B

Cp

VME

a

ARRAA

KP3A

1

firb(nbrctpm

DB

h0

ARTICLE IN PRESSG ModelIOMAC-4280; No. of Pages 6

International Journal of Biological Macromolecules xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

International Journal of Biological Macromolecules

j ourna l h o mepa ge: www.elsev ier .com/ locate / i jb iomac

ell system engineering to produce extracellularolyhydroxyalkanoate depolymerase with targeted applications

irginia Martínez1, Nina Dinjaski1, Laura I. de Eugenio, Fernando de la Pena,aría Auxiliadora Prieto ∗

nvironmental Biology Department, Centro de Investigaciones Biológicas, CSIC, Ramiro de Maeztu 9, 28040 Madrid, Spain

r t i c l e i n f o

rticle history:eceived 20 January 2014eceived in revised form 19 March 2014ccepted 5 April 2014vailable online xxx

eywords:olyhydroxyalkanoate depolymerase-Hydroxyalkanoic acidsntimicrobial activity

a b s t r a c t

Novel platforms based on the application of bacterial cell systems as factories for production of newbioproducts open avenues and dramatically expand the catalogue of existing biomaterials. Herein,we designed the strategy based on in vivo production of extracellular Pseudomonas fluorescens GK13(PhaZGK13) depolymerase to degrade previously biosynthesized polyhydroxyalkanotes (PHAs) or to obtain3-hydroxyalkanoic acids (HAs). With this aim, extracellular PhaZGK13 was produced in recombinantstrains and the optimal conditions for controlled release of HAs and oligomers by growing cells were setup with a particle suspension of 14C-labelled PHA, being maximal after 24 h of incubation. Genetic modi-fication of key factors involved in fatty acids metabolism revealed the influence of an active �-oxidationpathway on the extracellular degradation of PHA and subsequent HAs isolation. The highest HAs produc-tion was obtained using Pseudomonas putida KT2442 fadB mutant (0.27 mg/mL) due to the reduced ability

of this strain to metabolize the degradation products. The system was applied to produce new added valueHAs harboring thioester groups in the side chain from the functionalized mcl-PHA, PHACOS. Remarkably,hydrolyzed PHACOS showed greater potential to inhibit Staphylococcus aureusT growth when comparedto that of degradation products of non functionalized polyhydroxyoctanoate-co-hexanoate P(HO-co-HH).

. Introduction

The consumption of conventional petroleum based plasticsar exceeds its recycle, although the latter has significantlyncreased since the 90s. These non-biodegradable plastics rep-esent a serious environmental problem, and can be substitutedy environmentally friendly bioplastics, as polyhydroxyalkanoatesPHAs) [1,2]. PHAs are biodegradable polyesters, synthesized byumerous microorganisms and accumulated in a form of car-on storage intracellular granules [3] for self-consumption, oreleased by producer microorganisms after death to the extra-ellular space as carbon source for the microorganisms living in

Please cite this article in press as: V. Martínez, et al., Int. J. Biol. Macro

he same niche [4]. Described PHAs are mainly linear, head-to-tailolyesters composed of over 150 different chiral hydroxy fatty acidonomers with the hydroxyl-substituted carbon atom at 3 position

∗ Corresponding author at: Polymer Biotechnology Group, Environmental Biologyepartment, Centro de Investigaciones Biológicas, CSIC, C/Ramiro de Maeztu 9, POox 28040, Madrid, Spain. Tel.: +34 91 837 31 12; fax: +34 91 536 04 32.

E-mail address: auxi@cib.csic.es (M.A. Prieto).1 These authors contributed equally to this work.

ttp://dx.doi.org/10.1016/j.ijbiomac.2014.04.013141-8130/© 2014 Elsevier B.V. All rights reserved.

© 2014 Elsevier B.V. All rights reserved.

exclusively of the R configuration [5–7]. The number of carbonatoms forming R moiety can vary and, thus, short chain length PHAs(scl-PHAs) contain monomers with 4–5 carbon atoms, whereasmedium chain length PHAs (mcl-PHAs) are composed of monomerswith 6–14 carbon atoms [8]. One of the most promising applicationsof PHAs relies on these structural features, making the polymera potential source of great diversity of R-3-hydroxyalkanoic acids(HAs), hereinafter used to define both monomeric and oligomericbuilding blocks of the polymer [9–11].

HAs are valuable starting materials for pharmaceutical andmedical industries [12–14]. Thus, there is increasing demand fordeveloping novel methods that allow efficient production of HAs.In parallel, there is growing awareness of the importance of thesynthesis of enantiomerically pure mixture. This enantiomericalpurity is a key factor for accomplishing biological function of HAs.Currently, two main strategies are used to produce HAs, one basedon chemical synthesis and the other on biotransformation. How-ever, the drawbacks that follow chemical approaches substantially

mol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.04.013

limit their use. Namely, they tend to be expensive, not environ-mentally friendly, and the obtained yield of enantiomer pure HAsis often poor. Therefore, up-to-date tendency is to replace the con-ventional chemical processes by bio-based ones, which are less

ING ModelB

2 f Biolo

hTu

bmesPeaemdmmomaegrfv

pjct�fsdptl

2

2

3[mS((

2

moss(rwsoomcaw

ARTICLEIOMAC-4280; No. of Pages 6

V. Martínez et al. / International Journal o

armful for the environment and ensure the enantiomerical purity.his demand for more sustainable procedures led to the increasedse of microorganisms for biotechnological processes [15].

Different approaches to promote PHAs biotransformation haveeen reported, mainly based on genetic engineering of enzymaticachinery involved in PHA turnover process, consisting in over-

xpression of PHA degradation enzymes or mutagenesis of PHAynthases [10,11,16]. Biotransformation strategies used to obtainHA derived HAs are generally based on application of purifiednzyme (in vitro approach) or cell system as a factory (in vivopproach). Biodegradation of PHAs can be driven by intracellular orxtracellular depolymerases. The most extensively characterized,odel enzyme in extracellular mcl-PHAs degradation is Pseu-

omonas fluorescens GK13 depolymerase (PhaZGK13) [17,18]. Theain reaction products identified in in vitro systems of PHA depoly-erization of PhaZGK13 are dimers [17,19], whereas in the case

f in vivo intracellular polymer degradation by the PhaZ depoly-erase of Pseudomonas putida KT2442, the main reaction products

re monomers [10,13]. In contrast to costly methods involvingnzyme purification, heterologous PhaZGK13 expressing microor-anisms offer attractive approach regarding HAs production. In thisespect, engineering of metabolic pathway can be used as a power-ul tool to optimize the production process and increase the yieldsia evading HAs consumption by the producer.

Herein, we studied the effectiveness of recombinant PhaZGK13roducer strains to extracellularly degrade mcl-PHA, either aiming

ust to biodegrade it or with the goal of generating HAs. Spe-ific fatty acids metabolism modifications have been addressedo improve the HAs production. After analyzing the influence of-oxidation in the extracellular metabolism of PHA by fadR and

adB mutations, in Escherichia coli and P. putida, respectively, a newtrategy has been designed to obtain enantiopure mixture of PHAerived monomers and oligomers. Finally, under this scheme weroduced novel high added value HAs carrying thioester groups inhe side chain and their antibacterial performance against Staphy-ococcus aureusT has been evaluated.

. Material and methods

.1. Chemicals

Polyhydroxyoctanoate-co-hexanoate (P(HO-co-HH)) and poly--hydroxy-acetylthioalkanoate-co-3-hydroxyalkanoate PHACOS20] were kindly supplied by Biopolis S.L. (Valencia, Spain). Chro-

atography media were obtained from GE-Healthcare (Uppsala,weden). Most chemicals were obtained from Sigma–AldrichSt. Louis, MO). All other chemicals were purchased from MerckDarmstadt, Germany).

.2. Bacterial strains, media and growth conditions

E. coli MC4100 strain was used as host due to its mineraledium growth capability. P. putida KT2442 is a derivative strain

f the parental strain KT2440 whose complete genome nucleotideequence is accessible in the data bank [21]. Unless otherwisetated, E. coli and P. putida strains were grown in Lysogeny BrothLB) medium [22] or in M63 mineral medium [23] at 37 ◦C and 30 ◦C,espectively. The carbon sources added to the mineral mediumere glucose (20 mM) or glycerol (20 mM). Solid media were

upplemented with 1.5% (w/v) agar; agar plates containing anpaque-polymer top layer were prepared by mixing equal volumesf P(HO-co-HH) (5–8 g/L) in water and LB medium agar or mineral

Please cite this article in press as: V. Martínez, et al., Int. J. Biol. Macro

edium agar at about 50 ◦C and pouring 7 mL onto a pre-warmedorresponding bottom layer. The appropriate selection antibiotics,mpicillin (100 �g/mL) and gentamicin (10 �g/mL) were addedhen needed.

PRESSgical Macromolecules xxx (2014) xxx–xxx

2.3. DNA manipulation and plasmid construction

DNA manipulations and other molecular biology techniqueswere essentially performed as described previously [22]. Transfor-mation of E. coli cells was carried out by using the RbCl method or byelectroporation (GenePulser, Bio-Rad) [24]. Plasmid transferenceto the target Pseudomonas strains was done by the filter-matingtechnique [25]. DNA fragments were purified by standard proce-dures using Gene Clean (Bio 101, Inc., Vista, CA). To construct pLJ1,the 872 bp DNA fragment coding for the extracellular PhaZGK13depolymerase was PCR-amplified by using the oligonucleotidesPHAZGKF (5′ TCTAGAAGGAGATAAGTCATGC 3′) and PHAZGKR (5′

AAGCTTCCCGCGGTGGATCA 3′) using the total DNA of the strainP. fluorescens GK13 as template. For PCR amplification, we used2 units of AmpliTaq DNA polymerase (PerkinElmer Life Sciences),1 �g of DNA template, 1 �g of each deoxynucleoside triphosphate,and 2.5 mM MgCl2 in the buffer recommended by the manufacturer.Conditions for amplification were chosen according to the G + Ccontent of the corresponding oligonucleotides. The PCR productwas digested with the engineered endonucleases XbaI and HindIIIand cloned in pUC18.

For pIZPZ construction, pLJ1 plasmid was cut with SmaI andHindIII and the 882 bp fragment was subcloned in pIZ1016 plasmidcut with SalI, blunt extremes made with Klenow enzyme (accordingto manufacturer’s instructions) and digested with HindIII.

E. coli MF4100 was obtained by spontaneous mutagenesis offadR gene. Cells of parental strain MC4100 were grown on LBmedium for 48 h, washed with saline solution, plated onto M63-agar plates with octanoate 15 mM as sole carbon source andincubated for 11 days. The colonies able to grow were sequencedusing FADRR (5′ ATGGTCATTAAGGCGCAAAGC 3′) and FADRF (5′

TTATCGCCCCTGAATGGCTAA 3′) primers to confirm the fadR muta-tion.

Nucleotide sequences were determined directly with the sameoligonucleotides used for cloning. All the constructions were con-firmed by sequencing using an ABI Prism 3730 DNA Sequencer.

2.4. Depolymerase activity plates

For simple evaluation of enzyme activity, indicator plates wereprepared in the same way as P(HO-co-HH) agar plates, except thatthe nutrient medium was replaced with Tris–HCl buffer (15 mM,pH 8). The enzyme solution was dropped onto wells punched inthe plates and subsequently incubated at 37 ◦C. The diameters ofthe resulting clearing zones semiquantitatively indicated enzymeactivity.

2.5. 14C-labeled 3-hydroxyalkanoic release in growing cellssystem

MC4100, MC4100 (pLJ1), KT42FadB (pIZ1016) or KT42FadB(pIZPZ) cells were grown in M63 medium [23] with 20 mM glycerol(E. coli derived strains) or 20 mM glucose (Pseudomonas strains) ascarbon source. This medium was supplemented with 14C-labelledP(HO-co-HH)/water nano-particle emulsion (PHA*), prepared asdescribed by de Eugenio et al. [26], resulting in 34 cpm/�g. Cellswere incubated at 30 ◦C or 37 ◦C using an orbital shaker. Sampleswere taken at different time intervals, filtered and 200 �L of follow-throughs were analyzed using a scintillation counter.

2.6. 3-Hydroxyalkanoic release in growing cells system

mol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.04.013

MC4100 (pUC18), MC4100 (pLJ1), KTFadB (pIZ1016) orKT42FadB (pIZPZ) cells were grown in M63 medium with 20 mMglycerol (E. coli derived strains) or 20 mM glucose (Pseudomonasstrains) as carbon source. P(HO-co-HH)/water particle emulsion

ING ModelB

f Biolo

pcsac0r(

2

woot

spttraCmetto(2bifa7pwsui

2

tapp231am

cUta

2

mas

ARTICLEIOMAC-4280; No. of Pages 6

V. Martínez et al. / International Journal o

repared as described by de Eugenio et al. [26] was added to theseultures to a final concentration of 1.4 mg/mL. Appropriate expres-ion inducers were added when needed. Samples were collectedfter 23 h of incubation at 30 ◦C or 37 ◦C in an orbital shaker andentrifuged at 4 ◦C. Culture supernatants were filtered through.22 �m filters (Millipore Corp), lyophilized, and subsequentlyesuspended in THF for electrospray ionisation mass spectrometryESI-MS) analysis.

.7. Identification of the products released from PHA

As preliminary step to chromatographic separation, samplesere introduced to the ESI source in negative mode by continu-

us infusion by means of the instrument syringe pump at a ratef 3 mL/min. The source was operated at 4.5 kV and the capillaryemperature was set to 200 ◦C.

Chromatographic separation of the oligomers present in theamples was carried out on a Finnigan Surveyor (Thermo Electron)ump coupled with a Finnigan LCQ Deca (Thermo Electron) ionrap mass spectrometer. The separation was performed at roomemperature on a 100 × 2.1 mm (3 �m particle size) Hypersil HyPu-ity C18 column (Thermo Electron) at a flow rate of 100 �L/minnd an injection volume of 5 �L. The degradation products of PHA-OS were identified by analyzing the supernatant of the reactionixture by HPLC mass spectrometry (HPLC–MS). The HPLC–MS

xperiments were carried out on a Finnigan Surveyor (Thermo Elec-ron) pump coupled with a Finnigan LXQ TM (Thermo Electron) ionrap mass spectrometer. The separation was performed at 40 ◦Cn a 2.1 × 150 mm (3.5 �m particle size) XTerra MS C18 columnWaters) at a flow rate of 100 �L/min and an injection volume of5 �L. The mobile phase for analysis of degradation products ofoth P(HO-co-HH) and PHACOS was 0.1% ammonium hydroxide

n water (A) and 0.1% ammonium hydroxide in methanol (B). Theollowing elution program was used: at the start 95% A and 5% B;fter 3 min the percentage of B was linearly increased to 95% in

min, then kept constant for 20 min, ramped to the original com-osition in 5 min, and then equilibrated for 10 min. The detectionas monitored by MS-ESI spectrometry in negative mode at the

ame conditions of source voltage and capillary temperature thatsed in the step of continuous infusion. All spectra were recorded

n full scan mode (m/z 50–1500).

.8. Gas chromatography analysis of hydroxyl fatty acids

The R configuration of HAs was analyzed by GC–MS ofhe methanolysed supernatants where the control commerciallyvailable (R,S)-3-hydroxyoctanoic acid was used as describedreviously [10]. Methanolysis procedure was carried out by sus-ending 5–10 mg of dried aliquots in 2 mL of chloroform and

mL of methanol containing 15% sulfuric acid and 0.5 mg/mL-methylbenzoic acid (internal standard) and then incubated at00 ◦C for 4 h. After cooling, 1 mL of demineralized water was addednd the organic phase containing the resulting methyl esters ofonomers was analyzed.Separation of compounds was carried out in a Beta Dex 120

olumn (30 m × 0.25 mm di.d. × 0.25 mm film thickness) (Supelco,SA). Injector and detector temperature were kept at 300 ◦C. Oven

emperature was initially set at 95 ◦C, held isothermally for 10 minnd then increased at a rate of 0.5 ◦C/min up to 120 ◦C.

.9. Minimal inhibitory concentration (MIC) assay

Please cite this article in press as: V. Martínez, et al., Int. J. Biol. Macro

The minimal inhibitory concentration (MIC) of polymeronomers and oligomers was determined by microdilution assay

ccording to CLSI (Clinical and Laboratory Standards Institute)tandard procedure. Activity of compounds was tested on S. aureusT

PRESSgical Macromolecules xxx (2014) xxx–xxx 3

strain. PHACOS particle emulsion was prepared as described forP(HO-co-HH). To determine the antibacterial activity of hydrolyzedPHACOS, two reaction mixtures containing 3.4 mg/mL of PHACOSparticle suspension in 50 mM Tris–HCl (pH 8), and a control con-taining 3.0 mg/mL of P(HO-co-HH) particle suspension in 50 mMTris–HCl (pH 8) were separately subjected to enzymatic hydrolysisusing previously designed system based on application of mcl-PHAextracellular depolymerase PhaZGK13 (0.2 mg per assay) [27]. Fol-lowing 1 h incubation at 37 ◦C, supernatants of each reaction mixwere centrifuged and filtrated (0.2 �m pore size). The degradationproducts were identified and quantified by HPLC mass spectrome-try (HPLC–MS).

Stock solution of each compound was prepared in phosphate-buffered saline, pH 7.2 (PBS) with an initial concentration of 60 mM.Afterwards, serial dilutions were made in Muller Hinton BrothII (MHB II) to determine MIC values. Briefly, the sterile 96-wellround bottom clear polystyrene plates (Cultek S.L.U., Spain) wereprepared by dispensing 100 �L of appropriate dilution of testedcompound in culture broth per well. The S. aureusT inoculum inMHB II was added to each well, providing a final concentration of5 × 105 CFU/mL. A positive control (containing inoculum withouttested compound) and negative control (containing tested com-pound without inoculum) were included in each microplate. MICwas defined as the lowest concentration of tested substance thatinhibited visible growth of test bacteria.

3. Results and discussion

3.1. Switching E. coli and P. putida bacterial strains toextracellular PHA degraders

Although the application of purified and/or immobilizedenzymes in industrial processes implied a great advance forbiotechnology, advantages of biocatalysis are frequently compro-mised by poor enzymatic performance under non-native reactionconditions, the absence of enzymes with desired substrate or reac-tion specificities, and low metabolic fluxes or competing pathways[28,29]. Instead of using high cost strategies based on enzymepurification, we have developed an in vivo fermentation system,where by genetic engineering PHA degrading capacities of bacte-ria were improved. To this end, the prototype extracellular PHAdepolymerase, PhaZGK13 of P. fluorescens GK13 was introduced intoseveral expression systems. The pUC18 derived plasmid pLJ1 wasconstructed to allow the production of PhaZGK13 in E. coli MC4100.Although this recombinant strain was producing active depoly-merase, it was unable to grow on PHA-agar plates with the mediumchain length biopolymer as sole carbon source, since �-oxidationenzymes in E. coli are only induced in presence of long chain fattyacids [30].

The functionality of E. coli strain was improved via generat-ing an MC4100 derived strain by spontaneous mutation of fadRgene (E. coli MF4100) which allowed the constitutive expres-sion of the �-oxidation enzymes. fadR gene from MF4100 strainwas amplified and sequenced, showing that a mutant form ofFadR (FadR Arg57Leu) was produced. This regulatory mutationenabled MF4100 (pLJ1) to grow on P(HO-co-HH)-agar plates usingthe biopolymer as carbon source and degrade it forming clearcolony-surrounding zones (Fig. 1). In addition, E. coli MF4100 wascomplemented by in trans production of FadR giving rise to E. coliMF4100-FadR strain demonstrating the recovery of wild type phe-notype (data not shown).

mol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.04.013

The abilities of E. coli MF4100 (pLJ1) for degrading mcl-PHAwere compared to those of the model strain P. putida KT2440, anatural intracellular mcl-PHA degrader [26]. The use of this GRAScertified strain has several advantages, including fast growth in

ARTICLE IN PRESSG ModelBIOMAC-4280; No. of Pages 6

4 V. Martínez et al. / International Journal of Biological Macromolecules xxx (2014) xxx–xxx

Fig. 1. Extracellular P(HO-co-HH) degradation by E. coli MF4100 (pLJ1) strain. (A)Cells of E. coli MF4100 (pLJ1) growing on plates containing P(HO-co-HH) as singlecarbon source. (B) Detailed image of E. coli MF4100 (pLJ1) single colony with thesurrounding P(HO-co-HH) hydrolysis halo due to an active PhaZGK13 depolymerase.

Fig. 2. Turbidimetry and CFU number versus culture time of P(HO-co-HH) sup-plemented cultures. P(HO-co-HH)was prepared as a polymer/water emulsion andadded to the cultures to a final concentration of 2.4 mg/mL. Optical density and CFUof P. putida KT2442 (pIZPZ) (closed shapes) and E. coli MF4100 (pLJ1) (open shapes)cells pre-cultured in M63 with glucose [P. putida KT2442 (pIZPZ)] or glycerol [E. coliMF4100 (pLJ1)] and subsequently grown in M63 with P(HO-co-HH) as sole carbonsource was monitored throughout growth curve. Open and closed circles representCr

mmwbac

dUEiiHCfd(Ct(s

Fig. 3. Extracellular depolymerase activity assay of P. putida KT2442 (pIZPZ) andKT42FadB (pIZPZ) strains. Active P(HO-co-HH) depolymerase in P. putida KT2442 and

FU (left axis), while open and closed triangles represent OD600 (right axis). The dataeported are the averages of three different assays.

cl-fatty acids, availability of complete genome sequence andodels for mcl-PHA metabolism analysis [31,32]. P. putida KT2442as engineered to additionally degrade extracellular biopolymer

y expressing PhaZGK13 depolymerase [P. putida KT2442 (pIZPZ)]nd subsequently obtain enantiomerically pure HAs in a growingell system.

We compared the ability of the PhaZGK13 producing strains toegrade mcl-PHA in vivo determining OD600 and Colony Formingnits (CFU) of P(HO-co-HH) supplemented cultures. To this aim,. coli MF4100 (pLJ1) and P. putida KT2442 (pIZPZ) cells were grownn M63 medium with 2.4 mg/mL PHA as sole carbon source. Thenitial OD600 increased up to 2.5 due to the addition of P(HO-co-H) suspension but rapidly decreased during the first 24 h (Fig. 2).FU of KT2442 (pIZPZ) and MF4100 (pLJ1) increased 2.4 and 2.9

old, respectively (Fig. 2). The number of E. coli cells dramaticallyecreased during next 3 days of growth, while the one of KT2442pIZPZ) remained unchanged (Fig. 2). After 7 days, MF4100 (pLJ1)FU reached the minimum value being three orders of magni-

Please cite this article in press as: V. Martínez, et al., Int. J. Biol. Macro

ude lower than the initial inoculum, while the number of KT2442pIZPZ) cells remained constant. These results suggest that bothtrains are able to degrade extracellular mcl-PHA, but P. putida

KT42FadB cells carrying pIZPZ plasmid was detected as clearing zones surroundingthe bacterial mass growth on LB plus P(HO-co-HH)-agar plates (A) or on minimalmedium containing P(HO-co-HH) as single carbon source (B).

derived cells seemed to metabolize more efficiently the hydrolysisproducts.

3.2. Engineering Pseudomonas strains for HAs production

The second approach of this work consisted in the design of agrowing cell system for producing HAs based on the extracellularhydrolysis of mcl-PHA. Based on previous results showing that thePHA depolymerization products in P. putida KT2442 can be metab-olized through �-oxidation pathway [10], we focused our intereston designing specific mutations of fatty acid degradation machin-ery that could block self-utilization of released HAs. Thus, P. putidaKTFadB strain, a fadB gene mutant of P. putida KT2442 [20] wasselected to produce the PhaZGK13 depolymerase.

The ability of P. putida KT2442 (pIZPZ) and KTFadB (pIZPZ)strains to use HAs as carbon source was tested on P(HO-co-HH) toplayered LB-agar and M63-agar plates. While KT2442 (pIZPZ) strainwas growing and degrading mcl-PHA in both plates, FadB mutantstrain was only able to degrade mcl-PHA when other carbon sourcewas present to support cell growth (Fig. 3). These results confirmedthe PhaZGK13 depolymerase functionality in P. putida strains andthe inability of KTFadB mutant strain to efficiently catabolize HAs.Therefore, P. putida KT42FadB strain was targeted as a promisingcandidate for HAs recovery.

3.3. Radioactively labeled PHA hydrolysis as HAs release indicator

To determine the best conditions for HAs release, MC4100(pUC18), MC4100 (pLJ1), KT42FadB (pIZ1016) and KT42FadB(pIZPZ) cells were cultured in presence of 14C-labelled PHA* (seeSection 2). These strains were chosen given their inability to metab-olize the degradation products. Radioactivity present in the filteredsupernatants was measured at different time points. Exclusivelythe cells carrying pLJ1 or pIZPZ plasmids were able to releaseradioactive monomers and oligomers from 14C-labelled P(HO-co-HH) (Fig. 4). The liberation of HAs was maximal in MC4100(pLJ1) and KT42FadB (pIZPZ) strains after 24 h incubation, althoughhigher in the latter. However, the release of monomers/oligomersdecreased after 24 h in the KT42FadB (pIZPZ) strain, probably dueto the existence of an additional enzymatic set for fatty acid degra-dation [11,20,33].

3.4. Growing cell system as a bio-factory of PHA derived HAs andoligomers

mol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.04.013

After targeting the optimal conditions for HAs and oligomerrelease using PHA*, the degradation of non-labeled PHA was ana-lyzed and quantified. P. putida KT2442 (pIZPZ) and KTFadB (pIZPZ)

ARTICLE ING ModelBIOMAC-4280; No. of Pages 6

V. Martínez et al. / International Journal of Biolo

Fig. 4. HAs released by PhaZGK13 producer strains. Released HAs were quanti-fied determining present radioactivity in 200 �L of filtered supernatants of E. coliMC4100 (pUC18), MC4100 (pLJ1), P. putida KT42FadB (pIZ1016) and KT42FadB(pIZPZ) cells grown in presence of 14C-labelled P(HO-co-HH). Open circles, E. coliMC4100 (pLJ1) HAs production; closed circles, MC4100 control strain HAs release;open triangles, P. putida KT42FadB (pIZPZ) HAs release; closed triangles, KT42FadB(pIZ1016) control strain HAs release. The data reported are the averages of threedifferent assays.

Table 1HAs production by PhaZGK13 producer E. coli and P. putida strains.

Growing cellsa OD600 beforeaddingP(HO-co-HH)

Released HAs(mg/mL)

%Conversion

MC4100 1.97 ND NDMC4100 (pLJ1) 1.47 0.031 ± 0.001 2.25KT2442 3.70 ND NDKT2442 (pIZPZ) 2.10 0.010 ± 0.003 0.75KTFadB-

(pIZ1016)2.16 ND ND

KTFadB-(pIZPZ) 2.12 0.270 ± 0.040 19.16

ND – not detected.a Cells were grown in M63 medium with 20 mM glycerol (E. coli derived strains)

or 20 mM glucose (Pseudomonas strains) as carbon source for 23 h at 37 ◦C and 30 ◦C,rfi

aslGm1iimm

rh

(asscpawa

We thank B. Galán and V. Morales for helpful discussions. We

espectively. P(HO-co-HH)/water particle emulsion was added to all cultures to anal concentration of 1.4 mg/mL.

nd their respective control strains were grown in M63 mediumupplemented with glucose and PHA as described in Section 2. Fol-owing the incubation, supernatants were filtered and analyzed byC–MS. The highest production of HAs was obtained when FadButant strain P. putida KT42FadB (pIZPZ) was used (0.27 mg/mL,

9.16% hydrolysis) (Table 1). The reaction products obtained apply-ng the growing cell system strategy were identified by electrosprayonisation mass spectrometry (ESI-MS), HO–HO dimer being the

ost abundant compound in both supernatants (SFigs. 1–3). Allonomers obtained were of the R configuration (data not shown).See Figs. S1–S3 as supplementary file. Supplementary mate-

ial related to this article can be found, in the online version, atttp://dx.doi.org/10.1016/j.ijbiomac.2014.04.013.

These results demonstrated that KT2442 (pIZPZ) and MF4100pLJ1) producing PhaZGK13 can degrade the extracellular mcl-PHAnd reuse released monomers and oligomers as carbon and energyource, since these strains can grow using P(HO-co-HH) as soleubstrate. However, when �-oxidation system is slowed down theells can no longer exploit the degradation products for its ownrofit, and, thus, the yield of recovered HAs increase (Table 1). Thispproach permits recovery of enantiomerically pure HAs from PHA

Please cite this article in press as: V. Martínez, et al., Int. J. Biol. Macro

ith great potential for biomedical and pharmaceutical high valuepplications [13,14].

PRESSgical Macromolecules xxx (2014) xxx–xxx 5

3.5. Application of in vivo cell system to obtain added valuePHACOS derived HAs

After establishing the strategy to efficiently produce HAs, wehad keen interest in applying the system to obtain high-added valueproducts. Therefore, as a starting material we used PHACOS, func-tionalized PHA, previously obtained by metabolic engineering [20].Since, some PHA derived HAs have been shown bactericidal to cer-tain Gram-negative and Gram-positive bacteria [9,34–37,16] andanti Staphylococcal activity of PHACOS has been demonstrated [38],we examined if by using designed platform we can obtain highadded value product with antibacterial properties. Moreover, weinvestigated if the performance of obtained products had improvedantibacterial efficiency compared to the previously reported HAs(MIC ≈ 1–5 mM) [9,34–37,16].

For that we applied the developed system based on the strainKT42FadB (pIZPZ) for production of 3-hydroxyalkanoic acids (HAs)from PHACOS degradation. The identified reaction products ofhydrolyzed PHACOS were 63% trimers (three monomers of OH-6ATH fused) and 30% trimers containing mix of OH-C8 andOH-6ATH monomers, while the rest (7%) was a mix of all monomersand dimers (SFigs. 4 and 5). Depending on the depolymerase activ-ity, the end products of PHA degradation can be monomers, bothmonomers and dimers, or a mixture of oligomers as a result of theenzymatic PHA degradation [4]. As previously mentioned, mcl-PHAdepolymerases reaction products are mainly dimers (PhaZGK13 orPhaZBd) or monomers (PhaZKT2442). Interestingly, when PhaZGK13degrades this polymer containing thioester groups in the side chain,more than 90% of the released products are trimers, which werenever reported for this biocatalyst.

See Figs. S4 and S5 as supplementary file. Supplementary mate-rial related to this article can be found, in the online version, athttp://dx.doi.org/10.1016/j.ijbiomac.2014.04.013.

Inhibitory effect of PHACOS subunits was compared to thoseP(HO-co-HH) (SFig. 3) on bacterial growth was determined by mon-itoring MIC values according to microdilution CLSI (Clinical andLaboratory Standards Institute) standard protocol. MIC of PHACOSderived hydroxycarboxylic acid suspension showed much lowervalue (40 �M) compared to MIC of P(HO-co-HH) derived hydrox-ycarboxylic acid suspension (3 mM). Taken together, these resultssuggest that PhaZGK13 is suitable for hydrolyzing PHACOS and thatthe designed platform is useful for production of added value HAs.Finally, the difference in antimicrobial activity is due to the pres-ence of functionalized dimers and trimers in case of hydrolyzedPHACOS.

4. Conclusion

In this study we designed a growing cell system strategyto degrade mcl-PHA and obtain HAs based on the extracellularproduction of mcl-PHA depolymerase. Polymer was success-fully degraded by recombinant cells harboring P. fluorescensGK13 depolymerase. Importantly, modification of the �-oxidationpathway blocked utilization of PHAs degradation products and per-mitted efficient production of HAs. Taken together, these metabolicengineering assets were applied to obtain added value HAs. Finally,by using constructed cell system, PHACOS was degraded andobtained products were shown antibacterial against S. aureusT.

Acknowledgements

mol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.04.013

thank Biópolis S.L. for the supplied polymers. The technical work ofA. Valencia is greatly appreciated. This work was supported by the

ING ModelB

6 f Biolo

M2

R

[

[

[

[

[

[[

[[

[

[

[

[

[[[[

[

[[[[[

[

[

[

[

ARTICLEIOMAC-4280; No. of Pages 6

V. Martínez et al. / International Journal o

inisterio of Economía y Competitividad, Espana (BIO2010-21049,01120E092).

eferences

[1] M.A. Keane, ChemSusChem 2 (2009) 207–214.[2] G.Q. Chen, Chem. Soc. Rev. 38 (2009) 2434–2446.[3] M.A. Prieto, L.I. de Eugenio, B. Galán, J.M. Luengo, B. Witholt, in: J.L. Ramos, A.

Filloux (Eds.), Pseudomonas: A Model System in Biology. Pseudomonas, vol. V,Springer, Berlin, Germany, 2007.

[4] D. Jendrossek, R. Handrick, Ann. Rev. Microbiol. 56 (2002) 403–432.[5] K. Sudesh, H. Abe, Y. Doi, Prog. Polym. Sci. 25 (2000) 1503–1555.[6] A. Steinbüchel, T. Lütke-Eversloh, FEMS Microbiol. Lett. 221 (2003) 191–196.[7] M. Tortajada, L.F. da Silva, M.A. Prieto, Int. Microbiol. 16 (2013) 1–15.[8] L.L. Madison, G.W. Huisman, Microbiol. Mol. Biol. Rev. 63 (1999) 21–53.[9] K. Ruth, A. Grubelnik, R. Hartmann, T. Egli, M. Zinn, Q. Ren, Biomacromolecules

8 (2007) 279–286.10] L.I. de Eugenio, I.F. Escapa, V. Morales, N. Dinjaski, B. Galán, J.L. García, M.A.

Prieto, Environ. Microbiol. 12 (2010) 207–221.11] A.L. Chung, G.D. Zeng, H.L. Jin, Q. Wu, J.C. Chen, G.Q. Chen, Metab. Eng. 17 (2013)

23–29.12] T. Ohashi, J. Hasegawa, in: A.N. Collins, G.N. Sheldrake, J. Crosby (Eds.), Chirality

in Industry, vol. 1, John Wiley & Sons, New York, 1992, pp. 249–268.13] Q. Ren, K. Ruth, L. Thöny-Meyer, M. Zinn, Appl. Microbiol. Biotechnol. 87 (2010)

41–52.14] S. O’Connor, E. Szwej, J. Nikodinovic-Runic, A. O’Connor, A.T. Byrne, M. Devo-

celle, N. O’Donovan, W.M. Gallagher, R. Babu, S.T. Kenny, M. Zinn, Q.R. Zulian,K.E. O’Connor, Biomaterials 34 (2013) 2710–2718.

15] J.M. Otero, J. Nielsen, Biotechnol. Bioeng. 105 (2010) 439–460.16] A. Sandoval, E. Arias-Barrau, F. Bermejo, L. Canedo, G. Naharro, E.R. Olivera, J.M.

Luengo, Appl. Microbiol. Biotechnol. 67 (2005) 97–105.

Please cite this article in press as: V. Martínez, et al., Int. J. Biol. Macro

17] A. Schirmer, D. Jendrossek, J. Biotechnol. 176 (1994) 7065–7073.18] L.I. De Eugenio, J.L. García, P. García, M.A. Prieto, J.M. Sanz, Eng. Life Sci. 8 (2008)

260–267.19] J. Gangoiti, M. Santos, M.J. Llama, J.L. Serra, Appl. Environ. Microbiol. 76 (2010)

3554–3560.

[

[

PRESSgical Macromolecules xxx (2014) xxx–xxx

20] I.F. Escapa, V. Morales, V.P. Martino, E. Pollet, L. Avérous, J.L. García, M.A. Prieto,Appl. Microbiol. Biotechnol. 89 (2011) 1583–1598.

21] K.E. Nelson, C. Weinel, I.T. Paulsen, R.J. Dodson, H. Hilbert, V.A.P. Martins dosSantos, D.E. Fouts, S.R. Gill, M. Pop, M. Holmes, L. Brinkac, M. Beanan, R.T. DeBoy,S. Daugherty, J. Kolonay, R. Madupu, W. Nelson, O. White, J. Peterson, H. Khouri,I. Hance, P.C. Lee, E. Holtzapple, D. Scanlan, K. Tran, A. Moazzez, T. Utterback, M.Rizzo, K. Lee, D. Kosack, D. Moestl, H. Wedler, J. Lauber, D. Stjepandic, J. Hoheisel,M. Straetz, S. Heim, C. Kiewitz, J. Eisen, K.N. Timmis, Environ. Microbiol. 4 (2002)799–808.

22] J. Sambrook, D.W. Russell, Molecular Cloning. A Laboratory Manual, CSHL Press,Cold Spring Harbor, NY, 2001.

23] J.H. Miller, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1972.24] W.J. Dower, J.F. Miller, C.W. Ragsdale, Nucleic Acids Res. 16 (1998) 6127–6145.25] M. Herrero, V. de Lorenzo, K.N. Timmis, J. Bacteriol. 172 (1990) 6557–6567.26] L.I. de Eugenio, P. García, J.M. Luengo, J.M. Sanz, J.S. Román, J.L. García, M.A.

Prieto, J. Biol. Chem. 282 (2007) 4951–4962.27] V. Martínez, F. de la Pena, J. García-Hidalgo, I. de la Mata, J.L. García, M.A. Prieto,

Appl. Environ. Microbiol. 78 (2012) 6017–6026.28] A.J. Straathof, S. Panke, A. Schmid, Curr. Opin. Biotechnol. 13 (2002) 548–556.29] P.C. Cirino, L. Sun, Biotechnol. Prog. 24 (2008) 515–519.30] P.N. Black, C.C. DiRusso, Biochim. Biophys. Acta 1210 (1994) 123–145.31] J. Nogales, B.Ø. Palsson, I. Thiele, BMC Syst. Biol. 16 (2008) 79.32] I. Poblete-Castro, J. Becker, K. Dohnt, V.M. dos Santos, C. Wittmann, Appl. Micro-

biol. Biotechnol. 93 (2012) 2279–2290.33] E.R. Olivera, D. Carnicero, R. Jodra, B. Minambres, B. García, G.A. Abraham, A.

Gallardo, J.S. Román, J.L. García, G. Naharro, J.M. Luengo, Environ. Microbiol. 3(2001) 612–618.

34] G. Bergsson, J. Arnfinnsson, S.M. Karlsson, O. Steingrímsson, H. Thormar,Antimicrob. Agents Chemother. 42 (1998) 2290–2294.

35] G. Bergsson, O. Steingrímsson, H. Thormar, Antimicrob. Agents Chemother. 43(1999) 2790–2792.

36] G. Bergsson, J. Arnfinnsson, O. Steingrímsson, H. Thormar, Antimicrob. AgentsChemother. 45 (2001) 3209–3212.

mol. (2014), http://dx.doi.org/10.1016/j.ijbiomac.2014.04.013

37] G. Bergsson, O. Steingrímsson, H. Thormar, Int. J. Antimicrob. Agents 20 (2002)258–262.

38] N. Dinjaski, M. Fernández-Gutiérrez, S. Selvam, F.J. Parra-Ruiz, S.M. Lehman, J.San Román, E. García, J.L. García, A.J. García, M.A. Prieto, Biomaterials 35 (2014)14–24.