Biocatalytic induction of supramolecular order

Article abstract of DOI:10.1038/nchem.861

Supramolecular gels, which demonstrate tunable functionalities, have attracted much interest in a range of areas, including healthcare, environmental protection and energy-related technologies. Preparing these materials in a reliable manner is challenging, with an increased level of kinetic defects observed at higher self-assembly rates. Here, by combining biocatalysis and molecular self-assembly, we have shown the ability to more quickly access higher-ordered structures. By simply increasing enzyme concentration, supramolecular order expressed at molecular, nano- and micro-levels is dramatically enhanced, and, importantly, the gelator concentrations remain identical. Amphiphile molecules were prepared by attaching an aromatic moiety to a dipeptide backbone capped with a methyl ester. Their self-assembly was induced by an enzyme that hydrolysed the ester. Different enzyme concentrations altered the catalytic activity and size of the enzyme clusters, affecting their mobility. This allowed structurally diverse materials that represent local minima in the free energy landscape to be accessed based on a single gelator structure.

Full text of DOI:10.1038/nchem.861

ARTICLES
PUBLISHED ONLINE: 10 OCTOBER 2010 | DOI: 10.1038/NCHEM.861
Biocatalytic induction of supramolecular order
Andrew R. Hirst1,2†‡, Sangita Roy , Meenakshi Arora , Apurba K. Das , Nigel Hodson , Paul Murray ,
1‡
1
1†
3
4
4
5
5
6
6
Stephen Marshall , Nadeem Javid , Jan Sefcik , Job Boekhoven , Jan H. van Esch ,
7
†
7
1
Stefano Santabarbara , Neil T. Hunt and Rein V. Ulijn
*
Supramolecular gels, which demonstrate tunable functionalities, have attracted much interest in a range of areas, including
healthcare, environmental protection and energy-related technologies. Preparing these materials in a reliable manner is
challenging, with an increased level of kinetic defects observed at higher self-assembly rates. Here, by combining
biocatalysis and molecular self-assembly, we have shown the ability to more quickly access higher-ordered structures. By
simply increasing enzyme concentration, supramolecular order expressed at molecular, nano- and micro-levels is
dramatically enhanced, and, importantly, the gelator concentrations remain identical. Amphiphile molecules were prepared
by attaching an aromatic moiety to a dipeptide backbone capped with a methyl ester. Their self-assembly was induced by
an enzyme that hydrolysed the ester. Different enzyme concentrations altered the catalytic activity and size of the enzyme
clusters, affecting their mobility. This allowed structurally diverse materials that represent local minima in the free energy
landscape to be accessed based on a single gelator structure.
olecular self-assembly1–7 can be controlled using a variety We hypothesized that by taking advantage of the unique localized
8
,9
10
of stimuli, including chemical and mechanical triggers, nucleation and growth mechanism observed for biocatalysed
11
M
as well as X-rays . Although the traditional premise in self- systems, in combination with kinetic control, it would be possible
assembly suggests that supramolecular material properties can be to direct the self-assembling process towards such local thermodyn-
fully encoded into molecular building blocks, it is increasingly amic minima. Here, using the enzyme subtilisin as a biocatalyst, we
apparent that the chosen self-assembly pathway is central to the demonstrate the formation of non-equilibrium supramolecular
final structure and its material functionality. Biocatalytic control structures that self-assemble more quickly, and are more highly
of self-assembly systems is a novel direction for laboratory-based ordered as enzyme concentration increases.
self-assembly1 , although it is omnipresent in the biological
2–17
world. Indeed, enzymatically controlled self-assembly and dis- Results and discussion
assembly underlies vital processes such as cell movement, intracellu- Biocatalytic self-assembly and gelation. Self-assembling peptides
21–23
lar transport and muscle contraction. In chemists’ hands, the are versatile building blocks for the production of nanostructures
.
combination of biocatalysis and molecular self-assembly has recently Our system focuses on a number of related precursors based on
emerged as a powerful new approach to make novel stimuli-respon- aromatic peptide amphiphiles 6,24,25, namely Fmoc-dipeptide
sive molecular materials1 . We believe that catalytic control of methyl esters, where Fmoc is the chromophore N-(fluorenyl-9-
self-assembly provides important new methodology beyond such methoxycarbonyl), and the dipeptides are FF (1), FY (2), YL (3),
triggering of material transitions. In particular, the combination of VL (4) and FL (5), where F is the amino acid phenylalanine, Y is
biological selectivity, localized action and operation under constant, tyrosine, L is leucine and V is valine (see Fig. 1a, Table 1 and
physiological conditions provides a new methodology for bottom- Supplementary Table S1). Molecular self-assembly is initiated by
up nanofabrication of future soft materials and devices, allowing subtilisin—a hydrolytic enzyme from Bacillus licheniformis—
1
2–17
for unprecedented control of supramolecular order.
which hydrolyses the methyl ester to form a peptide derivative
Here, we focus on the control of supramolecular order with few that self-assembles (Fig. 1a). In these systems, nucleation and
defects. In principle, there are two possible approaches to defect early-stage structure growth are spatially confined at the site of
16
reduction—either improving the fidelity of the self-assembly enzyme action, as shown diagrammatically in Fig. 1b . Hence, it
process (avoiding defects) or using fully reversible systems that appears sensible to expect that the properties of the self-assembling
operate under thermodynamic control (repairing defects). The supramolecular structures could be affected by these catalytic centres.
latter approach is generally slow and only applicable to cases Figure 1c shows typical nucleation and initiation of fibre growth as
where the desired structure represents the global equilibrium state captured by atomic force microscopy (AFM). In each case, after 1 h
and where the system is fully reversible, that is, under thermodyn- of incubation at 55 8C and pH 8, followed by cooling to room
16
amic control . Many structures of interest, however, represent temperature, self-supporting gels were formed (Fig. 1d). Fourier
local thermodynamic minima1 . Such structures can potentially transform infrared (FTIR) analysis (Fig. 1e) revealed the formation of
8,19
2
0
be locked by appropriately tuning the assembling conditions . anti-parallel b-sheet structures upon cooling to room temperature.
1
2
WestCHEM, Department of Pure & Applied Chemistry, University of Strathclyde, Glasgow, Scotland, UK, Manchester Interdisciplinary Biocentre, University
3
4
of Manchester, UK, BioAFM Facility, Faculty of Medical and Human Sciences, The University of Manchester, Manchester, UK, Department of EEE,
5
University of Strathclyde, Glasgow, Scotland, UK, Department of Chemical and Process Engineering, University of Strathclyde, Glasgow, Scotland, UK,
6
7
†
Delft University of Technology, Delft, Netherlands, Department of Physics, University of Strathclyde, SUPA, Glasgow, Scotland, UK; Present address:
School of Physics and Astronomy, E C Stoner Building, University of Leeds, Leeds, LS2 9JT, UK (A.R.H.); Department of Chemistry, Indian Institute of
Technology Indore, IET, DAVV Campus, Khandwa Road, Indore 452017, India (A.K.D.); Istituto di Biofisica, Consiglio Nazionale delle Ricerche, Via Celoria
‡
2
6, 20133 Milano, Italy (S.S); These authors contributed equally to this work. *e-mail: rein.ulijn@strath.ac.uk
NATURE CHEMISTRY | VOL 2 | DECEMBER 2010 | www.nature.com/naturechemistry
2010 Macmillan Publishers Limited. All rights reserved.
1089
©

NATURE CHEMISTRY DOI: 10.1038/NCHEM.861
ARTICLES
a
c
b
2
O
R
H
N
O
OMe
N
H
1
O
R
O
III
III
2
O
R
H
N
O
OH
200nm
N
H
1
O
R
O
d
0U 1.5U 3U
6U
12U
e
f
60
Precursor
Gel
0
0
0
0
0
.25
.20
.15
.10
.05
1,624
5
5
4
5
0
5
1,682
18U
24U
30U
36U
40
1,600
1,650
1,700
1,750
0
5
10 15 20 25 30 35 40
Concentration of enzyme (U)
Wavenumber (cm–1
)
Figure 1 | Biocatalytic self-assembly and gelation. a, Chemical structure of the range of Fmoc-dipeptide methyl esters studied here (orange, fluorenyl moiety
of the Fmoc group; blue, dipeptide sequence; red, enzyme cleavable methoxy group). Subtilisin catalyses hydrolysis of these esters to Fmoc-peptide gelators.
b, Schematic of nucleation and growth mechanism of self-assembly controlled by subtilisin. c, AFM analysis of initial stages of the self-assembly process
(
12 U, 20 min at 55 8C, followed by cooling to room temperature and air drying. Here, 1 U is equivalent to the amount of enzyme that hydrolyses casein to
produce a colour equivalent to 1 mmol of tyrosine per minute at pH 7.5 at 37 8C), showing the formation of clusters and fibres. d, Self-supporting, optically
transparent gels are formed in each case. e, FTIR spectrum of gel-phase materials showing formation of the antiparallel b-sheet conformation. f, Melting
temperature (Tgel) of gels formed catalytically (black) and by a heating–cooling cycle (red) at different enzyme concentrations. The error bars show a
standard error corresponding to +0.5 8C.
The rate of catalytic self-assembly could be tuned simply by
changing the amount of enzyme present in the system. A range of
enzyme concentrations, between 1.5 and 36 U (units), were there-
Table 1 | Structures of the amino acids used for the
dipeptides.
1
2
R /R
Amino acid
fore selected but with otherwise identical treatment (1 h at 55 8C,
followed by cooling to room temperature, then left for 72 h to
ensure that structures were kinetically locked). During this time,
enzymatic conversion was dictated by enzyme concentration, with
conversions of between 55 and 99% after 20 min, reaching .96%
for all enzyme concentrations in each case, with complete conver-
sions (.99%) observed after cooling (Supplementary Fig. S1). The
resulting materials were structurally different, as was immediately
evident from a comparison of the melting behaviour of these gels,
which showed that higher enzyme concentrations gave rise to
increasingly stable gels with melting temperatures ranging from
OH
Tyrosine (Y)
Phenylalanine (F)
Leucine (L)
Valine (V)
Tgel ¼ 42 8C to 58 8C. These T values relate to a hard gel bound-
gel
ary point, that is, the temperature at which solvent is released from
the material, providing an indication of the network morphology.
When they were heated to 90 8C to disrupt the molecular order
and then cooled, the same Tgel values were observed in each case
Further analysis by circular dichroism (CD) and fluorescence (vide (58 8C; Fig. 1f). The differences in the properties of the enzymati-
infra) was consistent with previous observations reported for a series cally produced gels therefore result from their supramolecular
of closely related aromatic peptide amphiphiles1
6,24,25
. In these cases, organization. This observation provides strong evidence that kineti-
nano-sized fibres (ꢀ10 nm in diameter) were composed of several b- cally trapped structures are formed under catalytic control, but the
sheets that were locked together by means of p-stacking interactions. imposition of a subsequent heating/cooling cycle gives the thermo-
We will first discuss the final supramolecular structures that were dynamically favoured state. In a previous study by Pochan and col-
obtained at the different enzyme concentrations. The processes that leagues, different moduli were observed at faster assembly rates for
took place during the enzymatic conversion and cooling stages will salt-induced self-assembly and were related to molecular-level
2
6
then be discussed.
defects that had resulted in changes in the network properties .
1090
NATURE CHEMISTRY | VOL 2 | DECEMBER 2010 | www.nature.com/naturechemistry
©
2010 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.861
ARTICLES
a
30
b
0.65
0.60
0.55
c
3
2
0
0
0
0
1
6U
.5U
Ellipticity at 303nm
Emission ratio
2
2
1
1
5
0
5
0
5
1
2
3
2U
4U
6U
–40
0
.50
0.45
.40
–20
10
–
80
0U
1.5U
6U
12U
0
0
3
–
–
–
40
60
80
0.35
0.30
00 350 400 450 500 550
Wavelength (nm)
–120
–160
–200
24U
0
.25
0.20
.15
0.10
36U
0
0
3
00
350
400
450
500
550
0
5
10 15 20 25 30 35 40
240 260 280 300 320 340
Wavelength (nm)
Wavelength (nm)
Concentration of enzyme (U)
d
e
f
200nm
200nm
200nm
Figure 2 | Spectroscopic and structural evidence of catalytically induced supramolecular order. a, Emission spectra (excitation at 280 nm) of Fmoc-YL gel
obtained at different enzyme concentrations. Inset: Emission spectra of the gels after a heating2cooling cycle. b, Correlation of the increase in ellipticity at
303 nm with emission ratio. For ellipticity and emission ratio measurement, experimental data was acquired in triplicate. The average ellipticity at 303 nm
was plotted and the error bars show the standard deviation of the three measurements, corresponding to values in the range +(20.3 to 21.07). For
emission ratio, the average of the three experiments was plotted and error bars correspond to values in the range +0.02–0.04. c, CD spectra of
supramolecular hydrogel of Fmoc-YL, showing that higher-order chirality is induced with increasing enzyme concentrations. d–f, AFM images of the gels at
1
.5, 12 and 30 U (after 72 h).
Spectroscopic evidence of catalytically induced order. To acquire measurements, is expected to give rise to pronounced CD signals.
more insights at the supramolecular level, the gels were Figure 2c shows that not only is this the case at all concentrations
characterized by spectroscopic methods. The principal molecular of subtilisin, but the magnitude of CD is directly correlated with
interactions driving the self-assembly of aromatic peptide quenching of monomeric Fmoc emission at 320 nm (Fig. 2b). It
amphiphiles are weak hydrophobic forces, leading to p–p should be noted that no significant linear dichroism contribution
stacking between the fluorenyl chromophores, and hydrogen was observed (Supplementary Fig. S2). The amide regions of the
bonding between the dipeptides. Fluorescence emission spectra spectra (200–230 nm) were noisy due to high absorbance in this
(shown in Fig. 2a) principally monitored the emission of the region, with contributions from the enzyme and the peptide gelators
fluorenyl moieties, and in this case displayed two characteristic (Supplementary Fig. S2). However, we observed peaks at 280, 294
peaks, a narrow band centred at 320 nm and relatively broad, and 303 nm, which coincide with Fmoc absorption. CD spectra
redshifted structure displaying maxima between 420 and 440 nm. are typically recorded in large macromolecules, such as protein,
The former is assigned to emission of Fmoc-YL monomers, the DNA and even membrane ultrastructures, where the size of the
latter to the emission from the low-energy exciton state(s) supramolecular structure is close to, or exceeds that, of the measur-
2
6
assigned to excimer(s) promoted by p-stacking of Fmoc moieties. ing wavelength . This situation also occurs in the self-assembling
In addition to hydrogen bonding, p–p stacking between fluorenyl structures described here (see AFM data, Fig. 2d–f). In such
moieties is an important feature that underpins fibrillar molecular systems, CD signals result from chiral organization of the supra-
self-assembly, and it is observed in gels produced at every molecular assembly, rather than from inherent molecular chirality
concentration of subtilisin. Nevertheless, it is clear that the relative within the chromophore itself. From the results of Fig. 2c, two
intensity of the monomeric (320 nm) and excimeric (420 nm) main conclusions can be drawn. First, the sign of the CD is the
emission bands depends on the amount of catalyst present in the same for all enzyme concentrations tested, indicating a bias in the
reaction mixture. In particular, the monomeric emission is self-assembled structure for a single-handed structure. Second,
progressively quenched with respect to excimer emission by and more importantly, the level of chiral organization in the self-
increasing the quantities of enzyme (Fig. 2b) up to 36 U. No assembled state can be controlled by the amount of enzyme. This
further increase was observed at higher catalyst concentrations. suggests that increasing the amount of enzyme facilitates the tran-
The analysis of fluorescence emission spectra provides a clear scription of chiral information from the molecular-scale building
indication that enhanced catalyst concentrations promote the blocks to the self-assembled state, and in the presence of relatively
formation of extended p–p interactions among the fluorenyl high enzyme amounts the chirality of the system can be remarkably
moieties. This leads to the formation of extended one- enhanced. After melting and cooling of the gels, similar CD spectra
dimensional arrays that can be interpreted as a more ordered were observed for each sample (Supplementary Fig. S2c), but with
supramolecular structure. Upon heating of the locked gels above significant differences in the spectral features when compared
their Tgel, followed by cooling to room temperature, the same with the samples before melting.
spectra were observed in each case (Fig. 2a, inset), again
providing clear evidence of catalytically induced access of local Formation of catalytic clusters. Analysis by AFM revealed (Figs 2d–f,3a
thermodynamic minima that can be thermally unlocked.
and Supplementary Fig. S3) that the nanometre- to micrometre-scale
The occurrence of extensive excitonic interaction amongst the self-assembled network structure was directly controlled by the
Fmoc chromophores, as indicated by fluorescence emission amount of enzyme, with lower enzyme concentrations giving rise to
NATURE CHEMISTRY | VOL 2 | DECEMBER 2010 | www.nature.com/naturechemistry
2010 Macmillan Publishers Limited. All rights reserved.
1091
©

NATURE CHEMISTRY DOI: 10.1038/NCHEM.861
ARTICLES
a
80
b
2,500
2
1
1
,000
,500
,000
6
4
2
0
0
0
0
500
0
0
5
10
15
20
25
0
5
10
15
20
25
Concentration of enzyme (U)
Concentration of enzyme (U)
c
1
d
180
1
8 U
1.5 U
1
.5 U
1
2 U
1
1
50
20
3
6 U
9
6
3
0
0
0
0
0.1
0.000
0.001
0.002
0.003
0.004
0
10
20
30
40
50
60
Delay time (s)
Time (min)
Figure 3 | Evidence of catalytic clusters and cooperativity. a, Image analysis revealing the extent of bundling in relation to enzyme concentration. Error bars
show the standard deviation, corresponding to a value in the range +3–22 nm. b, DLS results show a slowing down of density fluctuation decay at higher
enzyme concentrations. The 30 and 36 U samples resulted in precipitates and were therefore excluded. c, Autocorrelation function showing a slower
relaxation mode for higher enzyme concentrations. The experiments were performed six times and the resulting autocorrelation functions were averaged;
standard deviations are shown. d, Quenching of intensity of the 320 nm peak for low (1.25 U), medium (12 U) and high (36 U) enzyme concentrations during
heating at 55 8C for 1 h. Experimental data for quenching was acquired in triplicate and the average intensity values of the three measurements were plotted
and error bars correspond to standard deviation in the range +1–6 AU.
shorter fibres that were less bundled (Fig. 2d). Quantitatively, this is likely that the mobility of enzyme clusters within these increasingly
supported by advanced digital image-processing techniques, which rigid interconnected networks is reduced, reducing discontinuities in
were used to generate a graph showing the width distribution of the fibre assembly and thereby favouring the formation of longer fibres, as
fibre bundles (Fig. 3a). We propose that a combination of two observed by AFM. Indeed, there is a clear inverse correlation between
factors dictates the remarkable catalytic induced long-range stacking the bundle widths observed and decay rates (Fig. 3a,b).
described here: enzymes acting cooperatively within catalytic clusters
and the restricted mobility of these clusters as enzyme Dynamics: higher supramolecular order faster. To gain further
concentrations increase. Dynamic light-scattering (DLS) analysis of insight into the processes that underpin structure formation,
enzyme solutions at 55 8C demonstrated that the initial decay rate of spectral changes were followed during enzymatic reaction using
density fluctuations decreased (Fig. 3b) and slower relaxation modes fluorescence. It was found that, for all enzyme concentrations, the
became more prominent (Fig. 3c) with increasing enzyme supramolecular organization of fluorenyl groups (as indicated by
concentrations. Samples with enzyme concentrations of 30 and 36 U quenching of the 320 nm band) occurred during enzymatic
resulted in precipitates and were therefore excluded from further conversion at 55 8C. Figure 3d shows spectral changes between low
analysis. DLS results showed that enzymes formed extended clusters (1.5 U), medium (12 U) and high (36 U) enzyme concentrations,
at all concentrations tested, because decay rates that are an order of clearly showing more rapid formation of quenched structures for
magnitude higher would be expected for relaxation due to Brownian the high concentration. These values showed only minor changes
motion of individual enzyme molecules. Assuming that the initial after cooling and for up to 72 h. In contrast, there was no
decay rates are due to unhindered Brownian motion of clusters, detectable b-sheet signal apparent from the FTIR results during
their mean hydrodynamic radius would vary from 85 to 260 nm the enzymatic conversion (Supplementary Fig. S4); however, clear
with increasing enzyme concentration. However, these clusters are peaks emerged upon cooling, with similar spectra appearing for all
not completely free to move around, as indicated by the presence of enzyme concentrations tested. Clearly, molecular order driven by
slow relaxation modes.
Slow relaxation modes in the DLS autocorrelation function into a b-sheet structure upon cooling.
Fig. 3c) were accompanied by considerable slow-down of the
For the Fmoc-YL-OMe (3) system described above, some gels
p-stacking interactions is induced before gelation, and it is locked
(
initial decay rates at higher enzyme concentrations (Fig. 3b). (24–36 U) were formed at or above their Tgel, whereas others
These observations indicate that extended interconnected networks (1–20 U) were formed below T . Although gelation is not observed
gel
are formed in these samples, with networks becoming more rigid before cooling, one might expect the Tgel compared to the reaction
with an increasing degree of interconnection, similar to previous temperature to have an impact on the gel properties. To study this
2
7–29
observations in colloidal, protein and polymer solutions
. It is effect, two additional gelators were investigated with different Tgel
1092
NATURE CHEMISTRY | VOL 2 | DECEMBER 2010 | www.nature.com/naturechemistry
©
2010 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.861
ARTICLES
values (ꢀ47 8C, Fmoc-FY; 42 8C, Fmoc-VL). These were studied at Atomic force microscopy (AFM). Glass cover slips (Agar Scientific) were cleaned
overnight in ethanol (Aldrich), air-dried and adhered to AFM support stubs (Agar
Scientific) using clear nail varnish. Gel samples were prepared as previously
described, deposited on the glass surface, and allowed to air dry overnight. Samples
were imaged by intermittent contact mode in air using a Veeco Multimode atomic
enzyme concentrations of 1.5, 6, 12, 24 and 36 U. Fmoc-FY showed
evidence of catalytic induced quenching of the monomeric peak,
combined with an enhanced excimer peak, as well as an enzymati-
cally induced Tgel that varied in value between ꢀ34 and 43 8C force microscope with a Nanoscope IIIa controller and an ‘E’ scanner. Imaging was
performed using silicon probes (OTESPA, Veeco Instruments S.A.S.) with a
(
Supplementary Fig. S5a–c). A more modest catalytically induced
21
nominal spring constant of 42 N m . Cantilever oscillation varied between 300 and
50 kHz and the drive amplitude was determined by the Nanoscope software. The
quenching of fluorescence and absence of induced Tgel was observed
for the lower melting Fmoc-VL, which is likely to be less structured
at the assembly temperature of 55 8C (Supplementary Fig. S5d–f).
3
setpoint was adjusted to just below the point at which tip 2 sample interaction was
2
lost. Height and phase images with scan sizes of either 5 or 2 mm were captured at a
These results suggest that catalytic locking of order is most pro- scan rate of 1.49 Hz and at a relative humidity of ,40%. Data were first-order
flattened using the Nanoscope software before image export. The instrument was
nounced under conditions where self-assembly, but not necessarily
gel formation, is favoured.
2
periodically calibrated using a grating with 180-nm-deep, 10-mm depressions.
Because of the correlation between the intensity of the CD signal,
the quenching of monomeric Fmoc emission (Fig. 2), the observed
gel melting temperatures (Fig. 1) and enhanced network formation
Fig. 2), it is apparent that the bundled chiral supramolecular struc-
FTIR spectroscopy. FTIR spectra were acquired in a Bruker Vertex spectrometer
21
with a spectral resolution of 2 cm . The spectra were obtained by averaging 64
interferograms for each sample. Measurements were performed in a standard IR
cuvette (Harrick Scientific), in which the sample was contained between two CaF2
(
tures represent a highly ordered state of these systems that is prefer- windows (thickness, 2 mm) separated by a 25-mm PTFE spacer. All sample
manipulations were performed in a glove box to minimize interference from carbon
entially obtained at higher catalyst concentrations. It is worth noting
dioxide and water vapour; D O was used as the solvent for all the IR measurements.
2
that the formation of well-aligned extended one-dimensional
30–33
p-stacked nanofibres is of relevance to supramolecular electronics
.
Dynamic light scattering (DLS). The required amounts of protease solution were
The results shown here imply that supramolecular electro-conduc-
tivity may be induced and controlled by catalytic self-assembly.
mixed with 100 mM phosphate buffer filtered with a 0.25 mm Anotop filter to
achieve the required protein concentrations of 1.5, 3, 6, 12, 18, 24, 30 and 36 U ml
21
.
In summary, we have demonstrated that by modulating the The samples were vortexed for 1 min and transferred to borosilicate glass cuvettes
amount of catalyst we can access a library of structurally diverse for scattering measurements. DLS experiments were carried out using a 3D LS
spectrometer (LS instruments) using vertically polarized He–Ne laser light (25 mW,
with a wavelength of 632.8 nm) with an avalanche photodiode detector at a
scattering angle of 908. The samples were heated to achieve a temperature of 55 8C
molecular and network assemblies in which, importantly, all
gelator concentrations are identical. The ability to kinetically induce
molecular order in a controlled and reproducible manner allows a
using a circulating water bath. The intensity autocorrelation function g (t) 2 1 was
2
high level of control to be exerted over the assembly process, recorded from time zero up to 25 min to determine the evolution of the average
therebyenabling accessto structurally diverse self-assembled materials apparent hydrodynamic radius with time. For all measurements, the decay of
g2(t) 2 1 was followed up to large values of delay time t to reach the baseline. The
autocorrelation functions were analysed by means of the cumulant method to
determine the average apparent diffusion coefficient D. The decay of the
that are inaccessible via conventional self-assembly. We provide
evidence of a mechanism in which the catalytic activity and mobility
of biocatalytic clusters dictate the induced supramolecular order
1/2
autocorrelation function was modelled as g (t) ¼ (g (t) 2 1) ¼ exp(2Gt), where
1
2
2
observed. These results suggest an important role for engineered G ¼ Dq is the decay rate, q ¼ (4pn/l)sin(u/2) is the scattering vector magnitude, n
catalytic particles in the molecular self-assembly of next-generation is the refractive index of the solvent, u is the scattering angle and l is the wavelength
of the laser. The average hydrodynamic radius rH can be calculated from the Stokes–
soft nanomaterials and devices.
Einstein equation, rH ¼ k T/6phD, where k is the Boltzmann constant, T is the
B
B
absolute temperature and h is the solvent viscosity at the given temperature.
Methods
21
Catalytic self assembly. Fmoc-YL-OMe (10 mmol kg ) was dispersed in a 1 ml
volume of 100 mM sodium phosphate buffer (pH 8) in the presence of varying
concentrations of subtilisin (1.25–60.0 ml (Sigma Aldrich, catalogue number P4860;
LOT 056K1213). The mixture was vortexed (30 s) and sonicated on ice for 20 min to
ensure that a homogeneous mixture was obtained, and the low temperature ensures
that no enzymatic conversion occurs up to this point. This was followed by heating,
either in an oil bath or within the spectrophotometer using a temperature-controlled
cuvet, at 55 8C for 60 min to allow enzymatic conversion to occur. The self-
assembling system was then allowed to cool to room temperature. The gel samples
were then left to stand for periods of 72 h before experimental measurements
were performed. Gelation was considered to have occurred when a homogeneous
Received 6 April 2010; accepted 23 August 2010;
published online 10 October 2010
References
1.
2.
3.
Estroff, L. A. & Hamilton, A. D. Water gelation by small organic molecules.
Chem. Rev. 104, 1201–1217 (2004).
Lehn, J. M. Supramolecular Chemistry—Concepts and Perspectives
(
VCH Weinheim, 1995).
Whitesides, G. M. & Boncheva, M. Beyond molecules: self-assembly of
mesoscopic and macroscopic components. Proc. Natl Acad. Sci. USA 99,
4769–4774 (2002).
‘
solid-like’ material was obtained that exhibited no gravitational flow.
The thermally reversible gel–sol transition temperature (Tgel) was determined
using a ball dropping methodology.
4
.
.
Hirst, A. R. et al. High-tech applications of self-assembling supramolecular
nanostructured gel-phase materials: from regenerative medicine to electronic
devices. Angew. Chem. Int. Ed. 47, 8002–8018 (2008).
Capito, R. M., Azevedo, H. S., Velichko, Y. S., Mata, A. & Stupp, S. I.
Self-assembly of large and small molecules into hierarchically ordered sacs and
membranes. Science 319, 1812–1816 (2008).
High-performance liquid chromatography. A Dionex P680 high-performance
liquid chromatography pump was used to quantify conversions of the enzymatic
reaction. A 50 ml sample was injected onto a Macherey-Nagel C18 column with a
length of 250 mm and an internal diameter of 4.6 mm and 5-mm fused silica
5
2
1
particles at a flow rate of 1 ml min . The eluting solvent system had a linear
gradient of 20% (v/v) acetonitrile in water for 4 min, gradually rising to 80% (v/v)
acetonitrile in water at 35 min. This concentration was kept constant until 40 min
when the gradient was decreased to 20% (v/v) acetonitrile in water at 42 min.
Sample preparation involved mixing 50 ml of gel with acetonitrile–water (950 ml,
6.
7.
8.
Wang, Q. et al. High-water-content mouldable hydrogels by mixing clay and a
dendritic molecular binder. Nature 463, 339–343 (2010).
Kiyonaka, S. et al. Semi-wet peptide/protein array using supramolecular
hydrogel. Nature Mater. 3, 58–64 (2004).
Jonkheijm, P., van der Schoot, P., Schenning, A. P. H. J. & Meijer, E. W. Probing
the solvent-assisted nucleation pathway in chemical self-assembly. Science 313,
5
0:50 mixture) containing 0.1% trifluoroacetic acid. The purity of each identified
peak was determined by UV detection at 280 nm.
80–83 (2006).
9
1
1
.
Lloyd, G. O. & Steed, J. W. Anion-tuning of supramolecular gel properties.
Nature Chem. 1, 437–442 (2009).
Circular dichroism (CD). Spectra were measured on a Jasco J815
spectropolarimeter with 1-s integrations, a step size of 1 nm and a single acquisition
with a slit width of 1 nm due to the dynamic nature of the system. Experimental data
were acquired in triplicate.
0. Carnall, J. M. A. et al. Mechanosensitive self-replication driven by self-
organization. Science 327, 1502–1506 (2010).
1. Cui, H. et al. Spontaneous and X-ray triggered crystallization at long range in
self-assembling filament networks. Science 327, 555–559 (2010).
Fluorescence spectroscopy. Fluorescence emission spectra were measured on a
Jasco FP-6500 spectrofluorometer with light measured orthogonally to the excitation
light, with excitation at 280 nm and an emission data range between 300 and
12. Winkler, S., Wilson, D. & Kaplan, D. L. Controlling b-sheet assembly in
genetically engineered silk by enzymatic phosphorylation/dephosphorylation.
Biochemistry 39, 12739–12746 (2000).
6
00 nm.
NATURE CHEMISTRY | VOL 2 | DECEMBER 2010 | www.nature.com/naturechemistry
2010 Macmillan Publishers Limited. All rights reserved.
1093
©

NATURE CHEMISTRY DOI: 10.1038/NCHEM.861
ARTICLES
1
3. Hu, B. H. & Messersmith, P. B. Rational design of transglutaminase substrate
peptides for rapid enzymatic formation of hydrogels. J. Am. Chem. Soc. 125,
27. Martin, J. E. & Wilcoxon, J. P. Critical dynamics of the sol–gel transition.
Phys. Rev. Lett. 61, 373–376 (1988).
1
4298–14299 (2003).
4. Yang, Z. M. et al. Enzymatic formation of supramolecular hydrogels. Adv. Mater.
6, 1440–1444 (2004).
5. Um, S. H. et al. Enzyme-catalysed assembly of DNA hydrogel. Nature Mater. 5,
97–801 (2006).
28. Fang, L., Brown, W. & Konak, C. Dynamic light scattering study of the sol–gel
transition. Macromolecules 24, 6839–6842 (1991).
1
1
1
1
1
29. Weijers, M., Visschers, R. W. & Nicolai, T. Light scattering study of heat-induced
aggregation and gelation of ovalbumin. Macromolecules 35, 4753–4762 (2002).
30. Yamamoto, Y. et al. Photoconductive coaxial nanotubes of molecularly
connected electron donor and acceptor Layers. Science 314, 1761–1764 (2006).
31. Schenning, A. P. H. J. & Meijer, E. W. Supramolecular electronics; nanowires
from self-assembled p-conjugated systems. Chem. Commun. 26,
3245–3258 (2005).
32. Xu, H. et al. An investigation of the conductivity of peptide nanotube networks
prepared by enzyme-triggered self-assembly. Nanoscale 2, 960–966 (2010).
33. Ashkenasy, N., Seth Horne, W. & Ghadiri, M. R. Design of self-assembling
peptide nanotubes with delocalized electronic states. Small 2, 99–102 (2006).
7
6. Williams, R. J. et al. Enzyme-assisted self-assembly under thermodynamic
control. Nature Nanotech. 4, 19–24 (2009).
7. Adler-Abramovich, L., Perry, R., Sagi, A., Gazit, E. & Shabat D. Controlled
assembly of peptide nanotubes triggered by enzymatic activation of
self-immolative dendrimers. ChemBioChem 8, 859–862 (2007).
8. Cui, H., Chen, Z., Zhong, S., Wooley, K. L. & Pochan, D. J. Block copolymer
assembly via kinetic control. Science 317, 647–650 (2007).
9. Yamamoto, T. et al. Stabilization of a kinetically favored nanostructure: surface
ROMP of self-assembled conductive nanocoils from a norbornene-appended
hexa-peri-hexabenzocoronene. J. Am. Chem. Soc. 128, 14337–14340 (2006).
0. Haines-Butterick, L. et al. Controlling hydrogelation kinetics by peptide design
for three-dimensional encapsulation and injectable delivery of cells. Proc. Natl
Acad. Sci. USA 104, 7791–7796 (2007).
1
1
Acknowledgements
The authors acknowledge the Engineering and Physical Sciences Research Council
2
(
EPSRC), Human Frontiers Science Programme (HFSP) and the Leverhulme Trust (UK)
for funding. The authors also thank L. Birchall and P.F. Caponi for assistance with graphics.
2
2
2
2
1. Zhang, S. Fabrication of novel materials through molecular self-assembly. Nature
Biotechnol. 21, 1171–1178 (2003).
2. Banwell, E. F. et al. Rational design and application of responsive alpha-helical
peptide hydrogels. Nature Mater. 8, 596–600 (2009).
3. Adler-Abramovich, L. et al. Self-assembled arrays of peptide nanotubes by
vapour deposition. Nature Nanotech. 4, 849–854 (2009).
4. Smith, A. M. et al. Fmoc-diphenylalanine self assembles to a hydrogel via a novel
architecture based on pi2pi interlocked beta-sheets. Adv. Mater. 20,
Author contributions
A.R.H., S.R., J.S., S.S. and R.V.U. conceived and designed the experiments. A.R.H., S.R.,
M.A., A.K.D., N.H., N.J. and J.B. performed the experiments. A.R.H., S.R., P.M., J.S., S.S.
and R.V.U. analysed the data. S.M., J.H.v.E. and N.T.H. contributed materials/analysis
tools. A.R.H., S.R., S.S. and R.V.U. co-wrote the paper.
3
7–41 (2008).
Additional information
2
5. Zhang, Y., Gu, H., Yang, Z. & Xu, B. Supramolecular hydrogels respond to
ligand2receptor interaction. J. Am. Chem. Soc. 125, 13680–13681 (2003).
6. Keller, D. & Bustamante, C. Theory of the interaction of light with large
inhomogeneous molecular aggregates. II. psi-type circular dichroism. J. Chem.
Phys. 84, 2972–2980 (1986).
The authors declare no competing financial interests. Supplementary information and
chemical compound information accompany this paper at www.nature.com/
naturechemistry. Reprints and permission information is available online at http://npg.nature.
com/reprintsandpermissions/. Correspondence and requests for materials should be
addressed to R.V.U.
2
1094
NATURE CHEMISTRY | VOL 2 | DECEMBER 2010 | www.nature.com/naturechemistry
©
2010 Macmillan Publishers Limited. All rights reserved.