Molecular cloning, expression and characterization of the first three genes in the mevalonate-independent isoprenoid pathway in Streptomyces coelicolor

Article abstract of DOI:10.1016/S0968-0896(01)00050-5

The mevalonate-independent biosynthetic pathway to isopentenyl diphosphate and dimethylallyl diphosphate, the universal precursors to the isoprenoids, operates in eubacteria, including Escherichia coli, in algae, and in the plastids of higher plants. A search of the Sanger Centre Streptomyces coelicolor genome database revealed open reading frames with ca. 40-50% identity at the deduced amino acid level to the first three E. coli enzymes of this pathway, corresponding to deoxyxylulose phosphate synthase, deoxyxylulose phosphate reductoisomerase and 2-C-methyl erythritol 4-phosphate cytidyltransferase. The coelicolor genes have been cloned and expressed in E. coli, and the recombinant proteins characterized physically and kinetically. The presence of the corresponding enzyme activities in Extracts of S. coelicolor CH999 further supports the operation of the mevalonate-independent pathway in this organism. Copyright

Full text of DOI:10.1016/S0968-0896(01)00050-5

Bioorganic & Medicinal Chemistry 9 (2001) 1467–1477
Molecular Cloning, Expression and Characterization of the First
Three Genes in the Mevalonate-Independent Isoprenoid Pathway
in Streptomyces Coelicolor
David E. Cane,* Cathy Chow, Antonietta Lillo and Ilgu Kang
Department of Chemistry, Box H, Brown University, Providence, RI 02912-9108, USA
Received 6 December 2000; accepted 15 January 2001
Abstract—The mevalonate-independent biosynthetic pathway to isopentenyl diphosphate and dimethylallyl diphosphate, the uni-
versal precursors to the isoprenoids, operates in eubacteria, including Escherichia coli, in algae, and in the plastids of higher plants.
A search of the Sanger Centre Streptomyces coelicolor genome database revealed open reading frames with ca. 40–50% identity at
the deduced amino acid level to the first three E. coli enzymes of this pathway, corresponding to deoxyxylulose phosphate synthase,
deoxyxylulose phosphate reductoisomerase and 2-C-methyl erythritol 4-phosphate cytidylyltransferase. The S. coelicolor genes have
been cloned and expressed in E. coli, and the recombinant proteins characterized physically and kinetically. The presence of the
corresponding enzyme activities in extracts of S. coelicolor CH999 further supports the operation of the mevalonate-independent
pathway in this organism. # 2001 Elsevier Science Ltd. All rights reserved.
Introduction
The first biochemical step in the mevalonate-indepen-
dent pathway involves the thiamin diphosphate-depen-
dent condensation of pyruvate and d-glyceraldehyde-3-
phosphate to give the five-carbon sugar derivative, d-1-
deoxyxylulose-5-phosphate (4, DXP). (Scheme 2) Inter-
estingly, the free alcohol, 1-deoxyxylulose (5, DX), has
been known for many years, having previously been
isolated from Streptomyces hygroscopicus, although the
function of this five-carbon ketose had been completely
obscure.3 The structural genes for DXP synthase were
first cloned and expressed from E. coli4a and from pep-
permint (Mentha x piperita).4b The E. coli DXP syn-
thase shows pronounced similarity to several
demonstrated or putative DXP synthases from a variety
of bacterial and plant species. Notably, DXP has also
recently been shown to be a key intermediate in forma-
tion of both pyridoxol phosphate (vitamin B6)5 and
thiamin (vitamin B1) itself in E. coli.6
For many years, following the discovery in 1956 that
mevalonic acid (1) is a precursor of cholesterol, the
central dogma of isoprenoid biosynthesis was that all
terpenoid natural products, of whatever structure—
from monoterpenes to triterpenes—and from whatever
source—from microbes to mammals—were derived
from the universal precursor, mevalonic acid.1 All the
steps involved in the conversion of mevalonate to the
biological isoprene unit, isopentenyl diphosphate (2,
IPP), have been identified and the relevant enzymes
thoroughly characterized (Scheme 1). Given the wide-
spread occurrence of the mevalonate pathway, it was
therefore especially startling when in the early 1990s two
research groups independently discovered a second bio-
chemical pathway to IPP that involves neither acetate
nor mevalonate.2 The mevalonate-independent pathway
was shown to begin with pyruvate and glyceraldehyde-
3-phosphate (G-3-P) (Scheme 2). Subsequent investiga-
tions have established that this alternative pathway is
not a rare exception, but in fact occurs widely in a range
of prokaryotic organisms, in algae, and in the
chloroplasts or plastids of numerous higher plants.
*Corresponding author. Tel.: +1-401-863-3588; fax: +1-401-863-
3556; e-mail: david_cane@brown.edu
Scheme 1. Conversion of acetyl-CoA to IPP by way of mevalonate
(1).
0968-0896/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00050-5

1468
D. E. Cane et al. / Bioorg. Med. Chem. 9 (2001) 1467–1477
Scheme 2. Mevalonate-independent biosynthetic pathway to IPP and isoprenoids, initiated by the TPP-dependent condensation of pyruvate and
G-3-P to give d-1-deoxyxylulose-5-phosphate (4, DXP).
In the next step in the mevalonate-independent pathway
to IPP, DXP is rearranged to 2-C-methyl-d-erythritol-4-
phosphate (6, MEP) by an NADPH-dependent reducto-
isomerase (DXP reductoisomerase).7 MEP has been
shown to undergo CTP-dependent conversion to 4-
diphosphocytidyl-2-C-methylerythritol (7, CDP-ME)
catalyzed by the E. coli protein, MEP cytidylyltransfer-
ase.8 Very recently, it has been established that the 2-
hydroxyl group of 7 is phosphorylated by the ychB gene
product, CDP-ME kinase, to produce 4-(cytidine 50-di-
phospho)-2-C-methyl-d-erythritol-2-phosphate (8, CDP-
ME2P).9 The last identified step in the mevalonate-
independent pathway involves the conversion of 8 to 2-
C-methyl-d-erythritol 2,4-cyclodiphosphate (9) with
concomitant elimination of cytidine 50-monophosphate,
catalyzed by the E. coli YgbB protein.10 The steps from
9 to isopentenyl diphosphate are as yet unknown and
are the object of intensive investigation in several
laboratories. Interest in the enzymes of this pathway has
been further heightened by the possibility that they
might serve as targets for antibacterial or antimalarial
drug therapy.11
Streptomyces are filamentous, Gram-positive soil bac-
teria. The Streptomycetes and related actinomycetes are
the source of two-thirds of all antibiotics. By far the
most thoroughly studied member of this genus is Strep-
tomyces coelicolor, which has been extensively char-
acterized genetically and serves as
a model of
morphological and physiological development in myce-
lial prokaryotes.15 Hopwood has recently established a
detailed physical and genetic map for S. coelicolor A3(2)
based on a complete set of ordered, overlapping cos-
mids that cover the entire 8 Mb linear chromosome.16
Using these cosmids, the sequencing of the S. coelicolor
genome has been undertaken by the Sanger Centre, and
has been mostly completed as of October 2000 with only
a few gaps remaining to be completed. The entire
sequence is expected to be completed by 2001. Using the
translated E. coli sequences as a query for a TBLASTN
search of the S. coelicolor genome database reveals open
reading frames with high levels of similarity to the E.
coli enzymes corresponding to dxps, dxr, ygbP and
ygbB. Interestingly, in the case of dxps, not one, but two
distinct open reading frames can be identified in S. coe-
licolor.
As part of our own long-standing interest in the bio-
synthesis of isoprenoids, we reported in 1981 the incor-
poration of [UL-13C6]glucose into a sesquiterpene
metabolite of Streptomyces UC5319, the antibiotic pen-
talenolactone (10).12 Although the observed pattern of
enhancements and couplings in the 13C NMR spectrum
of 10 was mostly as predicted for formation of pentale-
nolactone by the classical mevalonate pathway, an
anomalous set of 13C–13C couplings could not be satis-
factorily explained. The eventual discovery of the novel
pyruvate/G-3-P route to IPP cast these earlier observa-
tions on the labeling of pentalenolactone in an entirely
new light. Indeed, all of our earlier reported results
could be crisply accounted for by operation of the
mevalonate-independent pathway to generate IPP and
derived isoprenoid metabolites. In support of this pro-
posal, feeding experiments with Streptomyces UC5319
confirmed that labeled 1-deoxyxylulose (5) can serve as
a specific precursor of 10.13 (Scheme 3) In the meantime,
Seto has investigated isoprenoid biosynthesis in several
Streptomyces species, finding evidence for both the
mevalonate and the mevalonate-independent routes to
IPP and that these two pathways can sometimes co-exist
in the same species.14
We now report the PCR cloning and the expression in
E. coli of the S. coelicolor dxps, dxr and ygbP genes, and
the subsequent characterization of the respective pro-
teins. In complementary experiments, the activity of the
three native enzymes was also determined in cell-free
extracts of S. coelicolor CH999.
Scheme 3. Conversion of deuterated deoxyxylulose (5) to pentaleno-
lactone (10) by Streptomyces UC5319.

D. E. Cane et al. / Bioorg. Med. Chem. 9 (2001) 1467–1477
1469
Results
Purification and physical characterization of DXP syn-
thases—After optimization of the individual expression
conditions, both DXP synthases were purified to
homogeneity by similar protocols (Table 1). The
experimental level of expression of DXPS1 varied from
1–10% of total (soluble) protein in different runs, while
expression levels for DXPS2 appeared to be more
reproducible as ca. 20% of total soluble protein.
DXPS1 also appeared to be somewhat less stable upon
storage and frequently rapidly lost activity. The identity
of each of the purified proteins was verified by N-term-
inal sequencing, which gave results (DXPS1: TILE-
NIRGPRNLKALSEA; DXPS2: XXLTRITGPRN-
LNRLSL) consistent with the deduced amino acid
sequence for each protein after removal of the N-term-
inal methionine.18 ES–MS analysis of DXPS1 gave a
molecular mass of MD 69,416ꢀ13 (calcd: 69,513 Da)
and DXPS2 gave a molecular mass of MD 68,482ꢀ14
(calcd: 68,433 Da), again consistent with the absence of
the N-terminal Met. Gel filtration on a calibrated
Superose 12 Hr 10/30 column gave apparent molecular
weights of Mr 162,000 and 145,500 for DXPS1 and
DXPS2, respectively, indicating that both were homo-
dimers. Preparative scale incubation of pyruvate and
glyceraldehyde-3-phosphate with DXPS2 in the pre-
sence of thiamin diphosphate gave 1-deoxyxylulose-5-
phosphate (4), the structure of which, was directly con-
firmed by 1H NMR comparison with a synthetic reference
sample.19
Construction of plasmids and expression of dxps1 and
dxps2 from S. coelicolor—A TBLASTN search of the S.
coelicolor A3(2) genomic database compiled by the
Sanger Centre revealed the presence of two apparent
open-reading frames with ca. 50% identity at the
deduced amino acid level gene to the E. coli DXP syn-
thase protein.17 (Fig. 1) One of these open reading
frames, designated dxps1, was contained within S. coe-
licolor cosmid St6A5 (SC6A5;17 nt 16676–18646), while
the second, dxps2 matched sequences at the overlapping
ends of cosmids St7B7 (SC7B7;17 nt 12,742–13,800-end)
and SC1C3 (nt 158–1027. nt 1–157 reported for cosmid
SC1C3 overlap nt 13,644–13,800 at the 30-end of
SC7B7). In fact, the complete open reading frame for
dxps2 is actually included in the 30 end of cosmid St7B7.
Both of the apparent Streptomyces ORFs corresponded
to proteins similar in length to the E. coli DXS protein
and showed a typical transketolase signature motif.
After PCR cloning and expression, both sets of E. coli
BL21(DE3)/pLysS transformants expressed proteins
which migrated on 10% SDS–PAGE with the expected
relative masses, Mr 71 kDa (DXPS1, predicted:
69.36 kDa) and 70kDa (DXPS2, predicted: 68.43 kDa),
compared to a reference sample of E. coli deoxyxylulose
phosphate synthase, Mr 68 kDa (predicted: 67.61 kDa)
(data not shown). The majority of each of the recombi-
nant proteins was present in the soluble protein fraction.
Figure 1. Amino acid sequence alignment of deoxyxylulose-5-phosphate synthases from S. coelicolor (dxps1 and dxps2), Streptomyces sp. CL190
(cl190), and E. coli (dxs). Sequence identities: dxps1/dxps2, 80.1%; dxps1/cl190, 88.7%; dxps1/dxs, 57.7%; dxps2/cl190, 78.9%; dxps2/dxs, 57.3%;
cl190/dxs, 55.6%.

1470
D. E. Cane et al. / Bioorg. Med. Chem. 9 (2001) 1467–1477
Kinetic parameters for DXP synthases
vector, followed by transformation into E. coli BL21
strains for expression of proteins. While dxr was routi-
nely transformed into E. coli BL21(DE3)/pLysS com-
petent cells, examination of the ygbP sequence with
respect to codon usage by E. coli revealed several rare
codons (proline: CCC; arginine: CGG; glycine: GGG/
GGA) typically found in GC-rich genomes. Taking
account of this, commercially available E. coli
BL21(DE3)-RP competent cells were selected as the
recipient for the transformation of ygbP since this strain
was engineered to facilitate the reliable expression of
genes restricted by either AGG/AGA or CCC codons.
The steady state kinetic parameters for both enzymes
were very similar (Table 2). The kinetic parameters of a
sample of E. coli DXP synthase, which had not been
previously published,20 were similar in magnitude. Both
Streptomyces enzymes showed only slightly higher
turnover numbers than the E. coli enzyme and some-
what lower Km values for each substrate. Both DXP
synthases had broad pH optima, with maxima at pH 7.5
and 8.0, respectively.
Construction of plasmids and expression of S. coelicolor
dxr and ygbP
Purification and physical characterization of DXP
reductoisomerase and MEP cytidylyltransferase
A TBLASTN search of the S. coelicolor A3(2) genomic
database revealed the presence of an open-reading
frame contained within cosmid St5H4 (SC5H4;17 nt.
24,696–25,952) with ca. 50.1% similarity and 41.6%
identity at the deduced amino acid level to the E. coli
DXP reductoisomerase17 and an open-reading frame
contained within cosmid StD16A (SCD16A,17 nt. 4975–
5788) that had ca. 44.7% similarity and 36.9% identity
to the E. coli MEP cytidylyltransferase.17 After PCR
amplification, plasmids were constructed following
standard cloning procedures into the pET11 expression
The majority of the expressed DXP reductoisomerase
was found in the pellet fraction, in the form of inclusion
bodies, after cell lysis and centrifugation, with only ca.
10% of expressed reductoisomerase being soluble. After
optimization of the expression conditions, the soluble
DXP reductoisomerase fraction was purified to ca. 90%
homogeneity, as determined by SDS–PAGE, in one
chromatographic step using Reactive Red 120matrix,
which contains a ligand resembling the nicotinamide
ring. In this way, DXR protein, which is dependent
Table 1. Purification of recombinant S. coelicolor DXP synthases (DXPS1 and DXPS2) and MEP cytidylyltransferase (YgbP)
Enzyme
DXPS1
Purification step
Protein
(mg)
Activity
(nmol DXP or CDP-ME/min)
Spec. Act.
(nmol DXP or CDP-ME/min/mg)a
Purification (ꢁ-fold)
Crude extract
DE52
t-Bu-HIC
Resource Q
Crude extract
DE52
56
62.7
153
1.12
12.2
61.1
124
327
641
1202
1457
1548
0.056
0.30
0.77
3.35
—
1b
5b
10b
1
12.5
1.04
0.22
10.3
2.06
0.5
63.5
27.4
3368
1321
601
437
217
0.059
0.12
0.071
0.14
DXPS2
YgbP
2
t-Bu-HIC
3.7
4.5
4.7
1
5
14
60
Resource Q
Gel Filtration
Crude extract
Q-sepharose
Phenyl-sepharose
Superdex HiLoad
0.3
0.14
0.018
0.006
0.002
0.0007
aSpecific activity of purified DXPS1 for different batches of protein ranged from 124 to 1500 nmol DXP/min/mg, with the average ca. 500. The most
active batches of protein were used for the kinetic studies (Table 2).
bBased on the spec. activity of the DE52-purified DXPS1.
Table 2. Steady state kinetic parameters for recombinant S. coelicolor DXP synthases (DXPS1 and DXPS2), DXP reductoisomerase (DXR) and
MEP cytidylyl transferase (YgbP) and for the corresponding recombinant E. coli DXP enzymes
S. coelicolor
E. coli
Enzyme
DXPS1
Substrate
Km (mM)
kcat (sꢂ1
)
Km (mM)
kcat (sꢂ1
)
Pyruvate
G-3-P
Pyruvate
G-3-P
DXP
NADPH
MEP
CTP
0.30ꢀ0.01
0.60ꢀ0.07
0.57ꢀ0.03
0.20ꢀ0.03
0.19ꢀ0.01
0.19ꢀ0.02
3.26ꢀ0.09
1.45ꢀ0.03
1.1ꢀ0.1
1.1ꢀ0.1
2.40ꢀ0.34a
0.90ꢀ0.27a
0.8ꢀ0.2a
DXPS2
DXR
19.20ꢀ0.02
0.72ꢀ0.06
0.032ꢀ0.003b
21.7ꢀ0.02
16.8ꢀ0.2b
YgbP
1.4ꢀ0.1
6.0ꢀ0.05
aE. coli DXS.20
bRohdich et al.8a have reported Km values for recombinant E. coli YgbP of 3.14 mM (MEP) and 131 mM (CTP), and a kcat 9.9 sꢂ1, determined using a
spectrophotometric assay to monitor the production of inorganic pyrophosphate through enzyme-coupling to the reduction of NADP+ at 37 ꢃC.

D. E. Cane et al. / Bioorg. Med. Chem. 9 (2001) 1467–1477
1471
upon NADPH as a co-factor, was selectively bound to
the matrix and eluted only upon successive application
of high salt buffer. After optimization of expression
conditions, MEP cytidylyltransferase protein was pur-
ified in four steps (Table 1). The identity of the purified
MEP cytidylyltransferase was verified by N-terminal
Activity of mevalonate-independent pathway enzymes in
S. coelicolor CH999
The activities of native DXP synthase, DXP reducto-
isomerase and MEP cytidylyltransferase were deter-
mined for crude, cell-free extracts prepared from
submerged, liquid cultures of S. coelicolor CH999. The
magnitude of the Km values for each enzyme obtained
from radiochemical assays on the crude extracts were
comparable to those of the purified, recombinant
enzymes, although the value obtained for the Km of
CTP with crude YgbP was somewhat lower than that of
the purified protein (Table 3). Since there are pre-
sumably two functional DXP synthases present in S.
coelicolor, it is not possible to assign the observed
activity in the crude extracts to a specific DXP synthase.
Comparison of the Vmax values for crude S. coelicolor
DXPS, DXR, and YgbP with those of the correspond-
ing recombinant enzymes indicated that each of the
proteins dedicated to the mevalonate-independent
sequencing,
which
gave
results
(SBESRP-
SPAETPATTFXE) consistent with the deduced amino
acid sequence after removal of the N-terminal methio-
nine. ES–MS analysis of MEP cytidylyltransferase gave
a molecular mass of MD 27,709ꢀ13 (calcd: 27,712 Da)
again consistent with the absence of the N-terminal
Met.
Kinetic parameters for DXP reductoisomerase and MEP
cytidylyltransferase
The Streptomyces DXP reductoisomerase (DXR)
showed very similar turnover numbers to the E. coli
enzyme but a slightly lower Km value for DXP. (Table 2)
Comparison of the Km for MEP of the S. coelicolor
YgbP with that of the corresponding cytidylyltransfer-
ase from E. coli, however, showed that the E. coli
enzyme has a Km for MEP 100-fold lower than that
determined for S. coelicolor cytidylyltransferase (Table 2).
The Km for MEP (3 mM) reported earlier8a for the E.
coli YgbP is lower still by yet another factor of 10,
although the differences between the two sets of values
for the E. coli enzyme may reflect differences in the
enzyme preparation and assay methods used by the two
labs. The reported Km for CTP of the E. coli cytidylyl-
transferase was also smaller, by a tenth, than the value
of Km determined by us for the S. coelicolor enzyme.
The Streptomyces reductoisomerase showed a narrow
pH optimum of 8.0, while that of MEP cytidylyl-
transferase was broader, with a maximum at pH 7.9.
The optimal temperature range of the S. coelicolor
reductoisomerase was 25–30 ꢃC while the optimal assay
temperature for the cytidylyltransferase was 30 ꢃC.
pathway in S. coelicolor are each present at ca. 10ꢂ4
%
of total protein in liquid cultures of this organism.
(Table 4).
Discussion
Although isoprenoid metabolites have often been pre-
sumed to be relatively rare in Streptomyces, in fact sev-
eral examples of such compounds have been reported,
including the antibiotic pentalenolactone and related
metabolites,21 the sesquiterpene alcohol epicubenol,22
and several prenylated aromatic metabolites (e.g.,
naphterpin, carquinostatins, and novobiocin).14
Although the genus does not produce steroids, these
Gram-positive eubacteria do produce both menaqui-
nones14,23 and carotenoids.24 Very recently, Poralla has
reported the isolation of pentacyclic hopanoids from the
aerial mycelium of sporulating cultures of S. coelicolor
A3(2).25 These triterpene metabolites could not be
detected, however, in submerged, liquid culture.
Table 3. Km values for DXP synthase (DXPS), DXP reductoisome-
rase (DXR) and MEP cytidylyltransferase (YgbP) in cell-free extracts
of S. coelicolor CH999
Presence of two DXP synthases
Seto has shown that individual Streptomyces species
may utilize either the mevalonate or mevalonate-inde-
pendent pathway for the biosynthesis of isoprenoids,
with both pathways being found in some organisms.14
Seto has also recently reported the sequence of a
deoxyxylulose phosphate synthase17 from Streptomyces
sp. CL190and characterized the recombinant DXP
synthase.26 Interestingly, S. coelicolor DXPS1 has a
Enzyme
DXPS
Substrate
Km (mM)
Pyruvate
G-3-P
DXP
0.34ꢀ0.03
0.55ꢀ0.03
0.06ꢀ0.006
7.80ꢀ0.8
DXR
YgbP
MEP
CTP
0.098ꢀ0.005
Table 4. Comparison of the activities of S. coelicolor CH999 crude cell-free extract with S. coelicolor DXP synthase (DXPS), DXP reductoisome-
rase (DXR) and MEP cytidylyltransferase (YgbP) cloned in E. coli BL21 strains
CH999 CFE Activity
Substrate
Vmax (crude)a
(nmol/min/mg)
Vmax (pure)b
(nmol/min/mg)
Vmax (pure)/Vmax (crude)
DXPS
DXR
YgbP
Pyruvate
DXP
MEP
4ꢁ10ꢂ6
5ꢁ10ꢂ4
2ꢁ10ꢂ3
536
25,800
12,7507
3ꢁ107
5ꢁ107
ꢁ106
aApparent Vmax in the crude, cell-free S. coelicolor extract.
bVmax of the purified, recombinant protein.

1472
D. E. Cane et al. / Bioorg. Med. Chem. 9 (2001) 1467–1477
slightly higher level of sequence identity to the synthase
from Streptomyces sp. CL190than it does to S. coelico-
lor DXPS2. (Fig. 1).
selected recombinant proteins. The fact that all three
activities, DXPS, DXR, and YgbP, can also be detected
in cell-free extracts of S. coelicolor confirms that these
enzymes are indeed active in the native organism.
Although comparison of the observed Vmax values for
the crude S. coelicolor extracts with those of the corre-
sponding purified recombinant proteins indicates that
each enzyme is normally present as only a small fraction
of total protein in liquid cultures of S. coelicolor, the
observed Km values are comparable for native and
recombinant proteins, except for crude YgbP which
showed a somewhat lower apparent Km (Tables 3 and 4).
More significantly, the levels of activity of each enzyme
are sufficiently high to encourage the use of S. coelicolor
as an experimental vehicle to search for the enzymes of
the pathway catalyzing steps after that mediated by
YgbB and for detection of the corresponding inter-
mediates that lie between the cyclodiphosphate 9 and
IPP, consistent with the demonstrated biosynthesis of
isoprenoid metabolites in this and other Streptomyces
species.32
The finding that S. coelicolor harbors not one, but two,
dxps genes, is intriguing. Both encode fully functional
proteins and exhibit comparable kinetic parameters for
each of the two substrates, pyruvate and glycer-
aldehyde-3-phosphate. The regulation of the individual
genes is not yet known, nor is it known whether either
(or both) is essential for growth on minimal media.
Interestingly, both dxps genes are widely separated on
the physical map of the S. coelicolor genome, and nei-
ther is found close to the apparent structural genes
encoding the enzymes for the next two biochemical
steps in the IPP pathway, the dxr gene, corresponding
to the reductoisomerase that is located in Streptomyces
cosmid SC5H4, and the ygbP gene for the MEP cytidy-
lyltransferase in Streptomyces cosmid SCD16A. Thus,
to date, of five S. coelicolor genes implicated in the
mevalonate-independent IPP pathway (dxps1, dxps2,
dxr, ygbP, ychB, and ygbB), only ygbP and ygbB, which
are closely linked in all organisms examined to date, are
found together and none of the others are even remotely
close to one another in the S. coelicolor genome. On the
other hand, dxps1 (SC6A5.17) is located in cosmid
SC6A5 within 2855 bp of a 6741-bp cluster of five pre-
sumptive isoprenoid synthase genes that encode two
putative phytoene/squalene synthases (SC6A5.08 and
Experimental
Materials
Streptomyces coelicolor A3(2) cosmids St6A5
(SC6A5),17 St7B7 (SC7B7),17 St5H4 (SC5H4),17 and
StD16A (SCD16A)17 were gifts from Professor David
A. Hopwood and Dr. Helen Kieser of the John Innes
Institute, Norwich, UK. E. coli JM109/pJKR2 harbor-
ing the E. coli dxs gene cloned into pUC18 was prepared
by Dr. J. K. Robinson in our lab. The E. coli dxr and
YgbP genes were recloned by PCR, using the published
sequences,17 expressed in E. coli as pET-11 constructs,
and purified as previously described.7,8 E. coli XL1-
Blue, E. coli BL21(DE3)/pLysS and E. coli BL21(DE3)-
RP competent cells were purchased from Stratagene (La
Jolla, CA). S. coelicolor CH999, a derivative of S. coe-
licolor A3(2) in which the entire act cluster encoding the
genes for the biosynthesis of the highly blue-pigmented
antibiotic, actinorhodin, has been deleted, and in which
the biosynthesis of a second red-pigmented antibiotic,
undecylprodigiosin, has been blocked by a defined
mutation,28 was a gift from Professor Chaitan Khosla
of the Department of Chemical Engineering at Stanford
University. Culture media were obtained from Difco
(Sparks, MD). Silica TLC plates (250 mm) and pre-
swollen DE52 anion exchange resin were from What-
man (Clifton, NJ). SIL-NHR TLC plates were
purchased from Macherey-Nagel (Easton, PA). Cellu-
lose F254+366TLC plates (100 mm) were from Merck
(White House Station, NJ). The prepacked t-butyl
hydrophobic interaction column (t-but-HIC, 5 mL) was
obtained from BioRad (Hercules, CA). Sephadex G-25,
Resource Q anion exchange, Q-Sepharose, Phenyl-
Sepharose, Hi Load 16/60Superdex 20prep grade,
HR Superose gel filtration columns, and PD-10G25 gel
filtration columns were purchased from Pharmacia Bio-
tech Corp (Piscataway, NJ). Reactive Red 120matrix
and protease inhibitors were purchased from Sigma (St.
Louis, MA). Bradford protein assay reagent was from
SC6A5.09),
a
putative phytoene dehydrogenase
(SC6A5.11), a putative polyprenyl synthase (SC6A5.12),
and a putative squalene-hopene cyclase (SC6A5.13).17
Notably, a TBLASTN search of the S. coelicolor gen-
ome database using the Saccharomyces cerevisiae
HMG-CoA reductase17 as a query did not turn up any
statistically significant matches, suggesting strongly that
this key enzyme of the mevalonate pathway is absent in
S. coelicolor. In contrast, Seto has purified an HMG-
CoA reductase and isolated the corresponding struc-
tural gene from Streptomyces sp. CL190, an organism
which he has shown utilizes both the mevalonate and
mevalonate-independent pathways to biosynthesize
IPP.27
Comparison of activities between S. coelicolor crude,
cell-free extracts and the purified, recombinant proteins
The activities of the downstream enzymes of the meva-
lonate-independent IPP pathway have not been pre-
viously characterized in any Streptomyces species.
Although the observation of deoxyxylulose phosphate
synthase activity in S. coelicolor is significant, the first
committed step of the mevalonate-independent pathway
is catalyzed by the DXP reductoisomerase. The cloning
and expression of the genes corresponding to this step
and the subsequent YgbP-catalyzed reaction provides
strong evidence that S. coelicolor carries the necessary
complement of functional genes for the novel pathway.
To date, there have been few, if any, reports of the
detection of enzyme activities associated with the
mevalonate-independent pathway in cell-free prepara-
tions from the parent organisms, since most studies have
focused on molecular genetic analysis and expression of

D. E. Cane et al. / Bioorg. Med. Chem. 9 (2001) 1467–1477
1473
BioRad. PCR and sequencing primers were synthesized
by International DNA Technology (Coralville, IA). Taq
and Pfu DNA polymerases, all restriction enzymes, and
DNA ligase were purchased from either Stratagene or
Promega (Madison, WI). [2-14C]Pyruvate (15.0, 15.8, or
17.5 mCi/mmol) was purchased from New England
Nuclear (Boston, MA). All other chemicals were the
highest grade available.
ygbP: tatataggatccTCAGAACCCATCGTTCAGCCT-
C-30) including a BamH I site (italics) 30-of the stop
codon, to facilitate in-frame cloning into the pET11a
expression vector. The Streptomyces GTG start codons
found in each gene were replaced with an ATG codon in
each of the forward primers. PCR conditions: dena-
turation, 5 min (ygbP) or 10min ( dxps1, dxps2 and dxr)
at 95 ꢃC; 30( dxr) or 40 (dxps1, dxps2 and ygbP) cycles
of amplification (dxps1 and dxps2: denaturation, 95 ꢃC,
ꢃ
1 min; annealing, 65 ꢃC, 30s; extension, 72 C, 4 min;
Methods
dxr: denaturation, 95 ꢃC, 30s; annealing and extension,
75 ꢃC, 4 min; ygbP: denaturation, ꢃ95 ꢃC, 1 min 30s;
annealing, 65 ꢃC, 30s; extension, 72 C, 3 min) and pol-
ishing at 72 ꢃC for 2 min (ygbP), 5 min (dxr) or 10min
(dxps1 and dxps2); 10% DMSO in 1 X Taq polymerase
(dxps1 and dxps2) buffer or 10% DMF and 10% gelatin
in 1 X Pfu buffer (dxr, ygbP). The amplified PCR pro-
ducts were digested with Nde I and BamH I, ligated into
the corresponding sites in pET11a, and used to trans-
form E. coli XL1-Blue. The resulting plasmids were
purified and digested with Nde I and BamH I to verify
the presence of the desired insert and the sequences of
the cloned dxps1, dxps2, dxr and ygbP genes were con-
firmed. E. coli BL21(DE3)/pLysS was selected as the
expression host for transformation with plasmids
pET11-pDXPS1, pET11-pDXPS2, and pET11-pDXR
while E. coli BL21(DE3)-RP was used as the expression
host for transformation with plasmid pET11-pYgbP.
Expression of protein in Luria-Bertani medium supple-
mented with 3 mM ampicillin was induced by addition of
IPTG (DXPS1: 0.1 mM; DXPS2 and DXP reducto-
isomerase: 0.5 mM; MEP cytidylyltransferase: 0.4 mM)
at mid-log phase (OD 600 0.7) at 22 ꢃC (DXP syn-
thases), 30 ꢃC (DXP reductoisomerase) or 28 ꢃC (MEP
cytidylyltransferase). The cultures were incubated for an
additional 4 h (DXP synthases and MEP cytidylyl-
transferase) or 10h (DXP reductoisomerase) at their
respective induction temperatures.
Standard recombinant DNA manipulations were carried
out according to published procedures and as pre-
viously described.29,30 All protein purification steps were
carried out at 4 ꢃC. Protein concentrations were deter-
mined using the Biorad reagent according to the manu-
facturers directions, using bovine serum albumin as a
standard. PCR purification kits were purchased from
Qiagen (Valencia, CA). PCR was carried out with a
MiniCyler thermocycler from MJ Research (Waltham,
MA), equipped with a HotBonnet. Phosphoimaging
was performed using Molecular Imaging Screen-BI on a
BioRad GS-363 Molecular Imager System and the data
processed using Molecular Analyst software. Liquid
scintillation was performed on a Beckman Model LS
5801 liquid scintillation counter using Packard Omni-
Fluor scintillation cocktail. Spectrophotometric assays
were performed using a Hewlett-Packard 8452A Diode
Array spectrophotometer. dsDNA sequencing was per-
formed by the HHMI Biopolymer/Keck Foundation
Biotechnology Resource Laboratory, at the Yale Uni-
versity School of Medicine, New Haven, CT using the
dideoxy dye terminator method and automated fluor-
escent sequencing. N-terminal protein sequencing, ES–
MS, and MALDI–MS protein analysis were carried out
by the Keck Laboratory. Analysis of DNA and protein
sequences utilized the Compare, Bestfit, Pileup and
other programs in the Wisconsin sequence analysis
Package, version 10.0 (Unix), Genetics Computer
Group (GCG), Madison, WI. S. coelicolor cells were
disrupted using a French Pressure Cell Press from SLM-
Aminco Thermo Spectronic Instruments (Rochester,
NY). The Streptomyces coelicolor A3(2) Genomic
Database is available on the website of the Sanger Cen-
tre, Wellcome Trust Genome Campus, Hinxton, Cam-
bridge CB101SA: http://www.sanger.ac.uk/Projects/
S_coelicolor/.
Purification of DXP synthases
A 500 mL culture of IPTG-induced E. coli BL21(DE3)/
pLysS/pET11-pDXPS1 or /pET11-pDXPS2 was har-
vested by centrifugation, washed in buffer A (50mM
Tris–HCl, pH 7.9, 2 mM thiamin diphosphate, 5 mM
MgCl2, 1 mM dithiothreitol) supplemented with 5 mM
2-mercaptoethanol, resuspended in buffer A, and lysed
by sonication. The clear lysate was obtained by cen-
trifugation and passed through a 2.5ꢁ20cm column of
DE52 cellulose equilibrated with buffer B (40mM Tris–
HCl, pH. 7.5, 2 mM thiamin diphosphate, 5 mM MgCl2,
1 mM dithiothreitol). After washing with 200 mL of
buffer B, enzyme activity was eluted from the column by
a 1-L linear gradient of 0–500 mM NaCl in buffer B at a
flow rate of 2 mL/min. Enzyme activity pooled from
DE52 was mixed with an equal volume of 2 M ammo-
nium sulfate in buffer B to give a final ammonium sul-
fate concentration of 1 M and the mixture was applied
to a t-butyl hydrophobic interaction chromatography
column (t-but HIC, 5 mL) that had been equilibrated
with 1 M ammonium sulfate in buffer B. DXPS1 or
DXPS2 was eluted by applying 100 mL of a linear gra-
dient of 1.0–0 M ammonium sulfate in buffer B.
Cloning and expression of DXP synthases, DXP
reductoisomerase and MEP cytidylyl transferase
Primers for the amplification of each gene by PCR were
designed based on their respective entire nucleotide
sequences with the forward (50-end) primers (dxps1: 50-
tgctagccatATGACGATTCTGGAGAACATCCGGGG-
ACC-30; dxps2: 50-atatatcatATGCCGCTGCTGACCC-
GCAT-30; dxr: 50-gatccatATGAGCCACAGTCCAGC-
CCCACTCG-30; ygbP: 50-gcaaatcatATGTCTGACGA-
ATCGCGT-30) incorporating a Nde I restriction site
(italics) and the reverse (30-end) primers (dxps1: 50-cac-
tgctggatccTCAGTTCTCTCCTTGCGCTGGC-30; dxps2:
50-tatataggatccTCAGTCGCGCGCGGCCTCCA-30; dxr:
50-gatcggatccTCATGCACGGGCCTCCGCCGTCT-30;

1474
D. E. Cane et al. / Bioorg. Med. Chem. 9 (2001) 1467–1477
Ammonium sulfate in the pooled active fractions was
removed by G-25 gel filtration and the desalted filtrate
was applied to a 1-mL buffer-equilibrated Resource Q
anion exchange column. The desired enzyme activity
was eluted by applying 20mL of a linear gradient of
0–0.5 M NaCl in buffer B. DXPS2 was further purified
by gel filtration on a HiLoad 16/60Superdex 20col-
umn equilibrated with 0.5 M NaCl in buffer B, at a flow
rate of 0.2 mL/min. The progress of each purification
was monitored by SDS–PAGE. DXP synthase activity
was assayed after each purification step by a modification
of the previously described procedure.4a
ultrafiltration unit (Millipore YM-10). The purity of
protein was judged at each step by SDS-PAGE and
Biorad protein assay. After each purification step, MEP
cytidylyltransferase activity was assayed following a
previously described protocol.8a
Determination of native molecular weight
The native molecular weight of DXPS1 and DXPS2
were determined by gel filtration chromatography on an
HR Superose 12 gel filtration column equilibrated with
0.5 mL M NaCl in assay buffer A after calibration with
standard molecular weight size markers chymo-
trypsinogen (25 kDa), albumin (67 kDa), aldolase
(158 kDa), catalase (237 kDa).
Purification of DXP reductoisomerase
Reactive Red 120was packed into a glass column
(2.5ꢁ5 cm) and washed with deionized water until no
trace of red dye was found in the flow-through. A
500 mL culture of E. coli BL21(DE3)/pLysS/pET11-
DXR was harvested by centrifugation, washed and
resuspended in 50mL of buffer C (20mM Tris–HCl, pH
8.0, 5 mM MgCl2) containing 0.1 mM phenylmethane-
sulfonyl fluoride and 0.1 mM pepstatin A. Cell lysis was
performed by sonication, the clear lysate was obtained
by centrifugation and applied onto the column that had
been equilibrated with 200 mL of buffer C. After wash-
ing with 200 mL of buffer C, DXP reductoisomerase
activity was eluted with 2 M NaCl in buffer C, pooled
and concentrated using an ultrafiltration unit (Millipore
YM-30). Purity of the eluted protein was judged by
SDS–PAGE. Reductoisomerase activity was assayed
qualitatively after purification as previously described.7b
N-Terminal sequencing and molecular mass determination
Purified samples of DXPS1 were electroblotted to a PVDF
membrane and submitted for N-terminal sequencing.
The purified samples of DXPS2 and MEP cytidylyl-
transferase were precipitated with trichloroacetic acid,
the pellets were redissolved in 50% acetonitrile in water
and submitted for N-terminal sequencing and ES–MS.
Determination of steady-state kinetic parameters for
DXP synthases
Assays of DXPS1 and DXPS2 were carried out in
DXPS assay buffer (40mM Tris–HCl, 2 mM TPP,
5 mM MgCl2, 1 mM DTT) at 25 ꢃC. With pyruvate as
the variable substrate, the concentration of [2-14C]pyr-
uvate (0.64 mCi/mmol) was varied between 0.24 and
9.23 mM with a fixed concentration of glyceraldehyde-3-
phosphate of 8.4 mM. With glyceraldehyde-3-phosphate
as the variable substrate, the concentration of glycer-
aldehyde-3-phosphate was varied between 0.19 and
9.18 mM and the concentration of pyruvate (0.32 mCi/
mmol) was fixed at 2.53 mM. The amounts of protein
and the length of the incubation were adjusted for each
enzyme so as to give no more than 10% conversion.
Aliquots of 1 mL were analyzed by TLC4a followed by
phosphoimaging for 24 h. Production of [2-14C]DXP
was quantified using a [2-14C]pyruvate standard. The
measured intensities were also corrected by subtraction
of background at the same TLC Rf, generated by a
blank reaction in which assay buffer was used instead of
DXP synthase. The data for the formation of DXP as a
function of the concentration of each the variable sub-
strate were fit directly to the Michaelis-Menten equation
by non-linear least-squares regression using Kaleidagraph
software (Synergy Software).
Purification of MEP cytidylyltransferase. Q-Sepharose
Fast Flow resin was packed in a column (20ꢁ3 cm) and
equilibrated with buffer A. A 500 mL culture of IPTG-
induced E. coli BL21(DE3)-RP/pET11-pYgbP was har-
vested by centrifugation, washed and resusupended in
buffer A supplemented with 0.02% sodium azide, 2 mg
lysozyme and 0.1 mg DNAse. Cells were disrupted by
sonication, the clear lysate was obtained by centrifuga-
tion and loaded onto the column, which was washed
with a further 200 mL of buffer A. Enzyme activity was
eluted from the column by applying 600 mL of a linear
gradient of 0–1 M NaCl in buffer A, at a flow rate of
3 mL/min and pooled. In the next purification step,
pooled fractions of desired enzyme activity were mixed
with an equal volume of 2 M ammonium sulfate in buf-
fer A to give a final ammonium sulfate concentration of
1 M. This mixture was applied to a pre-packed, HiLoad
phenyl-sepharose column (20mL) that had been equili-
brated with 100 mL of 1 M ammonium sulfate in buffer
A. The column was washed with 100 mL of 1 M ammo-
nium sulfate in buffer A before the MEP cytidylyl-
transferase protein was eluted by applying 200 mL of a
linear gradient of 1–0M ammonium sulfate in buffer A,
at a flow rate of 2 mL/min. MEP cytidylyltransferase
was further purified by gel filtration on a HiLoad 16/60
Superdex 200 column equilibrated with buffer A.
Phenyl-sepharose purified enzyme was loaded onto the
column and eluted after 115 mL of buffer A, at a flow
rate of 0.3 mL/min. The desired enzyme activity was
pooled and concentrated from 4 to 0.5 mL using an
Determination of steady-state kinetic parameters for
DXP reductoisomerase
Assays of the reductoisomerase were carried out in
DXR assay buffer A (20mM Tris–HCl, 5 mM MgCl 2)
at 25 ꢃC. When DXP was the variable substrate its con-
centration was varied between 0.1 and 10 mM with a
fixed concentration of NADPH of 0.2 mM. Determina-
tion of the steady-state parameters for NADPH used a
starting NADPH concentration of 0.2 mM with DXP

D. E. Cane et al. / Bioorg. Med. Chem. 9 (2001) 1467–1477
1475
concentration fixed at 5 mM. In a typical assay,
Preparation of S. coelicolor CH999 cell free extract
NADPH and DXP were added to 1 mL of DXR assay
buffer A and the reaction initiated by addition of 10 mL
of DXP reductoisomerase solution and the oxidation of
NADPH was continuously monitored spectrophoto-
metrically at 340nm.
S. coelicolor cells were thawed and protease inhibitors,
leupeptin (4 mM) and pepstatin A (3 mM), were added.
Cells were disrupted by passage three times through a
French pressure cell at 1000 psi. The clear lysate was
obtained by centrifugation (19,500 g, 35 min) and de-
salted by G-25 gel filtration using pre-packed PD-10
columns.
Determination of steady-state kinetic parameters for
MEP cytidylyltransferase
Assays of MEP cytidylyltransferase were carried out at
30 ꢃC, using [2-14C]MEP (0.31 mCi/mmol). The MEP
concentration was varied between 0.5 and 23 mM with
the concentration of CTP fixed at 9 mM and the con-
centration of CTP was varied between 0.5 and 11.9 mM
with [2-14C]MEP concentration fixed at 10mM. In a
typical assay, CTP and MEP cytidylyltransferase were
added to YgbP assay buffer (100 mM Tris–HCl, pH 8.0,
5 mM MgCl2, 1 mM DTT) and equilibrated to 30 ꢃC for
1 min. Reactions were initiated by addition of
[2-14C]MEP. Aliquots of 1 ml were quenched at appro-
priate time intervals to allow ca. 10% conversion and
the reaction was analyzed by thin layer chromato-
graphy on SIL-NHR plates8a and quantitated by
phosphoimaging.
Assay of mevalonate-independent pathway activities in
S. coelicolor CH999
To determine the activity in S. coelicolor of mevalonate-
independent pathway enzymes, a 4 g wet cell pellet was
resuspended in 10 mL of 100 mM Tris–HCl (pH 8.0)
and lysed by passage three times through a French
pressure cell at 1000 psi. The clear lysate was obtained
upon centrifugation (19,000 g, 35 min) and desalted by
gel filtration with PD-10columns, in a total volume of
14 mL of 100 mM Tris–HCl (pH 8.0). The total protein
concentration of the desalted lysate was 3 mg/mL. The
S. coelicolor CH999 crude, cell-free extract in 100 mM
Tris–HCl, pH 8.0(14 mL) was incubated at 30 ꢃC with
7 mM [2-14C]DXP (15.7 mCi/mmol), 1 mM MnCl2 and
3 mM NADPH for 18 h, with reciprocal shaking
(200 rpm). Aliquots of 14 mL were analyzed at 2 h inter-
vals by thin-layer chromatography on cellulose, eluting
with THF/H2O (9:1)/0.1% v/v TFA, as an alternative
TLC system to the 1-propanol/ethyl acetate/water mix-
ture. The eluted TLC plate was analyzed by phospho-
imaging to determine the appearance of a new spot
which was identified as MEP by co-elution with a
[2-14C]MEP standard.
Enzymatic preparation of [2-14C]MEP
To a 50mM Tris–HCl solution (pH 8.0) containing
20mM [2- 14C]DXP (15.7 mCi/mmol), prepared accord-
ing to literature procedure,19 15 mM NADPH and
1 mM MnCl2, purified DXP reductoisomerase solution
was added. The mixture was incubated at 30 ꢃC for 8 h.
After removal of enzymes using an ultrafiltration unit
(Millipore, YM-10), the reaction mixture was analyzed
by TLC on SIL-NHR plates.8a The eluted TLC plate
was analyzed by phosphoimaging to quantify the con-
centration of [2-14C]MEP produced, using a calibration
curve generated with a [2-14C]pyruvate standard. In a
typical preparation, conversion of [2-14C]DXP to
[2-14C]MEP was 85%.
For determination of Km for DXP synthase, DXP
reductoisomerase and MEP cytidylyl transferase in
crude, cell-free extracts of S. coelicolor CH999, assays
were performed at 30 ꢃC with reciprocal shaking at
200 rpm. Typical total protein content in the cell-free
extracts, after de-salting, was 5 mg/mL in 100 mM Tris-
Cl. Aliquots of 10 mL were quenched with 20mM
EDTA (pH 8.0) and analyzed by thin layer chromato-
graphy on either cellulose (DXP synthase and DXP
reductoisomerase assays), eluting with THF/H2O (9:1)/
0.1% v/v TFA, or SIL-NHR (MEP cytidylyltransfer-
ase) plates eluting with 1-propanol/ethyl acetate/H2O
(6:1:3) (17). The formation of each product was quanti-
tated by phosphoimaging of the TLC plates, using a
[2-14C]pyruvate standard. In the assays of DXPS, the
concentration of pyruvate (0.22 mCi/mmol) was varied
between 0.05 and 1.66 mM while glyceraldehyde-3-
phosphate concentration was fixed at 2.5 mM, and the
concentration of glyceraldehyde-3-phosphate was varied
between 0.15 and 4.60 mM while pyruvate (0.43 mCi/
mmol) concentration was fixed at 2.40mM. The sub-
strates were mixed with 2 mM thiamin diphosphate,
1 mM MgCl2, 2 mM dithiothreitol, and 1 mM NaCl. To
initiate the reaction, this mixture was added to 8 mL of
CH999 desalted, cell-free extract in 100 mM Tris–HCl
(pH 7.5) at 30 ꢃC. For assays of S. coelicolor DXR,
[2-14C]DXP (0.42 mCi/mmol) concentration was varied
between 17 and 267 mM. [2-14C]DXP was mixed with
Growth of Streptomyces coelicolor CH999
One well-sporulated R2YE-agar31a culture of S. coeli-
color CH999 was used to inoculate 100 mL of sterilized,
supplemented liquid minimal media (SMM)31b that had
been modified slightly from the literature preparation by
the omission of glycine and antifoam. The inoculated
media was placed in an orbital shaker and incubated at
28 ꢃC with reciprocal shaking (325 rpm) for 72 h. Using
this saturated 100 mL culture, 2ꢁ400 mL of SMM were
in turn inoculated and incubated under the same condi-
tions. S. coelicolor cells were harvested by centrifugation
(6100 g, 1 h), washed twice with minimal volume of dis-
ruption buffer (100 mM Tris–HCl, pH 8.0, 20 mM
NaCl, 1 mM EDTA, 30% v/v glycerol) and resuspended
in disruption buffer to ca. 0.5 g wet cells per mL of buf-
fer. As a control, the dxr gene was successfully amplified
by PCR of S. coelicolor genomic DNA obtained from a
50-mL volume of S. coelicolor CH999 cell suspension by
the alkaline lysis/acetate precipitation method.31c S.
coelicolor CH999 cells were stored at ꢂ80 ꢃC until used.

1476
D. E. Cane et al. / Bioorg. Med. Chem. 9 (2001) 1467–1477
1 mM NaCl, 1 mM MnCl2 and 7.40 mM NADPH. To
initiate the reaction, this mixture was added to 6 mL of
desalted, crude cell-free extract of S. coelicolor CH999
in 100 mM Tris–HCl (pH 8.0) at 30 ꢃC. To assay native
S. coelicolor YgbP, the concentration of [2-14C]MEP
(1 mCi/mmol) was varied between 3 and 29 mM with the
concentration of CTP fixed at 7 mM; similarly, the con-
centration of CTP was varied between 0.24 mM and
0.94 mM while the concentration of [2-14C]MEP
(0.013 mCi/mmol) was fixed at 30 mM. To initiate the
reaction, the substrates in 5 mM MnCl2 and 10mM
NaCl were added to 2 mL of desalted, crude cell-free
Streptomyces extract in 100 mM Tris–HCl (pH 8.0) at
30 ꢃC.
M. H. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 11758. (b)
Kuzuyama, T.; Takagi, M.; Kaneda, K.; Watanabe, H.; Dairi,
T.; Seto, H. Tetrahedron Lett. 2000, 41, 703.
9. (a) Luttgen, H.; Rohdich, F.; Herz, S.; Wungsintaweekul,
J.; Hecht, S.; Schuhr, C. A.; Fellermeier, M.; Sagner, S.; Zenk,
M. H.; Bacher, A.; Eisenreich, W. Proc. Natl. Acad. Sci.
U.S.A. 2000, 97, 1062. (b) Kuzuyama, T.; Takagi, M.;
Kaneda, K.; Watanabe, H.; Dairi, T.; Seto, H. Tetrahedron
Lett. 2000, 41, 2925.
10. (a) Herz, S.; Wungsintaweekul, J.; Schuhr, C. A.; Hecht,
S.; Luttgen, H.; Sagner, S.; Fellermeier, M.; Eisenreich, W.;
Zenk, M. H.; Bacher, A.; Rohdich, F. Proc. Natl. Acad. Sci.
U.S.A. 2000, 97, 2486. (b) Takagi, M.; Kuzuyama, T.;
Kaneda, K.; Watanabe, H.; Dairi, T.; Seto, H. Tetrahedron
Lett. 2000, 41, 3395.
11. Jomaa, H.; Wiesner, J.; Sanderbrand, S.; Altincicek, B.;
Weidemeyer, C.; Hintz, M.; Turbachova, I.; Eberl, M.; Zei-
dler, J.; Lichtenthaler, H. K.; Soldati, D.; Beck, E. Science
1999, 285, 1573. Kuzuyama, T.; Shimizu, T.; Takahashi, S.;
Seto, H. Tetrahedron Lett. 1998, 39, 7913.
Acknowledgements
This work was supported by a National Institutes of
Health Merit Award Grant, GM30301, to D.E.C. The
authors would like to thank Dr. J. Kenneth Robinson
for preparation of E. coli JM109/pJKR2 harboring the
E. coli deoxyxylulose phosphate synthase and Prof.
David A. Hopwood and Dr. Helen Kieser for providing
gifts of the Streptomyces coelicolor cosmids used in this
study. Dr. Cathy Chow also acknowledges the support
of the Wellcome Trust in the form of a travel grant.
12. Cane, D. E.; Rossi, T.; Tillman, A. M.; Pachlatko, J. P. J.
Am. Chem. Soc. 1981, 103, 1838.
13. Cane, D. E. In Comprehensive Natural Products Chem-
istry. Isoprenoids Including Carotenoids and Steroids; Cane, D.
E., Vol. Ed.; Barton, D., Nakanishi, K., Meth-Cohn, O., Eds.;
Elsevier: Oxford, 1999; Vol. 2, pp 155–200. Unpublished
results of Arigoni, D.; Eppacher, S.; Schwarz, M. cited on pp
168–169 of ref 13.
14. Seto, H., Watanabe, H.; Furihata, K. Tetrahedron Lett.
37, 1996, 7979. Orihara, N.; Kuzuyama, T.; Takahashi, S.;
Furihata, K.; Seto, H. J. Antibiot. 51, 1998, 676. Orihara, N.;
Furihata, K.; Seto, H. J. Antibiot. 50, 1997, 979. Shin-ya, K.;
Furihata, K.; Hayakawa, Y.; Seto, H. Tetrahedron Lett. 31,
1990, 6025.
References and Notes
1. Tavormina, P. A.; Gibbs, M. H.; Huff, J. W. J. Am. Chem.
Soc. 1956, 78, 4498. Spurgeon, S. L.; Porter, J. W. In Bio-
synthesis of Isoprenoid Compounds; Porter, J. W., Spurgeon, S.
L., Eds.; Wiley: New York, 1981; pp 1–46.
15. Hopwood, D. A.; Chater, K. F.; Bibb, M. J. In Genetics
and Biochemistry of Antibiotic Production; Vining, L. C., Stut-
tard, C., Eds.; Butterworth-Heinemann: Philadelphia, 1994;
pp 65–102.
2. Rohmer, M. In Comprehensive Natural Products Chemistry.
Isoprenoids Including Carotenoids and Steroids; Cane, D. E.,
Vol. Ed.; Barton, D., Nakanishi, K., Meth-Cohn, O., Eds.;
Elsevier: Oxford, 1999; Vol. 2, pp 45–67. Rohmer, M. Nat.
Prod. Rep. 1999, 16, 565. Schwarz, M.; Arigoni, D. In Com-
prehensive Natural Products Chemistry. Isoprenoids Including
Carotenoids and Steroids; Cane, D. E., Vol. Ed.; Barton, D.,
Nakanishi, K., Meth-Cohn, O., Eds.; Elsevier: Oxford, 1999;
Vol. 2, pp 367–400.
3. Slechta, L.; Johnson, L. E. J. Antibiot. 1976, 29, 685.
4. (a) Sprenger, G. A.; Schorken, U.; Wiegert, T.; Grolle, S.;
De Graaf, A. A.; Taylor, S. V.; Begley, T. P.; Bringer-Meyer,
S.; Sahm, H. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 12857.
Lois, L. M.; Campos, N.; Putra, S. R.; Danielsen, K.; Roh-
mer, M.; Boronat, A. Proc. Natl. Acad. Sci. U.S.A. 1998, 95,
2105. (b) Lange, B. M.; Wildung, M. R.; McCaskill, D.; Cro-
teau, R. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 2100.
5. (a) Cane, D. E.; Du, S.; Robinson, J. K.; Hsiung, Y.-J.;
Spenser, I. D. J. Am. Chem. Soc. 1999, 121, 7722. (b) Laber,
B.; Maurer, W.; Scharf, S.; Stepusin, K.; Schmidt, F. S. FEBS
Lett. 1999, 449, 45.
6. Himmeldirk, K.; Kennedy, I. A.; Hill, R. E.; Sayer, B. G.;
Spenser, I. D. J. Chem. Soc. Chem. Commun. 1996, 1187.
7. (a) Duvold, T.; Bravo, J. M.; Pale-Grosdemange, C.; Roh-
mer, M. Tetrahedron Lett. 1997, 38, 4769. (b) Kuzuyama, T.;
Takahashi, S.; Watanabe, H.; Seto, H. Tetrahedron Lett. 1998,
39, 4509. Takahashi, S.; Kuzuyama, T.; Watanabe, H.; Seto,
H. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 9879. (c) Lange,
B. M.; Croteau, R. Arch. Biochem. Biophys. 1999, 365, 170.
8. (a) Rohdich, F.; Wungsintaweekul, J.; Fellermeier, M.;
Sagner, S.; Herz, S.; Kis, K.; Eisenreich, W.; Bacher, A.; Zenk,
16. Redenbach, M.; Kieser, H. M.; Denapaite, D.; Eichner,
A.; Cullum, J.; Kinashi, H.; Hopwood, D. A. Mol. Microbiol.
1996, 21, 77.
17. The following nucleotide sequences can be found in the
GenBank database: E. coli DXP synthase (dxs) (GenBank=-
GenBank Accession Number AF035440); E. coli DXP reduc-
toisomerase protein (dxr) (GenBank=GenBank Accession
Number AB013300) E. coli MEP cytidylyltransferase protein
(ygbP) (GenBank=GenBank Accession Number AE000358)
S. coelicolor cosmid SC6A5 (GenBank=GenBank Accession
Number AL049485); S. coelicolor cosmid SC7B7 (Gen-
Bank=GenBank Accession Number AL009199); S. coelicolor
cosmid SC1C3 (GenBank=GenBank Accession Number
AL023702); S. coelicolor cosmid SC5H4 (GenBank=Gen-
Bank Accession Number AL355913); S. coelicolor cosmid
SCD16A (GenBank=GenBank Accession Number AL
078618); Streptomyces sp. CL190DXS (GenBank=Genbank
Accession Number AB026631) Saccharomyces cerevisiae
HMG-CoA reductase (GenBank=Genbank Accession Num-
ber AAB667527).
18. The three amino acids designated as asparagines (italic N)
in the N-terminal sequences are actually aspartates, based on
the deduced amino acid sequences of DXPS1 and DXPS2.
19. Taylor, S. V.; Vu, L. D.; Begley, T. P.; Schorken, U.;
Grolle, S.; Sprenger, G.; Bringer-Meyer, S.; Sahm, H. J. Org.
Chem. 1998, 63, 2375.
20. Boronat and coworkers have measured a Km for pyruvate
of 2.9ꢀ0.5 mM for E. coli dXP synthase in 40mM Tris, pH
7.5 at 37 ꢃC (Querol, J.; Besumbes, O.; Lois, L. M.; Boronat,
A.; Imperial, S. 4th European Symposium on Plant Isoprenoids,
Barcelona, Spain, April 21–23, 1999) See also Seto et al. ref 27.

D. E. Cane et al. / Bioorg. Med. Chem. 9 (2001) 1467–1477
1477
21. Martin, D. G.; Slomp, G.; Mizsak, S.; Duchamp, D. J.;
Chidester, C. G. Tetrahedron Lett. 1970, 4901. Seto, H.;
Sasaki, T.; Uzawa, J.; Takeuchi, S.; Yonehara, H. Tetrahedron
Lett. 1978, 4411. Cane, D. E.; Oliver, J. S.; Harrison, P. H.
M.; Abell, C.; Hubbard, B. R.; Kane, C. T.; Lattman, R. J.
Am. Chem. Soc. 1990, 112, 4513.
22. Gerber, N. N. Phytochemistry 1971, 10, 185.
23. Yamada, Y.; Hou, C. F.; Sasaki, J.; Tahara, Y.; Yosh-
ioka, H. J. Gen. Appl. Microbiol. 1982, 28, 519.
24. Arcamone, F.; Camerion, B.; Franceschi, G.; Penco, S.
Gazz. Chim. Ital. 1970, 100, 581. Kato, F.; Hino, T.; Nakaji,
A.; Tanaka, M.; Koyama, Y. Mol. Gen. Genet. 1995, 247, 387.
25. Poralla, K.; Muth, G.; Hartner, T. FEMS Microbiol. Lett.
2000, 189, 93.
28. McDaniel, R.; Ebert-Khosla, S.; Hopwood, D. A.;
Khosla, C. Science 1993, 262, 1546.
29. Sambrook, J.; Fritsch, E. F.; Maniatis, T. Molecular
Cloning, A Laboratory Manual; 2nd ed; Cold Spring Harbor
Laboratory: Cold Spring Harbor, 1989.
30. Cane, D. E.; Kang, I. Arch. Biochem. Biophys. 2000, 376,
354.
31. Kieser, T.; Bibb, M. J.; Buttner, M. J.; Chater, K. F.;
Hopwood, D. A. Practical Streptomyces Genetics; The
John Innes Foundation: Norwich, 2000: (a) p 408; (b) p 413;
(c) p 184.
32. After submission of this manuscript, Poulter reported
the existence of two open reading frames encoding distinct
deoxyxylulose phosphate synthases in Rhodobacter capsulatus
and described the expression and characterization of the
two enzymes: Hahn, F. M.; Eubanks, L. M.; Testa, C. A.;
Blagg, B. S. J.; Baker, J. A.; Poulter, C. D. J. Bacteriol. 2001,
183, 1.
26. Kuzuyama, T.; Takagi, M.; Takahashi, S.; Seto, H. J.
Bacteriol. 2000, 182, 891.
27. Takahashi, S.; Kuzuyama, T.; Seto, H. J. Bacteriol. 1999,
181, 1256.