Transhalogenation Catalysed by Haloalkane Dehalogenases Engineered to Stop Natural Pathway at Intermediate

Article abstract of DOI:10.1002/adsc.201900132

Haloalkane dehalogenases (HLDs) are α/β-hydrolases that convert halogenated compounds to their corresponding alcohols. The overall kinetic mechanism proceeds via four steps: (i) binding of halogenated substrate, (ii) bimolecular nucleophilic substitution (S_N2) leading to the cleavage of a carbon-halogen bond and the formation of an alkyl-enzyme intermediate, (iii) nucleophilic addition of a water molecule resulting in the hydrolysis of the intermediate to the corresponding alcohol and (iv) release of the reaction products – an alcohol, a halide ion and a proton. Although, the overall reaction has been reported as irreversible, several kinetic evidences from previous studies suggest the reversibility of the first S_N2 chemical step. To study this phenomenon, we have engineered HLDs to stop the catalytic cycle at the stage of the alkyl-enzyme intermediate. The ability of the intermediate to exchange halides was confirmed by a stopped-flow fluorescence binding analysis. Finally, the transhalogenation reaction was confirmed with several HLDs and 2,3-dichloropropene in the presence of a high concentration of iodide. The formation of the transhalogenation product 3-iodo-2-chloropropene catalysed by five mutant HLDs was identified by gas chromatography coupled with mass spectrometry. Hereby we demonstrated the reversibility of the cleavage of the carbon-halogen bond by HLDs resulting in a transhalogenation. After optimization, the transhalogenation reaction can possibly find its use in biocatalytic applications. Enabling this reaction by strategically engineering the enzyme to stop at an intermediate in the catalytic cycle that is synthetically more useful than the product of the natural pathway is a novel concept. (Figure presented.).

Full text of DOI:10.1002/adsc.201900132

Title: Transhalogenation Catalysed by Haloalkane Dehalogenases
Engineered to Stop Natural Pathway at Intermediate
Authors: Andy Beier, Jiri Damborsky, and Zbynek Prokop
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Adv. Synth. Catal. 10.1002/adsc.201900132
Link to VoR: http://dx.doi.org/10.1002/adsc.201900132

10.1002/adsc.201900132
Advanced Synthesis & Catalysis
COMMUNICATION
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Transhalogenation Catalysed by Haloalkane Dehalogenases
Engineered to Stop Natural Pathway at Intermediate
Andy Beiera,b, Jiri Damborskya,b and Zbynek Prokopa,b*
a
Loschmidt Laboratories, Department of Experimental Biology and RECETOX, Faculty of Science, Masaryk
University, Kamenice 5A, 625 00 Brno, Czech Republic
International Clinical Research Center, St. Anne’s University Hospital, Pekarska 53, 656 91 Brno, Czech Republic
b
*
E-mail: zbynek@chemi.muni.cz. Tel: +420-5-4949-6667. FAX: +420-5-4949 6302.
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.
and pharmaceutical applications. However, to
achieve a good rational design, a detailed
understanding of the structural basis related to
enzyme function is critical. Model enzymes are of
great value for the investigation of molecular
mechanisms of enzyme catalysis as well as the
computational analysis of chemical concepts. One
class of enzymes, which is recognized as an
interesting model system, are haloalkane
Abstract. Haloalkane dehalogenases (HLDs) are α/β-
hydrolases that convert halogenated compounds to their
corresponding alcohols. The overall kinetic mechanism
proceeds via four steps: (i) binding of halogenated
substrate, (ii) bimolecular nucleophilic substitution (SN2)
leading to the cleavage of a carbon-halogen bond and the
formation
of
an
alkyl-enzyme
intermediate,
(iii) nucleophilic addition of a water molecule resulting in
the hydrolysis of the intermediate to the corresponding
alcohol and (iv) release of the reaction products - an
alcohol, a halide ion and a proton. Although, the overall
reaction has been reported as irreversible, several kinetic
evidences from previous studies suggest the reversibility of
the first SN2 chemical step. To study this phenomenon, we
have engineered HLDs to stop the catalytic cycle at the
stage of the alkyl-enzyme intermediate. The ability of the
intermediate to exchange halides was confirmed by a
stopped-flow fluorescence binding analysis. Finally, the
transhalogenation reaction was confirmed with several
HLDs with 2,3-dichloropropene in the presence of a high
concentration of iodide. The formation of the
dehalogenases (HLDs; EC 3.8.1.5).[1]
HLDs
catalyse the cleavage of a carbon-halogen bond of
organohalogen compounds by nucleophilic
substitution, followed by the hydrolysis of the
transhalogenation
product
3-iodo-2-chloropropene
catalysed by five mutant HLDs was identified by gas
chromatography coupled with mass spectrometry. Hereby
we demonstrated the reversibility of the cleavage of the
carbon-halogen bond by HLDs resulting in
a
transhalogenation. After optimization, the transhalogenation
reaction can possibly find its use in biocatalytic
applications. Enabling this reaction by strategically
engineering the enzyme to stop at an intermediate in the
catalytic cycle that is synthetically more useful than the
product of the natural pathway is a novel concept.
alkyl-enzyme
intermediates
producing
the
corresponding alcohols (Scheme 1A).
Scheme 1. Catalytic mechanisms of HLDs and their
engineered variants. (A) The reaction mechanism of
HLDs (adapted from Verschueren et al.).[2] (B) The
Keywords: enzyme catalysis; halide exchange; haloalkane
dehalogenase; nucleophilic substitution; redirection of
reaction; transhalogenation
-
-
kinetic mechanism of transhalogenation ([X ] >> [X ])
2
1
by engineered HLDs. Substitution of the catalytic
histidine prevents the hydrolytic step (red cross).
Halogenated substrate R-X, enzyme-substrate complex
E.R-X, complex of alkyl-enzyme intermediate with
bound halide ion E-R.X, enzyme-product complex E.R-
The field of protein engineering benefits from
directed evolution when the structural information
about the target enzyme is limited and the
structure-function relationship is not sufficiently
understood. A recent trend tends to prioritize a
semi-rational design to minimize the library size
and the screening effort. This greatly accelerates
the discovery of novel biocatalysts for biomedical
-
OH and halide ion X (X represents Cl, Br or I).
The kinetic mechanism of three representative
HLDs has been studied in detail, DhlA from
Xanthobacter autotrophicus GJ10,[3] DhaA from
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900132
and 1-chlorobutane (Supplementary Figure S3).
The equilibrium dissociation constants for the
specific binding of iodide (Kd = 1/Ka) calculated
from the fluorescence analysis were 2 ± 1, 200 ±
Advanced Synthesis & Catalysis
Rhodococcus rhodochrous NCIMB 13064[4] and
LinB from Sphingobium japonicum UT26.[5] For
LinB, kinetic signs for the reversibility of the first
chemical step (SN2) could be observed. To study
this phenomenon, we have investigated the
formation of the product of the reversed reaction by
performing the conversion of halogenated
substrates in the presence of a high concentration of
different halides. This design is based on the ability
of the alkyl-enzyme intermediate to exchange the
halide anion released from the substrate by a
different ion, which is added to the reaction mixture
in a large excess. In the case that the alkyl-enzyme
intermediate undergoes the reversed reaction, the
corresponding transhalogenated compound should
be detectable in the reaction mixture (Scheme 1B).
100 and 40 ± 20 mM for free DbjA H280F, and
alkyl-enzyme intermediate after the reaction with
The
wild
type
HLDs
DbeA
from
1,2-dibromoethane
respectively.
and
1-chlorobutane,
Bradyrhizobium elkani USDA94[6] and DbjA from
Bradyrhizobium japonicum USDA110[7] were
selected for the reversibility tests since the large
opening of their active sites makes the exchange of
halides for the enzyme in the state of the alkyl-
Figure 1. Halide binding to the engineered enzyme
DbjA H280F. (A) Dependence of amplitude of rapid
equilibrium fluorescence quenching recorded upon
mixing of DbjA H280F with chloride (open circle),
bromide (open triangle) and iodide (black square). (B)
The apparent association equilibrium constants of the
non-specific quenching (KQ) and the specific binding
(Ka) of halides were calculated by using extended Stern-
Volmer equation.
enzyme
intermediate
most
likely.
Gas
chromatography coupled with mass spectrometry
(GC-MS) was used to monitor the conversion of 1-
bromobutane with DbeA and DbjA in the presence
of a large excess (1 M) of chloride ions. Both wild
type enzymes catalysed the formation of the
corresponding
alcohol,
however,
the
After we confirmed the specific binding of the
iodide to DbjA H280F in both free and
intermediate states, the transhalogenation of 1-
transhalogenation product 1-chlorobutane was not
detected (Supplementary Figure S1). This result
is consistent with a previous kinetic analysis
indicating that hydrolysis is significantly faster in
comparison to the possible halide release from the
alkyl-enzyme intermediate or the rate of the
reversed reaction.[5]
chlorobutane,
1,2-dibromoethane,
bis(2-
chlorethyl)ether and 2,3-dichloropropene to their
corresponding iodinated analogues was tested
(Figure 2, Supplementary Table S1).
To maximize the chance of the reversed reaction
at the stage of the alkyl-enzyme intermediate, we
have constructed the variant DbjA H280F which
lacks the catalytic histidine. In such an HLD
mutant,[8] the hydrolytic step is prevented and the
alkyl-enzyme intermediate is trapped, which
provides enough time for a halide exchange and a
subsequent transhalogenation (Scheme 1B). Since
the reaction of this mutant with 1-bromobutane in
the presence of chloride still did not produce the
transhalogenation product, the halide exchange
appeared to be the limiting step.
The binding of chloride, bromide and iodide was
studied in detail with DbjA H280F by using the
stopped-flow fluorescence analysis. All the
obtained fluorescence values showed that the
equilibrium was reached rapidly, confirming that
there is no kinetic limitation for the halide
interaction with the enzyme (Supplementary
Figure S2). Both chloride and bromide showed a
nonspecific fluorescence quenching at increasing
concentrations, reflecting the role of these anions as
general fluorescence quenchers. The fluorescence
profile with iodide showed a strong additional
quenching activity, suggesting a specific binding to
DbjA_H280F (Figure 1).
Figure 2. Transhalogenation reaction catalysed by
engineered HLDs. (A)
Formation of the
transhalogenation product 3-iodo-2-chloropropene from
2,3-dichloropropene and iodide catalysed by the variant
DbjA H280F. (■) with DbjA H280F (□) without DbjA
H280F. (B) MS spectrum of the product of the
transhalogenation reaction (black) and 3-iodo-2-
chloropropene standard (red). The match of both MS
spectra is 98%. (C) Specific activities of different HLDs
with mutation in the catalytic histidine for the
conversion of 2,3-dichloropropene to 3-iodo-2-
chloropropene.
The specific binding of iodides was confirmed
also for the alkyl-enzyme intermediates after the
reaction of DbjA H280F with 1,2-dibromoethane
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900132
Advanced Synthesis & Catalysis
ampicillin (1 L) was inoculated with 5 mL of a freshly
grown preculture. The bacteria were grown at 37 °C,
until the optical density measured at 600 nm reached
approximately 0.6. Then the cultures were cooled to
20 °C, 0.5 mM IPTG was added, and growth continued
at 20 °C. After overnight growth, cells were harvested
using centrifugation for 10 min at 8000 rpm. A washing
step was carried out using approximately 125 mL of
washing buffer. Cells were harvested by centrifugation
(10 min at 8000 rpm). Cell pellets were resuspended in
40 mL of purification buffer A and then frozen at −80
°C. Cell disruption was achieved by sonication (4 × 8
min, 0.3 cycle, 85 amplitude). The crude extract was
centrifuged (20−60 min at 14000 rpm) to remove the cell
debris. The cell-free extracts were filtered through a
Rotilabo-Spritzenfilter (0.45 μm) to remove remaining
cell debris. Samples were purified using metal affinity
chromatography with a small adaptation of stepwise
increased imidazole concentration for washing (10 mL
of 4%, 10 mL of 6%, and 15 mL of 10%). Finally,
imidazole was removed by a double-dialysis step in 50
mM phosphate buffer, pH 7.5. Protein samples were
filtered through a sterile 0.22 μm Rotilabo-Spritzenfilter
for storage. The concentration of the prepared enzyme
solutions was determined in triplicate using Bradford
reagent, and the purity was checked by SDS-PAGE.
The formation of the iodinated analogue was
observed only in the reaction of DbjA H280F with
2,3-dichloropropene (Figure 2A). To identify the
transhalogenated product, we applied GC-MS
(Figure 2B) followed by a fragmentation analysis
and an analysis of isotopic clusters with the
assistance of the Mass Frontier 1.0 software
(HighChem, Slovak Republic). The heaviest ion at
the mass-to-charge ratio (m/z) 202 is corresponding
to the molecular weight of 3-iodo-2-chloropropene
(C3H4ICl). The isotopic cluster at m/z 202-204 with
an abundance of 100/31 (M/(M+2)) matches the
isotopic ratio 100/32 predicted by Mass Frontier for
the presence of one chlorine and one iodine atom in
the molecule. The m/z 127 represents iodine with
the characteristic monoisotopic abundance. The
m/z 75 (C3H4Cl) represents the rest of the molecule
after the cleavage of iodine. The m/z 49 represents
the fragment CH2Cl produced from m/z 75 by the
cleavage of C2H2. Both clusters m/z 75-77 and m/z
49-51 showed an isotopic abundance characteristic
for the presence of one chlorine in the molecule.
Additionally, the MS spectrum of the
transhalogenated product was compared to an MS
spectrum obtained for a commercially synthesized
chemical standard of 3-iodo-2-chloropropene with
98% match (CF Plus Chemicals, Czech Republic).
Finally, the conversion of 2,3-dichloropropene to
3-iodo-2-chloropropene was tested with a set of
selected HLDs carrying the mutation in the
catalytic histidine (DbjA H280F, LinB H272F,
DhlA H289F, DhaA H272F and DhaA31
H272F).[9] All the tested dehalogenases carrying the
mutation in the catalytic histidine were able to
Conversion of 1-bromobutane by DbeA_wt,
DbjA_wt and DbjA_H280F in presence of chloride.
Dehalogenation reactions were performed at 37°C in
25 ml Reacti Flasks closed by Mininert Valves. The
reaction mixtures were prepared by mixing of 15 ml of
100 mM glycine buffer containing 0.8 M NaCl (pH 8.6)
and 1-bromobutane as a substrate to a final concentration
of 9.3 mM. The reactions were initiated by addition of
300 μl enzyme in a concentration of 0.78 mg/ml
(DbeA_wt), 0.69 mg/ml (DbjA), or 1.10 mg/ml (DbjA
H280F). Buffer (300 μl) instead of enzyme was added to
the reaction mixture as a control. The reactions were
monitored by taking 1 ml samples at 5, 10, 15, 20, 25,
30, 40, 50 and 60 minutes from the reaction mixtures.
The reaction was stopped by addition of 0.5 ml
diethylether containing 1,2-dichloroethane as an internal
standard to the reaction mixture. After extraction,
samples were analysed using a gas chromatograph
(Agilent 7890, USA) equipped with the capillary column
DB-FFAP (30m x 0.25mm x 0.25μm, Phenomenex) and
connected with a mass spectrometer (Agilent 5975C,
USA). The 5 μl of sample were injected into a Split-
Splitless inlet at 250 °C, with split ratio of 1:20. The
temperature programme was isothermal at 40 °C for
1 min, followed by an increase to 140 °C at 20 °C/min.
The flow of carrier gas (He) was 1.2 ml/min. The MS
was operated at SCAN mode (30 to 240 amu). The
temperatures of the ion source and GC-MS interface
were 230 °C and 250° C, respectively.
catalyse
the
transhalogenation
of
2,3-
dichloropropene in the presence of iodides, even
though the overall reactions were slow. The most
efficient formation of 3-iodo-2-chloropropene was
observed with the engineered variants of DbjA and
DhaA, followed by LinB and DhaA31. The lowest
activity was shown with the DhlA variant (Figure
2C), very likely due to its closed active site cavity.
This view is supported by the difference in the rates
of DhaA and DhaA31, the latter mutant possessing
a more occluded active site cavity.[10]
Stopped-flow fluorescence analysis. The binding
of halides to the free DbjA H280F or its enzyme-alkyl
intermediates was performed with the stopped-flow
instrument SFM-300 (BioLogic, France) combined with
MPS-70. Fluorescence emission from tryptophan
residues was observed through a 320 nm cut-off filter
upon excitation at 295 nm. The reactions were
performed at 37°C in a glycine buffer pH 8.6. The
enzyme-alkyl intermediate was prepared by mixing the
enzyme with a solution of 1 mM 1-chlorbutane or 5 mM
1,2-dibromoethane in 50 mM glycine buffer pH 8.6 and
a subsequent incubation at room temperature for 10
minutes. Substrate that was not converted by the enzyme
was removed by evaporation. Final concentration of the
enzyme in the solution was 0.25 mg/ml. The chloride
and bromide data were fitted to Stern-Volmer quenching
equation:
In summary, the collected data provides a
direct experimental evidence for the reversibility of
the first chemical step (SN2) of the catalytic
mechanism of HLDs. These observations could
pave the way for a potential new chemistry and
reactions to synthetically more useful products
catalysed by HLDs by strategically stopping the
mechanistic cycle at the alkyl-enzyme intermediate.
However, more research and further optimization
would be needed to make the enzymatic
transhalogenation more efficient.
Experimental Section
1
퐹
Construction of DbjA H280F and DhlA H289F,
protein expression and purification. The recombinant
genes dbjA H280F and dhlA H289F were obtained by
site-directed mutagenisis. Expression of the HLDs was
achieved using E. coli BL21(DE3) containing the
constructed expression vectors. LB medium with
=
⁄
1 + 퐾푄. [푋−]
퐹0
where, F/F0 is the relative fluorescence, and KQ is the
quenching constant (which is the apparent association
equilibrium constant of the non-specific quenching
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900132
Advanced Synthesis & Catalysis
-
interaction between halide (X ) and enzyme. For the
iodide the extended Stern-Volmer was used:
[2] K. H. Verschueren, F. Seljée, H. J. Rozeboom, K. H.
Kalk, B. W. Dijkstra, Nature 1993, 363, 693-698.
1
1 + 푓. 퐾푎. [푋−]
[3] J. P. Schanstra, J. Kingma, D. B. Janssen, J. Biol.
Chem. 1996, 271, 14747-14753.
퐹
=
.
⁄
1 + 퐾푄. [푋−] 1 + 퐾푎. [푋−]
퐹0
[4] T. Bosma, M. G. Pikkemaat, J. Kingma, J. Dijk, D.
where, f is the fluorescence level of the halide-bound
enzyme complex, and Ka is the association equilibrium
constant of the specific binding of the halide to the
active site. Data fitting was performed by using the
software Origin 6.1 (OriginLab, USA).
B. Janssen, Biochemistry 2003, 42, 8047-8053.
[5] Z. Prokop, M. Monincova, R. Chaloupkova, M.
Klvana, Y. Nagata, D. B. Janssen, J. Damborsky, J.
Biol. Chem. 2003, 278, 45094-45100.
Reaction of selected substrates with engineered
HLD variants in the presence of iodide.
Transhalogenation reactions were performed at 37 °C in
25 ml Reacti Flasks closed by Mininert Valves. The
reaction mixtures were prepared by mixing 5 ml of 100
mM glycine buffer containing 200 mM KI (pH 8.6) and
3-12 mM substrate. The reactions were initiated by
addition of 5 ml enzyme in a concentration of 5 mg/ml.
Buffer instead of enzyme was added for the controls.
The reactions were monitored by taking 0.8 ml samples
from the reaction mixtures. The reaction was stopped by
gentle vortexing of the reaction mixture with 1.6 ml
diethylether containing 1,2-dichloroethane as an internal
standard. After extraction, the samples were analysed
using a gas chromatograph (Agilent 7890, USA)
equipped with the capillary column ZB-5 (30m x
0.25mm x 0.25μm, Phenomenex) and connected with a
mass spectrometer (Agilent 5975C, USA). For analysis,
1 μl of the sample was injected into a Split-Splitless inlet
at 250 °C, with a split ratio of 1:20. The temperature
programme was isothermal at 40 °C for 1 min, followed
by an increase to 250 °C at 20 °C/min. The flow of
carrier gas (He) was 1.2 ml/min. The MS was operated
at SCAN mode (15 to 250 amu). The temperature of the
ion source and GC-MS interface was 230 °C and 250 °C,
respectively.
[6] T. Prudnikova, T. Mozga, P. Rezacova, R.
Chaloupkova, Y. Sato, Y. Nagata, J. Brynda, M.
Kuty, J. Damborsky, I. Kuta Smatanova, Acta
Crystallogr., Sect. F: Struct. Biol. Cryst. Commun.
2009, 65, 353-356.
[7] Y. Sato, R. Natsume, M. Tsuda, J. Damborsky, Y.
Nagata, T. Senda, Acta Crystallogr., Sect. F: Struct.
Biol. Cryst. Commun. 2007, 63, 294-296.
[8] R. J. Pieters, M. Fennema, R. M. Kellogg, D. B.
Janssen, Bioorg. Med. Chem. Lett. 1999, 9, 161-166.
[9] a) A. Jesenská, J. Sꢀkora, A. Olzynska, J. Brezovskꢀ,
Z. Zdráhal, J. Damborskꢀ, M. Hof, J. Am. Chem.
Soc. 2008, 131, 494-501; b) S. Kaushik, Z. Prokop, J.
Damborsky, R. Chaloupkova, The FEBS journal
2017, 284, 134-148.
[10] M. Pavlova, M. Klvana, Z. Prokop, R.
Chaloupkova, P. Banas, M. Otyepka, R. C. Wade, M.
Tsuda, Y. Nagata, J. Damborsky, Nat. Chem. Biol.
2009, 5, 727-733.
Acknowledgements
We want to thank Professor Dick B. Janssen (Groningen
University, The Netherlands) for the fruitful discussion related
to the mechanism and kinetics of the HLD reaction, which led
to the idea of testing the reversibility of the SN2 chemical step
by halide exchange. We also thank our technician Hana
Moskalikova (Masaryk University, Czech Republic) for her
significant contribution to this study. Moreover, we would like
to express our gratitude to the Ministry of Education (LQ1605,
LM2015047,
LM20150550,
LM2015051CZ.02.1.01/0.0/0.0/16_013/0001761) and to the
European Union (SinFonia 814418) for the financial support.
References
[1] a) S. Nevolova, E. Manaskova, S. Mazurenko, J.
Damborsky, Z. Prokop, Biotechnol. J. 2018, 14,
e1800144; b) K. Beerens, S. Mazurenko, A. Kunka,
S. M. Marques, N. Hansen, M. Musil, R.
Chaloupkova, J. Waterman, J. Brezovsky, D. Bednar,
Z. Prokop, J. Damborsky, ACS Catal. 2018, 8, 9420-
9428; c) S. M. Marques, D. Bednar, J. Damborsky,
Front. Chem. 2019, 6, 650; d) M. Musil, H.
Konegger, J. Hon, D. Bednar, J. Damborsky, ACS
Catal. 2019, 9, 1033-1054; e) G. Jindal, K. Slanska,
V. Kolev, J. Damborsky, Z. Prokop, A. Warshel,
Proc. Natl. Acad. Sci. U. S. A. 2019, 116, 389-394; f)
P. Kokkonen, D. Bednar, V. Dockalova, Z. Prokop,
J. Damborsky, J. Biol. Chem. 2018, 29, 11505-
11512.
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900132
Advanced Synthesis & Catalysis
COMMUNICATION
Transhalogenation Catalysed by Haloalkane
Dehalogenases Engineered to Stop Natural
Pathway at Intermediate
TRANSHALOGENATION
Cl
Cl
Adv. Synth. Catal. 2019, Volume, Page – Page
-
-
Cl
Cl
Cl
OH
I
Andy Beier, Jiri Damborsky, Zbynek Prokop*
O
O
Cl
I
Engineered Haloalkane Dehalogenase (45 U/mg)
5
This article is protected by copyright. All rights reserved.