(Received for publication, January 21, 1997)
From the CTP-phosphoethanolamine cytidylyltransferase (ET)
is the enzyme that catalyzes the formation of CDP-ethanolamine in the
phosphatidylethanolamine biosynthetic pathway from ethanolamine. We
constructed a Saccharomyces cerevisiae mutant of which the
ECT1 gene, putatively encoding ET, was disrupted. This
mutant showed a growth defect on ethanolamine-containing medium and a
decrease of ET activity. A cDNA clone was isolated from a human
glioblastoma cDNA expression library by complementation of the
yeast mutant. Introduction of this cDNA into the yeast mutant
clearly restored the formation of CDP-ethanolamine and phosphatidylethanolamine in cells. ET activity in transformants was
higher than that in wild-type cells. The deduced protein sequence exhibited homology with the yeast, rat, and human CTP-phosphocholine cytidylyltransferases, as well as yeast ET. The cDNA gene product was expressed as a fusion with glutathione S-transferase in
Escherichia coli and shown to have ET activity. These
results clearly indicate that the cDNA obtained here encodes human
ET.
Phosphatidylethanolamine (PtdEtn)1 is
a major membrane constituent of both prokaryotic and eukaryotic cells.
In mammalian cells, this phospholipid can be synthesized de
novo via the CDP-ethanolamine pathway, by decarboxylation of
phosphatidylserine, and by the calcium-dependent exchange
of ethanolamine with the base moiety of pre-existing phospholipids (1).
Although the exact contributions of each of these pathways remain to be
established, the CDP-ethanolamine pathway probably plays an important
role in the synthesis of PtdEtn (2). CTP-phosphoethanolamine
cytidylyltransferase (ET; EC 2.7.7.14) catalyzes the synthesis of
CDP-ethanolamine from CTP and phosphoethanolamine, and is considered to
be an important regulatory enzyme in the CDP-ethanolamine pathway
through analogy with the CDP-choline pathway. However, little is known
about the regulation of its activity.
CTP-phosphocholine cytidylyltransferase (CT; EC 2.7.7.15) is a key
regulatory enzyme in the CDP-choline pathway. The properties of CT have
been extensively studied, and the role of CT in the regulation of
phosphatidylcholine (PtdCho) biosynthesis has been well established
(3). The sequences have been determined for several mammalian CT
cDNAs (4-7) and the gene from Saccharomyces cerevisiae
(8). CT exists in an inactive soluble form and an active membrane-bound
form. Regulation of the translocation process by changes in the lipid
composition of cellular membranes seems to be one of the major
mechanisms for the control of CT activity. Phosphorylation and
dephosphorylation of the enzyme are also correlated to the
translocation of CT to membranes; the soluble form is highly phosphorylated, and translocation of CT to membranes is accompanied by
extensive dephosphorylation. Multiple phosphorylation sites at the C
terminus have been identified (9, 10). CT is thought to have a
bipartite structure composed of a globular N-terminal catalytic domain
and an extended C-terminal domain. Between the catalytic domain and the
phosphorylation region is a sequence that is predicted to form
amphipathic Unlike for CT, the protein structure and regulatory properties of ET
have been studied much less extensively. ET has been purified
1,000-fold from a post-microsomal fraction of rat liver by Sundler
(13). van Golde et al. recently purified ET to homogeneity from rat liver (14). The purified protein has a molecular weight of
about 50,000 and appears to be a dimer. ET has long been tacitly assumed to be regulated like CT. However, there is accumulating evidence that the control mechanisms for the ET and CT activities are
different (15).
Only sequence information for the yeast S. cerevisiae
gene putatively encoding ET is available. Yeast cells possess
phospholipid-synthetic pathways quite similar to those found in other
eukaryotic cells. Yeast gene
ECT1,2 originally known as
MUQ1 (16), encodes a protein similar to CTs of yeast and
other species and, therefore, is thought to encode ET. Here, we report
the isolation and characterization of a human cDNA encoding ET by
genetic complementation of a yeast mutant defective in
ECT1.
Saccharomyces
cerevisiae DS15-3 (MAT The yeast ECT1 gene was
amplified from chromosomal DNA by the polymerase chain reaction (PCR).
The PCR primers used were
5 The 0.4-kbp
EcoRV/XhoI fragment of pUC-ECT located within the
ECT1 coding frame was replaced with a 1.3-kbp
HincII/XhoI fragment containing yeast
HIS3 to yield pUC-ect::HIS3. The 1.9-kbp
SphI/SalI fragment of this plasmid was used for
the transformation of DS15-3. His+ colonies were selected.
One disruptant strain, of which the gene disruption was confirmed by
Southern blot analysis, was designated as NA9 and used for the
isolation of human cDNA.
cDNA Cloning DNA was sequenced
manually using the dideoxynucleotide chain termination method (21) with
Klenow fragment (Takara Shuzo) and [32P]dCTP (Amersham
Corp.) on the M13 phage vector. The primers were either universal or a
series of oligonucleotides synthesized for the partial cDNA
sequence. Amino acid sequence similarities were found using the BLAST
program, and the translation of nucleotide sequences in the GenBank and
EMBL data bases. Alignments were further improved by eye. Only amino
acid identities were considered.
An adult human multitissue Northern
blot (Clontech) was hybridized with a random primed (Takara Shuzo)
32P-labeled entire cDNA fragment as a probe.
Hybridization was carried out at 42 °C overnight in a hybridization
solution containing 50% formamide, 50 mM sodium phosphate
(pH 6.5), 5 × SSC (1 × SSC: 0.15 M NaCl, 15 mM sodium citrate, pH7.0), 1 × Denhardt's (0.1% each of bovine serum albumin, polyvinylpyrrolidone, and Ficoll), and
0.1% SDS. The filter was washed twice at room temperature with 2 × SSC containing 0.1% SDS for 10 min, and then twice with 0.1 × SSC containing 0.1% SDS at 50 °C for 15 min.
Exponentially growing yeast cells were harvested by
centrifugation, washed with M-N medium, and then resuspended in the
same medium. To 1 ml of culture of each strain was added 7.4 kBq of [1,2-14C]ethanolamine and 0.4 µmol of ethanolamine, and
then the mixture was incubated at 30 °C. At the indicated times, 5 ml of 10% trichloroacetic acid was added to the culture. The mixture
was then filtered through GF/C glass-microfiber paper (Whatman). The
filter paper was washed three times with 5 ml of 10% trichloroacetic
acid and dried, and then radioactivity was counted after treatment with
Scintilamine-OH (Dojindo) as described previously (21). The uptake of
the labeled precursors into cells during the above incubation was
determined by counting the radioactivity in cells collected by
filtration through GF/C glass-microfiber paper as described previously
(22).
For the lipid analysis, cells were cultured under the specified
conditions, collected by centrifugation, and then suspended in M-N
medium. To 2 ml of cell suspension was added
[1,2-14C]ethanolamine or
[methyl-14C]choline (37 kBq each). After the
incubation at 30 °C for 60 min, cells were collected by
centrifugation, suspended in 100 µl of methanol, and then sonicated.
To the sonicated sample was added 200 µl of chloroform to extract the
lipids. The extract was washed once with 300 µl of saline, twice with
300 µl of methanol-saline (1:1, v/v), and then evaporated. Lipids
were dissolved in 50 µl of chloroform and chromatographed on a Silica
Gel 60 plate (Merk) with chloroform-methanol-acetic acid-formic
acid-water (35:15:6:2:1, by volume) as a developing solvent.
Radioactive materials were located by autoradiography.
For the analysis of water-soluble metabolites, 1 ml of a culture of
cells in M-N medium was incubated with 37 kBq of
[1,2-14C]ethanolamine for 60 min, and then the cells
were collected by centrifugation and suspended in 50 µl of 10%
trichloroacetic acid. After standing on ice for 20 min, the insoluble
materials were removed by centrifugation and the supernatant was
subjected to analysis by thin-layer chromatography on a Silica Gel 60 plate with ethanol and 2% ammonia (1:1, v/v). Radioactive materials were located by autoradiography.
A SmaI/EcoRI fragment excised from the
cDNA clone was ligated between the SmaI and
EcoRI sites of pGEX-2T (Pharmacia Biotech Inc.). The
plasmid, pGEX-hECT, thus obtained was used for expression in E. coli. Briefly, overnight cultures of E. coli JM103
harboring the pGEX-hECT plasmid were diluted in fresh medium and then
grown for another 2 h at 37 °C. Protein expression was induced
with 1 mM
isopropyl-1-thio- ET activity of fusion protein was determined in a reaction mixture
comprising 20 mM Tris-Cl (pH 7.5), 10 mM
MgCl2, 2 mM CTP, 5 mM
dithiothreitol, [1,2-14C]phosphoethanolamine, and
E. coli cell extract, in a total volume of 25 µl. After 30 min of incubation at 30 °C, the reaction was stopped by heating at
100 °C for 5 min. The reaction product was separated by thin-layer
chromatography on a Silica Gel 60 plate as described above. Protein
determination was carried out with BCA protein assay reagent
(Pierce).
Yeast cells were cultured in a minimal medium
on a Bio-Shaker BR-30 (Taitec) to the early stationary phase at
30 °C, harvested, and disrupted by vortexing vigorously with glass
beads (diameter, 0.3 mm) in 10 mM Tris-Cl containing 1 mM EDTA and 20% glycerol for 5 min. Disrupted cells were
centrifuged at 4,000 × g for 10 min, and the
supernatant fraction was used for the determination of ET and CT
activities as described previously (22) except that the total volume
was reduced to 50 µl. All enzyme reactions were carried out at
25 °C for 10 min or 20 min.
Restriction endonucleases and other nucleic
acid-modifying enzymes were purchased from Takara Shuzo. Taq
DNA polymerase was obtained from Boehringer Mannheim.
[ Based on the structural similarity, the S. cerevisiae ECT1 gene was thought to encode ET (16). To confirm
this, we constructed an yeast strain of which the ECT1 gene
was disrupted. Although mutants carrying the ect1 mutation
alone could exhibit no growth phenotype, the combination of the
ect1 and ise mutants allowed us to detect the
phenotype of the ect1 mutation. The ise mutation is a conditional choline- or ethanolamine-auxotrophic mutant of which
the growth is inhibited by high concentrations of inositol (18). As was
expected, an ect1 ise double mutant, NA9, we constructed did
not grow on inositol- and ethanolamine-containing media, although it
grew normally on inositol- and choline-containing media or on
inositol-free media. The mutant also showed the decreased levels of the
synthesis of CDP-ethanolamine and PtdEtn as well as low levels of ET
activities (see below). These results strongly indicate that the
ECT1 encodes ET.
Using the ect1 ise mutant, we next attempted to clone
the human cDNA for ET by the complementation method. Strain NA9 was transformed with a human cDNA expression library, and the colonies that grew on a minimal medium supplemented with inositol and
ethanolamine but not on a medium supplemented with inositol alone were
selected. Five clones were obtained. Based on their restriction maps,
the obtained clones could be classified into two groups: four clones (pH1-pH4) and one clone (pS4a). Fig. 1 shows the growth
phenotype of the yeast mutant transformed with the cloned plasmids.
Plasmids pH4 and pS4a clearly suppress the growth defect of strain NA9 on inositol- and ethanolamine-containing medium, although the suppression by pH4 is slightly weaker. Partial sequence analysis revealed that plasmid pH4 contains the cDNA for human choline kinase, which has already been isolated (26). On the other hand, plasmid pS4a was found to have a new cDNA. We therefore sequenced the entire region of the new cDNA.
A restriction map
of the cDNA insert of pS4a is shown in Fig.
2A. The nucleotide sequence of the cDNA
was determined on both strands. The cDNA insert consisted of a
total of 1,856 bases, with a single long open reading frame (ORF)
starting with an ATG codon at position 67 and ending at position 1,236 (Fig. 2B). Although it was not known whether translation
started at the first methionine shown in Fig. 2B or at a
further upstream one, we predicted that the first methionine shown in
Fig. 2B is the likely translation start site, as the
sequence flanking it conforms to the consensus sequence for translation
initiation proposed by Kozak (27), (A/G)XXATG(A/G)X(C/T). This was supported by
sequence comparison with the yeast ET (see below). The ORF encoded a
protein of 389 amino acids with a calculated molecular mass of 43.8 kDa. As was expected, the reading frame is in-frame with the
ADH1 N-terminal region of the expression vector, pADANS. The
consensus site for polyadenylation, 5
The deduced amino acid sequence for the
cDNA coding region was used to search a number of data bases
provided by the National Institute of Genetics, Japan. The cDNA
gene product was found to be similar to several cytidylyltransferases
and found to be most similar (36% identity) to the S. cerevisiae ET (Fig. 3A). The homology
extends across the entire lengths of both proteins. It is also
noteworthy that both proteins have large internal repeated sequences
(Fig. 3B). On the other hand, the sequence similarity between the cDNA gene product and CTs from several species,
including from humans, was limited to their N-terminal halves. Limited
sequence similarity was also observed at the C termini of human and
yeast ET, and yeast CT, but not at the C terminus of human CT (Fig. 3C). These results strongly suggested that the cDNA
obtained here is for human ET. We describe below biochemical evidence
demonstrating that this is indeed the case.
Northern analysis was carried out with a
variety of adult human tissues. For this study the entire 2-kbp human
cDNA fragment from pS4a was used as a probe. As shown in Fig.
4, a major transcript of approximately 2 kilobases was
detected among the RNAs from the several human tissues examined. The
size of the transcript is consistent with that of the cDNA
obtained. The transcript is very abundant in liver, heart, and skeletal
muscle tissues.
To confirm
that plasmid pS4a contains the cDNA for human ET, we examined
whether or not the cDNA could restore the activity of the
CDP-ethanolamine pathway in the yeast strain. The flux of ethanolamine
through the ET step was analyzed by determining the incorporation of
[14C]ethanolamine into PtdEtn. The overall rate of PtdEtn
synthesis was determined by measuring the radioactivity in lipids. As
shown in Fig. 5, the incorporation of radioactivity in
the ect1 mutant was very low. On the other hand, the
incorporation of radioactivity into lipids in the mutant strain
harboring plasmid pS4a had recovered to 81% (at 60 min) of that in the
wild type. It was also shown that the introduction of the human
cDNA for choline kinase (plasmid pH4) restored the incorporation of
radiolabeled ethanolamine into lipids, although the extent was slightly
low (43% of that in the wild type at 60 min).
We next analyzed the radioactive lipids derived from
14C-labeled ethanolamine and choline. Cells were cultured
with labeled compounds for 60 min and lipids were extracted and
analyzed by thin-layer chromatography. As shown in Fig.
6, the amount of labeled PtdEtn and PtdCho derived from
[14C]ethanolamine was very low in the mutant cells
(lane 5) when compared with those in wild-type cells
(lane 4). Introduction of human cDNA into the yeast
mutant clearly restored the synthesis of PtdEtn and PtdCho from
[14C]ethanolamine (lane 6). In contrast, there
were no significant differences in the amount of labeled PtdCho derived
from [14C]choline between these strains (lanes
1-3).
We next examined the incorporation of [14C]ethanolamine
into the intermediates of the CDP-ethanolamine pathway. The
ect1 mutant harboring vector plasmid pADANS and plasmid
pS4a, as well as the wild-type strain harboring pADANS were grown at
30 °C in the presence of [14C]ethanolamine for 60 min.
Trichloroacetic acid-soluble materials were analyzed by thin-layer
chromatography. As shown in Fig. 7, mutant cells
accumulated much higher levels of radioactive phosphoethanolamine than
wild-type cells. Conversely, the level of radioactive CDP-ethanolamine was very low in the mutant. Introduction of the human cDNA into the
yeast mutant clearly restored the level of CDP-ethanolamine. These
results strongly indicate that the cDNA product has ET activity. Further evidence supporting this conclusion was obtained by assaying the ET activity of the cDNA product.
ET and CT activities in the transformant
were determined and compared with those in the mutant and wild-type
cells. As shown in Table I, ect1 mutant
strain, NA9 having a vector alone, had a very low ET activity, when
compared with wild-type strain, DS15-3. On the other hand, the
introduction of human cDNA on a multicopy vector into the mutant
cell clearly restored the ET activity. The activity was higher than
that in wild-type cells. This might be the results of gene dosage
effect. CT activity was measured as a control, and we found no
significant differences in the enzyme activities between these
strains.
ET and CT activities in wild-type, mutant, and transformant cells
Department of Biochemical Engineering and
Science,
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENT
REFERENCES
-helices and to interact with the membrane bilayer of
activating phospholipids (11, 12).
Strain, Media, and Growth Conditions
ise leu2 his3) was used as
the wild-type strain and as the recipient for ECT1
disruption. DS15-3 is a segregant of the diploid strain, DS15 (17).
The ise mutant is a conditional ethanolamine auxotroph whose
growth is inhibited by high concentrations of inositol, and the growth defect is suppressed by the addition of ethanolamine to the culture medium (18). Yeast cells were cultured in either YPD or synthetic minimal medium at 30 °C. The compositions of the YPD and
inositol-free minimal media were reported previously (18). The
composition of nitrogen-free minimal medium (M-N) is given in Ref. 19.
Inositol, L-leucine and L-histidine were each
added to the culture media at the concentration of 20 µg/ml.
Ethanolamine was added to the culture media at the concentration of 10 µg/ml.
-TCTAGATGACGGTAAACTTAGATC-3
(forward primer) and
5
-ATGTAATACCAGTTTTCTTA-3
(reverse primer). The
complementary oligonucleotides contained additional sequences
(underlined), so there was a BamHI restriction site at the
5
end and a SalI site at the 3
end of the PCR products. An
about 1-kbp PCR product was digested with BamHI and
SalI, purified by gel electrophoresis, and then inserted
between the BamHI and SalI sites of pUC18 to
yield pUC-ECT. By comparing the restriction maps of the isolated clone
and the reported one, and partial sequencing of the cloned fragment,
plasmid pUC-ECT was confirmed to contain the yeast ECT1
gene.
A human cDNA library made from
glioblastoma cells was constructed on a yeast multicopy vector, pADANS,
which contains the yeast ADH1 promoter, the following small
part of the coding region, and the terminator (20). Inserted cDNAs
are designed to be expressed in yeast cells as a fusion protein with ADH1 N-terminal amino acids. NA9 was transformed with the
cDNA library, and the colonies that grew on a minimal medium
supplemented with inositol and ethanolamine were selected. Colonies
that grew on a medium supplemented with inositol alone were excluded.
Plasmids were recovered from the transformants and used for
Escherichia coli transformation.
-D-galactopyranoside (IPTG), and after a further 3 h of growth, the cells were pelleted and resuspended in 10 mM Tris-Cl buffer (pH 7.5) containing 1 mM EDTA and 20% glycerol. The cells were then lysed by
sonication and centrifuged at 8,000 × g for 10 min.
The supernatant was used as a crude cell extract for determination of
the enzyme activity. A part of the IPTG-treated and untreated cells,
respectively, was used for protein analysis. SDS-polyacrylamide gel
electrophoresis (SDS-PAGE) of total E. coli proteins was
performed in 10% polyacrylamide mini-slab gels, essentially as
described by Laemmli (23). Molecular weight references were purchased
from Sigma. Proteins were stained with Coomassie
Brilliant Blue using a Quick-CBB kit (Wako).
-32P]dCTP (110 TBq/mmol),
[methyl-14C]choline (2.04 GBq/mmol), and
[methyl-14C]phosphocholine (2.07 GBq/mmol)
were obtained from Amersham. [1,2-14C]Ethanolamine (170 MBq/mmol) was from DuPont NEN. CTP was from Yamasa. Choline chloride,
ethanolamine, phosphocholine, and phosphoethanolamine were from Tokyo
Kasei. Other chemicals were from Wako Chemicals. [1,2-14C]Phosphoethanolamine was prepared from
[1,2-14C]ethanolamine through the action of yeast choline
kinase (24) and purified as described by Sundler (25).
Isolation of Human cDNAs by Complementation in Vivo of a Yeast
Mutant
Fig. 1.
Suppression of the growth defect of yeast
cells by human cDNAs. ise ect1 double mutant NA9 was
transformed with the indicated plasmids, and then its growth phenotype
was examined on a minimal medium supplemented with or without inositol
plus ethanolamine. Plasmid pADANS is the multicopy vector used for the
construction of the human cDNA library.
[View Larger Version of this Image (51K GIF file)]
-AATAAA-3
, is present 579 base
pairs after a TAA termination codon.
Fig. 2.
Nucleotide and deduced amino acid sequences
of the human cDNA. A, restriction map of the cDNA in
plasmid pS4a. The hatched box represents the coding region.
The restriction sites are: E, EcoRI; H,
HindIII; N, NotI; P,
PstI; Pv, PvuII; S,
SacI; Sm, SmaI; Sp,
SphI. EcoRI, HindIII, and
NotI are cloning sites of vector pADANS. B, the
nucleotide sequence of the cDNA and the deduced amino acid sequence
of its product. The largest ORF is translated into an amino acid
sequence, which is shown below the nucleotide sequence in
the one-letter code. Residue numbers for nucleotides and
amino acids are shown on the right. The putative initiation
codon is indicated by the underline. The asterisk
denotes the termination codon. The double-underline
indicates a putative polyadenylation signal.
[View Larger Version of this Image (79K GIF file)]
Fig. 3.
Comparison of the deduced amino acid
sequences of human and yeast cytidylyltransferases. A,
sequences of ET from S. cerevisiae (16), and CTs of S. cerevisiae (8) and humans (7) were aligned in descending order
(yET, yCT, and hCT, respectively) with
respect to the similarity to the predicted sequence of the human ET
(hET). The bold letters show residues identical
to the hET sequence. B, sequence alignment of the N-terminal
and C-terminal halves of human and yeast ETs together with N-terminal
halves of human and yeast CTs. Residues identical among more than four sequences are indicated by bold letters. C,
schematic representation of the human and yeast ETs and CTs. The
hatched boxes represent the sequences shown in B.
The dotted boxes and closed boxes show the
sequences of local similarities.
[View Larger Version of this Image (65K GIF file)]
Fig. 4.
Tissue-specific expression of human ET
mRNA revealed by Northern blot analysis. A, a human
multitissue blot containing poly(A)+ RNA (Clontech) was
probed with a radiolabeled cDNA probe and then autoradiographed.
The positions of RNA size markers are indicated on the left.
B, the same blot was rehybridized with a
manufacturer-supplied -actin DNA as a probe.
[View Larger Version of this Image (48K GIF file)]
Fig. 5.
Incorporation of
[14C]ethanolamine into PtdEtn in the wild-type strain,
the ect1 mutant NA9, and the mutants harboring human cDNAs. To 1 ml of a culture of each strain was added 7.4 kBq of [14C]ethanolamine, and then the mixture was incubated
at 30 °C. At the indicated times, the incorporation of radioactivity
into lipids was determined as described under "Experimental
Procedures." A550 refers to the absorbance at
550 nm. , wild type with pADANS;
, mutant with pADANS;
,
mutant with pH4;
, mutant with pS4a. The total radioactivity taken
up into cells at 60 min was 3.3 × 104, 3.3 × 104, 5.0 × 104, and 3.5 × 104 dpm, for wild type with pADANS, mutant with pADANS,
mutant with pH4, and mutant with pS4a, respectively.
[View Larger Version of this Image (17K GIF file)]
Fig. 6.
Analysis of radiolabeled PtdEtn and
PtdCho. Yeast cells were incubated with 37 kBq of
[14C]choline (lanes 1-3) or 37 kBq of
[14C]ethanolamine (lanes 4-6) at 30 °C for
60 min. Lipids were extracted and analyzed by thin-layer chromatography
as described under "Experimental Procedures." Lanes 1 and 4, wild type; lanes 2 and 5,
mutant with pADANS; lanes 3 and 6, mutant with
pS4a.
[View Larger Version of this Image (52K GIF file)]
Fig. 7.
Analysis of the intermediates of the
CDP-ethanolamine pathway. Yeast cells were incubated at 30 °C
with 37 kBq of [14C]ethanolamine for 60 min. Radiolabeled
water-soluble metabolites were analyzed by thin-layer chromatography as
described under "Experimental Procedures." Lane 1,
mutant with pS4a; lane 2, mutant with pADANS; lane
3, wild type with pADANS.
[View Larger Version of this Image (58K GIF file)]
Strain
Plasmid
Specific
activitya
ET
CT
nmol/min/mg protein
DS15-3
None
1.13
± 0.26
0.87 ± 0.35
NA9
pADANS
0.07
± 0.05
1.38 ± 0.03
NA9
pS4a
2.68
± 0.62
1.65 ± 0.40
a
Specific activity data are the averages ± standard deviations from four independent experiments.
To determine the ET activity of the cDNA product, we expressed its protein product fused with GST in E. coli. A SmaI/EcoRI cDNA fragment was subcloned between the SmaI and EcoRI sites of expression vector pGEX-2T. The plasmid, pGEX-hECT, thus obtained resulted in an in-frame fusion between the N-terminal GST coding region and codons 18-390 of the ET coding region (including the stop codon). The construct was transformed into E. coli, and expression of the fusion protein was induced with IPTG. The production of the fusion protein was analyzed by SDS-PAGE. The IPTG-dependent induction of a large amount of the fusion protein was detected (data not shown). The molecular mass of the fusion protein appeared to be about 65 kDa, this being consistent with the calculated value.
We next determined the ET activity using
[14C]phosphoethanolamine as the substrate and an E. coli cell extract containing the fusion protein as the enzyme
source. Reaction products were analyzed by thin-layer chromatography
and located by autoradiography. As shown in Fig. 8, the
extract prepared from the pGEX-hECT-harboring E. coli cells
cultured in the presence of IPTG contained significant ET activity
(lane 2). The activity was dependent on CTP (lane 3) and Mg2+ (lane 4). When a boiled extract
or an extract prepared from vector-harboring cells was used as the
enzyme source, no activity was detected (lanes 6 and
1, respectively). These results clearly show that the
cDNA obtained here is of the structural gene for human ET.
We describe here the first isolation and characterization of the cDNA for ET from mammalian cells. The size of the mRNA for ET, as revealed by Northern blot analysis, was almost the same as that of the cDNA. The transcript is very abundant in liver, heart, and skeletal muscle tissues. The high level of expression in liver cells agrees with the fact that liver cells have high ET activity (2). The deduced polypeptide encoded by the cDNA consists of 389 amino acids with a molecular mass of about 43.8 kDa. This value is slightly less than that of ET purified from rat liver, i.e. approximately 49.5 kDa (11, 12).
The predicted amino acid sequence of the putative encoded protein shows a high degree of similarity to the yeast ET throughout the entire region. The putative human protein, however, is longer than the yeast protein in both the N- and C-terminal regions, especially at the C-terminal region. The two proteins are functionally similar. Indeed, the introduction of the human cDNA, even in a single copy, restored the growth defect of the yeast ect1 mutant (data not shown). An additional striking structural feature is that human and yeast ETs have a large repetitive sequence in their N-terminal and C-terminal halves, suggesting that the two halves are generated through a gene duplication event. The regions of greatest identity in the two halves of ETs are also homologous to the sequences in the N-terminal half of CTs. Based on the high degree of homology between the N-terminal domain of rat CT, a similarly located region in yeast CT, and a Bacillus subtilis glycerolphosphate cytidylyltransferase (4, 8, 28), it has been proposed that this region bears the active site of these cytidylyltransferases. This was confirmed by that the catalytic activity of a truncated form of rat CT, comprising residues 1-236, was quite similar to that of the wild-type CT (29). Therefore, human and yeast ETs might have two catalytic domains, and this might reflect the substrate specificity of the enzyme. Further studies involving structural and functional analyses of ETs are required.
In a previous study, we cloned the human cDNA for choline kinase from the same library as used here by complementing the yeast cki1 mutation (26). Choline kinase catalyzes the formation of phosphoethanolamine, which is one of the substrates for ET. As shown in Fig. 1, we again obtained the human choline kinase cDNA as a suppressor for the yeast ect1 mutation. It was also shown that the human choline kinase cDNA restored the defect of PtdEtn synthesis via CDP-ethanolamine when introduced into the yeast ect1 mutant (Fig. 5). These results strongly suggest that the introduction of multicopies of the choline kinase gene might result in overproduction of phosphoethanolamine, and the accumulation of phosphoethanolamine could bypass the defect of ect1 through the action of CT. Indeed, this is the case because we found that multicopies of the yeast CKI1 or CCT1 gene can also suppress the growth defect of the ect1 mutant (data not shown). Although phosphoethanolamine is a poor substrate for CT (22, 30), elevation of the enzyme activity of CT on introducing multicopies of CCT1 into cells could compensate for the defect of ET activity.
Based on the analogy between the CDP-choline and CDP-ethanolamine pathways, it has long been thought that ET would be regulated in a similar manner to CT. However, more recent studies summarized by Tijburg et al. (15) strongly suggest that, at least in liver, the CDP-choline and CDP-ethanolamine routes are subject to independent regulation. Further work on thorough comparison of the two cytidylyltransferases and the significance of structural differences that could be relevant for independent regulation of these enzymes are required. The cDNA and the information on the primary structure for ET obtained here will allow us to develop methods for studying the structural and functional relationship between ET and CT, as well as the regulatory mechanism for ET.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBankTM/EMBL Data Bank with accession number(s) D84307[GenBank].
We thank M. Wigler (Cold Spring Harbor Laboratory) for the cDNA library.