Polymeric "clickase" Accelerates the Copper Click Reaction of Small Molecules, Proteins, and Cells

Article abstract of DOI:10.1021/jacs.9b04181

Recent work has shown that polymeric catalysts can mimic some of the remarkable features of metalloenzymes by binding substrates in proximity to a bound metal center. We report here an unexpected role for the polymer: multivalent, reversible, and adaptive

Full text of DOI:10.1021/jacs.9b04181

Subscriber access provided by Bethel University
Article
Polymeric “Clickase” Accelerates the Copper Click
Reaction of Small Molecules, Proteins, and Cells
Junfeng Chen, Jiang Wang, Ke Li, Yuhan Wang, Martin
Gruebele, Andrew L. Ferguson, and Steven C. Zimmerman
J. Am. Chem. Soc., Just Accepted Manuscript • Publication Date (Web): 24 May 2019
Downloaded from http://pubs.acs.org on May 24, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 9
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
Polymeric “Clickase” Accelerates the Copper Click Reaction of Small Molecules,
Proteins, and Cells
Junfeng Chen,† Jiang Wang,‡ Ke Li,† Yuhan Wang,§ Martin Gruebele,†,‡,§ Andrew L. Ferguson,¶ and Steven C.
Zimmerman*,†,§
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
†Department of Chemistry, University of Illinois, Urbana, Illinois 61801, United States.
‡Department of Physics, University of Illinois, Urbana, Illinois 61801, United States.
§Center for Biophysics and Quantitative Biology, University of Illinois, Urbana, Illinois 61801, United States.
¶Institute for Molecular Engineering, The University of Chicago, Chicago, Illinois 60637, United States.
Supporting Information
ABSTRACT:
Recent work has shown that polymeric catalysts can mimic some of the remarkable features of metalloenzymes by binding
substrates in proximity to a bound metal center. We report here
an unexpected role for the polymer: multivalent, reversible, and
adaptive binding to protein surfaces allowing for accelerated
catalytic modification of proteins. The catalysts studied are a
group of copper-containing single-chain polymeric nanoparticles
(CuI−SCNP) that exhibit enzyme-like catalysis of the copper-
mediated azide-alkyne cyclo-addition reaction. The CuI−SCNP
use a previously observed “uptake mode,” binding small molecule
alkynes and azides inside a water-soluble amphiphilic polymer
and proximal to copper catalytic sites, but with unprecedented
rates. Remarkably, a combined experimental and computational
study shows that the same CuI−SCNP perform a more efficient
click reaction on modified protein surfaces and cell surface
glycans than do small molecule catalysts. The catalysis occurs
through an “attach mode” where the SCNP reversibly binds
protein surfaces through multiple hydrophobic and electrostatic contacts. The results more broadly point to a wider capability for polymeric
catalysts as artificial metalloenzymes, especially as it relates to bio-applications.
emerged as an especially promising approach.11-13 The advantages of
◼ INTRODUCTION
SCNPs include their modular synthesis,14 broad range of cross-
Synthetic transition metal catalysts are able to perform reactions
linking chemistries,11 adjustable size,14 tunable cell uptake.15 SCNP
in buffered solution, in cellulo, and even in vivo. As such, these
based catalysts have been reported to conduct reactions in aqueous
catalysts are important tools for chemical biology and potentially as
buffer and inside cells,16,17 and accelerate reactions by binding the
new therapeutic agents,1-3 with recent applications ranging from bio-
substrates supramolecularly, the latter demonstrating their artificial
conjugation reactions4 and cell surface editing5 to in situ drug
metalloenzyme behavior.18,19
synthesis.6 Many of the metal catalysts used were optimized for
We focused on SCNP-based catalysts for the copper-mediated
organic synthesis and, thus, function best in aprotic solvents and
alkyne-azide cycloaddition (CuAAC) “click” reaction, because it is
with reagent concentrations in the 0.1-1 M range. In competitive
arguably the most important (bio)orthogonal conjugation
biological media and at the low substrate concentrations frequently
reaction,20 yet still limited by copper toxicity and the need for low
used in bioapplications, the same catalysts often function poorly.
concentrations in biological applications.21,22 We previously used
Transition metal catalysis is common in biology with nature
water-soluble amphiphilic ROMP-derived polymers containing
managing these same challenges through the use of metalloenzymes.
amino acid side-chains for CuI binding. These copper-based SCNP
In particular, the precisely folded polypeptide scaffolds function to
protect the metal center and create a binding cavity to bring the
substrate and metal together.
I
(Cu −SCNP- ) functioned as “clickases,” showing enzyme-like click
1
catalysis and giving serviceable rates at micromolar substrate
concentrations and part-per-million levels of copper.16,18 Herein we
Considerable recent effort has focused on artificial metallo-
enzymes that combine non-natural transition metal catalysts with
either proteins7,8 or polymeric scaffolds.9,10 The surrounding protein
or polymer serves as a functional surrogate for the metalloenzyme
polypeptide backbone. Among the efforts to develop polymeric
transition metal catalysts, single chain nanoparticles (SCNPs) have
2
report a new class of polyacrylamide-based SCNP (SCNP- ) that
catalyze the CuAAC click reaction with rates significantly faster than
1
SCNP- and the best available tris(triazolylmethyl)amine ligands.
More strikingly, we report the unexpected discovery that
I
2
Cu −SCNP- adaptively and reversibly binds protein surfaces
ACS Paragon Plus Environment
1

Journal of the American Chemical Society
Page 2 of 9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. SCNP preparation.
electrostatic repulsion of unimolecular micelle-like structures allows higher concentration crosslinking. (b) Two-step conversion of P-
, DLS data and TEM image (inset) of SCNP- . Scale bar = 20 nm.
(
a) Illustration of SCNP preparation using a “folding and crosslinking” strategy wherein hydrophobic collapse and
1
to SCNP-
2
.
c
2
accelerating the click reaction of modified proteins and enabling
efficient surface glycan editing.23
To increase the propensity for intrachain crosslinking, a “folding
and crosslinking” strategy was pursued (Figure 1a). Thus,
1
poly(pentafluorophenyl acrylate) (P- ) was prepared with DP = 95,
Mn = 23 kDa, and PDI = 1.05 by RAFT polymerization.17 Successive
treatment with (10-aminodecyl)trimethylammonium chloride and
6-aminohexanoic acid followed by 3-azidopropyl-amine provided
◼ RESULTS AND DISCUSSION
Synthesis of SCNPs through “folding and crosslinking” strategy.
2
The development of SCNP- was enabled by three key innovations.
The first involved changing from the poly(norbornene imide)
2
2
polyacrylamide P- . In water, P- likely undergoes hydrophobic
collapse forming a unimolecular micelle-like structure. By using the
1
backbone used in SCNP- to a polyacrylamide-based SCNP related
1
hydrophobic crosslinker, diyne , P- could be covalently
crosslinked using the CuAAC reaction. The repulsion between
2
to that reported by Meijer and Palmans.17,19 This change was
motivated by the finding that the hydrophobic poly(norbornene
imide) backbone did not appear to participate in substrate binding.18
Second, we sought to overcome one of the major challenges of
synthesizing SCNPs on scale, the need for high dilution cross-
linking to minimize inter-chain coupling.24 Third, a covalent cross-
linking strategy was employed that installed a tris(triazolylmethyl)-
amine unit to serve as a more effective CuI ligand (see below).
2
polycationic charges on P- inhibits aggregation and allows the
crosslinking to occur intramolecularly at 0.2 mM a near 100-fold
higher concentration than used for our previously reported covalent
crosslinked SCNPs.14 The nanoparticles were imaged both by
transmission electron microscopy (TEM) and atomic force
microscopy (AFM) (Figure S2 and S3). The hydrodynamic
diameter of SCNPs were determined to be around 6 nm using
dynamic light scattering (DLS) and diffusion-ordered spectroscopy
Figure 2. Comparison reactions conducted in PBS buffer (1x, pH = 7.4) at room temp with
2
(20 μM), 3a (40 μM), sodium ascorbate (2 mM) and
the catalyst. [SCNP- ] = [SCNP- ] = 1, 2, 4 μM (ave. 10 copper ions per SCNP). [BTTAA] = [BTGTA] = [THPTA] = [TBTA] = 10, 20, 40
1
2
μM. Throughout a 1:1 ratio of ligand and CuSO4 used. The error bars for initial rates are the standard deviation of three independent experiments.
ACS Paragon Plus Environment
2

Page 3 of 9
Journal of the American Chemical Society
2
(DOSY) nuclear magnetic resonance (NMR). The shorter
relaxation time T2 measured by NMR indicated a more rigid
structure after the crosslinking, supporting the crosslinked structure.
Uptake Mode: The CuAAC Click Reaction of Small Molecules.
Inspired by the development of the BTTAA ligand by Marlow, Liu,
Wu, and coworkers,21,22 which is reported to give the fastest known
CuAAC reaction, 6-aminohexanoic acid was used in the preparation
ns was calculated. Not surprisingly, SCNP- exhibited a higher
degree of occupancy with anionic and longer chain alkynes (Figures
3b and S9). A general correlation between the uptake and reactivity
is observed and is consistent with enzyme-like activity. Experimental
1
2
3
4
5
6
7
8
3a
2
support for the encapsulation of alkyne by SCNP- was obtained
using two-dimensional nuclear Overhauser spectroscopy (2D-
NOESY) (Figure S8). The importance of substrate binding was also
I
2
1
2
of P- followed by the use of diyne in the click-mediated covalent
observed in the kinetic profile of Cu −SCNP- . The rates of the
I
CuAAC reaction of Cu −SCNP- with varying amounts of and
2
2
showed enzyme-like saturation kinetics in both alkyne and azide.
2
3a
crosslinking reaction. Thus, SCNP- provides a polymeric scaffold
surrounding the BTTAA-like units. The SCNPs could be prepared
with different degrees of crosslinking, but typically an average of ten
crosslinks and thus ten copper binding sites per nanoparticle were
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
I
2
used. To determine the catalytic performance of Cu −SCNP- , the
2
3a
1
rate of the fluorogenic click reaction between azide and alkyne
25
was measured. The rates were compared to those with SCNP- ,
our previously reported SCNP16 and high-performance small
molecule catalysts prepared from CuSO4 and several tris(triazolyl-
methyl)amine-based ligands, including TBTA,26 THPTA,26 and
BTTAA22 (Figure 2). Each reaction was performed at [CuSO4] = 10,
20 and 40 μM and a 1:1 ligand-copper ratio and in the presence of
sodium ascorbate as the reducing agent. Thus, the small molecule
catalyst runs were at 10-fold higher catalyst concentration because
each SCNP has on average ten copper binding sites.
I
2
2
Cu −SCNP- performs the CuAAC click reaction between and
3a
reaction rate is 10 times faster than the most efficient small molecule
significantly faster than the other catalysts. On average, its
I
I
1
catalyst Cu −BTTAA as well as Cu −SCNP- (Figure 2). A near
quantitative conversion was observed within 10 min with only 10
I
μM of copper. To our knowledge, Cu −SCNP- is the most efficient
2
catalyst reported to date for the “native” CuAAC reaction. The
I
utility of Cu −SCNP- for typical synthetic reactions was also
Figure 3. Uptake mode: CuAAC click reaction of small molecules. (a)
Substrates with different charge species and hydrophobicity. (b) Initial
2
studied in water with several hydrophobic alkyne and azide
rates of different substrates with the catalysis of CuI−SCNP-
binding number during the simulation. (c) Snapshot of the substrate
binding of SCNP- from MD simulation. The trimethyl ammonium
structures were colored red, the catalytic centers were colored blue,
other hydrophobic parts on SCNP- were colored grey and the
substrate (3a) was colored purple. (d) Fitting of CuI−SCNP-
kinetic
data to a random-sequential two-substrate Bi-Bi enzyme kinetics
equation fitting of CuI−SCNP-
kinetics data. Reaction rates
determinedby fluorimeter with CuI−SCNP-
(0.5 μM), (10 - 60 μM),
2
over their
I
2
substrates. Using 0.002 mol% of Cu −SCNP- , the reactions gave
isolated yields ranging from 92% to 98%, indicating high turnover
numbers (Table S1). Because the amount of polymer and copper
used is so small, we did not focus on recyclability. Nonetheless, after
the fluorogenic reaction, Cu −SCNP- could be isolated following
2
the click reaction of and and a second round of catalysis showed
a very similar kinetic profile (Figure S4).
2
I
2
2
2
3a
2
The positively charged surface and hydrophobic interior provide
2
I
Cu −SCNP- with the ability to bind substrates resulting in a high
2
2
local concentration of substrates in proximity to the catalytic CuI
centers (uptake mode).27,28 To probe the role of substrate charge and
hydrophobicity, the rate of the CuAAC reaction between
3a (10 - 200 μM), and NaAsc (2 mM) in PBS buffer (1x, pH = 7.4) at
room temperature.
I
2
Cu −SCNP- and alkynes and - were measured. These 14
alkynes can be assigned to four groups: neutral aromatic alkynes ( )
3
5 7
The reaction rate surface could be fit to a two-substrate enzyme
kinetics equation with random sequential Bi-Bi model (Figure 3d
and Figure S7, see detail information under fluorogenic reaction
with small molecules in the experimental section).29
Attach Mode: CuAAC Click Reaction on Protein Substrates.
Transition metal catalysts serve as important tools for modifying
proteins, however, protein concentrations typically in the
3
5
6
7
and cationic ( ), neutral ( ), and anionic ( ) aliphatic alkynes
I
2
(Figure 3a). Cu −SCNP- showed high CuAAC activity for all
5a c
(Figure 3b). The fastest rates are seen with the more hydrophobic
substrates except the low molecular weight cationic alkynes
-
I
2
substrates, although the high activity of Cu −SCNP- means that the
rate differences between substrates are much small than those
micromolar range require catalysts that are both highly active and
biocompatible.30-35 Given the high activity of Cu −SCNP- in the
I
I
1
observed with Cu −SCNP- (Figure S6). The macromolecular
2
I
architecture of Cu −SCNP- conferred an average rate acceleration
2
of 18-fold relative to the most efficient small molecule catalyst
CuAAC reaction, we sought to test its ability to operate on proteins.
To this end, the surface lysine groups of several proteins were
alkynylated with the NHS-ester of 4-pentynoic acid. Bovine serum
albumin (BSA-Al), human serum albumin (HAS-Al), protein A
(PA-Al), mCherry (mCh-Al), transferrin (TF-Al), and lysozyme
(Lz-Al), were selected because of their diverse size, shape, and
function. The average number of alkynes linked to each protein
ranged from two (Lz-Al) to thirteen (BSA-Al and HAS-Al) and was
determined by MALDI-MS (Figure S11). Two additional proteins
streptavidin (StAv-Al) and avidin (Av-Al) were also used, in this case
CuI−BTTAA for the same reaction.
Molecular dynamics (MD) simulations of SCNP- in a water box
2
were performed for 20 ns to better understand the nature of the
macromolecular effect. The polymer exhibited a globular structure
with a diameter of ca. 5.5 nm with a hydrophobic core and
ammonium ions residing on the surface (Figure 3c). Identical MD
2
simulations of SCNP- were performed but with alkyne substrates
present; the average number of molecules bound during the last 10
ACS Paragon Plus Environment
3

Journal of the American Chemical Society
Page 4 of 9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. CuAAC reactions on proteins and simulation between CuI−SCNP-2 and BSA protein. (a) Initial rates for the modification of 2 μM of
different proteins (for Lz-Al 5 μM) with CuI−SCNP-
2
(2 μM) or BTTAA (20 μM), of
pH = 7.4) at room temperature. The error bars stand for the standard deviation of three independent experiments. Zeta potential of proteins in PBS
buffer (1x, pH = 7.4). (b) MD simulation between a single SCNP- and BSA protein in PBS buffer (1x, pH = 7.4), and free energy was calculated
2
(40 μM) and sodium ascorbate (2 mM) in PBS buffer (1x,
2
from center of mass distance from 2.8 nm to 8.8 nm. The carboxylate groups on BSA protein were colored blue and the other parts were colored
yellow. Snapshots were shown at 3.1 nm, 4.2 nm and 7.4 nm with a blow-up image at 4.2 nm showing the ion pairs.
I
proceeded only 2-3 times faster with Cu −SCNP- and for Av-Al and
2
the alkynes were added as biotinylated substrates (see Supporting
Information for the structure).
Lz-Al, the rates with CuI−BTTAA and CuI−BTGTA were faster.
I
The superior bioconjugation performance of Cu −SCNP- with
2
All eight alkynylated proteins were clicked to coumarin azide in
PBS buffer, comparing the rate for catalysts Cu −SCNP- ,
2
I
2
the majority of the proteins was unexpected and suggested selective
complexation of the proteins by the nanoparticle. This model
involves a surface electrostatic complementarity, with accessible
copper centers in the polymer able to couple the protein alkyne
CuI−BTTAA and CuI−BTGTA at the same copper concentration
(Figure 4a). An increase in fluorescence indicates both the rate and
2
extent of coumarin azide conjugation to the protein alkyne groups.
2
In the case of BSA-Al, the reaction was further monitored by matrix-
assisted laser desorption ionization (MALDI) mass spectrometry.
Spectra taken before and after the CuAAC reaction of BSA-Al
indicated nearly complete reaction of the alkyne groups (Figure
groups to coumarin azide . Consistent with this model and the
observed selectivity, the zeta potentials of the five highly reactive
proteins (BSA-Al, HAS-Al, mCh-Al, PA-Al and TF-Al) ranged from
-19 to -25 mV indicating anionic charged surfaces (Figure 4a). In
contrast, StAv-Al had a nearly neutral surface, and Av-Al and Lz-Al
exhibited a polycationic surface. The measured zeta potentials are
consistent with the reported isoelectric points (PI) of the proteins.
We prepared BTGTA to evaluate the role of a single cationic
guanidinium group on the catalysis, given its known ability to bind
proteins. CuI−BTGTA performs only slightly faster than
CuI−BTTAA suggesting the importance of multivalent binding by
S13). For five of the proteins (BSA-Al, HAS-Al, PA-Al, mCh-Al and
I
2
TF-Al), the Cu −SCNP- catalyzed reaction was very fast with the
increase in fluorescence plateauing at ca. 10 min (Figure S12). More
importantly, the reaction rates were 10- to 26-times higher than for
I
Cu −BTTAA and Cu −BTGTA (Figure 4a). However, StAv Al
I
-
ACS Paragon Plus Environment
4

Page 5 of 9
Journal of the American Chemical Society
the hydrophobic regions. At distances >7 nm, the nanoparticle
became fully detached from the protein and the free energy reached
1
2
2
a plateau. Despite the ten cross-links, the shape of SCNP- was not
3
4
5
6
7
8
9
static, changing from globular when uncomplexed (CMD >7 nm) to
an elongated shape that maximized the intermolecular contacts
when bound to BSA (CMD = 3.1 nm). As a control, MD simulations
were performed with avidin, a positively charged protein, and a
significantly lower binding energy (-31 kcal/mol) was calculated
(Figure S10).
The results from the protein conjugation studies and MD
simulations prompted us to obtain experimental evidence for the
SCNP-protein interaction. We previously studied the encapsulation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
118
of small molecules within SCNP- by nuclear magnetic resonance
saturation transfer difference (NMR STD) spectroscopy,37 which
uses the intermolecular nuclear Overhauser effect (NOE). The same
technique was used here to examine the interaction between BSA
2
and SCNP- . The STD signals for peaks corresponding to the
trimethyl ammonium groups and the hydrophobic alkyl chains on
2
SCNP- were determined. As shown in Figure 5a and 5b, the STD
signal for the alkyl side-chain and trimethylammonium groups of
2
SCNP- increased with increasing saturation time indicating its
interaction with the protein, consistent with the MD simulations.
Fluorescence polarization experiments were conducted to
estimate the strength of the SCNP-protein interaction. Thus, the
polarization of fluorescein labelled BSA was measured with
2
increasing concentration of SCNP- . The apparent Kd calculated
2
was 26 μM for SCNP- (Figure 5c), which is higher than the working
concentration for the catalysis (2 μM). Only a very small degree of
fluorescence polarization was observed at 2 μM suggesting a
relatively weak and reversible SCNP-protein interaction under
Figure 5. Experimental evidence of SCNP-protein interaction. (a) STD
signals of CuI−SCNP-
(100 μM) with different irradiation time on
BSA (100 μM) at 7.0 ppm in deuterium PBS buffer (1x, pD = 7.4). (b)
Percentage of STD signals of the trimethyl ammonium structure and
hydrophobic chains on CuI−SCNP- (100 μM) or CuI−SCNP-3 (100
2
μM) with BSA (100 μM) protein in deuterium PBS buffer (1x, pD =
7.4) irradiated at 7.0 ppm. (c) Fluorescence polarization of fluorescein
labelled BSA (2 μM) with different concentration of CuI−SCNP-
2
in
2
PBS buffer (1x, pH = 7.4). The error bars represent the standard
deviation of three independent experiments.
I
2 36
surface complexation, which we refer to as “attach mode,” MD
Cu −SCNP- . To obtain a more detailed picture of the multivalent
Figure 6. Reactions on cell surface glycans. (a) Illustration of metabolic
labelling of cell surface glycan by using a mannose derivative with an
azido or alkyne group, and CuI−SCNP-2 mediated reactions on cell
membrane. (b) and (c) Confocal images of alkyne-labelled H460 cell
treated with CuI−SCNP-2 (1 μM) (b) or CuI−BTGTA (10 μM) (c), 2
(100 μM) and sodium ascorbate (2 mM) in PBS buffer (1x, pH = 7.4)
for 30 min. (d) and (e) Confocal images of azide-labelled H460 cell
treated with CuI−SCNP-2 (1 μM) (d) or CuI−BTGTA (10 μM) (e),
mCh-Al (1 μM) and sodium ascorbate (2 mM) in PBS buffer (1x, pH
= 7.4) for 30 min. Nucleus were stained with Hoechst for (d) and (e).
Scale bar = 20 μm.
2
simulations were conducted with a single SCNP- and BSA. A non-
equilibrium pulling calculation was performed over the center of
mass distance (CMD) between the two macromolecules from 2.8
nm to 8.8 nm to provide a semi-quantitative estimate of their
association free energy (Figure 4b). It was found that at a CMD of
ca. 3.1 nm, the system showed the lowest free energy at -122
2
kcal/mol. At this distance, SCNP- was attached to BSA protein by
a combination of apolar contacts and electrostatic interactions.
When the CMD was increased from 3.1 to 8.8 nm, the free energy
of the system gradually increased as the hydrophobic regions of
2
SCNP- detached from BSA protein (Movie S1). As shown in the
inset in Figure 4b, several trimethylammonium-carboxylate ion pairs
were found during this stage. This observation suggested that the
SCNP-protein interaction may initially be facilitated by electrostatic
interactions and further enhanced by subsequent contacts between
catalytic conditions, which is further supported by Förster resonance
energy transfer (FRET) experiments (Figure S16). The weak
binding did not result in aggregation as detected by DLS (Figure S15)
ACS Paragon Plus Environment
5

Journal of the American Chemical Society
Page 6 of 9
2
click reaction of 3-azido-7-hydroxy-coumarin ( ) forms compound
with restored fluorescence. The intensity was monitored using a
nor any detectable change in protein structure as seen by the circular
dichroism (CD) spectrum (Figure S17).
CuAAC Click Reaction on Cell Surfaces.
Cell surface glycans are
1
2
3
4
5
6
7
8
4
fluorimeter in kinetics mode, measuring the fluorescence intensity
every 10 s at λem = 488 nm with excitation at λex = 410 nm. The initial
rate was determined using the following procedure: Approximately
20-30 s after the start of the reaction, where the fluorescence signals
start to increase linearly over time, the slope of this linear part (5-15
data points) was calculated in counts per second (CPS) increase per
essential for cellular recognition, receptor signaling, and intracellular
trafficking.38 Given the high CuAAC activity observed with
I
2
Cu −SCNP- and its ability to function effectively in bioconjugation,
we examined the potential to perform catalysis on cell surface
glycans. Thus, H460 cells were incubated with Ac4ManNAl or
Ac4ManNAz separately to metabolically incorporate alkyne or azido
groups into the cell surface glycans.39 The alkyne-labelled cells were
2
3a
minute. For the reaction between and , the slope was calculated
into the initial reaction rate in “μM/min” from the observed
9
I
2
2
4
treated with coumarin azide and either Cu −SCNP- (1 μM) or
fluorescence intensity using pure as the standard. The initial
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CuI−BTGTA (10 μM). An even more demanding labeling was
examined by using the same two catalysts but treating the azide-
labeled cells with alkynylated mCherry (mCh-Al). Cu −SCNP-
again proved significantly more effective than CuI−BTGTA giving a
reaction rates for the other 13 alkyne substrates were calculated
based on the assumption that the reactions reach 100% conversion,
and the plateaued fluorescent signal corresponds to 20 μM of the
product. The fluorogenic CuAAC click reactions on protein were
performed using a similar procedure with the initial rate presented
directly as “CPS/min.” For enzyme-like kinetics fitting experiment,
I
2
2
strongly labeled cell membrane with coumarin azide (Figure 6b)
and mCh-Al (Figure 6d). Because the cell membrane and mCh-Al
have a highly negative surface charge the cationic nanoparticle is able
to bring them in proximity40 allowing the copper centers to perform
the CuAAC coupling. Indeed, this strategy significantly reduced the
required amount of cytotoxic copper ([CuI] = 10 μM) so that no
toxicity was observed (Figure S20).
2
the initial reaction rates of the reactions between (10, 20, 40, 60
I
μM) and (10, 40, 70, 150, 200 μM) with catalysis by Cu -SCNP-
3a
(0.5 μM) were measured. The initial reaction rates were fit to a
2
random sequential two-substrates enzyme kinetics equation:
[ ][ ]
ꢁ
[ ]
A B
ꢂꢃꢄ
ꢀ =
◼ CONCLUSION
ꢇ
ꢆ
ꢈꢇ[ ] [ ][ ]
ꢅ ꢅꢇꢈ + ꢅꢇꢈ A + ꢅ B + A
B
Our results indicate that combining transition metal catalytic
centers and polymeric scaffolds with substrate binding ability serves
Protein functionalization with alkyne groups.
In a 1.5-mL
I
as a powerful method to increase catalyst efficiency. Cu −SCNP-
2
centrifuge tube, protein (BSA, HSA, mCherry, protein A, or
transferrin) was dissolved in PBS buffer (1x, pH = 7.4, 1 mL) at a
concentration of 100 μM. Then, 2,5-dioxopyrrolidin-1-yl pent-4-
ynoate (0.2 mg) in DMSO (4 μL) was added to the protein solution.
The mixture was gently shaken for 6 h at room temperature and
small organic compounds were removed using an Amicon tube with
10 kDa cut-off. For lysozyme, due to the smaller molecular weight, it
was dissolved in PBS buffer (1x, pH = 7.4, 1mL) at the concentration
of 550 μM and 2,5-dioxopyrrolidin-1-yl pent-4-ynoate (0.2 mg) in
DMSO (4 μL) was added. The mixture was gently shaken for 6 h at
room temperature and purified by an Amicon tube with 3 kDa cutoff.
The molecular weight of the protein before and after
functionalization were determined using matrix-assisted laser
desorption/ionization (MALDI) mass spectrometry and the
average number of alkynes were determined by the molecular weight
increases. Functionalized proteins are referred to as BSA-Al, HSA-
Al, mCh-Al, PA-Al, TF-Al and Lz-Al. For functionalizing
streptavidin and avidin with three alkyne groups, each protein was
dissolved in PBS buffer (1x, pH = 7.4, 1 mL) at the concentration of
100 μM. Biotin-PEG4-alkyne was added to reach the concentration
of 300 μM (3 equivalents). The mixture was gently shaken for 10
min and used without further purification. The noncovalently
functionalized streptavidin and avidin are referred to as StAv-Al and
functions analogously to enzymes by binding substrates in proximity
to the metal center, thereby dramatically accelerating reaction rates
at low substrate concentrations. The single-chain nanoparticle was
also found to perform bioconjugation reactions to a range of protein
substrates. We propose that these SCNPs function as
supramolecular catalysts in two distinct ways. The first involves
encapsulation involving an uptake mode for small molecules. The
second was unexpected, an attach mode operating on protein
substrates. In this mode a multivalent surface binding process
overcomes what would otherwise be a sterically demanding reaction.
Thus, the semi-flexible synthetic polymer adapts to the surface of the
biomacromolecule and catalyzes the CuAAC reaction. The two
proposed modes by which the copper SCNP functions represent
limiting cases for small molecules and proteins and there will be a
spectrum of mechanisms depending on the specific substrate.
The demonstration of these SCNPs as highly efficient catalysts for
the “native” CuAAC reaction, suggest their utility in bioconjugation
reactions. More broadly, the polymeric structure of SCNP loosely
mimics the protein scaffold of metalloenzymes. Thus, the folded
amphiphilic polymer selectively and reversibly binds substrates in
aqueous buffer much like enzymes. Although well-defined binding
pocket are lacking, the polymer increases the local concentration of
substrate and the chance of collision with the metal site. As a result,
enzyme-like saturation kinetics are observed. Obvious extensions to
other catalytic reactions and even multi-step synthetic transform-
ations are easy to envision and a current focus in our laboratory.
Av-Al.
II
Cu -SCNP- was dissolved in D2O at
2D-NOESY experiments.
2
the concentration of 100 μM and the small molecules were added as
a 200 mM DMSO-d6 stock solution to reach a final concentration of
3 mM (30 equivalents). 2D-NOESY spectra were collected using a
water suppression 2D-NOESY method on a Bruker CB500. Spectra
were processed with base line correction and phase correction by
◼ EXPERIMENTAL SECTION
Fluorogenic reaction with small molecules
. The fluorogenic
3a
using MestReNova (v. 8.1). Substrate showed relatively strong
intermolecular NOE signals between its aromatic signals and the
CuAAC click reactions were performed in a 0.7-mL fluorimeter
cuvette. Catalyst ligand and CuSO4 in a 1:1 ratio and NaAsc were
dissolved in 0.5 mL of PBS buffer (1x, pH = 7.4) in the cuvette.
DMSO stock solutions of substrates were added to give a final
concentration of [NaAsc] = 2 mM and 2% (v/v) of DMSO. The
2
SCNP- alkyl signals at 1.2 ppm, and little cross-peak NOE signals
were observed from the other protons on SCNP- . Due to signal
2
5
overlap between the charged alkyne substrates (group and ) and
SCNP- , the intermolecular NOE signals between SCNP- with
7
2
2
ACS Paragon Plus Environment
6

Page 7 of 9
Journal of the American Chemical Society
charged molecule were studied with sodium benzenesulfonate and
polymers, and additional kinetic data along with details of the
computational methods (PDF)
1
2
3
benzyl trimethylammonium chloride. Using the above condition,
the intermolecular NOE signals were observed from both alkyl
region (1.2 ppm) and trimethyl ammonium (3.0 ppm) regions on
AUTHOR INFORMATION
4
5
6
7
8
9
2
SCNP- to benzenesulfonate, suggesting both electrostatic
Corresponding Author
interactions and hydrophobic interactions are involved in binding
anionic small molecules. Almost no NOE cross-peaks were observed
between benzyl trimethylammonium chloride and SCNP- ,
suggesting weak binding of cationic molecules.
* E-mail: sczimmer@illinois.edu.
2
ORCID
Junfeng Chen: 0000-0002-7100-0839
Jiang Wang: 0000-0002-3871-598X
Ke Li: 0000-0003-3851-7710
Yuhan Wang: 0000-0001-6807-3968
Martin Gruebele: 0000-0001-9291-8123
Andrew L. Ferguson: 0000-0002-8829-9726
Steven C. Zimmerman: 0000-0002-5333-3437
Saturation Transfer Difference (STD) experiment
. BSA with
II
II
2
3
Cu -SCNP- or Cu -SCNP- were dissolved in deuterium oxide
PBS buffer (1x, pD = 7.4) to reach a concentration of 100 μM for
both protein and nanoparticle. STD spectra were collected using a
water suppression STD method on a VNS750 spectrometer with the
bio-pack software. During the saturation period, the aromatic region
of BSA was irradiated at 7 ppm or -0.5 ppm, with the irradiation time
ranging from 0.5 s to 5 s. To minimize intramolecular signals from
BSA, a 15 ms relaxation T2 filter was applied during data acquisition.
Spectra were processed by MestReNova (v. 8.1), and the STD effect
intensity was calculated for the trimethyl ammonium peak at 3.0
ppm and alkyl chain peak at 1.2 ppm through the equation: STD =
(I0 – Isat)/I0.
Fluorescence Polarization.
dissolved in PBS buffer (1x, pH = 7.4) at the concentration of 2 μM
II
with the concentration of Cu -SCNP- ranging from 0 to 100 μM.
The solutions were transferred to a black 384-well plate, and 50 μL
of the solution was added to each well. The fluorescence polarization
of solutions in each well was measured on an Analyst HT plate
reader with the setup on fluorescein. The data was processed and fit
using OriginPro2017.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
This work was supported by the National Science Foundation
(NSF CHE-1709718) and the National Institutes of Health
(R01AR058361).
Fluorescein labelled BSA protein was
REFERENCES
(1) Bai, Y.; Chen, J.; Zimmerman, S.C. Designed transition metal
catalysts for intracellular organic synthesis. Chem. Soc. Rev.
47, 1811-1821.
(2) Yang, M.; Li, J.; Chen, P. R. Transition metal-mediated
bioorthogonal protein chemistry in living cells. Chem. Soc. Rev.
2
2018
,
2014
(3) Martínez-Calvo, M.; Mascareñas, J. L. Organometallic catalysis
, 43, 6511-6526.
Cell study.
In a 6-wells plate, 30000 of H460 cells were added in 3
mL of DMEM media (10% FBS) containing Ac4ManNAz (25 μM)
or Ac4ManNAl (25 μM) to each well with a glass slide. The cells were
incubated at 37 °C with 5% CO2 for 72 h for metabolic incorporation.
Cell media was removed, and each well was washed three times with
3 mL of PBS buffer (1x, pH = 7.4). DMEM media (10% FBS)
without Man-Al (3 mL) was added to each well and incubated for 1
h to deplete free Ac4ManNAz or Ac4ManNAl. Cell media was
removed and each well was washed three times with 3 mL of PBS
buffer (1x, pH = 7.4). For each well, 5 mL of PBS buffer (1x, pH =
2018
, 359,
in biological media and living settings. Coord. Chem. Rev.
57-79.
(4) Vinogradova, E. V.; Zhang, C.; Spokoyny, A. M.; Pentelute, B. L.;
Buchwald, S. L. Organometallic palladium reagents for cysteine
2015
, 526, 687-691.
bioconjugation. Nature
(5) Jiang, H.; López‐Aguilar, A.; Meng, L.; Gao, Z.; Liu, Y.; Tian, X.;
Yu, G.; Ovryn, B.; Moremen, K. W.; Wu, P. Modulating cell‐surface
receptor signaling and ion channel functions by in situ glycan editing.
2018
Angew. Chem. Int. Ed.
, 57, 967-971.
I
I
2
7.4) with/without Cu -SCNP- (1 μM), Cu -BTGTA (10 μM) or
Cu -BTTAA (10 μM), (100 μM) or mCh-Al (1 μM) and sodium
(6) Coverdale, J. P. C.; Romero-Canelón, I.; Scanchez-Cano, C.;
Clarkson, G. J.; Habtemariam, A.; Wills, M.; Sadler, P. J. Asymmetric
transfer hydrogenation by synthetic catalysts in cancer cells. Nat.
I
2
ascorbate (2 mM) was added. The cells were incubated at 37 °C with
5% CO2 for 0.5 h for the CuAAC reaction. The solutions were
removed and each well was washed three times with 3 mL of PBS
buffer (1x, pH = 7.4). For the cellular click reactions performed with
mCh-Al, the cell nucleus was stained with Hoechst 33342 (8 μM in
PBS buffer) for 2-3 min and subsequently washed three times with 3
mL of PBS buffer. The samples were imaged using a Leica SP8
UV/Visible Laser Confocal Microscope (excitation = 405 nm and
561 nm, blue channel: emission = 450-500 nm on the PMTs
detector, red channel: 600-650 nm on the high-sensitivity GaAsP
HyD detectors).
2018
Chem.
(7) Rebelein, J. G.; Ward, T. R. In vivo catalyzed new-to-nature
2018
, 10, 347-354.
reactions. Curr. Opin. Biotechnol.
, 53, 106-114.
(8) Yu, F.; Cangelosi, V. M.; Zastrow, M. L.; Tegoni, M.; Plegaria, J.
S.; Tebo, A. G.; Mocny, C. S.; Ruckthong, L.; Qayyum, H.; Pecoraro,
V. L. Protein design: toward functional metalloenzymes. Chem. Rev.
2014
(9) Cole, J. P.; Hanlon, A. M.; Rodriguez, K. J.; Berda, E. B. Protein‐
, 114, 3495-3578.
like structure and activity in synthetic polymers. J. Polym. Sci. Part
2017
A: Polym. Chem.
, 55, 191-206.
(10) Rothfuss, H.; Knofel, N. D.; Roesky, P. W.; Barner-Kowollik, C.
Single-chain nanoparticles as catalytic nanoreactors. J. Am. Chem.
ASSOCIATED CONTENT
Supporting Information
2018
Soc.
, 140, 5875-5881.
.
(11) Mavila, S.; Eivgi, O.; Berkovich, I.; Lemcoff, N. G.
Intramolecular cross-linking methodologies for the synthesis of
polymer nanoparticles. Chem. Rev.
(12) Rubio-Cervilla, J.; González, E.; Pomposo, J. A. Advances in
single-chain nanoparticles for catalysis applications. Nanomaterials
The Supporting Information is available free of charge on the ACS
Publications website at DOI:
2016
, 116, 878-961.
General experimental procedures and detailed synthetic
procedures and characterization data for small molecules and
2017
, 7, 341-360.
ACS Paragon Plus Environment
7

Journal of the American Chemical Society
Page 8 of 9
(13) Pomposo, J. A., Ed. Single-Chain polymer nanoparticles:
synthesis, characterization, simulations, and applications; John
Wiley & Sons, 2017.
(25) Sivakumar, K.; Xie, F.; Cash, B. M.; Long, S.; Barnhill, H. N.;
Wang, Q. A Fluorogenic 1,3-dipolar cycloaddition reaction of 3-
1
2
2004
azidocoumarins and acetylenes. Org. Lett.
, 6, 4603-4606.
3
4
5
6
7
8
9
(14) Bai, Y.; Xing, H.; Vincil, G. A.; Lee, J.; Henderson, E. J.; Lu, Y.;
Lemcoff, N. G.; Zimmerman, S. C. Practical synthesis of water-
soluble organic nanoparticles with a single reactive group and a
(26) Chan, T. R.; Hilgraf, R.; Sharpless, K. B.; Fokin, V. V.
Polytriazoles as copper(I)-stabilizing ligands in catalysis. Org. Lett.
2004
, 6, 2853-2855.
2014
functional carrier scaffold. Chem. Sci.
, 5, 2862-2868.
(27) Artar, M.; Souren, E. R. J.; Terashima, T.; Meijer, E. W.;
Palmans, A. R. A. Single chain polymeric nanoparticles as selective
(15) Bai, Y.; Xing, H.; Wu, P.; Feng, X.; Hwang K.; Lee, J. M.; Phang,
X. Y.; Lu, Y.; Zimmerman, S. C. Chemical control over cellular
uptake of organic nanoparticles by fine tuning surface functional
2015
hydrophobic reaction spaces in water. ACS Macro Lett.
1099-1103.
, 4,
2015
groups. ACS Nano
, 9, 10227-10236.
(28) Liu, Y.; Turunen, P.; de Waal, B. F. M.; Blank, K. G.; Rowan, A.
E.; Palmans, A. R. A.; Meijer, E. W. Catalytic single-chain polymeric
nanoparticles at work: from ensemble towards single-particle
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(16) Bai, Y.; Feng, X.; Xing, H.; Xu, Y.; Kim, B. K.; Baig, N.; Zhou,
T.; Gewirth, A. A.; Lu, Y.; Oldfield, E.; Zimmerman, S. C. A highly
efficient single-chain metal–organic nanoparticle catalyst for
alkyne–azide “click” reactions in water and in cells. J. Am. Chem. Soc.
2018
, 3, 609-618
kinetics. Mol. Syst. Des. Eng.
(29) Ivanetich, K. M.; Goold, R. D. A rapid equilibrium random
sequential bi-bi mechanism for human placental glutathione S-
transferase. Biochim. Biophys. Acta, Protein Struct. Mol. Enzymol.
2016
, 138, 11077-11080.
(17) Liu, Y.; Pujals, S.; Stals, P. J. M.; Paulöhrl, T.; Presolski, S. I.;
Meijer, E. W.; Albertazzi, L.; Palmans, A. R. A. Catalytically active
single-chain polymeric nanoparticles: exploring their functions in
1989
(30) Antos, J. M.; Francis, M. B. Transition metal catalyzed methods
, 998, 7−13.
2018
, 140, 3423-3433.
2006
complex biological media. J. Am. Chem. Soc.
for site-selective protein modification. Curr. Opin. Chem. Biol.
10, 253-262.
(31) Spicer, C. D.; Davis, B. G. Selective chemical protein
,
(18) Chen, J.; Wang, J.; Bai, Y.; Ke, L.; Garcia, E. S.; Ferguson, A. L.;
Zimmerman, S. C. Enzyme-like click catalysis by a copper-
2018
, 140,
2014
containing single-chain nanoparticle. J. Am. Chem. Soc.
13695-13702.
(19) Huerta, E.; Stals, P. J. M.; Meijer, E. W.; Palmans, A. R. A.
modification. Nat. Commun.
(32) Jbara, M.; Maity, S. K.; Brik, A. Palladium in the chemical
, 5, 4740-4753.
2017
,
synthesis and modification of proteins. Angew. Chem. Int. Ed.
56, 10644-10655.
Consequences of folding a water‐soluble polymer around an
2013
organocatalyst. Angew. Chem. Int. Ed.
, 52, 2906−2910.
(33) Koniev, O.; Wagner, A. Developments and recent
advancements in the field of endogenous amino acid selective bond
(20) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. A
Stepwise Huisgen cycloaddition process: copper(I)‐catalyzed
regioselective “ligation” of azides and terminal alkynes. Angew.
2015
forming reactions for bioconjugation. Chem. Soc. Rev.
5495-5551.
, 44,
2002
Chem.
, 114, 2708-2711.
(34) Boutureira, O.; Bernardes, G. J. L. Advances in chemical
2015
(35) Takaoka, Y.; Ojida, A.; Hamachi, I. Protein organic chemistry
(21) Rodionvo, V. O.; Presolski, S. I.; Gardinier, S.; Lim, Y. H.; Finn,
M. G. Benzimidazole and related ligands for Cu-catalyzed
protein modification. Chem. Rev.
, 115, 2174-2195.
2007
azide−alkyne cycloaddition. J. Am. Chem. Soc.
, 129, 12696-
and applications for labeling and engineering in live‐cell systems.
2013
, 52, 4088-4106.
12704.
Angew. Chem. Int. Ed.
(36) Mogaki, R.; Hashim, P. K.; Okuro, K.; Aida, T. Guanidinium-
based “molecular glues” for modulation of biomolecular functions.
(22) Besanceney-Webler, C.; Jiang, H.; Zheng, T.; Feng, L.; Soriano
del Arno, D.; Wang, W.; Klivansky, L. M.; Marlow, F. L.; Liu, Y.; Wu,
P. Increasing the efficacy of bioorthogonal click reactions for
2017
, 46, 6480-6491.
Chem. Soc. Rev.
2011
bioconjugation: a comparative study. Angew. Chem. Int. Ed.
50, 8051-8056 (2011).
,
(37) Mayer, M.; Meyer, B. Characterization of ligand binding by
saturation transfer difference NMR spectroscopy. Angew. Chem. Int.
1999
, 38, 1784-1788.
(23) Ovryn, B.; Li, J.; Hong, S.; Wu, P. Visualizing glycans on single
cells and tissues–Visualizing glycans on single cells and tissues. Curr.
Ed.
(38) Ohtsubo, K.; Marth, J. D. Glycosylation in cellular mechanisms
2006
2017
Opin. Chem. Biol.
, 39, 39-45.
of health and disease. Cell
, 126, 855-867.
(24) Frisch, H.; Menzel, J. P.; Bloesser, F. R.; Marschner, D. E.;
Mundsinger, K.; Barner-Kowollik, C. Photochemistry in confined
environments for single-chain nanoparticle design. J. Am. Chem.
(39) Prescher, J. A.; Dube, D. H.; Bertozzi, C.R. Chemical
2004
, 430, 873-
remodelling of cell surfaces in living animals, Nature
877.
2018
Soc.
, 140, 9551-9557.
(40) Verma, A.; Stellacci, F. Effect of surface properties on
2010
nanoparticle–cell interactions. Small
, 6, 12-21.
ACS Paragon Plus Environment
8

Page 9 of 9
Journal of the American Chemical Society
Table of contents (TOC) graphic
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
ACS Paragon Plus Environment