www.fems-microbiology.org

FEMS Microbiology Letters 239 (2004) 195–201

Engineered Aeromonas hydrophila for enhanced productionof poly(3-hydroxybutyrate-co-3-hydroxyhexanoate)

with alterable monomers composition

Jing Han a,b, Yuan-Zheng Qiu a, Dai-Cheng Liu b, Guo-Qiang Chen a,*

a Department of Biological Sciences and Biotechnology, Tsinghua University, Beijing 100084, Chinab School of Life Science, Shan Dong Normal University, Jinan 250014, China

Received 14 July 2004; received in revised form 2 August 2004; accepted 30 August 2004

First published online 11 September 2004

Edited by A. Steinbuchel

Abstract

Aeromonas hydrophila 4AK4 produces poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) (PHBHHx) containing 3-hydroxybuty-

rate (3HB) and about 15 mol% 3-hydroxyhexanoate (3HHx) from dodecanoate. To study the factors affecting the monomer com-

position and PHBHHx content, genes encoding phasin (phaP), PHA synthase (phaC) and (R)-specific enoyl-CoA hydratase (phaJ)

from Aeromonas punctata (formerly named Aeromonas caviae) were introduced individually or jointly into A. hydrophila 4AK4. The

phaC gene increased 3HHx fraction more significantly than phaP, while phaJ had little effect. Expression of phaC alone increased the

3HHx fraction from 14 to 22 mol%. When phaC was co-expressed with phaP and phaJ, the 3HHx fraction increased from 14 to 34

mol%. Expression of phaP or phaC alone or with another gene enhanced PHBHHx content up to 64%, cell dry weight (CDW) as

much as 4.4 g L�1 and PHBHHx concentration to 2.7 g L�1 after 48 h in shake flask culture. The results suggest that a higher PHA

synthase activity could lead to a higher 3HHx fraction and PHBHHx content. Co-expression of phaJ with phaC or phaP would

favor PHA accumulation, although over-expression of phaJ did not affect PHA synthesis much. In addition, inhibition of b-oxida-tion by acrylate in A. hydrophila 4AK4 enhanced PHBHHx content. However, no monomers longer than 3HHx were detected. The

results show that genetic modification of A. hydrophila 4AK4 enhanced PHBHHx production and altered monomer composition of

the polymer.

� 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved.

Keywords: Aeromonas hydrophila; Biopolyester; Polyhydroxyalkanoates; Poly(3-hydroxybutyrate-co-3-hydroxyhexanoate)

1. Introduction

Polyhydroxyalkanoates (PHAs) are polyesters pro-duced by a wide variety of bacteria as carbon and en-

ergy storage material under unbalanced growth

conditions. Over the last decade, they have attracted

0378-1097/$22.00 � 2004 Federation of European Microbiological Societies

doi:10.1016/j.femsle.2004.08.044

* Corresponding author. Tel.: +86 10 62783844; fax: +86 10

62788784.

E-mail address: chengq@mail.tsinghua.edu.cn (G.-Q. Chen).

academic and industrial attention because of their

potential use as biodegradable thermoplastics [1]. Bac-

terial PHA can be divided into two groups dependingon the number of carbon atoms in the monomeric

units: short-chain-length (SCL) and medium-chain-

length (MCL) PHA [2]. Industrial scale production

of three types of PHA have been reported, namely,

poly-3-hydroxybutyrate (PHB) [3], the copolyester of

3-hydroxybutyrate (3HB) and 3-hydroxyvalerate (3HV)

(PHBV) [4], and the copolyester of 3-hydroxyhexanoate

. Published by Elsevier B.V. All rights reserved.

196 J. Han et al. / FEMS Microbiology Letters 239 (2004) 195–201

(3HHx) and 3HB (PHBHHx) [5]. The latter was found

to have better biocompatibility and physical proper-

ties, such as enhanced flexibility and improved impact

strength [6], than either PHB or polylactic acid (PLA)

[7].

Many bacteria can synthesize PHB, however, only afew produce PHBHHx. Among the latter, Aeromonas

hydrophila 4AK4 [5,8] and Aeromonas punctata (for-

merly known as Aeromonas caviae) [6] are the best-

studied strains. Molecular studies have demonstrated

that phasin (PhaP), PHA synthase (PhaC) and (R)-spe-

cific enoyl-CoA hydratase (PhaJ) of A. hydrophila

4AK4 were very similar to the A. punctata proteins,

possessing 100%, 98%, and 97% homology, respec-tively [9,10]. This suggested that A. hydrophila 4AK4

and A. punctata have similar PHBHHx biosynthesis

pathways.

Aeromonas hydrophila 4AK4 produced PHBHHx

containing 12–18 mol% 3HHx. This strain could not

produce PHBHHx with 3HHx fraction below 10

mol% or above 20 mol% regardless of growth condi-

tions [8,22]. However, PHBHHx containing different3HHx fractions has different properties which could

support various applications [11]. It was also found

that variations in the 3HHx fraction conferred differ-

ent biocompatibilities on PHBHHx useful for tissue

engineering [12]. Therefore, it is desirable to produce

PHBHHx with a wider range of alterable monomer

composition. Recently, a recombinant A. hydrophila

4AK4 was constructed that produced PHBHHx con-taining 3–12 mol% 3HHx using various ratios of gluc-

onate and dodecanoate as substrates [13]. Other

recombinant strains have been constructed that pro-

duced PHBHHx with alterable 3HHx fraction. Fukui

et al. [14] reported that a transconjugant of A. punc-

tata harboring additional copies of phaPCJ genes

accumulated PHBHHx with much higher 3HHx frac-

tion (46–63 mol%) from fatty acids or olive oil. How-ever, the mechanism remains unclear. The effects of

phaP or phaJ alone on monomers composition were

not studied by the group.

Properties such as the robust growth, simple

growth requirements and convenience for genetic

engineering make A. hydrophila 4AK4 very useful

for the production of PHBHHx [5,8,13]. However, it

is not clear how phaP, phaC and phaJ affectPHBHHx accumulation and monomer composition

in this strain. In order to clarify these effects and

the mechanism behind, also as a preparation for

industrial exploitation, in this paper, recombinant

strains of A. hydrophila 4AK4 with enhanced

PHBHHx synthesis ability were constructed and used

to produce PHBHHx with variable 3HHx monomer

fractions. Additionally, the effect of acrylate, a b-ketothiolase (FadA) inhibitor, on PHBHHx synthesis

was also studied.

2. Materials and methods

2.1. Bacterial strains

Aeromonas hydrophila strain 4AK4 (Laboratory col-

lection, referred to [5,8]) was used for PHBHHx produc-tion. Escherichia coli JM109 was used for PHBHHx

production or plasmid construction and E. coli S17-1

(recA; harbors the tra genes of plasmid RP4 in the chro-

mosome; proA, thi-1) [15] was used as a vector donor in

conjugation. A. hydrophila 4AK4 and E. coli were grown

in Luria–Bertani (LB) medium or on LB agar plates at

30 �C and 37 �C, respectively. Kanamycin (50 mg L�1)

or ampicillin (60 mg L�1) was added when needed.

2.2. Plasmids construction

All plasmids used in this study are listed in Table 1.

Recombinant plasmid pEE32 contained phaP, phaC

and phaJ of A. punctata [10]. Plasmid pTG04 was a

derivative of pEE32 that self-ligated after digestion by

NcoI, resulting in a large deletion in phaC. All otherplasmids were derived from pBBR1MCS-2 [16] by

inserting a DNA fragment of pEE32 or pTG04 as

shown in Fig. 1. The large EcoRI fragment of pEE32

and small EcoRI fragment of pTG04 were inserted into

the same site of pBBR1MCS-2 to obtain pQHAc07 or

pQHAc08 and pQHAc17 or pQHAc18, respectively,

depending on the orientation of inserts. The inserts used

to construct other plasmids were obtained by PCR usingpEE32 or pTG04 as templates. Plasmid pQHAc06 was

constructed by inserting a PCR-amplified DNA frag-

ment including phaP into pQHAc03. All DNA manipu-

lations including restriction digestion, ligation, and

agarose gel electrophoresis were carried out using stand-

ard procedures [17]. The recombinant plasmids were

first introduced into E. coli S17-1 by electroporation,

then transconjugation of A. hydrophila 4AK4 and E. coli

S17-1 harboring recombinant plasmids were carried out

as described by Friedrich et al. [18].

2.3. Culture conditions

Recombinant A. hydrophila 4AK4 strains were first

cultivated in LB medium overnight and then transferred

into mineral salt (MS) medium (pH 7.2) supplementedwith 1 g L�1yeast extract and containing 8 g L�1 dode-

canoate as carbon source. When sodium acrylate was

used, the cells were cultured in MS medium containing

4 g L�1 dodecanoate and 1 g L�1 nutrition broth

(NB), and 0.25–0.75 mM sodium acrylate was added

after 8 hours cultivation. The cells were cultured at

30�C for 48 h on a rotating shaker at 200 rpm (NBS,

Series 25D, New Brunswick, USA). The MS mediumcontained 9.0 g L�1 Na2HPO4 Æ 12H2O, 1.5 g L�1

KH2PO4, 1.0 g L�1 (NH4)2SO4, 0.4 g L�1 MgSO4 Æ 7-

Table 1

Plasmids used in this study

Plasmids Relevant characteristics Source or reference

pEE32 5.9 Kbp; pha native promoter; phaPCJ, AmpR [10]

pTG04 5.0 Kbp; derivative of pEE32; phaC negative mutant; phaPC0J This study

pBBR1MCS-2 5.1 Kbp; broad-host-range plasmid, KamR [16]

pQHAc01 5.5 Kbp; lac promoter; phaP This study

pQHAc02 6.9 Kbp; lac promoter; phaC This study

pQHAc03 5.6 Kbp; lac promoter; phaJ This study

pQHAc04 7.4 Kbp; lac promoter; phaPC This study

pQHAc05 7.3 Kbp; lac promoter; phaCJ This study

pQHAc06 6.0 Kbp; lac promoter; phaJP This study

pQHAc07 8.3 Kbp; lac and pha native promoter; phaPCJ This study

pQHAc08 8.3 Kbp; pha native promoter; phaPCJ This study

pQHAc09 8.0 Kbp; lac promoter; phaPCJ This study

pQHAc17 7.4 Kbp; lac and pha native promoter; phaPC0J This study

pQHAc18 7.4 Kbp; pha native promoter; phaPC0J This study

pQHAc19 6.9 Kbp; lac promoter; phaPC0J This study

pQHAc07

phaP

phaCphaJ

rep

Km

Plac

Eco RI

Eco RI

pQHAc17

phaP

phaC'phaJ

rep

Km

Plac

Eco RI

Eco RI

pEE32

Am

phaP

phaCphaJ

PlacOri

Eco RI

Eco RI

pTG04

Am

phaJ

phaC'(Ac)

phaP

PlacOri

Eco RI

Eco RI

pBBR1MCS-2

replacZ

Km

Eco RI

Fig. 1. Construction of plasmids pQHAc07 and pQHAc17. Plasmids were constructed by inserting the large EcoRI fragment of pEE32 (comprising

phaP, phaC and phaJ genes from A. punctata) and the small EcoRI fragment of pTG04 (comprising phaP, negative mutant phaC 0 and phaJ from A.

punctata) into the EcoRI site of pBBR1MCS-2, respectively.

J. Han et al. / FEMS Microbiology Letters 239 (2004) 195–201 197

H2O, and 1% (vol/vol) trace element solution. The com-

position of trace element solution has been described

[13]. For maintenance of plasmids in A. hydrophila

4AK4, 50 mg L�1 kanamycin was added to culturemedium.

For recombinant E. coli, an overnight culture was

used to inoculate MS medium containing 4 g L�1 dode-

canoate, 1 g L�1 yeast extract and 60 mg L�1ampicillin,

and grown at 37 �C for 72 h.

2.4. Analysis of PHBHHx

Cells were harvested by centrifugation. Cellular PHA

content and its monomer composition were analyzed by

198 J. Han et al. / FEMS Microbiology Letters 239 (2004) 195–201

gas chromatography (GC) after methanolysis of lyophi-

lized cells in chloroform [19]. GC analysis was carried

out using a Hewlett–Packard 6890 equipped with 30 m

HP-5 capillary column.

3. Results

3.1. PHBHHx synthesized by various recombinant A.

hydrophila 4AK4 strains harboring heterogenous genes

To study the factors affecting PHBHHx content and

its monomer fraction in A. hydrophila 4AK4, DNA frag-

ments containing phaP, phaC, phaJ, phaPC, phaCJ andphaJP from A. punctata were inserted into pBBR1MCS-

2. The resulting recombinant plasmids were introduced

into A. hydrophila 4AK4. The PHBHHx produced by

these recombinant strains is shown in Table 2. Expres-

sion of phaC had a significant effect on PHBHHx mono-

mer composition. By introducing phaC, phaPC or

phaCJ, 3HHx fraction increased from the control value

of 14 mol% to as much as 23 mol%. However, 3HHxfraction only increased from 14 to 18 mol% by introduc-

ing phaP alone. Expression of phaJ had little effect on

3HHx fraction. Expression of phaP or phaC alone or

with another gene generally enhanced PHBHHx synthe-

sis, where PHBHHx content, CDW and PHBHHx con-

centration were elevated up to 64%, 4.4 and 2.7 g L�1,

respectively, compared with 52%, 3.5 and 1.8 g L�1 in

wild type strain. However, PHBHHx synthesis was notenhanced by introducing phaJ alone.

3.2. Comparison of PHBHHx synthesized in recombinant

strains harboring phaC or mutant phaC

To further characterize the effect of additional phaC

on PHBHHx synthesis, DNA fragments of phaPCJ with

or without its native pha promoter were cloned intopBBR1MCS-2, resulting in pQHAc07 (lac and pha pro-

moter), pQHAc08 (pha promoter) and pQHAc09 (lac

promoter), and their corresponding plasmids

(pQHAc17, pQHAc18 and pQHAc19, respectively) con-

taining mutant phaC (phaC 0) were also constructed. The

Table 2

PHBHHx synthesized by various recombinant A. hydrophila 4AK4 strains h

Strains CDW (g L�1) PHBHHx content (wt%)

4AK4 3.5 ± 0.2 51.8 ± 2.0

4AK4(pQHAc01) 4.4 ± 0.2 57.8 ± 1.7

4AK4(pQHAc02) 4.3 ± 0.1 58.9 ± 2.5

4AK4(pQHAc03) 3.7 ± 0.4 51.6 ± 0.2

4AK4(pQHAc04) 4.2 ± 0.1 58.3 ± 1.4

4AK4(pQHAc05) 4.3 ± 0.2 63.5 ± 0.9

4AK4(pQHAc06) 4.4 ± 0.1 60.3 ± 2.3

A. hydrophila 4AK4 strains were cultivated in MS medium containing 8 g L�1

obtained from three parallel studies (n = 3).

deletion of a 920 bp NcoI fragment from phaC resulted

in the loss of ability to synthesize PHA in phaC 0. As

shown in Table 3, E. coli JM109 harboring phaC (in

pEE32) yielded 8% PHBHHx, whereas no PHBHHx

was detected in E. coli JM109 harboring phaC 0 (in

pTG04). The results of PHBHHx synthesis in recombi-nant A. hydrophila 4AK4 strains harboring phaC or

phaC 0 are shown in Table 3. Introduction of intact phaC

significantly increased 3HHx fraction 1.8–2.4-fold ver-

sus the wild type strain, whereas the mutant phaC 0 only

slightly increased the 3HHx fraction. When the heterog-

enous phaC gene was under the control of its native pha

promoter, the 3HHx fraction increased by as much as 34

mol%. In all these recombinant strains, PHBHHx con-tent, cell dry weight (CDW) and PHBHHx concentra-

tion were elevated, independent of intact or mutant

phaC gene.

3.3. The effect of acrylate on PHBHHx synthesized in A.

hydrophila 4AK4

Acrylate, an inhibitor of b-ketothiolase, has beenused to direct intermediates of fatty acid b-oxidationto PHA biosynthesis in non-PHA producing E. coli

[20]. Ralstonia eutropha was able to synthesize PHA with

a low 3HHx fraction in the presence of acrylate [21]. In

A. hydrophila 4AK4 grown on dodecanoate, addition of

acrylate resulted in a higher PHBHHx content, which

was elevated from 42% up to 58% (Table 4). The

3HHx fraction was slightly reduced when the concentra-tion of acrylate was increased. However, no monomers

longer than 3HHx were detected. Cell growth was signif-

icantly inhibited by acrylate.

4. Discussion

In A. hydrophila 4AK4, the pha operon includingphaP, phaC and phaJ functions to synthesize PHBHHx

from dodecanoate, similar to A. punctata [9,10]. Eluci-

dating the factors affecting PHBHHx content and its

monomer composition would facilitate the industrial

development of this strain. In this study, the effects of

arboring heterogenous genes

3HHx fraction (mol%) PHBHHx concentration (g L�1)

14.0 ± 0.7 1.8 ± 0.2

17.8 ± 0.3 2.5 ± 0.1

22.0 ± 0.4 2.4 ± 0.2

16.3 ± 0.9 1.9 ± 0.3

20.8 ± 1.6 2.5 ± 0.1

22.7 ± 0.6 2.7 ± 0.1

16.8 ± 0.1 2.7 ± 0.0

dodecanoate and 1 g L�1 yeast extract at 30 �C for 48 h. All data were

Table 3

Comparison of PHBHHx synthesis in recombinant strains harboring intact phaC gene or mutant phaC gene

Strains CDW (g L�1) PHBHHx content (wt%) 3HHx fraction (mol%) PHBHHx concentration (g L�1)

E. coli JM109 (pEE32) 1.2 ± 0.1 8.1 ± 0.6 10.9 ± 0.4 0.1 ± 0.0

JM109 (pTG04) 0.9 ± 0.1 None None None

A. hydrophila 4AK4 3.9 ± 0.1 52.6 ± 3.5 14.3 ± 0.3 2.1 ± 0.2

4AK4(pQHAc07) 5.1 ± 0.2 62.7 ± 1.9 29.6 ± 0.8 3.2 ± 0. 2

4AK4(pQHAc08) 4.8 ± 0.0 62.7 ± 3.2 34.1 ± 0.5 3.0 ± 0.2

4AK4(pQHAc09) 4.6 ± 0.1 62.0 ± 3.9 25.6 ± 0.5 2.9 ± 0.1

4AK4(pQHAc17) 4.4 ± 0.2 60.4 ± 2.4 18.0 ± 0.9 2.7 ± 0.1

4AK4(pQHAc18) 4.4 ± 0.1 64.3 ± 2.0 19.0 ± 0.4 2.8 ± 0.1

4AK4(pQHAc19) 4.3 ± 0.1 63.9 ± 2.4 21.3 ± 0.6 2.8 ± 0.1

A. hydrophila 4AK4 strains were cultivated in MS medium containing 8 g L�1 dodecanoate and 1 g L�1 yeast extract at 30 �C for 48 h. Recombinant

E. coli strains were cultivated in MS medium containing 4 g L�1 dodecanoate and 1 g L�1 yeast extract at 37 �C for 72 h. All data were obtained from

three parallel studies (n = 3).

Table 4

PHA produced by A. hydrophila 4AK4 in the presence of various

acrylate concentrations

Acrylate (mM) CDW (g L�1) PHBHHx content

(wt%)

3HHx fraction

(mol%)

0 2.5 ± 0.1 41.9 ± 2.3 16.3 ± 0.2

0.25 1.1 ± 0.2 56.3 ± 4.1 14.6 ± 0.4

0.50 0.9 ± 0.1 52.8 ± 4.2 13.7 ± 0.6

0.75 0.8 ± 0.1 58.3 ± 2.0 13.3 ± 0.3

Cells were cultivated in MS medium containing 4 g L�1 dodecanoate, 1

g L�1 NB and acrylate of various concentrations at 30 �C for 48 h. All

data were obtained from three parallel studies (n = 3).

PhaJ

C6 C4 C12

PhaJ

PHBHHx

PhaC

PhaC

3HB-CoA3HHx-CoA

Wild type

PhaJ PhaJ

3HHx-CoA

C4C12 C6

3HB-CoA

Recombinant

Fig. 2. PHBHHx biosynthesis in wild type and recombinant A.

hydrophila 4AK4 strains from dodeconoate. Thickness of the black

arrows represents carbon fluxes. C12, C6 and C4 represent interme-

diates of 12, 6 and 4 carbon atoms in b-oxidation. The higher PhaC

activity in recombinants pulls more C6 intermediates into PHBHHx

synthesis prior to conversion into shorter C4 intermediates, resulting in

a higher 3HHx fraction and PHBHHx content than those in wild type

strain. PhaJ, (R)-specific enoyl-CoA hydratase; PhaC, PHA synthase.

J. Han et al. / FEMS Microbiology Letters 239 (2004) 195–201 199

phaP, phaC and phaJ from A. punctata on PHBHHx

synthesis in A. hydrophila 4AK4 were investigated. The

reason for using the phaP, phaC and phaJ from A. punc-

tata instead of A. hydrophila is that the three genes in A.

hydrophila showed high similarity to the corresponding

genes in A. punctata [9,10].

Among the three genes, phaC had the most significant

effect on PHBHHx monomer composition. The ob-served increase in the 3HHx fraction might be due to

a higher PHA synthase activity due to the additional

copy of phaC. The observation that a higher 3HHx frac-

tion in PHBHHx is promoted by increased PHA syn-

thase activity has also been reported by other

researchers [14]. We hypothesize this is due to the mech-

anism illustrated in Fig. 2. After three cycles of b-oxida-tion, a fatty acid (dodecanoate) is shortened from twelvecarbon atoms (C12) to six carbon atoms (C6) in length.

By the action of PhaJ, hexenoyl-CoA (a C6 intermedi-

ate) is converted to (R)-3-hydroxyhexanoyl-CoA

(3HHx-CoA) and presented to PHA synthase before it

is degraded to shorter C4 intermediates. In recombinant

strains, the higher PHA synthase activity will accelerate

polymerization of 3HHx-CoA, resulting in lower 3HHx-

CoA concentration. Then PhaJ pulls more proportionof C6 intermediates to 3HHx-CoA due to weakened

feedback inhibition. Since a higher proportion of C6

intermediates from b-oxidation pathway are used in

PHBHHx synthesis, carbon flux to C4 intermediates is

reduced, thereby diminishing 3HB-CoA production.

Although higher PHA synthase activity may also in-

crease polymerization of 3HB-CoA in PHBHHx synthe-

sis, the reduced flux to C4 intermediates precludes an

increase in the 3HB fraction. This mechanism is further

supported by the fact that recombinants harboring phaC

produced a higher 3HHx fraction than those harboringphaC 0 (Table 3). Since more carbon fluxes from b-oxida-tion are used for PHBHHx synthesis, accumulation of

this product is enhanced.

Phasin is a PHA granule-associated protein

[14,22,23]. Recombinant E. coli expressing phaP from

Rodococcus rubber have more PHA granules, although

these are smaller in size [23–25]. In this study, we found

that over-expression of phaP slightly increased the3HHx fraction and PHBHHx content. PhaP may in-

crease the 3HHx fraction and PHA content through

200 J. Han et al. / FEMS Microbiology Letters 239 (2004) 195–201

PhaC, which can affect the 3HHx fraction and PHA

content as mentioned above. The PhaP-induced increase

in PHA synthase activity had been reported in A. punc-

tata [14]. PhaP may also aid the polymerization process.

PhaP is proposed to be an amphiphilic protein whose

hydrophobic domains are attached to PHA granulesand hydrophilic domains are exposed to cytoplasm

[22]. During synthesis, hydrophobic interactions be-

tween PhaP and growing PHA chains on the synthase

might promote the formation of inclusion bodies, in

turn accelerating the polymerization process catalyzed

by PhaC. A faster polymerization rate would increase

the 3HHx fraction and PHA content as mentioned

above. Further studies are needed to clarify the effectof PhaP on 3HHx fraction and PHA content in recomb-

inant A. hydrophila.

Over-expression of phaJ alone did not have much ef-

fect on the 3HHx fraction. This result was consistent

with a previous study using A. punctata [14]. The hydra-

tion step catalyzed by PhaJ was not considered a rate-

limiting step in PHBHHx synthesis [14]. As shown in

Table 2, over-expression of phaJ alone did not increasethe PHBHHx content. It is interesting to note that in

recombinants A. hydrophila 4AK4 (pQHAc02) and

4AK4 (pQHAc05) or 4AK4 (pQHAc01) and 4AK4

(pQHAc06), the PHBHHx content increased when phaJ

was co-expressed with phaP or phaC genes. This was

also observed in recombinant A. hydrophila 4AK4

strains harboring phaPC 0J, in which the PHBHHx con-

tent also increased, although phaC 0 was inactivated (Ta-ble 3). As noted above, over-expression of phaC resulted

in higher PHA synthase activity and faster polymeriza-

tion of 3HHx-CoA and 3HB-CoA, which would make

supply of 3HHx-CoA and 3HB-CoA (the hydration step

catalyzed by PhaJ) a rate-limiting step in recombinant

strains, although it was not in the wild type strain. So

co-expression of phaJ with phaC would enhance both

the hydration step and polymerization step, resultingin higher PHBHHx content. This assumption may also

explain the higher PHBHHx content when phaJ was

co-expressed with phaP. Since over-expression of phaP

might enhance PHA synthase activity and/or assist the

polymerization process noted above, co-expression of

phaJ would increase supply of 3HHx-CoA and 3HB-

CoA, resulting in the enhancement of both the hydra-

tion and polymerization steps. Although PhaJ was notrate-limiting in PHBHHx synthesis, co-expression of

phaJ with phaP or phaC may be necessary to favor

PHBHHx accumulation.

A previous report found that the lac promoter acted

constitutively in A. hydrophila 4AK4 [13]. In this study,

we compared the lac and pha promoters. When phaPCJ

genes were under the control of native pha promoter, the

highest 3HHx fraction was observed. Therefore, in A.

hydrophila the native pha promoter better expresses het-

erogenous genes than the lac promoter.

It seems that expressionofheterogenousgenes doesnot

inhibit cell growth of A. hydrophila 4AK4. Except for the

phaJ gene, expression of genes individually or jointly leads

to bothhigher PHAcontent andhigherCDW, resulting in

significantly increased PHA concentration. Thus, A.

hydrophila 4AK4 is a good microorganism for further ge-netic modification to enhance PHA production.

Inhibition of FadA activity by acrylate will allow

intermediates of b-oxidation to accumulate, which can

be channeled to PHA biosynthesis in E. coli [20]. Acry-

late also led to the incorporation of monomers of longer

chain length into PHA in R. eutropha [21]. In this study,

we found that acrylate enhanced PHBHHx accumula-

tion, probably due to a decrease in b-oxidation. How-ever, no monomers longer than 3HHx were detected,

which suggested that the substrate specificity of PHA

synthase in A. hydrophila 4AK4 was restricted to mono-

mers of C4 and C6.

Acknowledgements

We are very grateful for the kind donation of plasmid

pEE32 from Dr. Y. Doi of RIKEN (Saitama, Japan)

and plasmid pBBR1MCS-2 from Dr. Philip Green of

Procter & Gamble (Cincinnati, USA). This study wasfinancially supported by Natural Science Foundation

of China Grant No. 30170017 and the State Outstand-

ing Young Scientist Award (No. 30225001).

References

[1] Doi, Y. (1990) Microbial Polyesters. VCH Publishers, New York,

NY.

[2] Steinbuchel, A. (2001) Perspectives for biotechnological produc-

tion and utilization of biopolymers: metabolic engineering of

polyhydroxyalkanoate biosynthesis pathways as a successful

example. Macromol. Biosci. 1, 1–24.

[3] Hrabak, O. (1992) Industrial production of poly-b-hydroxybuty-rate. FEMS Microbiol. Rev. 103, 251–256.

[4] Byrom, D. (1992) Production of poly-b-hydroxybutyrate:poly-b-hydroxyvalerate copolymers. FEMS Microbiol. Rev. 103, 247–

250.

[5] Chen, G.Q., Zhang, G., Park, S.J. and Lee, S.Y. (2001) Industrial

scale production of poly(3-hydroxybutyrate-co-3-hydroxyhexano-

ate. Appl. Microbiol. Biotechnol. 57, 50–55.

[6] Doi, Y., Kitamura, S. and Abe, H. (1995) Microbial synthesis and

characterization of poly(3-hydroxybutyrate-co-3-hydroxyhexano-

ate). Macromolecules 28, 4822–4828.

[7] Yang, X.S., Zhao, K. and Chen, G.Q. (2002) Effect of surface

treatment on the biocompatibility of microbial polyhydroxyalk-

anoates. Biomaterials 23, 1391–1397.

[8] Lee, S.H., Oh, D.H., Ahn, W.S., Lee, Y., Choi, J. and Lee, S.Y.

(2000) Production of poly by high-cell density cultivation of

Aeromonas hydrophila. Biotechnol. Bioeng. 67, 240–244.

[9] Zhang, G. (2002) Cloning of polyhydroxyalkanoates biosynthesis

genes from bacteria. Ph.D. thesis. Department of Biological

Sciences and Biotechnology, Tsinghua University, Beijing, China.

[10] Fukui, T. and Doi, Y. (1997) Cloning and analysis of the poly(3-

hydroxybutyrate-co-3-hydroxyhexanoate) biosynthesis genes of

Aeromonas caviae. J. Bacteriol. 179, 4821–4830.

J. Han et al. / FEMS Microbiology Letters 239 (2004) 195–201 201

[11] Feng, L., Watanabe, T., Wang, Y., Kichise, T., Fukuchi, T.,

Chen, G.Q., Doi, Y. and Inoue, Y. (2002) Studies on comonomer

compositional distribution of bacterial poly(3-hydroxybutryate-

co -3-hydroxyhexanoate)s and thermal characteristics of their

fractions. Biomacromolecules 3, 1071–1077.

[12] Wang, Y.W., Yang, F., Wu, Q., Cheng, Y.C., Yu, P.H.F., Chen,

J.C. and Chen, G.Q. (2005) Effect of composition of poly(3-

hydroxybutyrate-co-3-hydroxyhexanoate) on growth of fibroblast

and osteoblast. Biomaterials (in press).

[13] Qiu, Y.Z., Ouyang, S.P., Shen, Z.Y., Wu, Q. and Chen,

G.Q. (2004) Metabolic engineering for the production of

copolyesters consisting of 3-hydroxybutyrate and 3-hydroxy-

hexanoate by Aeromonas hydrophila. Macromol. Biosci. 4,

255–261.

[14] Fukui, T., Kichise, T., Lwata, T. and Doi, Y. (2001) Character-

ization of 13 kDa granule-associated protein in Aeromonas caviae

and biosynthesis of polyhydroxyalkanoates with altered molar

composition by recombinant bacteria. Biomacromolecules 2, 148–

153.

[15] Simon, R., Priefer, U. and Puhler, A. (1983) A broad host range

mobilization system for in vivo genetic engineering: transposon

mutagenesis in gram negative bacteria. Bio/Technology 1, 784–

791.

[16] Kovach, M.E., Elzer, P.H., Hill, D.S., Robertson, G.T., Farris,

M.A., Roop II, R.M. and Peterson, K.M. (1995) Four new

derivatives of the broad-host range cloning vector pBBR1MCS,

carrying different antibiotic-resistance cassettes. Gene 166, 175–

176.

[17] Sambrook, J. and Russell, D.W. (2001) Molecular Cloning: A

Laboratory Manual, Third ed. Cold Spring Harbor Laboratory

Press, Cold Spring Harbor, NY.

[18] Friedrich, B., Hogrefe, C. and Schlegel, H.G. (1981) Naturally

occurring genetic transfer of hydrogen-oxidizing ability between

strains of Alcaligenes eutrophus. J. Bacteriol. 147, 198–205.

[19] He, W.N., Tian, W.D., Zhang, G., Chen, G.Q. and Zhang, Z.M.

(1998) Production of novel polyhydroxyalkanoates by Pseudo-

monas stutzeri 1317 from glucose and soybean oil. FEMS

Microbiol. Lett. 169, 45–49.

[20] Qi, Q., Steinbuchel, A. and Rehm, B.H.A. (1998) Metabolic

routing towards polyhydroxyalkanoic acid synthesis in recomb-

inant Escherichia coli (fadR): inhibition of fatty acid b-oxidationby acrylic acid. FEMS Microbiol. Lett. 167, 89–94.

[21] Green, P.R., Kemper, J., Schechtman, L., Guo, L., Satkowski,

M., Fiedler, S., Steinbuchel, A. and Rehm, B.H.A. (2002)

Formation of short chain length/medium chain length poly-

hydroxyalkanoate copolymers by fatty acid b-oxidation inhibited

Ralstonia eutropha. Biomacromolecules 3, 208–213.

[22] Jurasek, L. and Marchessault, R.H. (2002) The role of phasins in

the morphogenesis of poly(3-hydroxybutyrate) granules. Biomac-

romolecules 3, 256–261.

[23] Wieczorek, R., Pries, A. and Steinbuchel, A. (1995) Analysis of a

24-kDa protein associated with the polyhydroxyalkanoic acid

granules in Alcaligenes eutrophus. J. Bacteriol. 177, 2425–2435.

[24] Pieper-Furst, U., Madkour, M.H., Mayer, H. and Steinbuchel, A.

(1995) Identification of the region of a 14-kDa protein of

Rhodococcus rubber that is responsible for the binding of this

phasin to polyhydroxyalkanoic acid granules. J. Bacteriol. 177,

2513–2523.

[25] Potter, M., Madkour, M.H., Mayer, F. and Steinbuchel, A.

(2002) Regulation of phasin expression and polyhydroxyalkano-

ate (PHA) granule formation in Ralstonia eutropha H16. Micro-

biology 148, 2413–2426.