- Email: [email protected]
Microbial secretion of lactate-enriched oligomers for efficient conversion into lactide: A biological shortcut to polylactide
Journal of Bioscience and Bioengineering VOL. xx No. xx, 1e5, 2017 www.elsevier.com/locate/jbiosc
Microbial secretion of lactate-enriched oligomers for efficient conversion into lactide: A biological shortcut to polylactide Camila Utsunomia,1 Ken’ichiro Matsumoto,1, 2 Sakiko Date,1 Chiaki Hori,1, 2 and Seiichi Taguchi1, 2, * Division of Applied Chemistry, Graduate School of Engineering, Hokkaido University, N13-W8, Kita-ku, Sapporo 060-8628, Japan1 and CREST, JST, 4-1-8 Honcho, Kawaguchi, Saitama 332-0012, Japan2 Received 9 February 2017; accepted 4 March 2017 Available online xxx
Recently, we have succeeded in establishing the microbial platform for the secretion of lactate (LA)-based oligomers (D-LAOs), which consist of D-LA and D-3-hydroxybutyrate (D-3HB). The secretory production of D-LAOs was substantially enhanced by the supplementation of diethylene glycol (DEG), which resulted in the generation of DEG-capped oligomers at the carboxyl terminal (referred as D-LAOs-DEG). The microbial D-LAOs should be key compounds for the synthesis of lactide, an important intermediate for polylactides (PLAs) production, eliminating the costly chemo-oligomerization step in the PLA production process. Therefore, in order to demonstrate a proof-of-concept, here, we attempted to convert the D-LAOs-DEG into lactide via metal-catalyzed thermal depolymerization. As a result, D-LAOs-DEG containing 68 mol% LA were successfully converted into lactide, revealing that the DEG bound to D-LAOs-DEG does not inhibit the conversion into lactide. However, the lactide yield (4%) was considerably lower than that of synthetic LA homooligomers (33%). We presumed that 3HB units in the polymer chain blocked the lactide formation, and therefore, we investigated the LA enrichment in the oligomers. As the results, the combination of an LA-overproducing Escherichia coli mutant (Ddld and DpflA) with the use of xylose as a carbon source exhibited synergistic effect to increase LA fraction in the oligomers up to 89 mol%. The LA-enriched D-LAOs-DEG were converted into lactide with greater yield (18%). These results demonstrated that a greener shortcut route for PLA production can be created by using the microbial D-LAOs secretion system. Ó 2017, The Society for Biotechnology, Japan. All rights reserved. [Key words: Biosynthesis; Secretory production; Bioprocess; Metal-catalyzed reaction; Polyhydroxyalkanoate; Biomass; Polylactic acid; Oligoester]
Polylactides (PLAs) are one of the most successful biobased polyesters with diverse applications due to physical properties, biocompatibility, and processability (1). Currently, biomass resources such as corn, cane sugar, potato starch, and tapioca starch are used as carbon sources for PLA production (2). The conventional process of PLA production from biomass involves (i) a bioprocess for the production of lactic acid from biomass sugars, and (ii) a multistep chemo-process containing the oligomerization of lactic acid to generate lactate (LA) oligomers, depolymerization of LA oligomers into lactide (cyclic dimer of LA), and the polymerization of lactide into high molecular weight PLA via ringopening polymerization (ROP) (3,4) (Fig. 1). Although this is the main industrial route, the multistep chemo-bio process is considered complex and expensive relative to petroleum-based polymers (5). Among the steps, the production of lactide from lactic acid, which comprises the oligomerization of lactic acid and the lactide synthesis from LA oligomers, contributes to 30% of total cost for PLA production (6).
* Corresponding author at: Department of Chemistry for Life Sciences and Agriculture, Faculty of Life Sciences, Tokyo University of Agriculture, 1-1-1 Sakuragaoka, Setagaya-ku, Tokyo 156-8502, Japan. Tel.: þ81 3 5477 2577; fax: þ81 3 5477 2627. E-mail address: [email protected] (S. Taguchi).
Recently, we have discovered the secretory production of D-LAbased oligomers (D-LAOs), which consist of D-LA and D-3hydroxybutyrate (D-3HB), by recombinant Escherichia coli grown on semisynthetic medium using glucose as carbon source (7). This bacterium expresses an engineered polyhydroxyalkanoate (PHA) synthase (PhaC), designated as D-specific LA-polymerizing enzyme (LPE). This enzyme had been formerly developed for the production of LA-based polyesters (8,9). The production of D-LAOs was remarkably increased by supplementing diethylene glycol (DEG) in the bacterial cultivation. The transference of the polymer chain took place from the LPE to DEG via chain transfer (CT) reaction when DEG was added in the culture. The high frequency of DEGmediated CT reaction drastically decreased the molecular weight of the intracellularly accumulated LA-based polyesters, and also enhanced the production and secretion efficiency of D-LAOs. The oligomers synthesized under DEG supplementation were nearly fully conjugated with DEG at the carboxyl terminal (termed as DLAOs-DEG). The aim of the present study is to verify the feasibility of the DLAOs-DEG on the conversion into lactide, since this is a key reaction to construct a shortcut route in the process of poly(D-lactide) (PDLA) production (Fig. 1). By establishing this new shortcut route, the laborious purification of lactic acid from the microbial culture broth (10) and the lactic acid oligomerization can be eliminated, increasing the sustainability of the material and the cost effectiveness of PDLA production. Moreover, the production of optically
1389-1723/$ e see front matter Ó 2017, The Society for Biotechnology, Japan. All rights reserved. http://dx.doi.org/10.1016/j.jbiosc.2017.03.002
Please cite this article in press as: Utsunomia, C., et al., Microbial secretion of lactate-enriched oligomers for efficient conversion into lactide: A biological shortcut to polylactide, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.03.002
2
UTSUNOMIA ET AL.
J. BIOSCI. BIOENG.,
Bio-process D-LAOs O
Biomass
H
O
O O
O
D-LA
Conventional Route
Microbial platform LPE Secretory New shortcut route production
D-3HB
O
O HO
OH
Polycondensation
D-LAOs
Depolymerization
O
O O
O
ROP
O
n
Lactide
D-Lactic acid
OH
DEG
y
x
O
Polylactides
Chemo-process FIG. 1. Scheme of the conventional route and new shortcut route for PDLA production. In the conventional process, D-lactate (LA) oligomers are chemically prepared from purified Dlactic acid, which is generated by bacterial fermentation, via a polycondensation reaction. In the shortcut route proposed in this study, highlighted in red color, D-LA-based oligomers (D-LAOs) are directly secreted by bacteria from renewable biomass. LPE, lactate-polymerizing enzyme; ROP, ring-opening polymerization; DEG, diethylene glycol; D-3HB, D-3-hydroxybutyrate. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
pure D-LAO is beneficial in view of the superior thermal properties of poly(L-lactide) (PLLA)/PDLA stereocomplex (11,12), and the higher production cost of D-LA compared to L-LA. Therefore, DLAOs-DEG were used as substrates for conversion into lactide via metal-catalyzed reaction. To examine the effect of the terminal DEG in D-LAOs-DEG on the lactide formation, free-form D-LAOs (synthesized without DEG addition) were used for comparison. Moreover, to evaluate the influence of 3HB units in D-LAOs-DEG on the lactide synthesis efficiency, we attempted to synthesize LAenriched oligomers through metabolic and fermentation engineering approaches.
MATERIALS AND METHODS Bacterial strains and plasmids E. coli BW25113 (13), and the dual-gene knockout mutant (Dpfla and Ddld) JWMB1 (14) were used as the host strains. The expression vector pTV118NpctphaC1Ps(ST/FS/QK)AB harboring pct, phaC1Ps(ST/FS/ QK), phaA, and phaB genes (9), was used for the production of D-LAOs. Culture conditions in test tubes Cultivations for D-LAOs-DEG production were carried out in 10 mL glass test tubes containing LuriaeBertani (LB) medium (1.7 mL) with 20 g/L glucose or xylose and 100 mg/L ampicillin at 30 C for 48 h with reciprocal shaking at 180 rpm. The cultivations were performed with the supplementation of 5% DEG (v/v). For observing the aeration effects on D-LAOs production, the cultivations of the dual-gene knockout mutant with xylose as carbon source, were also performed using 2.5, 3.4, and 5.1 mL medium. Shake flask cultures Free-form D-LAOs were produced in shake flask cultivation. Seed culture of recombinant E. coli was prepared using 2 mL LB medium containing 100 mg/L ampicillin in 10 mL glass test tubes and cultured at 30 C for 12 h with reciprocal shaking at 180 rpm. One milliliter of the seed culture was then transferred into 100 mL LB medium containing 20 g/L glucose, and 100 mg/L ampicillin in a 500 mL shake flask and cultured at 30 C for 48 h with reciprocal shaking at 120 rpm. Measurement of extracellular D-LAOs The cell-free culture supernatant was analyzed before and after HCl treatment. HCl was added to the supernatant at a final concentration of 2.0 M and incubated at 100 C overnight to hydrolyse the D-LAOs. Afterward, the hydrolysate was neutralized with 2.0 M NaOH. The estimation of D-LAOs in the culture medium was determined by liquid chromatography-mass spectrometry (LC-MS) (LCMS-8030, Shimadzu, Japan) based on the difference of lactic acid and 3-hydroxybutyric acid concentrations in the samples after and before HCl treatment, as described previously (7). Extraction of D-LAOs from the culture supernatant D-LAOs were concentrated from the cell-free culture supernatant by two-phase extraction using chloroform (CHCl3), with modifications on the previously established method (7). The extraction was performed by adding 1 volume of CHCl3 to 1 volume of supernatant and mixing vigorously. After the separation of CHCl3 and water layers, the CHCl3 phase was transferred to a new test tube. The extraction was performed three times. To remove excess DEG, lactic acid/3-hydroxybutyric acid monomers, and short oligomers, 1 volume of 0.9% NaCl solution pH 8 was added to the
resulting CHCl3 fraction. The washing step was repeated twice. The molecular weight distribution of extracted oligomers was determined by electrospray ionization-time-of-flight-mass spectrometry (ESI-TOF-MS), as described in the literature (7). Lactide synthesis The synthesis of lactide occurs via thermal depolymerization of the LA oligomers via a metalecatalyzed backbiting reaction of the eOH end groups (15). Approximately 40 mg of vacuum dried extracted D-LAOs containing 68, 78, and 89 mol% LA, were individually weighted together with 40 mg zinc oxide (ZnO, Kanto Chemical, Japan). The sample bottle containing D-LAOs and catalyst was placed inside a rotary type Sibata GTO-350RD glass oven (Sibata Scientific Technology, Japan). The reaction system was heated and kept at 180 C while being rotated in a circular motion for 1 h under vacuum. The vaporized lactide was condensed into a bottle cooled on ice, and recovered in chloroform. Watersoluble synthetic L-LA oligomers (Glart, Japan) and free-form D-LAO, were also converted into lactide as experimental controls. The lactide yield (%, 2 [mmol lactide/mmol initial oligomeric LA]) was calculated based on the amounts of produced lactide and initial oligomeric LA quantified using the 1H NMR analysis. 1 1 H NMR of D-LAOs and lactide H NMR of extracted oligomers and generated lactides were recorded in CDCl3 with tetramethylsilane as the internal reference using a JEOL JNM-ECS400 spectrometer (JEOL, Japan) at 400 MHz. Benzoic acid (Wako, Japan) was used as an internal standard to quantify oligomers and lactide based on the integral area of the methyl group (eCH3); 45 excitation pulse was used and relaxation delay was set to 10 s.
RESULTS AND DISCUSSION D-LAOs-DEG could be converted into lactide via metalcatalyzed backbiting reaction The synthesis of lactide from the extracted D-LAOs-DEG was undertaken to verify the applicability of the biosynthesized D-LAOs-DEG for subsequent PLA production. These oligomers were produced in the test tube by the wild-type strain BW25113 grown on glucose with 5% DEG supplementation. After extraction of the culture supernatant with chloroform, D-LAOs-DEG with 68 mol% LA and degree of polymerization (DP) of approximately trimer to 7 mer were recovered in chloroform phase (designated as extracted D-LAOsDEG). As controls, extracted free-form D-LAOs (61 mol% LA, DP w trimer to 16 mer) and synthetic L-LA homo-oligomers (L-LAOs) (100 mol% LA, DP w trimer to 14 mer), were also used as substrates for lactide synthesis. For generating lactides, the LA oligomers were heated with zinc oxide as catalyst, and the vaporized fraction was recovered by condensation in bottle 3 (Fig. 2A) and subjected to 1H NMR analysis (Fig. 2B). The 1H NMR spectrum (d in ppm) of the sample generated from D-LAOs-DEG exhibited signals at 5.0 ppm (1H, q, A) and 1.7 ppm (3H, d, B), which were identical to those of the standard Dlactide. The lactide generated from free-form D-LAOs also exhibited
Please cite this article in press as: Utsunomia, C., et al., Microbial secretion of lactate-enriched oligomers for efficient conversion into lactide: A biological shortcut to polylactide, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.03.002
VOL. xx, 2017
MICROBIAL LA-OLIGOMERS FOR SHORTCUT IN PLA PRODUCTION
A
Bottle 2
Bottle 3 Condensation bulb
Heater tube
LAO(a) + ZnO (d)
3
Lactide(e) (c)
(b)
Vacuum pump
Sample bulb Bottle 1
Cooling unit
B
A O
O
H
O CH3
H
H3 B CH O
D-Lactide standard
D-Lactide from D-LAOs-DEG
1.7
5.0
B
A 5.0
4.0
3.0 ppm
2.0
1.0
FIG. 2. Lactide synthesis from extracted D-LAOs-DEG. (A) Scheme of the lactide synthesis apparatus. The location of the initial oligomer in bottle 1 (a), condensed oligomer in bottle 2 (b), condensed oligomer in bottle 3 (c), residual oligomer in bottle 1 (d), and lactide (e), in the lactide synthesis apparatus, is represented here. (B) 1H NMR spectrum of crude lactide synthesized from D-LAOs-DEG 68 mol% LA. (C) Lactide yields from free-form D-LAOs, D-LAOs-DEG, and synthetic L-LAOs as substrates. Refer to Fig. 2A for obtaining details about samples localization in the lactide synthesis apparatus.
Please cite this article in press as: Utsunomia, C., et al., Microbial secretion of lactate-enriched oligomers for efficient conversion into lactide: A biological shortcut to polylactide, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.03.002
4
UTSUNOMIA ET AL.
J. BIOSCI. BIOENG.,
TABLE 1. Secretory production of D-LAOs-DEG from glucose and xylose by engineered BW25113 and dual-gene knockout mutant (Dpfla and Ddld) JWMB1. Strain
Carbon source
BW25113 Glucose Xylose JWMB1 Glucose Xylose a b
CDW (g/L)
4.4 5.3 2.7 4.2
0.2 0.1 0.5 0.2
D-LAOs-DEG
Extracted D-LAOs-DEG
Oligomer LA Recovery (%) Oligomer LA (g/L) (mol%)b (g/L) (mol%)a 6.6 4.6 5.9 8.1
1.2 1.1 0.7 2.9
90 97 78 97
4 2 5 1
3.5 1.6 3.0 4.5
68 62.5 76.5 83
53 35 51 56
Determined on the basis of LC/MS analysis. Determined on the basis of 1H NMR analysis.
the same resonances as the standard D-lactide (Supplementary Fig. S1). Synthetic L-LAOs were converted into lactides by using the same procedure. These results proved that the D-LAOs-DEG can be catalytically converted into lactide as well as free-form D-LAOs. The conclusion is in agreement with the fact that lactide is formed via a backbiting reaction involving the hydroxyl terminal of the oligomer as the active site (16), thus, the carboxyl-terminal structure of D-LAOs should not block the lactide formation. The products condensed in bottle 2 and the residual products in bottle 1 (Fig. 2A) were also analyzed by 1H NMR. Unreacted oligomers were found in bottles 1, 2, and 3 after the depolymerization reaction of D-LAOsDEG, free-form D-LAOs, and synthetic L-LAOs (Fig. 2C). In addition, the D-LAOs condensed in bottle 2 had higher 3HB fraction than the oligomers recovered in bottle 3 (Fig. 2C). The oligomers trapped in bottle 3 should possess lower boiling temperature compared to those in bottle 2, because temperature in bottle 3 is lower than that in bottle 2. LA fraction in D-LAOs-DEG was increased by using xylose as carbon source and a dual-gene knockout mutant as a host strain Despite the microbial D-LAOs were shown to serve as precursor for lactide, the lactide formation over the substrate consumption obtained from free-form D-LAOs (25%) and D-LAOsDEG (13%) were substantially lower than that of synthetic L-LAOs (77%) (Fig. 2C). This result is presumably attributable to the presence of 3HB units in the D-LAOs, which could act as a stopper of the backbiting reaction and decrease the conversion rate of the oligomers into lactide. In addition, the presence of 3HB units decreases the frequency of LAeLA dyad in the oligomers, which is essential for lactide formation. Accordingly, enhancing the LA fraction in the oligomers may further increase the efficiency of the lactide synthesis. Based on this idea, we attempted to produce LA-enriched oligomers by optimizing cultivation conditions such as carbon source, bacterial strain, and culture aeration. The effect of carbon source, glucose and xylose, on D-LAOs-DEG production and their monomer composition was investigated (Table 1). The total production of D-LAOs-DEG in xylose culture was 4.6 1.1 g/L, which was lower than that obtained from glucose (6.6 1.2 g/L). Nevertheless, the D-LAOs-DEG in xylose included a higher LA fraction (97 2 mol% LA) relative to the oligomers secreted in the glucose cultivation (90 4 mol% LA). These results demonstrated that xylose is indeed effective for enhancing the LA fraction in D-LAOs-DEG, although the total production of D-LAOsDEG was lower compared to the glucose culture. These findings are consistent with the previous report regarding the production of high-molecular-weight P(LA-co-3HB) (17). The LA enrichment in DLAOs caused by the consumption of xylose might be related to the different capacities of regenerating NADH and NADPH in the metabolism routes of xylose and glucose (18,19). As a consequence of that, the xylose metabolism is thought to have a higher LA units supplying rate than glucose (17). In order to address the influence of the host strain, D-LAOs-DEG were produced by the dual-gene knockout mutant (Dpfla and Ddld) JWMB1 from glucose or xylose in comparison with the parent strain
TABLE 2. Secretory production of D-LAOs-DEG from xylose by engineered E. coli JWMB1 under microaerobic condition. Cultivation volume (mL)
1.7 2.5 3.4 5.1
CDW (g/L)
D-LAOs-DEG Oligomer (g/L)
4.2 3.9 3.6 3.0
0.2 0.1 0.0 0.1
8.1 7.4 7.7 6.3
Extracted D-LAOs-DEG
LA (mol%)
2.9 1.6 1.3 1.1
97 97 93 92
Oligomer (g/L)
LA (mol%)
Recovery (%)
4.5 4.8 5.0 3.3
83 84 89 85
53 65 65 52
1 2 2 2
BW25113 (Table 1). The deletion of pflA is known to eliminate formate formation from acetyl-CoA, channeling the flux toward lactic acid (20). The dld mutation prevents lactic acid oxidation into pyruvate, improving intracellular availability of lactic acid (21). When glucose was used as a carbon source, JWMB1 produced similar amount of D-LAOs-DEG compared to BW25113, and LA fraction in the D-LAOs-DEG obtained using JWMB1 (78 mol%) was lower than that from BW25113 (90 mol%). In contrast, with the use of xylose, the D-LAOs-DEG production (8.1 2.9 g/L, 97 1 mol% LA) was higher than that of BW25113 (4.6 1.1 g/L, 97 2 mol% LA). This result indicates that there is a synergy between the dualmutation and use of xylose as a carbon source to increase both the production and LA fraction of the oligomers. Microaerobic conditions increased the recovery of extracted D-LAOs-DEG In order to further upregulate the LA fraction in DLAOs-DEG, the oligomers were produced under microaerobic conditions, which are well-known to promote lactic acid production (22,23) and LA fraction enrichment in LA-based polyester (24). To control the microaerobic conditions using simplified method, the volume of culture medium in a test tube was increased from 1.7 to 5.1 mL. As the result, recovery of D-LAOs-DEG in chloroform phase was improved by using the microaerobic conditions (Table 2). Among them, the highest LA fraction in extracted D-LAOs-DEG was 89 mol% LA (Fig. 3). The total oligomer productions and their respective LA fractions, were, however, rather decreased (Table 2). These results indicated that the hydrophobicity of the oligomers, which is determined by the molecular weight and monomer composition of the oligomers, tends to increase under microaerobic conditions. Here, it should be noticed that the molecular weight of extracted D-LAOs-DEG (DP w trimer to 7mer) was not significantly altered by using microaerobic conditions (Supplementary Fig. S2), but the amount of extracted D-LAOs-DEG was increased (Table 2). This phenomenon is probably due to the fact that the molecular weight of secreted oligomers is limited by the solubility of oligomer molecules in water.
3 H
O 2
2
O 1
x
D-3HB
A
O
O 1 O
C O
B y
OH D
DEG
D-LA LA(2)
DEG(A,C)
DEG(B)
DEG(D)
LA(1) 3HB(2)
3HB(3)
3HB(1) 5.0
4.0
3.0 ppm
2.0
1.0
FIG. 3. 1H NMR spectrum of D-LAOs-DEG containing 89 mol% LA.
Please cite this article in press as: Utsunomia, C., et al., Microbial secretion of lactate-enriched oligomers for efficient conversion into lactide: A biological shortcut to polylactide, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.03.002
VOL. xx, 2017
MICROBIAL LA-OLIGOMERS FOR SHORTCUT IN PLA PRODUCTION
5
JSPS Kakenhi [grant numbers 26660080 to S.T. and 26281043 to K.M.].
References
FIG. 4. Variation of the lactide yield (%) as a function of the LA fraction in extracted DLAOs-DEG. As a control, lactide was synthesized from synthetic L-LAOs.
Lactide yield was enhanced by LA-enrichment in oligomers The influence of the LA fraction in D-LAOs on the efficiency of lactide synthesis was evaluated using extracted LAenriched D-LAOs-DEG containing up to 89 mol% LA with nearly the same DP of approximately trimer to 7-mer. The LA fraction enrichment in D-LAOs-DEG significantly increased the lactide yields up to 18% (Fig. 4). Nevertheless, the lactide yield from LAenriched D-LAOs-DEG was still lower than that obtained from synthetic L-LAOs. Further increase in LA fraction of D-LAOs-DEG seems to be necessary to achieve more efficient lactide synthesis. In addition, an important factor determining the efficiency of lactide synthesis should be the molecular weight of D-LAOs-DEG. During lactide synthesis, significant amount of oligomers was condensed in bottles 2 and 3 (Fig. 2A and Supplementary Table S1), suggesting that short oligomers were lost by vaporization during heating (25). In the industrial process, oligomers with molecular weights around 400e2500 g/mol, which correspond to DP of approximately 5-mer to 34-mer, are used in the synthesis of lactide (6). Therefore, increasing the molecular weight of the secreted oligomers should be effective to improve the conversion efficiency of D-LAOs-DEG into lactide. In conclusion, here, we demonstrated the successful lactide synthesis by using D-LAOs-DEG secreted by engineered E. coli expressing evolved LPE. Notably, the backbiting reaction-based lactide conversion was not inhibited by the presence of DEG at the carboxyl-terminal of D-LAOs. This finding provides us a proofof-concept for establishing a shortcut route for PDLA production. Moreover, the improvement in the lactide synthesis efficiency was achieved via LA enrichment in the oligomers by the combination of xylose as carbon source with the use of a dual-gene knockout mutant (Dpfla and Ddld) strain. Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.jbiosc.2017.03.002. ACKNOWLEDGMENTS We thank Dr. Toshifumi Satoh, Hokkaido University, for helpful discussions about lactide synthesis, and Dr. Tomoki Erata, Hokkaido University, for his useful comments on NMR analysis. This study was partially supported by CREST, JST (to S.T.), ALCA, JST (to K.M.),
1. Liu, M.-J., Chen, S.-C., Yang, K.-K., and Wang, Y.-Z.: Biodegradable polylactide based materials with improved crystallinity, mechanical properties and rheological behaviour by introducing a long-chain branched copolymer, RSC Adv., 5, 42162e42173 (2015). 2. Shen, L., Worrell, E., and Patel, M.: Present and future development in plastics from biomass, Biofuels Bioprod. Biorefin., 4, 25e40 (2010). 3. Gupta, A. P. and Kumar, V.: New emerging trends in synthetic biodegradable polymersepolylactide: a critique, Eur. Polym. J., 43, 4053e4074 (2007). 4. Slomkowski, S., Penczek, S., and Duda, A.: Polylactidesdan overview, Polym. Adv. Technol., 25, 436e447 (2014). 5. Lasprilla, A. J., Martinez, G. A., Lunelli, B. H., Jardini, A. L., and Filho, R. M.: Poly-lactic acid synthesis for application in biomedical devices e a review, Biotechnol. Adv., 30, 321e328 (2012). 6. Van Wouwe, P., Dusselier, M., Vanleeuw, E., and Sels, B.: Lactide synthesis and chirality control for polylactic acid production, ChemSusChem, 9, 907e921 (2016). 7. Utsunomia, C., Matsumoto, K., and Taguchi, S.: Microbial secretion of Dlactate-based oligomers, ACS Sustain. Chem. Eng., 5, 2360e2367 (2017). 8. Taguchi, S., Yamada, M., Matsumoto, K., Tajima, K., Satoh, Y., Munekata, M., Ohno, K., Kohda, K., Shimamura, T., and Kambe, H.: A microbial factory for lactate-based polyesters using a lactate-polymerizing enzyme, Proc. Natl. Acad. Sci. USA, 105, 17323e17327 (2008). 9. Yamada, M., Matsumoto, K., Shimizu, K., Uramoto, S., Nakai, T., Shozui, F., and Taguchi, S.: Adjustable mutations in lactate (LA)-polymerizing enzyme for the microbial production of LA-based polyesters with tailor-made monomer composition, Biomacromolecules, 11, 815e819 (2010). 10. Li, Q.-Z., Jiang, X.-L., Feng, X.-J., Wang, J.-M., Sun, C., Zhang, H.-B., Xian, M., and Liu, H.-Z.: Recovery processes of organic acids from fermentation broths in the biomass-based industry, J. Microbiol. Biotechnol., 26, 1e8 (2016). 11. Ikada, Y., Jamshidi, K., Tsuji, H., and Hyon, S. H.: Stereocomplex formation between enantiomeric poly(lactides), Macromolecules, 20, 904e906 (1987). 12. Tsuji, H.: Poly(lactide) stereocomplexes: formation, structure, properties, degradation, and applications, Macromol. Biosci., 5, 569e597 (2005). 13. Baba, T., Ara, T., Hasegawa, M., Takai, Y., Okumura, Y., Baba, M., Datsenko, K. A., Tomita, M., Wanner, B. L., and Mori, H.: Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection, Mol. Syst. Biol., 2 (2006). 2006.0008. 14. Nduko, J. M., Matsumoto, K., Ooi, T., and Taguchi, S.: Enhanced production of poly(lactate-co-3-hydroxybutyrate) from xylose in engineered Escherichia coli overexpressing a galactitol transporter, Appl. Microbiol. Biotechnol., 98, 2453e2460 (2014). 15. Yoo, D. K., Kim, D., and Lee, D. S.: Synthesis of lactide from oligomeric PLA: effects of temperature, pressure, and catalyst, Macromol. Res., 14, 510e516 (2006). 16. Nishida, H., Mori, T., Hoshihara, S., Fan, Y., Shirai, Y., and Endo, T.: Effect of tin on poly(L-lactic acid) pyrolysis, Polym. Degrad. Stab., 81, 515e523 (2003). 17. Nduko, J. M., Matsumoto, K., Ooi, T., and Taguchi, S.: Effectiveness of xylose utilization for high yield production of lactate-enriched P(lactate-co-3hydroxybutyrate) using a lactate-overproducing strain of Escherichia coli and an evolved lactate-polymerizing enzyme, Metab. Eng., 15, 159e166 (2013). 18. Chin, J. W., Khankal, R., Monroe, C. A., Maranas, C. D., and Cirino, P. C.: Analysis of NADPH supply during xylitol production by engineered Escherichia coli, Biotechnol. Bioeng., 102, 209e220 (2009). 19. Lim, S.-J., Jung, Y.-M., Shin, H.-D., and Lee, Y.-H.: Amplification of the NADPHrelated genes zwf and gnd for the oddball biosynthesis of PHB in an E. coli transformant harboring a cloned phbCAB operon, J. Biosci. Bioeng., 93, 543e549 (2002). 20. Zhu, J. and Shimizu, K.: The effect of pfl gene knockout on the metabolism for optically pure D-lactate production by Escherichia coli, Appl. Microbiol. Biotechnol., 64, 367e375 (2004). 21. Zhou, L., Zuo, Z.-R., Chen, X.-Z., Niu, D.-D., Tian, K.-M., Prior, B. A., Shen, W., Shi, G.-Y., Singh, S., and Wang, Z.-X.: Evaluation of genetic manipulation strategies on D-lactate production by Escherichia coli, Curr. Microbiol., 62, 981e989 (2011). 22. Tarmy, E. M. and Kaplan, N. O.: Chemical characterization of D-lactate dehydrogenase from Escherichia coli B, J. Biol. Chem., 243, 2579e2586 (1968). 23. Mat-Jan, F., Alam, K. Y., and Clark, D. P.: Mutants of Escherichia coli deficient in the fermentative lactate dehydrogenase, J. Bacteriol., 171, 342e348 (1989). 24. Yamada, M., Matsumoto, K. I., Nakai, T., and Taguchi, S.: Microbial production of lactate-enriched poly[(R)-lactate-co-(R)-3-hydroxybutyrate] with novel thermal properties, Biomacromolecules, 10, 677e681 (2009). 25. Upare, P. P., Hwang, Y. K., Chang, J.-S., and Hwang, D. W.: Synthesis of lactide from alkyl lactate via a prepolymer route, Ind. Eng. Chem. Res., 51, 4837e4842 (2012).
Please cite this article in press as: Utsunomia, C., et al., Microbial secretion of lactate-enriched oligomers for efficient conversion into lactide: A biological shortcut to polylactide, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.03.002
Microbial secretion of lactate-enriched oligomers for efficient conversion into lactide: A biological shortcut to polylactide Camila Utsunomia,1 Ken’ichiro Matsumoto,1, 2 Sakiko Date,1 Chiaki Hori,1, 2 and Seiichi Taguchi1, 2, * Division of Applied Chemistry, Graduate School of Engineering, Hokkaido University, N13-W8, Kita-ku, Sapporo 060-8628, Japan1 and CREST, JST, 4-1-8 Honcho, Kawaguchi, Saitama 332-0012, Japan2 Received 9 February 2017; accepted 4 March 2017 Available online xxx
Recently, we have succeeded in establishing the microbial platform for the secretion of lactate (LA)-based oligomers (D-LAOs), which consist of D-LA and D-3-hydroxybutyrate (D-3HB). The secretory production of D-LAOs was substantially enhanced by the supplementation of diethylene glycol (DEG), which resulted in the generation of DEG-capped oligomers at the carboxyl terminal (referred as D-LAOs-DEG). The microbial D-LAOs should be key compounds for the synthesis of lactide, an important intermediate for polylactides (PLAs) production, eliminating the costly chemo-oligomerization step in the PLA production process. Therefore, in order to demonstrate a proof-of-concept, here, we attempted to convert the D-LAOs-DEG into lactide via metal-catalyzed thermal depolymerization. As a result, D-LAOs-DEG containing 68 mol% LA were successfully converted into lactide, revealing that the DEG bound to D-LAOs-DEG does not inhibit the conversion into lactide. However, the lactide yield (4%) was considerably lower than that of synthetic LA homooligomers (33%). We presumed that 3HB units in the polymer chain blocked the lactide formation, and therefore, we investigated the LA enrichment in the oligomers. As the results, the combination of an LA-overproducing Escherichia coli mutant (Ddld and DpflA) with the use of xylose as a carbon source exhibited synergistic effect to increase LA fraction in the oligomers up to 89 mol%. The LA-enriched D-LAOs-DEG were converted into lactide with greater yield (18%). These results demonstrated that a greener shortcut route for PLA production can be created by using the microbial D-LAOs secretion system. Ó 2017, The Society for Biotechnology, Japan. All rights reserved. [Key words: Biosynthesis; Secretory production; Bioprocess; Metal-catalyzed reaction; Polyhydroxyalkanoate; Biomass; Polylactic acid; Oligoester]
Polylactides (PLAs) are one of the most successful biobased polyesters with diverse applications due to physical properties, biocompatibility, and processability (1). Currently, biomass resources such as corn, cane sugar, potato starch, and tapioca starch are used as carbon sources for PLA production (2). The conventional process of PLA production from biomass involves (i) a bioprocess for the production of lactic acid from biomass sugars, and (ii) a multistep chemo-process containing the oligomerization of lactic acid to generate lactate (LA) oligomers, depolymerization of LA oligomers into lactide (cyclic dimer of LA), and the polymerization of lactide into high molecular weight PLA via ringopening polymerization (ROP) (3,4) (Fig. 1). Although this is the main industrial route, the multistep chemo-bio process is considered complex and expensive relative to petroleum-based polymers (5). Among the steps, the production of lactide from lactic acid, which comprises the oligomerization of lactic acid and the lactide synthesis from LA oligomers, contributes to 30% of total cost for PLA production (6).
* Corresponding author at: Department of Chemistry for Life Sciences and Agriculture, Faculty of Life Sciences, Tokyo University of Agriculture, 1-1-1 Sakuragaoka, Setagaya-ku, Tokyo 156-8502, Japan. Tel.: þ81 3 5477 2577; fax: þ81 3 5477 2627. E-mail address: [email protected] (S. Taguchi).
Recently, we have discovered the secretory production of D-LAbased oligomers (D-LAOs), which consist of D-LA and D-3hydroxybutyrate (D-3HB), by recombinant Escherichia coli grown on semisynthetic medium using glucose as carbon source (7). This bacterium expresses an engineered polyhydroxyalkanoate (PHA) synthase (PhaC), designated as D-specific LA-polymerizing enzyme (LPE). This enzyme had been formerly developed for the production of LA-based polyesters (8,9). The production of D-LAOs was remarkably increased by supplementing diethylene glycol (DEG) in the bacterial cultivation. The transference of the polymer chain took place from the LPE to DEG via chain transfer (CT) reaction when DEG was added in the culture. The high frequency of DEGmediated CT reaction drastically decreased the molecular weight of the intracellularly accumulated LA-based polyesters, and also enhanced the production and secretion efficiency of D-LAOs. The oligomers synthesized under DEG supplementation were nearly fully conjugated with DEG at the carboxyl terminal (termed as DLAOs-DEG). The aim of the present study is to verify the feasibility of the DLAOs-DEG on the conversion into lactide, since this is a key reaction to construct a shortcut route in the process of poly(D-lactide) (PDLA) production (Fig. 1). By establishing this new shortcut route, the laborious purification of lactic acid from the microbial culture broth (10) and the lactic acid oligomerization can be eliminated, increasing the sustainability of the material and the cost effectiveness of PDLA production. Moreover, the production of optically
1389-1723/$ e see front matter Ó 2017, The Society for Biotechnology, Japan. All rights reserved. http://dx.doi.org/10.1016/j.jbiosc.2017.03.002
Please cite this article in press as: Utsunomia, C., et al., Microbial secretion of lactate-enriched oligomers for efficient conversion into lactide: A biological shortcut to polylactide, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.03.002
2
UTSUNOMIA ET AL.
J. BIOSCI. BIOENG.,
Bio-process D-LAOs O
Biomass
H
O
O O
O
D-LA
Conventional Route
Microbial platform LPE Secretory New shortcut route production
D-3HB
O
O HO
OH
Polycondensation
D-LAOs
Depolymerization
O
O O
O
ROP
O
n
Lactide
D-Lactic acid
OH
DEG
y
x
O
Polylactides
Chemo-process FIG. 1. Scheme of the conventional route and new shortcut route for PDLA production. In the conventional process, D-lactate (LA) oligomers are chemically prepared from purified Dlactic acid, which is generated by bacterial fermentation, via a polycondensation reaction. In the shortcut route proposed in this study, highlighted in red color, D-LA-based oligomers (D-LAOs) are directly secreted by bacteria from renewable biomass. LPE, lactate-polymerizing enzyme; ROP, ring-opening polymerization; DEG, diethylene glycol; D-3HB, D-3-hydroxybutyrate. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
pure D-LAO is beneficial in view of the superior thermal properties of poly(L-lactide) (PLLA)/PDLA stereocomplex (11,12), and the higher production cost of D-LA compared to L-LA. Therefore, DLAOs-DEG were used as substrates for conversion into lactide via metal-catalyzed reaction. To examine the effect of the terminal DEG in D-LAOs-DEG on the lactide formation, free-form D-LAOs (synthesized without DEG addition) were used for comparison. Moreover, to evaluate the influence of 3HB units in D-LAOs-DEG on the lactide synthesis efficiency, we attempted to synthesize LAenriched oligomers through metabolic and fermentation engineering approaches.
MATERIALS AND METHODS Bacterial strains and plasmids E. coli BW25113 (13), and the dual-gene knockout mutant (Dpfla and Ddld) JWMB1 (14) were used as the host strains. The expression vector pTV118NpctphaC1Ps(ST/FS/QK)AB harboring pct, phaC1Ps(ST/FS/ QK), phaA, and phaB genes (9), was used for the production of D-LAOs. Culture conditions in test tubes Cultivations for D-LAOs-DEG production were carried out in 10 mL glass test tubes containing LuriaeBertani (LB) medium (1.7 mL) with 20 g/L glucose or xylose and 100 mg/L ampicillin at 30 C for 48 h with reciprocal shaking at 180 rpm. The cultivations were performed with the supplementation of 5% DEG (v/v). For observing the aeration effects on D-LAOs production, the cultivations of the dual-gene knockout mutant with xylose as carbon source, were also performed using 2.5, 3.4, and 5.1 mL medium. Shake flask cultures Free-form D-LAOs were produced in shake flask cultivation. Seed culture of recombinant E. coli was prepared using 2 mL LB medium containing 100 mg/L ampicillin in 10 mL glass test tubes and cultured at 30 C for 12 h with reciprocal shaking at 180 rpm. One milliliter of the seed culture was then transferred into 100 mL LB medium containing 20 g/L glucose, and 100 mg/L ampicillin in a 500 mL shake flask and cultured at 30 C for 48 h with reciprocal shaking at 120 rpm. Measurement of extracellular D-LAOs The cell-free culture supernatant was analyzed before and after HCl treatment. HCl was added to the supernatant at a final concentration of 2.0 M and incubated at 100 C overnight to hydrolyse the D-LAOs. Afterward, the hydrolysate was neutralized with 2.0 M NaOH. The estimation of D-LAOs in the culture medium was determined by liquid chromatography-mass spectrometry (LC-MS) (LCMS-8030, Shimadzu, Japan) based on the difference of lactic acid and 3-hydroxybutyric acid concentrations in the samples after and before HCl treatment, as described previously (7). Extraction of D-LAOs from the culture supernatant D-LAOs were concentrated from the cell-free culture supernatant by two-phase extraction using chloroform (CHCl3), with modifications on the previously established method (7). The extraction was performed by adding 1 volume of CHCl3 to 1 volume of supernatant and mixing vigorously. After the separation of CHCl3 and water layers, the CHCl3 phase was transferred to a new test tube. The extraction was performed three times. To remove excess DEG, lactic acid/3-hydroxybutyric acid monomers, and short oligomers, 1 volume of 0.9% NaCl solution pH 8 was added to the
resulting CHCl3 fraction. The washing step was repeated twice. The molecular weight distribution of extracted oligomers was determined by electrospray ionization-time-of-flight-mass spectrometry (ESI-TOF-MS), as described in the literature (7). Lactide synthesis The synthesis of lactide occurs via thermal depolymerization of the LA oligomers via a metalecatalyzed backbiting reaction of the eOH end groups (15). Approximately 40 mg of vacuum dried extracted D-LAOs containing 68, 78, and 89 mol% LA, were individually weighted together with 40 mg zinc oxide (ZnO, Kanto Chemical, Japan). The sample bottle containing D-LAOs and catalyst was placed inside a rotary type Sibata GTO-350RD glass oven (Sibata Scientific Technology, Japan). The reaction system was heated and kept at 180 C while being rotated in a circular motion for 1 h under vacuum. The vaporized lactide was condensed into a bottle cooled on ice, and recovered in chloroform. Watersoluble synthetic L-LA oligomers (Glart, Japan) and free-form D-LAO, were also converted into lactide as experimental controls. The lactide yield (%, 2 [mmol lactide/mmol initial oligomeric LA]) was calculated based on the amounts of produced lactide and initial oligomeric LA quantified using the 1H NMR analysis. 1 1 H NMR of D-LAOs and lactide H NMR of extracted oligomers and generated lactides were recorded in CDCl3 with tetramethylsilane as the internal reference using a JEOL JNM-ECS400 spectrometer (JEOL, Japan) at 400 MHz. Benzoic acid (Wako, Japan) was used as an internal standard to quantify oligomers and lactide based on the integral area of the methyl group (eCH3); 45 excitation pulse was used and relaxation delay was set to 10 s.
RESULTS AND DISCUSSION D-LAOs-DEG could be converted into lactide via metalcatalyzed backbiting reaction The synthesis of lactide from the extracted D-LAOs-DEG was undertaken to verify the applicability of the biosynthesized D-LAOs-DEG for subsequent PLA production. These oligomers were produced in the test tube by the wild-type strain BW25113 grown on glucose with 5% DEG supplementation. After extraction of the culture supernatant with chloroform, D-LAOs-DEG with 68 mol% LA and degree of polymerization (DP) of approximately trimer to 7 mer were recovered in chloroform phase (designated as extracted D-LAOsDEG). As controls, extracted free-form D-LAOs (61 mol% LA, DP w trimer to 16 mer) and synthetic L-LA homo-oligomers (L-LAOs) (100 mol% LA, DP w trimer to 14 mer), were also used as substrates for lactide synthesis. For generating lactides, the LA oligomers were heated with zinc oxide as catalyst, and the vaporized fraction was recovered by condensation in bottle 3 (Fig. 2A) and subjected to 1H NMR analysis (Fig. 2B). The 1H NMR spectrum (d in ppm) of the sample generated from D-LAOs-DEG exhibited signals at 5.0 ppm (1H, q, A) and 1.7 ppm (3H, d, B), which were identical to those of the standard Dlactide. The lactide generated from free-form D-LAOs also exhibited
Please cite this article in press as: Utsunomia, C., et al., Microbial secretion of lactate-enriched oligomers for efficient conversion into lactide: A biological shortcut to polylactide, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.03.002
VOL. xx, 2017
MICROBIAL LA-OLIGOMERS FOR SHORTCUT IN PLA PRODUCTION
A
Bottle 2
Bottle 3 Condensation bulb
Heater tube
LAO(a) + ZnO (d)
3
Lactide(e) (c)
(b)
Vacuum pump
Sample bulb Bottle 1
Cooling unit
B
A O
O
H
O CH3
H
H3 B CH O
D-Lactide standard
D-Lactide from D-LAOs-DEG
1.7
5.0
B
A 5.0
4.0
3.0 ppm
2.0
1.0
FIG. 2. Lactide synthesis from extracted D-LAOs-DEG. (A) Scheme of the lactide synthesis apparatus. The location of the initial oligomer in bottle 1 (a), condensed oligomer in bottle 2 (b), condensed oligomer in bottle 3 (c), residual oligomer in bottle 1 (d), and lactide (e), in the lactide synthesis apparatus, is represented here. (B) 1H NMR spectrum of crude lactide synthesized from D-LAOs-DEG 68 mol% LA. (C) Lactide yields from free-form D-LAOs, D-LAOs-DEG, and synthetic L-LAOs as substrates. Refer to Fig. 2A for obtaining details about samples localization in the lactide synthesis apparatus.
Please cite this article in press as: Utsunomia, C., et al., Microbial secretion of lactate-enriched oligomers for efficient conversion into lactide: A biological shortcut to polylactide, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.03.002
4
UTSUNOMIA ET AL.
J. BIOSCI. BIOENG.,
TABLE 1. Secretory production of D-LAOs-DEG from glucose and xylose by engineered BW25113 and dual-gene knockout mutant (Dpfla and Ddld) JWMB1. Strain
Carbon source
BW25113 Glucose Xylose JWMB1 Glucose Xylose a b
CDW (g/L)
4.4 5.3 2.7 4.2
0.2 0.1 0.5 0.2
D-LAOs-DEG
Extracted D-LAOs-DEG
Oligomer LA Recovery (%) Oligomer LA (g/L) (mol%)b (g/L) (mol%)a 6.6 4.6 5.9 8.1
1.2 1.1 0.7 2.9
90 97 78 97
4 2 5 1
3.5 1.6 3.0 4.5
68 62.5 76.5 83
53 35 51 56
Determined on the basis of LC/MS analysis. Determined on the basis of 1H NMR analysis.
the same resonances as the standard D-lactide (Supplementary Fig. S1). Synthetic L-LAOs were converted into lactides by using the same procedure. These results proved that the D-LAOs-DEG can be catalytically converted into lactide as well as free-form D-LAOs. The conclusion is in agreement with the fact that lactide is formed via a backbiting reaction involving the hydroxyl terminal of the oligomer as the active site (16), thus, the carboxyl-terminal structure of D-LAOs should not block the lactide formation. The products condensed in bottle 2 and the residual products in bottle 1 (Fig. 2A) were also analyzed by 1H NMR. Unreacted oligomers were found in bottles 1, 2, and 3 after the depolymerization reaction of D-LAOsDEG, free-form D-LAOs, and synthetic L-LAOs (Fig. 2C). In addition, the D-LAOs condensed in bottle 2 had higher 3HB fraction than the oligomers recovered in bottle 3 (Fig. 2C). The oligomers trapped in bottle 3 should possess lower boiling temperature compared to those in bottle 2, because temperature in bottle 3 is lower than that in bottle 2. LA fraction in D-LAOs-DEG was increased by using xylose as carbon source and a dual-gene knockout mutant as a host strain Despite the microbial D-LAOs were shown to serve as precursor for lactide, the lactide formation over the substrate consumption obtained from free-form D-LAOs (25%) and D-LAOsDEG (13%) were substantially lower than that of synthetic L-LAOs (77%) (Fig. 2C). This result is presumably attributable to the presence of 3HB units in the D-LAOs, which could act as a stopper of the backbiting reaction and decrease the conversion rate of the oligomers into lactide. In addition, the presence of 3HB units decreases the frequency of LAeLA dyad in the oligomers, which is essential for lactide formation. Accordingly, enhancing the LA fraction in the oligomers may further increase the efficiency of the lactide synthesis. Based on this idea, we attempted to produce LA-enriched oligomers by optimizing cultivation conditions such as carbon source, bacterial strain, and culture aeration. The effect of carbon source, glucose and xylose, on D-LAOs-DEG production and their monomer composition was investigated (Table 1). The total production of D-LAOs-DEG in xylose culture was 4.6 1.1 g/L, which was lower than that obtained from glucose (6.6 1.2 g/L). Nevertheless, the D-LAOs-DEG in xylose included a higher LA fraction (97 2 mol% LA) relative to the oligomers secreted in the glucose cultivation (90 4 mol% LA). These results demonstrated that xylose is indeed effective for enhancing the LA fraction in D-LAOs-DEG, although the total production of D-LAOsDEG was lower compared to the glucose culture. These findings are consistent with the previous report regarding the production of high-molecular-weight P(LA-co-3HB) (17). The LA enrichment in DLAOs caused by the consumption of xylose might be related to the different capacities of regenerating NADH and NADPH in the metabolism routes of xylose and glucose (18,19). As a consequence of that, the xylose metabolism is thought to have a higher LA units supplying rate than glucose (17). In order to address the influence of the host strain, D-LAOs-DEG were produced by the dual-gene knockout mutant (Dpfla and Ddld) JWMB1 from glucose or xylose in comparison with the parent strain
TABLE 2. Secretory production of D-LAOs-DEG from xylose by engineered E. coli JWMB1 under microaerobic condition. Cultivation volume (mL)
1.7 2.5 3.4 5.1
CDW (g/L)
D-LAOs-DEG Oligomer (g/L)
4.2 3.9 3.6 3.0
0.2 0.1 0.0 0.1
8.1 7.4 7.7 6.3
Extracted D-LAOs-DEG
LA (mol%)
2.9 1.6 1.3 1.1
97 97 93 92
Oligomer (g/L)
LA (mol%)
Recovery (%)
4.5 4.8 5.0 3.3
83 84 89 85
53 65 65 52
1 2 2 2
BW25113 (Table 1). The deletion of pflA is known to eliminate formate formation from acetyl-CoA, channeling the flux toward lactic acid (20). The dld mutation prevents lactic acid oxidation into pyruvate, improving intracellular availability of lactic acid (21). When glucose was used as a carbon source, JWMB1 produced similar amount of D-LAOs-DEG compared to BW25113, and LA fraction in the D-LAOs-DEG obtained using JWMB1 (78 mol%) was lower than that from BW25113 (90 mol%). In contrast, with the use of xylose, the D-LAOs-DEG production (8.1 2.9 g/L, 97 1 mol% LA) was higher than that of BW25113 (4.6 1.1 g/L, 97 2 mol% LA). This result indicates that there is a synergy between the dualmutation and use of xylose as a carbon source to increase both the production and LA fraction of the oligomers. Microaerobic conditions increased the recovery of extracted D-LAOs-DEG In order to further upregulate the LA fraction in DLAOs-DEG, the oligomers were produced under microaerobic conditions, which are well-known to promote lactic acid production (22,23) and LA fraction enrichment in LA-based polyester (24). To control the microaerobic conditions using simplified method, the volume of culture medium in a test tube was increased from 1.7 to 5.1 mL. As the result, recovery of D-LAOs-DEG in chloroform phase was improved by using the microaerobic conditions (Table 2). Among them, the highest LA fraction in extracted D-LAOs-DEG was 89 mol% LA (Fig. 3). The total oligomer productions and their respective LA fractions, were, however, rather decreased (Table 2). These results indicated that the hydrophobicity of the oligomers, which is determined by the molecular weight and monomer composition of the oligomers, tends to increase under microaerobic conditions. Here, it should be noticed that the molecular weight of extracted D-LAOs-DEG (DP w trimer to 7mer) was not significantly altered by using microaerobic conditions (Supplementary Fig. S2), but the amount of extracted D-LAOs-DEG was increased (Table 2). This phenomenon is probably due to the fact that the molecular weight of secreted oligomers is limited by the solubility of oligomer molecules in water.
3 H
O 2
2
O 1
x
D-3HB
A
O
O 1 O
C O
B y
OH D
DEG
D-LA LA(2)
DEG(A,C)
DEG(B)
DEG(D)
LA(1) 3HB(2)
3HB(3)
3HB(1) 5.0
4.0
3.0 ppm
2.0
1.0
FIG. 3. 1H NMR spectrum of D-LAOs-DEG containing 89 mol% LA.
Please cite this article in press as: Utsunomia, C., et al., Microbial secretion of lactate-enriched oligomers for efficient conversion into lactide: A biological shortcut to polylactide, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.03.002
VOL. xx, 2017
MICROBIAL LA-OLIGOMERS FOR SHORTCUT IN PLA PRODUCTION
5
JSPS Kakenhi [grant numbers 26660080 to S.T. and 26281043 to K.M.].
References
FIG. 4. Variation of the lactide yield (%) as a function of the LA fraction in extracted DLAOs-DEG. As a control, lactide was synthesized from synthetic L-LAOs.
Lactide yield was enhanced by LA-enrichment in oligomers The influence of the LA fraction in D-LAOs on the efficiency of lactide synthesis was evaluated using extracted LAenriched D-LAOs-DEG containing up to 89 mol% LA with nearly the same DP of approximately trimer to 7-mer. The LA fraction enrichment in D-LAOs-DEG significantly increased the lactide yields up to 18% (Fig. 4). Nevertheless, the lactide yield from LAenriched D-LAOs-DEG was still lower than that obtained from synthetic L-LAOs. Further increase in LA fraction of D-LAOs-DEG seems to be necessary to achieve more efficient lactide synthesis. In addition, an important factor determining the efficiency of lactide synthesis should be the molecular weight of D-LAOs-DEG. During lactide synthesis, significant amount of oligomers was condensed in bottles 2 and 3 (Fig. 2A and Supplementary Table S1), suggesting that short oligomers were lost by vaporization during heating (25). In the industrial process, oligomers with molecular weights around 400e2500 g/mol, which correspond to DP of approximately 5-mer to 34-mer, are used in the synthesis of lactide (6). Therefore, increasing the molecular weight of the secreted oligomers should be effective to improve the conversion efficiency of D-LAOs-DEG into lactide. In conclusion, here, we demonstrated the successful lactide synthesis by using D-LAOs-DEG secreted by engineered E. coli expressing evolved LPE. Notably, the backbiting reaction-based lactide conversion was not inhibited by the presence of DEG at the carboxyl-terminal of D-LAOs. This finding provides us a proofof-concept for establishing a shortcut route for PDLA production. Moreover, the improvement in the lactide synthesis efficiency was achieved via LA enrichment in the oligomers by the combination of xylose as carbon source with the use of a dual-gene knockout mutant (Dpfla and Ddld) strain. Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.jbiosc.2017.03.002. ACKNOWLEDGMENTS We thank Dr. Toshifumi Satoh, Hokkaido University, for helpful discussions about lactide synthesis, and Dr. Tomoki Erata, Hokkaido University, for his useful comments on NMR analysis. This study was partially supported by CREST, JST (to S.T.), ALCA, JST (to K.M.),
1. Liu, M.-J., Chen, S.-C., Yang, K.-K., and Wang, Y.-Z.: Biodegradable polylactide based materials with improved crystallinity, mechanical properties and rheological behaviour by introducing a long-chain branched copolymer, RSC Adv., 5, 42162e42173 (2015). 2. Shen, L., Worrell, E., and Patel, M.: Present and future development in plastics from biomass, Biofuels Bioprod. Biorefin., 4, 25e40 (2010). 3. Gupta, A. P. and Kumar, V.: New emerging trends in synthetic biodegradable polymersepolylactide: a critique, Eur. Polym. J., 43, 4053e4074 (2007). 4. Slomkowski, S., Penczek, S., and Duda, A.: Polylactidesdan overview, Polym. Adv. Technol., 25, 436e447 (2014). 5. Lasprilla, A. J., Martinez, G. A., Lunelli, B. H., Jardini, A. L., and Filho, R. M.: Poly-lactic acid synthesis for application in biomedical devices e a review, Biotechnol. Adv., 30, 321e328 (2012). 6. Van Wouwe, P., Dusselier, M., Vanleeuw, E., and Sels, B.: Lactide synthesis and chirality control for polylactic acid production, ChemSusChem, 9, 907e921 (2016). 7. Utsunomia, C., Matsumoto, K., and Taguchi, S.: Microbial secretion of Dlactate-based oligomers, ACS Sustain. Chem. Eng., 5, 2360e2367 (2017). 8. Taguchi, S., Yamada, M., Matsumoto, K., Tajima, K., Satoh, Y., Munekata, M., Ohno, K., Kohda, K., Shimamura, T., and Kambe, H.: A microbial factory for lactate-based polyesters using a lactate-polymerizing enzyme, Proc. Natl. Acad. Sci. USA, 105, 17323e17327 (2008). 9. Yamada, M., Matsumoto, K., Shimizu, K., Uramoto, S., Nakai, T., Shozui, F., and Taguchi, S.: Adjustable mutations in lactate (LA)-polymerizing enzyme for the microbial production of LA-based polyesters with tailor-made monomer composition, Biomacromolecules, 11, 815e819 (2010). 10. Li, Q.-Z., Jiang, X.-L., Feng, X.-J., Wang, J.-M., Sun, C., Zhang, H.-B., Xian, M., and Liu, H.-Z.: Recovery processes of organic acids from fermentation broths in the biomass-based industry, J. Microbiol. Biotechnol., 26, 1e8 (2016). 11. Ikada, Y., Jamshidi, K., Tsuji, H., and Hyon, S. H.: Stereocomplex formation between enantiomeric poly(lactides), Macromolecules, 20, 904e906 (1987). 12. Tsuji, H.: Poly(lactide) stereocomplexes: formation, structure, properties, degradation, and applications, Macromol. Biosci., 5, 569e597 (2005). 13. Baba, T., Ara, T., Hasegawa, M., Takai, Y., Okumura, Y., Baba, M., Datsenko, K. A., Tomita, M., Wanner, B. L., and Mori, H.: Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection, Mol. Syst. Biol., 2 (2006). 2006.0008. 14. Nduko, J. M., Matsumoto, K., Ooi, T., and Taguchi, S.: Enhanced production of poly(lactate-co-3-hydroxybutyrate) from xylose in engineered Escherichia coli overexpressing a galactitol transporter, Appl. Microbiol. Biotechnol., 98, 2453e2460 (2014). 15. Yoo, D. K., Kim, D., and Lee, D. S.: Synthesis of lactide from oligomeric PLA: effects of temperature, pressure, and catalyst, Macromol. Res., 14, 510e516 (2006). 16. Nishida, H., Mori, T., Hoshihara, S., Fan, Y., Shirai, Y., and Endo, T.: Effect of tin on poly(L-lactic acid) pyrolysis, Polym. Degrad. Stab., 81, 515e523 (2003). 17. Nduko, J. M., Matsumoto, K., Ooi, T., and Taguchi, S.: Effectiveness of xylose utilization for high yield production of lactate-enriched P(lactate-co-3hydroxybutyrate) using a lactate-overproducing strain of Escherichia coli and an evolved lactate-polymerizing enzyme, Metab. Eng., 15, 159e166 (2013). 18. Chin, J. W., Khankal, R., Monroe, C. A., Maranas, C. D., and Cirino, P. C.: Analysis of NADPH supply during xylitol production by engineered Escherichia coli, Biotechnol. Bioeng., 102, 209e220 (2009). 19. Lim, S.-J., Jung, Y.-M., Shin, H.-D., and Lee, Y.-H.: Amplification of the NADPHrelated genes zwf and gnd for the oddball biosynthesis of PHB in an E. coli transformant harboring a cloned phbCAB operon, J. Biosci. Bioeng., 93, 543e549 (2002). 20. Zhu, J. and Shimizu, K.: The effect of pfl gene knockout on the metabolism for optically pure D-lactate production by Escherichia coli, Appl. Microbiol. Biotechnol., 64, 367e375 (2004). 21. Zhou, L., Zuo, Z.-R., Chen, X.-Z., Niu, D.-D., Tian, K.-M., Prior, B. A., Shen, W., Shi, G.-Y., Singh, S., and Wang, Z.-X.: Evaluation of genetic manipulation strategies on D-lactate production by Escherichia coli, Curr. Microbiol., 62, 981e989 (2011). 22. Tarmy, E. M. and Kaplan, N. O.: Chemical characterization of D-lactate dehydrogenase from Escherichia coli B, J. Biol. Chem., 243, 2579e2586 (1968). 23. Mat-Jan, F., Alam, K. Y., and Clark, D. P.: Mutants of Escherichia coli deficient in the fermentative lactate dehydrogenase, J. Bacteriol., 171, 342e348 (1989). 24. Yamada, M., Matsumoto, K. I., Nakai, T., and Taguchi, S.: Microbial production of lactate-enriched poly[(R)-lactate-co-(R)-3-hydroxybutyrate] with novel thermal properties, Biomacromolecules, 10, 677e681 (2009). 25. Upare, P. P., Hwang, Y. K., Chang, J.-S., and Hwang, D. W.: Synthesis of lactide from alkyl lactate via a prepolymer route, Ind. Eng. Chem. Res., 51, 4837e4842 (2012).
Please cite this article in press as: Utsunomia, C., et al., Microbial secretion of lactate-enriched oligomers for efficient conversion into lactide: A biological shortcut to polylactide, J. Biosci. Bioeng., (2017), http://dx.doi.org/10.1016/j.jbiosc.2017.03.002