Peptide-modified dendrimers as templates for the production of highly reactive catalytic nanomaterials

Article abstract of DOI:10.1021/cm5007444

Peptide-driven nanomaterials synthesis and assembly has become a significant research thrust due to the capability to generate a range of multifunctional materials with high spatial precision and tunable properties. Despite the extensive amount of available literature, the majority of studies report the use of free peptides to drive synthesis and assembly. Such strategies are not an entirely accurate representation of nature, as many materials binding peptides found in biological systems are sterically constrained to a larger biological motif. Herein we report the synthesis of catalytic Pd nanomaterials using constrained peptides covalently attached to the surface of small, water-soluble dendrimers. Using the R5 peptide conjugated to polyamidoamine dendrimer as a bioconjugate, Pd nanomaterials were generated that displayed altered morphologies compared to nanomaterials templated with free R5. It was discovered that the peptide surface density on the dendrimer affected the resulting nanoscale morphology. Furthermore, the catalytic activities of Pd materials templated with R5/dendrimer are higher as compared to the R5-templated Pd materials for the hydrogenation of allyl alcohol, with an average increase in turnover frequency of ～1500 mol product (mol Pd × h)^-1. Small angle X-ray scattering analysis and dynamic light scattering indicate that Pd derived from R5/dendrimer templates remained less aggregated in solution and displayed more available reactive Pd surface area. Such morphological changes in solution are attributed to the constrained peptide binding motifs, which altered the Pd morphology and subsequent properties. Moreover, the results of this study suggest that constrained materials binding peptide systems can be employed as a means to alter morphology and improve resulting properties.

Full text of DOI:10.1021/cm5007444

Article
pubs.acs.org/cm
Peptide-Modified Dendrimers as Templates for the Production of
Highly Reactive Catalytic Nanomaterials
†
,‡
‡,∥
†
§
,†
Nicholas M. Bedford, , Rohit Bhandari, Joseph M. Slocik, Soenke Seifert, Rajesh R. Naik,*
and Marc R. Knecht*‡
†Materials and Manufacturing Directorate, Air Force Research Laboratory, Wright-Patterson Air Force Base, Ohio 45433, United
States
‡
Department of Chemistry, University of Miami, 1301 Memorial Drive, Coral Gables, Florida 33146, United States
§
X-ray Science Division, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439, United States
*
S Supporting Information
ABSTRACT: Peptide-driven nanomaterials synthesis and assembly has become a significant research thrust due to the capability
to generate a range of multifunctional materials with high spatial precision and tunable properties. Despite the extensive amount
of available literature, the majority of studies report the use of free peptides to drive synthesis and assembly. Such strategies are
not an entirely accurate representation of nature, as many materials binding peptides found in biological systems are sterically
constrained to a larger biological motif. Herein we report the synthesis of catalytic Pd nanomaterials using constrained peptides
covalently attached to the surface of small, water-soluble dendrimers. Using the R5 peptide conjugated to polyamidoamine
dendrimer as a bioconjugate, Pd nanomaterials were generated that displayed altered morphologies compared to nanomaterials
templated with free R5. It was discovered that the peptide surface density on the dendrimer affected the resulting nanoscale
morphology. Furthermore, the catalytic activities of Pd materials templated with R5/dendrimer are higher as compared to the
R5-templated Pd materials for the hydrogenation of allyl alcohol, with an average increase in turnover frequency of ∼1500 mol
−1
product (mol Pd × h) . Small angle X-ray scattering analysis and dynamic light scattering indicate that Pd derived from R5/
dendrimer templates remained less aggregated in solution and displayed more available reactive Pd surface area. Such
morphological changes in solution are attributed to the constrained peptide binding motifs, which altered the Pd morphology
and subsequent properties. Moreover, the results of this study suggest that constrained materials binding peptide systems can be
employed as a means to alter morphology and improve resulting properties.
INTRODUCTION
especially important considering that the noncovalent inter-
actions of peptide binding could potentially be tailored for
additional capabilities that are not possible using covalently
bound ligands. Furthermore, the ability to form multifunctional
structures via fusion peptides can also be achieved by
■
Peptide-driven fabrication methods represent an emerging
direction for the ambient synthesis and assembly of nano-
1
−3
structured materials.
Inspired by biomineralization, peptide
sequences discovered via biocombinatorial selection techniques
or isolated directly from an organism have been found to
fabricate a wide variety of nanoscale materials by binding to the
incorporating two material binding domains into a single
23−25
sequence.
Such ligand programmability and material
4
−8
inorganic surface. Using these approaches, metal,
metal
structural control ultimately allows for the fine-tuning of
emergent properties. Taken together, these biomimetic
materials possess important and tunable properties that could
9
−13
14−17
18−21
oxide,
semiconductor,
and carbon
nanomaterials
have been generated/assembled under energy-friendly and
aqueous reaction conditions. These ambient conditions,
coupled with inherent biological specificity, make peptide-
driven nanomaterial production an attractive alternative to
traditional routes that require energy-intensive reactions with
nonspecific organic ligands to stabilize the structures. This is
22
Received: March 1, 2014
Revised: June 16, 2014
Published: July 8, 2014
©
2014 American Chemical Society
4082
dx.doi.org/10.1021/cm5007444 | Chem. Mater. 2014, 26, 4082−4091

Chemistry of Materials
Article
1
7,26
27,28
be exploited for optical,
31,32
energy harvesting/storage,
applications.
were produced, which are not observed using the free peptide
template. Such morphological diversity is attributed to the
confined orientation of the R5 bound to the surface of the
dendrimer, where the peptide surface density was noted to
impact the overall metallic structure. Once produced, the
biotemplated materials were employed as catalysts for the
hydrogenation of allyl alcohol, from which the calculated
turnover frequency (TOF) values were greatly enhanced
compared to the materials generated using the dendrimer-free
R5 scaffold. The TOF values were exceedingly high, suggesting
that the dendritic template is enhanced for both materials
production and catalytic application. To study this effect, the
peptide−dendrimer templated Pd nanomaterials were exam-
ined using small-angle X-ray scattering (SAXS) and dynamic
light scattering (DLS), which demonstrated a more porous
macrostructure of the bio/dendrimer template in solution with
a higher degree of catalytic surface area for the Pd component
as compared to the structures encapsulated by the native R5
template. As a result, these studies suggest that peptide-
modified dendrimers represent innovative biomimetic tem-
plates where the molecular rigidity and global framework
directly affect the catalytic functionality.
2
9,30
sensing,
and catalytic
Materials binding sequences are commonly used as free
peptides in solution to generate nanostructures by binding to
the surface of the growing inorganic component.
can produce the desired materials, such flexibility is quite
different than that of the peptide morphology found in nature
where the N- and/or C-termini of the sequence are typically
5
−7
While this
12,33
attached to a larger protein construct.
This subtlety could
possess significant implications over nanomaterial synthesis and
assembly. For instance, sequences isolated from biomineralizing
species are commonly derived from membrane bound
while peptides discovered via biopanning
techniques, such as phage and cell-surface display, are anchored
1
2,33
proteins,
2
onto the surface of the panning moiety. Previous work by the
Belcher group has demonstrated that constraining such material
binding sequences along virus backbones can be employed to
template inorganic structures, resulting in the production of
1
6,24,27
anisotropic nanowires.
These structures have important
catalytic and energy harvesting capabilities that exploit the one-
dimensional nature of the virus. Furthermore, Rosi and
colleagues have used hybrid DNA-modified peptides that self-
34
assemble to generate helical nanoribbons of Au nanoparticles.
RESULTS AND DISCUSSION
This process provides enhanced control over both the global
and local assembly construct via incorporation of the peptide
into a supramolecular framework, where the biomolecule
retains its ability to recognize and bind Au nanoparticles.
In contrast, certain materials binding peptides can assemble
into larger macromolecular aggregates in solution depending on
the location of the amino acids in the primary sequence. One
example is the R5 peptide (SSKKSGSYSGSKGSKRRIL),
derived from the biosilica producing diatom Cylindrothica
■
Cysteine modified R5 peptides were covalently attached to the
surface of hydroxyl terminated G4 PAMAM dendrimers using
N-(p-maleimidophenyl) isocyanate (PMPI), a common hydrox-
yl-to-thiol coupling agent (Scheme 1). In this reaction, the
Scheme 1. Synthesis of the R5/Dendrimer Template and
Subsequent Pd Nanomaterial Generation
3
3
fusiformis. The presence of the RRIL motif at the C-terminal
region of the largely hydrophilic sequence likely drives peptide
aggregate formation, wherein the RRIL component is buried
within the core of the framework to display the highly polar
functionalities to solution. This architecture allows the peptide
to be exploited as a template for other nonsilica nanomaterials
3
5−40
as well.
One such example is the formation of catalytic Pd
nanomaterials, where previously a dependence upon the Pd:R5
ratio over the final morphology of the encapsulated metallic
38−41
component was demonstrated.
Using this sequence, the
morphologies of Pd nanomaterials ranged from spherical
nanoparticles at low ratios to linear nanoribbons and branching
3
8−40
nanoparticle networks (NPNs) at higher ratios.
These
structures have been shown to be highly reactive in catalytic
C−C coupling, 4-nitrophenol reduction, and olefin hydro-
38−40
genation reactions.
In this contribution, we report the use of R5-dendrimer
conjugates as templates for the fabrication of Pd nanomaterials,
where the biotemplate dramatically influences the catalytic
activity of the final structures. Dendrimers are monodisperse,
hydroxyl groups on the surface of the dendrimer react with the
isocyanate groups in PMPI to make the dendrimer suitable for
subsequent thiol attachment. Cysteine labeled R5 then couples
to the maleimide portion of PMPI to form a thioether linkage,
successfully conjugating the peptide to the surface of the
dendrimer (see Supporting Information, Scheme S1, for more
details). As R5 lacks a native thiol functionality, cysteine was
inserted onto the peptide sequence at the N-terminus, thus
displaying the RRIL motif to solution to promote self-assembly.
Once coupled, the peptide surface density could affect both
metal templating and the final nanomaterial morphology; thus,
peptide coverage of the dendrimer surface was studied at
R5:dendrimer ratios of 4, 8, and 16. The use of G4-PAMAM
42
spherical polymers radially grown from a central core. This
architecture yields a macromolecular structure of a semihollow
interior with tunable surface functionalities for potential
4
3−45
applications in nanomaterial production
cine.
and biomedi-
4
6−48
For this study, hydroxyl-terminated fourth generation
polyamidoamine (G4 PAMAM−OH) dendrimers were used as
a water-soluble framework to which the R5 peptide was
covalently attached, creating a unique biomimetic dendritic
scaffold for the templation of inorganic nanomaterials. Using
the R5/dendrimer templates, highly reactive Pd nanostructures
of multiple morphologies ranging from nanoparticles to NPNs
4
083
dx.doi.org/10.1021/cm5007444 | Chem. Mater. 2014, 26, 4082−4091

Chemistry of Materials
versus smaller dendrimer generations) allows these surface
Article
(
can only arise from the biomolecules bound to the dendrimer
periphery. As shown in the black plot of Figure 1b, when the
CR5W peptide was conjugated to the dendrimer surface, an
increase in the fluorescence intensity was observed proportional
to the calculated peptide surface density. This result indicates
that the peptide−dendrimer conjugates were achieved at the
approximate surface densities anticipated based upon the
reaction stoichiometry. Note that all of the conjugates have
significantly higher fluorescence intensities than the unmodified
dendrimer. Moreover, simply mixing the dendrimer and CR5W
together without PMPI results in negligible fluorescence signal
after dialysis (magenta plot). While a minor fluorescence signal
is noted suggesting nonspecific interactions between the
peptide and dendrimer, the intensity is much weaker than the
values observed with the chemically conjugated biotemplates.
This result, along with IR analysis and TEM studies (discussed
below), suggested that PMPI-based coupling strategies
effectively conjugated the peptide to the dendrimer to form
the biomacromolecular structures.
densities to be explored while minimizing steric crowding of the
peptide at the dendrimer surface. To differentiate between the
various materials prepared, the metallic structures are defined as
XR5(PdY), where X represents the number of peptides present
on the dendrimer surface (4, 8, or 16) and Y denotes the
Pd:peptide ratio employed to prepare the metallic nanostruc-
tures. Note that Pd:peptide ratios were used for the fabrication
of all of the inorganic materials and not Pd:dendrimer ratios.
Once the biomimetic template was generated, IR spectros-
copy was performed on lyophilized material to confirm peptide
Circular dichroism (CD) spectroscopy was used to probe the
conformational changes of the R5 peptide covalently attached
to the PAMAM surface. As shown in Figure 1c, the native CR5
peptide (red line) demonstrates a large negative peak at 200
nm, indicative of a random coiled structure. This structure is
52
consistent with previous CD studies of the parent R5 and
53
other short linear peptides. Upon conjugation to the rigid
dendrimer, the peptides undergo structural changes as observed
by drastic changes in the CD spectra for the 4R5, 8R5, and
1
6R5 species. A very broad peak centered ∼215 nm is found for
the 4R5 sample (green plot), suggesting that the peptide is
becoming more structured. This feature was also found in the
8R5 and 16R5 samples (blue and cyan plots, respectively),
becoming broader and/or shifting to higher wavelengths. These
shifts signify a structural transition from random coil to more
helical structures, suggesting that a more compact and
constrained orientation of the R5 sequence upon conjugation
to the dendrimer periphery. For reference, a spectrum of the
PAMAM dendrimer is also shown in Figure 1c, indicating that
the features found in the peptide−dendrimer conjugates arise
from the peptide only and not from the polymeric core. Such
results were anticipated based upon similar peptide structural
effects when bound to larger constructs, such as those observed
for the variable peptide region when integrated into the phage
Figure 1. Characterization of the R5/dendrimer conjugate formation,
peptide surface density, peptide confirmation, and aggregate size via
(
a) IR, (b) fluorescence, and (c) CD spectroscopy. Part (d) presents
the DLS size analysis of the generated biotemplates.
conjugation to the dendrimer periphery (Figure 1a). The bare
dendrimer spectrum (black plot) exhibits peaks at 1624 and
542 cm− corresponding to the amide stretches found
1
1
throughout the repeating, branched polymeric structure. For
the free CR5 peptide (red plot), amide peaks are also present at
−
1
1
618 and 1516 cm arising from the amides of the peptide
−1
backbone, along with a small peak at 1780 cm corresponding
49
to the thiol group of cysteine. Large absorptions at 1178 and
134 cm− are also present for CR5, corresponding to amide III
1
54
1
viral structure.
absorption from intra- and intermolecular interaction between
Deconvolution of the CD spectra assisted in further
characterization of the secondary structure of the peptides
(Supporting Information, Table S1). The helical content of the
unconstrained linear CR5 peptide was minimal, as expected.
Upon conjunction of CR5 to the dendrimer, the helical content
increases linearly with peptide surface density. This increase in
helical nature is also characteristic of other peptide-containing
50
amino acid side chains. Upon conjugation of four equivalents
of the CR5 to the dendrimer surface (green plot), the thiol
band disappears, concomitant with the production of two new
−1
peaks at 1042 and 976 cm . These new stretches appear from
the generation of the thioether group that conjugates the
51
biomolecules to the polymeric surface. Moreover, the amide
III absorptions are also masked, suggesting that the amino acid
interactions are substantially changed, likely due to steric
hindrance of the peptide, further indicating that the peptides
are coupled to the dendrimer surface.
55,56
dendrimers.
Similarly, the extent of unordered content
decreases with increasing peptide density, with the uncon-
strained CR5 peptide exhibiting the highest percentage of this
confirmation. Together, this analysis demonstrates that the R5
peptide is becoming more conformationally hindered at higher
surface conjugation ratios due to close proximity of the peptide
chains attached to the surface of dendritic core.
DLS was used to probe the aggregate size of the free
dendrimer and peptides, as well as the dendrimer/peptide
constructs in solution. As shown in Figure 1d, the G4 PAMAM
dendrimer displayed an aggregate size of 5.6 ± 0.2 nm,
consistent with previously measured dimensions for the
To determine the peptide surface density, a fluorescence
assay was conducted wherein tryptophan was used as a
fluorescent reporter incorporated at the opposite termini
from the cysteine residue (i.e., CR5W). Such a design should
allow for the coupling strategy to proceed unimpeded, while
allowing for monitoring of the degree of conjugation. After
peptide coupling to the dendrimer, the system was dialyzed to
remove excess free peptide, and thus any fluorescence signal
4
084
dx.doi.org/10.1021/cm5007444 | Chem. Mater. 2014, 26, 4082−4091

Chemistry of Materials
Article
5
7
monodisperse structure. When studying the CR5 peptides, an
aggregate size of 520 ± 60 nm was obtained. For reference,
under the same conditions, the native R5 that lacks the cysteine
residue of the present sequence displayed an aggregate size of
Table 1. Summary of TEM and Catalytic Observations for
the Pd Nanomaterials Templated with the R5−Dendrimer
Conjugates
3
7
peptide
9
60 ± 43 nm, suggesting that the positioning of the thiol
coverage-
TOF
a
−1
group modulated the self-assembly process. Peptide aggregation
is anticipated to originate from the RRIL motif, burying this
component into a hydrophobic core to expose the hydrophilic
(Pd amount)
morphology
particles
size (nm) mol product (mol Pd × h)
1
6R5(Pd1O)
6R5(Pd30)
2.7 ± 0.6
2.7 ± 0.6
4800 ± 400
4600 ± 200
1
particles
(57%)
5
8
residues. In the CR5, a modestly hydrophobic cysteine is
located on the N-terminus, opposite of the RRIL component,
thus slightly perturbing the assembly effects in the native
peptide, resulting in the observed smaller aggregate size. Once
the peptides were conjugated to the dendrimer, an interesting
assembly pattern was observed (Figure 1d). For the R5-based
conjugates, the sizes varied over a narrow range of 480 ± 70 nm
for the 16R5 species until a value of 540 ± 80 nm for the 4R5
conjugate. Such sizes were nearly identical to that of the free
CR5 in solution, suggesting that the dendrimer did not
significantly inhibit the assembly process. It is interesting to
note that despite the potential for extensive cross-linking in the
higher surface density conjugates, the size trend, on average,
decreases with increasing peptide density. Such results may be
due to the effect of the higher peptide surface density leading to
sterically restricted biomolecules, thus preventing them from
achieving appropriate conformations to drive structural
assembly. Note that no free peptide was present in solution
due to extensive dialysis, and thus the observed sizes were
directly attributed to the peptide/dendrimer conjugates only.
Interestingly the sizes obtained for the conjugates are similar to
ribbons
4.0 ± 0.7
(
43%)
1
6R5(Pd60)
R5(Pd30)
NPNs
3.9 ± 1.0
2.6 ± 0.9
4300 ± 200
4800 ± 200
8
particles
(
83%)
ribbons
(17%)
3.4 ± 0.8
2.4 ± 0.5
3.1 ± 0.9
3.0 ± 0.6
3.4 ± 0.6
8R5(Pd60)
particles
4800 ± 100
4500 ± 200
(
79%)
ribbons
21%)
(
8
R5(Pd90)
particles
23%)
(
ribbons
(19%)
NPNs (58%) 4.7 ± 0.9
4R5(Pd60)
particles
88%)
2.5 ± 0.5
3.4 ± 0.8
2.6 ± 0.5
3.3 ± 1.1
2.9 ± 0.5
4100 ± 400
4210 ± 90
4400 ± 120
(
ribbons
12%)
(
4
4
R5(Pd90)
particles
(80%)
ribbons
41
(
20%)
those reported for R5 peptide truncates, suggesting that the
assembly process can occur with a constricted RRIL motif but
that this portion facilitates the organization event.
R5(Pd120)
particles
57%)
(
NPNs (43%) 4.5 ± 0.8
Once the peptide/dendrimer scaffolds were confirmed, they
were employed for the production of zerovalent Pd nanoma-
a
Morphology populations estimated from at least five different TEM
2
+
images.
terials. In this regard, Pd was incubated with each of the
peptide/dendrimer complexes, followed by reduction via
NaBH . For these materials, the amount of Pd added to each
60, as presented in Figure 2a−c. For the 16R5(Pd10) materials,
spherical particles were exclusively generated, with an average
diameter of 2.7 ± 0.6 nm. Upon increasing the Pd:R5 ratio to
30 for the 16R5(Pd30) sample, a mixture of spherical particles
(57%) and nanoribbons (43%) were observed with an average
diameter of 2.7 ± 0.6 nm for the nanoparticles and an average
width of 4.0 ± 0.7 nm for the nanoribbons. Further increasing
the Pd concentration for the 16R5(Pd60) materials, metallic
NPNs were the only morphology observed with an average
width of 3.9 ± 1.0 nm. This varying morphology over a wide
range of Pd:R5 ratios likely arises from template-directing
effects, as discussed below.
In a similar fashion, Pd nanomaterials were fabricated using
the 8R5 templates (Figure 2d−f) at Pd:R5 ratios of 30, 60, and
90. Note that this is a similar template as to the one discussed
above; however, fewer peptides were displayed at the
dendrimer periphery. For this system at the two lower Pd:R5
ratios, particles and ribbons were typically observed. In general,
2.6 ± 0.9 nm nanoparticles were the main component of the
8R5(Pd30) sample at 83% of the material population; however,
17% of the sample was in the nanoribbon morphology (width =
3.4 ± 0.8 nm). For the 8R5(Pd60) sample, spherical
nanoparticles again were the predominate morphology with
interspersed nanoribbons (79% vs 21%, respectively) with
average dimensions of 2.4 ± 0.5 nm (nanoparticle diameter)
and 3.1 ± 0.9 nm (nanoribbon width). Interestingly, in the
8R5(Pd90) system, a mixture of all three morphologies was
4
system was determined based upon the number of R5 peptides
in the reaction and not the total number of conjugates. From
this, a series of Pd:R5 ratios were initially tested for all peptide/
dendrimer conjugates, where the upper limit was chosen based
on nanomaterial stability. Previous reports using the free R5
peptide have suggested that the ratio of Pd:R5 employed
altered the morphology of the templated Pd nanomaterials; at a
Pd:R5 ratio of 60 (termed Pd60), spherical nanomaterials were
obtained, while for materials prepared at ratios of 90 (Pd90)
and 120 (Pd120), anisotropic nanoribbons and NPN
morphologies were generated, respectively, due to increased
38−40
loading of Pd in the template.
To this end, as the loading
increases, the number of individual nanoparticles increases,
resulting in a controlled particle aggregate to form the linear
structures within the templates. Such effects are likely to be
attenuated by the bulky dendrimers conjugated into the
biotemplate, which was fully examined.
Using the series of conjugates prepared, the Pd nanomateri-
als were generated and analyzed using transmission electron
microscopy (TEM). The morphology of the resulting metallic
structures depended upon the Pd:peptide ratio, where spherical
nanoparticles, as well as linear nanoribbons and NPNs were
generated. Particle size and morphology information for all of
the stable structures produced using the different templates are
summarized in Table 1. Using the 16R5 templates, stable Pd
nanomaterials were synthesized at Pd:R5 ratios of 10, 30, and
4
085
dx.doi.org/10.1021/cm5007444 | Chem. Mater. 2014, 26, 4082−4091

Chemistry of Materials
Article
Figure 2. TEM images of Pd nanomaterials templated with (a) 16R5(Pd10), (b) 16R5(Pd30), (c) 16R5(Pd60), (d) 8R5(Pd30), (e) 8R5(Pd60), (f)
R5(Pd90), (g) 4R5(Pd60), (h) 4R5(Pd90), and (i) 4R5(Pd120). Arrows and outlines in the 16R5 Pd sample show examples of the morphologies
8
discussed in the text. HRTEM images (insets) indicate the presence of Pd lattice fringes for selected samples (8-R5(Pd30), 8-R5(Pd60), and 4-
R5(Pd60); inset scale bar = 5 nm).
noted, with NPNs becoming the dominant structure (58%) of
an average width of 4.7 ± 0.9 nm. Finally, Pd nanomaterials
templated with the 4R5 conjugates also displayed interesting
metal loading-dependent morphologies (Figure 2g−i). In this
regard, spherical nanoparticles (88%) and nanoribbons (12%)
were generated for the 4R5(Pd60) sample (2.5 ± 1.0 nm
diameter and 3.4 ± 0.8 nm width, respectively), while at higher
ratios, mixtures of nanoparticles, nanoribbons, and NPNs were
noted. For instance, spherical nanoparticles (80%, 2.6 ± 0.5
nm) and nanoribbons (20%, 3.3 ± 1.1 nm) were obtained from
the 4R5(Pd90) system, while NPNs (43%, 4.5 ± 0.8 nm) and
spherical nanoparticles (57%, 2.9 ± 0.5 nm) were found with
particles and nanoribbons, and nanoparticles and NPNs,
respectively. At Pd90, the R5 peptide produces only nanorib-
38−40
bons,
yet the 8R5 and 4R5 templates yielded multiple
morphologies. Similarly, NPNs are formed at a Pd:R5 ratio of
38−40
120 for the R5 only scaffold,
while most of the peptide/
dendrimer complexes are unable to form stable nanomaterials
at this ratio. Moreover, the clear transition between the
different morphologies at increasing Pd concentrations is not as
defined when using peptide/dendrimer conjugates. Such
instances were clearly observed in the majority of the R5/
PAMAM systems as the examined ratios displayed multiple
morphologies. This nanomorphology effect can likely be
attributed to the dendrimer within the bioscaffold, resulting
in significant structural and template modulations, as depicted
in Scheme 1. To this end, the bioframework is formed via
random assembly of the peptide-dendrimer conjugates. Upon
reduction of the Pd2+ ions sequestered within the structure,
spherical nanoparticles are initially formed. To generate the
nanoribbons and NPNs, linear particle aggregation must occur.
In the R5 only template, the peptide strands guide linear
nanoparticle aggregation; however, such a process is likely to be
4
R5(Pd120) template. High-resolution TEM (HRTEM)
images for selected samples (insets of Figure 2d,e,g) further
confirm the observed morphologies while simultaneously
displaying the lattice fringes associated with Pd.
Key morphological differences were observed as compared to
the Pd nanomaterials templated by the R5 only scaffold. For
example, Pd60 nanomaterials prepared with R5 templates
38−40
formed spherical particles exclusively,
while Pd60 materials
templated with 16R5, 8R5, and 4R5 yielded NPNs, nano-
4
086
dx.doi.org/10.1021/cm5007444 | Chem. Mater. 2014, 26, 4082−4091

Chemistry of Materials
Article
impeded by the dendrimers integrated into the present
templates, where the large organic macromolecules should
sterically prevent two Pd particles from interacting across the
polymeric component. As such, linear particle aggregation and
morphology formation is obstructed, resulting in multiple Pd
morphologies in a single sample. Such an event is in direct
contrast to the R5 only template where the dendrimers are not
present, resulting in a more morphologically homogeneous
sample set. Moreover, the lysine groups of the peptide on the
dendrimer periphery have a higher affinity for Pd than the
tertiary amines within the interior of the polymer. Such a bias
results in a template wherein a fraction of the total organic
material is unavailable to interact with the metal component,
effectively reducing the metal loading capacity of the template.
As a consequence, anisotropic morphologies tend to occur at
lower Pd:R5 ratios compared to the R5-only scaffold.
It is also worth mentioning that the formation of Pd
dendrimer-encapsulated nanoparticles (DENs) is unlikely using
these templates. Previous reports of Pd DENs demonstrated
the production of uniform particles that are much smaller in
size (1.2−1.5 nm) compared to materials templated by the R5−
5
9−61
dendrimer conjugates.
from the size histograms obtained from TEM analysis
Supporting Information, Figure S1). Moreover, the amount
of Pd per conjugate is substantially greater than what could be
Such size scales are largely absent
Figure 3. Catalytic analysis of the R5/dendrimer templated Pd
nanomaterials: (a) the reaction scheme and (b) TOF values for the
hydrogenation of allyl alcohol. The black dashed line represents the
average TOF value for Pd nanomaterials templated with free R5 at
(
4
5,62
loaded into the G4 PAMAM−OH dendrimer hosts,
would lead to Pd precipitation in the absence of the surface-
constrained peptides. Finally, the primary amines of the peptide
which
40
ratios of 60−120 incremented by 10.
2
+
also possess stronger interactions with the Pd metal ions
compared to the interior tertiary amines of the dendrimer, thus
inhibiting metal uptake into the polymer to prevent DEN
formation. Taken together, this suggests that attachment of the
biomolecules to the dendrimer controls the metallic morphol-
ogy based upon template effects, which are controlled by the
number and structure of surface bound peptides.
−
1
product (mol Pd × h) across all Pd:R5 ratios, as previously
40
reported. This average is represented by the dashed line in
Figure 3b for a direct comparison. To the best of our
knowledge, these TOFs represent some of the highest values
reported for the hydrogenation of allyl alcohol using
40,64−68
comparable Pd nanocatalysts.
In particular, Pd DENs
Once the nanomaterials were characterized, their catalytic
properties were tested using the hydrogenation of allyl alcohol
to 1-propanol (Figure 3a). Hydrogenation reactions are
created inside G4-PAMAM dendrimers yielded a TOF of 480
mol product (mol Pd × h)− for the hydrogenation of allyl
alcohol, which is significantly lower than the values obtained for
1
63
67
critically important on industrial scales, where the peptide−
dendrimer conjugate template could directly influence the
material properties. To this end, the presence of dendrimers
within the R5 template may perturb the scaffold, making it
more porous for enhanced substrate diffusion to the reactive
metal surface. This, in turn, would likely lead to higher degrees
of reactivity. For this reaction, 0.05 mol % Pd of each material
was bubbled with H for 30 min in H O at room temperature
the materials prepared using the R5/PAMAM conjugates.
Note that when the hydrogenation of allyl alcohol was studied
in the presence of the metal free dendrimer, R5 peptide, and
R5/dendrimer conjugates, no product formation or substrate
isomerization was observed.
While it is clear that significantly enhanced reactivity was
observed from the Pd nanomaterials templated with the R5/
dendrimer conjugates compared to their R5 only templated
counterparts, the metallic structures of both sets of materials
were similar in size, composition, and morphology as examined
by TEM. As such, the R5/dendrimer conjugate may promote
enhanced reactivity by altering the spatial arrangement of Pd
nanomaterials through changes in the template organic
framework. To probe these structural features, DLS and
SAXS studies were performed to provide size-scale structural
information on the Pd nanomaterials templated by both free
and dendrimer-bound R5 scaffolds. From this data, comparative
analysis of the global structural differences can be determined in
solution, including both the organic and inorganic, which is
difficult to assess using microscopy methods that typically
require a sample in the solid state. DLS data for Pd
nanomaterials templated with the R5-PAMAM conjugates
demonstrated size aggregates in solution ranging from 150 to
350 nm (Figure 4a). Such a size range is nearly 100-fold larger
than the inorganic components observed by TEM. Note this is
2
2
to saturate the metallic surface with H . Following the addition
2
of the allyl alcohol substrate, hydrogenation of the olefin to 1-
propanol occurred at the metallic surface. Formation of
propionaldehyde, an allyl alcohol isomer, was also observed
in minor amounts during the initial stages of the reaction,
where the isomer was eventually converted to the anticipated
final product.
To quantify the reaction efficiency, aliquots were extracted at
selected time intervals to determine TOF values for each
system. The reaction TOFs for the dendrimer/R5-templated
Pd nanomaterials are shown in Figure 3b and Table 1. In
general, the catalytic reactivities of the materials prepared at
different Pd:peptide ratios were quite similar, with average
TOFs ranging between ∼4000 and 4800 mol product (mol Pd
−
1
×
h) . Such values are significantly higher than those obtained
from the Pd materials templated by the free R5 peptide
template, which possessed an average TOF of 3000 ± 100 mol
4
087
dx.doi.org/10.1021/cm5007444 | Chem. Mater. 2014, 26, 4082−4091

Chemistry of Materials
Article
scaffolds be completely filled, significantly larger and randomly
aggregated bulk Pd would have been observed by TEM, which
was never observed for any of the samples.
To more fully evaluate the global structure of the
biotemplated materials, SAXS analysis was completed. SAXS
patterns (Supporting Information, Figure S2) were obtained for
the Pd nanostructures templated by both the R5/PAMAM and
unconstrained R5 frameworks. Due to the various morpholo-
gies obtained via these approaches (particularly those samples
containing anisotropic particles and/or a mixture of morphol-
ogies), simple analysis of the Guinier and Porod regions of the
SAXS pattern could not be achieved. Pair-distance distribution
function (PDDF) analysis, however, can be used on the
obtained SAXS patterns for materials displaying a higher degree
of inhomogeneity and comparative irregularity. In this method,
the SAXS pattern is fit to an autocorrelation function and
inverse Fourier transformed into real-space distances. The
resulting PDDF, P(r), describes the distances between
scattering objects (i.e., Pd atoms) in a larger macromolecular
69
structure. As shown in Figure 4b, PDDFs for the Pd
nanostructures templated by the R5/PAMAM conjugates
exhibited maximums of ∼3 nm for all samples and then
extended to larger pair distances. This particle size is consistent
with the size-scales observed for the metallic components by
TEM, an indication that the minimum possible size for Pd is
obtained in solution using R5/dendrimer templates and local
particle aggregation is avoided. Conversely, the Pd materials
templated by the free R5 scaffold possessed markedly different
PDDFs, where the maximums were between 6 and 10 nm for
all Pd:R5 ratios (Figure 4c). Such an increase in the first pair
distances suggests that differing degrees of local particle
aggregation occurred using the unconstrained peptide template.
Note that lower pair-distance shoulders are present for Pd
templated with unconstrained R5, indicating a degree of
nonaggregated nanomaterials in solution. The highly flexible R5
scaffolds can more easily participate in intermolecular
interactions compared to the conformationally hindered
peptides in the dendrimer conjugates. Such increases in
template−template interactions likely facilitates an enhanced
Pd aggregation as observed in the PDDF data for R5-templated
Pd.
It is worth noting that the PDDFs were extended to larger
sizes (Supporting Information, Figure S3), which indicated that
larger Pd superstructures were assembled, consistent with DLS
data (Figure 4a). Under these considerations and the
observations from the SAXS and DLS data, the availability of
the reactive Pd surface within the template is higher for
materials generated with R5−dendrimer conjugates. In
contrast, the aggregated Pd structures found within the R5
template were likely to impede substrate diffusion and present a
lower metallic surface area, thus diminishing the catalytic
reactivity for the hydrogenation of allyl alcohol as compared to
R5−dendrimer templated materials.
Figure 4. (a) DLS of Pd nanomaterials templated with the R5/
dendrimer bioconjugate and PDDFs of the Pd nanomaterials
templated with (b) the R5/dendrimer scaffold and (c) the
unconstrained R5 framework.
a similar aggregate size range for the Pd structures templated
40
CONCLUSIONS
with free R5 over a series of Pd:R5 ratios. Given the
observation achieved via TEM for the metallic component, the
overall larger structures measured via DLS, which represent the
entire templated structure including the bioframework, are
presumed to be semihollow, allowing for relatively easy
substrate diffusion, catalytic turnover, and product release.
This semihollow structure correlates well with the materials
growing within the biomimetic framework that is not
completely filled by the metallic components. Should the
■
In summary, PAMAM dendrimers were used as water-soluble
scaffolds for the attachment of R5 peptides, which together
mimic materials-binding peptide architectures found in nature.
These R5−dendrimer conjugates were capable of templating
Pd nanomaterials of varying morphologies, whose structures
were dependent on the Pd:R5 ratio and peptide surface density
on the dendrimer periphery. Furthermore, morphological
trends from R5/dendrimer-templated Pd nanomaterials differ
4
088
dx.doi.org/10.1021/cm5007444 | Chem. Mater. 2014, 26, 4082−4091

Chemistry of Materials
Article
from those created using the free R5 scaffold in solution,
indicating that the conformation of the peptide sequence and
the incorporated polymer component are critically important in
dictating both the Pd morphology and catalytic activity. Once
generated, the reactivity was probed using the hydrogenation of
allyl alcohol as a model system. For this, the majority of the Pd
materials templated with the peptide/dendrimer constructs1
demonstrated TOFs of >1000 mol product (mol Pd × h)−
units higher than the materials generated using the dendrimer-
free peptide templates. Similarities in particle size observed
from TEM indicated that the increase in catalytic activity was
due to differences in the arrangement of both the templates and
the Pd materials, which was confirmed via SAXS and DLS.
PDDF analysis from the SAXS patterns showed that the R5-
dendrimer-templated Pd displayed size-scales consistent with
TEM observations, while the free R5-templated materials
tended to aggregate into larger structures, effectively reducing
the available surface area for catalysis. More generally, this study
demonstrates the capability to alter the morphology and
properties of peptide-derived nanomaterials by simply choosing
a system that mimics the constrained nature of the material
binding peptide observed in nature.
Allyl Alcohol Hydrogenation. Catalytic hydrogenation of allyl
alcohol with the Pd nanoparticles was performed using previously
4
0,64
published methods.
Briefly, 0.05 mol % Pd of the different
materials was added to 25.0 mL of H O and 20.0 μL of antifoam in a
2
three-neck round-bottom flask. Under constant stirring, H gas was
2
bubbled through the solution using a glass gas dispersion tube at a
gauge pressure of 50 kPa. After 30 min, 25.0 mL of a 50.0 mM
aqueous allyl alcohol solution was added to the reaction flask, followed
immediately by the removal of a 1.00 mL aliquot to serve as an initial
time point. Subsequent aliquots were extracted at 1, 5, 10, 15, 20, 25,
and 30 min. The extracted aliquots were injected into a gas
chromatograph (GCAgilent 7820A) equipped with a DB-ALC1
column and a flame ionization detector (FID) without further
purification. GC response factors were determined using standard
solutions of allyl alcohol, n-propanol (product), and propionaldehyde
6
4
(isomer). With retention times and response factors established,
TOF values were calculated.
Characterization. The formation of the R5−dendrimer conjugates
was confirmed using attenuated total reflectance Fourier transform
infrared spectroscopy (ATR-FTIR, PerkinElmer Frontier) by analyz-
ing solid samples produced via lyophilization of aqueous solutions
(
Labconco, FreeZone 4.5). Comparison of the number of covalently
attached peptides to each PAMAM dendrimer was accomplished via
fluorescence measurements employing tryptophan as a fluorescent
label. Using the automated peptide synthesis techniques described
above, the fluorescent residue was incorporated at the opposite termini
METHODS
■
of the cysteine moiety. The fluorescence was measured with a λex
=
Materials. Hydroxyl-terminated G4 PAMAM (ethylenediamine
2
80 nm and a λem = 348 nm using a BioTek Synergy H1 Hybrid Multi-
core), K PdCl , NaBH , allyl alcohol, and antifoam SE-15 (nonionic
2
4
4
Mode Microplate Reader. CD spectroscopy was performed on a Jasco
J-815 CD spectrometer using a quartz cuvette with a 0.5 cm path
length from 190−300 nm at a scan rate of 20 nm/min. The peptide
concentration (not conjugate concentration) in solution was 50 μM.
Secondary structure contributions were deconvoluted from the
individual CD spectrum using CDPro software. DLS studies were
conducted using a Malvern Zetasizer nano ZS instrument. TEM
analysis of the Pd nanomaterial morphologies was examined using a
Phillips CM-200, while HRTEM was performed on a FEI Titan. Sizing
analysis was performed on at least five individual images using >100
different structures to complete the histograms. The relative
morphology population created per peptide−dendrimer conjugate
was determined by counting the number of structures in at least five
individual images. SAXS studies were performed at the 12-ID-B
beamline at the Advanced Photon Source (APS), Argonne National
Laboratory (ANL) using 12 keV irradiation and Pilatus area detector.
Samples were analyzed in 2 mm (outer diameter) quartz capillaries at
concentrations employed for sample preparation. PDDFs were
1
0% emulsion of silicone defoamer) were purchased from Sigma. N-
(p-maleimidophenyl)isocyanate was purchased from Thermo Scien-
tific, while dimethyl sulfoxide (DMSO) was purchased from BDH
Chemicals. All FMOC amino acids and chemicals used in peptide
synthesis were purchased from Advanced Chemtech. Ultrapure, 18.2
MΩ cm water was used for all experiments.
R5/Dendrimer Conjugation. All peptides were synthesized using
standard solid-phase FMOC automated peptide synthesis protocols
employing a TETRAS model peptide synthesizer (CreoSalus). Once
prepared, the crude peptides were purified via reverse-phase HPLC
and confirmed by MALDI TOF mass spectrometry. Cysteine-labeled
R5 peptides were covalently attached to the surface of hydroxyl-
terminated G4 PAMAM dendrimers using PMPI. For this, an aqueous
dendrimer solution was prepared at a concentration of 5 μM and a pH
of 9.00−9.25 using KOH to drive the coupling of PMPI to the surface
hydroxyls. Fresh PMPI stocks were made at 0.1 M in DMSO and
added to the PAMAM solutions at the desired stoichiometric ratio for
peptide coupling. Ratios of 4, 8, and 16 peptides per dendrimer were
studied. PMPI coupling to the dendrimer was performed under
magnetic stirring for at least 2.0 h at room temperature. Once the
PMPI was coupled to the dendrimer surface, the materials were
dialyzed using SnakeSkin dialysis tubing (7000 MWCO) for at least
7
0
71
calculated with the autocorrelation method of Moore using the
7
2
Irena SAXS data analysis macros written in Igor Pro.
ASSOCIATED CONTENT
5.0 h to remove any unreacted coupling reagent. Subsequently, the pH
■
of the PMPI-modified dendrimers was adjusted to 6.25−6.75 using
HCl, after which stoichiometric amounts of the cysteine-containing
peptides were added from a freshly prepared 10 mg/mL peptide stock
solution. After at least 2.0 h of reaction while vigorously stirring, the
R5−dendrimer conjugates were dialyzed for ≥12.0 h to remove any
unreacted peptide.
*
S
Supporting Information
A detailed schematic of R5-PAMAM conjugation, summaries of
R5 secondary structure for R5/PAMAM and CR5, size
histograms of Pd nanomaterials from TEM analysis, SAXS
patterns of Pd templated with R5-PAMAM and unconstrained
R5, and PDDFs derived from SAXS patterns for Pd
nanomaterials templated with free R5 and R5/dendrimer
conjugates examined at larger pair distances. This material is
available free of charge via the Internet at http://pubs.acs.org.
Pd Nanomaterial Synthesis. The Pd nanomaterials encapsulated
within the peptide/dendrimer template were synthesized at specific
metal:peptide ratios using previously described methods.
4
3
8−41
Briefly,
00 μL of 5 μM R5/dendrimer conjugate was added to a
predetermined volume of H O such that the final conjugate
2
concentration was 1 μM. Under magnetic stirring, freshly prepared,
aqueous 0.1 M K PdCl was added at selected Pd:peptide ratios. Note
that Pd:peptide ratios were used and not Pd:dendrimer ratios. After 15
AUTHOR INFORMATION
2
4
■
Corresponding Authors
*M.R.K.: Phone (305) 284-9351; e-mail knecht@miami.edu.
*R.R.N.: Phone (937) 255-9717; e-mail rajesh.naik@us.af.mil.
min of stirring, 75 μL of freshly prepared 0.1 M NaBH was added to
4
reduce the Pd2 to Pd. Nanomaterial samples that were not stable after
+
2
4 h were not used for characterization or catalytic analysis.
4
089
dx.doi.org/10.1021/cm5007444 | Chem. Mater. 2014, 26, 4082−4091

Chemistry of Materials
Present Address
R.B.: Department of Chemical and Materials Engineering,
University of Kentucky, Lexington, Kentucky 40506, United
States.
Article
(20) Kim, S. N.; Kuang, Z.; Slocik, J. M.; Jones, S. E.; Cui, Y.; Farmer,
B. L.; McAlpine, M. C.; Naik, R. R. J. Am. Chem. Soc. 2011, 133,
∥
1
(
4480−14483.
21) So, C. R.; Hayamizu, Y.; Yazici, H.; Gresswell, C.; Khatayevich,
D.; Tamerler, C.; Sarikaya, M. ACS Nano 2012, 6, 1648−1656.
(22) Tang, Z.; Palafox-Hernandez, J. P.; Law, W.-C.; E. Hughes, Z.;
Swihart, M. T.; Prasad, P. N.; Knecht, M. R.; Walsh, T. R. ACS Nano
Funding
This work was supported in part by the Air Force Office of
Scientific Research (RN) and National Science Foundation
2
(
(
013, 7, 9632−9646.
23) Slocik, J. M.; Naik, R. R. Adv. Mater. 2006, 18, 1988−1992.
24) Nam, K. T.; Kim, D.-W.; Yoo, P. J.; Chiang, C.-Y.; Meethong,
(
M.R.K.: DMR-1145175). Further financial support was
provided by University of Miami. N.M.B acknowledges
fellowship support from the National Research Council
Research Associateship award. Use of the Advanced Photon
Source, an Office of Science User Facility operated for the U.S.
Department of Energy (DOE) Office of Science by Argonne
National Laboratory, was supported by the U.S. DOE under
Contract No. DE-AC02-06CH11357.
N.; Hammond, P. T.; Chiang, Y.-M.; Belcher, A. M. Science 2006, 312,
85−888.
25) Kuang, Z.; Kim, S. N.; Crookes-Goodson, W. J.; Farmer, B. L.;
Naik, R. R. ACS Nano 2009, 4, 452−458.
26) Slocik, J. M.; Tam, F.; Halas, N. J.; Naik, R. R. Nano Lett. 2007,
8
(
(
7
, 1054−1058.
(27) Lee, Y. J.; Lee, Y.; Oh, D.; Chen, T.; Ceder, G.; Belcher, A. M.
Nano Lett. 2010, 10, 2433−2440.
28) Nam, Y. S.; Park, H.; Magyar, A. P.; Yun, D. S.; Pollom, T. S.;
Belcher, A. M. Nanoscale 2012, 4, 3405−3409.
29) Cui, Y.; Kim, S. N.; Naik, R. R.; McAlpine, M. C. Acc. Chem. Res.
2012, 45, 696−704.
30) Slocik, J. M.; Zabinski, J. S.; Phillips, D. M.; Naik, R. R. Small
008, 4, 548−551.
31) Coppage, R.; Slocik, J. M.; Sethi, M.; Pacardo, D. B.; Naik, R. R.;
Notes
(
The authors declare no competing financial interest.
(
ACKNOWLEDGMENTS
■
(
2
(
The authors acknowledge Ms. Beverley Briggs for assistance
with peptide synthesis.
Knecht, M. R. Angew. Chem., Int. Ed. 2010, 49, 3767−3770.
(32) Coppage, R.; Slocik, J. M.; Ramezani-Dakhel, H.; Bedford, N.
M.; Heinz, H.; Naik, R. R.; Knecht, M. R. J. Am. Chem. Soc. 2013, 135,
REFERENCES
■
(
1
1) Dickerson, M. B.; Sandhage, K. H.; Naik, R. R. Chem. Rev. 2008,
08, 4935−4978.
2) Sarikaya, M.; Tamerler, C.; Jen, A. K. Y.; Schulten, K.; Baneyx, F.
Nat. Mater. 2003, 2, 577−585.
3) Briggs, B. D.; Knecht, M. R. J. Phys. Chem. Lett. 2012, 3, 405−
18.
1
(
1
(
1048−11054.
33) Kroger, N.; Deutzmann, R.; Sumper, M. Science 1999, 286,
129−1132.
34) Chen, C.-L.; Zhang, P.; Rosi, N. L. J. Am. Chem. Soc. 2008, 130,
̈
(
(
4
(
(
1
(
(
3555−13557.
35) Sewell, S. L.; Wright, D. W. Chem. Mater. 2006, 18, 3108−3113.
4) Brown, S. Nat. Biotechnol. 1997, 15, 269−272.
5) Naik, R. R.; Stringer, S. J.; Agarwal, G.; Jones, S. E.; Stone, M. O.
36) Cole, K. E.; Ortiz, A. N.; Schoonen, M. A.; Valentine, A. M.
Chem. Mater. 2006, 18, 4592−4599.
37) Bhandari, R.; Knecht, M. R. Catal. Sci. Technol. 2012, 2, 1360−
366.
38) Jakhmola, A.; Bhandari, R.; Pacardo, D. B.; Knecht, M. R. J.
Nat. Mater. 2002, 1, 169−172.
6) Chiu, C.-Y.; Li, Y.; Ruan, L.; Ye, X.; Murray, C. B.; Huang, Y. Nat.
Chem. 2011, 3, 393−399.
7) Pacardo, D. B.; Sethi, M.; Jones, S. E.; Naik, R. R.; Knecht, M. R.
ACS Nano 2009, 3, 1288−1296.
8) Reiss, B. D.; Mao, C.; Solis, D. J.; Ryan, K. S.; Thomson, T.;
Belcher, A. M. Nano Lett. 2004, 4, 1127−1132.
9) Patwardhan, S. V.; Emami, F. S.; Berry, R. J.; Jones, S. E.; Naik, R.
R.; Deschaume, O.; Heinz, H.; Perry, C. C. J. Am. Chem. Soc. 2012,
34, 6244−6256.
10) Dickerson, M. B.; Jones, S. E.; Cai, Y.; Ahmad, G.; Naik, R. R.;
Kroger, N.; Sandhage, K. H. Chem. Mater. 2008, 20, 1578−1584.
11) Ahmad, G.; Dickerson, M. B.; Cai, Y.; Jones, S. E.; Ernst, E. M.;
(
1
(
(
(
Mater. Chem. 2010, 20, 1522−1531.
(39) Bhandari, R.; Knecht, M. R. ACS Catal. 2011, 1, 89−98.
(40) Bhandari, R.; Pacardo, D. B.; Bedford, N. M.; Naik, R. R.;
(
Knecht, M. R. J. Phys. Chem. C 2013, 117, 18053−18062.
(
(
(
41) Bhandari, R.; Knecht, M. R. Langmuir 2012, 28, 8110−8119.
42) Grayson, S. M.; Frechet, J. M. J. Chem. Rev. 2001, 101, 3819−
́
1
(
3
868.
(
43) Crooks, R. M.; Zhao, M.; Sun, L.; Chechik, V.; Yeung, L. K. Acc.
Chem. Res. 2000, 34, 181−190.
44) Myers, V. S.; Weir, M. G.; Carino, E. V.; Yancey, D. F.; Pande,
S.; Crooks, R. M. Chem. Sci. 2011, 2, 1632−1646.
45) Bronstein, L. M.; Shifrina, Z. B. Chem. Rev. 2011, 111, 5301−
344.
46) Esfand, R.; Tomalia, D. A. Drug Discovery Today 2001, 6, 427−
436.
(47) Mintzer, M. A.; Grinstaff, M. W. Chem. Soc. Rev. 2011, 40, 173−
190.
(48) Stiriba, S.-E.; Frey, H.; Haag, R. Angew. Chem., Int. Ed. 2002, 41,
̈
(
(
Vernon, J. P.; Haluska, M. S.; Fang, Y.; Wang, J.; Subramanyam, G.;
Naik, R. R.; Sandhage, K. H. J. Am. Chem. Soc. 2007, 130, 4−5.
12) Arakaki, A.; Webb, J.; Matsunaga, T. J. Biol. Chem. 2003, 278,
(
5
(
(
8
(
745−8750.
13) Thai, C. K.; Dai, H.; Sastry, M. S. R.; Sarikaya, M.; Schwartz, D.
T.; Baneyx, F. Biotechnol. Bioeng. 2004, 87, 129−137.
14) Flynn, C. E.; Mao, C.; Hayhurst, A.; Williams, J. L.; Georgiou,
G.; Iverson, B.; Belcher, A. M. J. Mater. Chem. 2003, 13, 2414−2421.
15) Whaley, S. R.; English, D. S.; Hu, E. L.; Barbara, P. F.; Belcher,
A. M. Nature 2000, 405, 665−668.
16) Mao, C.; Flynn, C. E.; Hayhurst, A.; Sweeney, R.; Qi, J.;
Georgiou, G.; Iverson, B.; Belcher, A. M. Proc. Nat. Acad. Sci. U.S.A.
003, 100, 6946−6951.
17) Spoerke, E. D.; Voigt, J. A. Adv. Funct. Mater. 2007, 17, 2031−
037.
18) Wang, S.; Humphreys, E. S.; Chung, S.-Y.; Delduco, D. F.;
(
(
1329−1334.
(
(49) Barth, A. Prog. Biophys. Mol. Biol. 2000, 74, 141−173.
(50) Jackson, M.; Mantsch, H. H. Crit. Rev. Biochem. Mol. Biol. 1995,
30, 95−120.
2
(
2
(
(51) Wood, R.; Piantadosi, C.; Snyder, F. J. Lipid Res. 1969, 10, 370−
373.
(52) Pender, M. J.; Sowards, L. A.; Hartgerink, J. D.; Stone, M. O.;
Lustig, S. R.; Wang, H.; Parker, K. N.; Rizzo, N. W.; Subramoney, S.;
Naik, R. R. Nano Lett. 2005, 6, 40−44.
Chiang, Y.-M.; Jagota, A. Nat. Mater. 2003, 2, 196−200.
(53) Nygren, P.; Lundqvist, M.; Broo, K.; Jonsson, B.-H. Nano Lett.
2008, 8, 1844−1852.
(
19) Cui, Y.; Kim, S. N.; Jones, S. E.; Wissler, L. L.; Naik, R. R.;
McAlpine, M. C. Nano Lett. 2010, 10, 4559−4565.
(54) Doi, N.; Yanagawa, H. Cell. Mol. Life Sci. 1998, 54, 394−404.
4
090
dx.doi.org/10.1021/cm5007444 | Chem. Mater. 2014, 26, 4082−4091

Chemistry of Materials
Article
(
55) Higashi, N.; Koga, T.; Niwa, M. ChemBioChem 2002, 3, 448−
54.
56) Javor, S.; Natalello, A.; Doglia, S. M.; Reymond, J.-L. J. Am.
Chem. Soc. 2008, 130, 17248−17249.
4
(
̈
(
2
(
3
(
3
(
57) Orberg, M.-L.; Schillen
007, 8, 1557−1563.
58) Knecht, M. R.; Wright, D. W. Chem. Commun. 2003, 3038−
039.
59) Zhao, M.; Crooks, R. M. Angew. Chem., Int. Ed. 1999, 38, 364−
66.
60) Wilson, O. M.; Knecht, M. R.; Garcia-Martinez, J. C.; Crooks, R.
M. J. Am. Chem. Soc. 2006, 128, 4510−4511.
61) Pande, S.; Weir, M. G.; Zaccheo, B. A.; Crooks, R. M. New J.
Chem. 2011, 35, 2054−2060.
62) Scott, R. W. J.; Ye, H.; Henriquez, R. R.; Crooks, R. M. Chem.
Mater. 2003, 15, 3873−3878.
63) Murphree, E. V.; Brown, C. L.; Gohr, E. J. Ind. Eng. Chem. 1940,
2, 1203−1212.
64) Bhattacharjee, S.; Bruening, M. L. Langmuir 2008, 24, 2916−
920.
́
, K.; Nylander, T. Biomacromolecules
(
(
(
3
(
2
(
65) Jiang, Y.; Gao, Q. J. Am. Chem. Soc. 2005, 128, 716−717.
(
66) Oh, S.-K.; Niu, Y.; Crooks, R. M. Langmuir 2005, 21, 10209−
1
0213.
(
1
(
67) Niu, Y.; Yeung, L. K.; Crooks, R. M. J. Am. Chem. Soc. 2001,
23, 6840−6846.
68) Ornelas, C.; Ruiz, J.; Salmon, L.; Astruc, D. Adv. Synth. Catal.
2
(
008, 350, 837−845.
69) Hong, X.; Hao, Q. Appl. Phys. Lett. 2009, 94, 083903−083903−
3.
(
70) Chan, W. C.; White, P. D. Fmoc Solid Phase Peptide Synthesis;
Oxford University Press: Oxford, 2004.
(
(
71) Moore, P. J. Appl. Crystallogr. 1980, 13, 168−175.
72) Ilavsky, J.; Jemian, P. R. J. Appl. Crystallogr. 2009, 42, 347−353.
4
091
dx.doi.org/10.1021/cm5007444 | Chem. Mater. 2014, 26, 4082−4091