Biosynthesis of Natural-Synthetic Hybrid Copolymers

Department of Chemistry, University of Alabama, Huntsville, Alabama 35899. Received November 14, 2003; Revised Manuscript Received December 22, 2003...
0 downloads 0 Views 133KB Size
Download PDF
Biomacromolecules 2004, 5, 643-649

643

Biosynthesis of Natural-Synthetic Hybrid Copolymers: Polyhydroxyoctanoate-Diethylene Glycol Vorapat Sanguanchaipaiwong,† Candace L. Gabelish,† James Hook,† Carmen Scholz,‡ and L. John R. Foster*,† Biopolymer Research Group, CAMD, School of Biotechnology and Biomolecular Sciences and Nuclear Magnetic Resonance Facility, University of New South Wales, Sydney, NSW 2052 Australia, and Department of Chemistry, University of Alabama, Huntsville, Alabama 35899 Received November 14, 2003; Revised Manuscript Received December 22, 2003

A new natural-synthetic hybrid biomaterial has been isolated from the growth of Pseudomonas oleoVorans in the presence of diethylene glycol (DEG). DEG was consumed by P. oleoVorans with 20 mM sodium octanoate in modified E* medium, but its presence in the fermentation medium retarded cell growth and viability, influencing production and composition of polyhydroxyalkanoates with medium chain length substituents (mclPHAs) and consequently attenuating PHA yield. DEG affected the composition of the mclPHA with an increase in the C8 component: polyhydroxyoctanoate (PHO). Gas chromatography-mass spectrometry (GC-MS) was used to quantitatively monitor DEG in the system and reveal its cellular adsorption and penetration. Intracellularly, the DEG significantly reduced the molar mass of the mclPHA; PHO with a bimodal distribution of high and low molecular weight fractions was observed. 1H NMR, 2-D COSY, and heteronuclear single quantum coherence spectra confirmed that the high molecular weight fraction consisted of PHO chains terminated by DEG. Thus, the synthesis of this natural-synthetic hybrid copolymer, PHODEG, opens the way for microbial synthesis of a wide variety of PHA-DEG copolymers with a range of bioactive properties. Introduction Polyhydroxyalkanoates (PHAs) are a diverse family of optically active polyesters synthesized by a wide range of microorganisms, usually under unfavorable growth conditions.1 Under such conditions PHAs are accumulated intracellularly as carbon and energy reserves and may also serve as ion sinks.2,3 As a consequence of their physiochemical properties and biodegradability, PHAs have received considerable commercial interest as alternatives to conventional thermoplastics.4,5 Their production from renewable or waste sources coupled with their disposal as biowaste makes them increasingly attractive in the pursuit of sustainable development.6-8 The first and most studied member of the PHA family, poly-β-hydroxybutyrate (PHB), possesses a relatively high crystallinity and brittle nature when processed through various melt crystallization or solution casting techniques.9-11 The physicomechanical properties of PHB were subsequently improved by the microbial synthesis of its random copolymer with polyhydroxyvalerate (PHV).12,13 The flexibility of the PHA biosynthetic pathway has been extensively explored and exploited to produce PHAs with over 100 different monomeric components.14 The properties of these biopolymers range from rigid and crystalline to flexible and elastomeric.1,2,15 * To whom correspondence should be addressed. E-mail: J.Foster@ unsw.edu.au. † University of New South Wales. ‡ University of Alabama.

Figure 1. Formula for PHAs with the variable substituent R in the stereoregular R configuration.

Flexibility in PHA synthesis is species dependent. Thus, whereas Ralstonia eutropha produces PHAs possessing relatively short length constituents in the side chain (sclPHA), such as the PHB, many of the pseudomonads belonging to the RNA homology group I produce PHAs with medium length alkyl side chains (mclPHA) such as poly-β-hydroxyoctanoate (PHO), as well as PHAs with chemically functional groups (fclPHA), for example poly-β-hydroxyphenylvalerate (PHPV).16,17 The general structure of β-linked PHAs is shown in Figure 1. In addition to PHA composition, the physicomechanical properties of PHA based devices can be affected by their processing.11 Recently, Gross and co-workers have reported the manipulation of PHB molecular weight through the addition of poly(ethylene glycol) (PEG) to microbial fermentations.18-21 Together with PHA composition, manipulation of molar mass plays a crucial role in the control of physicomechanical properties. In addition to altering PHB molecular mass, PEG also affects cell growth, polymer concentration, cellular productivity, cell viability, and polymer structure. These parameters are influenced by the molecular weight of the PEG, its concentration, and the PHB

10.1021/bm0344708 CCC: $27.50 © 2004 American Chemical Society Published on Web 02/06/2004

644

Biomacromolecules, Vol. 5, No. 2, 2004

producing microbial species.18-23 Furthermore, Gross and coworkers reported the potential for microbial species to incorporate PEG units at the synthetic terminus of the PHB chain, i.e., “PEGylation”.18 A family of PEGs includes members that are biocompatible and serve a number of important biomedical functions such as organ and tissue preservation.24 Furthermore, there is a wide range of PEG conjugates with useful properties and applications. Surfaces modified by covalent attachment of PEG chains are resistant to cell and protein adsorption and have consequently received considerable attention in the preparation of biomaterials.25,26 The biocompatibility of some PHAs coupled with their PEGylation, implies the potential to synthesize a range of new biologically active materials. Furthermore, the incorporation of PEG end groups into PHA chains may permit the preparation of telechelic polymers. Thus, the PEG group may be reacted, post extraction, with a range of molecules introducing the possibility of new conjugates between PHAs and other natural or synthetic polymers. Whereas the production of natural-synthetic hybrids of PHB-PEG have been reported for fermentations of R. eutropha and Alcaligenes latus, recent studies with strains of Pseudomonas oleoVorans have failed to induce PEGylation of mclPHAs.22,23 The mclPHAs constitute the majority component of the PHA family, with a greater diversity of physicomechanical properties.1 Thus, PEGylation of mclPHAs would introduce the possibility of producing a wide range of reactive PHA-PEG copolymers. Motivated by a desire to induce PEGylation in mclPHAs, we have investigated the effect of DEG (PEG106) on the growth of Pseudomonas oleoVorans and its synthesis of mclPHAs. Materials and Methods Materials. Diethylene glycol (DEG) was supplied as PEG106 by Sigma-Aldrich, (Sydney Australia). Octanoic acid was obtained as its sodium salt also from Sigma-Aldrich. All other chemicals were obtained from APS Chemicals, Seven Hills, Australia. All chemicals were at least 98% purity. Bacterial Strains, Inoculum Preparation, and Growth Conditions. Stock cultures of P. oleoVorans (ATCC 29347) were frozen at -10 °C from which working cultures were obtained every four weeks. The strain was maintained by successive subculture onto fresh agar modified E* medium plates containing 20 mM sodium salt of octanoic acid (sodium octanoate) and were stored at 4 °C.27 Modified E* medium consisted of (NH4)H2PO4, 5.94 g/L; K2HPO4, 5.8 g/L; KH2PO4, 3.7 g/L; 100 mM MgSO4, 20 mL/L and microelement solution, 1 mL/L. The micro-element solution contained 2.78 g/L FeSO4‚7H2O, 1.98 g/L MnCl2‚4H2O, 2.81 g/L CoSO4‚7H2O, 1.67 g/L CaCl2‚2H2O, 0.117 g/L CuCl2‚ 2H2O and 0.29 g/L ZnSO4 dissolved in 1 M HCl. P. oleoVorans was grown aerobically in a 500 mL baffled Erlenmeyer flask containing 100 mL liquid modified E* medium and 20mM sodium octanoate at 30 °C and agitation of 200 rpm. After the late log growth phase was achieved (about 22 h), this preseed culture was used as 10% (v/v)

Sanguanchaipaiwong et al.

inocula for four seed flasks with the same media compositions, working volume, and conditions to that of the preseed. The late log growth phase of these seed cultures was reached in 10 h after which 400 mL of the culture was transferred to a 5 L fermentor containing 3.6 L of liquid modified E* medium and 20 mM sodium octanoate. (Braun Biostat, Braun Scientific, Germany). The same protocol was simultaneously carried out with the addition of 2% (v/v) DEG in each of the seed flasks. Inoculation of the 5 L fermentors proceeded with similar total cell numbers. Validation of growth and polymer production trends was established by repeating the fermentations an additional two times. Polymer Isolation. The intracellular polymer was isolated from lyophilized cells by extraction with an excess of chloroform (80 mL per 1 g of biomass, 30 °C, 48h).22 The product was purified by methanol precipitation as crude polymer. The crude polymer displayed a bimodal molecular weight distribution. The relatively high molecular weight component (H) of the crude polymer was extracted and purified by the application of at least three methanol precipitation/washing cycles to the crude extract. The relatively low molecular weight component (L) was extracted and purified from the crude extract through size exclusion chromatography using a biologically inert molecular sieve (Bio-beads SX-1, Bio-Rad, CA). The residual methanolcrude polymer mix was stored at -4 °C for 24 h after removal of purified high molecular weight fraction polymer precipitate (H). After this period a second precipitate was observed and extracted (L2). Analytical Methods. During fermentation, duplicate volumes of 3 mL were sampled periodically until the beginning of the stationary phase. Viable cell numbers were determined in triplicate using a standard spread plate technique. Cells were harvested by centrifugation (6000 g, 15 min) and repeatedly washed in sterile fresh media with no carbon sources until no residual carbon, octanoate and DEG, were detected. The biomass was lyophilized for 24 h and then acclimatized under room-temperature conditions prior to determination of its dry weight. The dry cell weight was calculated by subtracting the weight of accumulated microbial polymer (see below) from the biomass weight. The supernatants were used to determine concentrations of octanoic acid and DEG (see below) Gas Chromatography-Mass Spectrometry. Polymer concentration was determined after methanolysis of the biomass by gas chromatography-mass spectrometry (GCMS) using a GC model 5890 Series II with a MS model 5972 Series Mass Selective Detector (Hewlett-Packard, Penn. USA). The GC was equipped with a BP5 (SGE, Sydney Australia) capillary column (25 m). Helium was used as the carrier gas (1.2 mL/min). The temperatures of the injector and detector were 250 and 275 °C, respectively. The following program was used: 90 °C for 1 min, 7 °C/min to 160 °C, held for 1 min, then 20 °C/min to 250 °C which was then held for 5 min. A 1.0 µL injection was made with a split ratio of 10:1. The mass spectrometer was autotuned with perfluorotributylamine (PFTBA; Sigma Sydney, Australia). Polymer components were identified and quantitated

Biosynthesis of Natural-Synthetic Hybrid Copolymers

based on their retention times against known standards and their m/e values (103, 74, and 71). The concentrations of DEG and octanoate were quantitatively determined in the supernatant samples and washings by GC-MS after a cleanup step. In addition, the aqueous phase from the methanolysis protocol was also examined to determine the presence and concentration of DEG. The pHs of the samples were reduced to 2.8 (below pKa of 4.89 for octanoic acid) so that all octanoate molecules were protonated prior to addition to Supleclean LC-18 SPE tubes (Supelco, Sydney, Australia). Undesired salts from the medium were thus removed before elution of analytes with methanol. Octanoate and DEG were separated and quantitatively detected by GC-MS using a BP21 (SGE) capillary column (30 m). Helium was used as the carrier gas (1.0 mL/min). For DEG, the temperature profile was 85 °C initially, then 10 °C/min to 160 °C, 5 °C/min to 180 °C, then 10 °C/min to a final temperature of 220 °C. For free octanoic acid, the same oven profile was used until 160 °C, then 5 °C/min to 190 °C, then 10 °C/min to 220 °C which was held for 30 s. A 1.0 µL injection was made with a split ratio of 10:1. The major fragments, which were consistently observed for free octanoic acid, had the following m/e values: 60, 73, and 101 and for DEG, m/e values: 45, 75, and 76. Gel Permeation Chromatography. GPC analyses of the crude and purified polymer components were performed on a modular system comprising the following units: an LC10ATVP Shimadzu solvent delivery system, a SIL-10ADVP Shimadzu auto-injector with a stepwise injection control motor with an accuracy of (1 µL, a column set which consisted of a PL 5.0 µm bead size guard column and a set of 3-5.0 µm PL linear columns (10e3, 10e4, 10e5 Å) kept at a constant 40 °C inside a CTO-10AC VP Shimadzu Column Oven and an RID-10A Shimadzu Refractive Index Detector. Tetrahydrofuran (THF) was utilized as the continuous phase at a flow rate of 1 mL/min. Nuclear Magnetic Resonance Spectroscopy. Samples of crude polymer were dissolved in deuterated chloroform (ca. 4 mg/mL) and examined using a Bruker DMX600, operating at 600.13 MHz for 1H and 150.92 MHz for 13C. 1H spectra were recorded with a pulse width of 4.5 µs (45° pulse), a spectral width of 6.6 kHz, an acquisition time of 2.5 s, a relaxation delay of 6 s with 64-256 scans for required signalto-noise and referenced internally to chloroform (7.26 ppm with respect to tetramethylsilane). Purified polymer extracts were examined using 1H NMR and 2-D spectroscopy (1H1 H, COSY, correlation spectra and1H-13C correlation spectroscopy (HSQC, heteronuclear single quantum coherence spectra)) on a Bruker DMX600. Results and Discussion P. oleoWorans Growth Studies. In an attempt to produce a natural synthetic hybrid copolymer of mclPHA-DEG, P. oleoVorans was cultivated in modified E* media containing 20 mM sodium octanoate and 2% (v/v) DEG. The growth parameters from this fermentation were compared to concurrent fermentations in the absence of DEG. In our investigation the cultivation of P. oleoVorans with 20 mM sodium octanoate as sole carbon source resulted in

Biomacromolecules, Vol. 5, No. 2, 2004 645

Figure 2. Fermentation of P. oleovorans with 20 mM sodium octanoate (a) dry cell concentration in g/L b, viable cells in log. No/ mL O, (b) Octanoate concentration in media in g/L 9, PHA concentration in g/L 4.

a maximum dry cell concentration of 2.27 ( 0.05 g/L after 10 h incubation, (5L Braun Biostat, 30 °C, 200rpm). The addition of 2% (v/v) DEG to the media had little effect on the duration of the exponential growth phase but significantly reduced cell growth rates from approximately 0.26 to 0.13 g/L/h, resulting in a cell concentration of 1.85 ( 0.05 g/L (Figures 2A and 3A). In contrast, Ashby et al. and Kim report that fermentations of P. oleoVorans (ATCC 29347, B14682, and B14683) in the presence of PEG200 and PEG400 showed negligible changes in cell concentrations.22,23 Similarly, 2-3% (v/v) PEG200 had little effect on the growth of R. eutropha and A. latus.18,21 Cell viability, as determined by colony forming units (Log no./mL), was also affected by the presence of DEG. The introduction of DEG reduced cell viability determined at the stationary phase from 1.95 ( 0.02 × 1010/mL to 2.57 ( 0.02 × 109/mL, nearly an 8-fold decrease. Using a similar type of P. oleoVorans (ATCC 29347) and various concentrations of PEG200 (1-16% v/v), Kim reported no effects on cell viability until a loading of 8% (v/v).23 Our results show that DEG significantly affected cell growth and viability compared to its higher molecular weight counterparts. Such effects may be due to the interesting interactions PEG molecules have with cell membranes, to which they readily adsorb.28,29 Furthermore, it has been shown that PEG molecules associate with the membrane phospholipid headgroups leading to membrane fluidity and consequently

646

Biomacromolecules, Vol. 5, No. 2, 2004

Sanguanchaipaiwong et al. Table 1. Non-PHA Cell and PHA Yield Coefficients, YX/S and YP/S, Based on a Carbon Balance after 8 Hours Incubation for Fermentations of P. oleovorans with and without DEG PHA sample

Yx/s

Yp/s

control (- DEG)a + DEG (2% v/v)b

0.47 0.11

0.11 0.06

a Octanoate as sole carbon source. b Octanoate and DEG as carbon sources.

Figure 4. Concentration of DEG, monitored as PEG106, removed from cell mass by washing in sterile media (g of DEG per g of dry cells). Figure 3. Fermentation of P. oleovorans with 20 mM sodium octanoate and 2% (v/v) DEG added prior to inoculation (a) dry cell concentration in g/L b, viable cells in log. No/mL O, (b) Octanoate concentration in media in g/L 9, PHA concentration in g/L 4, DEG concentration in media in g/L [.

increased permeability to ions and small metabolites.30 One can speculate that such an effect may facilitate the penetration of low molecular weight PEG into cells of P. oleoVorans, where they may influence PHA synthesis. A comparison of Figures 2B and 3B shows that DEG did indeed influence PHA yield. In the control fermentation (absence of DEG), the PHA yield determined through GCMS analysis of the methanolysed biomass was 0.366 ( 0.003 g/L (Figure 2B). This was reduced to 0.249 ( 0.008 g/L when P. oleoVorans was cultivated in the presence of DEG (Figure 3B). However, the PHA concentration relative to the biomass was only slightly reduced from 14 to 12%. In comparison, the addition of comparatively higher molecular weight PEGs to P. oleoVorans fermentations are reported to have significantly reduced PHA yields.22,23 DEG was quantitatively monitored in the media using GCMS. Figure 3B clearly shows a decrease in DEG concentration from 20 g/L to approximately 14.6 g/L during the first 6 h of the fermentation. This then gradually decreased to 13.9 ( 0.05 g/L at 8 h with no subsequent change. An uninoculated control showed a negligible decrease in DEG over a similar duration (0.008 ( 0.005 g/L). Our decision to use 2% (v/v) DEG was motivated by a desire to induce “DEGylation”. Given the effects on cell membranes by PEG together with a lack of PEGylation observed in P. oleoVorans using PEG200 and PEG400, it was anticipated that a reduction in PEG molecular weight would facilitate its penetration into cells. Similarly, previous reports on PEG modulated synthesis of sclPHAs have suggested that a loading of 2% (v/v) was the minimum concentration required

to induce PEGylation.18,19 It is interesting to note therefore that there was a decrease of only 30% in DEG concentration after 8 h of incubation in our fermentations. This period coincided with the almost complete utilization of octanoate in the system and maximum PHA concentration (Figure 3). Our results suggest that P. oleoVorans is able to cometabolize DEG in the presence of a more readily available carbon source such as octanoate; this suggestion is consistent with the report by Kawai demonstrating the metabolism of DEG by a number of pseudomonads.31 It is unclear if cometabolism by P. oleoVorans was directed toward cell maintenance and/or the generation of undetected extracellular carbon rich compounds. Consequently, cell and PHA yield coefficients based on the carbon balance, Yx/s and Yp/s, respectively, show significant reductions compared to fermentations in the absence of DEG (Table 1). Unfortunately, no PEG data is available for comparison with previous studies in this area.18-23 GC-MS analysis of the cell washings showed that DEG was also associated with the outer surface of the cells. The concentration of DEG associated with the cell mass increased from 80 mg of DEG per g of dry cells (mg/g) at the start of the fermentation to approximately 350 mg/g after 6 h of incubation (Figure 4). Interestingly, the concentration of cellassociated DEG was reduced after this period. Analysis of the aqueous fraction of the methanolysed biomass also demonstrated the presence of DEG, suggesting its penetration into the cell (Figure 5). The presence of intracellular DEG was later confirmed through NMR (Figure 7A). Thus, DEG was adsorbed onto the cell surface and entered into the cell. Polymer Composition. The composition of the Intracellular polymer synthesized by P. oleoVorans during fermentation on sodium octanoate and DEG was analyzed using GC-MS and compared to that cultivated on octanoate

Biosynthesis of Natural-Synthetic Hybrid Copolymers

Figure 5. GC-MS trace illustrating the presence of DEG in the aqueous phase of the biomass methanolysis, (a) GC-MS trace indicating peak x with m/e values equivalent to those of DEG, (b) single ion mass spectrum of peak x confirming DEG.

Figure 6. GPC traces of PHA extracted from DEG modulated fermentation of P. oleovorans; (a) crude polymer showing bimodal distribution with high “H” and low “L” molar mass fractions, (b) L fraction after purification and (c) H fraction after purification.

alone. The molecular weights of the extracted polymers were determined using GPC. Although the presence of DEG within cells of P. oleoVorans had little effect on PHA yield, it had a noticeable effect on polymer composition. When cultivated in the presence of sodium octanoate, P. oleoVorans synthesized a mclPHA:polyhydroxyoctanoate, (PHO). PHO from this fermentation possessed three monomeric components with hydroxyoctanoate (HO-C8) the main constituent (87.3%), together with hydroxyhexanoate (HX-C6) and hydroxydecanoate (HD-C10), 9.7% and 3.0% respectively. Intracellular DEG resulted in a noticeable shift toward the C8 component (93.9%; Table 2). These results are consistent with those reported for sclPHAs in R. eutropha, but contradict those regarding mclPHA production in P. oleoVorans with PEG200.20,23 The crude polymer was composed of both high (H) and low (L) molecular weight fractions and compositional analysis after their separation showed negligible difference between the two. Furthermore, a second precipitate (L2) was observed upon storage of the residual methanol from the polymer purification process at -4 °C for 24 h. The composition of this precipitate was also identified through GC as PHO with similar components as the H and L fractions (Table 2). Motivated by a desire to control PHA molecular masses,

Biomacromolecules, Vol. 5, No. 2, 2004 647

Gross and co-workers have reported the influence of various PEGs on sclPHA production and properties in a number of bacterial species. More recently, Ashby et al. and Kim have demonstrated similar PEG200 and PEG400 modulated control in mclPHAs. The mclPHA synthesized by P. oleoVorans in the presence of DEG showed significant reductions in molecular weight number (Mn) and polydispersity compared to the mclPHA produced in the absence of DEG, (PEG106; Table 2). These changes in molecular mass are consistent with those previously reported.22,23 More importantly, the molecular weights for the crude polymer extract from the DEG modulated fermentation showed a bimodal distribution consisting of high and low molecular weight fractions (Figure 6, Table 2). Ashby et al. report similar fractions in sclPHAs produced by A. latus, with the low molecular weight portion consisting of PEG200 terminated PHB. However, this only occurred when PEG200 was added during and not at initiation of the fermentation.19 Cold storage of the residual methanol resulted in the isolation of a second PHO precipitate with a similar molecular mass to that of the L fraction. NMR Studies. Purified polymer isolates from the DEG modulated fermentation were examined using 1H NMR spectroscopy (Figure 7). Individual polymer samples consisting of the relatively high (H) and low (L) molecular weights were examined. The NMR spectrum for both samples clearly demonstrated strong signals for all of the typical peaks associated with the mclPHA identified through GC-MS, that is, the methine signal at 5.2 ppm, the signal for the methylene protons in the main chain that are adjacent to the chiral carbon between 2.4 and 2.6 ppm, the methylene protons of the side chains between 1.2 and 1.4 ppm and terminal methyl groups in the side chains at 0.9 ppm. These are consistent with the overlap of peaks from the major monomer units found within the mclPHA sample (HX, HO, and HD). The 1H NMR spectrum for the “H” fraction polymer also showed noticeable signals at 3.6, 3.7, and 3.74 ppm, corresponding to protons of DEG without internal repeating unit. The signal at 4.2 ppm correlated to the covalent bond between DEG and mclPHA. Assuming that the contribution of the overlapping signals in the 3.6-4.3 ppm region can be estimated by Bernoullian curve fitting, the area under the peaks was measured by cutting and weighing. The integration results showed that the ratio of protons “b” to “d” to “a” to “c” was approximately 1:1:1:1; suggesting no unbound DEG in the purified mclPHA-DEG. Cultivations of P. oleoVorans with PEG200 and PEG400 produced similar spectra for mclPHA but failed to show any peaks corresponding to PEG and its covalent bonding to the mclPHA, i.e., PEGylation.23 Although 1H NMR is strongly suggestive of DEGylation of PHO chains, further corroboration was sought from 2-D NMR correlation experiments. 1H-1H COSY and1H-13C HSQC experiments were performed to confirm the nature of the covalent linkage between terminal units of mclPHA and DEG. The COSY spectrum (Figure 8) clearly demonstrates that the four sets of peaks in the 3.6-4.3 ppm region are linked together as two ethylene units, whereas the HSQC spectrum (Figure 9) confirms the link to 13C sites of the DEG moiety.

648

Biomacromolecules, Vol. 5, No. 2, 2004

Sanguanchaipaiwong et al.

Figure 7. 600 MHZ 1H NMR spectra of mclPHA purified from DEG modulated fermentation of P. oleovorans, (A) mclPHA composed of relatively high molecular mass, i.e., “H” fraction, and (B) mclPHA composed of relatively low molecular mass, i.e., “L” fraction.

The origin of the PHO possessing the relatively low molecular mass, i.e., the “L” fraction, is unclear. Although the proton NMR spectrum clearly shows the signals for a mclPHA, there are a number of unidentified peaks. The peaks in region 3.0-4.5 ppm are suggestive of a DEG residue. It is possible that the L fraction is the result of harvesting the cell mass at the onset of the stationary phase when some degree of polymer degradation and metabolism may have occurred.

Conclusions P. oleoVorans is capable of synthesizing a range of PHAs with diverse monomeric components. Control of the PHA molecular mass in P. oleoVorans can be achieved through the strategic addition of PEG.22,23 In our studies, we have investigated the influence of DEG on the growth of P. oleoVorans and its synthesis of mclPHAs. Our results suggest that DEG in the fermentation system was co-metabolized

Biomacromolecules, Vol. 5, No. 2, 2004 649

Biosynthesis of Natural-Synthetic Hybrid Copolymers Table 2. Composition and Molar Masses of PHAs Produced by P. oleovorans in the Presence and Absence of 2% (v/v) DEG composition (mol. %) PHA sample control (- DEG) + DEGa H (2T v/v) L Ppt. IIb L2

C6

C8

C10

9.7 4.3 3.9 4.2

87.3 93.9 95.0 94.5

3.0 1.8 1.1 1.3

Mw Mn (×103)c (×103) 235 100 8 8

142 77 6 6

PDI 1.65 1.30 1.25 1.31

a DEG was added to microbial growth medium prior to inoculation. Second precipitate from purification process. c Molar masses were determined by gel permeation chromatograph (GPC).

b

reducing their molecular mass and altering the PHA composition. More importantly, DEG, in contrast to previous studies with PEG200 and PEG400, was utilized by the cells to DEGylate the mclPHA chains. The microbial synthesis of a natural-synthetic hybrid of PHO-DEG introduces the potential for a wide range of PHA-PEG copolymers with a range of bioactive properties. Acknowledgment. The authors acknowledge Dr. L. Yee and J. Chiu from the Centre for Advanced Macromolecular Design (CAMD) for their assistance with GPC determination. Dr. H. Stender is acknowledged for expert assistance with acquisition of NMR spectra. References and Notes (1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

Figure 8. 600 MHz 1H-1H COSY spectrum of purified DEGylated PHO in CDCl3, showing the expansion of the region 3.5-4.3 ppm.

(11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25)

Figure 9. 600 MHz 1H-13C HSQC spectrum of purified DEGylated PHO in CDCl3, showing the expansion of the region 3.5-4.3 ppm and 60-75 ppm.

(26) (27) (28)

with the octanoate carbon source. DEG was adsorbed to the cell surfaces and penetrated into the cell and consequently reduced cell growth and viability. Intracellular DEG was also found associated with extracted PHA, suggesting its diffusion into the microbial inclusion bodies. DEG significantly affected the synthesis of mclPHAs by P. oleoVorans,

(29) (30) (31)

Doi, Y. Microbial Polyesters; VCH: New York, 1990. Anderson, A. J.; Dawes, E. A. Microbiol. ReV. 1990, 54, 450. Dawes, E. A.; Senior, P. J. AdV. Microbiol. Physiol. 1973, 10, 135. Byrom, D. In Plastics from microbes; Mobley, D. P., Ed.; Hanser Publishers: New York, 1994; p 5. Doi, Y., Steinbuchel, A., Eds.; Biopolymers Vol. 1-3; John Wiley & Sons: Weinheim, Germany, 2001. Budwill, K.; Fedorak, P. M.; Page, W. J. Appl. EnViron. Microbiol. 1992, 58, 1398. Mergaert, J.; Anderson, C.; Wouters, A.; Swings, J. J. Appl. EnViron. Microbiol. 1994, 2, 177. Satoh, H.; Mino, T.; Matsuo, T. Int. J. Biolog. Macromol. 1999, 25, 105. Lemoigne, M. Ann. Inst. Pasteur. 1925, 39, 144. Azuma, Y.; Yoshie, N.; Sakurai, M.; Inoue, Y.; Chujo, R. Polymer 1992, 33, 4763. Yasin, M.; Tighe, B. J. Plast. Rubb. Compos. Process. Appl. 1993, 19, 15. Howells, E. R. Chem. Ind. 1982, 15, 508. Holmes, P. A. Phys. Technol. 1985, 16, 32. Steinbuchel, A. In Biotechnology; Rehm, H.-J., Reed, G., Eds.; John Wiley & Sons: Weinheim, Germany, 1996; p 405. Brandl, H.; Gross, R. A.; Lenz, R. W.; Fuller, R. C. In AdVances in Biochemical Engineering/Biotechnology; Fichter, A., Ghose, T. K., Eds.; Springer: Berlin, Germany, 1990; p 77. Fritzsche, K.; Lenz, R. W.; Fuller, R. C. Makromol. Chem. 1990, 191, 1957. Curley, J. M.; Lenz, R. W.; Fuller, R. C.; Browne, S. E.; Gabriel, C. P.; Sapana, P. Polymer 1997, 38, 5313. Ashby, R. D.; Shi, F.; Gross, R. A. Tetrahedron 1997, 53 (45), 15209. Ashby, R. D.; Shi, F.; Gross, R. A. Biotech. Bioeng. 1999, 62, 106. Shi, F.; Gross, R. A.; Rutherford, D. R. Macromolecules 1996, 29, 10. Shi, F.; Ashby, R.; Gross, R. A. Macromolecules 1996, 29, 7753. Ashby, R. D.; Solaiman, D. K. Y.; Foglia, T. A. Appl. Microbiol. Biotechnol. 2002, 60, 154. Kim, O. J. Polym. Res. 2000, 7, 91. Harris, J. M. In: Poly(ethylene glycol) chemistry: biotechnical and biomedical applications; Plenum Press: New York, 1992; p 1. Inada, Y.; Matsushima, A.; Kodera, Y.; Nishimura, H. J. Bioact. Compat. Polym. 1990, 5, 343. Albertsson, P. A. Partition of cell particles and macromolecules; Wiley: New York, 1986. Foster, L. J. R.; Stuart, E. S.; Tehrani, A.; Lenz, R. W.; Fuller, R. C. Int. J. Biol. Macromol. 1996, 19, 177. Boni, L. T.; Hah, J. S.; Hui, S. W.; Mukherjee, P.; Ho, J. T.; Jung, C. Y. Biochim. Biophys. Acta 1984, 775, 409. Yamazaki, M.; Ito, T. Biochemistry 1990, 29, 1309. Ingram, L. O.; Buttke, T. M. In AdVances in Microbiology Physiology; Rose, A. H., Tempest, D. W., Eds.; Academic Press: New York, 1984; pp 25, 253. Kawai, F. Appl. Microbiol. Biotechnol. 1995, 44, 532.

BM0344708