Functional assembly of a multi-enzyme methanol scaffold for enhanced NADH production†

ChemComm
COMMUNICATION
Cite this: Chem. Commun., 2013,
49, 3766
Received 18th January 2013,
Accepted 20th March 2013
Functional assembly of a multi-enzyme methanol
oxidation cascade on a surface-displayed trifunctional
scaffold for enhanced NADH production†
Fang Liu,a Scott Bantab and Wilfred Chen*a
DOI: 10.1039/c3cc40454d
www.rsc.org/chemcomm
We report a simple and low-cost strategy that allows the sequential
and site-specific assembly of a dehydrogenase-based multi-enzyme
cascade for methanol oxidation on the yeast surface using the highaffinity interactions between three orthogonal cohesin–dockerin
pairs. The multi-enzyme cascade showed 5 times higher NADH
production rate than the non-complexed enzyme mixture, a result
of efficient substrate channeling.
Enzymatic biofuel cells have significant advantages over conventional chemical fuel cells as enzymes are highly specific,
and the devices can be operated under neutral pH and ambient
temperature.1 However, they can be limited by low power
densities as most enzyme fuel cells employ only a single
enzyme for partial fuel oxidation.2 A significant improvement
in the current density can be achieved by using a multi-enzyme
cascade capable of complete fuel oxidization.3 For example,
methanol can be fully oxidized to CO2 by three NAD+-dependent
dehydrogenases, alcohol dehydrogenase (ADH), formaldehyde
dehydrogenase (FALDH) and formate dehydrogenase (FDH)4,5 to
provide six electrons to the anode rather than only 2 electrons
based on a single-step oxidation.
Multi-enzyme cascades capable of completely oxidizing fuels have
been created. Kar et al.6 reported the use of tetrabutylammonium
bromide modified Nafion membranes to entrap dehydrogenases for
the complete oxidation of methanol to CO2, and a marked improvement in the methanol oxidation current density was achieved.
In another study,7 three dehydrogenases were encapsulated
within titania particles resulting in up to 5 times higher
yield than that of free enzymes. This substantial improvement
in the overall reaction rate is a direct result of enzyme clustering afforded by enzyme co-immobilization and confinement,
a
Department of Chemical and Biomolecular Engineering, University of Delaware,
Newark, DE 19716, USA. E-mail: [email protected]; Fax: +1 30 2831 1048;
Tel: +1 30 2831 6327
b
Department of Chemical Engineering, Columbia University, New York, NY 10027,
USA
† Electronic supplementary information (ESI) available: Experimental details of
materials and methods and specific activities of the three dehydrogenases data.
See DOI: 10.1039/c3cc40454d
3766
Chem. Commun., 2013, 49, 3766--3768
which effectively reduces the diffusional length scale of the
intermediates along the multi-reaction pathway.8 More recently,
three dehydrogenases have been engineered to self-assemble into
mixed proteinaceous hydrogels and this produced high current
densities in bioanodes.5 Unfortunately, these multiple enzymes
were randomly clustered in the above examples, and precise
control over the ratio and ordering of the enzymes were not
achieved. An improved design that provides the appropriate ratio
and ordering of enzymes will likely result in a further improved
oxidation rate because of the more efficient sequential utilization
of the substrate. This is demonstrated by the 77-fold improvement in the product titer using a synthetic protein scaffold
containing three mevalonate biosynthetic enzymes.9 It is easy
to envision that the ability to spatially control multiple oxidative
enzymes in a similar manner can be used to enhance the overall
current densities for enzymatic biofuel cell applications.
The cohesin–dockerin (Coh–Doc) pair is a high-affinity protein
complex responsible for the position-specific self-assembly of
cellulosomes for enhanced cellulose degradation.10 The Coh–Doc
pairs have extremely high binding affinity with a dissociation
constant (Kd) of 10 9 to 10 12 M and are species specific.11 This
unique Coh–Doc interaction has been exploited to assemble
enzyme complexes containing three enzymes from the glycolysis
and gluconeogenesis pathways and showed increased reaction
rates facilitated by the substrate channeling effect.12 Here, we
created a multi-enzyme cascade for complete methanol oxidation by docking the three required dehydrogenases onto a
trifunctional synthetic scaffold displayed on the yeast surface
(Fig. 1). The amount of NADH produced, which is proportional
to the amount of electrons that can be transferred to the anode,
was used to assess the potential benefit.
Previously, a miniscaffoldin consists of an internal cellulosebinding domain (CBD) and three divergent cohesin domains
was displayed on the yeast cell surface.13 While the CBD domain
is required to bind cellulose, the additional spacing can potentially increase the diffusion distance between the first two
dehydrogenases. To facilitate efficient substrate channeling,
the internal CBD domain was deleted and a new trifunctional
scaffold containing only three divergent cohesin domains was
This journal is
c
The Royal Society of Chemistry 2013
Communication
ChemComm
Fig. 1 A schematic representation for the co-immobilization of three dehydrogenases on a trifunctional scaffold displayed on the yeast surface. Three orthogonal cohesin–dockerin pairs from Clostridium cellulolyticum (CC), Clostridium
thermocullum (CT) and Ruminoccocus flavefaciens (RF) are used for the assembly.
Fig. 2 Phase-contrast and immunofluorescence micrographs of yeast cells displaying the trifunctional scaffold or assembled with dockerin-tagged enzymes.
(A) Yeast cells not displaying the scaffold or (B) displaying the scaffold were
probed with an anti-c-Myc serum and fluorescently stained with a goat-antimouse IgG conjugated with Alexa 488. Functionality of the cohesin domains was
confirmed by the correct assembly of three dockerin-tagged cellulases, (C) Ec,
(D) At, or (E) Gf, on the displayed scaffolds using anti-c-His6 antibodies.
constructed (Fig. 1). A c-Myc tag at the C terminus of the
scaffold was used to probe the surface localization using
immunofluorescence microscopy. Display of the full-length
scaffold was confirmed by observing brightly fluorescent cells
(Fig. 2B), while no detectable fluorescence was observed for
control cells not displaying the scaffold (Fig. 2A).
To check the functionality of each cohesin domain on the
displayed scaffolds, yeast cells displaying scaffolds were incubated
with E. coli cell lysates containing each of the three dockerintagged cellulases Ec, At, or Gf13 for 1 h. The correct assembly of
each enzyme was confirmed by probing the C-terminus His6 tag
using immunofluorescence microscopy (Fig. 2C–E), indicating that
each of cohesin domain on the displayed scaffolds was functional.
In this work, methanol was fully oxidized to CO2 via the
sequential reactions catalyzed by using the ADH from Bacillus
stearothermophilus,5 the FALDH from Pseudomonas putida,14
and the FDH from Saccharomyces cerevisiae.5 The three dehydrogenases were each tagged at the C-terminus with a distinct
dockerin domain from either C. cellulolyticum, C. thermocellum
or R. flavefaciens that binds specifically to the corresponding
cohesin domain in the scaffold (Fig. 1). The recombinant
dehydrogenases were expressed in E. coli (Fig. 3A), and active
enzymes were confirmed by the corresponding enzymatic assays
(Table S1, ESI†). Since the tetrameric ADH and FALDH and the
dimeric FDH are all found to be active, these results indicate that
addition of the dockerin domain to the C-termini did not
influence the folding and activities of the enzymes.
This journal is
c
The Royal Society of Chemistry 2013
Fig. 3 Expression of dockerin-tagged dehydrogenases and their functional
assembly. (A) SDS-PAGE analysis of E. coli cell lysates expressing different
dockerin-tagged dehydrogenases, ADH-DocCC, FALDH-DocCT and FDH-DocRF.
The arrows indicated the expected dehydrogenase fusion proteins; (B) phasecontrast and immunofluorescence micrographs of yeast cells displaying scaffolds
docked with three dockerin-tagged dehydrogenases (ADH-DocCC, FALDH-DocCT
or FDH-DocRF). Cells were incubated with anti-c-His6 serum and fluorescently
stained with a goat anti-mouse IgG conjugated with Alexa 488. Cells displaying
only the scaffold were used as controls.
A His6 tag was also added to the C terminus of each of the
dehydrogenase fusions for probing the surface assembly. Yeast
cells displaying the trifunctional scaffold on the surface were
incubated with an excess amount of E. coli cell lysates containing
either ADH-DocCC, FALDH-DocCT, or FDH-DocRF for 1 h. The
assembly of each dehydrogenase on yeast cells surface was
confirmed by immunofluorescence microscopy. Detection of
fluorescent cells (Fig. 3B) in all cases confirmed the functionality
of the dockerin domain on each dehydrogenase.
Each of the dehydrogenase fusion enzymes were individually
assembled onto the surface-displayed scaffolds using varying
amounts of cell lysates. After incubation, cells were washed twice
and resuspended in the assay buffer before the whole-cell activity
was measured. For each dehydrogenase, the whole-cell activity
continued to increase when increasing amount of cell lysates
was incubated until reaching a plateau when all the binding sites
on the scaffolds were occupied (Fig. 4). From the individual
enzyme binding experiments, the amount of cell lysates needed
to saturate the appropriate cohesin domain was determined
(Fig. 4) and used for assembly of the multi-enzyme cascades.
Similarly, the unbound enzymes were removed and the level of
enzyme bound on the scaffold was determined based on the
differences in enzyme activities before and after binding.
To assess the synergistic effects of enzyme assembly, the
overall reaction rate was measured using the production of
NADH as an indicator. After saturating yeast cells displaying
the scaffolds with ADH-DocCC, cells were washed and tested for
NADH production using methanol as the substrate. The wholecell NADH production rate was 1.7 times higher than that of the
same amount of free ADH-DocCC (Fig. 5). This slight increase is
likely due to the improved enzyme–substrate association to the
Chem. Commun., 2013, 49, 3766--3768
3767
ChemComm
Fig. 4 Saturation binding curves for each dehydrogenase fusion. The total protein
concentration was 3 mg ml 1 for ADH-DocCC and FALDH-DocCC and 20 mg ml 1 for
FDH-DocRF. The arrows indicate the amounts of enzyme cell lysates used to incubate
with the yeast cells displaying scaffolds for the multi-enzyme cascades assembly.
Communication
5-fold enhancement is within the range reported for similar
synthetic protein scaffolds containing three mevalonate biosynthetic enzymes, which vary anywhere from 1.4-fold increase
in mevalonate synthesis using a 1 : 1 : 1 enzyme ratio to over
77-fold using a 1 : 2 : 2 enzyme ratio.9 It is possible that further
enhancement can be achieved by changing enzyme ratios using
different scaffold architectures.
Since the amount of NADH produced is directly related to
the amount of electrons that can be provided to the anode in an
enzymatic biofuel cell, our results indicate the reported threeenzyme cascade could be used to significantly increase the
current density. Because of the ease in preparation without the
need of protein purification, this method may be generalized to
enable the assembly of other multi-enzyme cascades using the
yeast surface displayed scaffolds.
In summary, three recombinant dehydrogenases, ADH, FALDH,
and FDH, each fused with a different dockerin domain, were
functionally expressed in E. coli and assembled onto a trifunctional
scaffold displayed on the yeast surface through the specific cohesin–
dockerin interactions, resulting in a three-enzyme cascade for the
conversion of methanol to CO2. The position-specific and sequential
assembly of three dehydrogenases on the displayed scaffold facilitates the efficient substrate channeling, resulting in more than
5 times higher NADH production rate than that of the noncomplexed enzymes. This substantial enhancement in NADH production could be very promising in improving the current density
when used for methanol-based enzymatic biofuel cell applications.
Notes and references
Fig. 5 NADH production rate generated by dehydrogenase multi-enzyme
cascade or by the same amount of free enzymes.
cell-surface oriented ADH as described before.15 When ADHDocCC and FALDH-DocCT were co-assembled on the displayed
scaffolds, a 3.8-fold increase in the NADH production rate was
observed when compared with the same amount of the two
enzymes in solution. This higher level of synergy for the twoenzyme system indicates the importance of substrate channeling
in enhancing the overall reaction rate and is consistent with
other two-enzyme clusters used for enzymatic fuel cell applications reported in literature.16
Finally, the three-enzyme cascade was formed by incubating
yeast cells displaying the trifunctional scaffolds with a saturating
level of the three dehydrogenases. The multi-enzyme cascades
with all three enzymes assembled onto the scaffolds exhibited the
highest level of synergy with the overall rate 5.1 times that of the
non-complexed three-enzyme mixture (Fig. 5). This result
indicates the additional benefits of clustering the entire threeenzyme cascade because of the sequential substrate channeling
and the ability to drive the reaction to the forward direction due to
the reduction in formate inhibition on FALDH and conversion of
formate to CO2 by FDH.17 It should be noted that our observed
3768
Chem. Commun., 2013, 49, 3766--3768
1 M. H. Osman, A. A. Shah and F. C. Walsh, Biosens. Bioelectron., 2011,
26, 3087.
2 (a) M. Togo, A. Takamura, T. Asai, H. Kaji and M. Nishizawa,
J. Power Sources, 2008, 178, 53; (b) A. Habrioux, G. Merle,
K. Servat, K. B. Kokoh, C. Innocent, M. Cretin and S. Tingry,
J. Electroanal. Chem., 2008, 622, 97; (c) T. L. Klotzbach, M. Watt,
Y. Ansari and S. D. Minteer, J. Membr. Sci., 2008, 311, 81.
3 D. Sokic-Lazic, R. L. Arechederra, B. L. Treu and S. D. Minteer,
Electroanalysis, 2010, 22, 757.
4 G. T. R. Palmore, H. Bertschy, S. H. Bergens and G. M. Whitesides,
J. Electroanal. Chem., 1998, 443, 155.
5 Y. H. Kim, E. Campbell, J. Yu, S. D. Minteer and S. Banta, Angew.
Chem., Int. Ed., 2013, 52, 1437.
6 P. Kar, H. Wen, H. Li, S. D. Minteer and S. Calabrese Barton,
J. Electrochem. Soc., 2011, 158, B580.
7 Q. Sun, Y. Jiang, Z. Jiang, L. Zhang, X. Sun and J. Li, Ind. Eng. Chem.
Res., 2009, 48, 4210.
8 B. El-Zahab, D. Donnelly and P. Wang, Biotechnol. Bioeng., 2008, 99, 508.
9 J. E. Dueber, G. C. Wu, G. R. Malmirchegini, T. S. Moon,
C. J. Petzold, A. V. Ullal, K. L. J. Prather and J. D. Keasling, Nat.
Biotechnol., 2009, 27, 753.
10 E. A. Bayer, J.-P. Belaich, Y. Shoham and R. Lamed, Annu. Rev.
Microbiol., 2004, 58, 521.
11 S. Pagès, A. Bélaı̈ch, J. P. Bélaı̈ch, E. Morag, R. Lamed, Y. Shoham
and E. A. Bayer, Proteins, 1997, 29, 517.
12 C. You, S. Myung and Y.-H. P. Zhang, Angew. Chem., Int. Ed., 2012,
51, 8787.
13 (a) S.-L. Tsai, J. Oh, S. Singh, R. Chen and W. Chen, Appl. Environ.
Microbiol., 2009, 75, 6087; (b) S.-L. Tsai, G. Goyal and W. Chen, Appl.
Environ. Microbiol., 2010, 76, 7514.
14 K. Ito, M. Takahashi, T. Yoshimoto and D. Tsuru, J. Bacteriol., 1994,
176, 2483.
15 C.-S. Wu, C.-T. Wu, Y.-S. Yang and F.-H. Ko, Chem. Commun., 2008, 5327.
16 M. J. Moehlenbrock, M. T. Meredith and S. D. Minteer, ACS Catal.,
2012, 2, 17.
17 P. K. Addo, R. L. Arechederra and S. D. Minteer, Electroanalysis,
2010, 22, 807.
This journal is
c
The Royal Society of Chemistry 2013