Conversion of hydroxy nitriles to lactones using Rhodococcus rhodochrous whole cells

Full text of DOI:10.1021/jo9616623

9084
J . Org. Chem. 1996, 61, 9084-9085
focused on chemo-, regio-, and enantioselectivity.5-7 The
proposed mechanisms of enzyme-catalyzed nitrile hy-
drolysis has been discussed, and several species of
microorganisms have been identified to possess this
activity.8 In this paper we describe our initial efforts to
extend microbial hydrolysis of nitriles to substrates
containing a hydroxyl group in the 4- and 5-position of
the substrate such that a γ-butyrolactone or δ-valerolac-
tone is formed from the presumed hydroxy acid inter-
mediate.
Con ver sion of Hyd r oxy Nitr iles to La cton es
Usin g Rh od ococcu s r h od och r ou s Wh ole
Cells
Stephen K. Taylor,*,† Nikolas H. Chmiel,
Lloyd J . Simons, and J ames R. Vyvyan†
Department of Chemistry, Hope College,
Holland, Michigan 49422-9000
R. rhodochrous IFO 155649 was incubated at 30 °C in
a culture medium containing glycerol as a carbon source
and ꢀ-caprolactam as the source of nitrogen blending the
methods of Ohta5a and Mayaux.10 The harvested wet
cells were suspended in a phosphate buffer, the hydroxy
nitrile substrate 111,12 was introduced, and the reaction
was monitored by TLC. After workup, the products were
purified by chromatography on silica and characterized
by GC, IR, NMR, and chiral GC.13 Compounds 2a ,4a 2b,4a
2c,2a,14 2d ,2a,14 2e,15 2f,16 2g,17 2h ,18 and 2i4a have been
reported, and their identity was confirmed by comparison
with authentic standards and/or literature data.
As illustrated in Table 1, microbial hydrolysis of the
hydroxy nitriles 1 provides good yields of lactones 2.19
Younger, actively growing cell cultures (2-3 days) gave
modest enantioselectivity in the formation of 2 (Table 1,
entries 1 and 2), but at the expense of the chemical yield.
In these experiments a low mass recovery of crude
material was obtained, possibly as a result of further
metabolism of the substrate by the microbe. Older,
resting cells (6-10 days) provided very good yields of the
various lactones 2, although with little or no enantiose-
lectivity (Table 1, entries 3-8). Occasionally, traces of
an amide were seen in the 1H NMR spectrum of the crude
product mixture from the hydrolysis of 1a ,20 but this was
not consistently reproducible, and similar evidence of
Received August 29, 1996
Nitriles are one of the most synthetically versatile
functional groups in organic synthesis. Their ease of
preparation from varied precursors and their subsequent
reactions, particularly those of their enolates, allow rapid
assembly of structurally and stereochemically complex
compounds.1 Moreover, γ-butyro- and δ-valerolactones
are prominent moieties in biologically active compounds.
They are components of natural flavors and odors,
pheromones of several species of insects, and intermedi-
ates in the synthesis of natural products.2 Hydrolysis of
4- and 5-hydroxy nitriles and subsequent lactonization
of an intermediate hydroxy acid would seem an attrac-
tive, general route to these important molecules (eq 1).
Unfortunately, chemical conversion of a nitrile to its
corresponding carboxylic acid is often difficult, requiring
harsh conditions of concentrated acid or base and ex-
tended reaction times at elevated temperature.3 Such
vigorous conditions are not desirable for the preparation
of structurally elaborate and/or sensitive substrates.
Therefore, the hydrolysis of a hydroxynitrile to the
corresponding lactone by way of the intermediate hydroxy
acid has not been a transformation of choice for the
construction of substituted lactones.4 Herein we report
our preliminary results in the facile microbial hydrolysis
and lactonization of hydroxynitriles using Rhodococcus
rhodochrous whole cells. This high-yielding transforma-
tion takes place at pH 6 in a matter of minutes at 30 °C.
This method paves the way to a milder general prepara-
tion of substituted lactones from nitrile precursors.
Several reports of microbial hydrolysis of nitriles to
either their corresponding amides or carboxylic acids
have appeared recently, and many of the studies have
(5) Sugai, T.; Katoh, O.; Ohta, H. Tetrahedron 1995, 51, 11987 and
references cited therein.
(6) Crosby, J .; Moilliet, J .; Parratt, J . S.; Turner, N. J . J . Chem. Soc.,
Perkin Trans. 1 1994, 1679 and references cited therein.
(7) (a) de Raadt, A.; Klempier, N.; Faber, K.; Griengl, H. J . Chem.
Soc., Perkin Trans. 1 1992, 137. (b) Gradley, M. L.; Deverson, C. J . F.;
Knowles, C. J . Arch. Microbiol. 1994, 161, 246. (c) Layh, N.; Stolz, A.;
Bo¨hme, J .; Effenberger, F.; Knackmuss, H.-J . J . Biotechnol. 1994, 33,
175. (d) Bauer, R.; Hirrlinger, B.; Layh, N.; Stolz, A.; Knackmuss, H.-
J . Appl. Microbiol. Biotechnol. 1994, 42,1 and references cited in the
above works.
(8) Faber, K. Biotransformations in Organic Chemistry, 2nd ed.;
Springer-Verlag: New York, 1995; Chapter 2, pp 124-133 and
references therein.
(9) Purchased from Institute for Fermentation, Osaka (IFO), 17-85
J uso-honmachi 2-chome, Yodogawa-ku, Osaka 532, J apan.
(10) Mayaux, J .-F.; Cerbelaud, E.; Soubrier, F.; Yeh, P.; Blanche,
F.; Pe´tre´, D. J . Bacteriol. 1991, 173, 6694.
(11) Substrates 1a -h were prepared by opening the appropriate
epoxide with lithioacetonitrile. Preparation of these and other hydroxy
nitriles by this method, including examination of the diastereoselec-
tivity of this reaction will be described elsewhere. Compound 1i was
prepared by NaBH4 reduction of commercially available 5-oxo-
hexanenitrile.
(12) Larcheveque, M.; Debal, A. Synth. Commun. 1980, 10, 49.
(13) See Supporting Information for typical experimental procedure.
(14) De´ziel, R.; Malenfant, E. J . Org. Chem. 1995, 60, 4660.
(15) Heiba, E. I.; Dessau, R. M.; Rodewald, P. G. J . Am. Chem. Soc.
1974, 96, 7977.
† Camille and Henry Dreyfus Scholar/Fellow 1995-1997.
(1) (a) The Chemistry of the Cyano Group; Rappoport, Z., Ed.;
Interscience: New York, 1970. (b) Arseniyadis, S.; Kyler, K. S.; Watt,
D. S. In Organic Reactions; Dauben, W. G., Ed.; Wiley: New York,
1984; Vol. 31, Chapter 1.
(2) (a) Brown, H. C.; Kulkarni, S. V.; Racherla, U. S. J . Org. Chem.
1994, 59, 365. (b) Lin, G.-Q.; Xu, W.-C. Tetrahedron 1996, 52, 5907.
(c) Enders, D.; Knopp, M. Tetrahedron 1996, 52, 5805. (d) Marson, C.
M.; Randall, L.; Winter, M. J . Tetrahedron Lett. 1994, 35, 6717. (e)
Baskaran, S.; Islam, I.; Chandrasekaran, S. J . Org. Chem. 1990, 55,
891. (f) Utaka, M.; Watabu, H.; Takeda, A. J . Org. Chem. 1987, 52,
4363. (g) Cardellach, J .; Font, J .; Ortun˜o, R. M. J . Heterocycl. Chem.
1984, 21, 327 and references cited in the above works.
(3) For representative examples see: (a) Prout, F. S.; Hartman, R.
J .; Huang, E. P.-Y.; Korpics, C. J .; Tichelaar, G. R. Organic Syntheses;
Wiley: New York, 1963; Collect. Vol. IV, p 93. (b) Allen, C. F. H.;
J ohnson, H. B. Organic Syntheses; Wiley: New York, 1963; Collect.
Vol. IV, p 804. (c) McGuire, M. A.; Sorenson, E.; Owings, F. W.; Resnick,
T. M.; Fox, M.; Baine, N. H. J . Org. Chem. 1994, 59, 6683.
(4) For representative examples of this transformation, see: (a)
Gopalan, A.; Lucero, R.; J acobs, H.; Berryman, K. Synth. Commun.
1991, 21, 1321. (b) Matsuda, I.; Murata, S.; Ishii, Y. J . Chem. Soc.,
Perkin Trans. 1 1979, 26.
(16) Mattes, H.; Benezra, C. J . Org. Chem. 1988, 53, 2732.
(17) Chattopadhyay, S.; Mamdapur, V. R.; Chadha, M. S. Tetrahe-
dron 1990, 46, 3667.
(18) Alonso, D.; Font, J .; Ortun˜o, R. M. J . Org. Chem. 1991, 56, 5567.
(19) As a control, substrate 1a was placed in the pH 6 buffer at 30
°C for 18 h. No 2a was detected by TLC or GC, and 1a was recovered
quantitatively.
(20) Determined by comparison to the 1H NMR spectrum of 4-hy-
droxyhexanamide (3), which was prepared by partial chemical hy-
drolysis of 1a using the method of Rao: Rao, C. G. Synth. Commun.
1982, 12, 177.
S0022-3263(96)01662-3 CCC: $12.00 © 1996 American Chemical Society

Communications
J . Org. Chem., Vol. 61, No. 26, 1996 9085
Ta ble 1. Micr obia l Hyd r olysis of Hyd r oxyn itr iles
(path b). A hydroxy acid intermediate (cf. 6 in Scheme
1) was observed by TLC and IR analysis of the crude
reaction mixture in some runs.22
This suggests that path a is operational, but does not
rule out some lactonization by path b. Hydroxy acid
formation was especially prevalent in substrates that
lead to δ-valerolactones (see below). In those cases where
a hydroxy acid was observed, acidification of the reaction
mixture prior to extraction promoted lactonization and
allowed good recovery of 2. It is an open question
whether the modest enantiodifferentiation of the sub-
strate occurs prior or subsequent to (via path b) the
formation of 5.21
Comparison with an authentic standard showed that
the major enantiomer of 2a produced (Table 1, entries 1
and 2) had the 4R configuration, which is the natural
enantiomer of the Trogoderma beetle pheromone.24 Lac-
tone 2b is an attractant for the rice weevil,2f and
compound 2g is a pheromone for the rove beetle.17
The microbial hydrolysis proceeds rapidly on 5-hydroxy
nitriles as well, although large amounts of a hydroxy acid
intermediate are usually observed prior to acidification
of the reaction mixture.23,24 Workup and recovery of the
lactone 2i (Table 1, entry 12) is complicated by a
relatively slow lactonization and competing oligomeriza-
tion of the hydroxyacid intermediate. This is responsible
for the low yield of 2i, which we feel will be remedied in
the optimization phase of this work.
entry nitrile
R
n
product yielda (%) ee (%)
1
2b
3
4
5
6
7
8
9
1a
1a
1a
1b
1c
1d
1e
1f
Et
Et
Et
1
1
1
1
1
1
1
1
1
1
2
2a
2a
2a
2b
2c
2d
2e
2f
9
14
79
82
37
68
40
53
40
79
16d
62
56
0
12
6
4
0
8
C5H11
t-Bu
Ph
5-hexenyl
3-butenyl
C8H17
c-C5H10
Me
1g
1h
1i
2g
2h
2i
c
10
11
c
a
Unoptimized isolated yields of material that has been chro-
b
matographed on silica gel. Glucose was used in place of glycerol
c
d
in the culture. Not determined. Crude yield; see text regarding
isolation problems.
Sch em e 1
In summary, we have demonstrated that hydroxy
nitriles can be efficiently converted to the corresponding
lactones in one step using the microorganism R. rhodo-
chrous under very mild conditions. Current studies
include defining the scope of acceptable substrates,
optimizing workup procedures and yields, immobilizing
the microorganism to facilitate product isolation, and
improving the enantioselectivity of the hydrolyses. Our
immediate future plans include refining the proposed
mechanism of this transformation and isolating the
enzyme(s) responsible for this activity.
Ack n ow led gm en t. This work was supported by the
Camille and Henry Dreyfus Scholar/Fellow Program for
Undergraduate Institutions and a Cottrell College Sci-
ence Award of Research Corporation. Acknowledgment
is made to the donors of the Petroleum Research Fund,
administered by the American Chemical Society, for
partial support of this research. The authors also thank
Drs. Vicki Isola, Maria Burnatowska-Hledin, Virginia
McDonough, and Lois Tverberg of the Hope College
Biology Department for very helpful discussions con-
cerning culturing the microorganism.
amide formation was not observed with other sub-
strates.21 Reaction times ranged from 30 min to 3 h for
most substrates.
There are two possible pathways by which the lactone
2 could be formed within the framework of the proposed
mechanism of nitrilase enzymes (Scheme 1).8 The pro-
posed acyl-enzyme species 5 could be cleaved by water
(path a) to produce the hydroxy acid 6. Acid-catalyzed
dehydration of 6 would produce 2. Alternatively, the
tethered hydroxyl in 5 could attack to produce 2 directly
Su p p or tin g In for m a tion Ava ila ble: Typical experimen-
1
tal procedures and copies of H NMR spectra for compounds
2a -c,e-h and 3 and copies of 13C NMR spectra for 2c,e-g as
representative examples of product purity (13 pages).
J O9616623
(22) Evidence for the hydroxy acid included an unsymmetrical very
broad OH stretch (3500-3000 cm-1) and an acid carbonyl stretch (1715
cm-1) in the IR and a very polar spot by TLC that dissappeared upon
treatment with acid.
(23) Taylor, S. K.; Fried, J . A.; Grassl, Y. N.; Marolewski, A. E.;
Pelton, E. A.; Poel, T.-J .; Rezanka, D. S.; Whittaker, M. R. J . Org.
Chem. 1993, 58, 7304 and references therein.
(21) Various microorganisms including Rhodococcus sp. have been
shown to possess hydratase/amidase activity, nitrilase activity, or both
(see refs 5-8). Although we cannot rule out the two-step hydratase/
amidase mechanism, we have chosen to present our results within the
context of a nitrilase mechanism.
(24) Taylor, S. K.; Simons, L. J .; Vyvyan, J . R. Unpublished results.