European Polymer Journal 47 (2011) 524–534

Contents lists available at ScienceDirect

European Polymer Journal

journal homepage: www.elsevier .com/locate /europol j

Feature Article

Green polymer chemistry: Precision synthesis of novel multifunctionalpoly(ethylene glycol)s using enzymatic catalysis

Judit E. Puskas ⇑, Kwang Su Seo, Mustafa Y. SenDepartment of Polymer Science, The University of Akron, Akron, OH 44325-3909, USA

a r t i c l e i n f o a b s t r a c t

Article history:Available online 2 November 2010

Dedicated to Professor Nikos Hadjichristidisin recognition of his contribution to polymerscience.

Keywords:Green chemistryEnzymesPoly(ethylene glycol)FunctionalizationDendrimer

0014-3057 � 2010 Elsevier Ltd.doi:10.1016/j.eurpolymj.2010.10.015

⇑ Corresponding author.E-mail address: jpuskas@uakron.edu (J.E. Puskas

Open access under CC

This paper gives an overview about enzyme catalysis, and reports the precision synthesis ofmultifunctional poly(ethylene glycol)s using this green chemistry approach. Specifically,vinyl acrylate was transesterified with tetraethylene glycol (TEG) and a PEG withDPn = 23, and then (HO)2–TEG–(OH)2 and (HO)2–PEG–(OH)2 were synthesized by theMichael addition of diethanolamine to the acrylate double bonds. These structures willserve as the core of novel dendrimers designed for drug delivery applications.

� 2010 Elsevier Ltd. Open access under CC BY-NC-ND license.

1. Introduction

In tune of the globally increasing interest in ‘‘green orgreener’’ chemistry, including the efficient use of energy,hazard reduction, waste minimization and the use ofrenewable resources, designed to prevent pollution and re-duce resource consumptions [1,2], we have been exploringthe power of enzyme catalysis in polymer functionaliza-tion. Enzymatic catalysis in organic synthesis has emergedas an attractive ‘‘green chemistry’’ alternative to conven-tional chemical catalysis [3]. Enzymatic catalysis has nowbeen applied to polymer synthesis [4–6] and functionaliza-tion [7–12] with several advantages, including high effi-ciency, recyclability, the ability to operate under mildconditions, and environmental friendliness [3]. There areseveral excellent reviews on this topic [13–16]. This paperwill be discussing our latest results towards the designand synthesis of novel multifunctional poly(ethylene gly-col)s PEGs and PEG dendrimers for drug delivery applica-tions. Enzyme catalysis gives unprecedented control in

).

BY-NC-ND license.

polymer chemistry, similarly to living anionic polymerizationtechniques pioneered by Professor Nikos Hadjichristidis[17–21]. This paper is dedicated to him, on the occasionof his retirement.

2. Enzymes in organic chemistry

It was recognized over 30 years ago that enzymes werenot limited to operate in their native aqueous media and theycould efficiently act as catalysts for the biotransformation ofa wide range of substrates in organic solvents. So far, about3000 enzymes have been identified and classified by theInternational Union of Biochemistry and Molecular Biologyinto six categories according to the type of reactions theycan catalyze (Fig. 1) [3]. Among these, oxidoreductases andhydrolyses are the most widely used catalysts in biotransfor-mations. For example, in the 1987–2003 time periods about85% of all enzyme research was performed with hydrolasesand oxidoreductases (60% and 25%, respectively) [3].

Lipases belong to the group of hydrolyses, and are themost popular biocatalysts. They are widely used in esteri-fication, transesterification, aminolysis, and Michael addi-tion reactions in organic solvents [13].

Fig. 1. Classification of enzymes.

J.E. Puskas et al. / European Polymer Journal 47 (2011) 524–534 525

2.1. Lipase-catalyzed transesterification

Lipase-catalyzed transesterification reactions are especiallyuseful for the preparation of optically active compounds byasymmetrization or resolution of racemic or prochiral sub-strates (Fig. 2) [22]. Although both lipases and proteases(the enzymes that hydrolyse peptide bonds in vivo) havebeen used as transesterification catalysts, the latter foundlimited application due to high substrate selectivity [23–25].Several parameters such as the structure of acyl donor, theenzyme source, the structure of alcohol and the reactionstoichiometry must be taken into account for a successfultransesterification.

Transesterification reactions are generally reversible. Inorder to change the reversible nature of the reaction intoan irreversible type, the nucleophilicity of the leavinggroup of the acyl donor should be depleted by the intro-duction of electron-withdrawing groups such as trifluoro-ethyl or trichloroethyl into the ester [26]. Alternatively,the use of oxime esters [27], thioesters [28], and anhy-drides [29] as activated acyl donors have been proposed.The use of enol esters [30], such as vinyl or isopropenyl es-ters appears to be the most useful since they liberateunstable enols as by-products which rapidly tautomerizeto give the corresponding aldehydes or ketones (Fig. 2).

Therefore, the reaction becomes completely irrevers-ible. It was shown that acyl transfer reactions using enolesters are 100 to 1000 times faster than the reactions usingnon-activated esters such as ethyl acetate [30]. Vinyl estersare favored over isopropenyl esters because of less sterichindrance and thus higher reaction rates [31]. Acetalde-hyde, which forms during the reactions with vinyl esters,is known to inactivate the lipases from Candida rugosaand Geotrichum candidum by forming a Schiff’s base withthe lysine residues of the protein; however most lipases,including Candida antarctica lipase B (CALB), tolerate the lib-

Fig. 2. Asymmetrization of a prochiral diol by CALB-catalyzed enantio-selective transesterification to produce an intermediate for antifungalagents.

erated acetaldehyde [32]. Yadav and Trivedi [33] comparedthe catalytic activity of various commercially available li-pases in transesterification of vinyl acetate with n-octanol.CALB, Mucor meihei lipase and Pseudomonas lipase wereimmobilized on macroporous polyacrylic resin beads, anio-nic resin and diatomite, respectively, whereas Candida rug-osa lipase and porcine pancreatic lipase were in free form.It was observed that CALB was the most efficient lipase.The structure of the alcohol is another important parameteraffecting the initial rate and overall conversions in enzymatictransesterification. It was observed that straight-chain alco-hols gave better conversion compared to aromatic andbranched-chain alcohols in the CALB-catalyzed transesteri-fication of vinyl acetate due to less steric hindrance aroundthe hydroxyl group (Fig. 3) [33]. Furthermore, aromaticalcohols with saturated shorter side chains (e.g. benzylalcohol) were more reactive than those with longer butunsaturated side chains (e.g. cinnamyl alcohol); however,no explanation about this finding was given.

Although an increase in ester concentration was shownto increase the rate and conversion of transesterification[33,34], an increase in alcohol concentration might causereduced rates and conversions due to competitive inhibi-tion by the alcohol which can bind reversibly to the en-zyme active site and prevent the binding of the estersubstrate [33,35]. The driving force for alcohol bindingmight be the high polarity of the region around the activeserine site of the enzyme [33]. For example, Yadav andTrivedi [33] varied the concentration of n-octanol whilekeeping the amount of vinyl acetate (1 mol/L) constant inCALB-catalyzed transesterification. They observed an in-crease in the reaction rate when n-octanol concentrationwas increased from 0.25 to 1 mol/L; however, upon furtherincrease the rate decreased.

Comparison of the catalytic activities of CALB and dis-tannoxane, a conventional tin-based catalyst revealed that

Fig. 3. The reactivity of various alcohols in CALB-catalyzed transesteri-fication of vinyl acetate.

HO

Cln

t-BuOK

HO

n

O

n

O

+H

O

CALB, Hexane50 oC, 24 h

O

O

Fig. 5. Unsuccessful transesterification of vinyl methacrylate with dehy-drochlorinated PIB–OH prepared from the a-MSE/TiCl4 initiator system.

526 J.E. Puskas et al. / European Polymer Journal 47 (2011) 524–534

the transesterification of vinyl acetate with 2-phenyl-1-propanol (Fig. 4) catalyzed by CALB was complete in 2 h,while the traditional tin-based catalyst yielded 95% con-version in 12 h [36].

The reaction was irreversible because the vinyl alcoholproduct instantaneously tautomerized into acetaldehyde,which was easily removed from the system due to itslow volatility. However, when the methyl group was re-placed by a polyisobutylene chain, no reaction occurreddue to the steric hindrance (Fig. 5) [37].

The tertiary chloride chain end of PIB–OH (Mn = 8200 g/mol, Mw/Mn = 1.13) was removed by dehydrochlorinationusing t-BuOK [38]; and the resulting polymer was reactedwith vinyl methacrylate in the presence of CALB. 1H NMRspectroscopy revealed that the polymer remained intactas the spectrum of the reaction product was identical tothat of the starting material.

Another example of steric effect on the transesterifica-tion was demonstrated using primary hydroxy-functional-ized polystyrenes, which were prepared by end-capping ofpoly(styryl)lithium with ethylene oxide (Mn = 2100 g/mol,Mw/Mn = 1.07) and (CH3)2SiClH followed by hydrosilationwith allyl alcohol (Mn = 2600 g/mol, Mw/Mn = 1.06) [39],respectively. The ethylene oxide end-capped PS–OH didnot react while a PS–OH with a (CH3)2Si–CH2 spacer wasquantitatively methacrylated by transesterification of vinylmethacrylate within 48 h [10].

2.2. Lipase-catalyzed Michael addition

The Michael addition, one of the most fundamentalreactions in organic synthesis, is the 1,4-addition of anucleophile to an activated a,b-unsaturated carbonyl com-pound. The first enzymatic Michael addition was reportedin 1986, in which 2-aminophenol was reacted with 2-(tri-fluoromethyl) propenoic acid in a phosphate buffer in thepresence of the lipase from Candida cylindracea, reaching83% conversion [40]. However, only a few groups have fo-cused on enzyme-catalyzed Michael-type additions since

Fig. 4. Transesterification of vinyl acetate: com

then. Hydrolases, specifically lipases and proteases, havebeen used as Michael addition catalysts. Cai et al. [41]investigated the Michael addition of pyrimidine deriva-tives to acrylates using an alkaline protease from Bacillussubtilis (Fig. 6). Among several organic solvents with differ-ing polarity, DMSO was found to be the most effective sol-vent. Higher yields were obtained when the pyrimidinehad a more electron-withdrawing group and smaller sub-stituent and the acrylate had a shorter alcohol chain, butno explanation for these results were given.

The same researchers carried out a systematic study ofthe Michael-type addition of imidazoles to acrylates usingseveral commercially available serine hydrolases [42]. Inthe reaction of imidazole with methyl acrylate, Amano li-pase M from Mucor javanicus showed the highest catalyticactivity (96% conversion). In addition, higher conversionswere achieved in less polar solvents. However, in the reac-tion of imidazole derivatives containing a nitro group, thistrend was not observed due to their poor solubility in non-polar solvents. Donors with a more electron-withdrawinggroup and acceptors with a shorter alcohol chain gave bet-ter yields. When the acceptor had the same chain length,the yields with acrylates were higher than with methacry-lates. In another study, the initial reaction rate of the Mi-chael addition of 4-nitro-imidazole to methyl acrylate inDMSO in the presence of Amano acylase from Aspergillusoryzae, a protease containing a zinc ion in the active site,was found to be 638-fold faster than the reaction in the

parison of CALB and a tin-based catalyst.

+NH

NHR

O O

R: H, CH3, F, Br

O

OR'

50 oC

Protease fromBacillus subtilis

NH

NR

O O

O

O R'

R': CH3CH2CH3CH2CH2CH2CH3

Fig. 6. Enzymatic Michael addition of uracil and its derivatives toacrylates.

Fig. 8. CALB-catalyzed esterification of poly(acrylic acid) with variouspolyols in bulk.

J.E. Puskas et al. / European Polymer Journal 47 (2011) 524–534 527

absence of a biocatalyst [43]. This enzyme was also shownto effectively catalyze the addition of a variety of N-hetero-cycles to a,b-unsaturated compounds.

The addition of 1,3-dicarbonyl compounds to a,b-unsat-urated compounds was also reported (Fig. 7) [44]. Compar-ison of several hydrolases revealed that D-aminoacylasefrom Escherichia coli, which is another zinc-dependent pro-tease, was the best catalyst for the addition of ethyl aceto-acetate to methyl vinyl ketone in 2-methyl-2-butanol.When the reaction was carried out in different organic sol-vents, no correlation between the enzyme activity and thepartition coefficient (logP) of the solvent was observed.However, this enzymatic behavior could not be explained.

3. Enzymes for the functionalization of syntheticpolymers

Before our group has started working on the functional-ization of preformed synthetic polymers, only a few exam-ples were available in the literature. When polybutadienewas reacted with hydrogen peroxide and a catalyticamount of acetic acid in the presence of CALB, only about60% of the units in 1,4-enchainment (cis and trans) wereepoxidized, without any change to the pendant vinylgroups [9]. The inability to reach higher conversion was as-cribed to the conformational changes in the partially epox-idized polymer. Battistel et al. [45] reported the hydrolysisof the pendant nitrile groups of polyacrylonitrile fibers intothe corresponding amides using a nitrile hydratase enzymewith 16% conversion. The CALB-catalyzed esterification ofpoly(acrylic acid) with various polyols showed that thereactions were highly regioselective and that only one ofthe alcohol groups of the polyol reacted with a pendant ac-rylic acid group [46] (Fig. 8). Furthermore, the enzyme didnot catalyze intermolecular reactions between the free hy-droxyl groups of the esterified polyol and the carboxylicacid groups of another poly(acrylic acid) chain. Based on1H NMR spectroscopy, it was found that the maximumesterification with glycerol was 32%.

Heise and coworkers [47] reported the grafting of vinylacetate onto poly[styrene-co-(p-vinyl-2-phenylethanol)]containing 45% secondary OH groups by CALB-catalyzed

Fig. 7. Enzyme-catalyzed Michael addition of 1,3-dicarbonyl compoundsto a,b-unsaturated compounds.

transesterification in toluene (Fig. 9(A)). The enzyme stere-oselectively catalyzed the reaction with vinyl acetate in Rconfiguration. When a backbone containing 100% primaryhydroxy groups was used, 75% of the OH groups reactedwith vinyl acetate and changing the reaction parametersdid not increase the conversion significantly. Similarly,when a copolymer of styrene and 4-vinylbenzyl alcoholwith 10% primary hydroxyl functionality was used, vinylacetate grafting was 95% as determined by 1H NMR spec-troscopy [48]. It was also observed that when [poly(sty-rene-co-methyl-2-(4-styryl) acetate)] was reacted withbenzyl alcohol, there was no reaction at all [49]. This prob-lem was overcome by introducing a spacer in the ester-functionalized polymer (Fig. 9(B)), and up to 78% conversionwas reached [49].

Poly[N-(2-hydroxypropyl)-11-methacryloylaminoun-decanamide-co-styrene], a comb-like polymer, and its cor-responding monomer was acylated with vinyl acetate,phenyl acetate, 4-fluorophenyl acetate and phenyl stearatein the presence of a lipase from Pseudomonas fluorescens[50]. The copolymer was acylated with about 40% conver-sion when phenyl acetate was used as the acyl donor. Thehigher reactivity of the monomer compared to the copoly-mer indicated the effect of steric hindrance on the reactionkinetics. b-Galactosidase from Aspergillus oryzae was em-ployed as the catalyst for the functionalization of PEG withgalactosylate in water using lactose as the donor [51]. 1HNMR analysis revealed 30–50% conversion. Gross andcoworkers [52] prepared organosilicon carbohydrate con-jugates by CALB-catalyzed regioselective esterification ofsiloxanes having diacid end groups with a,b-ethyl glyco-side under vacuum in bulk (Fig. 10). The organosilicon re-acted with the primary hydroxyl group of the a,b-ethylglycoside; and the extent of esterification was determinedby electrospray ionization mass spectrometry as 99%.

Recently, the first examples of quantitative functionali-zation of synthetic polymers using CALB-catalyzed reactionswith and without organic solvents has been reported byour group [10–12,36,37]. The primary hydroxyl groups ofhydroxyl-functionalized polyisobutylenes (PIBs) were quan-titatively methacrylated by transesterification of vinyl meth-acrylate in the presence of CALB within 24 h in hexaneand 2 h in bulk, respectively (Fig. 11) [12,37]. Specifically,asymmetric methacrylation of a,x-hydroxy functionalizedPIB was achieved by the regioselective transesterificationof vinyl methacrylate using CALB in hexane within 24 h,leaving the sterically hindered hydroxyl group intact(Fig. 11) [10,37].

Commercially available polydimethylsiloxanes (PDMSs):HO–PDMS–OH (HO–[Si(CH3)2–O]n–Si(CH3)2–OH, Mn = 3200g/mol); PDMS-monocarbinol (HO–(CH2)2–O–(CH2)3–[Si(CH3)2

+Toluene

CALBn m

X OH

X: H or CH3

O

On m

X O

O

+H

O(A)

+CALB

n m

ROH:

+(B)

ROH

O OO

O

OH , OH

n m

O O R

O

O

CH3OH

Fig. 9. CALB-catalyzed transesterifications: (A) transesterification of vinyl acetate with poly(styrene-co-4-vinyl alcohol)s and (B) transesterification ofpoly[methyl-6-(2-(4-vinylphenyl)acetoxyl) hexanoate-co-styrene with benzyl alcohol and 1-phenylethanol.

Fig. 10. CALB-catalyzed esterification of diacid end-blocked siloxanes with a,b-ethyl glycoside.

528 J.E. Puskas et al. / European Polymer Journal 47 (2011) 524–534

–O]n–Si(CH3)2–C4H9, Mn = 5000 g/mol) and PDMS-dicarbi-nol (HO–(CH2)2–O–(CH2)3–Si(CH3)2–O–[Si(CH3)2–O]n–Si(CH3)2–(CH2)3–O–(CH2)2–OH, Mn = 4500 and 1000 g/mol)were quantitatively functionalized by transesterificationof 1.5 eq. of vinyl methacrylate under solventless condi-tions within 2 h in the presence of CALB (10 wt.% relativeto the total weight of reactants) [10,36].

PEGs with various molecular weights and molecularweight distributions were quantitatively methacrylatedor acylated by CALB-catalyzed transesterification in THFwithin 24 h (Fig. 12) [11].

Thymine-functionalized PEG was successfully preparedby the Amano lipase M-catalyzed Michael addition of thy-mine to PEG diacrylate within 72 h [10]. The PEG-diacry-

Fig. 11. CALB-catalyzed methacrylation of PIB–OHs. PIB–OH (Mn = 5200 g/mol,Mw/Mn = 1.34) made from Glissopal�2300 [53], and asymmetric telechelic HO–P

late was prepared by the transesterification of vinylacrylate with HO–PEG–OH (Mn = 2000 g/mol, Mw/Mw =1.91) in the presence of CALB. The CALB-catalyzed Michaeladdition of an amino terminated poly(ethylene glycol)methyl ether (Mn = 2000 g/mol, Mw/Mn = 1.05) to 1,3,5-tri-acryloylhexahydro-1,3,5-triazine yielded a monofunctionalproduct with two acrylate groups available for further con-jugation with other functional groups [10].

Recently, we were able to functionalize PEGs under sol-vent-free conditions within 4 h, by dissolving low molecu-lar weight HO–PEG–OH (Mn = 1050 and 2000 g/mol) in thecorresponding acyl donors at 50 �C (Fig. 13). 1H and 13CNMR along with MALDI-ToF confirmed quantitative con-version with the expected structures.

Mw/Mn = 1.09) made from PIB-allyl [53], Glissopal–OH (Mn = 3600 g/mol,IB–OH (Mn = 7200 g/mol, Mw/Mn = 1.04).

Fig. 12. CALB-catalyzed acylation and methacrylation of PEGs.

O

50 oC, 4 h

(5 eq. per -OH)

CALB (10 wt%)HO O O H

n

O O O n

O

O

H

O

O

(5 eq. per -OH)

(5 eq. per -OH)

O

O

O

O

O O O n

O O

O O O n

O O

Fig. 13. Quantitative functionalization of PEGs in bulk within 4 h.

J.E. Puskas et al. / European Polymer Journal 47 (2011) 524–534 529

Based on these results, we set out to design and con-struct PEG dendrimers using enzyme catalysis.

4. Dendrimer synthesis

Dendrimers have extensively been used in novel drugdelivery systems, due to multivalent surfaces, and highdrug carrying capacity [54–58]. However, the synthesis ofdendrimers remains an expensive and difficult processwith a complex mixture of products, resulting in stochasticconjugation of both targeting ligands and drugs [59–61].

Fig. 14. Synthetic processes for PEG dendrimers

PEG-based dendrimers [62–68] are outstanding candidatesfor targeted drug delivery, because of a unique combina-tion of the properties of PEG including water solubility,non-toxicity and limited recognition by the immune sys-tem [69–72].

Gnanou and et al. reported the synthesis of multifunc-tional PEGs prepared by an iterative divergent approachcombining anionic ring-opening polymerization (AROP)of ethylene oxide from multi-hydroxylated precursorsand subsequent branching of the PEO chain ends (Fig. 14)[66–68]. PEGG1(OH)3 (90% conversion) was prepared by

by anionic ring-opening polymerization.

Fig. 15. Dendrimer design for targeted drug delivery.

530 J.E. Puskas et al. / European Polymer Journal 47 (2011) 524–534

AROP of ethylene oxide with 1,1,1-tris(hydroxymethyl)ethane as an initiator, in the presence of diphenylmethylpo-tassium. The nucleophilic substitution of allyl chloridewith PEGG1(OH)3 followed by the oxidation of the allylgroups in the resulting polymer using OsO4 and subse-quent polymerization with ethylene oxide yielded PEG-G1(OH)6. PEG dendrimers up to Mn = 900000 g/mol (Mw/Mn = 1.28) with theoretically 384 outer hydroxy functionalgroups were obtained, but the overall efficiency was lessthan 20%.

Preliminary to the precision synthesis of PEG dendri-mers for targeted drug delivery applications (the designis shown in Fig. 15: the dendrimer should carry multipletargeting, therapeutic, imaging and diagnostic agents),here we report the precision synthesis of four-functionalTEG and PEG using enzymatic catalysis.

4.1. (HO)2–TEG–(OH)2

A four-functional tetraethylene glycol (TEG) was pre-pared by sequential transesterification of vinyl acrylate

Fig. 16. Enzyme-catalyzed synth

followed by the Michael addition of diethanolaminecatalyzed by CALB (Fig. 16(B); details are given in the Sup-porting Information). Monitoring the reaction with thinlayer chromatography (TLC) (Fig. 16(C)) revealed that bothtransesterification and Michael addition reactions weresuccessful as quantitative conversions were reached with-in 24 and 2 h, respectively.

Fig. 17 shows the 1H NMR spectra of the products ineach step. In the spectrum of the TEG diacrylate, the inte-gration ratios of the vinyl [d = 5.8 (3), d = 6.2 (2), andd = 6.3 (1)], and methylene protons [d = 4.2 (4)] were1:1:1:2 as expected (Fig. 17(A)). After the Michael addition,the vinyl protons from the acrylate groups disappeared andnew signals corresponding to the protons of the OH endgroups and that of newly formed chain end appeared withthe expected integration ratios of 1:1:1:2:2 at d = 4.2 ppmand d = 2.7 ppm (a), d = 2.4 ppm (b), d = 2.5 ppm (c), andd = 3.3 ppm (d), respectively. (Fig. 17(B)).

The 13C NMR spectrum of the TEG acrylate (A) and(HO)2–TEG–(OH)2 (B) also confirmed the expected struc-tures (Fig. 18). The carbons attached to the hydroxyl groupin the starting material at d = 60.13 ppm shifted downfieldto d = 63.41 ppm (d) after the reaction and the carbon res-onances of the acrylate group appeared at d = 165.47 ppm(c), d = 131.77 ppm (b) and d = 128.08 ppm (a) correspond-ing to carbonyl carbon, a-carbon and the vinyl carbonsconnected to the a-carbon, respectively (Fig. 18(A)). UponMichael addition of diethanolamine to the TEG diacrylate(2 h) the carbon resonances of the acrylate groups at d =172.23 ppm (c), d = 32.11 ppm (b) and d = 50.02 ppm (a)disappeared, and new signals appeared at d = 59.56 ppm(1) and d = 56.30 ppm (2), confirming the successful synthesisof the TEG dendrimer (Fig. 18(B)).

4.2. (HO)2–PEG–(OH)2

Using a similar synthetic strategy, a four-functional PEGwas also synthesized. a,x-hydroxy PEG (HO–PEG–OH,Mn = 1050 g/mol, Mw/Mn = 1.08) was reacted with vinylacrylate (3.0 equivalent per –OH of PEG) in the presence

esis of (HO)2–TEG–(OH)2.

Fig. 17. 1H NMR spectra of TEG acrylate (A) and (HO)2–TEG–(OH)2 (B).

Fig. 18. 13C NMR spectra of TEG acrylate (A) and (HO)2–TEG–(OH)2 (B).

J.E. Puskas et al. / European Polymer Journal 47 (2011) 524–534 531

Fig. 19. 1H and 13C NMR spectra of (HO)2–PEG–(OH)2 in DMSO-d6.

532 J.E. Puskas et al. / European Polymer Journal 47 (2011) 524–534

of CALB for 24 h at 50 �C. The resulting PEG diacrylate wasthen reacted with diethanolamine (1.0 equivalent per acry-late group of PEG) in DMSO at 50 �C for 24 h via CALB-cat-alyzed Michael addition to obtain (HO)2–PEG–(OH)2 (99.5%yield by 1H NMR). Fig. 19 shows the 1H and 13C NMR spec-tra of the (HO)2–PEG–(OH)2 product. The integral ratio ofthe chain end hydroxyl protons (8) to the methylene pro-tons next to the ester linkage (3) was 1:1 confirming theexpected structure. The 13C NMR spectrum of (HO)2–PEG–(OH)2 (Fig. 19(B)) was identical to that of (HO)2–TEG–(OH)2 (Fig. 18(B)).

5. Summary

In summary, we successfully synthesized (HO)2–TEG–(OH)2 and (OH)2–PEG–(OH)2 as a core of novel dendrimersusing sequential CALB-catalyzed transesterification andMichael addition. The use of enzymes provides us withunprecedented control in polymer functionalization andbuilding of novel architectures. Specifically, the designand synthesis of novel polymeric architectures by utilizingenzyme selectivity has the potential of revolutionizingpolymer science.

Acknowledgements

This work is supported by National Science Foundationunder DMR-509687 and DMR-0804878. We wish to thankThe Ohio Board of Regents and The National Science Foun-dation for funds used to purchase the NMR (CHE-0341701and DMR-0414599) instruments used in this work.

Appendix A. Supplementary data

Supplementary data associated with this article canbe found, in the online version, at doi:10.1016/j.eurpolymj.2010.10.015.

References

[1] Crabtree RH. In: Meyer H, Ghisalba O, Leresche JE, editors. Handbookof green catalysis Biocatalysis, vol. 3. Weinheim: Wiley-VCH VerlagGmbH & Co; 2009. p. 174.

[2] Roberto B. In: Lombardo M, Trombini C, editors. Eco-friendlysynthesis of fine chemicals. Cambridge: Royal Society ofChemistry; 2009. p. 3.

[3] Faber KM. In: Biotransformations in organic chemistry. NewYork: Springer-Verlag; 2004.

[4] Chaudry AK, Beckman EJ, Russell AJ. Rational control of polymermolecular weight and dispersity during enzyme-catalyzed polyestersynthesis in supercritical fluids. J Am Chem Soc 1995;117(13):3728–33.

[5] Svirkin YY, Xu J, Gross RA, Kaplan D, Swift G. Enzyme-catalyzedstereoelective ring-opening polymerization of a-methyl-b-propio-lactone. Macromolecules 1996;29(13):4591–7.

[6] Banerjee S, Ramannair P, Wu K, John VT, McPherson G, Akkara J, et al.In: Gross RA, Kaplan DL, Swift GS, editors. Enzymes in polymersynthesis. Washington, DC: ACS Symposium Series 684, AmericanChemical Society; 1998. p. 125.

[7] Chen HN, Gu QM. Biotransformation of polysaccharides. In: WangPG, Bertozzi CR, editors. Glycochemistry: principles, synthesis andapplications. New York: Marcel Dekker; 2001. p. 567.

[8] Gubitz GM, Cavaco-Paulo A. New substrates for reliable enzymes:enzymatic modification of polymers. Curr Opin Biotech 2003;14(6):577–82.

[9] Javie AWP, Overton N, St Pourcain CB. Enzyme catalysedmodification of synthetic polymers. J Chem Soc Perkin Trans 1999;1:2171–6.

[10] Puskas JE, Sen MY, Seo KS. Green polymer chemistry using nature’scatalysts: enzymes. J Polym Sci Part A: Polym Chem 2009;47:2959–76.

J.E. Puskas et al. / European Polymer Journal 47 (2011) 524–534 533

[11] Puskas JE, Sen MY, Kasper JR. Green polymer chemistry: telechelicpoly(ethylene glycols) via enzymatic catalysis. J Polym Sci Part A:Polym Chem 2008;46:3024–8.

[12] Sen MY, Puskas JE, Ummadisetty S, Kennedy JP. Green polymerchemistry: II Enzymatic synthesis of methacrylate-terminatedpolyisobutylenes. Macromol Rapid Commun 2008;29:1598–602.

[13] Gross RA, Kumar A, Kalra B. Polymer synthesis by in vitro enzymecatalysis. Chem Rev 2001;101:2097–124.

[14] Kobayashi S, Uyama H, Kimura S. Enzymatic polymerization. ChemRev 2001;101:3793–818.

[15] Varma IK, Albertsson AC, Rajkhowa R, Srivastava RK. Enzymecatalyzed synthesis of polyesters. Prog Polym Sci 2005;30(10):949–81.

[16] Kobayashi S, Makino A. Enzymatic polymer synthesis: anopportunity for green polymer chemistry. Chem Rev 2009;109(11):5288–353.

[17] Hadjichristidis N, Roovers JEL. Synthesis and solution properties oflinear, four-branched, and six-branched star polyisoprenes. J PolymSci Polym Phys Ed 1974;12(12):2521–33.

[18] Hadjichristidis N, Guyot A, Fetters LJ. Star-branched polymers. 1. Thesynthesis of star polyisoprenes using octa- and dodecachlorosilanesas linking agents. Macromolecules 1978;11(4):668–72.

[19] Hadjichristidis N, Pitsikalis M, Pispas S, Iatrou H. Polymers withcomplex architecture by living anionic polymerization. Chem Rev2001;101(12):3747–92.

[20] Hadjichristidis N, Iatrou H, Pitsikalis M, Mays J. Macromoleculararchitectures by living and controlled/living polymerizations. ProgPoly Sci 2006;31(12):1068–132.

[21] Matyjaszewski K, Müller AHE. Controlled and living polymeriza-tions. In: Hadjichristidis N, Pitsikalis M, Iatrou H, Sakellariou G,editors. Macromolecular architectures by living and controlled/living polymerizations. Weinheim: Wiley–VCH; 2009. p. 343–443.

[22] Morgan B, Dodds DR, Zaks A, Andrews DR, Klesse R. Enzymicdesymmetrization of prochiral 2-substituted-1, 3-propanediols: apractical chemoenzymic synthesis of a key precursor of SCH 51048,a broad-spectrum orally active antifungal agent. J Org Chem1997;62:7736–43.

[23] Gotor V. Pharmaceuticals through enzymatic transesterification andenzymatic aminolysis reactions. Biocatal Biotransform 2000;18:87–103.

[24] Santaniello E, Ferraboschi P, Grisenti P. Lipase-catalyzedtransesterification in organic solvents: applications to thepreparation of enantiomerically pure compounds. Enzyme MicrobTechnol 1993;15:367–82.

[25] Bordusa F. Proteases in organic synthesis. Chem Rev 2002;102:4817–67.

[26] Kirchner G, Scollar MP, Klibanov AM. Resolution of racemic mixturesvia lipase catalysis in organic solvents. J Am Chem Soc 1985;107:7072–6.

[27] Ghogare A, Kumar GS. Oxime esters as novel irreversible acyltransfer agents for lipase catalysis in organic media. J Chem SocChem Commun 1989:1533–5.

[28] Oehrner N, Martinelle M, Mattson A, Norin T, Hult K. Thioethyl-,vinyl-, ethyl octanoate esters and octanoic acid as acyl donors inlipase catalyzed acyl transfer reactions. Biocatalysis 1994;9:105–14.

[29] Bianchi D, Cesti P, Battistel E. Anhydrides as acylating agents inlipase catalyzed stereoselective esterification of racemic alcohols. JOrg Chem 1988;53:5531–4.

[30] Wang YF, Lalonde JJ, Momongan M, Bergbreiter DE, Wong CH.Lipase-catalyzed irreversible transesterifications using enol esters asacylating reagents: preparative enantio- and regioselective synthesesof alcohols, glycerol derivatives, sugars and organometallics. J AmChem Soc 1988;110:7200–5.

[31] Faber K, Riva S. Enzyme-catalyzed irreversible acyl transfer.Synthesis 1992:895–910.

[32] Weber HK, Stecher H, Faber K. Sensitivity of microbial lipases toacetaldehyde formed by acyl-transfer reactions from vinyl esters.Biotechnol Lett 1995;17:803–8.

[33] Yadav GD, Trivedi AH. Kinetic modeling of immobilized-lipasecatalyzed transesterification of n-octanol with vinyl acetate innon-aqueous media. Enzyme Microb Technol 2003;32:783–9.

[34] Yadav GD, Lathi PS. Lipase catalyzed transesterification of methylacetoacetate with n-butanol. J Mol Catal B: Enzym 2005;32:107–13.

[35] Rizzi M, Stylos P, Riek A, Reuss M. A kinetic study of immobilizedlipase catalyzing the synthesis of isoamyl acetate by transe-sterification in n-hexane. Enzyme Microb Technol 1992;14: 709–14.

[36] Puskas JE, Sen MY. Green polymer chemistry: enzymaticfunctionalization of liquid polymers in bulk. In: Cheng NH, Gross

RA, editors. Green polymer chemistry:biocatalysis and biomaterials.Washington, DC: ACS Symposium Series 1043, American ChemicalSociety; 2010. p. 417–24.

[37] Sen MY. PhD thesis. Akron OH: The University of Akron; 2009.[38] Kennedy JP, Chang VSC, Smith RA, Ivan B. New telechelic polymers

and sequential copolymers by polyfunctional initiator-transferagents (inifers). V. Synthesis of a-tert.-butyl-x-isopropenylpolyisobutylene and a, x- di(isopropenyl)polyisobutylene. Polym Bull1979;1:575–80.

[39] Quirk RP, Kim H, Polce MJ, Wesdemiotis C. Anionic synthesis ofprimary amine functionalized polystyrenes via hydrosilation ofallylamines with silyl hydride functionalized polystyrenes.Macromolecules 2005;38(19):7895–906.

[40] Kitazume T, Ikeya T, Murata K. Synthesis of optically activetrifluorinated compounds: asymmetric Michael addition withhydrolytic enzymes. J Chem Soc Chem Commun 1986:1331–3.

[41] Cai Y, Sun XF, Wang N, Lin XF. Alkaline protease from Bacillus subtiliscatalyzed Michael addition of pyrimidine derivatives to a, b-ethylenic compounds in organic media. Synthesis 2004:671–4.

[42] Cai Y, Wu Q, Xiao YM, Lv DS, Lin XF. Hydrolase-catalyzed Michaeladdition of imidazoles to acrylic monomers in organic medium. JBiotechnol 2006;121:330–7.

[43] Qian C, Xu JM, Wu Q, Lv DS, Lin XF. Promiscuous acylase-catalyzedaza-Michael additions of aromatic N-heterocycles in organic solvent.Tetrahedron Lett 2007;48:6100–4.

[44] Xu JM, Zhang F, Wu Q, Zhang QY, Lin XF. Hydrolase-catalyzedMichael addition of 1, 3-dicarbonyl compounds to a, b-unsaturatedcompounds in organic solvent. J Mol Catal B: Enzym 2007;49:50–4.

[45] Battistel E, Morra M, Marinetti M. Enzymatic surface modification ofacrylonitrile fibers. Appl Surf Sci 2001;177:32–41.

[46] Sahoo B, Gao W, Gross RA. Lipase-catalyzed esterification ofpolyacrylic acid with extraordinary selectivity. Polym Prepr2003;44(2):581–2.

[47] Duxbury CJ, Hilker I, de Wildeman SMA, Heise A. Enzyme-responsivematerials: chirality to program polymer reactivity. Angew Chem, IntEd 2007;46:8452–4.

[48] Duxbury CJ, Cummins D, Heise A. Selective enzymatic grafting bysteric control. Macromol Rapid Commun 2007;28:235–40.

[49] Padovani M, Hilker I, Duxbury CJ, Heise A. Functionalization ofpolymers with high precision by dual regio- and stereoselectiveenzymatic reactions. Macromolecules 2008;41:2439–44.

[50] Pavel K, Ritter H. Enzymes in polymer chemistry. 6. Lipase-catalyzedacrylation of comb-like methacrylic polymers containing OH groupsin the side chains. Makromol Chem 1992;193:323–8.

[51] Cheng HN, Gu QM. Biotransformation of polysaccharides. In: WangPG, Bertozzi CR, editors. Glycochemistry: principles, synthesis andapplications. New York: Marcel Dekker; 2001. p. 567–80.

[52] Sahoo B, Brandstadt KF, Lane TH, Gross RA. ‘‘Sweet silicones’’:biocatalytic reactions to form organosilicon carbohydratemacromers. Org Lett 2005;7:3857–60.

[53] Ummadisetty S, Kennedy JP. Quantitative syntheses of novelpolyisobutylenes fitted with terminal primary -Br, -OH, -NH2, andmethacrylate termini. J Polym Sci Part A: Polym Chem 2008;46:4236–42.

[54] Bosman AW, Jassen HM, Meijer EW. About dendrimers: structure,physical properties, and applications. Chem Rev 1999;99:1665–88.

[55] Svenson S, Tomalia DA. Dendrimers in biomedical applications—reflections on the field. Adv Drug Deliv Rev 2005;57(16):2106–29.

[56] Yang H, Kao WJ. Dendrimers for pharmaceutical and biomedicalapplications. J Biomater Sci Polym Ed 2006;17:3–19.

[57] Majoros IJ, Baker Jr JR. Dendrimer-based nanomedicine. Singapore:Pan Stanford Publishing; 2008.

[58] Medina SH, El-Sayed EH. Dendrimers as carriers for delivery ofchemotherapeutic agents. Chem Rev 2009;109:3141–57.

[59] Peterson J, Allikmaa V, Subbi J, Pehk T, Lopp M. Structural deviationin poly(amidoamine) dendrimers: a MALDI-ToF MS analysis. EurPolym J 2003;39:33–42.

[60] Islam MT, Majoros IJ, Baker Jr JR. HPLC analysis of PAMAM dendrimerbased on multifunctional devices. J Chromatogr B 2005;822:21–6.

[61] Hong S, Leroueil PR, Majoros IJ, Orr BG, Baker JR, Banaszak Holl MM.The binding avidity of a nanoparticle-based multivalent targeteddrug delivery platform. Chem Biol 2007;14:107–15.

[62] Okuda T, Kawakami S, Akimoto N, Niidome T, Yamashita F, HashidaM. PEGylated lysine dendrimers for tumor-selective targeting afterintravenous injection in tumor-bearing mice. J Controlled Rel2006;116:330–6.

[63] Guillaudeu SJ, Fox ME, Haidar YM, Dy EE, Szoka FC, Frechet JMJ.PEGylated dendrimers with core functionality for biologicalapplications. Bioconjugate Chem 2008;19:461–9.

534 J.E. Puskas et al. / European Polymer Journal 47 (2011) 524–534

[64] Greenwald RB, Choe YH, McGuire J, Conover CD. Effective drugdelivery by PEGylated drug conjugate. Adv Drug Deliv Rev2003;55:217–50.

[65] Kojima C, Kono K, Maruyama K, Takagishi T. Synthesis ofpolyamidoamine dendrimers having poly(ethylene glycol) graftsand their ability to encapsulate anticancer drugs. Bioconjugate Chem2000;11:910–6.

[66] Feng X, Taton D, Chaikof EL, Gnanou Y. Toward an easy access todendrimer-like poly(ethylene oxide)s. J Am Chem Soc2005;127(351):10956–66.

[67] Feng X, Taton D, Borsali R, Chaikof EL, Gnanou Y. PH responsivenessdendrimer-like poly(ethylene oxide)s. J Am Chem Soc2006;128(35):11551–62.

[68] Feng X, Taton D, Chaikof EL, Gnanou Y. Fast access to dendrimer-likepoly(ethylene oxide)s through anionic ring-opening polymerizationof ethylene oxide and use of nonprotected glycidol as branchingagent. Macromolecules 2009;42(19):7292–8.

[69] Duncan R. The drawing era of polymer therapeutics. Nature RevDrug Disc 2003;2:347–60.

[70] Duncan R. Polymer conjugates as anticancer nanomedicines. Nat RevCancer 2006;6:688–701.

[71] Farokhzad OC, Langer R. Nanomedicine: developing smartertherapeutic and diagnostic modalities. Adv Drug Deliv Rev2006;58:1456–9.

[72] Tong R, Cheng J. Anticancer polymeric nanomedicines. Polym Rev2007;47:345–81.

Dr. Puskas received a PhD in plastics andrubber technology in 1985, and an M. E. Sc inorganic and biochemical engineering in 1977,from the Technical University of Budapest,Hungary. Her advisors were Professors FerencTüdös and Tibor Kelen of Hungary, and Pro-fessor Joseph P. Kennedy at the University ofAkron, Ohio, USA, in the framework of col-laboration between the National ScienceFoundation of the USA and the HungarianAcademy of Sciences. She started her aca-demic career in 1996. Before that she was

involved in polymer research and development in the microelectronic,paint and rubber industries. Her present interests include green polymerchemistry, biomimetic processes and biomaterials, living/controlled

polymerizations, polymerization mechanisms and kinetics, thermoplasticelastomers and polymer structure/property relationships, and probingthe polymer-bio interface. She is one of the editors of the new Interdis-ciplinary Reviews in Nanomedicine and NanoBiotechnology WIRE, Pub-lished by Wiley-Blackwell, and a member of the Advisory Board, of theEuropean Polymer Journal. Until July of 2008 she was one of the tworegional Editors with the highest citation index. She was also member ofthe IUPAC Working Party IV.2.1 ‘‘Structure-property relationships ofcommercial polymers’’. Puskas has been published in more than 350

publications, including technical reports, is an inventor or co-inventor of26 U.S. patents and applications, and has been Chair or organizer of anumber of international conferences. Puskas has been awarded her fourthNSF and first NIH grant (the first ever in the Department of PolymerScience) in 2010. She is the recipient of several awards, including the1999 PEO (Professional Engineers of Ontario, Canada) Medal inResearch&Development, a 2000 Premier’s Research Excellence Award, the2004 Mercator Professorship Award from the DFG (Deutschen Fors-chungsgemeinschaft, German Research Foundation), the LANXESS (pre-viously Bayer) Industrial Chair 1998-2008, and the 2009 ‘‘Chemistry ofThermoplastic Elastomers’’ Award of the Rubber Division of the AmericanChemical Society. She was elected Fellow of the American Institute ofMedical and Biological Engineering AIMBE in 2010.

As a coinventor of the polymer used on the Taxus� coronary stent,Puskas helped the University of Akron generate more than $5 million inlicense fees. With her business partner they founded Biopolymers Inter-national Inc. to supply research quantities of functionalized polymersusing enzyme catalysis under license from the University of Akron.

Kwang Su Seo studied polymer chemistry atthe Chonbuk National University (SouthKorea), where he received his BS degree in2003 and MS degree in 2005 under thesupervision of Dr. Gilson Khang, Dr. Hai BangLee and Dr. Moon Suk Kim. He is currently aPh.D. student in the Puskas group in theDepartment of Polymer Science at The Uni-versity of Akron, where he works on the syn-thesis of PEG-based dendrimers using enzymecatalysis for targeted multivalent cancer drugdelivery.

Dr. Sen obtained his BS degree in Chemistry atBogazici University, Turkey in 2002 and MSdegree in Polymer Science at University ofAkron in 2005 under the supervision of Dr.Roderic P. Quirk. He then joined the Puskasgroup in December of 2005 and received hisPhD in 2009. He was the first student workingon the functionalization of polymers usingenzymatic catalysis, a new research directionin the Puskas group. His research interestsincluded the enzymatic functionalization ofbiomaterials and getting a better fundamental

understanding of enzyme catalysis. He currently works at Kordsa GlobalR&D Center in Turkey as a Project Leader.