Charge-Preserving Atom Transfer Radical Polymerization Initiator Rescues the Lost Function of Negatively Charged Protein-Polymer Conjugates

Article abstract of DOI:10.1021/acs.biomac.9b00379

When grown from the surface of proteins, negatively charged polymers cause irreversible inactivation, thereby limiting the breadth of the synthetic space that negatively charged protein-polymer conjugates can be applied to. More broadly speaking, independent of polymer and synthetic approach, almost all protein-polymer conjugates are less active than their precursors. After more than a decade without major advances in understanding why the attachment of some polymers so sharply deactivates enzymes, we focused our attention on a technique to protect enzymes from the growth of a deactivating polymer by restoring the charge at the protein surface during polymer attachment. We synthesized an amino-reactive positively charged atom transfer radical polymerization initiator that inserted a permanent positive charge at the site of bio-macroinitiator attachment. Preserving the surface charge through attachment of the permanent positively charged initiator led to the first observation of activity of enzymes that were coupled to negatively charged homopolymers.

Full text of DOI:10.1021/acs.biomac.9b00379

Subscriber access provided by Bibliothèque de l'Université Paris-Sud
Article
A charge-maintaining ATRP initiator rescues the lost
function of negatively charged protein-polymer conjugates
Stefanie L Baker, Hironobu Murata, Bibifatima Kaupbayeva,
Adina Tasbolat, Krzysztof Matyjaszewski, and Alan J Russell
Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.9b00379 • Publication Date (Web): 13 May 2019
Downloaded from http://pubs.acs.org on May 14, 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 58
Biomacromolecules
1
2
3
4
5
1
2
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
A charge-maintaining ATRP initiator rescues the lost function of negatively
charged protein-polymer conjugates
5
6
7
8
9
0
1
2
3
4
Stefanie L. Baker,1,2** Hironobu Murata,2** Bibifatima Kaupbayeva , Adina Tasbolat ,
2,3
4
2
,5
*1,2,5,6,7
Krzysztof Matyjaszewski , Alan J. Russell
1
1
1
1
1
1Department of Biomedical Engineering, Scott Hall 4N201, Carnegie Mellon University, 5000
Forbes Avenue, Pittsburgh, PA 15213, United States
15
1
1
1
1
2
6
7
8
9
0
2Center for Polymer-Based Protein Engineering, Carnegie Mellon University, 5000 Forbes
Avenue, Pittsburgh, PA 15213, United States
3Department of Biological Sciences, Carnegie Mellon University, 4400 Fifth Avenue, Pittsburgh,
PA 15213, United States
21
22
2
2
2
2
2
3
4
5
6
7
4Department of Chemistry and Chemical Technology, Al-Farabi Kazakh National University, 71
Al-Farabi Avenue, Almaty, 050040, Republic of Kazakhstan
5Department of Chemistry, Carnegie Mellon University, 4400 Fifth Avenue, Pittsburgh, PA
15213, United States
28
2
3
3
3
3
9
0
1
2
3
6Disruptive Health Technology Institute, Carnegie Mellon University, 5000 Forbes Avenue,
Pittsburgh, PA 15213, United States
7Department of Chemical Engineering, Carnegie Mellon University, 5000 Forbes Avenue,
Pittsburgh, PA 15213, United States
34
35
**These authors contributed equally to this work
3
3
6
7
3
8
39
1
ACS Paragon Plus Environment

Biomacromolecules
Page 2 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
4
0
1
ABSRACT
When grown from the surface of proteins, negatively charged polymers cause irreversible
inactivation, thereby limiting the breadth of the synthetic space that negatively charged protein-
polymer conjugates can be applied to. More broadly speaking, independent of polymer and
synthetic approach, almost all protein-polymer conjugates are less active than their precursors.
After more than a decade without major advances in understanding why the attachment of some
polymers so sharply deactivates enzymes, we focused our attention on a technique to protect
enzymes from the growth of a deactivating polymer by restoring the charge at the protein surface
during polymer attachment. We synthesized an amino-reactive positively-charged ATRP initiator
that inserted a permanent positive charge at the site of bio-macroinitiator attachment.
Maintaining surface charge through attachment of the permanent positively charged initiator led
to the first observation of activity of enzymes that were coupled to negatively charged
homopolymers.
42
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
4
4
4
4
4
3
4
5
6
7
8
49
5
5
0
1
52
5
5
5
5
5
5
3
4
5
6
7
8
Keywords: charged initiator, ATRP, protein-polymer conjugate, surface charge
59
6
6
0
1
2
ACS Paragon Plus Environment

Page 3 of 58
Biomacromolecules
1
2
3
6
6
2
3
The delicate balance of forces that maintain the structure, function, and dynamics of
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
enzymes is at the heart of their remarkable activity and bothersome instability. Although some
2
64
enzymes have evolved to survive in extreme environments , protein engineers that desire to
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
6
6
6
6
7
5
6
7
8
9
0
stabilize proteins for therapeutic or industrial uses have generally used molecular biology to
dramatically improve function.3–5 Another compelling approach to protein/enzyme stabilization
has been to covalently attach polymers to the protein surface.6–9 Polymers can either be
covalently coupled to the surface of proteins (commonly referred to as PEGylation;
PEG=poly(ethylene glycol)) or grown from the surface of proteins using controlled radical
polymerization from protein-initiator constructs, also known as bio-macroinitiators.10–20 ATRP
has been used to grow dense polymer coatings that “nano-armor” proteins.21,22 A wide variety of
polymers, over a range of molecular weights and densities, have been conjugated with proteins to
71
7
7
2
3
determine their impact on function, including random copolymers , block copolymers24–26,
2
3
thermoresponsive10,13,27 or pH responsive polymers , branched polymers , and charged
polymers.9,30,31 Unfortunately, however, the full diversity of polymers that can be attached to
proteins is inaccessible because some polymers almost instantaneously unfold and inactivate
proteins. In particular, the growth of negatively charged polymers from the surface of enzymes
is not tolerated by proteins.32
28
29
74
7
7
7
7
7
8
5
6
7
8
9
0
Interestingly, even while the conjugation of conventional PEG to a protein increases its
3
3
stability, it is always at the expense of bioactivity. Protecting proteins from conjugated
polymer-induced activity loss is a vexing challenge in the field of next generation protein-
polymer conjugate design.34 Since Jiang’s work in 2011, zwitterionic polymers have become
increasingly attractive for protein conjugation because they improve stability while preserving
81
8
8
2
3
3
5
84
some degree of bioactivity. Zwitterionic monomers have a one-to-one ratio of positively and
3
ACS Paragon Plus Environment

Biomacromolecules
Page 4 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
8
8
5
6
negatively charged groups yielding a net neutral molecule and their polymeric forms were found
to be stabilizing for proteins by mimicking stabilizing kosmotropic anions and chaotropic cations
as defined by the Hofmeister series.35,36 Previous studies by our group have shown the
stabilizing effect of zwitterionic poly(carboxybetaine methacrylate) (pCBMA) along with high
stabilities and activity retention with positively charged polymers, such as poly(quaternary
87
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
8
9
9
9
9
8
9
0
1
2
3
ammonium methacrylate) (pQA).9,32 These polymers contain stabilizing chaotropic cations,
+
NH , aligning with the findings of Jiang. We also found that at low pH, these polymers could
4
3
2
act as powerful intramolecular chaperones to drive protein folding and stabilization. We have
also shown, however, that negatively charged polymers, such as poly(sulfonate methacrylate)
(pSMA), are destabilizing and deactivating.16,32 pSMA is a hydrophilic, negatively charged
polymer containing kosmotropic sulfonate anions, as shown in Scheme 1, which are defined as
stabilizing by the Hofmeister series. Another study used the SMA monomer, copolymerized with
three other monomers of varying hydrophobicities, and found the random heteropolymer to be
94
9
9
5
6
97
23
9
9
8
9
stabilizing due to statistical monomer distribution. The first publication by our group on
grafted-from protein-polymer conjugates used a negatively charged homopolymer, poly(sodium
100
4-styrenesulfonate) to create conjugates with α-chymotrypsin (CT) that also sharply reduced
8
1
1
01
02
enzyme activity. To our knowledge, there are no reported protein-negatively charged
homopolymer conjugates that maintain enzyme stability or activity. Determining why negatively
charged homopolymers are deactivating and destabilizing to proteins may enable us to remediate
the problem and then use that knowledge to determine if more general protein conjugate-induced
activity loss can be reduced when using compatible polymers.
103
1
1
1
04
05
06
In order to design a polymer conjugation protective strategy, we have focused our
attention on how ATRP initiators are coupled to proteins. Our prior work has always focused on
107
4
ACS Paragon Plus Environment

Page 5 of 58
Biomacromolecules
1
2
3
1
1
08
09
the properties of next generation protein-polymer conjugates, but in doing that work, we
routinely used protein macroinitiator constructs as controls. We have observed that although
compatible polymers can stabilize enzymes, their biomacromolecular initiator precursors are
often less active and stable than the native protein or the conjugate.32 Protein-initiator constructs
are most often formed by reacting accessible surface amino groups with activated ester alkyl
halides. These reactions, and almost all common PEGylation coupling chemistries, sacrifice the
native electrostatic environment of the protein surface for the supposed benefit of the resulting
protein-polymer conjugate (Figure S1 and Table S1). The prevailing view of protein scientists
has been that maintaining protein surface charge-charge interactions is less important to protein
stability than maintaining the integrity of the hydrophobic core. Indeed, hydrophobic
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
110
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
1
1
1
1
11
12
13
14
15
16
117
-1
1
1
18
19
interactions within a protein contribute hundreds of kJ mol to maintaining a folded
conformation, whereas exposed surface charge-charge interactions only contribute a few kJ mol-
1.37–40 Surprisingly, however, rationally optimizing charge-charge interactions can be an effective
tool when designing proteins with increased stability.41–47 These charge-charge interactions can
either be optimized by increasing favorable electrostatic interactions or by decreasing the number
120
1
1
1
1
1
1
21
22
23
24
25
26
4
8
of unfavorable electrostatic interactions. It has also been observed that long-range electrostatic
4
7
interactions are just as important as short-range salt bridge interactions. Conversely, others
have questioned whether charge-charge interactions are important influencers of stability.49
Commercially, of the 14 FDA-approved therapeutic protein-polymer conjugates, Neulasta and
Plegridy are the only two examples where the nitrogen is still positively charged after PEG
attachment by utilizing aldehyde attachment chemistries that result in secondary amines (Figure
S1 and Table S1). Two natural questions arise: could we protect an enzyme from the growth of
an incompatible polymer by initiating the reaction from a charge-maintaining initiator? And, if
127
1
1
28
29
130
5
ACS Paragon Plus Environment

Biomacromolecules
Page 6 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
1
31
32
so, could that same strategy be used to further enhance the activity and stability of compatible
polymer-protein conjugates?
133
Modeled after successfully commercialized PEGylation reagents, alkyl halide ATRP-
initiators usually possess N-hydroxysuccinimide (NHS) groups that react with protein primary
amines, including the N-terminal and lysine residues. We have developed algorithms that can
predict the site of reaction on the protein surface as well as the initiation reaction rates at each
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
1
1
1
1
34
35
36
37
38
39
5
0
amino group. The amino group on the N-terminus has a pK in the range of 7.8-8.0 while the
a
pK ’s of lysine side chain amino groups range from approximately 10.5-12.0, depending on their
a
5
1
local environment. Therefore, at biologically relevant pH values (6-8), all accessible amino
groups will be positively charged almost all of the time (>95%). These positive charges are
unfortunately lost when neutral initiators are used to launch protein-ATRP as the nitrogens are
converted to amide bonds.52,53
140
1
1
41
42
143
We hypothesized that a positively charged ATRP-initiator would restore the positive
charges on an enzyme surface, thereby enhancing activity and stability of the enzyme-initiator
constructs and the protein-polymer conjugates derived from it. We describe the synthesis of the
first positively charged amine-reactive ATRP-initiator. We used this initiator to study the
growth of an incompatible polymer (pSMA) from the surfaces of α-chymotrypsin (CT), uricase,
acetylcholinesterase, lysozyme, and avidin. We now understand that the charge of an ATRP-
initiator is a crucial parameter to consider in the design of highly active and stable protein-
polymer conjugate variants and that negatively charged polymers can now lead to conjugates
with high activities and stabilities.
1
1
1
1
1
44
45
46
47
48
149
1
1
1
50
51
52
MATERIALS AND METHODS
6
ACS Paragon Plus Environment

Page 7 of 58
Biomacromolecules
1
2
3
1
1
53
54
Materials α-chymotrypsin (CT) from bovine pancreas (type II), acetylcholinesterase (AChE)
from Electrophorus electricus (electric eel, type VI-S), uricase from Candida sp., lysozyme from
chicken egg white, biotin, and HABA 4'-hydroxyazobenzene-2-carboxylic acid were purchased
from Sigma Aldrich (St. Louis, MO). Avidin from egg white was purchased from Lee
Biosolutions (Maryland Heights, MO). CT, lysozyme, and avidin were used as received. AChE
and uricase were dialyzed in 25 mM sodium phosphate (pH 7.0) using a 25 kDa molecular
weight cutoff dialysis tube in a refrigerator for 24 h and were then lyophilized. Copper (II)
chloride, sodium ascorbate, 1,1,4,7,10,10-hexamethyltriethylenetetramine (HMTETA), 2-bromo-
2-methylpropionyl bromide, 3-(dimethylamino)-1-propylamine, N,N’-diisopropylcarbodimine,
fluorescamine, acetylthiocholine iodide, and 5,5’-dithiobis(2-nitrobenzoic acid) (DTNB) were
purchased from Sigma Aldrich. 4-bromobutyric acid and N-hydroxysuccinimide were purchased
from TCI USA (Portland, OR). Micro BCA assay kit was purchased from Thermo Fisher
Scientific.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
155
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
1
1
1
1
56
57
58
59
60
61
162
1
1
63
64
165
166
Instrumentation and Sample Analysis Preparations
1
13
TM
1
1
67
68
H and C NMR were recorded on a spectrometer (500 MHz, 125 MHz, Bruker Avance 500)
in the NMR facility located in the Center for Molecular Analysis, Carnegie Mellon University,
169
Pittsburgh, PA, with deuterium oxide (D O) and DMSO-d . Routine FT-IR spectra were
2
6
1
1
70
71
obtained with a Nicolet Avatar 560 FT-IR spectrometer (Thermo) in the Department of Chemical
Engineering at Carnegie Mellon University. Ultraviolet-visible (UV−VIS) spectra were obtained
and used for enzyme activity determination using an UV−VIS spectrometer (Lambda 45,
PerkinElmer) with a temperature-controlled cell holder. Melting points (mp) were measured with
a Laboratory Devices Mel-Temp. Number and weight average molecular weights (M and M )
172
1
1
1
73
74
75
n
w
and the polydispersity index (M /M ) were estimated by gel permeation chromatography (GPC)
w
n
7
ACS Paragon Plus Environment

Biomacromolecules
Page 8 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
1
76
77
on a Water 2695 Series with a data processor, equipped with three columns (Waters
Ultrahydrogel Linier, 500 and 250), using Dulbecco’s Phosphate Buffered Saline with 0.02 wt%
sodium azide for pCBMA and 80 vol% of 100 mM sodium phosphate (pH 9.0) and 20 vol% of
acetonitrile for pSMA as an eluent at flow rate of 1.0 mL/min, with detection by a refractive
index (RI) detector. Pullulan standards (PSS-Polymer Standards Service – USA Inc, Amherst,
MA) were used for calibration. Matrix-Assisted Laser Desorption Ionization Time-of-Flight
Spectrometry (MALDI-ToF MS) was performed with a Perseptive Biosystems Voyager Elite
MALDI-ToF spectrometer in the Center for Molecular Analysis, Carnegie Mellon University.
Dynamic Light Scattering (DLS) data were collected on a Malvern Zetasizer nano-ZS, which
was located in the Department of Chemistry, Carnegie Mellon University. The concentration of
the sample solution was kept at 0.2 - 1.0 mg/mL. The hydrodynamic diameters of samples were
measured three times (15 runs for each measurement).
178
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
1
1
1
1
79
80
81
82
83
84
185
1
1
86
87
188
1
1
89
90
Positive initiator synthesis and characterization
4-bromobutyryl-N-oxysuccinimide ester synthesis
1
1
1
1
1
91
92
93
94
95
N,N’-diisopropylcarbodimine (8.5 mL, 55 mmol) was slowly added to the solution of 4-
bromobutyric acid (8.4 g, 50 mmol) and N-hydroxysuccinimide (4.3 g, 55 mmol) in
dichloromethane (100 mL) at 0 °C, and then the mixture was stirred at room temperature
overnight. Precipitated urea was filtered out and the filtrate was evaporated to remove the
solvent. 4-bromobutyryl-N-oxysuccinimine ester was isolated by recrystallization in 2-propanol;
1
196
yield 10.2 g (77 %), mp 49 – 52 °C. H NMR (500 MHz, DMSO-d ) δ 2.16 (m, 2 H,
6
1
1
97
98
C=OCH CH CH Br), 2.81 (s, 4 H, succinimide), 2.83 (t, 2 H, J = 7.0 Hz, C=OCH CH CH Br),
2
2
2
2
2
2
1
3
3.60 (t, 2 H, J = 7.0 Hz, C=OCH CH CH Br) ppm; C NMR (125 MHz, DMSO-d ) δ 25.9, 27.9,
2
2
2
6
8
ACS Paragon Plus Environment

Page 9 of 58
Biomacromolecules
1
2
3
1
2
99
00
29.5, 33.3, 168.7, 170.6 ppm; IR (KBr pellete) 3017, 2948, 2914, 2852, 1812, 1786, 1731, 1382,
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
-1
1360, 1311, 1202 and 1150 cm .
2
01
N-(3-N’,N’-Dimethylaminopropyl)-2-bromoisobutyramide synthesis
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
202
2-bromo-2-methylpropionyl bromide (3.4 mL, 27 mmol) was slowly added into the solution of
3-(dimethylamino)-1-propylamine (3.1 mL, 24.6 mmol) in deionized water (50 mL) at 0 °C, and
then the mixture was stirred at room temperature for 1 h. After the mixture was adjusted to pH
10 with 5 N NaOH aq. at 0 °C, the product was extracted with ethyl acetate (50 mL x 3). The
organic phase was washed with 20 wt% potassium carbonate aq. (50 mL × 3) and saturated NaCl
aq. (50mL × 2). The organic phase was dried with Na CO and evaporated to remove the solvent.
2
2
2
2
2
2
03
04
05
06
07
08
2
3
N-(3-N’,N’-Dimethylaminopropyl)-2-bromoisobutyramide was isolated in vacuo. oil compound;
1
209
yield 5.9 g (95 %), H NMR (500 MHz, DMSO-d ) δ 1.56 (m, 2 H,
6
2
2
10
11
(CH ) NCH CH CH NHC=O), 1.84 (s, 6 H, C=OC(CH ) Br), 2.11 (s, 6 H, (CH ) NCH CH ),
3
2
2
2
2
3 2
3 2
2
2
2.22 (t, 2 H, J = 7.0 Hz, (CH ) NCH CH ), 3.12 (td, 2 H, J = 5.5 Hz and J = 7.0 Hz,
3
2
2
2
1
3
212
CH CH NHC=O), 8.20 (broad t, 1 H, J = 5.5 Hz, amide) ppm; C NMR (125 MHz, DMSO-d )
2 2 6
2
2
13
14
δ 26.7, 31.6, 38.7, 45.5, 57.4, 61.5, 170.9 ppm; IR (NaCl plate) 3349, 2975, 2946, 2864, 2822,
-1
2800, 1661, 1537, 1465, 1370, 1294, 1263, 1195, 1161 and 1113 cm .
2
2
15
16
Positively charged ATRP initiator synthesis
N-(3-N’,N’-Dimethylaminopropyl)-2-bromoisobutyramide (1.9 g, 7.5 mmol) and 4-
bromobutyryl-N-oxysuccinimide ester (2.0 g, 7.5 mmol) were added in dried acetonitrile (50
mL) and bubbled with nitrogen gas for 10 min. The mixture was sealed and stirred at 40 °C
overnight. The positively charged ATRP initiator (1) was precipitated in a mixture of ethyl
217
2
2
18
19
220
acetate and diethyl ether (1:1 volume ratio), and the oil compound was isolated in vacuo; yield
9
ACS Paragon Plus Environment

Biomacromolecules
Page 10 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2
2
21
22
3.6 g (93 %) (Figure S2), H NMR (500 MHz, D O) δ 1.88 (s, 6 H, C=OC(CH ) Br), 1.91 – 2.24
2 3 2
+
(m, 4 H, C=OCH CH CH N CH CH CH NHC=O), 2.83 – 2.93 (m, 9 H, succinimide and
2
2
2
2
2
2
+
+
223
C=OCH CH CH N (CH )), 3.04 – 3.14 (m, 5 H, C=OCH CH CH N (CH ) ), 3.28 – 3.41 (m, 4
2 2 2 3 2 2 2 3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
+
13
2
2
2
24
25
26
H, N CH CH CH NHC=O) ppm (Figure S3); C NMR (125 MHz, D O) δ22.1, 24.0, 25.6, 30.6,
2 2 2 2
36.6, 42.9, 51.1, 55.4, 61.6, 62.1, 169.1, 173.2, 174.9 ppm (Figure S4); IR (NaCl plate) 3418,
-1
2969, 2708, 1813, 1780, 1734, 1653, 1536, 1472, 1371, 1298, 1210, 1113 and 1074 cm .
227
Electrospray ionization mass spectrometry (ESI-MS)
2
2
28
29
The positive initiator was characterized with ESI-MS using a Finnigan LCQ (Thermo Fisher)
quadrupole field ion trap mass spectrometer with an electrospray ionization source (Figure S5).
Each scan was acquired over 150-2000 m/z using a step of 0.5 u, dwell time of 1.5 ms, mass
defect of 50 pu, and 80 V orifice potential. The positive initiator was dissolved in methanol at
100 µM and eluted using methanol at a flow rate of 15 µL/min.
230
2
2
2
31
32
33
Protein-initiator synthesis and characterization
234
Chymotrypsin (CT): 200 µL of NHS-functionalized ATRP initiator solution in DMSO (168
µmol, 56 mg for neutral and 87 mg for positive initiator, respectively) was added to a CT
solution (60 mg in 30 mL of 100 mM sodium phosphate buffer (pH 8.0)) and stirred at 4 °C for 2
h. CT-initiator conjugates were dialyzed in deionized water using a 15 kDa molecular weight
cutoff dialysis tube in a refrigerator for 24 h and were then lyophilized.
2
2
35
36
237
2
2
38
39
Lysozyme: 200 µL of NHS-functionalized ATRP initiator solution in DMSO (172 µmol, 58 mg
for neutral and 89 mg for positive initiator, respectively) was added to a lysozyme solution (70
mg in 30 mL of 100 mM sodium phosphate buffer (pH 8.0)) and stirred at room temperature for
2 h. Lysozyme-initiator conjugates were dialyzed in deionized water using an 8 kDa molecular
weight cutoff dialysis tube in a refrigerator for 24 h and were then lyophilized.
240
2
2
2
41
42
43
10
ACS Paragon Plus Environment

Page 11 of 58
Biomacromolecules
1
2
3
2
2
44
45
Avidin: Neutral initiator (21 mg, 0.063 mmol) and avidin (20 mg, 0.0013 mmol protein, 0.013
mmol primary amine groups) were dissolved in 4 mL of 0.1 M sodium phosphate buffer (pH 8).
The positively charged ATRP initiator (27 mg, 0.063 mmol) was dissolved in 50 µL of DMSO
and added to a solution of avidin (20 mg, 0.013 mmol primary amine groups) in 4 mL of 0.1 M
sodium phosphate buffer (pH 8). The solutions were stirred at 4°C for 2 h and avidin conjugates
were purified by dialysis using 15 kDa molecular weight cutoff dialysis tube, in 25 mM sodium
phosphate (pH 8), for 24 h at 4°C and were then lyophilized.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
246
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
2
2
2
2
47
48
49
50
51
52
Uricase: 200 µL of NHS-functionalized ATRP initiator solution in DMSO (100 µmol, 34 mg for
neutral and 52 mg for positive initiator, respectively) was added to a uricase solution (20 mg of
uricase in 20 mL of 100 mM sodium phosphate buffer (pH 7.0)) and stirred at room temperature
for 2 h. Uricase-initiator conjugates were dialyzed in 25 mM sodium phosphate (pH 7.0) using a
25 kDa molecular weight cutoff dialysis tube in a refrigerator for 24 h and were then lyophilized.
The concentration of protein in the uricase-initiator conjugate was determined using a Micro
BCA Assay Kit (ThermoFisher Scientific).
253
2
2
54
55
256
2
2
2
2
2
2
57
58
59
60
61
62
Acetylcholinesterase (AChE): 100 µL of NHS-functionalized ATRP initiator solution in DMSO
(12.7 µmol, 4.3 mg for neutral and 5.5 mg for positive initiator, respectively) was added to an
AChE solution (7 mg of AChE in 10 mL of 100 mM sodium phosphate buffer (pH 7.0)) and
stirred at room temperature for 2 h. AChE-initiator conjugates were dialyzed in 25 mM sodium
phosphate buffer (pH 7.0) using a 25 kDa molecular weight cutoff dialysis tube in a refrigerator
for 24 h and were then lyophilized. The concentration of protein in the AChE-initiator conjugate
was determined using a Micro BCA Assay Kit.
263
2
64
2
2
65
66
BCA protein assay
The concentration of protein in the solution was determined using a Micro BCA protein Assay
2
67
Kit (ThermoFisher Scientific). The sample solution (25 L) and Micro BCA working reagent
11
ACS Paragon Plus Environment

Biomacromolecules
Page 12 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
2
68
69
(75 µL) were incubated at 60 °C for 1 h. After incubation, 900 µL of deionized water was added
and the absorbance at 562 nm was recorded by an UV-VIS spectrometer (Lambda 45, Perkin
Elmer). The standard curve was obtained using native protein.
270
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
71
72
Fluorescamine assay
A fluorescamine assay was used to determine the number of bound initiators on the protein
surface. 40 L of sample, 100 mM sodium phosphate buffer (40 µL, pH 8), and fluorescamine
273
2
2
74
75
solution in DMSO (20 L, 3 mg/mL) were added into a 96-well plate and incubated at room
temperature for 15 min. Fluorescence intensities were measured at the excitation of 390 nm and
emission of 470 nm with 10-nm bandwidths by a Safire Spike plate reader at the Molecular
Biosensor and Imaging Center, Carnegie Mellon University. Initiator concentration was
determined using a standard curve using native protein.
276
2
2
77
78
279
Trypsin digestion of protein-initiators Trypsin digestion was performed on protein-initiators to
generate peptide fragments to determine modification sites. Peptide fragments were analyzed
using matrix-assisted laser desorption/ionization time-of-flight (MALDI-ToF) mass
spectrometry. Native CT, CT(N), and CT(+) were digested according to the protocol described in
the In-Solution Tryptic Digestion and Guanidination Kit. 20 µg of protein or protein-initiator
constructs (10 µL of a 2 mg/mL protein solution in deionized water) were added to 15 µL of 50
mM ammonium bicarbonate with 1.5 µL of 100 mM dithiothreitol (DTT) in an Eppendorf tube.
The reaction was incubated for 5 min at 95 °C. 3 µL of 100 mM iodoacetamide aqueous solution
was then added and samples were incubated in the dark for 20 minutes at room temperature for
thiol alkylation. Next, 1 µL of 100 ng trypsin was added to the tube and the reaction was
incubated at 37 °C for 3 hours. Then, an additional 1 µL of 100 ng trypsin was subsequently
added. The trypsin digestion was terminated after a total reaction time of 12 hours by the
2
2
2
2
2
2
80
81
82
83
84
85
286
2
2
87
88
289
90
2
12
ACS Paragon Plus Environment

Page 13 of 58
Biomacromolecules
1
2
3
2
2
91
92
addition of trifluoroacetic acid (TFA). Digested samples were purified using ZipTipC microtips
1
8
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
and eluted with 2 µL of matrix solution (20 mg/mL sinapinic acid in 50 % acetonitrile with
0.1 % TFA) directly onto a MALDI-ToF plate. The molecular weights of the expected peptide
fragments before and after digestion were predicted using PeptideCutter on UniProt P00766
(ExPASy Bioinformatics Portal, Swiss Institute of Bioinformatics). CT-initiator digests were
compared to native CT digests. Modification at a particular amino group was determined by
either the loss of a peak of the CT-initiator in comparison to native CT or by the appearance of a
new peak that equaled the mass (or adducts) of the peptide fragment plus the mass of the initiator
(neutral initiator: 220 Da, positive initiator: 320 Da). Fragments below the MALDI-ToF lower
m/z=500 limit were not able to be detected.
293
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
2
2
2
2
94
95
96
97
98
99
300
3
3
01
02
MALDI-ToF analysis Protein solutions (1.0 mg/mL) were mixed with an equal volume of
matrix (Sinapinic acid (20 mg/mL) in 50% acetonitrile with 0.4% trifluoroacetic acid) and 2 µL
of the resulting mixture was loaded onto a silver sterling plate target. Apomyoglobin,
cytochrome C, and aldolase were used as calibration standards. To determine the extent of
initiator modification on protein-initiators, the m/z of the native protein was subtracted from the
m/z of the protein-initiator. The difference in m/z was then divided by the mass of the initiator
(neutral initiator: 220 Da, positive initiator: 320 Da; after attachment) to obtain the number of
initiators per protein. When analyzing trypsin digests of protein-initiators, Bradykinin fragment,
angiotensin II (human) and insulin oxidized B chain (bovine) were used as calibration standards.
MALDI-ToF data was collected on a PerSeptive Voyager STR MS with nitrogen laser (337 nm)
and 20 kV accelerating voltage with a grid voltage of 90 %. 300 laser shots covering the
complete spot were accumulated for each spectrum.
303
3
3
3
3
3
3
04
05
06
07
08
09
310
3
3
11
12
13
ACS Paragon Plus Environment

Biomacromolecules
Page 14 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
3
13
14
Isoelectric Focusing (IEF) gel Criterion IEF precast gels (pH 3-10, 12+2 well, polyacrylamide
gel, 13.3 x 8.7 cm) from Bio-Rad were used to determine the isoelectric point of proteins and
protein-initiators. Protein solutions (concentration depended on sample) were mixed with 50%
glycerol using a 1 to 10 ratio of protein sample to 50% glycerol. 30 μL were loaded into each
315
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
3
16
17
well. The IEF standards were prepared and loaded according to the Bio-Rad instruction manual.
The gel was run in a stepwise manner as follows: 100 V for 60 minutes, 250 V for 60 minutes,
500 V for 30 minutes. Gels were silver stained using the Pierce Silver Stain Kit following their
instructions.
318
3
3
19
20
321
322
Protein-pCBMA and pSMA conjugate synthesis
A solution of monomer (230 mg for CBMA and 246 mg for SMA, 1.0 mmol respectively) and
CT-initiator (23 mg for neutral and 25 mg for positive initiator, 10 µmol of initiator) in 100 mM
sodium phosphate (20 mL, pH 7) was sealed and bubbled with nitrogen gas in an ice bath for 30
3
3
23
24
4
5
325
min. 2 mL of deoxygenated catalyst solution (described previously ) was added to the
polymerization reactor under bubbling nitrogen. The mixture was sealed and stirred in a
refrigerator for 1 h. The conjugate was isolated by dialysis with a 25 kDa molecular weight
cutoff dialysis tube in deionized water or 25 mM sodium phosphate buffer in a refrigerator for 24
h and then lyophilized. Conjugate CT content was determined by BCA assay as described
3
3
3
3
3
3
26
27
28
29
30
31
3
2
previously. Other protein-polymer conjugates were prepared using the same ATRP reactant
ratios, but at ~5x lower concentration of all reactants, and analyzed by the same procedure as
CT-polymers. Monomer concentrations to target DPs of 100 were used for all conjugates except
avidin, which was targeted at 150. Monomer conversions are listed in Table S2.
Acid hydrolysis and characterization of cleaved polymer
332
3
3
33
34
14
ACS Paragon Plus Environment

Page 15 of 58
Biomacromolecules
1
2
3
3
3
35
36
The grafted polymer was cleaved by acid hydrolysis from the conjugate. Protein-polymer
conjugate (20 mg) and 6 N HCl aq. (5 mL) were placed in a hydrolysis tube. After three
freeze−pump−thaw cycles, the hydrolysis was performed at 110 °C for 24 h in vacuo. The
cleaved polymer was isolated by dialysis using a 1 kDa molecular weight cut off dialysis tube in
deionized water and was then lyophilized. The molecular weight of the cleaved polymer was
measured by GPC.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
337
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
3
3
3
3
3
38
39
40
41
42
43
Dynamic light scattering Dynamic light scattering data was collected on a Malvern Zetasizer
nano-ZS located in the Department of Chemistry, Carnegie Mellon University, Pittsburgh, PA.
The hydrodynamic diameters (number and volume distributions) of samples were measured three
times (5 runs per measurement) at room temperature.
344
3
3
45
46
Prediction of logD ChemAxon was used to draw the structure of and calculate the
hydrophobicity (logD) of lysine side chains and lysine-initiators.
347
Activity Assays
3
3
48
49
CT: N-succinyl-Ala-Ala-Pro-Phe-p-nitroanilide (Suc-AAPF-pNA) was used as a substrate for
CT. Substrate (0-20 mg/mL in DMSO, 30 μL) was added to a 1.5 mL cuvette with sodium
phosphate buffer (pH 8, 100 mM). Native CT, CT-initiators, or CT-polymers (0.1 mg/mL protein,
4 μM, 10 μL) was added to the cuvette with substrate and buffer. The initial substrate hydrolysis
rate was measured in triplicate by recording the increase in absorbance at 412 nm over the first
60 seconds after mixing using a Lambda 2 Perkin Elmer ultraviolet-visible spectrometer
equipped with a temperature-controlled cell holder at 37 °C. Michaelis-Menten parameters were
determined using nonlinear curve fitting of initial hydrolysis rate versus substrate concentration
in GraphPad.
350
3
3
51
52
353
3
3
54
55
356
15
ACS Paragon Plus Environment

Biomacromolecules
Page 16 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
3
57
58
Lysozyme: Activities of native, lysozyme-initiator, and polymer conjugates were determined by
turbidimetric assays. Lyophilized Micrococcus lysodeikticus (Sigma Aldrich) was used to
monitor enzymatic catalysis of cell wall lysis. Absorption at 450 nm of suspended M.
lysodeikticus (990 µL, 0.2 mg/mL) in 50 mM sodium phosphate (pH 6.0) was measured by an
UV-VIS spectrometer. 10 µL of native and lysozyme-initiator solutions (1.4 µM in 50 mM
sodium phosphate (pH 6.0)) were added and the change of absorbance at 450 nm at room
temperature was monitored.
359
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
3
3
3
3
3
60
61
62
63
64
65
Avidin: HABA is an azo-dye that binds to avidin and shows spectral changes, thus can be used
2
for determination of avidin binding affinity . The absorption at 500 nm of a 200 µM HABA
366
solution in phosphate buffered saline without calcium or magnesium (986 µL, Lonza) was
measured using an UV-VIS spectrometer. 16 µL of the conjugate solution (125 µM of avidin in
deionized water) was added to the HABA solution and incubated at room temperature for 1 min,
and then the absorption at 500 nm was measured. The change in absorbance at 500 nm was used
to determine the amount of bound HABA to the conjugate. It has been previously shown that
3
3
67
68
369
3
3
3
3
3
3
70
71
72
73
74
75
3
upon biotin binding, tryptophan fluorescence of avidin is decreased from 337 to 324 nm. We
detected changes in intrinsic tryptophan fluorescence confirming biotin binding and took biotin
binding kinetic measurements using a stopped-flow spectrometer with fluorescence detection
(Applied Photophysics SX20). The excitation wavelength was 295 nm with a 5 nm bandwidth.
The instrument was permitted to collect 1000 data points throughout the reaction (0.1-450 s).
For all experiments, the avidin concentration was 0.5 µM (final) and biotin concentration was 5.0
µM (final). Reactions were initiated by mixing equal volumes of avidin with its substrates in 0.1
M phosphate buffer, pH 8. Fluorescence was measured in volts. Data were fit to single
exponential functions using the F(t) = F -∆Fexp(-k t) equation, where, k is the observed first-
376
3
3
77
78
379
∞
obs
obs
16
ACS Paragon Plus Environment

Page 17 of 58
Biomacromolecules
1
2
3
3
3
80
81
order rate constant, F is the final value of fluorescence and ∆F is the amplitude. In case of
∞
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
native avidin kinetics, the data were fit to single exponential functions using F(t) = F +∆Fexp(-
∞
382
kobst), where kobs is the observed first-order rate constant, F is the final value of fluorescence,
∞
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
3
3
3
3
3
83
84
85
86
87
88
and ∆F is the amplitude.
Uricase: Uric acid (0 – 400 µL of 300 µM in 50 mM sodium tetraborate (pH 8.5)) was mixed
with 50 mM sodium phosphate buffer (990 – 580 µL, pH 8.5). Native, initiator, and polymer
conjugate solutions (10 µL, 20 µM of uricase) were added to the substrate solution. The initial
rate was monitored by recording the decreasing in absorbance at 290 nm using an UV-VIS
absorbance spectrometer with a temperature controlled cell holder at 37 °C. Michaelis-Menten
parameters were determined by nonlinear curve fitting of initial rate versus substrate
concentration plots using Prism 7 software (GraphPad).
389
3
3
90
91
AChE: Acetylthiocholine iodide (0 – 100 µL of 10 mM in 100 mM sodium phosphate buffer
(pH 7.4)) and 10 µL of DTNB solution (50 mM in DMSO) were mixed with 100 mM sodium
phosphate (980 – 880 µL, pH 7.4). Native, initiator, and polymer conjugate solutions (10 µL,
4.2 µM of AChE) were added to the substrate solution. The initial rate was monitored by
recording the increase in absorbance at 412 nm using an UV-VIS absorbance spectrometer with a
temperature controlled cell holder at 37 °C. Michaelis-Menten parameters were determined by
nonlinear curve fitting of initial rate versus substrate concentration plots using Prism 7 software
(GraphPad).
392
3
3
3
3
3
3
93
94
95
96
97
98
399
Residual activity assays
4
4
4
00
01
02
CT: Native CT, CT-initiators, and CT-polymers (1 mg/mL, 40 μM protein) were dissolved in
sodium phosphate buffer (pH 8, 100 mM). In triplicate, samples were then diluted to 4 μM for
incubation. For thermostability, samples were incubated at 50 °C and pH 8 in a circulating water
17
ACS Paragon Plus Environment

Biomacromolecules
Page 18 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
4
4
4
03
04
05
06
bath. For acid stability, samples were incubated at pH 1(167 mM HCl) and 37 °C. At specified
time points, aliquots of 10 μL were removed over 60 minutes and residual activity was measured
using Suc-AAPF-pNA as a substrate (6 mg/mL, 30 μL, 288 μM in DMSO) in sodium phosphate
buffer (pH 8, 100 mM, 37 °C, 960 μL). Initial hydrolysis rate was measured as the increase in
absorbance at 412 nm over 40 seconds and data was normalized to its optimal activity (pH 8,
37 °C) at time 0.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
407
4
4
08
09
Lysozyme: Native lysozyme, initiator, and polymer conjugates (14 µM of lysozyme) in 50 mM
sodium phosphate buffer (pH 6.0) were incubated at 80 °C. At given time points, aliquots (10
µL) were removed and activity was measured in 990 µL of suspended M. lysodeikticus (0.2
mg/mL) in 50 mM sodium phosphate (pH 6.0) at room temperature. Rates were monitored by
recording the decreasing in absorbance at 450 nm using an UV-VIS spectrometer. The residual
activity was calculated as a ratio of the initial rates of the reaction at the given time points over
the initial activity at time zero.
410
4
4
4
11
12
13
414
4
4
15
16
Uricase: Native uricase, initiator, and polymer conjugates (20 µM of uricase) in 50 mM sodium
borate buffer (pH 8.5) were incubated at 75 °C. At given time points, aliquots (10 µL) were
removed and activity was measured in 990 µL of 100 µM uric acid in 50 mM sodium borate
buffer (pH 8.5) at 37 °C. Rates were monitored by recording the decrease in absorbance at 290
nm using an UV-VIS spectrometer. The residual activity was calculated as a ratio of the initial
rates of the reaction at the given time points over the initial activity at time zero.
AChE: Native, initiator, and polymer conjugates (1.4 µM of AChE) in 100 mM sodium
phosphate buffer (pH 7.0) were incubated at 50 °C. At given time points, aliquots (10 µL) were
removed and added to a mixture of 930 µL of 100 mM sodium phosphate (pH 7.4), 50 µL of
acetylthiocholine iodide (10 mM in 100 mM sodium phosphate (pH 7.4) and 10 µL of DTNB
417
4
4
18
19
420
4
4
4
21
22
23
424
25
4
18
ACS Paragon Plus Environment

Page 19 of 58
Biomacromolecules
1
2
3
4
4
26
27
solution (50 mM in DMSO) at 37 °C. Activity rates were monitored by recording the increase in
absorbance at 412 nm using an UV-VIS spectrometer. The residual activity was calculated as a
ratio of the initial rates of the reaction at the given time point over the initial activity at time zero.
Tryptophan Fluorescence Fluorescence measurements were collected using a BioTek Synergy
H1 Plate Reader. Native CT, CT-initiators, and CT-polymers (1 mg/mL, 40 μM protein) were
dissolved in sodium phosphate buffer (pH 8, 100 mM). Samples were diluted to 0.1 mg/mL (4
μM protein) in a black round bottom 96 well plate in triplicate. For thermostability, samples
were incubated at 45 °C (maximum temperature setting for the BioTek Synergy H1 plate reader)
and pH 8. For acid stability, samples were incubated at pH 1 (167 mM HCl) and 37 °C.
Fluorescence intensity was measured every 2 minutes over 60 minutes (excitation at 270 nm,
emissions at 330 nm and 350 nm). The ratio of emission fluorescence intensities (350 nm/330
nm) was plotted over time with time 0 as the fluorescence intensity of the sample at pH 8 and
37 °C.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
428
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
4
29
30
431
4
4
32
33
434
4
4
35
36
437
4
4
4
4
4
4
38
39
40
41
42
43
Molecular Dynamics Simulation CT-positive initiator model was built with the Maestro
Schrodinger build toolkit using the crystal structure of CT as the initial structure (PDB: 4CHA).
Positive initiators were attached to the N-terminus and all 14 lysine residues to create a fully
modified CT-positive initiator construct. The molecule was subjected to a 1 ns simulated
annealing using Desmond. Simulated annealing was performed in 4 stages: linear increasing
temperature from 300-400 K over 0-100 ps, constant temperature at 400 K from 100-400 ps,
linear decreasing temperature from 400-300 K over 400-700 ps, and constant temperature at 300
K from 700-1000 ps. The simulation system was prepared in Desmond system builder and
consisted of OPLS 2005 force field, SPC water model, orthorhombic minimized box, and NaCl
ions to neutralize the box followed by the addition of 100 mM NaCl. NVT ensemble and
444
4
4
45
46
447
48
4
19
ACS Paragon Plus Environment

Biomacromolecules
Page 20 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
4
49
50
Berendsen thermostat were used to control temperature with a 1 ps relaxation time. The van der
Waals interaction had a cutoff of 9 Å and particle mesh Ewald was used for Coulomb
interactions with a 9 Å switching distance. The molecule was simulated using Desmond over 1
ns with a 1.2 ps recording energy interval and 5 ps trajectory recording. A molecular dynamics
simulation production run was performed on the final structure from simulated annealing. The
simulation was performed over 20 ns at 300 K with a time-step bonded of 2 fs and a NPT
ensemble (trajectories were recorded every 1.2 ps and energy was recorded every 4.8 ps). “Nose-
Hoover chain” thermostat and “Martyna-Tobia-Klein” Barostat methods were used with 2 ps
relaxation time and isotropic coupling. The default relaxation protocol was used with a 9 Å
cutoff for van der Waals interactions. The solute atoms were restrained first with a force constant
451
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
4
4
4
4
4
52
53
54
55
56
57
458
-1 -1
4
4
59
60
of 50.0 kcal mol Å . Next, the solute heavy atoms were restrained with a force constant of 50.0
-1
-1
kcal mol Å . After the production run, the trajectory was then loaded into Visual Molecular
Dynamics (VMD) software for further analysis. The VMD salt bridge plug-in was used to
monitor salt bridge formation and location over the 20 ns trajectory caused by initiator-induced
structural changes. Salt bridges were monitored between acidic and basic protein residues and
not from charges originating from the initiator itself.
461
4
4
4
4
4
4
62
63
64
65
66
67
Data Availability. All relevant data is included in the manuscript, supporting information, or
may be available upon author’s request.
RESULTS AND DISCUSSION
468
Bioconjugate Synthesis and Characterization
4
4
69
70
We first synthesized a positively charged ATRP-initiator (Figure S2). N-(3-N’,N’-
Dimethylaminopropyl)-2-bromoisobutyramide and 4-bromobutyryl-N-oxysuccinimide ester were
synthesized and then reacted with each other to form a positively charged ATRP-initiator. The
471
20
ACS Paragon Plus Environment

Page 21 of 58
Biomacromolecules
1
2
3
4
4
72
73
overall synthesis had a 68% yield. The positive charge was in the form of a quaternary
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
+
ammonium, NR , which was located approximately halfway between the protein-reactive NHS
4
474
group and the terminal alkyl halide. The positive charge on the initiator quaternary nitrogen atom
is therefore extended by 5.6 ± 0.2 Å from the original charge on the lysine NZ atom. We selected
a quaternary ammonium group as the source of the positive charge since it would remain
positively charged at all pH’s and would have a higher propensity of providing stabilization due
to its chaotropic cationic nature.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
4
4
4
4
4
75
76
77
78
79
80
We have previously studied polymer-based protein engineering of the serine protease
chymotrypsin. Therefore, we naturally began our exploration of the impact of a permanently
positively charged initiator on enzyme structure, dynamics and function by measuring the
kinetics and stability of chymotrypsin-initiator constructs and chymotrypsin-polymer conjugates.
Our previous studies have shown that chymotrypsin loses activity but gains stability when
poly(carboxybetaine methacrylate) (pCBMA) is grown from the surface. Conversely,
poly(sulfonate methacrylate) (pSMA) rapidly destroys the function of chymotrypsin (Scheme
1).9,32
481
4
4
82
83
484
4
4
85
86
4
4
4
4
4
4
87
88
89
90
91
92
Scheme 1. Synthetic approach to prepare grafted-from protein-polymer conjugates using neutral (N) or positively
charged (+) initiators. The initiators reacted with primary amino groups on the protein surface through NHS
chemistry. ATRP was then performed from the bio-macroinitiators using a zwitterionic polymer, pCBMA, or a
negatively charged polymer, pSMA.
21
ACS Paragon Plus Environment

Biomacromolecules
Page 22 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
4
93
94
4
4
4
95
96
97
4
98
The CT-initiator constructs were analyzed with matrix assisted laser
4
99
desorption/ionization time-of-flight (MALDI-ToF) mass spectroscopy to determine the average
number of amino groups that had been modified (Figure S6). CT has 15 total amino groups (N-
terminus and 14 lysines). The average numbers of initiators attached to CT were 14.1 and 10.6
for the neutral (N) and positively charged (+) initiators, respectively (Table 1). The slight
decrease in total number of positive initiator modifications in comparison to the neutral initiator
was likely due to its larger size and charge. The larger size could inhibit reactions with primary
amines that have decreased accessible surface areas while the positive charge could hinder
500
5
5
01
02
503
5
5
5
04
05
06
5
0
reactions with primary amines in positively charged regions of CT. In order to determine the
22
ACS Paragon Plus Environment

Page 23 of 58
Biomacromolecules
1
2
3
5
5
07
08
sites of modification for each protein-initiator constructs, trypsin digestion followed by analysis
of peptide fragments using MALDI-ToF was performed (Figure S7 and Tables S3 and S4).50
The same sites were modified with the neutral and positively charged initiators except K82, K84,
K90, K177, and K202. In general, these sites had decreased exposed surface areas which made
them unable to be modified by the positively charged initiator, which was larger in size than the
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
509
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
5
5
5
5
5
10
11
12
13
14
15
5
0
neutral initiator. We were first interested in determining how the different CT-initiators
impacted the isoelectric point (pI) of CT, which is the pH at which CT has no net electrical
charge. We used isoelectric focusing (IEF) gel electrophoresis with a pH 3-10 gradient (Figure
S8) to study the protein charge in the enzyme-initiator constructs. Native CT has a pI of
5
4
516
approximately 8.75. The pI of the CT(N) construct dropped to pI values ranging from ~3-6,
with the majority of the band intensity at the limit of the gel around pH 3. There were three
distinct bands: pI ~3, 5, and 6 which were most likely due to sub-populations of protein-initiators
that had different numbers of reacted initiators. While MALDI-ToF provided an average number
of modifications in the sample, the IEF gel allowed us to visualize the sub-populations with
different degrees of modification. The decrease in pI for CT(N) was expected since the protein
was losing positive charges and becoming more acidic. The pI values for CT(+) were increased
to ~5-7.5 from CT(N), but were still decreased from native CT. pI values arise from an average
5
5
17
18
519
5
5
5
5
5
5
20
21
22
23
24
25
of the individual residue pK values, which are highly sensitive to their local electrostatic
a
5
5
environment. It is also known that charge-charge interactions are the dominant factor that shift
5
6
526
pK values of ionizable groups on the protein surface. Since our positively charged initiator did
a
5
5
27
28
not have a pK it made sense that the pI of CT(+) would be restored to intermediate values
a,
relative to native CT. After verifying that pI values for CT(+) were increased to ~5-7.5 from
23
ACS Paragon Plus Environment

Biomacromolecules
Page 24 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5
5
29
30
CT(N), protein-polymer conjugates were synthesized with incompatible (pSMA) and
compatible (pCBMA) polymers (Table 1).
5
5
31
32
Table 1. Characterization and activities of native CT, CT-initiators, and CT-polymers.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Catalytic
Cleaved
polymer
D
h
(nm)
Turnover
number,
Michaelis
Constant,
KM (µM)
Efficiency,
Number
of
initiators
Conjugate
M
n
(kDa);
PDI
M
n
Number
Dist.
kcat/K
M
-1
-1 -1
*(kDa);DP
kcat (s )
26.6 ± 0.3
21.6 ± 0.6
18.9 ± 0.4
22.2 ± 0.2
3.0 ± 0.2
(µM s )
0.25 ± 0.01
0.36 ± 0.03
0.36 ± 0.03
0.20 ± 0.004
0.01 ± 0.002
0.42 ± 0.02
0.30 ± 0.01
Native
(+)
--
--
--
--
--
--
3.98 ± 0.48
4.18 ± 0.70
4.07 ± 0.31
14.3 ± 7.5
12.9 ± 6.9
18.3 ± 3.4
17.5 ± 7.7
107 ± 5
60 ± 7
10.6
14.1
10.6
14.1
10.6
14.1
(
N)
(+)-pSMA
N)-pSMA
(+)-pCBMA
--
53 ± 5
21.7; 1.44
19.9; 1.54
23.8; 1.57
18.1; 1.54
256: 82
109 ± 3
217 ± 35
73 ± 6
(
306; 75
278; 104
281; 79
30.5 ± 0.6
25.6 ± 0.2
(N)-pCBMA
84 ± 3
533
534
n n
* Conjugate M =(cleaved polymer M × number of initiators) + molar mass of CT.
The molecular weights of the polymers were kept constant (targeted degree of
polymerization of 100) in order to compare the activity and stability of each conjugate to that of
5
5
5
5
5
5
35
36
37
38
39
40
5
7
the native enzyme. After ATRP and purification of the conjugates via dialysis, protein-polymer
conjugates were characterized with a bicinchoninic acid (BCA) assay to determine protein
concentration from which conjugate molecular weight and degree of polymerization were
1
0
estimated. The polymers were also cleaved from the protein surface via acid hydrolysis and the
isolated polymers were analyzed by gel permeation chromatography for relative molecular
weight and polydispersity (PDI) from which conjugate molecular weight and degree of
polymerization were also calculated. Conjugate number-average molecular weights reported in
Table 1 are derived from GPC data, however, BCA and GPC results agreed well. Number
541
5
5
42
43
544
45
average hydrodynamic diameters (D ) were also measured using dynamic light scattering and
h
5
conjugates grew in size from 3.98 nm for native CT to approximately 13 nm for CT-pSMA and
24
ACS Paragon Plus Environment

Page 25 of 58
Biomacromolecules
1
2
3
5
5
46
47
approximately 18 nm for CT-pCBMA conjugates grown from either CT(+) or CT(N). CT-
pCBMA conjugates had slightly larger hydrodynamic diameters because pCBMA is super-
hydrophilic which would give CT-pCBMA a larger hydration layer than CT-pSMA conjugates.
Only single peaks were detected in DLS by number and volume distributions (Figures S9-S12)
and there were no signs of visible aggregation.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
548
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
5
5
5
5
5
49
50
51
52
53
54
Impact of charged initiator on activity of an incompatible protein-polymer conjugate
Attachment of polymers to proteins usually causes significant activity reductions, which have
been attributed to protein structural stiffening.58 The degree of protein modification and activity
loss are also tightly correlated. Negatively charged polymers, such as pSMA, have been
previously shown to inactivate CT rapidly.32 Indeed, in the previous study, tryptophan
fluorescence intensity at pH 8.0 increased after the CT(N)-pSMA conjugate was synthesized,
indicating that the conjugate was already partially unfolded even in its most optimal
555
5
5
56
57
558
32
5
5
5
5
5
5
59
60
61
62
63
64
environment. We next explored whether maintaining enzyme surface charge, using a charged
initiator, could protect the biocatalytic activity in CT-pSMA conjugates. Activities were
measured by Michaelis-Menten kinetics at pH 8 and 37 °C using Succinyl-Ala-Ala-Pro-Phe-p-
nitroanilide (Suc-AAPF-pNA), a hydrophobic and negatively charged substrate for CT, that
binds to the hydrophobic S binding pocket and is then cleaved by the catalytic triad (Ser 195,
1
-1
His 57, Asp 102). The turnover numbers (k , s ), Michaelis constants (K , μM), and overall
cat
M
-1 -1
5
5
65
66
catalytic efficiencies (k /K , μM s ) were first determined for native CT and CT-pSMA
cat
M
conjugates grown from either neutral or positive initiators (Table 1). The growth of pSMA from
CT(N) caused CT to lose 97% of its activity, matching results from previous work.9,32
Unfortunately, this degree of activity loss is not unprecedented for protein-polymer conjugates in
567
68
5
25
ACS Paragon Plus Environment

Biomacromolecules
Page 26 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5
5
69
70
general. We were delighted to observe that the CT(+)-pSMA conjugate showed similar activities
to native CT in terms of both k and K . It is important to note that the positively charged
cat
M
571
initiator modified less sites than the neutral initiator (10.6 vs. 14.1). In our prior work with
chymotrypsin, however, we have shown that the activities of CT modified at 11 sites and 13 sites
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
5
5
5
5
5
72
73
74
75
76
77
5
1
were similar , so we believe that the activity difference between the neutral and positively
charged initiator was related to charge versus number of modifications.
We next considered why a simple charged initiator might have such a profound effect on
protein surface electrostatics when surrounded by a vast array of negative charges in the
polymer. After all, the CT(+)-pSMA conjugate retained 10 positive charges at its surface, but
added over 1,000 negative charges to the molecular shell. Charged groups produce an electric
field due to interactions with other charged particles in close proximity. The electric field
578
5
5
79
80
strength at a surface with propagating point charges can be estimated using (equation 2),
푘푞
퐸 = 푟2
5
81
(2)
-1
2
-2
582
where E is electric field (NC ), k is Coulomb’s constant (9.0E9 Nm C ), q is the signed
magnitude of the point charge, and r is the distance between the charges. Therefore, the electric
field strength is proportional to the magnitude of the electric charge and inversely proportional to
the squared distance. A CT(+)-pSMA conjugate of one polymer chain was modeled to estimate
the electric field strength at the protein surface (Figure 1). Since the electric field is additive,
even if there were 100 negative charges (DP=100) following the positive charge, the electric
field strength at the protein surface would still be +0.77. This calculation assumes that the
polymer backbone is in a linear conformation, which might not be true for all polymer types, but
we have previously shown through molecular dynamics simulation that pSMA is a rather stiff,
5
5
83
84
585
5
5
5
86
87
88
589
5
5
90
91
3
2
inflexible polymer, increasing the validity of the assumption. This informative example
26
ACS Paragon Plus Environment

Page 27 of 58
Biomacromolecules
1
2
3
5
5
92
93
highlights the importance of maintaining optimal surface charge prior to growth of charged
polymers. Since the positively charged initiator was able to maintain high levels of activity of an
incompatible polymer-protein conjugate, we next explored whether it may also improve
conjugate stability when exposed to typical stressors, such as temperature and pH.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
594
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
5
95
96
5
97
5
5
6
6
98
99
00
01
Figure 1. CT(+)-pSMA conjugate modeled as point charges. The positive charge is from the quarternary ammonium
on the positive initiator while the remaining negative charges are the anionic sulfonate groups on SMA monomers.
Even if there were 100 negative charges to the right of the positive charge, the electric field strength at the protein
surface would still be +0.77.
6
02
6
03
Impact of charged initiator on stability of an incompatible protein-polymer conjugate
Various strategies have been used to stabilize protein solutions including adding excipients
(polyols and salts)36,59,60, immobilization onto solid supports, encapsulation into reverse
604
61
6
6
05
06
6
2
9
micelles , or by covalent attachment of polymer. CT has a two-step deactivation mechanism
607
08
609
where it undergoes complete deactivation via an intermediate transition state as follows (1),63
푘1 푘2
퐹 퐼 푈
6
(1)
where F, I, and U are the folded, intermediate, and unfolded conformational states, and k and k2
1
-1
6
6
10
11
are first order deactivation rate constants (min ). CT, when conjugated with polymer, resists
6
3
inactivation by either not unfolding or refolding at a high rate. We first determined the
resistance of CT-pSMA conjugates to inactivation by heat (50 °C, pH 8) and acid (pH 1, 37 °C)
612
27
ACS Paragon Plus Environment

Biomacromolecules
Page 28 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
6
6
13
14
(Figure 2). Deactivation via temperature and pH were chosen because they are two common
stressors for protein structural stability and are valuable parameters for both industrial and
therapeutic applications of enzymes. Covalently attached polymers have the ability to increase
the temperature and pH working ranges of the enzyme to increase their robustness during
industrial processing. At specified time points, aliquots were taken from the incubating samples
and activities were measured at pH 8 and 37 °C. CT(N)-pSMA irreversibly inactivated after just
10 minutes at 50 °C (Figure 2A). In another demonstration of the impact of the positive initiator,
CT(+)-pSMA was remarkably stable as it retained 60% residual activity at 60 minutes. Our prior
work has elucidated the mechanism of CT-polymer conjugate resistance to acid induced
irreversible inactivation where the polymers trap proteins in partially unfolded intermediate
615
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
6
6
6
6
6
16
17
18
19
20
21
622
32
6
6
23
24
states, prevent complete unfolding, and assist in refolding. In acid, CT(+)-pSMA also displayed
higher stability and was able to maintain 20% residual activity in comparison to CT(N)-pSMA,
which had immediately irreversibly inactivated (Figure 2B). These data, in combination with
Michaelis-Menten activity data, were the first observations to our knowledge of both high
activity and stability with a negatively charged protein-polymer conjugate. The stability data
further confirmed our view of the importance of net protein surface charge restoration when
growing polymers from protein surfaces. Next, we began to dissect the interactions between the
protein surface and the bio-macroinitiators.
625
6
6
6
6
6
6
26
27
28
29
30
31
28
ACS Paragon Plus Environment

Page 29 of 58
Biomacromolecules
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
32
6
6
6
6
6
6
33
34
35
36
37
38
Figure 2. A) CT-pSMA thermal stabilities at 50 °C and pH 8 and B) CT-pSMA acid stabilities at pH 1and 37 °C for
conjugates grown from (N) or (+) initiators. CT(N)-pSMA (○), and CT(+)-pSMA (○). CT-pSMA conjugates
synthesized using the positive initiator had increased thermal and acid stabilities in comparison to their neutral
initiator conjugate counterparts. Residual activities were normalized to activity at time 0 which was the conjugate’s
optimal conditions for activity at pH 8 and 37 °C. Error bars in all plots represent the standard error of the mean
from triplicate measurements.
639
640
6
6
6
41
42
43
Impact of charged initiator on activities and stabilities of protein-initiator constructs
Michaelis-Menten activities for CT-initiators were measured and are listed in Table 1. CT(N)
and CT(+) had similar activities, but both had higher overall catalytic efficiencies than native CT
due to a decrease in K . The observed decrease in K upon neutral initiator attachment could
644
M
M
6
6
45
46
have been the result of the hydrophobicity of the initiator. The octanol-water distribution
coefficient (log D), which takes into account the hydrophobicity of different ionization states of a
charged molecule, of a lysine side chain at pH 8.0 was -1.00 using ChemAxon. After covalent
attachment of the neutral initiator, the calculated log D value of a lysine-initiator construct rose
to 1.82. Considering that this reaction occurred on 14 out of the possible 15 amino groups, the
surface of CT would have undoubtedly become more hydrophobic which would have
strengthened the van der Waals interactions between the hydrophobic substrate and hydrophobic
S1 binding pocket to increase the affinity for the substrate. After attachment of the positive
initiator, however, the calculated log D of the lysine-initiator construct decreased to -1.98. The
647
6
6
6
6
6
6
48
49
50
51
52
53
29
ACS Paragon Plus Environment