Available online at www.sciencedirect.

com
Journal of Biotechnology 132 (2007) 251258
Large-scale production and efcient recovery of PHB with desirable
material properties, from the newly characterised
Bacillus cereus SPV
S.P. Valappil
a,b
, S.K. Misra
b
, A.R. Boccaccini
b
, T. Keshavarz
a
, C. Bucke
a
, I. Roy
a,
a
Department of Molecular and Applied Biosciences, School of Biosciences, University of Westminster,
115 New Cavendish Street, London W1W 6UW, UK
b
Department of Materials, Imperial College London, London SW7 2BP, UK
Received 26 September 2006; received in revised form 13 March 2007; accepted 15 March 2007
Abstract
A newly characterised Bacillus strain, Bacillus cereus SPV was found to produce PHB at a concentration of 38% of its dry cell weight in shaken
ask cultures, using glucose as the main carbon source. Polymer production was then scaled up to 20 L batch fermentations where 29% dry cell
weight of PHB was obtained within 48 h. Following this, a simple glucose feeding strategy was developed and the cells accumulated 38% dry
cell weight of PHB, an increase in the overall volumetric yield by 31% compared to the batch fermentation. Sporulation is the cause of low PHB
productivity fromthe genus Bacillus [Wu, Q., Huang, H., Hu, G.H., Chen, J., Ho, K.P., Chen, G.Q., 2001. Production of poly-3-hydroxybutyrate by
Bacillus sp. JMa5 cultivated in molasses media. Antonie van Leeuwenhoek 80, 111118]. However, in this study, acidic pH conditions (4.55.8)
completely suppress sporulation, in accordance with Kominek and Halvorson [Kominek, L.A., Halvorson, H.O., 1965. Metabolism of poly--
hydroxybutyrate and acetoin in Bacillus cereus. J. Bacteriol. 90, 12511259], and result in an increase in the yield of PHB production. This
observation emphasises the potential of the use of Bacillus in the commercial production of PHB and other PHAs. The recovery of the PHB
produced was optimised and the isolated polymer characterised to identify its material properties. The polymer extracted, was found to have similar
molecular weight, polydispersity index and lower crystallinity index than others reported in literature. Also, the extracted polymer was found to
have desirable material properties for potential tissue engineering applications.
2007 Elsevier B.V. All rights reserved.
Keywords: Bacillus cereus SPV; Polyhydroxybutyrate; Fermentation; Polymer extraction; Polydispersity index; Crystallinity index
1. Introduction
Polyhydroxyalkanoates (PHAs) are polyesters of 3-, 4-, 5-
and 6-hydroxyalkanoic acids, produced by a variety of bacterial
species and are classied as both biodegradable and biocompat-
ible polyesters (Chen and Wu, 2005). They are synthesised by
bacteria as storage compounds for energy and carbon, normally
in the presence of excess carbon with at least one nutrient essen-
tial for growth, such as nitrogen, phosphorus, sulphur or oxygen
present in limiting concentration (Anderson and Dawes, 1990).
The material properties of these polymers vary depending on

Corresponding author. Tel.: +44 20 79115000x3567; fax: +44 20 79115087.

E-mail address: royi@wmin.ac.uk (I. Roy).
the constituent monomers; hence, they are one of the largest
sources of diverse biomaterials. Of all the known PHAs, poly-
3-hydroxybutyrate, PHB, is the most commonly and widely
produced homopolymer by many bacteria. Current applications
of PHB-based polymers or composites include the packag-
ing industry, medicine, pharmacy, agriculture, food industry,
raw material for enantiomerically pure chemicals and the paint
industry (Anderson and Dawes, 1990).
Currently, PHBis produced in an industrial scale using Gram-
negative bacteria such as Cupriavidus necator (Vandamme and
Coenye, 2004), Alcaligenes latus and recombinant Escherichia
coli. However, PHB isolated from Gram-negative organisms
contain the outer membrane lipopolysaccharide (LPS) endotox-
ins, which are pyrogens known to copurify with the PHAs. The
presence of LPS induces a strong immunogenic reaction and
is therefore undesirable for the biomedical application of the
0168-1656/$ see front matter 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jbiotec.2007.03.013
252 S.P. Valappil et al. / Journal of Biotechnology 132 (2007) 251258
PHAs (Chen and Wu, 2005). Gram-positive bacteria lack LPS
and are hence potentially better sources of PHAs to be used for
biomedical applications (Valappil et al., 2007a). However, only
limited work has been carried out on the large-scale produc-
tion of PHB from the Gram-positive genus Bacillus. In the only
report so far, Wu et al., 2001, scaled up the production of PHB
by Bacillus sp. JMa5 using a molasses medium and achieved a
PHB content of 2535%dcw. It was reported that a high ratio
of carbon to nitrogen or carbon to phosphorus and low oxy-
gen supply, conditions known to increase PHB production in
other systems, triggered sporulation in Bacillus sp. JMa5 and
hence reduced rather than increased the yield of PHB (Wu et al.,
2001).
In the large-scale PHB production both the cost effective-
ness and pharmacological purity of PHB is mainly dependent
on the organism used and the extraction method employed
to isolate the polymer from the bacterial cells. Majority
of the separation processes proposed for the recovery of
PHB involve extraction using solvents such as methylene
chloride, propylene carbonate, dichloroethane or chloroform
(Anderson and Dawes, 1990), however, the method in
which boiling chloroform is used in a soxhlet apparatus
(Kannan and Rehacek, 1970; Manna et al., 1999) is most
popular.
In this paper, we report, the large-scale production of PHB
by Bacillus cereus SPV using a production medium used exclu-
sively by this group for the growth of Bacillus sp. Variation in
fermentation conditions has been explored to increase the yield
of PHB. The polymer produced has been isolated using different
methods to optimise the extraction. Finally, the effects of differ-
ent extraction techniques on the polymers properties have been
investigated.
2. Materials and methods
2.1. Chemicals
All chemicals were obtained from SigmaAldrich Company
Ltd. (Dorset, England) except yeast extract which was from
DIFCO (BD UK Ltd., Oxford, UK). GCMS was performed
using analytical grade reagents procured from SigmaAldrich
Company Ltd.
2.2. Bacterial strain and its maintenance
A PHB producing Bacillus spp., obtained from the Univer-
sity of Westminster culture collection was recently characterised
to be B. cereus SPV (Valappil et al., 2007b). This strain
has been used for the current study. Stock cultures were
grown at 30

C in nutrient broth (containing (g/L): Lab-

Lemco Powder, 1.0; yeast extract, 2.0; peptone, 5.0; sodium
chloride, 5.0) and maintained at 4

C after growth on
nutrient agar (containing (g/L): Lab-Lemco Powder, 1.0;
yeast extract, 2.0; peptone, 5.0; sodium chloride, 5.0; agar,
15.0).
2.3. Growth media and culture conditions
2.3.1. Shaken asks culture
A semi-dened PHB production medium, Kannan and
Rehacek medium (Kannan and Rehacek, 1970), was used with
modication, by removing the calcium carbonate from its origi-
nal composition, for the production of PHB in both shaken ask
and 20 L fermentor studies. The pH of the medium was adjusted
to 6.8 before autoclaving. The modied medium contained in
g/L, glucose, 20; Difco yeast extract, 2.5; potassium chloride, 3;
ammonium sulphate, 5.0; 100 mL of defatted soybean dialysate
(prepared from 10 g of defatted soybean meal in 1000 mL of
distilled water for 24 h at 4

C). The inoculum was prepared

in 250 mL Erlenmeyer asks containing 30 mL of sterile nutri-
ent broth (which were inoculated with single colonies from the
overnight grown nutrient agar plates). Flasks were incubated at
30

C for 24 h on a rotary shaker at 250 rpm. The seed cultures,

30 mL, were transferred into 300 mL of sterile modied Kan-
nan and Rehacek medium with glucose as the carbon source
and incubated at 30

C with a rotational speed of 250 rpm. The

temperature range for the growth of B. cereus SPV was found
to be from 25 to 45

C, with optimal growth at 30

C (data not
shown). In order to maintain reproducibility and comparable
conditions of aeration after sample removal, six asks for each
set (ask sets I, II and III in duplicates) were simultaneously
inoculated with the same inoculum. Samples were taken from
the ask set I immediately after inoculation, however, the ask
sets II and III were allowed to grow until 48 and 96 h, respec-
tively, before removing samples. A maximum of 50 mL culture
broth was removed aseptically from each ask for the analy-
sis of dry cell weight, PHB accumulation, glucose consumption
and pH variation, during the growth, so as to maintain simi-
lar media volume to head space ratio, crucial for aeration, in
each case.
2.3.2. Fermentor culture
The production of PHB using B. cereus SPV was carried out
in 20 L fermentors with a 14 L working volume. The fermentors
were sterilized while containing only the salts of the modied
Kannan and Rehacek medium (Kannan and Rehacek, 1970).
Separately sterilized glucose and soybean dialysate were added
to the fermentors aseptically before inoculation with 1.4 L of
a 24 h inoculum. The pH of the medium was checked before
inoculation, and was adjusted to 6.8 using IN NaOH. The dis-
solved oxygen tension (DOT) was set at 100% air saturation
at the beginning of the run. The pH, DOT, glucose concentra-
tion and cell optical density were monitored throughout the run.
The impeller speed was kept at 250 rpm, airow rate at 1.0 vvm
and the temperature at 30

C. During the fed-batch fermenta-

tion 150 10 mL of concentrated glucose solution (500 g/L)
was administered aseptically into the medium to maintain the
glucose concentration level above 10 g/L. This was carried out
at 20, 32, 40 and 54 h of growth. Dissolved oxygen concen-
tration was measured with polarographic oxygen-sensing probe
(Ingold, Mettler-Toledo Ltd., Beaumont Leys, Leicester, UK).
Antifoam (FG-10; Dow corning, Edison, NJ, USA) at a concen-
tration of 1:10 (v/v) in water was added after 12, 24 and 48 h
S.P. Valappil et al. / Journal of Biotechnology 132 (2007) 251258 253
of growth. Samples were withdrawn aseptically either at hourly
intervals or after a xed incubation period.
2.4. Qualitative analysis of endospores
Samples of B. cereus SPV were checked for the for-
mation of endospores during polymer production using the
SchaefferFulton method (Doetsch, 1981). Samples were air-
dried on a glass slide, heat xed, covered with lter paper,
saturated with aqueous malachite green dye (0.5%, w/v) and
placed over a boiling water bath for 5 min. The slide was washed
in tap water and counterstained with safranine for 30 s, followed
by washing under tap water and blot dried using absorbent paper.
The preparation was then examined using a ZEISS-Axioskop
microscope using an oil immersion objective (100).
2.5. Determination of glucose concentration
The Glucose assay kit (GAGO-20, SigmaAldrich, UK) was
used for the monitoring of glucose level.
2.6. Extraction of PHB
For the extraction of PHB, 300 mL of the cells were har-
vested by centrifugation at 5000 g and then lyophilized. The
following methods were then employed.
2.6.1. Soxhlet extraction
For the extraction of PHB, the cell mass was lysed in sodium
hypochlorite (16%, v/v in water) at 37

Cfor 12 h, centrifuged at
10,000 g, and the residue was washed twice each with water,
acetone, ethanol and diethyl ether (Ramsay et al., 1994; Manna
et al., 1999). Finally, the residue was dried and subjected to
soxhlet extraction for 24 h using chloroform as the solvent. The
chloroform containing the PHB was then concentrated by vac-
uum rotary evaporation and precipitated using 10 volumes of
ice-cold methanol. The precipitate obtained was centrifuged and
air-dried.
2.6.2. Chloroform extraction
The polymer was extractedfromcells bystirring1 gof freeze-
dried cell in 100 mL chloroform for 48 h at 37

C and puried
by re-precipitation with 10 volumes of ice-cold methanol (Hahn
et al., 1995).
2.6.3. Chloroformhypochlorite dispersion extraction
PHB was extracted by treating 1 g of freeze-dried cells with
a dispersion containing 50 mL of chloroform and 50 mL of
a diluted (30%) sodium hypochlorite solution in water, in an
orbital shaker at 100 rpm. The cell powder was treated at 38

C
for 1 h. The mixture obtained was then centrifuged at 4000 g
for 10 min, which resulted in three separate phases. PHB was
recovered from the bottom phase, i.e. that of chloroform by pre-
cipitation using 10 volumes of ice-cold methanol (Hahn et al.,
1995).
2.7. Characterisation and quantication of PHB
2.7.1. Solid-state
13
C NMR analysis
13
C NMR cross-polarization with magic-angle sample spin-
ning (CP/MAS) experiments were performed at 75.4 MHz on a
VarianUnityInova spectrometer usinga 5 mmmagic-angle spin-
ning (MAS) probe with the following acquisition and processing
parameters: spinning speed 5.11 kHz, number of acquisitions
316, recycle delay 5 s and contact time 1 ms. Selective spectra of
the quaternary and methyl carbons were obtained using the dipo-
lar dephasing (non-quaternary suppression) pulse sequence
(Opella andFrey, 1979). This studywas carriedout at the Depart-
ment of Chemistry, University of Durham, UK.
2.7.2. Gas chromatography (GC)
For the quantication of the PHB, following slight modi-
cation of the gas chromatographic method of Huijberts et al.,
1994, was employed. One milligramper millilitre of sample and
1 mg/mL methyl benzoate in chloroformwas added to a mixture
of 2 mL of 15% sulphuric acid in methanol (ratio 1:1) at 100

C
for 5 h in a reux. After the reaction, the tubes were cooled on ice
for 5 min, 1.0 mL distilled water was added and the tubes were
vortexed for 1 min. After phase separation the organic phase was
collected and dried over anhydrous sodium sulphate. The analy-
sis was carried out using a Sigma 3Bgas chromatograph (Perkin
Elmer, USA) equipped with a BP21 (SGE Europe Ltd., Kiln
Farm, Milton Keynes, UK) column (50 mlength, 0.25 mminter-
nal diameter and 0.25 mlmthickness). The sample (1 L), in
chloroform, was injected with helium (1 mLmin
1
) as the car-
rier gas. The injector temperature was 225

C and the column

temperature was increased from 40 to 240

C at 20

C/min and
held at the nal temperature for 10 min.
2.7.3. Fourier transform infrared spectroscopy (FTIR)
Freeze-dried, precipitated PHB from B. cereus SPV was
used to prepare KBr discs (sample:KBr, 1:100) (Kemp, 1989).
An FTIR spectrum 1720X spectrometer (Perkin Elmer, USA)
was used under the following conditions: spectral range,
4000400 cm
1
; window material, CsI; 16 scans; resolution
4 cm
1
; the detector was a temperature-stabilized, coated FR-
DTGS detector.
2.7.4. Molecular mass measurements using gel permeation
chromatography (GPC)
Molecular mass analysis was conducted with puried PHB,
which was dissolved in chloroform (1 mg/mL PHB) and intro-
duced into a GPC system. The GPC system was equipped with
a PLgel guard plus 2 mixed bed-B column (30 cm10 m).
The eluted polymer was detected with a differential refractome-
ter. The data were collected and analysed using Viscotek Trisec
2000 and Trisec 3.0 software. This work was carried out at
RAPRA, UK.
2.7.5. Differential scanning calorimetry
Thermal analysis was performed in a Perkin Elmer DSC 7
(Pyris software 3.81). Samples were heated from25 to 200

Cat
a heating rate of 10

C/min (Run 1). After being kept at 200

C
254 S.P. Valappil et al. / Journal of Biotechnology 132 (2007) 251258
for 2 min, the samples were rapidly quenched to 50

C. They
were then heated again at a rate of 10

C/min to 200

C (Run 2).
DSC curves were recorded at Run 2.
3. Results and discussion
3.1. PHB production in shaken asks
To optimise the chance of a successful fermentation at a larger
scale, the process parameters need to be set at the shaken ask
level and translated into a set of standard parameters in large-
scale fermentation.
In shaken ask cultures, the cell mass increased steadily, lead-
ing to a maximum cell density within 24 h of cultivation after
whichthere was a gradual decrease (Fig. 1). The pHof the culture
medium decreased during the growth, from its initial value of 7
to a minimumof 4.5. Cessation of logarithmic growth coincided
with the approach of the pH minimum and rapid consump-
tion of glucose. PHB accumulated rapidly during the stationary
phase and reached a maximum concentration of 38% of dry cell
weight (dcw) at 60 h of growth. We found modied Kannan and
Rehacek medium to be the best for PHB production using the
newly characterised B. cereus SPV (Valappil et al., 2007b) as
compared to other reported minimal media such as potassium
decient production medium (Wakisaka et al., 1982) and phos-
phate decient production medium(Lopez et al., 1986) (data not
shown). Another unique and important observation in this study
is that once maximal PHBconcentration was achieved, the PHB
concentration remained almost constant for a period of 64 h.
Thus, it seems that the PHB produced was not degraded in order
to be utilized for sporulation, in contrast to the report of Wu et
al., 2001. The extent of sporulation was tested at each time point
in this study using the SchaefferFulton method and the results
obtained were always negative. The unbuffered medium used in
this study led to low values of pH for the media (4.55.8) during
the stationary phase, resulting in prevention of PHB degrada-
tion. This observation was consistent to a previous report on
a Bacillus sp. where low pH conditions have been reported to
Fig. 1. Growth and PHB accumulation of the newly identied Bacillus cereus
SPV in shaken asks. Changes in the pH (), dry cell weight (), glucose
() and PHB () when grown in the Kannan and Rehacek medium. All the
measurements were duplicatedandthe relative standarddeviations were 12%.
Fig. 2. Growth and PHB accumulation of the newly characterised Bacillus
cereus SPV in a 20 L fermentor. Changes in the pH (), dry cell weight (),
dissolved oxygen tension (DOT, % air saturation) () and PHB () when the
newly identied Bacillus cereus SPV was grown in the Kannan and Rehacek
medium.
inhibit utilization of the polymer as well as spore formation in
B. cereus T (Kominek and Halvorson, 1965). This observation
in the present study is crucial for the production of PHB using
B. cereus SPV and other PHB producing Bacillus spp.
3.2. PHB production in fermentors (batch)
The successful production of PHBin shaken ask cultures led
to the large-scale production of PHB in batch cultures (Fig. 2).
The two main medium parameters that are known to affect the
PHB yield are the pH of the medium and the dissolved oxy-
gen tension (DOT) (Wu et al., 2001). The pH of the medium
was not controlled leading to a low pH of 5.2. As in the case
of the shaken ask cultures, PHB accumulation was found to
begin after the initial logarithmic growth, after the pHminimum
of the culture was achieved. As expected, PHB production was
accompanied by a simultaneous decrease in glucose concentra-
tion. The organism reached stationary phase at 23 h of growth
and PHB accumulation reached a maximum (29%dcw) at 48 h
of growth after which the concentration remained constant. The
extent of sporulation was also monitored throughout the fermen-
tation using the SchaefferFulton method. No sporulation was
observed; hence, as in the shaken ask cultures, there was no
degradation of the PHB, possibly due to the lack of sporulation.
However, the yield of PHBproduction was not very high leading
to the trial of fed-batch fermentation with an aim to increase the
PHB yield.
3.3. PHB production in fermentors (fed-batch)
The fermentation prole obtained from the batch grown B.
cereus SPV in 20 L fermentors along with the shaken ask
growth analysis was used to develop a simple nutrient feeding
strategy. As seen in Fig. 3, the cell mass increased steadily and
attained its maximum value within 24 h of cultivation. The cul-
ture reached stationary phase at 24 h, which coincided with an
increase in dissolved oxygen concentration. During the growth,
the pH of the culture medium decreased from its initial value of
7 to a minimum of 5.2 (Fig. 3).
S.P. Valappil et al. / Journal of Biotechnology 132 (2007) 251258 255
Fig. 3. Growth and PHB accumulation of the newly identied Bacillus cereus
SPV in 20 L fed-batch fermentation. Changes in pH (), dry cell weight (),
dissolved oxygen tension (DOT, %air saturation) () and PHB() when grown
in the Kannan and Rehacek mediumwith feeding of glucose in order to maintain
nal glucose concentration above 10 g/L.
The two major signicant differences between the batch and
fed-batch fermentations were the time required for maximal
accumulation of PHB, which showed a large shift from 48 to
32 h and the nal PHB yield, which increased from 29%dcw to
a much higher value of 38%dcw, an increase of 31%in the over-
all volumetric yield (Table 1). These two changes would lead to
a much higher level of PHB accumulation within a shorter time
interval leading to a useful decrease in the resources required
for scaling up the production of PHB. Hence, in the fed-batch
fermentation, feeding glucose to maintain a nal concentration
above 10 g/L has successfully improved the large-scale pro-
duction of PHB. PHB concentration has remained constant at
38%dcw once the maximum concentration has been achieved
even after 32 h of growth, consistent with observations made
in shaken ask and batch fermentations. We postulate that low
pH values have resulted in the inhibition of intracellular PHB
depolymerisation, hence preventing the wasteful degradation of
PHB after it has been synthesised.
Also, cell growth and PHAaccumulation need to be balanced
to avoid incomplete accumulation of PHA or premature termi-
nation of PHA synthesis at low cell concentration. The highest
residual cell concentration, practically achievable, with a high
Table 1
Comparison of the PHB yield (% dcw) obtained in the batch and fed-batch
fermentation of newly identied Bacillus cereus SPVin the Kannan and Rehacek
medium
Time (h) Dry cell weight (g/L) PHA yield (% dry cell weight)
Batch Fed-batch Batch Fed-batch
20 2.10 2.40 16.00 17.00
24 2.50 3.00 20.00 26.86
28 2.50 3.00 23.00 30.00
32 2.49 3.00 24.00 38.00
36 2.49 2.90 25.00 38.00
40 2.49 2.80 26.00 38.00
44 2.50 2.75 28.00 38.00
48 2.50 2.73 29.00 38.00
52 2.49 2.68 29.00 38.00
56 2.49 2.84 29.00 37.00
60 2.48 3.00 29.00 38.00
PHA content will give the best results. In the present study,
the highest cell concentration achieved using the production
medium was 2.5 g/L in the batch and 3 g/L in the fed-batch fer-
mentation. To achieve this, the fermentation should be stopped at
48 h in the batch culture (Y
P/X
=0.29 g/g; Y
P/S
=0.071g/g) and at
32 h in the fed-batch culture (Y
P/X
=0.38 g/g; Y
P/S
=0.114 g/g)
since the productivity is highest at these times. As the PHA
content remains stable from this point onwards prolonged
cultivation to achieve higher PHA concentration will be coun-
terproductive.
3.4. Extraction of PHB
After the successful scaling up of PHB production,
the efciency of polymer extraction using hypochlorite
and organic solvents were evaluated. Three different tech-
niques namely chloroform extraction (Ramsay et al., 1994),
chloroformhypochlorite dispersion extraction (Hahn et al.,
1993, 1994, 1995) and soxhlet extraction (Ramsay et al., 1994;
Manna et al., 1999) were tested for the rst time, for PHBextrac-
tion from B. cereus SPV. The soxhlet extraction involves the
extraction of the polymer fromdesiccated cells into chloroform.
The resulting solution is ltered to remove debris, concentrated,
and the polymer precipitated using methanol or ethanol, leav-
ing low molecular weight lipids in solution. In both the soxhlet
and chloroform hypochlorite dispersion techniques, the anionic
surfactant hypochlorite is used to remove cellular debris before
the intracellular polymer is dissolved in chloroform. Hypochlo-
rite extraction is known to cause severe degradation of PHB
leading to a signicant reduction of the polymer chain length
(Berger et al., 1989). However, the extent of this degradation
varies considerably between organisms. Within Gram-negative
organisms, the amount of PHB degraded to a lower molecular
weight compound in A. eutrophus (75% reduction in the num-
ber average molecular weight, M
N
) during the recovery process,
was signicantly higher compared to that in recombinant E. coli
(15%reduction in M
N
). The chloroformextraction method relies
solely on the solubility of the PHB in chloroform and does not
involve any treatment with sodium hypochlorite. This method
is widely used to recover PHB as it results in less polymer
degradation.
In this work, the chloroform extraction method gave the
highest crude yield of the polymer (31%dcw), followed by the
dispersion method (30%dcw) and the lowest yield (27%dcw)
was obtained using soxhlet extraction (Table 2). The level
of PHB purity was then calculated from the total amount
of PHB in the isolated powder as determined by GC com-
pared to the crude weight of polymer powder. The results
were veried by comparing it with standard PHB obtained
from SigmaAldrich Company Ltd. (Dorset, England). It was
found that the purity of PHB extracted was highest for soxh-
let extraction (99%), followed by the chloroformhypochlorite
dispersion extraction (95%) and nally the chloroform extrac-
tion method (92%). This could be attributed to the rigorous
washing with solvents and the physical separation of the
extract from the cellular debris in the soxhlet extraction
method.
256 S.P. Valappil et al. / Journal of Biotechnology 132 (2007) 251258
Table 2
Thermal and molecular characterisation of PHB samples extracted from the Bacillus cereus SPV cells using three different techniques
Sample Yield (%) Purity (%) M
w
PDI T
m
(

C) T
g
(

C) H
f
(J/g) X
c
(%) CI
Soxhlet 27 99 1,100,00 1.75 169.71 2.04 86.140 57.66 0.76
Chloroform 31 92 882,000 2.6 160.83 2.45 81.289 54.42 0.74
ChloroformHOCl dispersion 30 95 885,000 3.1 171.71 2.72 95.72 64.08 0.78
M
w
, molecular weight; PDI, polydispersity index; T
m
, melting temperature; T
g
, glass transition temperature; H
f
, heat of fusion; X
c
, crystallinity; CI, crystallinity
index.
3.5. Characterisation of the extracted PHB
3.5.1.
13
C NMR CP/MAS analysis
The NMR analysis was used to determine the structure of the
isolatedpolymer fromB. cereus SPVgrowninthe modiedKan-
nan and Rehacek medium. Four narrowlines appeared with very
strong intensities which were identical to the CP/MAS
13
CNMR
spectra of PHBreported previously (Doi et al., 1989). These four
peaks were assignable to the methyl (CH
3
; 21.2 ppm), methy-
lene (CH
2
; 42.7 ppm), methine (CH; 68.5 ppm) and carbonyl
(C=O; 169.7 ppm) carbon resonance of PHB (Doi et al., 1986,
1989). This analysis thus conrmed the molecular composition
of the polymer to be PHB.
3.5.2. Differential scanning calorimetry (DSC)
To corroborate the results described above and to exam-
ine possible morphological differences in the PHB obtained by
different extraction techniques, the melting temperatures and
enthalpies of fusion of the PHB samples obtained were deter-
mined using DSC. The thermal properties of the polymer such
as the glass transition temperature (T
g
) and the melting tem-
perature (T
m
) are crucial for polymer processing. The melting
temperature obtained for PHBextracted using the three different
methods were slightly lower compared to what is reported in the
literature, between 173 and 180

C (Hahn et al., 1995; Sudesh

et al., 2000; Williams and Martin, 2005) (Table 2). Among the
three techniques used, the polymer isolated using the soxhlet
and the chloroformhypochlorite dispersion methods gave sim-
ilar T
m
values (169.71 and 171.71

C, respectively) as compared
to a relatively lower value for the sample obtained using the
chloroform extraction technique (160.83

C). The glass transi-

tion temperature measured for the PHBextracted using the three
different methods were also slightly lower compared to what is
reported in the literature, 4

C (Sudesh et al., 2000; Williams

and Martin, 2005) (Table 2). As in the case of T
m
values, the T
g
values measured for the polymer isolated using the soxhlet and
the chloroformhypochlorite dispersion methods gave similar
values (2.04 and 2.72

C, respectively) as compared to a lower

value for the sample extracted using the chloroform extraction
technique (2.45

C). The chloroformextraction technique uses

the least amount of solvents during extraction and may allow
lipids, fatty acids and other hydrophobic cellular materials to
be extracted along with PHB. The presence of such impurities
might have led to the different thermal properties.
The percentage crystallinity of the polymer was also deter-
mined by measuring the enthalpy of fusion (H
fusion
) for the
extracted polymer by calculating the area under the melting
curve. The value of the H
fusion
for 100% crystalline PHB is
known to be 149.37 J/g (Barham et al., 1984), hence percentage
crystallinity of the isolated polymer can be calculated accord-
ingly. Although PHB found within the bacteria are amorphous,
when isolated from bacteria it is known to form crystalline
structure due to freeze-drying or aqueous solvent treatment
employed during recovery (Hahn et al., 1995). In the present
study, the crystallinity obtained for the PHB extracted using the
three different methods were found to be in the lower range of
that reported in the literature, 5580% crystallinity (Holmes,
1998) (Table 2). Among the three extraction techniques used,
the chloroformhypochlorite dispersion method gave the high-
est value (64.08%) compared to the soxhlet extraction (57.66%)
and the chloroform extraction (54.42%). This result indicates
that the crystallinity of the PHB extracted from B. cereus SPV
depends on the extraction technique.
3.5.3. FTIR analysis
The absorption bands at 1728 cm
1
, corresponding to the
ester carbonyl groupandat 1282 cm
1
correspondingtothe CH
group, characteristic of PHB (Hong et al., 1999), were identi-
ed in the polymer isolated from B. cereus SPV (Fig. 4). FTIR
analysis was then used to measure the degree of crystallinity of
the polymer (Galego et al., 2000). The crystallinity index (CI) is
dened as the ratio of the intensities of the bands at 1382 cm
1
(CH
3
), which are insensitive to the degree of crystallinity, to
Fig. 4. The FTIR spectrum of a PHB sample in the form of a KBr pellet.
S.P. Valappil et al. / Journal of Biotechnology 132 (2007) 251258 257
that at 1185 cm
1
(COC), which is sensitive to the amorphous
state (Galego et al., 2000). The CI of PHB isolated using all the
three techniques was determined to be in the range 0.740.78
(Table 2) which was much lower than the previously reported
value of 0.949 for PHB (Galego et al., 2000). Crystallinity of
a polymer is known to play a major role in the degradation of
a polymer (Iannace et al., 2001). The amorphous regions of a
polymer degrade at a much faster rate as compared to the crys-
talline regions. Also, the degraded crystalline by-products and
debris of polymer systems can induce an inammatory response
when used for tissue engineering (TE) applications (Yang et al.,
2001). Hence, the relatively lower crystallinity of the PHB iso-
lated from B. cereus SPV is a favourable property for use in TE
applications.
3.5.4. GPC analysis
GPC analysis showed that the PHB extracted using the soxh-
let extraction technique was of relatively high weight average
molecular weight (M
W
=1.1 10
6
) and number average molec-
ular weight (M
N
=0.339 10
6
) compared to the PHB produced
by Bacillus sp. INT005 grown on glucose containing medium
(M
W
=0.525 10
6
and M
N
=0.281 10
6
) (Tajima et al., 2003).
The polydispersity index (M
W
/M
N
) was measured to be 1.75,
a value comparable to that obtained for the PHB obtained
from Bacillus sp. INT005 using different carbon sources. Com-
paring the weight average molecular weights (Table 2) of
the polymer isolated using the soxhlet extraction technique
(M
W
=1.1 10
6
) to that of the chloroformhypochlorite disper-
sion method (M
W
=0.885 10
6
) and the chloroform extraction
method (M
W
=0.882 10
6
) conrms that the crystalline mor-
phology developed by freeze-drying and the aqueous solvent
treatment protected PHB molecules from being digested by
hypochlorite. Korsatko et al. (1987) evaluated the inuence of
PHB molecular weight on drug release, and found the release
of the antihypersensitive drug midodrin-HCl from compressed
tablets increased with the molecular weight of the polymer. Fur-
ther, the PHBcoatedgranules of the beta blocker, Celiprolol-HCl
prepared using uid bed dryer systemwere found to be effective
in slowrelease when polymer with lowpolydispersity index was
used (Korsatko et al., 1987). Hence, the lowpolydispersity index
of 1.75 obtained for the PHBproduced in this study compared to
what is reported for most other PHBproducing micro-organisms
(Doi, 1990) can further extend the use of this PHB in controlled
drug delivery applications.
4. Conclusion
The present study shows that in B. cereus SPV, acidic
pH ensured no degradation of the accumulated PHB which
are known to be depolymerised under alkaline pH conditions
(Nakata, 1963) and in turn support sporulation (Wu et al., 2001).
In order to increase the production of the PHB from B. cereus
SPV, a simple feeding strategy of glucose was attempted which
increased the overall volumetric yield by 31% compared to the
batch fermentation. This yield can now be further increased by
optimising fermentation parameters such as DOT levels. In this
work, we observe much lesser molecular degradation of PHB
by the hypochlorite process (polydispersity index of 1.75 using
16%hypochlorite concentration) in B. cereus SPV, as compared
to Gram-negative organisms reported previously (increase of
polydispersity index from 1.9 to 4.8 for A. eutrophus and 2 to
2.5 for E. coli as the concentration of hypochlorite increased
from 0 to 20% (Hahn et al., 1995)). This could be postulated
to be due to better protection of the polymer provided by the
Gram-positive cell wall as compared to the Gram-negative one
and needs further investigation. Moreover, the different extrac-
tion techniques employed in this study were found to have an
effect on the thermal and molecular properties of the PHB iso-
lated. Thus, the present study conrms the potential of B. cereus
SPV to produce PHB in a large scale, with a good volumet-
ric yield and without signicant intracellular degradation. The
polymer produced can then be extracted using any of the three
techniques, all yielding highly pure polymer and with relatively
lower crystallinity. The soxhlet technique yields polymer with
a slightly higher molecular weight and lower PDI value, prop-
erties preferred for certain applications such as controlled drug
release.
Acknowledgements
This work was supported by the EPSRC, UK grant no.
EP/C515617/1(P). S.P. Valappil was also provided nancial sup-
port by the University of Westminster, London, UK. S.K. Misra
was supported by ORS award. We thank Dr. Showan N. Nazhat
from Eastman Dental Institute, University College London, for
the analysis of PHB samples.
References
Anderson, A.J., Dawes, E.A., 1990. Occurrence, metabolism, metabolic role,
and industrial uses of bacterial polyhydroxyalkanoates. Microbiol. Rev. 54,
450472.
Barham, P., Keller, A., Otun, E.L., Holmes, P., 1984. Crystallization and mor-
phology of a bacterial thermoplastic: poly-3-hydroxybutyrate. J. Mater. Sci.
19, 27812794.
Berger, E., Ramsay, B.A., Ramsay, J.A., Chavarie, C., Braunegg, G., 1989. PHB
recovery and hypochlorite digestion of non-PHBbiomass. Biotechnol. Tech.
3, 227232.
Chen, G.Q., Wu, Q., 2005. The application of polyhydroxyalkanoates as tissue
engineering materials. Biomaterials 26, 65656578.
Doetsch, R.N., 1981. Determinative methods of light microscopy. In: Gerhardt,
P., Murray, R.G.E., Costilow, R.N., Nester, E.W., Wood, W.A., Krieg, N.R.,
Phillips, G.B. (Eds.), Manual of Methods for General Bacteriology. Ameri-
can Society for Microbiology, Washington, DC, pp. 2133.
Doi, Y., 1990. Microbial Polyesters. VCH Publishing, New York.
Doi, Y., Kawaguchi, Y., Nakamura, Y., Kunioka, M., 1989. Nuclear magnetic res-
onance studies of poly(3-hydroxybutyrate) and polyphosphate metabolism
in Alcaligenes eutrophus. Appl. Environ. Microbiol. 55, 29322938.
Doi, Y., Kunioka, M., Nakamura, Y., Soga, K., 1986. Conformational analysis
of a poly(-hydroxybutyrate) in Alcaligenes eutrophus by solid-state
13
C
NMR spectroscopy. Makromol. Chem. 7, 661664.
Galego, N., Rozsa, C., Sanchez, R., Fung, J., Vazquez, A., Tomas, J.S., 2000.
Characterization and application of poly(-hydroxyalkanoates) family as
composite biomaterials. Polym. Test. 19, 485492.
Hahn, S.K., Chang, Y.K., Kim, B.S., Chang, H.N., 1994. Optimization of
microbial poly (3-hydroxybutyrate) recovery using dispersions of sodium
hypochlorite solution and chloroform. Biotechnol. Bioeng. 44, 256261.
258 S.P. Valappil et al. / Journal of Biotechnology 132 (2007) 251258
Hahn, S.K., Chang, Y.K., Kim, B.S., Lee, K.M., Chang, H.N., 1993. The recovery
of poly (3-hydroxybutyrate) by using dispersion of sodium hypochlorite
solution and chloroform. Biotechnol. Tech. 7, 209212.
Hahn, S.K., Chang, Y.K., Lee, S.Y., 1995. Recovery and characterization of poly
(3-hydroxybutyric acid) synthesized in Alcaligens eutrophus and recombi-
nant Escherichia coli. Appl. Environ. Microbiol. 61, 3439.
Holmes, P.A., 1998. Biologically produced (R)-3-hydroxyalkanoate polymers
and copolymers. In: Bassett, D.C. (Ed.), Developments in Crystalline Poly-
mers. Elsevier, London, pp. 165.
Hong, K., Sun, S., Tian, W., Chen, G.Q., Huang, W., 1999. A rapid method
for detecting bacterial polyhydroxyalkanoates in intact cells by Fourier
transform infrared spectroscopy. Appl. Microbiol. Biotechnol. 51, 523
526.
Huijberts, G.N.M., Wal, H.V., Wilkinson, C., Eggink, G., 1994. Gas-
chromatographic analysis of poly(3-hydroxyalkanoates) in bacteria.
Biotechnol. Tech. 8, 187192.
Iannace, S., Maffezzoli, A., Leo, G., Nicolais, L., 2001. Inuence of crystal and
amorphous phase morphology on hydrolytic degradation of PLLAsubjected
to different processing conditions. Polymer 42, 37993807.
Kannan, L.V., Rehacek, Z., 1970. Formation of poly--hydroxybutyrate by
Actinomycetes. Indian J. Biochem. 7, 126129.
Kemp, W., 1989. Organic Spectroscopy. Macmillan Education Ltd., London.
Kominek, L.A., Halvorson, H.O., 1965. Metabolismof poly--hydroxybutyrate
and acetoin in Bacillus cereus. J. Bacteriol. 90, 12511259.
Korsatko, W., Korsatko, B., Lafferty, R.M., Weidmann, V., 1987. The molecular
weight of poly-d()-3-hydroxybutyric acid on its use as a retard matrix for
sustained drug release. In: Proceedings of the Third European Congress of
Biopharmacology and Pharmacokinetics, vol. 1, pp. 234242.
Lopez, J.G., Rubia, T.D., La Ballesteros, F., Cormenzana, A.R., 1986. Growth
of Bacillus megaterium in phosphate limited medium. Folia Microbiol. 31,
98105.
Manna, A., Banarjee, R., Paul, A.K., 1999. Accumulation of poly (3-
hydroxybutyric acid) by some soil Streptomyces. Curr. Microbiol. 39,
153158.
Nakata, H.M., 1963. Effect of pH on intermediates produced during growth and
sporulation of Bacillus Cereus. J. Bacteriol. 86, 577581.
Opella, S.J., Frey, M.H., 1979. Selection of nonprotonated carbon resonances in
solid-state nuclear magnetic resonance. J. Am. Chem. Soc. 101, 58545856.
Ramsay, J.A., Berger, E., Voyer, R., Chavarie, C., Ramsay, B.A., 1994. Extrac-
tion of poly-3-hydroxybutyrate using chlorinated solvents. Biotechnol. Tech.
8, 589594.
Sudesh, K., Abe, H., Doi, Y., 2000. Synthesis, structure and properties of poly-
hydroxyalkanoates: biological polyesters. Prog. Poly. Sci. 25, 15031555.
Tajima, K., Igari, T., Nishimura, D., Nakamura, M., Satoh, Y., Munekata, M.,
2003. Isolation and characterization of Bacillus sp INT005 accumulating
polyhydroxyalkanoate (PHA) from gas eld soil. J. Biosci. Bioeng. 95,
7781.
Valappil, S.P., Boccaccini, A.R., Bucke, C., Roy, I., 2007a. Polyhydroxyalka-
noates in Gram-positive bacteria: insights from the genera Bacillus and
Streptomyces. Antonie van Leeuwenhoek 91, 117.
Valappil, S.P., Peiris, D., Langley, G.J., Herniman, J.M., Boccaccini, A.R.,
Bucke, C., Roy, I., 2007b. Polyhydroxyalkanoate (PHA) biosynthesis from
structurally unrelated carbon sources by a newly characterized Bacillus spp.
J. Biotechnol. 127, 475487.
Vandamme, P., Coenye, T., 2004. Taxonomy of the genus Cupriavidus: a tale of
lost and found. Int. J. Syst. Evol. Microbiol. 54, 22852289.
Wakisaka, Y., Masaki, E., Nishimoto, Y., 1982. Formation of -endotoxin or
poly--hydroxybutyric acid granules by asporogenous mutants of Bacillus
thurungiensis. Appl. Environ. Microbiol. 43, 14731480.
Williams, S.F., Martin, D.P., 2005. Applications of PHAs in medicine and
pharmacy. In: Steinbuchel, A., Marchessault, R.H. (Eds.), Biopolymers
for Medical and Pharmaceutical Applications. WileyVCH, Weinheim, pp.
89125.
Wu, Q., Huang, H., Hu, G.H., Chen, J., Ho, K.P., Chen, G.Q., 2001. Production of
poly-3-hydroxybutyrate by Bacillus sp. JMa5 cultivated in molasses media.
Antonie van Leeuwenhoek 80, 111118.
Yang, S., Leong, K., Du, Z., Chua, C., 2001. The design of scaffolds for use in
tissue engineering Part I. Traditional factors. Tissue Eng. 7, 679689.