A one-pot multistep approach to α-azido-phosphonate and phosphonothioate diesters: Key intermediates in the synthesis of haptens for the generation of antibody ligases

Article abstract of DOI:10.1016/S0960-894X(00)00137-2

A four-step, one-pot synthesis of mixed α-azido-phosphonates and phosphonothioates 12a-d is described. This chemistry has provided a facile route to haptens 6a-b and 7 that have been employed for the elicitation of antibody ligases. Five hapten-specific antibodies have been identified as modest catalysts of a model peptide ligation reaction between thioester 1b and thiol 2 to give the amide product 5. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00137-2

Bioorganic & Medicinal Chemistry Letters 10 (2000) 915±918
A One-Pot Multistep Approach to ꢀ-Azido-phosphonate and
Phosphonothioate Diesters:
Key Intermediates in the Synthesis of Haptens for the Generation
of Antibody Ligases
Curtis W. Harwig, Timothy Z. Ho€man, Anita D. Wentworth and Kim D. Janda*
Department of Chemistry and The Skaggs Institute for Chemical Biology, The Scripps Research Institute,
10550 North Torrey Pines Road, La Jolla, California 92037, USA
Received 23 November 1999; accepted 16 February 2000
AbstractÐA four-step, one-pot synthesis of mixed a-azido-phosphonates and phosphonothioates 12a±d is described. This chemistry
has provided a facile route to haptens 6a±b and 7 that have been employed for the elicitation of antibody ligases. Five hapten-
speci®c antibodies have been identi®ed as modest catalysts of a model peptide ligation reaction between thioester 1b and thiol 2 to
give the amide product 5. # 2000 Elsevier Science Ltd. All rights reserved.
Chemical ligation is emerging as a valuable technique
for the convergent synthesis of large proteins from
unprotected peptide segments.1 The acceleration of
such chemoselective couplings of native or non-native
peptide fragments is being explored with enzymes,2
catalytic antibodies,3 5 and self-replicating peptide
sequences.6
6a±b and 7 were designed as transition state analogues
for the formation of thioester 3.13 The location of the
glutaroyl linker in 6a±b was chosen in an e€ort to pro-
mote recognition of the incoming nucleophile 2 in the
antibody binding site and thus to direct its attack on the
carbonyl center of 1a±b in the presence of 55 M water.14
Phosphonic acid derivatives have been widely utilized in
bioorganic chemistry, for example as mechanistic
probes15 and inhibitors16 of peptidases. Methodology
now exists whereby simple unsymmetrical phosphonate
diesters and phosphonamidate esters can be obtained in
`one-pot' with good selectivity, using the tetrazole-
catalyzed coupling of a phosphonyl dichloride with
either two di€erent alcohols,17 or an alcohol and an
amine.18 19 However, reported procedures for the
synthesis of highly functionalized mixed a-amino-phos-
phonates and phosphonamidates still employ a more
laborious approach, involving the stepwise derivatiza-
tion of polar mono- or bisphosphonic acids which
themselves require initial puri®cation by ion exchange
or reverse-phase chromatography.20 Consequently, the
demand for a less circuitous route to generate complex
mixed a-amido-phosphonate and phosphonothioates,
such as haptens 6a±b and 7, served as a pivotal chal-
lenge within this project. Herein we describe a four-step,
one-pot synthesis of mixed a-azido-phosphonate and
phosphonothioate esters (Scheme 2), and demonstrate
their smooth transformation into the respective a-amino
and a-amido counterparts.
Recently, the generation of peptides with native back-
bones has been achieved using an advanced chemical
ligation process which combines a facile transthioester-
i®cation reaction with a spontaneous rearrangement to
give an N-substituted amide precursor of the latent
peptide bond.7 9 We are seeking to exploit the known
transesteri®cation capabilities of antibodies10,11 to facil-
itate a similar ligation approach. By targeting this type
of cascade process, the desired product of antibody cat-
alysis is a labile intermediate in the reaction sequence
and thus its recognition should not inhibit turnover.
The model system outlined in Scheme 1 has been
devised to examine the feasibility of our strategy.12
The transesteri®cation reaction between (thio)ester 1a±b
and thiol 2 to generate the intermediate 3 with con-
comitant release of 4a or b represents the rate deter-
mining step in the formation of the ligation product 5.
Consequently, the mixed a-amidophosphonate diesters
*Corresponding author. Tel.: +1-619-784-2516; fax: +1-619-784-
2595; e-mail: kdjanda@scripps.edu
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00137-2

916
C. W. Harwig et al. / Bioorg. Med. Chem. Lett. 10 (2000) 915±918
Scheme 1. Substrates 1a±b and 2 and haptens 6a±b and 7 were employed in a catalytic antibody approach to chemical ligation via transesteri®cation.
Table 1. Four-step, one-pot approach to mixed a-azido-phosphonate and phosphonothioate diesters 12a±d
Entry
1
RXH
R0OH
Product
Yield (%)a
24
31P NMRb (ppm)
55.8(s), 55.2(s)
12a
2
3
12b
12c
25
21
57.9(s), 57.1(s)
23.7(s), 23.3(s)
4a23
13a23
4a23
4
12d
29
54.9(s), 54.5(s)
13b23
aIsolated yield after silica gel chromatography.
bRecorded on a Bruker AMX 400 NMR machine at 101 MHz; chemical shifts of diastereomers are reported down®eld from an external reference
(phosphoric acid).
Diethyl 1-hydroxy-2-phenethylphosphonate 8 was treated
with hydrazoic acid under Mitsunobu conditions to
generate the key phosphorus building block, azide 9.21
Then, in one reaction vessel azide 9 was ®rst treated
with excess bromotrimethylsilane22 to generate the die-
ster 10 which, in turn, was converted into phosphonyl
dichloride 11 using oxalyl chloride with a catalytic
amount of DMF.23 The coupling to 11 was initiated, in
the same pot, by the addition of either an alcohol or thiol
in the presence of 1H-tetrazole and DIPEA. After 15 min
the second alcohol was added and the reaction mixture
was stirred for a further 16 h. Silica gel chromatography
furnished the mixed a-azido-phosphonothioates 12a,b,d
and phosphonates 12c as diastereomeric mixtures in
respectable yields (Table 1).24 Easily accessible thiols
and alcohols were employed to assess the utility of this
approach (Table 1, entries 1 and 2) before the one-pot
coupling procedure was performed using synthetic
components 4a and 13a/b to generate hapten precursors
12c/d (Table 1, entries 3 and 4), respectively.25 This
multistep process was far superior to existing stepwise
methodology which, in our hands, furnished very poor
yields of 12a±d (0±10%). In addition, the polarity of the
monophosphonic acid intermediates generated during
the stepwise approach made their isolation troublesome.
The a-azido-phosphonate 12c was smoothly reduced to
its corresponding a-amino derivative using 1,3-propane-
thiol26 and subsequent acylation with glutaric anhy-
dride a€orded hapten 6a in acceptable yield (45%).
Hapten 7 was completed using an analogous reduction±
Scheme 2. Synthesis of a-azido-phosphonates and phosphonothioates.
Reagents and conditions: (i) hydrazoic acid, DEAD, PPh3, (ii)
TMSBr, (iii) (COCl)2, DMF (cat), CH2Cl2 (iv) iPr2NEt, 1H-tetrazole
(cat), CH2Cl2, RXH, (v) R0OH. Steps (ii)±(v) are carried out in `one-
pot'.

C. W. Harwig et al. / Bioorg. Med. Chem. Lett. 10 (2000) 915±918
917
acetylation transformation of 12c followed by attach-
ment of a six-carbon linker.27 The more robust azide
moiety in 12d was reduced with tin(II) chloride28 and
subsequent treatment with glutaric anhydride furnished
hapten 6b in 15% yield.
2, but their catalyzed-rates do not merit extensive
kinetic study. By incorporating the facile synthetic
procedures reported in this paper, we are at present
preparing new haptens in an e€ort to generate more
powerful antibody ligases.
The (thio)ester substrates 1a±b were formed by EDC
mediated esteri®cation of d,l-phenylalanine with 4a±b,
Acknowledgements
respectively.29 The thiol substrate 2 was synthesized
30
.
from S-(2-pyridylthio)cysteamine HCl 14 in two steps
This research was supported by the NIH (GM4385) and
the Skaggs Institute for Chemical Biology. The authors
thank Dr. Paul Wentworth, Jr. for his help during the
preparation of this manuscript.
(Scheme 3). Alkylation of 14 with t-butyl bromoacetate
in the presence of DIPEA yielded 15 (70%) which was
.
then treated with tris-(2-carboxyethyl)phosphine HCl
(TCEP) to give substrate 2. For ease of handling, thiol 2
was immediately converted into its corresponding di-
sul®de 16 (80%) by exposure to air. A fresh stock of 2
was prepared immediately prior to kinetic assays by in
situ treatment of 16 with an aqueous solution of TCEP.
References and Notes
1. Schnolzer, M.; Kent, S. B. H. Science 1992, 256, 221.
2. Abrahmsen, L.; Tom, J.; Burnier, J.; Butcher, K. A.;
Kossiako€, A.; Wells, J. A. Biochemistry 1991, 30, 4151.
3. Hirschmann, R.; Smith, III, A. B.; Taylor, C. M.; Benkovic,
P. A.; Taylor, S. D.; Yager, K. M.; Sprengeler, P. A.;
Benkovic, S. J. Science 1994, 265, 234.
4. Smithrud, D. B.; Benkovic, P. A.; Benkovic, S. J.; Taylor,
C. M.; Yager, K. M.; Witherington, J.; Philips, B. W.;
Sprengeler, P. A.; Smith, III, A. B.; Hirschmann, R. J. Am.
Chem. Soc. 1997, 119, 278.
5. Jacobsen, J. R.; Schultz, P. G. Proc. Natl. Acad. Sci. USA
1994, 91, 5888.
6. Lee, D. H.; Severin, K.; Yokobayashi, Y.; Ghadiri, M. R.
Nature (London) 1997, 390, 591.
7. Wieland, T.; Bokelmann, E.; Bauer, L.; Lang, H. U.; Lau,
H. Liebigs Ann. Chem. 1953, 583, 129.
8. Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B. H.
Science 1994, 266, 776.
9. Canne, L. E.; Bark, S. J.; Kent, S. B. H. J. Am. Chem. Soc.
1996, 118, 5891.
10. Wirsching, P.; Ashley, J. A.; Benkovic, S. J.; Janda, K. D.;
Lerner, R. A. Science 1991, 252, 680.
11. Lin, C.-S.; Ho€man, T. Z.; Xie, Y.; Wirsching, P.; Janda,
K. D. Chem. Commun. 1998, 1075.
12. This model system based on the work of Kent et al.9
allows a simple demonstration of the trans(thio)esterication-
rearrangement approach to chemical ligation.
Monoclonal antibodies with speci®c recognition of
haptens 6a±b or 7 were generated using established
protocols.31 These puri®ed immunoglobulins were
screened initially for their ability to accelerate the
release of 4a±b from esters 1a±b in the presence of thiol
2 at pH 7.0 (100 mM MOPS, 2.5% DMSO).32 Twelve
antibodies capable of catalyzing the expulsion of 4b
were further investigated to ascertain the nature of their
activity. By monitoring the formation of amide 5 (the
spontaneous rearrangement product of 3) during the
reaction between 1b and 2 in the presence of the anti-
bodies, it was possible to distinguish thioesterase and
transthioesterase mechanisms. This analysis revealed
that ®ve antibodies, three raised against 6a and two
raised against 6b, catalyze the transesteri®cation pro-
cess. The remaining seven antibodies simply catalyze the
hydrolysis of 1b. Preliminary kinetic studies have
revealed that the overall antibody-catalyzed rates (ca.
102) for the generation of 5 are encouraging but do not
warrant in-depth investigation.
In conclusion, we have developed a facile four-step,
one-pot coupling procedure for the synthesis of novel a-
azido-phosphonate and phosphonothioate diesters.
Representative examples 12c±d have been exploited as
intermediates for synthesis of haptens 6a±b and 7. Five
of the antibodies generated against haptens 6a±b cata-
lyze the transthioesteri®cation reaction between 1b and
13. Wentworth, Jr., P.; Janda, K. D. Curr. Opin. Chem. Biol.
1998, 2, 138.
14. It is well known, from structural studies, that antibody
recognition is minimal at the site of linker attachment and that
epitopes directly opposite to the linker can be deeply buried in
the antibody binding site: Haynes, M. R.; Stura, E. A.;
Hilvert, D.; Wilson, I. A. Science 1994, 263, 646.
15. Camp, N. P.; Hawkins, P. C. D.; Hitchcock, P. B.; Gani,
P. Bioorg. Med. Chem. Lett. 1992, 2, 1047.
16. Bartlett, P. A.; Marlowe, C. K. Science 1987, 235, 569.
17. Zhao, K.; Landry, D. W. Tetrahedron 1993, 49, 363.
18. Mlodnosky, K. L.; Holmes, H. M.; Lam, V. Q.; Berkman,
C. E. Tetrahedron Lett. 1997, 38, 8803.
19. Yang, G.; Zhao, K.; Landry, D. W. Tetrahedron Lett.
1998, 39, 2449.
20. Hirschmann, R.; Yager, K. M.; Taylor, C. M.; Witherington,
J.; Sprengeler, P. A.; Phillips, B. W.; Moore, W.; Smith, III, A. B.
J. Am. Chem. Soc. 1997, 119, 8177 (and references therein).
21. Gajda, T.; Matusiak, M. Synthesis 1992, 367. Introduction
of the latent a-amino functionality masked as an azide group
proved to be critical to the success of this coupling procedure.
22. McKenna, C. E.; Higa, M. T.; Cheung, N. H.; McKenna,
M.-C. Tetrahedron Lett. 1977, 155.
Scheme 3. Substrate synthesis. Reagents and conditions: (i) t-butyl
bromoacetate, DIPEA, DMF, (ii) TCEP, EtOAc±H2O, (iii) TCEP,
H2O.

918
C. W. Harwig et al. / Bioorg. Med. Chem. Lett. 10 (2000) 915±918
23. Removal of volatiles under a stream of argon furnished the
crude intermediates 10 and 11. 31P NMR was used to monitor
their formation: 10 (d 15.8 ppm) and 11 (d 47.2 ppm).
24. Only minor amounts (< 5%) of the symmetrically sub-
stituted phosphonate diesters were observed.
28. Maiti, S. N.; Singh, M. P.; Micetich, R. G. Tetrahedron
Lett. 1986, 27, 1423.
29. Thiol 4b was prepared from alcohol 4a25 in three steps
(71% overall). Alcohol 4a was activated as its alkyl iodide
(PPh3, I2, imidazole), before thiolation with sodium triisopro-
pylsilyl thiolate. The silyl protecting group was removed with
TFA.
25. Alcohol 4a was prepared by reduction of 4-methylsulfonyl-
benzaldehyde with NaBH4 in ethanol (95%). Alcohol 13a was
prepared in an overall yield of 14% by initial monoprotection of
ethylene glycol (NaH, TBDMSCl), followed by coupling with
t-butyl a-bromoacetate (50% aq NaOH, phase tranfer catalyst
Bu4NHSO4) with removal of the TBDMS ether (TBAF) in the
®nal step. Alcohol 13b was prepared in three steps also (63%
overall). The alcohol formed by the reaction of triphenyl-
methylmercaptan with 2-bromoethanol was etheri®ed using t-
butyl a-bromoacetate (vide supra). Finally, the S-trityl group
was removed with AgNO3. (Gupta, K. C.; Sharma, P.; Kumar,
P.; Sathyanarayana, S. Nucleic Acids Res. 1991, 19, 3019).
26. Bayley, H.; Standring, D. N.; Knowles, J. R. Tetrahedron
Lett. 1978, 18, 3633.
27. Hapten 7 was synthesized from 12c in ®ve steps (31%
overall). After reduction of 12c (propanethiol), acetylation was
achieved using acetic anhydride (pyridine, DMAP). Removal
of the t-butyl ester (TFA) allowed attachment of the protected
linker t-butyl 6-aminohexanoate (EDC, HOBT, DIPEA).
Deprotection of the resulting ester (TFA) furnished 7. t-Butyl
6-aminohexanoate was synthesized in 74% yield from 6-amino-
hexanoic acid using benzyloxycarbonyl protection of the
amine during t-butyl ester formation.
30. Ebright, Y. W.; Chen, Y.; Pendergrast, P. S.; Ebright, R.
H. Biochemistry 1992, 31, 10664.
31. KLH conjugates of 6a±b and 7 were used to immunize
129Gix mice. Monoclonal antibodies were generated using
hybridoma technology (Kohler, G.; Milstein, C. Nature 1975,
256, 495). Twenty-two monoclonal antibodies were elicited to
KLH-6a, 25 monoclonal antibodies were elicited to KLH-6b,
and 21 monoclonal antibodies were elicited to KLH-7.
32. In a typical screening assay, the disul®de 16 was mixed in
equimolar amounts with TCEP in water for 5 min. To initiate
the assay, this mixture was added to the aqueous bu€er system
[100 mM Bicine (pH 8.0), containing the esters or thioesters
1a±d (500 mM) and 2.5% DMSO in the presence or absence
of antibody (20 mM)] to give an initial thiol 2 concentration of
1 mM. The rates of reaction were measured by monitoring the
rate of formation of 4a±b (or 5) by HPLC (adsorbosphere-HS
column, acetonitrile:water (0.1% TFA) mobile phase, detec-
tion at 230 nm). The antibody-mediated rates (for < 5% of
reaction, during which the progress curves were linear) were
directly compared with the observed rate for the non-catalyzed
reaction under identical assay conditions.