On the bi-enzymatic behaviour of Saccharomyces cerevisiae-mediated stereoselective biotransformation of 2,6,6-trimethylcyclohex-2-ene-1,4-dione

Article abstract of DOI:10.1016/j.mcat.2018.01.002

Baker's yeast has been well-known to have the ability to reduce a variety of substrates into many optically active compounds. One of the important chemicals is (4R,6R)-4-hydroxy-2,2,6-trimethylcyclohexanone or in short, (4R,6R)-actinol, a product formed f

Full text of DOI:10.1016/j.mcat.2018.01.002

Molecular Catalysis 447 (2018) 56–64
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
On the bi-enzymatic behaviour of Saccharomyces cerevisiae-mediated
stereoselective biotransformation of
2
,6,6-trimethylcyclohex-2-ene-1,4-dione
Mohamad Hekarl Uzir , Nazalan Najimudinb
a,∗
a
School of Chemical Engineering, Engineering Campus, Universiti Sains Malaysia, Seri Ampangan, 14300 Nibong Tebal, Penang, Malaysia
School of Biological Sciences, Universiti Sains Malaysia, 11800 USM, Penang, Malaysia
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 23 September 2017
Received in revised form
Baker’s yeast has been well-known to have the ability to reduce a variety of substrates into many optically
active compounds. One of the important chemicals is (4R,6R)-4-hydroxy-2,2,6-trimethylcyclohexanone
or in short, (4R,6R)-actinol, a product formed from the reduction of 2,6,6-trimethylcyclohex-2-ene-1,4-
dione, (ketoisophorone). The work has successfully characterized the route of the reaction during the
conversion mediated by Saccharomyces cerevisiae and it was also experimentally observed the impor-
tance of cofactor (NADH/NADPH) availability and stability during the biotransformation. The presence
of cofactor has been proven to assist the product formation by keeping a continuous flow of hydrogen
ions to and from the reduction system, while the amount of enzymes present determined the rate of
both intermediates and product formed during the course of biotransformation. Results of the growing
cells, particularly during the exponential phase of the growth significantly differ to that of the stationary
phase system in terms of the formation of the intermediates and the final product. For the growing cell
biotransformation, the route only stopped at the intermediate compound (4) during the course of the
2
9 November 2017
Accepted 3 January 2018
Available online 4 February 2018
Keywords:
Saccharomyces cerevisiae
Whole-cell
Biotransformation
Stereoselective
Bi-enzymatic
3
0 h reaction, whereas for the stationary phase biotransformation, compound (5) was readily formed in
a range of time between 2 and 12 h of reaction, depending on the amount of cells available and glucose
supplied to the system. Within the yeast cell, there are two responsible enzymes that mediate the bio-
transformation; carbonyl reductase and enoate reductase that work both in parallel and sequential to
produce a chiral alcohol of (4R,6R)-actinol.
©
2018 Elsevier B.V. All rights reserved.
1
. Introduction
The reduction of ketoisophorone or 2,6,6-trimethylcyclohex-
2
-ene-1,4-dione (1) has long become an interest for its valuable
The application of biocatalysts to undergo chemical reactions
products as well as its interesting mechanism through bio-
transformation. The reduction of (1) was first synthesized and
reported for more than four decades ago and only recently, the
mechanism of the particular reduction is about to be fully under-
stood [7,8]. Fig. 1 shows the possible route of biotransformation
of ketoisophorone leading to the final product(s); (4R,6R)-4-
hydroxy-2,2,6-trimethylcyclohexanone, (5) ((4R,6R)-actinol) and
its corresponding enantiomer (6).
Various species of microorganisms have been screened for bet-
ter yield of compound (5), which is the precursor for a number
of optically active carotenoids and plant hormones such as zeax-
anthin and xanthoxin to name a few [9]. Zeaxanthin is a type of
carotenoid from the xanthrophyll group found both in human and
plants. Nowadays, zeaxanthin has been largely used as a natural
food colorant as well as an active ingredient in pharmaceutical
industry due to its high anti-oxidant properties [10]. It was also
has been widely practised by major chemical processing compa-
nies including food and pharmaceuticals [1]. This is due to the fact
that biocatalysts, in the form of either enzyme or whole-cell are
able to specifically target a particular bond or functional group of a
chemical compound [2,3]. However, in the natural environment of
a microorganism, there are intricate connections of enzymes that
work both in parallel and sequential in order to keep the biolog-
ical system alive. It is within these connections of enzymes that
one could exploit the ability of a microorganism into becoming an
important biocatalyst, which could reduce the cost of a process as
well as maintain a green environment [4–6].
∗ Corresponding author.
E-mail address: chhekarl@usm.my (M.H. Uzir).
https://doi.org/10.1016/j.mcat.2018.01.002
2
468-8231/© 2018 Elsevier B.V. All rights reserved.

M.H. Uzir, N. Najimudin / Molecular Catalysis 447 (2018) 56–64
57
Fig. 2. The use of cofactor as hydrogen ion transfers to both reductases-mediated
reductions. Compounds XRed and XOx represent the reduced and oxidized X respec-
tively, where X is a species that undergoes oxidation reaction.
of ketoisophorone (1) in 1988 [8]. Their observation showed that
the mycelium of Aspergillus niger released a plethora of enzymes
during the growth period, which resulted in the formation of com-
pounds (1)–(6) shown in Fig. 1. Prior to the work of Yamazaki, a
thermophilic bacterium from the species of Thermomonospora cur-
vata was used as the biocatalyst to perform the biotransformation.
The reaction however stopped only until compound (4) with the
concentration of 0.0065 mol/L (1.0 g/L) at a maximum cell growth
of 1.5 g/L [15]. The other compounds existed only at a trace level
within the reaction medium. The use of thermophile was further
extended with the combination of two different microorganisms.
Bacillus stearothermophilus was chosen to link with the T. curvata
system into a sequential biotransformation procedure. The reac-
tion then continued where compound (4) was further reduced into
compound (5) at a concentration of 1.5 g/L [16,25,26].
Fig. 1. Possible routes of ketoisophorone (1) reduction based on two enzymes natu-
rally produced by Saccharomyces cerevisiae. ER represents enoate reductase and CR
refers to carbonyl reductase.
reported that zaexanthin and other members of xanthrophylls fam-
ily could prevent the appearance of many diseases, for instance;
arteriosclerosis, cataracts, multiple sclerosis, macular degeneration
and even cancer [11]. Xanthoxin on the other hand is a type of plant
hormone that inhibits growth, found in abundant in plants such as
tomato and spinach [12]. In the previous report by Kobayashi and
co-workers, xanthoxin was also found in the extract of algae and
sea grass namely; Codium latum and Sargassum thunbergii, which
were collected from the coastal areas of Japan [13]. In some higher
plants biosynthetic pathway, xanthoxin acts as a precursor in the
formation of abscisic acid, in other words, a plant growth regulator
From the selected biocatalysts listed above, it could be sum-
marized that the rate of product formation is totally governed by
the two enzymes responsible in the reduction of ketoisophorone
[
14].
The application of whole-cell Baker’s yeast type-II, (Sac-
charomyces cerevisiae) has shown substantial productivity,
(1). However, for a reduction reaction in particular, the case is
a
rather challenging when the product formation is also dependent
on the amount of cofactor available within the cells [27–29]. The
cofactor that coupled the ketoisophorone (1) reduction is shown
in Fig. 2. Compounds XRed and XOx represented in the diagram
are the reduced and oxidized formed of X respectively, that occur
during the oxidation process. Since oxidation could take place by
many substrates within the cell, therefore, X represents an arbitrary
compound that undergoes such oxidation reactions.
The aim of this work is mainly to investigate the effect
of cofactor stability and enzymes availability towards bio-
transformation leading to the formation of (R)-4-hydroxy-2,2,6-
trimethylcyclohexanone, (5). The work also monitors the amount
of enzymes present within the cells and how they could eventu-
ally affect the route of the biotransformation. The formation of
particularly on the product’s regio- and stereoselectivity as com-
pared to other organisms such as Thermomonospora curvata and
Bacillus stearothermophilus [8,15–17]. Within these microorgan-
isms, there are two important enzymes responsible for the
formation of both products (5) and (6). Carbonyl reductase or com-
monly known as alcohol dehydrogenase (ADH), that mediates the
reduction of aldehyde and ketone to alcohol, while enoate reduc-
tase, normally termed as the Old Yellow Enzyme, (OYE) [18–20],
which catalyzes the reduction of the C C bond are the two enzymes
that stereoselectively reduced ketoisophorone (1). On separate
works on enzyme characterization, both carbonyl reductase and
enoate reductase have been successfully purified and character-
ized from different range of microorganisms. However, the main
focus was on the yeast genus including Baker’s yeast as well as
other species such as Saccharomyces carlsbergensis, Kazachstania sp.,
Kluyveromyces marxianus, Candida sp. and Zygosaccharomyces sp. to
name a few [21–24].
The possible routes of ketoisophorone (1) biotransformation
using different types of microorganisms are presented in Fig. 1.
However, the particular scheme only refers to Baker’s yeast-
mediated reduction, since the presence of compounds (3) and (6)
were not observed during the course of biotransformation. In the
earlier work of Leuenberger et al., S. cerevisiae was first used as
a biocatalyst in a number of experimental set-ups. The highest
fermentation volume of 200 L was used with the formation of inter-
mediate (4) at 60 g/L [7]. The work was later continued by many
Japanese researchers, applying different types of microorganisms
ranging from bacteria, fungi, different species of yeasts and plant
cells. Yamazaki et al. first reported the fungal-mediated reduction
(R)-4-hydroxy-3,3,5-trimethylcyclohexanone (6) was not followed
during this investigation due to the fact that S. cerevisiae-mediated
biotransformation of ketoisophorone (1) gave no trace of the com-
pound with no apparent peak on the chromatogram as separately
reported by Wada and Buque-Taboada et al. [30–32]. The outcome
of this work would provide a guideline to biochemists as well as
molecular biologists into carrying out molecular work on genes
expression for further product optimization.
2. Materials and method
2.1. Microorganism
Baker’s yeast type-II (Saccharomyces cerevisiae) in dried form
was purchased from Sigma-Aldrich, USA.

5
8
M.H. Uzir, N. Najimudin / Molecular Catalysis 447 (2018) 56–64
2.2. Chemicals
2.4.2. Stationary phase biotransformation at different glucose
concentrations
Standard ketoisophorone (1) (98%), potassium hydrogen phos-
For a stationary phase biotransformation, phosphate buffer was
used as an aqueous medium for cell suspension. Phosphate buffer
phate (K HPO ), potassium dihydrogen phosphate (KH PO ), d-
2
4
2
4
(
+)-glucose (>99.5%), Tris-HCl, bovine serum albumin (BSA), nicoti-
was prepared at pH 7 at a concentration of 0.1 M. 10.7 g of K HPO4
2
namide adenine dinucleotide (NADH), nicotinamide adenine dinu-
cleotide phosphate (NADPH), sodium hydroxide (NaOH) and ethyl
acetate were purchased from Sigma-Aldrich, USA. Luria-Bertani
and 5.24 g of KH PO4 were added and dissolved with 1 L dionized
2
water. 1 M sodium hydroxide was used to adjust the solution to the
required pH. The buffer solution was prepared with the addition of 3
different glucose concentrations (5, 10 and 15 g/L). This would pro-
vide an additional source of carbon especially for cofactor recycling
during the bioconversion process. 5 gdcw/L Baker’s yeast was pre-
pared in 250 mL buffer solution prior to biotransformation (with
conditions similar to the growing cell biotransformation). After
30 min of cells inoculation, 0.2 g/L ketoisophorone (1) was added
into the buffer solution and samples were withdrawn every hour
in order to check for substrate degradation, products formation and
cofactor availability and stability.
(
LB) growth medium which includes; yeast extract, peptone and
sodium chloride (NaCl) were also purchased from Sigma-Aldrich,
USA. Pure enantiomer of (S)-4-hydroxy-2,6,6-trimethylcyclohex-
2
-enone, (2) (>98%) and 2,2,6-trimethylcyclohexane-1,4-dione, (4)
>95%) were purchased from AAP Pharma Technologies (India) and
Angene International Limited (Hong Kong) respectively.
(
2.3. Growth of Baker’s yeast Type-II, Saccharomyces cerevisiae
The growth of S. cerevisiae was carried out in shake-flasks. LB
2.4.3. Stationary phase biotransformation at different cell
concentrations
medium as the growth nutrient was prepared with 10 g sodium
chloride, 10 g peptone and 5 g yeast extract dissolved in 1 L deion-
ized water. 5 g/L glucose was added as an additional carbon source
to the medium in a separate system for comparison purposes.
The experiment to determine the effect of different cell concen-
trations was also carried out in 250 mL phosphate buffer. 3 different
cell concentrations (7, 10 and 15 gdcw/L) were chosen with the addi-
tion of 5 g/L glucose. 0.2 g/L ketoisophorone (1) was introduced into
the cell suspension 30 min after the inoculation with biotransfor-
2
50 mL of the prepared medium was measured and poured into
◦
a baffled shake-flask. The flask was autoclaved at 121 C for 20 min.
Cooled medium was then ready to be inoculated with dried yeast
cells. 0.25 g of dried Baker’s yeast was measured and added to the
◦
mation conditions of 37 C and 150 rpm shaking speed. Samples
were withdrawn every hour in order to determine the remaining
substrate, the amount of products and the cofactor level during the
biotransformation.
2
50 mL LB medium resulting into 1 g/L initial cell concentration.
The procedure was carried out aseptically in a laminar flow cabinet
to ensure that the growth is free from contamination. Fermenta-
tion was carried out in an incubator-shaker (IKA, KS4000i Control,
2.5. Analytical methods
◦
Korea) at 37 C with the shaking speed of 150 rpm.
Samples from the fermentations were collected every hour in
order to observe the cell growth and later a profile was plotted
for both systems without and with the addition of glucose. The
fermentation was left to continue for about 6 days (154 h).
2.5.1. Substrate and products quantifications
Samples were withdrawn at least 2 times at every hour. For
a 0.5 mL sample withdrawn from the biotransformation medium,
0.5 mL ethyl acetate was added and the mixture was vigorously vor-
texed for at least 5 min. Sample was then centrifuged at 4000 rpm
for 10 min (Eppendorf 5702R, Germany) to separate the aqueous
and the organic phase. Once the separation was complete, the top
layer (organic phase) was removed and kept in a sample vial for gas
chromatographic analysis.
In order to determine the cell mass throughout the course of
fermentation, 1 mL of sample from the culture was withdrawn and
centrifuged at 4000 rpm for 10 min. Prior to centrifugation, 1 mL-
size of empty plastic Eppendorf tubes were dried in an oven and
weighed (dried-tube weight). Supernatant from the centrifuged
◦
sample was removed and the cell pellet was dried at 60 C for 1 h.
The dried cell within the tube was again weighed and the difference
between the tube with cell pellet and that with the empty ones were
calculated, which gives a measurement in terms of gdcw/L.
2.5.2. Cofactor (NADH/NADPH) analysis
Another 1 mL sample was taken from the reaction medium
and subjected to centrifugation for 10 min at 4000 rpm in a 1 mL
Eppendorf tube. The supernatant resulted after centrifugation was
removed and the cell pellet remained at the base of the tube was
added with 1 mL solution of BSA and Tris-HCl buffer. The mixture
was vortexed for 5 min until the pellet was completely suspended
in the buffer. The suspended cell was then undergone sonication
process in order to break open the cell membrane. Sonication
using Misonix Sonicator 3000 (Cole-Palmer, USA) was carried out
2.4. Shake-flask biotransformation
2.4.1. Growing cell biotransformation
The method was carried out in order to observe the cell capa-
bility in reducing ketoisophorone (1) into its respective products
during the growth period. This will provide a wider perspective in
terms of enzymes responsible for the conversion, cofactor availabil-
ity and inhibition effect by the substrate (1) should there be any.
By referring to the standard growth profiles of S. cerevisiae obtained
previously, fermentations were again repeated with a standard LB
◦
in ice (<5 C) for 10 cycles (5 s on and 5 s off) at the sonication
amplitude of 8 ␮m and later centrifuged at 4000 rpm for 10 min
to remove the cell debris. The supernatant was carefully trans-
ferred into a quartz cuvette and quickly checked for absorbance
using a UV–vis spectrophotometer (Cary 60, Agilent Technolo-
gies, USA) with the absorbance wavelength of NADH/NADPH set at
◦
medium and LB medium with additional 5 g/L glucose at 37 C and
1
50 rpm shaking speed. Growth is basically defined as cell division
3
40 nm. The method of cofactor quantification separately reported
where microorganisms actively divide themselves and later pro-
duce an exponential curve within the profile. The point where this
exponential curve begins marked the point where 0.2 g/L (≈1 mM)
of substrate (1) was introduced into the medium. As the fermen-
tation continued, samples were withdrawn from the shake-flasks
every hour to check for substrate consumption, products formation
and cofactor availability and stability.
by Walker and Rover Jr. with their co-workers was used with slight
modifications to suit with the yeast cells used in the experimenta-
tion [33,34].
2.5.3. Gas chromatographic analysis
Organic samples extracted from the biotransformation method
previously described in subsection 2.5.1 were kept in glass vials

M.H. Uzir, N. Najimudin / Molecular Catalysis 447 (2018) 56–64
59
Table 1
Specific growth rates of the growing S. cerevisiae without and with the addition of
5
g/L glucose.
Specific growth rate ␮ (h 1
−
)
Growth medium content
Without glucose
With glucose
0.078
0.129
Fig. 3. Growth profile of Saccharomyces cerevisiae with different phases denoted as
(
A) lag phase, (B) exponential phase, (C) stationary phase and (D) death phase. (ꢀ,
growth with 5 g/L glucose, ᭹, growth without 5 g/L glucose).
Fig. 4. Cofactor stability in a biotransformation during the exponential phase of S.
cerevisiae (ꢀ, cell growth with 5 g/L glucose, ᭹, cell growth without 5 g/L glucose,
ꢀ, cofactor absorbance from growth with 5 g/L glucose, ꢀ, cofactor absorbance from
growth without 5 g/L glucose).
for chromatographic analysis. Standards were performed prior
to samples analysis using 7820A Gas Chromatography, (Agi-
lent Technologies, USA) equipped with flame ionization detector
and a capillary column (0.25 ␮m, 0.25 mm 30 m, MEGA-DEX
◦
DMT Beta, Italy). The oven was initially set at 120 C for
to the cell, therefore, the phases did not differ much during the
period of 6 days of fermentation. For both fermentations, the spe-
cific growth rates, ␮ for medium without and with 5 g/L glucose are
◦
◦
2
min and ramped to 210 C at a rate of 2 C/min. The injector
◦
temperature was maintained at 220 C with helium as a car-
rier gas with a flow rate of 1.0 mL/min. Pure ketoisophorone
−
1
0.078 and 0.129 h respectively as listed in Table 1. These values
apparently show that the additional glucose in the growth medium
does not give much effect to the growth curve until at the very end
of the fermentation period.
(
1), (S)-4-hydroxy-2,6,6-trimethylcyclohex-2-enone, (2) and 2,2,6-
trimethylcyclohexane-1,4-dione, (4) were prepared according to
standard concentrations in ethyl acetate prior to injection. 1 ␮L of
each component was analyzed in triplicates with individual cal-
ibration curves quoted at 95% confidence interval based on five
From the growth phase diagram obtained, the lag phase (region
A) was observed where the cell took approximately 5 h to stabilize
and adjust to the surrounding new environment. Due to this reason,
biotransformation was only started by introducing the substrate (1)
at the fifth hour after the fermentation began. Samples were then
taken every hour to check for substrate consumption, product(s)
formation, cell growth as well as cofactor availability and stability.
Fig. 4 apparently represents the growth of cells and the availability
2
independently prepared standard solutions (0.1–6.0 g/L, R > 98%).
3
. Results and discussion
3.1. Biotransformation of ketoisophorone (1) during the growth
phase of Saccharomyces cerevisiae
of NADH/NADPH within the cells (in the form of absorbance, A340
)
Prior to carrying out biotransformation during the growth phase
of the cells, standard fermentation of S. cerevisiae was conducted in
a shake-flask. A standard LB medium was prepared as previously
described and 1 gdcw/L cell was inoculated in a 250 mL medium. The
at any particular times during biotransformation. The cell seems to
correspond considerably well to the substrate as the biotransfor-
mation progresses. It was earlier reported by Buque-Taboada et al.
that the substrate (1) would only affect the cells for concentration
above 80 mM (12.17 g/L) [32]. Therefore, the substrate concentra-
tion used in the present study would not show any inhibition and
toxicity effects as that observed in the previous investigation.
As the cells increase in density, the cofactor detection through
the UV–vis also follows the same trend. However, the only differ-
ence observed is on the effect of glucose added to the medium,
where it stabilizes the cofactor (NADH/NADPH) produced by the
cells. A sudden decrease of cofactor was observed after about thir-
teenth hour after the reaction started. This could be due to the fact
that the cells were still adjusting themselves after the addition of
ketoisophorone (1) into the medium and a sudden plunge of cofac-
tor absorbance provides a clear picture of the sudden use of the
available cofactor for substrate conversion before the cells need to
produce more for other cellular application. This is supported by
the appearance of 2,2,6-trimethylcyclohexane-1,4-dione, (4) at fif-
teenth and eleventh hours of biotransformation without and with
glucose respectively (Fig. 5(A and B)).
◦
culture was left in an incubator-shaker set at 37 C and 150 rpm for
3
0 min before the first sample was withdrawn for growth quantifi-
cation. The period of 30 min was given for the cells to be completely
mixed and suspended in the medium as well as adapted to the
controlled environment. A similar set of culture was also prepared
using LB medium with the addition of 5 g/L glucose. 3 sets of fer-
mentation cultures were run in order to obtain consistent phases
as depicted in Fig. 3. This phase diagram, particularly the expo-
nential phase of the yeast growth (region B), shows that the cells
were actively dividing until a period where the cell division stopped
and reached a stationary phase. From the growth phase diagrams
(
medium with and without 5 g/L glucose), the time at which the
phase shifts from one phase to the other is approximately simi-
lar for each set with different of only ± 1 h. Fig. 3 provides a clear
indication of the exact time where the substrate should be intro-
duced into the growing cell. Since the introduction of glucose in
the growth medium is only to provide an additional carbon source

6
0
M.H. Uzir, N. Najimudin / Molecular Catalysis 447 (2018) 56–64
Fig. 6. A simplified version of the glycolysis steps combined with the TCA cycle
showing the theoretical formation of cofactor (NADH) in yeasts/bacteria.
link between the redox reactions and later conclude that the com-
pound is also important in determining the rate of the reaction [36].
Fig. 6 provides a simplified version of the glycolysis steps and that
connected to the TCA (tricarboxylic acid) or the Kreb’s cycle, which
are the main source of cofactor regeneration. Theoretically, with the
total of 10 cofactor molecules formed from each glucose structure,
the additional glucose could therefore maintain the availability and
stability of cofactor needed for the redox biotransformation.
This experiment utilized phosphate buffer as the aqueous
medium for cell suspension, where the main purpose is to remove
all the nutrients required for cell division and growth could then
be suppressed. Since the maximum cell growth was approximately
5 gdcw/L in the previous experiments, thus, the amount of yeast
inoculated in the medium was also of the same amount.
Fig. 5. Biotransformation of 2,6,6-trimethylcyclohex-2-ene-1,4-dione (1) (᭹) with
the formation of intermediate 2,2,6-trimethylcyclohexane-1,4-dione (4) (᭿) in
growth media (A) without 5 g/L glucose and (B) with 5 g/L glucose.
Fig. 7 provides a clear picture of the progress of the bio-
transformation with three different glucose concentrations. The
consumption of ketoisophorone (1) was apparently observed after
the fourth hour of reaction, however, the intermediate product, (4)
was only detected 4 h later (Fig. 7(A)). The delay of the product
formed could be due to the insufficient enoate reductase to carry
out the conversion of substrate (1). The problem of inadequate
cofactor could be ruled out as the compound was considerably
stable with only minor fluctuations at the absorbance value of
approximately ± 0.25. The reaction was then prolonged for another
Throughout the course of biotransformation, either with the LB
medium with or without glucose, the yeast cells seem to produce
only the intermediate (4) after about 30 h of reaction. It was also
observed that with the maximum cell density of approximately
5
2
gdcw/L, the organism could only reduce the substrate (1) into
,2,6-trimethylcyclohexane-1,4-dione, (4), with enoate reductase
dominating the group of proteins formed during the growth period.
This observation is similar to that observed by Hori et al. on the
biotransformation of ketoisophorone (1) with T. curvata as a bio-
catalyst, where only compound (4) appeared for a 30 h period of
fermentation [15].
9
h where an apparent additional peak was observed on the chro-
matogram. The position and retention time of the peak were later
confirmed with that obtained by Yamazaki et al., which to be
(R)-4-hydroxy-2,2,6-trimethylcyclohexanone, (5) [8]. Since there
was no standard compound purchased for (R)-4-hydroxy-2,2,6-
trimethylcyclohexanone, the mol percent (mol%) of the compound
detected by the gas chromatograph was used instead. This result
also gives a preliminary information about the existence of car-
bonyl reductase at the very end of the biotransformation. Upon
comparing the three amounts of product (5) in (mol%) (data shown
separately in Fig. 7(A–C)), they were approximately similar (with
concentration of 12 ± 2 mol%). The only difference between the
three glucose concentrations is the rate at which the products
were formed. On the other hand, compound (R)-4-hydroxy-3,3,5-
trimethylcyclohexanone, (6) was not observed throughout the
3
.2. Biotransformation of ketoisophorone (1) with variable
glucose concentrations for the maintenance of cofactor
regeneration
In all biotransformation involving redox reaction, one of the
+
important aspects is the transfer of hydrogen ion (H ) to and from
the main reaction. The hydrogen transfer process from an oxidized
system to that of the reduced system is carried out by a cofactor and
the generation of cofactor is dependent on the most basic cellular
metabolism of glucose breakdown or glycolysis [35]. Therefore the
aim of varying glucose concentration is mainly to look into the abil-
ity of the cells to generate more cofactor (NADH/NADPH) in order to

M.H. Uzir, N. Najimudin / Molecular Catalysis 447 (2018) 56–64
61
Table 2
The initial rates of product(s) formation during the biotransformation of 2,6,6-trimethylcyclohex-2-ene-1,4-dione (1) in phosphate buffer at different glucose concentrations.
Glucose concentration (g/L)
Initial rate of products formation
2,2,6-trimethylcyclohexane-1,4-dione, (4) (g/L-h)
(R)-4-hydroxy-2,2,6-trimethylcyclohexanone, (5) (mol%/h)
−
−
4
4
4
5
5.2 × 10
4.0 × 10
0.37
0.91
0.84
1
1
0
5
9.0 × 10−
Table 3
The initial rate of intermediates and product formations during the biotransformation of 2,6,6-trimethylcyclohex-2-ene-1,4-dione (1) in phosphate buffer at different cell
concentrations.
Cell concentration (g/L)
Initial rate of products formation
S)-4-hydroxy-2,6,6-
(
2,2,6-trimethylcyclohexane-1,4-
dione, (4)
(g/L-h)
(R)-4-hydroxy-2,2,6-
trimethylcyclohexanone, (5)
(mol%/h)
trimethylcyclohex-2-enone, (2)
g/L-h)
(
5a
7
0
5.2 × 10
−4
0.37
0.87
1.04
1.60
−
−
2
3
−3
1.5 × 10
2.4 × 10
0
8.4 × 10
−2
1
1
0
5
1.4 × 10
−
3
8.9 × 10
a
Results quoted from the biotransformation of substrate using 5 gdcw/L cell concentration with 5 g/L glucose concentration in buffer solution (Table 2).
course of reaction as there was no apparent peak detected on
the chromatogram as earlier reported by Leuenberger and Buque-
Taboada with their co-workers, utilizing the same microorganism
ketoisophorone (1). A progress similar to that of the 5 gdcw/L cells
was observed with the appearance of a trace of (S)-4-hydroxy-
2,6,6-trimethylcyclohex-2-enone, (2) in the 7 gdcw/L cells, which
was not observed in the low concentration of cells. By referring to
Fig. 8(A), the intermediate detected at approximately eleventh hour
of the reaction shows that enough carbonyl reductase exists within
the cells to perform the reduction of the C O functional group [38].
The formation of (S)-4-hydroxy-2,6,6-trimethylcyclohex-2-enone,
(2) from the biotransformations using 7 and 10 gdcw/L cells listed
in Table 3 with the rates of 1.5 × 10 and 2.4 × 10 g/L-h respec-
tively, proved that higher amount of cells could provide enough
carbonyl reductase to mediate the reduction of the ketone func-
tional group. Moreover, with the amount of cells exceed that of
the glucose concentration, the biotransformation seems to main-
tain its production of compound (5) (Fig. 8(A and B)). Despite the
deteriorating cofactor stability after the eighth hour of reaction,
product (5) was still increasing as a result of the gradual decrease
of both intermediates (2) and (4). This is due to the fact that there
were still cofactor molecules available to transfer hydrogen ions
to the reduction system. Based on this observation, it could be
hypothesized that biotransformation could still proceed with the
minimum amount of cofactor for the cells to maintain their func-
tions. However, with such bare minimum cofactor available, the
rate of product formed would not be as high as that with the addi-
tional glucose.
[7,31,32].
The rates of product formation are listed in Table 2. Even
though the rates were calculated in different units, the speed could
however explain about the enzymes that mediate the biotransfor-
mation, in other words, the activity of the particular enzyme. The
more enzymes available in the cells, the faster a reaction proceeds.
This could be observed from the formation of product (4) in Table 2
−
2
−3
−4
at 5.2 × 10 g/L-h for 5 g/L glucose with a slight drop when glucose
was doubled in another set of reaction. Since the amount of glucose
was twice that of the cells, the extra carbon source could probably
be used by the cells to generate more enzymes, and in this case, the
carbonyl reductase used to reduce one of the C O bonds from the
structural formula. This is clearly depicted in the higher rate of for-
mation of product (5) in 10 g/L glucose as compared to that of 5 g/L.
A similar observation could also be seen for higher glucose con-
centration of 15 g/L. Compound (4) was formed at a rate more than
twice that of the 10 g/L glucose concentration, which result could
be deduced that enoate reductase was generated in greater amount
from the excess carbon source provided [37]. An excess amount of
glucose was also used to produce more carbonyl reductase, which
resulted in a more or less the same concentration of product (5) in
all three glucose concentrations.
The effect of even higher cell concentration, in the case of
1
5 gdcw/L gave a rather conclusive observation on the behaviour
3
.3. Effect of different cell concentrations during the
of this complex bi-enzymatic biotransformation. Fig. 8(C) provides
an interesting observation of the dynamics of the 2-enzyme system.
It is commonly known that for a system with higher cell concentra-
tion, the result of biotransformation would be improved in terms of
product formation and the rate that the product is formed, which
of course conformed by the current observation on the formation
of (R)-4-hydroxy-2,2,6-trimethylcyclohexanone, (5). However, by
looking at the progress of both intermediates (2) and (4), it is
apparent that intermediate (2) was only detected in traces amount
biotransformation of ketoisophorone (1)
A part from monitoring the effect of glucose concentrations
towards product formation, the amount of cells at a particular
instance would also give a different progress of biotransformation.
This effect could be slightly seen in the earlier reaction utilizing the
growing-cell system. The result shown in Fig. 5(B) of the growing
cell biotransformation clearly shows such effects where the grad-
ual increase of cells would also stabilize the cofactor level for the
H
(similar results for 10 gdcw/L cell concentration) as compared to
+
transfer processes (Fig. 4).
intermediate (4), where the accumulation was observed until the
end of the experiment. It is obvious from the plots that there are two
possibilities which could be deduced based on these intermediate
compounds. Firstly, the traces amount of intermediate (2) detected
Due to these observations, a separate set of experiments was
conducted to closely monitor the progress of biotransformation
at different amount of cell concentrations. The first cell con-
centration of 7 gdcw/L was used to carry out the reduction of

6
2
M.H. Uzir, N. Najimudin / Molecular Catalysis 447 (2018) 56–64
Fig. 7. Profiles of S. cerevisiae-mediated biotransformation of 2,6,6-
trimethylcyclohex-2-ene-1,4-dione (1), (᭹) in phosphate buffer, forming an
intermediate, 2,2,6-trimethylcyclohexane-1,4-dione (4), (᭿) and the final product
Fig. 8. Profiles of S. cerevisiae-mediated biotransformation of 2,6,6-
trimethylcyclohex-2-ene-1,4-dione (1), (᭹) in phosphate buffer, forming
intermediates, 2,2,6-trimethylcyclohexane-1,4-dione (4), (᭿) and (S)-4-
hydroxy-2,6,6-trimethylcyclohex-2-enone, (2), (ꢁ) and the final product
(R)-4-hydroxy-2,2,6-trimethylcyclohexanone (5), (ꢀ). The stability of cofactor
NADH/NADPH) during the course of reaction is given by (♦). (A) 5 g/L glucose, (B)
(
(
R)-4-hydroxy-2,2,6-trimethylcyclohexanone (5), (ꢀ). The stability of cofac-
tor (NADH/NADPH) during the course of reaction is given by (♦). (A) 7 gdcw/L cells,
B) 10 gdcw/L cells and (C) 15 gdcw/L cells.
1
0 g/L glucose and (C) 15 g/L glucose.
(
through the gas chromatographic analysis could be due to either
there was no or low activity of carbonyl reductase or the conver-
sion of intermediate (2) into the final product (5) was relatively
fast that it would not be able to accumulate within the reaction
medium, in other words, the conversion was fast enough that once
the intermediate was formed, the enoate reductase available would
immediately convert it into product (5).
Secondly, for the accumulated intermediate (4) in the reaction
medium, the readily available enoate reductase from the cell would
convert substrate (1) at a relatively higher rate of conversion, which
continuously increased the concentration of (4) in the reaction
medium. In addition, the rate of conversion of intermediate (4) to
the final product (5) was considerably slow, which could be related
to the slow activity of the existing carbonyl reductase within the
cells. These possibilities could be depicted in a form of diagram
given in Fig. 9. The diagram gives a clear scenario of the behaviour of
the bi-enzymatic biotransformation system discussed above. Even
though the final compound (5) was continuously accumulated as
the reaction progresses, the system could still be optimised in terms
of the availability and improved activity of carbonyl reductase.
Moreover, the reduced stability of cofactor as shown in Fig. 8(C)
could also be improved by perhaps introducing a continuous feed-
ing strategy of glucose throughout the course of biotransformation.

M.H. Uzir, N. Najimudin / Molecular Catalysis 447 (2018) 56–64
63
References
[
[
1] J. Hardinge, Flavorings-a recipe for regulation, Chem. Ind. (1990) 694.
2] D.W. Armstrong, H. Yamazaki, Natural flavors production—a biotechnological
approach, Trends Biotechnol. 4 (1986) 264–268.
[
[
3] D.W. Armstrong, B. Gillies, H. Yamazaki, Natural flavors produced by
biotechnological processing, ACS Symp. Ser. 388 (1989) 105–119.
4] H. Simon, J. Bader, H. Gunther, S. Neumann, J. Thanos, Chiral compounds
synthesized by biocatalytic reductions, Angew. Chem. Int. Ed. Engl. 24 (1985)
539–553.
[
[
[
5] F. Lopez-Gallego, C. Schmidt-Dannert, Multi-enzymatic synthesis, Curr. Opin.
Chem. Biol. 14 (2010) 174–183.
6] E. Ricca, B. Brucher, J.H. Schrittwieser, Multi-enzyme cascade reactions:
overview and perspectives, Adv. Synth. Catal. 353 (2011) 2239–2262.
7] H.G.W. Leuenberger, W. Boguth, E. Widmer, R. Zell, Synthesis of optically
active natural carotenoids and structurally related compounds, Helv. Chim.
Acta 59 (1976) 1832–1849.
[
[
8] Y. Yamazaki, Y. Hayashi, N. Hori, Y. Mikami, Microbial conversion of
Fig. 9. The possible scenarios of the bi-enzymatic behaviour at higher concentra-
tion of S. cerevisiae. ER represents enoate reductase and CR represents carbonyl
reductase. Compound (1) is the substrate 2,6,6-trimethylcyclohex-2-ene-1,4-dione,
compound (2) refers to (S)-4-hydroxy-2,6,6-trimethylcyclohex-2-enone, compound
4-oxoisophorone by Aspergillus niger, Agric. Biol. Chem. 52 (1988) 2919–2920.
9] H. Mayer, Synthesis of optically active carotenoids and related compounds,
Pure Appl. Chem. 51 (1979) 535–564.
[
10] L.C. Mata-Gomez, J.C. Montanez, A. Mendez-Zavala, C.N. Aguilar,
Biotechnological production of carotenoids by yeasts: an overview, Microb.
Cell Fact. 13 (2014) 1–11.
(4) refers to 2,2,6-trimethylcyclohexane-1,4-dione and compound (5) represents the
final product (R)-4-hydroxy-2,2,6-trimethylcyclohexanone.
[
[
[
11] P. Bhosale, P.S. Bernstein, Microbial xanthophylls, Appl. Microbiol. Biotechnol.
68 (2005) 445–455.
4
. Conclusions
12] S. Setha, N. Hirai, H. Ohigashi, S. Kondo, Changes of xanthoxin levels during
fruit development in sweet cherries, J. Jpn. Soc. Hort. Sci. 75 (2006) 185–187.
13] M. Kobayashi, A. Kawabata, A. Chindanonda, A. Sakurai, Isolation and
identification of xanthoxin from marine plants, Agric. Biol. Chem. 54 (1990)
The reduction of ketoisophorone (1) into a useful chiral inter-
mediate for the synthesis of carotenoids and other pharmaceutical
compounds using Baker’s yeast type-II as the biocatalyst provides
a complex and interesting reaction mechanism. The interaction
between cofactor recycling and enzymes formation within the
yeast cells is interesting and important information for a bi-
enzymatic system to be carefully checked and understood prior to
genes expression and cloning. Since the reaction is highly depen-
dent on the cofactor (NADH/NADPH) availability, therefore, carbon
source, such as glucose must be provided to ensure a continuous
product formation. Despite of the fact that cofactor is only required
in the recycling of the hydrogen ions within the redox system, its
present in excess through glucose metabolism could eventually sta-
bilize the system and consequently prolong the biotransformation
for certain period of time. The present work also showed that the
growing cell failed to continue the reduction towards the final prod-
uct (5), as cells were concentrating on dividing themselves with
only enoate reductase formed during the course of the growth
phase. It is suspected that carbonyl reductase would only form
once the cell reached the stationary phase with a continuous supply
of glucose, which work is still currently in progress. Additionally,
the amount of biocatalyst also plays an important role in the bio-
transformation reaction as it readily provides the initial enzyme
concentration, for substrate conversion.
For a biotransformation system that depends on two types of
enzymes, the information on which protein exists first and the
amount present within the cell is essential in determining which
genes needs to be enhanced in order to optimize the system [39].
Since the results also relate the carbon source to that of the pro-
teins formation, therefore, the energy landscape in protein folding
should also be taken into account in the analysis of reaction rate and
products formation [40]. The finding from this work is believed to
be the first reported work in understanding the behaviour of the
in-vivo bi-enzymatic ketoisophorone biotransformation system.
2
723–2724.
[14] K. Sakai, K. Takahashi, T. Nukano, Convenient systhesis of optically active
abscisic acid and xanthoxin, Tetrahedron 48 (1992) 8229–8238.
15] N. Hori, T. Hieda, Y. Mikami, Microbial conversion of 4-oxoisophorone by
[
thermophile, Thermomonospora curvata, Agric. Biol. Chem. 48 (1984) 123–129.
[16] K. Nishii, K. Sode, I. Karube, Sequential 2-step conversion of 4-oxoisophorone
to 4-hydroxy-2,2,6-trimethylcyclohexanone thermophilic bacteria, Appl.
Microbiol. Biotechnol. 33 (1990) 245–250.
17] G.K. Khor, M.H. Uzir, Saccharomyces cerevisiae: a potential stereospecific
reduction tool for biotransformation of mono- and sesquiterpenoids, Yeast 28
(2011) 93–107.
18] Y.S. Niino, S. Chakraborty, B.J. Brown, V. Massey, A new yellow enzyme of
saccharomyces cerevisiae, J. Biol. Chem. 270 (1995) 1983–1991.
19] K. Stott, K. Saito, D.J. Thiele, V. Massey, Old yellow enzyme: the discovery of
multiple isozymes and a family of related proteins, J. Biol. Chem. 268 (1993)
[
[
[
6
097–6106.
[
20] C. Drewke, M. Ciriacy, Overexpression: purification and properties of alcohol
dehydrogenase IV from Saccharomyces cerevisiae, Biochim. Biophys. Acta 950
(1988) 54–60.
21] E.D. Amato, J.D. Stewart, Applications of protein engineering to members of
the old yellow enzyme family, Biotechnol. Adv. 33 (2015) 624–631.
22] U. Lutstorf, R. Megnet, Multiple forms of alcohol dehydrogenage in
Saccharomyces cerevisiae, Arch. Biochem. Biophys. 126 (1968) 933–944.
23] R.E. Williams, N.C. Bruce, New uses for an old enzyme-the old yellow enzyme
family of flavoenzymes, Microbiology 148 (2002) 1607–1614.
24] S. Raimondi, D. Romano, A. Amaretti, F. Molinari, M. Rossi, Enoate reductase
from non-convertional yeast: bioconversion, cloning and functional
expression in Saccharomyces cerevisiae, J. Biotechnol. 156 (2011) 279–285.
25] K. Nishii, K. Sode, I. Karube, Microbial conversion of dihidrooxoisophorone
[
[
[
[
[
(
DIOP) to 4-hydroxy-2,2,6-trimethylcyclohexanone (4-HTMCH) by
thermophilic bacteria, J. Biotechnol. 9 (1989) 117–128.
[26] K. Nishii, K. Sode, I. Karube, Continuous two-step bioconversion of
-oxoisophorone (OIP) to 4-hydroxy-2,2,6-trimethylcyclohexanone
4-HTMCH) by thermophilic bacteria, J. Biotechnol. Lett. 17 (1995) 785–790.
4
(
[
[
[
[
27] W. Su, W.-H. Xiao, Y. Wang, D. Liu, X. Zhou, Y.-J. Yuan, Alleviating redox
imbalance enhances 7-dehydrocholestrol production in engineered
Saccharomyces cerevisiae, PLoS One 10 (2015).
28] M. Geier, C. Brnadner, G.A. Strohmeier, M. Hall, F.S. Hartner, A. Glieder,
Engineering Pichia pastoris for improved NADH regeneration: a novel chassis
strain for whole-cell catalysis, Beilstein J. Org. Chem. 11 (2015) 1741–1748.
29] S. Kara, J.H. Schrittwieser, F. Hollmann, M.B. Ansorge-Schumacher, Recent
trends and novel concepts in cofactor-dependent biotransformation, Appl.
Microbiol. Biotechnol. 98 (2014) 1517–1529.
30] M. Wada, A. Yoshizumi, Y. Noda, M. Kataoka, S. Shimizu, H. Takagi, S.
Nakamori, Production of a doubly chiral compound
Acknowledgements
(
4R,6R)-4-hydroxy-2,2,6-trimethylcyclohexanone by two-step enzymatic
asymmetric reduction, Appl. Environ. Microbiol. 69 (2003) 933–937.
The authors would like to thank the Ministry of Higher Edu-
cation (MoHE) for providing the Fundamental Research Grant
Scheme, (FRGS) (PJKIMIA.6071248). The help of Umapathy N.,
Muniandy H. R. S. and Zakaria N. Z. in the completion of this work
was greatly acknowledged.
[31] E.M. Buque-Taboada, A.J.J. Straathof, J.J. Heijnen, L.A.M. van der Wielen, In-situ
product removal using loop in asymmetric reduction of 4-oxoisophorone by
Saccharomyces cerevisiae, Biotechnol. Bioeng. 86 (2004) 795–800.
32] E.M. Buque-Taboada, A.J.J. Straathof, J.J. Heijnen, L.A.M. van-der-Wielen,
Substrate inhibition and product degradation during the reduction of
4-oxoisophorone by Saccharomyces cerevisiae, Enzyme Microb. Technol. 37
[
(
2005) 625–633.

6
4
M.H. Uzir, N. Najimudin / Molecular Catalysis 447 (2018) 56–64
[
[
33] J.R.L. Walker, Spectrophotometric determination of enzyme activity: alcohol
dehydrogenase (ADH), Biochem. Educ. 20 (1992) 42–43.
34] L. Rover Jr., J.C.B. Fernandes, G. de Olivera-Neto, L.T. Kubota, E. Katekawa,
S.H.P. Serrano, Study of NADH stability using ultraviolet-visible
[38] M.P. Sibi, J.W. Christensen, Asymmetric reduction of ketones using bakers’
yeast, in: S.M. Roberts, G. Poignant (Eds.), Catalysts for Fine Chemical
Synthesis: Hydrolysis, Oxidation and Reduction, John Wiley & Sons Ltd., 2002,
pp. 137–142.
spectrophotometric analysis and factorial design, Anal. Biochem. 260 (1998)
[39] J.D. Stewart, Organic transformations catalyzed by engineered yeast cells and
related systems, Curr. Opin. Biotechnol. 11 (2000) 363–368.
[40] J.D. Bryngelson, J.N. Onuchic, N.D. Socci, P.G. Wolynes, Funnels, pathways, and
energy landscape of protein folding: a synthesis, Proteins: Struct. Funct.
Genet. 21 (1995) 167–195.
5
0–55.
[
[
[
35] J.W. Chin, R. Khankal, C.A. Monroe, C.D. Maranas, P.C. Cirino, Analysis of
NADPH supply during xylitol production by engineered Escherichia coli,
Biotechnol. Bioeng. 102 (2009) 209–220.
36] Z. Xiao, C. Lv, C. Gao, J. Qin, C. Ma, Z. Liu, P. Liu, L. Li, P. Xu, A novel whole-cell
biocatalyst with NAD(+) regeneration for production of chiral chemicals, PLoS
One 5 (2010) 1–6.
37] R. Stuermer, B. Hauer, M. Hall, K. Faber, Asymmetric bioreduction of activated
C
C bonds using enoate reductase from the old yellow enzyme family, Curr.
Opin. Biotechnol. 11 (2007) 203–213.