Cyclic five-membered phosphinate esters as transition state analogues for obtaining phosphohydrolase antibodies

642

Cyclic five-membered phosphinate esters as

transition state analogues for obtaining

phosphohydrolase antibodies

Gabriel Hum, Krista Wooler, Jeremy Lee, and Scott D. Taylor

Abstract: Two novel cyclic five-membered phosphinate esters (6 and 7) were synthesized and used as transition state

analogues (TSAs) for raising phosphohydrolase antibodies. The key step in the syntheses was a McCormack reaction

between (2-chloroethyl)dichlorophosphite and 1,4-diphenyl-1,3-butadiene to form isomeric cyclic phosphinyl chlorides

which upon further elaboration yielded the TSAs. X-ray crystal structures of two cyclic phosphinate TSA precursors

(14a and 14b) show highly compressed CPC bond angles of 94°–97° indicating that the CPC bond angles in 6 and 7

are significantly distorted towards the ideal trigonal bipyramidal equatorial–apical transition state angle of 90° formed

during phosphate ester hydrolysis. Antibodies were raised to 6 and 7 using standard hybridoma technology. One anti-

body, Jel 541, an IgM class antibody, was obtained that was capable of catalyzing the hydrolysis of phosphonate sub-

strate 9 and, to a much lesser extent, phosphate substrate 8. This antibody bound TSA 6 approximately 100 times more

tightly than TSA 7 as determined by solid phase radio immune assays. The activity of the antibody-catalyzed reaction

was completely inhibited by the presence of TSA 6. Although the relative instability and poor solubility of Jel 541

made it impossible to perform a detailed kinetic analysis of the antibody-catalyzed reactions, these results demonstrate

that phosphohydrolase antibodies can indeed be obtained using cyclic five membered phosphinates as haptens.

Key words: transition state analogues, phosphate ester hydrolysis, catalytic antibodies.

Résumé : On a réalisé la synthèse de deux nouveaux esters phosphinates formant un cycle à cinq chaînons (6 et 7) et

on les a utilisés comme analogues de l’état de transition (AET) pour la production d’anticorps de la phosphohydrolase.

L’étape clé de la synthèse est une réaction de McCormack entre le dichlorophosphite de 2-chloroéthyle et le

1,4-diphénylbuta-1,3-diène pour former les chlorures de phosphinyle cycliques isomères qui, par réaction ultérieure,

conduisent aux AET. Les structures cristallines des deux phosphinates cycliques précurseurs dAET (14a et 14b), telles

que déterminées par diffraction des rayons X, montrent que les angles de liaison CPC sont fortement comprimés

(94–97°) et ceci indique que les angles de liaison dans les composés 6 et 7 sont fortement déformés vers l’angle idéal

de 90° de l’état de transition trigonal bipyramidal équatorial-axial qui caractérise l’hydrolyse des esters phosphates. Des

anticorps ont été produits à l’aide des produits 6 et 7 en faisant appel à la technologie hybridome standard. On a

obtenu un anticorps, Jel 541, un anticorps de la classe IgM, capable de catalyser l’hydrolyse d’un substrat phosphonate

9 et, à un degré moindre, celle du substrat phosphonate 8. Sur la base d’essais radioimmunologiques en phase solide,

cet anticorps se lie à l’AET 6 environ 100 fois plus fortement que l’AET 7. L’activité de la réaction catalysée par

l’anticorps est complètement inhibée par la présence de l’AET 6. Même si l’instabilité relative et la mauvaise solubilité

du Jel 541 empêchent toute analyse cinétique détaillée des réactions catalysées par l’anticorps, ces résultats démontrent

qu’il est possible d’obtenir des anticorps de la phosphohydrolase en utilisant des phosphinates cycliques à cinq

chaînons comme haptènes.

Mots clés : analogues d’états de transition, hydrolyse d’esters phosphates, anticorps catalytiques. Hum et al. 655

Introduction

therapeutic agents in the treatment or prevention of

organophosphorus poisoning such as paraoxon, tabun, sarin

etc. The most common means of obtaining catalytic antibod-

ies is to raise antibodies to compounds that mimic the transi-

tion state of the reaction of interest. In general, phosphate

triesters and phosphonate diesters hydrolyze via pentavalent,

Recently, several reports have appeared describing anti-

bodies capable of hydrolyzing phosphate triesters (1–4). In-

terest in generating these antibodies, as well as phosphonate-

hydrolyzing abzymes, is mainly due to their potential as

Received December 15, 1999. Published on the NRC Research Press website on May 12, 2000.

G. Hum and S.D. Taylor.^1,2Department of Chemistry, University of Toronto, Mississauga Campus, 3359 Mississauga Rd. N,

Mississauga, ON L5L 1C6, Canada.

K. Wooler and J. Lee. Dept. of Biochemistry, University of Saskatchewan, 107 Wiggins Rd., Saskatoon, SK S7N 5E5, Canada.

¹Current address: Department of Chemistry, University of Waterloo, 200 University Avenue West, Waterloo, ON N2L 3G1, Canada.

²Author to whom correspondence may be addressed. Telephone: (519) 888-4567 ext. 3325. Fax: (519) 746-0435.

e-mail: s5taylor@sciborg.uwaterloo.ca

Can. J. Chem. 78: 642–655 (2000)

Hum et al.

643

trigonal bipyramidal (TBP) transition states or intermediates

(5). However, most TBP species are either too toxic or too

unstable in aqueous solution to be used as haptens for immu-

nization. Consequently, recent reports on phosphotriesterase

antibodies have utilized a bait and switch approach (1–4) us-

ing tetrahedral N-oxides as transition state analogues (TSAs).

For example, Rosenblum et al. (1) have reported antibodies

capable of catalyzing the hydrolysis of the phosphotriester 1

with a rate enhancement of ~360-fold by raising antibodies

to the N-oxide 2. Janda and coworkers (2–4) have reported

the generation of antibodies capable of catalyzing the hydro-

lysis of the insecticide paraoxon, 3, utilizing employing

N-oxides 4 and 5 as haptens. With the cyclic hapten 4, the

largest rate enhancement obtained was ~500-fold (2, 3). The

rate enhancement was only 56-fold using the less con-

strained hapten 5 (4). On the basis of these results, Janda

and coworkers (4) concluded that constrained haptens elicit

superior catalysts for paraoxon hydrolysis.

As part of our research program on antibody-catalyzed re-

actions, we became interested in examining new haptens for

obtaining phosphohydrolase antibodies. It has been known

for many years that phosphate or phosphonate esters con-

tained in a five-membered ring can be as much as 10⁸-times

more reactive than their acyclic analogues (5–8). This phe-

nomenon has been studied in detail and has been attributed

mainly to these species having strained bond angles at phos-

phorus approaching that of the ideal TBP equatorial–apical

transition state configuration (5–9). Applying this knowledge

to TSA design, we propose that cyclic five-membered

phosphinic acids could be used as TSAs for raising phospho-

hydrolase abzymes. To test this strategy we have designed

two isomeric cyclic 5-membered phosphinate esters, 6 and 7,

as TSAs for the hydrolysis of model phosphonate and phos-

phate esters 8 and 9 (Scheme 1). Several features of these

TSAs warrant mention. First, compounds 6 and 7 should

have highly compressed endocyclic C—P—C bond angles of

about 95°–98° (9, 10). Thus, when antibodies raised to 6 or

7 bind acyclic substrates 8 or 9, the ArX—P—OAr bond in

8 or 9 will be compressed from the unstrained tetrahedral

angle (109°) to a strained angle of 95°–98° (towards the op-

timal unstrained TBP transition state angle of 90°) analo-

gous to that found in the highly reactive cyclic, five-membered

phosphate and phosphonate esters. This will result in the in-

duction of strain or stress upon the bound substrate thus pre-

venting the substrate from falling into a thermodynamic

“pit” upon binding to the abzyme. Attack of hydroxide or

water at phosphorus will result in the formation of the un-

strained TBP transition state. Second, although 8 and 9 are

neutral, the highly polar nature of the phosphoryl bond (11)

should result in the generation of complementary residues in

644

Can. J. Chem. Vol. 78, 2000

Scheme 1.

the antibody binding site which will aid in the stabilization

of the negatively charged TBP transition state. Third, by uti-

lizing both the trans (6, phenyl groups trans to one another)

and cis (7, phenyl groups cis to one another) isomers of the

TSA, crucial antigenic determinants can be ascertained. For

example, if an antibody raised to the cis hapten binds

equally well to the trans hapten (or vice versa) then this

would imply that the antibody does not bind strongly to both

phenyl rings and would probably not catalyze the reaction.

Finally, both TSAs have a hexanoic acid linker arm for at-

taching the TSAs to carrier proteins which is necessary for

immunization. The linker arm is attached to the phosphorus

atom and not the aryl rings to minimize any interference by

the linker arm with antibody recognition of the aryl rings.

In this report we present the synthesis of trans and cis

2,5-diarylphospholanate TSAs (6 and 7) and our attempts to

obtain phosphohydrolase antibodies using these compounds

as haptens.

Results and discussion

Synthesis of transition state analogues 6 and 7

Although there are a variety of methods for synthesizing

cyclic five-membered phosphinates (12, 13), we chose to

employ a McCormack reaction since this is the most com-

mon and versatile means of constructing this class of com-

pounds (13, 14). The McCormack reaction involves a

cycloaddition reaction between a P(III) halide or in situ-

generated phosphenium ion (15) and a 1,3-diene. Although

this reaction usually proceeds more readily with phosphen-

ium ions, we chose instead to employ Moedritzer’s modifi-

cation (16) of the McCormack reaction, which employs 2-

chloroethyldichlorophosphite as the P(III) halide and di-

rectly produces cyclic phosphinyl chlorides which can be re-

acted in situ with alcohols to form the esters. For our purposes,

1,4-diphenyl-1,3-butadiene would be employed as the diene.

However, there are very few examples in the literature of

McCormack reactions between 1,4-diaryl-1,3-butadienes and

P(III) halides and, in all cases reported to date, only alkyl-

phosphonic dichlorides have been used yielding phospholes

(17–19) or phospholenium salts (20). McCormack reactions

between diaryl-1,3-butadienes and chlorophosphites have

never been reported.

McCormack reactions are usually carried out neat or in an

aprotic solvent, such as benzene, at room temperature or,

more rarely, under reflux conditions (13). However, we were

unable to detect any reaction between 2-chloroethyldichloro-

phosphite, 10, and 1,4-diphenyl-1,3-butadiene, 11, when per-

formed neat (with heating) or in refluxing benzene. This is

most probably due to both steric effects and the deactivating

effect of the phenyl groups on the cycloaddition reaction

Scheme 2.

Hum et al.

645

Scheme 3.

(14).³However, performing the reaction in a pressure tube at

170°C (neat) or 175°C (benzene as solvent) followed by re-

action of the crude reaction mixture with excess ethanol in

benzene and 1 equiv. of triethylamine, yielded the 2,3-dihydro

cyclic phosphinate products 12a and 12b (Scheme 2) in an

overall yield of 30–37%. Although a McCormack reaction

of this type usually yields the 3,4-dihydro products, the for-

mation of the 2,3-dihydro products 12a and 12b was not en-

tirely unexpected since studies by other workers have shown

that double bond migration sometimes occurs when

McCormack reactions are conducted at high temperatures

(13, 21). Isomers 12a and 12b could be separated using sil-

ica gel chromatography yielding the trans isomer, 12a,

(phenyl group and ethoxy group trans to one another) and

cis isomer, 12b (phenyl group and ethoxy group cis) in a 1:1

ratio if performed neat or 2:1 in favor of 12a if performed in

benzene.

methyl and methylene protons of the ethoxy group appear as a

triplet at δ 1.15–1.25 and as a complex multiplet at δ 3.9–4.3

respectively and these chemical shifts are typical of ethoxy

groups in most simple ethoxy phosphinates. However, with

the cis isomer, 12b, the methyl protons appear as a triplet at

δ 0.85–0.95, considerably further upfield than those of the

trans isomer. In addition, the methylene protons appear as

two separate multiplets, one proton at δ 3.7–3.95 ppm and

the other at δ 3.4–3.65, again considerably further upfield

than those of the trans isomer. The upfield shift of these

protons in the cis isomers is a result of the shielding effect

of the phenyl moiety that is located at the 5-position on the

phospholene ring. This does not occur with the trans iso-

mers since the ethoxy and phenyl ring are on opposite sides

of the phospholene ring. A similar phenomenon has been re-

ported with some 2-phenyl-2-phospholene-1-oxides (22).

Hydrogenation (5% Pd/C) of the cis phospholenate ester,

12b, yielded the “all cis” phospholanate ester 13c (both

phenyl groups and the ethoxy group cis to one another) ex-

clusively in 82% yield (Scheme 3). Hydrogenation of 12a

The cis–trans relationship of the ethoxy and phenyl

groups in 12a and 12b was readily determined by conven-

1

tional H NMR spectroscopy. In the trans isomer, 12a, the

³U.S. Patents 2,663,736 and 3,663,737.

646

Can. J. Chem. Vol. 78, 2000

Scheme 4.

Scheme 5.

groups in all subsequent phospholanate esters reported in

this study were readily determined in the same manner sim-

ply by examining the chemical shift of the -CH₂-O-P pro-

tons and by ¹³C NMR.

gave both the “phenyl cis” phospholanate ester, 13b, (major

product, the two phenyl groups cis to one another and the

ethoxy group trans to the phenyl groups) and the “phenyl

trans” phospholanate ester, 13a as the minor product in an

overall 99% yield. These two isomers could be readily sepa-

rated from one another using conventional silica gel chroma-

tography. We found that sufficient quantities of the phenyl

trans isomer, 13a, could be obtained by isomerization of ei-

ther of the less stable cis esters, 13b and 13c in sodium

ethoxide solutions. For example, a solution of pure 13b was

isomerized in a 0.5 M solution of sodium ethoxide in etha-

nol in just a few hours to yield mainly 13a. This

isomerization could be followed by ³¹P NMR. After equilib-

rium is reached, the reaction was quenched with mild acid

and two isomers were readily separated using conventional

silica gel chromatography. Esters 13a–c were bis-para-ni-

trated to form 14a–c in moderate to good yields (55% for

14a, 64% for 14b, and 75% for 14c, Scheme 3) employing

classical nitration conditions (HNO₃/H₂SO₄).

A tert-butyl-protected linker arm was prepared as shown

in Scheme 4. Thus, caprolactone was reacted with tetra-

butylammonium hydroxide in water to form the carboxylate

salt, followed by reaction with benzyl chloride in DMF to

give the hydroxyester 15 in an overall yield of 89%. Com-

pound 15 was converted to a tert-butyl ester protected linker

arm by first protecting the alcohol moiety as the TBDMS

ether (16, 92%) followed by hydrogenolysis to obtain the

acid 17 (87% from 15). The acid was converted into the tert-

butyl ester using DCC–DMAP and t-butanol and the silyl

protecting group removed using TBAF to give the tert-butyl

ester protected linker arm 18 (38% from 17).

Trans ester, 14a, was converted into the free acid, 19a, us-

ing TMSI followed by hydrolysis of the resulting TMS ester

in an overall yield of 90% (Scheme 5). The linker arm was

attached via a Mitsonobu reaction (23) between 19a and 18

to give the protected racemic TSA 20a in an 80% yield. Re-

moval of the t-butyl group using 20% TFA in CH₂Cl₂pro-

ceeded smoothly to give the trans TSA, 6 (80% yield).

Compounds 14b or 14c were converted into the cis free

acid 19b using the same procedure described for the trans

acid 19a (Scheme 6). However, Mitsonobu coupling of 18 to

19b proceeded very sluggishly. Partial purification of the re-

action mixture revealed that approximately equal amounts of

isomers 20a–c were formed (Scheme 6) in an overall low

yield (45% yield). The fact that all possible isomers (20a–c)

were obtained suggested that not only was isomerization tak-

ing place, but that at least some of the all cis isomer 19c was

formed initially. This was confirmed by performing the

Mitsonobu reaction with the unnitrated acid, 21, which

yielded exclusively the all cis ester, 22b in a 60% yield

(Scheme 7). It also appears that the presence of the nitro

groups in 19b are responsible for the isomerization (by in-

creasing the acidity of the benzylic protons on the

phospholane ring) and that this occurs fairly readily.

The all cis isomers (13c and 14c) and phenyl cis isomers

(13b and 14b) could be readily distinguished from the

phenyl trans isomers (13a and 14a) since the cis isomers are

symmetrical meso compounds and their phospholane ring

carbons appear as only two sets of doublets in the ¹³C NMR

as opposed to four sets of doublets for the unsymmetrical

phenyl trans ester. The all cis and phenyl cis isomers were

1

readily distinguished on the basis of their H NMRs. For ex-

ample, in the all cis isomer, 13c, the methylene protons of

the ethoxy group appear considerably more upfield (δ 3.19–

3.35) than is normally found with a phosphorus ethyl ester

(typically δ 3.9–4.4) due to the shielding effect of the phenyl

rings and are even further upfield than the benzylic

phospholane ring protons (δ 3.33–3.51). In the case of the

phenyl cis isomer, 13b, the ethyl group is on the opposite

side of the phospholane ring as the phenyl groups. Thus, the

methylene protons appear much further downfield and have

chemical shifts that one would expect to have for a “normal”

phosphinate ethyl ester (δ 3.94–4.12) and are much further

downfield than the benzylic protons (δ 3.2–3.4). The

stereochemical relationship between the ethoxy and phenyl

Since the Mitsonobu procedure with 19b produces a mix-

ture of isomeric products, an alternative method for attaching

the linker arm was necessary. We attempted to synthesize

Hum et al.

647

Scheme 6.

Scheme 7.

Scheme 8.

20b by converting 18b into the acid chloride using oxalyl

chloride and cat. DMF, followed by reaction with 18 in the

presence of 1.0 equiv. amine base in benzene (Scheme 8).

However, isomerization occurred during the reaction and the

undesired “aryl trans” ester, 20a, was the predominant prod-

uct. However, this was not the case with the unnitrated acid

21 which yielded exclusively the desired phenyl cis ester

22a in a 54% yield. To obtain the cis TSA 7, we subjected

22a to HNO₃–H₂SO₄(Scheme 8) which we anticipated would

result in both ring nitration and loss of the tert-butyl protect-

ing group. This was successful, yielding TSA 7 in a 45%

yield and no products resulting from isomerization were ob-

tained.

sition state angle of 90°. To our knowledge, the only report

describing a crystal structure of a saturated phospholanate is

that of phospholanic acid, obtained by Alver and Kjoge (9)

in 1969. The CPC bond angle was reported to be 94.8°.

Since it was possible that the CPC bond angles in TSAs 6

and 7 may differ significantly from that found in phospho-

lanic acid due to the presence of the aryl groups at positions

2 and 5 in the phospholanate ring and the ester linkage, we

obtained the crystal structures of the TSA precursors, 14a

and 14b. The CPC bond angle for 14a is 97.4° while that for 14b

is 94.0°. It is unlikely that replacing the ethyl group in 14a

or 14b with the linker arm found in TSAs 6 and 7 would re-

sult in a significant change in the endocyclic CPC bond an-

gles. Thus, it may be concluded that the CPC bond angles in

6 and 7 are significantly distorted or strained towards the

optimal unstrained TBP transition state angle of 90°.

Crucial to the success of our approach to obtaining

phosphohydrolase antibodies is that the CPC bond angle in

TSAs 6 and 7 is compressed towards the optimal TBP tran-

648

Can. J. Chem. Vol. 78, 2000

Scheme 9.

The half-lives of 8 and 9 in 10 mM CHES buffer, 50 mM

NaCl at pH 9.0, were determined using spectrophotometry

by measuring the increase in absorbance at 400 nm. Under

these conditions, the half life of the phosphate 8 was 14 h

while that of the phosphonate 9 was 29 h.

Antibody production and catalytic screening

Compounds 6 and 7 were coupled to carrier proteins BSA

and KLH for monoclonal antibody production and screening

by converting their carboxylic acid moieties to activated es-

ters using N-hydroxy sulfosuccinimide and EDC as coupling

agent followed by reaction of the resulting activated esters

with BSA or KLH in 50 mM phosphate buffer. Monoclonal

antibodies to each hapten (KLH conjugates) were raised us-

ing standard hybridoma technology (24). After propagation

and subcloning of the hybridomas, seven hybridoma cell

lines were obtained that produced antibody capable of bind-

ing the TSAs as determined by solid phase radio immune as-

says (SPRIA). Isotyping of the seven antibodies revealed

that three were IgM’s and four were IgG’s. Only one of

these antibodies, Jel 541, was able to distinguish, in terms of

binding affinity, between the trans TSA and cis TSA, exhib-

iting a greater affinity for the trans hapten than the cis

hapten. Surprisingly, Jel 541 was obtained from mice immu-

nized with the cis hapten. To determine if this may have

been due to isomerization of the cis hapten to the thermody-

namically more stable trans hapten during the immunization

process, cis hapten 7 was incubated in 50 mM phosphate

buffer at pH 7.5, and then examined for the formation of the

cis TSA 6 by HPLC over a period of one week. After

1 week, 8% of 7 had isomerized to 6. These results suggest

that some isomerization of the cis to the trans hapten oc-

curred during the eight week immunization process.

Synthesis of substrates 8 and 9

Phosphate ester substrate 8 was synthesized by coupling

linker arm 18 to bis(p-nitrophenyl)phosphate using a

Mitsonobu reaction followed by removal of the tert-butyl

protecting group from the crude ester using TFA (73% over-

all yield, Scheme 9).

Phosphonate ester substrate 9 was synthesized as outlined

in Scheme 10. Thus, hydrolysis of phosphonate 23 in aque-

ous NaOH followed by acidification gave the phosphonic

acid 24 in 90% yield. EDC coupling of p-nitrophenol to 24

gave the ester 25 in 64% yield. Reaction of 25 with NaBr in

refluxing butanone followed by acidification gave

phosphonic acid 26 in 93% yield. The linker arm 18 was at-

tached to phosphonic acid 26 using a Mitsonobu reaction.

The resulting ester 27 was not isolated in pure form but in-

stead the crude was reacted with TFA–CH₂Cl₂to give sub-

strate 9 (64% from 26).

The hybridomas clones were grown intraperitoneally in

mice and the antibodies purified from ascitic fluid. Competi-

tive SPRIA experiments with purified Jel 541 indicated that

Jel 541 bound the trans TSA approximately 100 times more

tightly than the cis TSA. All seven antibodies were screened

for activity by incubating the antibodies (~5 µM) with sub-

strates 8 or 9 (500 µM) in 10 mM HEPES buffer at pH 7.0

Scheme 10.

Hum et al.

649

Fig. 1. Hydrolysis of phosphonate substrate 9 (0.5 mM) in

10 mM HEPES, 50 mM NaCl, pH 7.0 by buffer alone (᭹),

5 µM Jel 541 (᭺).

Fig. 2. Hydrolysis of phosphonate substrate 9 (0.5 mM) in

10 mM CHES, 50 mM NaCl, pH 9.0 by buffer alone (᭹), 5 µM

Jel 541 (᭺), 5 µM Jel 541 + 0.5 mM TSA 6 (ٗ).

to a much lesser extent the phosphate substrate 8. Although

it was not possible to characterize Jel 541 in detail, these re-

sults demonstrate that phosphohydrolase antibodies can in-

deed be obtained using haptens such as 6. We are currently

attempting to obtain more IgG class monoclonal antibodies

to 6, since this class of antibodies are much easier to charac-

terize than IgM class antibodies and will allow for a more

complete assessment of this approach to obtaining

phosphohydrolase abzymes.

and determining the absorbance of the reaction solution at

400 nm at various time intervals over a period of several

hours. Only Jel 541, one of the IgM class antibodies, cata-

lyzed the hydrolysis of phosphonate 9 (Fig. 1) This antibody

was also capable of functioning at pH 9.0 (Fig. 2). Jel 541

also catalyzed the hydrolysis of the phosphate substrate 8

(data not shown) at pH 9.0 but to a much lesser extent than

the phosphonate substrate. This result is consistent with the

fact that the phosphonate substrate is structurally more simi-

lar to the hapten than the phosphate substrate. The possibil-

ity that the observed activity was due to contamination by an

unknown murine phosphatase can be excluded for the fol-

lowing reasons. First, none of the other antibodies obtained

in this study showed activity yet were prepared in an identi-

cal manner to Jel 541. Second, the catalytic activity of Jel

541 is completely inhibited by the presence of the trans

TSA (Fig. 2). Although it is clear that Jel 541 is capable of

catalyzing the hydrolysis of the substrates, the relative insta-

bility and poor solubility of Jel 541 has prevented us from

performing an accurate and detailed kinetic analysis. Conse-

quently, we have been unable to determine the basic kinetic

parameters (k_cat/k_uncat, K_metc.) of the abzyme. These are

common problems when dealing with IgM class antibodies

(25) which have molecular weights of approximately

10³kDa. To avoid some of the problems associated with

IgM Mabs, we attempted to convert Jel 541 Mab into a more

soluble Fab fragment by digesting the Mab with trypsin in

the presence of a reducing agent such as cysteine followed

by alkylation with iodoacetamide and purification (25).

However, we were unable to obtain significant quantities of

pure Fab using this procedure. It is clear that an IgG class

abzyme will have to be obtained to fully assess the viability

of our approach to phosphohydrolase abzymes.

Experimental

General

All starting materials for organic synthesis were obtained

from commercial suppliers (Sigma–Aldrich Canada,

Oakville, Ontario, Canada or Lancaster Synthesis Inc.,

Windham, New Hampshire, U.S.A.) unless stated otherwise.

Solvents were purchased from Caledon Laboratories

(Georgetown, Ontario, Canada), Lancaster Synthesis Inc., or

BDH Canada (Toronto, Canada). Reagents for hybridoma

production were obtained from Gibco-BRL. Tetrahydrofuran

(THF) and diethyl ether were distilled over sodium metal in

the presence of benzophenone. Benzene and CH₂Cl₂were

distilled over calcium hydride. Ethanol was distilled over

magnesium turnings. Dry chloroform was obtained by distil-

lation over phosphorus pentoxide. All glassware was

predried prior to use and all liquid transfers were performed

using dry syringes and needles. Silica gel chromatography

was performed on 40–60µ particle silica gel obtained from

Toronto Research Chemicals (Toronto, Ontario, Canada).

Melting points were obtained on a Electrothermal Inc. melt-

1

ing point apparatus and are uncorrected. H and ³¹P NMR

spectra were recorded on a Varian 200-Gemini NMR ma-

chine at approximately 200 MHz, and 80 MHz respectively.

¹³C NMR spectra were recorded on a Varian-500 NMR at

126 MHz or on a Varian 200-Gemini NMR machine at

50.3 MHz. The abbreviations s, d, t, q, qt, m, dd, and b are

used for singlet, doublet, triplet, quartet, quintet, multiplet,

doublet of doublets and broad respectively. Coupling con-

stants are reported in Hertz (Hz). All NMRs were run using

Summary

In summary, two novel phospholanates 6 and 7 were pre-

pared and used as TSAs for obtaining phosphohydrolase an-

tibodies. One IgM class antibody, Jel 541, was obtained that

catalyzed the hydrolysis of the phosphonate substrate 9 and

650

Can. J. Chem. Vol. 78, 2000

CDCl₃as solvent unless stated otherwise. Chemical shifts (δ)

for H NMR spectra run in CDCl₃are reported in ppm rela-

50 mL), 5% NaHCO₃(3 × 50 mL), brine (3 × 50 mL), and

then dried (MgSO₄) and concentrated in vacuo to yield a

crude orange oil which was subjected to silica gel column

chromatography. Flash chromatography (silica, 90% CH₂Cl₂

– 10% EtOAc, three columns required) of the crude reaction

yielded 12a as a pale yellow solid (577 mg, 19.2%). TLC

(silica, 90% CH₂Cl₂– 10% EtOAc): R_f= 0.4; mp 82–86°C.

¹H NMR δ: 1.23 (3 H, t, J = 6.9 Hz), 2.81–2.96 (1 H, m),

3.10–3.54 (2 H, m), 3.98–4.23 (2 H, m), 7.03 (1 H, t, J =

3.3 Hz), 7.42 (8 H, m), 7.65 (2 H, d, J = 6.2 Hz). ¹³C NMR

δ: 142.20 (d, J = 36.1),138.35 (br d), 136.95 (d, J = 4.6 Hz),

133.59 (d, J = 9.1 Hz), 128.70, 128.39 (d, J = 11.5 Hz),

126.96, 126.68 (d, J = 5.5 Hz), 61.58 (d, J = 6.4 Hz), 42.28

(d, J = 92.8 Hz), 35.57 (d, J = 17.8 Hz), 16.66 (d, J = 5 Hz).

³¹P NMR δ: 60.41 (s). MS, m/z (relative intensity): 298 (64),

270 (36), 205 (100), 104 (31), 91 (41), 77 (30). HRMS

calcd. for C₁₈H₁₉O₂P: 298.1123; found 298.1127. From the

same column 12b was obtained as a white solid (533 mg,

17.8%). TLC (silica, 90% CH₂Cl₂– 10% EtOAc): R_f= 0.6;

1

tive to the internal standard tetramethylsilane (TMS). Chem-

ical shifts (δ) for ¹H NMR spectra run in CD₃OD are

reported in ppm relative to residual solvent protons (δ 4.87).

1

Chemical shifts (δ) for H NMR spectra run in DMSO_d–6are

reported in ppm relative to residual solvent protons (δ 2.50).

For ¹³C NMR spectra run in CDCl₃, chemical shifts are re-

ported in ppm relative to the CDCl₃residual carbons (δ 77.0

for the central peak). For ¹³C NMR spectra run in DMSO_d–6

,

chemical shifts are reported in ppm relative to the DMSO re-

sidual carbons (δ 39.5 for the central peak). For ³¹P NMR

spectra, chemical shifts are reported in ppm relative to 85%

phosphoric acid (external). Electron impact (EI) mass spec-

tra were obtained on a Micromass 70-S-250 mass spectrom-

eter. HPLC purifications and hapten isomerization studies

were performed on a Waters LC 450 System equipped with

an Waters 2487 absorbance detector using an Altech normal

phase silica preparative HPLC column (250 mm × 22 mm

i.d. with 10 µ Econosil silica, cat. no. 6259) or an Altech

normal phase analytical HPLC column (10 µ Econosil silica,

cat. no. 60090). All of the phospholanate esters reported

here were analyzed by HPLC using the above system

equipped with the above mentioned Altech normal phase an-

alytical HPLC column and were found to exhibit a single

peak in the HPLC chromatogram.

1

mp 102–104°C. H NMR δ: 0.91 (3 H, t, J = 5.9 Hz), 2.84–

3.93 (5.5 H, m), 7.08 (0.5 H, t, J = 3.75 Hz), 7.35 (8 H, m),

7.71 (2 H, d, J = 7 Hz). ¹³C NMR δ: 142.23 (d, J =

35.7 Hz), 137.58, 136.12 (d, J = 6.5 Hz), 135.20, 133.14 (br

d), 128.70 (d, J = 6.2 Hz), 128.34, 126.92, 126.66 (d, J =

3.7), 61.58 (d, J = 6.4), 43.44 (d, J = 90.6), 33.19 (d, J =

16.4) 16.16. ³¹P NMR δ: 60.05 (s). MS, m/z (relative inten-

sity) 298 (100), 270 (66), 254 (47), 205 (50), 194 (83), 91

(43), 77 (30). HRMS calcd. for C₁₈H₁₉O₂P: 298.1123; found

298.1134.

(±)-trans-1-Ethoxy-2,5-diphenyl-2,3-dihydro-1H-1λ⁵-

phosphol-1-one (12a) and (±)-cis-1-ethoxy-2,5-diphenyl-

2,3-dihydro-1H-1λ⁵-phosphol-1-one (12b)

Method B: The above McCormack reaction can also be per-

formed in a similar manner as described above using dry

benzene as solvent (approximately 3 mL of benzene per

gram of diene) and heating the reaction in the sealed tube

for 72 h at 185°C. This procedure produces 12a and 12b in a

2:1 ratio in yields that are slightly higher that obtained by

method A. CAUTION: When performing the reaction us-

ing benzene as solvent, for safety reasons, the total vol-

ume of the reaction in the pressure tube must not be

greater than half the total capacity of the pressure tube.

The reaction must be performed in a fumehood behind a

blast shield.

Method A: 1,4 Diphenyl-1,3-butadiene (2.06 g, 10 mmol, 1

equiv.) was combined with 2,6-di-tert-butyl-4-methyl phenol

(10 mg, 2 mol%) in a 15 mL ACE Glass pressure tube (cat.

no. 8648–75) equipped with a stir bar and then flushed thor-

oughly with Ar. Phosphorous trichloride (0.545 mL,

6.33 mmol, 0.66 equiv.) and tris(2-chloroethyl) phosphite

(0.635 mL, 3.33 mmol, 0.33 equiv.) was added to the tube

under a stream of Ar. The glass bomb was rapidly sealed

and the thick, mainly solid mixture was vigorously shaken

by hand for about 30 s to mix the contents. The tube was im-

mersed in an oil bath, stirred vigorously, and slowly heated

to 170°C and maintained at this temperature for 19 h during

which time the solid diene melted and the solution reaction

became an orange colour. CAUTION: This procedure

must be performed in a fumehood behind a blast shield.

In addition to several broad beaks, the ³¹P NMR of the crude

reaction mixture exhibited two sharp peaks of approximately

equal magnitude at 68.6 and 69.7 ppm. After removing the

oil bath and allowing the tube to cool to room temperature,

the tube was fitted with a septum and dry benzene (20 mL)

was added to the crude reaction mixture under argon. The

solution was transferred very slowly over a period of 30 min

via syringe to an ice cold solution of dry ethanol (5.8 mL,

100 mmol, 10 equiv.) and triethylamine (1.7 mL, 10 mmol, 1

equiv.) in dry benzene (20 mL) and the reaction was stirred

overnight at room temperature under Ar. The volume of the

crude reaction mixture was reduced to one quarter by rotary

evaporation and diluted with ether (75 mL) which resulted in

the precipitation of triethylamine hydrochloride which was

removed by suction filtration. The filter cake was washed

with ether and the filtrate was washed with 0.1 N HCl (3 ×

(±)-1-Ethoxy-trans-2,5-diphenyl-1λ⁵-phospholan-1-one

(13a) and (meso)-trans-1-ethyoxy-cis-2,5-diphenyl-1λ⁵-

phospholan-1-one (13b)

A solution of 12a (1.530 g, 5.03 mmol) in ethanol

(20 mL) containing 10% Pd/C (100 mg) was hydrogenated

overnight at 30 psi under a H₂atmosphere using a Parr hy-

drogenation apparatus. The Pd catalyst was removed by fil-

tration and the filtrate concentrated in vacuo leaving a pale

yellow oil which was subjected to column chromatography

(silica, 90% CH₂Cl₂– 10% EtOAc) to give pure 13a and

13b. Overall yield was 99%. 13a was obtained as a white

solid (432 mg, 27%). TLC (silica, 90% CH₂Cl₂– 10%

1

EtOAc): R_f= 0.5; mp 103–105°C. H NMR δ: 0.93 (3 H, t,

J = 7.2 Hz), 2.07–2.60 (4 H, m), 3.09–3.52 (3 H, m), 3.66–

3.85 (1 H, m), 7.35 (10 H, cm). ¹³C NMR δ: 137.13 (d, J =

5.5 Hz), 128.70 (d, J = 5.5 Hz), 127.52, 126.68, 61.37 (d,

J = 6.4 Hz), 43.59 (d, J = 85.1 Hz), 29.57 (d, J = 14.6 Hz),

16.69. ³¹P NMR δ: 61.73 (s). MS, m/z (relative intensity):

300 (87), 206 (42), 196 (36), 104 (100), 91 (28), 78 (22).

Hum et al.

651

HRMS calcd. for C₁₈H₂₁O₂P: 300.1279; found 300.1272.

13b was obtained as a white solid (1.10 g, 72%). TLC (sil-

ica, 90% CH₂Cl₂– 10% EtOAc): R_f= 0.3; mp 69–72°C.

¹H NMR δ: 1.24 (3 H, t, J = 7.1 Hz), 2.33–2.46 (4 H, m),

3.21–3.37 (2 H, m), 4.00 (2 H, m), 7.35 (10 H, m). ¹³C

NMR δ: 136.69 (bd), 136.22 (d, J = 2.7 Hz), 128.76,

128.56, 128.14 (d, J = 4.6 Hz), 126.81, 61.24 (d, J =

7.3 Hz), 46.68 (d, J = 84.1 Hz), 45.50 (d, J = 86.9 Hz),

30.40 (d, J = 12.8 Hz), 27.85 (d, J = 9.0 Hz), 16.26 (d, J =

4.6 Hz). ³¹P NMR δ: 62.37 (s). MS, m/z (relative intensity):

300 (91), 206 (41), 196 (26), 104 (100), 91 (30), 78 (20).

HRMS calcd. for C₁₈H₂₁O₂P: 300.1279; found 300.1294.

13.8 Hz), 129.45 (d, J = 5.5 Hz), 128.90, 123.73, 62.10 (d,

J = 5.4 Hz), 46.7 (d, J = 84.2 Hz), 45.5 (d, H = 86.1 Hz),

30.37, (d, J = 11 Hz), 27.73 (d, J = 11 Hz.), 16.30. ³¹P NMR

δ: 59.7 (s). MS, m/z (relative intensity): 390 (100), 362 (28),

296 (61), 149 (86), 119 (51), 103 (55), 77 (58). HRMS

calcd. for C₁₈H₁₉N₂O₆P: 390.0981; found 390.0994.

(meso)-trans-1-ethoxy-cis-2,5-di(4-nitrophenyl)-1λ⁵-phospho-

lan-1-one (14b): Obtained in 64% yield as a white solid

from 13b using the general procedure described above. Puri-

fication was achieved by recrystallization in EtOAc–hexane.

1

mp 190–192°C. H NMR δ: 1.24 (3 H, t, J = 7.2 Hz), 2.45

(6 H, m), 3.40 (2 H, m), 4.03 (2 H, m), 7.52 (4 H, dd, J =

2 Hz, J = 7 Hz), 8.22 (4 H, d, J = 8.7 Hz). ¹³C NMR

(CDCl₃) δ: 147.32, 144.16 (d, J = 5.4 Hz), 129.45, 123.73,

62.26 (d, J = 7.3 Hz), 43.66 (d, J = 86.1 Hz), 29.31 (d, J =

12.8 Hz), 16.60. ³¹P NMR δ: 60.49 (s). MS, m/z (relative in-

tensity): 390 (100), 362, (26), 296 (77), 148 (95), 119 (44),

103 (37), 77 (36). HRMS calcd. for C₁₈H₁₉N₂O₆P:

390.0981; found 390.0989.

(meso)-cis-1-Ethoxy-cis-2,5-diphenyl-1λ ⁵-phospholan-1-one

(13c)

A solution of 12b (820 mg, 2.75 mmol) in ethanol

(20 mL) containing 10% Pd/C (60 mg) was hydrogenated

overnight at 30 psi under a H₂atmosphere using a Parr hy-

drogenation apparatus. The Pd catalyst was removed by fil-

tration and the filtrate concentrated in vacuo leaving a pale

yellow solid. Column chromatography (silica, 100% EtOAc)

yielded pure 13c as a white solid (674 mg, 82%). TLC (sil-

(meso)-cis-1-Ethoxy-cis-2,5-di(4-nitrophenyl)-1λ⁵-phospho-

lan-1-one (14c): Obtained in 75% yield as a white solid

from 13c using the general procedure described above. Puri-

fication was achieved using flash chromatography (silica,

100% EtOAc). TLC (silica): R_f= 0.3; white solid, mp

1

ica, 100% EtOAc): R_f= 0.40; mp 133–137°C. H NMR δ:

0.64 (3 H, t, J = 7 Hz), 2.42 (4 H, m), 3.22–3.54 (4 H, m),

7.31 (10 H, m). ¹³C NMR δ: 135.54 (d, J = 5.5 Hz), 128.41,

127.94 (d, J = 3.7 Hz), 126.63, 61.20 (d, J = 7.4 Hz), 43.96

(d, J = 82.4 Hz), 27.45 (d, J = 12.8 Hz), 15.90. ³¹P NMR δ:

65.14 (s). MS, m/z (relative intensity): 300 (64), 206 (48),

196 (18), 104 (100), 91 (29), 78 (25)). HRMS calcd. for

C₁₈H₂₁O₂P: 300.1279; found 300.1290.

1

180–182°C. H NMR δ: 0.65 (3 H, t, J = 7), 2.45 (4 H, m),

3.38 (2 H, m), 3.59 (2 H, m), 7.51 (4 H, dd, J = 2.2 Hz, J =

9.1 Hz), 8.21 (4 H, d, J = 8.8 Hz). ¹³C NMR δ: 147.28,

143.54 (d, J = 8.2 Hz), 128.63 (d, J = 4.6 Hz), 123.62, 62.15

(d, J = 7.3 Hz), 44.10 (d, J = 81.5 Hz), 26.88 (d, J =

7.9 Hz), 15.94 (d, J = 3.7 Hz). ³¹P NMR δ: 59.9 (s).

MS, m/z (relative intensity): 390 (35), 362 (20), 296 (50),

149 (100), 119 (42), 103 (39), 77 (42). HRMS calcd. for

C₁₈H₁₉N₂O₆P: 390.0981; found 390.0992.

General procedure for the nitration of 2,5-diphenyl-1λ⁵-

phospholan-1-one esters 14a,⁴⁴14b,⁴and 14c

To an ice cold solution of 13a, 13b or 13c (100 mg,

0.33 mmol, 1 equiv.) in H₂SO₄(1.5 mL) was added a solu-

tion of HNO₃:H₂SO₄(1:1, 0.0180 mL, 0.86 mmol, 2.6 equiv.

HNO₃) dropwise of a period of 1 min. The reaction was

stirred at 0°C for 30 min and then poured into a beaker con-

taining ice which upon melting left a suspension. CHCl₃

(20 mL) was added and the two layers were separated. The

aqueous layer was further extracted with CHCl₃(3 × 20 mL)

and the combined organics were washed with brine (2 ×

30 mL), dried (MgSO₄) and concentrated in vacuo leaving

an oily solid residue. Pure nitrated esters (14a, 14b, and 14c)

were obtained by either silica gel chromatography or

recrystallization of the crude residue.

Benzyl 6-hydroxyhexanoate (15): To an aqueous solution of

20% tetrabutylammonium hydroxide (100 mL, 77.1 mmol, 1

equiv.) was added ε-caprolactone (8.79 g, 77.1 mmol, 1

equiv.) and the solution heated to 65°C for 2 h during which

time the solution became homogeneous. The solution was

concentrated by rotary evaporation leaving a white solid that

was dried under high vacuum for several hours. The solid

was dissolved in dry DMF (80 mL) and benzylchloride

(10.7 g, 84.8 mmol, 1.1 equiv.) was added and the reaction

stirred for 24 h. The reaction mixture was diluted with

80 mL water and the extracted with ether (4 × 150 mL). The

ether layers were combined and dried (Na₂SO₄), then con-

centrated leaving a pale yellow oil. Flash chromatography

(silica, 60% hexane – 40% EtOAc) of the oil yielded pure 15

as a colourless oil (15.2g, 89%). TLC (silica, 60% hexane –

40% EtOAc): R_f= 0.3. ¹H NMR δ: 1.33–1.71 (7 H, m), 1.91

(1 H, s), 2.37 (2 H, t, J = 7.3 Hz), 3.60 (2 H, t, J = 5.8 Hz),

5.10 (2 H, s), 7.34 (5 H, s). ¹³C NMR δ: 173.43, 136.23,

128.50, 128.08, 66.05, 62.30, 34.25, 32.30, 25.34, 24.71.

MS, m/z (relative intensity): 223 (3), 194 (12), 115 (59), 108

(±)-1-Ethoxy-trans-2,5-di(4-nitrophenyl)-1λ⁵-phospholan-1-

one (14a): Obtained in 55% yield as a white solid from 13a

using the general procedure described above. Purification

was achieved using flash chromatography (silica, 80%

EtOAc – 20% hexane). TLC (80% EtOAc – 20% hexane): R_f

1

= 0.70; mp 179–180°C. H NMR δ: 0.98 (3 H, t, J = 7 Hz),

2.22 (2 H, m), 2.56 (2 H, m), 3.45 (3 H, m), 3.83 (1 H, m),

7.54 (4 H, dd, J = 8 Hz, J = 2.1 Hz), 9.24 (4 H, d, J =

8.4 Hz). ¹³C NMR δ: 147.32, 144.00, 143.46 (br d, J =

⁴Table of crystallographic details, atomic coordinates, anisotropic displacement parameters, and bond distances and angles have been depos-

ited and may be purchased from: The Depository of Unpublished Data, Document Delivery, CISTI, National Research Council Canada, Ot-

tawa, Ontario, Canada, K1A 0S2. Table of bond distances and angles and H coordinates have also been deposited with the Cambridge

Crystallographic Data Centre and can be obtained on request from: The Director, Cambridge Crystallographic Data Centre, University

Chemical Laboratory, 12 Union Road, Cambridge, CB2 1Ez, U.K.

652

Can. J. Chem. Vol. 78, 2000

(52), 91 (100). HRMS calcd. for C₁₃H₁₈O₃m/z: 222.1256

(7 × 200 mL), 10% NaHCO₃(2 × 200 mL), saturated brine

(1 × 200 mL), dried (MgSO₄) and concentrated leaving a

yellow oil. Flash chromatography (silica, 70% hexane –

30% EtOAc) of the oil yielded pure 18 as a pale yellow oil

(1.15 g, 38%). TLC (silica, 70% hexane – 30% EtOAc): R_f=

(M⁺); found 222.1252.

Benzyl 6-(tert-butyldimethylsilyloxy)hexanoate (16): To

a

solution of 15 (7.0 g, 31.5 mmol) in dry DMF (15 mL) was

added imidazole (5.14 g, 75.6 mmol, 2.4 equiv.) and the so-

lution stirred until all of the imidazole dissolved. tert-Butyl-

dimethylsilylchloride (5.69 g, 37.8 mmol, 1.2 equiv.) was

added in portions over a period of approximately 4 min and

the solution stirred under argon overnight. The reaction was

diluted with ether (400 mL) and washed with water (2 ×

200 mL), 10% NaHCO₃(2 × 200 mL), saturated brine (2 ×

200 mL), dried (MgSO₄), and concentrated leaving a pale

yellow oil. Flash chromatography (silica, 90% hexane – 10%

EtOAc) of the crude oil yielded pure 16 as a colourless oil

(9.71 g, 92%). TLC (silica, 90% hexane – 10% EtOAc): R_f=

1

0.25. H NMR δ: 1.36–1.63 (6 H, m), 1.41 (9 H, s), 1.86

(1 H, s), 2.20 (2 H, t, J = 7.3 Hz), 3.60 (2 H, bt). ¹³C NMR

δ: 173.14, 79.90, 61.99, 35.37, 32.17, 27.95, 25.17, 24.71.

MS, m/z (relative intensity): 132 (8), 115 (48), 97 (34), 69

(35), 57 (100). HRMS calcd. for C₁₁H₂₃O₃Si m/z: 132.0777

(M⁺– C₄H₉); found 132.0786.

General procedure for the deprotection of esters 13b, 13c,

14a, 14b, and 14c

Trimethyl silyl iodide (1.2 equiv.) was added dropwise to

a solution of the ester (13b, 13c, 14a, 14b, or 14c) in dry

CH₂Cl₂(approx. 1 mmol ester/mL of solvent) and the reac-

tion stirred for 2 h under an Ar atmosphere. Additional

CH₂Cl₂was added and the product was extracted into a 0.1

N sodium hydroxide solution. The two layers were immedi-

ately separated and the aqueous layer rapidly acidified to

pH 2 (pH paper) with 6 N hydrochloric acid which resulted

in the formation of the desired acid product as a precipitate.

The product (19a, 19b, or 21) was recovered by suction fil-

tration. In some instances, further purification by HPLC was

necessary.

1

0.6. H NMR δ: 0.044 (6 H, s); 0.89 (9 H, s); 1.36–1.71

(6 H, m); 2.37 (2 H, t, J = 7.3 Hz); 3.59 (2 H, t, J = 5.9 Hz);

5.12 (2 H, s); 7.36 (5 H, s). ¹³C NMR δ: 173.17, 136.47,

136.38, 128.46, 128.04, 65.94, 62.90, 34.34, 32.48, 25.97,

25.53, 24.82, 18.29, –5.30. MS, m/z (relative intensity): 279

(13), 171 (4), 91 (100). HRMS calcd. for C₁₅H₂₃O₃Si m/z:

279.1416 (M⁺– C₄H₉); found 279.1411.

6-(tert-Butyldimethylsilyloxy)hexanoic acid (17): To a solu-

tion of 16 (9.71 g, 28.8 mmol) in EtOAc (100 mL) was

added 5% Pd/C (400 mg). The flask was fitted with a bal-

loon filled with H₂and the reaction stirred for 16 h. The re-

action was filtered through filter paper, concentrated, and

diluted with 50 mL EtOAc. The solution was filtered again,

concentrated and residual solvent removed under high

vacuum leaving pure 17 as a colourless oil (6.79 g, 95%).

¹H NMR δ: 0.044 (6 H, s), 0.89 (9 H, s), 1.39–1.69 (6

H, m), 2.36 (2 H, t, J = 7.3 Hz), 3.61 (2 H, t, J = 5.9 Hz),

11.05 (1 H, bs). ¹³C NMR δ: 179.18, 62.85, 33.97, 32.39,

25.89, 25.40, 24.51, 18.23, –5.39. MS, m/z (relative inten-

sity): 231 (2), 229 (4), 213 (7), 189 (37), 171 (89), 75 (100).

HRMS calcd. for C₁₁H₂₃O₃Si m/z: 231.1416 (M⁺– CH₃);

found 231.1416.

(±)-1-Hydroxy-trans-2,5-di(4-nitrophenyl)-1λ⁵-phospholan-1-one

(19a): Obtained in a 90% yield as amorphous yellow solid

from 14a using the general procedure for ester deprotection

1

described above; mp 271–273°C. H NMR (DMSO-d₆) δ:

2.04–2.45 (4 H, m), 3.40–3.64 (2 H, m), 7.62 (4 H, d, J =

5.8 Hz), 8.27 (4 H, d, J = 5.8 Hz). ¹³C (DMSO-d₆) δ: 146.26,

129.68 (d, J = 5.4 Hz), 123.17, 45.68 (d, J = 82.2 Hz), 28.54

(d, J = 10 Hz). ³¹P NMR (DMSO-d₆) δ: 58.60 (s). MS, m/z

(relative intensity): 362 (100), 298 (36), 149 (54), 119 (38),

103 (26), 77 (29). HRMS calcd. for C₁₆H₁₅N₂O₆P:

362.0668; found 362.0671.

1-Hydroxy-cis-2,5-di(4-nitrophenyl)-1λ⁵-phospholan-1-one (19b):

Obtained in a 90% yield as an amorphous light yellow solid

from 14b or 14c using the general procedure for ester

deprotection described above. Purification was achieved via

HPLC (isocratic, 99% CHCl₃– 1% MeOH, retention time =

tert-Butyl 6-hydroxyhexanoate (18): To a solution of 17 (4g,

16.2 mmol) in dry CH₂Cl₂(15 mL) was added DMAP

(396 mg, 20 mol%), t-BuOH (6.19 mL, 64.8 mmol, 4 equiv.)

and the solution was cooled in an ice bath. DCC was added

(3.52g, 17.0 mmol, 1.05 equiv.) in portions over a three min

period. The solution was stirred for 1 h at 0°C and then al-

lowed to warm to room temperature then stirred for a further

24 h. The reaction was filtered and the filtrate concentrated

leaving a pale yellow oil. The oil was dissolved in ether

(300 mL) and washed with water (2 × 150 mL), 10%

NaHCO₃(1 × 150 mL), saturated brine (1 × 150 mL), dried

(MgSO₄) and concentrated to approximately 20 mL. The so-

lution was filtered and the filtrate concentrated leaving a yel-

low oil. The oil was distilled (bp 65–68°C, 100 µm) to give

3.0 g of a colourless oil. In addition to the desired tert-butyl

1

13 min); mp 278–281°C. H NMR (DMSO-d₆) δ: 2.06–2.48

(4 H, m), 3.45–3.64 (2 H, m), 7.62 (4 H, d, J = 7.3 Hz), 8.22

(4 H, d, J = 8.8 Hz). ³¹P NMR (DMSO-d₆) δ: 59.17 (s). ¹³

C

(DMSO-d₆) δ: 146.24, 129.56 (d, J = 4.5 Hz), 123.15, 43.71

(d, J = 82.4 Hz), 28.52 (d, J = 3.3 Hz). MS, m/z (relative in-

tensity): 362 (36), 298 (19), 149 (100), 133 (43), 119 (20),

103 (32), 77 (43). HRMS calcd. for C₁₆H₁₅N₂O₆P:

362.0668; found 362.0681.

1-Hydroxy-cis-2,5-diphenyl-1λ⁵-phospholan-1-one (21):

Obtained in a 90% yield as a white solid from 13b or 13c

using the general procedure for ester deprotection described

1

ester of 17, the H NMR of the distilled material indicated

that small amounts (approximately 5%) of both DCC and

DCU were present. 2.5 g (approximately 8.25 mmol) of the

impure distilled oil was dissolved in dry THF (6 mL). To

this was added TBAF (24 mL of a 1.0 M solution in THF,

24 mmol) and the solution stirred for 48 h. The solution was

diluted with ether (300 mL) and then washed with water

1

above; mp 169–171°C. H NMR (DMSO-d₆) δ: 2.13–2.44

(4 H, m), 3.20–3.37 (2 H, m), 7.16–7.39 (10 H, m). ¹³C

(DMSO-d₆) δ: 138 (δ, J = 6.5 Hz), 128.42 (d, J = 5.5 Hz),

128.08, 126.01, 43.23 (d, J = 84.2 Hz), 27.61 (d, J = 12.9 Hz).

³¹P NMR (DMSO-d₆) δ 60.32 (s). MS, m/z (relative intensity):

Hum et al.

653

272 (57), 168 (23), 104 (100), 91 (24), 78 (25). HRMS

calcd. for C₁₆H₁₇O₂P: 272.0966; found 292.0976.

(15 H, m), 2.18 (2 H, t, J = 7.3 Hz), 2.40 (4 H, m), 3.29

(2 H, m), 3.91 (2 H, q, J = 6.6 Hz), 7.35 (10 H, m). ¹³C

NMR δ: 172.66, 137.07 (d, J = 4.6 Hz), 128.75, 128.63,

128.50, 126.68, 79.92, 65.18 (d, J = 6.5), 43.51 (d, J =

86.0 Hz), 35.39, 30.49 (d, J = 5.5 Hz), 29.51 (d, J = 14.7 Hz),

28.15, 25.13, 24.62. ³¹P NMR δ: 62.56 (s). MS, m/z (relative

intensity): 442, 386 (99), 369 (33), 272 (60), 104 (100).

HRMS calcd. for C₂₆H₃₅O₄P: 442.2297; found 442.2293.

(±)-tert-Butyl 6-{[trans-2,5-di(4-nitrophenyl)-1-oxo-1λ⁵-phospho-

lan-1-yl]oxy}hexanoate (20a) from 19a via the Mitsonobu

procedure: 18 (39 mg, 0.21 mmol, 1.5 eq), 19a (50 mg,

1.4 mmol, 1.5 equiv.), triphenylphosphine (54.3 mg,

0.21 mmol, 1.5 equiv.), and diisopropylazidodicarboxylate

(DIAD) (0.041 mL, 0.21 mmol, 1.5 equiv.) were dissolved

in dry THF (1.5 mL). The solution was stirred for 30 min at

room temperature followed by removal of the solvent by ro-

tary evaporation. Flash chromatography (silica, 50% EtOAc

– 50% hexane) of the crude residue yielded pure 20a as a

clear colourless oil (60.1 mg, 82%). TLC (silica, 50%

(meso)-trans-6-{[cis-2,5-Di(4-nitrophenyl)-1-oxo-1λ⁵-phospho-

lan-1-yl]oxy}hexanoic acid (7): 22a (120 mg, 0.27 mmol, 1

equiv.) was dissolved in 0.5 mL of H₂SO₄at 0°C. A solution

of HNO₃–H₂SO₄(1:1, 0.062 mL, 2.6 equiv. HNO₃) was

added and the reaction was stirred at 0°C for 30 min. The re-

action was poured into a beaker containing ice which upon

melting left a suspension. CHCl₃(5 mL) was added and the

two layers were separated. The aqueous layer was further

extracted with CHCl₃(4 × 10 mL) and the combined

organics were washed with brine (2 × 15 mL), dried

(MgSO₄), and concentrated in vacuo leaving an oily of solid

residue. Attempts to obtain pure product by subjecting the

crude residue to silica gel flash chromatography were unsuc-

cessful. Final purification of the desired product was accom-

plished using HPLC (isocratic, 70% EtOAc – 30% hexane,

retention time = 18 min) to give 7 as a clear colourless oil

1

EtOAc – 50% hexane): R_f= 0.5. H NMR δ: 1.00–1.48

(15 H, m), 2.00–2.74 (6 H, m), 3.19–3.86 (4 H, m), 7.45–

7.58 (4 H, m), 8.23 (4 H, dd, J = 0.9 Hz, J = 6.9 Hz). ¹³C

NMR δ: 172.42, 147.41, 143.84 (d, J = 3.7 Hz), 143.21 (d,

J = 6.4 Hz), 129.50 (d, J = 5.5 Hz), 128.82 (d, J = 4.6 Hz),

123.80, 123.77, 80.08, 65.77 (d, J = 7.3 Hz), 46.83 (d, J =

83.3 Hz), 45.51 (d, J = 86 Hz), 35.17, 30.50, 30.24, 28.11,

27.77, 27.57, 24.89, 24.38. ³¹P NMR δ: 59.77 (s). MS, m/z

(relative intensity): 476 (12), 459 (65), 363 (89), 280 (48),

216 (54), 57 (100). HRMS calcd. for C₂₆H₃₄N₂0₈P (-C₄H₈):

476.1348; found 476.1351.

1

(51 mg, 44%). H NMR δ: 1.23–1.68 (6H, m), 2.24 (2 H, t,

(±)-6-{[trans-2,5-Di(4-nitrophenyl)-1-oxo-1λ⁵-phospholan-

1-yl]oxy}hexanoic acid (6): 20a (35 mg, 0.066 mmol) was

dissolved in 20% TFA – CH₂Cl₂(1.5 mL) and stirred over-

night at room temperature. The CH₂Cl₂and TFA was re-

moved by rotary evaporation. Flash chromatography (silica,

80% EtOAc – 20% hexane) on the crude residue yielded

pure 6 as a clear colourless oil (25 mg, 80%). TLC (silica,

J = 7 Hz), 2.35–2.60 (4 H, m), 3.35–3.57 (2 H, m), 3.87–

4.03 (2 H, dd, J = 6.6 Hz, J = 7.7 Hz), 7.52 (4 H, dd, J =

1.8 Hz, J = 6.9 Hz), 8.19 (4 H, dd, J = 8.4 Hz). ¹³C NMR δ:

176.79, 147.28, 143.97 (d, J = 7.3 Hz), 129.46 (d, J =

5.5 Hz), 123.84, 66.23 (d, J = 7.3 Hz), 43.56 (d, J = 84.2 Hz),

33.48, 30.30 (d, J = 5.5 Hz), 29.26 (d, J = 14.7 Hz), 24.96,

24.1. ³¹P NMR (CDCl₃) δ: 61.11 (s). MS, m/z (relative in-

tensity): 363(53), 280(57), 149(52). HRMS calcd. for C₂₂H₂₅

N₂O₈P: 477.1427; found 477.1432.

1

80% EtOAc – 20% hexane): R_f= 0.5. H NMR δ: 1.09–

1.65 (6 H, m), 2.05–2.75 (6 H, m), 3.17–3.98 (4 H, m),

7.52 (4 H, m), 8.22 (4 H, m). ¹³C NMR δ: 175.92, 147.44,

143.65 (d, J = 3.7 Hz), 143.11 (d, J = 6.4 Hz), 129.53 (d, J

= 5.5 Hz), 128.87 (d, J = 3.7 Hz), 123.82, 65.83 (d, J =

7.3 Hz), 46.64 (d, J = 84.2 Hz), 45.50 (d, J = 85.1 Hz),

33.23, 30.33, 30.02, 24.80, 24.02. ³¹P NMR δ: 60.17 (s).

MS m/z (relative intensity): 477, 444 (8), 363 (83), 280 (84),

216 (53), 149 (54), 119 (56), 69 (82). HRMS calcd. for

C₂₂H₂₅N₂O₈P (-OH): 459.1321; found 459.1327.

(meso)-cis-tert-Butyl 6-{[1-oxo-cis-2,5-diphenyl-1λ⁵-phospholan-

1-yl]oxy}hexanoate (22b): 18 (60 mg, 0.33 mmol, 1.5 equiv.),

triphenyl phosphine (87 mg, 0.33 mmol, 1.5 equiv.), 21

(60 mg, 0.22 mmol, 1 equiv.), and DIAD (65 uL, 67 mg,

0.33 mmol, 1.5 equiv.) were dissolved in dry THF (1.5 mL)

and stirred at room temperature for 30 min. The solvent was

removed under reduced pressure. Flash chromatography of

the crude residue (silica, 60% EtOAc – 40% hexane) yielded

22b as a clear colourless oil (58.5 mg, 60%). TLC (silica,

(meso)-trans-tert-Butyl 6-{[1-oxo-cis-2,5-diphenyl-1λ⁵-phospholan-

1-yl]oxy}hexanoate (22a): To a solution of 21 (120 mg,

0.55 mmol, 1 equiv.) in dry CH₂Cl₂(2 mL) was added oxalyl

chloride (0.144 mL, 1.65 mmol, 3 equiv.) dropwise over a

period of 1 min and the solution was stirred for 3 h. The sol-

vent was removed by rotary evaporation and the crude acid

chloride subjected to high vacuum for 12 h. The crude acid

chloride was dissolved in dry benzene (1 mL) and a solution

of 18 (125 mg, 0.66 mmol, 1.1 equiv.) and triethylamine

(0.100 mL, 0.72 mmol, 1.2 equiv.) in dry benzene (1 mL)

was added dropwise over a period of 2 min and the reaction

was stirred for 24 h. Ether was added which resulted in the

precipitation of triethylamine hydrochloride which was re-

moved by vacuum filtration and the filtrate was concentrated

by rotary evaporation. Flash chromatography (silica, 90%

CH₂Cl2 – 10% EtOAc) on the crude residue yielded pure

22a as a clear colourless oil (140 mg, 57%). TLC (silica,

90% CH₂Cl₂– 10% EtOAc): R_f= 0.6. ¹H NMR δ: 1.23–1.65

1

60% EtOAc – 40% hexane): R_f= 0.4. H NMR δ: 0.65–0.85

(2 H, br m), 0.89–1.04 (2 H, br m), 1.20–1.26 (3 H, br m),

1.439, (7 H, s), 1.75 (1 H, s), 1.97 (3 H, t, J = 8.1 Hz), 2.35–

2.48 (4 H, br m), 3.11–3.20 (2 H, br m), 3.38–3.54 (2 H,

br m), 7.20–7.35 (10 H, br m). ¹³C NMR δ: 172.94, 136.29

(d, J = 8.1 Hz), 128.43 (d, J = 2.2 Hz), 127.80 (d, J =

5.1 Hz), 126.65 (d, J = 3.0 Hz), 79.83, 64.69 (d, J = 7.4 Hz),

43.74 (d, J = 82.0 Hz), 35.24, 29.87 (d, J = 5.1 Hz), 28.05,

27.48 (d, J = 7.2 Hz), 24.44, 24.34. ³¹P NMR δ: 61.75 (s).

MS, m/z (relative intensity): 443 (18), 387 (58), 272 (49),

104 (100). HRMS calcd. for C₂₆H₃₅O₄P: 442.2273; found

442.2264.

Formation of bovine serum albumin (BSA)–TSA and

keyhole lympet hemocyanin (KLH)–TSA conjugates

To a solution of TSAs 6 or 7 (~6 mg, ~0.013 mmol) in

DMF:water (3:2, 3 mL) was added N-hydroxysulfosuccinimide

654

Can. J. Chem. Vol. 78, 2000

(3.5 mg, 0.016 mmol) and EDC (3.1 mg, 0.016 mmol). The

reaction was stirred at rt for 3 h. The reaction mixture was

divided into two equal aliquots (1.5 mL each). One aliquot

was added to a solution of BSA (6.3 mg) in 50 mM phos-

phate buffer (pH 7.5, 3 mL) and the other aliquot was added

to a solution of KLH (6.1 mg) in the same buffer. The reac-

tion were shaken at 4°C for 16 h. Unconjugated hapten was

removed by dialysis (6 × 1 L, 3 h each) in 50 mM phosphate

buffer, pH 7.5 at 4°C. The number of haptens per BSA was

determined using the method of Habeeb (27). The number of

trans haptens 6 per BSA was 11 and the number of cis

haptens 7 per BSA was 19. We were unable to determine the

number of haptens per KLH due to the nonhomogeneous na-

ture of the KLH solutions. The conjugates were stored at

–20°C.

5.19 mmol). The reaction was stirred under argon for 72 h.

The solution was diluted with methylene chloride (20 mL),

washed with 5% sodium bicarbonate (2 × 50 mL), 0.1 N

HCl (2 × 50 mL), and saturated brine (1 × 50 mL), dried

(MgSO₄) and concentrated leaving a solid residue. Recry-

stallization of the crude in hexane–ethyl acetate yielded pure

25 as a white solid (64% yield); mp 112–114°C. ¹H NMR δ:

3.48 (2 H, d, J = 22.7 Hz), 3.80 (3 H, d, J = 11 Hz), 7.29

(2 H, d, J = 8.8 Hz), 7.51 (2 H, dd, J = 2.9 Hz, J = 5.9 Hz),

8.22 (4 H, d, J = 8.8 Hz). ¹³C NMR δ: 155.31 (d, J =

8.2 Hz), 147.48, 144.90, 138.09 (d, J = 9.1 Hz), 130.78 (d,

J = 6.4 Hz), 125.61, 123.86, 120.85 (d, J = 4.6 Hz), 53.91

(d, J = 6.4 Hz), 33.55 (d, J = 139.1 Hz). ³¹P NMR δ: 20.94.

MS, m/z (relative intensity): 352 (100), 335 (41), 305 (27),

257 (34), 136 (49). HRMS calcd. for C₁₄H₁₃O₇N₂P:

352.0460; found 352.0450.

6-{[Di(4-nitrophenoxy)phosphoryl]oxy}hexanoic acid (8):

To a solution of 18 (125 mg, 0.66 mmol) in dry THF (4 mL)

was added bis(4-nitrophenyl)phosphate (150 mg, 0.44. mmol),

triphenyl phosphine (173 mg, 0.66 mmol), and DIAD (130

uL, 130 mg, 0.66 mmol). The solution was stirred under ar-

gon for 30 min, the solvent was then removed by rotary

evaporation and the crude residue chromatographed (silica,

95% CH₂Cl₂– 5% EtOAc). TLC (silica, 95% CH₂Cl₂– 5%

EtOAc): R_f= 0.7. Due to slow product breakdown, it was

not possible to separate the tert-butyl-6-{[di(4-nitro-

phenoxy)phosphoryl]oxy}hexanoate product from trace

amounts of 4-nitrophenol. Thus, the impure product was dis-

solved in a solution of TFA:CH₂Cl₂(1:5) and stirred for

16 h. The solvent was removed by rotary evaporation and

the product obtained by flash chromatography (silica, 50%

hexane – 50% EtOAc) yielded 8 as a white solid (146 mg,

73% overall yield). TLC (silica, 50% hexane – 50% EtOAc):

4-nitrophenyl hydrogen (4-nitrobenzyl)-phosphonate (26):

To a solution of 25 (1.83 g, 5.4 mmol) in dry butanone

(40 mL) was added NaBr (833 mg, 8.10 mmol) and the re-

sulting solution refluxed for 16 h. Upon cooling the solution

to rt, a precipitate formed. The mixture was filtered and the

precipitate washed with cold butanone. The solid was dis-

solved in water and the solution was acidified with 6 N HCl

to pH 2 which resulted in the formation of a white precipi-

tate. The mixture was filtered and the precipitate washed

with cold water then dried under vacuum to give 26 as a

1

white solid (535 mg, 93% yield). H NMR (DMSO_d6) δ:

3.54 (2 H, dd, J = 8.1 Hz, J = 23 Hz), 7.38 (2 H, d, J =

8 Hz), 7.59 (2 H. d. J = 5.9 Hz), 8.19 (4 H, d, J = 9.5 Hz).

¹³C NMR δ: 156.20, 146.44, 143.68, 141.11 (d, J = 9.1 Hz),

131.07 (d, J = 6.4 Hz), 125.48, 123.19, 121.07 (d, J =

4.6 Hz), 34.35 (d, J = 133.6 Hz). ³¹P NMR δ: 21.66.

MS, m/z (relative intensity): 338 (33), 228 (18), 210 (24),

139 (100), 109 (83). HRMS calcd. for C₁₃H₁₁O₇N₂P:

338.0304; found 338.0310.

1

R_f= 0.2; mp 77–78°C. H NMR δ: 1.20–1.90 (7 H, c m),

2.35 (2 H, t, J = 7 Hz), 4.34 (2 H, q, J = 6.6 Hz), 7.40 (4 H,

d, J = 9.2 Hz), 8.26 (4 H, d, J = 7.3 Hz). ¹³C NMR δ:

178.16, 154.74 (d, J = 6.4 Hz), 145.40, 125.81, 120.71 (d,

J = 4.7 Hz), 70.49, 33.57, 29.81, 24.84, 24.02. ³¹P NMR δ:

–15.83 Hz (s). MS, m/z (relative intensity): 341 (80), 316

(51), 139 (100), 109 (82), 65 (70). HRMS calcd. for

C₁₈H₁₉0₁₀,N₂P: 455.08556; found 455.0830.

6-{[4-Nitrobenzyl)(4-nitrophenoxy)phosphoryl]oxy}hexanoic

acid (9): To a solution of hydroxy ester 18 (156 mg,

0.83 mmol) in dry THF (1.5 mL) was added phosphonic

acid 26 (180 mg, 0.55 mmol), triphenyl phosphine (218 mg,

0.83 mmol), and DIAD (180 ul, 180 mg, 0.83 mmol). The

reaction was stirred for 1 h. The solvent was removed by ro-

tary evaporation and the crude purified by chromatography

(silica. 40% hexane – 60% EtOAc). We were unable to sepa-

rate the phosphonate product 27 from small amounts of con-

taminating of p-nitrophenol. Consequently, the slightly

impure phosphonate 27 was dissolved in a solution of 20%

TFA–CH₂Cl₂(1 mL) and stirred for 16 h. The solvent was

removed by rotary evaporation and the crude residue

chromatographed (silica, 20% hexane – 80% EtOAc) to

yield pure 9 as a thick colourless oil (159 mg, 64% overall

yield). TLC (silica, 20% hexane – 80% EtOAc): R_f= 0.6. ¹H

NMR (DMSO_d6) δ: 0.75–1.75 (7 H, m), 2.30 (2 H, t, J =

7.2 Hz), 3.52 (2 H, J = 22.7 Hz), 4.15 (2 H, m), 7.29 (2 H,

d, J = 9.6 Hz), 7.51 (2 H, dd, J = 2.6 Hz, 8.20 (4 H, d, J =

8.8 Hz). ¹³C NMR δ: 177.47, 155.16 (d, J = 7.1 Hz), 147.50,

144.93, 138.07 (d, J = 9.1 Hz), 130.85 (d, J = 5.5 Hz),

125.63, 123.84, 121.00, 67.83 (d, J = 7.3 Hz), 33.80 (d, J =

138.1), 33.61, 29.92 (d, J = 4.6 Hz), 24.84, 24.07. ³¹P NMR

δ: 19.7. MS, m/z (relative intensity): 339 (64), 314 (58), 139

Methyl hydrogen (4-nitrobenzyl)phosphonate (24): A mix-

ture composed of dimethyl (4-nitrobenzyl)phosphonate (26)

(2.5 g, 10.2 mmol) and sodium hydroxide (816 mg,

20.4 mmol) in water (60 mL) was heated at 75°C until ho-

mogeneous. The solution was acidified to pH 1 with concen-

trated hydrochloric acid (6 M), filtered, the precipitate

washed with water giving 24 as a white solid (2.12 g, 90%

1

yield); mp 132–134°C. H NMR (DMSO_d6) δ: 3.19 (2 H, d,

J = 22.8 Hz), 3.59 (3 H, d, J = 11.3 Hz), 7.44 (2 H, dd, J =

2.5 Hz, J = 8.8 Hz), 8.17 (2 H, d, J = 8 Hz). ¹³C NMR δ:

146.28, 142.02 (d, J = 9.2 Hz), 130.90 (d, J = 5.5 Hz),

123.13, 51.72 (d, J = 6.4 Hz), 33.30 (d, J = 131.8). ³¹P NMR

δ: 25.48 (s). MS, m/z (relative intensity): 231 (50), 214

(100), 184 (31), 136 (31), 107 (62), 89 (52). HRMS calcd.

for C₈H₁₀O₅NP: 231.0297; found 231.0314.

Methyl (4-nitrophenyl) (4-nitrobenzyl) phosphonate (25): Com-

pound 24 (1 g, 4.33 mmol) was suspended in dry methylene

chloride (40 mL). To the suspension was added 4-nitro-

phenol (722 mg, 5.19 mmol) and EDC (995 mg,

Hum et al.

655

(77), 109 (51), 97 (66), 69 (100). HRMS calcd. for

C₁₉H₂₁0₉,N₂P: 453.1063; found 453.1086.

ume was 0.300 mL and the total amount of DMSO in the as-

say mix was 1% (v/v). The change in absorbance at 400 nm

was monitored at 25°C in a Gilford 260 spectrophotometer.

Kinetic studies with substrates 8 and 9 in the absence

of antibody

Acknowledgements

A 20 µL aliquot of a 1 mM stock solution of 8 or 9 in

DMF was added to a cuvette containing 980 µL of 10 mM

CHES, 50 mM NaCl, pH 9.0 preincubated at 25°C. The

cuvette was incubated at 25°C and the hydrolysis reaction

was followed by measuring the absorbance at 400 nm every

30 min using a Cary 1B UV–Vis spectrophotometer until the

reaction had proceeded to approximately 3–4 half-lives. The

data was fitted to the first-order rate equation using the GraFit

program (28). Each reaction was performed in duplicate.

We would like to thank the Medical Research Council of

Canada (MRC) for financial support. We thank Alan Lough

for performing X-ray structure analyses.

References

1. J. S. Rosenblum, L.-C. Lo, T. Li, K.D. Janda, and R.A. Lerner.

Angew. Chem. Int. Ed. Eng. 34, 2275 (1995).

2. B.J. Lavey and K.D. Janda. Bioorg. Med. Chem. Lett. 6, 1523

(1996).

3. B.J. Lavey and K.D. Janda. J. Org. Chem. 61, 7633 (1996).

4. D.A. Spivak, T. Hoffman, A.H. Moore, M.J. Taylor, and K.D.

Janda. Bioorg. Med. Chem. 7, 1145 (1999).

5. G.R.J. Thatcher and R. Kluger. Adv. Phys. Org. Chem. 25, 99

(1989).

6. F.H. Westheimer. Acc. Chem. Res. 1, 70 (1968).

7. R. Kluger and S. Taylor. J. Am. Chem. Soc. 112, 6669 (1990).

8. S. Taylor and R. Kluger. J. Am. Chem. Soc. 114, 3067 (1992).

9. Y.H. Chiu and W.N. Lipscomb. J. Am. Chem. Soc. 91, 4150

(1969).

Isomerization studies with 7

A 500 µM solution of 7 (3.0 mL) in 20 mM phosphate

buffer at pH 7.5 was incubated at 25°C. Aliquots (0.5 mL)

were removed at various time intervals and quenched with

1 M HCl (40 µL) and then extracted with EtOAc (0.5 mL).

The extent of isomerization to 6 was determined by normal

phase HPLC analysis of the EtOAc solutions using EtOAc

as eluant (1.0 mL/min for 5 min then switching to 2.0 mL

for 6 min) and the detector set at 280 nm. Under these con-

ditions, the retention time of the cis TSA 7 was 4.2 min and

the trans TSA 6 was 6.2 min.

10. E. Alver and H.M. Kjoge. Acta Chem. Scand. 23, 1101 (1969).

11. V.V. Prezhdo, O.V. Prezdho, and E.V. Vaschenko. J. Mol.

Struct. 385, 137 (1996).

12. S.M. Kalinov, M.F. Rostovskaya, and V.I. Vysotskii. Zh.

Obshch. Khim. 58, 783 (1988).

13. L.D. Quin. The Heterocyclic Chemistry of Phosphorus. Wiley,

New York. 1981. Chap. 2.

14. W.B. McCormack. Chem. Abstr. 49, 7601 (1955).

15. A.H. Cowley, R.A. Kemp, J.G. Lasch, N.C. Norman, C.A.

Stewart, B.R. Whittlesey, and T.C. Wright. Inorg. Chem. 25,

740 (1986).

16. K. Moedritzer. Synth. React. Inorg. Met. Org. Chem. 5, 45

(1975).

17. J.-C. Fiaud and J.-Y. Legros. Tetrahedron Lett. 32, 5089 (1991).

18. I.G.M. Campbell, R.C. Cookson, M.B. Hocking, and A.N.J.

Hughes. J. Chem. Soc. 2184 (1965).

19. B. Lukas, R.M.G. Roberts, J. Silver, and A.S. Wells. J.

Organometal. Chem. 256, 103 (1983).

20. J.M. Holland and D.W. Jones. J. Chem. Soc. Perkin Trans. 1,

927 (1973).

21. L.D. Quin, J.P. Gratz, and J. Barket. J. Org. Chem. 33, 1034

(1968).

Antibody production

Balb/c mice were immunized IP five times at ten day in-

tervals with 50 µg of the KLH conjugates. The first four in-

jections were given intraperitoneally after emulsification of

the conjugate with Freund’s adjuvant while the final injec-

tion was given intravenously with just conjugate in PBS.

Three days after the final injection, hybridoma production

was initiated as described previously (29, 30). The fusion

was screened by solid phase radio immune assays (SPRIA)

(29, 30) using PVC plates coated with the BSA conjugates.

Seven cell lines, two obtained from mice immunized with

the trans TSA 6 and five obtained from mice immunized

with the trans TSA 7, were cloned. Isotyping of the antibod-

ies obtained from mice immunized with the trans hapten 6

revealed that one was an IgM class and one an IgG class an-

tibody, while those obtained from mice immunized with the

cis hapten 7 revealed that three were IgG class antibodies

and two were IgM class antibodies. Ascites fluid was pre-

pared (29, 30) from these seven cell lines and the antibodies

were purified by gel exclusion chromatography on a

Sephracryl S-200 column as previous described (29, 30).

SDS-PAGE (12%) analysis of Jel 541 revealed only a single

band at the top of the gel (Coomassie Blue staining). Com-

petitive SPRIAs (29, 30) were performed with plates coated

with the BSA conjugate of the cis TSA 7.

22. R. Bodalski, K.M. Pietrusiewicz, and J. Koszuk. Tetrahedron,

31, 1907 (1975).

23. D.A. Campbell and J.C. Bermak. J. Org. Chem. 59, 658 (1994).

24. G. Kohler and C. Milstein. Nature (London), 256, 495 (1975).

25. J.M. Bidlack and P.C. Mabie. J. Immunol. Meth. 91, 157 (1986).

26. S.D. Taylor, C.C. Kotoris, A.N. Dinaut, and M.-J. Chen. Tetra-

hedron, 54 1691 (1998).

27. A.F. Habeeb. Anal. Biochem. 14, 328 (1966).

28. R.J. Leatherbarrow. GraFit Version 3.0, Erithacus Software

Ltd., Staines, U.K. 1992.

Antibody kinetics

29. J.S. Lee, M.L. Woodsworth, and L.J.P. Latimer. Biochemistry,

23, 3277 (1984).

30. J.S. Lee, D.F. Dombrowski, and T.R. Mosmann. Biochemistry,

21, 4940 (1982).

Purified antibody was dialyzed overnight into 10 mM

CHES, 50 mM NaCl, pH 9.0 or 10 mM HEPES, 50 mM

NaCl, pH 7.0. A stock solution of the substrates in DMSO

was added directly to the antibody solutions. The final vol-

Article Doi

Cyclic five-membered phosphinate esters as transition state analogues for obtaining phosphohydrolase antibodies

DOI: 10.1139/v00-068

Source and publish data:

Authors:

Article abstract of DOI:10.1139/v00-068

Full text of DOI:10.1139/v00-068

Products guided by the article

R&D Labs maybe for 158961-87-6

Relevant to this article

Hot Product