Mechanistic studies of 1-aminocyclopropane-1-carboxylate deaminase: Unique covalent catalysis by coenzyme B6

Full text of DOI:10.1021/ja960822p

J. Am. Chem. Soc. 1996, 118, 8763-8764
8763
Scheme 1
Mechanistic Studies of
1-Aminocyclopropane-1-carboxylate Deaminase:
Unique Covalent Catalysis by Coenzyme B6
Keqiang Li, Wensheng Du, Nanette Loida S. Que, and
Hung-wen Liu*
Department of Chemistry, UniVersity of Minnesota
Minneapolis, Minnesota 55455
ReceiVed March 13, 1996
The 1-aminocyclopropane-1-carboxylate (ACC) deaminase,
produced by Pseudomonads when ACC (1) is the sole nitrogen
source, is a PLP-linked catalyst with a unique capacity to break
the cyclopropane ring leading to the formation of R-ketobutyric
acid (4).1 Since ACC bears no abstractable R-H and the
carboxyl group is retained in the product, the ring opening must
be initiated without obvious accessibility to an R-carbanionic
intermediate which is the common entry of catalysis shared by
PLP-dependent enzymes.2 Precedence for such CR-Câ bond
cleavage among PLP-linked catalysts is found only in serine
hydroxymethyl transferase, in which the CR-Câ bond is cleaved
through a retro-aldol type process.3 However, such a mode of
action is apparently not feasible in the ACC case, and, thus,
ACC deaminase is likely to adapt an unusual mechanism to
accomplish this unique ring fragmentation reaction.
Early studies have established the competence of a variety
of cyclic and acyclic amino acids as alternate substrates or
inhibitors for ACC deaminase.4 It has also been shown that
the R-anion equivalent of the vinylglycyl-PLP aldimine (3) is a
key intermediate in ACC fragmentation.4 Although these
results, in conjunction with further work using labeled substrates,
had provided essential mechanistic insights into the sequence
of events occurring during ACC conversion (Scheme 1),5 the
actual mode of cyclopropane fragmentation, the most intriguing
and challenging facet of this catalysis, remained unresolved.
Two mechanistic variants had been hypothesized: (A) nucleo-
philic addition to open the ring, followed by â-proton abstraction
(Scheme 1(a)); (B) direct â-proton abstraction to initiate
cyclopropane cleavage.4b,6 Route B is highly unlikely since it
requires initiation of catalysis by abstraction of an inert proton
from the ring C-H group and such an anion induced ring
cleavage is stereoelectronically unfavored. However, route A
is also thought-provoking, since it invokes a unique covalent
catalysis considered novel in coenzyme B6 chemistry. Clearly,
the action of ACC deaminase imposes a challenge to our
understanding of the diverse roles of the B6 coenzyme in
biological transformations.
Scheme 2
Our strategy to gain insight into this puzzling problem was
to use modified substrates which, upon incubation with ACC
deaminase, would lead to either inhibition or turnover depending
(1) Honma, M.; Shimomura, T. Agric. Biol. Chem. 1978, 42, 1825.
(2) (a) Evangelopoulos, A. E. Chemical and Biological Aspects of Vitamin
B6 Catalysis; Liss, A. R., New York, 1984. (b) Vitamin B6 Pyridoxal
Phosphate, Chemical, Biochemical, and Medical Aspects; Dolphin, D.,
Poulson, R., Avramovic, O., Eds.; Wiley-Interscience: New York, 1986;
Parts A and B.
(3) (a) Schirch, L. AdV. Enzymol. 1982, 53, 83. (b) Palcic, M. M.; Floss,
H. G. In Vitamin B6 Pyridoxal Phosphate, Chemical, Biochemical, and
Medical Aspects; Dolphin, D., Poulson, R., Avramovic, O., Eds.;
Wiley-Interscience: New York, 1986; Parts A, p 25.
(4) (a) Honma, M.; Shimomura, T.; Shiraishi, K.; Ichihara, A.; Sakamura,
S. Agric. Biol. Chem. 1979, 43, 1676. (b) Walsh, C. T.; Pascal, R. A., Jr.;
Johnston, M.; Raines, R.; Dikshit, D.; Krantz, A.; Honma, M. Biochemistry
1981, 20, 7509. (c) Erion, M. D.; Walsh, C. T. Biochemistry 1987, 26,
3417.
(5) (a) Hill, R. K.; Prakash, S. R.; Wiesendanger, R.; Angst, W.;
Martinoni, B.; Arigoni, D.; Liu, H.-w.; Walsh, C. T. J. Am. Chem. Soc.
1984, 106, 795. (d) Liu, H.-w.; Auchus, R.; Walsh, C. T. J. Am. Chem.
Soc. 1984, 106, 5335.
(6) Liu, H.-w.; Walsh, C. T. In The Chemistry of the Cyclopropyl Group;
Rappoport, Z., Ed.; Wiley-Interscience: New York, 1987; Part 2, p 969.
on the modes of catalysis. For instance, judging from the
hypothesized mechanisms, it was perceived that 1-amino-2-
methylenecyclopropane-1-carboxylate (5, methylene-ACC) could
be a mechanism-specific probe. As illustrated in Scheme 1(b),
if the catalysis advances via mechanism A as predicted, reaction
with 5 would lead to irreversible inactivation, accompanied by
possible formation of compound 6 as a turnover product.7 On
the basis of this premise, compound 5 was synthesized by the
reactions delineated in Scheme 2. Although the enantiomeri-
cally pure samples were preferred, experiments performed with
a racemic mixture were able to provide ample information to
(7) If the reaction proceeds via the highly unlikely route B, the enzyme
may also be inactivated by 5 but with 2-oxo-3-pentenoic acid as a potential
turnover product.
S0002-7863(96)00822-0 CCC: $12.00 © 1996 American Chemical Society

8764 J. Am. Chem. Soc., Vol. 118, No. 36, 1996
Communications to the Editor
distinguish these mechanisms due to the distinct outcomes of
the incubation.
was accomplished by the same sequence shown in Scheme 2
starting with [1,1-2H2]propargyl alcohol (9) which was prepared
by LiAl2H4 reduction of the corresponding methyl ester. The
deuterium content in the final product was estimated to be
greater than 98% based on NMR integration. Analysis of the
turnover product isolated from the incubation of ACC deaminase
with labeled 7 revealed that the ratio between the integration
of the sum of the two olefinic signals and that of the methyl
signal in the NMR spectrum of 8 was 3.1:1. It is thus clear
from this data that addition to C-3 is the preferred mode of
nucleophilic attack in the active site of ACC deaminase.
It is worth mentioning that cyclopropanes can act as elec-
trophiles when electron-withdrawing substituents are attached.11
Since the role of PLP is to serve as an electron sink, the
cyclopropane ring of ACC is expected to be fairly electrophilic
due to double activation by both the carboxylate and the PLP-
aldimine groups. Hence, such nucleophilic addition to rupture
the electrophilic cyclopropyl ring of ACC is mechanistically
feasible. The fact that the â-H being abstracted in the ring-
opened covalent enzyme-substrate adduct (2) is an acidic
hydrogen lends further credence to the assigned mechanism.
Thus, the cleavage of ACC (1) to R-ketobutyrate (4) by ACC
deaminase clearly represents an intriguing conversion beyond
the common scope entailed by coenzyme B6 dependent catalysts.
Interestingly, this process, when considered in reverse, is
essentially the mechanism for the PLP-enzyme ACC synthase
and is clearly a highly relevant precedent.12,13
We observed that when ACC deaminase was exposed to 5,8
time-dependent inactivation occurred. This inactivation exhib-
ited Michaelis-Menten kinetics with a kinact of 1.25 h-1 and a
KI of 3.3 mM.9 Since the enzyme activity was not recovered
after extensive dialysis, the inactivation is clearly irreversible.
Furthermore, protection from inhibition was observed in the
presence of excess substrate, the effect of 5 on ACC deaminase
was thus concluded to be active-site directed. A partition ratio
of 8250 was deduced from a plot of the residual activity
observed versus total equivalents of 5 added. Interestingly,
NMR analysis of the crude incubation mixture revealed the
existence of a single turnover product which was elucidated as
2-oxo-3-methyl-3-butenoic acid (6) by spectroscopic charac-
terization.10 The identity of this compound was further con-
firmed by comparing its spectral properties with a standard
prepared by coupling ethyl oxalate with the corresponding
alkenyl Grignard reagent.10 This assignment is also supported
by the fact that the integration of the methyl peak at δ 1.90 is
equal to the sum of the methylene hydrogen signals at δ 6.05
and 6.20 when the incubation was conducted in deuterated
buffer. The isolation of 6 as the sole turnover product is most
consistent with route A as the inactivation mechanism and thus
provides compelling evidence for ring cleavage of ACC as a
nucleophilic-addition initiated event.
As shown in Scheme 1(b), both the ring methylene carbon
(C-3) and the terminal carbon (C-2′) of the exocyclic methylene
group are possible sites for the initial nucleophilic attack. While
the outcomes of these two modes of attack are identical, they
can be distinguished by using a probe having a dideuterated
exocyclic methylene moiety such as 7. Synthesis of labeled 7
Acknowledgment. We would like to express our appreciation to
Dr. Christine Shewmaker and Calgene Inc. for their generous gifts of
the plasmid pCGN1472, and to Professor Mamoru Honma for the
Pseudomonas sp. ACP strain. We thank Dr. Eugene Oh for his helpful
discussions. This work was supported by a National Institutes of Health
Grant (GM40541). H.-w.L. is a recipient of a National Institutes of
Health Research Career Development Award (GM 00559). K.L. is a
recipient of a National Research Service Award (NIGS GM17412).
(8) The ACC deaminase gene has been recently cloned, sequenced, and
expressed (Sheehy, R. E.; Honma, M.; Yamada, M.; Sasaki, T.; Martineau,
B.; Hiatt, W. R. J. Bacteriol. 1991, 173, 5260). A plasmid pCGN1472
containing the ACC deaminase gene has been kindly provided by Calgene.
The insert of pCGN1472 was subcloned into pET17b to give a new construct
pQNt23. This enzyme, isolated from the resulting overexpression E. coli
strain pQNt23/BL21(DE3)plysS, was purified via a sequence of DEAE-
Sepharose, AH-Agarose, and FPLC Mono-Q chromatography. The purified
JA960822P
(11) (a) Danishefsky, S. Acc. Chem. Res. 1979, 12, 66. (b) Stevens, R.
V. Pure Appl. Chem. 1979, 51, 1317. (c) Suckling, C. J. Angew. Chem.
Int. Ed. Engl. 1988, 27, 537. (d) Haddow, J.; Suckling, C. J.; Wood, H. C.
S. J. Chem. Soc., Perkin Trans. I 1989, 1297. (e) Silverman, R. B.; Ding,
C. Z.; Borrillo, L.; Chang, J. T. J. Am. Chem. Soc. 1993, 115, 2982.
(12) (a) Khani-Oskouee, S.; Ramalingam, K.; Kalvin, D.; Woodard, R.
W. Bioorg. Chem. 1987, 15, 92. (b) Sato, T.; Oeller, P. W.; Theologis, A.
J. Biol. Chem. 1991, 266, 3752. (c) White, M. F.; Vasquez, J.; Yang, S. F.;
Kirsch, J. F. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 12428.
(13) Since ACC is the precursor of an essential phytohormone, ethylene,
studies of the reaction catalyzed by ACC deaminase may also provide
information for designing methods to control and regulate the ethylene
biosynthetic process (Dong, G.; Fernandez-Maculet, J. C.; Yang, S. F. Proc.
Natl. Acad. Sci. U.S.A. 1992, 89, 9789; Klee, H. J.; Hayford, M. B.;
Kretzmer, K. A.; Barry, G. F.; Kishore, G. M. Plant Cell, 1991, 3, 1187).
enzyme has a specific activity of 4.2 U/mg,9 with a ratio of A280/A420
4.5.
)
(9) A standard 1 mL assay was performed in 50 mM potassium phosphate
buffer (pH 7.5) at 23 °C, containing substrate, 8-20 µg ACC deaminase,
125 µg of L-lactate dehydrogenase, and 0.1 µmol of NADH. Loss of NADH
absorbance at 340 nm (ꢀ 6220) was monitored. The unit of activity is defined
as µmol of NADH consumption per minute. The kinetic parameters of
inactivation were derived from plotting the apparent inactivation rate,
obtained by incubating ACC deaminase (90 µg) with 5 of varied concentra-
tion (1-10 mM), versus inhibitor concentration.
(10) Rambaud, M.; Bakasse, M.; Duguay, G.; Villieras, J. Synthesis 1988,
564.