Fine Tuning of Spatial Arrangement of Enzymes in a

PCNA-Mediated Multienzyme Complex Using a Rigid

Poly-L-Proline Linker
Tomoaki Haga1, Hidehiko Hirakawa1*, Teruyuki Nagamune1,2*
1 Department of Chemistry and Biotechnology, School of Engineering, the University of Tokyo, Hongo, Tokyo, Japan, 2 Department of Bioengineering, School
of Engineering, the University of Tokyo, Hongo, Tokyo, Japan

Abstract
Inspired by natural multienzyme complexes, many types of artificial multienzyme complexes have recently been
constructed. We previously constructed a self-assembled complex of a bacterial cytochrome P450 and its ferredoxin
and ferredoxin reductase partners using heterotrimerization of proliferating cell nuclear antigen (PCNA) from
Sulfolobus solfataricus. In this study, we inserted different peptide linkers between ferredoxin and the PCNA subunit,
and examined the effect on activity of the self-assembled multienzyme complex. Although the activity was affected by
the lengths of both the rigid poly-L-proline-rich linkers and the flexible Gly4-Ser repeating linkers, the poly-L-prolinerich linkers provided the greatest activity enhancement. The optimized poly-L-proline-rich linker enhanced the activity
1.9-fold compared with the GGGGSLVPRGSGGGGS linker used in the previously reported complex, while the Gly4Ser repeating linkers, (G4S)n (n = 16), did not yield higher activity than the maximum activity by the optimized poly-Lproline linker. Both the rigidity/flexibility and length of the peptide linker were found to be important for enhancing the
overall activity of the multienzyme complex.
Citation: Haga T, Hirakawa H, Nagamune T (2013) Fine Tuning of Spatial Arrangement of Enzymes in a PCNA-Mediated Multienzyme Complex Using a
Rigid Poly-L-Proline Linker. PLoS ONE 8(9): e75114. doi:10.1371/[Link].0075114
Editor: Mark Isalan, Center for Genomic Regulation, Spain
Received June 5, 2013; Accepted August 8, 2013; Published September 5, 2013
Copyright: 2013 Haga et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by a Grant-in-Aid for Young Scientists (B) from the Ministry of Education, Culture, Sports, Science and Technology
(#21760633). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
* E-mail: hirakawa@[Link] (HH); nagamune@[Link] (TN)

Introduction

reductase (PdR), using proliferating cell nuclear antigen

(PCNA) subunits from Sulfolobus solfataricus as a selfassembling scaffold [13] (Figure 1A). S. solfataricus PCNA is a
heterotrimeric ring-shaped protein that provides a scaffold for
DNA-related enzymes by encircling dsDNA [1416]. The PCNA
heterotrimer is composed of three distinct subunits, PCNA1,
PCNA2 and PCNA3, which can be separately expressed as
monomer proteins and then assembled to form the
heterotrimer. Thus, three genetic fusion proteins of the PCNA
subunits with PdR, PdX and P450cam, i.e. PCNA1-PdR,
PCNA2-PdX and PCNA3-P450cam, respectively, were used to
form a heterotrimeric complex termed PUPPET (PCNAutilized protein complex of P450 and its two electron transferrelated
proteins).
PUPPET
showed
much
higher
monooxygenase activity than an equimolar mixture of free
PdR, PdX and P450cam. This is because PdR, PdX and
P450cam are located in close proximity to each other on the
PCNA heterotrimer and thus electrons are more efficiently
relayed from PdR to P450cam via PdX. However, the peptide
linkers between the PCNA subunits and the three enzymes,

In nature, enzymes often work together in cascade or

coupled reactions by forming multienzyme complexes. This
helps to prevent degradation of unstable intermediate products
and to enhance mass transfer between enzymes. Inspired by
natural multienzyme complexes, artificial multienzyme
complexes have been constructed by gene fusion, coimmobilization,
co-entrapment,
and
scaffold-mediated
assembly [1]. Scaffold-mediated multienzyme complexes have
been constructed using biomaterials such as RNA [2], DNA
[37] and protein [812]. In these complexes, enzymes are
connected to single scaffolds by nucleic acids or peptide
linkers. The linkers regulate the relative spatial arrangements
of the enzymes, which affect the overall activities of the
multienzyme complexes. However, except for a few reports
[8,9], the linkers in artificial multienzyme complexes have not
been examined in detail.
Recently, we demonstrated selective complex formation of
Pseudomonas putida cytochrome P450 (P450cam) with its
redox partner proteins, putidaredoxin (PdX) and putidaredoxin

PLOS ONE | [Link]

September 2013 | Volume 8 | Issue 9 | e75114

Multienzyme Complex Tuned by Rigid Peptide Linker

Figure 1. Optimization of the PCNA2-PdX fusion protein linker in PUPPET. (A) Model depicting the self-assembly of PCNA1PdR, PCNA2-PdX and PCNA3-P450cam. (B) Schematic representation of 11 PCNA2-PdX fusion proteins (P1-P5 and G1-G6). The
calculated molecular masses of the PCNA2-PdX linker variants are listed in the tables. MM, molecular mass.
doi: 10.1371/[Link].0075114.g001

enzymes with protein scaffolds for the construction of

multienzyme complexes.

which are likely to affect the overall activity of PUPPET, were

not examined in this earlier study.
In this report, we examined the effect of the peptide linker
between PCNA2 and PdX on the monooxygenase activity of
PUPPET. An optimum poly-L-proline-rich linker increased the
activity 1.9-fold compared with the GGGGSLVPRGSGGGGS
linker used in the previous report. The increased activity
conferred by the poly-L-proline-rich linker could not be
achieved using the optimized Gly4-Ser repeating linker. The
poly-L-proline-rich linker provided a rigid spacer between PdX
and PCNA and thus may have enabled a spatially-appropriate
arrangement for more efficient interactions between PdX and
P450cam. Therefore, linkers comprising a rigid peptide
sequence are likely to be the best candidates for fusing

PLOS ONE | [Link]

Materials and Methods

Plasmid construction
The plasmids expressing PCNA2 protein fused to PdX were
constructed by inserting DNA sequences encoding peptide
linkers between the C-terminus of PCNA2 and the N-terminus
of the Cys73Ser/Cys85Ser mutant of PdX. First, two SapI sites
were inserted by PCR between the genes encoding PCNA2
and the Cys73Ser/Cys85Ser mutant of PdX in pHSG+PCNA2PdX
[17],
using
two
primers,
5GGATCCGCTCTTCAGGAGGTGGTGGCTCTATGTCTAAAG-

September 2013 | Volume 8 | Issue 9 | e75114

Multienzyme Complex Tuned by Rigid Peptide Linker

Enzyme assays

3
(forward)
and
5GAATTCGCTCTTCTACCGCCGTCCGCGC-3 (reverse). The
DNA fragment, containing two SapI sites between PCNA2 and
PdX, was amplified from the resulting plasmid by PCR using
two
primers,
5TATTTCCAGGGCCATATGATGAAAGCCAAAGTGATCG-3
(forward) and 5-ATGGCCCTGGAAATACAGG-3 (reverse). A
modified pET-15b(+) (Novagen, Darmstadt, Germany), termed
pET15T, which lacked a SapI site and had a TEV protease
recognition site between a thrombin recognition site and an
NdeI site, was linearized by PCR using two primers, 5GGATCCGGCTGCTAACAAAG-3
(forward)
and
5TTAGCAGCCGGATCCTTACCATTGCCTATCGGG-3
(reverse). The amplified DNA fragment was fused to the
linearized plasmid using the In-Fusion enzyme (Clontech,
Mountain View, CA, USA), to obtain p2SSX. The
oligonucleotide sequences encoding (G4S)n (n = 13) were
inserted between the sequences encoding PCNA2 and PdX
into p2SSX by PCR. The oligonucleotide sequences encoding
(G4S)n (n = 46) and G4S(P5)nG4S (n = 15) were inserted by
ligating hybridized oligonucleotides with sticky ends into SapIdigested p2SSX. The oligonucleotides used to insert the linkers
are summarized in Table S1.
A gene encoding the PCNA3-P450cam fusion protein was
inserted between the NdeI and BamHI sites of pET-15b(+). The
resulting plasmid, pET15+oP3C, was mutated by PCR using
two primers, 5-AGCACCCGTCGTGAATTTAACGTTCGC-3
(forward) and 5-TTCACGACGGGTGCTGCCGATAATGTT-3
(reverse). The resulting plasmid, pET15+P3 N106RC, expressed
PCNA3N106R-P450cam fusion protein.

Cytochrome c reduction rates were determined from the

absorption changes at 550 nm. The extinction coefficient
difference (cytochrome cred cytochrome cox) used to calculate
the rates was 21.1 mM-1 cm-1. The assays were performed in
50 mM potassium phosphate buffer, pH7.4, containing 150 mM
potassium chloride, 20 M cytochrome c (Sigma-Aldrich, St.
Louis, MS, USA), 50 M NADH (Sigma-Aldrich) and 4.5 nM
heterotrimeric complex. P450cam catalytic activities were
determined from O2 consumption rates, which were directly
measured using a Clark-type oxygen electrode, by subtracting
the values obtained in the absence of d-camphor from those in
the presence of d-camphor. The assays were performed in 50
mM potassium phosphate buffer, pH7.4, containing 150 mM
potassium chloride, 500 M d-camphor, 100 M NADH and
various concentrations of the heterotrimers. All assays were
conducted at 25C.

Results
Construction of PCNA2-PdX fusion proteins
To enhance the activity of PUPPET, we optimized the
peptide linker connecting PdX and PCNA2, because PdX
interacts with both PdR and P450cam, and the spatial
arrangement of PdX relative to these enzymes affects their
interactions in the PUPPET complex. We therefore introduced
various lengths of the poly-L-proline or Gly4-Ser repeating
linkers between PCNA2 and PdX. Rigid poly-L-proline peptides
have often been used as spacers [1822]. Flexible Gly4-Ser
repeating peptides have also frequently been used for antibody
engineering [23] and bifunctional fusion protein construction
[2426]. PCNA2-PdX fusion proteins, PCNA2-G 4S(P5)nG4SPdX (n = 15) and PCNA2-(G4S)n-PdX (n = 16), which have
proline-rich linkers, G4S(P5)nG4S, and GlySer-rich linkers,
(G4S)n, respectively (Figure 1B), were successfully expressed
and purified (Figure 2A). The fusion proteins retained
characteristic UVVis spectra for PdX (Figure 3A and Figure
S1).

Preparation of PUPPETs
PCNA1-PdR and PCNA2-PdX fusion proteins were
expressed and purified as described in a previous report [13],
except that BL21 Star (DE3) pLysS (Invitrogen, Carlsbad, CA,
USA) was used. PCNA3-P450cam fusion protein, in which
PCNA3 had a N106R mutation that improved the solubility of
the fusion protein but did not affect PUPPET formation
(unpublished data), was expressed and purified as described in
the above report, except that T7 Express Iq (New England
BioLabs, Ipswich, MA, USA) was transformed with pET15+P3
N106RC.
An approximately equimolar ratio of PCNA1-PdR and
PCNA2-PdX and a slightly excess amount of PCNA3-P450cam
were mixed and incubated on ice for 1 h. The mixture was
subjected to size-exclusion chromatography on a Superdex
200 10/300 GL column (1.0 30 cm; GE Healthcare, Little
Chalfont, Buckinghamshire, UK) with 50 mM potassium
phosphate buffer, pH7.4, containing 150 mM potassium
chloride and 2 mM d-camphor (Wako Pure Chemical
Industries, Osaka, Japan). The fractions with an Abs455/Abs392
ratio of 0.310.34 were collected. The collected fractions were
combined and concentrated using an Amicon Ultra-15
Centrifugal Unit (30,000 NMWL; Millipore, Billerica, MA, USA).
The d-camphor was then removed using a HiTrap desalting
column pre-equilibrated with 50 mM potassium phosphate
buffer, pH7.4, containing 150 mM potassium chloride.

PLOS ONE | [Link]

Preparation of the heterotrimers

Heterotrimeric complexes of PdR, PdX and P450cam were
prepared by mixing PCNA1-PdR, PCNA2-PdX and PCNA3P450cam. Each resulting protein complex, PUPPET-Pn (n = 1
5) and PUPPET-Gn (n = 16), comprising PCNA2-G
4S(P5)nG4S-PdX (n = 15) and PCNA2-(G4S)n-PdX (n = 16),
respectively (Figure 2B), showed similar UVVis spectra to a
linear combination of PCNA1-PdR, PCNA2-PdX and PCNA3P450cam, as previously reported [13] (Figure 3B and Figure
S2). The spectrum of each PUPPET had a slightly lower
absorbance around 390 nm and slightly higher absorbance
around 420 nm, compared with the linear combination spectra
of PCNA1-PdR, PCNA2-PdX and PCNA3-P450cam as
observed in the previous study. The linker between PCNA2
and PdX did not significantly affect the PUPPET spectrum
(Figure 3B and Figure S2).

September 2013 | Volume 8 | Issue 9 | e75114

Multienzyme Complex Tuned by Rigid Peptide Linker

PUPPET-Pn and PUPPET-Gn, the poly-L-proline-rich linkers,

G4S(P5)nG4S (n = 25), were more effective than the Gly4-Ser
repeating linkers in enhancing the activity of PUPPET. The
G4SP20G4S linker gave the highest activity (6.5 0.3 [Link]-1);
a level not achieved by the Gly4-Ser repeating linkers, (G4S)n (n
= 16). This activity is 1.9-fold greater than that of the
previously reported PUPPET complex that had a
GGGGSLVPRGSGGGGS linker (3.4 0.1 M min-1).
We compared the monooxygenase activities of PUPPET-P4
and PUPPET-G5 to the original PUPPET complex (Figure 6),
using various enzyme concentrations. In agreement with the
previous report [13], the initial activity of each PUPPET did not
have a linear relationship with the PUPPET concentration
(Figure 6A), and the apparent specific activity decreased
according to the decrease in PUPPET concentration (Figure
6B), probably because of the dissociation of PCNA3-P450cam
from PUPPET at lower PUPPET concentrations. At each
enzyme concentration, PUPPET-P4 showed higher activity
compared with PUPPET-G5. The apparent specific activities of
PUPPET-P4 and PUPPET-G5 showed similar PUPPET
concentration dependency to that of the original PUPPET, and
therefore the enhanced activity levels of PUPPET-P4 and
PUPPET-G5 compared with the original PUPPET complex are
independent of the PUPPET concentration.

Discussion
The functions of natural multi-domain proteins are often
regulated by properties of the interdomain-connecting peptide
linkers, such as their hydrophobicity and rigidity (reviewed in
28). The functions of artificial chimeric proteins are often
similarly dependent on the linker properties [29]. Recently,
artificial scaffold-mediated multienzyme complexes, in which
linkers connect enzymes with single scaffolds, have attracted
much attention because scaffolds can provide a basis for the
precise control of the relative position and orientation of
multiple enzymes [1]. Although the effect of linker length on
protein-ligand and proteinprotein associations [30,31] and
bifunctional multienzyme complexes [26,32,33] has been well
studied, only a few reports have investigated the effect of the
linker length on artificial scaffold-mediated multienzyme
complexes [8,9]. Moreover, the effect of the type of linker on
the overall activities of artificial scaffold-mediated multienzyme
complexes has never been examined. Linkers in artificial
multienzyme complexes [3,1012], which catalyze cascade
reactions, might not have been considered to significantly
affect the whole catalytic activities because cascade reactions
do not depend on direct physical contact between enzymes. In
this study, we examined the effect of the type and length of the
peptide linker connecting PdX with the protein scaffold on the
overall activity of the artificial protein complex comprising
P450cam, PdX and PdR.
The monooxygenase activity of PUPPET was observed to
depend on the characteristics of the linker between PCNA2
and PdX (Figure 5). This implies that the linker affected the
electron transfer process from PdR to P450cam, because the
electron transfer from PdX to P450cam during their transient
complex formation is a rate limiting step in the P450cam

Figure 2. SDS-PAGE analyses of PCNA2-PdX fusion

proteins and PUPPETs. (A) PCNA2-PdX fusion proteins: lane
15, PCNA2-G 4S(P5)nG4S-PdX (n = 15); lane 611, PCNA2(G4S)n-PdX (n = 16) (B) PUPPETs: lane 15, PUPPET-Pn (n =
15); lane 611, PUPPET-Gn (n = 16). P1-PdR, P2-PdX and
P3-P450 are PCNA1-PdR, PCNA2-PdX and PCNA3-P450cam,
respectively.
doi: 10.1371/[Link].0075114.g002

Cytochrome c reduction activities of PUPPET

Cytochrome c reduction activities were measured to evaluate
the effect of the linker on the electron transfer rate from PdR to
PdX in PUPPET. The cytochrome c reduction rate reflects the
electron transfer rate from PdR to PdX because direct
cytochrome c reduction by PdR is negligible compared with
that effected by reduced PdX [27]. The linker between PCNA2
and PdX was not observed to significantly affect the
cytochrome c reduction activity of PUPPET (Figure 4).

Effect of the linker on PUPPET activity

The effect of the linker on the monooxygenase activities of
PUPPET was evaluated by measuring the substrate-dependent
O2 consumption rate at the complex concentration of 18 nM
(Figure 5). In PUPPET-Pn, the activity increased with
increasing linker length in the range of n = 13, although
PUPPET-Pn (n = 35) showed similar activity (Figure 5A).
Likewise, in PUPPET-Gn, the activity increased with increasing
linker length in the range of n = 13, although PUPPET-Gn (n =
36) showed similar activity (Figure 5B). Although the linker
length had an effect on the monooxygenase activity of both

PLOS ONE | [Link]

September 2013 | Volume 8 | Issue 9 | e75114

Multienzyme Complex Tuned by Rigid Peptide Linker

Figure 3. UVVis spectra of PCNA2-G 4S-PdX and PUPPET-G1. (A) UVVis spectrum of PCNA2-G 4S-PdX. The peaks at 314,
412, and 460 nm, and the broad band in the long wavelength region, are specific to [2Fe-2S] cluster-containing proteins. UVVis
spectra of PCNA2-(G4S)n-PdX (n = 26) and PCNA2-G 4S(P5)nG4S-PdX (n = 15) are shown in Figure S1. (B) UVVis spectra of
PUPPET-G1 (solid line), and a linear combination of the individual component protein spectra (broken line). PUPPET-Gn (n = 26)
and PUPPET-Pn (n = 15) are shown in Figure S2.
doi: 10.1371/[Link].0075114.g003

PLOS ONE | [Link]

September 2013 | Volume 8 | Issue 9 | e75114

Multienzyme Complex Tuned by Rigid Peptide Linker

Figure 4. Cytochrome c reduction activities of the PUPPET linker variants. The activities of (A) PUPPET-Pn (n = 15) and (B)
PUPPET-Gn (n = 16) were evaluated in a reaction mixture containing 4.5 nM enzymes and 20 M cytochrome c. Error bars
represent the standard errors of three replicates.
doi: 10.1371/[Link].0075114.g004

PLOS ONE | [Link]

September 2013 | Volume 8 | Issue 9 | e75114

Multienzyme Complex Tuned by Rigid Peptide Linker

Figure 5. P450cam oxidation activities of the PUPPET linker variants. The activities of (A) PUPPET-Pn (n = 15) and (B)
PUPPET-Gn (n = 16) were evaluated in reaction mixtures containing 18 nM enzyme. The activity of the previously reported
PUPPET, containing a GGGGSLVPRGSGGGGS linker (indicated as O), and the activity of an equimolar mixture of PdR, PdX and
P450cam (indicated as R), were also evaluated under the same reaction conditions. Error bars represent the standard error of
three replicates; *P < 0.05, ***P < 0.001 (Students t-test).
doi: 10.1371/[Link].0075114.g005

PLOS ONE | [Link]

September 2013 | Volume 8 | Issue 9 | e75114

Multienzyme Complex Tuned by Rigid Peptide Linker

Figure 6. PUPPET concentration-dependent monooxygenase activities. (A) Initial rates and (B) specific activities were plotted
against the protein concentration of PUPPET-P4 (open squares), PUPPET-G5 (open triangles) and original PUPPET (closed circles).
Each specific activity was normalized using the activity at 72 nM in Figure 6B. Error bars represent the standard error of three
replicates.
doi: 10.1371/[Link].0075114.g006

PLOS ONE | [Link]

September 2013 | Volume 8 | Issue 9 | e75114

Multienzyme Complex Tuned by Rigid Peptide Linker

Figure 7. Models for binding between PdX and P450cam in PUPPET. (A) A docking model of P450cam and PdX. The docking
program GRAMM-X [42] was used to generate the model from crystal structures of P450cam (PDB: 1DZ4) and PdX (PDB: 1XLP)
according to a previous report [40]. (B, C) Spatial arrangement of P450cam and the PCNA ring when the PdX-binding site of
P450cam faces (B) in the same direction as or (C) in a perpendicular direction to the PCNA ring. The distance between the Ctermini of PCNA2 and PCNA3 was estimated from the crystal structure of the PCNA heterotrimer (PDB: 2NTI).
doi: 10.1371/[Link].0075114.g007

catalytic cycle [34]. The electron transfer process from PdR to

P450cam is divided into two steps: reduction of PdX by PdR,
and electron donation from reduced PdX to P450cam. The
cytochrome c reduction activity, which reflects the reduction
rate of PdX by PdR, was not significantly affected by changing
the PCNA2-PdX linker (Figure 4). The dependence of
cytochrome c reduction activity on the linker was not identical
to that of the monooxygenase activity. This result indicates that
the linker mostly affected the electron donation step from
reduced PdX to P450cam.
The linker length of the poly-L-proline-rich linker,
G4S(P5)nG4S (n = 15), had a marked effect on the
monooxygenase activity of PUPPET (Figure 5). Though the
dissociation of PCNA3-P450cam from PUPPET is known to
have a negative effect on the activity [13,17], the effect of the
linker length was not involved in this dissociation, because the
relationship between concentration and activity of PUPPET
was not affected by the linker variants (Figure 6B). The poly-Lproline linker has been used as a molecular ruler, because
poly-L-proline is expected to form an all-trans type II helix and
the length of the helix can be predicted as approximately 3
per residue [21,35]. Recent studies demonstrated that the endto-end distances of the poly-L-proline peptides are shorter than
those of perfectly rigid type II proline helices, probably because
of the small number of cis isomers interspersed in aqueous
solutions of poly-L-proline peptides [3639]. Nevertheless, the
end-to-end distance was reported to increase with an increase
in the number of proline residues, up to at least 24 amino acids
[38]. Therefore, the repeat number of the Pro residue in the
linker is expected to control the distance between PCNA2 and
PdX, and to subsequently affect the spatial arrangement of
PdX relative to P450cam in PUPPET-Pn. The strong
dependence of the monooxygenase activity on the repeat
number of the Pro residue in the G4S(P5)nG4S linker indicates

PLOS ONE | [Link]

that the relative spatial arrangement of PdX governs the

electron donation step from reduced PdX to P450cam in the
PUPPETs.
A previously reported docking model of the P450cam-PdX
electron transfer complex [40], demonstrated that the distance
between Leu11 of P450cam (of which residues Met1Asn10
are undefined in the crystal structure), and the N-terminus of
PdX, is approximately 65 in the electron transfer complex
(Figure 7A). We also estimated that the mean end-to-end
distance of helical linker Pro20 was approximately 50
according to a previous study [38]. PdX and P450cam are
located on the same side of the PCNA ring because they are
fused to the C-termini of PCNA2 and PCNA3, respectively. By
combining the above structural information, we suggest a
model of the electron transfer complex, in which the PdXbinding site of P450cam faces in a perpendicular direction
towards the PCNA ring (Figure 7B). In this model, the
G4SP20G4S linker between the C-terminus of PCNA2 and the
N-terminus of PdX is the most effective in allowing PdX close
to the P450cam binding site. On the contrary, the speculation
that the binding site faces in a parallel direction to the PCNA
ring as shown in Figure 7C, conflicts with the experimental data
showing that the activity of PUPPET-Pn increased with an
increase in the distance between the N-terminus of PdX and
the C-terminus of PCNA2, in the range of n = 14. Therefore,
the PdX-binding site of P450cam should preferably face in the
perpendicular direction to the PCNA ring in the complex.
Although the length of the Gly4-Ser repeating linker, (G 4S)n (n
= 16), also affected the monooxygenase activity, these linkers
did not allow these PUPPET complexes to achieve the activity
levels of even PUPPET-P2 (Figure 5). This result suggests that
the mean end-to-end distances of the (G4S)n linkers are shorter
than that of G4SP10G4S. Evers et al. demonstrated that the
elongation of a Gly- and Ser-rich (GGSGGS)n linker from n =

September 2013 | Volume 8 | Issue 9 | e75114

Multienzyme Complex Tuned by Rigid Peptide Linker

suggest that linkers containing rigid peptides are more suitable

for the control of the spatial arrangement of enzymes in
artificial multienzyme complexes in which component enzymes/
proteins need to interact with each other, and can thereby
result in improved catalytic activity.

16 only increased the distance between the two linked

proteins by approximately 6 [41]. The Gly4-Ser repeating
linker probably has similar characteristics to the (GGSGGS)n
linker. Therefore, the flexible (G4S)n linkers do not enable PdX
to be located at the appropriate distance away from the PCNA
ring in order for the electron transfer complex to form, as
shown in Figure 7B, and may instead cause PdX to interact
with P450cam in an unfavorable manner, as shown in Figure
7C. The difference in the possible spatial arrangements of PdX
in PUPPET-Pn and PUPPET-Gn might therefore result in the
differences observed in their activities.

Supporting Information
Figure S1. UVVis spectra of PCNA2-PdX linker variants.
UVVis spectra of PCNA2-G 4S(P5)nG4S-PdX (n = 15, P1-P5)
and PCNA2-(G4S)n-PdX (n = 26, G2-G6) are listed.
(TIF)

Conclusion

Figure S2. UVVis spectra of PUPPET linker variants. Solid

lines indicate each PUPPET-Pn (n = 15, P1-P5) and PUPPETGn (n = 26, G2-G6) spectrum. Broken lines indicate a linear
combination of the individual component protein spectra.
(TIF)

In this study, we successfully enhanced the monooxygenase

activity of the PCNA-mediated protein complex of PdR, PdX
and P450cam by optimizing the peptide linker between PCNA2
and PdX. The linker affected the electron donation step from
reduced PdX to P450cam in the complex. The G4SP20G4S
linker most effectively enhanced the activity, most likely
because the rigid Pro20 sequence contributed to PdX
positioning close to the PdX-binding site of P450cam. On the
other hand, the high activity obtained with this linker was not
achieved with the flexible (G4S)n linkers in the range of n = 16,
owing to the short end-to-end distances of these flexible
linkers. Recently, we have demonstrated that introduction of
disulfide bonds between the PCNA subunits prevented the
dissociation of PCNA3-P450cam from PUPPET and enhanced
the activity [17]. Therefore, introduction of disulfide bonds can
further enhaced the activity of PUPPET-P4. Our findings

Table S1. Oligo DNA sequences to insert peptide linkers

between PCNA2 and PdX.
(DOC)

Author Contributions
Conceived and designed the experiments: TH HH. Performed
the experiments: TH. Analyzed the data: TH. Contributed
reagents/materials/analysis tools: TH HH. Wrote the
manuscript: TH HH TN. Interpreted the data: TH HH TN.

References
1. Schoffelen S, Van Hest JCM (2012) Multi-enzyme systems: bringing
enzymes together in vitro. Soft Matter 8: 1736-1746. doi:10.1039/
c1sm06452e.
2. Delebecque CJ, Lindner AB, Silver PA, Aldaye FA (2011) Organization
of intracellular reactions with rationally designed RNA assemblies.
Science 333: 470474. doi:10.1126/science.1206938. PubMed:
21700839.
3. Conrado RJ, Wu GC, Boock JT, Xu H, Chen SY et al. (2012) DNAguided assembly of biosynthetic pathways promotes improved catalytic
efficiency. Nucleic Acids Res 40: 18791889. doi:10.1093/nar/gkr888.
PubMed: 22021385.
4. Erkelenz M, Kuo CH, Niemeyer CM (2011) DNA-mediated assembly of
cytochrome P450 BM3 subdomains. J Am Chem Soc 133: 16111
16118. doi:10.1021/ja204993s. PubMed: 21919448.
5. Fu J, Liu M, Liu Y, Woodbury NW, Yan H (2012) Interenzyme substrate
diffusion for an enzyme cascade organized on spatially addressable
DNA nanostructures. J Am Chem Soc 134: 55165519. doi:10.1021/
ja300897h. PubMed: 22414276.
6. Mller J, Niemeyer CM (2008) DNA-directed assembly of artificial
multienzyme complexes. Biochem Biophys Res Commun 377: 6267.
doi:10.1016/[Link].2008.09.078. PubMed: 18823945.
7. Wilner OI, Weizmann Y, Gill R, Lioubashevski O, Freeman R et al.
(2009) Enzyme cascades activated on topologically programmed DNA
scaffolds. Nat Nanotechnol 4: 249254. doi:10.1038/nnano.2009.50.
PubMed: 19350036.
8. Agapakis CM, Ducat DC, Boyle PM, Wintermute EH, Way JC et al.
(2010) Insulation of a synthetic hydrogen metabolism circuit in bacteria.
J Biol Eng 4: 3 doi:10.1186/1754-1611-4-3. PubMed: 20184755.
9. Caspi J, Barak Y, Haimovitz R, Irwin D, Lamed R et al. (2009) Effect of
linker length and dockerin position on conversion of a Thermobifida
fusca endoglucanase to the cellulosomal mode. Appl Environ Microbiol
75: 73357342. doi:10.1128/AEM.01241-09. PubMed: 19820154.
10. Dueber JE, Wu GC, Malmirchegini GR, Moon TS, Petzold CJ et al.
(2009) Synthetic protein scaffolds provide modular control over

PLOS ONE | [Link]

11.

12.

13.

14.

15.

16.

17.

18.

19.

10

metabolic flux. Nat Biotechnol 27: 753759. doi:10.1038/nbt.1557.

PubMed: 19648908.
You C, Myung S, Zhang Y-HP (2012) Facilitated substrate channeling
in a self-assembled trifunctional enzyme complex. Angew Chem Int Ed
51: 87878790 doi:10.1002/anie.201202441.
You C, Zhang Y-HP (2013) Self-Assembly of Synthetic Metabolons
through Synthetic Protein Scaffolds: One-Step Purification, Coimmobilization, and Substrate Channeling. Acs Synth Biol 2: 102110.
doi:10.1021/sb300068g. PubMed: 23656373.
Hirakawa H, Nagamune T (2010) Molecular assembly of P450 with
ferredoxin and ferredoxin reductase by fusion to PCNA. Chembiochem
11: 15171520. doi:10.1002/cbic.201000226. PubMed: 20607777.
Dionne I, Nookala RK, Jackson SP, Doherty AJ, Bell SD (2003) A
heterotrimeric PCNA in the hyperthermophilic archaeon Sulfolobus
solfataricus.
Mol
Cell
11:
275282.
doi:10.1016/
S1097-2765(02)00824-9. PubMed: 12535540.
Dionne I, Brown NJ, Woodgate R, Bell SD (2008) On the mechanism of
loading the PCNA sliding clamp by RFC. Mol Microbiol 68: 216222.
doi:10.1111/j.1365-2958.2008.06150.x. PubMed: 18312273.
Hlinkova V, Xing G, Bauer J, Shin YJ, Dionne I et al. (2008) Structures
of monomeric, dimeric and trimeric PCNA: PCNA-ring assembly and
opening. Acta Crystallogr D Biol Crystallogr 64: 941949. doi:10.1107/
S0907444908021665. PubMed: 18703842.
Hirakawa H, Kakitani A, Nagamune T (2013) Introduction of selective
intersubunit disulfide bonds into self-assembly protein scaffold to
enhance an artificial multienzyme complexs activity. Biotechnol Bioeng
110: 18581864. doi:10.1002/bit.24861. PubMed: 23404255.
Arora PS, Ansari AZ, Best TP, Ptashne M, Dervan PB (2002) Design of
artificial transcriptional activators with rigid poly-L-proline linkers. J Am
Chem Soc 124: 1306713071. doi:10.1021/ja0208355. PubMed:
12405833.
Blanco-Canosa JB, Medintz IL, Farrell D, Mattoussi H, Dawson PE
(2010) Rapid covalent ligation of fluorescent peptides to water

September 2013 | Volume 8 | Issue 9 | e75114

Multienzyme Complex Tuned by Rigid Peptide Linker

20.

21.
22.

23.

24.

25.

26.

27.

28.
29.
30.

31.

solubilized quantum dots. J Am Chem Soc 132: 1002710033. doi:

10.1021/ja910988d. PubMed: 20597509.
Sato S, Kwon Y, Kamisuki S, Srivastava N, Mao Q et al. (2007)
Polyproline-rod approach to isolating protein targets of bioactive small
molecules: isolation of a new target of indomethacin. J Am Chem Soc
129: 873880. doi:10.1021/ja0655643. PubMed: 17243824.
Stryer L, Haugland RP (1967) Energy transfer: a spectroscopic ruler.
Proc Natl Acad Sci U S A 58: 719726. doi:10.1073/pnas.58.2.719.
PubMed: 5233469.
Tanaka T, Nomura W, Narumi T, Masuda A, Tamamura H (2010)
Bivalent ligands of CXCR4 with rigid linkers for elucidation of the
dimerization state in cells. J Am Chem Soc 132: 1589915901. doi:
10.1021/ja107447w. PubMed: 20973474.
Huston JS, Levinson D, Mudgett-Hunter M, Tai MS, Novotn J et al.
(1988) Protein engineering of antibody binding sites: recovery of
specific activity in an anti-digoxin single-chain Fv analogue produced in
Escherichia coli. Proc Natl Acad Sci U S A 85: 58795883. doi:
10.1073/pnas.85.16.5879. PubMed: 3045807.
Arai R, Ueda H, Kitayama A, Kamiya N, Nagamune T (2001) Design of
the linkers which effectively separate domains of a bifunctional fusion
protein. Protein Eng 14: 529532. doi:10.1093/protein/14.8.529.
PubMed: 11579220.
Arai R, Wriggers W, Nishikawa Y, Nagamune T, Fujisawa T (2004)
Conformations of variably linked chimeric proteins evaluated by
synchrotron X-ray small-angle scattering. Proteins 57: 829838. doi:
10.1002/prot.20244. PubMed: 15390267.
Lu P, Feng MG (2008) Bifunctional enhancement of a beta-glucanasexylanase fusion enzyme by optimization of peptide linkers. Appl
Microbiol Biotechnol 79: 579587. doi:10.1007/s00253-008-1468-4.
PubMed: 18415095.
Roome PW, Peterson JA (1988) The oxidation of reduced
putidaredoxin reductase by oxidized putidaredoxin. Arc Biochem
Biophys 266: 4150. doi:10.1016/0003-9861(88)90234-2. PubMed:
2845865.
Wriggers W, Chakravarty S, Jennings PA (2005) Control of protein
functional dynamics by peptide linkers. Biopolymers 80: 736746. doi:
10.1002/bip.20291. PubMed: 15880774.
Chen X, Zaro JL, Shen W-C (2013) Fusion protein linkers: Property,
design and functionality. Adv Drug Deliv Rev (. (2012)) doi:10.1016/
[Link].2012.09.039. PubMed: 23026637.
Anthis NJ, Clore GM (2013) The length of the calmodulin linker
determines the extent of transient interdomain association and target
affinity. J Am Chem Soc 135: 96489651. doi:10.1021/ja4051422.
PubMed: 23782151.
Krishnamurthy VM, Semetey V, Bracher PJ, Shen N, Whitesides GM
(2007) Dependence of effective molarity on linker length for an

PLOS ONE | [Link]

32.

33.

34.
35.
36.

37.

38.

39.
40.

41.

42.

11

intramolecular protein-ligand system. J Am Chem Soc 129: 13121320.

doi:10.1021/ja066780e. PubMed: 17263415.
Albertsen L, Chen Y, Bach LS, Rattleff S, Maury J et al. (2011)
Diversion of flux toward sesquiterpene production in Saccharomyces
cerevisiae by fusion of host and heterologous enzymes. Appl Environ
Microbiol 77: 10331040. doi:10.1128/AEM.01361-10. PubMed:
21148687.
Carlsson H, Ljung S, Blow L (1996) Physical and kinetic effects on
induction of various linker regions in beta-galactosidase/galactose
dehydrogenase fusion enzymes. Biochim Biophys Acta 1293: 154160.
doi:10.1016/0167-4838(95)00240-5. PubMed: 8652621.
Brewer CB, Peterson JA (1988) Single turnover kinetics of the reaction
between oxycytochrome P-450cam and reduced putidaredoxin. J Biol
Chem 263: 791798. PubMed: 2826462.
Cowan PM, McGavin S (1955) Structure of Poly-L-Proline. Nature 176:
501503. doi:10.1038/176501a0.
Best RB, Merchant KA, Gopich IV, Schuler B, Bax A et al. (2007) Effect
of flexibility and cis residues in single-molecule FRET studies of
polyproline. Proc Natl Acad Sci U S A 104: 1896418969. doi:10.1073/
pnas.0709567104. PubMed: 18029448.
Doose S, Neuweiler H, Barsch H, Sauer M (2007) Probing polyproline
structure and dynamics by photoinduced electron transfer provides
evidence for deviations from a regular polyproline type II helix. Proc
Natl Acad Sci U S A 104: 1740017405. doi:10.1073/pnas.
0705605104. PubMed: 17956989.
Hanson JA, Brokaw J, Hayden CC, Chu J-W, Yang H (2012) Structural
distributions from single-molecule measurements as a tool for
molecular mechanics. Chem Phys 396: 6171. doi:10.1016/
[Link].2011.06.014. PubMed: 22661822.
Watkins LP, Chang H, Yang H (2006) Quantitative single-molecule
conformational distributions: a case study with poly-(L-proline). J Phys
Chem A 110: 51915203. doi:10.1021/jp055886d. PubMed: 16610843.
Kuznetsov VY, Poulos TL, Sevrioukova IF (2006) Putidaredoxin-tocytochrome P450cam electron transfer: differences between the two
reductive steps required for catalysis. Biochemistry 45: 1193411944.
doi:10.1021/bi0611154. PubMed: 17002293.
Evers TH, Van Dongen EMWM, Faesen AC, Meijer EW, Merkx M
(2006) Quantitative understanding of the energy transfer between
fluorescent proteins connected via flexible peptide linkers. Biochemistry
45: 1318313192. doi:10.1021/bi061288t. PubMed: 17073440.
Tovchigrechko A, Vakser IA (2006) GRAMM-X public web server for
protein-protein docking. Nucleic Acids Res 34: W310W314. doi:
10.1093/nar/gkj099. PubMed: 16845016.

September 2013 | Volume 8 | Issue 9 | e75114