Protein J

DOI 10.1007/s10930-017-9729-7

Characterization of the Dihydroorotase from Methanococcus

jannaschii
Jacqueline Vitali1,2   · Aditya K. Singh1,4 · Michael J. Colaneri3 

© Springer Science+Business Media, LLC 2017

Abstract  The gene that codes for the putative dihy- some of the factors that may confer thermostability—more
droorotase in the hyperthermophilic archaeon Methanococ- Lys and Ile, fewer Ala, Thr, Gln and Gly residues, and
cus jannaschii was subcloned in pET-21a and expressed in shorter N- and C-termini. Additional and better insight into
Escherichia coli. A purification protocol was devised. The the thermostabilization strategies adopted by this enzyme
purity of the protein was evaluated by SDS-PAGE and the will be provided when its crystal structure is determined.
protein was confirmed by sequencing using LC–MS. The
calculated molecular mass is 48104 Da. SEC-LS suggested Keywords  Dihydroorotase · Pyrimidine biosynthesis ·
that the protein is a monomer in solution. ICP-MS showed Methanococcus jannaschii · Enzyme kinetics ·
that there are two Zn ions per monomer. Kinetic analysis Thermostability · Homology modeling
of the recombinant protein gave hyperbolic kinetics with
­Vmax = 12.2 µmol/min/mg and K ­ m = 0.14 mM at 25 °C. Fur- Abbreviations
thermore the activity of the protein increased with tempera- A. aeolicus  Aquifex aeolicus
ture consistent with the hyperthermophilic nature of the ATCase Aspartate transcarbamoylase
organism. A homology model was constructed using the B. anthracis, Ba Bacillus anthracis
mesophilic Bacillus anthracis protein as the template. Resi- BME 2-Mercaptoethanol
dues known to be critical for Zn and substrate binding were BSA Bovine serum albumin
conserved. The activity of the enzyme at 85 and 90 °C was CA Carbamoyl aspartate
found to be relatively constant over 160 min and this corre- CAD Carbamoyl phosphate synthetase/
lates with the temperature of optimal growth of the organ- aspartate transcarbamoylase/dihy-
ism of 85 °C. The amino acid sequences and structures of droorotase protein
the two proteins were compared and this gave insight into CID Collision induced dissociation
DHO Dihydroorotate
DHOase Dihydroorotase
* Jacqueline Vitali
[email protected] E. coli, Ec  Escherichia coli
IPTG Isopropyl β-d-1-thiogalactopyranoside
1
Department of Physics, Cleveland State University, LB Luria–Bertani medium
Cleveland, OH 44115, USA
MES 2-(N-morpholino)ethanesulfonic acid
2
Department of Biology, Geology and Environmental M. jannaschii, Mj 
Methanococcus jannaschii
Sciences, Cleveland State University, Cleveland, OH 44115,
SRM Selective reaction monitoring
USA
3
SDS-PAGE Sodium dodecyl sulfate polyacryla-
Department of Chemistry and Physics, State University
mide gel electrophoresis
of New York at Old Westbury, Old Westbury, NY 11568,
USA S. aureus  Staphylococcus aureus
4 T. thermophilus  Thermus thermophilus
Present Address: Department of Pharmacology
and Toxicology, University of Texas Medical Branch, Tris Tris(hydroxymethyl)aminomethane
Galveston, TX 77555, USA

13
Vol.:(0123456789)
 J. Vitali et al.

1 Introduction with ATCase. ATCase catalyzes the formation of CA which

is the substrate of DHOase.
DHOase (EC 3.5.2.3) catalyzes the conversion of N-car- The basic mechanism of the DHOase reaction was deter-
bamoyl-l-aspartate to l-dihydroorotate (Fig. 1) in the third mined from the crystal structure of the type II Escherichia
step of the de novo biosynthesis of pyrimidines. It is a coli enzyme [5] and biochemically verified [6]. The CA
zinc-containing member of the amidohydrolase superfam- substrate binds in the active site to the two Zn ions and to a
ily of metalloenzymes [1]. Historically, dihydroorotases number of invariant residues as well as main chain atoms.
were classified into two distinct types [2]. Long (type I) An invariant aspartate abstracts a proton from the amide
evolved first and are larger with molecular weight ~45 kDa. nitrogen of CA facilitating a nucleophilic attack. This basic
Short (type II) evolved more recently and are smaller with mechanism is maintained in all DHOases as the residues
molecular weight ~38  kDa. The two classes have very involved are invariant and the key interactions with the sub-
low sequence identity (<20%). A more recent phyloge- strate are conserved in the known complexes [3, 5, 7, 8].
netic analysis [3] gave two highly divergent clusters con- Archaeal DHOases have not been previously stud-
sistent with this classification. Long DHOases include the ied. The phylogenetic analysis [3] suggests that archaeal
subclasses of archaeal, bacterial type I, bacterial type III, and bacterial type I DHOases are more closely related to
animal CAD and CAD-like inactive DHOases from fungi. each other than to the other types. Archaeal DHOases are
Short DHOases further subdivide in active DHOases from expected to have both similarities and differences from the
fungi, bacterial type II and plant DHOases. DHOase is prokaryotic and eukaryotic DHOases. In addition, many
believed to be the ancestral protein of the amidohydrolase archaea live in harsh environments and some of the dif-
superfamily in part because it is present in most life forms ferences correspond to adaptations to their physiological
and in part because it is a biosynthetic enzyme as compared environment. This paper is the first study of an archaeal
to the other members of the family that have a catabolic DHOase, that from the hyperthermophilic archaeon Meth-
role [2, 4]. Other members of the superfamily acting on a anococcus jannaschii. We are interested in studying this
cyclic amide ring include dihydropyrimidinases, d-hydan- dihydroorotase in order to gain further insight into the
toinases, and allantoinases [1, 4]. variability within the dihydroorotase family and in order to
All DHOases adopt the (βα)8-barrel fold (TIM barrel) understand how the structure and function of this enzyme
which is characteristic of the amidohydrolase superfam- adapt to the high temperatures that are the normal habitat
ily of proteins [1]. The differences between the two clus- of this organism. As a first step, the pyrC gene of M. jan-
ters are primarily structural [3]. Long DHOases have extra naschii was subcloned into an expression vector, the gene
residues at the N- and C-termini forming a two-layered β product was purified, and the enzyme was characterized. A
stranded domain adjacent to the TIM barrel compared to homology model was built based on the homologous meso-
short DHOases that essentially consist of the catalytic philic Bacillus anthracis DHOase.
core. DHOases have two or one Zn ions in their active site
as well as four conserved histidines and one aspartate that
coordinate them. In type II, III and CAD DHOases the two 2 Materials and Methods
Zn ions are bridged by a carboxylated Lys residue which is
conserved in these types. In type I DHOases, the two Zn 2.1 Plasmid Preparation
ions are bridged by an aspartate which is invariant in this
type. Type II, III and CAD DHOases have a long catalytic Clones of M. jannaschii genomic DNA encoding DHOase
flexible loop that covers the active site when the substrate were obtained from the American Type Culture Collec-
CA is bound. The corresponding loop is short in type I tion (ATCC number 624773). PCR was used to amplify
DHOases. Type I bacterial DHOases are often associated the open reading frame using PfuTurbo DNA polymerase
(Stratagene). The forward primer was GGA​ ATT​CCAT-
ATGCTA​TTA​AAA​AAC​TGT​AGA​AT (NdeI site under-
lined, ORF start codon bolded) and the reverse primer
was GCGGA​TCC​TTAACA​TTT​ACA​TCC​ATA​AGC​TTC
(BamHI site underlined, stop codon bolded). The result-
ing product was subcloned in pCR-Blunt II-TOPO vec-
tor (Invitrogen) and digested with NdeI and BamHI. The
digested fragment was ligated into pET-21a (Novagen)
using T4 DNA ligase (NEB). The ligated mix was trans-
Fig. 1  The reversible cyclization of N-carbamoyl-l-aspartate to
l-dihydroorotate catalyzed by DHOase. Colors used are dark grey for
formed in subcloning efficiency DH5α E. coli competent
C, blue for N and red for O. Figure was prepared with UCSF Chimera cells (Invitrogen). Positive clones were identified by colony

13
Characterization of the Dihydroorotase from Methanococcus jannaschii

PCR and the pET-21a plasmid harboring the M. jannaschii pure protein, as determined by SDS-PAGE, were pooled
pyrC gene was prepared using the QIAprep spin miniprep and concentrated.
kit (Qiagen). The construct was verified by sequencing at All centrifugations were done in in an Avanti J-26XP
the Genomics Core of Cleveland Clinic’s Lerner Research centrifuge and the hydrophobic interaction chromatography
Institute. The plasmid was transformed in Rosetta-gami 2 was performed using an ÄKTAprime system (GE Health-
(DE3) cells (Novagen) for protein expression. care). The purity of the protein throughout the purification
was evaluated using SDS-PAGE.
2.2 Expression of the Mj‑PyrC Gene Product
2.4 Liquid Chromatography–Mass Spectrometry
Analysis
Typically, an overnight culture of Rosetta-gami 2 (DE3)
cells harboring the pET-21a plasmid with the M. jannaschii
The protein was sequenced using LC–MS by the Lerner
pyrC gene in LB Lennox media supplemented with 50 µg/
Research Institute’s Proteomics Laboratory. The LC–MS
ml ampicillin, 34 µg/ml chloramphenicol, 50 µg/ml strep-
system was a Finnigan LTQ linear ion trap mass spec-
tomycin and 12.5 µg/ml tetracycline was used to inoculate
trometer system and the HPLC column was a self-packed
LB Lennox media supplemented with these antibiotics.
9 cm × 75 µm Phenomenex Jupiter C18 reversed-phase cap-
The inoculum was 1% of the media volume. The cells were
illary chromatography column. The sample used for this
grown at 37 °C, 200 rpm to an A­ 600 of ~0.6 in an Innova 44
analysis was an SDS-PAGE gel. The bands to be analyzed
incubator shaker from New Brunswick Scientific. Protein
were excised and digested in-gel with trypsin, chymot-
expression was then induced by the addition of isopropyl-β-
rypsin and endoproteinase GluC overnight at room temper-
d-1-thiogalactopyranoside to a final concentration of 1 mM
ature. The peptides formed in each digestion were analyzed
and the cell culture was grown for 16 additional hours at
by LC–MS. This analysis was done in a data dependent
18 °C. The cells were centrifuged at 4000 rpm for 20 min in
manner acquiring first a full mass scan to determine the
an Avanti J-26XP centrifuge and kept at −80 °C until ready
most abundant peptide ions eluting off the LC column fol-
to use.
lowed by four collision induced dissociation (CID) experi-
ments to determine amino acid sequence in successive
2.3 Purification instrument scans. For each digest, the data were analyzed
by using all CID spectra to search the NCBI non-redundant
Before use, the frozen pellets were thawed and resuspended protein database with the search program Mascot (Matrix
in 0.05 M Tris–Cl pH 7.5 buffer with 2 mM 2-mercaptoe- Science, London, UK) using a mammalian taxonomy fil-
thanol (BME) and 0.05  mM zinc acetate, sonicated, and ter. All matching spectra were verified by manual interpre-
the mixture was centrifuged at 15,000 rpm for 20 min. The tation. The interpretation process was aided by additional
cell supernatant was brought to 35% saturation of ammo- searches using the programs Sequest (Thermo Fisher Sci-
nium sulfate, stirred overnight at 4 °C and centrifuged entific, San Jose, CA, USA) and Blast (http://blast.ncbi.
at 16,000  rpm for 20  min. A heat step of the supernatant nlm.nih.gov). Targeted experiments were also performed
at 85 °C for 15  min followed and the mixture was centri- involving selective reaction monitoring (SRM) for specific
fuged at 16,000 rpm for 20 min. The supernatant was dia- peptides that include residues not determined from the data
lysed in 50  mM Mes–KOH 5.8 buffer with 2  mM BME dependent approach.
and 0.05  mM zinc-acetate followed by cation exchange
chromatography using a hand packed 9 ml SP column (Lab 2.5 Determination of Dihydroorotase Activity
Pack from GE Healthcare) and a gradient from 0 to 0.5 M
NaCl. The fractions containing DHOase were pooled and Dihydroorotase activity was studied in the reverse direc-
dialyzed in 40 mM K ­ H2PO4 pH 8 buffer with 2 mM BME tion in 100 mM Tris–acetate pH 8.3 using the colorimetric
and 0.05 mM zinc acetate. Ammonium sulfate was subse- assay of Prescott and Jones [9] and substrate concentra-
quently added to the protein solution to 1.3 M ammonium tions to 3.15 mM. Readings were made at a wavelength of
sulfate. The final purification step employed hydropho- 466  nm in a SPECTRAmax PLUS 384 microplate reader
bic interaction chromatography using a phenyl-Sepharose from Molecular Devices using the software SoftMax Pro
column [HiPrep Phenyl FF (high Sub) 16/10, GE Health- 6.3. The assay volume was 0.5  ml and the assay was car-
care] that was pre-equilibrated with the same buffer as the ried out at 25, 45 and 80 °C. The reaction was initiated
protein. The gradient employed was 40  mM K ­ H2PO4 pH by addition of the enzyme. The background hydrolysis of
8, 2 mM 2-mercaptoethanol, 0.05 mM zinc acetate, 1.3 M dihydroorotate to carbamoyl aspartate in the absence of
ammonium sulfate to 40 mM K ­ H2PO4, pH 8, 2 mM 2-mer- enzyme was monitored. All measurements were corrected
captoethanol, 0.05  mM zinc acetate. Fractions containing for the background absorbance in the absence of substrate.

13
 J. Vitali et al.

In addition the enzyme kinetic data were corrected for the was used as it proved satisfactory during analyses of 3 pro-
background reaction. tein standards: Aldolase (156  kDa), BSA (66  kDa) and
Kinetic parameters were calculated by fitting data to the Ovalbumin (43 kDa).
Michaelis–Menten equation using the MacDaesX program
by Evan Kantrowitz, Boston College. Protein concentration
2.9 Homology Modeling
was measured by the Bradford assay using the Coomassie
Plus Protein Assay reagent from Pierce with BSA as the
An initial homology model was generated with the
standard.
I-TASSER (Iterative Threading ASSEmbly Refinement)
server [11–14]. The amino acid sequence of the target
2.6 Thermal Stability of Dihydroorotase
was obtained from the ExPASy Bioinformatics resource
portal (http://www.expasy.org) with accession number
The thermal stability of the M. jannaschii DHOase was
UniProtKB: Q58885. I-TASSER identifies structural tem-
measured in the reverse direction by heating the enzyme at
plates from the Protein Data Bank using a multiple thread-
85, 90 and 98 °C as a function of time up to 160  min. At
ing approach and constructs full length atomic models by
each time point an aliquot was rapidly chilled and stored
iterative template fragment assembly simulations. The
at 4 °C followed by determination of enzymatic activity at
closest analog in the Protein Data Bank was identified by
37 °C using the colorimetric assay.
I-TASSER to be the DHOase from B. anthracis (pdb id:
3mpg). The model was examined in Chimera [15] and a
2.7 Zinc Inductively Coupled Plasma‑Mass
few adjustments were made to its sequence alignment with
Spectrometry Measurements
the B. anthracis protein to improve the alignment with
the aspartate bridging the two Zn ions in B. anthracis and
The zinc content of the purified recombinant M. jan-
known to be conserved in type I DHOases [3]. The final
naschii DHOase was obtained using ICP-MS (Intertek
model was created in Chimera using the graphical inter-
Inc., Whitehouse, NJ, USA). Prior to the measurements
face to Modeller [16]. Modeller models protein structure
the purified protein was dialyzed against three changes of
by satisfaction of spatial restraints. The structural template
50 mM Mes–KOH 5.8, 2 mM BME at 4 °C in the presence
was chain A of the B. anthracis DHOase and the revised
of a dialysis bag containing Chelex-100 so that fortuitously
alignment of the two proteins was used. The Zn ions from
bound metals would be removed. The protein concentra-
the template were transferred to the output models. Steric
tion of the dialyzed sample was determined by the Bradford
clashes were identified in the accepted model and relieved
assay. The samples were analyzed using an Agilent 7700X
using the “minimize” option in Chimera. The minimize
ICP-MS instrument. Dublicate ICP-MS analyses of the pro-
option used Amber ff14SB [17] parameters for standard
tein and dialysis buffer solutions were carried out.
residues and Amber’s Antechamber module [18] for non-
standard residues. The stereochemistry was evaluated with
2.8 HPLC Size Exclusion Chromatography/Laser
PROCHECK [19].
Light Scattering

High performance liquid chromatography (HPLC) size

exclusion chromatography/laser light scattering was per- 3 Results
formed by the Biophysics Resource of the Keck Facility
at Yale University [10]. The column used was a Superdex 3.1 Expression and Purification of M. jannaschii
S-200, 10/30, HR SEC (GE Healthcare) and was con- DHOase
nected to an HPLC system (Agilent 1200, Agilent Tech-
nologies). The elution from the column was monitored Methanococcus jannaschii DHOase was previously
by a photodiode array UV/VIS detector (Agilent Tech- sequenced and cloned as part of the study of the genome
nologies), differential refractometer (OPTILab-rEx Wyatt of this organism [20] but it was not expressed or ana-
Corp.), static and dynamic, multiangle laser light scattering lyzed. In this paper we subcloned the M. jannaschii pyrC
detector (HELEOS II with QELS capability, Wyatt Corp.). gene into the pET-21a vector, expressed it in Rosetta-gami
The SEC-UV/LS/RI system was equilibrated in a 50  mM 2 (DE3) E. coli and purified it. At the end of the purifica-
Tris–Cl pH 8, 150 mM NaCl, 2 mM BME, 0.05 mM zinc tion, SDS-PAGE gave a single band (Fig. 2a). Its molecular
acetate and the flow rate was 1.0 ml/min. The sample was weight was determined from a plot of the logarithm of the
filtered through a 0.22 µm filter before injecting to the col- molecular weight of protein standards versus their relative
umn. Weight average molecular masses were calculated by migration distance in the SDS-PAGE gel by interpolation
ASTRA. During data analysis, a dn/dc value of 0.165 ml/g (Fig. 2b). The value obtained of 47,503 Da is comparable

13
Characterization of the Dihydroorotase from Methanococcus jannaschii

Fig. 3  Composite sequence map. Peptides identified in tryptic diges-

tion only are in red, peptides identified in chymotryptic digestion only
are in blue, and peptides identified in GluC digestion only are in pur-
ple. Peptides identified in both tryptic and chymotryptic digestions
are in green. Shaded areas correspond to peptides not identified in the
data dependent analysis of the tryptic and chymotryptic digestions. Of
them, the N-terminal tripeptide was identified in the data dependent
analysis of the GluC digest, Lys137 was identified in low abundant
ions in the SRM analysis of both tryptic and chymotryptic digests and
Cys423 was identified in a low abundant ion in the SRM analysis of
the tryptic digest

acids were not identified in the data dependent analysis of

the tryptic and chymotryptic digests. These were the three
N-terminal amino acids, Lys137, and the C-terminal amino
acid Cys423. The inability to identify these areas could be
due to the formation of tryptic and/or chymotryptic pep-
tides that have poor LC–MS characteristics, the presence of
a protein modification, or their low abundance precluding
Fig.  2  a SDS-PAGE of purified recombinant M. jannaschii DHOase. selection for the collision induced dissociation experiments.
Lane A corresponds to the molecular weight markers, lane B has the A third band from the SDS-PAGE gel was digested
purified protein. The molecular weight markers were the Precision
Plus Protein unstained standards (Biorad). b The corresponding plot
with GluC. The digest was analyzed with LC–MS in a data
of the logarithm of the molecular weight versus relative migration dependent manner. This digest resulted in the identifica-
distance of the standards. Based on this plot, the molecular weight of tion of 17 peptides covering 61% of the protein sequence.
the protein was estimated to be 47,503 Da The three N-terminal amino acids were identified in this
analysis.
to the value of 48,104  Da for the M. jannaschii DHOase In order to determine if the Lys137, N- and C-terminal
computed from its amino acid sequence. peptides were not identified in the data dependent analysis
of tryptic and chymotryptic digests due to their low abun-
3.2 Identification of M. jannaschii Dihydroorotase dance, several SRM experiments were performed. Lys137
was identified in low abundant ions in the SRM analysis
The purified protein was sequenced with LC–MS in order of tryptic and chymotryptic digests. The terminal Cys423
to confirm its identity as M. jannaschii DHOase and to was identified from the SRM analysis of a low abundant
cover as much of the protein sequence as possible. C-terminal tryptic peptide. Several different forms of the
Two bands from the SDS-PAGE gel were initially N-terminal peptide were also targeted in the chymotryptic
digested, one with trypsin and the other with chymotrypsin, digest for SRM analysis but gave no indications of these
and the digests analyzed by LC–MS in a data depend- three N-terminal amino acids. Figure  3 gives a composite
ent manner. M. jannaschii DHOase was identified as the sequence map of the protein indicating the coverage by
major component in both digests. A total of 49 peptides trypsin and chymotrypsin as well as how the remaining five
was identified in the trypsin digest covering 85% of the pro- residues were determined.
tein sequence. The chymotrypsin digest gave 176 peptides Moreover, the ExPASy site suggested that it is possi-
and covered 96% of the protein sequence. The two together ble that Lys137 is carboxylated. In ExPASy, the Zn bind-
covered 99% of the protein sequence. A total of five amino ing site is propagated from the E. coli enzyme [5] to other

13
 J. Vitali et al.

DHOases using sequence alignments with the program

MUSCLE [21]. E. coli [5] has a carboxylated Lys residue
in the metal binding site of the enzyme at position 103 of
its sequence. According to these alignments, this position
corresponds to position 137 in M. jannaschii. However, this
region is highly variable among DHOases [3]. If Lys137
was carboxylated, it would result in the addition of 44 Da
­(CO2) to the lysine residue. However, the SRM analysis of
tryptic and chymotryptic digests was not able to identify
Lys + 44 Da containing peptides suggesting that Lys137 is
unmodified and that, in contrast to E. coli, it may not inter-
act with the Zn ions.

3.3 Steady State Kinetics of the M. jannaschii

Fig. 4  Saturation curves for M. jannascii DHOase in the degradative
Dihydroorotase direction at pH 8.3. Data are averages of four measurements at 25 °C,
three measurements at 45 °C, and two measurements at 80 °C
Table  1 gives the kinetic parameters of M. jannaschii
DHOase in the degradative direction at pH 8.3 at 25, 45
and 80 °C. The saturation curves from which these param- which its activity rapidly diminishes to 29% of its initial
eters were determined were averages of four measurements value over the following 60 min. This is consistent with the
at 25 °C, three measurements at 45 °C, and two measure- observation [22] that there is no growth of M. jannaschii at
ments at 80 °C and are shown in Fig. 4. At all temperatures 95 °C in defined medium.
the enzyme follows Michaelis–Menten kinetics.
3.5 Monomeric Structure of M. jannaschii
3.4 Thermal Stability of the M. jannaschii Dihydroorotase
Dihydroorotase
The SEC-LS/UV/RI data were analyzed with the ASTRA
The thermal stability of the M. jannaschii DHOase was software and suggested that the protein is a monomer in
evaluated at 85, 90 and 98 °C as the optimal growth temper- solution. As seen in Fig. 6, the molecular masses computed
ature of this organism is 85 °C. As seen in Fig. 5, at 85 and in slices of 1  s intervals across the eluting peak between
90 °C, the activity of the enzyme is constant over 160 min. 15.6 and 16.3  ml are nearly constant indicating that the
However at 98 °C, the enzyme is stable for 100  min after protein sample was monodisperse. The weight average

Table 1  Kinetic parameters of Organism Temperature Maximal velocity Km (mM) Typea

M. jannascii DHOase and of (°C) (µmol/min/mg)
selected dihydroorotases in the
degradative direction M. jannaschiib 25 12.2 ± 0.1 0.140 ± 0.009 Archaeal
45 44.7 ± 0.6 0.15 ± 0.01
80 248 ± 5 0.52 ± 0.03
B. anthracisc 25 2.4 ± 0.1 0.114 ± 0.016 Bacterial type I
37 8.6 ± 0.2 0.36 ± 0.06
B. caldolyticusd 70 77.4 ± 4.5 0.195 ± 0.028 Bacterial type I
Hamsterd 37 3.9 ± 0.5 0.022 ± 0.005 CAD
Humane 25 8.20 ± 0.24 0.028 ± 0.004 CAD
E. colif 30 155 ± 2 0.080 ± 0.001 Bacterial type II
a
 Classification from reference [3]
b
 Reported values are averages of four measurements at 25 °C, three measurements at 45 °C, and two meas-
urements at 80 °C.
c
 Values at 25 °C obtained from [7]. Values at 37 °C extracted from Fig. 10 of [25]
d
 Values are from [36]
e
 Values obtained from [3]
f
 Values obtained from [6]

13
Characterization of the Dihydroorotase from Methanococcus jannaschii

Fig. 5  Thermal stability of M. jannaschii DHOase at 85, 90 and Fig. 7  Calibration curve for Zn in ICP-MS analysis used to deter-
98 °C. Solutions of the enzyme were heated at the indicated tempera- mine the Zn content of M. jannaschii DHOase. Blue filled circles
ture in 0.1 M Tris–acetate pH 7.5, 150 mM NaCl, 2 mM BME. At the illustrate the Zn standards used to construct this curve. Red filled
indicated times, samples were removed and immediately placed on squares correspond to the protein samples and yellow filled squares
ice. The colorimetric assay was used to determine enzymatic activity correspond to the buffer samples. Both were diluted by a factor of 10.
at 37 °C in the degradative direction The green filled triangles are bracketing quality controls for the pro-
tein and buffer samples. 1 ppb corresponds to 1 µg/l

calibration curve, and the resulting concentrations were

recorded. These concentrations (Fig. 7) were then corrected
for dilution by a factor of 10. Each analysis was the average
of five replicate readings and each reading was the average
of 100 individual mass sweeps.
The ICP-MS runs gave 752 and 776 µg/l zinc in the
two protein samples of concentration 0.26 mg/ml. The Zn
concentrations obtained for the buffer were 46 and 31 µg/l.
These values give an average of 2.05 mol zinc per mol pro-
tein classifying the M. jannaschii DHOase as a two-zinc
center protein.
Fig. 6  Molecular mass distribution plot from SEC-UV/LS/RI analy-
sis for M. jannaschii DHOase. Molecular masses are plotted as red
filled triangles (for clarity only every 5th measurement of molecular 3.7 A Model of M. jannaschii Dihydroorotase
mass across the eluting peak is plotted); the blue line corresponds to
the UV trace of the protein eluting from the SEC column monitored
at 280 nm
The top I-TASSER model had a satisfactory C-score of
0.90. I-TASSER uses the C-score as the criterion for deter-
mining the best model. I-TASSER inferred the function of
molecular mass was 46,780  Da, in agreement with the the protein to be a dihydroorotase, identified the B. anthra-
theoretical value of 48,104  Da computed from the amino cis protein as the closest structural analog in the Protein
acid sequence, indicating that the protein is monomeric in Data Bank and predicted the α Zn binding site. As the pro-
solution. tein had two Zn ions and did not contain a carboxylated
lysine, the sequence alignment with B. anthracis predicted
3.6 Zinc Content of Purified M. jannaschii by I-TASSER was adjusted to improve the alignment with
Dihydroorotase the aspartate bridging the two Zn ions in the DHOase of
B. anthracis and known to be conserved in type I DHOa-
A set of Zn standards was prepared in an acid matrix and ses [3]. Among the 20 models generated by Modeller in
analyzed by ICP-MS. A calibration curve relating meas- Chimera, the one with the most negative zDOPE (normal-
ured intensity (counts per second) with concentration for ized Discrete Optimized Protein Energy) score of −0.51
Zn in solution was constructed (Fig.  7). The protein and was selected as the best model. The zDOPE score is used
buffer samples were also prepared by acid digestion and by Modeller in Chimera to evaluate the plausibility of a
analyzed under the same conditions as the standards. The model. The Ramachandran plot statistics [19] are satisfac-
sample intensities were measured and compared with the tory for the Chimera model with 98.4% of the residues in

13
 J. Vitali et al.

13
 Characterization of the Dihydroorotase from Methanococcus jannaschii

◂Fig.  8  a A schematic ribbons diagram illustrating the homology and 3 from the C terminal extension βVIII, βXI and βXII
model of the M. jannaschii DHOase superimposed on the enzyme (Fig. 8a, b). The C terminal extension also contributes a β
from B. anthracis (pdb id: 3mpg, chain A). M. jannaschii DHOase
is depicted in rainbow colors, from blue at the N-terminus to red
hairpin (βIX and βX) (Fig. 8b) that protrudes on one side
at the C-terminus. 3mpg, chain A is depicted in dark grey. In this of the TIM barrel toward the top of it (Fig. 8a). The rmsd
view, the β hairpin loop described in the text for M.jannaschii (βIX between corresponding Cα atoms of 410 atom pairs in the
and βX) is in orange in the back of the molecule. The homology two proteins within 2  Å from each other is 0.45  Å. The
model was calculated with Modeller in the UCSF chimera environ-
ment using the dihydroorotase of B. anthracis as the template. A
sequence identity between the two proteins is 32.4%.
stereo pair. Figure was drawn with the UCSF Chimera package. b The active site on the top of the TIM barrel is closely
Sequence alignment of the M. jannaschii (Mj) and B. anthracis (Ba) similar to Ba (Fig.  8c) and Ec (Fig.  8d). The model has
dihydroorotases. The figure was prepared with UCSF Chimera and the two Zn ions coordinated to the invariant His56, His58,
is based on the structural superposition of the two proteins. Identity
between the two sequences is emphasized by blocks drawn at the
His227, His168, Asp302 [3] and they are bridged by
suitable sequence positions. Invariant residues in all dihydroorotases the carboxylate of Asp146. Asp302 is the aspartate that
are highlighted; those that coordinate the two Zn ions in magenta, abstracts a hydrogen from the amide nitrogen to facilitate
those with side chains involved in substrate binding in blue and the the nucleophilic attack. In Ec, the two Zn ions are bridged
rest in green. Cyan corresponds to amino acids that are not con-
served but their positions are conserved and their main chain atoms
with the carboxyl group of the carboxylated Lys103 (E. coli
interact with the substrate. Residues Asp146 in Mj and Asp151 in Ba numbering) [5] and as seen in Fig. 8d, the carboxyl group
are outlined in black. Their carboxylate oxygens bridge the two Zn of the carboxylated Lys103 in Ec (Ec numbering) is at the
ions. An aspartate at this position is conserved in type I DHOases [3]. same position as the carboxyl group of the aspartate in the
The coral region covers corresponding residues in the two proteins
that are within 2  Å from each other. Sequence is colored according
Mj model. Figure 5b, c include also the invariant residues
to the secondary structure: red for α helices, blue for β strands and Arg60, Asn89, His306 that interact with the substrate in
black for loops. The secondary structural elements of the TIM barrel the known DHOases complexes [3, 5, 7, 8]. The substrate
are labeled from β1 to α8. The β strands in the adjacent domain are also interacts with the main chain at positions 275, 320 and
labeled as βI to βXII. Helices are not labeled unless they correspond
to the (βa)8 barrel. The two sequences exhibit 32% identity. c Active
321 (M. jannaschii numbering) [3]. A number of other resi-
site of the Mj model (cyan carbon atoms) superimposed on the active dues in the vicinity of the active site are also conserved.
site of Ba dihydroorotase (pdb id: 3mpg; grey carbon atoms). Zinc These are Pro88, Pro276, Pro305, Met87, Glu170, His257,
atoms are grey spheres. The residues shown are those that coordinate Ala304, Lys311 (M. jannaschii numbering) [3] and their
the Zn ions and those whose side chains are involved in interactions
with the substrate. D146 bridges the two Zn ions in the model. An
role may be to maintain the active site structure. All these
earlier multiple sequence alignment [3] showed that this aspartate in invariant residues are highlighted in Fig. 8b.
conserved in type I DHOases. In the superposition, the two Zn ions
in Mj are slightly displaced from the Zn ions in Ba by ~0.15  Å. d 3.8 Amino Acid Content of M. jannaschii DHOase
Superposition of the active sites of the Mj model (cyan carbon atoms)
and E. coli DHOase (pdb id: 1j79; grey carbon atoms). Zinc atoms
are grey spheres. The residues shown are those that coordinate the Zn Table  2 gives the amino acid composition of Mj and
ions and those whose side chains are involved in interactions with the Ba DHOases computed with the program protparam in
substrate. D146 bridges the two Zn ions in the model and the carbox- ExPASy [23]. As seen in this table, in Mj there is a substan-
ylated lysine KCX103 bridges the two Zn ions in E. coli. An earlier
multiple sequence alignment [3] showed that the carboxylated lysine
tial increase of charged Lys residues and large hydrophobic
is conserved in type II, III and CAD dihydroorotases and is shifted Ile residues and a decrease in Ala, Thr and Gly. The homol-
upstream relative to the conserved aspartate in type I dihydroorotases. ogy model shows that the Ile’s are mostly in the hydropho-
In the superposition, the two Zn ions in Mj are slightly displaced from bic core concealed from the solvent (Fig. 9a) and primarily
the Zn ions in Ec by ~0.22 Å
replace Val and to a lesser extent Ala, Leu and other amino
acids (Fig. 8b). Lys’ are mostly on the exterior of the mol-
the most favored and additional allowed regions and only ecule (Fig. 9b) and replace charged and polar amino acids,
0.5% in the disallowed regions. more than non-polar amino acids (Fig. 8b). Table 2 shows
The architecture of the Mj DHOase model is closely also a decrease in Gln and an increase in Asn residues in
similar to the template Ba, in particular in the secondary Mj compared to the mesophilic Ba DHOase.
structural elements with differences primarily in the loops
that connect them (Fig.  8a, b). Residues 51–340 form the
catalytic (βα)8 barrel and the adjacent β stranded domain is 4 Discussion
formed by residues 1–50 from the N-terminal extension and
residues 341–423 from the C-terminal extension (Fig.  8a, DHOases form a very diverse family of enzymes. Even
b). The outer antiparallel β sheet of the extra domain con- though they catalyze the same reaction, its members differ
sists of the 4 strands βI, βIV, βV and βVI of the N terminal in size, sequence, oligomeric structure, dimerization inter-
extension and the inner antiparallel β sheet consists of 6 faces, Zn content and association or not with ATCase [3].
strands, 3 from the N terminal extension, βII, βIII and βVII The current work is the first study of an archaeal DHOase

13
 J. Vitali et al.

Table 2  Comparison of the amino acid composition of Mj and Ba

DHOases
Mj Ba
Number % Number %

Ala 19 4.50 43 10.00

Arg 8 1.90 15 3.50
Asn 35 8.30 17 4.00
Asp 30 7.10 25 5.80
Cys 9 2.10 9 2.10
Gln 4 0.90 10 2.30
Glu 31 7.30 38 8.90
Gly 23 5.40 36 8.40
His 12 2.80 17 4.00
Ile 51 12.10 30 7.00
Leu 38 9.00 34 7.90
Lys 59 13.90 30 7.00
Met 4 0.90 12 2.80
Phe 18 4.30 14 3.30
Pro 15 3.50 19 4.40
Ser 16 3.80 10 2.30
Thr 12 2.80 31 7.20
Trp 3 0.70 3 0.70
Tyr 10 2.40 5 1.20
Val 26 6.10 30 7.00
Fig. 9  Stereo views. a The increased Ile content of Mj DHOase as
Total 423 428
compared to Ba. The α C backbone of Mj DHOase is shown in gold
and the Ile replacements are shown in light sea green. It can be seen
The amino acid composition was computed with the protparam tool
that most of these replacements are in the interior of the protein. b
in ExPASy [23]. “Number” refers to the number of amino acids of a
The increased Lys content of Mj DHOase as compared to Ba. The
given kind in the protein, and “%” to the corresponding percentage
α C backbone of Mj DHOase is shown in gold and the Lys replace-
ments are shown in blue. It can be seen that most of these residues are
on the surface of the protein
and gives some understanding how this enzyme resembles
and differs from other DHOases.
In the recent multiple sequence alignment [3], this car-
4.1 Structural Aspects boxylated lysine in types II, III and CAD is shifted
upstream two positions relative to the invariant aspartate
Archaeal DHOases are closest to bacterial type I enzymes. in type I.
The latter form dimers and frequently associate with Variations from the Zn—Asp—Zn scheme in type I have
ATCase to form a stoichiometric 1:1 complex e.g. in Aa been previously observed. For instance, Aquifex aeolicus
[24] and Ba [25]. It is assumed [3] that the stereochemistry DHOase has one Zn whether complexed (pdb id: 3d6n)
of the Aa complex [24] is adopted by the other DHOases [24] or uncomplexed (pdb id: 1xrf and 1xrt) [27] with
that associate with ATCase. In contrast, the Mj enzyme is a ATCase. The Staphylococcus aureus enzyme (pdb id: 3gri)
monomer in solution. In addition, a complex with ATCase also has one Zn center bridged to the conserved Asp via a
is not possible as M. jannaschii has both pyrI and pyrB water molecule. These alternate active sites suggest analo-
genes and their products associate to form a type B ATCase gous but somewhat different mechanisms. For instance,
holoenzyme [26]. theoretical calculations for A. aeolicus DHOase indicate
The active site has two Zn ions as most DHOases [3]. that a water in the crystal at the position of the β Zn low-
These are bridged by an aspartate Asp146 which is at the ers the activation energy of the reaction [8]. These findings
same position as the conserved aspartate in type I bridg- question the invariance of a two Zn center for DHOases.
ing the two Zn ions (Fig. 8c). This differentiates Mj and There has been recent discussion on the viability of the
type I DHOases from types II, III and CAD that have a observed mono-zinc active site in some type I DHOases.
carboxylated lysine bridging the two Zn ions (Fig.  8d) Grande-Garcia et  al. [3] suggested that all DHOases have
[3]. The modified Lys is conserved within these groups. a two-Zn center. They contend that the type I structures

13
Characterization of the Dihydroorotase from Methanococcus jannaschii

observed with one Zn-center in the crystalline state may are expected to be comparable [34]. The large difference in
represent an incomplete view of the active site where the the specific activities between Mj and Ba at 80 and 37 °C,
β Zn ion is not bound or binds with partial occupancy. In could be attributed to subtle differences in the vicinity of
these cases, it is possible that the more solvent-exposed β their binding sites. This explanation has been invoked for
Zn could dissociate during purification and crystallization. pairs of homologous proteins that have dissimilar activi-
This idea was supported by the fact that the β-Zn ligands ties at their physiological temperatures [34]. It was also
are conserved in the known structures and by the incorrect suggested [33] that these changes must alter the function
estimation of the Zn content in E. coli and human DHOa- by changing the mobility of the structures. Similar differ-
ses. An initial analysis of both found only one zinc but ences were found between the thermophilic bacterium B.
further study showed two. Typically, dialysis is performed caldolyticus and mesophilic hamster DHOase at their cor-
to remove fortuitously bound metal [28–30] before the responding physiological temperatures in [36] (Table 1).
analysis which could explain the loss of the second zinc. At 25 °C, the specific activity of Mj is higher than in Ba
However, the existence of fully functional DHOases with (Table  1). This observation may be unexpected because
a mononuclear metal center is supported by the fact that hypethermophilic enzymes are believed to have rigid struc-
deletion of the β-site in A. aeolicus DHOase [8] by replac- tures at ambient temperatures and catalysis requires flex-
ing the two histidines binding the β Zn with Ala affected ibility [34]. However, Elias et al. [35] discuss that the rigid-
neither catalysis nor active site structure. However, when ity of the fold of an enzyme does not necessarily affect the
the second metal binding site in amidohydrolases with a flexibility of its active site or its activity. Furthermore, there
binuclear metal center was eliminated, all catalytic activ- are reports of thermophilic enzymes that are as or more
ity ceased [31, 32]. Nonetheless, we have unambigiously active than their mesophilic counterparts [35] as it happens
shown that M. jannaschii DHOase contains two Zn ions. in Mj, at all temperatures. The specific activity of Mj is
Whether other archaeal DHOases have one or two Zn cent- higher than all the mesophilic proteins in Table 1.
ers remains to be seen.
4.3 Molecular Basis of Thermostability
4.2 Kinetics of M. jannaschii DHOase
The comparison of the amino acid compositions and
The kinetic studies of Mj DHOase in Fig.  4 and Table  1 sequences of the Mj and Ba dihydroorotases in Sect. 3.8
show that its specific activity increases with temperature, (Table  2; Fig.  8b) gave insight into some of the factors
from 12.2 µmol/min/mg at 25 °C to 248 µmol/min/mg at that may confer stability in the Mj enzyme. The amino
80 °C, which is a common feature to all enzymes. The rea- acid substitutions and their locations in Mj are consistent
son is that the increase in temperature increases the number with thermostabilization strategies adopted by hyperther-
of enzyme molecules with enough energy to undergo the mophilic enzymes [34, 37–39]. Thermostable enzymes
necessary catalytic conformational changes [33–35]. The typically have more ion pairs (salt bridges), especially
increase in K­ m with temperature (0.52 mM at 80 °C as com- in networks, more hydrophobic interactions in their inte-
pared to 0.14 mM at 25 °C) (Table 1; Fig. 4) in this organ- rior, and a larger number of bulkier hydrophobic side
ism is also consistent with observations in other proteins chains as compared to their mesophilic counterparts. It
[33], and indicates an increase in the molecular flexibility was also shown [40, 41] that hyperthermophilic archaea
of the active site. The increase in specific activity with tem- substitute non-polar amino acids with Ile and have an
perature resembles the related catalytic trimer of ATCase increased Lys content from their corresponding meso-
from the same organism [26]. The fold-increase in turnover philic homologues. Furthermore, charged residues are
rate every 10 °C increase in temperature (Q10) based on the overrepresented on the surface [40, 41] and it was sug-
specific activity was calculated to be 1.73 between 25 and gested [40] that a large number of exposed charged amino
80 °C for the Mj DHOase. This fits nicely within the range acids stabilize proteins at high temperatures by forming
of Q10 values observed in thermophilic enzymes of 1.2 to extended ion-pair networks. Such ion pairs and intercon-
2.3 based on specific activity [35]. Both K ­ m and specific nected salt bridge networks on the surface of hyperther-
activity values in the mesophilic counterpart from Ba show mophiles have been demonstrated [34]. The decrease in
also an increase with temperature (Table 1). Gln in Mj is consistent with other thermostable enzymes
Close to their corresponding physiological tempera- [37, 40]. However, the Asn content in hyperthermophilic
tures (85 °C for Mj and 37 °C for Ba) the activity of Mj is archaea is comparable to their mesophilic homologues
much higher than the activity of its mesophilic homolog and the reason for its increase in Mj is not clear. In addi-
(Table  1). This is in contrast to expectations as the active tion, shorter N- and C-termini increase thermostability
sites of the two are very similar and the flexibilities of [34, 37] and, as seen in Fig. 8a, b, both N- and C-termini
homologous proteins at their physiological temperatures are shorter in M. jannaschii as compared to B. anthracis.

13
 J. Vitali et al.

It may be noted that the close agreement in the secondary 6. Porter TN, Li Y, Raushel FM (2004) Mechanism of the dihy-
structure of the enzymes in M. jannaschii and B. anthra- droorotase reaction. Biochemistry 43:16285–16292
7. Rice AJ, Lei H, Santarsiero BD, Lee H, Johnson ME (2016)
cis indicate that the protein core of the mesophilic protein Ca-asp bound X-ray structure and inhibition of Bacil-
is already optimally stable. Some of the thermostabiliza- lus anthracis dihydroorotase (DHOase). Bioorg Med Chem
tion strategies of DHOase in M. jannaschii are similar to 24:4536–4543
those adopted by the catalytic subunit of ATCase in the 8. Edwards BF, Fernando R, Martin PD, Grimley E, Cordes
M, Vaishnav A, Brunzelle JS, Evans HG, Evans DR (2013)
same organism [42]. The mononuclear metal center of type-I dihydroorotase from
Additional and better insight on the stereochemistry of Aquifex aeolicus. BMC Biochem 14:1
Mj DHOase and its thermostabilization strategies will be 9. Prescott LM, Jones ME (1969) Modified methods for the deter-
provided when its crystal structure is determined. In addi- mination of carbamyl aspartate. Anal Biochem 32:408–419
10. Folta-Stogniew E, Williams KR (1999) Determination of

tion, more data on other archaeal DHOases are required molecular masses of proteins in solution: implementation of an
to generalize the differences and similarities between HPLC size exclusion chromatography and laser light scattering
archaeal DHOases and the other types. service in a core laboratory. J Biomol Tech 10:51–63
11. Yang J, Yan R, Roy A, Xu D, Poisson J, Zhang Y (2015) The
Acknowledgements  This work was supported in part by grant I-TASSER suite: protein structure and function prediction. Nat
GM071512 (JV) from the National Institutes of Health. Molecular Methods 12:7–8
graphics and analyses were performed with the UCSF Chimera pack- 12. Yang J, Zhang Y (2015) I-TASSER server: new development
age. Chimera is developed by the Resource for Biocomputing, Visu- for protein structure and function predictions. Nucleic Acids
alization, and Informatics at the University of California, San Fran- Res 43:W174–W181
cisco (supported by NIGMS P41-GM103311). We thank Dr. Belinda 13. Zhang Y (2008) I-TASSER server for protein 3D structure pre-
Willard of the Lerner Research Institute in Cleveland Clinic for the diction. BMC Bioinform 9:40-2105-9-40
LC–MS, Mr. Michael Murphy of Intertek Chemicals and Pharma- 14. Roy A, Kucukural A, Zhang Y (2010) I-TASSER: a unified
ceuticals for the ICP-MS and Dr. Ewa Folta-Stogniew of the Bio- platform for automated protein structure and function predic-
physics Resource of the Keck Facility at Yale for the SEC-LS. The tion. Nat Protoc 5:725–738
SEC-LS/UV/RI instrumentation was supported by NIH Award Num- 15. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt
ber 1S10RR023748-01. We also thank Dr. Bin Su of Cleveland State DM, Meng EC, Ferrin TE (2004) UCSF Chimera—a visuali-
University for use of his SPECTRAmax PLUS 384 microplate reader zation system for exploratory research and analysis. J Comput
and Drs. Barbara Zimmermann of Los Andes University and Evan Chem 25:1605–1612
Kantrowitz of Boston College for enlightening discussions. 16. Sali A, Blundell T (1994) Comparative protein modelling by
satisfaction of spatial restraints. Protein Struct Distance Anal
Compliance with Ethical Standards  64:C86
17. Maier JA, Martinez C, Kasavajhala K, Wickstrom L, Hauser
KE, Simmerling C (2015) ff14SB: improving the accuracy of
Conflict of interest  The authors declare that they have no conflicts protein side chain and backbone parameters from ff99SB. J
of interest. Chem Theory Comput 11:3696–3713
18. Wang J, Wang W, Kollman PA, Case DA (2006) Automatic
Ethical Approval  This article does not contain any studies with atom type and bond type perception in molecular mechanical
human participants or animals performed by any of the authors. calculations. J Mol Graph Model 25:247–260
19. Laskowski RA, Rullmann JAC, MacArthur MW, Kaptein R,
Thornton JM (1996) AQUA and PROCHECK-NMR: programs
for checking the quality of protein structures solved by NMR. J
References Biomol NMR 8:477–486
20. Bult CJ, White O, Olsen GJ, Zhou L (1996) Complete genome
1. Holm L, Sander C (1997) An evolutionary treasure: unification sequence of the methanogenic archaeon, Methanococcus jan-
of a broad set of amidohydrolases related to urease. Proteins naschii. Science 273:1058
28:72–82 21. Edgar RC (2004) MUSCLE: multiple sequence alignment

2. Fields C, Brichta D, Shepherdson M, Farinha M, O’Donovan with high accuracy and high throughput. Nucleic Acids Res
G (1999) Phylogenetic analysis and classification of dihy- 32:1792–1797
droorotases: a complex history for a complex enzyme. Paths 22. Jones W, Leigh JA, Mayer F, Woese C, Wolfe R (1983) Metha-
Pyrimidines 7:49–63 nococcus jannaschii sp. nov., an extremely thermophilic metha-
3. Grande-García A, Lallous N, Díaz-Tejada C, Ramón-Maiques nogen from a submarine hydrothermal vent. Arch Microbiol
S (2014) Structure, functional characterization, and evolu- 136:254–261
tion of the dihydroorotase domain of human CAD. Structure 23. Gasteiger E, Hoogland C, Gattiker A, Duvaud S, Wilkins MR,
22:185–198 Appel RD, Bairoch A (2005) Protein identification and analysis
4. Kim GJ, Kim HS (1998) Identification of the structural simi- tools on the ExPASy server. In: Walker JM (ed) The proteomics
larity in the functionally related amidohydrolases acting on the handbook. Humana Press, Totowa, pp 571–607
cyclic amide ring. Biochem J 330(Pt 1):295–302 24. Zhang P, Martin PD, Purcarea C, Vaishnav A, Brunzelle JS,
5. Thoden JB, Phillips GN Jr, Neal TM, Raushel FM, Holden Fernando R, Guy-Evans HI, Evans DR, Edwards BF (2009)
HM (2001) Molecular structure of dihydroorotase: a paradigm Dihydroorotase from the hyperthermophile Aquifiex aeolicus is
for catalysis through the use of a binuclear metal center. Bio- activated by stoichiometric association with aspartate transcarba-
chemistry 40:6989–6997 moylase and forms a one-pot reactor for pyrimidine biosynthesis.
Biochemistry 48:766–778

13
Characterization of the Dihydroorotase from Methanococcus jannaschii

25. Kankanala R (2011) Characterizing the oligomeric structure and 34. Feller G (2010) Topical review: protein stability and enzyme
catalytic activity of the dihydroorotase and aspartate transcarba- activity at extreme biological temperatures. J Phys Condens Mat-
moylase from the bacterium, Bacillus anthracis. Master’s Theses ter 22:323101
and Doctoral Dissertations. Paper 339. http://commons.emich. 35. Elias M, Wieczorek G, Rosenne S, Tawfik DS (2014) The uni-
edu/theses/339 versality of enzymatic rate–temperature dependency. Trends
26. Hack ES, Vorobyova T, Sakash JB, West JM, Macol CP, Herve Biochem Sci 39:1–7
G, Williams MK, Kantrowitz ER (2000) Characterization of the 36. Huang DT, Kaplan J, Menz RI, Katis VL, Wake RG, Zhao F,
aspartate transcarbamoylase from Methanococcus jannaschii. J Wolfenden R, Christopherson RI (2006) Thermodynamic analy-
Biol Chem 275:15820–15827 sis of catalysis by the dihydroorotases from hamster and Bacillus
27. Martin PD, Purcarea C, Zhang P, Vaishnav A, Sadecki S, Guy- caldolyticus, as compared with the uncatalyzed reaction. Bio-
Evans HI, Evans DR, Edwards BF (2005) The crystal structure chemistry 45:8275–8283
of a novel, latent dihydroorotase from Aquifex aeolicus at 1.7 Å 37. Vieille C, Zeikus GJ (2001) Hyperthermophilic enzymes:

resolution. J Mol Biol 348:535–547 sources, uses, and molecular mechanisms for thermostability.
28. Washabaugh MW, Collins KD (1984) Dihydroorotase from
Microbiol Mol Biol Rev 65:1–43
Escherichia coli. Purification and characterization. J Biol Chem 38. Petsko GA (2001) Structural basis of thermostability in hyper-
259:3293–3298 thermophilic proteins, or ‘there’s more than one way to skin a
29. Zimmermann BH, Evans DR (1993) Cloning, overexpression, cat’. Methods Enzymol 334:469–478
and characterization of the functional dihydroorotase domain 39. Ding Y, Cai Y, Han Y, Zhao B (2012) Comparison of the struc-
of the mammalian multifunctional protein CAD. Biochemistry tural basis for thermal stability between archaeal and bacterial
32:1519–1527 proteins. Extremophiles 16:67–78
30. Kelly RE, Mally MI, Evans DR (1986) The dihydroorotase
40. Mizuguchi K, Sele M, Cubellis MV (2007) Environment specific
domain of the multifunctional protein CAD. Subunit structure, substitution tables for thermophilic proteins. BMC Bioinform
zinc content, and kinetics. J Biol Chem 261:6073–6083 8(Suppl 1):S15
31. Wang C, Tsau H, Chen W, Huang C (2010) Identification and 41. Saelensminde G, Halskau Ø, Helland R, Willassen NP, Jonas-
characterization of a putative dihydroorotase, KPN01074, from sen I (2007) Structure-dependent relationships between growth
Klebsiella pneumoniae. Protein J 29:445–452 temperature of prokaryotes and the amino acid frequency in their
32. Ho Y, Huang Y, Huang C (2013) Chemical rescue of the post- proteins. Extremophiles 11:585–596
translationally carboxylated lysine mutant of allantoinase and 42. Vitali J, Colaneri MJ, Kantrowitz E (2008) Crystal structure of
dihydroorotase by metal ions and short-chain carboxylic acids. the catalytic trimer of Methanococcus jannaschii aspartate tran-
Amino Acids 44:1181–1191 scarbamoylase. Proteins 71:1324–1334
33. Fields PA (2001) Review: protein function at thermal extremes:
balancing stability and flexibility. Comp Biochem Physiol A
129:417–431

13