Synthetic Modularity of Protein-Metal-Organic Frameworks

Article abstract of DOI:10.1021/jacs.7b01202

Previously, we adopted the construction principles of metal-organic frameworks (MOFs) to design a 3D crystalline protein lattice in which pseudospherical ferritin nodes decorated on their C₃ symmetric vertices with Zn coordination sites were connected via a ditopic benzene-dihydroxamate linker. In this work, we have systematically varied both the metal ions presented at the vertices of the ferritin nodes (Zn(II), Ni(II), and Co(II)) and the synthetic dihydroxamate linkers, which yielded an expanded library of 15 ferritin-MOFs with the expected body-centered (cubic or tetragonal) lattice arrangements. Crystallographic and small-angle X-ray scattering (SAXS) analyses indicate that lattice symmetries and dimensions of ferritin-MOFs can be dictated by both the metal and linker components. SAXS measurements on bulk crystalline samples reveal that some ferritin-MOFs can adopt multiple lattice conformations, suggesting dynamic behavior. This work establishes that the self-assembly of ferritin-MOFs is highly robust and that the synthetic modularity that underlies the structural diversity of conventional MOFs can also be applied to the self-assembly of protein-based crystalline materials.

Full text of DOI:10.1021/jacs.7b01202

Article
pubs.acs.org/JACS
Synthetic Modularity of Protein−Metal−Organic Frameworks
Jake B. Bailey, Ling Zhang, Jerika A. Chiong, Sunhyung Ahn, and F. Akif Tezcan*
Department of Chemistry and Biochemistry, University of California at San Diego, 9500 Gilman Drive, La Jolla, California 92093,
United States
S
* Supporting Information
ABSTRACT: Previously, we adopted the construction
principles of metal−organic frameworks (MOFs) to design a
3D crystalline protein lattice in which pseudospherical ferritin
nodes decorated on their C3 symmetric vertices with Zn
coordination sites were connected via a ditopic benzene-
dihydroxamate linker. In this work, we have systematically
varied both the metal ions presented at the vertices of the
ferritin nodes (Zn(II), Ni(II), and Co(II)) and the synthetic
dihydroxamate linkers, which yielded an expanded library of 15 ferritin−MOFs with the expected body-centered (cubic or
tetragonal) lattice arrangements. Crystallographic and small-angle X-ray scattering (SAXS) analyses indicate that lattice
symmetries and dimensions of ferritin−MOFs can be dictated by both the metal and linker components. SAXS measurements on
bulk crystalline samples reveal that some ferritin−MOFs can adopt multiple lattice conformations, suggesting dynamic behavior.
This work establishes that the self-assembly of ferritin−MOFs is highly robust and that the synthetic modularity that underlies
the structural diversity of conventional MOFs can also be applied to the self-assembly of protein-based crystalline materials.
1b).41 Despite the remarkable size mismatch between the p-
INTRODUCTION
■
H2bdh linkers (ca. 9 Å long, 196 g/mol) and ferritin nodes
Periodic protein arrays constitute a major component of the
(120 Å across, >505 000 g/mol), their interactions were
cellular machinery and are widely utilized as platforms for nano-
sufficiently strong to dictate the arrangement of ferritin
and biotechnological applications due to their advanced
molecules into the desired body-centered cubic (bcc) lattice.
materials properties and precise display of diverse chemical
This system introduced a new class of tripartite hybrid materials
functionalities over the nm−μm scale.1−3 Accordingly, there
whose self-assembly is dependent on each of its components:
has been a growing interest in the construction of artificial
protein, metal, and organic linker. In this study, we set out to
protein assemblies.4−7 These efforts have engendered innova-
examine if the modularity that is inherent in conventional
tive design strategies and chemical approaches, which have led
MOFs and renders them particularly attractive as functional
to 0,8−11 1,12−15 2,12,16−18 and 3D19−23 assemblies with
materials also applies to protein−MOFs. Specifically, we asked
whether it is possible to synthetically modulate the formation
and structural parameters of protein−MOFs through discrete
molecular interactions.
complex structures and, in some cases, sophisticated24,25 and
evolvable functions,26 as well as emergent properties not yet
observed in biology.18,27 Despite these advances, the structural/
chemical heterogeneity of proteins still poses a substantial
challenge for designing supramolecular protein architectures
and arrays. While most design approaches are intended to be
generalizable, they still require adjustment of the protein
building blocks or the self-assembly procedures on a case-by-
case basis. In contrast, the thematically related field of
supramolecular chemistry has benefitted from synthetic access
to a large library of molecular building blocks and bonding
strategies,28,29 which can be mixed and matched to create
structural and functional diversity with relative ease. Such
synthetic versatility is aptly highlighted by metal−organic
frameworks (MOFs),30−40 a vast class of crystalline materials
composed of metal-based nodes and organic linkers that can be
combined in a modular fashion.
RESULTS AND DISCUSSION
■
Construction of Ferritin−MOF Components.
Metal−Ferritin Nodes. To enable a modular construction
approach, we sought to diversify the metal and the organic
linker components of the ferritin−MOFs (Figure 1). As
previously described,41 the 24meric human heavy-chain ferritin
(which we refer to as ferritin in this report) provides an
attractive building block for protein−MOFs for two primary
reasons. First, from a structural perspective, ferritin possesses an
octahedral (432) symmetry reminiscent of many metal nodes
employed in archetypal MOFs,42,43 which could, in principle,
provide access to different lattice arrangements (e.g., simple,
face-centered and body-centered cubic) depending on the
surface location of engineered metal binding sites. Second,
We recently adopted the construction strategies of MOFs to
create 3D protein lattices in which pseudospherical ferritin
molecules decorated on their outer surfaces with metal
coordination sites were bridged via ditopic metal-chelating
linkers (benzene-1,4-dihydroxamic acid; p-H2bdh or 1; Figure
Received: February 3, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.7b01202
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. Modular components of ferritin−MOFs. (a) The ferritin node (metal ions at the C3 vertices are shown as teal spheres). Close-up views of
the Zn(II), Ni(II), and Co(II) coordination environments. Water molecules are shown as red spheres. 2Fo−Fc maps are contoured to 1σ. (b) Ditopic
hydroxamic acid linkers.
ferritin is inherently functional and catalyzes iron biomineral-
ization in its hollow interior (8 nm in diameter). The native
ability of ferritin to act as a molecular template and container
has in fact been exploited in numerous applications.44−50 We
previously showed that, upon assembly into MOFs, the ferritin
nodes are still capable of iron mineralization, whereby the high
lattice porosity allows for rapid diffusion of Fe and its uptake
into the individual ferritin nodes, likely through the C4
symmetric pores.41,51
face-centered cubic (fcc; F432 space group) lattices by joining
ferritin molecules across their C2 symmetric interfaces via
coordination to pairs of D84 and engineered Q86 residues.52
The crystal structures of Ni− and Co−ferritin were determined
at 1.79 and 1.95 Å resolution, respectively (Table S2; PDB IDs:
5UP7 and 5VTD). These structures confirm that Ni(II) and
Co(II) ions are anchored by the tripodal H122 coordination
motifs at full occupancy, with three aquo ligands completing
nearly ideal octahedral coordination spheres. Relative to the
tetrahedral Zn coordination sites in Zn−ferritin, in Ni− and
Co−ferritin, the His122 side chains have moved slightly to
accommodate the octahedral geometry, with NHis−M−NHis
angles of 95°−96°, NHis−M−OH2 angles of 94° (Ni) and
87° (Co), and OH2−M−OH2 angles of 75° (Ni) and 82°
(Co). The M−NHis distances are 2.1 Å for both species, and the
M−OH2 distances are 2.1 Å for Ni−ferritin and 2.2 Å for Co−
ferritin. These observations thus establish that Ni− and Co−
ferritin nodes are poised for coordinating bridging linkers.
Synthetic Linkers. With the three distinct metallo−ferritin
nodes in hand, we next prepared a set of five ditopic ligands
(1−5) bearing hydroxamic acid head groups (Figure 1b), which
have been previously reported.41,53−56 For the synthesis of
ligands 1 and 4 (benzene-1,3-dihydroxamic acid; m-H2bdh), we
followed published protocols.41,53 Ligands 2 (E-ethylenedihy-
droxamic acid or E-H2edh), 3 (naphthalene-2,6-dihydroxamic
acid or 2,6-H2ndh), and 5 (xylene-1,4-dihydroxamic acid or p-
H2xdh) were prepared by the amidation of the respective
dicarboxylate-bearing precursors with O-tritylhydroxylamine,
followed by deprotection with trifluoroacetic acid to furnish the
dihydroxamic acid ligands (see the Supporting Information for
details on linker synthesis and characterization). Hydroxamate
functionalities were chosen due to their high affinity for Zn(II),
Ni(II), and Co(II), as well as their sterically unencumbered
nature compared to other commonly used aromatic chelates
(e.g., polypyridyl or catecholate-type ligands), allowing for
unhindered access to the surface-anchored metal ions without
bias from peripheral contacts.
In our initial studies, we deemed the C3 symmetric pores of
ferritin (essentially the eight vertices of a cube) as the most
suitable locations for the installment of metal anchoring sites
(Figure 1a). These corner locations allow for the installment of
surface-exposed tripodal coordination motifs that can bind
various metal ions with high affinity, while still presenting
outward-facing open coordination sites available for organic
linkers. While the C4 symmetric pores also permit stable, square
planar metal anchoring sites to be built, the appropriate
positions (residue 173 or 165) in these pores are somewhat
recessed from the protein surface and therefore not
immediately available as connection points for organic linkers.
Technically, tri- or tetradentate coordination sites could also be
incorporated onto the C2 symmetric surfaces of ferritin, but this
would require extensive design and engineering and result in
multiple metal binding sites on the same surface. To build
metal anchoring sites at the C3 symmetric pores, Thr122
residues lining the outer rim of these pores were replaced with
His. In our previous study, we established that the resulting
variant (T122Hferritin) bound Zn(II) ions in the expected
tetrahedral geometry with a tripodal base of three His side
chains and a single, solvent-bound coordination site pointing to
the protein exterior (Figures 1a and S1).41 We will hereafter
refer to this ferritin variant as Zn−ferritin.
In this study, we prepared the Ni− and Co−adducts of
T122Hferritin (Ni− and Co−ferritin) in a similar fashion as Zn−
ferritin (Table S1), whereby a large molar excess of Ni(II) or
Co(II) was added to ferritin solutions to saturate all available
metal binding sites within the protein cage, including the
installed tripodal coordination motifs at the C3 vertices. Ni−
and Co−ferritin were crystallized by vapor diffusion in the
presence of Ca2+, which selectively induces the formation of
The primary differences between ligands 1−5 can be
described by two parameters. First, the interhydroxamate
spacing (i.e., the linker length) of these ligands ranges from
<7.0 Å in 2 to >11 Å in 3 and 5. Second, the ligands offer four
B
DOI: 10.1021/jacs.7b01202
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 2. sc-XRD structures for (a) 1−Zn−ferritin and (b) 5−Zn−ferritin. Zinc atoms are highlighted as teal spheres. 2Fo−Fc electron density maps
are contoured at 1σ (blue) and 3σ (orange). The body diagonals of the unit cells are shown as purple lines.
different geometries in terms of the relative orientations of the
two hydroxamate head groups. Here, we define these
orientations with respect to the vector along the C−C bond
(highlighted in red in Figures 1b and S1) that appends the
hydroxamate moieties to the core of each ligand. In 1, the bond
vectors are collinear; in 2 and 3, they are parallel but offset; and
in 4, they form an obtuse angle. Finally, in 5, the vectors are not
fixed due to the rotational degrees of freedom about the
methylene spacers: they can yield parallel or bent orientations
but not collinear. Ligand 5 thus offers another variation among
the five linkers, namely increased conformational flexibility.
Self-Assembly of Ferritin−MOFs and Crystallographic
Characterization. Under previously established self-assembly
conditions,41 all 15 combinations of the three ferritin nodes
(Zn−ferritin, Ni−ferritin, and Co−ferritin) and five organic
linkers reproducibly yielded single-crystalline particles with
rhombic dodecahedral morphology (Figure S2). Self-assembly
of the crystalline lattices occurred within 12−24 h in aqueous
solutions containing 4.2 μM ferritin, 72 equiv (per 24meric
ferritin cage) of Zn(II), Ni(II), or Co(II) and 2 mM organic
linker at pH 9.5 and 23 °C. The sizes of these crystals ranged
from 5 to 200 μm. No crystals formed when any of the three
components (ferritin, metal, or linker) was omitted from the
self-assembly solutions, providing strong evidence for the
tripartite composition of the crystal lattices.
To understand the assembly of ferritin−MOFs in detail and
to examine the relationship between lattice symmetry and
metal/linker geometry, we pursued single-crystal X-ray
diffraction (sc-XRD) experiments. Ferritin−MOFs pose a
unique challenge in this regard in that the constituent building
blocks are very large molecules bridged by small linkers with
inherent flexibility and exceedingly small interaction footprints.
In terms of node/linker mass ratios, ferritin−MOFs (ca. 2500/
1) are akin to a lattice of regulation-size soccer balls held
together by wooden toothpicks. Remarkably, the total footprint
of hydroxamate−metal interactions (ca. 22 Å2 × 8 in bcc
arrangement) represents only 0.4% of the outer surface of a
ferritin cage (ca. 44,000 Å2). In a densely packed protein crystal
like that of the globular protein sperm whale myoglobin (PDB
ID: 5IKS, solvent content = 0.35), the fraction of the total
protein surface area engaged in lattice packing contacts is 16%.
In the case of a sparsely packed protein crystal (PDB ID: 1B5S,
solvent content = 0.89), this fraction is 8%, that is, 20-fold
higher than in ferritin−MOFs (PDB ID: 5CMR, solvent
content = 67%). These comparisons intimate that the ferritin−
MOFs should be quite dynamic and highly sensitive to external
perturbations and, at first glance, not conducive to high-
resolution structure determination by sc-XRD. Earlier screening
efforts led to single crystals of 1−Zn−ferritin which diffracted
to 3.8-Å resolution at best, with the great majority diffracting to
>6.0 Å.41 Here, we carried out a broader screen of the self-
assembly conditions, which enabled us to obtain midresolution
sc-XRD data for two ferritin−MOF variants and determine
their crystal structures (Table S1): 1−Zn−ferritin (improved
from 3.8 to 2.63 Å resolution, I432, a = 155.8 Å; PDB ID:
5UP8) and 5−Zn−ferritin (2.45 Å resolution, I4, a = 149.88 Å,
c = 162.23 Å; PDB ID: 5UP9).
In the bcc crystals of 1−Zn−ferritin, the C3 symmetry axes of
the individual ferritin molecules are perfectly aligned with the
crystallographic 3-fold symmetry axes that form the body
diagonals in the unit cells (Figure 2a). Consequently, the length
of each body diagonal (269.9 Å) is the sum of the diameters of
two ferritin molecules (i.e., 2 × 123.4 Å, as measured by the
distance between two Zn ions along the C3 axis within each
ferritin molecule) and twice the separation between the Zn ions
across the linker-mediated interface (i.e., 2 × 11.5 Å). Owing to
the cubic symmetry, each linker 1 is centered precisely at the
intersection of the crystallographic 2- and 3-fold symmetry axes.
This positioning, combined with the rotational averaging of
linker 1 about the 3-fold axis, gives rise to a diffuse electron
density for the linker, even at the considerably improved
resolution limit (Figure 2a).
These complications are eliminated in the case of the body-
centered tetragonal (bct) lattices of 5−Zn−ferritin which lack
crystallographic 3-fold symmetry and therefore present a
unique orientation of linker 5 between the two crystallo-
graphically nonequivalent Zn centers (Figure 2b). The
resultant, well-defined electron density allowed us to build an
unambiguous structural model of linker 5 and the Zn
coordination environments. This model confirms that the
hydroxamate head groups bind the Zn centers in a bidentate
geometry in the apical position, completing a pseudotetrahedral
C
DOI: 10.1021/jacs.7b01202
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 3. SAXS profiles of Zn−, Ni−, and Co−ferritin−MOFs mediated by linkers 1−5. Simulated diffraction patterns of the primary lattice are
shown in blue. Simulated patterns for any additional lattices are shown in orange or red. See Table 1 for the lattice parameters.
Zn coordination geometry, with both Fe−O coordination
distances at 2.2 Å. The xylene moiety of the linker assumes a
nearly parallel orientation with respect to the ferritin−ferritin
interfacial plane, giving rise to a large (ca. 3.75 Å) lateral
displacement of the Zn ions from the body diagonal of the bct
unit cell, thereby yielding a considerable cubic-to-tetragonal
distortion (Figure 2b). Due to the flat orientation of the linker,
the interfacial spacing between the ferritin molecules (as
defined by the projected Zn−Zn distance) is reduced to 9.5 Å
from 11.5 Å observed in the cubic 1−Zn−ferritin lattices
(Figure 2). This compaction brings a small number of
interfacial side chains into a distance range (3.8−5 Å) where
electrostatic interactions between ferritin molecules may be
invoked (Figure S3). However, based on our observation that
all three components of ferritin−MOFs are necessary to form
crystalline lattices and the fact that body-centered lattices (i.e.,
I432 and I4) have otherwise not been observed with human
heavy-chain ferritin, we can safely conclude that linker-
mediated interactions are the driving force for the formation
of the observed ferritin−MOF lattices.
Characterization of Ferritin−MOFs by Small-Angle X-
ray Scattering. Although sc-XRD experiments can provide
highly detailed structural information, they also require atomic-
level registry between the constituents of a lattice, which can be
challenging to achieve in the case of ferritin−MOFs. Yet, as the
cartoons in Figure 2 illustrate, there is a rather simple
relationship between the crystallographic parameters of
ferritin−MOFs (i.e., symmetry and unit cell parameters) and
the linker-mediated protein−protein interactions. Thus, we
turned to small-angle X-ray scattering (SAXS) experiments
which can furnish the desired crystallographic parameters
without the need for sc-XRD-quality crystals and enable bulk
measurements that can help identify any heterogeneity in the
D
DOI: 10.1021/jacs.7b01202
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 1. Unit Cell Parameters for Ferritin−MOFs Shown in Figure 3
a
b
c
Simulated SAXS pattern is shown in orange in Figure 3. Simulated SAXS pattern is shown in red in Figure 3. The additional peaks observed for
3−M−ferritin could be indicative of a primitive (P) tetragonal unit cell (see Figure S4).
crystalline samples which would be missed in sc-XRD
experiments.
All 15 ferritin MOFs produced well-defined SAXS patterns
(Figure 3), which allowed us to determine their lattice
symmetries and unit cell parameters (Table 1). All variants
possess bcc (cubic I) and/or bct (tetragonal I) symmetry and
or deviate from the body diagonals. It must be noted that the
unit cell parameters for the tetragonal sublattice of 5−Zn−
ferritin determined by room-temperature SAXS experiments (a
= 152.67 Å, c = 160.27 Å) show a smaller tetragonal distortion
compared to the sc-XRD case (a = 149.88 Å, c = 162.23 Å).
This points to a reduced lateral displacement and a ca. 1.1-Å
longer interfacial separation (projected Zn−Zn distance
estimated to be ca. 10.5 Å), which could be readily
accommodated by the tilting of linker 5 with respect to the
interfacial plane and bond rotations about the methylene
spacers (Figure S7). At this interfacial separation, there would
be no direct protein−protein contacts, suggesting that the
crystallographically observed interprotein contacts are a
consequence of the cryogenic temperatures used in sc-XRD
experiments.
A closer examination of the SAXS data reveals some trends
related to the metal and linker components of ferritin MOFs.
First, Zn− and Co−ferritin−MOF lattices are nearly identical
to one another in almost all instances but distinct from Ni−
ferritin−MOFs in the case of linkers 1, 4, and 5. Whereas linker
1 yields exclusively a cubic lattice with Zn− and Co−ferritin, it
produces tetragonal symmetry with Ni−ferritin. While the
primary lattices are nearly identical for linker 4, only the Ni−
ferritin forms an additional cubic lattice. The opposite trend is
observed with linker 5, where Ni−ferritin forms a single cubic
lattice while Zn− and Co−ferritin form a similar cubic (or near
cubic) lattice as well as a tetragonal lattice.
possess unit cell dimensions that fall within 155
10 nm.
These observations establish that each ferritin−MOF is formed
through the desired linker-mediated association of the metal
centers at the C3 vertices of ferritin nodes. Indeed, when the
metal-bound ferritins are treated with Ca2+ instead of the
linkers, a clear SAXS pattern for an fcc lattice is produced
(Figure S5), demonstrating the chemical selectivity of linker-
mediated ferritin self-assembly. To assess the accuracy of the
SAXS measurements, we compared the unit cell parameters of
1−Zn−ferritin determined by SAXS (bcc, a = 155.97 Å) to
those derived from sc-XRD (a = 155.81 Å), which revealed a
close correspondence between the two types of measurements.
An initial inspection of the SAXS data indicates that bulk
samples of most ferritin−MOF variants (10) consist of single,
body-centered lattices. However, certain metal/ligand combi-
nations (5) yield at least two distinct lattices (shown in
different colors in Figure 3 and separately listed in Table 1),
consistent with the expectation that ferritin−MOFs may be
inherently flexible and that subtle changes in the metal-linker
coordination and linker geometry can give rise to measurable
changes in the lattice parameters. The most prominent example
is provided by the simultaneous observation of three distinct
bct lattices of 1−Ni−ferritin whose unit cell parameters vary by
These observations can be ascribed to the similarities
between Zn(II) and Co(II) in terms of their coordination
preferences for a tetrahedral geometry (as well as similar ionic
radii and reactivity patterns),57 whereas Ni(II) typically tends
toward higher coordination numbers and octahedral geometry.
Thus, at least in some ferritin−MOF variants, the specific
combination of metal coordination geometry and the linker
structure dictate distinct outcomes in terms of the lattice
architecture. In the case of linker 1, we can propose a structural
model in which Zn(II) or Co(II) centers would accommodate a
ca. 7% (mean dimension: 154.3
10.0 Å). This variation
points to different extents of tetragonal distortion, i.e., the
difference between the a and c dimensions which is related to
the displacement of the linker-bound metal ions from the body
diagonals (see Figure S6 for structural models). In the bulk
samples of 4−Ni− and 5−Zn−ferritin crystals, both tetragonal
and cubic lattices are observed, meaning that the flexibility in
the linker geometry can allow the ferritin C3 axes to align with
E
DOI: 10.1021/jacs.7b01202
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
ORCID
head-on (apical) hydroxamate coordination to complete a
pseudotetrahedral geometry. This geometry, combined with the
linearity of linker 1, would align the C3 axes of the connected
ferritin nodes to yield cubic lattice symmetry, as observed in the
crystal structure of 1−Zn−ferritin. In contrast, for Ni−ferritin,
whose octahedral Ni(II) centers would prefer a “side-on”
attachment by the hydroxamate moiety, the linear linker 1
would offset the ferritin C3 axes to yield a tetragonal lattice
(Figure S6). Conversely, the “bent” linker 4 could yield a cubic
or tetragonal lattice symmetry through side-on Ni(II)
coordination, while the apical Zn(II) and Co(II) coordination
leads to a tetragonal lattice. Such models are more difficult to
propose for linker 5, whose inherently higher flexibility could
be expected to dampen the influence of metal−hydroxamate
coordination geometry on the lattice structure. Regardless,
further sc-XRD characterization will be necessary to probe the
validity of these structural models.
Interestingly, linkers 2 and 3 always yield tetragonal lattices
regardless of the metal component they are paired with. Since
these two ligands display the same relative hydroxamate
orientations (“parallel but offset”), they also provide a valid
point of comparison in terms of how the linker dimensions
affect lattice metrics. Indeed, we observe that the longer linker 3
consistently produces a larger tetragonal distortion (|a − c| ≥
11.8 Å) than 2 (|a − c| ≤ 5.0 Å) for all metallo−ferritin nodes.
These observations indicate that synthetic linkers can also
modulate the unit cell dimensions of ferritin−MOFs.
F. Akif Tezcan: 0000-0002-4733-6500
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Drs. Milan Gembicky, Curtis Moore, and Steven
Weigand for assistance with sc-XRD and SAXS measurements.
This work was primarily funded by the U.S. DOE (BES,
Division of Materials Sciences, Biomolecular Materials
Program, DE-FG02-10ER46677 to F.A.T.). Additional support
was provided by NSF (ligand synthesis, DMR-1602537 to
F.A.T.) Crystallographic data were collected at SSRL,
supported by the U.S. DOE (Office of Science, BES and
BER), as well as by the NIH. SAXS experiments were
performed at APS, a DOE Office of Science User Facility
operated by Argonne National Laboratory under Contract No.
DE-AC02-06CH11357.
REFERENCES
■
(1) Mann, S. Angew. Chem., Int. Ed. 2008, 47 (29), 5306−5320.
(2) Yang, L.; Liu, A.; Cao, S.; Putri, R. M.; Jonkheijm, P.;
Cornelissen, J. J. Chem. - Eur. J. 2016, 22 (44), 15570−15582.
(3) Ueno, T. Chem. - Eur. J. 2013, 19 (28), 9096−9102.
(4) Yeates, T. O.; Liu, Y.; Laniado, J. Curr. Opin. Struct. Biol. 2016,
39, 134−143.
(5) Kobayashi, N.; Arai, R. Curr. Opin. Biotechnol. 2017, 46, 57−65.
(6) Salgado, E. N.; Radford, R. J.; Tezcan, F. A. Acc. Chem. Res. 2010,
43 (5), 661−672.
CONCLUSIONS
■
(7) Radford, R. J.; Brodin, J. D.; Salgado, E. N.; Tezcan, F. A. Coord.
Chem. Rev. 2011, 255 (7−8), 790−803.
Here we have reported the construction and characterization of
a large library of 3D, crystalline protein−MOFs through a
combination of three different metallo−ferritin nodes and five
synthetic linkers bearing hydroxamate head groups. Our results
establish that the metal ion and the synthetic linker
components can be varied in a modular fashion to influence
the structural parameters (i.e., lattice symmetries and unit cell
dimensions) of protein−MOFs, akin to conventional MOFs.
Despite the remarkable size discrepancy between the ferritin
nodes and the organic linkers, the self-assembly of the ferritin−
MOFs is highly robust, emphasizing the utility of metal
coordination interactions in controlling protein self-assembly.
Excitingly, by virtue of the flexibility of the linkers and/or the
fluxionality of metal-linker bonds, several ferritin−MOFs have
been observed to adopt multiple lattice conformations, which
suggests that they may display dynamic behavior. Some of our
immediate goals include the utilization of linkers whose lengths
approximate the dimensions of the protein nodes and
investigating the emergent functional/physical properties of
protein−MOFs (e.g., their potential dynamic nature) arising
from their unique, tripartite composition.
(8) Lai, Y.-T.; Hura, G. L.; Dyer, K. N.; Tang, H. Y.; Tainer, J. A.;
Yeates, T. O. Sci. Adv. 2016, 2 (12), e1501855.
(9) King, N. P.; Bale, J. B.; Sheffler, W.; McNamara, D. E.; Gonen, S.;
Gonen, T.; Yeates, T. O.; Baker, D. Nature 2014, 510 (7503), 103−
108.
(10) Sciore, A.; Su, M.; Koldewey, P.; Eschweiler, J. D.; Diffley, K. A.;
Linhares, B. M.; Ruotolo, B. T.; Bardwell, J. C.; Skiniotis, G.; Marsh, E.
N. G. Proc. Natl. Acad. Sci. U. S. A. 2016, 113 (31), 8681−8686.
(11) Ni, T. W.; Tezcan, F. A. Angew. Chem., Int. Ed. 2010, 49 (39),
7014−7018.
(12) Brodin, J. D.; Ambroggio, X. I.; Tang, C. Y.; Parent, K. N.;
Baker, T. S.; Tezcan, F. A. Nat. Chem. 2012, 4 (5), 375−382.
(13) Brodin, J. D.; Smith, S. J.; Carr, J. R.; Tezcan, F. A. J. Am. Chem.
Soc. 2015, 137 (33), 10468−10471.
(14) Yang, G.; Zhang, X.; Kochovski, Z.; Zhang, Y.; Dai, B.; Sakai, F.;
Jiang, L.; Lu, Y.; Ballauff, M.; Li, X. J. Am. Chem. Soc. 2016, 138 (6),
1932−1937.
(15) Ballister, E. R.; Lai, A. H.; Zuckermann, R. N.; Cheng, Y.;
Mougous, J. D. Proc. Natl. Acad. Sci. U. S. A. 2008, 105 (10), 3733−
3738.
(16) Sinclair, J. C.; Davies, K. M.; Venien-Bryan, C.; Noble, M. E. M.
Nat. Nanotechnol. 2011, 6 (9), 558−562.
(17) Gonen, S.; DiMaio, F.; Gonen, T.; Baker, D. Science 2015, 348
(6241), 1365−1368.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.7b01202.
■
(18) Suzuki, Y.; Cardone, G.; Restrepo, D.; Zavattieri, P. D.; Baker,
T. S.; Tezcan, F. A. Nature 2016, 533 (7603), 369−373.
(19) Lanci, C. J.; MacDermaid, C. M.; Kang, S.-g.; Acharya, R.;
North, B.; Yang, X.; Qiu, X. J.; DeGrado, W. F.; Saven, J. G. Proc. Natl.
Acad. Sci. U. S. A. 2012, 109 (19), 7304−7309.
(20) Sakai, F.; Yang, G.; Weiss, M. S.; Liu, Y.; Chen, G.; Jiang, M.
Nat. Commun. 2014, 5, 4634.
(21) Kunzle, M.; Eckert, T.; Beck, T. J. Am. Chem. Soc. 2016, 138
S
Additional experimental details (Materials and Methods),
tables (Supporting Tables 1 and 2), and figures
(Supporting Figures 1−18) (PDF)
̈
(39), 12731−12734.
AUTHOR INFORMATION
Corresponding Author
*tezcan@ucsd.edu
■
(22) Kostiainen, M. A.; Hiekkataipale, P.; Laiho, A.; Lemieux, V.;
Seitsonen, J.; Ruokolainen, J.; Ceci, P. Nat. Nanotechnol. 2012, 8 (1),
52−56.
F
DOI: 10.1021/jacs.7b01202
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(23) Radford, R. J.; Lawrenz, M.; Nguyen, P. C.; McCammon, J. A.;
Tezcan, F. A. Chem. Commun. 2011, 47 (1), 313−315.
(24) Joh, N. H.; Wang, T.; Bhate, M. P.; Acharya, R.; Wu, Y.; Grabe,
M.; Hong, M.; Grigoryan, G.; DeGrado, W. F. Science 2014, 346
(6216), 1520−1524.
(25) Votteler, J.; Ogohara, C.; Yi, S.; Hsia, Y.; Nattermann, U.;
Belnap, D. M.; King, N. P.; Sundquist, W. I. Nature 2016, 540 (7632),
292−295.
(26) Song, W. J.; Tezcan, F. A. Science 2014, 346 (6216), 1525−
1528.
(27) Brodin, J. D.; Carr, J. R.; Sontz, P. A.; Tezcan, F. A. Proc. Natl.
Acad. Sci. U. S. A. 2014, 111 (8), 2897−2902.
(28) Chakrabarty, R.; Mukherjee, P. S.; Stang, P. J. Chem. Rev. 2011,
111 (11), 6810−6918.
(54) Breslow, R.; Jursic, B.; Yan, Z. F.; Friedman, E.; Leng, L.; Ngo,
L. Proc. Natl. Acad. Sci. U. S. A. 1991, 88, 5542−5546.
(55) Yamaye, M.; Nago, Y.; Sasaki, M.; Tsuru, T.; Mukae, K.;
Yoshinaga, T.; Murayama, R.; Tahara, C. Polym. Degrad. Stab. 2006,
91, 2014−2021.
(56) Saxena, N.; Juneja, H. D.; Munshi, K. N. J. Indian Chem. Soc.
1993, 70 (11−12), 943−949.
(57) Bertini, I.; Luchinat, C. The reaction pathways of zinc enzymes
and related biological catalysts. In Bioinorganic Chemistry; Bertini, I.,
Gray, H. B., Lippard, S. J., Valentine, J. S., Eds.; University Science
Books: Mill Valley, 1994; pp 37−106.
(29) Corbett, P. T.; Leclaire, J.; Vial, L.; West, K. R.; Wietor, J.-L.;
Sanders, J. K. M.; Otto, S. Chem. Rev. 2006, 106 (9), 3652−3711.
(30) Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M.
Science 2013, 341 (6149), 1230444.
(31) Zhou, H.-C.; Kitagawa, S. Chem. Soc. Rev. 2014, 43 (16), 5415−
5418.
(32) Eddaoudi, M.; Kim, J.; Rosi, N.; Vodak, D.; Wachter, J.;
O’Keeffe, M.; Yaghi, O. M. Science 2002, 295 (5554), 469−472.
(33) Howarth, A. J.; Peters, A. W.; Vermeulen, N. A.; Wang, T. C.;
Hupp, J. T.; Farha, O. K. Chem. Mater. 2017, 29 (1), 26−39.
(34) Chui, S. S.-Y.; Lo, S. M.-F.; Charmant, J. P. H.; Orpen, A. G.;
Williams, I. D. Science 1999, 283 (5405), 1148−1150.
(35) Li, J. R.; Sculley, J.; Zhou, H. C. Chem. Rev. 2012, 112 (2), 869−
932.
(36) Horike, S.; Shimomura, S.; Kitagawa, S. Nat. Chem. 2009, 1 (9),
695−704.
(37) Yoon, M.; Srirambalaji, R.; Kim, K. Chem. Rev. 2012, 112 (2),
1196−1231.
(38) Sumida, K.; Rogow, D. L.; Mason, J. A.; McDonald, T. M.;
Bloch, E. D.; Herm, Z. R.; Bae, T. H.; Long, J. R. Chem. Rev. 2012, 112
(2), 724−781.
(39) Horcajada, P.; Gref, R.; Baati, T.; Allan, P. K.; Maurin, G.;
Couvreur, P.; Ferey, G.; Morris, R. E.; Serre, C. Chem. Rev. 2012, 112
(2), 1232−1268.
(40) Wang, Z.; Cohen, S. M. Chem. Soc. Rev. 2009, 38 (5), 1315−
1329.
(41) Sontz, P. A.; Bailey, J. B.; Ahn, S.; Tezcan, F. A. J. Am. Chem. Soc.
2015, 137 (36), 11598−11601.
(42) Li, H.; Eddaoudi, M.; O’Keeffe, M.; Yaghi, O. M. Nature 1999,
402 (6759), 276−279.
(43) Barthelet, K.; Marrot, J.; Riou, D.; Ferey, G. Angew. Chem., Int.
Ed. 2002, 41 (2), 281−284.
(44) Abe, S.; Ueno, T. RSC Adv. 2015, 5 (27), 21366−21375.
(45) MaHam, A.; Tang, Z. W.; Wu, H.; Wang, J.; Lin, Y. H. Small
2009, 5 (15), 1706−1721.
(46) Meldrum, F. C.; Heywood, B. R.; Mann, S. Science 1992, 257
(5069), 522−524.
(47) Abe, S.; Niemeyer, J.; Abe, M.; Takezawa, Y.; Ueno, T.; Hikage,
T.; Erker, G.; Watanabe, Y. J. Am. Chem. Soc. 2008, 130 (32), 10512−
10514.
(48) Aime, S.; Frullano, L.; Geninatti Crich, S. Angew. Chem., Int. Ed.
2002, 41 (6), 1017−1019.
(49) Butts, C. A.; Swift, J.; Kang, S.-g.; Di Costanzo, L.; Christianson,
D. W.; Saven, J. G.; Dmochowski, I. J. Biochemistry 2008, 47 (48),
12729−12739.
(50) Swift, J.; Wehbi, W. A.; Kelly, B. D.; Stowell, X. F.; Saven, J. G.;
Dmochowski, I. J. J. Am. Chem. Soc. 2006, 128 (20), 6611−6619.
(51) Bernacchioni, C.; Ghini, V.; Theil, E. C.; Turano, P. RSC Adv.
2016, 6, 21219−21227.
(52) Lawson, D. M.; Artymiuk, P. J.; Yewdall, S. J.; Smith, J. M.;
Livingstone, J. C.; Treffry, A.; Luzzago, A.; Levi, S.; Arosio, P.;
Cesareni, G.; Thomas, C. D.; Shaw, W. V.; Harrison, P. M. Nature
1991, 349 (6309), 541−4.
(53) Bai, Y.; Guo, D.; Duan, C. Y.; Dang, D. B.; Pang, K. L.; Meng,
Q. J. Chem. Commun. 2004, 2, 186−187.
G
DOI: 10.1021/jacs.7b01202
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX