An efficient enzymatic synthesis of benzocispentacin and its new six- and seven-membered homologues

Article abstract of DOI:10.1002/chem.200501286

A very efficient enzymatic method was developed for the synthesis of new enantiomeric benzocispentacin and its six- and seven-membered homologues through the Lipolase (lipase B from Candida antarctica) catalyzed enantioselective (E > 200) ring opening of 3,4-benzo-6-azabicyclo[3.2.0]heptan-7-one, 4,5-benzo-7-azabicyclo[4.2.0]octan-8-one, and 5,6-benzo-8-azabicyclo[5.2.0] nonan-9-one with H₂O in iPr₂O at 60°C. The (1R,2R)-β-amino acids (ee ≥ 96%, yields ≥ 40%) and (1S,6S)-, (1S,7S)-, and (1S,8S)-β-lactams (ee > 99%, yields ≥ 44%) produced could be easily separated. The ring opening of racemic and enantiomeric β-lactams with 18% HCl afforded the corresponding β-amino acid hydrochlorides.

Full text of DOI:10.1002/chem.200501286

FULL PAPER
DOI: 10.1002/chem.200501286
An Efficient Enzymatic Synthesis of Benzocispentacin
and Its New Six- and Seven-Membered Homologues
Eniko˝ Forró and Ferenc Fülçp*[a]
Abstract: A very efficient enzymatic
method was developed for the synthe-
sis of new enantiomeric benzocispenta-
cin and its six- and seven-membered
homologues through the Lipolase
(lipase B from Candida antarctica) cat-
alyzed enantioselective (E>200) ring
opening of 3,4-benzo-6-azabicyclo-
azabicyclo[4.2.0]octan-8-one, and 5,6-
benzo-8-azabicyclo[5.2.0]nonan-9-one
with H2O in iPr2O at 608C. The
yieldsꢁ40%) and (1S,6S)-, (1S,7S)-,
and (1S,8S)-b-lactams (ee>99%,
yieldsꢁ44%) produced could be easily
separated. The ring opening of racemic
and enantiomeric b-lactams with 18%
HCl afforded the corresponding b-
amino acid hydrochlorides.
(1R,2R)-b-amino
acids
(eeꢁ96%,
Keywords: amino acids · cleavage
reactions enantioselectivity
enzymes · lactams
·
·
[3.2.0]heptan-7-one,
4,5-benzo-7-
Introduction
cently discovered a simple and efficient direct enzymatic hy-
drolysis method for the enantioselective (E>200) ring
cleavage of b-lactams in an organic medium (e.g., the syn-
thesis of cispentacin).[11] The great advantages of this
method are that the lactam ring need not necessarily be acti-
vated, and the b-amino acid and b-lactam products are ob-
tained in good chemical yields and can easily be separated.
Shintani and Fu[12] described a copper-catalyzed intramolec-
ular Kinugasa reaction (in the presence of planar chiral
phosphaferrocene-oxazoline as the ligand) and prepared tri-
cyclic b-lactams, our target b-lactam fused to a six-mem-
bered ring being one of them (ee=88%).
Cyclic b-lactams and b-amino acids have been intensively in-
vestigated owing to their potential biological activity (e.g.,
monobactams and cispentacin)[1] and their utility in synthetic
chemistry.[2] For example, they can serve as building blocks
for the synthesis of modified peptides with increased activity
and stability,[3–5] and with well-defined three-dimensional
structures (e.g., b-peptides with possible antibiotic activity)
similar to those of natural peptides.[6] Alicyclic b-amino
acids can also be used in heterocyclic[7] and combinatorial[8]
chemistry.
Several direct and indirect enzymatic methods have been
described for the preparation of enantiopure b-amino acids
or their derivatives. The indirect enzymatic method consist-
ing of the lipase-catalyzed asymmetric acylation of the pri-
mary hydroxy group of N-hydroxymethylated b-lactams, or
the lipase-catalyzed hydrolysis of the corresponding ester
derivatives, followed by ring opening to give the b-amino
ester or acid, respectively, are not too efficient and relatively
long procedures, but they ensure the simultaneous prepara-
tion of both b-lactam enantiomers.[9] The direct b-lactam
ring opening, through lipase-catalyzed enantioselective alco-
holysis, furnishes the b-amino acids in low yields.[10] We re-
Because of the earlier extensive investigations of alicyclic
b-lactams and b-amino acids, the synthesis of new benzocis-
pentacin homologues in both racemic and enantiopure
forms is very attractive. Our present aim was the synthesis
of the title bicyclic b-amino acids through the lipase-cata-
lyzed ring opening of the corresponding b-lactams.
Results and Discussion
Syntheses of (ꢀ)-4, (ꢀ)-5, and (ꢀ)-6: The b-lactams (ꢀ)-4
to (ꢀ)-6 were prepared by 1,2-dipolar cycloaddition of
chlorosulfonyl isocyanate (CSI) to the corresponding cyclo-
alkenes (1–3; Scheme 1), by using slightly modified litera-
ture procedures.[13–15] In order to synthesize the seven-mem-
bered b-lactam (ꢀ)-6, benzosuberene (3), the substrate for
CSI addition, was first prepared from 1-benzosuberone by
means of NaBH4 reduction and subsequent H2O elimina-
[a] Dr. E. Forró, Prof. Dr. F. Fülçp
Institute of Pharmaceutical Chemistry
University of Szeged, 6701 Szeged, PO Box121 (Hungary)
Fax: ( +36)62-545-705
E-mail: fulop@pharm.u-szeged.hu
Chem. Eur. J. 2006, 12, 2587 – 2592
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2587

Scheme 1. Syntheses of compounds (ꢀ)-4 to (ꢀ)-9.
tion.[16] All the reactions took place regio- and stereoselec-
tively: not even traces of other regio- or stereoisomers were
detected. As the racemic six- and seven-membered b-amino
acids are new compounds, the transformations of (ꢀ)-5 and
(ꢀ)-6 by ring opening with 18% HCl were also performed,
resulting in b-amino acid hydrochlorides (ꢀ)-8 and (ꢀ)-9,
respectively (Scheme 1).
Scheme 2. Lipase-catalyzed enantioselective ring opening of compounds
(ꢀ)-4 to (ꢀ)-6.
Lipase-catalyzed enantioselective ring opening of (ꢀ)-4,
(ꢀ)-5, and (ꢀ)-6: The earlier results[11] on the lipase-cata-
lyzed enantioselective hydrolysis of b-lactams suggested the
possibility of the enantioselective ring opening of (ꢀ)-4 to
(ꢀ)-6 with H2O in an organic solvent (Scheme 2).
Table 1. Conversion and enantioselectivity of ring opening of (ꢀ)-5.[a]
[c]
Entry Enzyme
T
H2O
Conv.
ees[b] eep
E
(30 mgmLꢂ1
)
[8C] [equiv] [%]
[%] [%]
1
2
3
4
5
6
7
8
Chirazyme L-2 50
1
–
1
2
4
1
–
1
2
1
1
9
24
24
19
11
27
24
23
10
32
32
23
12
37
31
29
20
>99 >200
>99 >200
>99 >200
>99 >200
>99 >200
>99 >200
>99 >200
>99 >200
>99 >200
Chirazyme L-2 60
Chirazyme L-2 60
Chirazyme L-2 60
Chirazyme L-2 60
Chirazyme L-2 70
We started with the ring cleavage of (ꢀ)-5 with one
equivalent of H2O, and Lipolase (lipase B from Candida ant-
arctica, produced by submerged fermentation of a genetical-
ly modified Aspergillus oryzae microorganism and adsorbed
on a macroporous resin) in iPr2O at 608C (Table 1, entry 8),
but also tested the reactions with Chirazyme L-2 (a carrier-
fixed lipase B from Candida antarctica; Table 1, entry 3) and
CAL-A (lipase A from Candida antarctica; Table 1, en-
tries 10 and 11). High enantioselectivities (E>200) were ob-
served in the cases of Chirazyme L-2 and Lipolase, whereas
CAL-A (neither the free enzyme nor the immobilized prep-
aration) did not exhibit any catalytic activity (no product
was detected after 15 h).
Lipolase
Lipolase
Lipolase
CAL-A
60
60
60
60
60
9
10
11
17
no reaction
no reaction
CAL-A[d]
[a] 0.05m substrate in iPr2O after 15 h. [b] According to GC. s=substrate.
[c] According to GC after double derivatization. p=product. [d] Contains
20% (w/w) of lipase adsorbed on Celite in the presence of sucrose.
The Chirazyme L-2 catalyzed reactions were performed
not only at 608C, but also at 508C (Table 1, entry 1) and
708C (Table 1, entry 6). With increasing temperature, the re-
action rate increased without a drop in enantioselectivity
(E>200).
A set of experiments was performed to determine the
effect of H2O present in the reaction medium on the enzy-
matic activity (Table 1, entries 4, 5 and 9). The catalytic ac-
tivities of the tested Chirazyme L-2 and Lipolase were pro-
gressively lowered on increasing the amount of H2O, al-
though the enantioselectivity was apparently not affected.
Furthermore, the hydrolysis of 5 in the presence of Chira-
zyme L-2 or Lipolase was complete even without the addi-
tion of any H2O (Table 1, entries 2 and 7), because the H2O
present in the enzyme preparation or in the solvent used
was sufficient for the b-lactam ring opening.
Even though the enantioselectivity in the Lipolase-cata-
lyzed ring opening of 5 was excellent (E>200), the reactivi-
ty for the hydrolysis clearly increased as the quantity of
enzyme was increased (Table 2). When merely 10 mgmLꢂ1
of enzyme was used, the reaction reached 50% conversion
over a relatively long period of time (236 h; entry 1). In
spite of the fact that the optimal enzyme quantity (resulting
in the shortest reaction time needed to reach 50% conver-
Abstract in Hungarian: HatØkony enzimes módszert dolgoz-
tunkik benzociszpentacin Øs homológjai enantiomertiszta
˝
˝
formµban tçrtØno eloµllítµsµra. A 3,4-benzo-6-azabiciklo-
[3.2.0]heptµn-7-on, 4,5-benzo-7-azabiciklo[4.2.0]oktµn-8-on
Øs 5,6-benzo-8-azabiciklo[5.2.0]nonµn-9-on enantioszelektív
˝ ˝
gyurunyitµsµt (E>200) vízzel Lipolase (Candida antarctica
B lipµz-katalízissel, diizopropil-Øterben, 608C-on vØgeztük. A
nagy enantiomerfelesleggel Øs jó termelØssel kapott (1R,2R)-
b-aminosav (eeꢁ96%, termelØsꢁ40%) Øs (1S,6S)-,
(1S,7S)-valamint (1S,8S)-b-laktµm (ee>99%, termelØsꢁ
˝
44%) enantiomerekelvµlasztµsa egyszeru en, szerves-vizes
extrakcióval tçrtØnt. Mind a racØm, mind az enantiomertiszta
˝
˝
b-laktµmokból eloµllítottuka megfelelo aminosav hidroklori-
dokat.
2588
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 2587 – 2592

Benzocispentacin and Its Homologues
FULL PAPER
Table 2. Effect of quantity of Lipolase on ring opening of (ꢀ)-5.[a]
Transformations of the enantiomers: The transformations
involving the ring opening of b-lactam enantiomers 13–15
with 18% HCl resulted in the enantiomers of the b-amino
acid hydrochlorides 19–21 (eeꢁ99%; Scheme 2). Treatment
of enantiomeric b-amino acids 10–12 with 18% HCl resulted
in amino acid hydrochlorides 16–18, respectively (eeꢁ
98%). As the b-homooligomers constructed from the build-
ing block trans-2-aminocycloalkanecarboxylic acid display
stable helical structures,[6] the preparation of trans-b-amino
acid hydrochlorides (racemic 26 and enantiopure 23 and 27)
was also investigated (Scheme 3). The transformation of 11
with SOCl2 afforded amino ester 22 (ee>99%), whereas
that of 14 by ring opening with 22% HCl/EtOH resulted in
the enantiomer of the b-amino ester 25 (ee=99%). NaOEt
isomerization of racemic 24 and enantiopure 22 and 25, fol-
lowed by acidic hydrolysis, resulted in the corresponding
trans-amino acid hydrochlorides 23, 26, and 27 (ee=99%,
containing ꢃ30% cis isomer; Scheme 3). The physical data
of the enantiomers prepared are reported in Table 5.
[c]
Entry
Lipolase
Conv.
[%]
ees[b]
eep
[%]
E
[gmLꢂ1
]
[%]
1
10
9
7
(98)
13
17
28
31
40
45
>99
(>99)
>99
>99
>99
>99
>99
>99
>200
(50 after 236 h)
(>200)
>200
>200
>200
>200
>200
>200
2
3
4
5
6
7
20
30
40
50
75
12
15
22
24
29
31
100
[a] 0.05m substrate in iPr2O, with 1 equiv H2O, at 608C, after 15 h. [b] Ac-
cording to GC. [c] According to GC after double derivatization.
sion) proved to be 100 mgmLꢂ1 (entry 7), for reasons of
economy, 50 mgmLꢂ1 Lipolase was chosen for use in the
preparative-scale resolutions of (ꢀ)-4 to (ꢀ)-6.
Several solvents were tested in order to study the solvent
effect in the Lipolase-catalyzed hydrolysis of (ꢀ)-5 at 608C
(Table 3). Lipolase was practically inactive in acetone
The absolute configuration in the case of 16 was proved
by comparing the [a] value with the literature data[13]
(Table 5), whereas for its six- and seven-membered homo-
logues the analyzed chromatograms indicated the same
enantioselectivity preference for Lipolase.
Table 3. Effect of solvent on ring opening of (ꢀ)-5.[a]
[c]
Entry
Solvent
Conv.
ees[b]
eep
E
[%]
[%]
[%]
1
2
3
4
5
6
7
acetone
no reaction
no reaction
33
no reaction
44
tetrahydrofuran
diethyl ether
tert-amyl alcohol
diisopropyl ether
toluene
Conclusion
49
>99
>200
In conclusion, an efficient direct enzymatic method was de-
veloped for the synthesis of benzocispentacin and its new,
optically pure six- and seven-membered homologues by
enantioselective ring cleavage of the corresponding b-lac-
tams in an organic medium. The Lipolase-catalyzed highly
enantioselective reactions (E>200) with H2O (1 equiv) as
the nucleophile in iPr2O at 608C produced b-amino acid and
b-lactam enantiomers (eeꢁ96%) in good chemical yields
(40–45%). The products were also easily separated. Trans-
formations by the ring opening of b-lactams with 18% HCl
and 22% EtOH/HCl resulted in the corresponding b-amino
acid and ester hydrochlorides, respectively (eeꢁ99%). Iso-
merization of the esters, followed by hydrolysis, resulted in
the corresponding trans-amino acid hydrochlorides. The
present method of formation of new enantiopure b-lactams
and b-amino acids (benzocispentacin and its homologues)
proved to be a very simple, inexpensive route, and could be
easily scaled up. The synthe-
75
44
>99
>99
>200
>200
31
n-hexane
no reaction
[a] 0.05m substrate in the solvent tested, 30 mgmLꢂ1 Lipolase, and
1 equiv H2O, at 608C, after 22 h. [b] According to GC. [c] According to
GC after double derivatization.
(entry 1), tetrahydrofuran (entry 2), and n-hexane (entry 7).
The reaction proceeded more slowly in Et2O (entry 3) and
toluene (entry 6) than in iPr2O (conversion 44% after 22 h).
On the basis of the results of the above experiments, the
gram-scale resolutions of (ꢀ)-4 to (ꢀ)-6 were performed in
iPr2O with Lipolase (50 mgmLꢂ1) as the catalyst and H2O
(1 equiv) as the nucleophile, at 608C. The results are pre-
sented in Table 4 and in the Experimental Section.
sized
enantiopure
b-amino
Table 4. Lipolase-catalyzed ring opening of (ꢀ)-4 to (ꢀ)-6.[a]
b-Amino acid (10–12)
acids are promising building
blocks for the synthesis of pep-
tides and peptidomimetics, as
potential pharmacons.
b-Lactam (13–15)
Isomer
t
Conv.
[%]
E
Yield
[%]
Isomer
ee[b]
[a]2D5
Yield
[%]
ee[c]
[a]2D5
[h]
[%]
[%]
(ꢀ)-4
(ꢀ)-5
(ꢀ)-6
6
54
51
50
50
50
>200
>200
>200
40
45
44
1R,2R
1R,2R
1R,2R
96
99
97
ꢂ6[d]
+25.1[f]
+7[h]
44
45
45
1S,5S
1S,6S
1S,7S
99
99
99
+224[e]
+313[g]
ꢂ137[i]
[a] 50 mgmLꢂ1 enzyme in iPr2O and 1 equiv H2O, at 608C. [b] Determined by GC after double derivatization
with 1) diazomethane or 2) acetic anhydride, in the presence of 4-dimethylaminopyridine and pyridine (see the
Experimental Section). [c] According to GC (see the Experimental Section). [d] c=0.11 in H2O. [e] c=0.33 in
CHCl3. [f] c=0.30 in H2O. [g] c=0.35 in CHCl3. [h] c=0.09 in H2O. [i] c=0.28 in CHCl3.
Experimental Section
Materials and methods: Lipolase
(lipase
B from Candida antarctica),
Chem. Eur. J. 2006, 12, 2587 – 2592
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2589

E. Forró and F. Fülçp
spectrometer. Melting points were de-
termined by using a Kofler apparatus.
Synthesis of racemic 3,4-benzo-6-
azabicyclo[3.2.0]heptan-7-one,
benzo-7-azabicyclo[4.2.0]octan-8-one,
and 5,6-benzo-8-azabicyclo[5.2.0]-
4,5-
nonan-9-one ((ꢀ)-4, (ꢀ)-5, and (ꢀ)-6)
Synthesis of (ꢀ)-4: By using the litera-
ture procedure[13] and indene (11.62 g,
0.10 mol), (ꢀ)-4 was obtained (11.44 g,
72%; recrystallized from AcOEt and
MeOH, m.p. 190–1918C).
Data for (ꢀ)-4: 1H NMR (400 MHz,
CDCl3, 258C, TMS): d=3.07 (dd, J=
17.5, 10.5 Hz, 1H; CH2), 3.35 (d, J=
17.3 Hz, 1H; CH2), 4.03 (d, J=
10.5 Hz, 1H; CHCO), 5.03 (d, J=
Scheme 3. Epimerization of esters.
4.2 Hz, 1H; CHN), 6.23 (brs, 1H;
NH), 7.21–7.34 ppm (m, 4H; C6H4);
13C NMR (100.62 MHz, CDCl3): d=
31.1, 54.9, 59.2, 125.8, 127.0, 127.8, 129.8, 141.2, 144.5, 172.0 ppm; elemen-
tal analysis calcd (%) for C10H9NO: C 75.45, H 5.70, N 8.80; found: C
75.39, H 5.60, N 8.85.
Table 5. Specific rotations of enantiomers prepared.
Amino acid hydrochloride
ee [%]
[a]2D5
(1R,2R)-16
98
ꢂ5.8 (c=0.4 in H2O)
ꢂ5.7 (c=0.5 in MeOH)[a]
+5.7 (c=0.4 in H2O)
+27.8 (c=0.35 in H2O)
ꢂ27.5 (c=0.37 in H2O)
+8.1 (c=0.4 in H2O)
ꢂ8 (c=0.12 in H2O)
Synthesis of (ꢀ)-5: A solution of CSI (15.57 g, 0.11 mol) in dry Et2O
(40 mL) was added dropwise to 1,2-dihydronaphthalene (13.02 g, 0.1 mol)
dissolved in dry Et2O (160 mL) at 08C. The solution was stirred for 0.5 h
at 08C, and then for 4 h at room temperature. The resulting liquid was
added dropwise to a vigorously stirred solution of Na2SO3 (1.87 g) and
K2CO3 (43.28 g) in H2O (200 mL). The organic layer was separated and
the phase was extracted with AcOEt. The combined organic layer was
dried (Na2SO4), filtered, and concentrated. The resulting yellowish oil
was crystallized from n-hexane to give (ꢀ)-5 (16.15 g, 78%; recrystallized
from iPr2O, m.p. 96–998C).
Data for (ꢀ)-5: 1H NMR (400 MHz, CDCl3, 258C, TMS): d=1.59–2.86
(m, 4H; 2”CH2), 3.71–3.73 (m, 1H; CHCO), 4.68 (d, J=4.9 Hz, 1H;
CHN), 5.96 (brs, 1H; NH), 7.19–7.3 ppm (m, 4H; C6H4); 13C NMR
(100.62 MHz, CDCl3): d=23.6, 27.5, 50.8, 52.2, 127.3, 129.2, 129.7, 130.2,
134.7, 140.0, 170.4 ppm; elemental analysis calcd (%) for C11H11NO: C
76.28, H 6.40, N 8.09; found: C 76.32, H 6.44, N 8.11.
(1S,2S)-19
(1R,2R)-17
(1S,2S)-20
(1R,2R)-18
(1S,2S)-21
(1R,2R)-22
(1S,2S)-25
>99
>99
99
98
99
>99
>99
+40 (c=0.15 in EtOH)
ꢂ39 (c=0.17 in EtOH)
[a] [a]2D5 =ꢂ5.3 (c=0.5 in MeOH).[13]
produced by the submerged fermentation of a genetically modified As-
pergillus oryzae microorganism and adsorbed on a macroporous resin,
was from Sigma-Aldrich. CAL-A (lipase A from Candida antarctica) and
Novozym 435 as an immobilized lipase (lipase B from Candida antarcti-
ca) on a macroporous acrylic resin were from Novo Nordisk. Chirazyme
L-2 (a carrier-fixed lipase B from Candida antarctica) was purchased
from Roche Diagnostics Corporation. Before use, CAL-A (5 g) was dis-
solved in Tris-HCl buffer (0.02m; pH 7.8) in the presence of sucrose
(3 g), followed by adsorption on Celite (17 g; Sigma). The lipase prepara-
tion thus obtained contained 20% (w/w) lipase. 1-Benzosuberone, CSI,
and the cycloalkenes (indene and 1,2-dihydronaphthalene) were from Al-
drich. The solvents were of the highest analytical grade.
Synthesis of (ꢀ)-6: A solution of CSI (2.79 mL, 32.1 mmol) in dry Et2O
(20 mL) was added dropwise to benzosuberene (1.56 g, 10.71 mmol), pre-
pared from 1-benzosuberone by means of NaBH4 reduction and subse-
quent H2O elimination,[16] dissolved in dry Et2O (20 mL) at room temper-
ature. The solution was stirred for 2 days at room temperature, and the
mixture was then quenched with H2O. The organic layer was separated
and poured into a solution of 20% Na2SO3 (24 mL). The pH was adjust-
ed to 8 by the addition of 10% KOH. The organic layer was separated
and the phase was extracted with Et2O. The combined organic layer was
dried (Na2SO4), filtered, and concentrated. The resulting white solid was
recrystallized from n-hexane and Et2O to give (ꢀ)-6 (0.52 g, 26%; m.p.
158–1618C).
Data for (ꢀ)-6: 1H NMR (400 MHz, CDCl3, 258C, TMS): d=1.31–2.75
(m, 6H; 3”CH2) 3.41 (d, J=13.4 Hz, 1H; CHCO), 4.95 (d, J=5.4 Hz,
1H; CHN), 6.79 (brs, 1H; NH), 7.04–7.29 ppm (4H; m, C6H4); 13C NMR
(100.62 MHz, D2O): d=21.7, 23.4, 31.0, 53.3, 54.2, 126.4, 127.5, 128.6,
129.4, 136.7, 137.1, 171.5 ppm; elemental analysis calcd (%) for
C12H13NO: C 76.98, H 7.00, N 7.48; found: C 76.91, H 6.88, N 7.36.
In a typical small-scale experiment, racemic b-lactam (0.05m solution) in
an organic solvent (2 mL) was added to the lipase tested (10, 20, 30, 40,
50, 75, or 100 mgmLꢂ1). H2O (0, 1, 2, or 4 equiv) was added. The mixture
was shaken at 50, 60, or 708C. The progress of the reaction was followed
by taking samples from the reaction mixture at intervals and analyzing
them by means of gas chromatography (GC). The ee values for the un-
reacted b-lactam enantiomers were determined by gas chromatography
on a Chromopack Chiralsil-DexCB column (25 m) [120 8C for 7 min!
1908C, temperature rise 208Cminꢂ1; 140 kPa; retention times (min): 13,
13.74 (antipode: 13.38); 14, 16.57 (antipode: 15.82); 15, 20.14 (antipode:
19.84)], while the ee values for the b-amino acids produced were deter-
mined by using a gas chromatograph equipped with a chiral column after
double derivatization with 1) diazomethane (Caution! The derivatization
with diazomethane should be performed under a well-working hood);
2) acetic anhydride in the presence of 4-dimethylaminopyridine and pyri-
dine [CP-Chirasil-DexCB column, 120 8C for 7 min!1908C, temperature
rise 208Cminꢂ1; 140 kPa; retention times (min): 10, 13.73 (antipode:
13.38); 11, 15.71 (antipode:15.52); 12, 17.86 (antipode: 17.09)].
Gram-scale resolution of racemic (ꢀ)-4, (ꢀ)-5, and (ꢀ)-6
Gram-scale resolution of (ꢀ)-4: Racemic 4 (1.00 g, 6.28 mmol) was dis-
solved in iPr2O (40 mL). Lipolase (2 g, 50 mgmLꢂ1) and H2O (113 mL,
6.28 mmol) were added and the mixture was shaken in an incubator
shaker at 608C for 6 h. The reaction was stopped by filtering off the
enzyme at 50% conversion. The solvent was evaporated off and the resi-
due (1S,5S)-13 was crystallized out [441 mg, 44%; recrystallized from
iPr2O [a]2D5 =+224 (c=0.33 in CHCl3); m.p. 232–2338C; ee=99%]. The
filtered enzyme was washed with distilled H2O (3”15 mL), and the H2O
was evaporated off, yielding the crystalline b-amino acid (1R,2R)-10
Optical rotations were measured with a Perkin–Elmer 341 polarimeter.
1H and 13C NMR spectra were recorded on a Bruker Avance DRX 400
2590
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 2587 – 2592

Benzocispentacin and Its Homologues
FULL PAPER
[446 g, 40%; [a]2D5 =ꢂ6 (c=0.11 in H2O); m.p. 263–2648C with sublima-
tion (recrystallized from H2O and Me2CO), ee=96%]. When 10
(200 mg) was treated with 22% HCl/EtOH (3 mL), (1R,2R)-16 was ob-
tained [181 mg, 75%; [a]D25 =ꢂ5.8 (c=0.4 in H2O); [a]2D5 =ꢂ5.7 (c=0.5 in
MeOH); ref. [13]: [a]2D5 =ꢂ5.3 (c=0.5 in MeOH); m.p. 210–2178C with
sublimation (recrystallized from EtOH and Et2O), ee=98%] was
formed.
Ring opening of (ꢀ)-4, (ꢀ)-5, (ꢀ)-6, (1S,5S)-13, (1S,6S)-14, and (1S,7S)-
15 with HCl: Racemic (ꢀ)-4 (500 mg, 3.14 mmol), (ꢀ)-5 (500 mg,
2.88 mmol), or (ꢀ)-6 (500 mg, 2.67 mmol) was dissolved in 18% HCl
(15 mL) and placed under refluxfor 3 h. The solvent was evaporated off,
and the product was recrystallized from EtOH and Et2O, which afforded
white crystals of (ꢀ)-7 (597 mg, 85%, m.p. 219–2228C), (ꢀ)-8 (571 mg,
87%, m.p. 237–2408C), or (ꢀ)-9 (520 mg, 85%, m.p. 158–1608C), respec-
tively.
1
Data for 13: The H NMR (400 MHz, CDCl3, 258C, TMS) data are simi-
lar to those for (ꢀ)-4; elemental analysis calcd (%) for C10H9NO: C
By following the procedure described above, the ring-opening reactions
of (1S,5S)-13 (200 mg, 1.25 mmol), (1S,6S)-14 (200 mg, 1.15 mmol), and
(1S,7S)-15 (100 mg, 0.53 mmol) afforded white crystals of (1S,2S)-19
[213 mg, 79%, [a]2D5 =+5.7 (c=0.4 in H2O); m.p. 204–2118C with subli-
mation; ee=99%], (1S,2S)-20 [229 mg, 87%; [a]2D5 =ꢂ27.5 (c=0.37 in
75.45, H 5.70, N 8.80; found: C 75.42, H 5.59, N 8.77.
Data for 10: 1H NMR (400 MHz, D2O, 25 8C, TMS): d=3.25 (d, J=
9.6 Hz, 1H; CH2), 3.3 (d, J=8.6 Hz, 1H; CH2), 3.52 (d, J=7.2 Hz, 1H;
CHCO), 4.85 (d, J=6.8 Hz, 1H; CHN), 7.35–7.53 ppm (m, 4H; C6H4);
elemental analysis calcd (%) for C10H11NO2: C 67.78, H 6.26, N 7.90;
found: C 67.47, H 6.25, N 7.98.
Data for 16: 1H NMR (400 MHz, D2O, 25 8C, TMS): d=3.38 (d, J=
8.9 Hz, 2H; CH2), 3.78 (dd, J=16, 8.6 Hz, 1H; CHCO), 5.01 (d, J=
6.8 Hz, 1H; CHN), 7.38–7.47 ppm (m, 4H; C6H4); elemental analysis
calcd (%) for C10H11NO2·HCl: C 56.21, H 5.66, N 5.56; found: C 55.88, H
5.40, N 5.58.
H2O); m.p. 260–2628C; ee 99%], and (1S,2S)-21 [102 mg, 79%; [a]D25
ꢂ8 (c=0.12 in H2O); m.p. 252–2598C; ee=99%], respectively.
=
The 1H NMR (400 MHz, D2O, 25 8C, TMS) data for (1S,2S)-19 and (ꢀ)-
7, (1S,2S)-20 and (ꢀ)-8, and (1S,2S)-21 and (ꢀ)-9 are similar to those for
(1R,2R)-16, (1R,2R)-17, and (1R,2R)-18.
Elemental analysis calcd (%) for (1S,2S)-19 (C10H11NO2·HCl): C 56.21,
H 5.66, N 5.56; found: C 56.11, H 5.64, N 6.69.
Gram-scale resolution of (ꢀ)-5: By using the procedure described above,
Elemental analysis calcd (%) for (ꢀ)-7 (C10H11NO2·HCl):
C 56.21,
the reaction of racemic
5 (1.00 g, 5.77 mmol) and H2O (103 mL,
H 5.66, N 5.56; found: C 55.96, H 5.60, N 6.44.
5.77 mmol) in iPr2O (40 mL) in the presence of Lipolase (2 g,
50 mgmLꢂ1) at 608C afforded unreacted (1S,6S)-14 [452 mg, 45%; [a]2D5
=
Elemental analysis calcd (%) for (1S,2S)-20 (C11H13NO2·HCl): C 58.03,
H 6.20, N 6.15; found: C 57.88, H 6.39, N 6.12.
+313 (c=0.35 in CHCl3); m.p. 142–1438C (recrystallized from iPr2O);
ee=99%] and b-amino acid (1R,2R)-11 [490 mg, 45%; [a]2D5 =+25.1 (c=
0.30 in H2O); m.p. 269–2718C with sublimation (recrystallized from H2O
and Me2CO); ee=99%] in 54 h. When 11 (200 mg) was treated with
22% HCl/EtOH (3 mL), (1R,2R)-17 [198 mg, 82%; [a]2D5 =+27.8 (c=
0.35 in H2O); m.p. 237–2408C with sublimation (recrystallized from
EtOH and Et2O), ee=99%] was formed.
Elemental analysis calcd (%) for (ꢀ)-8 (C11H13NO2·HCl):
C 58.03,
H 6.20, N 6.15; found: C 57.96, H 5.98, N 6.10.
Elemental analysis calcd (%) for (1S,2S)-21 (C12H15NO2·HCl): C 59.63,
H 6.67, N 5.79; found: C 59.51, H 6.86, N 5.64.
Elemental analysis calcd (%) for (ꢀ)-9 (C12H15NO2·HCl):
C 59.63,
H 6.67, N 5.79; found: C 59.53, H 6.64, N 5.77.
1
Data for 14: The H NMR (400 MHz, CDCl3, 258C, TMS) data are simi-
Preparation of esters (ꢀ)-24 and (1S,2S)-25: (ꢀ)-5 or (1S,6S)-14 (200 mg,
1.15 mmol) was dissolved in 22% EtOH/HCl (10 mL) and placed under
refluxfor 9 h. The solvent was evaporated off, and the product was re-
crystallized from EtOH and Et2O, which afforded white crystals of (ꢀ)-
24 [219 mg, 77%, m.p. 161–1648C with sublimation] or (1S,2S)-25
[230 mg, 80%, [a]2D5 =ꢂ39 (c=0.17 in EtOH); m.p. 188–1918C with subli-
mation], respectively.
Data for (ꢀ)-24: 1H NMR (400 MHz, CDCl3, 258C, TMS): d=1.25 (t,
J=7.1 Hz, 3H; CH3), 2.08–3.01 (m, 4H; 2”CH2), 3.22–3.25 (m, 1H;
CHCO), 4.21 (q, J=7.1 Hz, 2H; OCH2), 4.55 (brs, 1H; CHN), 7.11–7.26
(m, 3H; C6H4), 7.71 ppm (d, J=7.2 Hz, 1H; C6H4); elemental analysis
calcd (%) for C13H17NO2·HCl: C 61.05, H 7.09, N 5.48; found: C 61.00, H
6.93, N 5.49.
lar to those for (ꢀ)-5; elemental analysis calcd (%) for C11H11NO: C
76.28, H 6.40, N 8.09; found: C 76.39, H 6.55, N 7.94.
Data for 11: 1H NMR (400 MHz, D2O, 25 8C, TMS): d=1.98 (dd, J=9.9,
4.8 Hz, 1H; CH2), 2.22 (dd, J=11.1, 1.9 Hz, 1H; CH2), 2.84–3.01 (m, 3H;
CH2 and CHCO), 4.67 (brs, 1H; CHN), 7.29–7.39 ppm (m, 4H; C6H4);
13C NMR (100.62 MHz, D2O): d=21.7, 28.1, 43.4, 50.3, 127.0, 129.6,
129.8, 130.1, 131.3, 137.6, 181.4 ppm; elemental analysis calcd (%) for
C11H13NO2: C 69.09, H 6.85, N 7.32; found: C 69.11, H 6.85, N 7.33.
Data for 17: 1H NMR (400 MHz, D2O, 25 8C, TMS): d=2.06–2.31 (m,
2H; CH2), 2.91–2.99 (m, 2H; CH2), 3.13–3.18 (m, 1H; CHCO), 4.83 (d,
J=3.4 Hz, 1H; CHN), 7.24–7.37 ppm (m, 4H; C6H4); 13C NMR
(100.62 MHz, D2O): d=20.7, 27.5, 41.7, 50.0, 127.2, 129.2, 130.0, 130.2,
130.5, 137.1, 177.0; elemental analysis calcd (%) for C11H13NO2·HCl: C
58.03, H 6.20, N 6.15; found: C 58.17, H 6.31, N 6.10.
Data for (1S,2S)-25: The 1H NMR (400 MHz, D2O, 25 8C, TMS) data are
similar to those for (ꢀ)-24; elemental analysis calcd (%) for (1S,2S)-25
(C13H17NO2·HCl): C 61.05, H 7.09, N 5.48; found: C 60.88, H 7.02, N
5.56.
Gram-scale resolution of (ꢀ)-6: By using the procedure described above,
the reaction of racemic
6 (500 mg, 2.67 mmol) and H2O (48 mL,
2.67 mmol) in iPr2O (20 mL) in the presence of Lipolase (1 g,
50 mgmLꢂ1) at 608C afforded unreacted (1S,7S)-15 (219 mg, 45%; [a]2D5
=
Preparation of ester (1R,2R)-22: SOCl2 (0.21 mL, 2.86 mmol) was added
dropwise to dry EtOH (15 mL) at ꢂ158C. Amino acid 11 (250 mg,
1.3 mmol) was added in one portion to the mixture. After stirring for
30 min at 08C, and then for 3 h at room temperature, the mixture was
placed under refluxfor a further 30 min and then evaporated. Ester 22
was recrystallized from EtOH and Et2O [203 mg, 60%, [a]2D5 =+40 (c=
0.2 in EtOH); m.p. 189–1928C with sublimation].
ꢂ137 (c=0.28 in CHCl3); m.p. 2088C (recrystallized from iPr2O); ee=
99%] and b-amino acid (1R,2R)-12 [256 mg, 43%; [a]2D5 =+7 (c=0.12 in
H2O); m.p. 241–2438C (recrystallized from H2O and Me2CO); ee=97%]
in 51 h. When 12 (100 mg) was treated with 22% HCl/EtOH (3 mL),
(1R,2R)-18 [93 mg, 79%; [a]D25 =+8.1 (c=0.4 in H2O); m.p. 262–2678C
(recrystallized from EtOH and Et2O), ee=98%] was formed.
1
Data for 15: The H NMR (400 MHz, CDCl3, 258C, TMS) data are simi-
Data for (1R,2R)-22: The 1H NMR (400 MHz, D2O, 25 8C, TMS) data are
similar to those for (ꢀ)-24; elemental analysis calcd (%) for (1R,2R)-22
(C13H17NO2·HCl): C 61.05, H 7.09, N 5.48; found: C 60.91, H 7.22, N
5.38.
lar to those for (ꢀ)-6; elemental analysis calcd (%) for C12H13NO: C
76.98, H 7.00, N 7.48; found: C 76.88, H 7.36, N 7.41.
Data for 12: 1H NMR (400 MHz, D2O, 25 8C, TMS): d=1.73–2.18 (m,
4H; CH2) 2.86–2.93 (m, 3H; CH2 and CHCO), 4.83 (brs, 1H; CHN),
7.26–7.34 ppm (m, 4H; C6H4); elemental analysis calcd (%) for
C12H15NO2: C 70.22, H 7.37, N 6.82; found: C 70.24, H 7.18, N 6.55.
Data for 18: 1H NMR (400 MHz, D2O, 25 8C, TMS): d=1.67–2.89 (m,
6H; 3”CH2), 3.23 (brs, 1H; CHCO), 5.0 (brs, 1H; CHN), 7.28–7.34 ppm
(m, 4H; C6H4); elemental analysis calcd (%) for C12H15NO2·HCl: C
59.63, H 6.67, N 5.79; found: C 59.58, H 6.53, N 5.41.
Isomerization of (ꢀ)-24, (1R,2R)-22, and (1S,2S)-25: Na (0.1 g,
4.34 mmol) was dissolved in dry EtOH (5 mL). Base (ꢀ)-24 (500 mg,
2.28 mmol) was added to this solution and the mixture was refluxed for
9 h. After evaporation of the solvent, 18% HCl (5 mL) was added, and
then the mixture was refluxed for 10 h. After standing overnight, the so-
lution was filtered, and evaporated to dryness. MeOH (5 mL) was added,
and the solution was filtered again. The solvent was evaporated off, and
Chem. Eur. J. 2006, 12, 2587 – 2592
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2591

E. Forró and F. Fülçp
the residue was recrystallized from EtOH and Et2O, which afforded
arkat-usa.org/ark/journal/2002/I05 Moreno-Manas/MM-334C/
white crystals of (ꢀ)-26 (404 mg, 78%; containing ꢃ30% cis isomer).
MM-334C.pdf.
Data for (ꢀ)-26: 1H NMR (400 MHz, D2O, 25 8C, TMS): d=2.11–3.03
(m, 4H; 2”CH2), 3.17–3.22 (m, 1H; CHCO), 4.9 (d, J=5.5 Hz, 1H;
CHN), 7.26–7.44 ppm (m, 4H; C6H4); elemental analysis calcd (%) for
(ꢀ)-26 (C11H13NO2·HCl): C 58.03, H 6.20, N 6.15; found: C 58.14, H
6.11, N 6.13.
[5] F. Fülçp, E. Forró, G. K. Tóth, Org. Lett. 2004, 6, 4239–4241.
[6] a) D. Seebach, J. L. Matthews, Chem. Commun. 1997, 2015–2022;
b) D. H. Appela, L. A. Christianson, D. A. Klein, D. R. Powell, X. L.
Huang, I. J. Barchi, S. H. Gellman, Nature 1997, 387, 381–384;
c) S. H. Gellman, Acc. Chem. Res. 1998, 31, 173–180; d) E. A.
Porter, X. F. Wang, H. S. Lee, B. Weisblum, S. H. Gellman, Nature
2000, 404, 565–565; e) R. Günther, H. Hofmann, K. Kuczera, J.
Phys. Chem. B 2001, 105, 5559–5567; f) T. Martinek, G. K. Tóth, E.
Vass, M. Hollósi, F. Fülçp, Angew. Chem. 2002, 114, 1794–1797;
Angew. Chem. Int. Ed. 2002, 41, 1718–1721; g) A. Hetꢁnyi, I. M.
Mµndity, T. A. Martinek, G. K. Tóth, F. Fülçp, J. Am. Chem. Soc.
2005, 127, 547–553.
By following the procedure described above, base (1R,2R)-22 (150 mg,
0.68 mmol) and base (1S,2S)-25 (150 mg, 0.68 mmol) afforded white crys-
tals of (1R,2S)-23 (116 mg, 75%) and (1S,2R)-27 (139 mg, 89%) (both
containing ꢃ30% cis isomer), respectively.
The 1H NMR (400 MHz, D2O, 25 8C, TMS) data for (1R,2S)-23 and
(1S,2R)-27 are similar to those for (ꢀ)-26.
[7] a) Z. Szakonyi, T. Martinek, A. Hetꢁnyi, F. Fülçp, Tetrahedron:
Asymmetry 2000, 11, 4571–4579; b) S. G. Davies, G. Bhalay, Tetrahe-
dron: Asymmetry 1996, 7, 1595–1596.
[8] a) S. Gedey, J. van der Eycken, F. Fülçp, Org. Lett. 2002, 4, 1967–
1969; b) S. Gedey, J. van der Eycken, F. Fülçp, Lett. Org. Chem.
2004, 1, 215–217.
Acknowledgements
The authors acknowledge receipt of OTKA grants T 046440 and
[9] a) E. Forró, J. ꢂrva, F. Fülçp, Tetrahedron: Asymmetry 2001, 12,
643–649; b) E. Forró, F. Fülçp, Tetrahedron: Asymmetry 2001, 12,
2351–2358; c) J. Kµmµn, E. Forró, F. Fülçp, Tetrahedron: Asymme-
try 2000, 11, 1593–1600; d) P. Csomós, L. T. Kanerva, G. Bernµth, F.
Fülçp, Tetrahedron: Asymmetry 1996, 7, 1789–1796.
[10] S. Park, E. Forró, H. Grewal, F. Fülçp, R. J. Kazlauskas, Adv. Synth.
Catal. 2003, 345, 986–995.
[11] E. Forró, F. Fülçp, Org. Lett. 2003, 5, 1209–1212.
[12] R. Shintani, G. C. Fu, Angew. Chem. 2003, 115, 4216–4219; Angew.
Chem. Int. Ed. 2003, 42, 4082–4085.
[13] F. Fülçp, M. Palkó, J. Kµmµn, L. Lµzµr, R. Sillanpꢃꢃ, Tetrahedron:
Asymmetry 2000, 11, 4179–4187.
[14] K. Bakshi, K. J. Barakat, Y. Lai, R. P. Nargund, B. L. Palucki, M. K.
Park, A. A. Patchett, I. Sebhat, Z. Ye, PCT Int. Appl. (2002), WO
2002015909, p. 128.
[15] E. J. Moriconi, P. H. Mazzocchi, J. Org. Chem. 1966, 31, 1372–1379.
[16] P. A. Bonvallet, E. M. Todd, Y. S. Kim, R. J. McMahon, J. Org.
Chem. 2002, 67, 9031–9042.
T 049407, GVOP-3.1.1.-2004-05-0255/3.0, and a Bolyai Fellowship for
E.F.
[1] a) F. Fülçp, Chem. Rev. 2001, 101, 2181–2204; b) K. Park, M. J.
Kurth, Tetrahedron 2002, 58, 8629–8659; c) A. Kuhl, M. G. Hahn,
M. Dumic, J. Mittendorf, Amino Acids 2005, 29, 89–100.
[2] a) L. Kiss, E. Forró, G. Bernµth, F. Fülçp, Synthesis 2005, 1265–
1268; b) Z. Szakonyi, S. Gyónfalvi, E. Forró, A. Hetꢁnyi, F. Fülçp,
Eur. J. Org. Chem. 2005, 4017–4023; c) F. Fülçp, M. Palkó, E.
Forró, M. Dervarics, T. Martinek, Eur. J. Org. Chem. 2005, 3214–
3220.
[3] a) K. M. Brashear, C. A. Hunt, B. T. Kucer, M. E. Duggan, G. D.
Hartman, G. A. Rodan, S. B. Rodan, C. Leu, T. Prueksaritanont, C.
Fernandez-Metzler, A. Barrish, C. F. Homnick, J. H. Hutchinson,
P. J. Coleman, Bioorg. Med. Chem. Lett. 2002, 12, 3483–3486;
b) D. L. Steer, R. A. Lew, P. Perlmutter, A. I. Smith, M. I. Aguilar,
Curr. Med. Chem. 2002, 9, 811–822.
[4] a) C. Palomo, M. Oiarbide, A. Landa, A. Esnal, A. Linden, J. Org.
Chem. 2001, 66, 4180–4186; b) C. Palomo, I. Ganboa, M. Oiarbide,
G. T. Sciano, J. L. Miranda, ARKIVOC 2002, v, 8–16, http://www.
Received: October 17, 2005
Published online: December 29, 2005
2592
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 2587 – 2592