ON SOME PRACTICAL ASPECTS OF CLICK THIOL-ENE … · Click chemistry reactions have been used since then to produce dendrimers [2-4], to modify polymers [5-7] and to conjugate biomolecules

Global Journal of Pure and Applied Chemistry Research

Vol.6, No.1, pp.9-40, March 2018

___Published by European Centre for Research Training and Development UK (www.eajournals.org)

9

Print ISSN: 2055-0073(Print), Online ISSN: 2055-0081(Online)

ON SOME PRACTICAL ASPECTS OF CLICK THIOL-ENE REACTIONS FOR

POLYMER MODIFICATION AND CONJUGATION WITH BIOMOLECULES

Débora Vieira Way, Pedro Ivo T. R. Barbosa, José Carlos Pinto

Programa de Engenharia Química / COPPE, Universidade Federal do Rio de Janeiro, Cidade

Universitária, CP:68502, Rio de Janeiro – 21941-972 RJ, Brazil

ABSTRACT: Thiol-ene reactions are among the most popular click reactions, being

particularly appealing for functionalization of polymer materials and conjugation with

proteins. However, although the formal definition of "click chemistry" can be very strict, one

can easily arise some controversial issues related to the application of thiol-ene reactions in

actual process conditions. In order to illustrate some of these issues, thiol-ene reactions were

performed at conditions that simulate the conjugation of proteins with unsaturated polymers.

It is shown that the presence of a heterogeneous reacting system can exert enormous impact

on the performance of thiol-ene reactions, which has been largely overlooked in the literature.

It is also shown that the performance of heterogeneous click reactions depends on the many

constituents of the reacting system so that the design of click reactions requires detailed

analysis of the desired reaction system and may not necessarily constitute a mechanistic route

for immediate bioconjugation applications.

KEYWORDS: Click Chemistry, Thiol-Ene, Polymer, Bioconjugation, FTIR.

INTRODUCTION

Click chemistry reactions have been studied extensively for 20 years, but were formally defined

for the first time by KOLB and coworkers in 2001 [1]. These authors established that click

reactions must necessarily be “modular, wide in scope, give very high yields, generate only

inoffensive byproducts that can be removed by nonchromatographic methods, and be

stereospecific (but not necessarily enantioselective). The required process characteristics

include simple reaction conditions (ideally, the process should be insensitive to oxygen and

water), readily available starting materials and reagents, the use of no solvent or a solvent that

is benign (such as water) or easily removed, and simple product isolation.”[1].

Click chemistry reactions have been used since then to produce dendrimers [2-4], to modify

polymers [5-7] and to conjugate biomolecules to different substrates [8-12]. In the particular

case of polymer applications, the use of click chemistry reactions to modify the surface of

polymer particles and to conjugate proteins with polymer chains is appealing, given the

expected high yields and specificity of the chemical reactions.

The most commonly studied click chemistry reactions are: (i) Huisgen reactions; (ii) copper-

free azide-alkyne cicloadditions; (iii) Diels-Alder reactions, (iv) thiol-ene and thiol-yne

reactions; (v) condensation of aldehydes and ketones with heteroatom-bound amines; and (vi)

Staudinger ligation of azides and phosphines [13]. Each distinct click reaction presents

characteristic advantages and disadvantages that are briefly discussed below in the context of

polymer modification and conjugation with proteins.

http://www.eajournals.org/


Vol.6, No.1, pp.9-40, March 2018


10


CuAAc reaction, also known as the Huisgen reaction, is undoubtedly the most studied click

reaction. The use of a copper catalyst to perform the reaction constitutes simultaneously

important advantage and disadvantage of this click reaction [12]. For example, the use of a

copper catalyst makes the reaction 107 times faster than the respective copper-free azide-alkyne

cicloaddition. Besides, the use of the copper catalyst can enhance both reaction yields and

product selectivity by orders of magnitude, when compared to the uncatalyzed reaction [14,15].

Copper, however, is associated to many diseases, including hepatitis, Alzheimer’s and other

neurological diseases and cannot be removed from the product [16]. As a consequence, the use

of the resulting products in pharmaceutical and biological applications is not advisable.

Attempts have been made to develop and use heterogeneous copper catalysts in Huisgen

reactions, but reports indicate that these catalysts seem to slow down the reaction significantly

[17]. For this reason, currently the CuAAc reaction does not constitute a viable click reaction

option for development of biomedical applications, including conjugation of polymer materials

to proteins.

Copper-free azide-alkyne cicloadditions are more suitable for biomedical applications, due to

the absence of copper. However, long reaction times and high temperatures are usually needed

[15], which could constitute a major problem for biomolecule conjugations.

Regarding the Diels Alder reactions, one of its main disadvantages in bioconjugation

applications is the fact that it may be necessary to modify the structure of the biomolecule [18].

Although in some cases the modification seems trivial from the chemical point of view,

proteins, peptides and carbohydrates present complex three-dimensional structures that must

be preserved, if preservation of the biological activity is also sought. In these cases, small

changes of the chemical and geometrical structure of these molecules can result in an

irreversible loss of activity [19]. As a consequence, Diels Alder reactions are used mainly for

production of polymer materials with molecular structures that can be difficult to produce

otherwise [20].

Condensation of aldehydes and ketones with heteroatom-bound amines was the first click

reaction described as biorthogonal, which means that the reaction can occur inside a living

system without interfering with native biochemical processes [21]. However, despite aldehyde

and ketone functional groups are not present on the cell surface, they appear as intracellular

metabolites. Because of that, the use of this type of reaction for labeling of biomolecules within

living systems is limited [13,21]. Moreover, ketones and aldehydes react with amine

nucleophiles that are enhanced by the -effect, but also react with thiols and alcohols.

Therefore, for bioconjugation applications, even though the reactions with thiols and alcohols

are usually thermodynamically unfavorable in water, there is always a risk that residual

aldehyde or ketone groups present in the conjugate system might interact with the living system

[21,13]. Accordingly, condensation of aldehydes and ketones with heteroatom-bound amines

does not seem to be truly biorthogonal.

As described in the previous case, Staudinger reactions are also biorthogonal, since none of its

reactants or products interact with the biological environment [22]. However, handling of

organic azides should be made with extreme caution, because most of them are explosive [13].

As a consequence, based on the original definition proposed by KOLB and coworkers, it is

debatable whether the term "click" can be used to qualify this class of reactions.

Thiol-ene and thiol-yne reactions comprise addition reactions between a molecule containing

a thiol group and alkenes or alkynes, respectively [23]. Some authors report that thiol-yne



Vol.6, No.1, pp.9-40, March 2018


11


reactions are about three times slower than thiol-ene reactions, particularly because thiol-yne

reactions occur in two steps instead of one [24].

In the present manuscript, click thiol-ene reactions are discussed with more detail because they

are faster than thiol-yne reactions and because many biomolecules contain cysteine residues

[25], making biomolecule modification unnecessary. This can constitute a major advantage of

this particular click route for bioconjugation. Additionally, thiol-ene reactions can occur

through free radical initiation, be mediated by nucleophiles, acids and bases (known as thiol-

Michael reactions) or simply require the presence of polar solvents, such as water. Moreover,

the range of reactants that may be used is extensive and includes both activated and non-

activated alkenes [24]. One must also consider that polymer chains produced with vinyl

monomers normally present unsaturated pendant vinyl bonds in the molecule, encouraging the

use of the thiol-ene route for bioconjugation.

The procedure most widely adopted to perform thiol-ene reactions makes use of thermal or

photo radical initiators. The reaction pathway is similar to free radical polymerizations and

includes initiation, propagation and termination steps [24]. The reaction mechanism is

discussed in more detail further ahead, in a section specifically dedicated to this topic. It must

be noted that the addition occurs with anti-Markovnikov orientation [24] and that molecules

containing thiol groups act as chain transfer agents [26]. Reaction times are usually of the order

of seconds even under mild temperature and pressure conditions. It is also worth mentioning

that thiol-ene reactions are not usually affected by the presence of atmospheric oxygen, which

can constitute a major practical advantage in many applications [27].

In the field of bioconjugation, the idea that a simple reaction with high efficiency and that does

not require the use of toxic solvents can be exploited to immobilize biological assets onto the

surface of polymer particles is extremely attractive and justifies the general interest in click

chemistry. However, even though click reactions seem to be simple and effective, some authors

have discussed intrinsic limitations of click reactions [28-30], while many aspects related to

the practical implementation of click reactions in actual production environments have yet to

be discussed.

One interesting aspect that has been consistently overlooked in the literature, for example, is

the fact that most authors make use of toxic solvents in the proposed reaction schemes. This

practice seems to contradict the original definition proposed by KOLB and coworkers in 2001

[1], although many researchers report the use of dichloromethane [31,32], chloroform [15,33],

dimethylformamide [32,34], tetrahydrofuran [15,35,36], pyridine [15] and other toxic solvents

[15,35,37] to perform click reactions. It is worth mentioning that solvent residuals must be

completely removed from the final product when biomedical and pharmaceutical applications

are sought, in order to avoid intoxication and meet quality-control requirements [38]. Product

purification, however, should be performed with care (in order to preserve the biological

activity of the molecule) and may involve a complex series of unit operations (distillation,

filtration, centrifugation, among others) and, as a consequence, can be very expensive. For this

reason, the use of toxic solvents should be avoided and it might be argued if reactions

performed with toxic solvents should be really regarded as click chemistry procedures.

Another important point regards the fact that click chemistry reactions are normally performed

in solution (homogeneous media), although bioconjugation reactions between polymer

particles and biomolecules usually are performed in heterogeneous media. KOLB et al. (2001)

observed that, when water is used as solvent, solubilization of reactants in the reaction medium



Vol.6, No.1, pp.9-40, March 2018


12


may not be required [1]. As a matter of fact, KOLB et al. defended that efficient agitation

should be sufficient to maintain solid materials homogeneously suspended in water and

guarantee that reactions would take place efficiently [1,35]. DECAN et al. [39] studied the use

of copper nanoparticles as heterogeneous catalyst for CuAAc click reactions and reported that

only one catalytic event per nanoparticle took place at a given time and that product had a

residence time of 3 s after the surface reaction. ZHAO et al. [40] studied a photochemical

thiol-ene click modification of cellulose in its solid native state. In this reaction, solid cellulose

was added to a water solution of 2,2-dimethoxy-2-phenylacetophenone and irradiated for one

hour. Authors reported success in the proposed heterogeneous reaction.

In spite of that, the vast majority of the publications report that the analyzed reactions occur in

homogeneous media [31-37]; however, it is not possible to assume that results obtained in

solution would be similar in heterogeneous systems. As a matter of fact, some authors justify

that their thiol-ene reactions were unsuccessful because reagents were incompatible with

solvents, creating a heterogeneous reaction mixture [41]. As a consequence, it can be said that

very little is known about click reactions performed in heterogeneous systems, as required by

some bioconjugation reactions. It is important to emphasize that most polymer products are

prepared in heterogeneous media and obtained as fine solid suspensions in water.

Regarding more specifically the click thiol-ene reactions, an important practical issue is related

to the possible effect exerted by different free radical initiators on the course of the chemical

transformations. The usual click thiol-ene mechanism does not require the detailed

specification of the free radical initiator and this probably explains why AIBN

(azobisisobutyronitrile) is used almost exclusively in all published procedures. As a matter of

fact, few authors attempted to use other types of thermal initiators, such as potassium persulfate

and benzoyl peroxide, in their studies, although the use of these initiators is much more usual

in real production environments because of the reduced costs and safer handling procedures

[42].

Moreover, click chemistry reactions already reported (and especially the ones related to

polymer conjugation) are extremely complex and usually require multiple reaction and

purification steps [35,43]. This makes the whole procedure very hard or nearly impossible to

reproduce. In the present work, the main goal was to carry out heterogeneous thiol-ene

reactions in a very simple way using readily available and non-toxic reagents (always as

possible). To the best of our knowledge, no other published work has fulfilled these

requirements.

Based on the discussion presented in the previous paragraphs, click thiol-ene reactions were

performed between a divinyl monomer (divinylbenzene, 1,5-hexadiene or ethylene glycol

dimethacrylate) and a molecule containing free thiol groups (L-cysteine, cysteine

hydrochloride or 3-mercaptopropionic acid) in order to assess and illustrate some practical

issues related to the use of click thiol-ene reactions for bioconjugation in heterogeneus media.

The analyzed monomers are frequently used for reticulation of polymer particles and to provide

unsaturated vinyl bonds for posterior polymer fuctionalization. The analyzed biomolecules

frequently constitute proteins and are possible sources of thiol groups for bioconjugation.

Reactions were conducted in aqueous suspensions, as usually desired at plant site, although

toxic solvents were also used in some cases for benchmarking purposes. In order to illustrate

the importance of the free radical initiator, reactions were performed using different thermal

initiators (AIBN, benzoyl peroxide and potassium persulfate). Based on the obtained results,

some important practical gaps in the click thiol-ene literature are detected and discussed,



Vol.6, No.1, pp.9-40, March 2018


13


providing suggestions for research related to the use of click chemistry for conjugation of

polymer particles and biomolecules in the near future.

MATERIALS AND METHODS

Materials

Ethylene glycol dimethacrylate (EGDMA, with minimum purity of 98 wt%), 1,5-hexadiene

(HD, with minimum purity of 97 wt% and containing up to 3 wt% of other diolefins), 3-

mercaptopropionic acid (MPA, with minimum purity of 99 wt%), styrene (minimum purity of

99 wt%), dichloromethane (DCM, with minimum purity of 99,5 wt%), L-cysteine (Cys, with

minimum purity of 97 wt%) and cysteine hydrochloride (Cys-Cl, with minimum purity of 98

wt%) were purchased from Sigma-Aldrich (Rio de Janeiro, Brazil). Benzoyl peroxide (BPO),

sodium borohydride and N,N-dimethylformamide (DMF) were obtained from VETEC (Rio de

Janeiro, Brazil) with minimum purity of 99 wt%. Azobisisobutyronitrile (AIBN) was

purchased from Akzo Nobel (Netherlands) with minimum purity of 99 wt%. Divinylbenzene

(DVB) was obtained from Merck (Germany) with purity of 99 wt% (as a 65 wt% solution in

ethyl-styrene). Potassium persulfate was purchased from Proquimios (Brazil) with minimum

purity of 99 wt%. Tetrahydrofuran (THF) was obtained from Tedia (Brazil) and dimethyl

sulfoxyde (DMSO) from Nuclear (Brazil), both with minimum purity of 99 wt%. Deuterated

chloroform was purchased from Cambridge Isotope Laboratories (USA) with minimum purity

of 99 wt%. Unless stated otherwise, reagents were used as received.

Reactions

Thiol-ene Reaction with Styrene

2.36 g of AIBN were initially dissolved in 15 g of DMF. Then, 0.42 g of styrene and 6.09g of

MPA were added to this solution. The final solution was transferred to a round bottom flask

attached to a condenser and kept at 80 ºC for 4 hours under constant magnetic stirring (600

rpm). The product was dried for 72 hours in a recirculating oven at room temperature and then

washed with acetone.

Thiol-ene Reactions with L-Cysteine

10 g of distilled water, 1.21 g of L-cysteine and either HD (0.21 g), DVB (0.33 g) or EGDMA

(0.50 g) were added into a round bottom flash attached to a condenser. After mixing and

reaching the desired temperature, BPO (0.61 g), AIBN (0.41 g) or K2S2O8 (0.68 g) was fed into

the flask. The reaction mixture was kept under constant magnetic stirring (600 rpm) at 85 ºC

or at 70 ºC for 1h30min. The condenser temperature was kept at 10 ºC and products were dried

in a recirculating oven at room temperature for 24h.

Thiol-ene Reactions with Cysteine Hydrochloride

30 g of distilled water, 1.57 g of cysteine hydrochloride and 0.25 g of EGDMA were added

into a round bottom flash attached to a condenser. After mixing and reaching the desired

temperature, AIBN (0.82 g) or K2S2O8 (1.35 g) was fed into the flask. The reaction mixture

was kept under constant magnetic stirring (600 rpm) at 70 ºC for 4 h. The condenser

temperature was kept at 10 ºC and products were dried in a recirculating oven at room

temperature. In some cases, cysteine hydrochloride was reduced with sodium borohydride prior



Vol.6, No.1, pp.9-40, March 2018


14


to the reaction. When this procedure was conducted, Cys-Cl, water and 1.13 g of sodium

borohydride were kept under constant magnetic stirring for 2 h at ambient temperature. After

this step, the reaction proceeded as described previously.

Polymerization Reactions

Bulk polymerization reactions were conducted in test tubes. A solution containing 1 g of

benzoyl peroxide and 25 g of styrene was prepared (when copolymerization was carried out,

0.5 wt% of styrene was replaced by EGDMA). Then, 2 g of the solution were added to each

test tube and placed in an ethylene glycol bath heated at 85 °C. Each tube was removed from

the bath at specified reaction times (10 min, 30 min, 1 h, 1h30min, 2 h, 2h30min, 3 h, 3h30min,

4 h and 4h30min) and 10 drops of alcoholic hydroquinone solution (1% w/v) were added [44].

All products were dried first in a recirculating oven and then in a vacuum oven at room

temperature.

Thiol-ene Reactions with 3-Mercaptopropionic Acid

0.2 g of previously prepared poly(styrene-co-divinylbenzene), containing pendant double

bonds, 2.5 g of DMF, DCM or THF and 4 mg of MPA were added into a round bottom flash

attached to a condenser. After mixing and reaching the desired temperature, AIBN (3 mg) was

fed into the flask. The reaction mixture was kept under constant magnetic stirring (600 rpm) at

80 ºC for 4 hours.

Characterization

Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy (FTIR-ATR)

FTIR analyses were conducted in a Thermo Nicolet 6700 (Thermo Fisher Scientific, USA)

equipped with Smart OrbitTM, an ATR accessory comprising a diamond crystal and a swivel

pressure tower that ensures the application of consistent pressure on samples. Infrared analyses

were performed in the mid-infrared region with spectral resolution of 4 cm-1 at room

temperature. Spectral data were reported as averages of 128 scans.

Gel Permeation Chromatography (GPC)

GPC analyses were performed at 40 ºC with help of a Viscotek (Malvern, United Kingdom)

VE2001 chromatograph equipped with a Viscotek (Malvern, United Kingdom) VE3580

refractometric detector and a Viscotek (Malvern, United Kingdom) 2500 UV detector set at

255 nm and equipped with a deuterium lamp, a Shodex KF-G (Showa Denko K.K., Japan) pre-

column, two Shodex KF-804 (Showa Denko K.K., Japan) columns and one Shodex KF-805

(Showa Denko K.K., Japan) column. The equipment was calibrated with polystyrene standards

with molecular weights ranging from 376 to 1.0 x 106 Da. Samples were prepared with

concentration of 1 mg/mL in THF. Solutions were filtrated with a Teflon filter with pore size

of 0.45 m. The injection volume was equal to 200 L and the operating flow rate was equal

to 1 mL/min.

Solubility Tests

For solubility tests, 1 mL of solvent (water, THF, DMF, DMSO, hydrochloric acid, acetone or

methanol) was mixed with 10 mg of product and left under mild agitation at ambient

temperature for 48 hours. Then, samples were heated until 60 ºC. Solubility was checked

visually through identification of suspended solid material.



Vol.6, No.1, pp.9-40, March 2018


15


THIOL-ENE REACTION MECHANISM

In this section, the thiol-ene reaction mechanism is presented in more detail in order to

complement the introductory section and help explaining the results presented ahead. The thiol-

ene reaction has been known since the early 1900s and its mechanism has been described in

many excellent reviews [45-48]. The mechanism is summarized in

Figure 1, where the initiation, propagation, chain transfer and termination steps are illustrated.

As one can see by comparing

Figure 1 and Figure 2, the mechanism that describes the thiol-ene reaction is very similar to

the mechanism normally used to represent a free-radical polymerization reaction. The main

difference relies on the initiation step. In the polymerization mechanism, initiation consists in

the decomposition of the initiator, generating free radicals that readily react with monomers.

In the thiol-ene mechanism, the free radicals generated by the initiator decomposition react

with the molecules that contain SH groups. Besides, it is important to highlight that molecules

containing SH groups act as chain transfer agents.

In fact, the reaction of radical fragments with vinyl monomer molecules or thio-compounds

constitutes a random event. Therefore, one cannot discard beforehand the occurrence of

monomer initiation (or polymerization reactions) when thiol-ene reactions are pursued. In spite

of that, this important side reaction has been usually neglected or omitted from mechanisms

presented in publications concerning thiol-ene click reactions. CRAMER et al. studied the

occurrence of polymerization reactions in click chemistry systems and reported that, when

acrylate conversions reached 100%, thiol conversions were close to 50%, indicating the

importance of this side reaction step [50]. This result confirms that the polymerization reaction

occurred in a much bigger extent than the thiol-ene reaction. For this reason, the possible

occurrence of polymerization reactions (as side reactions) will always be considered in the

results presented in the following sections.



Vol.6, No.1, pp.9-40, March 2018


16


Figure 1. Schematic representation of the thiol-ene reaction mechanism.



Vol.6, No.1, pp.9-40, March 2018


17


Figure 2. Schematic representation of the free-radical polymerization reaction

mechanism [49].

RESULTS AND DISCUSSION

Thiol-ene Reactions with Styrene

The first experiments performed in the present work intended to describe a benchmark system

to be used as reference for all other proposed and analyzed reactions. As widely discussed in

the click chemistry literature, the reactivity of the double bond molecules follows the following

decreasing order of reactivity: norbonene > vinyl ether > vinyl ester > allylether > acrylate >

N-substituted maleimide > methacrylate > styrene > conjugated dienes [45-48]. According to

this scale, styrene can be successfully used in thiol-ene reactions and for this reason it was

selected as a reactant for reaction A, as styrene constitutes a cheap and well-studied model

polymerization system. After extensive review on thiol-ene reactions, it was observed that the

equivalence thiol:ene ratios usually range from 2:1 to 10:1 in publications that reported high

reaction yields and good selectivity [51,52]. Similarly, equivalence initiator:thiol ratios usually

range from 0.1 to 1 [51,52] and the most commonly used thermal initiator is AIBN [35,37,51].

Finally, many groups reported successful results when MPA was used as the source of thiol

groups [7,42,53]. Therefore, in reaction A, styrene and MPA were used in the equivalence

thiol:ene proportion of 10:1, using AIBN as the initiator with the equivalence initiator:thiol

ratio of 1:2.

Reaction A was conducted in solution, using DMF as solvent (as usual in the literature) and

the final product was dried in a recirculating oven at room temperature and then washed with

acetone before FTIR analysis. The FTIR analysis of the residual liquid product (after drying

for 72 hours, some residual liquid still remained in the flask) was also performed. Both spectra

are presented in



Vol.6, No.1, pp.9-40, March 2018


18


Figure 3.

As one can see in

Figure 3, a band placed at 699 cm-1 appears in both liquid and solid products of reaction A.

This is a clear indication that the click chemistry occurred, since the C-S FTIR bands appear at

692 cm-1 [54]. The observed shift is common in FTIR analyses and can be related to interactions

with other components. The residual liquid product of reaction A is a mixture of DMF, MPA

and click chemistry product. It is important to observe that neither the solid product nor the

liquid product present evidences of residual styrene double bonds, that should appear at 1630

cm-1, 910 cm-1 or 990 cm-1 [55]. Therefore it can be assumed that styrene molecules were

efficiently converted into click products.

2000 1500 1000 500

a.u.

Wavenumber (cm-1)

699 cm-1

MPA

DMF

Styrene

Reaction A_liquid

Reaction A_solid

Figure 3. FTIR spectra of liquid and solid products of reaction A and of MPA, DMF

and styrene.

In order to observe if polystyrene might have been produced as a side product, GPC analysis

was performed with the solid product and RI and UV signals were detected. The GPC results

for the solid product of reaction A are presented in

Figure 4 and in

Figure 5. As one can see in

Figure 4, only one product was detected in the GPC analysis and the calculated number average

molar mass was equal to 222 g/gmol. This is a clear indication that polymerization reactions

did not take place; otherwise, the average molar mass would be much higher. Besides, the

polydispersity index (PDI) was very small and equal to 1.08, indicating that nearly 100% of

the molecules have the same size. This also indicates polystyrene was not formed, since in non-

controlled polymerizations the PDI index is usually close to 2 or higher [49].

Figure 4 also indicates that the click product seems to be pure, since only one narrow peak was

detected during the whole analysis. This is confirmed in

Figure 5, which shows that the UV detector also detected a single peak that started to appear at

the exact time when the RI peak was detected. Finally, the molar mass obtained by the GPC

analysis was essentially equal to the sum of styrene and MPA molar masses (MPA= 106.14

g/gmol and Styrene = 104.15), suggesting once more that the reaction product was a click



Vol.6, No.1, pp.9-40, March 2018


19


product between styrene and MPA. Therefore, reaction A can be used as a successful

benchmark for the other reactions performed in the present work, since it seems to constitute a

successful thiol-ene system.

0 100 200 300 400 500 600

0,0

0,2

0,4

0,6

0,8

1,0R

I no

rma

lize

d r

esp

on

se (

a.u

.)

Molar Mass (g/gmol)

Figure 4. Molar mass distribution of the product obtained in reaction A

29 30 31 32

RI

UV

Elution time (min)

No

rma

lize

d r

esp

on

se (

a.u

.)

Figure 5. Responses of RI and UV GPC detectors for the solid product of reaction A.

In the following sections, several modifications are proposed in our benchmark reaction in

order to adapt it to more realistic conditions concerning bioconjugation applications. In each

section the modifications are presented and explained in detail.

Thiol-ene Reactions with L-Cysteine at 85 ºC

In bioconjugation reactions, the thiol concentration is expected to be small, as thiol groups are

usually provided by biomolecules (which are frequently extremely expensive) while the

unsaturated double bonds are normally provided by the polymer matrix. As a consequence,

thiol:ene ratios are expected to be much lower in real conjugation reactions involving polymer

matrices and proteins, for instance. Also, it is not viable to use such high concentrations of

toxic solvents as used in reaction A when biomedical and pharmaceutical applications are

pursued. In fact, the best solution is always to use aqueous environments. Moreover, MPA

decreases gamma-aminobutyric acid in the brain, thereby causing convulsions [56]. For this



Vol.6, No.1, pp.9-40, March 2018


20


reason, in this section MPA was replaced by L-cysteine, which is non-toxic and is present in

the chemical structure of many biomolecules. Additionally, styrene is a monounsaturated

monomer and cannot be used to insert pendant double bonds in a polymer matrix for further

bioconjugation. In order to do so, multifunctional monomers must be used. Therefore, 1,5-

hexadiene, divinylbenzene and ethylene glycol dimethacrylate were tested. Finally, all

reactions described in this section were performed with 3 different initiators in order to

compare the relative performances of the distinct initiation sytems.

Based on the modifications proposed above, a first group of reactions was performed with

different dienes, with equivalence thiol-ene ratio of 2:1 and using different thermal initiators

with equivalence initiator:thiol ratio of 0.5, as shown in Table 1. Reactions were conducted in

heterogeneous aqueous media at 85 ºC. It is important to highlight that in this proposed group

of reactions only K2S2O8 is completely water soluble, while L-cysteine is poorly soluble in

water. All other reagents are not water soluble, which means that in some cases the system is

composed of three different phases (organic liquid phase, aqueous liquid phase and solid

phase). Finally, L-cysteine is insoluble in HD, DVB and EGDMA. It is important to observe

that the heterogeneous nature of the reacting system should be expected in most bioconjugation

problems.

FTIR analyses of all obtained products and respective reagents were performed. Figure 6

shows the FTIR spectra of products obtained in runs 1, 2 and 3, performed with 1,5-hexadiene

and different initiators. Spectra of HD and L-cysteine are also shown in Figure 1 for the sake

of comparison. Figure 6 indicates that the chemical nature of the thermal initiators exerted an

enormous influence on the reaction course, since the FTIR spectra of the obtained products

were very different from each other. This point has been largely overlooked in the literature

and clearly indicates that different initiators can exert distinct effects on the click thiol-ene

reaction, making necessary the more detailed description of radical formation and radical

interaction with the remaining reagents. It is also important to observe that none of the obtained

solid products was soluble in the analyzed solvents, indicating the production of a complex

mixture of products, as also suggested by the obtained FTIR spectra, which has also been

overlooked in the literature. In other words, the expected high specificity of the click reactions

is not attained in the present case, despite the usual reaction conditions and recipe. Because of

the complexity of the obtained FTIR spectra, spectral analysis was focused mainly on the

characteristic bands that can be related directly to the thiol-ene reactions.

The characteristic band that indicates the presence of C-S bonds is positioned at 692 cm-1 [53].

Assuming that the click reaction occurred, one might expect the relative increase of this peak

at the end of reaction. However, as indicated by the dotted lines in Figure 6, this increase could

not be observed. It is also important to emphasize that the characteristic band related to S-S

bonds is located at 538 cm -1 [57], as also indicated by dotted lines in Figure 1. It is interesting

to observe that the relative intensity of this band remained virtually unchanged after reaction,

which apparently excludes the possible formation of cystine through reaction between two

cysteine molecules [58]. Thus, one may possibly conclude that the click thiol-ene reaction

between cysteine and 1,5-hexadiene, mediated by the analyzed thermal initiators, did not occur

in heterogeneous aqueous media, as the characteristic C-S band was not detected. As important

as that, the complexity of the FTIR spectra of final solid products and solubility tests indicated

that other complex reactions took place, suggesting that the thiol-ene reaction may not present

high specificity at the analyzed conditions. Figure 6 also indicates that 1,5-hexadiene double

bonds were consumed during reactions 1, 2 and 3, since the peaks located at 910 and 990 cm-1



Vol.6, No.1, pp.9-40, March 2018


21


[59] and related to the double bonds of HD do not appear in the products. This consumption

might be the result of 1,5-hexadiene polymerization, for example.

Table 1. Thiol-ene reactions with L-cysteine.

Reaction Alkene (ene) Initiator Thiol Temperature

1 HD BPO L-cysteine 85 ºC

2 HD AIBN L-cysteine 85 ºC

3 HD K2S2O8 L-cysteine 85 ºC

4 DVB BPO L-cysteine 85 ºC

5 DVB AIBN L-cysteine 85 ºC

6 DVB K2S2O8 L-cysteine 85 ºC

7 EGDMA BPO L-cysteine 85 ºC

8 EGDMA AIBN L-cysteine 85 ºC

9 EGDMA K2S2O8 L-cysteine 85 ºC

10 - BPO L-cysteine 85 ºC

11 - AIBN L-cysteine 85 ºC

12 - K2S2O8 L-cysteine 85 ºC

It is interesting to observe that the FTIR spectrum of the solid product prepared with 1,5-

hexadiene, L-cysteine and benzoyl peroxide in run 1 was more similar to the original FTIR

spectrum of L-cysteine, when compared to the other two products. This may be related to the

reaction temperature, as BPO decomposes more slowly than AIBN and potassium persulfate

at 85 °C [60]. The possible effect of the rate of decomposition of thermal initiators on the

course of click thiol-ene reactions has also been overlooked in the published material. The

faster rate of radical generation provided by AIBN and persulfate may have led to higher

concentrations of free radicals in the medium and possibly increased the importance of other

undesired side reactions.



Vol.6, No.1, pp.9-40, March 2018


22


4000 3500 3000 2500 2000 1500 1000 500

910 cm-1

a.u.

Wavenumber (cm-1)

HD

L-cysteine

Reaction 3

Reaction 2

Reaction 1

990 cm-1

Figure 6. FTIR spectra of the final solid products of runs 1, 2 and 3 and of reagents L-

cysteine and 1,5-hexadiene.

The FTIR spectra of solid products obtained through reactions of L-cysteine with DVB and

EGDMA are shown respectively in Figure 7 and Figure 8. Once again it is possible to observe

the distinct effects exerted by the different initiators on the reaction product and the fact that

the relative intensity of the characteristic C-S band placed at 692 cm-1 did not increase. As a

consequence, one can say that the observed initiator effects and the low specificity of the thiol-

ene reaction do not depend solely on the particular analyzed diene, which can be regarded as a

very important conclusion. Similarly, the final FTIR spectra of solid powders obtained with

BPO were more similar to the FTIR spectrum of L-cysteine than in the other cases, suggesting

that the observed BPO effect does not depend on the analyzed diene either. Additionally,

obtained solid products were not soluble in any of the analyzed solvents, as in the previous

case. It is also important to emphasize that, as shown in Figure 7, the band associated to DVB

pendant double bonds (985 cm-1) [61] does not appear on the products of reactions 4, 5 and 6.

This is a clear indication that the double bonds were consumed in side reactions that might

include DVB polymerization, for example.

Figure 8 shows similar results for EGDMA, as the band related to the EGDMA double bonds

(1641 cm-1) [62] and indicated by a dashed line does not appear on the products of reactions 4,

5 and 6. This is once more a clear indication that the double bonds were consumed in side

reactions that might include the EGDMA polymerization. Also, the presence of a band marked

by a dashed line at 1720 cm-1 [63], which can be associated to carbonyl groups, indicates the

presence of EGDMA in the final product. This result seems to corroborate the idea that one of

the side reactions that occur during the above described click chemistry attempts might be the

polymerization of the dienes.



Vol.6, No.1, pp.9-40, March 2018


23


1500 1000 500

a.u

.

Wavenumber (cm-1)

DVB

L-cysteine

Reaction 6

Reaction 5

Reaction 4

985 cm-1


cysteine and DVB.

1500 1000 500

a.u.

Wavenumber (cm-1)

EGDMA

L-cysteine

Reaction 9

Reaction 8

Reaction 7


cysteine and EGDMA.

Figure 9 shows that the FTIR spectra of the solid products obtained with BPO were relatively

similar, presenting lower sensitivity to the analyzed diene. The scenario is similar when the

other initiators are taken into consideration, as shown in Figure 10 and Figure 11. Figures 4, 5

and 6 reinforce that the reactions performed with different initiators can follow different

reaction mechanisms and suggest that the diene may not react with L-cysteine when the

reagents are contained in distinct phases, as the dienes were suspended in the aqueous phase as

small droplets.



Vol.6, No.1, pp.9-40, March 2018


24


Figure 12 shows FTIR spectra of solid products obtained in runs 10, 11 and 12, performed in

absence of the dienes. Based on the FTIR data, it can be observed that the obtained products

were different from the ones presented in Figures 1 to 6. This clearly shows that the dienes do

take part in the reaction, but illustrates once more that the different initiators interact differently

with L-cysteine, that the final solid material can be constituted by a mixture of different

products, that the thiol-ene reaction can present low specificity and that the characteristic C-S

bond may not be formed at all at the analyzed conditions. Additionally, obtained solid products

were not soluble in any of the analyzed solvents.

2000 1800 1600 1400 1200 1000 800 600 400

a.u

.

Wavenumber (cm-1)

BPO

L-cysteine

Reaction 7

Reaction 4

Reaction 1


cysteine and BPO.

1500 1000 500

a.u

.

Wavenumber (cm-1)

AIBN

L-cysteine

Reaction 8

Reaction 5

Reaction 2


cysteine and AIBN.



Vol.6, No.1, pp.9-40, March 2018


25


Based on the previous paragraphs, it sounds reasonable to say that thiol-ene reactions do not

present the characteristic features of click reactions at the analyzed conditions. Besides, it

seems clear that both initiator and diene affect the course of the reactions, which may lead to

formation of a complex mixture of products, with low specificity towards formation of the

desired C-S bond. These effects had never been discussed in the open literature and are very

important for the proper design of heterogeneous bioconjugation systems.

4000 3500 3000 2500 2000 1500 1000 500

a.u

.

Wavenumber (cm-1)

K2S

2O

8

L-cysteine

Reaction 9

Reaction 6

Reaction 3


cysteine and K2S2O8.

4000 3500 3000 2500 2000 1500 1000 500

a.u.

Wavenumber (cm-1)

Reaction 10

Reaction 11

Reaction 12

L-cysteine

BPO

AIBN

K2S

2O

8

Figure 12. FTIR spectra of the final solid products of runs 10, 11 and 12 and of reagents

L-cysteine, BPO, AIBN and K2S2O8.



Vol.6, No.1, pp.9-40, March 2018


26


Thiol-ene Reactions with L-Cysteine at 70 ºC

In order to evaluate the effect of temperature (and of rate of free radical generation),

experiments were performed with AIBN and K2S2O8 at 70 °C, as presented in Table 2. BPO

was not used because the rate of thermal decomposition of BPO at 70 °C is very low. DVB and

EGDMA were used as monomers in these tests because these monomers are used more

frequently as crosslinking agents in most commercial polymerization reactions. Once again it

is important to highlight that in this second group of reactions performed at 70 °C only K2S2O8

and EGDMA are completely water soluble, while L-cysteine is poorly soluble in water. All

other reagents are not soluble in water, which means that in some cases the system is composed

of three different phases.

Table 2. Group of experiments performed at 70 ºC.

Reaction Alkene (ene) Initiator Thiol Temperature

13 DVB AIBN L-cysteine 70 ºC

14 EGDMA AIBN L-cysteine 70 ºC

15 - AIBN L-cysteine 70 ºC

16 DVB K2S2O8 L-cysteine 70 ºC

17 EGDMA K2S2O8 L-cysteine 70 ºC

18 - K2S2O8 L-cysteine 70 ºC

As one can see in Figure 13 and in Figure 14, formation of the characteristic C-S bond could

not be detected in any run. Residual double bonds could not be detected either (band at 1641

cm-1 for reactions 14 and 17 and 985 cm-1 for reactions 13 and 16) [61,62], indicating that the

dienes were consumed and formed byproducts. In reactions 14 and 17 the presence of the band

at 1720 cm-1 [62], which can be assigned to carbonyl groups, indicates the presence of EGDMA

in the final product. One can also observe that the reaction temperature clearly affected the

FTIR spectra of the final solid products, reinforcing the idea that the rate of free radical

generation does affect the course of the thiol-ene reaction. Although this point has been largely

overlooked in previous studies, it is not surprising, as the rate of radical generation affects the

concentration of free radicals and the possible occurrence of undesired side reactions. Figure

15 illustrates the effect of temperature on FTIR spectra of the final solid powder.

One of the possible causes for absence of the characteristic thiol-ene reactions in the previously

reported experiments could be the lack of free thiol in commercial L-cysteine. As reported in

the literature, L-cysteine can react to form dimers, with formation of disulfide bonds [58]. As

one can see in Figure 16, disulfide bonds can be detected through the characteristic S-S band

placed at 534 cm-1. Therefore, it may be important to reduce L-cysteine prior to use in click

thiol-ene reactions, in order to eliminate possible disulfide bonds. This makes the commercial

process less competitive and does not change the fact that a complex network of reactions

seems to take place when the reagents are put in contact.



Vol.6, No.1, pp.9-40, March 2018


27


2000 1500 1000 500

a.u.

Wavenumber (cm-1)

L-cysteine

Reaction 13

Reaction 14

Reaction 15

1720 cm-1

Figure 13. FTIR spectra of the final solid products of runs 13, 14 and 15 and of reagent

L-cysteine.

2000 1500 1000 500

a.u

.

Wavenumber (cm-1)

L-cysteine

Reaction 16

Reaction 17

Reaction 18

1720 cm-1

Figure 14. FTIR spectra of the final solid products of runs 16, 17 and 18 and of reagent

L-cysteine.

Another possible cause for absence of the characteristic thiol-ene reactions in the previously

reported experiments could be the fact that reactions that were performed with AIBN (the most

successfully used thermal initiator reported in the literature for thiol-ene reactions) were

conducted in heterogeneous aqueous media, instead of homogeneous organic media, as

reported by most authors. However, as discussed before, performing the thiol-ene reaction in

homogeneous medium would constitute a major disadvantage for conjugation of polymer

materials and biomolecules, as medical and pharmaceutical applications usually require



Vol.6, No.1, pp.9-40, March 2018


28


production of micro- and nanoparticles, which are produced in heterogeneous media during

polymerization.

2000 1800 1600 1400 1200 1000 800 600 400

a.u

.

Wavenumber (cm-1)

L-cysteine

Reaction 11

Reaction 15

Figure 15. FTIR spectra of the final solid products of runs 11 and 15 and of reagent L-

cysteine.

4000 3500 3000 2500 2000 1500 1000 500

a.u

.

Wavenumber (cm-1)

2551 cm-1

534 cm-1

Figure 16. FTIR spectra of commercial L-cysteine.



Vol.6, No.1, pp.9-40, March 2018


29


Thiol-ene Reactions with Cysteine Hydrochloride

As L-cysteine is not very soluble in water, cysteine hydrochloride was used to perform the

experiments presented in Table 3 in order to evaluate whether results could be different when

molecules containing thiol groups were solubilized in the aqueous medium. In all these

experiments, cysteine hydrochloride was first reduced with borohydride in order to prevent the

existence of disulfide bonds and increase the amount of free thiol groups, as discussed

previously. Cysteine hydrochloride, EGDMA and K2S2O8 are water soluble. All other reagents

are not soluble in water, which means that in reactions 19 and 20 the system was composed of

two different phases, while in reactions 21 and 22 all reagents were water soluble, as in the

benchmark click reaction performed with styrene, MPA and AIBN in DMF.

Table 3. Reactions with Cys-Cl.

Reaction Alkene (ene) Initiator Thiol Borohydride

19 EGDMA AIBN Cys-Cl No

20 EGDMA AIBN Cys-Cl Yes

21 EGDMA K2S2O8 Cys-Cl No

22 EGDMA K2S2O8 Cys-Cl Yes

FTIR analyses of the final solid products of reactions performed with AIBN and potassium

persulfate are shown respectively in Figure 17 and Figure 18. Results indicate that double

bonds contained in EGDMA (1641 cm-1) apparently were not completely consumed and that

C-S bonds could not be detected (692 cm-1). Incomplete consumption of vinyl bonds is possibly

related to the fact that the main reagents were present in different phases. Based on the FTIR

spectra, it is possible to infer that chloroalkanes were formed because of the characteristic band

placed at 700 cm-1 [63], indicating once more the formation of byproducts instead of the

expected click reaction. It is also important to note that cysteine hydrochloride reduction with

sodium borohydride did not lead to the expected high specificity of the click reactions,

indicating that the results reported here were not related necessarily with the oxidation state of

L-cysteine.

Thiol-ene Reactions with a Polymer Containing Pendant Double Bonds

In order to investigate the reasons why no effective click chemistry reaction was observed in

the previous tests (with exception of the benchmark reaction presented in Section 4.1), a

previously reported experiment was carried out. In the reported experiment, MPA was reacted

with unsaturated polystyrene in presence of AIBN at 80 ºC and in solution of dichloromethane

[42]. Authors reported 87% conversion after 4 h of reactions [42]. For this reason, MPA and

poly(styrene-co-ethylene glycol dimethacrylate) were selected as the thiol-ene pair and AIBN

was used as thermal initiator at the same conditions reported previously. Besides



Vol.6, No.1, pp.9-40, March 2018


30


dichloromethane, two common additional solvents were also tested, as described in the

experimental section.

The preparation of the copolymer material is described in Section 2.2.4. EGDMA was chosen

as a comonomer because it is a crosslinking agent widely used for biomedical applications [64].

It was expected that at least a fraction of the double bonds of the final copolymer chain would

be available after the copolymerization for later use in thiol-ene click reaction and

bioconjugation. This hypothesis is consistent with the theory of FLORY [65] and

STOCKMAYER [66], who proposed that chain crosslinking takes place in three steps: (i)

linear structures are formed having pendant side groups; (ii) the polymer structure grows and

branches begin to appear; (iii) crosslinking finally occurs. This strategy was already proven to

be efficient by other authors who also used unsaturated copolymers in their researches [42,67].

There are many published works about styrene copolymerizations with EGDMA, but to the

best of our knowledge there are no references to the presence of pendant double bonds in these

materials.

2000 1500 1000 500

a.u.

Wavenumber (cm-1)

EGDMA

~700 cm-1

1641 cm-1

Reaction 20

Reaction 19

Figure 17. FTIR spectra of the final solid products of runs 19 and 20.

2000 1500 1000 500

a.u.

Wavenumber (cm-1)

EGDMA

Reaction 21

Reaction 22

1641 cm-1

Figure 18. FTIR spectra of the final solid products of runs 21 and 22.



Vol.6, No.1, pp.9-40, March 2018


31


One of the reasons why styrene was selected as the main monomer to carry out the tests is the

fact that the styrene polymerization kinetics is well-known and can also be used for benchmark

purposes [67,68]. For example, Figure 19 shows the monomer conversions in polystyrene (PS)

and poly(styrene-co-ethylene glycol dimethacrylate) (P(S-co-EGDMA)) polymerizations. As

one can observe in Figure 19, the reaction rates for the two reaction systems are very similar,

mainly because only 0.5 wt% of EDGMA were added to the copolymerization system.

Obtained results indicated that both reactions reached conversions of almost 100%. Therefore,

in order to determine whether the intended pendant double bonds were present in the polymer

structure, FTIR analyses were conducted and results are presented in Figure 20.

0 50 100 150 200 250 300

0

20

40

60

80

100

PS

P(S-co-EGDMA)Con

vers

ion

(%)

Time (min)

Figure 19. Comparison between monomer conversions for polystyrene and

poly(styrene-co-ethylene glycol dimethacrylate).

4000 3500 3000 2500 2000 1500 1000 500

a.u

.

Wavenumber (cm-1)

1723 cm-1

PS

P(S-co-EGDMA)

Figure 20. FTIR spectra of PS and P(S-co-EGDMA).



Vol.6, No.1, pp.9-40, March 2018


32


As one can note in Figure 20, the FTIR spectra of the produced polymers were very similar

and this is not difficult to explain, since only 0.5 wt% of EGDMA were added into the reacting

mixture. However, an increase in the band located at 1723 cm-1 can be observed. IR bands of

-unsaturated and benzoate esters occur at 1730-1715 cm-1 [69]. Therefore, the band

located in the copolymer at 1723 cm-1 -

unsaturated esters. This indicates that EGDMA was inserted in the polymer chains and that

there are pendant double bonds in the copolymer structure. Accordingly, the copolymer

produced is suitable for thiol-ene reactions that are presented in Table 4.

Table 4. Reactions performed with P(S-co-EGDMA) and MPA as the thiol-ene pair.

Reaction Alkene (ene) Initiator Thiol Solvent

23 P(S-co-EGDMA) AIBN MPA DCM

24 P(S-co-EGDMA) AIBN MPA THF

25 P(S-co-EGDMA) AIBN MPA DMF

Click reactions were performed in different toxic solvents, as commonly reported in the

literature. FTIR analyses of the final products are shown in Figure 21. It is important to

highlight that reactions 23 and 24 were performed in homogeneous media, while reaction 25

was performed in heterogeneous media, because the polymer was not soluble in DMF. Results

presented in Figure 21 indicate that the band that refers to the pendant double bonds (1723 cm-

1) remains in the polymer structure after reactions 23 and 25, which is an indication that the

thiol-ene bonds were not formed. For reaction 24, consumption of the double bonds may

possibly have occurred; however, due to the interference of the polymer on the band that

characterizes the C-S bonds (692 cm-1), it was not possible to confirm unequivocally the

occurrence of the click reaction.

It is important to highlight that reaction 23, for which the click reaction did not occur, was

similar to a reaction reported as successfully performed by other research group [42]. The main

difference was the composition of the polymer material that contained the pendant double

bonds, poly(styrene-allyl bromide). However, this difference was not expected to interfere with

the final results, as the P(S-co-EGDMA) used in the present work also contained ene groups

and was also soluble in DCM. However, once more this indicates that click chemistry reactions

depend on the nature and concentration of the reacting mixture, even when the reactants

constitute a homogeneous liquid phase.

Thiol-ene Reaction with DVB

After this very long series of click reaction experiments, reaction A was reproduced, although

replacing styrene by DVB in order to observe how the thiol-ene reaction would occur when the

vinyl monomer was replaced by a divinyl monomer. In order to maintain similar concentrations

of vinyl bonds, 0.27 g of DVB were added to the reacting mixture, instead of 0.42 g of styrene

as described in the experimental section. This new experiment was named reaction 26 and

FTIR results are shown in Figure 22.



Vol.6, No.1, pp.9-40, March 2018


33


2000 1500 1000 500

a.u

.

Wavenumber (cm-1)

P(S-co-EGDMA)

Reaction 25

Reaction 24

Reaction 23

1723 cm-1

692 cm-1

Figure 21. FTIR spectra of reactions 23, 24, 25 and of P(S-co-EGDMA).

2000 1800 1600 1400 1200 1000 800 600 400

a.u

.

Wavenumber (cm-1)

MPA

DMF

DVB

R.26_liquid

R.26_solid

Figure 22. FTIR spectra of liquid and solid products of reaction 26 and of MPA, DMF

and DVB for comparison.

As in reaction A, FTIR analyses were performed for the solid product and the residual liquid

of reaction 26. Surprisingly, no indication of the characteristic C-S bond that characterizes the

click thiol-ene reaction was observed at 692 cm-1. Moreover, the residual liquid in this case

was composed almost exclusively of DMF. Therefore, it seems clear that the source of the vinyl

bond affects significantly the course of the click reaction.

GPC analysis of the solid product was performed and is presented in Figure 23. The obtained

number average molar mass was equal to 302 g/gmol, with polydispersity index of 1.07.



Vol.6, No.1, pp.9-40, March 2018


34


Differently from what had been observed in reaction A, this time the average molar mass of

the product did not correspond to the sum of the molar masses of DVB and MPA (130.19

g/gmol and 106.14 g/gmol respectively), being closer to the DVB dimer, especially if one

considers the addition of a radical fragment produced by AIBN (NC4H6 = 68 g/mol).

200 400 600 800 1000

No

rma

lize

d r

esp

on

se

(a

.u.)

Molar mass (g/gmol)

Figure 23. Molar mass distribution of the product obtained in reaction 26.

Observing in more detail the UV result presented in Figure 24, one can see that there is a large

band that was also detected with the RI detector and is associated to products with average

molar mass of 302 g/gmol as already discussed. However, the UV detector also provided a

small band, shifted towards lower molar masses. This band can probably be related to the

presence of decomposed AIBN. AIBN typically presents an absorption band located in 365 nm

[70]. But during AIBN decomposition, the traditional absorption band located in 365 nm and

reported by the literature is gradually shifted towards lower wavelengths close to 255 nm

[70,71], in the same region where the GPC analyses were conducted. Therefore, the small band

presented in Figure 24 can probably be caused by the presence of decomposed AIBN, which

would be in accordance with the extremely small molar mass observed.

However, the small band could also be related to the presence of residual reagents. To exclude

this possibility, the reaction reagents were solubilized in THF and injected in the

chromatograph. Results shown in Figure 25 and in Figure 26 indicate that none of the bands

observed for Reaction 26 seem to be related to either DMF, MPA, AIBN or DVB, that present

completely different profiles when compared to Reaction 26. Nevertheless, the fact that part of

the AIBN seems to be decomposed and that its profile is different from the profile of Reaction

26 does not indicate that the small band observed in Figure 24 is not related to AIBN. As

already discussed, the UV results reported in the literature clearly indicate that the absorption

of AIBN changes during its decomposition reaction, and that the final result is a monomodal

band detected ~255nm. So the small band present in Figure 24 is still probably associated to

decomposed AIBN obtained after Reaction 26.



Vol.6, No.1, pp.9-40, March 2018


35


28 29 30 31 32 33

No

rma

lize

d r

esp

on

se

(a

.u.)

RI

UV

Elution time (min)

Figure 24. RI and UV GPC results for reaction 26.

26 28 30 32 34 36 38 40 42 440

200

400

600

800

Det

ecto

r re

sponse

(m

V)

Elution time (minutes)

Styrene

MPA

Reaction26

Figure 25. UV GPC results for reagents and product of Reaction 26 (Graph 1).

Finally, the UV signal provided by the GPC detector (shown in Figure 24) indicated the

existence of multiple reaction products with molecular weight difference of approximately 150

g/mol, which clearly indicates that chain growth may take place and compete with the desired

click reaction. Therefore, keeping a broad perspective, it is not possible to assure that the main

click reaction constitutes the fastest and preferential reaction path in this complex reacting

system in all cases. Perhaps the click reaction is not necessarily the preferred reaction path in

most polymerization reaction systems.



Vol.6, No.1, pp.9-40, March 2018


36


26 28 30 32 34 36 38 40 42 440

200

400

600

800

Det

ecto

r re

spon

se (

mV

)

Elution time (minutes)

DVB

AIBN

DMF

Reaction 26

Figure 26. UV GPC results for reagents and product of Reaction 26 (Graph 2).

Reaction of Styrene with L-cysteine

Styrene apparently exerts a fundamental role in the click reaction scheme among the analyzed

reaction systems, since only the benchmark reaction A seemed to unequivocally produce thiol-

ene products. Therefore, reaction 5 was reproduced, but replacing DVB by styrene in order to

observe if the previously discussed issues (heterogeneous media, thiol reduction, initiatior type,

thiol:ene ratio) would still be important for other styrene reactions. FTIR spectra of the product

of reaction 27 are shown in

Figure 27. As marked in the graph, a band placed at 696 cm-1 could be detected for the product

of reaction 27; however, due to the characteristic styrene and polystyrene bands at the same

spectral region [72], unequivocal characterization of the C-S bands is not possible.

Nevertheless, GPC analyses of the solid product could not be performed because the solid

product was not soluble in the usual analytical solvents, which makes GPC and NMR analyses

not possible.

2000 1800 1600 1400 1200 1000 800 600 400

a.u.

Wavenumber (cm-1)

L-cysteine

Styrene

Polystyrene

Reaction 27696 cm

-1



Vol.6, No.1, pp.9-40, March 2018


37


Figure 27. FTIR spectra of the product of reaction 27 and styrene and L-cysteine for

comparison purposes.

CONCLUSION

In the present work a series of experiments was conducted and multiple issues were raised

concerning thiol-ene click reactions. According to the obtained results, it seems clear that the

efficiency of thiol-ene click reactions depend on the particular reagents used to perform the

reactions, including solvents, sources of vinyl bonds, sources of thiol groups and free-radical

initiators. This particular issue has been largely overlooked in the open literature. In fact, the

dienes used in the present manuscript (HD, DVB and EGDMA) apparently do not form

thioether bonds in the tested conditions (heterogeneous reactions performed at mild

temperature conditions), which are typical for conduction of bioconjugation. Even more

important, in cases where occurrence of the thiol-ene click reaction was not observed, other

side reactions (such as chain growth) took place and caused the formation of extremely

complex mixtures of products. However, when the click reaction was successful (as in the

benchmark reaction between styrene and MPA), the final product seemed to be very pure,

which is in accordance with the click chemistry definition. Therefore, a successful choice of

reagents and reaction conditions not only makes thiol-ene click chemistry possible, but also

seems to inhibit other side reactions, as polymerization, for example.

Based on the obtained results, apparently thiol-ene click reactions should preferentially be

conducted in homogeneous media, which is not clearly discussed in the literature and is

prejudicial for most bioconjugation problems, where the unsaturated polymer matrix

constitutes a solid phase and the protein is dissolved in the suspending medium. Also, it seems

that the choice of the reactants and reaction conditions strongly impact the course of the click

reaction, so that the reacting system must be carefully designed for the thiol-ene click reaction

to be successful. As widely discussed throughout this text, click chemistry reactions are not

always so simple as usually reported in the literature. It seems clear that many aspects of the

click reaction chemistry must be discussed before these reactions can find widespread use for

bioconjugation of biomolecules to polymer matrices.

Acknowledgments

The authors thank CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico,

Brazil) and FAPERJ (Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio

de Janeiro, Brazil) for supporting this work and providing scholarships.

REFERENCES

[1] Kolb, H.C., Finn, M.G. and Sharpless, B. (2001) Click Chemistry: Diverse Chemical

Function from a Few Good Reactions, Angewandte Chemie International Edition, 40

2004 – 2021.

[2] Arseneault, M., Wafer, C. and Morin, J. (2015) Recent Advances in Click Chemistry

Applied to Dendrimer Synthesis, Molecules, 20 9263-9294.

[3] Nguyen, T.L.A., Furrer, J., Nierengarten, J., Barbera, J. and Deschenaux, R. (2016)

Liquid-Crystalline Dendrimers Designed by Click Chemistry Sebastiano Guerra,

Macromolecules, 49 3222–3231.


http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1521-3773


Vol.6, No.1, pp.9-40, March 2018


38


[4] Killops, K.L., Campos, L.M. and Hawker, C.J. (2008) Robust, Efficient, and Orthogonal

Synthesis of Dendrimers via Thiol-ene “Click”, Journal of the American Chemical

Society, 130 5062–5064.

[5] Akeroyd, N. and Klumperman, B. (2011) The combination of living radical

polymerization and click chemistry for the synthesis of advanced macromolecular

architectures, European Polymer Journal, 47 1207–1231.

[6] Vandernbergh, J., Tura, T., Baeten, E. and Junkers, T. (2014) Polymer End Group

Modifications and Polymer Conjugations via “Click” Chemistry Employing

Microreactor Technology, Journal of Polymer Science Part A: Polymer Chemistry, 52

1263–1274.

[7] Arnold, R.M., Patton, D.L., Popik, V.V. and Locklin J. (2014) A Dynamic Duo: Pairing

Click Chemistry and Postpolymerization Modification To Design Complex Surfaces,

Accounts of Chemical Research, 47 2999-3008.

[8] Spicer, C.D. and Davis, B.G. (2014) Selective Chemical Protein Modification, Nature

Communications, 5 1-14.

[9] Wängler, C., Schäfer, M., Schirrmacher, R., Bartenstein, P. and Wängler B. (2011) DOTA

derivatives for site-specific biomolecule-modification via click chemistry: Synthesis and

comparison of reaction characteristics, Bioorganic & Medicinal Chemistry, 19 3864–

3874.

[10] New, K. and Brechbiel, M.W. (2009) Growing Applications of “Click Chemistry” for

Bioconjugation in Contemporary Biomedical Research, Cancer Biotherapy and

Radiopharmaceuticals, 24 289–302.

[11] Li, M., De, P., Gondi, S. and Sumerlin, B. (2008) Responsive Polymer-Protein

Bioconjugates Prepared by RAFT Polymerization and Copper-Catalyzed Azide-Alkyne

Click Chemistry, Macromolecular Rapid Communications, 29 1172–1176.

[12] Droumaguet, B..L. and Velonia, K. (2008) Click Chemistry: A Powerful Tool to Create

Polymer-Based Macromolecular Chimeras, Macromolecular Rapid Communications,

29 1073–1089.

[13] Lahann, J. (2009) Click Chemistry for Biotechnology and Materials Science, John Wiley

& Sons, Michigan, USA.

[14] Canduzini, H.A. (2012) Síntese e Funcionalização de 1,2,3-triazóis via reação de

cicloadição [3+2] de azidas e acetilenos teminais, M.Sc. Dissertation, USP, São Paulo,

SP, Brasil.

[15] Freitas, L., Ruela, F., Pereira, G., Alves, R., Freitas, R. and Santos, L. (2011) A reação

‘click’ na síntese de 1,2,3-triazóis: aspectos químicos e aplicações, Química Nova, 34

1791-1804.

[16] Strausak, D., Mercer, J.F.B., Dieter, H.H., Stremmel, W. and Multhaup, G. (2001)

Copper in disorders with neurological symptoms: Alzheimer’s, Menkes, and Wilson

diseases, Brain Research Bulletin, 55 175–185.

[17] Jiang, Y., Kong, D., Zhao, J., Zhang, W., Xu, W., Li, W. and Xu, G. (2014) A simple,

efficient thermally promoted protocol for Huisgen-click reaction catalyzed by

CuSO4_5H2O in water, Tetrahedron Letters, 55 2410–2414.

[18] Hermanson, G.T. (2013) Bioconjugate Techniques, Academic Press, Oxford, UK.

[19] Lodish, H., Berk, A., Zipursky, S.L., Matsudaira, P., Baltimore, D. and Darnell J. (2000)

Molecular Cell Biology, W.H. Freeman, New York.

[20] Hizal, G., Tunca, U. and Sanyal, A. (2011) Discrete Macromolecular Constructs via the

Diels-Alder “Click” Reaction”, Journal of Polymer Science Part A: Polymer

Chemistry, 49 4103-4120.


http://www.sciencedirect.com/science/journal/00143057

http://www.sciencedirect.com/science/journal/00143057/47/6

http://www.ncbi.nlm.nih.gov/pubmed/25127014

http://www.liebertpub.com/overview/cancer-biotherapy-and-radiopharmaceuticals/8/

http://www.liebertpub.com/overview/cancer-biotherapy-and-radiopharmaceuticals/8/


Vol.6, No.1, pp.9-40, March 2018


39


[21] Sletten, E.M. and Bertozzi, C.R. (2009) Bioorthogonal chemistry: fishing for selectivity

in a sea of functionality. Angewandte Chemie International Edition in English, 48 6974-

98.

[22] Saxon, E. and Bertozzi, C. (2000) Cell Surface Engineering by a Modified Staudinger

Reaction, Science, 287 2007-2010.

[23] Brosnan, S. and Schlaad, H. (2014) Modification of Polypeptide Materials by Thiol-X

Chemistry, Polymer, 55 5511-5516.

[24] Lowe, A.B., Hoyle, C.E. and Bowman, C.N. (2010) Thiol-yne Click Chemistry: A

Powerful and Versatile Methodology for Materials Synthesis, Journal of Materials

Chemistry, 20 4745–4750.

[25] Casini, A., Scozzafava, A. and Supuran, C.T. (2002) Cysteine-Modifying Agents: A

Possible Approach for Effective Anticancer and Antiviral Drug, Environmental Health

Perspectives, 110 801-806.

[26] Cramer, N.B. and Bowman, C.N. (2001) Kinetics of Thiol–Ene and Thiol–Acrylate

Photopolymerizations with Real-Time Fourier Transform Infrared, Journal of Polymer

Science: Part A: Polymer Chemistry, 39, 3311–3319.

[27] Nair, D.P., Podgórski, M., Chatani, S., Gong, T., Xi, W., Fenoli, C and Bowman, C.

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

in Materials Chemistry, Chemistry of Materials, 26 724-744.

[28] O´Connor, A., Marsat, J.N., Mitrugno, A., Flahive, T., Moran, N., Brayden, D. and

Devocelle, M. (2014) Poly(Ethylene glycol)-based backbones with high peptide loading

capacities, Molecules, 19 7559-7577.

[29] Hein, C.D., Liu, X. and Wang, D. (2008) Click Chemistry, a Powerful Tool for

Pharmaceutical Sciences, Pharmaceutical Research, 25 2216–2230.

[30] Shinde, V.S., Lende, D.B. and Parkale, V.V. (2015) Click Reaction: A New Approach in

Chemistry, International Journal of Science, Engineering and Technology Research, 4

821-823.

[31] Arslan, M., Gok, O., Sanyal, R. and Sanyal, A. (2014) Clickable Poly(ethylene glycol)-

Based Copolymers Using Azide–Alkyne Click Cycloaddition-Mediated Step-Growth

Polymerization, Macromolecular Chemistry and Physics, 215 2237−2247.

[32] Al-Kaabi, K. and Van Reenen, A. (2009) Synthesis of Poly(methyl methacrylate-g-

glycidyl azide) Graft Copolymers Using N, N-Dithiocarbamate-Mediated Iniferters,

Journal of Applied Polymer Science, 114 398-403.

[33] Pipkorn, R., Wiessler, M., Waldeck, W., Lorenz, P., Muelhausen, U., Fleischhacker, H,

Koch, M. and Braun, K. (2011) Enhancement of the Click Chemistry for the Inverse

Diels Alder Technology by Functionalization of Amide-Based Monomers, International

Journal of Medical Sciences, 8 387-396.

[34] Farley, R. and Saunders, B.R. (2014) A general method for functionalisation of microgel

particles with primary amines using click chemistry, Polymer, 55 471-480.

[35] Campos, L.M., Killops, K.L., Sakai, R., Paulusse, J.M.J., Damiron, D., Drockenmuller,

E., Messmore, B.W. and Hawker, C.J. (2008) Development of Thermal and

Photochemical Strategies for Thiol-Ene Click Polymer Functionalization,

Macromolecules, 41 7063-7070.

[36] Claudino, M., Van der Meulen, I., Trey, S., Jonsson, M., Heise, A. and Johansson, M.

(2012) Photoinduced Thiol–Ene Crosslinking of Globalide/e-Caprolactone

Copolymers: Curing Performance and Resulting Thermoset Properties, Journal of

Polymer Science Part A: Polymer Chemistry, 50 16–24.

[37] Goldman, A.S., Walther, A., Nebhani, L., Joso, R., Ernst, D., Barner-Kowollik, C.,

Barner, L. and Muller, A.H.E. (2009) Surface Modification of Poly(divinylbenzene)


http://onlinelibrary.wiley.com/doi/10.1002/anie.v36:24/issuetoc


Vol.6, No.1, pp.9-40, March 2018


40


Microspheres via Thiol-Ene Chemistry and Alkyne-Azide Click Reactions,

Macromolecules, 42 3707-3714.

[38] RESIDUAL SOLVENTS USP 30 Chemical Tests (2007). Access in June 19th 2017.

Available in:

http://www.usp.org/sites/default/files/usp_pdf/EN/USPNF/generalChapter467Current.p

df

[39] Decan, M.R., Impellizzeri, S., Marin, M.L. and Scaiano, J.C. (2014) Copper

nanoparticle heterogeneous catalytic ‘click’ cycloaddition confirmed by single-

molecule spectroscopy, Nature Communications, 5 1-8.

[40] Zhao, G., Hafrén, J., Deiana, L. and Córdova, A. (2010) Heterogeneous ‘‘Organoclick’’

Derivatization of Polysaccharides: Photochemical Thiol-ene Click Modification of

Solid Cellulose, Macromolecular Rapid Communications, 31 740–744.

[41] Magenau, A.J.D., Chan, J.W., Hoyle, C.E. and Storey, R.F. (2010) Facile

polyisobutylene functionalization via thiol–ene click chemistry, Polymer Chemistry, 1

831–833.

[42] Uygun, M., Tasdelen, M.A. and Yagci, Y. (2010) Influence of Type of Initiation on

Thiol–Ene ‘‘Click’’ Chemistry, Macromolecular Chemistry and Physics, 211 103–110.

[43] Boyer, C., Granville, A., Davis, T. and Bulmus V. (2009) Modification of RAFT-

Polymers via Thiol-Ene Reactions: A General Route to Functional Polymers and New

Architectures, Journal of Polymer Science: Part A: Polymer Chemistry, 47 3773–3794.

[44] Cirilo, L. (2013) Síntese de novos suportes com diferentes composições e graus de

hidrofobicidade visando avaliar a influência na imobilização da lipase B de Cândida

antártica, Undergraduate Thesis, Instituto de Química, UFRJ, Rio de Janeiro, Brasil.

[45] Cole, M., Jankousky, K., Bowman, C. and Christopher, N. (2013) Redox Initiation of

Bulk Thiol-Ene Polymerizations, Polymer Chemistry, 4 1167-1175.

[46] Hoyle, C., Bowman, E. and Christopher, N. (2010) Thiol-ene click chemistry,

Angewandte Chemie International, 49 1540-1573.

[47] Northop, B.H. and Coffey, R.N. (2012) Thiol–Ene Click Chemistry: Computational and

Kinetic Analysis of the Influence of Alkene Functionality, Journal of the American

Chemical Society, 134 13804-13817.

[48] Hoyle, C.E., Lee, T.Y. and Roper, T. (2004) Thiol–enes: Chemistry of the past with

promise for the future, Journal of Polymer Science: Part A: Polymer Chemistry, 42

5301-5338.

[49] Odian, G. (2004) Principles of Polymerization, John Wiley & Sons, New Jersey, USA.

[50] Cramer, N.B. and Bowman, N. (2011) Kinetics of thiol-ene and Thiol-Acrylate

photopolymerizations with Real-Time Fourier Transform Infrared, Journal of Polymer

Science: Part A: Polymer Chemistry, 39 3311-3319.

[51] Koo, S., Stamenovic, M., Prasath, R., Inglis, A. and Du Prez, F. (2010) Limitations of

radical thiol-ene reactions for polymer–polymer conjugation, Journal of Polymer

Science Part A: Polymer Chemistry, 48 1699–1713.

[52] Illy, N., Robitzer, M., Auvergne, R., Caillol, S., David, G. and Boutevin B. (2014)

Synthesis of Water-Soluble Allyl-Functionalized Oligochitosan and its Modification by

Thiol-ene Addition in Water, Journal of Polymer Science, Part A: Polymer Chemistry,

52 39-48.

[53] Arnold, R.M., Patton, D.L., Popik, V.V. and Locklin J. (2014) A Dynamic Duo: Pairing

Click Chemistry and Postpolymerization Modification To Design Complex Surfaces,

Accounts of Chemical Research, 47 2999−3008.

[54] Chandran, A., Sheena, M., Varghese, H.T., Panicker, C.Y, Manojkumar, T.K., Van

Alsenoy, C. and Rajendran, G. (2012) Vibrational Spectroscopic Study of (E)-4-



Vol.6, No.1, pp.9-40, March 2018


41


(Benzylideneamino)-N-Carbamimidoyl Benzenesulfonamide, ISRN Analytical

Chemistry, 2012 1-11.

[55] Bartholin, M. (1981) Styrene-Divinylbenzene Copolymers, 3. Revisited IR Analysis, Die

Makromolekulare Chemie, 182 2075-2085.

[56] Dericioglu, N., Gargante, C.L., Petroff, O.A., Mendelsohn, D. and Williamson, A.

(2008) Blockade of GABA Synthesis Only Affects Neural Excitability Under Activated

Conditions in Rat Hippocampal Slices, Neurochemistry International, 53 22-32.

[57] Devi, G., Renuga, T.S. and Gunasekaran, S. (2010) FTIR Spectroscopic Analysis of

Normal and Cancerous Human Breast Tissues between 450 cm-1 and 1100 cm-1 using

Trend Analysis, International Journal of ChemTech Research, 2 1426-1433.

[58] Wojcik, J., Altman, K.H. and Scheraga, H.A. (1990) Helix-Coil Stability Constants for

the Naturally Occurring Amino Acids in Water. XXIV. Half-Cystine Parameters from

Random Poly(Hydroxybutylglutamine-coS-Methylthio-L-Cysteine), Biopolymers, 30

121-134.

[59] Moiniddin, S. (2011) Municipal Waste Plastic conversion into Different Category Liquid

Hydrocarbon Fuel, In Chemistry, Emission Control, Radioactive Pollution and Indoor

Air Quality (Ed. Dr. Nicolas Mazzeo, InTech, Shangai, China, 45-80.

[60] Brandrup, J., Immergut, E. and Grulke, E. (1999) Polymer Handbook, Wiley-

Interscience, USA.

[61] Ihre, H. and Larson, A. (2003) Post-modification of a porous support, Patent number

WO 2003046063 A1.

[62] Urban, V.M., Machado, A.L., Vergani, C.E., Jorge, E.G., Santos, P.S., Leite, E.R. and

Canevarolo, S.V. (2007) Degree of Conversion and Molecular Weight of One Denture

Base and Three Reline Resins Submitted to Post-polymerization Treatments, Materials

Research, 10 191-197.

[63] Hussien, S.S. (2014) Biosorption of lanthanum from aqueous solution using Pleurotus

ostreatus basidiocarps, International Journal of Biotechnology Research, 2 26-36.

[64] Schmalz, G. and Bindslev, D.A. (2009) Biocompatibility of dental materials, Springer,

Berlin, Germany.

[65] Flory, R.J. (1953) Principles of Polymer Chemistry, Cornell University Press, Ithaca,

New York.

[66] Stockmayer, W. (1944) Theory of Molecular size distribution and gel formation in

branched-chain polymers, The Journal of Chemical Physics, 12 45-55.

[67] Salamone, J. (1996) Polymeric Materials Encyclopedia, CRC Press, Florida, USA.

[68] Soljic, I., Penovic, T., Juric, A. and Janovic, Z. (2009) Kinetic Study of Free Radical

Copolymerization of Dodecyl Methacrylate and Styrene Using Diperoxide Iniciator,

AIDIC Conference Series, 9 283-291.

[69] Sharma, B.K. (2005) Instrumental Methods of Chemical Analysis, Goel Publishing

House, Meerut, India.

[70] Zhenzhong, L., Zhang, G., Lu, W., Huang, Y., Zhang, J. and Chen T. (2015), UV light-

initiated RAFT polymerization induced self-assembly, Electronic Supplementary

Material (ESI) for Polymer Chemistry, 6 6129-6132.

[71] Pabin-Szafko, B. and Wisniewska, E. (2011) Applying of spectroscopic methods in the

investigation of radical polymerization processes and products, CHEMIK, 65 637-642.

[72] Urbaniak-Domagala, W. (2012) The Use of the Spectrometric Technique FTIR-ATR to

Examine the Polymer Surface, In Advanced Aspects of Spectroscopy (Ed. Dr.

Muhammad Akhyar Farrukh), Intech, Shangai, China, 85-104.


http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)0025-116X

http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)0025-116X

Documents

ON SOME PRACTICAL ASPECTS OF CLICK THIOL-ENE … · Click chemistry reactions have been used since then to produce dendrimers [2-4], to modify polymers [5-7] and to conjugate biomolecules