Adequate Reducing Conditions Enable Conjugation of Oxidized

Subscriber access provided by UNIV OF NEWCASTLE

Article

ADEQUATE REDUCING CONDITIONS ENABLE CONJUGATION OF OXIDIZED PEPTIDES TO POLYMERS BY ONE-POT THIOL CLICK CHEMISTRY Gwendoline Tallec, Celestine Jia Ling Loh, Benoît Liberelle, Araceli Garcia Ac, Sung Vo Duy, Sébastien Sauvé, Xavier Banquy, Frederic Murschel, and Gregory De Crescenzo Bioconjugate Chem., Just Accepted Manuscript • DOI: 10.1021/acs.bioconjchem.8b00684 • Publication Date (Web): 16 Oct 2018 Downloaded from http://pubs.acs.org on October 21, 2018

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 70 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

Bioconjugate Chemistry

Graphical Table of Content 539x318mm (96 x 96 DPI)

ACS Paragon Plus Environment

Bioconjugate Chemistry 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

Page 2 of 70

1

Adequate Reducing Conditions Enable Conjugation of Oxidized

2

Peptides to Polymers by One-Pot Thiol Click Chemistry

3

4

AUTHOR NAMES

5

Gwendoline Tallec‡, Celestine Loh†, Benoit Liberelle‡, Araceli Garcia-Ac§, Sung

6

Vo Duy¶, Sébastien Sauvé¶, Xavier Banquy§, Frederic Murschel§,* and Gregory

7

De Crescenzo‡,*

8

AUTHOR ADDRESSES

9

‡Department

of Chemical Engineering, Groupe de Recherche en Sciences et

10

Technologies Biomédicales (GRSTB), Bio-P2 Research Unit, École

11

Polytechnique de Montréal, P.O. Box 6079, succ. Centre-Ville, Montréal, QC,

12

Canada H3C 3A7.

13

§Faculty

14

Montreal, Quebec, Canada, H3C 3J7.

of Pharmacy, Université de Montréal, 2900 Edouard-Montpetit Blvd.,

1 ACS Paragon Plus Environment

Page 3 of 70 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


1

¶Department

of Chemistry, Université de Montréal, C.P. 6128, succ. Centre-Ville,

2

Montreal, QC H3C 3J7

3

†Division

4

University, 50 Nanyang Avenue, Singapore, Singapore, 639798.

5

*These authors contributed equally to this work. Corresponding authors e-mail

6

addresses: [email protected]; [email protected]

of Chemical and Biomolecular Engineering, Nanyang Technological

7 8

9

10

ABSTRACT

Thiol(-click)

chemistry

has

been

extensively

investigated

to

conjugate

11

(bio)molecules to polymers. Handling of cysteine-containing molecules may

12

however be cumbersome, especially in the case of fast-oxidizing coiled-coil-

13

forming peptides. In the present study, we investigated the practicality of a one-

14

pot process to concomitantly reduce and conjugate an oxidized peptide to a

15

polymer.

Three

thiol-based

conjugation

chemistries

(vinyl

sulfone

(VS), 2



Page 4 of 70

1

maleimide and pyridyldithiol) were assayed along with three reducing agents

2

(tris(2-carboxyethyl)phosphine

3

Seven out of the 9 possible combinations significantly enhanced the conjugation

4

yield, provided that an adequate concentration of reductant was used. Among

5

them, the co-incubation of an oxidized peptide with TCEP and a VS-modified

6

polymer displayed the highest level of conjugation. Our results also provide

7

insights into two topics that currently lack of consensus: TCEP is stable in 10

8

mM phosphate buffered saline and it reacts with thiol-alkylating agents at sub-

9

millimolar concentrations, and thus should be carefully used in order to avoid

10

(TCEP),

dithiothreitol

and

β-mercaptoethanol).

interference with thiol-based conjugation reactions.

11


Page 5 of 70 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

1


INTRODUCTION

2

Considerable efforts have been put over the last decades into developing

3

functional biomaterials based on bioconjugates for all fields of life sciences.

4

Combining the most interesting properties of natural or synthetic polymers and

5

peptides or proteins, bioconjugates may indeed provide a suitable solution for

6

numerous applications. Their preparation currently remains the subject of many

7

studies, with emerging strategies focusing on site-specific and bioorthogonal

8

reactions to be carried out in mild conditions to preserve the biological activity

9

and the integrity of the (bio)molecules of interest 1. “Click” chemistry reactions

10

have been increasingly developed in that endeavor, since they enable fast,

11

controlled, oriented, reproducible and high-yield conjugation of chemically-

12

delicate molecules

13

Solid-phase

2, 3.

peptide

synthesis

enables

the

introduction

of

virtually

any

14

functional/reactive group into a peptide via the side-chain of an unnatural amino

15

acid. New methods, such as native chemical ligation, also allow for the



Page 6 of 70

1

synthesis of increasingly long and complex peptide chains

4.

Nonetheless,

2

biological production of peptides and proteins remains the gold standard,

3

especially for mass production of complex proteins 5. The sulfhydryl group that

4

is borne by one natural amino acid, cysteine, has probably been the second

5

most common functional group used for biomolecules crosslinking or conjugation

6

6.

7

“thiol-click” chemistry

8

biologically- and chemically-produced peptides/proteins and does not require

9

chemical modification prior to conjugation. Furthermore, given that most proteins

It has been the target of an ever-increasing number of strategies based on 7, 8.

Indeed, thiol(-click) chemistry can be applied to both

10

rarely display more than one free cysteine residue

9,

the use of thiol groups

11

often provides site-specificity and orientation. If need be, a cysteine residue can

12

be introduced in the sequence of a recombinant protein to provide a single

13

reactive site, usually at the N- or C-terminus

10, 11.

14

The intertwining of 2 or more α helices, or α-helical coiled-coil, is an

15

ubiquitous motif found in proteins that has been extensively studied and is

16

among the best-understood protein folds

12.

The rational design of de novo 5


Page 7 of 70 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


1

coiled-coil multimers usually comprises 2 to 6 repeats of a 7 amino-acid

2

sequence, denoted (gabcdef)n. Solvent-exposed polar residues occupy positions

3

e and g, participate in intra- and interhelical polar/electrostatic interactions and

4

largely determine the homo- vs heteromultimerization state. Nonpolar and mostly

5

aliphatic residues are positioned in positions a and d to form an interhelical

6

hydrophobic core that is the main determinant for stability

7

peptides have recently (re)gained keen interest as a powerful tool for the

8

development of bioconjugates that may “solve problems across many different

9

biological systems”

14.

13.

Coiled-coil

To form fibers, hydrogels and dendrimers, or even to act

10

as linker between subcellular structures, coiled-coil-forming peptides have been

11

conjugated to polymers, recombinant protein fusions and nanoparticles

12

the past few years, a specific E/K heterodimeric coiled-coil complex has been

13

extensively studied: the peptides sequences are based on 3 to 5 repeats of the

14

(EVSALEK) and (KVSALKE) heptads or on close analogs that mostly differ in

15

the abcd region (VSAL, VAAL, IAAL, LAAI or IAAI)

16

for an increasing control over parallelism and oligomerization state

12, 14, 15, 17, 19.

15-18.

In

This allowed 14, 17,

as well 6



12, 19, 20.

Page 8 of 70

1

as specificity and stability

Our work on the coiled-coil-forming peptides

2

(KVSALKE)5 and (EVSALEK)5, or K5 and E5, has enabled the immobilization

3

and controlled release of recombinant growth factor-E5 chimeras on K5-

4

conjugated polymeric substrates

5

based on thiol-ene chemistry using cysteine-tagged K5 peptides that were either

6

chemically synthesized or produced in bacteria 5.

21, 22.

Our bioconjugation strategy has been

7

Any thiol-containing molecule loses reactivity over time, in air or in an

8

aqueous solution, due to oxidation, be it to reversible disulfide or to irreversible

9

sulfinic/sulfonic acid

23.

However, unlike most peptides, cysteine-tagged coiled-

10

coil-forming peptides are highly prone to oxidation, which can be complete

11

within a few hours in aqueous solutions even in mild conditions

12

be explained by the propensity of the coil peptide to homodimerize, even if it is

13

carefully designed with charged residues introduced to promote preferential

14

heterodimerization via electrostatic repulsion

15

peptides require careful handling for optimal results, which include cumbersome

16

procedures and specialized equipment, such as degassed solutions and an inert

15.

24.

This could

Nonetheless, all thiol-containing


Page 9 of 70 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


1

atmosphere (see anaerobic chambers). Moreover, to prevent oxidation, most

2

suppliers recommend storing peptides in lyophilized form under argon at -80°C

3

and avoiding repeated freeze-thaw cycles, which strongly deviates from the

4

genuine needs of early research in many laboratories that only require, and

5

may afford, minimal amounts of peptides for experimentation.

6

The most commonly reported strategy to reverse the oxidation of cysteine

7

residues is to incubate the peptide or protein in a concentrated solution of a

8

sulfhydryl-containing

9

mercaptoethanol (BME) then remove the excess reductant by size-exclusion

reducing

agent

such

as

6.

dithiothreitol

(DTT)

or

β-

10

chromatography or dialysis prior to conjugation

The water-soluble reductant

11

tris(2-carboxyethyl)phosphine (TCEP) has also been used in that endeavour

12

and, according to many sources, can eliminate the need for lengthy purification

13

before conjugation, in particular with maleimides

14

peptides and/or for specific applications that inevitably require TCEP removal,

15

including gel electrophoresis and protein labeling, agarose gels have been

16

commercialized with the reductant being covalently immobilized to the matrix;

6, 25, 26.

For fast oxidizing



Page 10 of 70

1

they have even been successfully applied to coiled-coil-forming peptides

2

After incubation, a simple centrifugation step is performed to pellet the slurry

3

and collect a solution of reduced sample with no reductant. Interestingly,

4

Zwyssig and colleagues presented in 2017 their work on magnetic cobalt

5

nanoparticles functionalized with TCEP, an approach minimizing the loss of

6

sample due to adsorption that is often encountered with the agarose beads

7

Henkel and colleagues followed by Kantner et al. reported, in 2016 and 2017,

8

respectively, elegant approaches based on the quenching of TCEP in situ with

9

4-azidobenzoic acid and with diazido-polyethylene glycol, respectively, that

10 11

circumvents the need for TCEP removal Nonetheless,

all

these

efforts

27.

28.

29, 30

neither

resolve

cysteine

oxidation

during

12

conjugation, which may strongly affect yield, nor build consensus around

13

whether non-sulfhydryl containing reductants interfere or are compatible with

14

thiol chemistry. Indeed, the need to remove TCEP before thiol-ene conjugation

15

remains a topical bone of contention

16

around reductant compatibility with thiol(-click) chemistry, we here investigated

31, 32.

In an effort to build consensus


Page 11 of 70 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


1

the thiol-based covalent grafting of a peptide to a polymer in highly unfavorable

2

conditions. More precisely, and to the extent of our knowledge, we carried out

3

a set of unprecedented experiments by examining the conjugation of a coil

4

peptide (fast oxidizing peptide, starting with a fully oxidized stock) to a surface-

5

immobilized polymer (kinetics of thiol reaction limited by diffusion) in a simple

6

aqueous solution (no other solvent, no degassing and no inert atmosphere).

7

Such unfavourable conditions can indeed be brought together when developing

8

biofunctionalized

surfaces,

9

immunoassays

Our rationale was to determine the feasibility of one-pot thiol

33.

e.g.

for

medical

implants,

biosensors

and

10

chemistry by co-incubating the peptide with (i) a commonly used reducing agent

11

(TCEP, DTT or BME, see Figure 1.B) and (ii) a model polymer that was

12

modified with a pyridyldithiol group, to obtain a cleavable disulfide bond, or with

13

the thiol-alkylating maleimide and vinyl sulfone groups (see Figure 1.A).

14



Page 12 of 70

1 2

Figure 1. Reaction schemes of one-pot thiol chemistry for the grafting of the K5 peptide

3

(R2-SH) to dextran (R1) in presence of reducing agents. (A) Reactions between the K5

4

peptide and dextran modified with vinyl sulfone and maleimide moieties for thiol-

5

alkylation or with pyridyldithiol moieties for thiol-disulfide interchange. (B) Mechanism of

6

oxidized K5 peptide reduction by a phosphine, tris(2-carboxyethyl)phosphine (TCEP), or

7

by thiol-containing reductants, dithiothreitol (DTT) and β-mercaptoethanol (BME).


Page 13 of 70 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


1

RESULTS AND DISCUSSION

2

Compatibility of in situ reduction with thiol-disulfide interchange and thiol

3

alkylation.

4

A covalent layer of low-fouling dextran bearing either vinyl sulfone or

5

carboxymethyl groups was first generated in polystyrene-based 96-well plates.

6

The surfaces with carboxymethylated dextran were further modified with a

7

crosslinker, namely EMCH or PDPH, to introduce maleimide or pyridyldithiol

8

moieties, respectively. The cysteine-terminated K5 peptide was then incubated

9

on the three reactive surfaces for 6 hours with different concentrations of a

10

reducing agent among tris(2-carboxyethyl)phosphine (TCEP), dithiothreitol (DTT)

11

or β-mercaptoethanol (BME) in phosphate buffered saline (PBS, 10 mM, pH

12

7.4). It is here noteworthy that the K5 peptide was largely oxidized at the

13

beginning of the experiments (less than 1.2% of free thiol groups, as assessed

14

using Ellman’s reagent, data not shown). The resulting K5 peptide density in

15

the wells was evaluated by incubating the E5-tagged epidermal growth factor 12 ACS Paragon Plus Environment


Page 14 of 70

1

(E5-EGF), in order to capture EGF via the formation of the E5/K5 coiled-coil

2

complex. The amount of tethered EGF was then assayed by performing an

3

ELISA against EGF. For each type of thiol-reactive surface, the concentration

4

ranges with the three reducing agents were performed and analyzed within the

5

same plate, and the data are presented as relative to the maximal ELISA

6

response of said plate (Figure 2). A series of controls was performed by

7

incubating the K5 peptide in PBS or PBS supplemented with 100 µM of TCEP

8

on thiol-reactive and quenched surfaces (Table 1).

9

The discussion of the data is here presented in two steps: first, a global

10

interpretation of the curve profiles shown in Figure 2 in light of the controls in

11

Table 1, with a distinction between the thiol alkylating agents (maleimide and

12

vinyl sulfone) and pyridyldithiol and second, an in-depth analysis of the data

13

from Figure 2, with a comparison between the thiol-containing reducing agents

14

(DTT and BME) and the phosphine (TCEP).


Page 15 of 70 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


1 2

Figure 2. Grafting of oxidized K5 peptide in PBS (10 mM, pH 7.4) co-incubated with

3

reducing agents on (A) vinyl sulfone, (B) maleimide and (C) pyridyldithiol-derivatized

4

dextran surfaces, as monitored by ELISA. The lines are guides for the eyes only.

5



1 2

Page 16 of 70

Incubation with reductants promotes covalent grafting of the K5 peptide in 7 out of 9 cases

3

Regarding the thiol alkylating agents (vinyl sulfone and maleimide), the

4

incubation of oxidized K5 peptides in the absence of reducing agent induced a

5

very low ELISA response (6 ± 4% and 5 ± 2% for vinyl sulfone and

6

maleimide, respectively, Table 1). The data thus indicated that there was little

7

to no reactivity of the oxidized peptide towards those functional groups. The

8

addition of BME to the K5 peptide on dextran-maleimide did not significantly

9

improve the ELISA response which remained minimal, in good agreement with

10

the high reactivity of BME towards maleimide at neutral pH

34.

Bell-shaped

11

curves were obtained for all other reductant/thiol alkylating agent combinations.

12

Importantly, the incubation of K5 peptides on unreactive surfaces induced a

13

relative response inferior to 4%, be it in plain PBS or in PBS supplemented

14

with 100 µM TCEP (Table 1). The evolution of the ELISA response on Figure

15

2.A and Figure 2.B thus depicts an increased covalent grafting of the K5

16

peptide via thiol chemistry and not its adsorption, as expected on dextran

21.


Page 17 of 70 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


1 2

Table 1. Relative peptide density obtained after incubating the K5 peptide in PBS or PBS

3

supplemented with 100 µM TCEP on vinyl sulfone, maleimide and pyridyldithiol-

4

derivatized surfaces and on unreactive control surfaces, as monitored by ELISA. Vinyl sulfone

5 6 7 8

Maleimide

Pyridyldithiol

Active

Blockeda

Active

Blockeda

Active

No linkerb

K5 in PBS

6±4

3.0 ± 0.1

5±2

1.7 ± 0.8

43 ± 3

25 ± 9

K5 in PBS with 100 µM TCEP

100 ± 5

3.2 ± 0.2

89 ± 4

1.6 ± 0.2

26 ± 2

26 ± 2

aBlocked

surface indicates that the thiol alkylating agent was incubated with 50 mM of BME for 30

minutes then thoroughly rinsed prior to K5 incubation. bBare

carboxymethylated dextran (no linker) was used as control for pyridyldithiol instead of a linker

previously incubated with 50 mM of BME to avoid possible bias due to disulfide exchange.

9

10

As for the pyridyldithiol-derivatized surfaces, the unreacted linker was neither

11

blocked with 50 mM of BME nor stored overnight in PBS after the K5 peptide

12

incubation, so as to prevent the release of the peptide grafted on the surface

13

via a reversible disulfide bond. In this context, the signal on control unreactive

14

surfaces (bare carboxymethylated dextran) reached ca. 25 % of the maximum

15

level (see Table 1). In terms of raw absorbance values, this was ca. 18 times

16

more than what was observed on the carboxymethylated dextran surfaces with

17

blocked maleimide surfaces, which is indicative of non-negligible adsorption in 16 ACS Paragon Plus Environment


Page 18 of 70

1

this particular setup. Nonetheless, the response for the oxidized K5 peptide on

2

the pyridyldithiol linker in PBS with no reductant was significantly higher than

3

on bare CMD (43 ± 3% of the maximal value of the plate, see Table 1 and

4

Figure 2.C). Here, free thiol groups that could interchange with the cystine

5

disulfide can be generated on the surface by a partial degradation of the linker

6

(liberation of the pyridine-2-thione). When the peptide was co-incubated with the

7

reducing agents, bell-shaped curves were obtained on pyridyldithiol surfaces,

8

except for BME which only decreased the ELISA signal.

9 10

TCEP interferes with thiol coupling reactions, but to a lesser extent than thiolcontaining reductants

11

Regarding DTT (and BME on vinyl sulfone only), the bell-shaped curves

12

shown in Figure 2 and the negative controls shown in Table 1 confirmed that:

13

(i) thiol-disulfide exchange occurred between the K5 peptide and the thiol-

14

containing reducing agents and (ii) an increase in reductant concentration

15

promoted

16

concentrations, the two agents most likely competed with the free thiol of the

peptide

grafting,

until

an

optimum

was

reached.

At

high


Page 19 of 70 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


1

cysteine residue for the maleimide, vinyl sulfone and pyridyldithiol moieties.

2

Interestingly, for most cases, peptide grafting was completely thwarted only

3

when the reducing agent was in large excess, that is, when the agent-to-

4

peptide ratio was over 10 for DTT on maleimide, 500 for both DTT and BME

5

on pyridyldithiol and even 105 for both agents on vinyl sulfone. The capacity of

6

the K5 peptide to react with the thiol-reactive moieties in the presence of a

7

high excess of reducing agent might be explained by an even higher excess of

8

thiol-reactive moieties on the surfaces compared to the reducing agent. Another

9

explanation would be that the K5 thiol is more reactive than BME and DTT

10

towards vinyl sulfone, maleimide and pyridyldithiol. More precisely, the N-

11

terminal cysteine of the K5 moiety may be, at pH 7.4, less protonated than

12

DTT and BME, and the deprotonated thiolate group R-S- is known to be

13

dramatically more reactive than the thiol group R-SH

14

feature pKa values of 9.2 and 9.6, respectively

15

higher than the cysteine value which is usually reported at 8.3-8.5

16

present study, the cysteine pKa value could be even lower, insofar as the

35.

Indeed, DTT and BME

36, 37,

which are significantly 35, 38.

In the



Page 20 of 70

1

residue is located at the N-terminus of a peptide with high helical propensity

39.

2

A similar reasoning could explain the overall lower peptide densities obtained

3

with BME, when compared to DTT, since BME features a higher pKa value and

4

lower disulfide reduction potential than DTT

37.

5

Regarding TCEP, similar results were obtained, although the phosphine

6

consistently led to higher peptide density for the three chemistries, at the

7

optimal concentration. In the case of pyridyldithiol, it could be argued that

8

TCEP concomitantly reduced the peptide disulfide and cleaved the linker

9

disulfide on the surface, thus rendering it inert towards the reduced peptide.

10

This would explain why the optimal TCEP concentration was very close to the

11

peptide concentration in this mixture (2 µM and 1 µM, respectively, Figure 2.C)

12

and that high TCEP concentrations thwarted peptide grafting. As for surfaces

13

derivatized with maleimide and vinyl sulfone moieties, the phosphine at high

14

concentration appeared to have prevented their reaction with the peptide. This

15

result is in disagreement with the oversimplified presentation that TCEP and

16

alkylating agents are compatible during thiol chemistry, which remains often 19 ACS Paragon Plus Environment

Page 21 of 70 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


1

found in the literature, especially of commercial sources

2

with several reports investigating and demonstrating TCEP and maleimide

3

cross-reactivity, starting from Shafer and colleagues who obtained by-products

4

when mixing TCEP and N-ethylmaleimide

5

confirmed the early observations, showing that an excess of the reducing agent

6

is detrimental to one-pot thiol chemistry

7

used successfully in one-pot thiol chemistry with other alkenes, including vinyl

8

sulfone

9

agent but as a catalyst for the thiol-ene reaction. In our study, it however

10

appeared that an excess amount of TCEP is also detrimental for thiol/vinyl

11

sulfone conjugation.

43,

vinyl phenyl

44

and acrylate

34.

It however agrees

Many others have since then

31, 32, 41, 42.

45

40.

Conversely, TCEP has been

moieties, and not only as reducing

12



1

Page 22 of 70

Competitive reactions between reducing agents and dextran-vinyl sulfone

2

The data shown on Figure 2 demonstrated that the conjugation of a fully

3

oxidized K5 peptide to a thiol-reactive polymer can be performed in the

4

presence of a well-chosen reducing agent at an adequate concentration, via

5

vinyl sulfone, maleimide or even pyridyldithiol chemistry.

6

In terms of stability, it is here noteworthy that the disulfide bond obtained with

7

pyridyldithiol is susceptible to breaking or reforming in the presence of

8

competing thiols or reducing agents. Regarding the thiol-Michael addition

9

reactions, vinyl sulfones are more reactive towards thiols than (meth)acrylates

10

and acrylamides, second only to maleimides

46.

Maleimides however yield a

11

reversible succinimide bond that features a considerably higher susceptibility to

12

hydrolytic degradation than the thioether sulfone obtained with vinyl sulfones

13

47.

14

reactive moiety compatible with the three assayed reducing agents, and VS

15

moieties can be readily integrated in polysaccharides using inexpensive divinyl

46,

Moreover, vinyl sulfone (VS) was, in the present study, the only thiol-


Page 23 of 70 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


1

sulfone, in a controlled manner over a wide range of grating ratio and without

2

side crosslinking reactions

3

characterization.

21.

Dextran-VS was therefore selected for further

4

K5 peptides were then incubated at different concentrations over dextran-VS

5

well-plates, with various concentrations of TCEP, and grafting density was

6

monitored by ELISA (Figure 3). Interestingly, the maximal amount of grafted

7

peptide, which was always obtained with 100 µM of TCEP, increased with

8

peptide concentration, by 2.8 ± 0.1, 5.1 ± 0.4 and 6.4 ± 0.2-fold when increasing the

9

concentration from 1 µM of K5 to 2, 5 and 10 µM of K5, respectively. The

10

data were thus in agreement with the VS moieties being in large excess in the

11

wells compared to the peptide. Further characterization was carried out by

12

incubating a high concentration of cysteamine on dextran-VS and by quantifying

13

the amount of resulting amino

14

estimated to be between 4000 and 5800 pmol in each well (data not shown).

15

This further confirmed that the VS moieties were in large excess when

groups. Quantities of VS moieties were



Page 24 of 70

1

compared to the cysteine-tagged peptide (50 µL x 1 µM = 50 pmol) but not

2

necessarily compared to the reducing agents.

3

Dextran-VS (17.4 mM of VS moieties) was incubated at room temperature

4

with the reductants TCEP, DTT and BME at a concentration of either 0.1 mM

5

or in excess at 50 mM for 6 and 24 hours. Unreacted reducing agents and

6

salts were then thoroughly removed by centrifugal filtration and the polymer

7

was lyophilized before 1H NMR spectroscopy in D2O (Figure 4). NMR analysis

8

of dextran-VS revealed a series of peaks between 3 and 4.5 ppm that

9

correspond to the hydrogen atoms within the saccharide unit, except for the

10

atom labeled c in Figure 4 which was well separated at 4.8-5.2 ppm and thus

11

used for normalization. The Michael addition of divinyl sulfone to the saccharide

12

unit was responsible for the doubling of peak c and for the apparition of the

13

peaks corresponding to the three hydrogen atoms of the VS carbon-carbon

14

double bond which can be seen in Figure 4: 1 H atom a and 2 H atoms b

15

integrated in the ranges 6.77-7 ppm and 6.2-6.45 ppm, respectively

21.


Page 25 of 70 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


1 2

Figure 3. Influence of peptide and TCEP concentration on the grafting of oxidized K5

3

peptide in PBS (10 mM, pH 7.4), on vinyl sulfone-derivatized dextran surfaces, as

4

monitored by ELISA. The data are relative to the ELISA signal obtained for 1 µM of K5

5

co-incubated with 100 µM of TCEP. The lines are guides for the eyes only.

6



Page 26 of 70

1 2

Figure 4. 1H NMR spectroscopy of vinyl sulfone-derivatized dextran incubated for 6 or 24

3

hours with (A) TCEP, (B) DTT and (C) BME at 0.1 and 50 mM. H atoms of interest are

4

highlighted on the schematic compounds on the right. % values in insets are NMR areas

5

under curve divided by the number of equivalent H atoms, relative to dextran saccharide

6

units (calculated from area under curve of the H atoms labelled “c”). 25 ACS Paragon Plus Environment

Page 27 of 70 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


1

Regarding the thiol-containing reductants, the incubation of dextran-VS with 50

2

mM of either DTT or BME caused the complete disappearance of the VS

3

carbon-carbon double bond after 6 hours, with little to no change after 24

4

hours (Figure 4.B and Figure 4.C). This observation was associated with the

5

apparition of the 2 hydrogen atoms d of the β-thiosulfonyl linkage at ca. 3 ppm,

6

as well as a peak at ca. 2.7 ppm that could be attributed to the 4 hydrogen

7

atoms f of DTT and the 2 hydrogen atoms g of BME, respectively, using the

8

EPFL

9

reductants were either predicted to be between 3.5 and 5 ppm, and therefore

10

embedded in the dextran signal, or undetectable when using the D2O solvent

11

(for hydroxyl groups). Unsurprisingly, when normalized per number of hydrogen

12

atoms (see insets in Figure 4), the integral values were consistent with a

13

complete and equimolar conjugation of DTT/BME with dextran-VS via thiol

14

alkylation, that is, all the vinyl sulfone moieties (17.4 mM) reacted within 6

15

hours with the thiol-containing reducing agent (50 mM). Interestingly, although

16

the integration was too low to be reliable, the peaks corresponding to the β-

1H

NMR online prediction tool. All other hydrogen atoms for the



linkage

and

the

reductant

started

Page 28 of 70

1

thiosulfonyl

to

appear

after

a

6-hour

2

incubation with only 0.1 mM of DTT or BME. This might indicate their

3

conjugation to the polymer in the aforementioned conditions, which would agree

4

with our interpretation of ELISA results on the possible competition for grafting

5

between the K5 peptide and both reductants at low concentrations (Figure 2.A).

6

As for TCEP, when incubating dextran-VS with 50 mM of the reducing agent

7

during 6 or 24 hours, the peaks corresponding to the VS carbon-carbon double

8

bond completely disappeared and the peak corresponding to the VS carbon-

9

carbon single bond appeared with a similar magnitude (Figure 4.A). The same

10

NMR spectra also featured a high intensity peak at 2.3-2.7 ppm, accounting for

11

12 hydrogen atoms that could not be attributed to the polymer. A peak at the

12

same displacement was also detected when incubating TCEP at the above-

13

determined optimal concentration for K5 peptide grafting with dextran-VS (100

14

µM), although the integration was too low to reliably estimate the reaction yield.

15

Two possible reaction products between VS and TCEP were hypothesized.

16

One possible reaction product is a phosphonium ion adduct based on Wang et 27 ACS Paragon Plus Environment

Page 29 of 70 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


1

al.’s work

2

ppm (6 H atoms closer to the phosphorus) and 3.25 ppm (6 H atoms closer to

3

the carboxylic acids), which strongly deviates from our data. The second

4

conjectured

5

schematized in Figure 4.A, with 12 atoms denoted e and highlighted in blue.

6

This structure is inspired by the work of Kantner et al. which revealed a

7

cyclization reaction involving the formation of an intermolecular covalent bond

8

between the phosphorus atom of TCEP and the terminal carbon of VS as well

9

as the formation of an intramolecular bond between the same phosphorus atom

48.

However, the peaks predicted for this product are reported at 2.78

product

contains

a

pentacoordinate

atom,

as

10

and one of the carboxylic acids oxygen atoms

11

of the pentacoordinate product over the phosphonium ion adduct was confirmed

12

by

13

accounting for 11.84 H atoms with a ca. 2.6 ppm displacement, as observed in

14

our NMR spectrum (Figure 4.A). In order to validate the formation of a

15

pentacoordinate product, NMR and LC-MS-MS analysis were performed on a

16

mixture of DVS and TCEP (without dextran). Of interest, NMR spectrum

infrared

spectroscopy,

and

their

NMR

49.

phosphorus

In their work, the formation

spectrum

displayed

one

peak



Page 30 of 70

1

analysis was clarified in the 3-4.5 ppm region, previously hidden by the

2

hydrogen atoms on the dextran (compare Figure 4 and Figure S1). When DVS

3

and DVS+TCEP spectrum were compared (Figure S1), we observed that the

4

peaks attributed to DVS disappeared for the DVS+TCEP mixture, and two

5

triplets appeared at 3.1 and 3.4 ppm. These two triplets are consistent with the

6

formation of a new bond between the pentacoordinate phosphorus atom of the

7

TCEP

8

electrospray ionization coupled to a high-resolution mass spectrometry, revealed

9

an abundant peak at m/z = 277.1 that was attributed to a fragment containing

10

the pentacoordinate phosphorus (Figure S2). Altogether, NMR and LC-MS-MS

11

analysis confirmed the structure of the conjectured product (Figure 4).

and

the

final

carbon

of

VS.

Furthermore,

liquid

chromatography

12

The results from the present study demonstrated that TCEP does react with

13

the thiol alkylating agent vinyl sulfone and cannot be merely said to be

14

compatible with thiol-ene chemistry. The mechanisms and products of reaction

15

between

16

clarification.

TCEP

and

VS

may

however

require

further

investigation

and


Page 31 of 70 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


1



Page 32 of 70

1

Optimization of one-pot thiol conjugation of oxidized K5 peptides with dextran-

2

vinyl sulfone

3

Despite the side-reaction between TCEP and vinyl sulfone, the phosphine

4

reductant was proven to dramatically enhance the conjugation efficiency of the

5

initially oxidized K5 peptide to dextran-VS within 6 hours in PBS, to a higher

6

extent than DTT and BME (ca. 16, 12 and 9-fold increase in signal between

7

PBS only and PBS supplemented with 100 µM TCEP, 100 µM DTT and 500

8

µM BME, respectively, Figure 2). The superiority of the phosphine group in

9

PBS could be explained by: (i) faster kinetics of K5 reduction by TCEP, (ii)

10

slower kinetics of competitive side-reactions between TCEP and VS and (iii) a

11

possible catalysis of thiol-VS Michael addition by TCEP

12

examined

13

specifically explored different buffering agents at various pH values to try to

14

improve grafting efficiency of the peptide on dextran-VS-modified surfaces

the

mechanisms

of

the

K5/VS/TCEP

48.

We therefore further

one-pot

incubation.

We


Page 33 of 70 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


1

(Figure 5) and studied the kinetics of K5/VS and TCEP/VS conjugation (Figure

2

6).

3

4 5

Figure 5. Grafting of K5 peptides on dextran-VS in presence or absence of 100 µM TCEP

6

in various buffers – Phosphate, HEPES, Tris and Borate Buffered Saline – as monitored by

7

ELISA.

8

Using 100 µM of TCEP in phosphate and HEPES buffered saline (pKa values

9

of 7.21 and 7.55 at 20°C, respectively), the ELISA signal obtained with 1 µM

10

of K5 peptides incubated on dextran-VS increased with pH between 6.5 and

11

7.5, with no significant difference between pH 7.5 and 8.5 (p > 0.66, see

12

Figure 5). A similar but offset trend was obtained with 100 µM of TCEP in tris



Page 34 of 70

1

buffered saline (TBS, pKa of 8.08 at 20°C), that is, an increase in signal

2

between pH 7.5 and 8.5 and then no statistical difference between pH 8.5 and

3

9.5 (p = 0.37). Conversely, the incubation of the K5 peptide in presence of

4

TCEP in borate buffered saline (pKa of 9.14 at 20°C) decreased dramatically

5

between pH 8.5 and 9.5 (p < 0.0001).

6

Regarding the levels of K5 grafting in absence of TCEP, a clear trend of

7

increased signal with increasing pH values was observed: 2.0 ± 0.5%, 3.2 ±

8

0.2%, 6 ± 1% and 14 ± 3% of the maximal value at pH 6,5, 7.5, 8.5 and 9.5

9

respectively. The elevated peptide density could be attributed to changes in the

10

adsorption level or, more probably, due to the undesirable reaction between VS

11

and residues other than cysteine, including the primary amino group of the

12

lysine side-chain and the aliphatic alcohol group of the serine side-chain

13

Altogether

the

data

seemed

to

indicate

that

the

maximal

50.

amount

of

14

immobilized K5 peptides is obtained with tris buffered saline at pH 8.5. The

15

increase in signal without TCEP however suggested non-specific covalent

16

grafting or adsorption. Moreover, the ELISA signal for that condition was not 33 ACS Paragon Plus Environment

Page 35 of 70 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


1

statistically different from the results obtained with HEPES buffered saline at pH

2

7.5 and 8.5 (p values of 0.07 and 0.16, respectively) and almost not

3

significantly different from those obtained with PBS at pH 7.5 (p = 0.049). Our

4

results can be explained by the higher instability of TCEP in PBS, when

5

compared to HEPES, tris and borate buffers

6

oxidation of TCEP was found to be the fastest in PBS at near neutral pH

7

values, although 72 hours were required to reach 57% oxidation at pH 7.4 in

8

150 mM PBS. We thus kept PBS as diluting buffer and investigated the

9

kinetics of the three reactions at hand, that is, reduction of the oxidized peptide

10

by TCEP, conjugation of the reduced peptide to dextran-VS and side-reaction

11

between TCEP and dextran-VS (Figure 6).

51.

In Han and Han’s report, the

12



Page 36 of 70

1 2

Figure 6. Kinetics of (A) K5 conjugation to dextran-VS in 10 mM PBS, pH 7.4, with or

3

without 100 µM TCEP, and kinetics of (B) TCEP oxidation in test tubes or in dextran-VS-

4

modified well plates, with or without 1 µM K5. Monitoring was performed by ELISA and

5

by reduction of Ellman’s reagent, respectively. The red dashed lines are guides for the eye

6

only whereas the black dashed lines are linear fits.

7 8

The data in Figure 6.A showed that, in presence of an initial concentration of

9

100 µM of TCEP, the grafting of the K5 peptide on dextran-VS continued over

10

a relatively long period of time. More precisely, the final K5 densities obtained

11

for the 1, 2, 4, 6 and 24-hour time points were all significantly different from

12

one another (p < 0.05), except for the 4 and 6-hour time points (p = 0.15,

13

filled squares). Conversely, although the ELISA response in absence of TCEP

14

slightly increased with time, there was no statistical difference in signal between 35 ACS Paragon Plus Environment

Page 37 of 70 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


1

4, 6 and 24 hours (p > 0.07, empty squares). The ratio between the K5

2

peptide density obtained in PBS only and the density obtained with PBS supplemented

3

with 100 µM of TCEP also consistently decreased over time and reached 5 ± 1% after 24 hours.

4

The data further demonstrated that the grafting via thiol-ene chemistry and/or

5

adsorption of the oxidized peptide in absence of reductant was not favourable

6

even after a long duration. This result was indicative of little rebalancing

7

between the oxidized and reduced forms of the peptide cysteine residues in

8

PBS without reductant, and thus indicative of the high stability of the coiled-coil

9

disulfide bond.

10 11

Given that the oxidized peptide could be grafted for at least 24 hours in

12

presence of TCEP, we examined the stability of the reducing agent in our

13

experimental conditions by quantifying its potential to reduce 5,5'-dithiobis-(2-

14

nitrobenzoic acid) (DTNB) into two 2-nitro-5-thiobenzoate molecules that adsorb

15

strongly at 405 nm

16

µM) was here neglected.

51.

The reduction of DTNB by the reactive K5 peptides (≤ 1



Page 38 of 70

1

When solutions of 100 µM TCEP in PBS, with or without 1 µM K5 peptide,

2

were incubated in polypropylene test tubes, DTNB reduction decreased very

3

slowly (● and ○, respectively, Figure 6.B). Without K5 peptide, 82 ± 2 µM of

4

active reductant were still detected after 10 days, which indicated that oxidation

5

of TCEP was almost negligible in 10 mM PBS, pH 7.4. When compared to the

6

results obtained by Han and Han

7

PBS, our data demonstrated that the phosphine can be fairly stable in a

8

phosphate buffered saline when the latter is not too concentrated. Interestingly,

9

when the same experiment was performed with DTT or BME, both reductants

10

oxidized much faster and active reductant could not be detected after 48 and

11

72 hours, respectively (Supporting Information, Figure S3). When 1 µM of K5

12

peptide was added to the 100 µM of TCEP in polypropylene test tubes, a

13

significant decrease in absorbance signal was measured starting from 48 hours,

14

and the active TCEP concentration after 10 days was evaluated at 49 ± 1 µM

15

(●, Figure 6.B). More precisely, curve-fitting revealed a linear decrease of

16

active TCEP concentration with steady rates of - 0.0568 and - 0.1889 µM.h-1,

51,

who used between 150 and 350 mM of


Page 39 of 70 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


1

in absence and presence of K5 peptides, respectively (R2 ≥ 0.98). The data

2

thus indicated that the K5 peptides contributed to the consumption of 0.1321

3

µM of TCEP per hour, which is equivalent to a 3.8-hour cycle of 1 µM K5

4

peptide oxidation and reduction (based on a consumption of 0.5 µM of TCEP

5

to form 1 µM of free thiol groups). This is further indicative of the fast oxidation

6

rate of the coiled-coil-forming K5 peptides, despite the electrostatic repulsion

7

expected from the lysine side-chains. The results are also in good agreement

8

with previous work, which showed more than 50% oxidation of the K5 peptide

9

within 150 minutes, in 10 mM PBS, pH 7.4, at 25°C

24.

10 11

Interestingly, when the same solutions (100 µM TCEP in PBS, with or without

12

1 µM K5 peptide) were incubated in dextran-VS well-plates, a dramatically

13

faster decrease in active reductant concentration was observed (■ and □,

14

respectively, Figure 6.B). A drop from 100 µM to ca. 67 µM was measured

15

within the first 6 hours, and complete disappearance of active TCEP occurred

16

within 4.5 days. In this time frame, the co-incubation with 1 µM of K5 peptide 38 ACS Paragon Plus Environment


Page 40 of 70

1

did not significantly influence DTNB reduction. Considering the data obtained in

2

the test tubes, the decrease in active TCEP concentration that was observed in

3

dextran-VS wells was not attributed to the phosphine oxidation. It was much

4

more likely due to TCEP reacting with the VS moieties, as previously observed

5

(see Figure 4 and discussion thereof). Within 6 hours, ca. 33 µM of the

6

phosphine covalently reacted with vinyl sulfone, which product we conjectured

7

to be a pentacoordinate conjugate (cf. scheme in Figure 4.A). A small amount

8

of TCEP (≤ 1 µM) could also have been consumed to catalyze the reaction of

9

K5 with VS

48.

It is here worth mentioning that the reaction product obtained by

10

incubating TCEP with the VS-modified polymer does not contain any reactive

11

moieties other than carboxylic groups. The incubation of DTT or BME in the

12

dextran-VS well-plates also led to a dramatic decrease in active reductant

13

concentration, faster than TCEP (complete reaction within 24 hours, Supporting

14

Information, Figure S3). The drop was particularly fast for β-mercaptoethanol:

15

95% of the reductant reacted with VS within 6 hours. The kinetics we observed

16

could further explain why TCEP improved the grafting of K5 on dextran-VS to 39 ACS Paragon Plus Environment

Page 41 of 70 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


1

an higher extent than the thiol-containing reducing agents, especially BME

2

(Figure 2.A).



1

2

Page 42 of 70

CONCLUSION

The present work demonstrated the practicality of the concomitant reduction of

3

an

oxidized/fast-oxidizing

4

polysaccharide via thiol-based chemistry. Using dextran that was previously

5

modified with either VS, maleimide or pyridyldithiol moieties and a reductant

6

among TCEP, DTT or BME, a dramatic increase in fully oxidized K5 peptide

7

conjugation

8

concentration of the reducing agent was adequately chosen. The VS/TCEP

9

combination provided the best results. Further investigation shed light on

10

unclear statements in the literature. Our data indeed confirmed that TCEP does

11

react with thiol-alkylating agents, although it can be used as a reductant if the

12

alkylating agent is in excess. Moreover, the side-reaction between TCEP and

13

VS is limited to the generation of carboxyl moieties on the polymer. We here

14

also demonstrated that TCEP is stable in the commonly used Dulbecco’s

15

Phosphate Buffered Saline (10 mM phosphate, pH 7.4).

was

peptide

demonstrated

in

and

7

out

its

of

covalent

9

cases,

conjugation

provided

to

that

a

the


Page 43 of 70 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


1



1

EXPERIMENTAL PROCEDURES

2

Materials

Page 44 of 70

3

Milli-Q grade water (18.2 MΩ.cm, referred to here as water) was generated

4

with a Millipore Gradient A 10 purification system (Etobicoke, ON). CellBIND

5

microplates

6

poly(allylamine)

7

(Warrington, PA). 500-kDa dextran (technical grade T) was purchased from

8

Pharmacosmos A/S (Holbaek, Denmark). EMCH (3,3’-N-[ε-maleimidocaproic acid]

9

hydrazide,

were

obtained

(PAAm,

trifluoroacetic

25

from kDa)

acid

salt)

Corning was

Inc.

(Corning,

obtained

and

from

PDPH

NY).

Linear

Polysciences,

Inc.

(3-(2-pyridyldithio)propionyl

10

hydrazide) were purchased from Pierce Biotechnology (Rockford, IL). Sodium

11

hydroxide (NaOH, 99.3% purity) and hydrochloric acid (HCl, 37.7% v/v) were

12

obtained from VWR International, Ltd. (Mont-Royal, QC). Deuterium oxide (99%

13

purity) was purchased from Cambridge Isotope Laboratories, Inc. (Andover,

14

MA).

15

MilliporeSigma

Unless

mentioned

otherwise,

(Oakville,

ON).

all

chemicals

Anti-human

EGF

were

ELISA

purchased DuoSet

kit

from was 43


Page 45 of 70 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


1

purchased from R&D systems (Minneapolis, MN). All absorbance measurements

2

were carried out on a Victor V Multilabel Counter from PerkinElmer Inc.

3

(Woodbridge, ON).

4 5

Biomolecule and bioconjugate synthesis, purification and characterization

6

Peptides and Proteins

7

Cysteine-tagged K5 peptides (CGG-(KVSALKE)5) were synthesized at the

8

peptide facility at the University of Colorado, Denver (> 98% purity). Chimeric

9

protein E5-EGF, consisting of an epidermal growth factor protein fused to a N-

10

terminal E5 tag (EVSALEK)5, was produced by transient transfection of HEK

11

293-6E cells, purified by immobilized metal-ion affinity and size-exclusion

12

chromatographies,

13

performed by ELISA and all peptides and proteins were stored at -80°C until

14

use.

15

as

previously

described

20.

Protein

quantification

was

Dextran-Vinyl Sulfone



1

Page 46 of 70

Technical grade T dextran (500 kDa) was modified with divinyl sulfone (DVS)

2

as previously described

3

of Milli-Q water containing 30 mM of NaOH. 13.5 mL of a freshly prepared

4

solution of DVS (0.11 M in water containing 30 mM NaOH) was added to the

5

dextran

6

temperature under vigorous vortexing and was quenched by adding 0.3% v/v of

7

pure HCl (12.1 M). The preparations were then desalted against water five

8

times

9

MilliporeSigma). Purified dextran-VS was lyophilized and stored at 4°C until use.

solution.

using

The

10-kDa

21.

Briefly, 450 mg of dextran were dissolved in 9 mL

reaction

was

centrifugal

carried

filter

out

devices

for

3

minutes

(Amicon

at

Ultra-15

room

from

10

1H

NMR was performed on the purified preparation and peak integration

11

revealed that ca. 33 % of the saccharide units were modified with vinyl sulfone.

12

Carboxymethylated Dextran

13

Dextran carboxymethylation and characterization were performed as previously

14

described

52.

Briefly, 400 mg of dextran was dissolved in 10 mL of a solution

15

containing 3 M NaOH and 1 M monochloroacetic acid. The solution was stirred

16

for 2 hours at room temperature and the reaction was quenched by the 45 ACS Paragon Plus Environment

Page 47 of 70 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


1

addition of 40 mg of NaH2PO4 followed by pH adjustment to neutrality using 18

2

M H2SO4. The solution was then filtered through a 0.2-µm PTFE filter and

3

dialyzed five times against Milli-Q water for 1 hour to remove reagents and

4

salts. Purified carboxymethylated dextran was lyophilized and stored at 4°C until

5

use. 1H NMR revealed that ca. 46% of the saccharide units were modified with

6

carboxyl moieties.

7 8

9 10

Thiol-Reactive Surface Preparation Amination Carboxylated CellBIND® 96-well microplates were used to prepare a reactive

11

surface for thiol-based K5 peptide grafting as previously described

12

the

13

carbodiimide chemistry by exposing the wells to 50 µL of a solution containing

14

380 µM PAAm, 40 mM EDC and 10 mM NHS in 100 mM MES buffer pH 4.7

15

for 22 hours. The wells were rinsed three times with Dulbecco’s Phosphate

surfaces

were

first

modified

with

poly(allylamine)

20.

(PAAm)

Briefly, through



Page 48 of 70

1

Buffered Saline (PBS; 10 mM with 137 mM NaCl, 2.7 mM KCl, pH 7.4). Free

2

amines on the surface were then used to covalently graft a sulfhydryl reactive

3

moiety.

4

Vinyl Sulfone

5

Dextran-VS was dissolved at 10 mg/mL in 50 mM borate pH 10. 50 µL of

6

the solution were added to each PAAm-coated well for 16 hours, which was

7

found to be the optimal incubation duration (data not shown). The wells were

8

rinsed three times with PBS before K5 incubation.

9

Pyridyldisulfide and Maleimide

10

Carboxymethylated dextran (CMD) was dissolved at 2 mg/mL in MilliQ water

11

containing 50 mM EDC and 12.5 mM NHS. 50 µL of the NHS-activated CMD

12

solution were added to each PAAm-coated well for 16 h. The wells were rinsed

13

three times with PBS and three times with MilliQ water. The free carboxyl

14

groups on CMD were then reactivated by a solution containing 167 mM EDC,

15

42 mM NHS and 17 mM MES for 10 minutes. The solution was discarded

16

before adding 50 µL of EMCH or PDPH (heterobifunctional linkers that both 47 ACS Paragon Plus Environment

Page 49 of 70 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


1

feature a hydrazide moiety as well as a maleimide and a pyridyldithiol moiety,

2

respectively). The linkers were incubated for 1 hour at a concentration of 1 mM

3

in PBS containing 10% v/v DMSO. The wells were rinsed three times with

4

PBS. 50 µL of ethanolamine (100 mM, pH 8.5) was then added to each well to

5

deactivate the remaining COO-NHS groups for 15 minutes. The wells were

6

finally rinsed three times with PBS before K5 incubation.

7 8

9

K5 Conjugation and Characterization

Thiol-based

grafting

of

cysteine-tagged

K5

peptides

was

performed

by

10

incubating, in each well, 50 µL of a solution containing 1 µM of K5 peptide,

11

unless otherwise mentioned, and reducing agents (DTT, BME or TCEP) at

12

concentrations ranging from 0 to 10 mM in PBS at room temperature, for 6

13

hours unless otherwise indicated. Other conditions include 10 mM buffered

14

saline (150 mM NaCl) of HEPES, borate and tris base with adjustment to

15

indicated pH with concentrated HCl or NaOH solutions. The surfaces were then



Page 50 of 70

1

rinsed three times with PBS and remaining sulfhydryl reactive moieties were

2

blocked by incubating 50 mM of BME for 30 min (for VS and EMCH only)

3

before rinsing three times with PBS.

4

The amount of grafted peptide was measured by its capacity to capture its

5

biological partner, E5. The nonspecific binding sites on the surfaces were first

6

blocked with 10% fetal bovine serum in PBS prior to a 1-h incubation of 500

7

pM E5-tagged epidermal growth factor (E5-EGF) in PBS supplemented with

8

0.1% v/v bovine serum albumin (PBS-BSA). The amount of immobilized EGF

9

was then quantified by a direct enzyme-linked immunosorbent assay (ELISA)

10

according to a protocol adapted from the manufacturer’s instructions. Briefly, the

11

surfaces were incubated with 50 µL of biotinylated anti-EGF antibody in PBS-

12

BSA for 30 minutes, then with 50 µL of horseradish-streptavidin conjugate in

13

PBS-BSA for 20 minutes. The wells were rinsed three times with PBS

14

supplemented with 0.05% Tween 20 between the different steps. Finally, 50 µL

15

of

16

tetramethylbenzidine) was added to each well and substrate oxidation was

the

substrate

solution

(50:50

v/v

mixture

of

hydrogen

peroxide

and


Page 51 of 70 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


1

monitored

for

10

2

Absorbance slopes were compiled and are referred to as ELISA signal in the

3

manuscript.

4

NMR Analysis

5

Dextran-VS

was

min

by

repetitious

dissolved

at

a

absorbance

concentration

readings

of

10

at

mg/mL

630

in

nm.

PBS

6

supplemented with either TCEP, DTT or BME at the indicated concentration

7

(0.1 mM or 50 mM). The reaction was carried out at room temperature for 6 to

8

24 hours. The samples were then purified against Milli-Q water using 10-kDa

9

centrifugal filters and freeze dried. The samples were examined by 1H NMR as

10

5

mg/mL

solutions

in

D2O.

11

(http://www.nmrdb.org/new_predictor/)

12

prediction.

The

EPFL

was

used

1H

for

NMR H

atom

Predict

tool

displacement

13 14

Thiol and reductant quantification



Page 52 of 70

1

The concentrations of free thiol groups of the K5 peptide and of active

2

reductant were measured using Ellman's reagent (5,5'-dithiobis-(2-nitrobenzoic

3

acid), DTNB)

4

µL of DTNB (1 mg/mL in PBS supplemented with 1 mM EDTA) and 10 µL of

5

Tween 20 (0.05% v/v in water). The concentration of the 2-nitro-5-thiobenzoate

6

product was measured in 96-well plates at an absorbance of 405 nm.

51.

Briefly, 140 µL of sample diluted in PBS were mixed with 60

7

8

Data Treatment

9

All data are represented as the mean value ± standard deviation and, when

10

indicated, statistical analysis was carried out by independent two-sample t-test,

11

with (in)equality of variances based on an F-test.

12


Page 53 of 70 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

1


ACKNOWLEDGMENTS

2

This work was supported by the Canada Research Chairs on Protein-

3

enhanced Biomaterials (G.D.C) and on Bioinspired Materials (X.B.), by the

4

Natural Sciences and Engineering Research Council of Canada Discovery Grant

5

(G.D.C and X.B.), by the Biomedical Science and Technology Research Group

6

(G.T.,

7

TransMedTech Institute (A.G.A., B.L. and F.M.) and its main funding partner,

8

the Canada First Research Excellence Fund. We would like to thank S.

9

Bilodeau for NMR measurements. The authors are grateful to Dr. Alexandra

10

Fürtös at the Department of Chemistry, Université de Montréal for her helpful

11

advice on LC-MS analyses. We thank the Canadian Foundation for Innovation

12

for the support for the mass spectrometry instrumentation.

13

CONFLICT OF INTEREST STATEMENT

14

The authors declare no competing financial interest.

B.L.,

F.M.

and

G.D.C.).

The

present

work

was

funded

by

the



Page 54 of 70

1

SUPPORTING INFORMATION

2

NMR and LC-MS-MS analysis of a mixture of DVS+CEP. Stability of DTT and

3

BME in PBS or in dextran-VS-modified well plate.

4

AUTHOR INFORMATION

5

Corresponding Authors

6

*E-mail: [email protected];

7

*E-mail: [email protected]

8 9

ORCID

10

Xavier Banquy: 0000-0002-3342-3179

11

Gregory De Crescenzo: 0000-0002-6280-1570

12

Frederic Murschel: 0000-0002-6735-8934

13 14

ABBREVIATIONS


Page 55 of 70 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


1

BME, β-mercaptoethanol; CMD, carboxymethylated dextran; Dex, dextran;

2

DTT, dithiothreitol; (D)VS, (di)vinylsulfone; EGF, epidermal growth factor; ELISA,

3

enzyme-linked immunosorbent assay; PBS, phosphate buffered saline; TCEP,

4

tris(carboxyethyl)phosphine

5



1

REFERENCES

2

(1)

Braun, A. C., Gutmann, M., Luhmann, T., and Meinel, L. (2018) Bioorthogonal

3

Strategies for Site-Directed Decoration of Biomaterials with Therapeutic Proteins. J.

4

Control. Release 273, 68-85.

5

(2)

6 7

Page 56 of 70

McKay, C. S., and Finn, M. G. (2014) Click Chemistry in Complex Mixtures: Bioorthogonal Bioconjugation. Chem. Biol. 21, 1075-1101.

(3)

Zou, Y., Zhang, L., Yang, L., Zhu, F., Ding, M., Lin, F., Wang, Z., and Li, Y. (2018)

8

"Click" Chemistry in Polymeric Scaffolds: Bioactive Materials for Tissue Engineering. J.

9

Control. Release 273, 160-179.

10

(4)

11 12

Thapa, P., Zhang, R. Y., Menon, V., and Bingham, J. P. (2014) Native Chemical Ligation: a Boon to Peptide Chemistry. Molecules 19, 14461-14483.

(5)

Riahi, N., Cappadocia, L., Henry, O., Omichinski, J., and De Crescenzo, G. (2016)

13

Soluble Expression, Purification and Functional Characterization of a Coil Peptide

14

Composed of a Positively Charged and Hydrophobic Motif. Amino Acids 48, 567-577.

15

(6)

16 17 18

Hermanson, G. T. (2013) Functional Targets for Bioconjugation. Bioconjugate Techniques (Third edition) pp 127-228, Chapter 2, Academic Press, Boston.

(7)

Dondoni, A. (2008) The Emergence of Thiol-ene Coupling as a Click Process for Materials and Bioorganic Chemistry. Angew. Chem. Int. Ed. Engl. 47, 8995-8997.


Page 57 of 70 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

1


(8)

2 3

Connect Polymers with Nature’s Building Blocks. ACS Macro Letters 2, 14-18. (9)

4 5

Stenzel, M. H. (2012) Bioconjugation Using Thiols: Old Chemistry Rediscovered to

Thornton, J. M. (1981) Disulphide Bridges in Globular Proteins. J Mol Biol 151, 261287.

(10)

Riahi, N., Murschel, F., Lerouge, S., Durocher, Y., Henry, O., and De Crescenzo, G.

6

(2017) Bioavailability of Immobilized Epidermal Growth Factor: Covalent Versus

7

Noncovalent Grafting. Biointerphases 12, 010501.

8

(11)

9

Immobilization of Active Vascular Endothelial Growth Factor via a Cysteine-Containing

10 11

Backer, M. V., Patel, V., Jehning, B. T., Claffey, K. P., and Backer, J. M. (2006) Surface

Tag. Biomaterials 27, 5452-5458. (12)

Thomas, F., Boyle, A. L., Burton, A. J., and Woolfson, D. N. (2013) A Set of de Novo

12

Designed Parallel Heterodimeric Coiled Coils with Quantified Dissociation Constants in

13

the Micromolar to Sub-nanomolar Regime. J. Am. Chem. Soc. 135, 5161-5166.

14

(13)

Monera, O. D., Sereda, T. J., Zhou, N. E., Kay, C. M., and Hodges, R. S. (1995)

15

Relationship of Sidechain Hydrophobicity and Alpha-Helical Propensity on the Stability

16

of the Single-Stranded Amphipathic Alpha-Helix. J. Pept. Sci. 1, 319-329.

17

(14)

Fletcher, J. M., Boyle, A. L., Bruning, M., Bartlett, G. J., Vincent, T. L., Zaccai, N. R.,

18

Armstrong, C. T., Bromley, E. H. C., Booth, P. J., Brady, R. L., et al. (2012) A Basis Set

19

of de Novo Coiled-Coil Peptide Oligomers for Rational Protein Design and Synthetic

20

Biology. ACS Synthetic Biology 1, 240-250. 56 ACS Paragon Plus Environment


1

(15)

Page 58 of 70

Gerling-Driessen, U. I., Mujkic-Ninnemann, N., Ponader, D., Schone, D., Hartmann, L.,

2

Koksch, B., Gerling-Driessen, U. I., Schone, D., Koksch, B., Ponader, D., et al. (2015)

3

Exploiting Oligo(Amido Amine) Backbones for the Multivalent Presentation of Coiled-

4

Coil Peptides. Biomacromolecules 16, 2394-2402.

5

(16)

Fortier, C., De Crescenzo, G., and Durocher, Y. (2013) A Versatile Coiled-Coil Tethering

6

System for the Oriented Display of Ligands on Nanocarriers for Targeted Gene Delivery.

7

Biomaterials 34, 1344-1353.

8

(17)

Pechar, M., Pola, R., Laga, R., Braunová, A., Filippov, S. K., Bogomolova, A.,

9

Bednárová, L., Vaněk, O., and Ulbrich, K. (2014) Coiled Coil Peptides and Polymer–

10

Peptide Conjugates: Synthesis, Self-Assembly, Characterization and Potential in Drug

11

Delivery Systems. Biomacromolecules 15, 2590-2599.

12

(18)

Jing, P., Rudra, J. S., Herr, A. B., and Collier, J. H. (2008) Self-Assembling Peptide-

13

Polymer Hydrogels Designed from the Coiled Coil Region of Fibrin. Biomacromolecules

14

9, 2438-2446.

15

(19)

Aronsson, C., Danmark, S., Zhou, F., Oberg, P., Enander, K., Su, H., and Aili, D. (2015)

16

Self-Sorting Heterodimeric Coiled Coil Peptides with Defined and Tuneable Self-

17

Assembly Properties. Sci. Rep. 5, 14063.

18

(20)

Murschel, F., Fortier, C., Jolicoeur, M., Hodges, R. S., and De Crescenzo, G. (2017) Two

19

Complementary Approaches for the Controlled Release of Biomolecules Immobilized via

20

Coiled-Coil Interactions: Peptide Core Mutations and Multivalent Presentation.

21

Biomacromolecules 18, 965-975. 57 ACS Paragon Plus Environment

Page 59 of 70 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

1


(21)

Noel, S., Fortier, C., Murschel, F., Belzil, A., Gaudet, G., Jolicoeur, M., and De

2

Crescenzo, G. (2016) Co-Immobilization of Adhesive Peptides and VEGF Within a

3

Dextran-Based Coating for Vascular Applications. Acta Biomater. 37, 69-82.

4

(22)

Addi, C., Murschel, F., and De Crescenzo, G. (2017) Design and Use of Chimeric

5

Proteins Containing a Collagen-Binding Domain for Wound Healing and Bone

6

Regeneration. Tissue Eng Part B Rev 23, 163-182.

7

(23)

Bagiyan, G. A., Koroleva, I. K., Soroka, N. V., and Ufimtsev, A. V. (2003) Oxidation of

8

Thiol Compounds by Molecular Oxygen in Aqueous Solutions. Russ. Chem. Bull. 52,

9

1135-1141.

10

(24)

Noel, S. (2015) Développement de Revêtements Polymériques Bioactifs pour Prothèses

11

Vasculaires Synthétiques. Institut de génie biomédical, 266. École polytechnique de

12

Montréal.

13

(25)

Visser, C. C., Voorwinden, L. H., Harders, L. R., Eloualid, M., van Bloois, L.,

14

Crommelin, D. J., Danhof, M., and de Boer, A. G. (2004) Coupling of Metal Containing

15

Homing Devices to Liposomes via a Maleimide Linker: Use of TCEP to Stabilize Thiol-

16

Groups Without Scavenging Metals. J. Drug Target. 12, 569-573.

17

(26)

ThermoScientific. (2013) User Guide: TCEP HCl. Pierce Biotechnology Inc., Rockford

18

(IL). https://assets.thermofisher.com/TFS-

19

Assets/LSG/manuals/MAN0011306_TCEP_HCl_UG.pdf.



1

(27)

Oude Blenke, E. E., van den Dikkenberg, J., van Kolck, B., Kros, A., and Mastrobattista,

2

E. (2016) Coiled Coil Interactions for the Targeting of Liposomes For Nucleic Acid

3

Delivery. Nanoscale 8, 8955-8965.

4

(28)

Zwyssig, A., Schneider, E. M., Zeltner, M., Rebmann, B., Zlateski, V., Grass, R. N., and

5

Stark, W. J. (2017) Protein Reduction and Dialysis-Free Work-Up through Phosphines

6

Immobilized on a Magnetic Support: TCEP-Functionalized Carbon-Coated Cobalt

7

Nanoparticles. Chemistry (Easton) 23, 8585-8589.

8

(29)

9

Henkel, M., Rockendorf, N., and Frey, A. (2016) Selective and Efficient Cysteine Conjugation by Maleimides in the Presence of Phosphine Reductants. Bioconjug. Chem.

10 11

27, 2260-2265. (30)

Kantner, T., Alkhawaja, B., and Watts, A. G. (2017) In Situ Quenching of

12

Trialkylphosphine Reducing Agents Using Water-Soluble PEG-Azides Improves

13

Maleimide Conjugation to Proteins. ACS Omega 2, 5785-5791.

14

(31)

15 16

Page 60 of 70

Sanchez, A., Pedroso, E., and Grandas, A. (2013) Oligonucleotide Cyclization: the ThiolMaleimide Reaction Revisited. Chem. Commun. (Camb.) 49, 309-311.

(32)

Armstrong, A. E., Zerbes, R., Fournier, P. A., and Arthur, P. G. (2011) A Fluorescent

17

Dual Labeling Technique for the Quantitative Measurement of Reduced and Oxidized

18

Protein Thiols in Tissue Samples. Free Radic. Biol. Med. 50, 510-517.


Page 61 of 70 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

1


(33)

Liberelle, B., Bartholin, L., Boucher, C., Murschel, F., Jolicoeur, M., Durocher, Y.,

2

Merzouki, A., and De Crescenzo, G. (2010) New ELISA Approach Based on Coiled-Coil

3

Interactions. J. Immunol. Methods 362, 161-167.

4

(34)

Shafer, D. E., Inman, J. K., and Lees, A. (2000) Reaction of Tris(2-

5

carboxyethyl)phosphine (TCEP) with Maleimide and Alpha-Haloacyl Groups:

6

Anomalous Elution of TCEP by Gel Filtration. Anal Biochem 282, 161-164.

7

(35)

8 9

Free Radic. Biol. Med. 80, 148-157. (36)

10 11

Riemer, J., and Herrmann Johannes, M. (2011) 6 Protein Import into the Intermembrane Space of Mitochondria, in Methods in Protein Biochemistry.

(37)

12 13

Poole, L. B. (2015) The Basics of Thiols and Cysteines in Redox Biology and Chemistry.

Lukesh, J. C., 3rd, Palte, M. J., and Raines, R. T. (2012) A Potent, Versatile DisulfideReducing Agent from Aspartic Acid. J Am Chem Soc 134, 4057-4059.

(38)

Tajc, S. G., Tolbert, B. S., Basavappa, R., and Miller, B. L. (2004) Direct Determination

14

of Thiol pKa by Isothermal Titration Microcalorimetry. J Am Chem Soc 126, 10508-

15

10509.

16

(39)

Kortemme, T., and Creighton, T. E. (1995) Ionisation of Cysteine Residues at the

17

Termini of Model Alpha-Helical Peptides. Relevance to Unusual Thiol pKa Values in

18

Proteins of the Thioredoxin Family. J Mol Biol 253, 799-812.

19 20

(40)

Kantner, T. (2015) Bioconjugation Strategies Through Thiol-Alkylation of Peptides and Proteins. Doctor of Philosophy (PhD), 295. University of Bath. 60 ACS Paragon Plus Environment


1

(41)

Schumacher, F. F., Nobles, M., Ryan, C. P., Smith, M. E., Tinker, A., Caddick, S., and

2

Baker, J. R. (2011) In Situ Maleimide Bridging of Disulfides and a New Approach to

3

Protein PEGylation. Bioconjug. Chem. 22, 132-136.

4

(42)

Page 62 of 70

Maret, B., Regnier, T., Rossi, J. C., Garrelly, L., Vial, L., and Pascal, R. (2014)

5

Reduction with Tris(2-carboxyethyl)phosphine (TCEP) Enables the Use of an S-

6

Sulphonate Protecting Group for Thiol-Mediated Bioconjugation. Rsc Advances 4, 7725-

7

7728.

8

(43)

9

of Thiol Terminated Acrylate Polymers with Divinyl Sulfone to Generate Well-Defined

10 11

Grover, G. N., Alconcel, S. N., Matsumoto, N. M., and Maynard, H. D. (2009) Trapping

Semi-Telechelic Michael Acceptor Polymers. Macromolecules 42, 7657-7663. (44)

Zhang, L., Vila, N., Klein, T., Kohring, G. W., Mazurenko, I., Walcarius, A., and

12

Etienne, M. (2016) Immobilization of Cysteine-Tagged Proteins on Electrode Surfaces by

13

Thiol-Ene Click Chemistry. ACS Appl. Mater. Interfaces 8, 17591-17598.

14

(45)

Jones, M. W., Mantovani, G., Ryan, S. M., Wang, X., Brayden, D. J., and Haddleton, D.

15

M. (2009) Phosphine-Mediated One-Pot Thiol-Ene "Click" Approach to Polymer-Protein

16

Conjugates. Chem. Commun. (Camb.), 5272-5274.

17

(46)

Nair, D. P., Podgorski, M., Chatani, S., Gong, T., Xi, W. X., Fenoli, C. R., and Bowman,

18

C. N. (2014) The Thiol-Michael Addition Click Reaction: A Powerful and Widely Used

19

Tool in Materials Chemistry. Chem. Mater. 26, 724-744.


Page 63 of 70 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

1


(47)

2 3

Adducts in Reducing Environments. Bioconjug. Chem. 22, 1946-1953. (48)

4 5

Baldwin, A. D., and Kiick, K. L. (2011) Tunable Degradation of Maleimide-Thiol

Wang, C., and Qi, C. (2013) Mechanistic Insights into N- or P-Centered Nucleophile Promoted Thiol–Vinylsulfone Michael Addition. Tetrahedron 69, 5348-5354.

(49)

Kantner, T., and Watts, A. G. (2016) Characterization of Reactions between Water-

6

Soluble Trialkylphosphines and Thiol Alkylating Reagents: Implications for Protein-

7

Conjugation Reactions. Bioconjug. Chem. 27, 2400-2406.

8

(50)

Morales-Sanfrutos, J., Lopez-Jaramillo, J., Ortega-Munoz, M., Megia-Fernandez, A.,

9

Perez-Balderas, F., Hernandez-Mateo, F., and Santoyo-Gonzalez, F. (2010) Vinyl

10

Sulfone: a Versatile Function for Simple Bioconjugation and Immobilization. Org.

11

Biomol. Chem. 8, 667-675.

12

(51)

Han, J. C., and Han, G. Y. (1994) A Procedure for Quantitative Determination of Tris(2-

13

carboxyethyl)phosphine, an Odorless Reducing Agent more Stable and Effective than

14

Dithiothreitol. Anal Biochem 220, 5-10.

15

(52)

Fortier, C., Louvier, E., Durocher, Y., and De Crescenzo, G. (2015) Tailoring the Surface

16

of a Gene Delivery Vector with Carboxymethylated Dextran: A Systematic Analysis.

17

Biomacromolecules 16, 1671-1681.

18 19 62 ACS Paragon Plus Environment


1

Page 64 of 70

TABLE OF CONTENTS GRAPHIC

2


Page 65 of 70 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. Reaction schemes of one-pot thiol chemistry for the grafting of the K5 peptide (R2-SH) to dextran (R1) in presence of reducing agents. (A) Reactions between the K5 peptide and dextran modified with vinyl sulfone and maleimide moieties for thiol-alkylation or with pyridyldithiol moieties for thiol-disulfide interchange. (B) Mechanism of oxidized K5 peptide reduction by a phosphine, tris(2-carboxyethyl)phosphine (TCEP), or by thiol-containing reductants, dithiothreitol (DTT) and β-mercaptoethanol (BME). 335x555mm (300 x 300 DPI)



Figure 2. Grafting of oxidized K5 peptide in PBS (10 mM, pH 7.4) co-incubated with reducing agents on (A) vinyl sulfone, (B) maleimide and (C) pyridyldithiol-derivatized dextran surfaces, as monitored by ELISA. The lines are guides for the eyes only. 84x140mm (300 x 300 DPI)


Page 66 of 70

Page 67 of 70 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 3. Influence of peptide and TCEP concentration on the grafting of oxidized K5 peptide in PBS (10 mM, pH 7.4), on vinyl sulfone-derivatized dextran surfaces, as monitored by ELISA. The data are relative to the ELISA signal obtained for 1 µM of K5 co-incubated with 100 µM of TCEP. The lines are guides for the eyes only. 69x56mm (600 x 600 DPI)



Figure 4. 1H NMR spectroscopy of vinyl sulfone-derivatized dextran incubated for 6 or 24 hours with (A) TCEP, (B) DTT and (C) BME at 0.1 and 50 mM. H atoms of interest are highlighted on the schematic compounds on the right. % values in insets are NMR areas under curve divided by the number of equivalent H atoms, relative to dextran saccharide units (calculated from area under curve of the H atoms labelled “c”). 177x189mm (300 x 300 DPI)


Page 68 of 70

Page 69 of 70 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 5. Grafting of K5 peptides on dextran-VS in presence or absence of 100 µM TCEP in various buffers – Phosphate, HEPES, Tris and Borate Buffered Saline – as monitored by ELISA. 84x69mm (300 x 300 DPI)



Figure 6. Kinetics of (A) K5 conjugation to dextran-VS in 10 mM PBS, pH 7.4, with or without 100 µM TCEP, and kinetics of (B) TCEP oxidation in test tubes or in dextran-VS-modified well plates, with or without 1 µM K5. Monitoring was performed by ELISA and by reduction of Ellman’s reagent, respectively. The red dashed lines are guides for the eye only whereas the black dashed lines are linear fits. 176x71mm (300 x 300 DPI)


Page 70 of 70

Adequate Reducing Conditions Enable Conjugation of Oxidized

Recommend Documents