CHCl3), [a]2D5 ¼ þ31.7 (c ¼ 1.00, acetone). IR (KBr): 3672w, 3480w, 3005w, 2870m, 1604w, 1475w,
1450w, 1325s, 1310s, 1090m, 1070s, 1055m, 1007m, 889m, 868m, 635w. 1H-NMR (300 MHz, CDCl3): 2.16
(m, t-like, HÀC(6), HeqÀC(10)); 1.91 (m, d-like, w1/2 ꢀ 20, HeqÀC(9)); 1.81 – 1.71 (m, HeqÀC(7),
3
HeqÀC(8)); 1.53 (br. t, 2J ꢀ J(10ax,9ax) ꢀ 12, HaxÀC(10))25); 1.41 – 1.22 (m, HaxÀC(8), HaxÀC(9)); 1.00
3
(qd, 2J ¼ 3J(7ax,6) ꢀ J(7ax,8ax) ¼ 12.5, 3J(7ax,8eq) ¼ 3.5, HaxÀC(7)). 13C-NMR (75.4 MHz, CDCl3): ca.
19
3
3
84 (m, C(1))19); ca. 73 (m, C(5 )) ); 39.2 (d, J(6,P) ¼ 12.9, C(6)); 32.4 (d, J(10,P) ¼ 6.3, C(10)); 25.9
24
)
)
The signal of C(5) did not exceed the noise.
3
3
Line broadening due to J(10ax,2HaxÀC(1)); J(10ax,9eq) was not resolved.
25

Helvetica Chimica Acta – Vol. 90 (2007)
2083
4
1
(C(7)); 23.9 (C(8)); 23.6 (d, J(9,P) ¼ 1.9, C(9)). 31P-NMR (121.5MHz, CDCl 3): À 14.9 (d, J(P,F) ¼
992). 19F-NMR (282.4 MHz, CDCl3): À 70.0 (d, J(F,P) ¼ 992)). ESI-MS: 197 (2, Mþ, C7H9D3FO3Pþ),
1
154 (9), 153 (7), 141 (3), 128 (2), 116 (4), 101 (15), 97 (89), 83 (26), 82 (85), 81 (100), 80 (89), 79 (26), 70
(24), 69 (22), 68 (26), 67 (17), 58 (20), 57 (17), 56 (23), 55 (28), 54 (24), 53 (16). CI-MS (NH3): 409 (11,
[2M þ NH4]þ), 215(100, [ M þ NH4]þ, [C7H9D3FO3P þ NH4]þ).
(À)-(1S,3R,6R)-3-Fluoro-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Oxide ((À)-11b):
[a]D25 ¼ À26.2 (c ¼ 1.00, CHCl3), [a]2D5 ¼ À32.2 (c ¼ 1.00, acetone). All other data: identical with those
of (þ)-11b.
(þ)-(1S,3S,6S)-3-Fluoro-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Oxide ((þ)-12a):
Colorless, viscous oil. Rf (hexane/Et2O 1:3) 0.32. [a]2D5 ¼ þ21.4 (c ¼ 1.00, CHCl3), [a]D25 ¼ þ17.7 (c ¼
1.00, acetone). IR (CHCl3): 3033w, 2944m, 2871w, 1473w, 1444m, 1434w, 1371w, 1330s, 1093s, 1067s,
1
3
3
1014s, 985s, 95 9m, 899s, 847m. H-NMR (300 MHz, CDCl3): 2.08 (dd, J(6,7ax) ¼ 10.6, J(6,7eq) ¼ 2.5,
HÀC(6)); 1.96 – 1.83 (m, HeqÀC(9), HeqÀC(10)); 1.74 (m, dt-like, w1/2 ꢀ 20, HeqÀC(7)); 1.64 – 1.51 (m,
CH2(8), HaxÀC(9), HaxÀC(10)); 1.36 (m, qd-like, w1/2 ꢀ 20, HaxÀC(7)). 13C-NMR (75.4 MHz, CDCl3):
ca. 81 (m, C(1))19); ca. 74 (m, C(5 )) ); 35.7 (d, J(6,P) ¼ 5.3, C(6)); 31.1 (d, (10,P) ¼ 9.3, C(10)); 24.5
19
3
3
(C(7)); 22.8 (C(8)); 18.6 (C(9)). 31P-NMR (121.5MHz, CDCl 3): À 15.5 (d, J(P,F) ¼ 1005). 19F-NMR
1
(282.4 MHz, CDCl3): À 86.4 (d, 1J(F,P) ¼ 1005)). EI-MS: 197 (1, Mþ, C7H9D3FO3Pþ), 154 (10), 153 (8),
141 (4), 128 (2), 116 (4), 101 (8), 97 (82), 83 (20), 82 (75), 81 (100), 80 (82), 79 (29), 70 (24), 69 (25), 68
(29), 67 (15), 58 (14), 57 (12), 56 (23), 55 (21), 54 (22), 53 (17). CI-MS (NH3): 409 (100, [2M þ NH4]þ),
215(100, [ M þ NH4]þ, [C7H9D3FO3P þ NH4]þ).
(À)-(1R,3R,6R)-3-Fluoro-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Oxide ((À)-12a):
[a]D25 ¼ À21.2 (c ¼ 1.00, CHCl3), [a]2D5 ¼ À17.6 (c ¼ 1.00, acetone). All other data: identical with those
of (þ)-12a.
(þ)-(1R,3S,6R)-3-Fluoro-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Oxide ((þ)-12b):
Colorless, tiny prisms (from Et2O/hexane). M.p. 62 – 648. Rf (hexane/Et2O 1 :3) 0.22. [a]2D5 > 0
(epimerization in CHCl3), [a]2D5 ¼ þ7.9 (c ¼ 1.00, acetone). IR (KBr): 2950m, 2865w, 1480w, 1453w,
1339m, 1325s, 1150m, 1116w, 1089m, 1062s, 1018m, 1000m, 985m, 948m, 922m, 892m, 865m, 823w, 628w.
1H-NMR (300 MHz, (CD3)2CO)26): 2.47 (t, 3J(6,7ax) ꢀ J(6,7eq) ꢀ 5 , HÀC(6)); 1.99 – 1.95( m, CH2(10));
3
1.82 – 1.65( m, CH2(7), HeqÀC(9)); 1.50 – 1.40 (m, CH2(8), HaxÀC(9)). 13C-NMR (75.4 MHz,
(CD3)2CO)20): 35.5 (d, 3J(6,P) ¼ 7.8, C(6)); 29.4 (C(10)); 25.4 (C(7)); 23.2 (C(8)); 22.4 (C(9)). 31P-
NMR (121.5MHz, (CD 3)2CO): À 15.9 (d, 1J(P,F) ¼ 980). 19F-NMR (282.4 MHz, (CD3)2CO): À 77.1 (d,
1J(F,P) ¼ 980). EI-MS: 197 (1, Mþ, C7H9D3FO3Pþ), 154 (6), 153 (5), 140 (2), 128 (1), 116 (3), 102 (7),
101 (7), 97 (74), 83 (21), 82 (81), 81 (100), 80 (91), 79 (36), 70 (25), 69 (21), 68 (34), 67 (24), 58 (14), 57
(10), 56 (23), 55 (20), 54 (27), 53 (16). CI-MS (NH3): 409 (100, [2M þ NH4]þ), 215(100, [ M þ NH4]þ,
[C7H9D3FO3P þ NH4]þ).
(À)-(1S,3R,6S)-3-Fluoro-2,4-dioxa-3-phospha(1,5,5-2H3)bicyclo[4.4.0]decane 3-Oxide ((À)-12b):
[a]D25 < 0 (epimerization in CHCl3), [a]2D5 ¼ À8.2 (c ¼ 1.00, acetone). All other data: identical with those
of (þ)-12b.
6.4. trans- and cis-N-[(Benzyloxy)carbonyl]-O-(3-oxido-2,4-dioxa-3-phospha(1,5,5-2H3)bicy-
clo[4.4.0]dec-3-yl)-l-serine Methyl Esters (14a/14a’, 14b/14b’, 15a/15a’, and 15b/15b’). To a soln. of
POCl3 (273 ml (457 mg), 3.1 mmol) and pyridine (256 ml (251 mg), 3.17 mmol) in anh. Et2O (6 ml)
prepared at 08 in a glove box (N2 atmosphere), a soln. containing (þ)-N-[(benzyloxy)carbonyl]-l-serine
methyl ester (805mg, 3.17 mmol; Bachem C-2605) was added (in situ formation of 13). After stirring for
2 h at 08, (Æ)-4 (400 mg, 3.0 mmol) and pyridine (485 ml (476 mg), 6.0 mmol) in anh. Et2O (4 ml) were
added, and the mixture was stirred for 12 h at r.t. and then evaporated. CC (SiO2, CHCl3/AcOEt 1:1) of
the crude product (1.83 g) yielded the less polar axial epimers 14a/14a’ (205mg, 16%) followed by the
equatorial epimers 14b/14b’ (268 mg, 21%), both as a ca. 1:1 mixture of diastereoisomers.
The analogus procedure with (Æ)-5 (400 mg, 3 mmol) in pyridine (256 ml (251 mg), 3.2 mmol),
POCl3 (273 ml (457 mg), 3.1 mmol), and (þ)-N-[(benzyloxy)carbonyl]-l-serine methyl ester (806 mg,
3.17 mmol) furnished, after CC (SiO2, Et2O/MeOH 50 :1) of the crude product (1.3 g), the cis-
configurated diastereoisomers (each ca. 1:1) 15a/15a’ (260 mg, 20%) and 15b/15b’ (180 mg, 14%).
26
)
Due to epimerization in CDCl3, the NMR spectra were recorded in (CD3)2CO.

2084
Helvetica Chimica Acta – Vol. 90 (2007)
N-[(Benzyloxy)carbonyl]-O-[(1RS,3RS,6SR)- and (1RS,3SR,6SR)-3-oxido-2,4-dioxa-3-phos-
pha(1,5,5-2H3)bicyclo[4.4.0]dec-3-yl]-l-serine Methyl Esters (14a/14a’ and 14b/14b’, resp.). Data of 14a/
14a’: Colorless solid. M.p. 125– 130 8. Rf (CHCl3/AcOEt 1:1) 0.31. 1H-NMR (600 MHz, CDCl3)27): 7.38 –
7.32 (m, PhCH2); 5.84 (br. d, 3J(NH,2) ¼ 7.7, NH); 5.15, 5.13 (AB, 2J ¼ 12.5, PhCH2); 4.63 (X of ABX-P,
br. d-like, w1/2 ꢀ 15, HÀC(2)); 4.50 (A of ABX-P, tq-like, HAÀC(3)); 4.38 (B of ABX-P, tt-like,
3
HBÀC(3)); 3.81 (s, MeO); 2.04 (dt, 2J ¼ 11.5, 3J(10’eq,9’ax) ꢀ J(10’eq,9’eq) ꢀ 2, HeqÀC(10’)); 1.87 (m, br.
3
t-like, HÀC(6’), HeqÀC(9’)); 1.73 (m, HeqÀC(8’)); 1.63 (dq, 2J ¼ 12, 3J(7’eq,6’) ꢀ J(7’eq,8’ax) ꢀ
2
3
3
3J(7’eq,8’eq) ꢀ 2, HeqÀC(7’)); 1.50 (td, J ꢀ J(10’ax,9’ax) ꢀ 12, J(10’ax,9’eq) ¼ 3.5, HaxÀC(10’)); 1.31 –
1.23 (m, HaxÀC(8’), HaxÀC(9’)); 0.87 (br. qd, 2J ꢀ J(7’ax,6’) ꢀ J(7’ax,8’ax) ꢀ 12, 3J(7’ax,8’eq) ꢀ 4,
HaxÀC(7’)). 13C-NMR (150.9 MHz, CDCl3): 169.5 (C(1)); 155.7 (OCON); 136.0 (arom. C); 128.5 (2C),
3
3
128.2, 128.1 (2C) (arom. CH); 81.6 (t-like, 3J(1’,P) ꢀ J(1’,2H) ꢀ 5, C(1’)); 71.5( quint.-like, 3J(5’,P) ꢀ
1
1J(5’,2H) ꢀ 5 , C(5’)); 67.2 (PhCH2); 66.7, 66.6 (each d, J(3,P) ¼ 5.1, 4.9, C(3))28); 54.5 (d, J(2,P) ¼ 6.7,
C(2)); 52.9 (MeO); 40.5 (d, 3J(6’,P) ¼ 5.9, C(6’)); 32.2 (d, 3J(10’,P) ¼ 9.1, C(10’)); 25.2 (C(7’)); 24.2
2
3
(C(8’)); 23.8 (C(9’)). 31P{1H}-NMR (242.9 MHz, CDCl3): À 6.66 (t-like, J(P,2H) ¼ 3.5). 31P{2H}-NMR
3
(242.9 MHz, CDCl3): À 6.66 (t, J(P,HÀC(3)) ¼ 7.0 ) . 31P{1H}-NMR (242.9 MHz, CD3CN/D2O/0.2m Tris
3
(pH 7.8) 11 :44 :45): À 5.12 (s, w1/2 ꢀ 10). ESI-MS (MeOH/CHCl3/NaI): 453 (100, [M þ Na]þ,
[C19H23D3NO8P þ Na]þ).
1
Data of 14b/14b’: Slightly greenish viscous oil. Rf (CHCl3/AcOEt 1:1) 0.16. H-NMR (600 MHz,
CDCl3)27): 7.36 – 7.30 (m, PhCH2); 5.80 (t, 3J(NH,2) ꢀ J(NH,3) ꢀ 7.1, NH); 5.13 (s, PhCH2); 4.57 (X of
4
ABX-P, dd-like, w1/2 ꢀ 15, HÀC(2)); 4.50 (A of ABX-P, tt-like, HAÀC(3)); 4.37 (B of ABX-P, ddd-like,
HBÀC(3)); 3.77 (s, MeO); 2.04 (m, br. dd-like, HeqÀC(10’)); 1.89 – 1.80 (m, HÀC(6’), HeqÀC(9’)); 1.72
(m, br. d-like, HeqÀC(8’)); 1.63 (m, dq-like, HeqÀC(7’)); 1.43 (qd-like, HaxÀC(10’)); 1.32 – 1.20 (m, quint.-
like, HaxÀC(8’), HaxÀC(9’)); 0.92 (m, qq-like, HaxÀC(7’)). 13C-NMR (150.9 MHz, CDCl3): 169.2 (C(1));
155.8 (OCON); 136.1 (arom. C); 128.5 (2C), 128.2, 128.1 (2C) (arom. CH); 81.6 (t-like, 3J(1’,P) ꢀ
1J(1’,2H) ꢀ 5, C(1’)); 71.5( quint.-like, 3J(5’,P) ꢀ J(5’,2H) ꢀ 5 , C(5’)); 67.8, 67.9 (each d, 2J(3,P) ¼ 5.5, 6.0,
1
C(3))28); 67.1 (PhCH2); 54.3 (d, 3J(2,P) ¼ 6.6, C(2)); 52.8 (MeO); 40.5 (d, J(6’,P) ¼ 6.7, C(6’)); 32.4 (d,
3
3J(10’,P) ¼ 8.2, C(10’)); 25.5 (C(7’)); 24.3 (C(8’)); 23.8 (C(9’)). 31P{1H}-NMR (242.9 MHz, CDCl3):
À 4.12 (t-like, 3J(P,2H) ¼ 2.5). 31P{2H}-NMR (242.9 MHz, CDCl3): À 4.11, À 4.13 (each t,
3J(P,HÀC(3)) ¼ 6.5). 31P{1H}-NMR (242.9 MHz, CD3CN/D2O/0.2m Tris (pH 7.8) 11:44 :45): À 4.10 (s,
w1/2 ꢀ 10). ESI-MS (MeOH/CHCl3/NaI): 453 (100, [M þ Na]þ, [C19H23D3NO8P þ Na]þ).
N-[(Benzyloxy)carbonyl]-O-[(1RS,3RS,6SR)- and (1RS,3SR,6RS)-3-oxido-2,4-dioxa-3-phos-
pha(1,5,5-2H3)bicyclo[4.4.0]dec-3-yl]-l-serine Methyl Esters (15a/15a’ and 15b/15b’, resp.). Data of 15a/
1
15a’: Colorless solid. M.p. 81 – 848. Rf (CHCl3/AcOEt 1:1) 0.32, Rf (Et2O/MeOH 50 :1) 0.16. H-NMR
(600 MHz, CDCl3)27): 7.29 – 7.25( m, PhCH2); 5.82 (br. d, J(NH,2) ¼ 7.7, NH); 5.07, 5.05 (AB, J ¼ 13,
PhCH2); 4.53 (X of ABX-P, br. dd-like, w1/2 ꢀ 18, HÀC(2)); 4.39 (A of ABX-P, td-like, HAÀC(3)); 4.30
(B of ABX-P, br. t-like, w1/2 ꢀ 20, HBÀC(3)); 3.72 (s, MeO); 1.87 (br. td-like, 2J ꢀ 13, 2J ꢀ 3, HeqÀC(9’),
HeqÀC(10’)); 1.74 (br. d-like, 2J ꢀ 13, HÀC(6’)); 1.57 – 1.38 (m, not resolved, HeqÀC(7’), CH2(8’),
3
2
2
3
3
3
HaxÀC(9’), HaxÀC(10’)); 1.21 (qt-like, J ꢀ J(7’ax,6’) ꢀ J(7’ax,8’ax) ꢀ 13, J(7’ax,8’eq) ꢀ 4, HaxÀC(7’)).
13C-NMR (150.9 MHz, CDCl3): 169.5 (C(1)); 155.7 (OCON); 136.0 (arom. C); 128.4 (2C), 128.1, 128.0
(2C) (arom. CH); ca. 78 (m, C(1’))19); ca. 73 (m, C(5’))19); 67.1 (PhCH2); 66.7, 66.5(each d, 2J(3,P) ¼ 4.3,
4.2, C(3))28); 54.4 (d, 3J(2,P) ¼ 5.8, C(2)); 52.8 (MeO); 35.7 (d, 3J(6’,P) ¼ 5.1, C(6’)); 31.1 (d, 3J(10’,P) ¼
8.6, C(10’)); 24.5(C(7 ’)); 22.8 (C(8’)); 18.6 (C(9’)). 31P{1H}-NMR (242.9 MHz, CDCl3): À 6.49 (s-like,
w1/2 ꢀ 8). 31P{2H}-NMR (242.9 MHz, CDCl3): À 6.49 (m, w1/2 ꢀ 22). 31P{1H}-NMR (161.9 MHz, CD3CN/
D2O/0.2m Tris (pH 7.8) 11:44 :45): À 5.38 (t-like, w1/2 ꢀ 10). ESI-MS (CH2Cl2/MeCN/NaI): 453 (100,
[M þ Na]þ, [C19H23D3NO8P þ Na]þ).
27
)
The diastereoisomerism of these mixtures was significantly revealed in the 1H-NMR at 600 MHz as
compared to that at 300 MHz. Most of the signals were doubled but the individual lines were
overlapping, a fact that resulted in an overall lower resolution compared to spectra at 300 MHz. A
full interpretation of the data is only made in those cases where the coupling situation was
unambiguous.
28
)
Double signals due to diastereoisomerism.

Helvetica Chimica Acta – Vol. 90 (2007)
2085
Data of 15b/15b’: Colorless viscous oil. Rf (CHCl3/AcOEt 1:1) 0.16, Rf (Et2O/MeOH 50 :1) 0.05.
1H-NMR (600 MHz, CDCl3)27): 7.29 – 7.21 (m, PhCH2); 5.82 (br. d, 3J(NH,2) ¼ 7.5, NH); 5.06 (s, PhCH2);
4.52 (X of ABX-P, br. dt-like, w1/2 ꢀ 16, HÀC(2)); 4.46 (A of ABX-P, tt-like, HAÀC(3)); 4.33 (B of ABX-
P, ddd-like, HBÀC(3)); 3.71 (s, MeO); 2.09 – 1.14 (m, not resolved, HÀC(6’), CH2(7’), CH2(8’), CH2(9’),
CH2(10’)). 13C-NMR (150.9 MHz, CDCl3): 169.1 (C(1)); 155.7 (OCON); 136.0 (arom. C); 128.4 (2C),
128.1, 127.9 (2C) (arom. CH); ca. 77 (m, C(1’))19); ca. 71 (m, C(5’))19); 67.9 (d, 2J(3,P) ¼ 5.3, C(3)); 67.0
(PhCH2); 54.4 (d, 3J(2,P) ¼ 5.8, C(2)); 52.8 (MeO); 35.5 (d, 3J(6’,P) ¼ 5.1, C(6’)); 30.3 (d, 3J(10’,P) ¼ 7.1,
C(10’)); 23.9 (C(7’)); 23.7 (C(8’)); 19.9 (C(9’)). 31P{1H}-NMR (242,9 MHz, CDCl3): À 3.93 (s-like, w1/2
ꢀ
8). 31P{2H}-NMR (242.9 MHz, CDCl3): À 3.93 (m, w1/2 ꢀ 20). 31P{1H}-NMR (242.9 MHz, CD3CN/D2O/
0.2m Tris (pH 7.8) 11:44 :45): À 3.77, À 3.95(each s, w1/2 ꢀ 10). ESI-MS (CH2Cl2/MeCN/NaI): 453 (100,
[M þ Na]þ, [C19H23D3NO8P þ Na]þ).
7. 2D 31P-NMR Experiments with d-Chymotrypsin. Sample Preparation. The freshly prepared soln. of
one of the enantiomerically pure inhibitors 9 – 12 (0.6 mg, 1.65 mmol) in CD3CN (60 ml) was quickly
added to a freshly prepared soln. of d-chymotrypsin (40 mg, 1.6 mmol) in D2O (240 ml)/Tris buffer (250 ml;
pH 7.8, 0.2m) and mixed in a Vortexꢀ. The mixture (yellow with the 2,4-dinitrophenoxy substituted
inhibitors 9 and 10 due to partial hydrolysis) was transferred into an NMR tube (5mm i.d.) and kept for
2 d at r.t. in the dark29).
NMR Parameters. General: Bruker DRX-600 spectrometer, 242.9 MHz, T 300 Æ 1 K; d(31P) in ppm
rel. to 85% H3PO4 (¼0 ppm) as external reference; window function in F1 and F2 ¼ sine 2; 1000
transients. 2D H-1/X Correlation via heteronuclear zero and double quantum coherence; phase sensitive
using TPPI with decoupling during acquisition; peak type selection by using gradient pulses with
coherence selection step after t1.
Inhibition Experiments with the 2,4-Dinitrophenoxy-Substituted Inhibitors 9 and 10: Relaxation delay
1
1.5s; acquisition time 0.213 s; spectral widths 1440 Hz (2.4 ppm ( F1, H)), 1950 Hz (8.0 ppm (F2,31P));
time domain in F1 ¼ 31.
Inhibition Experiments with the Fluoro-Substituted Inhibitors 11 and 12: Relaxation delay 1.5s;
acquisition time 0.1316 s; spectral widths 6600 Hz (11.0 ppm (F1, 1H)), 1950 Hz (8.0 ppm, (F2,31P)); time
domain in F1 ¼ 2048.
REFERENCES
[1] W. Ganci, E. J. M. Meier, F. Merckling, G. Przibille, U. Ringeisen, P. Rüedi, Helv. Chim. Acta 1997,
80, 421.
[2] S. Furegati, W. Ganci, G. Przibille, P. Rüedi, Helv. Chim. Acta 1998, 81, 1127.
[3] S. Furegati, W. Ganci, F. Gorla, U. Ringeisen, P. Rüedi, Helv. Chim. Acta 2004, 87, 2629.
[4] S. Furegati, M. Binder, A. Linden, P. Rüedi, Helv. Chim. Acta 2006, 89, 1351, and refs. cit. therein.
[5] M. J. Stçckli, P. Rüedi, Helv. Chim. Acta 2001, 84, 106.
[6] J. Sanchez Real, J. Pascual, An. Fis. Quim., B 1953, 49, 445; M. A. Febrer, P. Gomis, J. Pascual, An.
Fis. Quim. 1964, 60, 671; P. Gomis Torne, Rev. Real. Acad. Cienc. Exactas, Fis. Natur. Madrid 1966,
60, 419; K. Chilina, U. Thomas, A. F. Tucci, K. D. Michael, C. M. Stevens, Biochemistry 1969, 8, 2846;
R. U. Lemieux, J. T. Brewer, Adv. Chem. Ser. 1973, 117, 121.
[7] G. Przibille, Internal Progress Reports, University of Zurich, 1995, 1996; S. Furegati, A. Linden, G.
Przibille, D. Rentsch, P. Rüedi, ꢀ1st Electronic Conference on Synthetic Organic Chemistryꢁ
(ECSOC-1), 1 – 30 September 1997; http://unibas.ch/mdpi/ecsoc/a0034/a0034.htm.
[8] M. Wächter, ꢀ Herstellung optisch aktiver Organophosphate mit cis- und trans-Decalingerüst zur
Untersuchung der Inhibition von Acetylcholinesteraseꢁ, M. Sc. Thesis, University of Zurich, 2005.
[9] N. K. Schaffer, C. S. May, W. H. Summerson, J. Biol. Chem. 1953, 202, 67; D. M. Blow, J. J. Birktoft,
B. S. Hartley, Nature 1969, 221, 337; D. M. Blow, Acc. Chem. Res. 1976, 9, 145; J. Kraut, Annu. Rev.
Biochem. 1977, 46, 33; P. A. Frei, S. A. Whitt, J. B. Tobin, Science 1981, 264, 1927; T. A. Steitz, R. G.
29
)
Due to turbidity of the sample, no NMR spectra could be recorded when the soln. was kept at lower
temperature.

2086
Helvetica Chimica Acta – Vol. 90 (2007)
Shulman, Annu. Rev. Biophys. Bioeng. 1982, 11, 419; L. Polgar, in ꢀ Hydrolytic Enzymesꢁ, Eds. A.
Neuenberger and K. Brocklehurst, Elsevier, Amsterdam, 1987.
[10] F. Merckling, P. Rüedi, Tetrahedron Lett. 1996, 37, 2217.
[11] G. R. J. Thatcher, R. Kluger, Adv. Phys. Org. Chem. 1989, 25, 99.
[12] F. H. Westheimer, Acc. Chem. Res. 1968, 1, 70; R. F. Hudson, C. Brown, Acc. Chem. Res. 1972, 5, 204;
P. Gillespie, F. Ramirez, I. Ugi, D. Marquarding, Angew. Chem., Int. Ed. 1973, 12, 109; S. Tripett,
Pure Appl. Chem. 1974, 40, 595.
[13] F. Eckstein, P. J. Romaniuk, B. A. Connolly, Methods Enzymol. 1982, 87, 197; S. A. Buchwald, D. E.
Hansen, A. Hassett, J. R. Knowles, Methods Enzymol. 1982, 87, 279; G. Lowe, Acc. Chem. Res. 1983,
16, 244.
[14] R. Wolfenden, Acc. Chem. Res. 1972, 5, 10; G. E. Lienhard, Science 1973, 180, 149; R. Huber, W.
Bode, Acc. Chem. Res. 1978, 11, 114; A. Warshel, G. Naray-Szabo, F. Sussman, J. K. Hwang,
Biochemistry 1989, 28, 3629.
[15] N. S. Sampson, P. A. Bartlett, Biochemistry 1991, 30, 2255; R. Bone, N. S. Sampson, P. A. Bartlett,
D. A. Agard, Biochemistry 1991, 30, 2263.
[16] C. M. Thomson, S. Ryu, C. E. Berkman, J. Am. Chem. Soc. 1992, 114, 10710.
[17] R. F. Toia, J. E. Casida, Biochem. Pharmacol. 1979, 28, 3307; A. C. M. van der Drift, H. C. Beck,
W. H. Dekker, A. G. Hulst, E. R. J. Wils, Biochemistry 1985, 24, 6894; J. Grunwald, Y. Segall, E.
Shirin, D. Waysbort, N. Steinberg, I. Silman, Y. Ashani, Biochem. Pharmacol. 1989, 38, 3157.
[18] C. A. Bunton, E. J. Fendler, J. H. Fendler, J. Am. Chem. Soc. 1967, 89, 1221.
[19] C. J. Wilkins, J. Chem. Soc. 1951, 2726.
Received July 2, 2007