-
Biocatalysis, 1987, Vol. 1, pp. 23-36 Photocopying permitted by
license only
0 1987 hanvood academic publishers GmbH Printed in the United
Kingdom
STUDIES ON THE REGIOSELECTIVITY AND STEREOSELECTIVITY OF THE
SOLUBLE
METHANE MONOOXYGENASE FROM METHYLOCOCCUS CAPSULATUS (BATH)
DAVID J. LEAK* and HOWARD DALTON Department of Biological
Sciences, University of Warwick, Coventry CV4 7A L,
England, UK (Received 4 July 1986; in final form 12 September
1986)
Alkyl substituted derivatives of cyclohexane and cyclohexene
have been used as active site probes of the soluble methane
monooxygenase (MMO) from Methylococcus capsulam (Bath). It is
proposed that the products obtained are those that would be
predicted on the grounds of chemical reactivity modulated by two
enzymic constraints (i) steric hindrance favouring hydroxylation at
positions distal to bulky substituents and (ii) limited penetration
of the substrate beyond the active site of oxygen insertion.
Evidence for inversion of stereochemistry during the hydroxylation
of ck- dimethylcyclohexane and rearrangement during the
hydroxylation of 3-methyl-1-cyclohexene supports the suggestion
that a stepwise mechanism (hydrogen abstraction and hydroxylation)
operates in the hydroxylation of aliphatic carbons.
KEY WORDS Monooxygenase, Methylococcus capsulatus, hydroxylation
selectivity, active site
INTRODUCTION
The soluble methane monooxygenase from Methylococcus capsulatus
(Bath) is comprised of three components. Component C (Lund and
Dalton, 1985) is the NADH: acceptor reductase that transfers
electrons to component A (Woodland and Dalton, 1984), the oxygenase
that interacts with the hydrocarbon substrate. Component B is
involved in coupling substrate oxidation to electron transfer
(Green and Dalton, 1985). Component A is a non-heme iron protein of
relative molecular mass 210000 comprising three subunits of Mr = 54
OOO, 42 000 and 17 OOO, suggesting an cu,, p2, y2 arrangtment
containing approximately 2 mol/mol iron and OSmol/mol zinc
(Woodland and Dalton, 1984). Studies with cell extracts and whole
cells have demonstrated that this enzyme has low substrate
specificity, being capable of hydroxylation of aliphatic, alicyclic
and aromatic hydrocarbons (Dalton, 1980). However, some of the
whole cell studies must be treated with caution following the
demonstration that two forms of methane monooxygenase with
different substrate specificities may co-exist in some
methanotrophs (Stanley et al., 1983; Burrows et al., 1984).
In recent years several patents have been issued indicating the
potential of methane-oxidizing bacteria in the bioconversion of
several of the above- mentioned substrates (Higgins et al., 1980).
If the full capabilities of this system
* Present address: Centre for Biotechnology, Imperial College of
Science and Technology, London SW7 2AB, England, UK.
23
Bio
cata
l Bio
tran
sfor
mat
ion
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
15/1
2Fo
r pe
rson
al u
se o
nly.
-
24 D. J . LEAK AND H. DALTON
are to be realized in practice, a detailed understanding of the
mechanism of action of the enzyme will be essential. There are
basically three approaches that can be adopted to reach this goal.
These are: physico-chemical characterization of the enzyme active
site; kinetic studies on the interaction of the individual
components and their prosthetic groups; and evaluation of active
site chemistry through observation of products formed from
substrate hydroxylations.
Hydroxylation of short chain alkanes gives rise to primary and
secondary alcohols (Dalton, 1980). However, this observation
provides little information on preferential sites of hydroxylation
due to the flexibility and capacity for carbon-carbon bond rotation
in these substrates. The semi-rigid structure of cyclohexane
derivatives, however, allows differentiation of regioselectivity
and stereoselectivity of hydroxylation as well as an assessment of
steric constraints on substrate utilization. This capability has
previously been exploited in probing the active site of the
paraffin hydroxylase from Pseudomonas aeruginosu (van Ravenswaay
Claasen and van der Linden, 1971), and a similar approach has been
utilized in comparing the major constraints in hydroxylation by
cytochrome P-450 isoenzymes P-450,,, and P-450LM2 (White et a f . ,
1984). In this paper we report on the use of alicyclic compounds as
methane monooxygenase substrates to help determine the nature of
the active site.
MATERIALS AND METHODS
Crude soluble extract was prepared from 1001 batch cultures of
M. cupsufutus (Bath) as previously described (Stanley et ul., 1983;
Colby and Dalton, 1978). All experiments were done using extract
from a single batch, which was stored as frozen pellets at -80°C.
The protein concentration was 66 mg ml-' determined by the method
of Bradford (1976) with bovine serum albumen as standard.
Bioconversions were done at 45°C in 5 ml McCartney bottles with
neoprene stoppers. Incubation mixtures (1 ml) contained 5 mM
magnesium chloride, 2 pl of hydrocarbon substrate and 6.6 or 13.2
mg protein in 20 mM Tris-chloride buffer pH 7.0. After one minute
pre-incubation, the bioconversion was initiated by the addition of
5 mM NADH (ethanol free) plus 10 mM potassium formate. For rate
determination, duplicate bottles containing 6.6 mg protein ml-'
were incubated for three, six and nine minutes. For analysis of
total products, bottles containing 13.2 mg protein ml-' were
incubated for 3 minutes. Assays were terminated by plunging the
bottle into ice, followed by extraction of products into dichloro-
methane. Control experiments to demonstrate the involvement of MMO
in the bioconversion contained 0.2 ml acetylene, a potent inhibitor
(Dalton and Whit- tenbury, 1976).
Products were analysed by gas chromatography (Pye Unicam GCV) on
Carbowax 20M or Carbowax 20M TPA (10% on chromosorb W 80/100 mesh)
and on SP2100 (3% on chromosorb W 80/100 mesh). All analyses were
isothermal with temperatures in the range 100-140°C depending on
the products analysed. Nitrogen carrier gas was maintained at 30 ml
min-'. Quantitation was by internal standardization using a Hewlett
Packard 3390A integrator, and identification by co-retention with
authentic standards, chemical reactivity and electron impact mass
spectrometry (Carlo Erba GC-MS 80 Kratos) with separation on
Carbowax 20M. For initial assay optimization, products were
Bio
cata
l Bio
tran
sfor
mat
ion
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
15/1
2Fo
r pe
rson
al u
se o
nly.
-
HYDROXYLATION SELECTIVITY OF MMO 25
analyzed directly (without extraction) on Tenax GC (isothermal
190°C with N2 carrier gas at 20 ml min-’).
Chemicals Nicotinamide adenine dinucleotide, reduced form (NADH)
was obtained from BDH Chemicals Ltd, Atherstone, Warwickshire, UK;
the Bradford protein reagent was obtained from Bio-Rad Labs Ltd,
Watford, Herts, UK.
Cyclohexane, cyclohexanol, cyclohexanone, cyclohexene,
methylcyclohexane, and 2-methylcyclohexanol were obtained from BDH
Chemicals Ltd, Atherstone, Warwickshire, UK.
2-cyclohexen-l-ol, cyclohexene oxide, 1-, 3- and
4-methylcyclohexanol, cyclo- hexylmethanol,
cyclohexanecarboxaldehyde, cyclohexanecarboxylic acid, ethyl-
cyclohexane, 2- and 4-ethylcyclohexanol, 1-cyclohexylethanol,
t-butylcyclohex- ane, 3-methyl-l-cyclohexene, and
3-methyl-2-cyclohexen-1-01 were obtained from Aldrich Chemical Co
Ltd, Gillingham, Dorset, UK.
2-cyclohexylethanol, cis- and trans-l,4-dimethylcyclohexane
(separate and mixed isomers), 2,5-dimethylcyclohexanol (mixed
isomers), l-methyl-l-cyclohex- ene, and 4-methyl-1-cyclohexene were
obtained from Fluorochem Ltd, Glossop, Derbyshire, UK.
RESULTS AND DISCUSSION
Preliminary Studies Preliminary experiments with cyclohexane as
substrate demonstrated that the maximum rate of hydroxylation was
obtained with the assay technique described. With the formate +
formate dehydrogenase (present in the soluble extract) NADH
regeneration system, the rate of product formation was linear for
at least nine minutes. Doubling the substrate concentration or
halving the protein concentration reduced the specific activity. A
requirement for a minimum protein concentration for maximum
specific activity has previously been observed using this
system.
Cyclohexanol was found to be the sole product in these assays,
judged by co-retention with an authentic standard. No cyclohexanone
was detected, although this has been observed in long-term whole
cell assays.
Methylcyclohexane The products from hydroxylation of
methylcyclohexane were separated into three peaks on Carbowax 20M
TPA. Using authentic standards of 1-, 2-, 3- and
4-methylcyclohexanol, cyclohexylmethanol, cyclohexylcarboxaldehyde
and cyclo- hexylcarboxylic acid, the major product
co-chromatographed with cis-3- and trans-4-methylcyclohexano1,
which could not be differentiated on either of the columns
utilized. Pairs of stereoisomers in each standard were separated on
Carbowax 20M P A and were assigned according to the criteria of
Paris and Alexandre (1972) (this assignment has since been
confirmed with “pure” isomers). Although cis-3- and
trans-4-methylcyclohexanols had identical retention times, they
were distinguishable by their fragmentation patterns obtained by
mass spectrometry (Figures l(a) and (b)) notably in their major
fragmentation ion.
Bio
cata
l Bio
tran
sfor
mat
ion
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
15/1
2Fo
r pe
rson
al u
se o
nly.
-
26 D. J . LEAK AND H. DALTON
u 0
Bio
cata
l Bio
tran
sfor
mat
ion
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
15/1
2Fo
r pe
rson
al u
se o
nly.
-
HYDROXYLATION SELECTIVITY OF MMO 27
Inspection of the major enzyme product indicated that it was
predominantly cis-3-methylcyclohexanol (Figure l(c)) although there
was a small contribution from the trans-derivative. On the basis of
the major peaks M/Z 57 and M/Z 71, the contribution of
trans-4-methylcyclohexano1 was estimated at 12%. The dominance of
cis-3- and trans-4-methylcyclohexanols over trans-3- and cis-4-
derivatives (not detected in this assay, but see the analysis of
ethylcylcohexanols) indicates that, while the MMO is regioselective
rather than regiospecific for the 3 position, an overriding feature
is the formation of planar products, cis-3- and trans-4- positions
being equatorial with respect to an equatorial methyl sub-
stituent. While this could reflect a strict enzymic control on
hydroxylation, equatorial products would also be the major products
based on chemical criteria if one assumed a stepwise mechanism
involving “random” hydroxylation of an intermediate formed by
hydrogen abstraction, the stereoselectivity of hydroxyla- tion
being determined solely by steric interactions with the
substrate.
Of the two minor product peaks, that with the longest retention
time co-chromatographed with cyclohexylmethanol, and this was
supported by mass spectral analysis. The early peak
co-chromatographed with 1-methylcylcohexanol, although previous
studies suggested that this might have been difficult to separate
from 2-methy1cyc10hexanone7 which was not available for comparison.
A molecular ion at M/Z 114 in the mass spectral analysis, however,
indicated the former. The total composition of products from the
hydroxylation of methylcy- clohexane is presented in Table 1.
Higher A lkylcylcohexanes The major product from
ethylcyclohexane co-chromatographed with the later of the two peaks
obtained with a standard of 4-ethylcyclohexanol (cis and trans) on
Carbowax 20M TPA. On the basis of criteria already outlined this
was assigned as trans-4- and/or c~-3-ethylcyclohexanol. In the
absence of an authentic standard of 3-ethylcyclohexanol it was
impossible to assess the contribution made by each isomer. A small
peak that co-chromatographed with the early isomer, assumed to be
cis-4- or truns-3-ethylcyclohexanol, was also evident. It is not
known whether this reflected a genuine difference between the types
of products obtained with methyl- and ethylcyclohexane, as the
separation of ethylcyclohexanols was better than for
methylcyclohexanols on this systeni. A small amount of cis-4- or
trans-3-methylcyclohexanols may therefore have been masked in the
previous analyses.
Both of the side chain hydroxylated products 1- and
2-cyclohexylethanol, identified using authentic standards, were
present as minor products in similar amounts. Two further peaks
were present comprising approximately 2% of total products. Neither
co-chromatographed with cis- and trans-2-ethylcyclohexanol and in
the absence of a standard of 1-ethylcyclohexanol were not
identified. The rate of hydroxylation of ethylcyclohexane, based on
total product accumulation, was about half that with
methylcyclohexane.
The effect of steric hindrance of the alkyl side chain on
hydroxylation was examined using t-butylcyclohexane as substrate.
Even in extended assays no products were obtained, indicating that
a bulky side chain restricts access of the substrate to the active
site, prohibiting hydroxylation even at a distal point on the
cyclohexane ring. This is consistent with previous evidence for
lack of hydroxyla- tion of t-butylbenzene.
Bio
cata
l Bio
tran
sfor
mat
ion
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
15/1
2Fo
r pe
rson
al u
se o
nly.
-
28 D. J. LEAK AND H. DALTON
Table 1 Products formed from cyclohexane, cyclohexene and alkyl
substituted derivatives by the soluble methane monooxygenase (MMO)
from Methylococcus capsulatus (Bath).
Substrate Prod u c t.i Amount(%) Maximum activity t
Cyclohexane cyclohexanol
Methylcyclo hexane
Ethylcyclohexane
cis-3-methylcyclohexanol rram-4-methylcyclohexanoI
cyclohexylmethanol 1-methylcyclohexanol
cis-3-/tram-4-ethylcyclohexanol
cis-4-/fruans-3-ethylcyclohexanol 1-cyclohexylethanol
2-cyclohexylethanol 1-ethylcyclohexanol (?)
t-butylcyclohexane - cis -1,4-dimethyIcyclohexane 1
,tram-4-dimethylcyclohexanol
1 ,cis-4-dimethylcyclohexanol cis-4-rnethylcyclohexylmethanol
r-1, c-2,c-5-dimethylcyclohexanol
1 ,-cis-4-dimethylcyclohexanol r-1,
r-2,~-5-dirnethylcyclohexanol (1 ,tram-4-dimethylcyclohexanol?)
mm.s-l,4-dimethylcyclohexane nam-4-methylcyclohexylmethanol
Cyclohexene
1 methyl- 1-cyclohexene
3-methyl-1-cyclohexene
4-me thylcyclohexene
2-cyclohexenol cyclohexene oxide cyclohexanone
3-methyl-2-cyclohexen- 1-01 l-cyclohexene carboxaldehyde
1-methyl-2-cyclohexen- 1-01 2-methyl-2-cyclohexen-1-01
cis-6-methyl-2-cyclohexen-l-ol traans-6-methyl-2-cyclohexen-l-ol
3-methyl-1-epoxycyclohexane [saturated ketone]
[secondary ring alcohol]' [secondary ring a l c o h ~ l ] ~ 4
methyl-1-epoxycyclohexane [saturated ketone]
100 8.2
83.9 11.4 5.0 2.8 1.8
76.8 17.9 2.1 1.5 0.5
2.25
0
2.5
1.6
27.6
57.3 35.3 4.2 3.5
86.9 10.9 1.9 0.3
85.5 11.5 3 .O
36.8 33.4 ND 15.5 14.3
78.4 12.3 10.4 4.7 4.6
43.3 38.8 6.7 15.7 2.2
t Maximum activity (nmollminlmg protein) was that measured with
2 pllml substrate and 6.6 mglml protein in all cases I]? some
produa identifications are tentativelinfompletethew are discussed
in the text. ND = not determined.
and signify the order of elution from Carbowax 20M TPA. i.e.
=early = late
Dimethylcyclohexane The effect of a second ring substitutent was
analysed using cis- and trans-1,4- dimethylcyclohexane as
substrates. Although only a single standard of 2,5-
dimethylcyclohexanol (mixed isomers) was available, all four
isomers were separated on Carbowax 20MTPA and were assigned on the
basis of retention indices according to the findings of Paris and
Alexandre (1972). Due to the
Bio
cata
l Bio
tran
sfor
mat
ion
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
15/1
2Fo
r pe
rson
al u
se o
nly.
-
HYDROXYLATION SELECTIVITY OF MMO 29
inherent symmetry of the substrate, 1,4-dimethylcyclohexane,
several potential hydroxylation points are equivalent. Thus, of
singly hydroxylated products, only three positional isomers were
possible, i.e. 2,5-dimethylcyclohexanol (two epime- ric pairs a-d),
1,4-dimethylcyclohexanol (cis and trans e, f) and 4-methylcyclohex-
anols would be expected to have the shortest retention time
followed by the 2,5-dimethylcyclohexanols and finally the
4-methylcyclohexylmethanols. Possible products are illustrated in
Figure 2.
The major products from cis-1,4-dimethylcyclohexane were
identified as 1,4-dimethylcyclohexanols (both isomers e and f) on
the basis of retention time and mass spectrometry. The major
fragmentation ions of these two products were identical, and the
presence of a molecular ion at M/Z 128 but no peak at M/Z 110
(M-18) was indicative of a tertiary alcohol. The assignment was
supported by the lack of oxidation with Jones reagent (0.1 M sodium
dichromate in 2.5 M sulphuric acid). The presence of both isomers
indicated that hydroxylation had proceeded with partial inversion
of stereochemistry, supporting the suggestion that hydroxylation of
aliphatic carbons involves a stepwise mechanism. By comparison with
the products obtained from trans-l,4-dimethylcyclohexane and
reference to previous studies, the product with the shortest
retention time was assigned as 1 ,cis-4-dimethylcyclohexanol (f).
Two minor products were also evident, one correlating with r-1,
c-2, c-5-dimethylcyclohexanol (d) and a late running peak presumed
to be cis-4-methylcyclohexylmethanol (h). Relative amounts of each
product (based on peak area) are presented in Table 1.
The products from trans-l,4-dimethylcyclohexane differed both
quantitatively and qualitatively. The major product has a long
retention time on Carbowax 20M TPA, was oxidizable with Jones
reagent and had a mass spectrum consistent with its assignment as
trans-4-methylcyclohexylmethanol (8) (peaks at M-18 and M-31 were
prominent but the molecular ion was absent). Two early peaks with
identical retention times to those produced from
cis-1,4-dimethylcyclohexane were also evident and were assigned as
1,4-methylcyclohexanols. The earlier of the two peaks, assumed to
be 1 ,cis-4-dimethylcyclohexanol (f), was prediminant, while the
later peak was also evident in similar proportions in the starting
material, indicating that it was probably not a genuine product. A
minor product was also evident which co-chromatographed with r-1,
t-2, c-5-dimethylcyclo- hexanol (a). Relative amounts and rates of
hydroxylation are presented in Table 1. It is interesting to note
that all of the products mentioned were obtained in similar
proportions when the substrate was a mixture of cis and trans
isomers, indicating that the affinity of MMO for the two isomers
does not differ greatly.
Cy clohexene Three products obtained with cyclohexene as
substrate were identified as 2-cyclohexene-1-01, cyclohexene oxide
and cyclohexanone (in order of decreasing abundance) by correlation
(GC-MS) with authentic standards. The rate of formation of the
major product 2-cyclohexen-1-01 was almost three times as high as
cyclohexane oxidation, suggesting that the allylic position is more
amenable to attack than a fully saturated hydrocarbon. However, the
greater solubility of the substrate (cyclohexene us cyclohexane)
may also have contributed to the increased rate.
It is useful to compare the types of products obtained with
cyclohexene as
Bio
cata
l Bio
tran
sfor
mat
ion
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
15/1
2Fo
r pe
rson
al u
se o
nly.
-
30 D. J. LEAK AND H. DALTON
a
b
C
d
e
f
h
doH HoH OH
CH20H
CH20H
Figure 2 Potential products from the hydroxylation of cis- and
rrans-dimethycyclohexane: (a) r-l,t-2,c-S-di- methylcyclohexanol;
(b) r-l,c-2,t-S-dimethylcyclohex- anol; (c) r - 1 , c-2,
f-5-dimethylcyclohexanol; (d) r- 1 , c-2, t-
5-dimethylcyclohexanol; (e) 1 ,trans-4-dimethylcyclo- hexanol (f) 1
,cis-4-dimethylcyclohexanol; (g) lrum-4- methylcyclohexylmethanol;
(h) cis-4-methylcyclohexyl- methanol.
Bio
cata
l Bio
tran
sfor
mat
ion
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
15/1
2Fo
r pe
rson
al u
se o
nly.
-
HYDROXYLATION SELECTIVITY OF MMO 31
a b C r------ - - - -7 ,- - - - - - - - - - --_I
Figure 3 Proposed active site geometry of MMO illustrat- ing
substrate orientation and the site of oxygen insertion (A) for
alkylcyclohexanes (a), trans-2-butene (b) and cis-2- butene
(c).
substrate and those from cis- and trans-Zbutenes. Although one
might expect a similarity in the active site configuration for
cis-2-butene and cyclohexene, the epoxide was the major product
from the former. However, the allylic hydroxyla- tion product
(crotyl alcohol) predominated with the trans isomer. (Statistical
weighting to take account of the equivalence of 1 and 4 positions
would, in fact, alter this order but the argument remains
qualitatively the same.) This gives us some insight into the active
site geometry (Figure 3).
Visualizing the substrate fairly tightly enclosed within the
active site of the enzyme (broken line), the results from
alkylcyclohexane oxidation suggest a preference for hydroxylation
at a point delineated A (Figure 3). Assuming the substrate can
penetrate sufficiently, this gives rise predominatly to cis-3-
alkylcyclohexanols. When the structures of cis- and trans-Zbutene
are placed into this framework it is evident that the double bond
in the cis is much more amenable to attack from position A than the
trans isomer, which would explain the higher levels of epoxide
formation in the former. (A preference for epoxidation over
terminal allylic hydroxylation is already evident from the
oxidation of propylene, where propylene oxide is the sole product.)
On the basis of this model, however, one would predict that the
epoxide should also comprise the major product in the case of
cyclohexene oxidation. Two possible explana- tions for this
discrepancy may be entertained.
1. Allylic positions in cyclohexene are secondary and would
therefore form charged intermediates more readily than primary
allylic carbons. Combined with their resonance stabilization a
stepwise mechanism by this route could become more favourable than
expoxidation.
2. 2-cyclohexen-1-01 could arise from the rearrangement of
cyclohexene oxide in the active site. (An alternative possibility,
that rearrangement occurs after release of the product from the
enzyme or by subsequent attack, was discounted by demonstrating the
stability of cyclohexene oxide under the assay conditions
described.) This would be envisaged as epoxide formation followed
by hydrogen abstraction.
It should be noted that there was no evidence for such a
rearrangement with cis- and trans-2-butene as substrates,
indicating that the propensity for this proposed rearrangement
reflected the greater reactivity of secondary over primary
aliphatic carbons. Both possiblities could involve double bond
shifting but in mechanism 2 this would be obligatory (i.e. loo%),
whereas this need not
Bio
cata
l Bio
tran
sfor
mat
ion
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
15/1
2Fo
r pe
rson
al u
se o
nly.
-
32 D. J . LEAK AND H. DALTON
be so in mechanism 1. Analysis of these two possibilities will
require specifically labelled ("C) cyclohexene.
Alkyl Cyclohexenes
Although only a single standard for potential products was
available, namely 3-methy1-2-cyc1ohexeno1, it was possible to
obtain positive or tentative identifica- tion of most products by a
combination of gas chromatography, mass spectro- metry and chemical
reactivity. Quantification was based on the amount/area ratio for
3-methyl-2-cyclohexenol.
Some commercial samples of 1-methyl-1-cyclohexene appeared to
react spon- taneously in the assay buffer giving a complex mixture
of products by a process that has not been further investigated.
However, enzyme dependent assays were achieved with one batch,
which yielded four major products. On the basis of previous
analyses of products from alkylcyclohexanes and cyclohexene, it was
apparent that the order of elution from Carbowax 20M TPA for
products derived from a single substrate is 1) epoxides, 2) ring
ketones, 3) ring alcohols and 4) side chain products. While the
earliest product from 1-methyl-1-cyclohexene was stable to acid
treatment, its mass spectrum was inconsistent with assignment as
2-methylcyclohexanone. A peak at M/Z 79 (M-33, 24% abundance) is
charac- teristic of methylcyclohexenols, arising from loss of H 2 0
and CH3 and is not a feature of ketones. This product was therefore
assigned to 1-methyl-Zcyclo- hexenol on the basis of its short
retention time. Major fragmentation ions were also consistent with
the predicted retro-Diels-Alder degradation product (M-28) and
subsequent loss of CH3 (M-43). The most likely route to this
product is considered to be via epoxidation (transient?) and
rearrangement, the latter step possibly being purely chemical
(Figure 4).
Of two GC peaks evident in the secondary ring alcohol region,
the later, major peak was identical in all respects to the standard
3-methyl-2-cyclohexenol. The mass spectrum of the earlier peak
differed in the relative abundance of fragmentation ions, notably
that of the molecular ion M/Z 112 and the major fragment M/Z 97
(M-15). Both spectra, however, contained peaks at M-28 indicative
of a retro-Diels-Alder elimination of GH, (Figure 5). Retention of
the hydroxyl group in the larger fragment indicated that this
product was probably 2-methyl-2-cyclohexenol.
(3-methyl-3-cyclohexenol was also excluded on the basis of its 8
peak index.)
On the basis of earlier results the final peak was suspected of
being 1-cyclohexenemethanol. However, the mass spectrum of this
product revealed a
J
Figure 4 Possible mechanism for formation of 1-
methyl-2-cyclohexen-1-01 from l-methyl-l-cyclo- hexene catalyzed by
MMO.
Bio
cata
l Bio
tran
sfor
mat
ion
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
15/1
2Fo
r pe
rson
al u
se o
nly.
-
HYDROXYLATION SELECTIVITY OF MMO 33
Figure 5 Retro-Diels-Alder elimination of C,H, from the
molecular ion of 3-methyl-2-cyclohexen-1-01.
molecular ion at M/Z 110 and major fragment at M/Z = 82 (M-28)
ascribable to retro-Diels-Alder elimination as previously outlined.
The third most abundant fragment at M/Z = 54 could be [HC=C-CHO]+,
which would be consistent with the proposed assignment as
1-cyclohexene carboxaldehyde. If correct, this is an interesting
observation, as methanol dehydrogenase (a non-specific primary
alcohol dehydrogenase) would be inactive under these assay
conditions. This suggests that both the initial hydroxylation and
further oxidation of the methyl group were catalyzed by the MMO.
With 3- and 4-methylcyclohexenes as substrates no comparable peak
was observed, suggesting that hydroxylation of the methyl group was
dependent on its allylic position. Furthermore, the absence of side
chain hydroxylation with these substrates indicates that
unsaturation has a considerable orientating effect determining
hydroxylation specificity.
With 3-methyl-1-cyclohexene as substrate, two minor products
were identified as the epoxide (acid sensitive) and a saturated
ketone. Although positive identification of the latter was not
made, its mass spectrum differed from that ascribed to
1-methyl-2-cyclohexen-1-01. In fact, from the three possible
methyl- cyclohexanones the spectrum was closest to that of
4-methylcyclohexanone (data from “Eight Peak Index”). However, the
possibility of mixed , methylcyclo- hexanones has not been
examined. The two major products from 3-methyl-l- cyclohexene had
retention times consistent with assignment as ring alcohols (mass
fragment ions at M-18 supported this) and almost identical mass
spectra suggesting that they were epimers. The major fragment at
M/Z70 (M-42) was indicative of a neutral elimination process. It is
difficult to envisage an alternative to retro-Diels-Alder
elimination of C H 4 H 2 - C H 3 but this would require that the
product be either 5-methyl-2-cyclohexen-1-01 or
6-methyl-2-cyclohexen-l-ol, neither of which would have been
expected. Indeed, 4-methyl-2-cyclohexen-l-ol, the expected product
from allylic hydroxylation of 3-methyl-1-cyclohexene would be
expected to yield a fairly abundant fragment at M-28 by the same
process. This was completely absent. On this basis it is evident
that some rearrangement process must have occurred. Although
further analysis will be required for confirmation, the most likely
rearrangement is considered to be that yielding
6-methyl-2-cyclohexen-1-01 from a transient epoxide. According to
the work of Paris and Alexandre (1972), the major product would
then be the cis isomer.
The epoxide (acid sensitive) was a considerably more abundant
product from 4-methyl-1-cyclohexene than the previous two
substrates, indicating that the relative position of the methyl
group is important in determining the extent of epoxidation.
Unfortunately, identification of the remaining products was ham-
pered by the lack of mass spectral data. However, some tentative
suggestions may
Bio
cata
l Bio
tran
sfor
mat
ion
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
15/1
2Fo
r pe
rson
al u
se o
nly.
-
34 D. J. LEAK AND H. DALTON
be made. A minor product with similar retention time to the
ketone identified with the previous substrate was probably also a
ketone, 3- or 4-methylcyclo- hexanone being the likely candidates.
The two major products had retention times in the range ascribed to
secondary ring alcohols. Given the inherent symmetry of the allylic
intermediate formed by hydrogen removal at the 6 position, it is
probable that at least one of these products is 5-methyl-2-
cyclohexen-1-01. (A mechanism involving epoxide rearrangement would
give the same product.)
Active Site Modelling and Predictions of Reactivity The soluble
MMO from both M. capsulatus (Bath) (Dalton et al. , 1981) and
Methylosinus trichosporium OB3b (Jezequel and Higgins, 1983)
produce phenols from aromatic substrates via epoxide intermediates,
which subsequently re- arrange with concomitant hydride shift.
However, contradictory evidence has been presented for the
mechanism of hydroxylation at aliphatic carbons. The partial
inversion of stereochemistry observed in this study during the
hydroxylation of cis-1,4-dimethylcyclohexane at the 1 position
supports the conclusion that a stepwise mechanism is involved
(Jezequel and Higgins, 1983). Partial epimeriza- tion is also a
feature of some cytochrome P450 systems (Groves et al., 1978), with
which parallels may be drawn.
It is evident from the products obtained from
monoalkylcyclohexanes that alkyl groups have an orientating effect
on substrate approach to the active site, the site of attack being
distal to the alkyl group. Additionally, the evidence for steric
hindrance with t-butylcyclohexane suggests that during
hydroxylation a substrate the size of methylcyclohexane is
completely enclosed within the enzyme and that this binding site is
too narrow to accommodate a t-butyl group. Thus, in the case of
alkylcyclohexanes the regioselectivity of hydroxylation is
determined primarily by the steric constraints on substrate
binding. However, given this constraint, the evidence for a
stepwise mechanism indicates that the stereoselectivity of
hydroxylation results primarily from chemical criteria; hydroxyl
group addition occurs from the least hindered direction, giving
rise to predominantly diequatorial products. A more detailed
analysis will be necessary to ascertain any enzymic contribution in
this respect.
The requirement for substrate burial within the active site
provides a possible explanation for the hydroxylation products from
the dimethylcyclohexanes. Trarzs-l,4-dimethylcyclohexane is
naturally diequatorial in the energetically favourable cyclohexane
ring “chair” conformation and this is presumably the structure
present in the active site. However, cis-l,4-dimethylcyclohexane
can only be diequatorial in the “boat” conformation (Figure 6 (b)).
From the evidence for a preferential site of attack at the 3
position of methylcyclohexane, the products from cis- and
trans-l,4-dimethylcyclohexane can be readily under- stood if it is
assumed that:
1. penetration of the substrate beyond the active site is
limited, resulting in predominantly subterminal products where
steric constraints allow:
2. steric constraints in the active site favour the substrate
orientations depicted in Figure 6 (a) and (b) as opposed to those
in 6 (c) and (d) relative to an active site A; (energetically a
“skew-boat’’ conformation would be more likely than the full boat
of Figure 6 (b), but the outcome would be the same);
Bio
cata
l Bio
tran
sfor
mat
ion
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
15/1
2Fo
r pe
rson
al u
se o
nly.
-
HYDROXYLATION SELECTIVITY OF MMO 35
a
z3a A A C d
A A Figure 6 (a-d) Possible binding conformations of trans- (a
and c) and cis- (b and d), 1,4-dimethylcyclohexanes relative to the
active site 'A' of methane monooxygenase.
3. the position of the active site A is primarily the site of
hydrogen abstraction; subsequent hydroxylation is less
stereospecific, giving rise to the sterically most favourable
products.
Inversion of stereochemistry during hydroxylation at the 1
position with cis- but not truns-l,4-dimethylcyclohexane probably
reflects the tendency of the proposed planar intermediate to flip
to the energetically more favourable chair conformation.
The authors appreciate that this explanation can be no more than
a working model at this stage and that the energetic barrier to
substrates binding in the skew-boat or boat conformation is
considerable, particularly in the face of the low substrate
specificity of the MMO.
While the low percentage of hydroxylation at the 1 position with
the trans substrate may reflect the proportion of substrate bound
in conformation c (Figure 6) exposing the C1 hydrogen to the active
site (the major product presumably arises with substrate bound in
conformation a but incompletely buried), the greater relief of
steric strain in conversion of the cis isomer to a tertiary planar
intermediate should also be taken into account. The failure of
Dalton et ul. (1981) to observe ring opening during the
hydroxylation of cyclopropane and methylcy- clopropane, which led
to the proposal of direct oxygen insertion, suggests that the
intermediates formed during aliphatic hydroxylation may be
stabilized within the active site.
Alkyl substitution also has an orientating effect in determining
the products from methylcyclohexenes, except with
1-methyl-1-cyclohexene where the allylic position of the methyl
group is evidently both sterically and electronically more
favourable for hydroxylation. Given the models proposed in Figures
3 and 6, the greater proportion of epoxide formed from
4-methyl-compared to 3-methyl-l- cyclohexene would be consistent
with the proposal of Jezequel and Higgins (1983) that the same
active oxygen species could be responsible for hydrogen abstraction
or epoxidation depending on the nature of the substrate. However,
further work will be required on the nature of some of the
rearrangements taking place before firm conclusions can be
drawn.
Bio
cata
l Bio
tran
sfor
mat
ion
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
15/1
2Fo
r pe
rson
al u
se o
nly.
-
36 D. J . LEAK AND H. DALTON
Acknowledgements The authors would like to thank Donna Balaam
for technical assistance and Mr I. Kaytal for running the mass
spectra. DJL is grateful to Allelix Inc., Mississauga, Ontario,
Canada, for financial support.
References
Bradford, M. M. (1976) A rapid and sensitive method for the
quantitation of microgram quantities of protein utilising the
principle of protein-dye binding. Anal. Biochem., 72, 248-254
Burrows, K. J . , Cornish, A., Scott, D . and Higgins, I . J.
(1984) Substrate specificities of the soluble and particulate
methane mono-oxygenases of Methylosinus trichosporium OB3b. J .
Gen. Microbiol., 130, 3327-3333
Colby, J. and Dalton, H. (1978) Resolution of the methane
mono-oxygenase from Methylococcus capsulatus strain Bath into three
components; purification and properties of component C, an
iron-containing flavoprotein. Biochem. J . , 171, 461-468
Dalton, H. (1980) Oxidation of hydrocarbons by methane
mono-oxygenase from a variety of microbes. Adv. Appl. Microbiol.,
26, 71-87
Dalton, H., Golding, B. J. , Waters, B. W . , Higgins, R . and
Taylor, J. A. (1981) Oxidations of cyclopropane, methyl
cyclopropane, and arenes with the monooxygenase system from
Methylococcus capsulatus. J . Chem. SOC. Chem. Comm., 189,
482-483
Dalton, H. and Whittenbury, R. (1976) The acetylene reduction
technique as an assay for nitrogenase in the methane oxidizing
bacterium Methylococcus capsulatus strain Bath. Arch. Microbiol.,
109,
Green, J. and Dalton, H. (1985) Protein B of soluble methane
mono-oxygenase from Methylococcus capsulatus (Bath). J . Bid.
Chem., 260, 15795-15801
Groves, J. T., McClusky, G. A., White. R. E. and Coon, M. J .
(1978) Aliphatic hydroxylation by highly purified liver microsomal
cytochrome P450. Evidence for a carbon radical intermediate.
Biochem. Biophys. Res. Commun., 81, 154-160
Higgins, I . J., Best, D. J . and Hammond, R. C. (1980) New
findings in methane utilizing bacteria highlight their importance
in the biosphere and their commercial potential. Nature (London),
286,
Jezequel, S. G . and Higgins, I . J . (1983) Mechanistic aspects
of biotransformations by the mono-oxygenase system of Methylosinus
trichosporium OB3b. J . Chem. Tech. Biotechnol., 33B,
Lund, J. and Dalton, H. (1985) Further characterization of the
FAD and Fe,S, redox centres of component C, the NADH acceptor
reductase of soluble MMO of Methylococcuy capsulatus (Bath). Eur. J
. Biochem., 147, 291-296
Paris, C. and Alexandre, P. (1972) Stereochemical investigation
of cyclohexane and terpene compounds by gas chromatography. J .
Chromatog. Sci., 10, 402-41 1
Royal Society of Chemistry (1983) Eight Peak Index of Mass
Spectra, 3rd edn. Nottingham Staniey, S. H., Prior, S. D., Leak. D.
J . and Dalton, H. (1983) Copper stress underlies the
fundamental change in intracellular location of methane
monooxygenase in methane oxidizing organisms: studies in batch and
continuous cultures. Biorechnol. Lett., 5 , 487-492
Van Ravenswaay Claasen, J . C. and van der Linden, A. C. (1971)
Substrate specificity of the paraffin hydroxylase of Pseudomonas
aeruginosa. Ant. uan Leeuw., 37, 339-352
White, R. E., McCarthy, M. B., Egebert, K. E. and Sligar, S. G.
(1984) Regioselectivity in the cytochromes P-450: control by
protein constraints and by chemical reactivities. Arch. Biochem.
Biophys., 228, 493-502
Woodland, M. P. and Dalton. H. (1984) Purification and
properties of component A of the methane mono-oxygenase from
Methylococcus capsulatus (Bath). J . Biol. Chem., 259, 53-60
147-151
561-564
139-144
Bio
cata
l Bio
tran
sfor
mat
ion
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
15/1
2Fo
r pe
rson
al u
se o
nly.