Archives of Biochemistry and Biophysics 403 (2002) 284–291

ABB
www.academicpress.com

Biochemical and enzymological properties of the

polyhydroxybutyrate synthase from the extremely halophilic
archaeon strain 56
uchel, and Bernd H.A. Rehm*
Francis F. Hezayen, Alexander Steinb€
Institut f€
ur Mikrobiologie der Westf€alischen, Wilhelms-Universit€at M€unster, Corrensstr. 3, D-48149 M€unster, Germany

Received 17 April 2002, and in revised form 17 May 2002

Abstract

Some members of the archaebacterial family Halobacteriaceae have been determined to accumulate polyhydroxyalkanoate
(PHA) and poly(3-hydroxybutyrate) (PHB). The extremely halophilic archaebacterium strain 56 is capable of accumulating large
amounts of PHB. Since measurements of enzyme activities related to archaebacterial PHB biosynthesis have never been achieved, we
investigated the enzymology of PHB biosynthesis in strain 56. Crude extracts of strain 56 cultivated under accumulating conditions
showed PHB synthase activity, whereas neither b-ketothiolase nor NADH/NADPH-dependent acetoacetyl–CoA reductase activity
was detectable. An 80-kDa protein, cross-reacting with the anti-PHB synthase antibodies raised against the PHB synthase from
Ralstonia eutropha, was identified in the crude extract and was strongly enriched by purification of PHB granules. The granule-
associated PHB synthase was enzymologically characterized. Enzyme kinetics showed a specific activity of about 4.6 U/mg and Hill
plot analysis revealed a K0:5 of 56 lM with (R)-3-hydroxybutyryl–CoA employed as substrate. A Hill coefficient of 1.75 indicated that
the PHB synthase exhibited positive cooperativity. The thioesters 3-hydroxyvaleryl–CoA, 4-hydroxybutyryl–CoA, and 3-hy-
droxydecanoyl–CoA were not accepted as substrates. Moreover, the PHB synthase was found to be competitively inhibited by CoA,
showing an IC50 of 160 lM. The PHB synthase was stable up to 60 °C and still exhibited about 90% of the maximum enzyme activity,
which was obtained at 40 °C. In contrast to the soluble PHB synthase, the granule-bound PHB synthase was almost independent of
the salt concentration. The PHB synthase could not be released from the PHB granules, indicating a covalent attachment to the PHB
core. This is the first description of an archaebacterial PHA synthase. Ó 2002 Elsevier Science (USA). All rights reserved.

Keywords: PHA synthase; PHA; Polyhydroxyalkanoate; Polyhydroxybutyrate; PHB; Extremely halophilic; Archaebacteria

Archaea are a group of ancient organisms that live in structurally intact and functionally active at high salt
adverse environmental conditions such as high salinity, concentrations only [5–10].
high temperature, and high or low pH conditions [1]; the Some members of the family Halobactericeae have
haloarchaea grow in nearly NaCl-saturated envi- been shown to accumulate large amounts of the intra-
ronments and require at least 1.5 M NaCl for growth [2– cellular reserve polymer poly(3-hydroxybutyrate) (PHB)1
4]. This group of organisms is known as ‘‘extreme [11–13]. PHB belongs to the polyhydroxyalkanoates
halophiles.’’ The intracellular salt concentration was (PHAs), that are common carbon energy storage mate-
found to be very high in extreme halophiles [2]. As a rials produced by numerous bacterial species [14]. PHA is
consequence, universally conserved molecules such as now under intensive investigation because of its inherent
proteins and nucleic acids may have also adapted to property as a biodegradable thermoplastic. More than 54
function at such a high salt concentration [5]. Indeed, PHA synthase genes which can be broadly classified into
several of the functional proteins in haloarchaea have three different classes based on their subunit composition
been reported to be halo-adapted in that they are
1
Abbreviations used: PHB, poly(3-hydroxybutyrate); PHA, polyhy-
*
Corresponding author. Fax: +49-251-833-8388. droxyalkanoate; scl, short-chain-length; Buk, butyrate kinase; Ptb,
E-mail address: [email protected] (B.H.A. Rehm). phosphotransbutyrylase; DTNB, 5,50 -dithiobis(2-nitrobenzoic acid).

0003-9861/02/$ - see front matter Ó 2002 Elsevier Science (USA). All rights reserved.
PII: S 0 0 0 3 - 9 8 6 1 ( 0 2 ) 0 0 2 3 4 - 5
 F.F. Hezayen et al. / Archives of Biochemistry and Biophysics 403 (2002) 284–291 285

and substrate specificities have been assigned and adjusted to 7.2 with 1 N NaOH, and, if required to
characterized [15,16]. Class I PHA synthase is active to- generate PHB-accumulating conditions, the strain was
ward short-chain-length (scl) (R)-hydroxyacyl–CoA grown in basal salt medium contained the following
thioesters, consisting of 3 to 5 carbon atoms, and is rep- ðg=L1 Þ: NaCl, 250; MgSO4  7  H2 O, 20; KCl, 2; yeast
resented by the PHA synthase of Ralstonia eutropha. extract, 1; and supplemented with 1% (w/v) sodium ac-
Class II, which is represented by the PHA synthase from etate and 1% (v/v) butyric acid [13]. The pH was ad-
Pseudomonas aeruginosa, is active toward medium-chain- justed to the proper value using NaOH and the media
length (R)-3-hydroxyacyl–CoA thioesters, containing 6 were sterilized by autoclaving. Growth was conducted at
to 14 carbon atoms. PHA synthases of class I and class II 40 °C for 8 days at 200 rpm.
are composed of a single subunit (PhaC), whereas class III
PHA synthases, represented by the Allochromatium Isolation and purification of PHB granules
vinosum PHA synthase, comprise two not identical su-
bunits, PhaC and PhaE, exhibiting activity toward scl (R)- Cells were cultivated under PHB-accumulating
3-hydroxyacyl–CoA. Although numerous PHA synthase conditions and harvested by centrifugation. Cells were
genes have been cloned and assigned, only a few PHA subjected to either cell lysis in distilled water or cell
synthases have been purified and enzymatically charac- disrupture 2 by using a French press. For disrupture
terized, such as PHA synthases from R. eutropha, by using a French press, cells were resuspended in
A. vinosum, and P. aeruginosa [17–21]. 50 mM Tris–HCl, pH 7.5, containing 25% (w/v) NaCl,
The biosynthesis pathway of PHB has been studied in 0.2% (w/v) KCl, 0.5% (w/v) MgSO4  ðH2 OÞ7 . Dis-
detail in R. eutropha [22–26]. In this bacterium, biosyn- rupted cells were subjected to centrifugation 10,000g,
thesis is initiated by the condensation of two acetyl–CoA 30 min) to sediment intact cells and cell debris. PHB
molecules to acetoacetyl–CoA catalyzed by the enzyme granules were then sedimented by ultracentrifugation
b-ketothiolase (EC 2.3.1.9). The gene for this enzyme is (100,000g, 1 h). To enhance purity, PHB granules were
designated phbA. Acetoacetyl–CoA is then reduced to washed in the same buffer and centrifuged again
(R)-3-hydroxybutyryl–CoA by the NADPH-dependent (100,000g, 30 min). The granules were suspended in a
acetoacetyl–CoA reductase (EC 1.1.1.36); the gene was suitable volume of the same buffer for further use.
designated phbB. These enzymes have been found and
studied in several PHB-accumulating bacteria [15]. Sodium dodecyl sulfate (SDS)–polyacrylamide gel elec-
In this study, we investigated enzymes involved in PHB trophoresis (PAGE) and Western immunoblotting
biosynthesis of the extremely halophilic archaebacterium
strain 56, which accumulated PHB contributing to 53% of SDS–PAGE was performed according to Sambrook
cellular dry weight [13,27]. A protocol for the isolation of et al. [28]. Proteins were separated in 12.5 or 15% (w/v)
PHB granules was established, and the in vitro activity of SDS–polyacrylamide gels and stained with Coomassie
the granule-associated PHB synthase, the first PHA syn- brilliant blue R-250. Western blotting was performed
thase from an archaebacterium, was detected. Further- using the Semidry Fastblott (Biometra, Germany). On
more, the granule-associated PHB synthase was Western blots [29] using nitrocellulose membranes, anti-
biochemically and enzymologically characterized. PHA synthase antibody cross-reacting proteins in crude
extracts from strain 56 were detected by applying the
respective monospecific, polyclonal anti-PhaC antise-
Materials and methods rum raised against the PHA synthase from R. eutropha
and an alkaline-phosphatase-antibody conjugate as
Bacterial strain second antibody. Phasin antibody cross-reacting pro-
teins in crude extracts from strain 56 were detected by
Strain 56 (DSM 13151) was recently isolated from a applying the respective monospecific, polyclonal anti-
soil sample collected from the surface of hyper-saline GA24 antiserum raised against the phasin GA24 from
soil close to Aswan city (Egypt). The isolate is related to R. eutropha and an alkaline-phosphatase-antibody
extremely halophilic archaebacteria based on 16S rDNA conjugate as second antibody. Bound antibodies were
sequencing, lipid analysis and physiological properties detected using nitro blue tetrazolium chloride and the
[27]. toluidine salt of 5-bromo-4-chloro-3-indolyl phosphate.

Media and growth conditions Purification of denatured PHB synthase and N-terminal
amino acid sequencing
Strain 56 was isolated and grown in S-G medium [1]
which contained the following ðg=L1 Þ: NaCl, 250; PHB granules were isolated from PHB-accumulating
MgSO4  7  H2 O, 20; KCl, 2; trisodium citrate, 3; yeast strain 56 as described above and were subjected to
extract, 10 (Difco); and casamino acids, 7.5. The pH was preparative SDS–PAGE. Proteins were separated in
286 F.F. Hezayen et al. / Archives of Biochemistry and Biophysics 403 (2002) 284–291

12.5 or 15% (w/v) SDS–polyacrylamide column gels by b-Ketothiolase assay

applying the PrepCell 491 (Bio-Rad) and subsequent
fractionation. Crude extracts of strain 56 were subjected to the b-
ketothiolase assay. The assay was conducted according to
Gas chromatographic (GC) analysis of polyester in cells Nishimura et al. [35]. Briefly, the reaction mixture con-
tained 50 mM Tris–HCl, pH 7.5, 50 lM MgCl2 , 50 lM
PHA was qualitatively and quantitatively analyzed acetoacetyl–CoA, 100 lM CoA, 0.2 % (w/v) KCl, 25%
by gas chromatography. Liquid cultures containing (w/v) NaCl, and crude extract. The assay was started by
PHB-accumulating cells were centrifuged after 8 days the addition of CoA, and enzyme activity was monitored
incubation at 10,000g for 15 min, then the cells were by measuring spectrophotometrically at 303 nm the
washed twice in saline and lyophilized overnight and decrease of the acetoacetyl–CoA concentration. The
8–10 mg lyophilized cell material was subjected to conversion of 1 lmol acetoacetyl–CoA to 2 acetyl–CoA
methanolysis in the presence of 15% (v/v) sulfuric acid. per minute corresponded to an enzyme activity of 1U.
The resulting methyl esters of the constituent 3-hy-
droxyalkanoic acids were assayed by GC according to PHA synthase assay
Brandl et al. [30] and as described in detail previously
[31]. GC analysis was performed by injecting 3 ll of PHA synthase was determined spectrophotometrical-
sample into a Perkin–Elmer 8420 gas chromatograph ly by monitoring the release of CoA at 412 nm [36]. The
(U€ berlingen, Germany) using a 0.5-lm-diameter standard assay contained 1 mM DTNB, 20 mM MgCl2 ,
Permphase PEG 25 Mx capillary column 60 m in 25% (w/v) NaCl, 0.2% (w/v) KCl, 0.5% (w/v) MgSO4 
length. ðH2 OÞ7 , purified granules and 50 lM of the respective
CoA thioester in 150 mM Tris–HCl, pH 7.5. One U of
Enzymatic preparation of 3HB-CoA, 4HB-CoA, and enzyme is defined as the amount which produced 1 lmol
3HV-CoA product per minute under the assay conditions employed.

3HB-CoA, 4HB-CoA, and 3HV-CoA could be pre-

pared from the corresponding hydroxy fatty acids by Results
coupling butyrate kinase (Buk) and phosphotransbu-
tyrylase (Ptb) from Clostridium acetobutylicum. These Analysis of enzyme activities related to PHB biosynthesis
enzymes were purified according to [32,33]. With Buk,
the hydroxy fatty acids are converted to the corre- Since the extremely halophilic archaeon strain 56 is
sponding hydroxyacylphosphates at the expense of ATP capable of synthesizing large amounts of PHB [13] when
consumption. Ptb then converts the hydroxyacylphos- excess carbon source, such as, e.g., starch or butyrate, is
phates and CoA to the corresponding hydroxyacyl–CoA available, we analyzed crude extracts of this archaebac-
thioesters. As shown for the preparation of 3HB-CoA, terium with respect to enzyme activities that are involved
more ATP than CoA was consumed, indicating excess in PHB biosynthesis [26]. Crude extracts from strain 56
formation of 3-hydroxybutyrylphosphate. The decrease grown under conditions which favor PHB accumulation
of CoA was a direct indication of the formation of 3HB- contributing to about 50% of cellular dry weight were
CoA. The synthesis was performed with 40 mM 3HA or subjected to enzymatic analysis. PHB accumulation was
4HA, 40 mM ATP, and 10 mM CoA, in a total volume confirmed by GC analysis. Enzymatic analysis of the PHB
of 2.5 ml as described previously [32]. synthase activity clearly showed an enzymatic activity of
about 3 U/mg crude extract protein. In contrast, analysis
Acetoacetyl–CoA reductase assay of the b-ketothiolase activity using acetoacetyl–CoA as
substrate in the presence or absence of NaCl or KCl
Crude extracts of strain 56 were subjected to the indicated no b-ketothiolase activity. Analysis of aceto-
acetoacetyl–CoA reductase assay. The assay was con- acetyl–CoA reductase activity employing either NADH
ducted according to Lynen and Wieland [34]. Briefly, the or NADPH as cofactor in the presence or absence of NaCl
reaction mixture contained 50 mM Tris–HCl, pH 7.5, or KCl also did not indicate acetoacetyl–CoA reductase
250 lM NAD(P)H, 50 lM acetoacetyl–CoA, 25% (w/v) activity. Enzymatic analysis of the PHB synthase activity
NaCl, 0.2% (w/v) KCl, 0.5% (w/v) MgSO4  ðH2 OÞ7 , and clearly showed an enzymatic activity of about 3 U/mg
crude extract. The assay was started by the addition of crude extract protein.
acetoacetyl–CoA, and enzyme activity was monitored by
measuring spectrophotometrically the change of ab- Isolation of PHB granules and immunological analysis
sorption at 365 nm due to the oxidation of NAD(P)H to
NADðPÞþ . The oxidation of 1 lmol NAD(P)H per min- Since PHB synthases have been described as being
ute corresponded to an enzyme activity of 1 unit (U). attached to the granule surface [15,16], and to enrich the
 F.F. Hezayen et al. / Archives of Biochemistry and Biophysics 403 (2002) 284–291 287

PHB synthase for enzymological analysis, two protocols but only in crude extracts from PHB-accumulating cells
for the isolation of PHB granules were developed. One and with isolated PHB granules (data not shown).
protocol took advantage of the possibility to simply lyse
the cells in distilled water and separate cell debris and Purification of the putative PHB synthase
PHB granules by differential centrifugation. This pro-
tocol might cause inactivation of salt-dependent en- The strongly cross-reacting 80-kDa protein was pu-
zymes. Thus, a second protocol was set up using a rified by preparative SDS–PAGE to electrophoretic
French press with subsequent differential ultracentrifu- homogenity and subjected to N-terminal amino acid
gation. Both protocols resulted in enrichment of pro- sequence analysis using Edman degradation. The fol-
teins, the synthesis of which was induced under PHB lowing N-terminal sequence was obtained: M E N D I A
accumulation conditions as shown by SDS–PAGE T V L D V Q E V A ? N ? (T) A. Blast (using the ‘‘short
analysis (Fig. 1). Moreover, both preparations resulted but nearly exact match’’ option) and Pattern search
in a protein sample which exhibited PHB synthase ac- analysis employing the currently publicly available da-
tivity. Crude extracts obtained from cells grown either tabases and the incomplete Haloferax volcanii genome
under non-PHB-accumulating conditions (S-G medium) database (Integrated Genomics) did not reveal any sig-
or under PHB-accumulating conditions (basic medium nificant sequence similarity. Refolding experiments
with additional carbon source) [13] and isolated PHB removing the SDS by dialysis against 50 mM potassium
granules were subjected to SDS–PAGE and immuno- phosphate buffer, pH 7.0, containing 25% (w/v) NaCl,
blotting analysis. SDS–PAGE analysis showed the 0.2% (w/v) KCl, 0.5% (w/v) MgSO4  ðH2 OÞ7 did not
enrichment of an about 80 kDa protein in crude extracts result in an active enzyme (data not shown).
from PHB accumulating cells and of about 50- and
28-kDa proteins in isolated PHB granules (Fig. 1). Im- Enzymological characterization of the granule-bound
munoblotting analysis was performed using the mono- PHB synthase
specific anti-PHB synthase antiserum raised against the
PHB synthase from R. eutropha. An about 80-kDa PHB granules were isolated as described above using a
protein was detected, which showed strong cross-reac- French press and subjected to enzymological character-
tion with the anti-PHB synthase antiserum (Fig. 1). ization of the PHB synthase. Kinetics of the 3HB-CoA
Moreover, we employed the anti-phasin antiserum, turnover indicated that the enzyme did not show a lag
which was raised against the phasin GA24 from R. eu- phase, exhibiting a PHB synthase activity of 4.6 U/mg
tropha. A cross-reacting 70-kDa protein was detected, granule-bound protein and an estimated specific PHB
synthase activity of 30 U/mg (Fig. 2). Enzyme kinetics
measuring the activity of the granule-bound PHB syn-
thase at various 3HB-CoA concentrations showed
sigmoidal kinetics, and Hill plot analysis revealed a K0:5
of 56 lM for 3HB-CoA (Fig. 3). Furthermore, a Hill
coefficient of 1.75 indicated that the PHB synthase
exhibited positive cooperativity. Further thioesters, such
as 3-hydroxyvaleryl–CoA, 4-hydroxybutyryl–CoA, and
3-hydroxydecanoyl–CoA, were also tested as substrates
for the granule-bound PHB synthase, but none of them

Fig. 1. SDS–PAGE and immunoblotting analysis of crude extracts and

PHB granules from strain 56. Lanes: 1, crude extract from cells grown
under non-PHB-accumulating conditions; 2, crude extract from cells
grown under PHB-accumulating conditions; 3, PHB granules; 4 and 5, Fig. 2. Time course of CoA release by granule-bound PHB synthase
immunoblotting with anti-PHB synthase antibodies corresponding to from 75 lM 3-hydroxybutyryl–CoA. CoA release was monitored in a
lanes 1and 2. The arrow indicates the position of the anti-PHB syn- continuous assay using DTNB [34]; 10 lg=ml PHB granule-bound
thase antibody cross-reacting protein. Molecular weight standards are PHB synthase was applied in the presence of 25% (w/v) NaCl, 0.2%
indicated on the left. (w/v) KCl, and 0.5% (w/v) MgSO4 .
288 F.F. Hezayen et al. / Archives of Biochemistry and Biophysics 403 (2002) 284–291

(Fig. 4B). Strain 56 was able to grow at 55 °C, suggesting

that enzymes of this archaebacterium are rather stable
even at higher temperatures. Thus, we recorded a tem-
perature profile of the PHB synthase (Fig. 5), which
showed that the granule-bound enzyme still exhibited
about 50% of its activity at 80 °C. However, stability of
the PHB synthase was strongly affected above 60 °C
(Fig. 5). The maximum PHB synthase activity was ob-
tained at 40 °C, which coincided with the best growth
temperature of strain 56 (Fig. 5). Strain 56 is related to
the extremely halophilic archaebacteria and required for
growth at least 10% (w/v) NaCl [27]. Enzymes from these
Fig. 3. In vitro activity of granule-bound PHB synthase from strain 56 bacteria are usually strictly dependent on high salt
employing various concentrations of the substrate 3-hydroxybutyryl– concentrations. Therefore, the dependency of the PHB
CoA; 10 lg=ml PHB granule-bound PHB synthase was applied in the synthase on the salt concentration was analyzed. Inter-
presence of 25% (w/v) NaCl, 0.2% (w/v) KCl, and 0.5% (w/v)
MgSO4  7H2 O. The inset represents the corresponding Hill plot.
estingly, the granule-bound PHB synthase activity was
almost independent of the NaCl concentration in the
presence of 0.2% (w/v) KCl, and 0.5% (w/v) MgSO4 
led to detectable enzyme activity. Since some PHA syn- ðH2 OÞ7 (Table 1; Fig. 6). In the absence of salts the
thases have been found to be inhibited by CoA, we granule-bound PHB synthase still exhibited about 40%
investigated the effect of CoA on the archaebacterial of the maximum activity. Enzyme activity was strongly
granule-bound PHB synthase. Increasing CoA concen- enhanced by slightly increasing the ionic strength inde-
trations in the PHB synthase assay showed an inhibitory pendent of the salt (Table 1). In contrast to the granule-
effect on the granule-bound PHB synthase, and a IC50 of
160 lM was determined (Fig. 4A). To investigate the
mode of inhibition by CoA, the PHB synthase activity
was measured in the presence of a defined CoA
concentration and increasing substrate concentration

Fig. 5. PHB synthase activity and stability as a function of tempera-

ture; 10 lg=ml PHB granule-bound PHB synthase were applied in the
presence of 25% (w/v) NaCl, 0.2% (w/v) KCl, 0.5% (w/v)
MgSO4  7H2 O, and 75 lM 3-hydroxybutyryl–CoA as substrate.

Table 1
Effect of various salts on PHA synthase activity
NaCla KCla MgSO4  ðH2 OÞ7 a Relative PHA
synthase activity (%)
0 0 0 40
0 0 0.5 70
0 0.2 0.5 90
3 0.2 0.5 90
Fig. 4. Inhibition of the granule-bound PHB synthase by CoA. (A) 3 20 0.5 85
PHB synthase activity as a function of CoA concentration; (B) 25 0 0 90
competitive inhibition of the PHB synthase demonstrated by the ap- 25 0.2 0 97
plication of various substrate concentrations at defined CoA concen- 25 0 0.5 92
trations; 10 lg=ml PHB granule-bound PHB synthase was applied in 25 0.2 0.5 100
a
the presence of 25% (w/v) NaCl, 0.2% (w/v) KCl, and 0.5% (w/v) Salt concentrations were provided as % of w/v.
MgSO4  7H2 O. For assay conditions see Materials and methods.
 F.F. Hezayen et al. / Archives of Biochemistry and Biophysics 403 (2002) 284–291 289

Fig. 6. PHB synthase activity as a function of NaCl concentration;

10 lg=ml PHB granule-bound PHB synthase were applied in the pre-
sence of variable concentrations of NaCl and 0.2% (w/v) KCl, 0.5% (w/v)
MgSO4  ðH2 OÞ7 , and 75 lM 3-hydroxybutyryl–CoA as substrate.

bound PHB synthase, the soluble PHB synthase, which

Fig. 8. Analysis of the solubilization of granule-associated proteins by
resided in the cytosolic fraction (supernatant after ul-
SDS–PAGE and immunoblotting. Purified PHB granules were
tracentrifugation), was strictly dependent on the salt subjected to solublization in 8 M urea in the presence of 5 mM DTE.
concentration. In the presence of 25% (w/v) NaCl, 0.2% Lanes: M, molecular weight standard; 1, purified PHB granules; 2,
(w/v) KCl, and 0.5% (w/v) MgSO4  ðH2 OÞ7 the soluble solubilized proteins (supernatant after 100,000g centrifugation for 1 h);
PHB synthase exhibited a specific activity of about 3 and 4, immunoblot of lanes 1 and 2 employing the anti-PHB syn-
thase antibodies raised against the PHB synthase from R. eutropha.
120 mU/mg cytosolic protein, whereas in the absence of
these salts no PHB synthase activity was detectable (data
not shown). The pH optimum of the granule-bound PHB
N-terminal sequencing of the major granule-associated
synthase was 7.5, and rather high activity was obtained
28-kDa protein
for the pH range 7–8.5 (Fig. 7). Efforts to enrich or pu-
rify the soluble PHB synthase failed.
The 28-kDa protein was purified by preparative
SDS–PAGE to electrophoretic homogenity and
Attachment of PHB synthase to PHB granules
subjected to N-terminal amino acid sequence analysis
using Edman degradation. The following N-terminal
To evaluate the mode of attachment of the PHB
sequence was obtained: (A) D S Q Q P A Q.
synthase and other granule-associated proteins to the
PHB granule, we applied the chaotropic reagents urea
(8 M) or guanidine–HCl (6 M) in absence or presence of
Discussion
5 mM DTE for solubilization experiments. Except for
the PHB synthase, all other granule-associated proteins
In this study, we demonstrated for the first time the
were released from the PHB granules (Fig. 8).
PHB synthase activity in an archaebacterium. This en-
zyme was found to be mainly granule associated and its
production was induced under PHB-accumulating
conditions. Immunological analysis suggested that a
strongly cross-reacting and inducible 80-kDa protein
corresponded with PHB synthase (Fig. 1). However,
further evidence by N-terminal amino acid sequencing
was not achievable, due to a lack in sequence similarity.
The class I PHA synthase from Caulobacter crescentus is
so far the largest PHA synthase exhibiting a similar
molecular weight of about 73 kDa [37]. The enzymo-
logical analysis of the granule-bound PHB synthase re-
vealed some unusual features of this new PHA synthase.
The granule-bound enzyme showed a high thermosta-
bility and exhibited 40% of the maximum PHB synthase
Fig. 7. PHB synthase activity as a function of pH; 10 lg=ml PHB
granule-bound PHB synthase was applied in the presence of 25% (w/v)
activity in the absence of salts (Figs. 5 and 6; Table 1).
NaCl, 0.2% (w/v) KCl, 0.5% (w/v) MgSO4  ðH2 OÞ7 , and 75 lM Almost full enzyme activity was achieved when
3-hydroxybutyryl–CoA as substrate. increasing the ionic strength independent of the salt ion
290 F.F. Hezayen et al. / Archives of Biochemistry and Biophysics 403 (2002) 284–291

(Table 1). Interestingly, the addition of Mg2þ ions, halobacteria might have developed a different metabolic
which have been found to serve as cofactors for PHA route toward PHB biosynthesis.
synthases, showed the strongest effect. However, high
enzyme activity without Mg2þ ions was observed, when
Acknowledgments
other ions were present (Table 1), suggesting that Mg2þ
ions do not serve as cofactors. Generally, enzymes from
This study was supported by a fellowship provided by
extremely halophilic archaebacteria require high salt
the Egyptian Government to Francis F. Hezayen and by
concentrations for their biological activity and stability.
a grant from the Deutsche Forschungsgemeinschaft
High salt concentrations are required to compensate the
provided to Bernd H.A. Rehm.
high protein surface charge of these enzymes [10]. Ac-
cordingly, the soluble PHB synthase did not exhibit
enzyme activity in the absence of salts. However, the References
soluble PHA synthase was only present at very low copy
[1] S.N. Sehgal, N.E. Gibons, Can. J. Microbiol. 6 (1960) 165–
number as indicated by a very weak enzyme activity. 169.
This suggests that the enzyme rapidly converts into the [2] M. Ginzburg, L. Sachs, Ginzburg B.Z, J. Gen. Physiol. 55 (1970)
highly active granule-associated enzyme, when substrate 187–207.
is available. In this study, we investigated the mode of [3] D.J. Kushner, in: D.J. Kushner (Ed.), Microbiol Life in
attachment, and evidence that PHA synthase is cova- Extreme Environments, Academic Press, London, 1978, pp.
317–368.
lently linked to the PHB core of the granule was [4] M. Kamekura, Extremophiles 2 (1998) 289–295.
provided (Fig. 8). Based on the postulated catalytic [5] H. Eisenberg, M. Mevarech, G. Zaccai, Adv. Protein Chem. 43
mechanism, which uses a conserved cysteine residue as (1992) 1–62.
catalytic nucleophile, the enzyme might be covalently [6] J.K. Lanyi, Bacteriol. Rev. 38 (1974) 272–290.
attached via a thioester bond. The other major granule- [7] K. Hecht, T. Langer, A. Wrba, R. Jaenicke, Biol. Chem. Hoppe
Seyler 371 (1990) 515–519.
associated proteins, presumably involved in granule [8] G. Krishnan, W. Altekar, Biochemistry 32 (1993) 791–798.
structure and functionally resembling the phasins, were [9] A. Madan, H.M. Sonawat, Physiol. Chem. Phys. Med. NMR 28
released by treatment with chaotropic reagents, which (1996) 15–28.
indicated that these proteins were noncovalently at- [10] M. Mevarech, F. Frolow, L.M. Gloss, Biophys. Chem. 86 (2000)
tached (Fig. 8). PHA synthases show a rather broad 155–164.
[11] R. Fernandez-Casatillo, F. Rodrıguez-Valera, J. Gonzales-Ra-
substrate specificity, and PHB synthases from most mos, F. Ruiz-Berraquero, Appl. Environ. Microbiol. 51 (1986)
bacteria, such as e.g. R. eutropha and A. vinosum, use 3- 214–216.
hydroxyacyl–CoA thioesters with three to five carbon [12] J.G. Lillo, F. Rodrıguez-Valera, Appl. Environ. Microbiol. 56
atoms [15,16]. The haloarchaeal PHB synthase did not (1990) 2517–2521.
accept even 3-hydroxyvaleryl–CoA as substrate, which [13] F.F. Hezayen, B.H.A. Rehm, R. Eberhardt, A. Steinb€ uchel, Appl.
Microbiol. Biotechnol. 54 (2000) 319–325.
indicated a more narrow substrate specificity. Kinetic [14] A. Steinb€uchel, in: H.J. Rehm, G. Reed, A. P€ uhler, P Stadler
analysis showed that the haloarchaeal PHB synthase (Eds.), Biotechnology, vol. 6, Wiley, Heidelberg, Germany, 1996,
exerted no lag phase, which resembles the substrate p. 403.
conversion kinetics of the class III PHB synthase from [15] B.H.A. Rehm, A. Steinb€ uchel, Int. J. Biol. Macromol. 25 (1999)
A. vinosum (Fig. 2) [18]. From substrate conversion ki- 3–19.
[16] B.H.A. Rehm, A. Steinb€ uchel, in: A. Steinb€ uchel, Y. Doi
netics and substrate saturation kinetics, a Vmax of 4.6 U/ (Eds.), Handbook ‘‘Biopolymers,’’ vol. 3a, Wiley, 2001, pp.
mg granule-bound protein and a K0:5 of 56 lM were 173–215.
derived, respectively, which represented an about 20- [17] T.U. Gerngross, K.D. Snell, O.P. Peoples, A.J. Sinskey, E.
fold higher enzyme activity and affinity toward 3-hy- Csuhai, S. Masamune, J. Stubbe, Biochemistry 33 (1994) 9311–
droxybutyryl–CoA than has been described for the 9320.
[18] M. Liebergesell, I.L. Sonomoto, M. Madkour, F. Mayer, A.
granule-bound PHB synthase from R. eutropha (Fig. 3) Steinb€uchel, Eur. J. Biochem. 226 (1994) 71–80.
[38]. A Hill coefficient of 1.75 indicated positive coo- [19] R. Jossek, R. Reichelt, A. Steinb€uchel, Appl. Microbiol. Biotech-
perativity exhibited by the PHB synthase. CoA was nol. 49 (1998) 258–266.
found to be a competitive inhibitor of the haloarchaeal [20] Q. Qi, A. Steinb€uchel, B.H.A. Rehm, Appl. Microbiol. Biotech-
PHB synthase. However, the affinity of the haloarchaeal nol. 54 (2000) 37–43.
[21] B.H.A. Rehm, Q. Qi, B. Beermann Br., H.-J. Hinz, A.
PHB synthase to CoA represented by a IC50 of 160 lM Steinb€uchel, Biochem. J. 358 (2001) 263–268.
was rather low as compared to the P. aeruginosa PHA [22] O.P. Peoples, A.J. Sinskey, J. Biol. Chem. 264 (1989) 15298–
synthase PhaC1 and the A. vinosum PHA synthase 15303.
PhaE/C exhibiting Ki s of 85 and 100 lM, respectively [23] O.P. Peoples, A.J. Sinskey, J. Biol. Chem. 264 (1989) 15293–
[20,39]. Overall, this archaebacterial PHB synthase 15297.
[24] P. Schubert, A. Steinb€ uchel, H.G. Schlegel, J. Bacteriol. 170
represents the most unique PHB synthase so far (1988) 5837–5847.
characterized. Interestingly, no evidence for other PHB [25] S.C. Slater, W.H. Voigt, D.E.. Dennis, J. Bacteriol. 170 (1988)
biosynthesis enzymes was obtained, which indicated that 4431–4436.
 F.F. Hezayen et al. / Archives of Biochemistry and Biophysics 403 (2002) 284–291 291

[26] A. Steinb€uchel, H.G. Schlegel, Mol. Microbiol. 5 (1991) 523–542. [33] S.J. Liu, A. Steinb€
uchel, Appl. Microbiol. Biotechnol. 53 (2000)
[27] Hezayen, F.F., Tindall, B.J., Steinb€ uchel, A., Rehm, B.H.A., 545–552.
2002. Int. J. Syst. Evol. Microbiol. (submitted). [34] F. Lynen, H. Wieland, Meth. Enzymol. 1 (1955) 566–
[28] J. Sambrook, E.F. Fritsch, T. Maniatis, Molecular Cloning: 573.
A Laboratory Manual, second ed., Cold Spring Harbor [35] T. Nishimura, T. Saito, K. Tomita, Arch. Microbiol. 116 (1978)
Laboratory Press, Cold Spring Harbor, NY, 1989. 21–27.
[29] H. Towbin, T. Staehelin, J. Gordon, Proc. Natl. Acad. Sci. USA [36] H.E. Valentin, A. Steinb€ uchel, Appl. Microbiol. Biotechnol. 40
76 (1979) 4350–4354. (1994) 699–709.
[30] H. Brandl, R.A. Gross, R.W. Lenz, R.C. Fuller, Appl. Environ. [37] Q. Qi, B.H.A. Rehm, Microbiology 14 (7) (2001) 3353–
Microbiol. 54 (1988) 1977–1982. 3358.
[31] A. Timm, A. Steinb€ uchel, Appl. Environ. Microbiol. 56 (1990) [38] G.W. Haywood, A.J. Anderson, L. Chu, E.A. Dawes, FEMS
3360–3367. Microbiol. Lett. 57 (1989) 1–6.
[32] J.W. Cary, D.J. Petersen, E.T. Papoutsakis, G.N. Bennett, J. [39] R. Jossek, A. Steinb€
uchel, FEMS Microbiol. Lett. 168 (1998) 319–
Bacteriol. 170 (1988) 4613–4618. 324.