Cloning, mechanistic and functional analysis of a fungal sterol C24-methyltransferase implicated in brassicasterol biosynthesis

Article abstract of DOI:10.1016/j.bbalip.2010.06.007

The first committed step in the formation of 24-alkylsterols in the ascomycetous fungus Paracoccidiodes brasiliensis (Pb) has been shown to involve C24-methylation of lanosterol to eburicol (24(28)-methylene-24,25-dihydro-lanosterol) on the basis of metabolite co-occurrence. A similarity-based cloning strategy was employed to obtain the cDNA clone corresponding to the sterol C24-methyltransferase (SMT) implicated in the C24-methylation reaction. The resulting catalyst, prepared as a recombinant fusion protein (His/Trx/S), was expressed in Escherichia coli BL21(C43) and shown to possess a substrate specificity for lanosterol and to generate a single exocyclic methylene product. The full-length cDNA has an open reading frame of 1131 base pairs and encodes a protein of 377 residues with a calculated molecular mass of 42,502Da. The enzymatic C24-methylation gave a K_mapp of 38μM and k_catapp of 0.14min^-1. Quite unexpectedly, "plant" cycloartenol was catalyzed in high yield to 24(28)-methylene cycloartanol consistent with conformational arguments that favor that both cycloartenol and lanosterol are bound pseudoplanar in the ternary complex. Incubation of [27-¹³C]- or [24-²H]cycloartenol with PbSMT and analysis of the enzyme-generated product by a combination of ¹H and ¹³CNMR and mass spectroscopy established the regiospecific conversion of the pro-Z methyl group of the δ²⁴⁽²⁵⁾-substrate to the pro-R isopropyl methyl group of the product and the migration of H24 to C25 on the Re-face of the original substrate double bond undergoing C24-methylation. Inhibition kinetics and products formed from the substrate analogs 25-azalanosterol (K_i 14nM) and 26,27-dehydrolanosterol (K_i 54μM and k_inact of 0.24min^-1) provide direct evidence for distinct reaction channeling capitalized by structural differences in the C24- and C26-sterol acceptors. 25-Azalanosterol was a potent inhibitor of cell growth (IC₅₀, 30nM) promoting lanosterol accumulation and 24-alkyl sterol depletion. Phylogenetic analysis of PbSMT with related SMTs of diverse origin together with the results of the present study indicate that the enzyme may have a similar complement of active-site amino acid residues compared to related yeast SMTs affording monofunctional C₁-transfer behavior, yet there are sufficient differences in its overall amino acid composition and substrate-dependent partitioning pathways to group PbSMT into a fourth and new class of SMT.

Full text of DOI:10.1016/j.bbalip.2010.06.007

Biochimica et Biophysica Acta 1801 (2010) 1163–1174
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta
journal homepage: www.elsevier.com/locate/bbalip
Cloning, mechanistic and functional analysis of a fungal sterol C24-methyltransferase
implicated in brassicasterol biosynthesis
Maristela Pereira a,1, Zhihong Song b,1, Ludier Kesser Santos-Silva a,1, Mathew H. Richards b,1
,
Thi Thuy Minh Nguyen b, JiaLin Liu b, Celia Maria de Almeida Soares a, Aline Helena da Silva Cruz a,
Kulothungan Ganapathy b, W. David Nes b,
⁎
a
Laboratorio de Biologia Molecular, Instituto de Ciencias Biologicas, Universidade Federal de Goias, 74001-970, Goiania, Goias, Brazil
b
Department of Chemistry and Biochemistry, Texas Tech University, Lubbock, TX 79409-1061, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 22 April 2010
Received in revised form 15 June 2010
Accepted 18 June 2010
Available online 17 July 2010
The first committed step in the formation of 24-alkylsterols in the ascomycetous fungus Paracoccidiodes
brasiliensis (Pb) has been shown to involve C24-methylation of lanosterol to eburicol (24(28)-methylene-
24,25-dihydro-lanosterol) on the basis of metabolite co-occurrence. A similarity-based cloning strategy was
employed to obtain the cDNA clone corresponding to the sterol C24-methyltransferase (SMT) implicated in
the C24-methylation reaction. The resulting catalyst, prepared as a recombinant fusion protein (His/Trx/S),
was expressed in Escherichia coli BL21(C43) and shown to possess a substrate specificity for lanosterol and to
generate a single exocyclic methylene product. The full-length cDNA has an open reading frame of 1131 base
pairs and encodes a protein of 377 residues with a calculated molecular mass of 42,502 Da. The enzymatic
C24-methylation gave a Kmapp of 38 μM and kcatapp of 0.14 min−1. Quite unexpectedly, “plant” cycloartenol
was catalyzed in high yield to 24(28)-methylene cycloartanol consistent with conformational arguments
that favor that both cycloartenol and lanosterol are bound pseudoplanar in the ternary complex. Incubation
of [27-13C]- or [24-2H]cycloartenol with PbSMT and analysis of the enzyme-generated product by a
combination of 1H and 13CNMR and mass spectroscopy established the regiospecific conversion of the pro-Z
methyl group of the Δ24(25)-substrate to the pro-R isopropyl methyl group of the product and the migration
of H24 to C25 on the Re-face of the original substrate double bond undergoing C24-methylation. Inhibition
kinetics and products formed from the substrate analogs 25-azalanosterol (Ki 14 nM) and 26,27-
Keywords:
Paracoccidiodes brasiliensis brassicasterol
Ergosterol
Membrane
Cloning
Antifungal drug
Sterol metabolome
Ergosterol evolution
Sterol methylation mechanism
Erg6p
Ergosterol biosynthesis inhibitor
Lanosterol catalysis
dehydrolanosterol (Ki 54 μM and kinact of 0.24 min−1
) provide direct evidence for distinct reaction
channeling capitalized by structural differences in the C24- and C26-sterol acceptors. 25-Azalanosterol was a
potent inhibitor of cell growth (IC50, 30 nM) promoting lanosterol accumulation and 24-alkyl sterol
depletion. Phylogenetic analysis of PbSMT with related SMTs of diverse origin together with the results of the
present study indicate that the enzyme may have a similar complement of active-site amino acid residues
compared to related yeast SMTs affording monofunctional C1-transfer behavior, yet there are sufficient
differences in its overall amino acid composition and substrate-dependent partitioning pathways to group
PbSMT into a fourth and new class of SMT.
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
with cycloartenol (plant) or lanosterol (fungi and animals) precursors
and end with the formation of Δ5-sterols that possess a 24-H
The sterol family of natural products is characterized by a limited
array of ring structures yet contain a broad range of side chain skeletal
themes that originate in organisms of a photosynthetic or non-
photosynthetic lineage; the phylogenetically distinct pathways start
(cholesterol), β-CH3 (ergosterol) or α-C2H5 group (sitosterol)
(Fig. 1) [1–3]. Sterol metabolism is an ancient biochemistry that is
different in humans and fungi that infect them through the rate-
determining steps that control flux to the final product(s) [4,5].
Unique to fungi (and plants) is the critical slow enzyme involved with
phytosterol diversity, the sterol C24-methyltransferase (SMT) [6,7].
The central role of this catalyst in the evolution of the different sterol
metabolomes that occur across Kingdoms has stimulated considerable
interest in the mechanism and stereochemistry of C24-methylation
reactions as well as to provide a novel target for rational drug design
[6–12].
Abbreviations:SMT, Sterol C24-methyltransferase; GC-MS, Gas-chromatography-mass
spectroscopy; Pb, Paracoccidiodes brasiliensis; Sc, Saccharomyces cerevisiae
⁎
Corresponding author. Tel.: +1 806 742 1673; fax: +1 806 742 1289.
E-mail address: wdavid.nes@ttu.edu (W.D. Nes).
Authors contributed equally to this work.
1
1388-1981/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.bbalip.2010.06.007

1164
M. Pereira et al. / Biochimica et Biophysica Acta 1801 (2010) 1163–1174
Fig. 1. Diversity of 24-alkyl sterol biosynthesis scheme. The final sterols brassicasterol and stigmasterol can be formed by different routes that occur in protozoa, fungi and plants.
Sterols play a wide range of cellular functions in fungi, as they do in
animals, including that of a structural component of membranes [13],
in pheromone signaling [14], as sex hormones [15] and in “sparking”
cells to proliferate [16]. An important distinction for sterol usage in
fungi is that these microbes must retain the 24-alkyl group in the
sterol side chain for the sterol to carry out its individual functions
other than that of a bulk membrane insert [17]. Fungal growth and
maturation reportedly are affected by various types of sterol
biosynthesis inhibitors that act on the sterol C14-demethylase
(SDM) [18] or SMT enzymes [9], consistent with the essential role
of these enzymes in sterol-controlled physiology. Infection too can be
influenced by the sterol homeostasis of dimorphic fungi which
includes Candida albicans and Paracoccidiodes brasiliensis. Notably, P.
brasiliensis, the most frequent systemic mycoses in rural Latin America
and the causative agent of paracoccidioidmycosis, is closely aligned
phylogenetically to C. albicans [19]. The two ascomycetous fungi
possess similar thermo-regulated dimorphic life cycles and sterol
pathways, except that in the case of P. brasiliensis, the 24-alkyl sterol
mixture contains significant brassicasterol (ergosta-5,22-dienol) in
addition to ergosterol (ergosta-5,7-22-trienol) [20–22]. As observed
by treating C. albicans infections with drugs targeted at ergosterol
synthesis and action, P. brasiliensis treatment with amphotericin B and
azoles led to renal side effects and emergence of resistance [19].
Therefore, there is a renewed interest and search for sterol
biosynthesis inhibitors that can be developed as chemotherapeutic
agents against these pathogens.
order that catalyst-based sterol analogs can be designed to control flux
at the rate-determining step of the pathway [1,5]. In general, SMT
control governs the C24-methylation pathway with exquisite precision
to generate single or multiple products; progressive inhibition of these
enzymes can result in depletion of cellular 24alkyl-sterols (such as
ergosterol) accompanied by growth inhibition. Several previously
characterized SMTs are shown to be membrane-bound proteins with
monomer Mr=38 to 43 kDa of tetrameric organization, have a pH
optima in the range from 6.0 to 8.8 and require AdoMet cofactor for
catalysis. The catalytic competence of cloned SMTs of plant, fungal or
protozoa origin is similarly slow, kcat=0.6 min−1, and they recognize
optimal substrates with similar binding affinities, approximate
Km =35 μM and Kd =4 μM [1]. Comparison of the SMT primary
structures demonstrated 49 to 65% sequence identity among the
amino acids in them. Consistent with the significant overall sequence
relatedness, these enzymes all share in four conserved substrate binding
regions in the active site, three of them are sterol binding segments and
one of them possess an α/β-like Rossmann-fold that serves as the
AdoMet binding motif. To date, based on their substrate specificity,
SMTs can be grouped into three families that relate to fungal-plant
taxonomy, zymosterol (EC. 2.1.1.41), cycloartenol (EC 2.1.1.41) or 24
(28)-methylene lophenol (EC. 2.1.1.143) [1].
In contrast, much to our surprise, as described herein, upon initiating
studies to establish the importance of the C24-methylation reaction
involved in brassicasterol formation in P. brasiliensis, we discovered that
the cloned P. brasiliensis SMT (PbSMT) favored lanosterol, equally well
to cycloartenol and that zymosterol, the optimal substrate for
Candida albicans and Saccharomyces cerevisiae SMT (ScSMT), was a
poor substrate. Moreover, we found that under physiological
For some time, we have been interested in unearthing the molecular
libraries (metabolite structure and amounts) that characterize the sterol
metabolomes involved in growth and developmental regulation in

M. Pereira et al. / Biochimica et Biophysica Acta 1801 (2010) 1163–1174
1165
conditions 25-azalanosterol is a potent inhibitor of P. brasiliensis growth,
2.4. Heterologous expression and purification of PbSMT
much greater than that of the related yeasts S. cerevisiae and C. albicans
[23,24]. The combination of metabolite and enzymatic results point to a
new class of SMTs that prefer lanosterol as substrate and may represent
a novel cellularpathway for the designof tractable sterol analogs to treat
paracoccidioidmycosis.
PbSMT gene prepared by amplification of the corresponding cDNA
fragment was introduced into plasmid pET32a, and the resulting
protein expressed in E. coli BL21 (C43) competent cells (prepared in
Brazil at U.F.G.) or BL21(DE3) (prepared in the United States at T.T.U.)
under control of a T7 promoter (insert size 1131 bp). For BL21 (C43)
or BL21(DE3) harboring the PbSMT gene construct, a single colony
was inoculated into 250 mL of Luria–Bertani (LB) broth (1% tryptone,
1% NaCl and 0.5% yeast extract) (pH 7.5) supplemented with
ampicillin (50 μg/mL) and 20 mM glucose and grown for 10 h at
30 °C at 200 rpm shaking.. A 20 mL aliquot of this starter culture was
added to 2.8 L Erlenmeyer flasks containing 1 L of LB medium
supplemented with 1 mL of the stock ampicillin. The inoculum was
grown in the medium for approximately 3 h to A600nm 0.5 at 30 °C,
then 400 μL of 1 M isopropyl-1-thio-β-D-galactoside (IPTG) (Research
Products International Corp.) was added to the culture flask and the
cells further incubated for another 3 h at 30 °C with moderate shaking
(150 rpm) to induce recombinant PbSMT expression. The resulting
cells were harvested and the pellet centrifuged at 4 °C, 10,000×g,
10 min. The cell paste was stored at 20 °C for short term storage or
used directly for the enzyme studies. For PbSMT purification, the
recombinant protein was purified using the Ni-NTA resin (Invitrogen)
following the manufacturer's instructions. The levels of expression,
purity and size of the recombinant protein were monitored
chromatographically on a 12% sodium dodecyl sulfate-polyacrylamide
gel electrophoresis (SDS-PAGE) followed by Coomassie Blue or silver
staining. Total protein concentrations were determined by the
Bradford method with commercial reagents (Bio-Rad) using bovine
γ-globulin as standard [26].
2. Materials and methods
2.1. Pathogen and culture procedures
P. brasiliensis cells (strain Pb0, ATCC MYA-826) were grown in 1 L
flasks containing 250 mL Sabouraud Dextrose Medium (SDM, Difco)
for 72 h. at 36 °C. In the control, growth arrest developed after a short
lag phase of 24 h; followed by a log to stationary phase growth that
populated over the range from 2×105 cells/mL to 7×106 cells/mL.
Cell numbers were estimated by the use of an hemocytometer
(Bright-line counting chamber, Fisher). The cells were harvested and
the neutral lipid fraction was isolated and processed to yield total
sterols by procedures described elsewhere [25]. In a separate set of
experiments, the concentration in μM of the test compound inhibiting
growth to 50% of control values (IC50) at growth arrest was
determined from the inhibition profile. Inhibitors were dissolved in
a minimum amount of dimethyl sulfoxide (b1% v/v) in the ranges
from 10 nM to 1.0 μM or 1 μM to 100 μM.
2.2. Enzymes, substrates and reagents
Unless otherwise indicated, enzymes and vectors were obtained
from Invitrogen. AdoMet as the iodide or chloride salts was purchased
from Sigma and the sterol substrates (b98% by GLC analysis) were
obtained from our collection [25 and references cited therein].
Cycloartenol samples isotopically labeled with 2H (b98% 2H enriched)
or 13C (N99% 13C enriched at C27 and C26 in 8 to 2 ratio) were prepared
from cycloartenol as described previously [25]. 26,27-Dehydrolanos-
terol and 25-azalanosterol were prepared fresh according to the
reported synthetic routes [26,27]. [methyl-3H3]AdoMet, diluted with
AdoMet iodide salt to a sp. act. of 10 μCi/μmol, was purchased from
Perkin-Elmer (Waltham, MA) and [methyl-2H3]AdoMet (99.3% 2H
enrichment) was purchased from MSD Isotopes, Canada. All other
reagents and chemicals were purchased from Fisher, Aldrich or Sigma,
unless noted otherwise, and used without further purification.
2.5. Site-directed mutagenesis of PbSMT
The QuickChangeTM(QC) site-directed mutagenesis kit (Strata-
gene) was used to produce site-specific mutations at Tyr88 (Tyr81 in
the ERG6 nomenclature) positions of pET32a(+)-PbSMT using
appropriate primers. Polymerase chain reaction amplification with a
specific primer was performed according to manufacturers' instruc-
tion. The plasmid was isolated using the Promega SV wizard miniprep
kit and the mutation confirmed by sequencing, using the appropriate
primer (T7 promoter). The mutation converting Tyr88 to leucine
(Tyr88Leu) was obtained with the 45 mer primer (corresponding to
codons 78–88): 5′ GCTACCGATTTCCTGGAGTATGGCTGG3′. The muta-
tion converting Tyr88 to phenylalanine (Tyr81Phe) was obtained with
the primer corresponding to codons 78–88 of 5′ GCTACCGATTTCTTC-
GAGTATGGCTGG. The codon corresponding to the introduced muta-
tion is underlined. Each of the mutant constructs were transformed
into E. coli strain B21(DE3) and the cells grown to express
recombinant PbSMT as described above.
2.3. Cloning full-length PbSMT
PbSMT gene was cloned from an infection cDNA library of P.
brasiliensis described by Costa et al. [28]. PbSMT sequence was
identified from bioinformatics resources of several databases, (http://
www.ncbi.nlm.nih.gov/BLAST/ or www.tigr.org) using a BLAST
search against several SMTs, including the ERG6 gene corresponding
to the Erg6p. Degenerate primers related to the ERG6 gene were
designed as ERG6-F (5′-ATAACCGAATTCATGGCGCC-3′) and ERG6-R
(5′-ACGTGGTCGACACTTTACTC-3′) and contained engineered EcoRI
and SalI restriction sites (underlined), respectively. An 1131 bp cDNA
fragment containing the complete coding region was amplified by PCR
using these primers. The reaction program consisted of an initial
denaturation step at 94 °C for 3 min, 30 cycles of 94 °C for 90 s, 48 °C
for 2 min and 72 °C for 90 s and a final elongation at 72 °C for 10 min.
The PCR product was subcloned into pGEM-T Easy vector (Promega
Corp. Madison WI). The correctness of the insert was confirmed by
DNA sequencing. The cDNA was excised from pGEM-T Easy vector by
digestion with EcoRI and SalI, and cloned into the EcoRI and SalI sites
pET32a (+) (Novagen, EMD Biosciences, San Diego, CA) containing
one each of the S- and Trx- and His tags for antibody detection and
protein purification, respectively, and transformed into the expres-
sion host E. coli BL21(C43) (Supplemental Fig. 1).
2.6. Phylogenetic analyses
The PbSMT deduced amino acid sequence was aligned with other
SMTs available at the Pfam database and phylogenetic relationships
were generated with 41 deduced complete sequences from fungi,
plant and protozoa SMTs. A phylogenetic tree was constructed by
multiple alignments using the CLUSTAL X program and by the
neighbor-joining method visualized using the Tree View software.
Robustness of branches was estimated using 100 bootstrapped
replicates and indicates the percentage of times that all species
appear as a monophyletic cluster.
2.7. SMT assay and inhibitors studies
Cell paste (5 g) containing the recombinant PbSMT was re-
suspended in 30 mL of buffer (50 mM Tris–HCl, 2 mM MgCl2, 2 mM

1166
M. Pereira et al. / Biochimica et Biophysica Acta 1801 (2010) 1163–1174
β-mercaptoethanol, and 20% glycerol (v/v), pH 7.5), and the
suspension lysed by passage through a French pressure cell at
20,000 psi. Insoluble protein and cell debris were removed by
centrifugation (4 °C, 100,000×g, 1 h), and a portion of the superna-
tant (approximately 25 mL) assayed. The standard assay for enzyme
activity was carried out in 10 mL test tubes in final volume of 600 μL
containing fixed amounts of sterol (100 μM) or the sterol concentra-
tion was varied from 5 to 150 μM in 12 μL Tween 80 (5% of an ethanol
solution, v/v). Sterol packaged in Tween 80 was delivered to the
bottom of the test tube and dried under a stream of nitrogen. To this
film deposit was added 500 μL of 1.2 mg of soluble protein in buffer
(50 mM Tris–HCl, 2 mM MgCl2, 2 mM β-mercaptoethanol, and 20%
glycerol (v/v), pH 7.5, 5% glycerol (v/v)). The reaction was initiated by
the addition of 100 μL AdoMet (100 μM; [methyl-3H3]AdoMet, 0.6 μCi)
and incubated for 45 min at 30 °C. The incubation mixture was
terminated by addition of a solution (600 μL) of 10% methanolic-KOH.
The methylated sterol products were extracted in hexane, dried and in
the case of the radioactive samples analyzed by scintillation counting to
determine the conversion rate.
The steady-state kinetic parameters, Km, Vmax and kcat, estab-
lished under initial velocity conditions performed on soluble
protein were calculated by fitting the liquid scintillation data to
the Michaelis–Menten equation using the Enzyme Kinetic Module
from SSPS Inc. Inhibition of SMT using the soluble protein was
assessed by graphical procedures provided by Enzyme Kinetic Module,
and rate data were analyzed by linear regression analysis affording R2
values of 0.95 to 0.97 [27]. PbSMT steady-state kinetics, run under linear
conditions, were performed at varied lanosterol concentrations from
5 μM to 100 μM, or at varied 25-azalanosterol concentrations from
15 nM to 100 nM, or 26,27-dehydrolanosterol concentrations from 25
to 100 μM and [methyl-3H3]AdoMet fixed at 100 μM. Boiled controls
evidenced negligible activity with lanosterol and [methyl-3H3]
AdoMet as substrates; in experiments with soluble PbSMT approxi-
mately 500–1000 dpm were detected in assays with [methyl-3H3]
AdoMet and no sterol; this radioactivity was background corrected in all
the experimental samples. For determination of the pH, temperature,
and optimal amount of enzyme, single time point assays containing
1.12 mg of soluble PbSMT were incubated for 45 min at 30 °C.
2.9. Time-dependent inactivation
Experiments were performed with a 100,000×g supernatant
fraction containing the recombinant PbSMT in two major steps:
firstly, a pre-incubation experiment was established; to a 9 mL test
tube containing analog at the test concentration (varied between 0
and 100 μM) was added 2.5 mg of total protein, 1.5 mL buffer and a
fixed concentration of AdoMet (100 μM). The preparation was
allowed to incubate for up to 10 min at 35 °C. Secondly, at the times
indicated, aliquots (200 μL) of the pre-incubation preparation were
placed in a pre-cooled (dry ice ethanol) test tube to prevent catalysis.
The frozen samples were then individually thawed and filtered three
times with buffer to remove excess or unbound inhibitor using
Centricon YM30 (30 kDa cut off) ultrafiltration device by spinning at
5000×g at 4 °C in a Beckman Centrifuge. The final volume was
adjusted to 500 μL with buffer. The remaining enzyme activity in each
case was determined by assaying the samples at saturating concen-
trations of lanosterol and [methyl 3H3]AdoMet (100 μM; 0.6 μCi)
under standard assay conditions for 45 min. The logarithmic percent-
age of remaining enzyme activity was plotted against incubation time
of enzyme–inhibitor mixture to determine the half-life (t1/2) of
inactivation. Kitz and Wilson plot was then generated using t1/2 at
each inhibitor concentration and inverse of inhibitor concentration
on Sigmaplot 2001 according to the equation — t1/2 =0.69/kinact
+
0.69/kinact ⁎KI/I where kinact is the rate of inactivation and KI is the
concentration of inactivator that produces half maximal rate of
inactivation. The point of intersection with the ordinate is the t1/2 at
saturation from which kinact can be calculated while the intercept
along the X-axis gives −1/KI from which KI can be calculated [27].
For PbSMT labeling, 1 mg soluble protein was treated with 100 μM
of inhibitor, 100 μM of [methyl-3H3]AdoMet (1 μCi) in 600 μL standard
buffer overnight. After dialysis and filtration (Centricon, YM10) to
concentrate the protein and remove any unbound inhibitor, the
protein sample was boiled in SDS buffer for 10 min prior to analysis by
chromatographic separation on 12% SDS-PAGE gels. The gel was
stained with Gel-CodeTM (Pierce, Rockford, IL) stain solution for
20 min. The gel was saturated with fluorescent enhancer (Fluoro-
hance, Research Product International Corp., Mt. Prospect, IL) for
30 min and then dried by a gel drier at 80 °C for 2 h. Gels were
prepared for radiofluorography (Kodak XAR-5 X-ray film, Eastman
Kodak company, Rochester, NY) for 4 weeks at −80 °C. Unmodified
controls were included in all experiments.
2.8. Metabolite identification
For sterol quantification, GC was performed with a HP 5890 gas
chromatograph fitted with a 3 ft spiral column filled with 3% SE-30
(Mesh size 100/120 on chromsorb W; Alltech, Deerfield, IL). The GC
oven was maintained at 245 °C. Substrate and biomethyl products,
referenced to the retention time of cholesterol (RRTc), were
quantified by integration of the detector signal from 4 to 20 min.
HPLC was performed using semi-preparative or analytical C18 TSK gel
columns (Tosohaas) eluted at 1 mL/min with methanol at ambient
temperature. For combined GC-MS analysis was employed, a Hewlett-
Packard 6890 GC system coupled to a 5937 mass selective detector
with on-column injection. Chromatography was performed on a HP-5
column (30 m×0.25 mm×0.25 mm) with the temperature program
ramped from 170 °C to 280 °C at 20 °C/min with Helium carrier at
1.2 mL/min; MS was operated at an ionization voltage, 70 eV (EI) and
interface temperature, 250 °C. The product distribution generated by
wild-type or mutant PbSMT was determined by incubating saturating
levels of the co-substrates with sufficiently large preparations of
protein (multiple incubations with 5–9 mg soluble protein/assay)
overnight and the enzyme-generated products analyzed by GC-MS
and 1H or 13CNMR. The NMR spectra were recorded in CDCl3. at
500 MHz with TMS as an internal standard. Retention times in GC and
HPLC are relative to the retention of cholesterol (RRTc and αc,
respectively). Product peaks were quantified by integration of peak
area based internal standardization.
3. Results
3.1. Sterol composition of control and treated cultures
Phytosterol biosynthesis consists of a series of distinct enzymatic
reactions, usually involving 10 to 12 sterol genes in a given organism,
that converge on a few common intermediates to form one or a
mixture of 24-alkyl sterol end products. Although ergosterol is
typically viewed as the major sterol synthesized in fungi, a recent
study of P. brasiliensis revealed that ergosterol is accompanied by
brassicasterol, in a precursor–product relationship [20]. From the
previous GC-MS analysis of the sterol composition of P. brasiliensis
three sterols were detected in control, lanosterol, ergosterol and
brassicasterol in a ratio of 4:27:69. We reexamined the sterol
composition of the fungus in order to undertake a more detailed
analysis of the amounts and identities of the intermediates. Analysis
of the total sterol fraction by GC-MS showed that the fungus
synthesized 12 compounds of which lanosterol, ergosterol and
brassicasterol make up approximately 80% of the sterol mixture
(Supplemental Fig. 2, panel A) The chromatographic behavior and
mass spectra of the 12 sterols detected in the P. brasiliensis match
authentic specimens available in our steroid collection [29–31] and
for compounds reported in the literature [32–34]. Lanosterol 1,

M. Pereira et al. / Biochimica et Biophysica Acta 1801 (2010) 1163–1174
1167
eburicol 4, ergosterol 10 and brassicasterol 14 (Fig. 1) were puri-
fied from the fungal sterol fraction by HPLC and analyzed by 1HNMR
(s, singlet, d, doublet,m, mutliplet): Compound 1, δ, H18 (0.69, s),
H19 (1.00), H21 (0.91), H26 (1.61, d), H27 (1.67, d), H30 (0.98, s),
H31 (0.81, s) and H32 (0.88,s); 4, H18 (0.69, s), H19 (1.00, s), H21
(0.92, d), H26 (1.03, d), H27 (1.03, d), H28 (4.65, d), H30 (0.98, s),
H31 (0.81, s), H32 (0.88, s); 10, H18 (0.63, s), H19 (0.95, s), H21
(1.04, d), H26 (0,82, d), H27 (0.84, d), H28 (0.92, d); H6 (5.57, m),
H7 (5.37, m), H22 (5.19, m), H23 (5.18, m); 14, H18 (0.69, s), H19
(1.10, s), H21 (1.01, d), H26 (0.82, d), H27 (0.83, d), H6 (5.35, m),
H22 (5.23, m), H23 (5.30, m). The chemical shifts for the side chain
structures of ergosterol, brassicasterol and eburicol agree for them to
possess the 24β-methyl group in 10 and 14 and exocyclic methylene
group at C24 in 4 [35].
Based on these findings, we surmise that the first metabolic step in
the 24-alkyl sterol synthetic pathway of P. brasiliensis involves the
C24-methylation of lanosterol to eburicol (Supplemental Fig. 3),
similar to the ergosterol pathway in the filamentous Ascomycetes
Aspergillus fumigatus [32] and Basidiomycetes Cryptococcus neofor-
mans [25]. The final product of the pathway is brassicasterol, in
agreement with the natural flux of intermediates in which the sterol
nucleus of Δ8 converts to the Δ7 to Δ5,7 to Δ5-bond system [36]. In
contradistinction, in the Ascomycetes C. albicans lanosterol is
demethylated in the nucleus at C4 and C14 before C24-transalkylated
of zymosterol 7 to form fecosterol 8 (Fig. 1) [36]. Interestingly, not
only is the location of the C24-methylation step in 24-alkyl sterol
synthesis different in the Ascomycetes, but P. brasiliensis may have
evolved another sterol enzyme, i.e., the Δ7-reductase, to complete the
pathway to form a Δ5-monoene as can occur in the Zygomycetes
fungi, e.g., Mortierella alpina that accumulate cholesta-5,24-dienol
(desmosterol) [37,38].
In P. brasiliensis, thermodynamics dictate the direction and regula-
tory capacity of the 24-alkyl sterol metabolic pathway with the SMT
catalyzed reaction apparently operating far from equilibrium as the
first committed step. It would be expected that metabolic flux
through the phytosterol pathway can be controlled at the stage of
sterol C24-methylation through inhibitors, allosterism or genetic
control [1,6]. Indeed, inhibitors designed to interfere with P. brasiliensis
growth and the sterol C24-methylation reaction, such as 22-piperidin-
3-yl-pregnan-22(S)-3β-ol, were shown to be effective at an IC50 value of
250 nM resulting in the accumulation of lanosterol in the cells [20].
Thus, on mechanistic reasoning, we incubated different sterol analogs of
the SMT catalyzed step with P. brasiliensis; 26,27-dehydrolanosterol 16
failed to inhibit fungal growth at the highest concentration of drug
tested (100 μM), likely due to a lack of significant absorption into the
cells. Alternatively, 25-azalanosterol 17 inhibited growth generating
IC50 value of 35 nM 5 nM. The sterol composition of the treated cells at
IC50 showed lanosterol to accumulate (b85% of total sterols, Supple-
mental Fig. 2, panel B), consistent with the substrate specificity of SMT
predicted from the sterol pathway of control.
common consensus sterol binding segments (Regions I, III and IV)
(Fig. 2) in the primary structure. In addition, PbSMT possessed the
same set of 8 conserved signature residues that in the SCSMT upon
mutation afford altered substrate specificity and product distribution
[39].
3.3. Expression, purification and functional confirmation of recombinant
PbSMT
Recombinant PbSMT was prepared as a triple tagged fusion
enzyme containing Trx, S-tag- and N-terminal His6 sites for protein
purification in E. coli BL21(C43). The target protein was purified to
near homogeneity by Ni+ 2-NTA agarose chromatography [40] and
shown to migrate in SDS-polyacrylamide gel at Mr 60 kDa. The
chromatographic behavior of this specimen is close to the calculated
migration of 60.37 kDa for the fusion protein of predicted mass Mr
42,502 plus three attachments (14.33 kDa) and associated amino acids
(3.53 kDa) originating at the N-terminus of the pET32a(+) vector (Fig. 3).
Removal of the His-tag from the recombinant protein reduced enzyme
activity therefore further work of the PbSMT catalytic competence of wild-
type and mutant PbSMT was performed using a soluble preparation from
E. coli BL21 (DE3).
PbSMT was over-expressed as the recombinant protein in E. coli
BL21(C43) cells. As judged by GC-MS, the soluble enzyme incubated
with saturating amounts of lanosterol and AdoMet generated a single
product, eburicol (24(28)-methyene-24,25-dihydro-lanosterol) in
55% yield (Fig. 4). In a similar incubation of soluble PbSMT originating
in BL21(DE3) cells, the catalyst generated eburicol in a 45% yield. The
Michaelis constant for lanosterol was determined in the standard
manner from double-reciprocal plots yielding an approximate Km
value of 38 μM 2 μM and Vmax of 50 pmol/min/mg (Supplemental
Fig. 5). Assay of varied concentrations of lanosterol and fixed AdoMet
(100 μM) as the iodide or chloride salt with PbSMT generated similar
kinetic constants (data not shown). The specific activity of soluble
PbSMT determined for the C24-methylation reaction was established
in two separate experiments from which the turnover number can be
calculated using the equation kcat =Vmax [SMT]. In the first experi-
ment performed using the standard assay, the rate of lanosterol
conversion to methyl product afforded a Vmax of 50 pmol/min. In the
second experiment, the amount of PbSMT in the soluble protein
preparation was estimated by determining the amount of “diol”
product released from the PbSMT-sterol complex (cf. Section 3.6) in
similar fashion to the process reported for the soybean SMT2 [41]. For
recombinant PbSMT, the total enzyme level was determined to be
approximately 7.6 μg or for the native protein to be 5.9 μg. Using the
calculated Mr of PbSMT of 42,505 Da and assuming the construct to be
a tetramer, the yield of native enzyme in the assay is estimated to be
348 pmol. The calculated kcat for the native protein is therefore
50 pmol/min/348 pmol=0.14 min−1, providing a slow turnover
number similar to that reported previously for the purified SMT
enzymes from several different sources [6,40,41]. The optimum pH for
PbSMT activity was determined to be 7.5 to 8.0, with half maximum
velocities at 6.5 and 9.0 and a temperature optimum at 30 °C. The
conversion rate for the soluble fraction was linear with increasing
time up to 240 min in a total protein concentration of 1.4 mg/mL.
These parameters are consistent with other SMTs [6].
3.2. Cloning and sequencing of the PbSMT gene
Using a similarity-based cloning approach, the full-length cDNA
sequence of the PbSMT was identified and subcloned (Supplemental
Figs. 1 and 4). The resulting fragment of 1131 base pairs furnished an
open reading frame that encodes an apparent protein of 377 amino
acids with a calculated molecular mass of 42,502 Da and theoretical pI
of 6.08 (Fig. 2). The sequence of the putative PbSMT shares significant
sequence homology with SMT1 genes from ascomycetous fungi (b66%
similarity) and to a lesser extent to other SMT1 and SMT2 sequences
from plants and protozoa (b34% similarity) deposited in the GenBank
data base. Comparison of the deduced amino acid sequence of PbSMT
with other members of the AdoMet-dependent methyltransferase
super family showed conservation of the AdoMet binding residues
observed to be in contact with AdoMet (Region II) and to contain
3.4. Substrate specificity
To determine the optimal substrate for PbSMT, a series of sterol
acceptors that differ in the structure of the nucleus and side chain
were assayed with the soluble recombinant protein. The first set of
test compounds was selected based on the known substrate
preferences for the three SMT isoforms, classed as SMT1fungi/
protozoa, SMT1plant and SMT2 plant. Enzymatic assay of lanosterol
1 generated a Km of 38 μM and Vmax of 50 pmol/min which

1168
M. Pereira et al. / Biochimica et Biophysica Acta 1801 (2010) 1163–1174
Fig. 2. Alignment of sterol C24-methyltransferase amino acid sequences (GenBank accession numbers) from Saccharomyces cerevisiae (NP_013706), Candida albicans (XP_721708),
Cryptococcus neoformans (XP_568887), Pneumocystis carinii (AAK54439), Glycine max SMT1 (AAB04057), Glycine max SMT2-1 (FJ483973) and Trypanosoma brucei AAZ40214).
Identical residues conserved in the primary structure are shaded. The sequences were aligned using Align (Informax Inc.) with defaulted parameters. The deduced substrate
preference of SMT that catalyzes the first (Δ24(25)-substrate) or second (Δ24(28)-substrate) C1-transfer reaction, SMT1 or SMT2 is reported. Sterol and AdoMet binding segments are
indicated as Regions I, III, and IV and Region II, respectively.
corresponded to the optimal catalytic competence Vmax/Km =0.76.
Zymosterol 7 was a poor substrate generating a Km 35 μM/Vmax
2.0 pmol/min. Comparison of the properties of cycloartenol 2 and
lanosterol reveals that the two substrates bind in a similar orientation
in the ternary complex of the active site (Km = 36 μM/V-
max =48 pmol/min). Consistent with the kinetic observations, an
overnight incubation of cycloartenol with PbSMT yielded a similar GC
pattern of substrate-product to that of the lanosterol assay (Fig. 4).
PbSMT incubation of 26,27-dehydrolanosterol 16 in which the
terminal side chain carbons at C26 and C27 are fused yielded an
apparent Km 19 μM and Vmax 6 pmol/min. Catalysis of 31-norlanos-
terol and 14-methyl zymosterol generated specificity constants of
35 μM/45 pmol/min and 30 μM/2 pmol/min, respectively, suggest-
ing that neither a C4β-methyl or 14-methyl group attached to the
lanosterol is harmful to activity. However, the C4α-methyl group on
the lanosterol structure is crucial for optimal catalytic competence. A
lack of productive binding of 24(28)-methylenelophenol 6, fecosterol
8 or eburicol 4 (no activity) (Fig. 1) indicates that PbSMT cannot
perform the second alkylation step, consistent with the absence of
these sterols in wild-type cells.
SMTs, as reviewed in reference 6. However, their similar binding
properties and other chemical data showing that they exist in solution
and solid state in a pseudoplanar conformation [6] suggest that the
likely shape of cycloartenol recognized by the PbSMT is flat, like that of
lanosterol. Accordingly, the structural feature that is paramount in
SMT recognition of lanosterol and cycloartenol versus zymosterol is
the C3-OH group, whose orientation in the A-ring and hydrogen-
bonding ability can affect productive binding of the acceptor
molecule.
3.5. Inhibition of PbSMT
In an effort to develop fungal-specific drugs that target the SMT
and trap the putative 24-alkyl sterol intermediate in the C24-
methylation reaction possible inhibitors, 26,27-dehydrolanosterol
16 and 25-azalanosterol 17 were constructed and fed to the fungus
and independently tested with the isolated enzyme. Azasterols and
related substrate analogs that possess a methylenecyclopropane
structure in the sterol side chain have been used as probes of sterol
biosynthesis. In the first case, they are believed to be reversible
“transition state analogs” in which the protonated, ammonium form
of the molecule mimics the putative carbonium ion intermediate in
the methylation-reduction catalysis of the Δ24-bond. In the second
case, the active-site directed analog is designed to be an irreversible
“mechanism-based” inhibitor specifically for the SMT [9]. Notably,
under physiological conditions, these inhibitors may impair growth
From these kinetic data it would appear that the natural substrate
of PbSMT is lanosterol, consistent with the pathway of lanosterol
conversion to eburicol determined from metabolite profiling of the P.
brasiliensis sterol metabolome. The sterols tested here possessing the
lanosterol or cycloartenol frame have been considered to exist in
different conformations, bent or flat and that these structural
differences may explain why plant sterols cannot bind to the fungal

M. Pereira et al. / Biochimica et Biophysica Acta 1801 (2010) 1163–1174
1169
respectively. The kinetic behavior and resemblance of the inhibitor to
the true substrate suggests tight binding in the active site but differs
from it so as to be unreactive. On the other hand, if the inhibitor
combines irreversibly with the PbSMT the kinetic pattern can
approximate that of noncompetitive inhibition since the net effect is a
loss of active enzyme. Usually, this type of inhibition can be
distinguished from the noncompetitive, reversible inhibition case
since the reaction of inhibitor with enzyme (and/or enzyme–substrate)
is not instantaneous. Instead, there is a time-dependent decrease in
enzymatic activity as enzyme plus inhibitor to enzyme–inhibitor
complex proceeds. The resulting covalent labeling of the enzyme should
be detectable by SDS-PAGE electrophoresis and further it is expected
that the enzyme inactivates such that the inactivated enzyme has 1 mol
of inhibitor covalently bound per mol of active sites [40].
Incubation of increasing concentrations of 16 resulted in pseudo-
first-order, time-dependent inactivation of the PbSMT, as evidenced by
the linear dependence of the log residual activity against time (Fig. 6).
The rate of inactivation by 16 was saturable, with a maximum rate of
inactivation, kinact, of 0.24 min−1 0.01 min−1 (t1/2=2.86 min) at
0.52 μM PbSMT and a Ki for 16 of 54 0.02 μM. These values compare
favorably with the steady-state kinetic parameters for the normal
substrate lanosterol 1 (kcat =0.14 min-1 0.3 min−1, Km=38 μM
3 μM). Co-incubation with 10 μM of normal substrate at 50 μM and
100 μM, lanosterol, afforded protection against inactivation, generating
52% and 75% C-methylation activity, respectively. As expected, no
inactivation occurred when the incubations were carried out lacking
AdoMet, consistent with the absolute dependence of PbSMT on the
cofactor.
To examine the specificity in the presumptive covalent modification
of the PbSMT, soluble protein typically studied in the standard assay was
treated with 16 in the presence of AdoMet diluted with catalytic
amounts of [methyl-3H3]AdoMet (1 μCi), and the resulting protein
assayed and radiofluorography (Fig. 3). Despite the fact that the soluble
protein fraction contained numerous protein species as demonstrated
by Coomassie staining, only a single protein was labeled and this
radioactive component eluted with mobility identical to that of the
purified recombinant PbSMT harboring three tags (ca. 55–60 kDa).
Fig. 3. SDS-PAGE analysis of recombinant PbSMT purified by Ni-chromatography and
corresponding fluorogram of the chemical affinity labeled soluble protein. The lanes
indicated are the E. coli cell lysate after the addition of IPTG (lane 1), the recombinant
PbSMT purified by Ni-NTA resin (Lane 2) and the soluble preparation of PbSMT assayed
with 26,27-dehydrolanosterol paired with [methyl-3H3]AdoMet. The migration of
protein standards (in kilodaltons, kDa) is also indicated.
by blocking enzyme action differently; through tight binding 17 or to
decrease the concentration of active enzyme 16.
Effects of 25-azalanosterol on the parameters of the Michaelis–
Menten equation were determined in the presence of lanosterol or
cycloartenol yielding similar competitive-type patterns relative to the
test sterol and Ki values of 14 nM (Fig. 5) and 17 nM (data not shown),
3.6. Mechanistic rationalization of 26,27-dehydrolanosterol catalysis
The reaction of an affinity label with the PbSMT involves initial
formation of a reversibly bound enzyme–substrate complex, followed
by catalysis to an intermediate that can be converted to a methyl
product or the intermediate can be intercepted through covalent
modification and hence irreversible inhibition. PbSMT incubation of
Fig. 4. GLC separation of the PbSMT conversion of lanosterol to eburicol (solid line) and
cycloartenol to 24(28)-methylene cycloartanol (dotted line). Product isolation and
conditions for the capillary GC analysis are described under “Materials and methods”.
Fig. 5. Inhibition of PbSMT by 25-azalanosterol. In the Lineweaver–Burk plot shown,
lanosterol is the varied substrate at 0, 5 μM, 10 μM, 20 μM, 50 μM and 100 μM and
AdoMet is fixed at 100 μM, and the inhibitor is varied at 0, 15 nM, 25 nM and 100 nM.

1170
M. Pereira et al. / Biochimica et Biophysica Acta 1801 (2010) 1163–1174
involving these substrates catalyzed by PbSMT, large-scale incuba-
tions were performed and the distribution and structures of the
enzyme-generated products analyzed by HPLC, MS and NMR.
PbSMT assayed with lanosterol 1 paired with [methyl-3H3]AdoMet
revealed a single radioactive product in the HPLC that co-migrates
with authentic eburicol (Fig. 8, panel A). PbSMT assay of lanosterol
paired with non-radioactive AdoMet afforded a product that in the
1HNMR analysis, δ C3 (3.23, m), C18 (0.69, s), C19 (1.00,s), C21 (0.92,
d), C26 (1.03 d), C27 (1.03, d), C28 (4.65, d), C31α (0.98, s), C31β
(0.81, s), and C32 (0.88, s).ppm, confirmed the eburicol structure.
These results show that C24-methylation of lanosterol to eburicol is
regiospecific and direct thereby eliminating the possible intermediacy
of a C25 cation 19 that can form hydroxylated intermediate 30 (Fig. 7,
path a).
PbSMT incubation of 26,27-dehydrolanosterol 16 paired with
[methyl-3H3]AdoMet afforded two radioactive products, one com-
pound elutes as a polar specimen significantly before the elution time
of the substrate and a second compound elutes in the tail region of the
substrate (Fig. 8, panel B.). The ratio of the two enzyme-generated
products, referred to as (unknown) U-1 and U-2 is ca. 10:1,
respectively. The experiment was repeated with 16 and non-
radioactive AdoMet and the sterol fractions in HPLC monitored
using a multiple wavelength diode-array detector to detect UV
absorption for the number and type of double bonds that might be
Fig. 6. Inactivation of PbSMT with the mechanism-based inactivator, 26-27-dehydrola-
nosterol. Semi-log plots of residual activity versus time at 0 μM, 25 μM, 50 μM and 100 μM
concentrations of the inhibitor. The inset shows a replot of enzyme half-life (t1/2) for
inactivation versus 1/[I].
lanosterol or 26,27-dehydrolanosterol can lead to distinct intermedi-
ates that convert to methylated sterol products that reflect whether
the intermediate turned over (“monol”) or covalently bound to the
enzyme (“diol”) as outlined in Fig. 7. To determine the reaction course
Fig. 7. Alternate pathways and mechanisms for the conversion of Δ24(25)-containing sterol side chains to methylated product. Formation of these monol (referring to the sterol
containing a 3β-OH group) and diol (referring to the sterol containing a 3β-OH group and a second OH group in the sterol side chain) species requires distinct enzyme-bound
cationic intermediates. Methyl from AdoMet addition to the sterol side chain can occur at C24 to form a C28 methyl group in structure 29 or at C26 to form a C33 methyl group in
structures 25 and 27. Interception of the enzymatic conversion of 16 to 24 and 27 can occur to inactivate the SMT; saponification of the quenched enzyme assay 31 and 32, ligands
bound in the active site of the PbSMT. Note that the stereochemistry of the natural methyl product 29 is 25R and can be demonstrated by incubation of the relevant 13C27 sterol
acceptor molecule.

M. Pereira et al. / Biochimica et Biophysica Acta 1801 (2010) 1163–1174
1171
the structure C2H3 rather than the structure CH2H2 as would be
expected for methyl attack at C24.
Incubation of 16 produced significant amounts of the polar
metabolite U-1 relative to U-2 (tentatively the diol and monol
metabolites reported previously using the ScSMT) and much more
diol than reported in the incubation of ScSMT with 26,27-dehydro-
zymosterol [27,42,43]. The UV of U-1 possessed end absorption and
the mass spectrum possessed Mr of 456 and other diagnostic ions at
441, 423, 405, 398, 383 and 365, that suggests the sterol was
methylated in the side chain, contained one double bond in the side
chain and contained two hydroxyl groups in the molecule, i.e. a “diol”
(Supplemental Fig. 8). The 1HNMR of this sample was δ C3 (3.23, m),
C18 (0.69, s), C19 (1.00 s), C21 (0.93,d), C24 (5.34, m), C26 (1.68 m),
C27 (3.65, m), C28 (0.894, t), C30α (0.98, s), C31β (0.81,s), C32 (0.88,
s).The NMR spectrum showed the presence of a remote ethyl group
(0.887, t, j 7.5 Hz), an olefinic proton at δ 5.34 (m) coupled to the
primary alcohol group (C27 which resonated at 3.65, δ, J=10.5 Hz)
and a secondary alcohol group associated with C3 (multiplet at
3.23 ppm). The shifts of the signals for the methyl groups at C4 and
C14 in the NMR spectra were consistent with retention of the
lanosterol frame. The only possible structure for this new sterol is 32.
Notably, U-2 can only be recovered from the incubation mixture after
treatment with methanolic-KOH, suggesting that the sterol is bound
to the enzyme in an ester linkage and that the formation of the “extra”
alcohol group in the sterol side chain is chemically induced during the
saponification process. Further incubations of PbSMT with [methyl-
2H3]AdoMet paired with lanosterol afforded a single product of M+
442 amu whereas [methyl-2H3]AdoMet paired with 26,27-dehydrola-
nosterol afforded two products that contained molecular ion values at
M+ 441 amu and 459 amu, respectively. Comparison of the molecular
weights of incubation products of sterol acceptor 1 and 16 paired with
AdoMet versus that paired with [methyl-2H3]AdoMet, revealed a mass
increase in the high mass end of the spectrum only [M+ and M+-CH3]
associated with the addition of two 2H atoms in the side chain of
eburicol 4 and of three 2H atom increase in the side chain of 27,
respectively. The labeling studies provide support that the methyl
from AdoMet introduced to the side chain of 26,27-dehydrolanosterol
occurs at C26 to form a 26(33)-ethyl side chain, elongated compared
to the typical phytosterols ethylated at C24(28) as in stigmasterol 15.
Taken together, the evidence indicates that the monol and diol
products formed by catalysis of 16 involves partitioning of the
methylated cation 22 in the direction of cation 24 which can form an
alkylated enzyme with ligand structure 26 or convert to methyl
product 27 after rearrangement and deprotonation of C23. This
reaction path is a marked departure from the one favored by the
SCSMT incubated with 26,27-dehydrozymosterol in which cation 22
proceeds down the path to cation 23 generating 25 and 31 [27].
Fig. 8. HPLC chromatogram of the total sterol fraction of soluble PbSMT incubated with
saturating amounts of lanosterol paired with [methyl-3H3]AdoMet (panel A showing
enzyme-generated radioactivity of 4 incubations) and saturating amounts of 26,27-
dehydrolanosterol paired with [methyl-3H3]AdoMet (panel B showing half of the enzyme-
generated radioactivity of 8 incubations) as described in the “Materials and methods”
section. The dotted line shows the elution profile of the substrate and the solid line shows
the elution profile of the radioactive methylated product(s). Note that in the bottomfigure,
the methylated diol elutes between 8 and 14 min and the methylated monol elutes
between 32 and 36 min; based on the recovery of radioactivity in the fractions
corresponding to the diol the PbSMT kcat can be determined as discussed in the Results
section. Approximately 90% of the radioactivity injected into the column is recovered
within the time frame of 3 to 60 min.
formed in the enzyme-generated metabolites. The relevant U-1 and
U-2 compounds were recovered from the HPLC fractions and further
analyzed by GC-MS (Supplemental Fig. 6) and in the case of U-2
additionally by 1HNMR. For U-2, the major set of peaks in the high end
of the mass spectrum were 438 (M+), 423 (M+-CH3) 420 (M+-H20,
not detected), 409 (M+-C2H5, allylic cleavage of the C26–C28 system),
391 (M+-CH3-H20-CH3)), 405 (M+-CH3-H20), 382 (M+-C4H8, allylic
cleavage thru C24/C25); the UV spectrum was bimodal with end
absorption and a second peak centered at 235-240 nm indicating a
Δ8-methylated sterol with a side chain of two double bonds in
conjugation. These results are consistent with U-2 to be a new
C26 methyl compound 27 (Fig. 7). Incubation of 16 paired with
[methyl-2H3]AdoMet formed U-2. The mass spectrum of this sample was
composed of ions at m/z values of 441, 426, 409, 405, 391.The major ion
of the molecular ion cluster in the high mass end of the spectrum has an
m/z value of 441, three mass units greater than that of unlabelled U-2,
consistent with the addition of three 2H atoms to the sterol side chain.
There is little doubt that the biosynthetically introduced 2H atoms are
located in the extra methyl attached to the 26-methyl group which has
3.7. Stereochemistry of electrophilic alkylation
The alkylating mechanism for the formation of C28 in brassicas-
terol involves, as a first stage, the formation of a 24(28)-methylene
sterol intermediate through positional selectivity of the normal
deprotonation step of the reaction (i.e. regioselectivity for deprotona-
tion at C28 rather than C27 exemplified in the C-methylation reaction
performed by the P. wickerhamii SMT, [44]) coupled to the migration
of a hydrogen atom from C24 to C25. From a stereochemical point of
view, this migration directed by the electrophilic alkylation attack of
C24 can occur in two ways which lead to opposite configurations at
C25. To probe the stereochemistry of the migration of H24 to C25, we
separately incubated [24-2H]cycloartenol and [13C-27]cycloartenol
with PbSMT. Preparative incubation of PbSMT with [24-2H]cycloarte-
nol generated a single product that was separated from the substrate
by HPLC. MS of the enzyme-generated product gave M+ 441 amu
consistent with the methyl group and one deuterium atom addition to
the sterol side chain. The 500 MHz 1HNMR spectrum of the 2H-labeled

1172
M. Pereira et al. / Biochimica et Biophysica Acta 1801 (2010) 1163–1174
C24-methylene sterol compared to that of 24(28)-methylenecycloar-
tanol was different mainly in the chemical shifts for C26 and C-27 at
1.035 (d) and 1.22 (d) that moved downfield to δ 1.23 and 1.018 ppm
due to an α-deuterium isotope effect and the signals collapsed from
doublets (in the parent compound) to singlets (in the product). The
retention of a single deuterium (N99%) with [24-2H]cycloartenol and
the detection of that deuterium at C25 in the enzyme-generated
product establish that the methyl-methylene elimination at C28 and
the H24 of the substrate shift to form a new stereogenic center at C25
in a stereocontrolled manner. Further PbSMT incubation of [27-13C]
cycloartenol afforded a 13C-labeled methyl product of M+ 441 amu
(Supplemental Fig. 7, panel A). The signals for individual carbon
atoms matched the spectrum previously determined in this labora-
tory [45]: :1NMR δ ppm, 0.97 (H-18, s), 0.56 and 0.33 (H-19, d,
4.2=endo, d, 4.2=exo), 0.90 (H-21, d), 1.02 (H-26, dd), 1.02 (H-27,
dd), 0.97 (H-30, s), 0.81 (H-31, s), 0.90 (H-32, s), 3.21 (H-3, m), 4.72,
4.67 (H-28, d, d); 13CNMR (terminal side chain carbon atoms)
125 MHz δ ppm C24 (156.94), C25 (33.65), C26 (21.86), C27 (21.99)
and C28 (105.90). In the 13CNMR spectrum of [27-13C]24(28)-
methylene cycloartanol, the major enhanced signal at 21.90 ppm
(Supplemental Fig.7, panel B) corresponds to C 27, therefore C25 in
this specimen possesses the R-stereochemistry. These results indicate
that PbSMT catalysis of the sterol acceptor lead to a C24-methyl
product in which the 1,2-hydride migration of C24 to C25 occurs
specifically from the Re-face of the original substrate double bond
undergoing transalkylation as outlined in Fig. 7, path a 1→18→29.
3.8. Directed mutagenesis at Tyr88
The PbSMT amino acid at position Tyr88 is a critical amino acid in the
active sites of SMTs and corresponds to the highly conserved Tyr81 in the
fungal ScSMT and Tyr83 in plant GmSMT primary structures. The effect of
Fig. 9. Unrooted phylogenetic tree of ERG6 genes across eukaryote Kingdoms. Complete amino acid sequences of SMTs are available in the Pfam database. The phylogenetic tree was
built by the neighbor-joining method Representative organisms and GenBank accession number: Leishmania donovani (Q6RW42); L. major (Q4Q1I3); Trypanosoma brucei (Q4FKJ2);
T. cruzi (Q4CMB7); Dictyostelium discoideum (Q54I98); Aspergillus nidulans (Q5AX34); A. oryzae1 (Q2UBN2); A. oryzae2 (Q2U0V6); A. fumigatus (Q4W9V1); Gibberella zeae1
(Q4IAL8); G. zeae2 (Q4IJ25); Neurospora crassa (Q9P3R1); Candida albicans (O74198); C. glabrata (Q6FRZ7); Clavispora lusitaniae (Q875K1); Magnaporthe grisea (Q5EN22); Yarrowia
lipolytica (Q6C2D9); Debaryomyces hansenii (Q6BRB7); Kluyveromyces lactis (Q6CYB3); Ustilago maydis (Q4P9N1); Saccharomyces cerevisiae (P25087); Pneumocystis carinii
(Q96WX4); Eremothecium gossypii (Q759S7); Cryptococcus neoformans (Q5KM39); Schizosaccharomyces pombe (O14321); Arabidopsis thaliana1 (Q9LM02); A. thaliana2 (Q39227);
A. thaliana3 (Q94JS4); Gossypium hirsutum (Q2QDF7); Glycine max (Q43445); Ricinus communis (O24328); N. tabacum1 (O82434); N. tabacum2 (O24154); N. tabacum3 (O82720);
Triticum aestivum (Q41586); Oryza sativa1 (Q84M50); O. sativa2 (Q53MP5); O. sativa3 (Q6ZIX2); O. sativa4 (O82427); Zea mays (O49215). Robustness of branches was estimated
using 100 bootstraped replicates and indicates the percentage of times that all species appear as a monophyletic cluster.

M. Pereira et al. / Biochimica et Biophysica Acta 1801 (2010) 1163–1174
1173
mutation on substrate binding and catalysis resulting from the
substitution at Tyr-81 (Erg6p nomenclature) with leucine is a preferential
change in affinity for Δ24-sterols that affords alter partitioning in the
direction of 24-ethyl(idene) product formation. Remarkably, when a
point mutation at Tyr88 of phenylanine or leucine is introduced in the
primary structure of PbSMT catalytic competence is minimally affected
(90% and 50% activity relative to the activity of lanosterol controls,
respectively); GC-MS analysis (data not shown) of the enzyme-
generated hexane-extractable lipids show a single 24-methylene sterol
product. Although substrate affinity is not attenuated in the Tyr88Phe
mutant, it is in the Tyr88Leu mutant, as indicated by the 50% loss in
activity, suggesting that this residue in the native protein can contribute
to stabilization of the active-site topography.
outlined in Fig. 7. In addition, the labeling results rule out an alternative
generation of 19 such that the methyl addition from AdoMet approaches
from the Re-face (a-face methylation) of the substrate double
bond and firmly exclude the C24-methylation mechanism to create the
24α-methyl stereochemistry previously suggested for plant SMTs that
recognize cycloartenol as substrate [49–51].
While mechanistically similar to other SMTs, the PbSMT recognizes
and acts on distinct substrate features and can catalyze unique
partitioning that results in novel methylated products. These unusual
SMT properties may be considered in the definition of PbSMT into a
separate and fourth family of SMTs. One implication of the sequence
similarity and genomic organization of known SMT genes [1,39] is that
regions of sequence conservation may correspond to functional
domains and those functional domains may mediate catalysis of
particular reactions common to all SMTs across kingdoms. Conversely,
molecular analysis of the amino acid sequences of these SMTs coupled
with the variant substrate specificity for the corresponding enzyme
affords a phylogenetic tree representing the different SMT clans, in
which PbSMT is closely aligned to a specific Ascomycetes group as well
as to the Basidiomycetes (Fig. 9). Notably, the catalytic competence
carried out by SMT generates the phytosterol diversity, but how and
when the requisite substrate requirements evolved and the ordering of
different C-methylation pathways remains enigmatic. To this end,
determination of the SMT three-dimensional structure and unearthing
the roots of SMT catalysis provides an empirical approach to purse
rigorous investigations of structure-guided studies to develop SMT-
specific inhibitors which are currently underway.
4. Discussion
The work described here was directed toward characterizing the
sterol pathway and to establish the protein sequence and kinetic
parameters of the sterol methylating enzyme necessary in the
growth of P. brasiliensis. By examining the recombinant PbSMT
properties and establishing the sterol metabolome that defines
brassicasterol biosynthesis, it has been possible for the first time not
only to deduce the order of intermediates in sterol biosynthesis of
this fungal pathogen, but also to decipher new evolutionary
relationships of a crucial house keeping sterol catalyst using
substrate-based cataloging. Correlation of the features giving rise to
C28 and C26-methylated sterol products as well as directed
mutational analysis of a key amino acid residue shown in other
work to promote gain-in-function activities (acquired C2-transfer
activity) [39] revealed an unexpected influence of the activated
complex architecture to guide partitioning toward enzyme inactiva-
tion in a substrate- and phyla-specific manner.
We searched the P. brasiliensis genome for enzyme sequences
involved in sterol (http://www.broadinstitute.org/annotation/ge-
nome/paraccidioides_brasiliensis/Blast.html) biosynthesis and of the
10 ergosterol genes in the post-lanosterol pathway detailed for the S.
cerevisiae proteome all of them have been detected. In addition, the late
step Δ7-reductase gene detected in animal systems and most recently
reported in the fungus Mortierella alpina [46] that gives rise to the Δ5-
monoene system was also evident in the search. The proposed
brassicasterol pathway shown in Supplemental Fig. 3 is much the
same as the ergosterol pathway recently described in C. neoformans [25]
but is distinct from the ergosterol-brassicasterol pathway identified in
the Ascomycetes Giberella fujikori that proceeds from lanosterol through
obtusifoliol (C4α nor-eburicol) to form fecosterol [35]. Thus there are
separate pathways that can converge on the same set of 24-alkyl sterol
end products.
To enable a functional genomics approach to phytosterol biosynthe-
sis, an infection cDNA library from P. brasiliensis [28] was used togenerate
a full-length reading frame required in the cloning of PbSMT. The
resulting fusion protein was shown to catalyze a C1-transfer reaction
utilizing lanosterol or cycloartenol as the optimal substrate. Identification
by GC-MS and NMR of the enzyme-generated products demonstrated
specificity and regioselectivity of the PbSMT. When investigating the
apparent second-order rate constants associated with the test sterols,
steady-state discrimination of 31-norlanosterol compared to zymosterol
showed the former is the markedly superior substrate and that it is
similar to lanosterol. Moreover, cycloartenol was accepted as an optimal
substrate equal in activity to lanosterol, consistent with the product
profile analyses. These results indicate that the PbSMT active site is
distinctly different from that of the other SMTs studied to date, including
the related yeast P. carinii that is bifunctional (capable of C1- and C2-
transfer activities) and catalyzes 24(28)-methyelene lanosterol produc-
ing multiple 24-ethyl(idene) lanostenols [47,48]. The NMR analyses of
PbSMT catalyzed cycloartenol substrates support the proposed mecha-
nism for the C24-methylation of Δ24-sterols in eukaryotes [6,7], as
Acknowledgements
The authors acknowledge financial support from the National
Science Foundation (MCB-0920212 and MCB-0417436) and the
Welch Foundation (D-1276) to WDN and the Conselho Nacional de
Desenvolvimento Centifico e Tecnologico (CNPq) and International
Foundation for Science (IFS), Stockholm, Sweden to MP. LKSS and
AHSC are the recipients of a fellowship from the Coordenação de
Aperfeiçoamento de Ensino Superior (CAPES).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the
online version, at doi:10.1016/j.bbalip.2010.06.007.
References
[1] J. Jayasimha, C.B. Bowman, J.M. Pedroza, W.D. Nes, Engineering pathway enzymes
to understand the function and evolution of sterol structure and activity, Rec. Adv.
Phytochem. 40 (2006) 211–251.
[2] F. Bouvier, Al. Rahier, B. Camara, Biogenesis, molecular regulation and function of
plant isoprenoids, Prog. Lipid Res. 44 (2005) 357–429.
[3] P. Benveniste, Biosynthesis and accumulation of sterols, Annu. Rev. Plant Biol. 55
(2004) 429–457.
[4] J.J. Brocks, G.A. Logan, R. Buick, R.E. Summons, Archean molecular fossils and the
early rise of eukaryotes, Science 285 (1999) 1033–1036.
[5] W.D. Nes, Control of sterol biosynthesis and its importance to developmental
regulation and evolution, Rec. Adv. Phytochem. 24 (1990) 283–327.
[6] W.D. Nes, Sterol methyltransferase: enzymology and inhibition, Biochim. Biophys.
Acta 1529 (2000) 63–88.
[7] D. Arigoni, Stereochemical studies of enzymatic C-methylation reactions, Ciba
Found. Symp. 60 (1978) 243–258.
[8] J.A. Urbina, Ergosterol biosynthesis and drug development for Chagas disease,
Mem. Inst. Oswaldo Cruz 104 (2009) 311–318.
[9] Z. Song, W.D. Nes, Sterol biosynthesis inhibitors: potential for transition state
analogs and mechanism-based inactivators targeted at sterol methyltransferase,
Lipids 42 (2007) 15–33.
[10] G. Visba, G. San-blas, J. Murgich, H. Franco, Paracoccidiodes brasiliensis,
Paracoccidiomycosis and antifungal antibiotics, Curr. Drug Targets Infect. Disord.
5 (2005) 211–226.
[11] J. Burbiel, F. Bracher, Azasteroids as antifungals, Steroids 68 (2003) 587–594.
[12] E.S. Kaneshiro, Sterol biosynthesis in Pneumocystis: unique steps that define
unique targets, Drug Resist. Updat. 5 (2002) 259–268.

1174
M. Pereira et al. / Biochimica et Biophysica Acta 1801 (2010) 1163–1174
[13] R. Bitmann, Has nature designed the cholesterol side chain for optimal interaction
with phospholipids, Subcell. Biochem. 28 (1997) 145–171.
[33] S.W. Ellis, M.E. Rose, M. Grindle, Identification of a sterol mutant of Neurospora
crassa deficient in D14(15)-reductase activity, J. Gen. Microbiol. 137 (1991)
2627–2630.
[34] L. Mejanelle, J.F. Loez, N. Gunde-Cimerman, J.O. Grimalt, Ergosterol biosynthesis in
novel melanized fungi from hypersaline environments, J. Lipid Res. 42 (2001)
352–359.
[35] W.D. Nes, P.H. Le, Evidence for separate intermediates in the biosynthesis of 24β-
methylsterol end products by Gibberella fujikuroi, Biochim. Biophys. Acta 1042
(1990) 119–125.
[36] W.R. Nes, M.L. McKean, Biochemistry of Steroids and Other Isopentenoids,
University Park Press, Baltimore, 1977, pp. 325–455.
[14] H. Jin, J.M. McCaffery, E. Grote, Ergosterol promotes pheromone signaling and
plasma membrane fusion in mating yeast, J. Cell Biol. 180 (2008) 813–826.
[15] S.A. Brunt, R. Riehl, J.C. Silver, Steroid hormone regulation of the Achlya
ambisexualis 85-kilodalton heat shock protein, a component of the Achlya steroid
receptor complex, Mol. Cell. Biol. 10 (1990) 271–281.
[16] R.J. Rodriguez, C. Low, C.D.K. Bottema, L.W. Parks, Multiple functions for sterols in
Saccharomyces cerevisiae, Biochim. Biophys. Acta 837 (1985) 336–343.
[17] W.J. Pinto, W.R. Nes, Stereochemical specificity for sterols in Saccharomyces
cerevisiae, J. Biol. Chem. 258 (1983) 4472–4476.
[18] J.D. Weete, Mechanism of fungal growth suppression by inhibitors of ergosterol
biosynthesis, ACS Symp. Ser. 325 (1987) 268–285.
[19] G. San-Blas, G. Nino-Vega, Paracoccidioides brasiliensis: chemical and molecular
tools for research on cell walls, antifungals, diagnosis, taxonomy, Mycopathologia
165 (2008) 183–195.
[37] S. Shimizu, H. Kawashima, M. Wada, H. Yamada, Occurrence of a novel sterol,
24,25-methylenecholest-5-en-3β-ol in Mortierella alpina,1S-4, Lipids 27 (1992)
481–483.
[38] W.D. Nes, S.D. Nichols, Phytosterol biosynthesis pathway in Mortierella alpina,
Phytochemistry 67 (2006) 1716–1721.
[20] G. Visba, A. Alvarez, B. Moreno, G. San-Blas, S-Adenosyl-L-methionine inhibitors
Δ24-sterol methyltransferase and Δ24(28)-sterol methylreductase as possible agents
against Pareacoccidiodes brasiliensis, Antimicrob. Agents Chem. 47 (2003) 2966–2970.
[21] A.M. Pierce, R.B. Mueller, A.M. Unrau, A.C. Oehlschlager, Metabolism of Δ24-sterols
by yeast mutants blocked in removal of the C14-methyl group, Can. J. Chem. 56
(1978) 794–800.
[22] R. Kanagasabai, W. Zhou, J. Liu, T.T.M. Nguyen, P. Veeramachaneni, W.D. Nes,
Disruption of ergosterol biosynthesis, growth and the morphological transition in
Candida albicans by sterol methyltransferase inhibitors containing suflur at C25 in
the sterol side chain, Lipids 39 (2004) 737–746.
[23] T.T.M. Nguyen, Influence of Inhibitors of Sterol Biosynthesis of Growth and Sterol
Metabolism in Candida albicans, Master of Science, Texas Tech University, 2003,
pp. 1–89.
[24] A.C. Oehschlager, R.H. Angus, A.M. Pierce, H.D. Pierce, R. Srinivasan, Azasterol
inhibition of Δ24-sterol methyltransferase in Saccharomyces cerevisiae, Biochem-
istry 23 (1984) 3582–3589.
[39] K. Ganapathy, C.W. Jones, C.M. Stephens, R. Vatsyayan, J.M. Marshall, W.D. Nes,
Molecular probing of the Saccharomyces cerevisiae sterol 24-C-methyltransferase
reveals multiple amino acid residues involved with C2-transfer activity, Biochim.
Biophys. Acta 1781 (2008) 344–351.
[40] W. Zhou, G.I. Lepesheva, M.R. Waterman, W.D. Nes, Mechanistic analysis of a
multiple product sterol methyltransferase implicated in ergosterol biosynthesis
from Trypanosoma brucei, J. Biol. Chem. 281 (2006) 6290–6296.
[41] A.K. Neelakandan, Z. Song, J. Wang, M.H. Richards, X. Wu, B. Valliyodan, H.T.
Nguyen, W.D. Nes, Cloning, functional expression, and phylogenetic analysis of
plant 24C-methyltransferases involved in sitosterol biosynthesis, Phytochemistry
70 (2009) 1982–1998.
[42] J.A. Marshall, W.D. Nes, Isolation and characterization of an active-site peptide
from a sterol methyl transferase with a mechanism-based inhibitor, Bioorg. Med.
Chem. Lett. 9 (1999) 1533–1536.
[43] W.D. Nes, J.A. Marshall, Z. Jia, T.T. Jaradat, Z. Song, P. Jayasimha, Active site
mapping and substrate channeling in the sterol methyl transferase pathway, J.
Biol. Chem. 277 (2002) 42549–42556.
[25] W.D. Nes, W. Zhou, K. Ganapathy, J. Liu, R. Vatsyayan, S. Chamala, K. Hernandez, M.
Miranda, Sterol 24-C-methyltransferase: an enzymatic target for the disruption of
ergosterol biosynthesis and homeostasis in Cryptococcus neoformans, Arch.
Biochem. Biophys. 481 (2009) 210–218.
[44] A.T. Mangla, W.D. Nes, Sterol C-methyl transferase from Prototheca wickerhamii:
mechanism, sterol specificity and inhibition, Bioorg. Med. Chem.
925–936.
8 (2000)
[26] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein dye-binding, Anal. Biochem.
72 (1976) 248–254.
[45] A.L. Dennis, W.D. Nes, Sterol methyl transferase. Evidence for successive C-methyl
transfer reactions rgenerating Δ24(28) and Δ25(27)-olefins by
single plant
enzyme, Tetrahedron Letts. 43 (2002) 7017–7021.
-
a
[27] W.D. Nes, P. Jayasimha, Z. Song, Yeast sterol C24-methyltransferase: role of highly
conserved tyrosine-81 in catalytic competence studied by site-directed mutagenesis
and thermodynamic analysis, Arch. Biochem. Biophys. 477 (2008) 313–323.
[28] M. Costa, C.L. Borges, A.M. Bailao, G.V. Meirelles, Y.A. Mendonca, S.F. Dantas, F.P. de
Faria, M.S. Felipe, E.E. Molinari-Madlum, J. Mendes-Giannini, R.B. Fuza, W.S.
Martins, M. Pereira, C.M.A. Soares, Transcriptome profiling of Paracoccidiodes
brasiliensis yeast-phase cells recovered from infected mice brings new insight into
fungal response upon host interaction, Microbiology 153 (2007) 4194–4207.
[29] W.D. Nes, G.G. Janssen, F.G. Crumley, M. Kalinowska, T. Akihisa, The structural
requirements of sterols for membrane function in Saccharomyces cerevisiae, Arch.
Biochem. Biophys. 300 (1993) 724–733.
[46] S. Zhang, E. Sakuradani, S. Shimizu, Identification of a sterol D7-reductase gene
involved in desmosterol biosynthesis in Mortierella alpina 1S-4, Appl. Environ.
Microbiol. 73 (2007) 1736–17241.
[47] E.S. Kaneshiro, J.A. Rosenfeld, M. Brasselin-Eiweida, J.R. Stringer, S.P. Keely, A.G.
Smulian, J.L. Giner, The Pneumocystis carinii drug target S-adenosyl-L-methionine:
sterol methyltransferase has a unique substrate preference, Mol. Microbiol. 44
(2002) 989–999.
[48] W. Zhou, T.T.M. Nguyen, M.S. Collins, M.T. Cushion, W.D. Nes, Evidence for
multiple sterol methyltransferse pathways in Pneumocystis carinii, Lipids 37
(2002) 1177–1186.
[49] S. Seo, A. Uomori, Y. Yashimura, K.J. Takeda, Stereospecificity in the biosynthesis of
phytosterol side chains: carbon 13 NMR signal assignments of C26 and C27, J.
Amer. Chem. Soc. 105 (1983) 6343–6344.
[30] W. Zhou, G.A.M. Cross, W.D. Nes, Cholesterol import fails to prevent catalyst-
based inhibition of ergosterol synthesis and cell proliferation of Trypanosoma
brucei, J. Lipid Res. 48 (2007) 665–673.
[50] A. Rahier, J.C. Genot, F. Schuber, P. Benveniste, A.S. Narula, Inhibition of S-
adenosyl-L-methionine sterol-C-24-methyltransferase by analogues of a carbo-
cationic ion high energy intermediate. Structure activity relationships for C-25
heteroatoms (N, As, S) substituted triterpenoid derivatives, J. Biol. Chem. 259
(1984) 15215–15223.
[31] M. Venkatramesh, D. Guo, Z. Jia, W.D. Nes, Mechanism and structural require-
ments for transformation of substrates by the (S)-adenosyl-L methionine: Δ24(25)
-
sterol methyl transferase from Saccharomyces cerevisiae, Biochim. Biophys. Acta
1299 (1996) 313–324.
[32] L. Alcazar-Fuoli, E. Mellado, G. Garcia-Effron, J.F. Lopez, J.O. Grimalt, J.M. Cuenca-
Estrella, J.L. Rodriguez-Tudela, Ergosterol biosynthesis pathway in Aspergillus
fumigatus, Steroids 73 (2008) 339–347.
[51] T.W. Goodwin, Biosynthesis of sterols, in: P.K. Stumpf, E.E. Conn (Eds.), The
Biochemistry of Plants: A Comprehensive Treatise, vol 4, Academic Press, New
York, 1980, pp. 485–507.