Process Biochemistry 45 (2010) 153–163

Contents lists available at ScienceDirect

Process Biochemistry
journal homepage: www.elsevier.com/locate/procbio

Microbial biodegradable plastic production from a wheat-based biorefining

strategy
Y. Xu a, R.-H. Wang a, A.A. Koutinas a,b,*, C. Webb a
a
Satake Centre for Grain Process Engineering, School of Chemical Engineering and Analytical Science, The University of Manchester, PO Box 88, Manchester M60 1QD, United Kingdom
b
Department of Food Science and Technology, Agricultural University of Athens, Iera Odos 75, 118 55 Athens, Greece

A R T I C L E I N F O A B S T R A C T

Article history: Restructuring the current fermentation and recovery practices employed for the production of
Received 17 September 2008 polyhydroxyalkanoates is essential for the commercialisation of environmentally benign and cost
Received in revised form 28 August 2009 competitive biodegradable plastics. This study presents the potential of a wheat-based biorefinery for
Accepted 2 September 2009
the production of poly(3-hydroxybutyrate) (PHB). Fed-batch bioconversions using Wautersia eutropha
growing on wheat-derived media led to the production of 162.8 g/l PHB. A high PHB to total dry weight
Keywords: (TDW) yield of 93% (w/w) was achieved due to microbial autolysis at the end of fermentation. Images of
Wheat-based biorefinery
bacterial cells taken with a Transmission Electron Microscope (TEM) indicated the potential of bacterial
Wautersia eutropha
autolysis as a mean to shorten downstream processing for PHB purification. The consumption of amino
Poly(3-hydroxybutyrate)
acids and peptides derived from wheat gluten hydrolysis resulted in a high glucose to PHB conversion
yield of 0.47 g/g. The respective yield regarding the amount of wheat used for the production of enzymes
and PHB was around 0.3 g PHB/g wheat, which corresponds to 82.8% of the maximum theoretical
conversion yield. The productivity achieved was around 0.9 g/l h. Fermentations carried out on wheat-
derived media and media formulated with various commercial sources of nutrients (glucose, yeast
extract, soy-protein acid hydrolysate, casein hydrolysates, corn steep liquor and various inorganic
chemicals) showed that the proposed wheat-based biorefinery strategy enhanced PHB production.
ß 2009 Elsevier Ltd. All rights reserved.

1. Introduction purified carbon sources (i.e. glucose, sucrose), commercial pure

Polyhydroxybutyrate (PHB) belongs to the general group of
polyhydroxyalkanoates (PHAs) and has numerous potential uses,
such as a biodegradable substitute for petroleum-derived plastics,
biocomposite production, speciality biopolymer for medical appli-
cations and a source of the platform molecule 3-hydroxybutyric acid
[1–4]. However, industrial PHB production is hindered by unfavour-
able PHB production costs [5,6]. Low cost and environmentally
friendly PHB production is dependent on the restructuring of
traditional fermentation and recovery technologies [7,8].
Reduction of microbial PHB production costs could be achieved
by utilising renewable raw materials, such as potato processing
waste, pulp fiber sludge, whey, green grass, silage, and cereals [9–
13]. Solaiman et al. [14] presented a review on microbial PHA
production from agricultural feedstocks and co-products, such as
intact triacylglycerols (vegetable oils and animal fats), dairy whey,
molasses, and meat-and-bone meal. Targeting the substitution of
* Corresponding author at: Department of Food Science and Technology,
Agricultural University of Athens, Iera Odos 75, 118 55 Athens, Greece.
Tel.: +30 210 529 4729; fax: +30 210 529 4729.
E-mail address: akoutinas@aua.gr (A.A. Koutinas).
starch is one of the carbon sources used in many publications for
microbial PHB production [15–19]. Yu et al. [16] showed that PHB
fermentations on starch hydrolysates using a recombinant
Escherichia coli strain resulted in higher PHB concentration and
PHB content as compared to fermentations carried out on glucose.
A PHB concentration of 167.6 g/l at a productivity of 3.05 g/l h has
been reported by Yu et al. [16] when a semi-defined medium with
starch hydrolysate as carbon source was used. Quillaguaman et al.
[17] reported the utilisation of starch hydrolysates as carbon
source in fermentations carried out with Halomonas boliviensis to
produce a PHB content of 35%. Huang et al. [18] used extruded rice
bran as carbon source in a fed-batch fermentation carried out by
Haloferax mediterranei to produce a total dry weight of 140 g/l with
a PHA content of 55.6%. Commercial protein hydrolysates (i.e. yeast
extract, casein hydrolysates) and by-products of agro-industrial
origin (i.e. corn steep liquor) have been employed as sources of
nitrogen. Several studies have stressed the potential of amino acids
and peptides to enhance microbial PHB production [20–25].
Bormann et al. [23] reported the production of 65 g/l total dry
weight with a PHB content of 60–80% when 20–30 g/1 casein
peptone or casamino acids were used as sole sources of nitrogen. In
the majority of these studies, mineral supplementation is achieved
by the addition of inorganic chemicals.

1359-5113/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.procbio.2009.09.001
154 Y. Xu et al. / Process Biochemistry 45 (2010) 153–163
The establishment of bioprocesses as a viable alternative to
(petro)chemical routes for chemical production is dependent on
the development of integrated biomass-based biorefineries. A
wheat-based biorefining strategy has been proposed for the
production of a spectrum of chemicals with significant economic
improvements [13,26–32]. Koutinas et al. [33] presented various
alternative schemes of the generic biorefining concept, depend-
ing on the targeted product (i.e. bioethanol, PHB, succinic acid).
Fig. 1 presents the proposed process in the case of microbial PHB
production. On-site fungal bioconversions by Aspergillus awamori
is used for the production of wheat hydrolysates by exploiting
the enzymatic consortia (rich mainly in amylolytic enzymes)
produced by the fungus to hydrolyse mainly the starch content in
wheat (gluten and phytic acid are also partly hydrolysed) [34].
Fungal autolysis is used for the production of nutrient supple-
ments [35] that can be mixed with wheat hydrolysates to
formulate nutrient-complete media for a range of microbial
bioconversions. The optimised combination of gluten hydrolysis
and fungal autolysis would generate mixtures of amino acids and
peptides that would not only provide a nitrogen source but also
an additional carbon source. In conventional fermentation
practices, the protein fraction in cereal grains is rarely hydro-
lysed to enrich fermentation media in amino acids and peptides,
which could improve microbial fermentations significantly.
Fouache et al. [36] demonstrated that the use of gluten
hydrolysates as nutrient supplements for lactic acid bioconver-
sions resulted in higher productivities when compared against
yeast extracts and corn steep liquor. Wheat hydrolysates and
fungal autolysates would also contribute vitamins, minerals and
trace elements.
The scheme presented in Fig. 1 offers many advantages as
compared to conventional fermentation and recovery strategies for
PHB production. Wheat components (starch, gluten, bran) are used
for the production of at least three end-products (i.e. PHB,
recombinant proteins from surplus gluten and various value-
added products from bran-rich pearlings). Wheat fractionation
steps generate streams with optimum quantities of starch, gluten
Fig. 1. Proposed wheat-based biorefinery for PHB production.
and bran. A stream containing bran-rich pearlings could be used for
the production of various products such as ferulic acid or
arabinoxylans. Starch- and gluten-rich streams are used for the
production of hydrolysates that are subsequently used for the
production of PHB and recombinant proteins. Shake flask cultures
of recombinant E. coli carried out in both enriched LB and gluten
hydrolysates containing 2 g/l of glucose resulted in the production
of similar intracellular SUMO-EAK16 fusion protein (30–35% of the
total bacterial protein) [37]. Future research should focus on the
identification and optimisation of recombinant protein production
from gluten hydrolysates. The combined production of commodity
(i.e. biodegradable plastics) and speciality (i.e. pharmaceuticals)
products in the same biorefinery would improve process
economics and add value to the initial raw material.
Fungal fermentations by Aspergillus oryzae provide crude
enzyme-containing solutions (rich mainly in proteolytic enzymes)
that could enhance gluten hydrolysis and W. eutropha cell
disruption to facilitate PHB purification. It should be stressed that
A. oryzae is more efficient in protein hydrolysis, whereas A.
awamori in starch hydrolysis. In addition, preliminary results (not
published yet) showed that crude enzyme-containing solutions
from A. oryzae solid state fermentations led to the disruption of 70%
(w/w) of residual cell weight (non-PHB cell mass) increasing the
purity of PHB to 96% (w/w). These steps facilitate the creation of an
integrated biorefining concept that leads to the production of
many added-value products, targets diversified market outlets,
reduces waste production through recycling of nutrients, and
exploits all the components of a renewable resource.
Koutinas et al. [13] showed in shake flask fermentations that
wheat could be used as the sole source of nutrients for microbial
PHB production. The main aim of this study was to demonstrate
that efficient PHB production could be achieved in a bioreactor by
using the wheat-based biorefinery presented in Fig. 1. The nutrient
composition (glucose, free amino nitrogen, total nitrogen, phos-
phorus) in the wheat-derived media varied by mixing different
amounts of wheat hydrolysates and fungal autolysates. The
superiority of the wheat-derived feedstock against conventional
 Y. Xu et al. / Process Biochemistry 45 (2010) 153–163
media based on glucose and commercial nutrient supplements has
been elucidated via appropriate comparative studies.
2. Materials and methods
2.1. Microorganisms
An industrial strain of Aspergillus awamori 2B.361 U2/1 (ABM Chemicals Ltd.,
Woodley, Cheshire, UK) was utilised in fungal fermentations to produce the
enzymatic consortia for hydrolysis of wheat macromolecules. Fungal cultures were
stored at 4 8C in the form of spores in slopes containing solid medium made of 50 g/l
whole wheat flour and 20 g/l agar.
PHB bioconversions were conducted with W. eutropha NCIMB 11599 (formerly
classified as Ralstonia eutropha). W. eutropha is also a heterotypic synonym of
Cupriavidus necator [38]. Lyophilized bacterial cells were reactivated at 30 8C for
24 h in a growth medium that contained MRS broth (Oxoid) and glucose (5 g/l).
Bacterial cultures were stored at 4 8C in slopes containing 5 g/l glucose, 10 g/l yeast
extract and 20 g/l agar. Fermentation inocula were prepared by transferring
bacteria cells using a wire loop into liquid medium containing 5 g/l glucose and
10 g/l yeast extract.
2.2. Commercial culture media
Protein hydrolysates purchased from commercial catalogues were yeast extract
(YE), corn steep liquor (CSL), soy-protein acid hydrolysate (SH) and casein
hydrolysate (CH). YE, CSL and SH were purchased from Sigma and CH was
purchased from Fluka.
2.3. Processing of wheat and fungal fermentations
Wheat kernels from a soft variety of Consort (Fisher Seed and Grain Limited,
Cranswick, East Yorkshire, UK) were pearled for 20 s using a Satake Abrasive Test
Mill (model TM05) to remove up to 3% of the kernel. Pearled grains were milled into
flour using a hammer mill (Falling Number AB) with a screen size of 500 mm. Bran-
free (endosperm-rich) wheat flour (BFWF) was produced by milling pearled wheat
grains through a Buhler laboratory mill. The compositions of whole wheat flour,
pearled wheat flour and BFWF have been presented in a previous publication [34].
Fungal fermentations were carried out on 85 g/l (on a dry basis, db) initial
pearled wheat flour concentration. Detailed description of flour media preparation,
bioreactor design and experimental results from fungal fermentations has been
presented in previous publications [27,34]. The broth from fungal fermentations
was collected aseptically after 120 h and the fermentation solids were separated
aseptically from the liquid by vacuum filtration (Whatman No. 1).
2.4. Production of wheat-based fermentation media
The wheat-derived media used in W. eutropha bioconversions for PHB
production were mixtures of wheat hydrolysates and fungal autolysates. Detailed
description of individual experiments for wheat flour hydrolysis and fungal
autolysis has been presented in previous publications [13,34,35]. In this study,
wheat flour hydrolysates were produced in a 1 l glass reaction vessel by mixing the
filtrate from a A. awamori fermentation with various concentrations of BFWF (200–
350 g/l, db). Higher wheat starch to glucose (90–97%) conversion efficiencies were
achieved in this study as compared to the results presented by Koutinas et al. [34]
due to mechanical agitation with a paddle mixer used during flour hydrolysis. At the
end of the reaction (around 24 h), remaining solids were removed by vacuum
filtration (Whatman No. 1). No starch was detected in the remaining solids after
wheat flour hydrolysis.
The methodology described by Koutinas et al. [13] was followed for the
production of fungal autolysates. At the end of the reaction (around 24 h),
remaining solids were removed by vacuum filtration (Whatman No. 1).
Both wheat hydrolysates and fungal autolysates were filter sterilised using a
0.2 mm filter unit (Polycap 36 AS, Fisher). The pH of wheat hydrolysates and fungal
autolysates was adjusted to the optimum pH range (6.7–6.9) for W. eutropha growth
with 10 M NaOH.
2.5. Bacterial bioconversions
Bacterial fermentations in 500 ml Erlenmeyer flasks were carried out on a
200 rpm rotary shaker (Infors AG, CH-4103 Bottmingen) at 30 8C and initial pH in
the range of 6.5–6.8. All media prepared contained the same initial concentrations
of glucose (30 g/l) and free amino nitrogen (FAN) (500 mg/l). Fermentations on the
wheat-derived feedstock were formulated by mixing wheat hydrolysates, fungal
extracts and sterile tap water at appropriate volumes in order to acquire the
targeted nutrient concentration. Fermentations on mixtures of commercial glucose
and protein hydrolysates (including YE, SH, CH and CSL) were carried out with and
without mineral supplementation. In the case of mineral supplementation,
fermentation media contained 1 g/l KH2PO4, 0.005 g/l CaCl2 and 0.1 ml trace
element solution. The stock solution of trace elements contained (g/l): 71.2
MgSO4
7H2O, 0.44 ZnSO4
7H2O, 0.812 MnSO4
4H2O, 0.785 CuSO4
5H2O, 0.252
155
Na2MO4
2H2O, 4.98 FeSO4
7H2O, and 1.02 H3BO3. The same volume (50 ml) of
fermentation media and the same initial inoculum (1 ml) were used in each flask. All
measurements presented from shake flask experiments were taken when glucose was
completely consumed. Each shake flask fermentation was repeated three times.
Batch and fed-batch bacterial bioconversions were carried out in a 1.5 l
bioreactor (Electrolab Ltd, Tewkesbury, UK). Prior to fermentation, the bioreactor
was sterilised at 121 8C for 20 min. Temperature and pH were controlled at 30 8C
and 6.7–6.9, respectively. The pH was controlled by adding 10 M NaOH and 10%
H2SO4. Air sparging was carried out at a rate of 1 vvm. The dissolved oxygen (DO) in
the broth was maintained at values higher than 40% of saturation by varying
automatically the agitation speed in the range of 300–1200 rpm. An antifoam probe
was used to detect foam formation and antifoam solution (Dow Corning; BDH) was
used to break the foam whenever it was formed. Microbial growth was monitored
by measuring the CO2 content in the exhaust gas from the fermentation with a gas
analyser (Electrolab Ltd, Tewkesbury, UK). Wheat-derived media were pumped
aseptically into the bioreactor at the beginning of each fermentation.
Fed-batch fermentations FB1–FB6 were carried out by feeding a pure glucose
solution (450 g/l) at a constant rate of 2 ml/h when the residual glucose
concentration was less than 38 g/l. At that time, the CO2 evolution had already
reached a constant value indicating that the exponential growth phase had finished
due to FAN/TN depletion from the fermentation broth. The initial media were
mixtures of wheat hydrolysates and fungal autolysates with glucose and FAN
concentrations in the range of 63–95 g/l and 720–1244 mg/l, respectively. The
fermentations were stopped when CO2 evolution started to drop.
Fed-batch fermentations FB7–FB11 were carried out with two types of wheat-
based media and one pure glucose solution. The initial medium (a mixture of wheat
hydrolysates and fungal autolysates) contained varying initial glucose concentra-
tions (50–100 g/l) and a FAN concentration around 960 mg/l. The first feeding
medium was a wheat hydrolysate (340–380 g/l glucose and 1–1.2 g/l FAN
concentrations) and the second feeding medium was a pure glucose solution
(450 g/l). The highest glucose concentration (270 g/l) in wheat hydrolysates
produced during wheat flour hydrolysis was considered low as a feeding solution
for fed-batch fermentations. For this reason, pure glucose was added into the wheat
hydrolysate to enhance the final glucose concentration. Enhancement of glucose/
nutrient concentration in the wheat hydrolysate could also be achieved via vacuum
evaporation. Since wheat hydrolysates contained nutrients essential for cell
growth, feeding started (12–24 h) when the CO2 evolution was almost constant
indicating that exponential cell growth had finished and PHB accumulation had
been initiated. The feeding in fermentations FB7–FB11 started at three different
flow rates (2, 2.5 or 3 ml/h) and was modified manually on an ad hoc basis
depending on the trend of glucose consumption and CO2 evolution. The main aim
was to observe PHB production at varying nutrient feeding rates that led to changes
in microbial metabolism. The overall feeding rate in fermentations FB7–FB11 was
gradually decreased. This resulted in a slower rate of nutrient addition from
fermentation FB7 to FB11. Fermentation FB11 was carried out with only two feeding
rates (2 and 3 ml/h).
Due to the many steps required to prepare a bacterial fermentation on wheat-
based media in the bioreactor, many fermentations (especially fed-batch cultures)
were not repeated. However, the trends in the results against time and the various
studied effects showed the reliability of the results.
2.6. Analytical methods
An Analox GL6 analyzer was used to measure glucose concentration. FAN
concentration in liquid samples was analyzed by the ninhydrin colorimetric
method promulgated in the European Brewery Convention [39]. It should be
stressed that the term FAN describes the nitrogen from free amine groups contained
in amino acids and peptides. Phosphorus in liquid samples was analysed by the
method described by Herbert et al. [40]. Total nitrogen (TN) in liquid samples was
analysed by a TNM-1 Total Nitrogen analyser integrated with a chemiluminescence
detector and a AS1-V Autosampler (Shimadzu, Japan).
Samples from bacterial fermentations were taken at varied reaction intervals. Each
sample was centrifuged at 3000  g for 10 min and the sediment was washed with
distilled water and centrifuged two consecutive times. The supernatant was used for
the analysis of glucose, FAN and TN concentrations. The solids were re-suspended in
acetone and transferred into universal bottles. Total dry weight (TDW) measurements
were carried out by drying the solids at 50 8C and cooling in a desiccator to constant
weight. Residual cell weight (RCW) in each sample was determined by substracting
the PHB concentration measured by GC analysis from the TDW.
PHB was measured by following the analytical method proposed by Riis and Mai
[41]. A GC analyser (Hewlett Packand 5890, series II) with a poraplot column Q-HT
(10 m  0.32 mm) and a flame ionisation detector (FID) were used. The software
employed was Chemistation Version 6.03. The carrier gas was helium. Injection
temperature was 230 8C, detection temperature was 200 8C and initial temperature
was 120 8C.
2.7. Transmission electron microscopy
Images of W. eutropha cells containing PHB were obtained by transmission
electron microscopy (TEM). Samples were collected at various fermentation times
156 Y. Xu et al. / Process Biochemistry 45 (2010) 153–163

and harvested cells were isolated via centrifugation at 3000  g for 10 min in 0.1 M
phosphate buffer (pH 7). Samples (0.5 ml) were subsequently fixed with 0.5 ml of a
solution consisted of 2% glutaraldehyde, 4% formaldehyde, 2 mM calcium chloride,
0.15 M sucrose in 0.1 M sodium cacodylate buffer at pH 7.3. Fixed samples were
spinned down to form a pellet for 10 min at 3000  g. Then they were postfixed
with 1% OsO4 in water for 1 h, dehydrated in ascending concentrations of the
ethanol, transferred in propylene oxide, infiltrated with TAAB LV resin and
polymerized at 60 8C for 24 h. The polymerized blocks were cut on Reichert Ultracut
ultramicrotome (Austria, Wein) to get ultrathin sections of 70 nm thick. The
sections were collected on 200 mesh electron microscopy copper grids and stained
with 1% uranyl acetate and 0.3% lead citrate. The sections were observed in FEI
Tecnai 12 Biotwin microscope (FEI, Eindhoven, The Netherlands).

3. Results and discussion

3.1. Batch fermentations for PHB production using wheat-derived

media

The effect of the nutrient content in wheat-derived media on W.

eutropha growth and PHB synthesis was initially elucidated via
batch fermentations in a 1.5 l bioreactor. Table 1 presents the
results from seven batch fermentations that were conducted in
wheat-derived media with varying initial compositions of glucose
(19.5–120 g/l) and FAN (100–1200 mg/l). These experimental
results indicate that increasing glucose and FAN concentrations
at the beginning of batch fermentations lead to increasing TDW,
PHB concentration, RCW, and productivity (QPHB, i.e. PHB produc-
tion per hour). A glucose concentration higher than 110 g/l
resulted in a longer lag phase, whereas at glucose concentrations
higher than 120 g/l microbial growth was not observed (data not
shown). At FAN concentrations higher than approximately
1100 mg/l (experiment B7) no PHB accumulation was observed
(Table 1). Three additional batch fermentations (results not
presented) were conducted using glucose and FAN concentrations
around 70–100 g/l and 1100–1300 mg/l respectively in order to
verify that PHB accumulation was not achieved at FAN concentra-
tions higher than approximately 1100 mg/l. GC analysis showed
that a range of organic acids were produced at initial FAN
concentration higher than 1100 mg/l.
An average specific growth rate of 0.35 h 1 (less than 15%
deviation from the average) was calculated during the exponential
growth phase of fermentations B3–B6 from logarithmic plots of
CO2 evolution. This specific growth rate corresponds to the
fermentation period that microbial cells were growing exponen-
tially and PHB accumulation had not been initiated yet. At FAN
concentrations higher than 1100 mg/l, the average specific growth
rate during the exponential growth phase was around 0.9 h 1 (less
Fig. 2. Profiles of TDW, PHB, RCW, PHB content in microbial cells, CO2 content in
than 5% deviation from the average). These results indicate that exhaust gas and glucose/FAN/phosphorus consumption throughout batch
PHB accumulation by W. eutropha cells growing on wheat-based fermentation B4.
media is influenced by the specific growth rate.
Fig. 2 presents the profiles of experimental results from content in the medium has been consumed, the exponential
fermentation B4. Similar trends were observed in all batch growth phase of W. eutropha was terminated. The profiles of FAN
fermentations. Microbial growth is mainly associated with FAN and TN consumption illustrate that W. eutropha cells consume both
and TN consumption. When the majority of the FAN and TN amino acids and peptides. From glucose and PHB profiles, it can be

Table 1
Batch fermentations carried out on wheat-derived media.

Batch Initial glucose Initial FAN Ft (h)a TDWa Maximum PHB Maximum PHB to Maximum RCWb YPHB QPHB
content (g/l) content (mg/l) (g/l) contenta (g/l) TDW ratio (g/g) (g/l) (g/g) (g/l h)

B1 19.5 108 46 10.5 7 0.67 3.5 0.36 0.15

B2 34.4 190 42 19.3 11.5 0.6 7.8 0.34 0.27
B3 48.5 305 48 27.5 17.6 0.64 9.9 0.36 0.37
B4 66 560 63 42 23.9 0.57 18.1 0.36 0.38
B5 65.3 720 53 44.5 25.8 0.58 18.8 0.4 0.49
B6 97.4 951 69 63 41.5 0.66 21.8 0.43 0.6
B7 105 1200 33 26.2 0 0 26.2 0 0
a
Total dry weight (TDW) and maximum PHB content correspond to the fermentation time (Ft) that either glucose was completed consumed or PHB degradation was
initiated as was observed after quantitative analysis of fermentation samples.
b
Maximum residual cell weight (RCW) corresponds to the maximum value of non-PHB microbial biomass observed during each fermentation.
 Y. Xu et al. / Process Biochemistry 45 (2010) 153–163
observed that the detectable depletion of glucose in the medium
can be associated with PHB accumulation. A residual amount of
FAN, TN and phosphorus was observed at the end of all
fermentations. The profile of CO2 content in the outlet gas from
fermentation B4 shows that the CO2 content acquires a constant
value (indicating the end of the exponential cell growth phase)
approximately the same time that the majority of FAN and TN has
been consumed (around 22 h). The stationary phase of CO2 content
in the exhaust gas coincides with PHB accumulation and glucose
consumption.
It is interesting to note that microbial growth (6.4 g/l RCW) took
place in the first 17 h without glucose consumption. This means
that W. eutropha cells used other nutrients as carbon sources. The
close association of FAN/TN consumption with microbial growth
indicates that W. eutropha cells used peptides and amino acids as
carbon sources. In addition, the medium may contain some
oligosaccharides (e.g. maltose) but W. eutropha cells cannot
consume maltose [42]. The glucose to PHB conversion yields
(YPHB) calculated for fermentations B1–B6 (Table 1) indicate that
higher initial FAN concentrations result in increasing yields.
Fig. 2 also presents that phosphorus consumption occurs during
fermentation. Phosphorus is released during wheat flour hydro-
lysis and fungal autolysis. The microbial growth and nutrient
Fig. 3. TEM images of W. eutropha in fermentation B4; (A) 24 h, (B) 62 h, (C) 70 h, and (D) 82 h (bar = 2 mm).
157
consumption profiles observed in batch fermentations B1–B6
indicate that wheat-derived media provide adequate quantities of
FAN and phosphorus to support the growth of W. eutropha. By
manipulating the amount of FAN and/or phosphorus release during
wheat flour hydrolysis we could produce wheat hydrolysates with
high glucose concentration, and varying FAN and phosphorus
concentrations. The level of FAN and phosphorus concentration in
wheat hydrolysates can be manipulated by varying the extraction
of gluten and bran from wheat before wheat flour hydrolysis.
Wheat hydrolysates with high content of FAN and phosphorus
could be used to formulate starting media for PHB fermentations
[27]. Wheat hydrolysates with low content of FAN and/or
phosphorus could be subsequently used as feeding media for
fed-batch fermentations to induce PHB accumulation and provide
sufficient amount of nutrients to support microbial cell main-
tenance. It should be stressed that besides nitrogen, phosphorus
limitation could also induce PHB accumulation [43].
Koutinas et al. [13] reported that shake flask fermentations
using wheat-derived media with increasing FAN concentration and
constant glucose concentration resulted in higher RCW. The main
target of batch fermentations was to establish the optimum initial
FAN concentration that would lead to maximum RCW. Increasing
the RCW would eventually lead to higher PHB concentration
158 Y. Xu et al. / Process Biochemistry 45 (2010) 153–163

during fed-batch fermentations by feeding glucose-rich media.
From the batch fermentations conducted in this study, it could be
concluded that an initial FAN concentration in the range of 900–
1000 mg/l could lead to maximum RCW (20–22 g/l).
Fig. 3 presents TEM images of samples taken at different times
during fermentation B4. By comparing images 3a with b (taken at
24 and 62 h respectively) an increased PHB granule formation
during glucose consumption (see Fig. 2) could be observed. The
white granules in the cells are PHB. It can be observed that at 24 h
when only 4.4 g/l glucose had been consumed, PHB granules have
already been formed inside W. eutropha cells. Towards the
completion of glucose consumption at 62 h, the vast majority of
microbial cells contained many PHB granules.
Images 3c and d were taken at 70 and 82 h when glucose
consumption had stopped. The fermentation was not stopped
immediately after all glucose was consumed at 62 h in order to
investigate the possibility of cell autolysis and PHB release. From
images 3c and d, it can be observed that cell autolysis occurred. The
arrows are pointing at fragments of cell walls that were generated
through cell wall disruption by autolysins. Especially at 82 h
(image 3d), many PHB granules were free in the fermentation
broth. It was also observed that at image 3d some cells were free of
PHB. The observation of PHB-free cells approximately 20 h after
glucose consumption ended indicates the appearance of either
new cells or cells that have consumed the accumulated PHB
granules for maintenance. The consumption of PHB as a carbon
source and nutrients released from autolysed microbial cells may
lead to cell maintenance. However, close examination of image 3d
reveals the existence of dark-stained structures near the centre of
some PHB-free cells. Tian et al. [44] also reported the appearance of
similar dark-stained structures or mediation elements near the
centre of the cross-section of many W. eutropha cells at the early
stages of fermentations (contrary to this study where they were
observed during the cell autolysis stage). The dark mediation
elements gradually disappeared or became obscured at later
fermentation times when PHB granules were properly formed [44].
A separate study is required in order to elucidate cell autolysis,
potential growth of W. eutropha cells and formation of mediation
elements at the end of fermentations on wheat-based media.
Jung et al. [45] have developed a novel cultivation method
based on modulation of the initial inoculum size and the medium
composition that enables autolysis of recombinant E. coli cells and
spontaneous extraction of PHB granules. In many fermentations
presented in this study, PHB purity increased to values higher than
90% (PHB to TDW weight ratio) at the end of fermentations without
simultaneous reduction of PHB concentration. This indicates that
the application of cell autolysis for PHB extraction and purification
should be further examined in the wheat-based biorefinery
proposed in Fig. 1.
3.2. Fed-batch fermentations for PHB production using
wheat-derived media
The PHB content in microbial cells at the end of each batch
fermentation (Table 1) was lower than 67% (w/w). This is lower
than the highest content of PHB (80–90%) that can be
accumulated in microbial cells [16,23,46]. Operating the
fermentation system in a fed-batch mode and adding a
concentrated glucose solution more PHB could be accumulated
in W. eutropha cells. To justify this, six fed-batch fermentations
(FB1–FB6) were carried out at varying initial glucose and FAN
concentrations (Table 2). In each one of these fermentations, a
pure glucose solution (450 g/l) was fed at a constant rate of 2 ml/
h when the glucose concentration was reduced to less than 38 g/
l. The results presented in Table 2 illustrate that increasing initial
FAN concentration led to higher PHB production, RCW, PHB
content in microbial cells, and consumption of higher amounts of
glucose. The average glucose to PHB conversion yield (YPHB) and
productivity as related to PHB (QPHB) for fermentations FB1–FB5
were 0.44 g/g and 0.94 g/l h, respectively. Fermentations FB1–
FB5 resulted in enhanced PHB production as compared to batch
cultures B1–B6.
Although glucose feeding was initiated at different times, the
trends of PHB accumulation from fermentations FB1 to FB5 were
similar. This result indicates that feed starting time over the range
used does not have a significant impact on PHB accumulation rate.
PHB was not produced in FB6 due to the high initial FAN
concentration as observed in batch fermentations.

Table 2
Fed-batch fermentations carried out on wheat-derived media.

Fed-batch Total glucose Initial FAN Ft (h)b TDWb Maximum PHB Maximum PHB to Maximum RCWc YPHB QPHB
consumeda (g) content (mg/l) (g/l) contentb (g/l) TDW ratio (g/g) (g/l) (g/g) (g/l h)

Fed-batch fermentations using a feeding solution containing only glucose

FB1 64.7 720 53 54.2 44.1 0.81 17.1 0.46 0.83
FB2 72 784 61 63.8 51.7 0.81 17.8 0.45 0.85
FB3 82.7 809 60 64.5 56.5 0.88 18.3 0.44 0.94
FB4 85.2 853 61 71.6 65.5 0.92 19.1 0.44 1.08
FB5 100.5 960 70 73.2 68.2 0.93 24.1 0.41 0.97
FB6 50 1244 30 27.1 0 0 27.1 0 0

Fed-batche Total glucose Total FAN Ft (h)b TDWb Maximum PHB Maximum PHB to Final RCWf YPHB QPHB
consumeda(g) consumed (mg)d (g/l) contentb (g/l) TDW ratio (g/g) (g/l) (g/g) (g/l h)

Fed-batch fermentations using two feeding solutions, a wheat hydrolysate and a pure glucose solution
FB7 45 445 32 53.2 26 0.49 24.8 0.46 0.81
FB8 60 503 44 70 45.3 0.65 24.7 0.47 1.04
FB9 84.5 663 72 98.4 66.9 0.68 31.5 0.48 0.93
FB10 175.1 983 183 175.2 162.8 0.93 12.4 0.47 0.89
FB11 186.2 782 171 162.6 130.2 0.80 32 0.47 0.68
a
In all fed-batch fermentations the total amount of glucose consumed has been calculated by taking into consideration the initial glucose concentration at the beginning of
the fermentation, the glucose added during the feeding stage and the remaining glucose concentration at the end of the fermentation.
b
Total dry weight (TDW) and maximum PHB content correspond to the fermentation time (Ft) that glucose was completely consumed, or PHB degradation was initiated, or
production of extracellular polymeric material was initiated.
c
Maximum residual cell weight (RCW) in fermentations FB1–FB6 corresponds to the maximum value of non-PHB microbial biomass observed during each fermentation.
d
In fed-batch fermentations FB7–FB11 the total amount of FAN consumed has been calculated by taking into consideration the initial FAN concentration at the beginning
of the fermentation, the FAN added during the feeding stage and the remaining FAN concentration at the end of the fermentation.
e
The order that fermentations FB7–FB11 are presented coincides with the application of a slower overall feeding rate of wheat hydrolysates.
f
For fed-batch fermentations FB7–FB11, RCW corresponds to the fermentation time (Ft) that the maximum PHB concentration was observed.
 Y. Xu et al. / Process Biochemistry 45 (2010) 153–163
The highest PHB concentration (68.2 g/l) and glucose con-
sumption (100.5 g) were achieved at FB5 when an initial FAN
concentration of 960 mg/l was used (Fig. 4). Addition of the pure
glucose solution was initiated at 42 h. FAN and phosphorus
consumption ceased at approximately 52 and 61 h, respectively
(Fig. 4b). The profile of RCW at Fig. 4a shows that the RCW begun
to decrease when FAN and phosphorus consumption ceased.
This phenomenon occurred in all fed-batch fermentations (FB1–
FB5) carried out with addition of pure glucose. This indicates
that FAN availability during PHB accumulation is essential for
the survival of W. eutropha cells. Koutinas et al. [13] showed in
shake flask fermentations that the addition of wheat hydro-
lysates during the feeding stage of a fed-batch fermentation
prolonged the survival of microbial cells and led to higher PHB
concentrations.
Fermentations FB1–FB5 showed that feeding of a pure glucose
solution enhanced PHB formation but led to microbial cell lysis due
to the absence of nutrients essential for cell maintenance. Feeding
a solution containing essential nutrients for cell maintenance
could prolong the viability of W. eutropha cells during the PHB
production phase. However, PHB accumulation could be compro-
mised by the supply of nutrients that stimulate microbial growth.
In order to evaluate the effect of nutrient addition during the PHB
accumulation phase, a second set of fed-batch fermentations (FB7–
FB11) was carried out with two feeding media, a wheat
hydrolysate and a pure glucose solution (Table 2). The main
difference in these experiments was the trend of the wheat
hydrolysate feeding rate used in each fermentation. Fermentations
FB7–FB11 are presented in Table 2 in an order that a slower overall
feeding rate of wheat hydrolysates was applied. As shown in
Table 2 slower addition of wheat hydrolysates in fermentations
Fig. 4. Profiles of TDW, PHB, RCW, PHB content in microbial cells, and glucose/FAN
consumption throughout fed-batch fermentation FB5 using pure glucose as feeding
medium.
159
FB7–FB10 led to higher glucose/FAN consumption, PHB produc-
tion, and PHB to TDW ratio.
The duration of fermentations FB7–FB11 (Ft) presented in
Table 2 corresponds to the time that the maximum PHB
concentration was achieved. In fermentations FB7–FB9, PHB
production may have finished at 32, 44 and 72 h respectively,
but each one of these fermentations continued with the excretion
of unidentified extracellular polymer-like materials. The viscosity
of the fermentation broth gradually increased significantly and its
colour changed to yellow. Wang and Yu [47] reported that bacterial
growth and increased nitrogen availability led to enhanced
production of extracellular polymeric substances during W.
eutropha fermentation. In addition, extracellular polymeric sub-
stances may have been produced to supply an additional source of
nitrogen for the synthesis of PHB synthases by W. eutropha cells
[47]. The results from the fed-batch experiments presented in this
study also indicate that PHB synthesis and secretion of extra-
cellular polymer-like materials are closely associated with
bacterial growth. A further study is required to identify the nature
of these extracellular products and the association of microbial
metabolism with the secretion of such polymer-like substances.
Fermentation FB11 resulted in lower PHB production, PHB to
TDW ratio and productivity as compared to fermentation FB10. The
feeding rate of FB11 started at 2 ml/h at 22 h and increased to 3 ml/
h at 46 h until the end of the fermentation. After 171 h in FB11, PHB
accumulation ceased and organic acid production commenced by
W. eutropha cells. The results obtained in fermentations FB7–FB11
indicate that the feeding flow rate of wheat hydrolysates has a
significant effect on microbial metabolism. A fed-batch fermenta-
tion not presented in this study showed that PHB synthesis
stopped when glucose concentration in the broth during the
feeding stage was reduced to less than 10 g/l.
The highest PHB concentration (162.8 g/l) was achieved in FB10
(Fig. 5) when the feeding profile shown in Fig. 5c was used. The
utilisation of wheat hydrolysate as feeding solution led to the
maintenance of microbial activity as shown by the CO2 profile
(Fig. 5c). When feeding of wheat hydrolysate stopped at 119 h and
feeding of pure glucose solution was initiated, the CO2 profile
started to decline. This indicated that the absence of nutrients from
the fermentation broth led to a gradual reduction in bacterial
viability. The RCW in fermentation FB5 (Fig. 4) was reduced when a
pure glucose solution was added into the fermentation broth.
Therefore, using wheat hydrolysates as feeding media improves
significantly PHB production due to the retention of microbial
activity for a longer time.
The productivity achieved in FB10 (0.89 g/l h) should be
increased in order to facilitate industrial implementation of the
wheat-based biorefinery proposed in Fig. 1. This could be achieved
by feeding a more concentrated wheat hydrolysate so as to prevent
the dilution of the microbial population in the fermentation broth.
More concentrated wheat hydrolysates could be produced either
through evaporation of the wheat hydrolysates used in this process
or by improving the wheat hydrolysis process. The latter is feasible
when higher enzyme activities are used. Higher enzyme activities
could be produced from continuous Aspergillus awamori fermenta-
tions [27]. Optimising the feeding flow rate of wheat hydrolysates
could also increase productivity.
The conversion yield from glucose to PHB achieved in most fed-
batch fermentations is close to the theoretical one 0.48 g/g [48].
This could be attributed to the consumption of peptides and amino
acids. Certain amino acids could be used for PHB production
[22,24,25]. Lee et al. [22] supported that the stimulatory effects of
amino acids on PHB synthesis could be attributed to the
availability of more acetyl-CoA and/or NADPH. Therefore, the
exploitation of the protein content of cereals could lead to
enhanced PHB production yields when compared to fermentations
160 Y. Xu et al. / Process Biochemistry 45 (2010) 153–163

carried out with purified glucose or starch hydrolysates mixed

with synthetic chemicals.
Fig. 6 presents TEM images of samples taken at 62, 103, 142, and
183 h during fermentation FB10. These images demonstrate the
gradual increase of PHB accumulation during fermentation FB10.
The obvious cell autolysis that is depicted in image 6d explains the
gradual increase of PHB to TDW content up to a value of 93% (w/w)
at the end of fermentation.
It would be more useful in the case of industrial implementa-
tion to present the PHB production yield as related to the total
amount of wheat consumed for both fungal (enzyme production)
and bacterial (PHB production) fermentations. Therefore, for
fermentation FB10 the PHB production yield from wheat was
around 0.3 g PHB/g wheat, corresponding to 82.8% of the
theoretical conversion yield (0.362 g PHB/g wheat). These yields
were calculated by taking into consideration a 93% starch to
glucose conversion efficiency during wheat flour hydrolysis.
Maltose or other oligosaccharides produced during hydrolysis of
starch cannot be consumed by W. eutropha [42].
In the results presented above, the media produced by the
proposed wheat-based biorefinery resulted in high glucose to PHB
conversion yield (0.47 g/g), PHB concentration (162.8 g/l), and PHB
to TDW content (93% w/w). However, the PHB productivity (0.89 g/
l h) achieved was not high enough. A potential industrialisation of
the biorefinery proposed in Fig. 1 can only be evaluated by
conducting a life cycle assessment and an economic analysis. This
will be the focus of a future study.

3.3. Comparison of wheat-derived media with commercial complex

media for PHB production

Wheat-based media derived from the process presented in

Fig. 5. Profiles of TDW, PHB, RCW, PHB content in microbial cells, and glucose/FAN
consumption throughout fed-batch fermentation FB10 using wheat hydrolysate
and pure glucose as feeding media.
Fig. 1 have been compared to media formulated with commercial
sources of nutrients (including glucose, YE, SH, CH, CSL, and various
inorganic chemicals) for PHB production during W. eutropha
fermentations in both shake flask and a 1.5 l bioreactor. Table 3
summarises the results from two sets of shake flask fermentations.
At the same initial concentrations of glucose (30 g/l) and FAN
(500 mg/l), mineral supplementation enhanced TDW, PHB accu-
mulation and RCW in the fermentations that protein hydrolysates
and commercial glucose was used. Furthermore, mineral supple-
mentation reduced the time required to consume all the glucose.
The fermentation times (Ft) listed in Table 3 correspond to the time
required to consume all the glucose. The highest PHB concentra-
tions were obtained from wheat-derived media (10.9 g/l), CH with
mineral supplementation (10.9 g/l) and SH with mineral supple-
mentation (10.1 g/l). Fermentations with only commercial glucose
and protein hydrolysates resulted in lower microbial growth and
PHB synthesis. These results indicate that commercial protein
hydrolysates do not contain adequate quantities of minerals to

Table 3
Shake flask fermentations carried out on wheat-derived media and commercial sources of nutrients.

Media Ftb (h) TDW (g/l) PHB (g/l) PHB to TDW ratio (g/g) RCW (g/l)

Initial glucose and FAN concentrations: 30 g/l and 0.5 g/l

Wheat-based 31 15.8 10.9 0.69 4.9
G/YEa 37.5 13.6 8.1 0.59 5.5
G/SHa 85 15.5 10.1 0.65 5.4
G/CHa 61 16.3 10.9 0.67 5.4
G/CSLa 88 11 6.4 0.58 4.6

G/YE 44 12.6 10.6 0.84 2.0

G/SH 125 12.4 10 0.81 2.4
G/CH 72 7.3 5.6 0.77 1.7
G/CSL 109 5.8 3.7 0.64 2.1
a
Addition of minerals.
b
Fermentation time (Ft) corresponds to the time that all glucose has been consumed.
 Y. Xu et al. / Process Biochemistry 45 (2010) 153–163
Fig. 6. TEM images of W. eutropha cells in fed-batch culture FB10; (A) 62 h, (B) 103 h, (C) 142 h, and (D)183 h (bar = 2 mm).
support sufficient microbial growth and PHB synthesis as in the
case of wheat-derived media. Therefore, the utilisation of the
proposed wheat-based biorefinery will not require the addition of
any inorganic chemicals as sources of minerals. Furthermore, the
duration of fermentations carried out with wheat-based media
was significantly lower than fermentations carried out with
commercial media.
Based on the results from shake flask fermentations, batch
fermentations were also carried out in a 1.5 l bioreactor to justify
the superiority of wheat-based to commercial media. Three
fermentations were conducted using wheat-based and commercial
media (Fig. 7). The initial glucose and FAN concentrations were
around 62.5 g/l (61.6 g/l for wheat-derived media and 63.7 g/l for
YE- and SH-based media) and 812 mg/l (809, 821, and 803 mg/l for
wheat-, YE-, and SH-based media) respectively. The protein
hydrolysates used were YE and SH. CSL was not used due to the
poor performance in shake flask fermentations.
CH was not used because similar fermentations have been
presented by Bormann et al. [23]. According to Bormann et al. [23],
the highest TDW (65 g/l), PHB concentration (52.2 g/l) and PHB to
TDW ratio (0.8 g/g) were obtained in a medium with a CH
161
concentration of 30 g/l, which corresponds to more than 800 mg/l
FAN according to the FAN analytical technique used in this study.
The amount of glucose consumed in this fermentation was not
clearly stated. Fermentations FB3–FB5 (Table 2) using wheat-
derived media resulted in higher TDW (64.5–73.2 g/l), PHB
concentrations (56.5–68.2 g/l) and PHB to TDW content (0.88–
0.93 g/g) as compared to the results published by Bormann et al.
[23]. However, a higher productivity (1.2 g/l h) has been reported
by Bormann et al. [23].
The experimental results presented in Fig. 7 indicate that
wheat-based media provide a better medium for PHB production
as compared to media formulated with commercial glucose, CH,
SH and various inorganic chemicals. The highest TDW (49.5 g/l),
PHB concentration (28 g/l), and PHB to TDW ratio (0.56 g/g)
were achieved in a shorter fermentation time (60 h) when
wheat-based media were used. Therefore, the trend of the
results obtained in shake flask fermentations is verified in
fermentations carried out in a controlled environment using a
1.5 l bioreactor.
It should be stressed that the nutrient content of the
hydrolysates used is highly complex and FAN has been used as
162 Y. Xu et al. / Process Biochemistry 45 (2010) 153–163

Fig. 7. Profiles of TDW, PHB, RCW, PHB content in microbial cells, and glucose/FAN consumption throughout batch fermentation on various media. Fermentation media: (*)
glucose + YE + minerals; (~) glucose + SH + minerals; (~) wheat-derived media.

a basis for comparison. The experimental results presented in 4. Conclusions

Fig. 7 indicate that despite the fact that the initial glucose and
FAN concentrations were similar in these experiments, differ-
ences in specific nutrient composition between hydrolysates led
to significantly different experimental results. The differences
observed in the profiles of RCW throughout the three fermenta-
tions (Fig. 7c) indicate the influence of the nutrient content
from each medium to microbial growth. However, the trends of
FAN consumption in wheat-based media and YE were similar
(Fig. 7f).
The calculated PHB yield (0.45 g/g) and productivity (0.46 g/l h)
for wheat-based media were also higher than YE-based (0.36 g/g
and 0.33 g/l h) and SH-based (0.34 g/g and 0.17 g/l h) media. The
productivities were calculated by taking into consideration the
highest PHB concentration achieved in each fermentation.
This study presents a novel wheat-based biorefinery for PHB
production via microbial fermentation using wheat as the sole
source of nutrients. Processing advantages introduced by this
concept could lead to viable production of PHB. The addition of
nitrogen sources provided from gluten hydrolysis increases the
overall glucose to PHB conversion yield. The recycle of nutrients via
fungal autolysis would result in waste minimisation. The
exploitation of all wheat components for the production of
value-added products in conjunction with PHB (especially in the
case that gluten is used for the production of recombinant
proteins) will boost biorefinery development. Wheat-based media
result in improved PHB production as compared to commercial
sources of nutrients. Bacterial autolysis at the end of the
 Y. Xu et al. / Process Biochemistry 45 (2010) 153–163
fermentation could be exploited to increase the purity of PHB prior
to its purification. Besides PHB production, preliminary results
have indicated that PHB purification could be achieved via
microbial cell lysis induced by enzymes excreted by A. oryzae
solid state bioconversions. This may reduce the usage of chemicals
and/or energy for PHB purification.
Acknowledgements
This work was supported by the Engineering and Physical
Science Research Committee grant gr/s24909/01 in association
with the Crystal Faraday Partnership. The authors gratefully
acknowledge the generous contribution of the Satake Corporation
of Japan in providing financial support for much of the research
carried out in the Satake Centre for Grain Process Engineering
(SCGPE, University of Manchester, UK).
The authors thank the staff in the EM facility in the Faculty of
Life Sciences for their assistance, and the Wellcome Trust for
equipment grant support to the EM facility.
References
[1] Suriyamongkol P, Weselake R, Narine S, Moloney M, Shah S. Biotechnological
approaches for the production of polyhydroxyalkanoates in microorganisms
and plants—a review. Biotechnol Adv 2007;25:148–75.
[2] Pietrini M, Roes L, Patel MK, Chiellini E. Comparative life cycle studies on
poly(3-hydroxybutyrate)-based composites as potential replacement for con-
ventional petrochemical plastics. Biomacromology 2007;8:2210–8.
[3] Tokiwa Y, Ugwu CU. Biotechnological production of (R)-3-hydroxybutyric acid
monomer. J Biotechnol 2007;132:264–72.
[4] Novikova LN, Pettersson J, Brohlin M, Wiberg M, Novikov LN. Biodegradable
poly-b-hydroxybutyrate scaffold seeded with Schwann cells to promote
spinal cord repair. Biomaterials 2008;29:1198–206.
[5] Gerngross TU. Can biotechnology move us toward a sustainable society? Nat
Biotechnol 1999;17:541–4.
[6] Salehizadeh H, Van Loosdrecht MCM. Production of polyhydroxyalkanoates by
mixed culture: recent trends and biotechnological importance. Biotechnol Adv
2004;22:261–79.
[7] Doi Y, Steinbuchel A. Biopolymers: polyesters II—properties and chemical
synthesis. Wiley–VCH; 2001.
[8] Kim S, Dale BE. Life cycle assessment study of biopolymers (polyhydroxyalk-
anoates) derived from no-tilled corn. Int J Life Cycle Assess 2005;10:200–10.
[9] Rusendi D, Sheppard JD. Hydrolysis of potato processing waste for the produc-
tion of poly-b-hydroxybutyrate. Bioresour Technol 1995;54:191–6.
[10] Zhang S, Norrlow O, Wawrzynczyk J, Dey ES. Poly(3-hydroxybutyrate) bio-
synthesis in the biofilm of Alcaligenes eutrophus, using glucose enzymatically
released from pulp fiber sludge. Appl Environ Microbiol 2004;70:6776–82.
[11] Koller M, Bona R, Hermann C, Horvat P, Martinz J, Neto J, et al. Biotechnological
production of poly(3-hydroxybutyrate) with Wautersia eutropha by applica-
tion of green grass juice and silage juice as additional complex substrates.
Biocatal Biotransform 2005;23:329–37.
[12] Nikel PI, de Almeida A, Melillo EC, Galvagno MA, Pettinari MJ. New recombi-
nant Escherichia coli strain tailored for the production of poly(3-hydroxybu-
tyrate) from agroindustrial by-products. Appl Environ Microbiol
2006;72:3949–54.
[13] Koutinas AA, Xu Y, Wang R-H, Webb C. Polyhydroxybutyrate production from
a novel feedstock derived from a wheat-based biorefinery. Enzyme Microb
Technol 2007;40:1035–44.
[14] Solaiman DKY, Ashby RD, Foglia TA, Marmer WN. Conversion of agricultural
feedstock and coproducts into poly(hydroxyalkanoates). Appl Microbiol Bio-
technol 2006;71:783–9.
[15] Lapointe R, Lambert A, Savard L. Process for production of biopolymer. World
Intellectual Property Organization. Publication number WO0222841; 2002.
[16] Yu H, Shi Y, Yin J, Shen Z, Yang S. Genetic strategy for solving chemical
engineering problems in biochemical engineering. J Chem Technol Biotechnol
2003;78:283–6.
[17] Quillaguaman J, Hashim S, Bento F, Mattiasson B, Hatti-Kaul R. Poly(b-hydro-
xybutyrate) production by a moderate halophile, Halomonas boliviensis LC1
using starch hydrolysate as substrate. J Appl Microbiol 2005;99:151–7.
[18] Huang T-Y, Duan K-J, Huang S-Y, Chen CW. Production of polyhydroxyalk-
anoates from inexpensive extruded rice bran and starch by Haloferax medi-
terranei. J Ind Microbiol Biotechnol 2006;33:701–6.
[19] Chen CW, Don T-M, Yen H-F. Enzymatic extruded starch as a carbon source for
the production of poly(3-hydroxybutyrate-co-3-hydroxyvalerate) by Halo-
ferax mediterranei. Process Biochem 2006;41:2289–96.
163
[20] Fujita M, Nakamura K, Kuroki H, Yoshie N, Inoue Y. Biosynthesis of polyesters
from various amino acids by Alcaligenes eutrophus. Int Biol Macromol
1993;15:253–5.
[21] Lee SY, Chang HN. Effect of complex nitrogen source on the synthesis and
accumulation of poly(3-hydroxybutyric acid) by recombinant Escherichia coli
in flask and fed-batch cultures. J Environ Polym Degrad 1994;2:169–76.
[22] Lee SY, Lee YK, Chang HN. Stimulatory effects of amino acids and oleic acid on
poly(3-hydroxybutyric acid) synthesis by recombinant Escherichia coli. J Fer-
ment Bioeng 1995;79:177–80.
[23] Bormann EJ, Leihner M, Roth M, Beer B, Metzner K. Production of polyhy-
droxybutyrate by Ralstonia eutropha from protein hydrolysates. Appl Micro-
biol Biotechnol 1998;50:604–7.
[24] Mahishi LH, Rawal SK. Effect of amino acid supplementation on the synthesis
of poly(3-hydroxybutyrate) by recombinant Escherichia coli. World J Microbiol
Biotechnol 2002;18:805–10.
[25] Kimura H, Mouri K, Takeishi M, Endo T. Production and characterization of
poly(3-hydroxybutyric acid-co-3-hydroxyvaleric acid) from L-valine by Ral-
stonia eutropha. Bull Chem Soc Jpn 2003;76:1775–81.
[26] Wang R-H, Dominguez-Espinosa RM, Leonard K, Koutinas AA, Webb C. The
application of a generic feedstock from wheat for microbial fermentations.
Biotechnol Progr 2002;18:1033–8.
[27] Koutinas AA, Wang R–H, Webb C. Restructuring upstream bioprocessing:
technological and economical aspects for the production of a generic microbial
feedstock from wheat. Biotechnol Bioeng 2004;85:524–38.
[28] Webb C, Koutinas AA, Wang R–H. Developing a sustainable bioprocessing
strategy based on a generic feedstock. Adv Biochem Eng/Biotechnol 2004;
87:196–268.
[29] Koutinas AA, Wang R–H, Campbell GM, Webb C. A whole crop biorefinery
system—a closed system for the manufacture of nonfood products from
cereals. In: Kamm B, Gruber PR, Kamm M, editors. Biorefineries—industrial
processes and products, vol. 1. Wiley–VCH; 2006. p. 165–91 [Chapter 8].
[30] Arifeen N, Wang R-H, Kookos IK, Webb C, Koutinas AA. Process design and
optimization of novel wheat-based continuous bioethanol production system.
Biotechnol Progr 2007;23:1394–403.
[31] Arifeen N, Wang R-H, Kookos IK, Webb C, Koutinas AA. Optimisation and cost
estimation of novel wheat biorefining for continuous production of fermenta-
tion feedstock. Biotechnol Progr 2007;23:872–80.
[32] Du C, Lin SKC, Koutinas A, Wang R-H, Webb C. Succinic acid production from
wheat using a biorefining strategy. Appl Microbiol Biotechnol 2007;76:1263–
70.
[33] Koutinas AA, Wang R, Webb C. The biochemurgist–bioconvertion of agricultural
raw materials for chemical production. Biof Bioprod Bioref 2007;1:24–38.
[34] Koutinas AA, Arifeen N, Wang R-H, Webb C. Cereal-based biorefinery devel-
opment: Integrated enzyme production for cereal flour hydrolysis. Biotechnol
Bioeng 2007;97:61–72.
[35] Koutinas AA, Wang R–H, Webb C. Development of a process for the production
of nutrient supplements for fermentations based on fungal autolysis. Enzyme
Microb Technol 2005;36:629–38.
[36] Fouache C, Seigueilha L, Dubois E. Nitrogenous composition resulting from the
hydrolysis of maize gluten and a process for the preparation thereof. US Patent
6,495,342; 2002.
[37] Satakarni M, Koutinas AA, Webb C, Curtis R. Enrichment of fermentation media
and optimization of expression conditions for the production of EAK16 peptide
as fusions with SUMO. Biotechnol Bioeng 2009;102:725–35.
[38] Vandamme P, Coenye T. Taxonomy of the genus Cupriavidus: a tale of lost and
found. Int J Syst Evol Microbiol 2004;54:2285–9.
[39] Lie S. The EBC-ninhydrin method for determination of free alpha amino
nitrogen. J Inst Brew 1973;79:37–41.
[40] Herbert D, Phipps PJ, Strange RE. Chemical analysis of microbial cells. In: Norris
JR, Ribbons DW, editors. Methods in microbiology, vol. 5B. New York:
Academic Press; 1971.
[41] Riis V, Mai W. Gas chromatographic determination of poly-b-hydroxybutyric
acid in microbial biomass after hydrochloric acid propanolysis. J Chromatogr
1988;445:285–9.
[42] Haas R, Jin B, Zepf FT. Production of poly(3-hydroxybutyrate) from waste
potato starch. Biosci Biotechnol Biochem 2008;72:253–6.
[43] Ryu HW, Hahn SK, Chang YK, Chang HN. Production of poly(3-hydroxybuty-
rate) by high cell density fed-batch culture of alcaligenes eutrophus with
phosphate limitation. Biotechnol Bioeng 1997;55:28–32.
[44] Tian J, Sinskey AJ, Stubbe JA. Kinetic studies of polyhydroxybutyrate granule
formation in Wautersia eutropha H16 by transmission electron microscopy. J
Bacteriol 2005;187:3814–24.
[45] Jung IL, Phyo KH, Kim KC, Park HK, Kim IG. Spontaneous liberation of
intracellular polyhydroxybutyrate granules in Escherichia coli. Res Microbiol
2005;156:865–73.
[46] Lee SY, Choi J, Wong HH. Recent advances in polyhydroxyalkanoate production
by bacterial fermentation: mini-review. Int J Biol Macromol 1999;25:31–6.
[47] Wang J, Yu H-Q. Biosynthesis of polyhydroxybutyrate (PHB) and extracellular
polymeric substances (EPS) by Ralstonia eutropha ATCC 17699 in batch cul-
tures. Appl Microbiol Biotechnol 2007;75:871–8.
[48] Yamane T. Yield of poly-D( )-3 hydroxybutyrate from various carbon sources:
a theoretical study. Biotechnol Bioeng 1993;41:165–70.