Synthesis of a (+)-anatoxin-a analogue for monoclonal antibodies production

Article abstract of DOI:10.1016/j.tetlet.2009.05.094

Anatoxin-a analogue was prepared by resolution of aminoacid 1 using chiral (R)-4-phenyl-oxazolidin-2-thione as derivatizing agent. X-ray diffraction of a diastereomer allowed us to determine its absolute configuration. The synthesis could then be complete

Full text of DOI:10.1016/j.tetlet.2009.05.094

Tetrahedron Letters 50 (2009) 4554–4557
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthesis of a (+)-anatoxin-a analogue for monoclonal antibodies production
Mickael Marc a, Francis Outurquin a, Pierre-Yves Renard a, Christophe Créminon b, Xavier Franck a,
*
a UMR-CNRS6014 & FR3038 COBRA, Université de Rouen, INSA de Rouen, IRCOF, 1 rue Tesnière, 76130 Mont Saint Aignan, France
b CEA iBiTecS, Service de Pharmacologie et d’Immunoanalyse, Gif sur Yvette F-91191, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Anatoxin-a analogue was prepared by resolution of aminoacid 1 using chiral (R)-4-phenyl-oxazolidin-2-
thione as derivatizing agent. X-ray diffraction of a diastereomer allowed us to determine its absolute
configuration. The synthesis could then be completed in few steps followed by introduction of an amide
linker bearing a terminal alkyne in order to attach a carrier protein for monoclonal antibodies production.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 20 April 2009
Revised 20 May 2009
Accepted 26 May 2009
Available online 29 May 2009
Keywords:
Anatoxin, Aminoacid resolution
Monoclonal antibodies
Toxin analogue
Anatoxin-a is a highly toxic secondary amine alkaloid produced
by cyanobacteria (blue green algae). Anatoxin’s toxicity relies on
its ability to bind to nicotinic acetylcholine receptors (agonist)
with a lethal dose LD50 of 0.2–0.5 mg/kg IP in mice, it is also de-
signed as Very Fast Death Factor (VFDF). The presence of this toxin
in drinking waters is responsible for water poisoning, thus creating
a major threat on public health and making detection and quanti-
fication of anatoxin-a in water a big issue.1 For such a purpose,
monoclonal antibody technology can be very useful for the devel-
opment of a detection assay. The production of monoclonal anti-
bodies that are able to bind to anatoxin requires the
immunization of mice with an hapten composed of a protein (usu-
ally BSA or KLH) and the toxin or a close analogue. In order to focus
mAbs recognition onto this small hapten structure, we decided to
use the extracyclic ketone as the anchor for the attachment of
the carrier protein. Yet, two key structural features on anatoxin-a
are detrimental towards the commonly used techniques for the
bio-conjugation of this small hapten onto carrier proteins: the
key secondary and reactive amine (thus any attachment using acti-
vated esters or Schiff base formation are unappropriate), and the
presence of an activated alpha-beta unsaturated ketone, prevent-
ing the use of soft nucleophiles such as thiols. We thus decided
to undertake the synthesis of an anatoxin analogue 12 that will
be attached onto the carrier protein through click chemistry
(Fig. 1).2
mented reaction,3 thereafter, to attach the analogue with the pro-
tein by click chemistry, we decided to functionalize the analogue
with a terminal alkyne and link this alkyne to the anatoxin core
by an amide function. In order to have the partner azide function
on the carrier protein, BSA will be modified prior to the bio-conju-
gation with readily obtained activated ester of 6-azido hexanoic
acid (which could be obtained in one step from commercially avail-
able 6-bromohexanoic acid).
In order to stress the immunological response on the bicyclic
core and hence the mab recognition, we chose to replace the meth-
ylketone of anatoxin by a disubstituted amide, thus preventing the
planar arrangement of the C@O and the intracyclic double bond in
order to (1) have substantial difference in the side chain compared
to anatoxin and (2) lower the toxicity of the hapten caused by an
irreversible linkage to nicotinic acetylcholine receptors through
1,4-addition on the
a
,b-unsaturated carbonyle function.
O
H
O
H
N
N
5
N
anatoxin-a
analogue 12
N
N
N
Among those bioorthogonal bio-conjugation means, the [3+2]
Huisgen cycloaddition reaction is by far the most used and docu-
protein
O
N
H
N
5
hapten
* Corresponding author. Tel.: +33 0 2 35522403; fax: +33 0 2 35522959.
E-mail address: xavier.franck@insa-rouen.fr (X. Franck).
Figure 1. Anatoxin-a and analogue.
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.05.094

M. Marc et al. / Tetrahedron Letters 50 (2009) 4554–4557
4555
Anatoxin-a posseses an azabicyclo-[4.2.1] nonane ring system
that has led to many racemic and enantioselective approaches.4
Among these approaches, the construction of the azabicyclo-
[4.2.1] nonane ring starting from a cyclooctene ring by aminocyc-
lisation is one of the most popular but leads usually to racemic
syntheses.4c,i,5 Recently Fülöp described a deracemisation of a
cyclooctene fused b-lactam by enzymatic resolution but did not
complete the synthesis.6 We herein report on the construction of
the azabicyclo-[4.2.1] nonane ring by resolution of racemic amino-
acid using chiral auxiliaries such as oxazolidin-2-thiones and the
completion of the synthesis of an anatoxin analogue.
The racemic b-aminoacid 1 was readily prepared from the cor-
responding b-lactam after methanolysis, Boc-protection of the
amine and saponification. Then resolution was done by derivatiza-
tion of the acid with chiral (R)-4-phenyl-oxazolidin-2-thione and
separation of diastereomers 3a and 3b by simple chromatography
on silica (Scheme 1). Fortunately, we were able to obtain X-ray
structure of diastereomer 3b that allowed us to determine the
absolute configurations of the aminoacid moiety (Fig. 2).7
Having identified isomer 3b as 1R,8S isomer, we selected isomer
3a (1S,8R) to continue the synthesis of anatoxin analogue, as the
amino group has the right (R) absolute configuration. Thus, 3a
was converted into its methyl ester 4 (imidazole, MeOH and DMAP
cat.) and subjected to selenium-assisted aminocyclisation, giving
the azabicyclo-[4.2.1] nonane 5 as a single regioisomer4i in 75%
yield. Instead of removing the selenium residue by oxidative elim-
ination followed by hydrogenolysis (low yielding), we found that
Raney Nickel smoothly converted 5 into its reduced derivative 6
in a quantitative yield, thus greatly improving the overall yield of
the transformation (Scheme 2).
Figure 2. X-ray structure of diastereomer 3b.
S
Starting from ester 6, two different strategies were assessed in
order to access the targeted unsaturated amide 12.8 The first one
envisioned the early formation of the amide before the introduc-
tion of the unsaturation, whereas the second one relied on the
early introduction of the unsaturation before forming the amide
bond (Scheme 3).
O
O
O
BocHN
N
BocHN
OMe
MeOH, imidazole
DMAP cat. (89%)
(R)
(R)
Ph
(S)
(S)
3a
4
Following the first way, after saponification of ester 6 to carbox-
ylic acid 7, amide 8 was obtained by coupling with secondary
amine 13 in the presence of coupling agent PyBrOP8 in 54% yield
(no reaction occurred with DCC). Secondary amine 13 was pre-
pared in 5 steps from octynol after silylation of the alkyne, mesy-
lation and iodination of the hydroxyle followed by displacement
with methylamine (29% overall yield). Unfortunately, all attempts
PhSeBr, Na2CO3,
MeCN (92%)
O
O
Boc
N
Boc
N
OMe
OMe
Raney Ni, MeOH (quant.)
6
5
to introduce the unsaturation by a-phenylselenenylation of amide
PhSe
Scheme 2. Synthesis of ester 6.
8 proved to be unsuccessful. Indeed, enolization at ꢀ78 °C followed
by addition of PhSeBr led only to the recovery of starting material,
whereas reaction performed at 0 °C gave only degradation prod-
ucts. Therefore, we turned our attention to the second strategy.
Introduction of the unsaturation was thus performed prior to
the introduction of the amide side chain. -phenylselenenylation
a
of 6 (LDA and PhSeBr) followed by oxidative elimination (H2O2
and NaHCO3) gave ester 9 in a modest 25% yield over these two
steps (low yields were always obtained for b-elimination whatever
the conditions). Saponification of the methyl ester yielded the car-
boxylic acid 10 in almost quantitative yield. Introduction of the
amide function proved again to be difficult and only PyBrOP8 al-
lowed us to obtain 11 in a modest 40% yield. Thereafter, targeted
anatoxin-a analogue 12 could be obtained after desilylation and
Boc deprotection through TFA treatment in 78% yield for the last
two steps (Scheme 3).9–11
The synthesis of (+)-anatoxin-a analogue 12 has been accom-
plished by resolution of racemic aminoacid 1 and selenium-as-
sisted cyclisation. Its terminal alkyne function will allow the
linkage with a protein before production of monoclonal antibodies.
This last step is now under way and the results will be reported in
due course.
1) ClSO2NCO
2) MeOH/HCl
3) Boc2O
BocHN
COOH
4) NaOH
1
S
O
HN
Ph
DCC/DMAP
S
2
S
O
O
O
N
O
BocHN
BocHN
(S)
N
(R)
Ph
Ph
)
(
(S)
R
+
3b 38%
3a 49%
Scheme 1. Resolution of the racemic aminoacid 1.

4556
M. Marc et al. / Tetrahedron Letters 50 (2009) 4554–4557
O
O
O
Boc
N
Boc
Boc
N
1) LDA, PhSeBr (76%)
2) H2O2, NaHCO3 (31%)
N
OMe
OH
OMe
LiOH then H+
7
9
6
SiMe3
NaOH, THF
then H+
N
13
54%
5
PyBroP, DIEA, MeCN
SiMe3
H
SiMe3
O
Boc
N
O
SiMe3
O
Boc
N
Boc
N
N
OH
N
5
5
N
1) LDA, PhSeBr
2) H2O2
5
H
PyBroP, DIEA, MeCN
40%
10
8
11
1) K2CO3/MeOH
2) TFA
78%
H
O
H
N
N
5
12
1) BuLi
1) MsCl/Et3N
2) NaI/acetone/60°C
MeNH2
THF, 60°C
SiMe3
SiMe3
2) TMSCl then H+
H
SiMe3
N
HO
HO
I
5
13
5
5
5
42%
83%
H
83%
Scheme 3. Synthesis of (+)-anatoxin-a analogue 12.
175.5, 154.6, 138.1, 130.1, 129.2, 128.9, 128.7, 126.0, 79.2, 73.7, 62.4, 50.0, 44.2,
33.1, 28.4, 26.5, 25.8 23.7; IR (CH2Cl2): 3428, 2932, 2360, 1714, 1694, 1504,
Acknowledgements
m
1455, 1367, 1161, 1086, 1019, 737, 699 cmꢀ1; MS (ESI): m/z 431 ([M+H]+), 453
([M+Na]+); ½a 2D0
ꢀ84.8 (c 1.0, CHCl3).
ꢁ
We gratefully acknowledge the Direction Générale de l’Arme-
ment (DGA) for a fellowship to M.M., and Région Haute-Normandie
for financial support.
Compound 3b: (Rf = 0.36); 1H NMR: (300 MHz, CDCl3) d 7.30–7.16 (m, 5H),
5.70–5.62 (m, 1H), 5.57–5.49 (m, 1H), 5.41 (br d, J = 7.3 Hz, 1H), 5.09 (br d,
J = 11.3 Hz, 1H), 4.79 (t, J = 8.4 Hz, 1H), 4.70 (d, J = 9.9 Hz, 1H), 4.31 (dd, J = 8.4,
2.2 Hz, 2H), 3.50–2.20 (m, 2H), 2.12–1.81 (m, 3H), 1.72–1.52 (m, 3H), 1.37 (s,
9H); 13C NMR : (75 MHz, CDCl3) d 186.0, 175.7, 155.4, 139.4, 131.5, 129.2,
128.6, 127.2, 126.0, 79.5, 74.3, 63.8, 51.5, 43.4, 33.4, 30.2, 28.5, 26.0, 25.0; IR
References and notes
(KBr):
m
3386, 2975, 2940, 2888, 1691, 1654, 1515, 1367, 1152, 974 cmꢀ1; MS
(ESI) : m/z 431 ([M+H]+); ½a D20
ꢁ
21.2 (c 1.0, CHCl3).
1. (a) Selwood, A. I.; Holland, P. T.; Wood, S. A.; Smith, K. F.; McNaab, P. S. Environ.
Sci. Technol. 2007, 41, 506–510; (b) Spivak, C. E.; Witkop, B.; Albuquerque, E. X.
Mol. Pharmacol. 1980, 18, 384–394.
2. Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed. 2001, 40, 2004–
2021.
Compound 4: 1H NMR: (300 MHz, CDCl3) d 5.73–5.57 (m, 2H), 5.03 (br d, J =
8.2 Hz, 1H), 4.14–4.09 (m, 1H), 3.70 (s, 3H), 2.84–2.82 (m, 1H), 2.49–2.42 (m,
1H), 2.30–2.21 (m, 1H), 2.18–2.08 (m, 2H), 2.07–1.93 (m, 1H), 1.89–1.71 (m,
3H), 1.41 (s, 9H) ; 13C NMR : (75 MHz, CDCl3) d 174.8, 155.1, 130.5, 129.3, 79.3,
51.7, 50.2, 48.1, 33.0, 28.5, 27.4, 24.5, 23.5; IR (CH2Cl2):
1714, 1694, 1504, 1455, 1366, 1249, 1170 cmꢀ1; MS (ESI) m/z 284 ([M+H]+);
+55.7 (c 1.0, CHCl3).
Compound
(2 rotamers 66:34): 1H NMR: (300 MHz, CDCl3) d 4.69 (d,
m 3372, 2934, 2360,
3. Medal, M.; Torn, C. W. Chem. Rev. 2008, 2952–3015.
4. For recent syntheses see: (a) Roe, S. J.; Stockman, R. A. Chem. Commun. 2008, 29,
3432–3434; (b) Tomita, T.; Kita, Y.; Kitamura, T.; Sato, Y.; Mori, M. Tetrahedron
2006, 62, 10518–10527; (c) Ho, T.-L.; Zinurova, E. Helv. Chim. Acta 2006, 89,
134–137; (d) Hjelmgaard, T.; Søtofte, I.; Tanner, D. J. Org. Chem. 2005, 70, 5688–
5697; (e) Muniz, M. N.; Kanazawa, A.; Greene, A. E. Synlett 2005, 1328–1330; (f)
Brenneman, J. B.; Martin, S. F. Org. Lett. 2004, 6, 1329–1331; (g) Brenneman, J.
B.; Machauer, R.; Martin, S. F. Tetrahedron 2004, 60, 7301–7314; (h) Wegge, T.;
Schwarz, S.; Seitz, G. Tetrahedron: Asymmetry 2000, 11, 1405–1410; (i) Parsons,
P. J.; Camp, N. P.; Edwards, N.; Sumoreeah, L. R. Tetrahedron 2000, 56, 309–315;
(j) Aggarwal, V. K.; Humphries, P. S.; Fenwick, A. Angew. Chem., Int. Ed 1999, 38,
1985–1986; (k) Trost, B. M.; Oslob, J. D. J. Am. Chem. Soc. 1999, 121, 3057–3064.
5. (a) Ho, C.-Y.; Kim, K.-S.; Ham, W.-H. Tetrahedron Lett. 1998, 39, 2133–2136; (b)
Parsons, P. J.; Camp, N. P.; Underwood, J. M.; Harvey, D. M. Tetrahedron 1996,
52, 11637–11642; Parsons, P. J.; Camp, N. P.; Underwood, J. M.; Harvey, D. M.
Chem. Commun. 1995, 14, 1461–1462.
6. Forró, E.; Árva, J.; Fülöp, F. Tetrahedron: Asymmetry 2001, 12, 643–649.
7. Cambridge crystallographic data base access for 3b: CCDC 724405.
8. Coste, J.; Frerot, E.; Jouin, P. J. Org. Chem. 1994, 59, 2437–2446.
9. 1H NMR data of compound 12 are similar to literature data concerning
anatoxin,10 except mainly for H-1 which suffers from a shielding due to the
presence of the amide moiety as its carbonyl is not planar with the internal
C@C bond. The broadness of the signals is also due to the presence of this
moiety.
½ ꢁ
a 2D0
6
J = 8.8 Hz, 0.34H), 4.52 (d, J = 9.2 Hz, 0.66H), 4.35 (d, J = 8.8 Hz, 0.66H), 4.21–
4.13 (m, 0.34H), 3.70, (s, 1H), 3.68 (s, 2H), 2.47–2.29 (m, 2H), 2.10–1.64 (m,
6H), 1.62–1.53 (m, 3H), 1.42 (s, 3H), 1.39 (s, 6H); 13C NMR: (75 MHz, CDCl3)
d 175.0 (mino), 174.8 (majo), 153.5, 79.6 (majo), 79.1 (mino), 57.2 (mino), 56.8
(majo), 56.3, 53.3, 52.1 (mino), 51.7 (majo), 35.5 (majo), 34.1 (mino), 33.7
(mino), 32.8 (majo), 28.6 (mino), 28.5 (majo), 27.0, 26.8 (mino), 26.5 (majo),
21.9 (majo), 21.8 (mino); IR (CH2Cl2):
m 3367, 2934, 1738, 1693, 1404, 1173,
1112 cmꢀ1; MS (ESI) : m/z 284 ([M+H]+); ½a 2D0
ꢀ30.7 (c 1.07, CHCl3).
ꢁ
Compound 9 (2 rotamers 70:30): 1H NMR: (300 MHz, CDCl3) d 6.97 (t, J = 5.9 Hz,
1H), 5.14 (d, J = 7.5 Hz, 0.3H), 5.02 (d, J = 8.8 Hz, 0.7H), 4.40–4.38 (m, 0.7H), 4.31–
4.22 (m, 0.3H), 3.73 (s, 3H), 2.42–2.23 (m, 2H), 2.19–2.00 (m, 3H), 1.78–1.53
(m, 3H), 1.44 (s, 2.7H), 1.39 (s, 6.3H); 13C NMR : (75 MHz, CDCl3) d 167.3 (majo),
167.1 (mino), 153.3 (majo), 153.2 (mino), 141.9 (majo), 141.5 (mino), 140.7
(majo), 138.4 (mino), 79.4, 55.7 (majo), 55.4 (mino), 55.1 (mino), 54.7 (majo),
52.0, 32.7 (mino), 31.7 (majo), 31.1 (mino), 30.5 (majo), 30.0, 28.8 (mino), 28.6
(majo), 24.0; IR (CH2Cl2):
m 3372, 2975, 1714, 1694, 1404, 1366, 1237, 1170,
1111, 770 cmꢀ1; MS (ESI) m/z 282 ([M+H]+); ½a 2D0
ꢀ18.4 (c 1.0, CHCl3).
ꢁ
Compound 10 (2 rotamers 55:45): 1H NMR: (300 MHz, CDCl3) d 7.12 (t, J = 6.0 Hz,
0.55H), 6.88 (t, J = 6.1 Hz, 0.45H), 5.03 (d, J = 8.5 Hz, 0.55H), 4.91 (d, J = 8.3 Hz,
0.45H), 4.42 (br d, J = 8.5 Hz, 0.55H), 4.38–4.30 (m, 0.45H), 2.46–2.41 (m, 2H),
2.35–2.04 (m, 2H), 2.00–1.58 (m, 4H), 1.46 (s, 5H), 1.41 (s, 4H); 13C NMR:
(75 MHz, CDCl3) d 171.4, 154.3 (mino), 153.3 (majo), 144.5, 141.9 (majo), 140.1
(mino), 80.7 (majo), 79.7 (mino), 56.4 (mino), 55.7 (majo), 54.9 (mino), 54.4
(majo), 31.7 (majo), 31.5 (mino), 31.1, 30.9 (majo), 30.4 (mino), 28.6 (majo), 28.5
10. Koskinen, A. M. P.; Rapoport, H. J. Med. Chem. 1985, 28, 1301–1309.
11. Spectral data of selected compounds: Diastereomers 3a and 3b were separated
by column chromatography on silica using cyclohexane/EtOAc (8:2) as the
solvent.
(mino), 24.2 (majo), 23.8 (mino); MS (ESI): m/z 268 ([M+H]+); ½a D20
ꢀ17.9 (c 1.0,
ꢁ
Compound 3a: (Rf = 0.27); 1H NMR: (300 MHz, CDCl3) d 7.41–7.28 (m, 5H),
5.76–5.58 (m, 3H), 5.14 (br d, J = 7.3 Hz, 1H), 4.77 (t, J = 9.0 Hz, 1H), 4.43 (m,
1H), 4.38 (dd, J = 9.0, 4.5 Hz, 1H), 4.27 (m, 1H), 2.50–2.40 (m, 2H), 2.20–2.03
(m, 3H), 1.94–1.78 (m, 3H), 1.35 (s, 9H); 13C NMR: (75 MHz, CDCl3) d 185.2,
MeOH).
Compound 11 (2 rotamers 55:45): 1H NMR: (300 MHz, CDCl3) d 5.70 (t, J = 5.2 Hz,
0.55H), 5.58 (t, J = 5.2 Hz, 0.45H), 4.50 (d, J = 7.7 Hz, 1H), 4.37 (m, 0.55H), 4.23

M. Marc et al. / Tetrahedron Letters 50 (2009) 4554–4557
4557
(m, 0.45H), 2.97–2.83 (m, 3H), 2.28–2.22 (m, 3H), 2.15–2.12 (m, 4H), 2.00–1.95
(m, 2H), 1.75–1.55 (m, 3H), 1.53–1.42 (m, 4H), 1.45 (s, 4.5H), 1.40 (s, 4.5H), 1.30–
1.15 (m, 4H), 0.08 (s, 9H); 13C NMR : (75 MHz, CDCl3) d 171.7, 153.2 (mino), 152.8
(majo), 144.3, 143.9, 130.5 (majo), 128.4 (mino), 107.4, 84.4, 84.3, 79.4, 79.1,
77.4, 60.4, 57.3, 55.4, 51.0, 46.9, 37.6 (mino), 37.2 (majo), 32.6 (mino), 32.3
(majo), 31.4, 29.5, 28.5 (majo), 28.2 (mino), 26.8 (mino), 26.3 (majo), 23.5, 21.1
Compound 12 (TFA salt): 1H NMR: (300 MHz, CDCl3) d 9.81 (br s, 1H), 9.17 (br s,
1H), 6.12 (t, J = 5.2 Hz, 1H), 4.43 (br s, 1H), 4.29 (br s, 1H), 3.40–3.20 (m, 2H),
3.05–2.82 (m, 3H), 2.57–2.41 (m, 3H), 2.39–2.25 (m, 1H), 2.18–2.12 (m, 4H),
2.00–1.88 (m, 2H), 1.85–1.73 (m, 1H), 1.55–1.44 (m, 4H), 1.41–1.33 (m, 2H),
1.31–1.14 (m, 2H); 13C NMR: (75 MHz, CDCl3) d 170.5, 161.4 (q, TFA), 137.5,
136.4, 84.5, 68.4, 59.0, 57.1, 47.8, 37.7, 31.6, 28.4, 28.3, 27.7, 26.9, 26.6, 26.3, 23.2,
(mino), 19.8 (majo), 14.2, 0.2; IR (CH2Cl2):
m
2932, 2860, 2173, 1693, 1621, 1404,
18.3; IR (CH2Cl2): m 3411, 2943, 1694, 1674, 1651, 1615, 1455, 1206, 1138, 845,
1364, 1248, 1172, 1110, 842 cmꢀ1; MS (ESI) : m/z 461 ([M+H]+); ½a 2D0
ꢁ
+32.9
800, 724 cmꢀ1; MS (ESI): m/z 289.5 ([M+H]+); HRMS calcd for C18H29N2O:
(c 1, CHCl3).
289.2280. Found: 289.2283; ½a D20
ꢁ
+24.7 (c 0.5, CHCl3).