Research Article

Biocatalytic route for the synthesis of oligoesters of hydroxy-fatty acids and ε-

caprolactone†
Anamaria Todea1

Diana Aparaschivei1

Valentin Badea1

Carmen G. Boeriu2

Francisc Peter1

1Faculty of Industrial Chemistry and Environmental Engineering, University Politehnica

Timişoara, Timişoara, Romania

2Wageningen Food & Biobased Research, Wageningen, The Netherlands.

Correspondence: [Francisc Peter, Faculty of Industrial Chemistry and Environmental

Engineering, University Politehnica Timişoara, Carol Telbisz 6, 300001 Timişoara,

Romania; Carmen G. Boeriu, Wageningen Food & Biobased Research, P.O. Box 17, 6700 AA

Wageningen, The Netherlands].

E-mail: francisc.peter@upt.ro; carmen.boeriu@wur.nl

Keywords: biocatalysis, biopolymers, lipase, hydroxy-fatty acids, ε-caprolactone

†
This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process, which may lead to
differences between this version and the Version of Record. Please cite this article as doi:
[10.1002/biot.201700629].

This article is protected by copyright. All rights reserved

Received: October 31, 2017 / Revised: February 17, 2018 / Accepted: March 5, 2018

1
Abstract

Developments of past years placed the bio-based polyesters as competitive substitutes for

fossil-based polymers. Moreover, enzymatic polymerization using lipase catalysts has

become an important green alternative to chemical polymerization for the synthesis of

polyesters with biomedical applications, as several drawbacks related to the presence of

traces of metal catalysts, toxicity and higher temperatures could be avoided.

Copolymerization of ε-caprolactone (CL) with four hydroxy-fatty acids (HFA) from

renewable sources, e.g. 10-hydroxystearic acid, 12-hydroxystearic acid, ricinoleic acid, and

16-hydroxyhexadecanoic acid, was carried out using commercially available immobilized

lipases from Candida antarctica B, Thermomyces lanuginosus, and Pseudomonas stutzeri, as

well as a native lipase. MALDI-TOF-MS and 2D-NMR analysis confirmed the formation of

linear/branched and cyclic oligomers with average molecular weight around 1200 and

polymerization degree up to 15. The appropriate selection of the biocatalyst and reaction

temperature allowed the tailoring of the non-cyclic/cyclic copolymer ratio and increase of

the total copolymer content in the reaction product above 80%. The catalytic efficiency of

the best performing biocatalyst (Lipozyme TL) was evaluated during 4 reaction cycles,

showing excellent operational stability. The thermal stability of the reaction products was

assessed based on TG and DSC analysis. This new synthetic route for biobased oligomers

with novel functionalities and properties could have promising biomedical applications.

Abbreviations: ECL, ε-caprolactone; PCL, poly(ε-caprolactone); HFA, hydroxy-fatty acid;

10HSA, 10-hydroxystearic acid; 12HSA, 12-hydroxystearic acid; 16HHDA, 16-

hydroxyhexadecanoic acid; RCA, ricinoleic acid; PDI, polydispersity index; Dpmax, maximum

polymerization degree; NC, non-cyclic copolymer; CC, cyclic copolymer; Mn(C), number

average molecular weight of the copolymer; Mw(C), weight average molecular weight of the

copolymer.

2
1 Introduction

The bio-based polymer synthesis has emerged as a topic of interest, due to the real demand

for bio-plastic products. Among this arsenal of valuable materials polyesters are the most

promising, based on their unique and various properties suitable for development of novel

manufacturing technologies [1].

Poly(ε-caprolactone) (PCL) was one of the major synthetic biocompatible polymers, used in

some application in the biomedical field until other materials with superior properties

became available. Particularly, poly(lactic) acid and its copolymers or functionalized

derivatives demonstrated excellent properties for biomedical applications [2,3]. However,

the development of tissue engineering gained revived interest for PCL in the last period, due

to its good biocompatibility and superior rheologic and viscoelastic properties compared to

other aliphatic polyesters [4]. Moreover, recent findings seem to ensure a privileged

position for caprolactone-based polymers in the forthcoming era of biobased products and

technologies. PCL and its monomer ε-caprolactone (ECL) own physical and chemical

properties allowing the design of a wide range of tunable biomaterials and innovative

polymer architectures [5]. Recently, Schmidt et al. reported an enzyme cascade synthesis of

ε-caprolactone and its oligomers from cyclohexanone [6]. As cyclohexanone can be easily

synthesized from phenol, resulted by the pyrolysis of lignin, this finding can open the route

for biobased PCL. Although PCL is a valuable biomaterial with important applications for

drug-delivery systems [7] and bioresorbable scaffolds [8], several drawbacks such as low

hydrophilicity, low melting point (~60°C), and relatively slow biodegradability impeded its

extensive development [4]. To overcome these drawbacks, basically two solutions are

available: blending with other biomaterials, like hydrophilic polymers [9] or synthesis of

copolymers with improved properties [10]. In the same time, melting point in the

physiological range can be advantageous for specific biomedical applications, while slow

degradation rate results in longer stability of the implants [5]. Therefore, the optimal

3
properties of these polymers depend mainly on the specific requirements of the targeted

application.

Vegetable oils are a high-class renewable source for synthesis of various polymers,

including polyesters [11,12]. Among them, the materials resulted by polycondensation of

hydroxy-fatty acids proved to be sustainable replacements for fossil-based polymers and

were suitable for specialty and commodity applications [13,14]. Notwithstanding of these

benefits, improvement of catalysts and process conditions is still needed, as suggested by

several reports [15,16].

The use of enzymes as catalysts for polyester synthesis is an attracting alternative to

chemical polymerization. Lipases have been framed as powerful tool for polyester

synthesis, including the ring-opening polymerization of lactones, and several

comprehensive reviews were addressed in this topic [17,18,19,20]. Polyester synthesis

catalyzed by lipases also opens the route for the manufacture of fully biobased polymers

obtained from renewable resources [21], but also for specialty products like aromatic-

aliphatic oligoesters [22] and functional aliphatic oligoesters [23]. The lipase from Candida

antarctica B demonstrated special ability to catalyze the synthesis of linear and star

oligomers. Some drawbacks of the traditional chemical methods related to traces of metal

catalysts, toxicity and high reaction temperatures can be avoided using the biocatalytic

route [24]. Our group reported the lipase-catalyzed synthesis of several estolides, oligomers

of hydroxy-fatty acids derived from vegetable oils, in organic reaction media [25]. Apart of

lipase, cutinase, another lipolytic enzyme, was also found able to catalyze the

homopolymerization of ω-hydroxy acids, with different chain specificity than Novozym 435

[26]. Recent advances in this field confirmed that cutinases can catalyze the synthesis of

various polyesters by polycondensation of bio-based monomers, but their potential still

need to be fully exploited [27].

4
The copolymers of fatty acids and hydroxy-fatty acids have promising applications as drug

delivery systems, biodegradable lubricants, functional fluids, cosmetics, inks or coatings

[28]. Despite the well-known advantages of biocatalysis for the manufacture of innovative

polymeric materials, reports about the enzymatic synthesis of hydroxy-fatty acids

copolyesters are scarce. Random copolyesters were obtained from vegetable oil derivatives,

a vinylsulfide-containing hydroxy acid and a ketone-containing hydroxy ester, in

diphenylether, at temperatures up to 90C, using Novozym 435 lipase [29]. The enzyme-

catalyzed copolymerization of ECL with hydroxy-fatty acids was not yet reported, but it may

be an interesting approach and a promising solution to improve the properties in relation

to both homopolymer products, PCL and estolides. The chemical synthesis of these

copolymers was recently accomplished by our group [30], but the main interest is to

develop an entirely biobased route, using biocatalysis. The biocatalytic copolymerization

may result in copolymers with lower molecular mass compared to the chemical catalysis,

but the well-known selectivity of the enzymes could allow better control of the synthesized

compounds structure. In the present work, four structurally different hydroxy-fatty acids

(HFAs), with 16 or 18 carbon atoms, namely 10-hydroxystearic acid (10HSA), 12-

hydroxystearic acid (12HSA), 12-hydroxy-9-cis-octadecenoic acid (ricinoleic acid, RCA),

and 16-hydroxyhexadecanoic acid (16HHDA), were investigated as co-substrates for the

synthesis of copolymers with ECL.

2 Materials and methods

2.1 Materials

ε-Caprolactone (ECL), 12-hydroxyoctadecanoic acid (99%), 16-hydroxyhexadecanoic acid

(98%), 12-hydroxy-cis-9-octadecenoic acid (98%), bis(trimethylsilyl)trifluoroacetamide-

trimethylchlorosilane (BSTFA:TMCS, 99:1), trans-2-[3-(4-t-butyl-phenyl)-2-methyl-2-

propenylidene] malononitrile (DCTB), potassium trifluoroacetate (KTFA), as well as the

lipase from Pseudomonas fluorescens (Amano AK), were purchased from Sigma Aldrich.

5
Novozym® 435 (lipase from Candida antarctica B immobilized on acrylic resin) and

Lipozyme® TL-IM (lipase from Thermomyces lanuginosus immobilized on a non-

compressible silica gel carrier), were products of Novozymes (Denmark). Two CLEA

Pseudomonas stutzeri lipases, with general applicability (CLEA P. stutzeri) and designated

for organic media (CLEA P. st. OM), were obtained from CLEA Technologies (Delft, The

Netherlands). The organic solvents (toluene, tetrahydrofuran) were purchased from Merck

and their water content (determined by Karl Fischer titration) was below 0.01%.

2.2 10-Hydroxystearic acid synthesis

Since it is not commercially available, 10-hydroxystearic acid was synthesized in our

laboratory by using cell-free extract of E. coli TOP10 cells containing the plasmid pBAD-

HISA-OH. The recombinant cells containing the oleate hydratase and expression were

performed as previously reported [31]. The oleic acid conversion reaction was carried out

24 h at 30C and 200 rpm in 50mM TRIS-HCl buffer, pH 8 by using 0.5 ml cell free extract

(protein concentration 48 mg ml-1) and 0.6% (v/v) oleic acid.

Samples were taken at designated time intervals and a derivatization assay was performed

prior to the HPLC analysis, as described elsewhere [31]. Separation was performed by using

a C18-RP column (4.6x50 Merck ChromolithSpeedROD), with binary gradient elution (A:

H2O with 0.1% v/v trifluoroacetic acid, B: ACN) at 50°C, 1ml min-1 flow rate and UV detection

at 242 nm. The gradient was started with 50% B for 3 min, 50-80% B 2 min, and isocratic

elution (80% B) over 7 min. When a full conversion was indicated by the HPLC results, the

pH was adjusted to 2.0 with HCl, the mixture was filtered through Whatman filter paper and

the white solid precipitate was dissolved in acetone afterwards. The insoluble residues

were removed by filtration, and the acetone was evaporated by using rotary evaporation.

The product was dried overnight at 40C and stored at -20C until use. Structural

characterization of the synthesized 10HSA was accomplished by GC-MS, FT-IR and NMR

(see supporting information). The m/z peaks at 215, 229 and 331 in the mass spectrum

6
resulted from the GC-MS analysis (Fig. 1S, supporting information) are in accordance with

the values reported before for the trimethylsilyl derivative of 10HSA [32].

Fourier Transform Infrared (FT-IR) spectra were obtained in attenuated total reflectance

(ATR) mode on a Bruker Vertex 70 (Bruker Daltonik GmbH, Germany) spectrometer

equipped with a Platinium ATR, Bruker Diamond Type A225/Q. Spectra were collected in

the range 4000-400 cm-1 with a resolution of 4 cm-1 and with 64 co-added scans.

FTIR (cm−1): 3399 (ν(O-H)), 2913 (ν(C-H)), 1696 (ν (C-O) ester);

10HSA NMR: 1H (400 MHz, pyridine-D5) δ(ppm): 3.89-3.83 (m, 1H, CHOH), 2.51 (t, J=7.3Hz,

2H, CH2CO), 1.68-1.51 (m, 28H), 0.85 (t, J=7.2Hz, 3H, CH3);

13C (100.62 MHz, pyridine-D5) δ(ppm): 176.4(COOH), 71.3(CH-OH), 38.9 (11-CH2,9-CH2),

35.3 (CH2-COOH), 32.5 (16-CH2), 30.6(13-CH2), 30.5(6-CH2), 30.4 (14-CH2), 30.3(5-

CH2,15-CH2), 29.9 (4-CH2), 26.8 (8-CH2,12-CH2), 26.0 (3-CH2), 23.3 (17-CH2), 14.7 (CH3)

2.2 Enzymatic synthesis of the oligoesters

The syntheses were performed by adding the lipase, about 50 U/mmol substrate, to a molar

ratio of 1:1 ECL:HFA (30-100 mM), brought to a final volume of 2 ml, using toluene as

solvent. An Eppendorf Thermomixer Comfort (Eppendorf, Germany) heating shaker has

been used, with circular mixing movement to ensure an efficient mass transfer, without

grinding the support particles. The reactions have been carried out at temperatures

between 45°C and 85°C, under continuous stirring at 350 rpm for 24 h. At the end of the

reaction the enzyme was removed by centrifugation (13,000 rpm, 3 min)/filtration, and the

reaction mixture was analyzed by GC-MS and MALDI-TOF MS.

2.3 Characterization of the reaction products

MALDI-TOF MS analysis was carried out for composition analysis of polymers, using an

UltrafleXextreme Bruker spectrometer with FlexControl and FlexAnalysis software

packages for acquisition and processing of the data (BrukerDaltonics, Germany), at an

acceleration voltage of 25 kV, using DCTB as matrix and KTFA as ionization agent. The

7
  

sample preparation and analysis were performed as previously described [22]. The number

average molecular weight Mn(C), weight average molecular weight Mw(C) of the copolymers

(non-cyclic and cyclic), as well as the polydispersity index PDI have been calculated as

described elsewhere [33, 34].

GC-MS analysis was performed using a Thermo Analytic GC-MS system composed from

Trace 1310 gas chromatograph, MS module and TriPlus RSH autosampler. Thermo

Scientific TG-5ms 30 m x 0.25 mm x 0.25 µm capillary column was employed in the following

conditions: oven temperature 100–300⁰C with 10⁰C min-1 heating rate, injector

temperature 300⁰C, carrier gas (helium) flow 1.0 mL min-1. The HFA conversion/formation

was determined after derivatization with BSTFA+TMCS (99:1), at 2:1 reagent: sample ratio

(w/w), for 1h at 95C, as previously described [31]. Hexadecane was used as internal

standard.

NMR spectra of the isolated product were recorded on a Bruker AVANCE III spectrometer

operating at 500.0 MHz (1H) and 125.0 MHz (13C). The samples were dissolved in

tetrahydrofurane-d8 and the chemical shifts δ are given in ppm, relative to TMS.

The thermogravimetric analysis of the oligoesters were performed by using TG 209 F1 Libra

thermogravimetric analyzer (Netzsch, Germany) under nitrogen atmosphere, heating rate

10C min-1 in the temperature range 20-590C.

The differential scanning calorimetry characterization was performed by an DSC 204 F1

Phoenix differential scanning calorimeter (Netzsch, Germany) under nitrogen atmosphere,

heating rate 10Cmin-1, in the temperature range 20-600C.

3 Results and discussion

3.1 Biocatalytic synthesis of ECL_HFA copolymers

To the best of our knowledge, this is the first report about the biocatalytic copolymerization

of ECL and HFAs. Three of the selected HFAs, specifically 16HHDA, 12HSA and RCA, have

been used in our previous study [25] as substrates for the synthesis of estolides

8
(homopolymers of HFA) catalyzed by lipases, but not for copolymers. Although the

enzymatic synthesis of 10-hydroxystearic acid (10HSA), starting from oleic acid and oleate

hydratase from different sources was reported by several groups [35, 36], 10HSA based

polyester synthesis and characterization was not yet accomplished. Since 10HSA is not

commercially available, it was synthesized in our laboratory by biocatalytic hydration of

oleic acid, as presented in the Methods section, followed by characterization by means of

FT-IR, GC-MS, 1H-NMR, and 13C-NMR (see supporting information). The purity of the

obtained 10HSA was higher than 95% and it was used for the synthesis of copolymers with

ECL without further purification.

The copolymerization reactions were carried out adding the hydroxy-fatty acid (10HSA,

16HHDA, 12HSA, or RCA) to ECL dissolved in toluene and were started by the addition of a

lipase, at different temperatures in the range 40-80°C, for 24 h. At the end of the reaction,

the enzyme was removed by centrifugation, the product was dried at 25°C and analyzed by

FT-IR, MALDI-TOF MS, and NMR techniques, to identify the formed compounds. The

reaction time of 24h was selected based on preliminary experiments (data not presented),

showing that in this time interval an almost complete conversion of ECL (the more reactive

monomer) and high conversion of HFA were achieved. Toluene has been used as reaction

medium in all experiments, due to the acceptable solubility of all substrates in this solvent.

The insertion of hydroxy-fatty acids unit into the PCL backbone was demonstrated by

MALDI-TOF MS analysis. The MALDI-TOF MS spectra allow distinguishing between cyclic

and non-cyclic reaction products, chain length and dispersity, based on the similarity

between the calculated and identified molecular masses of the different oligomeric products

present in the reaction mixture. The identified copolymers were linear (in the case of

16HDDA), branched (in the case of 10HSA, 12HSA and RCA), and cyclic oligoesters with

different number of HFA units in the backbone. Figure 1a and 1b present two typical MALDI-

TOF MS spectra of the products formed in the reactions catalyzed by Lipozyme TL

9
(copolymers of 12HSA_ECL and RCA_ECL, respectively), displaying the peaks corresponding

to the appropriate oligoesters. As example, the m/z values of 1050.1, 1278.5 and 1392.8

represent the K+ adducts of the (RCA)n-(ECL)m copolymer products with inserted RCA units,

where n=3 and m=3-5. The m/z values 1116.1, 1290.4 and 1458.9 demonstrate the

formation of the (12HSA)n-(ECL)m copolymers of ECL with 12HSA (also as K+ adducts),

where n=1-2 and m=6-8. In accordance with our previous results concerning the chemical

synthesis of ECL copolymers with HFA [30], the formation of both homopolymers, PCL and

estolides, was also detected (m/z values of 1121, 1235, 941, 1216 in Figure 1). The reaction

products identified by MALDI-TOF MS are depicted in Figure 2. Even if the synthesis of PCL

and estolides as secondary reaction products could not be avoided, their formation in the

lowest possible amount was one of the objectives of this work, capitalizing the expected

higher selectivity of enzymes compared to chemical catalysts.

3.2 Screening of lipases to achieve high copolymer yield

The biocatalyst with the best catalytic efficiency for the synthesis of the targeted

copolymers was selected by screening four commercially available immobilized lipases:

Novozym 435 (from Candida antarctica B), Lipozyme TL (from Thermomyces lanuginosus),

as well as two CLEA Pseudomonas stutzeri lipases (one specific for organic media and one

with general applicability). Additionally, the native lipase from Pseudomonas fluorescens

was also tested, based on our previous study which proved it as the best-performing

biocatalyst for estolides synthesis [25]. The conversions were monitored by GC-MS analysis,

using hexadecane as internal standard. After 24h reaction time, the ε-caprolactone (ECL)

conversions were higher than 98%, regardless of the HFA co-substrate used, due to the

higher reactivity of the 7-member cycle lactone monomer compared to the hydroxy acids.

Consequently, the obtained copolymers contained more inserted ECL units, forming the

backbone of the products. Equimolar ratio of monomers was used in all experiments, to

10
keep the PCL copolymer amount in the reaction mixture as low as possible and to obtain

copolymers with more inserted HFA units.

The results presented in Table 1 show that higher HFA co-monomer conversions and higher

average molecular weights were achieved when 16HHDA was used as co-substrate. Among

the tested lipases, Novozym 435 and Lipozyme TL were the most efficient for all selected

substrates, while the CLEA biocatalysts led to lower molecular weight products. Although

displaying good catalytic activity in the polyesterification reaction, the native lipase from

Pseudomonas fluorescens was less efficient than the best immobilized lipases. The

composition of the reaction mixture was estimated based on the MALDI-TOF data, as

previously described [36]. Compared to Novozym 435, Lipozyme TL demonstrated better

selectivity for the formation of non-cyclic (linear or branched) copolymers, excepting those

with RCA. The highest relative amount of non-cyclic copolymer was 83%, in the

copolymerization product with 10HSA, while the highest polymerization degree was

between 11 and 15, depending on the nature of HFA used. Despite this variance, it is obvious

that structural differences between the HFA have no essential influence on the

copolymerization reaction catalyzed by lipases, therefore this experimental protocol is valid

for a variety of HFAs and the optimal conditions can be tailored for every substrate.

Compared to the chemical synthesis of 16HHDA and RCA copolymers, when tin(II) 2-

ethylhexanoate was used as catalyst at similar ECL: HFA molar ratio and higher

temperatures [30], the biocatalytic route led to higher average molecular weights (Mw

around 1300, using Novozym 435 as catalyst) and comparable content of oligoesters with

HFA units (between 73% and 83%, depending on the enzyme and HFA investigated). Only

for the oligomers of 12HSA, the average molecular weights of the products of the

biocatalytic route were lower compared to the chemical catalysis. It is noteworthy the high

selectivity of Lipozyme TL towards the branched copolymer formation in the case of

saturated hydroxy acids with secondary hydroxy groups (12HSA and 10HSA), relative to

11
  

the cyclic copolymer. These results prove the ability of lipases to catalyze these reactions

with high conversions and substrate-dependent specificities, in mild reaction conditions.

Particularly, the synthesis of polymers involving 10HSA was not reported before, neither by

chemical nor by biocatalytic route.

3.3 Influence of temperature on the molecular weight and product composition

The effect of increasing reaction temperatures on the conversion and polymerization

degree was studied considering the copolymerization of 16HHDA and ECL, at 1:1 molar

ratio and 30 mM substrate concentration, in the 40-80C temperature range. The selected

biocatalyst was Lipozyme TL, due to the higher relative content of non-cyclic copolymer in

the product, compared to Novozym 435, as shown in Table 1. The results, depicted in Figure

3 indicate an increase of the average molecular weight up to 35% when the temperature

was raised in the range 40C - 80C. Behind this overall increase of the molecular weight,

the selectivity of the lipase for the non-cyclic related to the cyclic form of the copolymer

decreased at higher temperature (Table 1S, supporting information), while the relative

content of copolymer reached more than 80% at 80C. The increase of the polymerization

degree at higher temperature is normal, although the reaction is also influenced by the

thermostability of the enzyme. Noteworthy, the influence of the temperature on the non-

cyclic/cyclic polymer ratio is also of special interest, since an important goal of the

enzymatic copolymerization is to achieve better control of the polymerization degree and

product composition than obtained by chemical catalysis. Considering this influence and

the selectivity of the lipase, a tailored non-cyclic/cyclic copolymer ratio can be designed,

according to the requirements of the manufactured product.

For a possible further increase of the average molecular weight of the products, a second

temperature study has been carried out in two steps. In this regard, the reactions were

performed first at 80C for 24h, then the temperature has been increased at 120C for

additional 24h. The aim was to check whether the increase of the temperature beyond the

12
  

inactivation limit of the enzyme could favor the chain elongation of the copolymer. The

results showed an increase of not more than 15% of the average molecular weight and

about 10% of the relative content of the total copolymer in the reaction product (Table 2S,

supporting information). It can be concluded that the thermal inactivation of the Lipozyme

TL enzyme occurred before significant additional chain elongation could happen.

3.4 Structural characterization of the reaction products

MALDI-TOF MS and NMR spectroscopy were used for the structural characterization of the

reaction products. It should be pointed out that the products were not fractionated;

therefore, signals assigned to the homopolymers were present in the spectra, as well.

However, the insertion of hydroxy-fatty acid moieties into the hydrophobic PCL backbone

could be proved by 2D NMR spectra 1H-13C HMBC. In a typical example (Figure 6S-8S,

supporting information), showing the 16HHDA_ECL reaction product, the distance coupling

of the carbon atom from the carbonyl moiety from 173.7 ppm (C-1) and the protons from

4.18 ppm (H2C-33, 3J), 2.4 ppm (H2C-2, 2J) and with the protons from 1.69 ppm (H2C-3,

3J).

The 13C NMR spectra revealed the presence of several signals in the region specific for

carbonyl groups which indicate the presence of the several products species, as the MALDI-

TOF spectra also revealed.

3.5 Thermal properties of the oligoesters

The thermal stability of the co-polyesters was evaluated by TG analysis (Figure 4),

compared with the homopolymers (PCL and the appropriate estolides). Unlike PCL, the

profiles of the reaction product and the correspondent estolide showed multistep

degradation processes. The first degradation step (up to 300C) corresponds to the loss of

small oligomers and the results confirm that the 16HDDA_ECL samples contain higher

molecular weight polyesters. The overlaid TG curves (Figure 2 and Figures 9S-14S in

supporting information) clearly indicate that the HSA-ECL copolyesters start to degrade at

13
  

lower temperatures compared to PCL. The same behavior was observed when poly(ε-

caprolactone-co-ε-thiocaprolactone) copolymers were synthesized [37]. The thermal

stability of the copolymers was higher compared to the estolides, as 35% mass was lost up

to 300C, compared to 45% registered for the 10HSA and 12HSA estolides. Compared to the

PCL homopolymer, the thermal stability of the copolymers was lower, excepting the

16HHDA_PCL reaction product.

Contrary to the parent homopolymer (PCL), the TGA curve of the copolymer clearly shows

two degradation steps, which indicates the probability of a block structure.

The melting temperature (Tm), the melting enthalpy (Hm) and degree of crystallinity were

determined by DSC measurements in the 20-590C range, at 10Kmin-1 heating rate. The

degree of crystallinity was calculated through the relation between Hm of each sample and

the theoretical value of a 100% crystalline PCL sample [38]. As a general tendency, the

copolyesters showed higher melting temperature values compared to PCL (58.9C), but

lower than the corresponding estolides. Among the four new synthesized materials, the

highest Tm values were registered for the 16HHDA_ECL product (Table 3S, supporting

information). However, important differences of the degree of crystallinity were not

observed when 10HSA and 12HSA were used as co-monomer, compared to the value

calculated for 16HHDA based products. These results are in concordance with the

crystallinity values of linear and branched polymers, reported elsewhere [38].

3.6 Biocatalyst stability in multiple reaction cycles

An important advantage of using immobilized enzymes is the possibility to operate them in

successive reaction cycles without important loss of activity and consequently reducing the

production costs. The operational stability of Lipozyme TL was investigated in four

consecutive reaction cycles for the synthesis of the 16HHDA_ECL copolymer, at 80C. This

study was carried out at higher temperature to explore the behavior of the biocatalyst at

prolonged utilization in more adverse conditions (according to the producer, the optimum

14
temperature of this enzyme is in the range of 50-75C). The reaction products were

analyzed by MALDI-TOF MS spectrometry. The results, presented in Table 2, indicate a slow

increase of the calculated average molecular weights during the biocatalyst reuse. The

increase of the polymerization degree during the reuse of the biocatalyst is positive and

demonstrates the high operational stability of the biocatalyst, although it was not significant

(less than 20%).

The relative copolymer content was higher than 80% throughout all reaction cycles,

although an increase of the relative content of cyclic copolymers to the detriment of the

linear copolymers was observed after each reuse. This observation is in accordance with

the increase of the relative amount of cyclic copolymer at higher temperature, as discussed

earlier. Such an increase of the cyclic/non-cyclic copolymer ratio during the repeated use of

a biocatalyst at the same temperature was not yet reported and currently there is no

reasonable explanation of this behavior.

4. Conclusion

Linear, branched, and cyclic oligoesters were successfully obtained for the first time from

different biobased HFAs and ECL, using a biocatalytic route. The synthesis of copolymers

of 10HSA is of special interest, considering that this renewable monomer was obtained from

oleic acid by a biocatalytic process, as well. Among the tested biocatalysts, the commercially

available lipases from Candida antarctica B (Novozym 435) and Thermomyces lanuginosus

(Lipozyme TL) were the most efficient, leading to higher average molecular weight and

copolymer content of the synthesized products. Thermal analysis revealed superior thermal

stability of the reaction products compared to the correspondent estolides. The structural

analysis of the copolymers by MALDI-TOF MS and NMR demonstrated that several HFA

units were inserted into the polymer backbone. The sequence of ECL and HFA units in the

polymer chains looks random and any control of the insertion mechanism was not yet

possible. However, we proved that the biocatalytic process allows the synthesis of these

15
oligoesters with higher selectivity compared to the chemical route. Particularly, a new

oligoester of ε-caprolactone and 10-hydroxystearic acid was obtained using Lipozyme TL at

> 88% HFA conversion and > 80% branched copolymer content in the reaction product.

Moreover, considering the very plausible manufacture of ECL from biobased resources in

the near future, the present process can open the way for an entirely biobased route for ECL

copolymers with new functionalities, providing very attractive applications in the field of

drug-delivery systems and bioresorbable scaffolds. Functionalization of the obtained

oligomers by synthesis of grafted copolymers or enzymatic epoxidation of the olefinic

double bond of the RCA_ECL copolymer, followed by the insertion of a specific functional

group, could be another important future research direction. Our studies will be focused on

applications, functionalization and degradability studies of these compounds.

Acknowledgement

This work was supported by a grant of the Romanian Authority for Scientific Research and

Innovation, CNCS/CCCDI - UEFISCDI, project number PN-III-P2-2.1-PED-2016-0168, within

PNCDI III. The authors are grateful to Prof. Isabel Arends and Dr. Linda Otten (University of

Delft) for supplying the oleate hydratase from Elizabethkingia meningoseptica, used for the

synthesis of 10-hydroxystearic acid.

Conflict of interest

The authors declare no financial or commercial conflict of interest.

5 References

[1] C. Vilela, A. F. Sousa, A. C. Fonseca, A. Serra, J. F. J. Coelho, C. S. R. Freire, A. J. D. Silvestre,

Polym. Chem. 2014, 5, 3119.

[2] A. Pellis, L. Silvestrini, D. Scaini, J. M. Coburn, L. Gardossi, D. L. Kaplan, E. H. Acero, G. M.

Guebitz. Process Biochem. 2017, 59, 77.

[3] B. Tyler, D. Gullotti, A. Mangraviti, T. Utsuki, H. Brem, Adv. Drug Deliv. Rev. 2016, 107, 163.

[4] M.A. Woodruff, D.W. Hutmacher, Prog. Polym. Sci. 2010, 35, 1217.

16
  

[5] A.L. Sisson, D. Ekinci, A. Lendlein, Polym. 2013, 54, 4333.

[6] S. Schmidt, C. Scherkus, J. Muschiol, U. Menyes, T. Winkler, W. Hummel, H. Gröger, A.

Liese, H.G. Herz, U.T. Bornscheuer, Angew. Chem. Int. Ed. 2015, 54, 2784.

[7] E. Schlesinger, N. Ciaccio, T.A. Desai, Mat. Sci. Eng. C 2015, 57, 232.

[8] G. Rasperini, S.P. Pilipchuk, C.L. Flanagan, C.H. Park, G. Pagni, S.J. Hollister, W.V.

Giannobile, J. Dent. Res. 2015, 94(9, Suppl. 2), 153S.

[9] A. Gholipour-Kanani, S.H. Bahrami, M.T. Joghataie, A. Samadikuchaksaraei, H. Ahmadi-

Taftie, S. Rabbani, A. Kororian, E. Erfani, IET Nanobiotechnol. 2014, 8, 123.

[10] J.J. Wurth, N.R. Blumenthal, V.P. Shastri, PLOS One 2014, 9, e99157, doi:

10.1371/journal.pone.0099157.

[11] C. Zhang, T.F. Garrison, S.A. Madbouly, M.R. Kessler, Prog. Polym. Sci. 2017, 71, 91.

[12] A. Gandini, T.M. Lacerda, A.J.F. Carvalho, E. Trovatti, Chem. Rev. 2016, 116, 1637.

[13] J. Jose, G. Pourfallah, A.L. Leao, S.S Narine, Polym. Int. 2014, 63, 1902.

[14] J. Jose, G. Pourfallah, D. Merkley, S. Li, L. Bouzidi, A. Lopes Leao, S. S. Narine, Polym.

Chem. 2014, 5, 3203.

[15] A. Mahapatro, A. Kumar, R. Gross, Biomacromol. 2004, 5, 62.

[16] J. Zhang, H. Shi, D. Wu, Z. Xing, A. Zhang, Y. Yang, Q. Li, Process Biochem. 2014, 49, 797.

[17] S. Kobayashi, Polym. Adv. Technol. 2015, 26, 677.

[18] S. Shoda, H. Uyama, J. Kadokawa, S. Kimura, S. Kobayashi, Chem. Rev. 2016, 116, 2307.

[19] S. Sen, J.E. Puskas, Molecules 2015, 20, 9358.

[20] M. Labet, W. Thielemans, Chem. Soc. Rev. 2009, 38, 3505.

[21] Y. Jiang, A.J.J. Woortman, G.O.R. Alberda van Ekenstein, K. Loos, Polym. Chem. 2015, 6,

5451.

[22] A. Pellis, A. Guarneri, M. Brandauer, E.H. Acero, H. Peerlings, L. Gardossi, G. M. Guebitz,

Biotechnol. J. 2016, 11, 642.

[23] S. Kobayashi, Proc. Jpn. Acad., Ser. B 2010, 86, 338.

17
  

[24] A. Kundys, E. Białecka-Florjańczyk, A. Fabiszewska, J. Małajowicz, J. Polym. Environ.

2017, DOI: 10.1007/s10924-017-0945-1.

[25] A. Todea, L.G. Otten, A.E. Frissen, I.W.C.E. Arends, F. Peter, C.G. Boeriu, Pure Appl. Chem.

2015, 87, 51.

[26] D. Feder, R.A. Gross, Biomacromolecules 2010, 11, 690.

[27] V. Ferrario, A. Pellis, M. Cespugli, G.M. Guebitz, L. Gardossi, Catalysts 2016, 6, 205;

doi:10.3390/catal6120205.

[28] L. Montero De Espinosa, M. R., Meier, European Polym. J. 2011, 47, 5, 837.

[29] Z. Beyazkilic, G. Lligadas, J.C. Ronda, M. Galia, V. Cadiz, Polym. 2015, 79, 290.

[30] D. Aparaschivei, A. Todea, I. Păușescu, V. Badea, M. Medeleanu, E. Sisu, M. Puiu, A.

Chirita-Emandi, F. Peter, Pure Appl. Chem. 2016, 88, 1191.

[31] A. Todea, A. Hiseni, L.G. Otten, I.W.C.E. Arends, F. Peter, C.G. Boeriu, J. Mol. Cat. B:

Enzymatic 2015, 119, 40.

[32] J. A. Hudson, C. A. M. Mackenzie, K. N. Joblin, Appl. Microbiol. Biotechnol. 1995, 44, 1.

[33] A. Todea, V. Badea, L. Nagy, S. Kéki, C. G. Boeriu, F. Péter, Acta. Biochim. Pol. 2014, 61,

205.

[34] S. Kakasi-Zsurka, A. Todea, A, But, C. Paul, C. G. Boeriu, C. Davidescu, L. Nagy, Á. Kuki, S.

Kéki, F. Péter, J. Mol. Cat. B: Enzymatic 2011, 71, 22.

[35] L. E. Bevers, M. W. H. Pinkse, D. E. M. Peter, W. R. Hagen, P. D. E. M. Verhaert, J.

Bacteriolog. 2009, 191, 10.

[36] A. Hiseni, I. W. C. E. Arends, L. G. Otten, ChemCatChem 2015, 7, 29.

[37] S.W. Duchiron, E. Pollet, S. Givry, L. Avérous, Eur. Polym. J. 2017, 87, 147.

[38] S. I. Erdagi, E. Doganci, C. Uyanik, F. Yilmaz, React. Funct. Polym. 2016, 99, 49.

18
  

Table 1.

The ability of various lipases to catalyze the synthesis of oligoesters of ECL and hydroxy-

fatty acids, in toluene, at 80C

HFA Lipase HFA Mn[C] Mw[C] PDI DPmax NCa) CCa)

co- conversion, [%] [%]
substrate [%]
10HSA P. fluorescens 73.3 727 793 1.09 8 51.8 17.0
Novozym 435 88.3 1215 1284 1.05 9 46.7 19.6
Lipozyme TL 88.6 1052 1146 1.09 11 83.1 0.4
CLEA P. stutzeri 67.2 663 702 1.06 6 59.2 15.7
CLEA P. st OM 75.9 741 826 1.11 10 56.6 18.9
12HSA P. fluorescens 75.8 855 887 1.03 9 52.6 20.4
Novozym 435 87.2 1169 1399 1.19 14 50.7 17.5
Lipozyme TL 89.7 913 1002 1.09 13 60.7 0.2
CLEA P. stutzeri 86.4 806 908 1.12 11 54.6 19.1
CLEA P. st OM 83.3 693 751 1.08 11 57.8 15.3
RCA P. fluorescens 76.9 1304 1426 1.09 11 18.8 16.8
Novozym 435 91.4 1145 1291 1.12 10 59.9 16.3
Lipozyme TL 94.3 917 1009 1.10 9 55.1 21.1
CLEA P. stutzeri 74.6 997 1086 1.09 8 32.8 21.8
CLEA P. st OM 83.2 694 730 1.05 7 37.3 18.6
16HHDA P. fluorescens 98.3 1177 1294 1.09 12 57.5 18.2
Novozym 435 99.1 1321 1479 1.12 15 54.7 21.9
Lipozyme TL 99.1 1101 1232 1.12 15 62.4 18.5
CLEA P. stutzeri 98.7 724 825 1.14 9 78.8 1.9
CLEA P. st OM 99.3 677 764 1.13 8 76.5 4.5
a) the estimated copolymer content was calculated from MALDI-TOF MS data

19
  

Table 2.

The efficiency of the Lipozyme TL lipase in multiple polyesterification reaction cycles of ECL

and 16HHDA, in toluene, at 80C. The length of each reaction cycle was 24 h.

Cycle Mn Mw PDI NC CC CPa) LHb) CHc) DPmax

No. [%] [%] [%] [%] [%]
1 1101 1232 1.12 62.4 18.5 80.9 18.9 0.3 15
2 1030 1219 1.18 67.4 22.1 89.5 4.8 5.5 13
3 1151 1342 1.16 61.3 29.1 84.2 5.1 4.3 14
4 1257 1462 1.16 58.5 33.8 92.3 4.6 3.0 15
a) total copolymer; b) linear homopolymer; c) cyclic homopolymer

20
  

Figure legends

Figure 1. MALDI-TOF MS spectra of the copolymerization products of ε-caprolactone with

(A) ricinoleic acid (RCA_ECL) and (B) 12-hydroxystearic acid (12HSA_ECL)

Figure 2. Reaction scheme for the synthesis of copolyesters from hydroxy-fatty acids and

ε-caprolactone, emphasizing the products obtained with 10-HSA. Dotted curves signify the

formation of cyclic copolymers and homopolymers, respectively

Figure 3. Effect of temperature on the copolymerization of ECL and 16HHDA, in toluene.

Yellow triangle symbols represent the values of the total copolymer content in the reaction

mixture

Figure 4. The TG (A) and DSC (B) curves of the 16HHDA_ECL copolyester (brown),

compared to the 16HHDA (purple) and ECL (orange) homopolymers

21
  

Figure 1a

22
  

Figure 1b

Figure 2

23
  

Figure 3

Figure 4a

24
Figure 4b

25