Genetic Engineering of Enterobacter asburiae Strain JDR-1 for ...

puffautomaticΒιοτεχνολογία

10 Δεκ 2012 (πριν από 4 χρόνια και 4 μήνες)

221 εμφανίσεις


Published Ahead of Print 17 July 2009.
10.1128/AEM.01180-09.
2009, 75(18):5743. DOI:Appl. Environ. Microbiol.
Preston
Changhao Bi, Xueli Zhang, Lonnie O. Ingram and James F.

Hydrolysates
Production of Ethanol from Hemicellulose
Strain JDR-1 for Efficientasburiae
EnterobacterGenetic Engineering of
http://aem.asm.org/content/75/18/5743
Updated information and services can be found at:
These include:
REFERENCES
http://aem.asm.org/content/75/18/5743#ref-list-1at:
This article cites 27 articles, 11 of which can be accessed free
CONTENT ALERTS
more»articles cite this article),
Receive: RSS Feeds, eTOCs, free email alerts (when new
http://journals.asm.org/site/misc/reprints.xhtmlInformation about commercial reprint orders:
http://journals.asm.org/site/subscriptions/To subscribe to to another ASM Journal go to:
on December 9, 2012 by guesthttp://aem.asm.org/Downloaded from
A
PPLIED AND
E
NVIRONMENTAL
M
ICROBIOLOGY
,Sept.2009,p.5743–5749 Vol.75,No.18
0099-2240/09/$08.00￿0 doi:10.1128/AEM.01180-09
Copyright © 2009,American Society for Microbiology.All Rights Reserved.
Genetic Engineering of Enterobacter asburiae Strain JDR-1 for Efficient
Production of Ethanol from Hemicellulose Hydrolysates
￿
Changhao Bi,Xueli Zhang,Lonnie O.Ingram,and James F.Preston*
Department of Microbiology and Cell Science,University of Florida,Gainesville,Florida 32611
Received 21 May 2009/Accepted 11 July 2009
Dilute acid pretreatment is an established method for hydrolyzing the methylglucuronoxylans of hemicel-
lulose to release fermentable xylose.In addition to xylose,this process releases the aldouronate methylglucu-
ronoxylose,which cannot be metabolized by current ethanologenic biocatalysts.Enterobacter asburiae JDR-1,
isolated from colonized wood,was found to efficiently ferment both methylglucuronoxylose and xylose in acid
hydrolysates of sweet gumxylan,producing predominantly ethanol and acetate.Transformation of E.asburiae
JDR-1 with pLOI555 or pLOI297,each containing the PET operon containing pyruvate decarboxylase (pdc)
and alcohol dehydrogenase B (adhB) genes derived fromZymomonas mobilis,replaced mixed-acid fermentation
with homoethanol fermentation.Deletion of the pyruvate formate lyase (pflB) gene further increased the
ethanol yield,resulting in a stable E.asburiae E1(pLOI555) strain that efficiently utilized both xylose and
methylglucuronoxylose in dilute acid hydrolysates of sweet gum xylan.Ethanol was produced from xylan
hydrolysate by E.asburiae E1(pLOI555) with a yield that was 99% of the theoretical maximum yield and at a
rate of 0.11 g ethanol/g (dry weight) cells/h,which was 1.57 times the yield and 1.48 times the rate obtained with
the ethanologenic strain Escherichia coli KO11.This engineered derivative of E.asburiae JDR-1 that is able to
ferment the predominant hexoses and pentoses derived from both hemicellulose and cellulose fractions is a
promising subject for development as an ethanologenic biocatalyst for production of fuels and chemicals from
agricultural residues and energy crops.
Lignocellulosic resources,including forest and agricultural
residues and evolving energy crops,offer benign alternatives to
petroleum-based resources for production of fuels and chem-
icals.As renewable resources,these lignocellulosic materials
are expected to decrease dependence on exhaustible supplies
of petroleum and mitigate the net release of carbon dioxide
into the atmosphere.The development of economically accept-
able bioconversion processes requires pretreatments that release
the maximal quantities of hexoses (predominantly glucose re-
leased from cellulose) and pentoses (arabinose and xylose) from
hemicelluloses and also requires microbial biocatalysts that effi-
ciently convert these compounds to a single targeted product.
As one of three main components of lignocellulosics,hemi-
cellulose contains polysaccharides comprised of pentoses,hex-
oses and sugar acids that account for 20 to 35% of the total
biomass from different sources (21).Methylglucuronoxylans
(MeGAX
n
),consisting of long chains of as many as 70 ￿-xy-
lopyranose residues linked by ￿-1,4-glycosidic bonds (25),are
the predominant components in the hemicellulose fractions of
agricultural residues and energy crops,including corn stover,
sugarcane bagasse,poplar,and switchgrass (7,18,23,24).In
hardwood and softwood xylans,a 4-O-methylglucuronic acid is
attached at the 2￿ position of every sixth to eighth xylose
residue (12,15).Dilute acid hydrolysis is commonly used to
make the monosaccharides comprising hemicellulose accessi-
ble for fermentation (7,22).However,the ￿-1,2 glucuronosyl
linkage in xylan is resistant to dilute acid hydrolysis,which
results in the release of methylglucuronoxylose (MeGAX)
along with xylose and other monosaccharides.MeGAX is not
fermented by bacterial biocatalysts currently used to convert
hemicellulose to ethanol,such as Escherichia coli KO11 (2,6).
In sweet gum xylan,as much as 27%of the carbohydrate may
be in this unfermentable fraction after dilute acid pretreatment
(2,20).Complete utilization of all hemicellulosic sugars can
improve the efficiency of conversion of lignocellulosic materi-
als to fuel ethanol and other value-added products.
Our previous research on the processing of hemicelluloses
for fermentation led to isolation of Enterobacter asburiae strain
JDR-1.This isolate performed mixed-acid fermentation of the
principal hexoses and pentoses that can be derived from cel-
lulose and hemicellulose fractions of lignocellulosic biomass
and exhibited a novel metabolic potential based on its ability to
ferment MeGAX and xylose to ethanol and acetate as major
fermentation products fromsweet gumMeGAX
n
hydrolysates
generated by dilute acid pretreatment (2).This strain has been
genetically modified to produce
D
-(￿)-lactate as the predom-
inant product from acid hydrolysates of MeGAX
n
(3).
In this study,the PET operon containing the pdc,adhA,and
adhB genes from Zymomonas mobilis (10,11) was incorpo-
rated into a pflB E.asburiae JDR-1 isolate by plasmid trans-
formation to construct homoethanologenic strains.The result-
ing recombinant strains were compared with E.asburiae
wild-type strain JDR-1 and the ethanologenic strain E.coli
KO11 to evaluate their efficiencies of production of ethanol
from dilute acid hydrolysates of sweet gum MeGAX
n
.
MATERIALS AND METHODS
Bacterial strains,media,and growth conditions.The bacterial strains con-
structed and used in this study are listed in Table 1.E.asburiae JDR-1 served as
a starting point for genetic engineering.During strain construction,cultures were
* Corresponding author.Mailing address:Department of Microbi-
ology and Cell Science,University of Florida,Bldg.981,MuseumRd.,
Gainesville,FL 32611-0700.Phone:(352) 392-5923.Fax:(352) 392-
5922.E-mail:jpreston@ufl.edu.
￿
Published ahead of print on 17 July 2009.
5743
on December 9, 2012 by guesthttp://aem.asm.org/Downloaded from
grown aerobically at 30°C,37°C,or 39°C in Luria broth (10 g liter
￿1
Difco
tryptone,5 g liter
￿1
Difco yeast extract,5 g liter
￿1
NaCl) containing either 2%
(wt/vol) glucose,5% sucrose,or 3% (wt/vol) arabinose.Ampicillin (50 mg li-
ter
￿1
),tetracycline (12.5 mg liter
￿1
),kanamycin (20 mg liter
￿1
and 50 mg
liter
￿1
),apramycin (20 mg liter
￿1
),and chloramphenicol (10 mg liter
￿1
and 40
mg liter
￿1
) were added when they were needed.
Sweet gum MeGAX
n
was prepared from sweet gum (Liquidambar styraciflua)
stem wood as previously described and was characterized by
13
C nuclear mag-
netic resonance (9,15).Dilute acid hydrolysates of MeGAX
n
were prepared by
acid hydrolysis of 1% (wt/vol) sweet gum xylan with 0.1 N H
2
SO
4
at 121°C for
60 min,followed by neutralization with BaCO
3
.The total carbohydrate concen-
trations of substrate preparations were determined by the phenol-sulfuric acid
assay (8) with xylose as a reference or by high-performance liquid chromatog-
raphy (HPLC) (2).Minimal media were supplemented with Zucker-Hankin
mineral salts (ZH salts) at pH 7.4 (28).Growth media were buffered with 100
mM sodium phosphate buffer (pH 7.0) or 100 mM 3-(N-morpholino)propane-
sulfonic acid (MOPS) buffer (pH 7.0) when necessary.
Genetic methods.Standard methods were used for most of the genetic ma-
nipulations.Qiagen kits were used for extraction of genomic DNA and plasmids
(Qiagen,Valencia,CA).PCR amplification was performed with an I-cycler
thermal cycler (Bio-Rad,Hercules,CA) and primers synthesized by Operon
(Huntsville,AL).Topo cloning kits were used for cloning (Invitrogen,Carlsbad,
CA).Electroporation was performed with Gene pulser Xcell (Bio-Rad,Hercu-
les,CA).Restriction endonucleases were purchased from New England Biolabs
(Ipswich,MA).DNA sequencing was performed by the University of Florida
Interdisciplinary Center for Biotechnology Research.
Fermentation.Batch fermentations were carried out in screw-cap tubes (16 by
100 mm) filled with nitrogen and sealed with rubber stoppers.The tubes were set
in a Glas-Col minirotator rotating at 60 rpm in a 30°C incubator.Neutralized
sweet gum xylan acid hydrolysate (0.5%,wt/vol),5%
D
-glucose,or 4%
D
-xylose
was added to 2￿ ZH salts directly to obtain a growth medium buffered with
100 mMphosphate buffer or MOPS buffer at pH7.0.Fermentation hydrolysates
were inoculated to obtain an initial optical density at 600 nm of 1.0 (determined
using a Beckman DU500 series spectrophotometer).For analysis of fermenta-
tion products,cultures were centrifuged,and the supernatants were passed
through 0.22-￿m filters and subjected to HPLC.Products were resolved on a
Bio-Rad HPX-87H column with 0.01 N H
2
SO
4
at 65°C.Samples were delivered
with a 710B WISP automatic injector,and chromatography was controlled with
a Waters 610 solvent delivery system using a flow rate of 0.5 ml/min.Products
were detected by differential refractometry with a Waters 2410 RI detector.Data
analysis was performed with Waters Millennium software.A quantitative rela-
tionship between E.asburiae JDR-1 cell dry weight and culture optical density at
600 nm was determined.For calculation of specific consumption rates and
specific production rates,the cell dry weight was determined based on the optical
density at 600 nm of the fermentation culture,which was 1.0 (0.51 g liter
￿1
)
initially and did not change appreciably during the fermentation with 0.5%xylan
hydrolysate.
Transformation of E.asburiae JDR-1 with plasmids carrying the PET operon.
E.asburiae JDR-1 was grown with one of several antibiotics at different concen-
trations on LB and minimal media in agar plates or in liquid media to test its
antibiotic resistance.Based on sensitivity to chloramphenicol and tetracycline,
plasmids pLOI555 (Cm
r
) and pLOI297(Tet
r
),respectively,both containing the
PET operon,were transformed into E.asburiae JDR-1 or E.asburiae E1 by
electroporation in a 100-￿l cuvette using 1.8 kV,a capacitance of 25 ￿F,and a
resistance of 200 ￿.For electroporation competent cells from25-ml exponential-
phase cultures were washed three times by suspension and centrifugation with
cold 10%glycerol.Cultures were plated on LB agar containing 2%glucose and
tetracycline (12.5 mg liter
￿1
) and on LB agar containing 2%glucose and chlor-
amphenicol (40 mg liter
￿1
) to select E.asburiae JDR-1 and E1 derivatives
carrying pLOI297 and pLOI555,respectively.Plasmids were extracted to confirm
their presence in E.asburiae cells.
Deletion of the pflB gene in E.asburiae JDR-1.The method used for gene
deletion in E.coli has been described previously (13,27),and minor modifica-
tions were made for E.asburiae JDR-1.The pflB gene in E.asburiae JDR-1 was
also selected as an integration site for the PET operon.Several sets of primers
were designed based on sequences of pflB orthologs in other Enterobacter spp.to
amplify this gene fragment from E.asburiae JDR-1.Only one set of primers
derived from E.coli B was found to amplify the E.asburiae JDR-1 pflB gene
fragment.The amplified E.asburiae JDR-1 DNA sequence and E.coli K-12 pflB
sequence were found to have 93%identity.The plasmids constructed are listed
in Table 1.A partial 1,031-bp sequence of the E.asburiae JDR-1 pflB gene
(GenBank accession number EU719655) was determined using a DNAfragment
amplified by PCR with specific primers based on the E.coli pflB sequence.
Because the pflB gene in E.coli K-12 is 2,280 bp long,the partial 1,031-bp E.
asburiae JDR-1 pflB gene sequence may represent 45% of the whole gene.The
3-kb cat-sacB cassette containing a chloramphenicol resistance gene (cat) and a
levansucrase gene (sacB) was obtained by digesting pLOI4162 with SmaI and
SfoI and used in subsequent ligations.The pflB gene fragment amplified fromE.
asburiae JDR-1 was cloned into the pCR 4-TOPO vector (Invitrogen) to obtain
plasmid pTOPOpfl.This plasmid was diluted 500-fold and served as a template
for inside-out PCR amplification using the pflB inside-out primers.The resulting
5.5-kb fragment containing the replicon was ligated to the blunt-end cat-sacB
cassette from pLOI4162 to produce a new plasmid,pTOPO4162pfl.This 5.5-kb
fragment was also used to construct a second plasmid,pTOPODpfl,by phos-
phorylation and self-ligation.Both pTOPO4162pfl and pTOPODpfl were then
digested with XmnI and diluted 500-fold,and they were used as templates for
amplification using the pflB primer set to produce linear DNA fragments for
integration step 1 (pfl￿-cat-sacB-pfl￿) and step 2 (pfl￿-pfl￿),respectively.After
electroporation of the step 1 fragment into E.asburiae JDR-1 containing
pLOI3240,cells were incubated for 2 h at 30°C.The recombinant candidates
were selected based on chloramphenicol (20 mg liter
￿1
) resistance in Luria broth
plates after overnight incubation (15 h) at 39°C.Colonies were patched on both
plates containing kanamycin (50 mg liter
￿1
) and plates containing chloramphen-
icol (40 mg liter
￿1
).The colonies that grewon plates containing chloramphenicol
(40 mg liter
￿1
) but not on plates containing kanamycin (50 mg liter
￿1
) were
TABLE 1.Bacterial strains and plasmids used for engineering ethanolgenic E.asburiae
Strain or plasmid Relevant characteristic(s)
a
Source or reference
Strains
E.coli Top10 Used for general cloning Invitrogen
E.coli KO11 pfl::(pdc adhB cat) ￿frd 16
E.asburiae JDR-1 Wild type This study
E.asburiae E1 E.asburiae JDR-1 ￿pflB This study
Plasmids
PLOI3240 Am
r
red,red recombinase protein 26
pLOI297 Tc
r
pdc
￿
adhB
￿
colEl 1
pLOI555 Cm
r
pdc
￿
adhB
￿
17
pLOI4162 bla cat,cat-sacB cassette 13
pCR 4-TOPO bla kan amp,TOPO TA cloning vector Invitrogen
pTOPOpfl pflB fragment (PCR) amplified from E.asburiae JDR-1 and cloned into
PCR4-TOPO vector
This study
pTOPO4162pfl cat-sacB cassette cloned into pflB in pTOPOpfl This study
pTOPODpfl PCR fragment amplified from pTOPOpfl,kinase treated,and self-ligated This study
a
Abbreviations:Am,ampicillin;Tc,tetracycline;Cm,chloramphenicol.
5744 BI ET AL.A
PPL
.E
NVIRON
.M
ICROBIOL
.
on December 9, 2012 by guesthttp://aem.asm.org/Downloaded from
subjected to PCR confirmation.The confirmed mutant colonies were trans-
formed with pLOI3240 and prepared for electroporation with the step 2 frag-
ment (pfl￿-pfl￿).After electroporation,cells were incubated at 30°C for 4 h and
then transferred into a 250-ml flask containing 100 ml of LB medium without
NaCl but with 10% sucrose.Following overnight incubation (30°C),colonies
were streaked on plates containing LB medium without NaCl but with 6%
(wt/vol) sucrose and incubated at 39°C for 16 h.Colonies were tested to deter-
mine the loss of apramycin and chloramphenicol resistance,and the results were
confirmed by PCR.The resulting strain,E.asburiae E1,had a disrupted pflB gene
without detectable heterologous DNA sequences.
Plasmid stability in E.asburiae JDR-1.E.asburiae JDR-1 harboring either
pLOI555 or pLOI297 was serially transferred in Luria broth containing 2%
glucose without antibiotics for more than 72 generations aerobically at 30°C.One
generation was defined as a twofold increase in the culture turbidity.Appropriate
dilutions of cultures were plated on Luria agar with and without antibiotics;the
colonies formed were counted,and the ratio of the number of cells with antibi-
otic resistance to the total number of cells was determined.Ten colonies that
exhibited antibiotic resistance (and therefore were presumed to contain
pLOI555 or pLOI297) after 72 generations were used in fermentation experi-
ments to test their ethanol-producing abilities.
Assay of PDC activity.Pyruvate decarboxylase (PDC) activity in engineered E.
asburiae JDR-1 derivatives was assayed by monitoring the pyruvate-dependent
oxidation of NADH with alcohol dehydrogenase as a coupling enzyme (4,17).
Exponential-phase anaerobic cultures were harvested,and cells were disrupted
using the FastPrep bead mill MP system(MP Biomedicals,Irvine,CA) in 0.05 M
phosphate buffer.Each supernatant was collected after 15 min of centrifugation
at 1,800 rpm (Eppendorf centrifuge 5414).The entire process was carried out at
4°C.Heat treatment for 15 min at 60°C was used to inactivate competing native
enzymes of E.asburiae JDR-1 which might have affected quantitative measure-
ments of PDC activities in transformants.The PDC enzyme activity assay was
performed using a reaction mixture containing 1.0 mMthiamine pyrophosphate,
1.0 mMMgCl
2
,0.40 mMNADH,20 mMsodium pyruvate,and 0.05 Msodium
phosphate buffer (pH 6.5).The assay was started by adding 20 ￿l crude cell
extract.The protein concentration of the crude extract was determined with a
bicinchoninic acid protein assay reagent kit (Pierce Chemical Co.,Rockford,IL).
RESULTS
Fermentation characteristics of wild-type strain E.asburiae
JDR-1.E.asburiae JDR-1 performed a mixed-acid fermenta-
tion with a low substrate concentration.When growing with
2.5%(wt/vol)
D
-glucose or 2%(wt/vol)
D
-xylose,the wild-type
strain produced a wide range of products,including succinate,
lactate,acetate,formate,2,3-butanediol,and ethanol (Table
2).During
D
-glucose fermentation,succinate and acetate were
produced at low concentrations,approximately 1 mM.Lactate
was produced at a concentration of approximately 10 mM,and
the major products were formate,2,3-butanediol,and ethanol,
each of which was produced at a concentration of approxi-
mately 40 mM.More acetate and less 2,3-butanediol were
produced during
D
-xylose fermentation (Table 2).In both
batch fermentations buffered with 0.1 M sodium phosphate
(pH 7.0),the wild-type strain failed to utilize all of the sub-
strates during the 48-h experiment.Even in the buffered me-
dium the pH after fermentation decreased to 4.8,which sug-
gested that acid production might have been the main factor
preventing the cells from utilizing all of the substrate.
The components of the mediumcontaining 0.5%sweet gum
hemicellulose hydrolysate were determined by HPLC to be 20
mMxylose,1.4 mMMeGAX,and a small amount of MeGAX
2
(Fig.1).Previous studies suggested that MeGAX was metab-
olized by E.asburiae JDR-1 into methanol,glucuronate,and
xylose.Glucuronate fermentation by E.asburiae JDR-1 gener-
ated acetate with a nearly 100% yield,indicating that more
reduced fermentation products (ethanol and lactate) could
come only from the free xylose and the xylose released from
MeGAX (2).Therefore,the theoretical maximum yield of
TABLE 2.Comparison of sugar fermentation products of wild-type and genetically engineered E.asburiae JDR-1
Fermentation
Concn of fermentation product (mM)
a
Ethanol yield
(% of theoretical
maximum amt)
b
Succinate Lactate Formate Acetate 2,3-Butanediol Ethanol
D
-Glucose (2.5%,wt/vol)
E.asburiae JDR-1
c
2.0 9.6 39.1 1.0 45.9 45.0 25.6
E.asburiae JDR-1(pLOI297) 1.8 4.7 9.4 3.8 ND 261.6 94.1
E.asburiae JDR-1(pLOI555) 1.6 2 7.7 3.4 ND 265 95.3
D
-Xylose (2%,wt/vol)
E.asburiae JDR-1
c
12.7 5.6 15.0 25.2 13.4 42.6 32.8
E.asburiae JDR-1(pLOI555) 2.2 1.2 3.6 4.2 ND 217.4 98.0
a
Fermentations were carried out at 30°C in minimal media with ZH salts for 48 h as described in Materials and Methods.ND,not detected.
b
Ratio of the amount of ethanol produced to the theoretical maximumamount,expressed as a percentage.A yield of 100%is defined as 2 mol ethanol/mol glucose
or 5 mol ethanol/3 mol xylose.
c
E.asburiae JDR-1 did not completely utilize the substrates within 48 h.
FIG.1.HPLCprofiles of fermentation media of E.asburiae JDR-1,
E.coli KO11,and E.asburiae E1(pLOI555) with 0.5%sweet gumxylan
hydrolysate and 0.1 M MOPS buffer after 48 h of fermentation.The
unlabeled peaks with retention times of 11 min and 21 min are peaks
for salts and buffers.
V
OL
.75,2009 MeGAX FERMENTATION BY ENTEROBACTER 5745
on December 9, 2012 by guesthttp://aem.asm.org/Downloaded from
ethanol from this hydrolysate was calculated to be 35.7 mM
based on the total amount of xylose present in the hydrolysate.
E.asburiae JDR-1 was able to completely utilize the MeGAX
in the 0.5% hydrolysate in about 12 h and was able to com-
pletely utilize the xylose in 20 h after a period of several hours
for adaptation to the hydrolysate medium.Similar amounts of
ethanol (15.6 mM) and acetate (20 mM) were produced along
with small a amount of formate,but no detectable 2,3-butane-
diol was produced;the ethanol yield was 44.2% of the theo-
retical maximum yield (Table 3 and Fig.1 and 2A).The spe-
cific consumption rates for xylose and MeGAX in the
hydrolysate and the specific production rates for acetate and
ethanol are shown in Table 4.
Fermentation characteristics of E.asburiae JDR-1(pLOI297)
and E.asburiae JDR-1(pLOI555).Plasmids pLOI297 andpLOI555
were transformed into E.asburiae JDR-1 for overexpression
of the pdc and adh genes.Both transformed strains were
able to completely utilize 2.5% (wt/vol)
D
-glucose or 2%
(wt/vol)
D
-xylose within 48 h,and ethanol was the predom-
inant fermentation product.The ethanol yields for
D
-glu-
cose fermentation were 94.1% and 95.3% for E.asburiae
JDR-1(pLOI297) and E.asburiae JDR-1(pLOI555),respec-
tively (Table 2).E.asburiae JDR-1(pLOI555) was tested
further for xylose fermentation,and the ethanol yield was
even higher,more than 98% of the theoretical value.Other
fermentation products were also present at concentrations
less than 10 mM (Table 2).
E.asburiae JDR-1(pLOI555) and JDR-1(pLOI297) were
tested for fermentation of dilute acid hyrolysates of sweet
gum MeGAX
n
.Both strains consumed MeGAX,as well as
xylose,within 18 h,and fermentation was complete within
25 h [Fig.2C shows the results for JDR-1(pLOI555);data
for JDR-1(pLOI297) are not shown].The xylose-specific
consumption rate of JDR-1(pLOI555) was similar to that of
TABLE 3.Fermentation products obtained from acid hydrolysates
of sweet gum xylan
a
Strain
Concn of fermentation
product (mM)
Ethanol yield
(% of theoretical
maximum amt)
b
Formic
acid
Acetic acid Ethanol
E.asburiae JDR-1 4.9 ￿0.4 20.0 ￿0.7 15.6 ￿0.8 44 ￿2
E.coli KO11 5.9 ￿1.0 10.6 ￿0.3 22.5 ￿0.2 63 ￿1
E.asburiae JDR-1
(pLOI555)
4.0 ￿0.4 13.5 ￿0.5 26.7 ￿1.0 75 ￿3
E.asburiae JDR-1
(pLOI297)
3.8 ￿0.3 9.9 ￿0.3 30.0 ￿1.5 84 ￿5
E.asburiae
E1(pLOI555)
0 4.5 ￿0.2 35.5 ￿1.1 99 ￿3
a
Fermentations were carried out at 30°C in minimal media with ZH salts for
48 h as described in Materials and Methods.The data are averages ￿ standard
deviations of three experiments.
b
Ratio of the amount of ethanol produced to the theoretical maximum
amount,expressed as a percentage.A yield of 100%is defined as 2 mol ethanol/
mol glucose or 5 mol ethanol/3 mol xylose.
FIG.2.Fermentation time courses for E.asburiae JDR-1 (A),E.coli KO11 (B),E.asburiae JDR-1(pLOI555) (C),and E.asburiae
E1(pLOI555) (D) in media containing buffered sweet gum xylan hydrolysate.The concentrations of substrates and fermentation products were
determined.Symbols:},xylose;f,MeGAX;‚,acetic acid;E,formic acid;￿,ethanol.
5746 BI ET AL.A
PPL
.E
NVIRON
.M
ICROBIOL
.
on December 9, 2012 by guesthttp://aem.asm.org/Downloaded from
the parent strain,but the MeGAX-specific consumption rate
was lower.Ethanol was the major fermentation product,and
the yield was much higher than that obtained with the par-
ent strain.However,both strains produced substantial
amounts of acetate (approximately10 mM),and the yields of
ethanol were lower than the yields when either
D
-xylose or
D
-glucose was the substrate (Table 4).
Comparison of the fermentation characteristics of E.as-
buriae E1(pLOI555) with those of E.coli KO11 and other E.
asburiae JDR-1 derivatives.Neither 2,3-butanediol nor lactic
acid was produced in a hydrolysate fermentation by either E.
asburiae JDR-1(pLOI297) or JDR-1(pLOI555).This result
indicated that only the acetate production pathway initiated
by pyruvate formate lyase competed for pyruvate and de-
creased the ethanol yield.In order to direct greater carbon
flux from pyruvate to ethanol,the pflB gene of E.asburiae
JDR-1 was deleted to obtain strain E.asburiae E1,which
was followed by pLOI555 transformation.When this strain
was tested in hydrolysate fermentation experiments,no for-
mic acid was produced,and only a small amount of acetate
was produced (4.5 mM).After several hours of adaption,the
MeGAX portion was consumed in 12 h and the xylose por-
tion was consumed in 20 h (Fig.2D).While the specific rates
of consumption of the substrates were close to those of
the parent strain and JDR-1(pLOI555),E.asburiae
E1(pLOI555) had a much higher specific rate of production
of ethanol (0.11 ￿ 0.01 g ethanol/g [dry weight] of cells/h)
and a much lower specific rate of production of acetate
(0.022 ￿ 0.003 g acetate/g [dry weight] of cells/h).Most of
the carbon sources in the hydrolysates were converted to
ethanol,resulting in a yield that was 99% of the maximum
theoretical yield (Tables 3 and 4 and Fig.1).
E.coli KO11,which was reported to be able to produce
0.54 g ethanol per g glucose (16),could produce ethanol at
a level that was only 63% of the theoretical maximum level
in the sweet gum xylan hydrolysate medium and accumu-
lated a substantial amount (10.6 ￿ 0.3 mM) of acetate (Fig.
1 and 2C).The sums of the ethanol and acetate concentra-
tions were 33.1 mM for E.coli KO11,40.2 mM for JDR-
1(pLOI555),39.9 mM for JDR-1(pLOI297),and 40.5 mM
for E1 (pLOI555) (Table 3).These results indicated that E.
coli KO11 utilized less substrate in the hydrolysate than the
three engineered E.asburiae strains utilized and produced
smaller amounts of products as a result of its inability to
utilize MeGAX in the hydrolysate (Fig.1 and 2B).The rate
of ethanol-specific production for E.coli KO11 (0.074 ￿
0.006 g ethanol/g [dry weight] cells/h) was much lower than
that for E.asburiae E1(pLOI555) (0.11 ￿ 0.01 g ethanol/g
[dry weight] cells/h) (Table 4).Compared with E.coli KO11,
E.asburiae E1(pLOI555) utilized more substrate in sweet
gum hydrolysate and was able to produce 57.8% more eth-
anol at a higher rate.
PDC activities in E.asburiae strains.Because of the relative
thermal stability of PDCencoded by the pdc gene of Z.mobilis,
heat treatment at 65°C for 15 min was used to inactivate com-
peting native enzymes,including activities that were associated
with the PDC complex and could affect measurements of PDC
activity (4,17).While crude extracts from all strains showed
pyruvate-dependent NADH oxidase activity before heat treat-
ment (data not shown),the wild-type strains were unable to
oxidize NADH after the heat treatment.However,all three
strains carrying a plasmid with the PET operon showed sub-
stantial PDC activities after heat treatment,indicating that
PDC encoded by pdc genes derived from Z.mobilis was
present in these E.asburiae strains.The specific activities of
PDC in cell crude extracts from E.asburiae JDR-1-derived
strains were as follows (averages ￿ standard deviations of
three experiments;1 Uis defined as the amount of enzyme that
catalyzed the conversion of 1 ￿mol of substrate per min at
roomtemperature):E.asburiae JDR-1,no activity;E.asburiae
JDR-1(pLOI297),1.02 ￿0.12 U/mg of cell protein;E.asburiae
JDR-1(pLOI555),0.77 ￿ 0.13 U/mg of cell protein;and E.
asburiae E1(pLOI555),0.53 ￿0.10 U/mg of cell protein.These
results indicate that successful expression of the Z.mobilis pdc
gene in E.asburiae strains is the key factor for predominant
ethanol production.
Plasmid stability in E.asburiae JDR1.The pLOI297 trans-
formant was relatively unstable,and only 10.7% of trans-
formed E.asburiae JDR-1 cells retained tetracycline resistance
after cultivation for 72 generations without antibiotic selection
pressure.The pLOI555 transformant,however,was quite sta-
ble,and 98.1% of pLOI555-transformed E.asburiae JDR-1
cells retained chloramphenicol resistance after growth for 72
generations in the absence of the antibiotic.The stabilities of
pLOI297 and pLOI555 in E.asburiae JDR-1,expressed as the
percentages of cells that retained antibiotic resistance,were as
follows (averages ￿standard deviations of three experiments):
for pLOI297,29.5%￿1.3%after 36 generations and 10.7%￿
2.6% after 72 generations;and for pLOI555,100.0% ￿ 2.8%
after 36 generations and 98.1%￿ 11.8%after 72 generations.
Fermentation analysis of 10 descendant colonies that retained
antibiotic resistance was also performed to confirmthat strains
with antibiotic resistance also retained the homoethanologenic
phenotype.
TABLE 4.Specific consumption rates and specific production rates with acid hydrolysates of sweet gum xylan (5 g/liter)
a
Strain
Specific consumption rate
(g/g ￿dry wt￿ cells/h)
Specific production rate
(g/g ￿dry wt￿ cells/h)
Xylose MeGAX Acetate Ethanol
E.asburiae JDR-1 0.33 ￿0.04 0.087 ￿0.012 0.13 ￿0.01 0.060 ￿0.009
E.coli KO11 0.38 ￿0.04 ND 0.11 ￿0.01 0.074 ￿0.006
E.asburiae JDR-1(pLOI555) 0.29 ￿0.03 0.058 ￿0.012 0.14 ￿0.02 0.052 ￿0.004
E.asburiae E1(pLOI555) 0.32 ￿0.28 0.077 ￿0.13 0.022 ￿0.003 0.11 ￿0.01
a
The data are averages ￿ standard deviations of three experiments.ND,not determined.
V
OL
.75,2009 MeGAX FERMENTATION BY ENTEROBACTER 5747
on December 9, 2012 by guesthttp://aem.asm.org/Downloaded from
DISCUSSION
In the research described here,a wild-type E.asburiae strain
whose genetic and physiological properties are not well known
was genetically engineered to obtain new metabolic potential.
The methodology and protocols developed in this study may be
useful for engineering other wild-type Enterobacter spp.While
E.asburiae JDR-1 was determined to be relatively resistant to
ampicillin and probably other ￿-lactam antibiotics,it was sen-
sitive to tetracycline (12.5 mg liter
￿1
),kanamycin (20 mg
liter
￿1
and 50 mg liter
￿1
),apramycin (20 mg liter
￿1
),and
chloramphenicol (10 mg liter
￿1
and 40 mg liter
￿1
).To deter-
mine if a plasmid-based system developed for use with E.coli
could be maintained and function in E.asburiae JDR-1,plas-
mid pCR4-TOPO with a small insertion was electroporated
into competent cells,and transformants could be selected on a
plate containing kanamycin (50 mg liter
￿1
).The transformed
pCR4-TOPO plasmid was qualitatively determined by DNA
gel electrophoresis to have a lower concentration in E.asburiae
JDR-1 than in the E.coli Top10 host (data not shown).While
not unexpected,these results show that plasmids may be de-
veloped in E.coli for introduction of genes encoding a desired
metabolic potential in E.asburiae JDR-1 and may have appli-
cations in the metabolic transformation of other Enterobacter
spp.as well.
With these transformation systems,E.asburiae JDR-
1(pLOI297) and E.asburiae JDR-1(pLOI555),developed with
the strategy used to genetically engineer the first ethanologenic
strains of E.coli (11),were able to produce ethanol at yields
that were 94.1% and 95.3% of the theoretical maximum yield
in the presence of glucose,but such high yields were not ob-
tained with dilute acid hydrolysates of MeGAX
n
.To decrease
the formation of the organic acids acetate and formate,the
pflB gene was deleted.The convenient one-step gene inactiva-
tion method successfully used for E.coli (5) failed to knock out
the pflB gene in E.asburiae JDR-1,so a different protocol had
to be developed.The alternative gene deletion method used
PCR fragments with several hundred bases of homologous
sequence at both ends instead of the 40 bp used by the one-
step method (13).Longer homologous sequences are expected
to increase the rate of homologous recombination (19) and
may delay complete degradation of the linear DNA by exo-
nucleases,thus increasing the probability of recombination
events.Recombinants were not recovered on plates containing
the levels of antibiotics used for selection of E.coli recombi-
nants,and lower concentrations of kanamycin (20 mg liter
￿1
)
and chloramphenicol (10 mg liter
￿1
) had to be used.This was
likely the basis for growth of nonrecombinant as well as re-
combinant colonies and required a second selection step that
involved patching colonies onto plates containing kanamycin
(50 mg liter
￿1
) and chloramphenicol (40 mg liter
￿1
).By max-
imizing the DNA concentration at approximately 5 ￿g/￿l and
using a concentration of 10
10
cells/100 ￿l for electroporation
transformation,usually three to six E.asburiae JDR-1 recom-
binants could be obtained by this procedure.The methodology
developed here might also be used to engineer other Entero-
bacter spp.with genetic manipulations developed for E.coli.
The E.asburiae strain with a genomic pflB deletion was
transformed with a plasmid,pLOI555,to obtain E.asburiae
E1(pLOI555),a strain capable of efficiently converting the
xylose residues derived from MeGAX
n
to ethanol,resulting in
a yield that was 99%of the theoretical maximum yield.In this
respect this strain was able to outperform the commercial
ethanologenic biocatalyst E.coli KO11 in a mediumcontaining
sweet gum xylan hydrolysate as the sole source of carbon.
The PDC specific activities measured for transformed E.
asburiae strains were noticeably lower than those measured for
the engineered strain Klebsiella oxytoca M5A1 (17),possibly
due to lower copy numbers of plasmids pLOI297 and pLOI555
in E.asburiae JDR-1.However,as found with engineered K.
oxytoca strains,E.asburiae JDR-1(pLOI297) had higher activ-
ity than E.asburiae JDR-1(pLOI555),which may have been
due to the presence of the colEl replicon in pLOI297,resulting
in a copy number higher than that in the strain transformed
with pLOI555.It was found that E.asburiae E1(pLOI555),
which had the highest ethanol yield with hydrolysates,had the
lowest PDC activity in a glucose culture.
The contribution of the adh gene from pLOI1555 is likely
critical to homoethanol production in E.asburiae E1,as it was
in the initial generation of the ethanologenic strains of E.coli
(10,11).When selected genes in E.asburiae JDR-1 were de-
leted to produce E.asburiae L1 with lactate as the predominant
product,fermentation was slow and incomplete without sup-
plementation with Luria-Bertani medium (3),supporting the
conclusion that efficient fermentation to produce a targeted
product requires a high level of expression of the gene encod-
ing the oxidoreductase responsible for generating the final
fermentation product during the reoxidation of NADH.
Plasmid stability is critical for biocatalysts engineered with
genes conferring a desired metabolic potential confined in a
plasmid,as consistent traits are required for long-term appli-
cations.High copy numbers of plasmid pLOI297,containing
the colEl replicon,were present in E.coli strains,but this
plasmid was unstable in K.oxytoca M5A1.pLOI555 derived
from cryptic low-copy-number plasmids in E.coli B (ATCC
11303),however,was very stable in K.oxytoca M5A1 (17).
Similar to the results of studies of K.oxytoca,plasmid pLOI555
was found to be more stable than pLOI297 in E.asburiae
JDR-1.Although plasmid pLOI555 is relatively stable in E.
asburiae E1,introduction of the pdc and adh genes into the
chromosome,which has been done during development of
ethanologenic strains of E.coli (14),may further increase the
stability of an ethanologenic E.asburiae strain.
ACKNOWLEDGMENTS
We thank K.T.Shanmugam for helpful assistance and discussions.
This research was supported by U.S.Department of Energy grants
DE FC36-99GO10476 and DE FC36-00GO10594,by The Consortium
for Plant Biotechnology research project GO12026-198 (DE FG36-
02GO12026),and by the Institute of Food and Agricultural Sciences,
University of Florida Experiment Station,as CRIS project MCS 3763.
REFERENCES
1.Alterthum,F.,and L.O.Ingram.1989.Efficient ethanol production from
xylose,lactose,and glucose by recombinant Escherichia coli.Appl.Environ.
Microbiol.55:1943–1948.
2.Bi,C.,J.D.Rice,and J.F.Preston.2009.Complete fermentation of xylose
and methylglucuronoxylose derived from methylglucuronoxylan by Entero-
bacter asburiae strain JDR-1.Appl.Environ.Microbiol.75:395–404.
3.Bi,C.,X.Zhang,J.D.Rice,L.O.Ingram,and J.F.Preston.6 June 2009.
Genetic engineering of Enterobacter asburiae strain JDR-1 for efficient
D
(￿)-
lactic acid production from hemicellulose hydrolysate.Biotechnol.Lett.
[Epub ahead of print.] doi:10.1007/s10529-009-0044-z.
5748 BI ET AL.A
PPL
.E
NVIRON
.M
ICROBIOL
.
on December 9, 2012 by guesthttp://aem.asm.org/Downloaded from
4.Conway,T.,G.W.Sewell,Y.A.Osman,and L.O.Ingram.1987.Cloning and
sequencing of the alcohol dehydrogenase II gene from Zymomonas mobilis.
J.Bacteriol.169:2591–2597.
5.Datsenko,K.A.,and B.L.Wanner.2000.One-step inactivation of chromo-
somal genes in Escherichia coli K-12 using PCR products.Proc.Natl.Acad.
Sci.USA 97:6640–6645.
6.Dien,B.S.,M.A.Cotta,and T.W.Jeffries.2003.Bacteria engineered for
fuel ethanol production:current status.Appl.Microbiol.Biotechnol.63:258–
266.
7.Dien,B.S.,H.J.G.Jung,K.P.Vogel,M.D.Casler,J.F.S.Lamb,L.Iten,
R.B.Mitchell,and G.Sarath.2006.Chemical composition and response to
dilute-acid pretreatment and enzymatic saccharification of alfalfa,reed ca-
narygrass,and switchgrass.Biomass Bioenergy 30:880–891.
8.Dubois,M.,K.A.Gilles,J.K.Hamilton,P.A.Rebers,and F.Smith.1956.
Colorimetric method for determination of sugars and related substances.
Anal.Chem.28:350–356.
9.Hurlbert,J.C.,and J.F.Preston.2001.Functional characterization of a
novel xylanase from a corn strain of Erwinia chrysanthemi.J.Bacteriol.
183:2093–2100.
10.Ingram,L.O.,and T.Conway.1988.Expression of different levels of eth-
anologenic enzymes from Zymomonas mobilis in recombinant strains of
Escherichia coli.Appl.Environ.Microbiol.54:397–404.
11.Ingram,L.O.,T.Conway,D.P.Clark,G.W.Sewell,and J.F.Preston.1987.
Genetic engineering of ethanol production in Escherichia coli.Appl.Envi-
ron.Microbiol.53:2420–2425.
12.Jacobs,A.,P.T.Larsson,and O.Dahlman.2001.Distribution of uronic
acids in xylans from various species of soft- and hardwood as determined by
MALDI mass spectrometry.Biomacromolecules 2:979–990.
13.Jantama,K.,M.J.Haupt,S.A.Svoronos,X.Zhang,J.C.Moore,K.T.
Shanmugam,and L.O.Ingram.2008.Combining metabolic engineering and
metabolic evolution to develop nonrecombinant strains of Escherichia coli C
that produce succinate and malate.Biotechnol.Bioeng.99:1140–1153.
14.Jarboe,L.R.,T.B.Grabar,L.P.Yomano,K.T.Shanmugan,and L.O.
Ingram.2007.Development of ethanologenic bacteria.Adv.Biochem.Eng.
Biotechnol.108:237–261.
15.Kardosova,A.,M.Matulova,and A.Malovikova.1998.(4-O-methyl-alpha-
D
-glucurono)-
D
-xylan from Rudbeckia fulgida,var.sullivantii (Boynton et
Beadle).Carbohydr.Res.308:99–105.
16.Ohta,K.,D.S.Beall,J.P.Mejia,K.T.Shanmugam,and L.O.Ingram.1991.
Genetic improvement of Escherichia coli for ethanol production chromo-
somal integration of Zymomonas mobilis genes encoding pyruvate decarbox-
ylase and alcohol dehydrogenase II.Appl.Environ.Microbiol.57:893–900.
17.Ohta,K.,D.S.Beall,J.P.Mejia,K.T.Shanmugam,and L.O.Ingram.1991.
Metabolic engineering of Klebsiella oxytoca M5A1 for ethanol production
from xylose and glucose.Appl.Environ.Microbiol.57:2810–2815.
18.Pordesimo,L.O.,B.R.Hames,S.Sokhansanj,and W.C.Edens.2005.
Variation in corn stover composition and energy content with crop maturity.
Biomass Bioenergy 28:366–374.
19.Puchta,H.,and B.Hohn.1991.A transient assay in plant cells reveals a
positive correlation between extrachromosomal recombination rates and
length of homologous overlap.Nucleic Acids Res.19:2693–2700.
20.Rodriguez,M.,E.A.Martinez,S.W.York,K.Zuobi-Hasona,L.O.Ingram,
K.T.Shanmugam,and J.F.Preston.2001.Properties of the hemicellulose
fractions of lignocellulosic biomass affecting bacterial ethanol production,p.
535.Abstr.101st Gen.Meet.Am.Soc.Microbiol.American Society for
Microbiology,Washington,DC.
21.Saha,B.C.2003.Hemicellulose bioconversion.J.Ind.Microbiol.Biotech-
nol.30:279–291.
22.Saha,B.C.,L.B.Iten,M.A.Cotta,and Y.V.Wu.2005.Dilute acid
pretreatment,enzymatic saccharification,and fermentation of rice hulls to
ethanol.Biotechnol.Progr.21:816–822.
23.Sun,J.X.,X.F.Sun,R.C.Sun,and Y.Q.Su.2004.Fractional extraction and
structural characterization of sugarcane bagasse hemicelluloses.Carbohydr.
Polym.56:195–204.
24.Sun,R.C.,J.M.Fang,J.Tomkinson,Z.C.Geng,and J.C.Liu.2001.
Fractional isolation,physico-chemical characterization and homogeneous
esterification of hemicelluloses from fast-growing poplar wood.Carbohydr.
Polym.44:29–39.
25.Timell,T.E.1964.Wood hemicelluloses.1.Adv.Carbohydr.Chem.19:247–
302.
26.Wood,B.E.,L.P.Yomano,S.W.York,and L.O.Ingram.2005.Develop-
ment of industrial-medium-required elimination of the 2,3-butanediol fer-
mentation pathway to maintain ethanol yield in an ethanologenic strain of
Klebsiella oxytoca.Biotechnol.Progr.21:1366–1372.
27.Zhang,X.,K.Jantama,J.C.Moore,K.T.Shanmugam,and L.O.Ingram.
2007.Production of
L
-alanine by metabolically engineered Escherichia coli.
Appl.Microbiol.Biotechnol.77:355–366.
28.Zucker,M.,and L.Hankin.1970.Regulation of pectate lyase synthesis in
Pseudomonas fluorescens and Erwinia carotovora.J.Bacteriol.104:13–18.
V
OL
.75,2009 MeGAX FERMENTATION BY ENTEROBACTER 5749
on December 9, 2012 by guesthttp://aem.asm.org/Downloaded from