Nagamune Nano Convergence (2017) 4:9

DOI 10.1186/s40580-017-0103-4

REVIEW Open Access

Biomolecular engineering fornanobio/

bionanotechnology
TeruyukiNagamune*

Abstract
Biomolecular engineering can be used to purposefully manipulate biomolecules, such as peptides, proteins, nucleic
acids and lipids, within the framework of the relations among their structures, functions and properties, as well as
their applicability to such areas as developing novel biomaterials, biosensing, bioimaging, and clinical diagnostics and
therapeutics. Nanotechnology can also be used to design and tune the sizes, shapes, properties and functionality of
nanomaterials. As such, there are considerable overlaps between nanotechnology and biomolecular engineering, in
that both are concerned with the structure and behavior of materials on the nanometer scale or smaller. Therefore,
in combination with nanotechnology, biomolecular engineering is expected to open up new fields of nanobio/
bionanotechnology and to contribute to the development of novel nanobiomaterials, nanobiodevices and nanobio-
systems. This review highlights recent studies using engineered biological molecules (e.g., oligonucleotides, peptides,
proteins, enzymes, polysaccharides, lipids, biological cofactors and ligands) combined with functional nanomaterials
in nanobio/bionanotechnology applications, including therapeutics, diagnostics, biosensing, bioanalysis and biocata-
lysts. Furthermore, this review focuses on five areas of recent advances in biomolecular engineering: (a) nucleic acid
engineering, (b) gene engineering, (c) protein engineering, (d) chemical and enzymatic conjugation technologies,
and (e) linker engineering. Precisely engineered nanobiomaterials, nanobiodevices and nanobiosystems are antici-
pated to emerge as next-generation platforms for bioelectronics, biosensors, biocatalysts, molecular imaging modali-
ties, biological actuators, and biomedical applications.
Keywords: Engineered biological molecules, Therapy, Diagnosis, Biosensing, Bioanalysis, Biocatalyst, Nucleic acid
engineering, Gene engineering, Protein engineering, Conjugation technologies

1Introduction Nanobiotechnology is used in relation to the ways

Nanotechnology is the creation and utilization of mate- in which nanotechnology is used to create materials,
rials, devices, and systems through controlling matter devices and systems for studying biological systems and
on the nanometer scale, and it is the key technology of developing new biological assay, diagnostic, therapeutic,
the twenty-first century. The ability to exploit the struc- information storage and computing systems, among oth-
tures, functions and processes of biological molecules, ers. These systems use nanotechnology to advance the
complexes and nanosystems to produce novel functional goals of biological fields. Some nanobiotechnologies scale
nanostructured biological materials has created the from the top down, such as from microfluidics to nano-
rapidly growing fields of nanobiotechnology and bion- fluidic biochips (e.g., lab-on-a-chip for continuous-flow
anotechnology, which are fusion research fields of nano- separation and the detection of such macromolecules
technology and biotechnology [1]. Although these words as DNA and proteins [2], point-of-care biosensors for
are often used interchangeably, in this review, they are detecting biomarkers and clinical diagnosis [37], and
utilized in terminologically different ways, as follows. solid-state nanopore sensors for DNA sequencing [8]).
Other nanobiotechnologies scale from the bottom up
for the fabrication of nanoscale hybrid materials, such as
*Correspondence: nagamune@bio.t.utokyo.ac.jp
Department ofChemistry andBiotechnology, Graduate School
complexes consisting of nanoparticles (NPs) (e.g., mag-
ofEngineering, The University ofTokyo, Tokyo, Japan netic NPs, AuNPs and AgNPs, silica NPs, quantum dots

Korea Nano Technology Research Society 2017. This article is distributed under the terms of the Creative Commons Attribu-
tion 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the
Creative Commons license, and indicate if changes were made.
Nagamune Nano Convergence (2017) 4:9 Page 2 of 56

(QDs), polymeric micelles, liposomes, dendrimers, and 2Application ofengineered biological molecules
fullerenes) and biological molecules, which are highly tonanobio/bionanotechnology
useful for biosensing, bioimaging, diagnostic and thera- Nanobio/bionanotechnology has created new opportuni-
peutic applications in healthcare [915]. ties for advances in diverse fields, including life science,
On the other hand, bionanotechnology refers to the medicine, electronics, engineering, and biotechnology.
ways in which biotechnology is used to improve exist- Nanoscale materials [e.g., NPs, nanowires, nanofibers,
ing or create new nanotechnologies through the study and nanotubes (NTs)] combined with various engineered
of how biological systems work and the applications biological molecules (e.g., proteins, enzymes, oligonu-
of biological molecules and systems to nanotechnol- cleotides, polysaccharides, lipids, biological cofactors and
ogy. DNA and RNA nanotechnologies, the utiliza- ligands) have been explored in many biological applica-
tion of the base-pairing and molecular self-assembly tions (e.g., therapy, diagnosis, bioimaging, biosensing,
properties of nucleic acids to create useful materials, bioanalysis, biocatalysis, cell and organ chips, bioelec-
such as DNA origami, DNA nanomachines, DNA scaf- tronic devices, and biological separation) (Fig. 1). Their
folds for electronics, photonics and protein arrays, and novel and unique properties and functions, such as high
DNA and RNA aptamers, ribozymes and riboswitches, volume-to-surface ratio, improved solubility, quantum
are important examples of bionanotechnology [16, size, macroscopic quantum tunnel and multifunctional-
17]. Another important area of research involves tak- ity, result in nanobiomaterials that are drastically differ-
ing advantage of the self-assembly properties of pep- ent from their corresponding bulk materials.
tides, proteins and lipids to generate well-defined 3D The current review is focused on advances in the devel-
structures, functional protein complexes, nanofilms opment of nanobiomaterials for applications in ther-
and other nanostructures, such as micelles, reverse apy, diagnosis, biosensing, bioanalysis and biocatalysis
micelles and liposomes, which could be used as novel because nanobiomaterials for cell and organ chips [22
approaches for the large-scale production of program- 25], bioelectronic devices [26, 27] and biological separa-
mable nanomaterials [1820]. The application of car- tion [28] have recently been reviewed in this journal.
bohydrate polymers combined with nanotechnology
in tissue engineering and medicine are also potential 2.1Nanobiomaterials fortherapy anddiagnosis
research fields for the development of novel biomate- Smart therapeutic and diagnostic or bioimaging NPs car-
rials for biosensing, bioimaging, diagnostic and drug- rying cargo materials, such as drugs, DNAs, RNAs, pro-
delivery systems [21]. teins, and imaging reagents, have been widely developed
With either nanobiotechnology or bionanotech- [11, 13, 2933]. To achieve intracellular NP and drug
nology, biological molecules are indispensable build- delivery, many strategies for overcoming various bio-
ing blocks for fabricating functional nanomaterials, logical barriers are needed, including the following: (i)
nanodevices and nanosystems. However, from the preventing removal from the circulation by cells of the
viewpoint of applying biological materials to nanotech- reticuloendothelial system; (ii) targeting specific cells; (iii)
nology, biological materials found in nature always have
sufficient functions and properties. Recent advances in
biomolecular engineering, such as genetic engineer-
ing, DNA and RNA engineering, protein engineering,
site-specific chemical and enzymatic conjugation tech-
nologies, self-assembly technology and massive high-
throughput screening (HTS) methods, have enabled us
to improve, stabilize, integrate and alter the functions
and properties of biological materials. Thus, it is possi-
ble to create engineered biological materials with func-
tions and properties that are optimized for various uses
in the fields of bioelectronics, biosensors, biocatalysis,
molecular imaging, biological actuators, drug delivery
systems, biomaterials for tissue engineering and regen-
erative medicine.
In this review, recent studies applying engineered bio-
logical materials to nanobio/bionanotechnology are
discussed, and various biomolecular engineering tech-
nologies are highlighted. Fig.1 A summary of nanobiomaterials and their applications
Nagamune Nano Convergence (2017) 4:9 Page 3 of 56

internalization into cells; (iv) escaping from endosomes; growth factor, platelet-derived growth factor, and nuclear
(v) trafficking to specific organelles; and (vi) controlling factor kappa-light-chain-enhancer of activated B cells.
the release of payloads (e.g., drugs, DNAs or RNAs). Specific aptamers for targets can be selected from a large
number of random sequences (libraries of 1015 random
2.1.1Preventing removal fromthe circulation oligonucleotides) via the systematic evolution of ligands
NPs made of hydrophobic synthetic polymers, metals by exponential enrichment (SELEX) [39]. Aptamers
or inorganic materials are usually not blood compat- generally have less immunogenicity, which can lead to
ible. Their injection into the body can provoke a coagu- improved biodistribution in the human body. NP sur-
lation response and activate the complement cascade; faces can easily be conjugated with aptamers, and the
subsequently, they can be recognized by phagocytes and conjugates show efficient cancer cell targeting and inter-
macrophages, rendering them useless or harmful. The nalization [40]. Small molecules, peptides and aptamers
surface modification of NPs with hydrophilic synthetic are preferred for targeting and imaging ligands because
or biological polymers, such as polyethylene glycol (PEG) they can be simply conjugated to NPs via facile chemical
[34], heparin [35] or dextran [36], forms a steric brush conjugation methods.
that imparts resistance to protein adsorption. This type Transferrin (Tf ) is a monomeric glycoprotein that can
of surface modification shows increased intrinsic antico- transport iron atoms into cells. Upon the binding of Tf to
agulant and anti-complement properties, as well as other the Tf receptor (TfR), the Tf/TfR complex is internalized
biological activities; in addition, it extends the circulation by cells through receptor-mediated endocytosis. TfR has
half-life and reduces the immunogenicity of NPs in the been explored as a target for delivering anti-cancer drugs
human body. The conformation of polymer chains on the into cancer cells due to its overexpression by malignant
surface also influences the pharmacokinetics and biodis- tumor cells. TfR can be targeted by direct interaction
tribution of NPs. with Tf displayed on the surface of NPs [41].
Monoclonal IgG antibodies (mAbs) have been the pre-
2.1.2Targeting specific cells ferred targeting molecules for receptors, membrane pro-
The surface modification of NPs with biological ligands, teins and glyco-antigens on the surface of cancer cells.
such as folate, arginine-glycine-aspartate (RGD) pep- Because many breast cancer cells overexpress human
tides, aptamers, transferrin, antibodies or small antibody epidermal growth factor receptor-2 (HER-2), NPs coated
fragments, facilitates NP targeting, imaging and inter- with anti-HER-2 antibodies can target breast cancer cells
nalization into specific cells, e.g., cancer cells, and tumor with high specificity. Similarly, epidermal growth factor
tissues. receptor (EGFR) can be targeted by anti-EGFR antibod-
Folate is a well-known small molecule frequently used ies. Despite the immense efforts directed toward their
as a cancer cell-targeting ligand that binds to folate development, mAb-conjugated NPs still encounter many
receptors with high affinity. The chemical conjugation of challenges and limitations, such as the difficulty or cost
folate onto the surface of NPs can significantly promote of manufacturing, immunogenicity, and penetration into
their targeted delivery into cancer cells that overexpress tumor tissues, as mAbs are very large (150170 kDa,
folate receptors [37]. 1520 nm in diameter) and complex molecules. Alter-
Proliferating tumors are known to generate new blood natively, after proper engineering, small antibody frag-
vessels. This process is an important feature of tumor ments [e.g., antigen-binding fragment (Fab: 55 kDa)
development characterized by the unique overexpres- and variable fragment (Fv: 27 kDa)] can be used as
sion of the integrins 3 and 5 by nascent endothelial they can retain the targeting affinity and specificity of the
cells during angiogenesis in various tumors, but not by original whole antibody (Fig.2a). For example, the single-
ordinary endothelial cells. Peptides possessing the RGD chain variable fragment (scFv: 28kDa) that consists of
sequence bind the integrins 3 and 5 with high affin- variable heavy- and light-chain domains connected with
ity. Cyclic RGD peptides show higher affinity and stability a flexible peptide linker can be used to target cells with
than do linear RGD peptides, which allows their use for high binding affinity and specificity.
developing integrin-selective, targeting NPs [38]. Additionally, many alternative molecular scaffolds to
Aptamers are short, single-stranded RNA or DNA oli- mAbs have been investigated and developed in recent
gonucleotides (1540 bases) that can bind to target mol- years, largely by the pursuit of much smaller (<20 kDa)
ecules with high affinity and specificity due to the ability targeting molecules with their putatively superior trans-
of the molecules to fold into unique conformations with port properties (Fig. 2b) [42]. These scaffolds include
three-dimensional (3D) structures. A large number of affibodies (8 kDa) with three-helix bundles struc-
aptamers have been screened against aberrantly activated ture derived from the Z domain of protein A, DARPin
proteins in cancer cells, such as vascular endothelial with three or more repeated small domains (6 kDa)
Nagamune Nano Convergence (2017) 4:9 Page 4 of 56

pVEC, transportan, MPG, Pep-1 and polyarginines, could

facilitate the internalization of NPs into cells through
either direct entry into the cytosol or endosomal path-
ways. The Tat peptide, penetrain and pVEC are short
peptides (20-mers) derived from the basic domain of
the HIV-1 trans-activator of transcription (Tat) protein,
the third helix of the Antennapedia homeodomain and
cadherin, respectively. Transportan, MPG and Pep-1 are
chimeric peptides (30-mers) that are formed by the
fusion of two natural sequences derived from galanin/
mastoparan, HIV-gp41/SV40 T-antigen and HIV-reverse
transcriptase/SV40 T-antigen, respectively. These CPPs
mostly bear a net positive charge and consist of amino
acid (AA) sequences with repeated basic AA units and
hydrophobic or aromatic AAs. The repeated basic AA
units might contribute to not only the binding of CPPs
to the negatively charged cell surface but also the endoso-
mal escape of CPPs via conformational change under the
acidic pH conditions of late endosomes.

2.1.4Endosomal escape
The endosomal-escape ability of NPs is indispensa-
ble for the delivery of NPs into the cytosol and to orga-
nelles within the cell. Peptide-based endosomal-escape
agents have been developed, and these are derived from
the small-peptide domains of several viral, bacterial and
human sources [44]. For example, the HA2 subunit of the
Haemophilus influenzae hemagglutinin (HA) protein of
Fig.2 Targeting molecules. a IgG and its small fragments, b small the influenza virus with a short chain of an N-terminal
molecular-binding scaffolds anionic peptide has shown fusogenic activity. At a low
pH, the protonation of the glutamate (Glu) and the aspar-
tate (Asp) causes a conformational change of this peptide
consisting of two -helices separated by a -turn derived from a random coil into an amphiphilic -helical struc-
from ankyrin repeat proteins, and monobody with seven ture. This change allows the amphiphilic -helical peptide
-sheets forming a -sandwich and three exposed loops to bind to the endosomal membrane, causing membrane
from the 10th human fibronectin extracellular type III disruption. A pH-sensitive peptide GALA with repeat-
domain (10 kDa). These scaffolds are lacking disulfide ing glutamate-alanine-leucine-alanine (Glu-Ala-Leu-Ala)
bonds that make it possible to produce functional scaf- units could disturb the lipid bilayer by the same mecha-
folds regardless of the redox potential of the cellular nism and facilitate the endosomal escape of GALA-mod-
environment, including the reducing environment of ified NPs at acidic pH values. Arginin (Arg)-rich peptides
the cytoplasm and nucleus. Another scaffold is knottins and cationic peptides, also derived from viral proteins,
(3.5 kDa) comprising a family of exceptionally small could mimic the endosomal-disruptive properties of viral
and highly stable proteins found in many species with particles [45]. Several chemical polymers, such as poly-
structural homology involving a triple-disulfide stabilized ethylenimine- and imidazole-containing polymers, with
knot motif. The randomization of loops or surfaces in endosomal-disruptive properties have been reported.
conjunction with phage, ribosome or cell surface display These polymers have a buffering capacity ranging from
technologies is used to engineer these molecular scaf- pH 5.07.2 and can promote endosome osmotic swell-
folds and select binders to target molecules from many ing and disruption via the proton sponge effect [46].
random libraries. Recently, a conformation-switchable synthetic lipid
consisting of two alkyl chains on a di(methoxyphenyl)-
2.1.3Internalization intocells pyridine (pH-switchable unit) and a polar head group at
The surface modification of NPs with cell-penetrating the para position to the pyridine N atom was reported;
peptides (CPPs) [43], such as the Tat peptide, penetrain, upon protonation, hydrogen bonding induced a relative
Nagamune Nano Convergence (2017) 4:9 Page 5 of 56

orientation change of the two alkyl chains, which dis- bonding of the polymer molecules are favorable com-
turbed the lipid packing of the membranes and conferred pared to the solubilization of the polymers by water.
endosomal-escape properties [47]. Examples of thermo-sensitive polymers are poly(N-iso-
propyl acrylamide) (PNIPAAm), poly(N,N-diethyl acryla-
2.1.5Trafficking tospecific organelles mide) (PDEAAm), poly(methyl vinylether) (PMVE),
In eukaryotic cells, proteins are specifically sorted dur- poly(N-vinyl caprolactam) (PVCL), and poly(ethylene
ing or after translation and delivered from the cytosol to oxide)-poly(propylene oxide)-poly(ethylene oxide)
target organelles, such as the nucleus, endoplasmic retic- (PEO-PPO-PEO).
ulum, peroxisomes and mitochondria. These proteins In the case of polymerdrug conjugates, pH-sensitive
contain organelle-targeting peptide signals often found at linkages, such as oxime (pH < 5), hydrazone (pH < 5),
the N-terminal extension consisting of a short, positively hydrazide (pH < 5) and acetal (pH < 45), have been
charged stretch of basic AAs and a long -helical stretch used to directly attach drug molecules to polymers. The
of hydrophobic AAs [48, 49], and a database of protein use of light as a stimulus to trigger drug release has been
localization signals has been constructed based on exper- actively explored owing to its high spatiotemporal resolu-
imental protein localization [50]. Gene delivery systems tion. Photosensitivity is often introduced to NPs through
for the gene therapy of chromosomal and mitochondrial functional groups that can change their conformations
DNA have been developed by chemically conjugating and structures (e.g., azobenzene, pyrene, nitrobenzene
nuclear and mitochondrial targeting signal peptides to and spirobenzopyran groups) or break their chemical
NPs consisting of therapeutic DNAs [51]. bonds (e.g., arylcarbonylmethyl, nitroaryl, arylmethyl and
coumarin-4-ylmethyl groups) upon irradiation [54, 55].
2.1.6Controlling payload release Enzymes perform a vast array of important functions
In many cases, NPs in the endosomes or the cytoplasm inside our body. For example, hydrolytic enzymes over-
must collapse to allow the release of their payloads. Sev- expressed in cancer cells and tumor tissue can break
eral strategies using stimulus-responsive moieties built certain bonds (e.g., ester, amide, glucuronide and phos-
into NPs have been utilized to improve the efficiency of phodiester bonds) within biopolymers, causing polymer
controlled release [31]. These include pH-sensitive and structure disassembly or destruction. Notable examples
thermal-sensitive polymers, which control interactions of these enzymes are esterase, matrix metalloproteinase,
between payloads and NPs [52], and external stimulus- -glucuronidase and alkaline phosphatase. These enzy-
sensitive crosslinkers, which conjugate payloads with matic reactions can be utilized to trigger drug release
NPs [53], such as pH-labile linkers, photosensitive- and [56].
enzyme-cleavable linkers, and disulfide crosslinkers that
are sensitive to a reducing intracellular environment. 2.1.7Recent advances intargeted drug delivery
The difference in pH values existing between healthy andbioimaging
tissues (pH 7.4) and the extracellular environment of A major challenge of targeted drug delivery and bioim-
solid tumors (pH 6.56.8), as well as between the cytosol aging in therapeutics and diagnostics is the fabrication
(pH 7.4) and endosomes (pH 56), has been extensively of NPs modified with various functional biomolecules
utilized to trigger the release of drugs into a specific for overcoming the above-mentioned biological barriers
organ or intracellular compartment. Polymers with func- with a triggered cargo release system. Pluronic polymer-
tional groups that can alter the structure and hydropho- based micelles, to which folic acid (FA), redox-sensitive
bicity of NPs as a result of protonation or deprotonation thiol groups and the anti-cancer drug doxorubicin (DOX)
in response to pH variation can be utilized in pH-sensi- are chemically conjugated with pH-sensitive linkers,
tive polymeric NPs. Notable examples of pH-sensitive could be successfully delivered into multidrug-resistant
polymers include poly(acryl amide) (PAAm), poly(acrylic (MDR) tumors in mice and exerted high cytotoxicity
acid) (PAA), poly(methacrylic acid) (PMAA), poly(methyl in the DOX-resistant MDR tumors by bypassing MDR
acrylate) (PMA), poly(diethylaminoethyl methacrylate) efflux [57]. The carboxylate graphene oxide (GO)-based
(PDEAEMA), poly(diallyl dimethylammonium chloride) nanocarrier was multifunctionalized by poly(ethylene
(PDDA) and poly(dimethyl aminoethyl methacrylate) glycol) (PEG) terminated with an amino group and an
(PDMAEMA). FA group (FAPEGNH2) via the amidation reaction.
Temperature-sensitive polymers and hydrogels exhibit The GO-based nanocarrier could adsorb large amounts
a volume phase transition at a certain temperature, which of DOX on the GO surface via stacking interac-
causes a dramatic change in the hydration state. This tions at a neutral pH but release it at an acidic pH. The
phase transition reflects competing hydrogen-bonding DOX-loaded FAPEG-modified GO-based nanocarrier
properties, where intra- and intermolecular hydrogen not only showed stable dispersibility and targetability to
Nagamune Nano Convergence (2017) 4:9 Page 6 of 56

cancer cells with high FA receptor expression levels but oligodeoxynucleotide-modified AuNPs with a library of
also exhibited the low pH-activated controlled release of functional molecule-conjugated complementary pep-
DOX in the endosomes of cells [58]. tide nucleic acids (PNAs). The PNAs were functionalized
Nanohydrogels composed of filamentous bacterio- by conjugation with 1,4,7,10-tetraazacyclododecane-
phages and AuNPs, which were self-assembled via elec- 1,4,7,10-tetraacetic acid for chelating 64Cu for posi-
trostatic interactions between the phage-capsid proteins tron emission tomography imaging, PEG for conferring
and imidazole-modified AuNPs, have been developed stealth properties, and Cy5 for fluorescent imaging.
and utilized for noninvasive imaging and targeted drug These NPs demonstrated good stability invivo by show-
delivery in preclinical mouse models of breast and pros- ing biodistribution behavior in mice [60].
tate cancer. The phage-based nanohydrogels could be Recently, streptavidin (SA)-containing multifunction-
multifunctionalized by fusing peptides, e.g., tumor-tar- alized NPs for carrying various biotinylated functional
geting ligands and CPPs, to phage-capsid proteins and biomolecules have been reported. SA is a homo-tetramer
by incorporating temperature-sensitive liposomes or protein, and each subunit can tightly bind to biotin mol-
mesoporous silica NPs containing imaging reagents and ecule. We developed an SA-based cell-permeable nano-
drugs. Because AuNPs packed densely within the nano- carrier equipped with photosensitizers as a versatile
hydrogel, their surface plasmon resonance shifted to vehicle for spatiotemporally controlled cargo protein
the near-infrared (NIR) range, thereby allowing the NIR delivery into the cytosol (Fig. 3a) [61]. These nanocarri-
laser-mediated spatiotemporal photothermal release of ers can be prepared by attaching photosensitizer (Alexa
cargo from temperature-sensitive liposomes [59]. Mul- Fluor 546: AF546)-modified biotinylated CPPs (oligo-
tifunctionalized AuNPs are generally constructed by the arginine peptide R9 or R15) to a few biotin-binding sites
covalent assembly of an Au core with thiolated ligands. of SA. Furthermore, a biotinylated target cargo protein
Novel multifunctionalized AuNPs have been assembled is also loaded onto this carrier complex by using the
in one step by the nucleic acid hybridization of thiolated remaining biotin-binding site of SA. Conjugation with

Fig.3 Protein transduction using the streptavidin based nano-carrier. a Schematic illustration of protein transduction using the streptavidin based
nano-carrier. b (1) Effect of the conjugation ratio of R15 peptides to SA on the fluorescence intensity of HeLa cells after uptake of AF546-labeled
SAR15 complex. (2) Effects of the length of Rpep on the fluorescence intensity of HeLa cells after uptake of AF546-labeled R pep itself ant SARpep
complex (Figure reproduced with permission from: Ref. [61]. Copyright (2015) with permission from Elsevier)
Nagamune Nano Convergence (2017) 4:9 Page 7 of 56

more than three CPPs per SA significantly raised the cell- and biotinylated AF647. These P-Aggs multifunctional-
permeability of the SACPP complexes into HeLa cells ized with Tf, Alexa Fluor 647 and DOX were introduced
(Fig.3b). Under optimized conditions, the SACPP (R15) into human colon cancer cells by endocytosis via TfR, fol-
complex could be delivered into cells with both high effi- lowed by the selective release of DOX from the P-Aggs
ciency and low cytotoxicity. Furthermore, the internal- in light-irradiated cells, resulting in the spatiotemporal
ized AF546-modified SA complex could spatiotemporally induction of target cancer cell apoptosis (Fig.5).
escape from the endosome in a light-irradiated area. We also developed a method for preparing SA-immo-
Photolytic protein aggregates (P-Aggs) for light-con- bilized redox-sensitive nanohydrogels via peptide tag-
trollable nanocarriers have also been developed using induced disulfide formation mediated by horseradish
SA [62]. Submicron-scaled P-Aggs were constructed by peroxidase (HRP) (Fig. 6a) [63]. In this system, the pep-
mixing SA and cargo proteins labeled with a biotinylated tides with sequences of HHHHHHC (C-tag) and GGGGY
caging reagent (BCR) and were utilized as a facile and (Y-tag) were genetically fused to the N- and C-termini of
versatile platform for the light-induced release of cargo SA (C-SA-Y), respectively. Here, H, C, G and Y denote
proteins (Fig.4). The size of P-Aggs could be controlled histidine, cystein, glycine and tyrosine, respectively. The
either by adding an excess of biotin to the above mix- C-SA-Y was mixed with HRP- and thiol-functionalized
ture to stop the increase in P-Agg size or by conducting 4-arm PEG to yield a C-SA-Y-immobilized hydrogel
a mixing reaction in a water pool of reverse micelles and (C-SA-Y gel) crosslinked with redox-sensitive disulfide
adding biotinylated-PEG to stop the increase in P-Agg bonds. The C-SA-Y immobilized in the hydrogel retained
size. For example, P-Aggs were prepared by mixing SA, a its affinity for biotin, allowing the incorporation of any
BCR-caged transferrin-doxorubicin conjugate (Tf-DOX) biotinylated functional biomolecules or synthetic chemical

Fig.4 Schematic illustration of photolytic P-Aggs formation and light-induced release of active proteins. a The chemical structure of BCR 1 consist-
ing of a biotinylated photo-cleavable protection group (red) and an amino-reactive group (black). b Schemes of P-Aggs formation. c Protein photo-
liberation from P-Aggs (Figure reproduced with permission from: Ref. [62]. Copyright (2016) with permission from John Wiley and Sons)
Nagamune Nano Convergence (2017) 4:9 Page 8 of 56

2.2Nanobiomaterials forbiosensing andbioanalysis

Biosensing and bioanalysis based on new nanomateri-
als and nanotechnology in the areas of nanoelectron-
ics, nanooptics, nanopatterns and nanofabrication have
a wide range of promising applications in point-of-care
diagnostics, earlier disease diagnosis, pathological test-
ing, food testing, environmental monitoring, drug
discovery, genomics and proteomics. The rapid develop-
ment of nanotechnology has resulted in the successful
synthesis and characterization of a variety of nanomate-
rials, making them ideal candidates for signal generation
and transduction in sensing. In other words, the unique
properties and functionalization of biomaterial-conju-
gated nanostructures make them very useful for signal
amplification in assays, other biomolecular recognition
events and fabricating functional nanostructured bioint-
erfaces [64, 65]. Therefore, nanomaterials and nanofab-
rication technologies play significant roles in fabricating
biosensors and biodevices (e.g., colorimetric, fluores-
cent, electrochemical, surface-enhanced Raman scatter-
ing, localized surface plasmon resonance, quartz crystal
microbalance and magnetic resonance imaging (MRI)),
Fig.5 Light-induced cellular uptake of Tf or a chemotherapeutic including implantable devices [66] for the detection of a
drug through degradation of P-Aggs. ac Confocal microscopy broad range of biomarkers with ultrahigh sensitivity and
images of DLD1 cells treated with P-Aggs consisting of SA and selectivity and rapid responses.
AF647-labeled caged Tf before light irradiation. df Those after light
irradiation at 8Jcm2. a, d AF647-fluorescence images, b, e differen-
tial interference contrast (DIC) images, c, f each merged image of (a, 2.2.1Nanomaterials forenhancing sensitivity ofbiosensing
b) or (d, e), respectively. The scale bars are 50m. g Cell viabilities of andbioanalysis
the DLD1 cells treated with doxorubicin-modified Tf (Tf-DOX) or with During the last decade, several promising nanomateri-
P-Aggs consisting of SA and the caged Tf-DOX before and after light als (e.g., QDs, NPs, carbon nanotubes (CNTs) and gra-
irradiation at 8Jcm2 (Figure reproduced with permission from: Ref.
[62]. Copyright (2016) with permission from John Wiley and Sons)
phene) with biomolecule-modified surfaces have been
widely used in the fields of biosensing, bioanalysis and
diagnostics [6770]. For example, one of the first nano-
agents into the hydrogel via biotin-SA interaction. The materials to have an impact on amperometric biosen-
C-SA-Y gel was further prepared within a reverse micelle sors was CNTs, which have such advantages as a small
system to yield a nanosized hydrogel, rendering it a poten- size with a large surface area, an excellent electron trans-
tial drug delivery carrier. A C-SA-Y nanogel functionalized fer ability, and easy biomolecule immobilization. CNT-
with biotinylated CPP (biotin-G3R15GYC-Alexa Fluor modified electrodes improved current densities and
546) and Alexa Fluor 488-labeled saporin was prepared, enhanced the reactivity of biomolecules, redox cofactors
and we investigated its efficacy as a drug delivery system and redox enzymes. In addition, aligned CNT forests
for human colon adenocarcinoma DLD1 cells (Fig. 6a). facilitated direct electron transferwith the redox centers
The cell viability assay results revealed that the C-SA-Y of enzymes, resulting in improved overall performance of
nanogel prepared without CPP or saporin hardly affected enzyme electrodes and enzyme-labeled immunosensors
the viability of DLD1 cells. In contrast, treatment of the [71].
human colon cancer cells with the C-SA-Y gel function- Several nanomaterials have shown great promise in
alized with both CPP and saporin resulted in a marked imaging due to their intrinsic imaging characteristics,
decrease in cell proliferation (Fig.6b). These results indi- such as their brightness, sharp bandwidth and long-term
cated that the C-SA-Y nanogel had been internalized into stability (e.g., fluorescent agents, such as QDs [72], mag-
the cells via the CPP, reducing cell variability by cytotoxic- netic NPs in MRI [73] and colloidal AuNPs [74]). For
ity of saporin. The internalization ability of CPP and cyto- imaging, nanomaterials can be targeted to specific dis-
toxicity of saporin were therefore successfully integrated in ease sites within the body by conjugating the materials
the C-SA-Y nanogel, with both properties working coop- to biomarker-specific biomolecules. These biomaterial-
eratively to yield a cytotoxic C-SA-Y nanogel. based imaging agents can also provide information in
Nagamune Nano Convergence (2017) 4:9 Page 9 of 56

Fig.6 Peptide tag-induced HRP-mediated preparation of a streptavidin-immobilized redox-sensitive hydrogel. a Schematic illustration of HRP-
mediated preparation of a streptavidin-immobilized redox-sensitive hydrogel and intracellular delivery. b Cytotoxicity assay of DLD1 cells incubated
with C-SA-Y nanogel functionalized with CPP and saporin. The viability of cells without any treatment was set as 100%. Cells without any treatment
(1) and treated with C-SA-Y nanogel (2), C-SA-Y nanogel with CPP (3), C-SA-Y nanogel with saporin (4), C-SA-Y nanogel with CPP and saporin (5), and
saporin (6) (Figure reproduced with permission from Ref. [63]. Copyright (2016) with permission from American Chemical Society)

addition to anatomical data, e.g., information relating to immunoassay devices equipped with capture-antibody-
physiology and function, which enables more accurate decorated single-walled carbon nanotube (SWCNT) for-
and early disease diagnosis, such as the highly sensitive ests on pyrolytic graphite chips have been developed. The
detection of early-stage cancer [75]. [Ru(bpy)3]2+-doped silica NPs covered with thin hydro-
philic polymer films prepared by the sequential layer-by-
2.2.2Nanofabrication technologies forbiosensing layer deposition of positively charged PDDA and negatively
andbioanalysis charged PAA were used as ECL labels in these systems
Microarrays [76] and microfluidic [77, 78] platforms cou- for highly sensitive two-analyte detection. Antibodies
pled with biomolecule-conjugated nanomaterials (e.g., to prostate specific antigen (PSA) and interleukin (IL)-6
QDs, NPs, or CNTs conjugated with enzymes, antibodies, were chemically conjugated to either SWCNTs or poly-
DNAs, or aptamers) have enabled the simultaneous mul- mer-coated RuBPY-silica-Ab2 NPs via amidization with
tiplex detection of many disease biomarkers for cancer, 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydro-
infectious diseases, diabetes, cardiovascular diseases and chloride (EDC) and N-hydroxysulfosuccinimide (NHSS).
Alzheimers disease. For example, novel electrochemilumi- The microfluidic immunoassay device provided the simulta-
nescence (ECL) microwell array [79] and microfluidic [80] neous detection of the biomarker proteins PSA and IL-6 in
Nagamune Nano Convergence (2017) 4:9 Page 10 of 56

serum, demonstrating high sensitivity and detection limits bioconjugation chemistry through molecular recognition
in the low femtogram per milliliter range (1021 M range) for the enzymatic generation of covalent bonds (Fig. 8)
(Fig.7) [80]. These platforms explored the detection of ultra- [83]. These self-assembly and enzymatic assembly meth-
low concentrations of target biomarkers and have realized ods also provide mechanisms for construction over a
rapid, ultrasensitive and cost-effective bioassays requiring hierarchy of length scales. Bionanofabrication will enable
minimum sample volumes, which will enable primary care the effective interfacing of biomolecules with nanomate-
physicians and patients to perform assays in their respec- rials to create implantable devices.
tive settings, using so-called point-of-care diagnostics. The
detection of cancer biomarkers by immunoassays and sen- 2.3Nanobiomaterials forbiocatalysis
sors using these engineered nanomaterials could also enable The use of nanomaterials for enzyme immobilization
the diagnosis of cancer at very early stages [81, 82]. and stabilization is highly effective not only in stabilizing
Fabrication must employ strategies to control chem- the enzyme activity but also in developing other advan-
istry to ensure not only that patterns and structures are tageous properties, including high enzyme loading and
generated at the desired location and within an appro- activity, an improved electron transfer rate, low mass
priate time frame but also that undesired side reactions transfer resistance, high resistance to proteolytic diges-
are prevented. Bionanofabrication, the use of biological tion and the easy separation and reuse of biocatalysts by
materials and mechanisms for the construction of nan- magnetic force [84]. The immobilization or entrapment
odevices for biosensing and bioanalysis, offers conver- of enzymes on the surface or interior of nanocarriers has
gent approaches for building nanointerfaces between been accomplished using various nanomaterials, such as
biomolecules and devices by either enzymatic assembly polymer NPs (e.g., polylactic acid, polystyrene, polyvinyl
or self-assembly. For example, film-forming pH-sensitive alcohol, and chitosan), magnetic and superparamagnetic
chitosan directly assembles on electrodes under physi- NPs, polymer nanofibers (e.g., nylon, polyurethane, poly-
ological conditions in response to electrode-imposed carbonate, polyvinyl alcohol, polylactic acid, polystyrene,
voltages (i.e., electrodeposition). Through recombinant and carbon), CNTs, GO nanosheets, porous silica NPs,
technology, biomolecular engineering allows target pro- solgel NPs and viral NPs [8587].
teins to be endowed with peptide tags [e.g., a Glutamine
(Gln)-tag for transglutaminase-mediated crosslinking 2.3.1Enzyme immobilization
between the side chains of Gln and Lysine (Lys) residues] There are considerable advantages of effectively immo-
for assembly, which enables fabrication and controls bilizing enzymes for modifying nanomaterial surface

Fig.7 Design of microfluidic ECL array for cancer biomarker detection. (1) syringe pump, (2) injector valve, (3) switch valve to guide the sample to
the desired channel, (4) tubing for inlet, (5) outlet, (6) poly(methylmethacrylate) plate, (7) Pt counter wire, (8) Ag/AgCl reference wire, (9) polydi-
methylsiloxane channels, (10) pyrolytic graphite chip (black), surrounded by hydrophobic polymer (white) to make microwells. Bottoms of microw-
ells (red rectangles) contain primary antibody-decorated SWCNT forests, (11) ECL label containing RuBPY-silica nanoparticles with cognate secondary
antibodies are injected to the capture protein analytes previously bound to cognate primary antibodies. ECL is detected with a CCD camera (Figure
reproduced with permission from: Ref. [80]. Copyright (2013) with permission from Springer Nature)
Nagamune Nano Convergence (2017) 4:9 Page 11 of 56

Fig.8 Biofabrication for construction of nanodevices. Schematic of the procedure for orthogonal enzymatic assembly using tyrosinase to anchor
the gelatin tether to chitosan and microbial transglutaminase to conjugate target proteins to the tether (Figure adapted with permission from: Ref.
[83]. Copyright (2009) American Chemical Society)

properties and grafting desirable functional groups onto as dicyclohexylcarbodiimide/N-hydroxysuccinimide

their surface through chemical functionalization tech- (NHS) and EDC/NHS, have been commonly utilized
niques. The surface chemistry of a functionalized nano- for crosslinking. However, in some cases, enzymes dra-
material can affect its dispersibility and interactions with matically lose their activities because many conventional
enzymes, thus altering the catalytic activity of the immo- enzyme immobilization approaches, which rely on the
bilized enzyme in a significant manner. Toward this end, nonspecific absorption of enzymes to solid supports or
much effort has been exerted to develop strategies for the chemical coupling of reactive groups within enzymes,
immobilizing enzymes that remain functional and sta- have inherent difficulties, such as protein denaturation,
ble on nanomaterial surfaces; various methods includ- poor stability due to nonspecific absorption, variations in
ing, physical and/or chemical attachment, entrapment, the spatial distances between enzymes and between the
and crosslinking, have been employed [86, 88, 89]. In enzymes and the surface, decreases in conformational
certain cases, a combination of two physical and chemi- enzyme flexibility and the inability to control enzyme
cal immobilization methods has been employed for sta- orientation.
ble immobilization. For example, the enzyme can first To overcome these problems, many strategies for
be immobilized by physical adsorption onto nanomate- enzyme immobilization have been developed. One
rials followed by crosslinking to avoid enzyme leaching. approach is known as single-enzyme nanoparticles
Both glutaraldehyde and carbodiimide chemistry, such (SENs), in which an organicinorganic hybrid polymer
Nagamune Nano Convergence (2017) 4:9 Page 12 of 56

network less than a few nanometers in thickness is directly immobilized on a substrate surface with desired
built up from the surface of an enzyme. The synthesis orientations and without the need for substrate surface
of SENs involves three reactions: first, amino groups on modification or complicated conjugation processes. For
the enzyme surface react with acryloyl chloride to yield example, an Au-binding peptide was applied to direct the
surface vinyl groups; then, free-radicals initiate vinyl self-assembly of organophosphorus hydrolase onto an
polymerization from the enzyme surface using a vinyl AuNP-coated graphene chemosensor. This electrochemi-
monomer and pendant trimethoxy-silane groups; finally, cal biosensor system could detect pesticides with a fast
orthogonal polymerization occurs via silanol condensa- response time, low detection limit, better operating sta-
tion reactions to crosslink the attached polymer chains bility and high sensitivity [97].
into a network (Fig. 9). It was demonstrated that SENs The amphiphilic protein HFBI (7.5kDa), class II hydro-
can be immobilized in mesoporous silica; additionally, phobin, that is produced by Trichoderma reesei adheres
this method of immobilization was shown to provide to solid surfaces and exhibits self-organization at water
a much more stable immobilized enzyme system than solid interfaces. A fusion protein between HFBI and glu-
that of native enzymes immobilized by either adsorption cose oxidase (GOx-HFBI) with a 21-AA flexible linker
or covalent bonding in the same material [90]. Another (linker sequence: SGSVTSTSKTTATASKTSTST) was
approach is to introduce molecular interfaces between constructed. This fusion protein exhibited the highest
a solid surface and enzymes. Several methods based on levels of both protein adsorption and high GOx activ-
this approach have been reported, such as the surface ity owing to the presence of the HFBI spacer and flexible
modification of solid supports with hydrophilic synthetic linker, which forms a self-organized protein layer on solid
polymers [91, 92] and peptides [93] with specificities and surface and enables the GOx component in the fusion
affinities toward enzymes, and the fusion of enzymes protein to be highly mobile, respectively [95].
with peptide tags [94] or anchor proteins [95, 96]. The crystalline bacterial cell surface layer (S-layer)
Peptides with an affinity for nanomaterials have been proteins of prokaryotic organisms constitute a unique
identified from a combinatorial peptide library, and self-assembly system that can be employed as a pattern-
these peptides are promising tools for bottom-up fab- ing element for various biological molecules, e.g., gly-
rication technology in the field of bionanotechnol- cans, polysaccharides, nucleic acids, and lipids. One of
ogy. Through the use of these peptides, enzymes can be the most excellent properties of S-layer proteins is their

Fig.9 Illustration of armored single-enzyme nanoparticle. a Schematic of preparation of the single-enzyme nanoparticles. b Chemistry for the
synthesis of single-enzyme nanoparticles (Figure adapted with permission from Ref. [90]. Copyright (2003) American Chemical Society)
Nagamune Nano Convergence (2017) 4:9 Page 13 of 56

capability to self-assemble into monomolecular protein carboxylase, reaction intermediates are covalently
lattices on artificial surfaces (e.g., plastics, noble metals or attached to functional domains or subunits and trans-
silicon wafers) or on Langmuir lipid films or liposomes. ferred between domains or subunits. Alternatively, sub-
A fusion protein between the S-layer protein SbpA from strate channeling in such multienzyme complexes as
Bacillus sphaericus CCM 2177 and the enzyme lamina- metabolons, including by glycolysis, the Calvin and Krebs
rinase (LamA) from Pyrococcus furiosus fully retained cycles, tryptophan synthase, carbamoyl phosphate syn-
the self-assembly capability of the S-layer moiety, and the thetase, and dhurrin synthesis, is utilized to prevent the
catalytic domain of LamA was exposed at the outer sur- loss of low-abundance intermediates, to protect unsta-
face of the formed protein lattice. The enzyme activity of ble intermediates from interacting with solvents and to
the S-layer fusion protein monolayer on silicon wafers, increase the effective concentration of reactants. Addi-
glass slides and different types of polymer membranes tionally, scaffold proteins are involved in many enzymatic
was compared with that of only LamA immobilized cascades in signaling pathways (e.g., the MAPK scaffold
with conventional techniques. LamA aligned within the in the MAPK phosphorylation cascade pathway) and
S-layer fusion protein lattice catalyzed two-fold higher metabolic processes (e.g., cellulosomes from Clostrid
glucose release from the laminarin polysaccharide sub- ium thermocellum). From a practical point of view, there
strate compared with the randomly immobilized enzyme. are several obstacles for the genetic fusion of over three
Thus, S-layer proteins can be utilised as building blocks enzymes to construct multienzyme complexes. First,
and templates for generating functional nanostructures large recombinant fusion proteins are easily misfolded
at the meso- and macroscopic scales [98]. and subsequently are either proteolyzed or form inac-
tive inclusion bodies in E. coli. Furthermore, the opti-
2.3.2Multienzyme complex systems mum refolding conditions of each enzyme motif in fusion
In nature, the macromolecular organization of multi- proteins are not always identical. Last, rational design
enzyme complexes has important implications for the methods for peptide linkers between enzymes that ena-
specificity, controllability, and throughput of multi-step ble control or linker spatial arrangement and orientation
biochemical reaction cascades. This nanoscale macro- have not yet been developed [106]. Additionally, engi-
molecular organization has been shown to increase the neering the required interfacial interactions for efficient
local concentrations of enzymes and their substrates, to enzyme clustering is extremely challenging. Therefore,
enhance intermediate channeling between consecutive flexible post-translational methods using enzymatic site-
enzymes and to prevent competition with other intra- specific proteinprotein conjugation and synthetic scaf-
cellular metabolites. The immobilization of an artificial folds by employing orthogonal interaction domains for
multienzyme system on a nanomaterial to mimic natu- assembly have been particularly attractive because of the
ral multienzyme organization could lead to promising modular nature of biomolecular design [103].
biocatalysts. However, the above-mentioned immobili-
zation methods for one type of enzyme on nanomateri- 2.3.2.1Posttranslational enzymatic modificationbased
als cannot always be applied to multienzyme systems in multienzyme complexes Many proteins are subjected
a straightforward manner because it is very difficult to to post-translational enzymatic modifications in nature.
control the precise spatial placement and the molecular The natural post-translational processing of proteins is
ratio of each component of a multienzyme system using generally efficient and site-specific under physiological
these methods. Therefore, strategies have been developed conditions. Therefore, invitro and invivo enzymatic pro-
for the fabrication of multienzyme reaction systems [99, tein modifications have been developed for site-specific
100], such as genetic fusion [101], encapsulation [102] in proteinprotein conjugation. The applications of enzy-
reverse micelles, liposomes, nano/mesoporous silica or matic modifications are limited to recombinant proteins
porous polymersomes, scaffold-mediated co-localization harboring additional protein/peptide tags. However, pro-
[103], and scaffold-free, site-specific, chemical and enzy- tein assembly using enzymatic modifications (e.g., inteins,
matic conjugation [104, 105]. sortase A, and transglutaminase) is a promising method
In many organisms, complex enzyme architectures because it is achieved simply by mixing proteins without
are assembled either by simple genetic fusion or enzyme special techniques [106].
clustering, as in the case of metabolons, or by coopera- Recently, we demonstrated a covalently fused multien-
tive and spatial organization using biomolecular scaf- zyme complex with a branched structure using micro-
folds, and these enzyme structures enhance the overall bial transglutaminase (MTGase) from Streptomyces
biological pathway performance (Fig.10) [103, 106, 107]. mobaraensis, which catalyzes the formation of an -(-
In metabolons, such as nonribosomal peptide synthase, glutamyl) lysine isopeptide bond between the side chains
polyketide synthase, fatty acid synthase and acetyl-CoA of Gln and Lys residues. A cytochrome P450 enzyme
Nagamune Nano Convergence (2017) 4:9 Page 14 of 56

a b
E4 E1 E1 E3
E3 E2
E2 E2

E4
E1
E1 E3
E1
E3 E2 E2 E4

c d
E1 E2 E1 E2 E3 E4

E1 E2 E3
E1 E2 E3 E4

E1 E2 E3 E4

Fig.10 Illustration of different modes of organizing enzyme complexes. a Free enzymes, b metabolon (enzyme clusters), c fusion enzymes, d scaf-
folded enzymes

from Pseudomonas putida (P450cam) requires two solu- P450cam system were incorporated into the same aque-
ble redox proteins, putidaredoxin (PdX) and putidare- ous pool of micelles at a 1:1:1 ratio (Fig. 11b) and ena-
doxin reductase (PdR), to receive electrons from NADH bled both extremely high local protein concentrations
for its catalytic cycle, in which PdX reduced by PdR with and efficient electron transfer to P450cam, resulting in a
NADH activates P450cam. Therefore, it has been sug- reaction activity higher than that of a reverse micelle sys-
gested that the complex formation of P450cam with PdX tem composed of an equimolar mixture of PdR, PdX and
and PdR can enhance the electron transfer from PdR to P450cam (Fig.11c) [109].
PdX and from PdX to P450cam. This unique multien-
zyme complex with a branched structure that has never 2.3.2.2 Scaffold proteinbased multienzyme com
been obtained by genetic fusion showed a much higher plexes Scaffold proteins enable the precise spatial place-
activity than that of tandem linear fusion P450cam genet- ment of the components of a multienzymatic reaction
ically fused with PdX and PdR (Fig.11a) [108]. This mul- cascade at the nanometer scale. Scaffolds are involved
tienzyme complex with a branched structure was further in many enzymatic reaction cascades in signaling path-
applied to a reverse micelle system. When the solubil- ways and metabolic processes [110], and they can provide
ity of substrate is quite low in an aqueous solution, the advantages over reactions catalyzed by freely diffusing
reverse micelle system is often adopted for simple, one- enzymes by segregating reactions, increasing throughput
step enzymatic reactions because the substrate can be and providing modularity for the construction of novel
solubilized at a high concentration in an organic solvent, reaction networks. Recently, various multienzyme sys-
subsequently accelerating the reaction rate. In the case of tems have been developed using natural scaffold proteins
a multienzyme system, especially systems including elec- [111] and synthetic scaffolds [112] composed of elements
tron transfer processes, such as the P450cam system, the of natural scaffold proteins, such as cellulosomes [113]
reverse micelle system is difficult to apply because each and signal transduction scaffolds [114].
component is usually distributed into different micelles Proliferating cell nuclear antigen (PCNA) is a DNA-
and because the incorporation of all components into the sliding clamp that forms a symmetrical ring-shaped
same aqueous pool of micelles is very difficult. Unlike the structure encircling double-stranded DNA (dsDNA)
natural P450cam system, all components of the branched and acts as a scaffold for DNA-related enzymes, such as
Nagamune Nano Convergence (2017) 4:9 Page 15 of 56

b c

Fig.11 The branched fusion protein construction by MTGase-mediated site-specific protein conjugation. a A fusion protein of putidaredoxin
reductase (PdR) and P450cam linked with a peptide containing a reactive Gln residue and putidaredoxin attached K-tag generated a three-way
branched fusion protein by MTGase. b Reaction scheme for d-camphor hydroxylation by branched P450cam with cofactor regeneration in a
reversed micellar system. c Effect of W0 on the initial activities of branched P450cam (open circles) and an equimolar mixture of PdR, PdX and
P450cam (closed circles) (a adapted with permission from: Ref. [106]. Copyright (2012) Springer, b, c adapted with permission from Ref. [109]. Copy-
right (2010) Oxford University Press)

DNA polymerase and helicase. The archaeon Sulfolo and phosphite was slightly smaller than that of an equi-
bus solfataricus has three distinct PCNA genes with the molar mixture of PUPPET and PTDH (69.7 4.8 M).
three expressed PCNA proteins, PCNA1, PCNA2 and This result indicates that the oxidation of NADH by the
PCNA3, which form a heterotrimeric complex. These PdR domain in PTDH-PUPPET might increase the effec-
three PCNAs were fused to the three component pro- tive local concentration of N AD+ around the PTDH
teins (i.e., PdR, PdX, and P450cam) composing the P. domain and that this proximity effect on cofactor chan-
putida P450 system (Fig.12a). The resulting fusion pro- neling could potentially be improved by optimizing the
teins, PCNA1-PdR, PCNA2-PdX and PCNA3-P450cam, arrangement of PTDH and PdR on the PCNA scaffold.
completely retained the functions of the component pro- Designer cellulosomes containing four different
teins, including the heterotrimerization of the PCNAs, enzymes (two cellulases and two xylanases) from Ther
the catalytic activities of PdR and P450cam, and the elec- mobifida fusca have been reported, where four dock-
tron transfer function of PdX. The three fusion proteins erin-fused cellulolytic enzymes were incorporated into
immediately formed a heterotrimeric complex invitro by specific locations on an artificial, chimeric scaffold con-
mixing. Compared to an equimolar mixture of PdR, PdX taining four cohesins corresponding to each dockerin. As
and P450cam, the complex showed a 52-fold enhance- expected, compared to their free enzyme mixture system
ment in the monooxygenase activity of P450cam because without the chimeric scaffolding, the resulting multien-
of efficient electron transfer within the complex from zyme complexes exhibited enhanced activity (~2.4-fold)
PdR to PdX and from PdX to P450cam [111]. This system on wheat straw as a complex cellulosic substrate [116].
based on the PCNA scaffold was further extended to a Recently, Deuber et al. demonstrated in vivo multi-
phosphite-driven self-sufficient P450cam system invitro enzyme complex formation in E. coli cells via synthetic
by incorporating phosphite dehydrogenase (PTDH) for protein scaffold expression. Protein scaffolds with vari-
cofactor NADH regeneration (Fig. 12b) [115]. The Km ous arrangements of fusion domains were built from the
value of PTDH-incorporated PUPPET (PTDH-PUPPET) interaction domains of signaling proteins, the mouse SH3
for NAD+ (51.02.7M) in the presence of d-camphor and PDZ domains and the rat GTPase protein-binding
Nagamune Nano Convergence (2017) 4:9 Page 16 of 56

Fig.12 Schematic illustration of PCNA-mediated multienzyme complex formation. a Self-assembly of PCNA-based heterotrimeric complex
(PUPPET) consisting of P450cam, its electron transfer-related proteins PdR and PdX that catalyzes the hydroxylation of d-camphor. b PTDH-PUPPET
complex that catalyzes the hydroxylation of d-camphor by regenerating NADH with consumption of phosphite (a reproduced with permission
from: Ref. [111]. Copyright (2010) WileyVCH. b Reproduced with permission from: Ref. [115]. Copyright (2013) WileyVCH)

domain (GBD). The three enzymes acetoacetyl-CoA thi- makes it difficult to utilize this assembly method invivo.
olase, hydroxymethylglutaryl-CoA synthase and hydrox- Currently, there are several methodologies for conjugat-
ymethylglutaryl-CoA reductase, which catalyze a cascade ing proteins with DNA [117]. Proteins have been assem-
reaction from acetyl-CoA to mevalonate, were genetically bled onto DNA scaffolds through intervening adapter
tagged with their cognate peptidyl ligands. These protein molecules, such as biotinstreptavidin, NiNTA-hexa-
scaffolds and enzymes with peptidyl ligands were co- histidine, antibodies-haptens and aptamers. Alternatively,
expressed in E. coli cells. A significant 77-fold increase in direct covalent conjugation with DNA can be achieved by
mevalonate production was achieved by the expression of modifying cysteine (Cys) or Lys residues via disulfide or
the optimized scaffold: (GBD)1-(SH3)2-(PDZ)2 [114]. maleimide coupling, as well as by bioorthogonal chemis-
try, such as expressed protein ligation, Staudinger ligation
2.3.2.3 Oligonucleotide scaffoldbased multienzyme com and Huisgen cycloaddition.
plexes DNA has numerous attractive features as a scaf- By utilizing DNA nanostructures as assembly scaffolds,
fold for multienzyme complexes. Its properties, such as it has become feasible to organize the DNA-directed
high rigidity, programmability, complexity and assembly assembly of artificial multienzyme complexes. DNA-
through complementary hybridization, allow DNA to mediated assembly was employed to control the activity
form excellent scaffolds with linear, two-dimensional (2D) of a multidomain enzyme. Cytochrome P450 BM3 (P450
and 3D structures (e.g., simple dsDNA helices, Holliday BM3) is composed of two domains, a flavin adenine dinu-
junctions, DNA tiles, and DNA origami) for arranging cleotide and flavin mononucleotide-containing reductase
multiple enzymes with controlled spacing in linear, 2D domain (BMR) and a heme-containing monooxygenase
or 3D geometric patterns and for constructing interactive domain (BMP). P450 BM3 shows monooxygenase activ-
multienzyme complexes and networks [117120]. DNA ity by transferring electrons to BMP from NADPH
protein conjugates are necessary to achieve DNA-directed through BMR. Both subdomains were genetically fused
protein assembly for the fabrication of multienzyme com- to the HaloTag protein, a self-labeling enzyme, enabling
plexes on DNA scaffolds. However, this requirement bioconjugation with chloroalkane-modified DNAs and
Nagamune Nano Convergence (2017) 4:9 Page 17 of 56

subsequently reconstituting BM3 activity by DNA-medi- 2O2 along the hydration

in the restricted diffusion of H
ated assembly. The arrangement of the two domains on layer of the contacted protein surfaces and enhancing the
a DNA scaffold can control the distance between them. enzyme cascade reaction activity (Fig.13d, e) [123].
The distance-dependent activity of multidomain P450 An enzyme cascade nanoreactor was constructed by
BM3 complexes was investigated by varying the length of coupling GOx and HRP using both a planar rectangular
spacing scaffolds between the BMR and BMP domains. orientation and short DNA origami NTs. Biotinylated
The resulting changes in distance between the redox GOx and HRP were positioned on the streptavidin-
centers of the two domains regulated the efficiency of decorated planar rectangular DNA sheet via the biotin
electron transfer and thus the enzymatic activity of the avidin interaction with a specific interenzyme distance
reconstituted P450 BM3 [121]. (i.e., the distance between GOx and HRP) of 15nm. This
2D DNA nanostructures provide an even greater DNA sheet equipped with GOx and HRP was then rolled
opportunity to organize multienzyme systems into more into a confined NT, resulting in the encapsulation of the
complicated geometric patterns. Thiolated nucleic acids enzymes in a nanoreactor. Remarkably, the enzymatic
were covalently linked to glucose oxidase (GOx) and coupling efficiency of this enzyme cascade within short
horseradish peroxidase (HRP) by using N-[(1-maleim- DNA NTs was significantly higher than that on the pla-
idocapropyloxy)sulphosuccinimide ester] as a bifunc- nar rectangular DNA sheet alone. When both enzymes
tional crosslinker. The GOx/HRP enzyme cascade was were confined within the DNA NTs, H2O2 could not dif-
organized on 2D hexagonal DNA strips via self-assem- fuse out of the diffusion layer, which was much thicker
bly. The distance between two enzymes was controlled than the diameter of the DNA NTs (20nm), resulting in a
by varying the positions of two free DNA tethers on the high coupling of the reaction intermediate H2O2 between
hexagonal DNA strips. The complementary DNA-con- the enzymes [124].
jugated enzymes organized on the two-hexagon strips A similar modular type of enzyme cascade nanoreactor
(shorter distances) showed 1.2-fold higher activity than was constructed using 3D DNA origami building blocks.
the four-hexagon strips. With shorter distances, interme- Each of the DNA origami units contained three biotin-
diate (H2O2) diffusion was more efficient, which there- conjugated strands protruding from the inner surface of
fore resulted in increased cascade reaction efficiency. the tubular structure. The deglycosylated avidin and NTV
However, the enzyme cascade was not activated in the were immobilized on the inner surface of the units via the
absence of the DNA scaffolds or in the presence of for- biotinavidin interaction to facilitate the further binding
eign DNA [122]. These observations indicate that spatial of biotinylated enzymes. Biotinylated GOx and HRP were
arrangement at the nanometer scale using a 2D nano- anchored inside the origami compartment with the help
structure comprising a rigid DNA duplex could control of NTV. The resulting GOx- and HRP-immobilized tubu-
the flux of an intermediate from a primary enzyme to a lar DNA origami structures were connected together by
secondary enzyme and that the flux control dominated hybridizing 32 short (36 bases) sequences. The GOx/
the multienzyme cascade reaction rate. HRP cascade reaction of the assembled dimer nanoreac-
More accurate distance control of the GOx/HRP tor showed significantly higher activity than that without
enzyme cascade was realized using DNA origami tiles as a DNA scaffold [125].
a scaffold. The distance between enzymes was systemati- Engineered RNA modules were assembled into dis-
cally varied from 1065nm, and the corresponding activ- crete (0D), one-dimensional (1D) and 2D scaffolds with
ities were evaluated. The study revealed the existence distinct protein-docking sites (duplexes with aptamer
of two different distance-dependent kinetic processes sites) and used to control the spatial organization of a
associated with the assembled enzyme pairs. Strongly hydrogen-producing pathway in bacteria. The 0D, 1D
enhanced activity was observed when the enzymes and 2D RNA scaffolds were assembled in vivo through
were closely spaced, while the activity decreased drasti- the incorporation of two orthogonal aptamers for cap-
cally for enzymes as little as 20nm apart. Increasing the turing the target phage-coat proteins MS2 and PP7. Cells
spacing further showed much weaker distance depend- expressing the designed RNA scaffold modules and both
ence (Fig. 13ac). This study revealed that intermediate ferredoxin/MS2 (FM) and [FeFe]-hydrogenase/PP7 (HP)
transfer between enzymes might occur at the connected fusion proteins showed remarkable increases in hydrogen
hydration shells for closely spaced enzymes. This mecha- production. Namely, 4-, 11- and 48-fold enhancements in
nism was verified by constructing different sizes of non- hydrogen production compared with that of control cells
catalytic protein bridges (-galactosidase (-Gal) and were observed from the RNA-templated hydrogenase
NeutrAvidin (NTV)) between GOx and HRP to facilitate and ferredoxin cascade reactions in cells expressing 0D,
intermediate transfer across protein surfaces. The bridg- 1D and 2D RNA scaffolds, respectively. This study sug-
ing protein changed the Brownian diffusion, resulting gests that a metabolic engineering approach can be used
Nagamune Nano Convergence (2017) 4:9 Page 18 of 56

Fig.13 Schematic illustration of interenzyme substrate diffusion for an enzyme cascade organized on spatially addressable DNA nanostructures.
a DNA nanostructure-directed coassembly of GOx and HRP enzymes with control over interenzyme distances and details of the GOx/HRP enzyme
cascade. b Spacing distance-dependent effect of assembled GOx/HRP pairs as illustrated by plots of product concentration (Absorbance of A BTS)
vs time for various nanostructured and free enzyme samples. c Enhancment of the activity of the enzyme pairs on DNA nanostructures compared
to free enzyme in solution. d The design of an assembled GOx/HRP pair with a protein bridge used to connect the hydration surfaces of GOx and
HRP. e Enhancement in the activity of assembled GOx/HRP pairs with -Gal and NTV bridges compared to unbridged GOx/HRP pairs (Figure repro-
duced with permission from: Ref. [123]. Copyright (2012) American Chemical Society)

to introduce structural nucleic acid nanostructures inside the basic building blocks of biological systems, and there
cells for the organization of multienzyme reaction path- are many new advantages available to nanotechnology by
ways [126]. manipulating their structures, functions and properties.
Since every biomolecule is different, there are a number
3Biomolecular engineering fornanobio/ of technologies used to manipulate each one individually.
bionanotechnology Biomolecules have various outstanding functions,
Biomolecular engineering addresses the manipulation such as molecular recognition, molecular binding, self-
of many biomolecules, such as nucleic acids, peptides, assembly, catalysis, molecular transport, signal transduc-
proteins, carbohydrates, and lipids. These molecules are tion, energy transfer, electron transfer, and luminescence.
Nagamune Nano Convergence (2017) 4:9 Page 19 of 56

These functions of biomolecules, especially nucleic acids and simple one-pot 2D DNA origami method named
and proteins, can be manipulated by nucleic acid (DNA/ scaffolded DNA origami, which involves the folding of
RNA) engineering, gene engineering, protein engineer- a long single strand of viral DNA into a DNA scaffold
ing, chemical and enzymatic conjugation technologies of a desired shape, such as a square, rectangle, triangle,
and linker engineering. Subsequently, engineered bio- five-pointed star, and even a smiley face using multi-
molecules can be applied to various fields, such as ther- ple short staple strands [130]. To fabricate and stabilize
apy, diagnosis, biosensing, bioanalysis, bioimaging, and various shapes of DNA tiles, crossover motifs have been
biocatalysis (Fig.14). designed through the reciprocal exchange of DNA back-
bones. Branched DNA tiles have also been constructed
3.1Nucleic acid engineering using sticky ends and crossover junction motifs, such as
Nucleic acids, such as DNA and RNA, exhibit a wide tensegrity triangles (rigid structures in a periodic-array
range of biochemical functions, including the storage and form) and algorithmic self-assembled Sierpinski triangles
transfer of genetic information, the regulation of gene (a fractal with the overall shape of an equilateral triangle).
expression, molecular recognition and catalysis. Nucleic These DNA tiles can further self-assemble into NTs, helix
acid engineering based on the base-pairing and self- bundles and complex DNA motifs and arrays [17]. 3D
assembly characteristics of nucleic acids is key for DNA/ DNA origami structures can be designed by extending
RNA nanotechnologies, such as those involving DNA/ the 2D DNA origami system, e.g., by bundling dsDNAs,
RNA origami, aptamers, and ribozymes [16, 17, 127]. where the relative positioning of adjacent dsDNAs is con-
trolled by crossovers or by folding 2D origami domains
3.1.1DNA/RNA origami into 3D structures using interconnection strands [131].
DNA/RNA origami, a new programmed nucleic acid 3D DNA networks with such topologies as cubes, poly-
assembly system, uses the nature of nucleic acid comple- hedrons, prisms and buckyballs have also been fabricated
mentarity (i.e., the specificity of WatsonCrick base pair- using a minimal set of DNA strands based on junction
ing) for the construction of nanostructures by means of flexibility and edge rigidity [17].
the intermolecular interactions of DNA/RNA strands. 2D Because the folding properties of RNA and DNA are
and 3D DNA/RNA nanostructures with a wide variety of not exactly the same, the assembly of RNA was gener-
shapes and defined sizes have been created with precise ally developed under a slightly different perspective due
control over their geometries, periodicities and topolo- to the secondary interactions in an RNA strand. For this
gies [16, 128, 129]. Rothemund developed a versatile reason, RNA tectonics based on tertiary interactions

Fig.14 Overview of biomolecular engineering for enhancing, altering and multiplexing functions of biomolecules, and its application to various
fields
Nagamune Nano Convergence (2017) 4:9 Page 20 of 56

have been introduced for the self-assembly of RNA. In Synthetic DNA pool
particular, hairpinhairpin or hairpinreceptor inter-
actions have been widely used to construct RNA struc- T7 RNA Constant
polymerase sequence
Random
sequence
Constant
sequence
tures [16]. However, the fundamental principles of DNA promoter
sequence
PCR
origami are applicable to RNA origami. For example, the
Clone PCR
use of three- and four-way junctions to build new and Aptamers ds-DNA pool cDNA
diverse RNA architectures is very similar to the branch- Transcribe Reverse
transcribe
ing approaches used for DNA. Both RNA and DNA can
form jigsaw puzzles and be developed into bundles [17]. RNA Enriched
Binding RNA
One of the most important features of DNA/RNA ori- selection
Activity
gami is that each individual position of the 2D structure selection

contains different sequence information. This means that Fig.15 The general procedure for the invitro selection of aptamers
or ribozymes
the functional molecules and particles that are attached
to the staple strands can be placed at desired positions
on the 2D structure. For example, NPs, proteins or dyes
were selectively positioned on 2D structures with pre- transcribed, amplified by PCR, and transcribed; then, the
cise control by conjugating ligands and aptamers to the entire cycle is repeated. After multiple rounds of selec-
staple strands. These DNA/RNA origami scaffolds could tion (generally 618 rounds), quite large populations
be applied to selective biomolecular functionalization, (>1013 different sequences) can be sieved, the ratio of
single-molecule imaging, DNA nanorobot, and molecu- active-to-inactive RNA sequences increases and finally
lar machine design [131]. The potential use of DNA/RNA the pool becomes dominated by molecules that can bind
nanostructures as scaffolds for X-ray crystallography and the target molecule.
nanomaterials for nanomechanical devices, biosensors, Chemically modified nucleotides provide several
biomimetic systems for energy transfer and photonics, advantages, such as enhanced nuclease resistance, an
and clinical diagnostics and therapeutics have been thor- improved binding affinity, increased oligonucleotide pool
oughly reviewed elsewhere [16, 17, 127129]; readers are diversity and improved success rate of selection. There-
referred to these studies for more detailed information. fore a modified oligonucleotide pool is becoming more
popular for aptamer selection. Although chemically
3.1.2Aptamers modified nucleotides and deoxynucleotide triphosphates
Aptamers are single-stranded nucleic acids (RNA, DNA, cannot be recognized by wild-type T7 RNA polymerases
and modified RNA or DNA) that bind to their targets and A-type DNA polymerases, such as Taq polymerase,
with high selectivity and affinity because of their 3D fortunately, modified nucleotide triphosphates (2-fluoro
shape. They are isolated from 1012 to 1015 combinato- pyrimidines, 2-O-methyl nucleotides) and functional-
rial oligonucleotide libraries chemically synthesized by ized 2-deoxynucleotide triphosphates with AA-like
invitro selection [132]. Many protocols, including high- residues (e.g., indole, benzyl, or alkyne moieties) can be
throughput next-generation sequencing and bioinfor- recognized by some mutant RNA polymerases [137] and
matics for the in vitro selection of aptamers, have been B-type polymerases, and Pwo and Vent (exo-) DNA poly-
developed and have demonstrated the capacity of aptam- merases [138], respectively.
ers to bind to a wide variety of target molecules, rang- Specific aptamers against diverse targets have been
ing from small metal ions, organic molecules, drugs, and developed and aptamer-conjugated nanomaterials such
peptides to large proteins and even complex cells or tis- as drug-encapsulated polymer NPs, CNTs, AuNPs, QDs
sues [39, 133136]. The general in vitro selection pro- and DNA origami demonstrated potential in applications
cedure for an aptamer, SELEX (Fig. 15), is as follows: a ranging from therapy, targeted drug delivery, sensors and
synthetic DNA pool is prepared by chemical synthesis. diagnostic reagents to aptamer-directed protein arrays
DNAs consist of a random or mutagenized sequence on DNA nanostructures. The details of such applications
region flanked on each end by a constant sequence and will not be covered in this review; readers are referred to
with a T7 RNA polymerase promoter at the 5 end. This several recently published reviews [2931, 40, 64, 68, 132,
DNA is amplified by a few cycles of polymerase chain 139, 140].
reaction (PCR) and subsequently transcribed in vitro to
make the RNA pool. The RNA molecules are then selcted 3.1.3Ribozymes
based on their binding affinity to the target molecule, for Natural ribozymes are RNA molecules that have enzy-
example, by passing them through a target-immobilized matic activity for cleaving phosphodiester linkages.
affinity column. The retained RNAs are eluted, reverse Therefore, ribozymes have significant potential for use
Nagamune Nano Convergence (2017) 4:9 Page 21 of 56

in cancer, genetic disease, and viral therapeutics by spe- well as off-target activity, resistance to serum and cellu-
cifically inhibiting gene expression through cleaving RNA lar nucleases, and cell-specific, targeted delivery, need to
substrates, such as mRNA, with the viral genome of RNA be addressed and overcome in order to utilize ribozymes
containing a sequence complementary to the catalytic in therapies. Ribozymes can be hardly incorporated into
center of the ribozymes [141]. cells in their naked forms and often required a vehicle for
Natural ribozymes bind to substrate RNAs through efficient delivery. Many classes of nanomaterials includ-
WatsonCrick base pairing, which offers the sequence- ing cationic liposomes, cationic polymer micelles [147]
specific cleavage of substrate RNAs. Two ribozymes, the and spherical nucleic acids composed of inorganic core
hammerhead ribozyme and the hairpin ribozyme, have and densely packed, highly oriented nucleic acid shell
been extensively studied [142]. The catalytic motif of a [148] have been used as delivery vehicles to prevent
ribozyme is surrounded by a flanking sequence that is nuclease-dependent degradation and to enhance cell-tar-
responsible for guiding the ribozyme to its target RNA geting and intracellular transduction [143].
and giving stability to the structure. With the hammer-
head ribozyme, cleavage is dependent on divalent metal 3.2Gene engineering
ions, such as magnesium, and can occur after any NUH Gene engineering is a powerful tool for creating artificial
triplet (where N = any nucleotide and H = A, C or U) genes for proteins and enzymes with desired, improved
within the target RNA sequence. The kinetics of the reac- and multiple properties such as molecular recognition,
tion can vary significantly (up to one or more orders of molecular binding, self-assembly, catalysis, molecular
magnitude) with different triplet-flanking sequence com- transport, signal transduction, energy transfer, electron
binations; thus, the choice of an appropriate ribozyme transfer, and luminescence, which contribute to develop
cleavage site is the first and most important step in ham- novel nanobiomaterials, nanobiodevices and nanobio-
merhead ribozyme design [143]. systems. This technology has been employed to evolve
Artificial ribozymes with catalytic properties have been genes in vitro through an iterative process consisting of
isolated by invitro selection from random or combinato- recombinant generation. Coupled with the powerful HTS
rial nucleic acid libraries. Variations of the aptamer selec- or selection methods, gene engineering has been widely
tion strategies can be used to isolate catalytic nucleic acid applied to solve problems in protein engineering. This
sequences by changing the binding selection step of the technology includes technologies for direct gene manipu-
aptamer selection process to an activity selection step lation, such as gene mutagenesis, DNA sequence amplifi-
(Fig.15). Such approaches have been used to change the cation [e.g., PCR and rolling circle amplification (RCA)],
function of known ribozymes and to create completely DNA shuffling and gene fusion.
new ones from a random or combinatorial nucleic acid There are many methods to generate genetic diver-
pool [144]. A broad range of chemical reactions could sity and to create combinatorial libraries. For example,
be catalyzed, such as the formation, cleavage and rear- a variety of in vitro gene manipulation techniques have
rangement of various types of covalent bonds. Examples been developed over the past decade that allow vari-
including not only the cleavage or ligation of RNA sub- ous types of directed changes in a gene by modifying
strates by phosphoester transfer at the phosphorus center (inserting, deleting or replacing) one or more codons
[144, 145] but also DielsAlder reactions, N-glycosidic (gene mutagenesis), swapping domains between related
bond formation, alkylations, acylations, and amide bond functional gene sequences (DNA shuffling) and fus-
formations at the carbon centers [144, 146] have been ing domains from different functional gene sequences
reviewed. The catalytic performance, nuclease resistance (gene fusion), resulting in the creation of diverse collec-
and diversity of the oligonucleotide pools of ribozymes tions of mutant gene clones. There are two main types of
could also be enhanced by the incorporation of chemi- mutagenesis, i.e., random and site-directed mutagenesis.
cally modified nucleotides, as utilized in aptamer selec-
tion protocols [146]. 3.2.1Random mutagenesis
Ribozymes can be expressed from a vector, which With random mutagenesis, point mutations are intro-
offers the advantage of the continued intracellular pro- duced at random positions in a gene of interest, typically
duction of these molecules. However, the turnover rates through error-prone PCR mutagenesis, in which MnCl2
of ribozymes are rather low in some cases, since disso- is added to the reaction mixture to cause a reduction in
ciation from the cleavage product is the rate-limiting the fidelity of the DNA amplification [149]. The modi-
step that controls their usefulness. Furthermore, some fied error-prone PCR method, which achieves higher
ribozymes require high divalent metal ion concentrations frequencies of base substitutions and both transition
for efficient substrate cleavage, which may limit their use and transversion mutations, was developed using mix-
in intracellular environments. All of these concerns, as tures of triphosphate derivatives of nucleoside analogs
Nagamune Nano Convergence (2017) 4:9 Page 22 of 56

[150, 151]. An error-prone RCA method, which is an iso- DNA-shuffling method can be used; this method involves
thermal DNA amplification method with the addition of the fragmentation of the parental genes using restric-
MnCl2 to the reaction mixture, was also developed for tion enzymes rather than DNase I [156] or uses single-
random mutagenesis [152]. Different in vitro chemical stranded DNA (ssDNA) templates rather than dsDNA
mutagenesis methods have also been used to introduce templates for DNase I fragmentation [157]. Since the
random mutations into a gene of interest. In these meth- use of ssDNA as templates will decrease the probability
ods, bases of DNA are modified by chemical mutagens, of homo-duplex formation, the percentage of the paren-
such as nitrous acid, bisulfate, hydroxylamine and ethyl tal genes in the shuffled library should be significantly
methane sulfonate, and these methods have less bias reduced.
than does mutagenesis using PCR-based methods [153]. DNA shuffling has been extended to distantly or com-
Randomized sequences are then cloned into a suitable pletely unrelated gene families, which require methods
expression vector, and the resulting mutant libraries can that do not rely on homologous recombination because
be screened to identify mutants with altered or improved of the degree of sequence divergence. Sequence homol-
properties. ogy-independent protein recombination [158] and incre-
mental truncation for the creation of hybrid enzymes
3.2.2Sitedirected mutagenesis lead to the formation of chimeric genes (Fig.16b) [159].
Site-directed mutagenesis is a method for altering a gene The rearrangement of these chimeras by shuffling yields
sequence at a selected location by using overlapping functional hybrids [160]. The main advantage of these
extension PCR. Point mutations, insertions, or deletions methods is that knowledge about detailed protein struc-
are introduced by incorporating DNA primers contain- ture is not required [161].
ing the desired modification with a DNA polymerase in Exon shuffling is a natural molecular mechanism for
an amplification reaction. Site-saturation mutagenesis the formation of new eukaryotic genes. New exon com-
further allows the substitution of predetermined protein binations can be generated by recombination within the
sites against all twenty possible AAs at once by employ- intervening intron sequences, yielding new rearranged
ing degenerate primers in which the three bases of the genes with altered functions. The natural process of exon
targeted codon are replaced by mixtures, most com- shuffling can be mimicked in vitro by generating librar-
monly NNN or NNK (N=A, C, G or T; K=G or T). A ies of exon-shuffled genes and subsequently screening
completely randomized codon, NNN, results in a library target DNA from libraries [162]. In this method, exons
size of 64 different sequences encoding all 20 AAs and or combinations of exons that encode protein domains
3 stop codons. On the other hand, NNK codons reduce are amplified by PCR using mixtures of chimeric oli-
the library size by half, still encoding 20 AAs, with the gonucleotides that determine which exons are spliced
advantage of having only one stop codon. In this config- together. By means of a self-priming overlap polymerase
uration, the AAs W, F, I, Y, Q, N, H, K, D, E, M and C are reaction, mixtures of these PCR fragments are combi-
encoded by a single codon, while G, A, V, P, and T, and L, natorially assembled into full-length genes. Recombina-
S, and R are encoded by two and three codons, respec- tion is performed by connecting an exon from one gene
tively [154]. to an exon from a different gene. In this way, two or
more exons from different genes can be combined
3.2.3DNA shuffling togetherectopically, or the same exon can be duplicated,
DNA shuffling is a method for the in vitro recom- to create a new exonintron structure.
bination of homologous genes to quickly generate a
large library of chimeric progeny genes incorporating 3.2.4Gene fusion
sequence fragments from a number of parent genes by Fusion genes are created by genetically fusing the open
random fragmentation though DNase I and PCR exten- reading frames of two or more genes in-frame through
sion without primers for reassembly; this process is fol- ligation or overlap extension PCR. To construct such
lowed by PCR amplification with primers to generate fusion genes, two types of connection are possible. One
full-length chimeras suitable for cloning into an expres- is end-to-end fusion, in which the 5 end of one gene is
sion vector (Fig.16a) [155]. One significant drawback of linked to the 3 end of the other gene. The second is inser-
this DNA-shuffling method is the low frequency of chi- tional fusion, in which one gene is inserted in-frame into
meric genes in the shuffled library, which may be due to the middle of the other parent gene [163]. These methods
the homo-duplex formation of DNA fragments derived offer various advantages for producing fusion genes with
from the same parental genes at the annealing step, the high throughput in different orientations and including
probability of which is much higher than that of hetero- linker sequences to maximize the performance of fusion
duplex formation. To address this problem, a modified partners [164].
Nagamune Nano Convergence (2017) 4:9 Page 23 of 56

Fig.16 Illustrations of genetic recombination methods for protein evolution. a DNA shuffling (in vitro recombination of homologous genes). b
ITCHY (in vitro recombination of homology-independent genes) (Figure adapted from Ref. [172])

3.3Protein engineering [161]. However, the major drawback of protein redesign

The field of protein engineering has always played a is that detailed structural knowledge of a protein is often
central role in biological science, biomedical research, unavailable, and, even when it is available, substitutions
and biotechnology. Protein engineering is also indis- at sites buried inside proteins are more likely to break
pensable technology to design useful and valuable their structures and functions. Therefore, it is still very
building blocks for nanobio/bionanotechnology to fab- difficult to predict the effects of various mutations on
ricate a variety of artificial self-assembled protein sys- the structural and functional properties of the mutated
tems with nanoscale structures [18, 19], proteins with protein, although many studies have been done to pre-
tagged peptides for immobilization on NPs [94] and dict the effects of AA substitutions on protein functions
engineered proteins for applications to bioelectronic [168]. Another rational protein design method is compu-
devices [23, 26, 27], therapy [42, 44, 45, 67, 165], bio- tational protein design, which aims to design new protein
imaging [67, 166], biosensing [83, 97, 167], and bioca- molecules with a target folding protein structure, novel
talysis [87, 89, 95, 98, 101, 103, 108, 110116]. There function and/or behavior. In this approach, proteins can
are two general strategies for protein engineering, i.e., be designed by transcendentally setting AA sequences
rational protein design and directed evolution (high- compatible with existing or postulated template back-
throughput library screening- or selection-based bone structures (de novo design) or by making calculated
approaches) (Fig.17). variations to a known protein structure and its sequence
(protein redesign) [169]. Rational protein design
3.3.1Rational protein design approaches make predicted AA sequences of protein that
In rational protein design (Fig.17, the left panel), detailed will fold into specific 3D structures. Subsequently, these
knowledge of the structure and function of a protein is predicted sequences should be validated experimentally
used to make desired changes to the protein. In general, through the chemical synthesis of an artificial gene, fol-
this approach has the advantage of creating functionally lowed by protein expression and purification. The details
improved proteins easily and inexpensively, since site- of computational protein design methods will not be
directed mutagenesis techniques allow precise changes covered in this review; readers are referred to several
in AA sequences, loops and even domains in proteins recently published reviews [170, 171].
Nagamune Nano Convergence (2017) 4:9 Page 24 of 56

Fig.17 Two general strategies and their procedures for protein engineering

3.3.2Directed evolution (protein engineering based diversity-generation methods and HTS/selection meth-
onhighthroughput library screening or selection) ods. The key technology of HTS/selection methods is
The directed evolution approach (Fig.17, the right panel) the linkage of the genotype (the nucleic acid that can be
involves many technologies, such as gene library diver- replicated) and the phenotype (the functional trait, such
sification, genotypephenotype linkage technologies, as binding or catalytic activity). Aptamer and ribozyme
display technologies, cell-free protein synthesis (CFPS) selection from nucleic acid libraries can be performed
technologies, and phenotype detection and evaluation much faster than those of functional proteins because the
technologies [172]. This approach mimics the process of nucleic acids themselves have binding or catalytic activi-
natural selection (Darwinian evolution) to evolve pro- ties (i.e., selectable phenotypes), such that the genotype
teins toward a target goal. It involves subjecting a gene and phenotype are identical. However, since proteins can-
to iterative rounds of mutagenesis (creating a molecular not be amplified, it is necessary to have a linkage between
library with sufficient diversity for the altered function), the phenotype exhibited by the protein and the genotype
selection (expressing the variants and isolating members (mRNA or DNA) encoding it to evolve proteins.
with the desired function), and amplification (generating Many genotypephenotype linkage technologies have
a template for the next round). This process can be per- been developed; these link proteins to their correspond-
formed in vivo (in living cells), or in vitro (free in solu- ing genes (Fig.18) [172174]. Genotypephenotype link-
tions or microdroplets). Molecular diversity is typically age technologies can be divided into invivo and invitro
created by various random mutagenesis and/or in vitro display technologies. In vitro display technologies can
gene recombination methods, as described in Gene be further classified into RNA display and DNA display
engineering. technologies.
Functionally improved variants are identified by an In vivo display technology includes phage display [175]
HTS or selection method and then used as the par- and baculovirus display [176], in which a protein gene
ents for the next round of evolution. The success of designated for evolution is fused to a coat protein gene
directed evolution depends on the choices of both and expressed as a fusion protein on the surface of phage
Nagamune Nano Convergence (2017) 4:9 Page 25 of 56

Fig.18 Various genotypephenotype linkage technologies. a Phage display technology. b Cell surface display technologies: invivo display on the
surface of bacteria, yeast or mammalian cell. c RNA display technology

and virus particles. Cell surface display technologies are toxic to cells and can cover quite large libraries (1015) by
also in vivo display technologies and use bacteria [177, bypassing the restricted library size bottleneck of invivo
178], yeast [179, 180] and mammalian cells [181] as host display technologies (Table1).
cells, in which the fusion gene resulting from a protein There are several in vitro DNA display technologies,
gene and a partial (or full) endogenous cell surface pro- such as CIS display [187], M. Hae III display [188], STA-
tein gene is expressed and displayed on the cell surface. BLE display [189], microbead display [190] and in vitro
These in vivo display technologies can indirectly link compartmentalization (IVC) [191]. CIS display non-
a protein designated for evolution and its gene through covalently links RepA (DNA-binding protein) fusion
the display of the protein on biological particles or cells. protein and its coding DNA template through the inter-
However, the library sizes of invivo display technologies action between RepA and the CIS element of the DNA
are usually restricted to the 1 081011 size range by the template. For M. Hae III display, the DNA methyltrans-
efficiency of the transformation and transduction steps of ferase M. HaeIII covalently links a protein and its DNA
their encoding plasmids. template. IVC technology uses the aqueous droplets in
In vitro display technologies are based on CFPS sys- wateroil emulsions to compartmentalize individual
tems. Recent advances in CFPS technologies and appli- genes and gene products. STABLE display and microbead
cations have been reviewed elsewhere [182185]. RNA display technologies utilize noncovalent biotinstrepta-
display technology includes mRNA display and ribosome vidin binding to link biotin-labeled DNA templates and
display [186]. mRNA display covalently links a protein streptavidin-fused proteins.
to its coding mRNA through a puromycin linker that is The details of HTS and selection methods, such as flu-
covalently attached to the protein via ribosome-catalyzed orescence-activated cell sorting-based phenotype detec-
peptide bond formation. Ribosome display noncovalently tion and evaluation technologies coupled with these
links a protein to its coding mRNA genetically fused to a display technologies [172, 192195], as well as the appli-
spacer sequence lacking a stop codon through a ribosome cations of the directed evolution of enzymes, antibodies,
because the nascent protein does not dissociate from receptors and other proteins in such areas as environ-
the ribosome. Such display technologies using in vitro mental issues, catalysis, gene therapy, and therapeutic
translation reactions can screen proteins that would be protein and vaccine development will not be covered in
Nagamune Nano Convergence (2017) 4:9 Page 26 of 56

Table1 Various display technologies

Technology (typical number ofsequences Description Strengths or weaknesses
screened perlibrary)

Bacterial cell display (108109) Fusion gene libraries of the target proteins and bacte- Selects proteins displayed on bacterial cell
rial surface proteins surfaces
Fusion proteins are displayed on bacterial cell surface Flow cytometry allows multiparameter, quanti-
tative screening
Smaller library size
Cannot screen proteins that would be toxic to cells
Yeast or mammalian cell display (1081010) Fusion gene libraries of the target protein and cell Selects proteins displayed on eukaryotic cell
surface proteins of yeast or mammalian cells surfaces
Fusion proteins are displayed on cell surface Flow cytometry allows multiparameter, quanti-
tative screening
Smaller library sizes
Cannot screen proteins that would be toxic to cells
Phage or baculovirus display (1011) Fusion gene libraries of the target protein and phage Robust and quick
or virus coat proteins Cannot screen proteins that would be toxic to cells
Infected bacteria produces phage or virus particles
displaying fusion protein libraries on the surface
Ribosome display (1015) mRNA-target protein complexes are displayed on Large library size
stalled ribosomes in cell free protein synthesis Can screen proteins that would be toxic to cells
system Requires stringent conditions and stable proteins
Reverse-transcription PCR allows amplification after
rounds of selections
mRNA display (1015) mRNA-target protein fusions are synthesized in cell Large library size
free protein synthesis system by conjugating them Can screen proteins that would be toxic to cells
through a puromycin linker Works well with small proteins but not large ones
Reverse-transcription PCR allows amplification after Requires stringent conditions
rounds of selections

this review; readers are referred to several recently pub- organic materials for use in nanobio/bionanotechnology.
lished articles and reviews [42, 172, 175, 177, 180, 181, These technologies range from classical chemical bio-
186, 187, 193208]. conjugation technologies targeting natural AAs to more
Recent, significant advances in protein engineer- sophisticated approaches, such as unnatural AA (UAA)
ing have come through computational methods, such incorporation based on amber stop codon suppression,
as SCHEMA, ProSAR, and ROSETTA. Computational bioorthogonal chemical conjugations, protein chemical
design based on these methods greatly decreases the ligations and enzymatic conjugations.
need for probing randomized sequence space, rendering
the route to novel biocatalysts much more efficient [195, 3.4.1Chemical conjugation technologies targeting natural
209212]. Therefore, in the future, more detailed knowl- AAs
edge about the relationship between protein structures Standard chemical conjugation technologies for proteins
and functions, as well as advancements in high-through- target the side chains of natural AAs, such as the primary
put technology, may greatly expand the capabilities of amine groups (RNH2) of Lys residue and the N-termi-
protein engineering. nus, the carboxylic acid groups (RCOOH) of Asp, Glu
and the C-terminus, the thiol group (RSH) of Cys, the
3.4Chemical andenzymatic conjugation technologies phenyl ring of tyrosine (Tyr) and the indole ring of tryp-
In the current postgenomic era, many studies require tophan (Trp) (Fig.19) [213].
chemically modified proteins or protein bioconjugates Lys is one of the most common AA residues in pro-
that are impossible to prepare via standard ribosomal teins with an average abundance of approximately 6%
synthesis. Conjugation technologies to site-specifically and is often surface-exposed due to its hydrophilicity;
modify proteins with diverse natural and unnatural func- therefore, it is an excellent target site for conjugation.
tionalities have been developed in the last two decades. On the other hand, the N-terminus provides a more site-
These technologies have been widely utilized to fabri- selective location but is not always surface-exposed. The
cate hybrid biomolecular material, such as protein/pro- primary amine of Lys has been predominantly function-
tein, protein/peptide, protein/nucleic acid, protein/lipid, alized with N-hydroxysuccinimidyl-esters (NHS-esters),
protein/oligosaccharide, and protein/ligand hybrids, and NHS-ester sulfates or isothiocyanates. In these electro-
hybrid materials comprising biomolecules and inorganic/ philic reagents, NHS-esters are highly used for primary
Nagamune Nano Convergence (2017) 4:9 Page 27 of 56

Fig.19 Standard chemical conjugation technologies for proteins targeting at side chains of natural AA (Figure adapted with permission from: Ref.
[213]. Copyright (2015) American Chemical Society)

amine-targeted functionalization because of the reaction conjugation because its average abundance in naturally
simplicity. A limitation of NHS-esters is a side reaction of occurring proteins is estimated to be approximately
hydrolysis in water (<5 h half-life), which accelerates as 1%. The relatively low abundance of Cys facilitates the
the pH increases above 7. This hydrolysis competes with genetic modification of the protein sequence to intro-
desired reactions and reduces reaction efficiency [214]. duce a unique Cys. The nucleophilic side chain of Cys
The N-terminus can be selectively targeted for modifica- can be site-selectively targeted to create a well-defined
tion when it is sufficiently accessible and not post-transla- conjugate. At slightly basic pH levels, the thiolate moiety
tionally modified. The transamination reaction mediated can be modified with disulfides, maleimides, thiol-ene,
by pyridoxal-5-phosphate can be applied to the modifica- dibromo-maleimides or bis-sulfone. Modification with
tion of the N-terminal residue without the presence of toxic disulfide (under mild oxidative condition) and maleim-
Cu(II) or denaturing organic cosolvents, although proteins ide (Michael addition) reagents produces disulfide and
possessing N-terminal serine (Ser), threonine (Thr), Cys, or thiosuccinimide bond linkages that are not stable in
Trp residues will be incompatible with this method because the presence of free thiols, such as reduced glutathione
of known side reactions with aldehydes [215]. (GSH) abundant in the cytoplasm of cells [213]. This
Asp and Glu are also the most common AA residues in GSH-sensitive conjugation property has been positively
naturally occurring proteins; they have an average abun- utilized for the release of drug delivery system payloads
dance of approximately 12%, are often surface-exposed in the cytoplasm. In contrast, the ring-opening hydroly-
and are excellent target conjugation sites. The carbox- sis of thiosuccinimide using maleimide derivative incor-
ylic acid side chains of Asp, Glu and the C-terminus can porating a basic amino group adjacent to the maleimide,
be functionalized by carbodiimide chemistry, typically positioned to provide intramolecular catalysis of thiosuc-
using EDC, which has been widely used for covalently cinimide ring hydrolysis, yields a stable conjugate (e.g., an
crosslinking a carboxylic acid and amine. However, the antibodydrug conjugate) [216].
relatively high abundance of Lys, Asp and Glu and the Methods for the conjugation of Tyr, which has an aver-
high solvent accessibility of their side chains make it age abundance of 3% in proteins, have also been devel-
impossible to modify a single site on the protein surface oped. In the presence of strong oxidizing agents (e.g.,
using these methods. H2O2) and appropriate catalysts, the phenolic side chain
Cys is not definitively hydrophilic or hydrophobic, of the Tyr residue can crosslink with other phenolic com-
and it is an attractive residue site for directed target pounds. The oxidizing agents required to catalyze these
Nagamune Nano Convergence (2017) 4:9 Page 28 of 56

reactions are not discerning, and there is concern over The incorporation of multiple different UAAs has been
causing undesired side reactions to other portions of pro- achieved by the extension of codon-anticodon pairs using
teins. To overcome this problem, a Tyr coupling reaction a different four-base codon for each tRNA [222]. Tech-
has been developed; it involves an electrophilic reagent, nology using acylating ribozyme (flexizyme) instead of
imines formed in situ from aldehydes and electron-rich ssRS has been developed for in vitro semi-enzymatic
anilines. This three-component Mannich-type coupling synthesis and acylation [223]. Therefore, SSI is minimally
reaction is highly selective for Tyr and proceeds under invasive and allows the incorporation of any UAA into a
mild conditions [217]. specific site of a protein with minor effects.
Conventional methods for the conjugation of Trp,
which has an average abundance of approximately 1%, 3.4.3Bioorthogonal chemical conjugation technologies
require toxic heavy metals or biochemically incompat- UAA-incorporated proteins or peptides can be chemose-
ible conditions. Some of these methods also exhibit lectively conjugated with other biomolecules and syn-
cross reactivity with other AAs (particularly Tyr), thus thetic inorganic/organic materials bearing bioorthogonal
limiting the range of applications. Recently, a transition functional groups. Other biomolecules, such as nucleic
metal-free method using 9-azabicyclo[3.3.1]nonane-3- acids, lipids, and synthetic inorganic/organic materials
one-N-oxyl (keto-ABNO) for the conjugation of Trp was modified with bioorthogonal functional groups, can also
reported. This new method showed novel features, such be subjected chemoselectively to a bioconjugation reac-
as high Trp selectivity, the formation of single conjugates tion. This reaction should occur in an aqueous solution
with high homogeneity, facile conjugation at an ambient at near physiological pH, have rapid kinetics even with
temperature and nearly neutral pH and a short reaction submillimolar concentrations of reactants, and occur at
time [218]. physiological temperatures. The chemical reactions that
satisfy these requirements include the keton/hydroxy-
3.4.2Chemical conjugation technologies targeting UAAs amine condensations, Huisgen [3+2] cycloaddition, the
Although the site-specific substitution of AAs with rare Staudinger ligation, the DielsAlder cycloadditions and a
AAs, such as Cys or Tyr, provides multiple options for photo-Click cycloadditions (Fig.20) [224, 225].
the covalent functionalization of proteins, it is possible
to damage the folding or function of the proteins when 3.4.3.1 Conjugation reactions through aldehydes
the genetic modifications for such chemical conjugations and ketones The carbonyl groups of aldehydes and
become too extensive. The incorporation of UAAs into ketones can specifically react with strong -effect nucleo-
proteins allows for greater flexibility in protein modifica- philes, such as hydrazides (RNNH2) and alkoxyamines
tions. Unlike the natural AA residues, these UAAs can (RONH2), under acidic conditions (pH 46) to produce
contain entirely unique reactive moieties, such as azide, stable hydrazones and oximes, respectively. Since these
cyano, iodo, bromo, boc, and dansyl groups, and thereby ketone/aldehyde condensations show rather slow kinet-
afford completely bioorthogonal reactivity that can occur ics with second-order rate constants, large excesses of the
inside of living systems without interfering with native conjugation reagent are necessary in order to achieve good
biochemical processes. The site-specific incorporation labeling efficiency. In general, ketone/aldehyde condensa-
(SSI) of UAAs utilizes nonsense codons to incorporate tions are best suited for conjugation reactions invitro or
one or a few UAAs into a single or multiple defined loca- at the cell surface because the reaction requires an acidic
tions in a protein. The amber stop codon suppression is pH and high concentrations of the labeling reagent, which
commonly employed, in which the target-location codon are problematic in terms of cell toxicity [224].
is mutated to an amber stop codon [219]. To incorpo-
rate the UAA, a tRNA and aminoacyl tRNA-synthetase 3.4.3.2 Conjugation reactions throughazides The azide
(aaRS) are engineered to pair with the UAA and rec- group is truly orthogonal in its reactivity to the major-
ognize the amber codon as a sense codon. SSI substi- ity of biological functionalities. The Huisgen 1,3-dipo-
tutes a natural AA with an unnatural analog, where the lar cycloaddition of alkynes and azides has been widely
endogenous aminoacyl tRNA-synthetase (aaRS) accepts adopted since this reaction is both selective and high
the UAA and aminoacylates the appropriate tRNA with yielding when catalyzed by Cu(I) for Cu-chelating pro-
the UAA. Incubating an auxotrophic expression host pylene derivatives or by strain release for strained cyclo-
in culture media containing the analog UAA instead heptyne derivatives. Another cycloaddition reaction, the
of the specific natural AA leads to UAA incorporation inverse-electron-demand DielsAlder reaction between
at nearly every location where the specific natural AA tetrazines and strained alkenes or alkynes, yields dihydro-
would have been incorporated [220]. The SSI of UAAs pyridazines or pyridazines with nitrogen gas as the only
has been further extended to CFPS systems [219, 221]. byproduct. These reactions have recently been explored
Nagamune Nano Convergence (2017) 4:9 Page 29 of 56

Fig.20 Chemoselective bioconjugation reactions. a Ketone/hydroxylamine condensations. b Copper-catalyzed alkyneazide Huisgen cycload-
ditions. c Strain-promoted alkyne-azide cycloadditions. d Staudinger ligation. e DielsAlder cycloadditions. f Photo-click cycloadditions (Figure
adapted with permission from: Ref. [224]. Copyright (2014) American Chemical Society)

as chemoselective reactions for labeling and manipulat- simple filtration or extraction without chromatography or
ing biomolecules in their native states. The reactions are recrystallization. Many other bioorthogonal conjugation
extraordinarily fast (up to 105M1s1) and have improved reactions have been reported; readers can refer to recent
second-order kinetics relative to the chelating Cu(I)-cata- reviews [224, 225].
lyzed azide-alkyne cycloaddition (10200M1s1) [224].
The 1,2,3-triazole linkage formed in the cycloaddition 3.4.4Chemical ligation technologies
reaction (click reaction) is thermodynamically and hydro- Native chemical ligation (NCL) has become the most gen-
lytically stable. This reaction is insensitive to aqueous con- eral and robust method for the conjugation of protein
ditions and pH levels ranging from 4 to 12, succeeds over peptide, proteinprotein, proteinDNA, and proteinNP
a broad temperature range, and tolerates a wide variety of materials because it is simple, general, and has a high yield
functional groups. Pure products can be easily isolated by efficiency [226]. NCL is a chemoselective coupling reaction
Nagamune Nano Convergence (2017) 4:9 Page 30 of 56

that generates a native peptide bond by a reversible tran- can be used to generate a C-terminal -thioester of a pro-
sthioesterification between a peptide fragment containing tein from protein-intein fusion. In EPL, one or more of
an N-terminal Cys residue (-Cys) and another peptide the peptides is of recombinant origin, but the actual liga-
fragment bearing a C-terminal -thioester group, followed tion step is still a chemical process and can be performed
by an irreversible intramolecular N-S acyl shift (Fig. 21). under a wide range of reactions to introduce a variety of
This reaction proceeds efficiently under physiological con- functional materials, such as fluorophores, UAAs, iso-
ditions and is compatible with all natural AA side chains. topic labels, and post-translational modifications, into a
Therefore, through the recombinant preparation of pro- large number of proteins [228]. By contrast, PTS post-
teins having an -Cys residue, NCL can be used to gener- translationally links two recombinant protein fragments.
ate proteins containing modifications at their N-termini. It An intein domain is split into two fragments (split intein
is advantageous to conduct NCL in an aqueous solution at or trans-splicing intein), IntN and IntC, which are fused
a neutral pH even though a C-terminal -thioester deriva- to the flanking polypeptides, termed the N and C exteins
tive can be competitively hydrolyzed. Recent extensions (ExN and E xC). The ligation step in PTS must be per-
of NCL, such as ligation rate acceleration, chemoselective formed under conditions compatible with protein folding
post-ligation modifications, and the streamlined ligation of because the process involves the functional reconstitu-
multiple peptide fragments, have been reviewed [227]. xNIntN and I ntCExC
tion of a split intein. In this step, E
Expressed protein ligation (EPL) and protein trans- associate, fold to form a functional intein, restore auto-
splicing (PTS) are both intein-based chemical conjuga- catalytic protein splicing activity to excise the I ntNIntC,
tion technologies that permit the assembly of a protein and ligate the flanking ExN and ExC with a peptide bond
from smaller synthetic and/or recombinant unprotected of Cys.
polypeptide building blocks (Fig. 22). An intein is an Although the advances in NCL, EPL and PTS made
internal protein domain that can autocatalytically excise it possible to precisely introduce a variety of func-
itself from a precursor protein. The cis-splicing of intein tional materials into peptides and proteins, these tech-
by the addition of high concentrations of thiol derivatives nologies also have some drawbacks, as follows. (1) The

Fig.21 Native chemical ligation. Native chemical ligation (NCL) is a chemoselective coupling reaction that links a peptide fragment containing an
N-terminal Cys (-Cys) residue and another peptide fragment bearing a C-terminal -thioester group by a native peptide bond (Figure reproduced
with permission from: Ref. [106]. Copyright (2012) Springer)
Nagamune Nano Convergence (2017) 4:9 Page 31 of 56

Fig.22 Intein-based chemical conjugation. a Expressed protein ligation (EPL) is a semisynthetic version of NCL in which synthetic and recombi-
nant polypeptides are chemically ligated together. Proteins (A) expressed as intein fusions can be cleaved from the intein with a variety of thiols to
give the corresponding -thioester derivative. Proteins (B) containing N-terminal Cys can be made recombinantly by masking the Cys with a pro-
tease tag that can be later removed. b Protein trans-splicing (PTS) post-translationally links two protein fragments. An intein domain is split into two
fragments, IntN and IntC, which are fused to the flanking exteins, ExN and ExC. ExNIntN and IntCExC associate and fold to form a functional intein.
This functional intein can restore protein splicing activity to excise itself, and to conjugate ExN and ExC with a peptide bond (Figures adapted with
permission from: Ref. [106]. Copyright (2012) Springer)

preparation of synthetic peptide -thioesters is still tech- catalyze the covalent addition of some chemical groups
nically difficult. (2) Since the ligation process is a chemi- (e.g., phosphate, acetate, amide, and methyl groups and
cal reaction, the higher concentrations of both or either biotin, flavins, carbohydrates and lipids) to the N- or
of the reactants are required. (3) The application of EPL C-terminus or a side chain of an AA residue at specific
to many disulfide bond-containing proteins is restricted site in a protein; these enzymes can also catalyze the
or complicated since the use of high concentrations (usu- cleavage and ligation of peptide backbones in proteins.
ally more than several tens of mM) of thiol derivatives is Natural post-translational modifications of proteins are
needed to induce thiolysis of the protein-intein fusions. generally efficient under physiological conditions and
(4) The expression of intein-based fusion proteins often site-specific. Therefore, a variety of transferase or ligase
results in the formation of inclusion bodies due to the enzymes have been repurposed for site-specific protein
large protein sizes and poor solubility, which requires modification. Typically, a small tag peptide sequence
additional refolding steps. incorporated into the target protein is recognized by the
post-translational modification enzyme as a substrate
3.4.5Enzymatic conjugation technologies and then transfers functional moieties from an analog
In nature, numerous proteins are post-translationally of its natural substrate onto the tag (Fig. 23). Examples
modified by enzymes and play important roles in control- include formylglycine-generating enzyme (FGE), pro-
ling cellar processes, such as metabolism, signal trans- tein farnesyltransferase (PFTase), N-myristoyltransferase
duction, gene expression, and cell differentiation. These (NMTase), biotin ligase (BirA), lipoic acid ligase (LAL),
enzymes participating in post-translational modifications microbial transglutaminase (MTGase), sortase A (SrtA),
Nagamune Nano Convergence (2017) 4:9 Page 32 of 56

glutathione S-transferase (GST), SpyLigase, and several Enzymatic protein conjugation technologies, including
engineered self-labeling protein tags. Except for self-labe- non-site-specific crosslinking by such oxidoreductases as
ling protein tags, a primary benefit of this approach is the peroxidase, laccase, tyrosinase, lysyl oxidase, and amine
small size of the peptide tag that must be incorporated oxidase, are reviewed elsewhere [105]. Here, we briefly
into proteins, which ranges from 3 to 15 residues. Some review recent enzymatic conjugation technologies for
enzymes only recognize the tag peptide at a specific posi- site-specific protein conjugation and crosslinking of bio-
tion in the primary sequence of the protein (often the N- molecules and synthetic materials. The applications of
or C-terminus), while others are not inherently limited by enzymatic conjugations and modifications of proteins
tag position. with other biomolecules and synthetic materials are

Fig.23 Chemoenzymatic labeling strategies of the protein of interest (POI) using post-translational modification enzymes. a Formylglycine
generating enzyme (FGE) recognizes LCXPXR peptide motif and converts the side chain of Cys residue into an aldehyde group. The POI fused to
the aldehyde tag can be further functionalized with aminooxy or hydrazide probes. b Farnesyltransferase (FTase) recognizes the four AAs sequence
CA1A2X (A1 and A2 are non-charged aliphatic AAs and X is C-terminal Met, Ser or Phe) at the C-terminus and catalyzes the attachment of the farnesyl
isoprenoid group to the Cys residue. The POI can be further labeled by bioorthogonal chemical conjugation of the farnesyl moiety functionalized
with azide or alkyne. c N-Myristoyl transferase (NMT) recognizes the GXXXS peptide motif at the N-terminus and attaches a myristate group to an
N-terminal Gly residue. The POI can be further labeled by bioorthogonal chemical conjugation of myristate moiety functionalized with azide or
alkyne. d Biotin ligase recognizes the GGLNDIFEAQKIEWH peptide motif derived from biotin carboxyl carrier protein and catalyzes the transfer of
biotin from an ATP intermediate (biotinyl 5-adenylate) to Lys residue. Biotinylated POI can then be labeled with streptavidin conjugated with a vari-
ety of chemical probes. e Lipoic acid ligase recognizes the GFEIDKVWYDLDA peptide motif and catalyzes the attachment of lipoic acid or its deriva-
tives to Lys residue in the motif. f Transglutaminase (TGase) catalyzes the transamination reaction and forms an iso-peptide bond between Gln in
POI and Lys residue-functionalized small molecule probes, peptides or proteins. g Sortase cleaves LPXTG peptide tag fused to POI between Thr and
Gly residue and conjugates oligo Gly-functionalized small molecule probes, peptides or proteins to POI by forming a peptide bond between Thr
and Gly residues. h GST catalyzes Cys arylation and conjugates probes bearing a 4-mercaptoperfluorobiphenyl moiety to the N-terminal -Glu-Cys-
Gly sequence of POI. i SpyLigase catalyzes the generation of an isopeptide bond between Lys residue in KTag and Asp residue in SpyTag
Nagamune Nano Convergence (2017) 4:9 Page 33 of 56

limited to recombinant proteins harboring additional eral method for N-terminal-specific recombinant protein
protein/peptide tags. However, protein functionaliza- labeling [235].
tion using enzymatic conjugations is a promising method
because it is achieved simply by mixing proteins without 3.4.5.4BirA BirA from E. coli catalyzes the adenosine
special techniques. The details of enzymatic conjuga- triphosphate (ATP)-dependent amide bond formation
tion technology applications will not be covered in this between the carboxylic group of biotin and the -amino
review; readers are referred to several recently published group of a Lys in an acceptor peptide sequence (23 AA
reviews [229232]. residues) (Fig. 23d). This acceptor sequence was fur-
ther optimized to a 15-AA acceptor peptide sequence
3.4.5.1FGE The FGE oxidizes Cys or Ser residue to (GLNDIFEAQKIEWHE) [236]. BirA can be used to site-
formylglycine (FGly) within a conserved 13-AA consen- specifically conjugate a biotin moiety to recombinant pro-
sus sequence found in prokaryotic Type I sulfatases. The teins by the genetic fusion of the BirA recognition acceptor
modification is thought to occur co-translationally, before peptide sequence with the target protein. The enzymatic
protein folding. The consensus sequence can be incor- biotin labeling to a protein allows the subsequent forma-
porated into heterologous proteins expressed in E. coli, tion of very strong noncovalent conjugate with avidin due
where it is modified efficiently by a co-expressed bacterial to the low dissociation constant between biotin and avi-
FGE. Furthermore, the minimized core motif sequence din (1015 M). Another orthogonal acceptor sequence
CX(P/A)XR or SXPXR, derived from the most highly for yeast BirA has been further developed to enable two-
conserved portion of the FGE recognition site, directed color imaging [237]. The substrate tolerance of BirA was
the efficient conversion of Cys or Ser to FGly. The alde- also expanded to biotin analogs, including ketone, azide,
hyde-bearing residue FGly can be subsequently used for and alkyne groups, which contain alternative functionali-
covalent conjugation using complementary aminooxy- ties suitable for bioorthogonal reactions [238].
or hydrazide-functionalized moieties by ketone-reactive
chemistries (Fig.23a) [233]. 3.4.5.5LAL LAL from E. coli catalyzes the ATP-
dependent amide bond formation between the carbox-
3.4.5.2PFTase PFTase is an / heterodimer enzyme ylic group of lipoic acid and the -amino group of a lysine
that catalyzes the transfer of a farnesyl isoprenoid group in an optimized 13-AA recognition acceptor sequence
from farnesyl pyrophosphate (FPP) through a thioether (GFEIDKVWYDLDA) (Fig. 23e). The Trp 37 residue at
bond to a sulfur atom on a Cys in a tetrapeptide sequence the lipoic acid-binding pocket of LAL was substituted
(denoted as a C A1A2X-box, here C is Cys, A1 and A2 are with small AA residues to accept a wider range of lipoic
aliphatic AAs, and X is one of a variety of AAs) four resi- acid analogs containing an aliphatic azide, aryl-aldehyde,
dues from the C-terminus (Fig. 23b). Since PFTase can or aryl-hydrazine moiety [239]. These lipoic acid analogs
tolerate many simple modifications to the aldehyde-con- are attached to a Lys residue in the acceptor sequence of a
taining isoprenoid substrate, it can be used to introduce protein and are then used to conjugate diverse functional
a diverse range of functionalities into proteins containing molecules by bioorthogonal reactions.
a CA1A2X-box positioned at the C-terminus. Subsequent
chemoselective reactions with the resulting protein can 3.4.5.6MTGase Transglutaminase is a unique enzyme
then be used for a wide range of applications. The catalytic that catalyzes the acyl-transfer reaction between the
activity of PFTase toward various FPP analogs has been -carboxyamide group of a Gln residue in proteins and
greatly improved by site-directed mutagenesis around the a wide variety of unbranched primary amines, com-
substrate-binding pocket of PFTase [234]. monly the -amino group of a Lys residue, and forms an
isopeptide bond between the side chain of Gln residues
3.4.5.3NMTase NMTase from Candida albicans cata- and primary amines (Fig.23f ). Because this conjugation
lyzes the acyl transfer of myristic acid from myristoyl- reaction is irreversible, involves the release of ammonia
CoA to the amino group of an N-terminal glycine (Gly) and proceeds quickly even under low temperature con-
residue of a protein to form an amide bond. NMTase ditions (~15 C), the conjugation product is stable, and
recognizes the sequence GXXX(S/T), where X can be a high yield can be obtained. MTGase is isolated from
any AA (Fig.23c). This enzyme can successfully transfer Streptomyces mobaraensis, which is widely used in the
alkyne- and azide-containing myristic acid analogs that food industry, and recognizes various peptide sequences
incorporated the bioorthogonal groups at the distal end consisting of Gln residues. A notable correlation was
of the lipid to the N-terminal Gly residue of recombinant observed between the polypeptide chain regions of high
proteins containing an N-terminal myristoylation motif. temperature factor (B-factor) determined crystallograph-
This method provides a convenient and potentially gen- ically and the MTGase attacking sites, thus indicating
Nagamune Nano Convergence (2017) 4:9 Page 34 of 56

the role of polypeptide chain mobility or local unfolding mediated transpeptidation are very short, these motifs
in dictating site-specific enzymatic modifications [240]. can be easily incorporated into proteins or polypeptides
Consequently, enhanced MTGase polypeptide chain flex- either by standard genetic means or chemical peptide
ibility limits the enzymatic reaction with Gln residues on synthesis. Benefiting from its simplicity and specificity, a
rigid polypeptide in globular proteins. Therefore, it is pos- soluble truncated Staphylococcus aureus SrtA that lacks
sible to predict the site(s) of Gln residue modifications by the N-terminal membrane-anchoring motif has begun to
MTGase on the basis of local structure and dynamics of be applied for a wide variety of protein engineering and
polypeptide chain containing Gln residue. bioconjugation purposes, including the in situ site-spe-
Because of its openness with regard to the primary cific fluorescent labeling of membrane proteins [248252]
amine substrate, MTGase is an attractive catalyst for and the fabrication of an electrochemically active protein
generating protein conjugates with small functional bilayer on electrodes [253].
molecules, lipids, nucleic acids, synthetic polymers, e.g., Unfortunately, since this conjugation reaction is revers-
PEG, peptides and even other proteins. Although the ible and the acyl-enzyme intermediate is hydrolyzed by
substrate specificity of MTGase toward the polypep- water even in the presence of enough oligo-Gly nucleo-
tide sequence containing a Gln residue (Q-tag) has not philes, the conjugation reaction does not proceed to
yet been clarified, the Q-tag derived from the polypep- completion. However, we have overcome this limitation
tides of globular proteins, the ribonuclease S-peptide by introducing a -hairpin structure around the ligation
(KETAAAKFERQHMDS and its Lys to Ala-substituted site of products and preventing substrate recognition by
peptide AETAAAAFERQHMDS), the F-helix peptide SrtA, thereby successfully stabilizing conjugation prod-
of horse heart myoglobin (PLAQSH) or the designed ucts and providing a high yield [254]. S. aureus SrtA
N-terminal oligo-Gly tag (N-Gly5), which are recog- needs Ca2+ for stabilizing the active site conformation,
nized as a Gln-substrate by MTGase, can be utilized and its strong Ca2+ dependency makes S. aureus SrtA
as Q-tag substrates [108, 241244]. For protein modi- difficult for use under low Ca2+ concentrations and in
fication by MTGase, these Q-tags are incorporated at the presence of Ca2+-binding substances. To overcome
the N- or C-terminus or inside the loop region of pro- this problem, we designed an S. aureus SrtA heptamu-
teins by genetic means. Subsequently, MTGase can tant (P94R/E105K/E108A/D16N/D165A/K190E/K196T)
site-specifically conjugate the Q-tag in the protein with that exhibited a high Ca2+-independent catalytic activ-
a primary amine-containing short synthetic linker or a ity and successfully catalyzed a selective proteinpro-
Lys residue-containing polypeptide tag (KTag) harbor- tein ligation in living cells, which usually retain low Ca2+
ing a functional moiety. However, one of the drawbacks concentrations [255]. These recent advances in S. aureus
of conjugating proteins possessing many Lys and Gln SrtA-mediated ligation will contribute to the develop-
residues is that the activity of MTGase toward Gln and ment and design of many other protein conjugates and
Lys residues makes it difficult to control the site(s) of multienzyme complexes both invitro and invivo.
modification.
3.4.5.8GST GST catalyzes conjugation reactions
3.4.5.7SrtA SrtAs are cell envelope-bound housekeep- between the Cys residue of glutathione (GSH, -Glu-Cys-
ing transpeptidases from gram-positive bacteria. SrtA Gly) and various electrophiles and allows the cell to detox-
attaches surface proteins, such as virulence factors, to the ify xenobiotics in vivo (Fig. 23h). The ubiquitous nature
penta-Gly motif of branched lipid II, the peptidoglycan of GST facilitates this bioconjugation with polypeptides
precursor. SrtA recognizes the peptide sequence (LPXTG) bearing an N-terminal GSH in aqueous media and ena-
and catalyzes the cleavage of the amide bond between the bles the chemo- and regioselective functionalization of a
Thr and Gly residues by means of an active site Cys resi- single Cys thiol group of GSH based on a nucleophilic aro-
due (Cys184) (Fig.23g). This process generates a covalent matic substitution reaction between Cys residues and per-
acyl-enzyme intermediate. The carboxyl group of the Thr fluoroarenes, even in the presence of other unprotected
of the thioester intermediate then undergoes nucleophilic Cys residues and reactive functional groups on the same
attack by an amino group of the oligo-Gly substrates, pro- polypeptide chain. This conjugation reaction can be car-
ducing ligated products and altering the primary struc- ried out over a wide range of temperatures (460C) and
ture. Recent reports have demonstrated that the -amino in co-solvent system with the addition of organic solvents
group of Lys residues can also act as a nucleophile instead (up to 20%) [256]. However, this technology is currently
of the -amino group of oligo-Gly [245]. Since both of the limited to peptide-based couplings due to the require-
optimized recognition peptide sequences, LPETGG [246] ment for both an N-terminal -Glu-Cys-Gly sequence and
and oligo-Gly with more than two repeats [247], for SrtA- a perfluoraryl reaction partner.
Nagamune Nano Convergence (2017) 4:9 Page 35 of 56

3.4.5.9SpyLigase SpyLigase is an artificial ligase obtained 3.4.6Selflabeling protein tagbased chemoenzymatic

by engineering a domain (CnaB2) from the fibronectin conjugation technologies
adhesion protein FbaB of Streptococcus pyogenes (Spy), Chemoenzymatic labeling methods exploit the exquisite
which is essential for the bacteria to invade human cells. molecular recognition mechanism between substrates/
Within CnaB2, there is a post-translational modification inhibitors and enzymes to create a new specific covalent
to form an isopeptide bond between Lys31 and Asp117 linkage between them by engineering enzymes (Fig. 24)
residues, which is catalyzed by an apposed Glu77 resi- [229].
due. Based on the 3D structure and isopeptide bond
formation mechanism of CnaB2, the domain was ration- 3.4.6.1SNAPtag SNAP-tag (20kDa) was derived from
ally split into three parts, SpyTag (AHIVMVDAYKPTK), the human DNA repair protein O6-alkylguanine-DNA
KTag (ATHIKFSKRD) and SpyLigase (11kDa, containing alkyl-transferase (AGT). The normal function of AGT is
the catalytic Glu77 residue). SpyLigase was derived from to repair O6-alkylated guanine in DNA by transferring
CnaB2 first by the removal of SpyTag and KTag, and then the alkyl group in an SN2 reaction to a reactive Cys145
by circular permutation via replacing residues from the residue in AGT. The repair mechanism is unusual because
C-terminus of CnaB2 with a Gly/Ser linker, followed by the protein is irreversibly inactivated. Consequently, the
N-terminal CnaB2 residues. SpyLigase not only can ligate reaction of AGT-fusion proteins with O6-benzylguanine
KTag and SpyTag fused at the C- or N-terminus of peptides (BG) derivatives harboring functional moieties leads to
but can also direct the ligation of KTag to SpyTag inserted the irreversible and covalent labeling of the fusion pro-
in the middle of a protein (Fig.23i). The yield of conjuga- teins since the functional moieties on BG are transferred
tion products decreased from approximately 5010% by along with the benzyl group of BG to the reactive Cys,
elevating the reaction temperature from 4 to 37C, likely creating a stable thioether covalent bond. The SNAP-tag-
due to a dynamic change in the secondary structure of mediated labeling of proteins in bacteria and yeast is spe-
SpyLigase [257]. cific, since the respective endogenous AGTs do not accept

Fig.24 Self-labeling protein tags. a, b Both SNAP- and CLIP-tag derive from O6-methylguanine-DNA methyltransferase with C145 as the active site.
c The Halo-tag derives from haloalkane dehalogenase whose active site D106 forms an ester bond with the chloroalkane linker. d The TMP-tag non-
covalently binds with trimethoprim and brings the , -unsaturated carbonyl (i) or sulfonyl (ii) into proximity of the engineered reactive Cys (L28C)
(Figure adapted with permission from: Ref. [229]. Copyright (2017) American Chemical Society)
Nagamune Nano Convergence (2017) 4:9 Page 36 of 56

BG as substrates, whereas AGT-deficient cell lines should tein can be covalently labeled with a variety of functional
be used for labeling in mammalian cells [258]. group-modified chloroalkane linkers and can be applied
to a wide range of fluorescent labels, affinity handles, or
3.4.6.2CLIPtag Subsequently, AGT mutant-based solid supports.
CLIP-tag, which reacts specifically with O2-benzylcyto-
sine (BC) derivatives, was developed by directed evolu- 3.4.6.4Trimethoprim (TMP)tag TMP-tag (18 kDa)
tion. To generate a mutant library of AGT, AA residues was derived from E. coli dihydrofolate reductase (eDHFR),
at positions with indirect proximity to BG bound in the which binds the small-molecule inhibitor TMP with high
active site were chosen with the aid of the crystal structure affinity (1nM KD) and selectivity (affinities for mamma-
of wild-type AGT. After two-step library screenings using lian DHFRs are KD>1M). The first-generation TMP-tag
yeast and phage display, CLIP-tag, the eight-point mutant harnessed the high-affinity interaction between eDHFR
of AGT (Met60Ileu, Tyr114Glu, Ala121Val, Lys131Asn, and TMP to form long-duration and yet reversible binding
Ser135Asp, Leu153Ser, Gly157Pro, Glu159Leu) was without covalent bond formation. The second-generation,
selected. CLIP-tag with potent catalytic activity exhib- engineered, self-labeling TMP-tag (Leu28Cys) exploited
ited a 105-fold change in substrate specificity and a 100- a proximity-induced Michael addition reactivity between
fold greater preference for BC over BG [259]. The mutual a Cys28 residue engineered on the eDHFR surface near
orthogonality of the SNAP- and CLIP-tags enables the the TMP binding site and a mild electrophile, such as
simultaneous labeling of multiple proteins in the same an , -unsaturated carbonyl moiety, e.g., the -carbon
cellular context. of acrylamide, or a sulfonyl group installed on the TMP
derivatives. To optimize the positioning of the Cys resi-
3.4.6.3HaloTag Rhodococcus haloalkane dehalogenase due nucleophile and the acrylamide electrophile of the
(DhaA) removes halides from aliphatic hydrocarbons by TMP derivatives, the site of point mutation on the eDHFR
a nucleophilic displacement mechanism. A covalent ester surface and the atom length of the spacer between the
bond is formed during catalysis between an Asp106 resi- 4-OH group of the TMP and the reactive -carbon of the
due in the enzyme and the hydrocarbon substrate. The acrylamide functional group were investigated based on
base-catalyzed hydrolysis of this covalent intermediate the molecular modeling of the eDHFR and TMP deriva-
subsequently releases the hydrocarbon as an alcohol and tive complexes. After subsequent combinatorial screening
regenerates the Asp106 nucleophile for additional rounds invitro, the combination of the TMP-tag (Leu28Cys) and
of catalysis. The based-catalyzed cleavage is mediated the TMP derivatives with a 10-atom spacer was selected
by a conserved His272 residue located near the Asp106 and exhibited superior specificity and efficiency in protein
nucleophile. HaloTag (33kDa) was derived from a mutant labeling with fluorophores for live cell imaging [261]. Since
DhaA, whose catalytic His272 residue is substituted with the covalent TMP-tag is based on a modular organic reac-
a Phe residue and does not exhibit the enzymatic activ- tion rather than a specific enzyme modification, it is easier
ity of intermediate cleavage. However, the apparent bind- to build additional features into the covalent TMP-tag.
ing rates of haloalkanes to this mutant are low compared Self-labeling protein tags, such as SNAP-, CLIP-, Halo-
to those of common affinity-based interactions, such as and TMP-tags, feature exquisite specificity and broad
biotinstreptavidin, potentially hampering the practical applicability to the areas of subcellular protein imaging in
utility of this mutant as a protein tag. To overcome this live cells, the fabrication of proteinDNA, proteinpep-
issue, several variants with dramatically improved bind- tide and proteinprotein complexes, and protein immo-
ing rates were identified using a semi-rational strategy, bilization on solid materials, but they are limited by their
proteinligand binding complex modeling, site-satura- large molecular size (2030kDa) and expensive substrate
tion mutagenesis, and HTS for faster binding kinetics. A derivatives, except for HaloTag.
mutant with three point substitutions, Lys175Met/Cys-
176Gly/Tyr273Leu, i.e., HaloTag, has a high apparent sec- 3.5Linker engineering
ond-order rate constant, thus allowing the labeling reac- Linker engineering is also an important technology for
tion to reach completion even under low haloalkane ligand controlling the distances, orientations and interactions
concentrations [260]. Covalent bond formation between among functional components crosslinked in conjugates.
the HaloTag and chloroalkane linker (14 atoms long with Linkers are indispensable units for the fabrication of mul-
6 carbon atoms proximal to the terminal chlorine) func- tidimensional biomaterials or complexes of bio/organic/
tionalized with small synthetic molecules is highly spe- inorganic materials. Such linkers can be classified as
cific, occurs rapidly under physiological conditions and is chemical or biological linkers, such as oligonucleotides or
essentially irreversible. Therefore, the HaloTag-fused pro- polypeptides.
Nagamune Nano Convergence (2017) 4:9 Page 37 of 56

3.5.1Chemical linkers programed structures [117120, 280]. These DNA linkers

Chemical linkers have been widely used to modify or have been utilized to immobilize functional materials (e.g.,
crosslink biomolecules, such as proteins, peptides, DNA, aptamers, peptides, proteins, antibodies, enzymes,
nucleic acids and drugs, synthetic polymers and solid sur- and NPs) on complementary DNA-modified solid sup-
faces with functional molecules and materials. Chemical ports for bioanalysis [117, 281], to fabricate multifunc-
linkers can be characterized by the following properties: tional NPs for biosensing and bioimaging [65, 68, 77, 79],
chemical specificity, reactive groups, spacer arm length, for DNA origami, and for placing cascading multienzyme
water solubility, cell membrane permeability, spontane- complexes on DNA scaffolds [120, 122125]. Although
ously reactive or photoreactive groups, and cleavabil- short DNA linkers display a relatively high physicochemi-
ity by such stimuli as pH, redox, and light. Particularly, cal stability in vitro, some approaches, such as the utili-
spacer arm length and water solubility are important zation of unnatural base DNA or PNA, are required for
parameters for protein modifications and crosslinking invivo applications to prevent degradation by nucleases.
using chemical linkers. For example, when biomolecules PNA is a DNA analog with a noncyclic, peptide-like
are functionalized with small molecules, such as fluoro- backbone (Fig.25). Owing to its flexible and neutral back-
phores or bioorthogonal functional groups, rigid, short bone instead of a negatively charged deoxyribose phos-
methylene arms are utilized as spacers. Various photo- phate backbone, PNA exhibits very good hybridization
cleavable, short chemical linkers were also developed to properties with DNA, RNA, PNA, and DNA duplexes at
control the functions of crosslinked biomolecules [54, low and even high ion concentrations, as well as a higher
262, 263]. In contrast, when proteins are functionalized temperature stability than the corresponding pure nucleic
with hydrophobic or large materials, hydrophilic, flex- acid complexes. Therefore, PNA can highly discriminate
ible, long spacer arms formed from PEG chains are often mismatched DNA and has a stronger binding affinity for
utilized to increase the water solubility of functionalized complementary DNA than does its DNA counterpart.
chemical linkers and to avoid steric hindrance between PNA also displays a very high stability against enzy-
proteins and functionalized materials. We utilized PEG matic degradation due to its peptide-like backbone [282].
chains as chemical linkers to prepare a Fab-green fluo- Applications of PNA linkers in the fields of therapy, diag-
rescent protein (GFP) immunoconjugate for a homo- nosis, and biosensing have been reviewed [282284]. For
geneous immunoassay [264], an enzyme-streptavidin example, coupling a radioactively labeled PNA to a TfR
conjugate for enzyme activity control [265, 266], and a
Synechocystis sp. DnaB intein-TMP conjugate for invitro
protein ligation [267], and the results showed that the
length of the PEG chemical linkers affected both the con-
jugation efficiency and the controllability of protein func-
tion. We also produced antibody-lipid and peptide-lipid
conjugates for cell surface display [268270] using PEG
chain linkers. Although there are enormous bioconjuga-
tion applications for biomolecules using chemical linkers,
the details of recent applications are reviewed elsewhere
[271279].

3.5.2Biological linkers
3.5.2.1 Oligonucleotide linkers In the bottom-up fabri-
cation of nanoscale systems, synthetic DNA oligonucleo-
tides are extraordinarily useful as a construction unit. The
extremely high specificity of WatsonCrick base pairing
allows one to readily design DNA linkers by using the pre-
dictable adeninethymine (AT) and guaninecytosine
(GC) hydrogen-bonding interaction between comple-
mentary nucleic acids. In practice, short DNA oligomers
with approximately 1030 nucleotides (mostly 21 nucleo-
tides forming a 7-nm long base pair segment) have been
utilized as linkers to noncovalently conjugate comple- Fig.25 Schematic chemical structures of PNA and DNA. The circles
mentary oligonucleotide-modified materials by hybridi- show the different backbone linkages of PNA and DNA. A, T, G, and C
zation and facilitate the fabrication of a wide variety of denote adenine, thymine, guanine and cytosine, respectively
Nagamune Nano Convergence (2017) 4:9 Page 38 of 56

mAb rendered the antisense agent transportable through space between the functional units of a fusion protein for
the bloodbrain barrier [285]. The identification and dif- correct folding. However, if the N- or C-terminus is not
ferentiation of tuberculous and nontuberculous myco- flexible or not long enough to prevent steric hindrance,
bacteria in liquid cultures were clinically evaluated by a this effect will reduce the degrees of freedom of units
fluorescence hybridization assay using PNA [286]. The in fusion protein dynamics and may cause unfavorable
detection of a complementary oligonucleotide at a fem- results, such as inclusion body formation derived by pro-
tomolar (~1015 M) level was accomplished based on the tein misfolding, a loss of function and a low yield of func-
ion-channel sensor technique using Au electrodes modi- tional fusion proteins. For this reason, longer peptide
fied with self-assembled monolayers of the PNA probe linkers are generally inserted between functional units
and 8-amino-1-octanethiol [287]. [290].
Peptide linkers are generally classified into three
3.5.2.2 Three classes ofpeptide linkers The concepts of groups according to their structures: flexible linkers, rigid
the protein domains and modules were first proposed in linkers, and site-specific linkers cleavable by proteolytic
1973 by Wetlaufer [288] and 1981 by Go [289], respec- enzyme digestion. In addition to the basic role of linking
tively. These concepts gave insights into domains and functional units together or releasing functional units
modules as the basic structural, functional or evolution- (e.g., toxin release in drug delivery systems, affinity tag
ary units of proteins. A wide variety of naturally occurring cleavage from tag-fused recombinant pharmaceutical
multidomain fusion proteins with different architectures proteins in the purification process), peptide linkers may
have been generated through evolution and characterized offer many other advantages for the production of fusion
to meet the functional requirements of living organisms proteins, such as improving biological activity and struc-
at the molecular level [290]. The strategies used by nature tural stability and achieving desirable biopharmaceutical
to evolve fusion proteins have been mimicked by the con- pharmacokinetic profiles [324]. Therefore, peptide link-
struction of hybrid or chimeric proteins using molecular ers play a variety of structural and functional roles in
biology techniques. Inspired by natural fusion proteins, fusion proteins.
synthetic fusion proteins have been designed to achieve
synergistically improved bioactivities or to generate novel 3.5.2.3 Flexible peptide linkers Flexible linkers are fre-
functional combinations derived from each of their com- quently adopted as natural inter-domain peptide linkers
ponent moieties, which are integrated into one molecule in multidomain proteins when the joined domains require
by peptide linkers. The fusion proteins have been widely a certain degree of movement or interaction. Based on the
applied in various areas, including recombinant protein analysis of AA preferences for residues contained in these
production by the tag-mediated enhancement of protein natural flexible linkers, it has been revealed that they are
expression, solubility and high-throughput purification generally composed of small, nonpolar (e.g., Gly) or polar
[291, 292], fluorescent protein-mediated molecular imag- (e.g., Ser, Thr) residues [325]. The small size of these AA
ing [293], advanced biocatalysis [101, 108, 111, 115, 164, residues provides flexibility and enables the mobility of
290, 294297], biosensing and bioelectronic materials the connected functional units. The incorporation of Ser
[290, 298300], pharmaceuticals, diagnostics and thera- or Thr can maintain the stability of the peptide linker in
peutics [208, 290, 301, 302], reporter protein-mediated aqueous solutions by forming hydrogen bonds with water
immunoassays [303310], the chimeric receptor-medi- molecules, thereby reducing unfavorable interactions
ated control of cell fate, e.g., growth, death, migration or between the linker and protein moieties. The most widely
differentiation [311319], the library selection of antibod- used synthetic flexible linker is the G4S-linker, (G4S)n,
ies [203, 320, 321] and antibody-mediated drug delivery where n indicates the number of G 4S motif repeats. By
[218, 322, 323]. changing the repeat number n, the length of this G 4S
Genetic fusion and enzymatic conjugation technolo- linker can be adjusted to achieve appropriate functional
gies have been commonly adopted for the construction unit separation or to maintain necessary interactions
of fusion proteins. Among them, an end-to-end genetic among units, thus allowing proper folding or achieving
fusion is the simplest method for constructing a fusion optimal biological activity [324]. Poly-Gly ( Gn) linkers also
protein, where the coding genes of functional units form an elongated structure similar to that of the unsta-
are combined together and expressed in a suitable host ble 310-helix conformation. Since Gly has the greatest
organism. Direct tandem genetic fusion through restric- freedom in backbone dihedral angles among the natural
tion enzyme sites is simple; the flexible and unstructured AAs, Gn linkers can be assumed to be the most flexible
N- or C-terminal regions of the component proteins polypeptide linkers [326]. In addition to the G 4S linkers
and additional short peptides derived from restriction and poly-Gly linkers, many other flexible linkers, such
enzyme sites act as a peptide linker to provide enough as KESGSVSSEQLAQFRSLD and EGKSSGSGSESKST
Nagamune Nano Convergence (2017) 4:9 Page 39 of 56

for the construction of a single-chain variable fragment fusion tags include factor Xa (I(E/D)GRX), enterokinase
(scFv), have been designed by searching libraries of 3D (DDDDKX), thrombin (LVPRGS), tobacco etch virus

derivative of human rhinovirus 3C protease, PreScission

peptide structures derived from protein data banks for protease (ENLYFQ(G/S)) and a genetically engineered
crosslinking peptides with proper VH and VL molecular
dimensions [327]. These flexible linkers are also rich in (LEVLFQGP) [329, 330]. Chemicals that are specific and
small or polar AAs, such as Gly, Ser, and Thr, and they efficient for the chemical cleavage of proteins in solution
contain additional AAs, such as Ala, to maintain flex- are CNBr (Met), 2-(2-nitrophenylsulfonyl)-3-methyl-
ibility, as well as large polar AAs, such as Glu and Lys, to 3-bromoindolenine (Trp), 2-nitro-5-thiocyanobenzoic
increase the solubility of fusion proteins. acid (Cys), formic acid (AspPro) and hydroxylamine
(AsnGly) [331]. Here, the down arrow and X in paren-
3.5.2.4Rigid peptide linkers Rigid linkers act as stiff thesis indicate the cleavage site of the recognition site
spacers between the functional units of fusion proteins and any AA, respectively. In general, enzymatic cleavage
to maintain their independent functions. The typical rigid is site-specific and can be carried out under mild condi-
linkers are helix-forming peptide linkers, such as the poly- tions. However, cleavage efficiency may vary with differ-
proline (Pro) helix (Pn), poly-Ala helix (An) and -helix- ent fusion proteins. Steric hindrance or the presence of
forming Ala-rich peptide (EA3K)n, which are stabilized by unfavorable residues around the cleavage site could result
the salt bridges between G lu and L
ys+ within the motifs in inefficient processing. In contrast to enzymatic cleav-
[328]. Fusion proteins with helical linker peptides are age, chemical cleavage offers a less expensive alternative
more thermally stable than are those with flexible link- but requires harsh conditions that may cause side-chain
ers. This property was attributed to the rigid structure of modifications. Furthermore, since chemical cleavage usu-
the -helical linker, which might decrease interference ally targets specific residues or dipeptide linkages, the fre-
among the linked moieties, suggesting that changes in quent presence of the single- or double-residue site rec-
linker structure and length could affect the stability and ognized by these chemicals within the AA sequence of the
bioactivity of functional moieties. target protein limits its use [332].
The Pro-rich peptide (XP)n, with X designating any AA, Self-cleaving tags are a special group of fusion tags
preferably Ala, Lys, or Glu, can also constrain the linker that are based on protein modules (e.g., intein, SrtA, the
to an extended conformation with relatively limited flex- FrpC module, and the Cys protease domain) and possess
ibility. The Pro residue is a very unique AA; it is a cyclic inducible proteolytic activities. Fusion proteins contain-
AA, and its side chain cyclizes back to the amide on the ing them can be site-specifically self-cleaved by the trig-
backbone, which restricts the confirmation of its back- ger of a low molecular weight compound or a change in
bone to a small range of backbone angles. Since the Pro its conformation. Combined with appropriate affinity
residue has no amide hydrogen to form a hydrogen bond tags, self-cleaving tags enable fusion protein purification,
with other AAs, it can avoid ordered structures and pre- cleavage and target protein separation to be achieved in a
vent interactions between the linkers and neighboring single step [332].
domains [324, 329]. Therefore, Pro resides in (XP)n link- In the case of Intein-tag, the target protein is fused to
ers can increase the linker stiffness and effectively sepa- the N-terminus of intein [e.g., VMA intein from Saccha
rate neighboring domains. romyces cerevisiae (51kDa) or DnaB intein from Synecho
cystis sp. strain PCC6803 (17kDa)], whose C-terminus is
3.5.2.5Sitespecific, cleavable peptide linkers Genetic conjugated with an affinity tag (Fig.26a). Intein-mediated
fusion technology provides an effective means for recom- site-specific cleavage can be triggered by thiol reagents,
binant protein expression and purification. A comprehen- such as dithiothreitol or -mercaptoethanol.
sive review of affinity tags can be found elsewhere [292, As for SrtA-tag, the fusion protein consists of an N-ter-
330]. Examples of affinity tags include poly-His, FLAG, minal affinity tag, a SrtA catalytic core, the LPXTG motif
HA, strep II, the calmodulin-binding peptide and the chi- and the target protein (Fig. 26b). On-resin cleavage can
tin-binding domain. These tags specifically interact with be induced by incubation in a Ca2+ ion-containing buffer,
their partner molecules and allow the fused protein to be and the released target protein, with an extra Gly residue
captured by corresponding partner molecule-modified at its N-terminus, can then be collected. However, this
matrices. In most cases, the tags are removed from the system has a potential drawback. Although the activity of
fusion proteins after an affinity tag-assisted purification SrtA from S. aureus is inducible by Ca2+ ions and mod-
process to obtain the final product consisting of pure erate conditions, it is not completely suppressed during
target protein. This is usually achieved by enzymatic or protein expression because abundant soluble M g2+ ions
3 4
chemical cleavage at the junction between the tag and the (10 - to 1 a2+ ions)
0 -fold greater in concentration than C
2+
target protein. Endoproteases commonly used to cleave in the cytosol can partly replace C a ions in function
Nagamune Nano Convergence (2017) 4:9 Page 40 of 56

[333], which causes unwanted fusion cleavage at an early

stage.
a The FrpC module is an iron-regulated protein pro-
duced by the gram-negative bacterium Neisseria menin
b gitides. The fusion construct contains the target protein,
which is at the N-terminal moiety, and the affinity-tagged
c self-processing module (SPM) (Fig.26c). The DNA cod-
ing sequence for the first four AAs of the SPM, which
are Asp-Pro-Leu-Ala, contains an NheI restriction site
d
that can be used for cloning. The Ca2+ ion-addition
induces SPM-mediated cleavage, resulting in the release
e of the target protein with an extra Asp residue at the
C-terminus.
Vibrio cholerae secretes a toxin with large, multifunc-
tional, auto-processing repeats; this toxin undergoes pro-
teolytic cleavage during translocation into host cells. The
proteolysis of the toxin is mediated by a conserved inter-
nal Cys protease domain (CPD), which is activated upon
the binding of the small molecule inositol polyphosphate
(IP6). Affinity-tagged CPD can be fused to the C-ter-
f minus of the target protein (Fig. 26d). The IP6-addition
triggers CPD-mediated cleavage, which allows the tar-
get protein to be released. Depending on the cloning site
used, one or more additional residues may be appended
to the C-terminus of the target protein.
Other applications of cleavable linkers are drug deliv-
ery systems to release free functional units of fusion pro-
teins invivo. These linkers are designed to cleave under
specific conditions, such as the presence of reducing
reagents or proteases. This linker system enables fusion
proteins to reduce steric hindrance and improve both
the independent actions and bioactivities of individual
Fig.26 Schematic representation of the construction of self- functional units after in vivo cleavage. The reduction of
cleaving fusion systems. Filled triangle indicates cleavage sites and X
disulfide bonds in vivo has been widely applied for the
stands for any AA. a The construct of the original C-terminal intein
fusion in which the target protein is fused to the N-terminus of the release of payloads from drug delivery systems fabricated
CBD-tagged intein. b The SrtA fusion construct that contains an by chemical conjugation technologies. Similarly, disulfide
N-terminal affinity-tag, SrtA catalytic core, the LPXTG motif and the linkers cleavable in vivo were designed for recombinant
target protein. Cleavage at the LPXTG site allows the release of the fusion proteins [334, 335]. One such disulfide linker
target protein with an extra N-terminal glycine. c The FrpC fusion
(LEAGCKNFFPRSFTSCGSLE) is based on a dithi-
construct that consists of the target protein and the affinity-tagged
SPM. Cleavage at the AspPro site (the first two AAs of SPM) results in ocyclopeptide containing an intramolecular disulfide
the release of the target protein with an extra aspartate residue at its bond formed between two Cys residues on the linker, as
C-terminus. d The CPD fusion construct in which the affinity-tagged well as a thrombin recognition sequence (PRS) between
CPD is fused to the C-terminus of the target protein. The VD double the two Cys residues (Fig. 26e). Another disulfide linker
residue in the linker sequence comes from the SalI restriction site
(CRRRRRREAEAC) also contains an intramolecular
used for cloning whereas ALADGK are residues contained within the
CPD. e The dithiocyclopeptide linker with one protease-sensitive site. disulfide bond and a peptide sequence sensitive to the
The fusion protein is linked via a dithiocyclopeptide linker containing secretion signal-processing proteases of the yeast secre-
a thrombin-specific sequence, PRS. The design of dithiocyclopeptide tory pathway. During protein expression, this linker is
linker was based on the structure of the cyclopeptide, somatostatin, first cleaved by the protease Kex2 at CRRRRRREAEAC,
with the replacement of AA residues 810, WKT, by a thrombin-
followed by the removal of the dipeptides RR and EA by
specific cleavage sequence, PRS. f The dithiocyclopeptide linker with
three secretion signal processing protease-sensitive sites. The fusion the secretion signal-processing proteases Kex1 and Ste13
protein is linked via a dithiocyclopeptide linker containing Kex1, Kex2 (CRRRRRR, EAEAC), respectively (Fig. 26f ). As
and Ste13-specific cleavage sequences. Kex2 cleaves RRE. Kex1 and a result, the AAs between the two Cys residues in the
Ste13 remove C-terminal RR and N-terminal EA, respectively linker were completely removed during secretion, and
Nagamune Nano Convergence (2017) 4:9 Page 41 of 56

the disulfide linked fusion protein was directly expressed (SpA) between the SpG and the (G4S) linker successfully
by Pichia pastoris. recovered the binding affinity of SpG to the CH1 domain
of IgG [339].
3.5.2.6 The effect oflinker composition, flexibility/rigidity Fusion protein pairs for noncompetitive and homoge-
and length on the functions and conformations of fusion neous immunoassays were developed by optimizing the
proteins The folding, stability, proteolytic sensitivity and flexible G4S linker length of each fusion protein. This
function of fusion proteins might be affected by the AA assay system is based on the antigen-dependent reas-
composition and the flexibility/rigidity and length of the sociation of antibody variable regions (VH, VL) and the
peptide linkers. For example, fusion proteins consisting subsequent complementation of the -Gal domains
of a cellulose-binding domain of Neocallimastix patri and . The best pair was found to be V H-(G4S)2-
ciarum cellulase A (Cel6A) and lipase B from Candida and VL-(G4S)1-, which, at its optimal concentration,
antarctica were constructed by connecting two functional showed a 2.5-fold increase in -Gal activity upon antigen
units with different linker peptides (444 AA residues, addition [340].
different Asn residue numbers and positions for poten- Chimeric receptors (chimeras of anti-fluorescein
tial N-glycosylation sites) derived from the natural pep- (FL) scFv and an engineered c-Mpl receptor possessing
tide linker contained in Cel6A. Analyses of linker stability only signaling mediator STAT3-binding motifs) were
toward proteolysis and the cellulose-binding activity and designed by changing the peptide linker length between
lipase activity of the fusion proteins were conducted; the the binding motifs of JAK and STAT3 using flexible link-
results revealed that fusion proteins with shorter linkers ers (G4S)n (n=0, 3, 6, 9). The activation level of STAT3
(416 AA residues) were more stable against proteolysis was quantitatively evaluated by detecting the level of
but had slightly lower cellulose-binding capacities than phosphorylated STAT3 after the stimulation of chimeric
those containing longer linkers. However, all fusion pro- receptor-expressing cells with FL-labeled bovine serum
teins retained the lipase-specific activity of the wild-type albumin (BSA-FL). The results showed that the STAT3
protein [336]. activation levels were 0.8-, 1.5- and 1.4-fold greater with
Bifunctional fusion proteins composed of the catalytic (G4S)3, (G4S)6 and (G4S)9, respectively, than without a
domains of endoglucanase (Endo5A) and -glucosidase linker. Therefore, changes in the distance from the JAK-
(Gluc1C) from a Paenibacillus strain were constructed by binding domain to the STAT3-binding motif exerted
changing the connection order of two domains and link- relatively minor effects on the phosphorylation level of
ing them with flexible peptide linkers of different lengths STAT3 [341].
(G4S)n (n=03). The results indicated that the substrate Helical poly-Ala linkers (Ala)n (n = 04) were
affinity Km and catalytic efficiency kcat/Km of Gluc1C inserted between the transmembrane and intracellular
were sensitive to its position, as it showed a decline in domains of a chimeric receptor (a tandem fusion pro-
both affinity and catalytic efficiency when Gluc1C was tein of anti-FL scFv/intracellular domain-truncated
placed at the N-terminus of the fusion protein. However, EpoR/gp130 intracellular domain), and the effect of
there was no direct relationship of linker length with linker length on cell proliferation was investigated by
either Endo5A or Gluc1C activity [337]. stimulating chimeric receptor-expressing cells with
Tandem fusion proteins of human serum albumin and BSA-FL. A periodic enhancement in cell proliferation
onconase (ONC) with flexible linkers (G4S)n (n = 03) was induced by the insertion of one to four Ala resi-
were constructed and expressed in P. pastoris. The dues. The chimeric receptors with linkers (Ala)n (n=0,
expression level of the fusion proteins had no relationship 1) transduced a growth signal, while growth activity was
with the linker length. However, while the ONC moiety lost when (Ala)n (n = 24) linkers were inserted. Fur-
of the fusion protein without a linker (n=0) showed no thermore, the extracellular EpoR D1 domain-truncated
cytotoxicity toward tumor cells, this gradually improved chimeric receptor showed different patterns in the
with increasing linker length [338]. periodic enhancement of cell proliferation by the inser-
For the development of a bifunctional immunoreagent, tion of one to four Ala residues. In this case, the chi-
the B1 domain of Streptococcal protein G (SpG), which meric receptors with linkers (Ala)n (n=0, 3, 4) failed to
binds to the Fc region and CH1 domain of IgG, was fused transduce a growth signal, whereas growth activity was
with luciferase from Vargula hilgendorfii (Vluc) using restored when one or two Ala residues were inserted.
flexible peptide linkers (G4S)n (n = 01). The resulting These results clearly demonstrate the importance of
fusion protein, SpG-(G4S)n-Vluc, retained the biolumi- intracellular domain orientation for the activation of
nescence activity of the Vluc moiety but lost the bind- chimeric receptors, which is readily controlled by the
ing affinity of SpG to IgG. However, inserting the three 109 rotation of the -helix Ala linker with each incre-
-helices bundle D domain of protein A from S. aureus ment of one Ala residue [342].
Nagamune Nano Convergence (2017) 4:9 Page 42 of 56

To construct a ligand-inducible scFv dimer, anti-ErbB2 heterotrimer PCNA from Sulfolobus solfataricus to form
scFv was fused with FKBPF36V, which is a mutant of FK- a self-assembling scaffold [111]. To enhance the activity
binding protein 12 that can be dimerized by the syn- of this self-assembled multienzyme complex, the peptide
thetic homodimeric ligand AP20187. The three kind linker connecting PdX with PCN2 was optimized using
of linkers, i.e., flexible (G4S)3, rigid -helix (EA3K)3 and various peptide linkers, such as flexible linkers (G4S)n
DKTHCP(G4S)2, derived from the hinge region of IgG (n=16), helical and rigid Pro-rich linkers (G4S(P5)n
were inserted between scFv and FKBPF36V, and the effect G4S) (n = 15) and other linkers (G4SLVPRGSG4S).
of linker properties on the activity of the fusion protein Although the activity was affected by the lengths of both
dimer, which can dimerize the artificial chimeric receptor the rigid Pro-rich linkers and the flexible linkers, the
ErbB2-gp130 expressed on the cell surface and induce cell Pro-rich linkers provided the greatest activity enhance-
proliferation signaling from the dimerized chimeric recep- ment. The optimized Pro-rich linker (G4S(P5)4G4S)
tor, were investigated. The results showed that the fusion enhanced the activity by 1.9-fold compared with the
protein with the hinge linker was the best for activating G4SLVPRGSG4S linker, while the (G4S)n (n = 16)
ErbB2-gp130 chimera-induced cell proliferation [320]. linker did not yield activity higher than the maximum
It has been demonstrated that the selective complex activity of the optimized Pro-rich linker. Both peptide
formation of P450cam with its redox partner proteins, linker rigidity/flexibility and length were found to be
PdX and PdR, can be achieved by fusing each com- important for enhancing overall multienzyme complex
ponent to the C-terminus of a different subunit of the activity (Fig.27) [343].

Fig.27 Optimization of the PCNA2-PdX fusion protein linker in PUPPET. a P450cam oxidation activities of the PUPPET linker variants, PUPPET-Pn
(n=15). b P450cam oxidation activities of the PUPPET linker variants, PUPPET-Gn (n=16). c A docking model of P450cam and PdX. d Spatial
arrangement of P450cam and the PCNA ring when the PdX-binding site of P450cam faces in the same direction to the PCNA ring. e Spatial arrange-
ment of P450cam and the PCNA ring when the PdX-binding site of P450cam faces in a perpendicular direction to the PCNA ring (Figures repro-
duced from Ref. [343])
Nagamune Nano Convergence (2017) 4:9 Page 43 of 56

The tandem fusion proteins -glucanase (Gluc)/ than the SL, again indicating that the flexible linkers had
xylanase (Xyl) were constructed using peptide linkers, a random, coiled conformation [346]. The real insitu con-
such as flexible linkers (G4S)n (n = 03), -helical link- formations of these fusion proteins and structures of the
ers (EA3K)n (n = 03) and others (MGSSSN designed linkers were further analyzed using synchrotron X-ray
using the software of the web server LINKER [344], and small-angle scattering (SAXS). The SAXS experiments
TGSRKYMELGATQGMGEALTRGM derived from the indicated that the fusion proteins with flexible link-
two -helix bundle of Humicola insolens endocellulase). ers assume an elongated conformation (Fig. 28a) rather
The effects of the linkers on the thermal stability and cat- than the most compact conformation (Fig.28b) and that
alytic efficiency of both enzymes were analyzed. The Gluc the distance between EBFP and EGFP was not regulated
moieties of most fusion constructs showed greater stabil- by the linker length. On the other hand, fusion proteins
ity at 4060C than did the parental Gluc and the linker- with helical linkers [LA(EA3K)nAAA n=4, 5] were more
free fusion protein. All the Xyl moieties showed thermal elongated than were those with flexible linkers, and the
stabilities similar to that of the parental Xyl, at 60C. It high-resolution models (Fig. 29) showed that the helical
was also revealed that the catalytic efficiencies of the Gluc linkers connected the EBFP and EGFP domains diago-
and Xyl moieties of all the fusion proteins were 3.04- to nally (Fig. 28c) rather than longitudinally (Fig. 28d).
4.26-fold and 0.82- to 1.43-fold those of the parental moi- However, in the case of the shorter helical linkers (n=2,
eties, respectively. The flexible linker (G4S)2 resulted in 3, especially n = 2), fusion protein multimerization was
the best fusion proteins, whose catalytic efficiencies were observed. Since most residues of the short helical linkers
increased by 4.26-fold for the Gluc moiety and by 1.43- are situated closer to the two domains of the fusion pro-
fold for the Xyl moiety. The Gluc and Xyl moieties of the tein, the charged residues, Glu and Lys in the (EA3K) unit
fusion protein with the rigid linker (EA3K)3 also showed are likely to form ion pairs with the oppositely charged
3.62- and 1.31-fold increases in catalytic efficiency [345].
Aiming to clarify the criteria for designing peptide
linkers for the effective separation of the domains in
a bifunctional fusion protein, a systematic investiga-
tion was carried out. As a model, the fusion proteins
of two Aequorea GFP variants, enhanced GFP (EGFP)
and enhanced blue fluorescent protein (EBFP), were
employed. The secondary structure of the linker and the
relative distance between EBFP and EGFP were exam-
ined using circular dichroism (CD) spectra and fluo-
rescent resonance energy transfer (FRET), respectively.
The following AA sequences were designed and utilized
as peptide linkers: a short linker (SL); LAAA (4 AAs)
(derived from the cleavage sites for HindIII and NotI);
flexible linkers (G4S)nAAA (n = 3, 4); -helical link-
ers LA(EA3K)nAAA (n=35); and a three -helix bun-
dle from the B domain of SpA (LFNKEQQNAFYEILH
L P N L N E E Q R N G F I Q S L K D D P S Q S A N L L A E A K-
KLNDAQAAA). The differential CD spectra analysis
suggested that the LA(EA3K)nAAA linkers formed an
-helix and that the -helical contents increased as the
number of the linker residues increased. In contrast, the
flexible linkers formed a random, coiled conformation.
The FRET from EBFP to EGFP decreased as the length
of the helical linkers increased, indicating that distances
increased in proportion to the length of the linkers. The Fig.28 Schematic illustrations of various conformations of the fusion
results showed that the helical linkers could effectively proteins. a EBFP (blue) and EGFP (green) are situated in a straight
separate the neighboring domains of the fusion protein. line, with the flexible linker (red) between the two domains. b EBFP
and EGFP reside side by side, for the most compact conformation
In the case of the fusion proteins with the flexible linkers,
with the flexible linker. c The helical linker connects EBFP and EGFP
the FRET efficiency was not sensitive to linker length and diagonally. d The helical linker and the long axes of EBFP and EGFP
was highly comparable to that of the fusion proteins with are situated in a straight line (Figure adapted with permission from:
the SL, although the flexible linkers were much longer Ref. [347]. Copyright (2004) John Wiley & Sons)
Nagamune Nano Convergence (2017) 4:9 Page 44 of 56

rigid helical linkers to control the distance and reduce the

interference between the domains [347]. This study is the
first example of modeling in situ fusion protein confor-
mations and linker structures by combining SAXS data of
fusion proteins, structural information of the functional
units from the Brookhaven Protein Data Bank (PDB),
and molecular dynamics calculations of peptide linker
structures. Recently, this modeling method was applied
to evaluate the in situ conformations and structures of
fusion proteins composed of a de novo two-helix bundle
protein and a single trimeric foldon domain of fibritin
from the bacteriophage T4 connected by a short peptide
linker (KLAAA). Size exclusion chromatography, multi-
angle light scattering, analytical ultracentrifugation, and
SAXS analyses indicated that the small (S form), middle
(M form), and large (L form) forms of the fusion protein
oligomers exist as 6-, 12-, and 18-mers, respectively. The
SAXS data further suggested that the S and M forms have
barrel- and tetrahedron-like shapes, respectively [348].
The selection of a suitable peptide linker, which allows a
desirable conformation and interaction among functional
units in fusion proteins, is key to the successful design of
fusion proteins. Generally, rigid linkers exhibit relatively
stiff structures by adopting -helical structures or by con-
taining multiple Pro residues with the cis isomer of the
peptide bond. Under many circumstances, they can sepa-
rate the functional domains in fusion protein more effi-
ciently than do flexible linkers. The length of the linkers
can be easily adjusted by changing the linker-unit repeat-
number, such as (EA3K), to achieve an optimal distance
between functional units. As a result, when the spatial
separation of the functional units is critical to avoid steric
hindrance and to preserve the folding, stability and activ-
ity of each unit in the fusion proteins, rigid linkers would
be selected. However, there are other types of fusion
Fig.29 High-resolution models (cartoon representation) of the EBFP
and EGFP connected with the helical linkers. BH4G and BH5G proteins, in which functional units are required to have
indicate EBFPLA(EA3K)nAAAEGFP (n=4, 5), respectively. Low-res- a certain degree of movement/interaction or a precise
olution models based on only SAXS data are shown as wire-frames. proximal spatial arrangement and orientation to form
The linker and the two domains are modeled and two different views complexes. In such cases, flexible linkers are often cho-
are shown (Figure reproduced with permission from: Ref. [347]. Copy-
sen because they can serve as a passive linker to maintain
right (2004) John Wiley & Sons)
a distance or to adjust the proximal spatial arrangement
and orientation of functional units. However, optimiz-
ing the peptide linker sequence and predicting the spatial
residues on the top surfaces of EBFP and EGFP. Conse- linker arrangement and orientation are more difficult for
quently, this ion pairs formation causes destabilization flexible linkers than for rigid linkers. Current strategies
of the short helix and melted helix linkers may act as are mostly empirical and intuitive and have a high uncer-
attractants for the attachment of neighboring molecules tainty. Therefore, computational simulation technologies
due to their charges and hydrophobicity, thereby caus- for predicting fusion protein conformations and linker
ing multimerization of fusion protein. On the other hand, structures would potentially encourage rational flexible
in the case of the fusion protein with the longer helical linker design with improved success rates.
linkers (n=4, 5), the linkers retained the -helix struc-
ture and could solvate monomeric fusion proteins. These 3.5.2.7Rational algorithms and software for designing
results clearly suggested the outstanding ability of the linker sequences and structures The rational design of
Nagamune Nano Convergence (2017) 4:9 Page 45 of 56

fusion proteins with desired conformations, proper- bulky hydrophobic residues) to further select their link-
ties and functions is a challenging issue. Most current ers of interest. The output of LINKER included a list of
approaches to linker selection and design processes for peptide sequences with the specified lengths, sequence
fusion proteins are still largely dependent on experience characteristics and chemical features of every linker
and intuition; such selection processes often involve sequence shown by hydrophobicity plots [344, 349].
great uncertainty, particularly in the case of longer flex- However, although the PDB database has expanded tre-
ible linker selection, and many unintended consequences, mendously during the last decade, no further updates or
such as the misfolding, low yield and reduced functional improvements were made to the LINKER website since it
activity of fusion proteins may occur. This is mostly was created, and it is no longer accessible.
because of our limited understanding of the sequence The web-based program LinkerDB (http://www.ibi.
structurefunction relationships in these fusion proteins. vu.nl/programs/linkerdbwww/) also provides a database
To overcome this problem, the computational prediction containing linker sequences with various confirmations
of fusion protein conformation and linker structure can and a search engine. The search algorithm accepts several
be considered a cost-effective alternative to experimental query types (e.g., PDB code, PDB header, linker length,
trial-and-error linker selection. Based on the structural secondary structure, sequence or solvent accessibility).
information of individual functional units and linkers The program can provide the linker sequences fitting the
(either from the PDB or homology modeling), consider- searching criteria as well as other information, such as
able progress has been made in predicting fusion protein the PDB code and a brief description of the source pro-
conformations and linker structures [290]. Approaches tein, the linkers position within the source protein, linker
for the design or selection of flexible linker sequences length, secondary structure, and solvent accessibility.
to connect two functional units can be categorized into Users can search for sequences with desired properties
two groups. The first group comprises library selection- and obtain candidate sequences from natural multid-
based approaches, in which a candidate linker sequence omain proteins [329].
is selected from a loop sequence library without consid- Another server website for facilitating linker selec-
eration of the conformation or placement of functional tion and fusion protein modeling is SynLinker (http://
units in the fusion proteins. The second group comprises bioinfo.bti.a-star.edu.sg/linkerdb). It contains informa-
modeling-based approaches, in which functional unit tion regarding 2260 linkers, consisting of natural linkers
conformation and placement and linker structure and AA extracted from multidomain proteins in the latest PDB,
composition would be optimized by simulation. as well as artificial and empirical linkers collected from
Regarding the first approach, a computer program the literature and patents. A user may specify multiple
called LINKER was developed. This web-based program query criteria to search SynLinker, such as the PDB ID
(http://astro.temple.edu/feng/Servers/Bioinformatic- of the source proteins, protein names, the number of AA
Servers.htm) automatically generated a set of peptide residues in a linker, and/or the end-to-end distance of a
sequences based on the assumption that the observed linker conformation in Angstroms (). Additionally, the
loop sequences in the X-ray crystal structures or the user can choose a linker starting residue, ending residue,
nuclear magnetic resonance structures were likely to AA enrichment, AA depletion and/or protease sensitiv-
adopt an extended conformation as linkers in a fusion ity as a desired linker property in the recombinant fusion
protein. Loop linker sequences of various lengths were protein. Once a query is submitted, both the natural and
extracted from the PDB, which contains both globu- artificial/empirical linkers in SynLinker are searched
lar and membrane proteins, by removing short loop simultaneously, yielding a list of potential linker candi-
sequences less than four residues and redundant dates satisfying the desired selection criteria together
sequences. LINKER searched its database of loop linker with information about the AA composition radar chart
sequences with user-specified inputs and outputted sev- and the conformation of the selected linker, as well as the
eral candidate linker sequences that meet the criteria. fusion protein structure and hydropathicity plot [350].
The basic input to the program was the desired length As for modeling-based approaches, the conformation
of the linker, expressed as either the number of residues and placement of functional units in fusion proteins,
or a distance in angstroms. Additional input parameters of which 3D structures are available from the PDB or
included potential cleavage sites for restriction endonu- homology modeling, can be predicted by computer-aided
cleases or proteases to avoid such that the selected link- modeling. A modeling tool known as FPMOD was devel-
ers would be resistant against the restriction enzymes oped and can generate fusion protein models by con-
and the specified protease during the DNA cloning and necting functional units with flexible linkers of proper
protein purification process, respectively. The users could lengths, defining regions of flexible linkers, treating
also include AA composition preferences (e.g., eliminate the structures of all functional units as rigid bodies and
Nagamune Nano Convergence (2017) 4:9 Page 46 of 56

rotating each of them around their flexible linker to pro- of IL-2 with the N-terminus of CelTOS. The tertiary struc-
duce random structures. This tool can extensively test the tures of the fusion proteins were predicted in silico by
conformational space of fusion proteins and finally gen- the I-TASSER online server (http://zhanglab.ccmb.med.
erate plausible models [351]. This tool has been applied umich.edu/I-TASSER/) [355]. The model with the high-
to designing FRET-based protein biosensors for C a2+ ion est confidence score (C-score: a scoring function based
by qualitatively predicting their FRET efficiencies, and on the relative clustering structural density and the con-
the predictions strongly agreed with the experimental sensus significance score of multiple threading templates)
results [352]. was considered as the best model. The selected structures
A similar modeling tool was developed for assem- of the fusion proteins with different linkers were then
bling structures of isolated functional units to constitute validated and analyzed using a Ramachandran plot assess-
multidomain fusion proteins. However, this approach of ment [356]. All the results verified the (G4S)3 linker as the
assembling functional units is different from the method most suitable for separating these proteins [357].
of testing conformational space. In this method, an The important issue to be addressed in structure pre-
ab initio protein-modeling method is utilized to predict diction is the method of searching the large and complex
the tertiary structure of fusion proteins, the conforma- conformational space to rapidly reach at the minimum
tion and placement of functional units and the linker energy structure, which is presumed to be the native fold.
structure. This method samples the degrees of free- The genetic algorithm combined with an extremely fast
dom of the linker (in other words, domain assembly as technique to search the conformation space exhaustively
a linker-folding problem) rather than those of the rigid and build a library of possible low-energy local structures
bodies, as adopted in FPMOD. The method consists of for oligopeptides (i.e., the MOLS method), was applied
an initial low-resolution search, in which the conforma- to the protein structure prediction. At the first step, the
tional space of the linker is explored using the Rosetta protein sequence was divided into short overlapping
de novo structure prediction method. This is followed by fragments, and then their structural libraries were built
a high-resolution search, in which all atoms are treated using the MOLS method. At the second step, the genetic
explicitly, and backbone and side chain degrees of free- algorithm exploited the libraries of fragment structures
dom are simultaneously optimized. The obtained models and predicted the single best structure for the protein
with the lowest energy are often very close to the correct sequence. In the application of this combined method
structures of existing multidomain proteins with very to peptides and small proteins, such as the avian pancre-
high accuracy [353]. atic polypeptide (36 AAs), the villin headpiece (36 AAs),
A method called pyDockTET (tethered-docking) uses melittin (26 AAs), the transcriptional activator Myb (52
rigid-body docking to generate domaindomain com- AAs) and the Trp zipper (16 AAs), it could predict their
plexes that are scored by the electrostatic and desolvation near-native structures [358].
energy terms, as well as a pseudo-energy term reflecting The computer-aided rational design methods for fusion
restraints from linker end-to-end distances; in this man- proteins are promising because these methods allow us
ner, near-native pair-wise domain poses are selected. to easily predict the desired conformation and placement
The optimal linker sequence length (in the number of of the functional units and linker structures of fusion
residues) with the linker ends (defined as the distance proteins, and consequently select suitable candidate
between the C atoms of the two ends of a linker) is linker sequences. However, it is difficult to determine
selected from a flexible linker database, which consists the unique conformation of flexible linkers due to many
of 542 linkers with sequence lengths ranging from 2 to local minima in free energy. Furthermore, if changes in
29 AAs derived from the inter-domain linkers of multid- the conformation or arrangement of functional units are
omain structures in the PDB [354]. essential to display their activity, the linker conformation
A fusion protein consisting of a protein called cell-tra- should also be changed to allow the movement of func-
versal protein for ookinetes and sporozoites (CelTOS) tional units, e.g., the N-terminal ATP-binding domain
antigen from Plasmodium falciparum (the deadliest of and unfolded substrate protein-binding domain con-
malaria species) and human IL-2 as an adjuvant was nected with a hydrophobic peptide linker in heat shock
designed to develop a candidate vaccine against malaria. protein 70 [359]. This complicated conformational tran-
CelTOS and IL-2 were linked together directly or by using sition issue makes it difficult to design optimum linkers
different flexible linkers, including (G)8, (G4S) and (G4S)3. for fusion proteins with multiple conformations. There-
Since the N-terminus of IL-2 and the C-terminus of Cel- fore, the rational design of fusion proteins with desired
TOS are critical to preserve their stability and bioactivity, properties and predictable behavior remains a daunting
the fusion protein was designed by linking the C-terminus challenge.
Nagamune Nano Convergence (2017) 4:9 Page 47 of 56

4Conclusion exclusive, and the relative weaknesses of each approach

This review highlighted some of the recent develop- will be solved in the near future. Given the rapid recent
ments in studies related to nanobio/bionanotechnology, progress in the biomolecular engineering and nanotech-
including the applications of engineered biological mol- nology fields, the design of completely novel biomate-
ecules combined with functional nanomaterials in ther- rial-based molecular devices and systems with functions
apy, diagnosis, biosensing, bioanalysis and biocatalysis. tailored for specific applications seems to be much easier
Furthermore, this review focused on recent advances in and more feasible than before.
biomolecular engineering for nanobio/bionanotechnol-
Competing interests
ogy, such as nucleic acid engineering, gene engineering, The author declares that he has no competing interests.
protein engineering, chemical and enzymatic conjugation
technologies, and linker engineering. Funding
This research was supported partly by Grants-in-Aid for Scientific Research (A)
Based on creative chemical and biological technolo- from Japan Society for the Promotion of Science (JSPS) (15H02319), the Center
gies, manipulation protocols for biomolecules, espe- for NanoBio Integration (CNBI) in The University of Tokyo, and Translational Sys-
cially nucleic acids, peptides, enzymes and proteins, tem Biology and Medicine Initiative from The Ministry of Education, Culture,
Sports, Science and Technology (MEXT).
were described. We also summarized the main strate-
gies adopted in nucleic acid engineering, gene engi-
Publishers Note
neering, protein engineering, chemical and enzymatic Springer Nature remains neutral with regard to jurisdictional claims in pub-
conjugation technologies and linker engineering. lished maps and institutional affiliations.
Nucleic acid engineering based on the base-pairing
Received: 17 February 2017 Accepted: 29 March 2017
and self-assembly characteristics of nucleic acids was
highlighted as a key technology for DNA/RNA nano-
technologies, such as DNA/RNA origami, aptamers,
ribozymes. Gene engineering includes direct manipu-
lation technologies for genes, such as gene mutagen- References
1. W.C.W. Chan, Bionanotechnology progress and advances. Biol. Blood
esis, DNA sequence amplification, DNA shuffling and Marrow Transplant. 12, 8791 (2006). doi:10.1016/j.bbmt.2005.10.004
gene fusion, which are powerful tools for generating 2. M.K. Kovarik, S.C. Jacobson, Nanofluidics in lab-on-a-chip devices. Anal.
enzymes, proteins, entire metabolic pathways, or even Chem. 81, 71337140 (2009). doi:10.1021/ac900614k
3. J. West, M. Becker, S. Tombrink, A. Manz, Micro total analysis systems:
entire genomes with desired or improved properties. latest achievements. Anal. Chem. 80, 44034419 (2008). doi:10.1021/
Two general strategies for protein engineering, i.e., ac800680j
rational protein design and directed evolution (i.e., 4. P. Yager, T. Edward, E. Fu, K. Helton, K. Nelson, M.R. Tam, B.H. Weigl,
Microfluidic diagnostic technologies for global public health. Nature
high-throughput library screening- or selection-based 442, 412418 (2006). doi:10.1038/nature05064
approaches) were discussed. Conjugation technolo- 5. C.D. Chin, V. Linder, S.K. Sia, Commercialization of microfluidic point-of-
gies to site-specifically modify proteins with diverse care diagnostic devices. Lab Chip 12, 21182134 (2012). doi:10.1039/
c2lc21204h
natural and unnatural functionalities have been devel- 6. A. Gencoglu, A.R. Minerick, Electrochemical detection techniques in
oped in the last two decades. These technologies range micro- and nanofluidic devices. Microfluid. Nanofluid. 17, 781807
from classical chemical bioconjugation technologies, (2014). doi:10.1007/s10404-014-1385-z
7. M.P. McRae, G. Simmons, J. Wong, J.T. McDevitt, Programmable bio-
bioorthogonal chemical conjugations, protein chemical nanochip platform: a point-of-care biosensor system with the capacity
ligations and enzymatic conjugations, which were over- to learn. Acc. Chem. Res. 49, 13591368 (2016). doi:10.1021/acs.
viewed. Linker engineering for controlling the distance, accounts.6b00112
8. S. Howorka, Z. Siwy, Nanopore analytics: sensing of single molecules.
orientation and interaction between functional com- Chem. Soc. Rev. 38, 23602384 (2009). doi:10.1039/b813796j
ponents crosslinked in conjugates is also an important 9. O.V. Salata, Applications of nanoparticles in biology and medicine. J.
technology. The design and optimization strategies of Nanobiotechnol. 2, 3 (2004). doi:10.1186/1477-3155-2-3
10. W.T. Liu, Nanoparticles and their biological and environmental applica-
chemical and biological linkers, such as oligonucleo- tions. J. Biosci. Bioeng. 102, 17 (2006). doi:10.1263/jbb.102.1
tides and polypeptides, were overviewed. 11. M. Rawat, D. Singh, S. Saraf, S. Saraf, Nanocarriers: promising vehicle for
A variety of strategies are now available for design- bioactive drugs. Biol. Pharm. Bull. 29, 17901798 (2006). doi:10.1248/
bpb.29.1790
ing and fabricating novel nanobiomaterials with highly 12. D. Knopp, D.P. Tang, R. Niessner, Bioanalytical applications of biomole-
ordered dimension and complexity based on biomolecu- cule -functionalized nanometer-sized doped silica particles. Anal. Chim.
lar self-assembly characteristics governed by molecu- Acta 647, 1430 (2009). doi:10.1016/j.aca.2009.05.037
13. R.A. Petros, J.M. Desimone, Strategies in the design of nanoparticles
lar interactions among nucleotides, peptides, proteins, for therapeutic applications. Nat. Rev. Drug Discov. 9, 615627 (2010).
lipids and small ligands, each of which focuses on design doi:10.1038/nrd2591
simplicity, high structural and functional control, or 14. M. Colombo, S. Carregal-Romero, M.F. Casula, L. Gutierrez, M.P. Morales,
I.B. Boehm, J.T. Heverhagen, D. Prosperi, W.J. Parak, Biological applica-
high fabrication accuracy [1620, 106, 127, 132, 360 tions of magnetic nanoparticles. Chem. Soc. Rev. 41, 43064334 (2012).
365]. Fundamentally, these properties are not mutually doi:10.1039/c2cs15337h
Nagamune Nano Convergence (2017) 4:9 Page 48 of 56

15. D.E. Lee, H. Koo, I.C. Sun, J.H. Ryu, K. Kim, I.C. Kwon, Multifunctional 38. K. Chen, X. Chen, Integrin targeted delivery of chemotherapeutics.
nanoparticles for multimodal imaging and theragnosis. Chem. Soc. Rev. Theranostics 1, 189200 (2011). doi:10.7150/thno/v01p0189
41, 26562672 (2012). doi:10.1039/c2cs15261d 39. R. Stoltenburg, C. Reinemann, B. Strehlitz, SELEXa (r)evolutionary
16. A.V. Pinheiro, D. Han, W.M. Shih, H. Yan, Challenges and opportunities for method to generate high-affinity nucleic acid ligands. Biomol. Eng. 24,
structural DNA nanotechnology. Nat. Nanotechnol. 6, 763772 (2011). 381403 (2007). doi:10.1016/j.bioeng.2007.06.001
doi:10.1038/NNANO.2011.187 40. X. Wu, J. Chen, M. Wu, J.X. Zhao, Aptamers: active targeting ligands
17. P. Guo, The emerging field of RNA nanotechnology. Nat. Nanotechnol. for cancer diagnosis and therapy. Theranostics 5, 322344 (2015).
5, 833842 (2010). doi:10.1038/NNANO.2010.231 doi:10.7150/thno.10257
18. E.H.C. Bromley, K. Channon, E. Moutevelis, D. Woolfson, Peptide and 41. T.R. Daniels, T. Delgado, G. Helguera, M.L. Penichet, The transfer-
protein building blocks for synthetic biology: from programming rin receptor part II: targeted delivery of therapeutic agents into
biomolecules to self-organized biomolecular systems. ACS Chem. Biol. cancer cells. Clin. Immunol. 121, 159176 (2006). doi:10.1016/j.
3, 3850 (2008). doi:10.1021/cb700249v clim.2006.06.006
19. D. Papapostolou, S. Howorka, Engineering and exploiting protein 42. E.T. Boder, W. Jiang, Engineering antibodies for cancer therapy.
assemblies in synthetic biology. Mol. BioSyst. 5, 723732 (2009). Annu. Rev. Chem. Biomol. Eng. 2, 5375 (2011). doi:10.1146/
doi:10.1039/b902440a annurev-chembioeng-061010-114142
20. S. Mashaghi, T. Jadidi, G. Koenderink, A. Mashaghi, Lipid nanotechnol- 43. C. Bechara, S. Sagan, Cell-penetrating peptides: 20years later, where
ogy. Int. J. Mol. Sci. 14, 42424282 (2013). doi:10.3390/ijms14024242 do we stand? FEBS Lett. 587, 16931702 (2013). doi:10.1016/j.
21. Y. Miura, Y. Hoshino, H. Seto, Glycopolymer nanobiotechnology. Chem. febslet.2013.04.031
Rev. 116, 16731692 (2016). doi:10.1021/acs.chemrev.5b00247 44. A.K. Varkouhi, M. Scholte, G. Storm, H.J. Haisma, Endosomal escape
22. K.Y. Baik, S.Y. Park, S. Namgung, D. Kim, D.G. Cho, M. Lee, S. Hong, pathways for delivery of biological. J. Control. Release 151, 220228
Synthetic nanowire/nanotube-based solid substrates for controlled cell (2011). doi:10.1016/j.jconrel.2010.11.004
growth. Nano Converg. 1, 28 (2014). doi:10.1186/s40580-014-0028-0 45. A.E. Sayed, S. Futaki, H. Harashima, Delivery of macromolecules using
23. M.A. Kafi, H.Y. Cho, J.W. Choi, Engineered peptide-based nanobio- arginine-rich cell-penetrating peptides: ways to overcome endosomal
materials for electrochemical cell chip. Nano Converg. 3, 17 (2016). entrapment. AAPS J. 11, 1322 (2009). doi:10.1208/s12248-008-9071-2
doi:10.1186/s40580-016-0077-7 46. Y.H. Li, J. Wang, M.G. Wientjes, J.L.S. Au, Delivery of nanomedicines to
24. S.T.D. Chueng, L. Yang, Y. Zhang, K.B. Lee, Multidimensional nanoma- extracellular and intracellular compartments of a solid tumor. Adv. Drug
terials forthe control ofstem cell fate. Nano Converg. 3, 23 (2016). Deliv. Rev. 64, 2939 (2012). doi:10.1016/j.addr.2011.04.006
doi:10.1186/s40580-016-0083-9 47. W. Viricel, A. Mbarek, J. Leblond, Switchable lipids: conformational
25. J.H. Choi, J. Lee, W. Shin, J.W. Choi, H.J. Kim, Priming nanoparticle- change for fast pH-triggered cytoplasmic delivery. Angew. Chem. Int.
guided diagnostics and therapeutics towards human organs-on-chips Ed. 54, 1274312747 (2015). doi:10.1002/anie.201504661
microphysiological system. Nano Converg. 3, 24 (2016). doi:10.1186/ 48. D.H. Kim, I. Hwang, Direct targeting of proteins from the cytosol to
s40580-016-0084-8 organelles: the ER versus endosymbiotic organelles. Traffic 14, 613621
26. W. Jo, U.K. Cheang, M.J. Kim, Development of flagella bio-templated (2013). doi:10.1111/tra.12043
nanomaterials for electronics. Nano Converg. 1, 10 (2014). doi:10.1186/ 49. L.D. Field, J.B. Delehanty, Y.C. Chen, I.L. Medintz, Peptides for specifically
s40580-014-0010-x targeting nanoparticles to cellular organelles: Quo vadis? Acc. Chem.
27. G. Faccio, K. Gajda-Schrantz, J. Ihssen, F. Boudoire, Y. Hu, B.S. Mun, D.K. Res. 48, 13801390 (2015). doi:10.1021/ar500449v
Bora, L. Thny-Meyer, A. Braun, Charge transfer between photo- 50. S. Negi, S. Pandey, S.M. Srinivasan, A. Mohammed, C. Guda, LocSigDB:
synthetic proteins and hematite in bio-hybrid photoelectrodes for a database of protein localization signals. Database 2015, 17 (2015).
solar water splitting cells. Nano Converg. 2, 9 (2015). doi:10.1186/ doi:10.1093/database/bav003
s40580-014-0040-4 51. P. Verderio, S. Avvakumova, G. Alessio, M. Bellini, M. Colombo, E. Galbiati,
28. D.B.T.G. Raj, N.A. Khan, Designer nanoparticle: nanobiotechnol- S. Mazzucchelli, J.P. Avila, B. Santini, D. Prosperi, Delivering colloidal
ogy tool for cell biology. Nano Converg. 3, 22 (2016). doi:10.1186/ nanoparticles to mammalian cells: a nanobio interface perspective.
s40580-016-0082-x Adv. Healthcare Mater. 3, 957976 (2014). doi:10.1002/adhm.201300602
29. F. Gu, R. Karnik, A. Wang, F. Alexis, E.L. Nissenbaum, S. Hong, R. Langer, 52. D. Schmaljohann, Thermo- and pH-responsive polymers in drug
O.C. Farokhzad, Targeted nanoparticles for cancer therapy. Nano Today delivery. Adv. Drug Deliv. Rev. 58, 16551670 (2006). doi:10.1016/j.
2, 1421 (2007). doi:10.1016/S1748-0132(07)70083-X addr.2006.09.020
30. M.K. Yu, J. Park, S. Jon, Targeting strategies for multifunctional nano- 53. E. Fleige, M.A. Quadir, R. Haag, Stimuli-responsive polymeric nanocar-
particles in cancer imaging and therapy. Theranostics 2, 344 (2012). riers for the controlled transport of active compounds: concepts and
doi:10.7150/thno.3463 applications. Adv. Drug Deliv. Rev. 64, 866884 (2012). doi:10.1016/j.
31. T. Sun, Y.S. Zhang, B. Pang, D.C. Hyun, M. Yang, Y. Xia, Engineered nano- addr.2012.01.020
particles for drug delivery in cancer therapy. Angew. Chem. Int. Ed. 53, 54. C. Brieke, F. Rohrbach, A. Gottschalk, G. Mayer, A. Heckel, Light-con-
1232012364 (2014). doi:10.1002/anie.201403036 trolled tools. Angew. Chem. Int. Ed. 51, 84468476 (2012). doi:10.1002/
32. P.K. Ba, B.H. Chung, Multiplexed detection of various breast cancer cells anie.201202134
by perfluorocarbon/quantum dot nanoemulsions conjugated with 55. P. Klan, T. Solomek, C.G. Bochet, A. Blanc, R. Givens, M. Rubina, V. Popik,
antibodies. Nano Converg. 1, 23 (2014). doi:10.1186/s40580-014-0023-5 A. Kostikov, J. Wirz, Photoremovable protecting groups in chemistry
33. W.K. Rhim, M. Kim, K.L. Hartman, K.W. Kang, J.M. Nam, Radionuclide- and biology: reaction mechanisms and efficacy. Chem. Rev. 113,
labeled nanostructures for invivo imaging of cancer. Nano Converg. 2, 119191 (2013). doi:10.1021/cr300177k
10 (2015). doi:10.1186/s40580-014-0041-3 56. Y.H. Yang, H. Aloysius, D. Inoyama, Y. Chen, L.Q. Hu, Enzyme-mediated
34. D.E. Owens III, N.A. Peppas, Opsonization, biodistribution, and pharma- hydrolytic activation of prodrugs. Acta Pharmaceutica Sinica B 1,
cokinetics of polymeric nanoparticles. Int. J. Pharm. 307, 93102 (2006). 143159 (2011). doi:10.1016/j.apsb.2011.08.001
doi:10.1016/j.ijpharm.2005.10.010 57. D.H. Nguyen, J.S. Lee, J.W. Bae, J.H. Choi, Y. Lee, J.Y. Son, K.D. Park,
35. M.M. Kemp, R.J. Linhardt, Heparin-based nanoparticles. WIREs Targeted doxorubicin nanotherapy strongly suppressing growth of
Nanomed. Nanobiotechnol. 2, 7787 (2010). doi:10.1002/wnan.068 multidrug resistant tumor in mice. Int. J. Pharm. 495, 329335 (2015).
36. K. Alhareth, C. Vauthier, F. Bourasset, C. Gueutin, G. Ponchel, F. Moussa, doi:10.1016/j.ijpharm.2015.08.083
Conformation of surface-decorating dextran chains affects the pharma- 58. X. Zhao, P. Liu, Biocompatible graphene oxide as a folate receptor-
cokinetics and biodistribution of doxorubicin-loaded nanoparticles. Eur. targeting drug delivery system for the controlled release of anti-cancer
J. Pharm. Biopharm. 81, 453457 (2012). doi:10.1016/j.ejpb.2012.03.009 drugs. RSC Adv. 4, 2423224239 (2014). doi:10.1039/c4ra02466d
37. P.S. Low, W.A. Henne, D.D. Doorneweerd, Discovery and development 59. H. Hosoya, A.S. Dobroff, W.H.P. Driessen, V. Cristini, L.M. Brinker, F.I.
of folic-acid-based receptor targeting for imaging and therapy of Staquicini, M. Card-Vila, S. DAngelo, F. Ferrara, B. Proneth, Y.S. Lin,
cancer and inflammatory diseases. Acc. Chem. Res. 41, 120129 (2008). D.R. Dunphy, P. Dogra, M.P. Melancon, R.J. Stafford, K. Miyazono, J.G.
doi:10.1021/ar7000815 Gelovani, K. Kataoka, C.J. Brinker, R.L. Sidman, W. Arapc, R. Pasqualini,
Nagamune Nano Convergence (2017) 4:9 Page 49 of 56

Integrated nanotechnology platform for tumor-targeted multimodal 79. N.P. Sardesai, J.C. Barron, J.F. Rusling, Carbon nanotube microwell array
imaging and therapeutic cargo release. Proc. Natl. Acad. Sci. USA 113, for sensitive electrochemiluminescent detection of cancer biomarker
18771882 (2016). doi:10.1073/pnas.1525796113 proteins. Anal. Chem. 83, 66986703 (2011). doi:10.1021/ac201292q
60. Z. Zhang, Y. Liu, C. Jarreau, M.J. Welchb, J.S.A. Taylor, Nucleic acid- 80. N.P. Sardesai, K. Kadimisetty, R. Faria, J.F. Rusling, A microfluidic
directed self-assembly of multifunctional gold nanoparticle imaging electrochemiluminescent device for detecting cancer biomarker
agents. Biomater. Sci. 1, 10551064 (2013). doi:10.1039/c3bm60070j proteins. Anal. Bioanal. Chem. 405, 38313838 (2013). doi:10.1007/
61. K. Minamihata, Y. Maeda, S. Yamaguchi, W. Ishihara, A. Ishiwatari, S. s00216-012-6656-5
Takamori, S. Yamahira, T. Nagamune, Photosensitizer and polycationic 81. Y.E. Choi, J.W. Kwak, J.W. Park, Nanotechnology for early cancer detec-
peptide-labeled streptavidin as a nano-carrier for light-controlled pro- tion. Sensors 10, 428455 (2010). doi:10.3390/s100100428
tein transduction. J. Biosci. Bioeng. 120, 630636 (2015). doi:10.1016/j. 82. R.V. Devi, M. Doble, R.S. Verma, Nanomaterials for early detection of
jbiosc.2015.04.001 cancer biomarker with special emphasis on gold nanoparticles in
62. A. Ishiwatari, S. Yamaguchi, S. Takamori, S. Yamahira, K. Minamihata, T. immunoassays/sensors. Biosens. Bioelectron. 68, 688698 (2015).
Nagamune, Photolytic protein aggregates: versatile materials for con- doi:10.1016/j.bios.2015.01.066
trolled release of active proteins. Adv. Healthcare Mater. 5, 10021007 83. X. Yang, X.W. Shi, Y. Liu, W.E. Bentley, G.F. Payne, Orthogonal enzymatic
(2016). doi:10.1002/adhm.201500957 reactions for the assembly of proteins at electrode addresses. Langmuir
63. M. Mishina, K. Minamihata, K. Moriyama, T. Nagamune, Peptide tag- 25, 338344 (2009). doi:10.1021/la802618q
induced horseradish peroxidase-mediated preparation of a streptavi- 84. J.B. Kim, J.W. Grate, P. Wang, Nanobiocatalysis and its potential
din-immobilized redox-sensitive hydrogel. Biomacromol 17, 19781984 applications. Trends Biotechnol. 26, 639646 (2008). doi:10.1016/j.
(2016). doi:10.1021/acs.biomac.6b00149 tibtech.2008.07.009
64. X. Zhang, Q. Guo, D. Cui, Recent advances in nanotechnology applied 85. P. Wang, Nanoscale biocatalyst systems. Curr. Opin. Biotechnol. 17,
to biosensors. Sensors 9, 10331053 (2009). doi:10.3390/s90201033 574579 (2006). doi:10.1016/j.copbio.2006.10.009
65. M. Holzinger, A.L. Goff, S. Cosnier, Nanomaterials for biosens- 86. M.L. Verma, C.J. Barrow, M. Puri, Nanobiotechnology as a novel
ing applications: a review. Front. Chem. 2, 63 (2014). doi:10.3389/ paradigm for enzyme immobilisation and stabilisation with potential
fchem.2014.00063 applications in biodiesel production. Appl. Microbiol. Biotechnol. 97,
66. S. Vaddiraju, I. Tomazos, D.J. Burgess, F.C. Jaind, F. Papadimitrakopoulos, 2339 (2013). doi:10.1007/s00253-012-4535-9
Emerging synergy between nanotechnology and implantable biosen- 87. D. Cardinale, N. Carette, T. Michon, Virus scaffolds as enzyme
sors: a review. Biosens. Bioelectron. 25, 15531565 (2010). doi:10.1016/j. nano-carriers. Trends Biotechnol. 30, 369376 (2012). doi:10.1016/j.
bios.2009.12.001 tibtech.2012.04.001
67. J. Nam, N. Won, J. Bang, H. Jin, J. Park, S. Jung, S. Jung, Y. Park, S. Kim, Sur- 88. M.L. Verma, M. Puri, C.J. Barrow, Recent trends in nanomaterials immobi-
face engineering of inorganic nanoparticles for imaging and therapy. lised enzymes for biofuel production. Crit. Rev. Biotechnol. 36, 108119
Adv. Drug Deliv. Rev. 65, 622648 (2013). doi:10.1016/j.addr.2012.08.015 (2016). doi:10.3109/07388551.2014.928811
68. S.S. Agasti, S. Rana, M.H. Park, C.K. Kim, C.C. You, V.M. Rotello, Nanopar- 89. A.A. Homaei, R. Sariri, F. Vianello, R. Stevanato, Enzyme immobili-
ticles for detection and diagnosis. Adv. Drug Deliv. Rev. 62, 316328 zation: an update. J. Chem. Biol. 6, 185205 (2013). doi:10.1007/
(2010). doi:10.1016/j.addr.2009.11.004 s12154-013-0102-9
69. H.S. Song, O.S. Kwon, S.H. Lee, S.J. Park, U.K. Kim, J. Jang, T.H. Park, 90. J. Kim, J.W. Grate, Single-enzyme nanoparticles armored by a nanom-
Human taste receptor-functionalized field effect transistor as a eter-scale organic/inorganic network. Nano Lett. 3, 12191222 (2003).
human-like nanobioelectronic tongue. Nano Lett. 13, 172178 (2013). doi:10.1021/nl034404b
doi:10.1021/nl3038147 91. A. Butt, A. Farrukh, A. Ghaffar, H. Duran, Z. Oluz, H. ur Rehman, T. Hussain,
70. O.S. Kwon, H.S. Song, S.J. Park, S.H. Lee, J.H. An, J.W. Park, H. Yang, H. R. Ahmad, A. Tahir, B. Yameen, Design of enzyme-immobilized polymer
Yoon, J. Bae, T.H. Park, J. Jang, An ultrasensitive, selective, multiplexed brush-grafted magnetic nanoparticles for efficient nematicidal activity.
superbioelectronic nose that mimics the human sense of smell. Nano RSC Adv. 5, 7768277688 (2015). doi:10.1039/c5ra10063a
Lett. 15, 65596567 (2015). doi:10.1021/acs.nanolett.5b02286 92. G. Bayramoglu, M.Y. Arica, Reversible immobilization of catalase on
71. K. Balasubramanian, M. Burghard, Biosensors based on carbon fibrous polymer grafted and metal chelated chitosan membrane. J. Mol.
nanotubes. Anal. Bioanal. Chem. 385, 452468 (2006). doi:10.1007/ Catal. B Enzym. 62, 297304 (2010). doi:10.1016/j.molcatb.2009.11.013
s00216-006-0314-8 93. J. Fu, J. Reinhold, N.W.Woodbury, Peptide-modified surfaces for enzyme immo-
72. B.A. Kairdolf, A.M. Smith, T.H. Stokes, M.D. Wang, A.N. Young, S. Nie, bilization. PLoS ONE 6, e18692 (2011). doi:10.1371/journal.pone.0018692
Semiconductor quantum dots for bioimaging and biodiagnostic 94. A. Care, P.L. Bergquist, A. Sunna, Solid-binding peptides: smart tools
applications. Annu. Rev. Anal. Chem. 6, 143162 (2013). doi:10.1146/ for nanobiotechnology. Trends Biotechnol. 33, 259268 (2015).
annurev-anchem-060908-155136 doi:10.1016/j.tibtech.2015.02.005
73. T.H. Shin, Y. Choi, S. Kim, J. Cheon, Recent advances in magnetic nano- 95. Y. Takatsujia, R. Yamasakia, A. Iwanaga, M. Lienemann, M.B. Linder, T.
particle-based multi-modal imaging. Chem. Soc. Rev. 44, 45014516 Haruyama, Solid-support immobilization of a swing fusion protein for
(2015). doi:10.1039/c4cs00345d enhanced glucose oxidase catalytic activity. Colloids Surf. B Biointer-
74. E. Boisselier, D. Astruc, Gold nanoparticles in nanomedicine: prepara- faces 112, 186191 (2013). doi:10.1016/j.colsurfb.2013.07.051
tions, imaging, diagnostics, therapies and toxicity. Chem. Soc. Rev. 38, 96. U.B. Sleytr, B. Schuster, E.M. Egelseer, D. Pum, S-layers: princi-
17591782 (2009). doi:10.1039/b806051g ples and applications. FEMS Microbiol. Rev. 38, 823864 (2014).
75. S.K. Vashist, A.G. Venkatesh, K. Mitsakakis, G. Czilwik, G. Roth, F. von doi:10.1111/1574-6976.12063
Stetten, R. Zengerle, Nanotechnology-based biosensors and diag- 97. M.H. Yang, B.G. Choi, T.J. Park, N.S. Heo, W.H. Hong, S.Y. Lee, Site-specific
nostics: technology push versus industrial/healthcare requirements. immobilization of gold binding polypeptide on gold nanoparticle-
Bionanoscience 2, 115126 (2012). doi:10.1007/s12668-012-0047-4 coated graphene sheet for biosensor application. Nanoscale 3,
76. J.F. Rusling, C.V. Kumar, J.S. Gutkind, V. Patel, Measurement of biomarker 29502956 (2011). doi:10.1039/c1nr10197h
proteins for point-of-care early detection and monitoring of cancer. 98. H. Tschiggerl, A. Breitwieser, G. de Roo, T. Verwoerd, C. Schaffer, U.B.
Analyst 135, 24962511 (2010). doi:10.1039/c0an00204f Sleyt, Exploitation of the S-layer self-assembly system for site directed
77. S.T. Sanjay, G. Fu, M. Dou, F. Xu, R. Liu, H. Qie, X.J. Li, Biomarker detection immobilization of enzymes demonstrated for an extremophilic lami-
for disease diagnosis using cost-effective microfluidic platforms. Analyst narinase from Pyrococcus furiosus. J. Biotechnol. 133, 403411 (2008).
140, 70627081 (2015). doi:10.1039/c5an00780a doi:10.1016/j.jbiotec.2007.09.018
78. D.M. Bruls, T.H. Evers, J.A.H. Kahlman, P.J.W. van Lankvelt, M. Ovsyanko, 99. S. Schoffelen, J.C.M. van Hest, Chemical approaches for the construc-
E.G.M. Pelssers, J.J.H.B. Schleipen, F.K. de Theije, C.A. Verschuren, T. van tion of multi-enzyme reaction systems. Curr. Opin. Struct. Biol. 23,
der Wijk, J.B.A. van Zon, W.U. Dittmer, A.H.J. Immink, J.H. Nieuwenhuis, 613621 (2013). doi:10.1016/j.sbi.2013.06.010
M.W.J. Prins, Rapid integrated biosensor for multiplexed immunoassays 100. A. Kchler, M. Yoshimoto, S. Luginbhl, F. Mavelli, P. Walde, Enzymatic
based on actuated magnetic nanoparticles. Lab Chip 9, 35043510 reactions in confined environments. Nat. Nanotechnol. 11, 409420
(2009). doi:10.1039/b913960e (2016). doi:10.1038/NNANO.2016.54
Nagamune Nano Convergence (2017) 4:9 Page 50 of 56

101. J.A. Mullaney, B.H.A. Rehm, Design of a single-chain multi-enzyme nanostructures. J. Am. Chem. Soc. 134, 55165519 (2012). doi:10.1021/
fusion protein establishing the polyhydroxybutyrate biosynthesis path- ja300897h
way. J. Biotechnol. 147, 3136 (2010). doi:10.1016/j.jbiotec.2010.02.021 124. Y. Fu, D. Zeng, J. Chao, Y. Jin, Z. Zhang, H. Liu, D. Li, H. Ma, Q. Huang, K.V.
102. M.C.M. van Oers, F.P.J.T. Rutjes, J.C.M. van Hest, Cascade reactions in Gothelf, C. Fan, Single-step rapid assembly of DNA origami nanostruc-
nanoreactors. Curr. Opin. Biotechnol. 28, 1016 (2014). doi:10.1016/j. tures for addressable nanoscale bioreactors. J. Am. Chem. Soc. 135,
copbio.2013.10.011 696702 (2013). doi:10.1021/ja3076692
103. R. Chen, Q. Chen, H. Kim, K.H. Siu, Q. Sun, S.L. Tsai, W. Chen, Biomolecular 125. V. Linko, M. Eerikainen, M.A. Kostiainen, A modular DNA origami-based
scaffolds for enhanced signaling and catalytic efficiency. Curr. Opin. enzyme cascade nanoreactor. Chem. Commun. 51, 53515354 (2015).
Biotechnol. 28, 5968 (2014). doi:10.1016/j.copbio.2013.11.007 doi:10.1039/c4cc08472a
104. T.W. Muir, Semisynthesis of proteins by expressed protein ligation. 126. C.J. Delebecque, A.B. Lindner, P.A. Silver, F.A. Aldaye, Organization of
Annu. Rev. Biochem. 72, 249289 (2003). doi:10.1146/annurev. intracellular reactions with rationally designed RNA assemblies. Science
biochem.72.121801.161900 333, 470474 (2011). doi:10.1126/science.1206938
105. T. Heck, G. Faccio, M. Richter, L. Thny-Meyer, Enzyme-catalyzed 127. N.C. Seeman, Nanomaterials based on DNA. Annu. Rev. Biochem. 79,
protein crosslinking. Appl. Microbiol. Biotechnol. 97, 461475 (2013). 6587 (2010). doi:10.1146/annurev-biochem-060308-102244
doi:10.1007/s00253-012-4569-z 128. N.C. Seeman, P.S. Lukema, Nucleic acid nanostructures: bottom-up con-
106. H. Hirakawa, T. Haga, T. Nagamune, Artificial protein complexes for bioca- trol of geometry on the nanoscale. Rep. Prog. Phys. 68, 237270 (2005).
talysis. Top. Catal. 55, 11241137 (2012). doi:10.1007/s11244-012-9900-5 doi:10.1088/0034-4885/68/1/R05
107. O. Idan, H. Hess, Engineering enzymatic cascades on nanoscale 129. H. Kim, Y. Park, J. Kim, J. Jeong, S. Han, J.S. Lee, J.B. Lee, Nucleic acid
scaffolds. Curr. Opin. Biotechnol. 24, 606611 (2013). doi:10.1016/j. engineering: RNA following the trail of DNA. ACS Comb. Sci. 18, 8799
copbio.2013.01.003 (2016). doi:10.1021/acscombsci.5b00108
108. H. Hirakawa, N. Kamiya, T. Tanaka, T. Nagamune, Intramolecular electron 130. P.W.K. Rothemund, Folding DNA to create nanoscale shapes and pat-
transfer in a cytochrome P450cam system with a site-specific branched terns. Nature 440, 297302 (2006). doi:10.1038/nature04586
structure. Protein Eng. Des. Sel. 20, 453459 (2007). doi:10.1093/ 131. M. Endo, Y. Yang, H. Sugiyama, DNA origami technology for biomaterials
protein/gzm045 applications. Biomater. Sci. 1, 347360 (2013). doi:10.1039/c2bm00154c
109. H. Hirakawa, N. Kamiya, Y. Kawarabayasi, T. Nagamune, Artificial self- 132. E.J. Cho, J.W. Lee, A.D. Ellington, Applications of aptamers as sen-
sufficient P450 in reversed micelles. Molecules 15, 29352948 (2010). sors. Annu. Rev. Anal. Chem. 2, 241264 (2009). doi:10.1146/annurev.
doi:10.3390/molecules15052935 anchem.1.031207.112851
110. M.C. Good, J.G. Zalatan, W.A. Lim, Scaffold proteins: hubs for control- 133. M. Darmostuk, S. Rimpelova, H. Gbelcova, T. Ruml, Current approaches
ling the flow of cellular information. Science 332, 680686 (2011). in SELEX: an update to aptamer selection technology. Biotechnol. Adv.
doi:10.1126/science.1198701 33, 11411161 (2015). doi:10.1016/j.biotechadv.2015.02.008
111. H. Hirakawa, T. Nagamune, Molecular assembly of P450 with ferredoxin 134. H. Sun, Y. Zu, A highlight of recent advances in aptamer technology
and ferredoxin reductase by fusion to PCNA. ChemBioChem 11, and its application. Molecules 20, 1195911980 (2015). doi:10.3390/
15171520 (2010). doi:10.1002/cbic.201000226 molecules200711959
112. W.R. Whitaker, J.E. Dueber, Metabolic pathway flux enhancement by 135. M. Blind, M. Blank, Aptamer selection technology and recent advances.
synthetic protein scaffolding. Methods Enzymol. 497, 447468 (2011). Mol. Ther. Nucleic Acids 4, e223 (2015). doi:10.1038/mtna.2014.74
doi:10.1016/B978-0-12-385075-1.00019-6 136. A.D. Keefe, S. Pai, A. Ellington, Aptamers as therapeutics. Nat. Rev. Drug
113. C. You, S. Myung, Y.H.P. Zhang, Facilitated substrate channeling in a self- Discov. 9, 537550 (2010). doi:10.1038/nrd3141
assembled trifunctional enzyme complex. Angew. Chem. Int. Ed. 51, 137. G.M. Stovall, R.S. Bedenbaugh, S. Singh, A.J. Meyer, P.J. Hatala, A.D.
87878790 (2012). doi:10.1002/anie.201202441 Ellington, B. Hall, Invitro selection using modified or unnatural
114. J.E. Dueber, G.C. Wu, G.R. Malmirchegini, T.S. Moon, C.J. Petzold, A.V. Ullal, nucleotides. Curr. Protoc. Nucleic Acid Chem. 56, 9.6.19.6.33 (2014).
K.L.J. Prather, J.D. Keasling, Synthetic protein scaffolds provide modular doi:10.1002/0471142700.nc0906s56
control over metabolic flux. Nat. Biotechnol. 27, 753761 (2009). 138. S. Jger, G. Rasched, H. Kornreich-Leshem, M. Engeser, O. Thum, M.
doi:10.1038/nbt.1557 Famulok, A versatile toolbox for variable DNA functionalization at high
115. H. Watanabe, H. Hirakawa, T. Nagamune, Phosphite-driven self- density. J. Am. Chem. Soc. 127, 1507115082 (2005). doi:10.1021/
sufficient cytochrome P450. ChemCatChem 5, 38353840 (2013). ja051725b
doi:10.1002/cctc.201300445 139. S. Song, L. Wang, J. Li, J. Zhao, C. Fan, Aptamer-based biosensors. Trends
116. S. Moras, Y. Barak, J. Caspi, Y. Hadar, R. Lamed, Y. Shoham, D.B. Wilson, Anal. Chem. 27, 108117 (2008). doi:10.1016/j.trac.2007.12.004
E.A. Bayer, Cellulase-xylanase synergy in designer cellulosomes for 140. Y. Liu, C. Lin, H. Li, H. Yan, Aptamer-directed self-asssembly of protein
enhanced degradation of a complex cellulosic substrate. mBio 1, arrays on a DNA nanostructure. Angew. Che. Int. Ed. 44, 43334338
e0028510 (2010). doi:10.1128/mBio.00285-10 (2005). doi:10.1002/anie.200501089
117. C.M. Niemeyer, Semisynthetic DNAprotein conjugates for biosens- 141. J. Mulhbacher, P. St-Pierre, D.A. Lafontaine, Therapeutic applications of
ing and nanofabrication. Angew. Chem. Int. Ed. 49, 12001216 (2010). ribozymes and riboswitches. Curr. Opin. Pharmacol. 10, 551556 (2010).
doi:10.1002/anie.200904930 doi:10.1016/j.coph.2010.07.002
118. C. Teller, I. Willner, Organizing proteinDNA hybrids as nanostructures 142. K.Y. Jen, A.M. Gewirtz, Suppression of gene expression by targeted
with programmed functionalities. Trends Biotechnol. 28, 619628 disruption of messenger RNA: available options and current strategies.
(2010). doi:10.1016/j.tibtech.2010.09.005 Stem Cells 18, 307319 (2000). doi:10.1634/stemcells.18-5-307
119. J. Fu, M. Liu, Y. Liu, H. Yan, Spatially-interactive biomolecular net- 143. J. Rossi, Ribozymes, genomeics and therapeutics. Chem. Biol. 6, R33
works organized by nucleic acid nanostructures. Acc. Chem. Res. 45, R37 (1999). doi:10.1016/S1074-5521(99)80001-5
12151226 (2012). doi:10.1021/ar200295q 144. D.S. Wilson, J.W. Szostak, Invitro selection of functional nucleic acids.
120. Y.R. Yang, Y. Liu, H. Yan, DNA nanostructures as programmable Annu. Rev. Biochem. 68, 611647 (1999). doi:10.1146/annurev.
biomolecular scaffolds. Bioconjugate Chem. 26, 13811395 (2015). biochem.68.1.611
doi:10.1021/acs.bioconjchem.5b00194 145. G.F. Joyce, Directed evolution of nucleic acid enzymes. Annu.
121. M. Erkelenz, C.H. Kuo, C.M. Niemeyer, DNA-mediated assembly of Rev. Biochem. 73, 791836 (2004). doi:10.1146/annurev.
cytochrome P450 BM3 subdomains. J. Am. Chem. Soc. 133, 16111 biochem.73.011303.073717
16118 (2011). doi:10.1021/ja204993s 146. A. Jschke, Artificial ribozymes and deoxyribozymes. Curr. Opin. Struct.
122. O.I. Wilner, Y. Weizmann, R. Gill, O. Lioubashevski, R. Freeman, I. Willner, Biol. 11, 321326 (2001). doi:10.1016/S0959-440X(00)00208-6
Enzyme cascades activated on topologically programmed DNA scaf- 147. J.H. Jeong, Y.W. Cho, B. Jung, K. Park, J.-D. Kim, Self-assembled nano-
folds. Nat. Nanotechnol. 4, 249254 (2009). doi:10.1038/NNANO.2009.50 particles of ribozymes with poly(ethylene glycol)-b-poly(l-lysine)
123. J. Fu, M. Liu, Y. Liu, N.W. Woodbury, H. Yan, Interenzyme substrate diffu- block copolymers. Jpn. J. Appl. Phys. 45, 591595 (2006). doi:10.1143/
sion for an enzyme cascade organized on spatially addressable DNA JJAP.45.591
Nagamune Nano Convergence (2017) 4:9 Page 51 of 56

148. J.L. Rouge, T.L. Sita, L. Hao, F.M. Kouri, W.E. Briley, A.H. Stegh, C.A. Mirkin, 171. R.J. Pantazes, M.J. Grisewood, C.D. Maranas, Recent advances in com-
Ribozyme-spherical nucleic acids. J. Am. Chem. Soc. 137, 1052810531 putational protein design. Curr. Opin. Struct. Biol. 21, 467472 (2011).
(2015). doi:10.1021/jacs.5b07104 doi:10.1016/j.sbi.2011.04.005
149. R.C. Cadwell, G.F. Joyce, Randomization of genes by PCR mutagenesis. 172. M.S. Packer, D.R. Liu, Methods for the directed evolution of proteins. Nat.
Genome Res. 2, 2833 (1992). doi:10.1101/gr.2.1.28 Rev. Genet. 16, 379394 (2015). doi:10.1038/nrg3927
150. H. Ueda, M. Nakanishi, E. Suzuki, T. Nagamune, Enhancement of 173. H. Leemhuis, V. Stein, A.D. Griffiths, F. Hollfelder, New genotypepheno-
mutation frequency with nucleotide triphosphate analogs in PCR type linkages for directed evolution of functional proteins. Curr. Opin.
random mutagenesis. J. Ferment. Technol. 79, 303305 (1995). Struct. Biol. 15, 472478 (2005). doi:10.1016/j.sbi.2005.07.006
doi:10.1016/0922-338X(95)90624-9 174. P.A.G. Tizei, E. Csibra, L. Torres, V.B. Pinheiro, Selection platforms for
151. M. Zaccolo, D.M. Williams, D.M. Brown, E. Gherardi, An approach to ran- directed evolution in synthetic biology. Biochem. Soc. Trans. 44,
dom mutagenesis of DNA using mixtures of triphosphate derivatives 11651175 (2016). doi:10.1042/BST20160076
of nucleoside analogues. J. Mol. Biol. 255, 589603 (1996). doi:10.1006/ 175. H.M.E. Azzazya, W.E. Highsmith Jr., Phage display technology: clinical
jmbi.1996.0049 applications and recent innovations. Clin. Biochem. 35, 425445 (2002).
152. R. Fujii, M. Kitaoka, K. Hayashi, Error-prone rolling circle amplification: doi:10.1016/S0009-9120(02)00343-0
the simplest random mutagenesis protocol. Nat. Protoc. 1, 24932497 176. A.R. Mkel, C. Oker-Blom, Baculovirus display: a multifunctional tech-
(2006). doi:10.1038/nprot.2006.403 nology for gene delivery and eukaryotic library development. Adv. Virus
153. Y.P. Lai, J. Huang, L.F. Wang, J. Li, Z.R. Wu, A new approach to ran- Res. 68, 91112 (2006). doi:10.1016/S0065-3527(06)68003-2
dom mutagenesis invitro. Biotechnol. Bioeng. 86, 622627 (2004). 177. S.Y. Lee, J.H. Choi, Z. Xu, Microbial cell-surface display. Trends Biotechnol.
doi:10.1002/bit.20066 21, 4552 (2003). doi:10.1016/S0167-7799(02)00006-9
154. B. Sullivan, A.Z. Walton, J.D. Stewart, Library construction and evaluation 178. M. Desvaux, E. Dumas, I. Chafsey, M. Hbraud, Protein cell surface dis-
for site saturation mutagenesis. Enzyme Microb. Technol. 53, 7077 play in Gram-positive bacteria. FEMS Microbiol. Lett. 256, 115 (2006).
(2013). doi:10.1016/j.enzmictec.2013.02.012 doi:10.1111/j.1574-6968.2006.00122.x
155. W.P.C. Stemmer, DNA shuffling by random fragmentation and reassem- 179. T. Tanaka, R. Yamada, C. Ogino, A. Kondo, Recent developments in
bly: invitro recombination for molecular evolution. Proc. Natl. Acad. Sci. yeast cell surface display toward extended applications in biotechnol-
USA 91, 1074710751 (1994). doi:10.1073/pnas.91.22.10747 ogy. Appl. Microbiol. Biotechnol. 95, 577591 (2012). doi:10.1007/
156. M. Kikuchi, K. Ohnishi, S. Harayama, Novel family shuffling methods s00253-012-4175-0
for the invitro evolution of enzymes. Gene 236, 159167 (1999). 180. S.A. Gai, K.D. Wittrup, Yeast surface display for protein engineering
doi:10.1016/S0378-1119(99)00240-1 and characterization. Curr. Opin. Struct. Biol. 17, 467473 (2007).
157. M. Kikuchi, K. Ohnishi, S. Harayama, An effective family shuffling doi:10.1016/j.sbi.2007.08.012
method using single-stranded DNA. Gene 243, 133137 (2000). 181. K.D. Wittrup, Protein engineering by cell-surface display. Curr. Opin.
doi:10.1016/S0378-1119(99)00547-8 Biotechnol. 12, 395399 (2001). doi:10.1016/S0958-1669(00)00233-0
158. V. Sieber, C.A. Martinez, F.H. Arnold, Libraries of hybrid proteins from 182. M. He, Cell-free protein synthesis: applications in proteomics and
distantly related sequences. Nat. Biotechnol. 19, 456460 (2001). biotechnology. New Biotechnol. 25, 126131 (2008). doi:10.1016/j.
doi:10.1038/88129 nbt.2008.08.004
159. S. Lutz, M. Ostermeier, S.J. Benkovic, Rapid generation of incremental 183. E.D. Carlson, R. Gan, C.E. Hodgman, M.C. Jewett, Cell-free protein
truncation libraries for protein engineering using -phosphothioate synthesis: applications come of age. Biotechnol. Adv. 30, 11851194
nucleotides. Nucleic Acids Res. 29, 16e (2001). doi:10.1093/nar/29.4.e16 (2012). doi:10.1016/j.biotechadv.2011.09.016
160. S. Lutz, M. Ostermeier, G.L. Moore, C.D. Maranas, S.J. Benkovic, Creating 184. W.D. Fessner, Systems biocatalysis: development and engineering of
multiple-crossover DNA libraries independent of sequence iden- cell-free artificial metabolisms for preparative multi-enzymatic synthe-
tity. Proc. Natl. Acad. Sci. USA 98, 1124811253 (2001). doi:10.1073/ sis. New Biotechnol. 32, 658664 (2015). doi:10.1016/j.nbt.2014.11.007
pnas.201413698 185. S.M. Schinn, A. Broadbent, W.T. Bradley, B.C. Bundy, Protein synthesis
161. J.A. Brannigan, A.J. Wilkinson, Protein engineering 20years on. Nat. Rev. directly from PCR: progress and applications of cell-free protein synthe-
Mol. Cell Biol. 3, 964970 (2002). doi:10.1038/nrm975 sis with linear DNA. New Biotechnol. 33, 480487 (2016). doi:10.1016/j.
162. J.A. Kolkman, W.P.C. Stemmer, Directed evolution of proteins by exon nbt.2016.04.002
shuffling. Nat. Biotechnol. 19, 423428 (2001). doi:10.1038/88084 186. D. Lipovsek, A. Plckthun, In-vitro protein evolution by ribosome
163. N. Doi, H. Yanagawa, Insertional gene fusion technology. FEBS Lett. 457, display and mRNA display. J. Immunol. Methods 290, 5167 (2004).
14 (1999). doi:10.1016/S0014-5793(99)00991-6 doi:10.1016/j.jim.2004.04.008
164. S. Elleuche, Bringing functions together with fusion enzymesfrom 187. R. Odegrip, D. Coomber, B. Eldridge, R. Hederer, P.A. Kuhlman, C. Ullman,
natures inventions to biotechnological applications. Appl. Microbiol. K. FitzGerald, D. McGregor, CIS display: Invitro selection of peptides
Biotechnol. 99, 15451556 (2015). doi:10.1007/s00253-014-6315-1 from libraries of proteinDNA complexes. Proc. Natl. Acad. Sci. USA 101,
165. S.A. Navai, N. Ahmed, Targeting the tumour profile using broad 28062810 (2004). doi:10.1073/pnas.0400219101
spectrum chimaeric antigen receptor T-cells. Biochem. Soc. Trans. 44, 188. J. Bertschinger, D. Neri, Covalent DNA display as a novel tool for
391396 (2016). doi:10.1042/BST20150266 directed evolution of proteins invitro. Protein Eng. Des. Sel. 17,
166. A. Miyawaki, Y. Niino, Molecular spies for bioimagingfluorescent 699707 (2004). doi:10.1093/protein/gzh082
protein-based probes. Mol. Cell 58, 632643 (2015). doi:10.1016/j. 189. N. Doi, H. Yanagawa, STABLE: proteinDNA fusion system for screening
molcel.2015.03.002 of combinatorial protein libraries invitro. FEBS Lett. 457, 227230
167. H. Mattoussi, J.M. Mauro, E.R. Goldman, G.P. Anderson, V.C. Sundar, (1999). doi:10.1016/S0014-5793(99)01041-8
F.V. Mikulec, M.G. Bawendi, Self-assembly of CdSeZnS quantum dot 190. O. Nord, M. Uhln, P.A. Nygren, Microbead display of proteins by
bioconjugates using an engineered recombinant protein. J. Am. Chem. cell-free expression of anchored DNA. J. Biotechnol. 106, 113 (2003).
Soc. 122, 1214212150 (2000). doi:10.1021/ja002535y doi:10.1016/j.jbiotec.2003.09.002
168. R.A. Studer, B.H. Dessailly, C.A. Orengo, Residue mutations and their 191. D.S. Tawfik, A.D. Griffiths, Man-made cell-like compartments for
impact on protein structure and function: detecting beneficial and molecular evolution. Nat. Biotechnol. 16, 652656 (1998). doi:10.1038/
pathogenic changes. Biochem. J. 449, 581594 (2013). doi:10.1042/ nbt0798-652
BJ20121221 192. H. Lin, V.W. Cornish, Screening and selection methods for large-
169. M. Surez, A. Jaramillo, Challenges in the computational design of pro- scale analysis of protein function. Angew. Chem. Int. Ed. 41,
teins. J. R. Soc. Interface 6, S477S491 (2009). doi:10.1098/rsif.2008.0508. 44024425 (2002). doi:10.1002/1521-3773(20021202)41:23<4402:AID-
focus ANIE4402>3.0.CO;2-H
170. J.G. Saven, Computational protein design: engineering molecular 193. S. Becker, H.U. Schmoldt, T.M. Adams, S. Wilhelm, H. Kolmar, Ultra-high-
diversity, nonnatural enzymes, nonbiological cofactor complexes, throughput screening based on cell-surface display and fluorescence-
and membrane proteins. Curr. Opin. Chem. Biol. 15, 452457 (2011). activated cell sorting for the identification of novel biocatalysts. Curr.
doi:10.1016/j.cbpa.2011.03.014 Opin. Biotechnol. 15, 323329 (2004). doi:10.1016/j.copbio.2004.06.001
Nagamune Nano Convergence (2017) 4:9 Page 52 of 56

194. G. Yang, S.G. Withers, Ultrahigh-throughput FACS-based screening for reaction. Angew. Chem. Int. Ed. 45, 53075311 (2006). doi:10.1002/
directed enzyme evolution. ChemBioChem 10, 27042715 (2009). anie.200600368
doi:10.1002/cbic.200900384 216. R.P. Lyon, J.R. Setter, T.D. Bovee, S.O. Doronina, J.H. Hunter, M.E. Ander-
195. A. Currin, N. Swainston, P.J. Dayace, D.B. Kell, Synthetic biology for son, C.L. Balasubramanian, S.M. Duniho, C.I. Leiske, F. Li, P.D. Senter,
the directed evolution of protein biocatalysts: navigating sequence Self-hydrolyzing maleimides improve the stability and pharmaco-
space intelligently. Chem. Soc. Rev. 44, 11721239 (2015). doi:10.1039/ logical properties of antibody-drug conjugates. Nat. Biotechnol. 32,
c4cs00351a 10591062 (2014). doi:10.1038/nbt.2968
196. S. Kumar, Engineering cytochrome P450 biocatalysts for biotechnology, 217. N.S. Joshi, J.R. Whitaker, M.B. Francis, A three-component Mannich-type
medicine, and bioremediation. Expert. Opin. Drug. Metab. Toxicol. 6, reaction for selective tyrosine bioconjugation. J. Am. Chem. Soc. 126,
115131 (2010). doi:10.1517/17425250903431040 1594215943 (2004). doi:10.1021/ja0439017
197. H. Zhao, K. Chockalingam, Z. Chen, Directed evolution of enzymes and 218. Y. Seki, T. Ishiyama, D. Sasaki, J. Abe, Y. Sohma, K. Oisaki, M. Kanai, Transi-
pathways for industrial biocatalysis. Curr. Opin. Biotechnol. 13, 104110 tion metal-free tryptophan-selective bioconjugation of proteins. J. Am.
(2002). doi:10.1016/S0958-1669(02)00291-4 Chem. Soc. 138, 1079810801 (2016). doi:10.1021/jacs.6b06692
198. A. Fernandez-Gacio, M. Uguen, J. Fastrez, Phage display as a tool for the 219. C.C. Liu, P.G. Schultz, Adding new chemistries to the genetic code.
directed evolution of enzymes. Trends Biotechnol. 21, 408414 (2003). Annu. Rev. Biochem. 79, 413444 (2010). doi:10.1146/annurev.
doi:10.1016/S0167-7799(03)00194-X biochem.052308.105824
199. J. McCafferty, D. Schofield, Identification of optimal protein binders 220. J.A. Johnson, Y.Y. Lu, J.A.V. Deventer, D.A. Tirrell, Residue-specific incor-
through the use of large genetically encoded display libraries. Curr. poration of non-canonical amino acids into proteins. Curr. Opin. Chem.
Opin. Chem. Biol. 26, 1624 (2015). doi:10.1016/j.cbpa.2015.01.003 Biol. 14, 774780 (2010). doi:10.1016/j.cbpa.2010.09.013
200. I. Chen, B.M. Dorr, D.R. Liu, A general strategy for the evolution of bond- 221. A.R. Goerke, J.R. Swartz, High-level cell-free synthesis yields of proteins
forming enzymes using yeast display. Proc. Natl. Acad. Sci. USA 108, containing site-specific non-natural amino acids. Biotechnol. Bioeng.
1139911404 (2011). doi:10.1073/pnas.1101046108 102, 400416 (2009). doi:10.1002/bit.22070
201. M. Ho, I. Pastan, Mammalian cell display for antibody 222. T. Hohsaka, M. Sisido, Incorporation of non-natural amino acids into
engineering. Methods Mol. Biol. 525, 337352 (2009). proteins. Curr. Opin. Chem. Biol. 6, 809815 (2002). doi:10.1016/
doi:10.1007/978-1-59745-554-1_18 S1367-5931(02)00376-9
202. P.M. Bowers, R.A. Horlick, T.Y. Neben, R.M. Toobian, G.L. Tomlinson, J.L. 223. T. Passioura, H. Suga, Reprogramming the genetic code invitro. Trends
Dalton, H.A. Jones, A. Chen, L. Altobell III, X. Zhang, J.L. Macomber, Biochem. Sci. 39, 4004008 (2014). doi:10.1016/j.tibs.2014.07.005
I.P. Krapf, B.F. Wu, A. McConnell, B. Chau, T. Holland, A.D. Berkebile, S.S. 224. K. Lang, J.W. Chin, Bioorthogonal reactions for labeling proteins. ACS
Neben, W.J. Boyle, D.J. King, Coupling mammalian cell surface display Chem. Biol. 9, 1620 (2014). doi:10.1021/cb4009292
with somatic hypermutation for the discovery and maturation of 225. Y. Gong, L. Pan, Recent advances in bioorthogonal reactions for
human antibodies. Proc. Natl. Acad. Sci. USA 108, 2045520460 (2011). site-specific protein labeling and engineering. Tetrahedron Lett. 56,
doi:10.1073/pnas.1114010108 21232132 (2015). doi:10.1016/j.tetlet.2015.03.065
203. R. Yoshida, M. Kawahara, T. Nagamune, A novel platform for antibody 226. P.E. Dawson, S.B.H. Kent, Synthesis of native proteins by chemical liga-
library selection in mammalian cells based on a growth signalobody. tion. Annu. Rev. Biochem. 69, 923960 (2000). doi:10.1146/annurev.
Biotechnol. Bioeng. 111, 11701179 (2014). doi:10.1002/bit.25173 biochem.69.1.923
204. R. Yoshida, M. Kawahara, T. Nagamune, Domain structure of growth 227. L.R. Malins, R.J. Payne, Recent extensions to native chemical ligation for
signalobodies critically affects the outcome of antibody library selec- the chemical synthesis of peptides and proteins. Curr. Opin. Chem. Biol.
tion. J. Biochem. 157, 497506 (2015). doi:10.1093/jb/mvv008 22, 7078 (2014). doi:10.1016/j.cbpa.2014.09.021
205. K. Soga, H. Abo, S.Y. Qin, T. Kyoutou, K. Hiemori, H. Tateno, N. Matsu- 228. V. Muralidharan, T.W. Muir, Protein ligation: an enabling technology for
moto, J. Hirabayashi, K. Yamamoto, Mammalian cell surface display as a the biophysical analysis of proteins. Nat. Methods 3, 429438 (2006).
novel method for developing engineered lectins with novel character- doi:10.1038/nmeth886
istics. Biomolecules 5, 15401562 (2015). doi:10.3390/biom5031540 229. H. Tian, A. Frstenberg, T. Huber, Labeling and single-molecule methods
206. A.D. Griffiths, D.S. Tawfik, Directed evolution of an extremely fast phos- to monitor G protein-coupled receptor dynamics. Chem. Rev. 117,
photriesterase by invitro compartmentalization. EMBO J. 22, 2435 186245 (2017). doi:10.1021/acs.chemrev.6b00084
(2003). doi:10.1093/emboj/cdg014 230. M. Rashidian, J.K. Dozier, M.D. Distefano, Enzymatic labeling of proteins:
207. M. Naimuddin, T. Kubo, A high performance platform based on cDNA techniques and approaches. Bioconjugate Chem. 24, 12771294
display for efficient synthesis of protein fusions and accelerated (2013). doi:10.1021/bc400102w
directed evolution. ACS Comb. Sci. 18, 117129 (2016). doi:10.1021/ 231. J.M. McFarland, D. Rabuka, Recent advances in chemoenzymatic bioconjuga-
acscombsci.5b00139 tion methods. Org. Chem. Insights 5, 714 (2015). doi:10.4137/OCI.S17957
208. S.A. Richman, D.M. Kranz, Display, engineering, and applications of 232. S.A. Walper, K.B. Turner, I.L. Medintz, Enzymatic bioconjugation of nano-
antigen-specific T cell receptors. Biomol. Eng. 24, 361373 (2007). particles: developing specificity and control. Curr. Opin. Biotechnol. 34,
doi:10.1016/j.bioeng.2007.02.009 232241 (2015). doi:10.1016/j.copbio.2015.04.003
209. C. Jckel, D. Hilvert, Biocatalysts by evolution. Curr. Opin. Biotechnol. 21, 233. P. Wu, W. Shui, B.L. Carlson, N. Hu, D. Rabuka, J. Lee, C.R. Bertozzi, Site-
753759 (2010). doi:10.1016/j.copbio.2010.08.008 specific chemical modification of recombinant proteins produced
210. A.S. Bommarius, J.K. Blum, M.J. Abrahamson, Status of protein in mammalian cells by using the genetically encoded aldehyde
engineering for biocatalysts: how to design an industrially useful tag. Proc. Natl. Acad. Sci. USA 106, 30003005 (2009). doi:10.1073/
biocatalyst. Curr. Opin. Chem. Biol. 15, 194200 (2011). doi:10.1016/j. pnas.0807820106
cbpa.2010.11.011 234. J.K. Dozier, S.L. Khatwani, J.W. Wollack, Y.C. Wang, C. Schmidt-Dannert,
211. E.M. Brustad, F.H. Arnold, Optimizing non-natural protein function M.D. Distefano, Engineering protein farnesyltransferase for enzymatic
with directed evolution. Curr. Opin. Chem. Biol. 15, 201210 (2011). protein labeling applications. Bioconjugate Chem. 25, 12031212
doi:10.1016/j.cbpa.2010.11.020 (2014). doi:10.1021/bc500240p
212. Y. Li, P.C. Cirino, Recent advances in engineering proteins for biocataly- 235. W.P. Heal, S.R. Wickramasinghe, P.W. Bowyer, A.A. Holder, D.F. Smith, R.J.
sis. Biotechnol. Bioeng. 111, 12731287 (2014). doi:10.1002/bit.25240 Leatherbarrow, E.W. Tate, Site-specific N-terminal labelling of proteins
213. A.C. Obermeyer, B.D. Olsen, Synthesis and application of protein- invitro and invivo using N-myristoyl transferase and bioorthogonal
containing block copolymers. ACS Macro Lett. 4, 101110 (2015). ligation chemistry. Chem. Commun. 480482 (2008). doi:10.1039/
doi:10.1021/mz500732e B716115H
214. M.T. Smith, A.K. Hawes, B.C. Bundy, Reengineering viruses and virus-like 236. I. Chen, M. Howarth, W.Y. Lin, A.Y. Ting, Site-specific labeling of cell sur-
particles through chemical functionalization strategies. Curr. Opin. face proteins with biophysical probes using biotin ligase. Nat. Methods
Biotechnol. 24, 620626 (2013). doi:10.1016/j.copbio.2013.01.011 2, 99104 (2005). doi:10.1038/nmeth735
215. J.M. Gilmore, R.A. Scheck, A.P. Esser-Kahn, N.S. Joshi, M.B. Francis, 237. I. Chen, Y.A. Choi, A.Y. Ting, Phage display evolution of a peptide
N-terminal protein modification through a biomimetic transamination substrate for yeast biotin ligase and application to two-color quantum
Nagamune Nano Convergence (2017) 4:9 Page 53 of 56

dot labeling of cell surface proteins. J. Am. Chem. Soc. 129, 66196625 257. J.O. Fierer, G. Veggiani, M. Howarth, SpyLigase peptidepeptide liga-
(2007). doi:10.1021/ja071013g tion polymerizes affibodies to enhance magnetic cancer cell capture.
238. S.A. Slavoff, I. Chen, Y.A. Choi, A.Y. Ting, Expanding the substrate toler- Proc. Natl. Acad. Sci. USA 111, E1176E1181 (2014). doi:10.1073/
ance of biotin ligase through exploration of enzymes from diverse spe- pnas.1315776111
cies. J. Am. Chem. Soc. 130, 11601162 (2008). doi:10.1021/ja076655i 258. A. Juillerat, T. Gronemeyer, A. Keppler, S. Gendreizig, H. Pick, H. Vogel, K.
239. J.D. Cohen, P. Zou, A.Y. Ting, Site-specific protein modification using Johnsson, Directed evolution of O6-alkylguanine-DNA alkyltransferase
lipoic acid ligase and bis-aryl hydrazone formation. ChemBioChem 13, for efficient labeling of fusion proteins with small molecules invivo.
888894 (2012). doi:10.1002/cgic.201100764 Chem. Biol. 10, 313317 (2003). doi:10.1016/S1074-5521(03)00068-1
240. A. Fontana, B. Spolaore, A. Mero, F.M. Veronese, Site-specific modi- 259. A. Gautier, A. Juillerat, C. Heinis, I.R. Corra Jr., M. Kindermann, F.
fication and PEGylation of pharmaceutical proteins mediated by Beaufils, K. Johnsson, An engineered protein tag for multiprotein
transglutaminase. Adv. Drug Deliv. Rev. 60, 1328 (2008). doi:10.1016/j. labeling in living cells. Chem. Biol. 15, 128136 (2008). doi:10.1016/j.
addr.2007.06.015 chembiol.2008.01.007
241. N. Kamiya, T. Tanaka, T. Suzyuki, T. Takazawa, S. Takeda, K. Watanabe, T. 260. G.V. Los, L.P. Encell, M.C. McDougall, D.D. Hartzell, N. Karassina, C. Zim-
Nagamune, S-peptide as a potent peptidyl linker for protein crosslink- prich, M.G. Wood, R. Learish, R.F. Ohana, M. Urh, D. Simpson, J. Mendez,
ing by microbial transglutaminase from Streptomyces mobaraensis. K. Zimmerman, P. Otto, G. Vidugiris, J. Zhu, A. Darzins, D.H. Klaubert,
Bioconjug. Chem. 14, 351357 (2003). doi:10.1021/bc025610y R.F. Bulleit, K.V. Wood, HaloTag: a novel protein labeling technology for
242. T. Tanaka, N. Kamiya, T. Nagamune, Peptidyl linkers for protein heterodi- cell imaging and protein analysis. ACS Chem. Biol. 3, 373382 (2008).
merization catalyzed by microbial transglutaminase. Bioconjug. Chem. doi:10.1021/cb800025k
15, 491497 (2004). doi:10.1021/bc034209o 261. Z. Chen, C. Jing, S.S. Gallagher, M.P. Sheetz, V.W. Cornish, Second-gen-
243. T. Takazawa, N. Kamiya, H. Ueda, T. Nagamune, Enzymatic labeling eration covalent TMP-tag for live cell imaging. J. Am. Chem. Soc. 134,
of a single chain variable fragment of an antibody with alkaline 1369213699 (2012). doi:10.1021/ja303374p
phosphatase by microbial transglutaminase. Biotechnol. Bioeng. 86, 262. K. Katayama, S. Tsukiji, T. Furuta, T. Nagamune, A bromocoumarin-based
399404 (2004). doi:10.1002/bit.20019 linker for synthesis of photocleavable peptidoconjugates with high
244. T. Tanaka, N. Kamiya, T. Nagamune, N-terminal glycine-specific protein photosensitivity. Chem. Commun. 53995401 (2008). doi:10.1039/
conjugation catalyzed by microbial transglutaminase. FEBS Lett. 579, b812058g
20922096 (2005). doi:10.1016/j.febslet.2005.02.064 263. S. Yamaguchi, Y. Chen, S. Nakajima, T. Furuta, T. Nagamune, Light-acti-
245. S. Dasgupta, S. Samantaray, D. Sahal, R.P. Roy, Isopeptide ligation vated gene expression from site-specific caged DNA with a biotinylated
catalyzed by quintessential sortase A: mechanistic cues from cyclic and photolabile protection group. Chem. Commun. 46, 22442246 (2010).
branched oligomers of indolicidin. J. Biol. Chem. 286, 2399624006 doi:10.1039/b922502a
(2011). doi:10.1074/jbc.M111.247650 264. Y. Ohiro, H. Ueda, N. Shibata, T. Nagamune, Enhanced fuorescence reso-
246. S. Pritz, Y. Wolf, O. Kraetke, J. Klose, M. Bienert, M. Beyermann, Synthesis nance energy transfer immunoassay with improved sensitivity based
of biologically active peptide nucleic acid-peptide conjugates by on the Fab-based immunoconjugates. Anal. Biochem. 360, 266272
sortase-mediated ligation. J. Org. Chem. 72, 39093912 (2007). (2007). doi:10.1016/j.ab.2006.10.025
doi:10.1021/jo062331l 265. S. Takamori, S. Yamaguchi, N. Ohashia, T. Nagamune, Sterically bulky
247. X. Huang, A. Aulabaugh, W. Ding, B. Kapoor, L. Alksne, K. Tabei, G. caging for light-inducible protein activation. Chem. Commun. 49,
Ellestad, Kinetic mechanism of Staphylococcus aureus sortase SrtA. 30133015 (2013). doi:10.1039/c3cc38026b
Biochemistry 42, 1130711315 (2003). doi:10.1021/bi034391g 266. Y. Oshiba, T. Tamaki, H. Ohashi, H. Hirakawa, S. Yamaguchi, T. Nagamune,
248. T. Tanaka, T. Yamamoto, S. Tsukiji, T. Nagamune, Site-specific protein T. Yamaguchi, Effect of length of molecular recognition moiety on
modification on living cells catalyzed by sortase. ChemBioChem 9, enzymatic activity switching. J. Biosci. Bioeng. 116, 433437 (2013).
802807 (2008). doi:10.1002/cbic.200700614 doi:10.1016/j.jbiosc.2013.04.003
249. T. Yamamoto, T. Nagamune, Expansion of the sortase-mediated labeling 267. T. Ando, S. Tsukiji, T. Tanaka, T. Nagamune, Construction of a small-mol-
method for site-specific N-terminal labeling of cell surface proteins on ecule-integrated semisynthetic split intein for invivo protein ligation.
living cells. Chem. Commun. 10221024 (2009). doi:10.1039/b818792d Chem. Commun. 49954997 (2007). doi:10.1039/b712843f
250. S. Tsukiji, T. Nagamune, Sortase-mediated ligation: a gift from gram- 268. H.A. Chung, K. Tajima, K. Kato, N. Matsumoto, K. Yamamoto, T. Nagamune,
positive bacteria to protein engineering. ChemBioChem 10, 787798 Modulating the actions of NK cell-mediated cytotoxicity using lipid-
(2009). doi:10.1002/cbic.200800724 PEG(n) and inhibitory receptor-specific antagonistic peptide conjugates.
251. N. Hirota, D. Tasuda, T. Hashidate, T. Yamamoto, S. Yamaguchi, T. Naga- Biotechnol. Prog. 21, 12261230 (2005). doi:10.1021/bp049646b
mune, T. Nagase, T. Shimizu, M. Nakamura, Amino acid residues critical 269. U. Tomita, S. Yamaguchi, Y. Sugimoto, S. Takamori, T. Nagamune,
for endoplasmic reticulum export and trafficking of platelet-activating Poly(ethylene glycol)-lipid-conjugated antibodies enhance dendritic
factor receptor. J. Biol. Chem. 285, 59315940 (2010). doi:10.1074/jbc. cell phagocytosis of apoptotic cancer cells. Pharmaceuticals 5, 405416
M109.066282 (2012). doi:10.3390/ph5050405
252. W. Lan, S. Yamaguchi, T. Yamamoto, S. Yamahira, M. Tan, N. Murakami, J. 270. U. Tomita, S. Yamaguchi, Y. Maeda, K. Chujo, K. Minamihata, T. Naga-
Zhang, M. Nakamura, T. Nagamune, Visualization of the pH-dependent mune, Protein cell-surface display through insitu enzymatic modifica-
dynamic distribution of G2A in living cells. FASEB J. 28, 39653974 tion of proteins with a poly (ethylene glycol)-lipid. Biotechno. Bioeng.
(2014). doi:10.1096/fj.14-252999 110, 27852789 (2013). doi:10.1002/bit.24933
253. T. Lee, J. Min, H. Hirakawa, T. Nagamune, J.W. Choi, Fusion protein bilayer 271. P. Farka, B. Systrick, Chemical conjugation of biomacromol-
fabrication composed of recombinant azurin/cytochrome P450 by the ecules: a mini-review. Chem. Pap. 64, 683695 (2010). doi:10.2478/
sortase-mediated ligation method. Colloids Surf. B Biointerfaces 120, s11696-010-0057-z
215221 (2014). doi:10.1016/j.colsurfb.2014.03.034 272. J. Kalia, R.T. Raines, Advances in bioconjugation. Curr. Org. Chem. 14,
254. Y. Yamamura, H. Hirakawa, S. Yamaguchi, T. Nagamune, Enhancement of 138147 (2010)
sortase A-mediated protein ligation by inducing a -hairpin structure 273. N. Stephanopoulos, M.B. Francis, Choosing an effective protein bio-
around the ligation site. Chem. Commun. 47, 47424744 (2011). conjugation strategy. Nat. Chem. Biol. 7, 876884 (2011). doi:10.1038/
doi:10.1039/c0cc05334a nchembio.720
255. H. Hirakawa, S. Ishikawa, T. Nagamune, Ca2+-independent sortase-A 274. S.S. van Berkel, M.B. van Eldijk, J.C.M. van Hest, Staudinger ligation as
exhibits high selective protein ligation activity in the cytoplasm of a method for bioconjugation. Angew. Chem. Int. Ed. 50, 88068827
Escherichia coli. Biotechnol. J. 10, 14871492 (2015). doi:10.1002/ (2011). doi:10.1002/anie.201008102
biot.201500012 275. E. Hemaprabha, Chemical crosslinking of proteins: a review. J. Pharm.
256. C. Zhang, A.M. Spokoyny, Y. Zou, M.D. Simon, B.L. Pentelute, Enzymatic Sci. Innov. 1, 2226 (2012)
click ligation: selective cysteine modification in polypeptides enabled 276. C.S. McKay, M.G. Finn, Click chemistry in complex mixtures: bioorthogo-
by promiscuous glutathione S-transferase. Angew. Chem. Int. Ed. 52, nal bioconjugation. Chem. Biol. 21, 10751101 (2014). doi:10.1016/j.
1400114005 (2013). doi:10.1002/anie.201306430 chembiol.2014.09.002
Nagamune Nano Convergence (2017) 4:9 Page 54 of 56

277. D.C. Goldstein, P. Thordarson, J.R. Peterson, The bioconjugation of redox 299. J.W. Choi, Y.S. Nam, B.H. Lee, D.J. Ahn, T. Nagamune, Charge trap in
proteins to novel electrode materials. Aust. J. Chem. 62, 13201327 self-assembled monolayer of cytochrome b562-green fluorescent
(2009). doi:10.1071/CH09240 protein chimera. Curr. Appl. Phys. 6, 760765 (2006). doi:10.1016/j.
278. E. Jabbari, Bioconjugation of hydrogels for tissue engineering. Curr. cap.2005.04.035
Opin. Biotechnol. 22, 655660 (2011). doi:10.1016/j.copbio.2011.01.003 300. Y. Lu, M.B. Lerner, Z.J. Qi, J.J. Mitala Jr., J.H. Lim, B.M. Discher, A.T.C.
279. J.B. Blanco-Canosa, M. Wu, K. Susumu, E. Petryayeva, T.L. Jennings, P.E. Johnson, Graphene-protein bioelectronic devices with wavelength-
Dawson, W.R. Algar, I.L. Medintz, Recent progress in the bioconjuga- dependent photoresponse. Appl. Phys. Lett. 100, 033110 (2012).
tion of quantum dot. Coord. Chem. Rev. 263264, 101137 (2014). doi:10.1063/1.3678024
doi:10.1016/j.ccr.2013.08.030 301. S.R. Schmidt, Fusion-proteins as biopharmaceuticals: applications
280. C.M. Niemeyer, Self-assembled nanostructures based on DNA: towards and challenges. Curr. Opin. Drug Discov. Dev. 12, 284295 (2009).
the development of nanobiotechnology. Curr. Opin. Chem. Biol. 4, doi:10.13140/2.1.4445.5366
609618 (2000). doi:10.1016/S1367-5931(00)00140-X 302. N. Ilk, E.M. Egelseer, U.B. Sleytr, S-layer fusion proteinsconstruction
281. T.A. Taton, R.C. Mucic, C.A. Mirkin, R.L. Letsinger, The DNA-mediated principles and applications. Curr. Opin. Biotechnol. 22, 824831 (2011).
formation of supramolecular mono- and multilayered nanoparticle doi:10.1016/j.copbio.2011.05.510
structures. J. Am. Chem. Soc. 122, 63056306 (2000). doi:10.1021/ 303. C. Suzuki, H. Ueda, E. Suzuki, T. Nagamune, Construction and charac-
ja0007962 terization of anti-hapten scFv-alkalinephosphatase chimeric protein. J.
282. P.E. Nielsen, Peptide nucleic acid: a versatile tool in genetic diagnos- Biochem. 122, 322329 (1997). (PMID: 9378709)
tics and molecular biology. Curr. Opin. Biotechnol. 12, 1620 (2001). 304. C. Suzuki, H. Ueda, K. Tsumoto, W.C. Mahoney, I. Kumagai, T. Naga-
doi:10.1016/S0958-1669(00)00170-1 mune, Open sandwich ELISA with VH-/VL-alkaline phosphatase fusion
283. J. Kurreck, Antisense technologies: improvement through novel proteins. J. Immunol. Methods 224, 171184 (1999). doi:10.1016/
chemical modifications. Eur. J. Biochem. 270, 16281644 (2003). S0022-1759(99)00020-4
doi:10.1046/j.1432-1033.2003.03555.x 305. T. Yokozeki, H. Ueda, R. Arai, W.C. Mahoney, T. Nagamune, A homogene-
284. R.P. Singh, B.K. Oh, J.W. Choi, Application of peptide nucleic acid ous noncompetitive immunoassay for the detection of small haptens.
towards development of nanobiosensor arrays. Bioelectrochemistry 79, Anal. Chem. 74, 25002504 (2002). doi:10.1021/ac015743x
153161 (2010). doi:10.1016/j.bioelechem.2010.02.004 306. Y. Ohiro, R. Arai, H. Ueda, T. Nagamune, A homogeneous and noncom-
285. N. Shi, R.J. Boado, W.M. Pardridge, Antisense imaging of gene expression petitive immunoassay based on the enhanced fluorescnce reso-
in the brain invivo. Proc. Natl. Acad. Sci. USA 97, 1470914714 (2000). nance energy transfer by leucine zipper interaction. Anal. Chem. 74,
doi:10.1073/pnas.250332397 57865792 (2002). doi:10.1021/ac0203387
286. P. Hongmanee, H. Stender, O.F. Rasmussen, Evaluation of a fluorescence 307. N. Komiya, H. Ueda, Y. Ohiro, T. Nagamune, Homogeneous sandwich
insitu hybridization assay for differentiation between tuberculous immunoassay based on the enzymatic complementation induced
and nontuberculous Mycobacterium species in smears of Lowen- by single-chain Fv fragments. Anal. Biochem. 327, 241246 (2004).
steinJensen and mycobacteria growth indicator tube cultures using doi:10.1016/j.ab.2004.01.035
peptide nucleic acid probes. J. Clin. Microbiol. 39, 10321035 (2001). 308. R. Abe, H. Ohashi, I. Iijima, M. Ihara, H. Takagi, T. Hohsaka, H. Ueda,
doi:10.1128/JCM.39.3.1032-1035.2001 Quenchbodies: quench-based antibody probes that show antigen-
287. H. Aoki, Y. Umezawa, Trace analysis of an oligonucleotide with a specific dependent fluorescence. J. Am. Chem. Soc. 133, 1738617394 (2011).
sequence using PNA-based ion-channel sensors. Analyst 128, 681685 doi:10.1021/ja205925j
(2003). doi:10.1039/B300465A 309. K.N. Islam, M. Ihara, J. Dong, N. Kasagi, T. Mori, H. Ueda, Direct construc-
288. D.B. Wetlaufer, Nucleation, rapid folding, and globular intrachain tion of an open-sandwich enzyme immunoassay for one-step noncom-
regions in proteins. Proc. Natl. Acad. Sci. USA 70, 697701 (1973) petitive detection of thyroid hormone T4. Anal. Chem. 83, 10081014
289. M. Go, Correlation of DNA exonic regions with protein structural units (2011). doi:10.1021/ac102801r
in haemoglobin. Nature 291, 9092 (1981). doi:10.1038/291090a0 310. H.J. Jeong, Y. Ohmuro-Matsuyama, H. Ohashi, F. Ohsawa, Y. Tatsu, M.
290. K. Yu, C. Liu, B.G. Kim, D.Y. Lee, Synthetic fusion protein design and Inagaki, H. Ueda, Detection of vimentin serine phosphorylation by
applications. Biotechnol. Adv. 33, 155164 (2015). doi:10.1016/j. multicolor Quenchbodies. Biosens. Bioelectron. 40, 1723 (2013).
biotechadv.2014.11.005 doi:10.1016/j.bios.2012.06.030
291. K. Terpe, Overview of tag protein fusions: from molecular and biochem- 311. T. Sogo, M. Kawahara, H. Ueda, M. Otsu, M. Onodera, H. Nakauchi, T.
ical fundamentals to commercial systems. Appl. Microbiol. Biotechnol. Nagamune, T cell growth control using hapten-specific antibody/inter-
60, 523533 (2003). doi:10.1007/s00253-002-1158-6 leukin-2 receptor chimera. Cytokine 46, 127136 (2009). doi:10.1016/j.
292. S. Costa, A. Almedia, A. Castro, L. Domingues, Fusion tags for protein cyto.2008.12.020
solubility, purification and immunogenicity in Escherichia coli: the novel 312. M. Kawahara, H. Ueda, T. Nagamune, Engineering cytokine receptors
Fh8 system. Front. Microbiol. 5, 63 (2014). doi:10.3389/fmicb.2014.00063 to control cellular functions. Biochem. Eng. J. 48, 283294 (2010).
293. A. Mller-Taubenberger, K.I. Anderson, Recent advances using green doi:10.1016/j.bej.2009.09.010
and red fluorescent protein variants. Appl. Microbiol. Biotechnol. 77, 313. M. Kawahara, J. Chen, T. Sogo, J. Teng, M. Otsu, M. Onodera, H. Nakauchi,
112 (2007). doi:10.1007/s00253-007-1131-5 H. Ueda, T. Nagamune, Growth promotion of genetically modified
294. H. Yang, L. Liu, F. Xu, The promises and challenges of fusion constructs hematopoietic progenitors using an antibody/c-Mpl chimera. Cytokine
in protein biochemistry and enzymology. Appl. Microbiol. Biotechnol. 55, 402408 (2011). doi:10.1016/j.cyto.2011.05.024
100, 82738281 (2016). doi:10.1007/s00253-016-7795-y 314. M. Kawahara, T. Nagamune, Engineering of mammalian cell membrane
295. G. Gilardi, Y.T. Meharenna, G.E. Tsotsou, S.J. Sadeghi, M. Fairhead, S. Gian- proteins. Curr. Opin. Chem. Eng. 1, 411417 (2012). doi:10.1016/j.
nini, Molecular Lego: design of molecular assemblies of P450 enzymes coche.2012.05.002
for nanobiotechnology. Biosens. Bioelectron. 17, 133145 (2002). 315. K. Saka, M. Kawahara, T. Nagamune, Reconstitution of a cytokine
doi:10.1016/S0956-5663(01)00286-X receptor scaffold utilizing multiple different tyrosine motifs. Biotechnol.
296. S.J. Sadeghi, G. Gilardi, Chimeric P450 enzymes: activity of artificial redox Bioeng. 110, 31973204 (2013). doi:10.1002/bit.24973
fusions driven by different reductases for biotechnological applications. 316. Y. Tone, M. Kawahara, D. Kawaguchi, H. Ueda, T. Nagamune, Death
Biotechnol. Appl. Biochem. 60, 102110 (2013). doi:10.1002/bab.1086 signalobody: inducing conditional cell death in response to a specific
297. K.Y. Choi, E.O. Jung, D.H. Jung, B.P. Pandey, N. Lee, H. Yun, H. Park, B.G. Kim, antigen. Hum. Gene Ther. Methods. 24, 141150 (2013). doi:10.1089/
Novel ironsulfur containing NADPH-reductase from Nocardia farcinica hgtb.2012.147
IFM10152 and fusion construction with CYP51 lanosterol demethylase. 317. Y. Tone, M. Kawahara, J. Hayashi, T. Nagamune, Cell fate conversion by
Biotechnol. Bioeng. 109, 630636 (2012). doi:10.1002/bit.24359 conditionally switching the signal-transducing domain of signalbodies.
298. B. Lee, S. Takeda, K. Nakajima, J. Noh, J.W. Choi, M. Hara, T. Nagamune, Biotechnol. Bioeng. 110, 32193226 (2013). doi:10.1002/bit.24985
Rectified photocurrent in a protein based molecular photo-diode 318. M. Kawahara, A. Hitomi, T. Nagamune, S-Fms signalobody enhances
consisting of a cytochrome b 562-green fluorescent protein chimera. Bio- myeloid cell growth and migration. Biotechnol. J. 9, 954961 (2014).
sens. Bioelectron. 19, 11691174 (2004). doi:10.1016/j.bios.2003.11.016 doi:10.1002/biot.201300346
Nagamune Nano Convergence (2017) 4:9 Page 55 of 56

319. H. Nakabayashi, S. Aoyama, M. Kawahara, T. Nagamune, Differentiation 340. H. Ueda, T. Yokozeki, R. Arai, K. Tsumoto, I. Kumagai, T. Nagamune, An
signalobody: demonstration of antigen-dependent osteoclast differen- optimized homogeneous noncompetitive immunoassay based on
tiation from a progenitor cell line. J. Biosci. Bioeng. 122, 357363 (2016). the antigen-driven enzymatic complementation. J. Immunol. Methods
doi:10.1016/j.jbiosc.2016.02.010 279, 209218 (2003). doi:10.1016/S0022-1759(03)00256-4
320. T. Miura, T. Nagamune, M. Kawahara, Ligand-induced dimeric antibody 341. K. Saka, M. Kawahara, T. Nagamune, Quantitative control of intracellular
for selecting antibodies against a membrane protein based on mam- signaling activity through chimeric receptors incorporating multiple
malian cell proliferation. Biotechnol. Bioeng. 113, 11131123 (2016). identical tyrosine motifs. Biotechnol. Bioeng. 111, 948955 (2014).
doi:10.1002/bit.25858 doi:10.1002/bit.25151
321. S. Lee, Y. Kaku, S. Inoue, T. Nagamune, M. Kawahara, Growth signalo- 342. W. Liu, M. Kawahara, H. Ueda, T. Nagamune, Construction of a
body selects functional intrabodies in the mammalian cytoplasm. fluorescein-responsive chimeric receptor with strict ligand depend-
Biotechnol. J. 11, 565573 (2016). doi:10.1002/biot.201500364 ency. Biotechnol. Bioeng. 101, 975984 (2008). doi:10.1002/bit.21961
322. J. Winkler, Nanomedicines based on recombinant fusion proteins for tar- 343. T. Haga, H. Hirakawa, T. Nagamune, Fine tuning of spatial arrangement
geting therapeutic siRNA oligonucleotides. Ther. Deliv. 2, 891905 (2011) of enzymes in a PCNA-mediated multienzyme complex using a rigid
323. W.M. Pardridge, Bloodbrain barrier drug delivery of IgG fusion proteins poly-l-proline linker. PLoS ONE 8, e75114 (2013). doi:10.1371/journal.
with a transferrin receptor monoclonal antibody. Expert Opin. Drug pone.0075114
Deliv. 12, 207222 (2015). doi:10.1517/17425247.2014.952627 344. F. Xue, Z. Gu, J.A. Feng, LINKER: a web server to generate peptide
324. X. Chen, J.L. Zaro, W.C. Shen, Fusion protein linkers: property, design and sequences with extended conformation. Nucleic Acids Res. 32, W562
functionality. Adv. Drug Deliv. Rev. 65, 13571369 (2013). doi:10.1016/j. W565 (2004). doi:10.1093/nar/gkh422
addr.2012.09.039 345. P. Lu, M.G. Feng, Bifunctional enhancement of a -glucanase-xylanase
325. P. Argos, An investigation of oligopeptides linking domains in protein fusion enzyme by optimization of peptide linkers. Appl. Microbiol.
tertiary structures and possible candidates for general gene fusion. J. Biotechnol. 79, 579587 (2008). doi:10.1007/s00253-008-1468-4
Mol. Biol. 211, 943958 (1990). doi:10.1016/0022-2836(90)90085-Z 346. R. Arai, H. Ueda, A. Kitayama, N. Kamiya, T. Nagamune, Design of the
326. S. Ohnishi, H. Kamikubo, M. Onitsuka, M. Kataoka, D. Shortle, Conforma- linkers which effectively separate domains of a bifunctional fusion
tional preference of polyglycine in solution to elongated structure. J. protein. Protein Eng. 14, 529532 (2001). doi:10.1093/protein/14.8
Am. Chem. Soc. 128, 1633816344 (2006). doi:10.1021/ja066008b 347. R. Arai, W. Wriggers, Y. Nishikawa, T. Nagamune, T. Fujisawa, Conformations
327. R. Bird, K. Hardman, J. Jacobson, S. Johnson, B. Kaufman, S. Lee, T. Lee, of variably linked chimeric proteins evaluated by synchrotron X-ray small-
S. Pope, G. Riordan, M. Whitlow, Single-chain antigen-binding proteins. angle scattering. Proteins 57, 829838 (2004). doi:10.1002/prot.20244
Science 242, 423426 (1988). doi:10.1126/science.3140379 348. N. Kobayashi, K. Yanase, T. Sato, S. Unzai, M.H. Hecht, R. Arai, Self-assem-
328. S. Marqusee, R.L. Baldwin, Helix stabilization by GluLys+ salt bridges bling nano-architectures created from a protein nano-building block
in short peptides of de novo design. Proc. Natl. Acad. Sci. USA 84, using an intermolecularly folded dimeric de novo protein. J. Am. Chem.
88988902 (1987). doi:10.1073/pnas.84.24.8898 Soc. 137, 1128511293 (2015). doi:10.1021/jacs.5b03593
329. R.A. George, J. Heringa, An analysis of protein domain linkers: their clas- 349. C.J. Crasto, J.A. Feng, LINKER: a program to generate linker sequences
sification and role in protein folding. Protein Eng. 15, 871879 (2002). for fusion proteins. Protein Eng. 13, 309312 (2000). doi:10.1093/
doi:10.1093/protein/15.11.871 protein/13.5.309
330. C.L. Young, Z.T. Britton, A.S. Robinson, Recombinant protein expres- 350. C. Liu, J.X. Chin, D.Y. Lee, SynLinker: an integrated system for designing
sion and purification: a comprehensive review of affinity tags and linkers and synthetic fusion proteins. Bioinformatics 31, 37003702
microbial applications. Biotechnol. J. 7, 620626 (2012). doi:10.1002/ (2015). doi:10.1093/bioinformatics/btv447
biot.201100155 351. J. Chiang, I. Li, E. Pham, K. Truong, FPMOD: a modeling tool for sampling
331. D.L. Crimmins, S.M. Mische, N.D. Denslow, Chemical cleavage of the conformational space of fusion proteins. Conf. Proc. IEEE Eng. Med.
proteins in solution. Curr. Protoc. Protein Sci. Chapter11, 11.4.111.4.11 Biol. Soc. 1, 41114114 (2006). doi:10.1109/IEMBS.2006.259224
(2005). doi: 10.1002/0471140864.ps1104s40 352. E. Pham, J. Chiang, I. Li, W. Shum, K. Truong, A computational tool for
332. Y. Li, Self-cleaving fusion tags for recombinant protein production. designing FRET protein biosensors by rigid-body sampling of their
Biotechnol. Lett. 33, 869881 (2011). doi:10.1007/s10529-011-0533-8 conformational space. Structure 15, 515523 (2007). doi:10.1016/j.
333. W.J. Bradshaw, A.H. Davies, C.J. Chambers, A.K. Roberts, C.C. Shone, K.R. str.2007.03.009
Acharya, Molecular features of the sortase enzyme family. FEBS J. 282, 353. A.M. Wollacott, A. Zanghellini, P. Murphy, D. Baker, Prediction of
20972114 (2015). doi:10.1111/febs.13288 structures of multidomain proteins from structures of the individual
334. X. Chen, Y. Bai, J. Zaro, W.C. Shen, Design of an invivo cleavable disulfide domains. Protein Sci. 16, 165175 (2007). doi:10.1110/ps.062270707
linker in recombinant fusion proteins. Biotechniques 49, 513518 354. T.M. Cheng, T.L. Blundell, J. Fernandez-Recio, Structural assembly of
(2010). doi:10.2144/000113450 two-domain proteins by rigid-body docking. BMC Bioinform. 9, 441
335. H.L. Zhao, C. Xue, J.L. Du, M. Ren, S. Xia, Z.M. Liu, Balancing the pharma- (2008). doi:10.1186/1471-2105-9-441
cokinetics and pharmacodynamics of interferon-alpha2b and human 355. Y. Zhang, I-TASSER server for protein 3D structure prediction. BMC
serum albumin fusion protein by proteolytic or reductive cleavage Bioinform. 9, 40 (2008). doi:10.1186/1471-2105-9-40
increases its invivo therapeutic efficacy. Mol. Pharm. 9, 664670 (2012). 356. S. Lovell, I. Davis, B. Arendall III, P. Bakker, M. Word, M. Prisant, J. Rich-
doi:10.1021/mp200347q ardson, D. Richardson, Structure validation by C-geometry: , and
336. M. Gustavsson, J. Lehti, S. Denman, T.T. Teeri, K. Hult, M. Martinelle, C deviation. Proteins Struct. Funct. Bioinform. 50, 437450 (2003).
Stable linker peptides for a cellulose-binding domain-lipase fusion doi:10.1002/prot.10286
protein expressed in Pichia pastoris. Protein Eng. 14, 711715 (2001). 357. S. Shamriz, H. Ofoghi, N. Moazami, Effect of linker length and residues
doi:10.1093/protein/14.9.711 on the structure and stability of a fusion protein with malaria vac-
337. N. Adlakha, S. Sawant, A. Anil, A. Lali, S.S. Yazdani, Specific fusion of cine application. Comput. Biol. Med. 76, 2429 (2016). doi:10.1016/j.
-1,4-endoglucanase and -1,4-glucosidase enhances cellulolytic activ- compbiomed.2016.06.015
ity and helps in channeling of intermediates. Appl. Environ. Microbiol. 358. J. Arunachalam, V. Kanagasabai, N. Gautham, Protein structure predic-
78, 74477454 (2012). doi:10.1128/AEM.01386-12 tion using mutually orthogonal Latin squares and a genetic algorithm.
338. G.G. Yang, X.Y. Xu, Y. Ding, Q.Q. Cui, Z. Wang, Q.Y. Zhang, S.H. Shi, Z.Y. Biochem. Biophys. Res. Commun. 342, 424433 (2006). doi:10.1016/j.
Lv, X.Y. Wang, J.H. Zhang, R.G. Zhang, C.S. Xu, Linker length affects bbrc.2006.01.162
expression and bioactivity of the onconase fusion protein in Pichia 359. A. Nicola, P. Delarue, P. Senet, Decipher the mechanisms of protein
pastoris. Genet. Mol. Res. 14, 1936019370 (2015). doi:10.4238/2015. conformational changes induced by nucleotide binding through free-
December.29.46 energy landscape analysis: ATP binding to Hsp70. PLoS Comput. Biol. 9,
339. Y. Maeda, H. Ueda, J. Kazami, G. Kawano, E. Suzuki, T. Nagamune, e1003379 (2013). doi:10.1371/journal.pcbi.1003379
Engineering of functional chimeric protein G-Vargula luciferase. Anal. 360. N.P. King, Y.T. Lai, Practical approaches to designing novel protein
Biochem. 249, 147152 (1997). doi:10.1006/abio.1997.2181 assemblies. Curr. Opin. Struct. Biol. 23, 632638 (2013). doi:10.1016/j.
sbi.2013.06.002
Nagamune Nano Convergence (2017) 4:9 Page 56 of 56

361. R. Wang, Y. Zhang, D. Lu, J. Ge, Z. Liu, R.N. Zare, Functional protein- 364. Q. Luo, C. Hou, Y. Bai, R. Wang, J. Liu, Protein assembly: versatile
organic/inorganic hybrid nanomaterials. WIREs Nanomed. Nanobio- approaches to construct highly ordered nanostructures. Chem. Rev.
technol. 5, 320328 (2013). doi:10.1002/wnan.1210 116, 1357113632 (2016). doi:10.1021/acs.chemrev.6b00228
362. L.H.L. Lua, N.K. Connors, F. Sainsbury, Y.P. Chuan, N. Wibowo, A.P.J. Mid- 365. S. Hiyama, Y. Moritani, R. Gojo, S. Takeuchi, K. Sutoh, Biomolecular-
delberg, Bioengineering virus-like particles as vaccines. Biotechnol. motor-based autonomous delivery of lipid vesicles as nano- or micro-
Bioeng. 111, 425440 (2014). doi:10.1002/bit.25159 scale reactors on a chip. Lab Chip 10, 27412748 (2010). doi:10.1039/
363. D. Romera, P. Couleaud, S.H. Mejias, A. Aires, A.L. Cortajarena, Biomo- c004615a
lecular templating of functional hybrid nanostructures using repeat
protein scaffolds. Biochem. Soc. Trans. 43, 825831 (2015). doi:10.1042/
BST20150077