Substrate specificity of a novel squalene-hopene cyclase from Zymomonas mobilis

Article abstract of DOI:10.1016/j.molcatb.2012.02.007

Squalene-hopene cyclases (SHC; EC 5.4.99.17) catalyze the cyclization of triterpenoids via cationic intermediates in one of the most complex reactions known in biochemistry. In this study, we report the functional expression of a novel SHC from the ethanol producing bacterium Zymomonas mobilis (ZmoSHC1; YP-163283.1). Biochemical characterization of ZmoSHC1 uncovered unique substrate activity patterns compared to the previously reported AacSHC from Alicyclobacillus acidocaldarius and ZmoSHC2, the second squalene-hopene cyclase from Z. mobilis. ZmoSHC1 showed cyclization of the non-natural substrates homofarnesol (C₁₆) and citronellal (C₁₀) in addition to hopene formation from squalene (C₃₀). Moreover, ZmoSHC1 turned out to reveal high biocatalytic stability during long-term incubations. Remarkably, ZmoSHC1 exhibited a shift of activity towards substrates of shorter chain lengths, displaying over 50-fold higher conversion of homofarnesol and more than 2-fold higher conversion of citronellal in comparison to squalene conversion.

Full text of DOI:10.1016/j.molcatb.2012.02.007

Journal of Molecular Catalysis B: Enzymatic 84 (2012) 72–77
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Substrate specificity of a novel squalene-hopene cyclase from Zymomonas mobilis
Miriam Seitza , Janosch Klebensbergera , Sascha Siebenhallera , Michael Breuerb , Gabriele Siedenburgc ,
Dieter Jendrossekc, Bernhard Hauera,∗
a Institute of Technical Biochemistry, Universitaet Stuttgart, Allmandring 31, D-70569 Stuttgart, Germany
b BASF SE, Fine Chemicals and Biocatalysis Research, D-67056 Ludwigshafen, Germany
c Institute of Microbiology, Universitaet Stuttgart, Allmandring 31, D-70569 Stuttgart, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Squalene-hopene cyclases (SHC; EC 5.4.99.17) catalyze the cyclization of triterpenoids via cationic
intermediates in one of the most complex reactions known in biochemistry. In this study, we report
the functional expression of a novel SHC from the ethanol producing bacterium Zymomonas mobilis
(ZmoSHC1; YP 163283.1). Biochemical characterization of ZmoSHC1 uncovered unique substrate activity
patterns compared to the previously reported AacSHC from Alicyclobacillus acidocaldarius and ZmoSHC2,
the second squalene-hopene cyclase from Z. mobilis. ZmoSHC1 showed cyclization of the non-natural
substrates homofarnesol (C16) and citronellal (C10) in addition to hopene formation from squalene
(C30). Moreover, ZmoSHC1 turned out to reveal high biocatalytic stability during long-term incubations.
Remarkably, ZmoSHC1 exhibited a shift of activity towards substrates of shorter chain lengths, displaying
over 50-fold higher conversion of homofarnesol and more than 2-fold higher conversion of citronellal in
comparison to squalene conversion.
Received 9 December 2011
Received in revised form 8 February 2012
Accepted 17 February 2012
Available online 25 February 2012
Keywords:
Squalene-hopene cyclase
Zymomonas mobilis
Homofarnesol
Citronellal
Biocatalysis
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
termination step of the cyclization reaction by an intramolecular
nucleophilic attack of the final carbocation forming a heterocyclic
The large group of triterpene cyclases has been investigated
profoundly during the last decades. In particular, the family of
squalene-hopene cyclases (SHC) has been widely used in muta-
tional and biochemical analysis elucidating the highly complex
mechanism by which these enzymes catalyze the reaction of the
natural substrate squalene 1 to its pentacyclic products hopene
2 and hopanol 3 (Scheme 1). It was the pioneering work of both
organic and biological chemists in this field, which led to a consid-
erable understanding on the structural diversity and the cyclization
process attained by a single enzyme [1–12]. Interestingly, other
related molecules, like linear terpenoids of different chain lengths
containing additional functional groups have also been shown
to serve as substrates for the well investigated SHC from the
thermophilic bacterium Alicyclobacillus acidocaldarius (AacSHC) in
several studies [13–18]. Among these, our attention was attracted
ring. It is also worth noting that ambroxan 5 is a valuable chemical
used in the flavor and fragrance industry and thus, an alternative
biochemical access based on easy available educts could be of eco-
nomic interest. Despite the fact that the conversion rates reported
for this substrate are extremely low, these results triggered the
question whether SHCs might generally have a much broader cat-
alytic potential and, thus, could represent useful biocatalysts for
challenging cyclization reactions [19–21].
Mining genome databases we now found several genes
annotated as squalene-hopene cyclases whose corresponding pro-
teins have not been characterized biochemically so far [22]. The
genome of a very interesting example, found in Zymomonas mobilis,
was completely sequenced in 2005 and recently its annotation
was revised [23,24]. This Gram-negative bacterium is considered
to be one of the most efficient hopanoid producers known to
date which encodes for two different SHC enzymes ZmoSHC1
(YP 163283.1) and ZmoSHC2 (AAF12829.1) [25–28]. One of these
(ZmoSHC2) was functionally expressed and partially characterized
in Escherichia coli in a previous publication [29]. In an accompa-
nying study it was reported that ZmoSHC1 offers the capacity to
accept the C10 monoterpenoic aldehyde citronellal 6 as substrate,
forming the cyclic product 2-isopropenyl-5-methyl-cyclohexanol
(IMC) 7 [30].
by a work of Neumann et al., in which the cyclization of the C16
-
alcohol homofarnesol 4 to the tricyclic ambroxan 5 was reported
[14].
Beside the fact that the chain length of this substrate is roughly
half the size of the “natural” C30 substrate squalene, it is also
interesting that an alcohol group is supposed to take part in the
Based on these precognitions, we decided to compare the enzy-
matic activity patterns and specificites of ZmoSHC1, ZmoSHC2 and
∗
Corresponding author. Tel.: +49 0 711 685 63193.
E-mail address: bernhard.hauer@itb.uni-stuttgart.de (B. Hauer).
1381-1177/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2012.02.007

M. Seitz et al. / Journal of Molecular Catalysis B: Enzymatic 84 (2012) 72–77
73
expression vector pET-22b(+) from Novagen by using the restric-
tion sites BamHI and NdeI (more detailed information is provided
in Supplementary data S1).
2.3. Culture conditions and protein expression
For plasmid maintenance, protein expression and biotransfor-
mations with cells, the E. coli strain BL21(DE3) was used and
100 ␮g/mL ampicillin was added to all growth media. Precul-
tures of the strains were grown overnight in 5 mL LB medium at
37 ◦C with shaking at 180 rpm (Multitron HT, Infors AG, Bottman-
ningen, Switzerland). These precultures were used to inoculate
three independent 200 mL TB cultures in 2 L Erlenmeyer flasks
(without baffles) with an optical density of OD600 = 0.05 and the
cultures were incubated as described above. When an OD600 = 0.5
was reached, SHC expression was induced by the addition of
isopropyl-ˇ-D-thiogalactopyranoside (IPTG) in a final concentra-
tion of 0.2 mM. 4 h after induction, the three cultures were mixed,
the cells were harvested by centrifugation at 13,670 × g for 20 min
at 4 ◦C (Centrifuge RC 6 Plus, Sorvall, Langenselbold, Germany),
pooled and washed twice with phosphate-buffer (100 mM pH 7).
After the final washing step, the cells were aliquoted and stored at
−20 ◦C until further usage. For controlling successful expressions
of the enzymes, aliquots were routinely analyzed by SDS-PAGE
(Supplementary data S2) using standard protocols [31].
Scheme 1. Model transformations of squalene 1 to hopene 2 and hopanol 3;
homofarnesol 4 to ambroxan 5 and (S)-citronellal 6 to 2-isopropenyl-5-methyl-
cyclohexanol (IMC) 7 for the characterization of the squalene-hopene cyclases
ZmoSHC1, ZmoSHC2 and AacSHC.
the well-studied AacSHC. In this report, we describe the expres-
sion, overproduction and partial purification of these enzymes from
E. coli and the applicability of these cyclases on the cyclization reac-
tions with the model substrates squalene 1, homofarnesol 4 and
(S)-Citronellal 6 (Scheme 1). We also studied the characterization
of the biocatalytic stability of ZmoSHC1 in long-term incubations in
two different setups.
2.4. Biotransformations with cell suspensions
For all the biotransformations, emulsions of 100 mM substrate
in solubilization buffer (50 mM citrate-buffer pH 6 containing 1%
Triton X-100, 10 mM MgCl2) were prepared. This emulsion was
mixed thoroughly prior to addition. The biotransformations were
performed in a final volume of 1 mL in a 2 mL reaction tube. All
experiments were performed as biological and technical tripli-
cates. Biotransformations with E. coli cells were performed with cell
suspensions after partial disrupture of the cells by three consecu-
tive freeze-thawing cycles. After dilution and addition of 10 mM
substrate (resulting parameters: cell density OD600 = 10, 50 mM
citrate-buffer containing 0.2% Triton X-100, 2 mM MgCl2, pH 6 for
conversions with homofarnesol and squalene or pH 4.5 for conver-
sions with (S)-citronellal) the biotransformations were carried out
for 20 h at 30 ◦C for conversions with cells expressing ZmoSHC1
and ZmoSHC2 and 60 ◦C for conversions with cells expressing
AacSHC, respectively, with shaking at 1400 rpm in an Eppendorf
Thermomixer. For biocatalytic stability assays, 900 ␮L of reaction
volume containing cells and buffers (50 mM citrate-buffer pH 6
containing 0.2% Triton X-100, 2 mM MgCl2, cell density OD600 = 80)
were mixed in 2 mL Eppendorf tubes and pre-incubated at 30 ◦C for
12, 24, 48, 72 and 96 h. After this time, 100 ␮L of substrate emulsion
was added and the mixture was further incubated for 5 h at 30 ◦C
and 180 rpm in a shaking incubator. As negative control, cells of
E. coli BL21(DE3) carrying the empty pET-22b(+) vector were used.
2. Experimental
2.1. General experimental information, materials and methods
The chemicals and solvents were purchased from
Sigma–Aldrich. Homofarnesol was provided by BASF SE.
The SHC from A. acidocaldarius was obtained from Karl
Poralla, University of Tuebingen (Germany). All products
were identified with GC–MS or GC–FID analysis by using
reference material. The systematic nomenclature for the
relevant chemical compounds are squalene [(6E,10E,14E,18E)-
2,6,10,15,19,23-hexamethyltetracosa-2,6,10,14,18,22-hexaene],
hopene [hop-22(29)-ene], hopanol [hopan-22-ol], homofar-
nesol [(3E,7E)-4,8,12-trimethyltrideca-3,7,11-trien-1-ol, >90%
E,E-homofarnesol], ambroxan [(3␣R,5␣S,9␣S,9ßR)-3␣,6,6,9␣-
tetramethyl-dodecahydronaphto[2,1-ß]furan],
(S)-citronellal
[(3S)-3,7-dimethyloct-6-enal] and IMC [2-isopropenyl-5-methyl-
cyclohexanol]. GC–MS analyses were performed with a Shimadzu
GC2010 system equipped with a GCMS-QP2010 mass-selective
detector (electron impact, 70 eV) and an AOC-5000 auto injector
using a 5% phenyl polysil–phenylene–siloxane phase column
(CS-Chromatographie, Langerwehe, Germany, 30 m, 0.25 mm,
0.25 ␮m). Helium was used as carrier gas (linear velocity 30 cm/s).
GC–FID analyses were carried out with a Shimadzu GC2010 and
an AOC-20i auto injector by using H2 as carrier gas (linear velocity
30 cm/s). Diastereomers of IMC 7 were identified by comparison
with authentic reference material via co-injection on achiral GC
(see Supplementary data S5B). Biocatalytic activity was deter-
mined by substraction of the mean of negative controls from
biotransformations.
2.5. Partial purification of the membrane-bound protein fraction
For the partial purification of the SHCs, thawed cells
(OD600 = 100) were resuspended in 20 mM citrate-buffer pH 6
containing 100 mM EDTA. The cells were disrupted by ultrasonic
treatment (Branson Sonifier 250, duty cycle 35%, output control 4)
for six times one minute on ice, with cooling on ice for at least one
minute in between the sonification steps. The cell debris was col-
lected by centrifugation at 38,000 × g for 60 min at 4 ◦C (Centrifuge
RC 6 Plus, Sorvall, Langenselbold, Germany); the supernatant was
discarded. The cell debris was resuspended in solubilization buffer
(composition see Section 2.4) and incubated for 1 h mixing gently
at 4 ◦C, in order to solubilize the membrane-bound proteins from
2.2. Molecular methods
Squalene-hopene cyclases 1 and 2 from Z. mobilis (ZmoSHC1:
YP 163283.1; and ZmoSHC2: AAF12829.1) and A. acidocaldar-
ius (ZP 03492960.1) were amplified and subcloned into the

74
M. Seitz et al. / Journal of Molecular Catalysis B: Enzymatic 84 (2012) 72–77
the membranes with the detergent (1% Triton X-100). Afterwards,
the mixture was centrifuged (conditions see above) and the super-
natant was used for the biotransformations (partially purified
enzyme).
period, 100 ␮L of substrate emulsion (10 mM) was added. The sub-
sequent reactions were performed as described in Section 2.4.
2.9. Sample preparation
2.6. Protein determination of the partially purified enzyme
After incubation, 100 ␮L of internal standard 1 (50 mM 1-
decanol in solubilization buffer, ISTD1) was added to the samples
and mixed thoroughly. Subsequently, 750 ␮L of n-heptane contain-
ing internal standard 2 (end concentration 0.5 mM 1-dodecanol,
ISTD2) was added. After mixing and centrifugation at 20,800 × g
for 10 min at room temperature (Eppendorf, Centrifuge 5417R), the
organic phases were transferred into glass GC-Vials (WICOM, WIC
42000). The extraction procedure was performed two times lead-
ing to a final volume of 1.5 mL. The analysis was performed by gas
chromatography coupled with mass spectrometry (GC–MS) or a
flame ionization detector (GC–FID). More detailed information is
provided in the Supplementary data S5.
The total protein concentration was determined using the Brad-
ford Ultra assay (expedeon). The determination was performed in
two different dilutions in triplicate using bovine serum albumin
(New England Biolabs) containing 0.1% Triton X-100 as a standard.
2.7. Densitometric determination of the cyclase
The percentage of SHC within the total protein mixture was
determined densitometrically from SDS-gels (Supplementary data
S2). The optical measurements were performed by a digital photo
station (Vilber Loumat; software Quantum ST4 15.12) and densito-
metric analysis of the gels was performed using the ImageJ software
version 1.44p (Wayne Rasband).
3. Results and discussion
3.1. Expression and partial purification of SHCs
2.8. Biotransformations with partially purified enzyme
In a first approach, an expression system for ZmoSHC1 in E.
coli and two different biotransformation setups, one based on
cell suspensions and another one using partially purified enzyme,
were successfully established. For the partial purification of the
membrane-bound ZmoSHC1, a purification protocol was used
according to previous studies [13]. By using this method, the
membrane anchored ZmoSHC1 enzyme could be solubilized and
partially purified finally exceeding 85–95% of the total protein in
this fraction (see Supplementary data S1). The same procedures
resulting comparable protein and enzyme concentrations were
performed using ZmoSHC2 and AacSHC, respectively.
The partially purified enzyme was mixed with solubilization
buffer (composition see Section 2.4) in order to use the same total
protein concentrations for comparable assays (protein concentra-
tion 1.5 mg/mL). 100 ␮L of this enzyme solution was mixed with
800 ␮L of 50 mM citrate-buffer (resulting pH 6 for conversions
with homofarnesol or squalene and pH 4.5 for conversions with
(S)-citronellal) and 100 ␮L substrate emulsion (composition see
Section 2.4). For biocatalytic stability assays, enzyme solution and
buffer were mixed in 2 mL Eppendorf tubes and pre-incubated at
30 ◦C for 12, 24, 48, 72 and 96 h. After this pre-incubation time
Fig. 1. Biotransformations with (S)-citronellal 6 ( ), homofarnesol 4 ( ), and squalene 1 ( ) to the corresponding cyclic products IMC 7, ambroxan 5 and hopene and
hopene 2 and hopanol 3 (shown as sum of peak areas) A using cells (OD600 = 10) expressing ZmoSHC1, ZmoSHC2 and AacSHC, B using 0.1 mg of the partially purified cyclases
ZmoSHC1, ZmoSHC2 and AacSHC for 20 h. Data represent the percentage of the products in relation to total peak area of substrates and products, derived from GC–MS or
GC–FID analysis. n.d.: product not detected.

M. Seitz et al. / Journal of Molecular Catalysis B: Enzymatic 84 (2012) 72–77
75
3.2. Comparison of the substrate specificity of cyclases
higher transformation of homofarnesol 4 compared to squalene 1.
In opposite to cell suspensions, partially purified enzymes exhibit
a 2-fold enhanced transformation of citronellal 6 and 47.5-fold
increase of conversion of homofarnesol 4. For more detailed infor-
mation about enzymatic conversion rates and standard deviations
see Supplementary data S3. The catalytic activity towards citronel-
lal 6 by ZmoSHC1 is generally remarkable, as to our knowledge,
this is the only triterpenoid cyclase known to date which demon-
strates the cyclization of a monoterpenoid which is not activated
by a pyrophosphate leaving group [30].
While cells expressing ZmoSHC2 showed no conversion of
homofarnesol 4 or (S)-citronellal 6, cells expressing AacSHC showed
low but significant conversion of homofarnesol 4 (3.4%) in accor-
dance with the results of Neumann et al. [14]. In contrast to AacSHC,
which has a 28-fold higher conversion rate with squalene 1 than
with homofarnesol 4, ZmoSHC1 preferentially transforms homo-
farnesol 4. This compound was converted 80 times better than
squalene 1, the alleged natural substrate of ZmoSHC1.
However, these bioconversions being performed with equal
amounts of cells expressing the different cyclases, cannot be
compared quantitatively due to different expression levels of
the enzymes. In order to evaluate these data, an additional
experimental setup with identical, carefully leveled quanti-
ties of partially purified SHCs was used. Using this protocol,
comparability of the activities of the cyclases towards the dif-
ferent substrates was generated. As shown in Fig. 1B, activity
patterns for all of the three enzymes were similar to those
using cell suspensions, namely efficient conversion of homo-
In order to compare the substrate specificities of the SHCs
ZmoSHC1, ZmoSHC2, and AacSHC, three model reactions were
selected, namely the cyclization of squalene 1 to hopene 2 or
hopanol 3, homofarnesol 4 to ambroxan 5, and (S)-citronellal 6 to
IMC 7. The cyclization of homofarnesol 4 to ambroxan 5 as well
as the conversion of citronellal 6 to IMC 7 could be economically
attractive, as both products are used in the flavour and fragrance
industry. In addition, IMC 7 from these biotransformations may also
allow an alternative access for the production of menthol in place
of traditional metal-catalyzed transformations [32].
In bioconversion reactions performed for 20 h using E. coli cells
expressing ZmoSHC1, ZmoSHC2 and AacSHC, the cyclization of
squalene 1 could be observed for all cyclases tested (Fig. 1A).
Despite the very low conversion of squalene 1 observed with cells
expressing ZmoSHC1 or ZmoSHC2 in comparison to cells expressing
AacSHC (96.9%), these results demonstrated functional expression
of all enzymes. In bioconversions with ZmoSHC1 as biocatalyst, pre-
dominant hopene formation was observed. In contrast to ZmoSHC1,
ZmoSHC2 and AacSHC converted squalene 1 to the cyclization prod-
ucts hopene 2 and hopanol 3 with ratios of approximately 7:1 for
ZmoSHC2 and 6:1 for AacSHC, respectively.
However, the difference in activity of the enzymes towards a
given substrate was surprising. We found efficient conversion of
homofarnesol 4 to ambroxan 5 (22.9%) and low conversion of (S)-
citronellal 6 to IMC 7 (0.8%) only with cells expressing ZmoSHC1.
Control experiments (E. coli cells harboring the empty vector in the
presence of substrate, as well as substrate only) showed that auto-
cyclization of (S)-citronellal 6 to the corresponding cyclic product
diastereomers of IMC 7, isopulegol, neo-isopulegol, iso-isopulegol
and neoiso-isopulegol, occured. However, the selectivity and activ-
ity obtained with ZmoSHC1 was greatly increased in comparison
to control experiments performed. Bioconversions with ZmoSHC1
demonstrated the predominant formation of one isomer of IMC 7,
iso-isopulegol, after 20 h of incubation (see Supplementary data
S5B for further description of the stereoisomers). A direct com-
parison of the conversions of squalene 1, homofarnesol 4 and
citronellal 6 reveals, that biotransformations with cells of ZmoSHC1
show a 2.6-fold increased conversion of citronellal 6 and 76.3-fold
farnesol
4 (9.5%) and significant activity with citronellal 6
in cells expressing ZmoSHC1 (0.4%) and a 40–400-fold decreased
activity of ZmoSHC2 (1.8%) and ZmoSHC1 (0.2%) towards squalene
1 when compared to AacSHC (78.9%). The only significant differ-
ence to the results from the experiments with cell suspensions was
the missing activity of the partially purified AacSHC protein solu-
tion towards homofarnesol 4 under the given reaction conditions.
However, this observation can be explained by the lower concen-
tration of cyclase in this setup, and thus, the formation of ambroxan
5 by AacSHC was most likely below the detection limit.
In order to test whether the substrate conversion rates with
ZmoSHC1 could be increased, the density of cells was increased
Fig. 2. Genomic regions surrounding ZmoSHC1 (A) and ZmoSHC2 (B) ( ); genes associated to the hopene metabolism marked in gray ( ); genes not associated to the
hopene metabolism marked in white ( ) [23,24,26,28].

76
M. Seitz et al. / Journal of Molecular Catalysis B: Enzymatic 84 (2012) 72–77
Fig. 3. Stability assay: enzymatic activity of ZmoSHC1 towards homofarnesol 4 in relation to the pre-incubation time A using cells (OD600 = 80) expressing ZmoSHC1, B using
0.1 mg of partially purified ZmoSHC1; conversions after 5 h at 30◦. Data represent the percentage of the products in relation to total peak area of substrates and products,
derived from GC–FID analysis.
from OD600 = 10 to OD600 = 80 in the whole cell bioconversion setup.
Under these conditions, homofarnesol 4 and citronellal 6 con-
versions could be improved significantly. Hence, the conversion
rates of homofarnesol 4 and citronellal 6 reached 50.3% and 16.4%,
respectively (data not shown).
biocatalyst in different experimental setups. The biocatalytic stabil-
ity of ZmoSHC1 in long-term incubations (up to 96 h) with dense cell
suspensions and partially purified enzyme was investigated. Sur-
prisingly, without any pre-incubation, lower conversion rates were
observed in both setups. We assume that these lower conversion
rates result from the poor formation of emulsions within the first
hours of the stability assays. In the experiment with dense cell sus-
pensions, the highest activity of the enzyme towards homofarnesol
4 was detected after 12 h of pre-incubation. With increasing pre-
incubation time, a slight decrease of the enzymatic activity could be
observed, finally resulting in 30.4% conversion of homofarnesol 4
(Fig. 3A) after 96 h. For details on the conversion rates of the stability
assay see Supplementary data S4.
The observed activity pattern of ZmoSHC1 is surprising, as squa-
lene 1 is considered to be the “natural” substrate of this enzyme
annotated as SHC. However, our results indicate, that despite the
fact that squalene 1 can be converted, the naturally occurring sub-
strate for ZmoSHC1 within the cell might be different. In contrast
to ZmoSHC1, the activity pattern of ZmoSHC2 suggests that this
enzyme is indeed a true squalene-hopene cyclase. This hypothesis
is further supported by the genomic context surrounding ZmoSHC2,
as it is situated in close proximity to other genes involved in
the biosynthesis of hopanoids (Fig. 2B) including two subunits
of the squalene synthase, a squalene associated FAD-dependent
desaturase, and a hopanoid associated phosphorylase. The genomic
context of ZmoSHC1 on the other hand has no similarity to genes
involved in the biosynthesis of hopanoids (Fig. 2A). This, together
with the observed shift in its substrate preference towards linear
isoprenoids of shorter chain lengths, indicates that in vivo ZmoSHC1
is acting as an isoprenoid cyclase but is most likely not involved in
the biosynthesis of squalene 1. However, the true role of ZmoSHC1
in Z. mobilis needs to be determined by further investigations
[23,24,26,28].
Pairwise alignment of the corresponding amino acid sequences
revealed a global sequence identity of 33.6% between ZmoSHC1
and ZmoSHC2 and the existence of a 26 residues long fragment in
the C-terminal end of ZmoSHC1, which was missing in ZmoSHC2.
Furthermore, the alignment has highlighted that both AacSHC and
ZmoSHC2 lack 52 and 45 amino acids, respectively, at the N-
terminus compared to ZmoSHC1 (see Supplementary data S6A–D).
A further analysis of the family distribution using the recently
established TTCED database revealed a distant phylogenetic rela-
tionship of the two enzymes, assigning them into two different fam-
ilies, namely the SHC 1 family for ZmoSHC1 and the SHC 9 family
for ZmoSHC2 [22]. In comparison to this, AacSHC, classified in family
SHC 1, reveals global sequence identities of 40.6% with ZmoSHC1
and 37.2% with ZmoSHC2. Thus, the low identity and phylogenetic
relation of the two enzymes ZmoSHC1 and ZmoSHC2 might provide
a further explanation for their different biochemical properties.
In contrast to this, the catalytic activity of the membrane-bound
protein fraction of cells expressing ZmoSHC1 remained stable dur-
ing the full time of the experiment with activities ranging from 7.9%
to 8.3% (Fig. 3B), demonstrating a high stability of the catalyst under
two different reaction conditions.
4. Conclusions
Summarizing our results, we cloned, heterologously expressed
and partially purified the novel triterpene cyclase ZmoSHC1 from
Z. mobilis showing new catalytic properties. We further established
two biotransformation setups and characterized this enzyme. In
these experiments, unique substrate activity patterns of ZmoSHC1
compared to ZmoSHC2 and AacSHC were observed. Furthermore,
increased enzymatic activity towards the non-natural substrates
homofarnesol and citronellal has been demonstrated. From our
study, it can be concluded that the squalene-hopene cyclase
ZmoSHC1 from Z. mobilis represents a very interesting novel
enzyme uncovering a shift of substrate acceptance to shorter chain
linear isoprenoids containing hetero atoms. Therefore, squalene-
hopene cyclases seem to be a very promising family of proteins in
order to identify novel enzymatic activities for the biosynthesis of
complex natural as well as non-natural compounds.
Acknowledgments
This work was financially supported by Bundesministerium für
Bildung und Forschung (BMBF). We thank B. Nestl, C. Geinitz,
P.-O. Syrén, S. Hammer, and S. Racolta for inspiring discussions and
critical revision of the manuscript.
3.3. Biocatalytic stability of ZmoSHC1
Among the most important parameters for the use of enzymes
are the conversion rate of the substrate and the stability of the

M. Seitz et al. / Journal of Molecular Catalysis B: Enzymatic 84 (2012) 72–77
77
Appendix A. Supplementary data
[16] H. Tanaka, H. Noguchi, I. Abe, Org. Lett. 7 (2005) 5873–5876.
[17] H. Tanaka, H. Noma, H. Noguchi, I. Abe, Tetrahedron Lett. 47 (2006) 3085–3089.
[18] H. Tanaka, H. Noguchi, I. Abe, Tetrahedron Lett. 45 (2004) 3093–3096.
[19] Y. Yonemura, T. Ohyama, T. Hoshino, Org. Biomol. Chem. 10 (2012) 440–446.
[20] H. Atsuko, I. Kazuaki, Kao Corp, patent: The method for producing (-)-3a,6,6,9a-
tetramethyldecahydronaphtho[2,1-b]furan includes making SHC act on the
homofarnesol, in a solvent at a pH of 5. 2-6.6, Classifications: International
(2006.01): C12P 17/18; C12N 15/09, Application Number: JP2007000228877.
[21] T. Sato, H. Hoshino, S. Yoshida, M. Nakajima, T. Hoshino, J. Am. Chem. Soc. 133
(2011) 17540–17543.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molcatb.2012.02.007.
References
[1] K.U. Wendt, G.E. Schulz, Structure 6 (1998) 127–133.
[2] D.W. Christianson, Chem. Rev. 106 (2006) 3412–3442.
[3] I. Abe, M. Rohmer, G.D. Prestwich, Chem. Rev. 93 (1993) 2189–2206.
[4] T. Sato, M. Kouda, T. Hoshino, Biosci. Biotechnol. Biochem. 68 (2004) 728–738.
[5] K.U. Wendt, G.E. Schulz, E.J. Corey, D.R. Liu, Angew. Chem. Int. Ed. 39 (2000)
2812–2833.
[6] G. Siedenburg, D. Jendrossek, Appl. Environ. Microbiol. 77 (2011) 3905–3915.
[7] T. Hoshino, T. Sato, Chem. Commun. (2002) 291–301.
[8] J.C. Medina, K.S. Kyler, J. Am. Chem. Soc. 110 (1988) 4818–4821.
[9] E.J. Corey, S.K. Gross, J. Am. Chem. Soc. 89 (1967) 4561–4562.
[10] E.J. Corey, S.C. Virgil, S. Sarshar, J. Am. Chem. Soc. 113 (1991) 8171–8172.
[11] E.J. Corey, H. Cheng, Tetrahedron Lett. 37 (1996) 2709–2712.
[12] E.J. Corey, S.C. Virgil, H. Cheng, C.H. Baker, S.P.T. Matsuda, V. Singh, S. Sarshar,
J. Am. Chem. Soc. 117 (1995) 11819–11820.
[22] http://www.ttced.uni-stuttgart.de.
[23] J.S. Seo, H. Chong, H.S. Park, K.O. Yoon, C. Jung, J.J. Kim, J.H. Hong, H. Kim, J.H.
Kim, J.I. Kil, C.J. Park, H.M. Oh, J.S. Lee, S.J. Jin, H.W. Um, H.J. Lee, S.J. Oh, J.Y. Kim,
H.L. Kang, S.Y. Lee, K.J. Lee, H.S. Kang, Nat. Biotechnol. 23 (2005) 63–68.
[24] S. Yang, K.M. Pappas, L.J. Hauser, M.L. Land, G.L. Chen, G.B. Hurst, C. Pan, V.N.
Kouvelis, M.A. Typas, D.A. Pelletier, D.M. Klingeman, Y.J. Chang, N.F. Samatova,
S.D. Brown, Nat. Biotechnol. 27 (2009) 893–894.
[25] B. Seckler, K. Poralla, Biochim. Biophys. Acta 881 (1986) 356–363.
[26] M.A.F. Hermans, B. Neuss, H. Sahm, J. Bacteriol. 173 (1991) 5592–5595.
[27] E. Douka, A.-I. Koukkou, C. Drainas, C. Grosdemange-Billiard, M. Rohmer, FEMS
Microbiol. Lett. 199 (2001) 247–251.
[28] S. Bringer, T. Härtner, K. Poralla, H. Sahm, Arch. Microbiol. 140 (1985) 312–316.
[29] I.G. Reipen, K. Poralla, H. Sahm, G.A. Sprenger, Microbiology 141 (1995)
155–161.
[30] G. Siedenburg, D. Jendrossek, M. Breuer, B. Juhl, J. Pleiss, M. Seitz, J. Klebens-
berger, B. Hauer, Appl. Environ. Microbiol. 78 (2012) 1055–1062.
[31] U.K. Laemmli, Nature 227 (1970) 680–685.
[13] D. Ochs, C. Kaletta, K.-D. Entian, A. Beck-Sickinger, K. Poralla, J. Bacteriol. 174
(1992) 298–302.
[14] S. Neumann, H. Simon, Biol. Chem. Hoppe-Seyler 367 (1986) 723–729.
[15] T. Hoshino, Y. Kumai, I. Kudo, S. Nakano, S. Ohashi, Org. Biomol. Chem. 2 (2004)
2650–2657.
[32] N. Ravasio, N. Poli, R. Psaro, M. Saba, F. Zaccheria, Top. Catal. 13 (2000) 195–199.