Y.K. Dhande et al. / Process Biochemistry 47 (2012) 1965–1971
1967
to 2-ketoisocaproate through a 3-step chain elongation cycle cat-
alyzed by 2-isopropylmalatesynthase (LeuA), isopropyl malate
isomerase complex (LeuC, LeuD) and 3-isopropylmalate dehydro-
genase (LeuB). It was previously demonstrated that LeuA, LeuB,
LeuC and LeuD are promiscuous enough to allow 2KB to go through
the same elongation cycle to produce 2-ketovalerate (2KV) and fur-
ther to 2-ketocaproate (2KC) (4). 2KC can then be decarboxylated
by a DC into valeraldehyde and oxidized subsequently to PA by a
DH.
appropriate aldehyde dehydrogenase and corresponding 2-keto acid in the range
of 1–8 mM in assay buffer (50 mM NaH2PO4, pH 6.8, 1 mM MgSO4, 0.5 mM ThDP)
with a total volume of 78 l. To start the reaction, 2 l of 1 M IPDC was added and
generation of NADH was monitored at 340 nm. Kinetic parameters (kcat and KM)
were determined by fitting initial rate data to the Michaelis–Menten equation.
3. Results and discussion
3.1. Construction of metabolic pathways for biosynthesis of 2MB
and PA
2. Materials and methods
All the enzymes downstream of aspartate biosynthesis were
2.1. Bacterial strains, reagents, media and cultivation
overexpressed from three synthetic operons. The constructed oper-
ons for PA and 2MB are shown in Fig. 2A and B respectively. The
genes thrA, thrB and thrC involved in threonine synthesis consti-
tuted the first operon under the regulation of PLlacO1 promoter
on a low copy plasmid carrying spectinomycin resistance marker.
For 2MB synthesis, ilvA, ilvG, ilvM, ilvC and ilvD were cloned on a
medium copy plasmid with a kanamycin resistance marker. For
synthesis of PA, ilvA, leuA, leuB, leuC and leuD were cloned on a
medium copy plasmid carrying a kanamycin resistance marker.
Various aldehyde dehydrogenases and ketoacid decarboxylases
were present in the transcriptional order DC-DH on high copy plas-
mids carrying an ampicillin resistance marker.
Since threonine is a common intermediate in both the path-
ways, a threonine overproducer E. coli strain ATCC98082 was used
in the study. The strain had threonine exporter gene rhtA removed
to ensure high intracellular level of threonine [8]. The alcohol dehy-
drogenase yqhD gene deletion was performed to eliminate the side
reactions leading to production of alcohols. The resultant strain is
referred to hereafter as the PA1 strain.
The E. coli strain used in this study was a threonine overproducer strain
ATCC98082 which had threonine and homoserine exporter gene rhtA knocked out
to ensure high intracellular levels of threonine [8]. The yqhD gene deletion strain
was obtained from the Keio collection [20]. It was transformed with plasmid pCP20
to remove the kanamycin resistance marker. The various strains and plasmids used
in this study are listed in Table 1. The resultant strain (PA1) was transformed with
plasmids pIPA1, pIPA2 and any one of the pIPA4 to pIPA15 for production of 2MB.
For production of PA, it was transformed with pIPA1, pIPA3 and any one of the pIPA4
to pIPA15.
XL1-Blue and XL10-Gold competent cells from Stratagene (La Jolla, CA) were
used for propagation of plasmids while BL21 (DE3) competent cells from New Eng-
land Biolabs (Ipswich, MA) were used for protein expression. All the restriction
TM
®
enzymes, Quick ligation
New England Biolabs.
kit and Phusion high-fidelity PCR kit were also from
A 2×YT rich medium (16 g/L Bacto-tryptone, 10 g/L yeast extract and 5 g/L NaCl)
◦
was used to culture the E. coli strains at 37 C and 250 rpm. Antibiotics were added
as needed (100 mg/L ampicillin, 25 mg/L kanamycin and 25 mg/L spectinomycin).
2
.2. Production experiments and HPLC analysis
Production experiments were carried out in triplicates. Three independent
colonies were picked from freshly transformed E. coli strains and cultured overnight
in 2 ml 2×YT medium containing appropriate antibiotics. 250 l of overnight cul-
tures were transferred into 125 ml conical flasks containing 5 ml M9 medium
supplemented with 40 g/L glucose, 5 g/L yeast extract, 10 mg/L thiamine, 100 mg/L
ampicillin, 25 mg/L kanamycin and 25 mg/L spectinomycin. Protein expression was
induced by adding 0.1 mM isopropyl--d-thiogalactoside (IPTG). 0.2 g CaCO3 was
added into the flask for neutralization of acids produced. After incubation for 48 h
3
.2. Synthesis of C5 carboxylic acids
The key steps in the designed pathways are decarboxylation of
ketoacids KMV and 2KC into respective aldehydes, followed by oxi-
dation to carboxylic acids. The success was dependent on discovery
of enzymes that can catalyze these reactions. Based on our previ-
ous work on producing isobutyric acid [18], we cloned wild-type
◦
at 30 C and 250 rpm, samples were centrifuged, supernatants were collected and
diluted twice for analysis. Analysis was done using an Agilent 1260 Infinity HPLC
containing an Aminex HPX 87H column (Bio-Rad, USA) equipped with a refractive-
index detector. The mobile phase was 5 mM H2SO4 at a flow rate of 0.6 ml/min. The
2
-ketoisovalerate decarboxylase KIVD from Lactococcus lactis [21]
◦
◦
column temperature was 35 C and detection temperature was 50 C. The loss due to
vaporization was estimated to be less than 5%. The data are presented as the mean
values with error bars indicating the standard error.
and phenylacetaldehyde dehydrogenase PadA [22] from E. coli to
check for the production of our target chemicals. KIVD has been
previously shown to have a wide substrate range for converting
2-ketoacids into aldehydes [3,5]. The PA1 strain was transformed
with plasmids pIPA1, pIPA2 and pIPA4 for 2MB and plasmids pIPA1,
pIPA3 and pIPA4 for PA synthesis. Shake flask experiments were
2.3. Protein expression and purification
The enzyme AldH was purified by cloning the gene into an expression plasmid
having N-terminal 6xhis-tag (pIPA16). This plasmid was then transformed into E. coli
◦
carried out with the recombinant strain at 30 C and samples were
strain BL21. Cells were inoculated from an overnight pre-culture at 1/300 dilution
◦
analyzed by HPLC as described in the methods section. Through this,
productions of 2.26 g/L for 2MB and 2.1 g/L for PA were achieved
which demonstrated the feasibility of our biosynthetic approach.
and grown at 30 C in 300 ml 2xYT rich medium containing 100 mg/L ampicillin.
When the OD reached 0.6, IPTG was added for induction of protein expression. Cell
pellets were lysed by sonication in a buffer (pH 9.0) containing 250 mM NaCl, 2 mM
DTT, 5 mM imidazole and 50 mM Tris. The enzyme was purified from crude cell
lysate through Ni-NTA column chromatography and buffer-exchanged using Ami-
con Ultra centrifugal filters (Millipore). 50 M tris buffer (pH 8), 1 mM MgSO4 and
3.3. Screening of aldehyde dehydrogenases
2
0% glycerol was used for storage of AldH. 100 l of concentrated protein solutions
◦
were aliquoted into PCR tubes and flash frozen at −80 C for long term storage. Pro-
tein concentration was determined by measuring UV absorbance at 280 nm. Purified
KDHba and IPDC were available from another study [19].
In order to improve production titers, the effect of choosing dif-
ferent aldehyde dehydrogenases was examined (Fig. 3A and B). It
was speculated that some of the aldehyde dehydrogenases would
be promiscuous enough to catalyze these conversions. Six such
enzymes (including PadA) were selected as candidate enzymes
for this study. They were acetaldehyde dehydrogenase AldB [23],
2.4. Enzymatic assay
Enzymatic assay of KDHba consisted of 0.5 mM NAD+ and valeraldehyde in the
range of 50 to 400 M in assay buffer (50 mM NaH2PO4, pH 8.0, 1 mM DTT) with a
total volume of 78 l. To start the reaction, 2 l of 1 M KDHba was added and gener-
3
-hydroxypropionaldehyde dehydrogenase AldH [24], phenylac-
−
1
−1
etaldehyde dehydrogenase PadA [22], succinate semialdehyde
dehydrogenase GabD [25], ␥-aminobutyraldehyde dehydrogenase
YdcW [26] from E. coli and ␣-ketoglutaric semialdehyde dehy-
drogenase KDHba [24] from Burkholderia ambifaria. Strains were
constructed with three synthetic operons as shown in Fig. 2. All
the enzymes except DHs were the same and wild-type KIVD was
ation of NADH was monitored at 340 nm (extinction coefficient, 6.22 mM cm ) at
room temperature. Similar protocol was used for AldH with 2-methyl butyraldehyde
concentrations in the range of 1–6 mM.
The activity of IPDC was measured using a coupled enzymatic assay method.
Excess of appropriate aldehyde dehydrogenase (AldH for 2-keto-3-methylvalerate
and KDHba for 2-ketocaproate) was used to oxidize aldehyde into acid while cofactor
NAD+ was reduced to NADH. The assay mixture contained 0.5 mM NAD+, 0.1 M