J. Breitenbach et al. / Archives of Biochemistry and Biophysics 529 (2013) 86–91
87
mixture with diethyl ether/petroleum ether (b.p. 35–60 °C) (1:1, v/
v) and analyzed in HPLC system II as described below.
Experimental procedures
For determination of the Km value for b-carotene and echine-
none, substrate concentrations varied in the range from 1 to
Growth of organisms
20 lM. Km and Vmax values were obtained as mean from three dou-
Cultivation of E. coli was in darkness in LB medium either rou-
tinely at 37 °C for 3 days or for only 1 day for better accumulation
and isolation of 4-hydroxy-b-carotene or for protein expression.
The strain DH5a was used for cloning and synthesis of carotenoids
and JM101 for protein expression.
ble reciprocal Lineweaver–Burk plots of the reaction rates versus
the substrate concentration with five data points and a correlation
coefficient for linearity of >0.97. I50 values for inhibition of the
ketolase reaction by diphenylamine (DPA)2 were determined with
concentrations from 50 to 600 lM.
Plasmids
Carotenoid analysis
The expression vector pPEU30 was performed constructed by
PCR amplification of the whole promoter region of the carotenogen-
ic gene cluster from Pantoea ananatis formerly Erwinia uredovura
out of plasmid pACCRT-EBI [13] with primers forward
TACCGCTCGAGCTGCCAA and revers GAATTCTCGGGCTGTCCTTAT
generating a XhoI and EcoRI restriction site, subcloning into the
T-overhang vector pMON38201 [14], cut out with XhoI/EcoRI and
ligated into the XhoI/EcoRI sites of plasmid pPQE30 (Quiagen,
Hilden, Germany). Then, the crtO coding region was PCR amplified
from plasmids pslr0088 [9] with primers forward CCATGGGATC-
CATCACCACCG and revers TTCTGCAGTTACCAAAAACGACG generat-
ing a BamHI and PstI restriction site, and cloned into the BamHI/PstI
sites of pPEU30 generating the expression vector pPEU30crtO.
Other plasmids used for co-expression in E. coli were pAC-
Carotenoids were extracted from freeze-dried cells by heating
in acetone (60 °C, 20 min) and partitioning into 10% ether in petrol.
The carotenoids from the upper phase were either used directly for
analysis or fractionated by TLC to concentrate the ketolation inter-
mediates 4-hydroxy-b-carotene and 4-hydroxy-echinenone. TLC
was carried out on activated silica plates with the solvent tolu-
ene/ethylacetate/methanol (90:9:1, v/v/v). The TLC band contain-
ing the 4-hydroxy derivatives with an Rf-value of 0.45 above a
0.35 band containing cis cantaxanthin was scraped off and the
carotenoids eluted with acetone. The fraction from a 1-day old cul-
ture was used to identify 4-hydroxy-b-carotene and a 3-day old
culture for 4-hydroxy-echinenone. This was achieved on a HPLC
system I on a 15 cm Nucleosil C18, 3l column with a mobile phase
CAR16
containing a combined lycopene
pACCRT-EBI [13] for the synthesis of
D
crtX to generate a b-carotene background [13], pQE970
- and b-cylase [15], together with
-carotene, pRKcrtR [16] for
of acetonitrile/2-propanol/methanol/water (82:5:10:3, v/v/v/v),
flow 1 ml/min, at 10 °C column temperature. The other carotenoid
separations were carried out with HPLC system II on a 25 cm Vydac
e
a
the synthesis of b-cryptoxanthin from b-carotene, and pCRBKT for
the synthesis of canthaxanthin from b-carotene [17].
201TP C18, 10l column with 2% water in methanol, flow 1 ml/min,
at 25 °C. Spectra were recorded on-line with a Kontron DAD 440
diode array detector (Kontron Instruments, Neufahrn, Germany).
Peaks were identified by co-chromatography with reference com-
pounds together with their spectra. Authentic carotenoid stan-
dards for HPLC were generated by combinatorial biosynthesis in
E. coli [19] or in case of 4-hydroxy-b-carotene, 4-hydroxy-40-
keto-b-carotene and 4,40-dihydroxy-b-carotene by reduction of
echinenone or canthaxanthin with NaBH4 [20].
Enzyme purification after heterologous expression in E. coli
Escherichia coli cells were re-suspended 1:30 of the culture vol-
ume in 100 mM phosphate buffer pH 7.8 containing 0.1% Nonidet
P40, broken in a French pressure cell at 95 MPa and the homoge-
nate centrifuged at 40,000 g for 20 min at 4 °C. After incubation
of 15 ml of the supernatant with 1.2 ml of the talon resin (BD Clon-
tech, Heidelberg, Germany) for 45 min on ice, the mixture was
packed in a column and non-bounded proteins were washed four
times with 5 ml buffer containing 100 mM potassium phosphate
buffer pH 7.8, 300 mM sodium chloride, 10% glycerol and 10 mM
imidazole. For elution of proteins (in 1.5 ml fractions), increasing
imidazole concentrations (50, 100, 150 and 200 mM) were used.
The 200 mM fraction used for in vitro assays was desalted on
Sephadex G25 (PD 10 column, Pharmacia). For the documentation
Software
Data base searches were carried out with the similarity search
tool BLAST P 2.2.10 [21]. Phylogenetic analysis of amino acid se-
quences were performed with the program Clustal X [22] and the
alignments visualized with TreeView.
of the different purification steps 270 ll of each fraction was pre-
Results
cipitated after addition of 6% of trichloric acid and applied for sep-
aration on a 12.5% SDS–PAG [18]. The relative amounts of the
expressed proteins were estimated after separation on the SDS-
gel by staining with Coomassie Brilliant Blue and scanning using
a densitometric software. Protein concentrations were determined
with the BioRad protein assay.
The phylogenetic tree of Fig. 1 with CrtI-related enzymes from
bacteria consists of four branches starting from a common origin.
CrtO type ketolases from different prokaryotes group well to-
gether. CrtO from Synechocystis is closest to the other cyanobacte-
rial monoketolase from Gloeobacter violaceus. Protein identity
among the cyanobacterial enzymes in this cluster is around 60%.
The cyanobacterial ketolases join with CrtOs from other bacteria
to a common branch. Another cluster represents bacterial phyto-
ene desaturases (CrtI). It is close to the 4,40-diapophytoene desatu-
rases (CrtN). In a functional sense, these enzymes differ by their
specificity for a C40 or C30 substrate. Connected to them are the
CrtNb ketolases which show sequence homology to these desatu-
rases [11]. In contrast to the ketolation activity of CrtO, CrtNb en-
zymes form an aldehyde group at the terminal C-atom of acyclic
In vitro enzyme assay
The ketolase assays contained 230
imidazole fraction (equivalent to 250
l
l of the desalted 200 mM
l
g protein), 4 g of b-caro-
l
tene or other carotenoids as indicated, 0.1% Nonidet P40, 1 mM
cofactor (in 20 l buffer or 10 l methanol for ubiquinone and plas-
l
l
toquinone) in a final volume of 1.5 ml 0.1 M potassium phosphate
buffer pH 7.8. After incubation for 16 h at 28 °C, the reaction was
terminated by addition of 2.5 ml of methanol. The remaining sub-
strate and the products formed were extracted from the incubation
2
Abbreviations used: DPA, diphenylamine.