 |
Printable
PDF Version
Choline Transport for
Phospholipid Synthesis
VERA MICHEL,* ZONGFEI
YUAN,* SHOBHA RAMSUBIR,
AND MARICA BAKOVIC*,1
*Department of Human
Health and Nutritional
Sciences, University of
Guelph, Guelph, Ontario,
Canada N1G 2W1; and
Ontario Cancer
Institute, Division of
Stem Cell and
Differentiation,
University Health
Network, Toronto,
Ontario, Canada M5G 2M1
Choline is an essential
nutrient for all cells
because it plays a role
in the synthesis of the
membrane phospholipid
components of the cell
membranes, as a
methyl-group donor in
methionine metabolism as
well as in the synthesis
of the neurotransmitter
acetylcholine. Choline
deficiency affects the
expression of genes
involved in cell
proliferation,
differentiation, and
apoptosis, and it has
been associated with
liver dysfunction and
cancer. Abnormal choline
transport and metabolism
have been implicated in
a number of
neurodegenerative
disorders such as
Alzheimer’s and
Parkinson’s disease.
Therefore, the study of
choline transport and
the characteristics of
choline transporters are
of central importance to
understanding the
mechanisms that underlie
membrane integrity and
cell signaling in such
disorders. Kinetic
studies with
radiolabeled choline and
inhibitors distinguish
three systems for
choline transport:
(i)low-affinity
facilitated diffusion,
(ii) high-affinity,
Na+-dependent transport,
and (iii)
intermediate-affinity,
Na+-independent
transport. It is only
recently, however, that
the proteins having
transport
characteristics of at
least one of these
systems have been
identified. They include
(i) polyspecific organic
cation transporters (OCTs)
with low affinity for
choline, (ii)
high-affinity choline
transporters (CHT1s),
and (iii)
intermediate-affinity
choline transporter-like
(CTL1) proteins. CHT1
and CTL1 but not OCT
transporters are
selectively inhibited
with hemicholinium- 3
and essentially display
characteristics of
specialized transporters
for targeted choline
metabolism. CHT1 is
abundant in neurons and
almost exclusively
supplies choline for
acetylcholine synthesis.
The focus here is more
on newly-discovered CTL1
choline transporters.
They are expressed in
different organisms and
cell types, apparently
not for the biosynthesis
of acetylcholine but for
the production of the
most abundant metabolite
of choline, the membrane
lipid
phosphatidylcholine. Exp
Biol Med 231:490–504,
2006 Key words: choline;
phospholipids; choline
transport; CTL1
Importance of Choline
In 1998, choline was
identified as an
essential nutrient for
humans by the National
Academy of Sciences,
United States (1).
Choline is a quaternary
amine (trimethyl-bhydroxy-ethylammonium)
(2) and its
concentration in plasma
has a physiological
range of 10–50 lM (3).
Choline is predominantly
utilized for the
synthesis of essential
lipid components of the
cell membranes,
phosphatidylcholine (PC)
and sphingomyelin, and
for the production of
potent lipid mediators
such as
platelet-activating
factor and
lysophosphatidylcholine
(4). Quantitatively, PC
is the most important
metabolite of choline
and accounts for
approximately one-half
of the total membrane
lipid content (5).
Similarly to folate,
choline is a source of
labile methyl groups
(6). In the liver and
kidney mitochondria,
choline is oxidized to
betaine (trimethylglycine),
which then enters the
one-carbon cycle and
serves as a methyl donor
in the remethylation of
homocysteine to
methionine, to
ultimately generate the
methylation agent S-adenosylmethionine
(SAM; Ref. 7). Finally,
choline is best known
for its key role in
neurons as the precursor
of the neurotransmitter
acetylcholine (8, 9).
Acetylcholine could also
be synthesized and
released from
nonneuronal cells;
however, its role
outside of neurons is
not clearly defined
(10). The adequate
intake level for choline
is 550 mg/day for men
and 425 mg/day for women
(11). Choline content
varies considerably
among common foods (12)
and dietary choline
supplementation could
improve multiple body
functions, including
cognition (13),
learning, and memory
(14, 15). Therefore,
choline is an essential
dietary component for
the normal functioning
of organisms. Here, we
focus on specific
aspects of choline
research in recent
years, including choline
deficiency and choline
metabolic pathways and
transport, with emphasis
on the emerging new
families of transporters
necessary for the
production of PC, a
lipid essential for the
structural integrity and
functioning of cell
membranes.
Choline Deficiency
Healthy individuals meet
or exceed the adequate
intake levels for
choline (16); however,
certain population
groups are at risk for
choline deficiency.
These include pregnant
and lactating women,
infants, cirrhosis
patients, and patients
depending on parenteral
nutrition (17). Dietary
choline deficiency in
animal models causes
spontaneous carcinoma of
the liver and generally
increases sensitivity to
carcinogens (18). It is
the only type of
nutritional deficiency
known to have such
deleterious effect (19).
A gradual reduction in
choline supplementation
initially causes
apoptosis in rat
hepatocytes (20, 21);
however, when cells
adapt to a low choline
intake, they become
resistant, and a
tumorigenic
transformation is
frequently observed
(21). Apoptosis induced
by choline deficiency is
p53-independent and is
linked to TGFb1
signaling and generation
of reactive oxygen
species, which disrupt
the membrane potential
and eventually lead to
mitochondrial
dysfunction and cell
death (21, 22). Adaptive
responses to deleterious
effects of choline
deficiency include
increased transcription
of the proto-oncogenes
c-myc, c-Ha-ras and c-fos,
and increased levels of
proinflammatory
cytokines such as TNF-a
(23). Furthermore, it
has recently been shown
that choline deficiency
affects the expression
of 1000 genes in neural
precursor cells, with
one-third of the genes
involved in the cell
proliferation,
differentiation,
methyl-group metabolism,
and apoptosis (24),
demonstrating the
essentiality of choline
for proper cell
function. Because it is
predominantly utilized
for membrane PC, choline
has a significant impact
on liver function and
systemic lipid
metabolism. In
experimental animals as
well as in humans
(25–28), choline
deficiency results in
the formation of
‘‘nonalcoholic fatty
liver,’’ associated with
accumulation of hepatic
lipids and an increased
sensitivity to
inflammation (29). In
knockout mouse models
lacking liver
CTP:phosphocholine
cytidylyltransferase
(Pcyt1),
which is the major
regulatory enzyme in PC
synthesis, hepatic PC
decreases and an
accumulation of
triglycerides in the
liver is observed. At
the same time, plasma
lipids as well as plasma
lipoprotein content
become reduced (30). It
is well established that
homocysteine levels are
elevated after a
methionine load in
choline-deficient
individuals (26). The
interaction of choline
with other nutrients in
one-carbon metabolism
and homocysteine
production is becoming
an increasingly
important issue to be
examined more actively
in the future (31). The
close relationship
between choline and
folate/methionine
metabolism offers an
opportunity for improved
assessment of human
susceptibility to
choline status,
development of
nonalcoholic fatty
liver, and development
of cancer. The risk
assessment based on
genetic variations may
offer novel preventive
strategies for similar
diseases. The
relationship between
folate deficiency and
polymorphisms of the
5,10-methylenetetrahydrofolate
reductase (MTHFR) gene
is well-known; it
resulted in widespread
supplementation of food
products with folate to
prevent development of
birth defects (32).
Choline deficiency is
not related to the
variations in the MTHFR
gene; however, the
polymorphism of
5,10-methylenetetrahydrofolate
dehydrogenase (MTHFDH)
is a significant risk
factor. When placed on a
choline-deficient diet,
carriers of the most
common allele,
MTHFDH1958A, become
depleted of choline
faster than noncarriers
(27). During pregnancy,
fetal plasma choline
levels are several-fold
higher compared with
choline levels in
maternal plasma, which
may result in a
depletion of maternal
choline stores (29),
implicating the
importance of choline
for the developing
fetus. In a normal
pregnancy, plasma
choline levels rise
constantly throughout
the gestation period
(33), coinciding with
elevated homocysteine
levels (34), which
indicates that the high
fetal demand for choline
may stimulate an
increased choline
synthesis in the
maternal liver (34).
Choline supplements
during pregnancy have
been shown to improve
lifelong memory in rats
(35). Adult offspring of
mothers that had
received choline
supplements were more
adept at tasks involving
spatial and temporal
memory and attention
(35–37), and these
behavioral effects
persisted beyond the age
of 2 years. These data
suggest that choline has
a permanent
developmental effect on
brain organization and
function (1).
Furthermore, consuming
foods high in choline
during pregnancy may
reduce the risk of
neural tube defects;
this is attributed
mainly to the role of
choline as a
methyl-group donor (38).
More on choline
deficiency and the
relationship between
choline, the one-carbon
cycle, and lipid
metabolism is discussed
in the following
sections on choline
metabolic pathways and
transport.
Choline Metabolic
Pathways
The majority of cellular
choline is
phosphorylated by
choline kinase (CK) to
phosphocholine, to which
CTP can then be added by
Pcyt1 to yield
CDP-choline. The
formation of PC results
from the reaction of
CDP-choline with
diacylglycerol (DAG),
catalyzed by CDP-choline:
DAG
cholinephosphotransferase
(CPT; Fig. 1). This
pathway is known as the
Kennedy (CDP-choline)
pathway for de novo
synthesis of PC and is
essential for the
formation of membrane PC
in all nucleated cells.
Since its discovery in
the 1950s by Eugene
Kennedy, this pathway
has been extensively
studied, and most
aspects of its
regulation are well
established. In recent
years, focus is more on
cloning and
characterization of
genes participating in
the pathway (39–43),
with emphasis on the
rate-regulatory gene
Pcyt1 (44–46); the
Kennedy pathway is not
considered to be
regulated by other steps
in the pathway, nor by
the rate of choline
transport. During
evolution, the liver has
retained a backup
pathway for choline
production from the
second most abundant
membrane phospholipids,
phosphatidylethanolamine
(PE; Fig. 1), to provide
this essential
metabolite when dietary
choline is limited, as,
for example, during
starvation, embryonic
development, pregnancy,
or lactation. PE is
transformed to PC in a
three-step methylation
by SAM, which is
catalyzed by
phosphatidylethanolamine-N-methyltransferase
(PEMT). The PE
methylation pathway has
been extensively studied
by Dr. Dennis Vance and
his collaborators
(for reviews see Refs.
47, 48). More recently,
his laboratory developed
a PEMT knockout mouse
model, providing
valuable data on the
role of this pathway in
the liver, PC synthesis,
lipoprotein secretion,
and bile and
homocysteine production
(42, 49). Liver choline/phospholipid
and methionine/homocysteine
pathways are
interrelated by their
need for methyl groups
provided by SAM (Fig.
1). When PE methylation
is blocked, as it is in
PEMT knockout mice (PEMT
/ ), plasma homocysteine
is 50% lower than in
wild-type PEMTž/ž mice
(50). On the other hand,
mice having Pcyt1a
conditionally deleted in
macrophages have
elevated plasma
homocysteine because of
an increased methylation
of PE, to compensate for
the dysfunctional
CDP-choline pathway
(49). When fed a choline-deficient
diet, the PEMT / mouse
dies because of hepatic
depletion of PC. When
fed a regular diet, the
mouse is normal, because
the PEMT / liver can
utilize dietary choline
via the CDP-choline
pathway. The PEMT /
animals develop an
anomalous lipoprotein
profile and a
‘‘fat-liver’’ phenotype
when fed a high-fat/highcholesterol,
‘‘Western-style’’ diet
(51). Surprisingly,
doubleknockout
PEMT / /MDR2 / mice
(MDR2 is a hepatic
Pglycoprotein that
exports PC into bile)
survive on a
cholinedeficient diet
(52). In these animals,
choline is recycled via
increased PC metabolism
and increased flux
through the CDP-choline
pathway, as well as by
suppression of choline
oxidation to betaine
(Fig. 1). Therefore, it
becomes apparent that
PEMT / mice die on a
choline-deficient diet
because these pathways
do not compensate the PC
drain into bile.
Together, these
invaluable studies
strongly demonstrate the
importance of proper
metabolic balances
between choline, SAM,
PE, and PC in liver
function, and make the
pathways (shown in Fig.
1) for their regulation
significant risk factors
for development of
‘‘fatty-liver’’
steatosis,
hyperlipoproteinemia,
hyperhomocyteinemia, and
cardiovascular disease.
Importance of Choline
Transport Separately
from choline metabolism,
disturbed choline
transport has been
proposed to play an
important role in
multiple clinical
manifestations,
particularly in
neurological, muscular,
and immunological
disorders (53, 54) and
in various cancers,
including breast and
ovarian cancer (55, 56).
However, choline
transport has been
studied predominantly in
neurodegenerative
disorders, particularly
in Alzheimer’s disease.
Alzheimer’s disease is
manifested with the
specific loss of
cholinergic neurons, and
choline transport is an
attractive target for
drug development to
treat the disease (57).
It is now clear that
choline is provided to
cholinergic neurons by a
high-affinity transport
system, which controls
the formation of
acetylcholine (57) and
as such is of the
greatest interest in
this disorder (58). In
addition, a low-affinity
choline transport system
unrelated to cholinergic
transport has often been
irregular in the
erythrocytes of
Alzheimer’s disease
patients (59). A reduced
choline level in the
cerebrospinal fluid of
Parkinson’s disease and
Huntington’s disease
patients is a result not
only of impaired choline
transport but also of a
modified
phospholipid output (60,
61). A lower
phospholipid output
could result in membrane
damage, in addition to
the disturbances in
neurotransmitter
production that are very
common in neurological
disorders. In
Friedreich’s ataxia, the
PC molecules
incorporated into the
membrane of certain
cells are abnormal in
that they contain a low
percentage of linoleic
acid (62). Emerging new
data (discussed below)
suggest that choline
carriers that
participate in PC
metabolism, not only in
neuronal cells but in
most cells, are distinct
from cholinergic
carriers. These cells
typically have a medium
to low affinity for
choline and could be
choline-specific,
polyspecific for various
organic cations, or
both. Therefore, besides
the independent
regulation of
intracellular choline
utilization for PC
production in the
Kennedy pathway and in
liver PE methylation,
complex transport
systems for choline to
enter and/or exit
specialized cells are
required in order to
adjust free choline with
specific cellular
demands for choline,
which, when disturbed,
could have serious
implications for
membrane fluidity and
function in all tissues.
Systems for Choline
Transport
Choline is a positively
charged quaternary amine
and requires a
protein-mediated
mechanism to effectively
pass the membrane lipid
barrier. The mechanisms
underlying choline
transport have not been
completely elucidated;
however, three
protein-mediated and
thus saturable uptake
systems, following
Michaelis-Menten
kinetics, are well
documented (63–68). The
first mechanism is the
facilitated diffusion
that has an apparent Km
of 10 lM of choline,
driven by a choline
concentration gradient
as described in red
blood cells (64). The
second is a
high-affinity, Naž- and
energy-dependent
transport system with a
Km of 0.5–3 lM that is
sensitive to inhibition
by low concentration of
hemicholinium-3 (HC-3,
Ki of 1–3 lM; Ref. 65).
This ‘‘active
transport’’ system is
coupled to the
biosynthesis of
acetylcholine and is
primary to neuronal
tissues (65). The third,
also ‘‘active
transport,’’ system,
which is somewhat lower
in affinity for choline
(Km & 20–200 lM), is
ubiquitously
distributed, and less
effectively inhibited by
HC-3 (Ki & 20–200 lM)
than the neuronal system
(66–68). This system
operates in most cells
as a means of choline
uptake for the purpose
of phospholipid
synthesis. Many organs,
such as lung, brain,
placenta, kidney, and
liver have acquired one
or a combination of
several choline
transport systems, which
allow them to fulfill
specific functions. The
alveoli of the lungs are
lined with pulmonary
surfactant, which is a
PC phospholipid-protein
complex that promotes
lung stability (69). The
physiological activity
of the surfactant is
largely attributable to
the unique PC component
of the surfactant,
disaturated
phosphatidylcholine
(69). In all other cells
in the lung, the major
source of PC is the
Kennedy (CDP-choline)
pathway, which requires
the circulating choline
to be transported into
the lung by the alveolar
epithelium. It has been
suggested that this
transcellular transport
process may be
rate-limiting for the
biosynthesis of
pulmonary PC (70, 71).
In addition, alveolar
type II epithelial cells
have developed two types
of ‘‘active’’ choline
transport mechanisms to
regulate the pulmonary
choline content (68):
The first is a
high-affinity,
Naždependent system (Km
of 1.5 lM; Ki of 1.7 lM
for HC-3) resembling the
neuronal transport, and
the second is a
Nažindependent system
that has a lower
affinity for choline (Km
of 19 lM; Ki of 12 lM
for HC-3), resembling
the cholinespecific but
nonneuronal transport.
There is also a third,
very slow residual
transport, which is
presumed to be by
passive diffusion.
The central nervous
system requires choline
for the synthesis of
membrane phospholipids
and for acetylcholine.
In the brain, choline
that is transported via
the ‘‘highaffinity’’
system is destined for
acetylcholine synthesis,
whereas the
‘‘lower-affinity’’
system supplies choline
for phospholipid
synthesis (72). The
ranges of affinity for
these systems could vary
and depend on the region
of the brain in which
they are located,
suggesting that a
combination of choline
carriers could be
involved. Because the
brain cannot
synthesize de novo
choline, choline uptake
from extracellular
fluids is essential
(73). Choline enters the
brain from the systemic
circulation by a
saturable transport at
the blood-brain barrier
(BBB) at a rate
proportional to changes
in the blood choline
concentration, thus
representing a
facilitated diffusion
process (72). Human
keratinocytes express a
Naž-independent,
polyspecific choline
transport system similar
to the choline carrier
in the BBB and
intestinal epithelial
cells. This choline
carrier has relevance
both for the
biosynthesis of skin
lipids and for the
uptake of cationic drugs
into the keratinocytes,
as it does at the BBB
(28).
The importance of
placental choline has
been demonstrated in
studies in which mouse
embryos showed growth
retardation and
developmental defects of
the neural tube and face
during neurulation when
either choline uptake or
metabolism was inhibited
(7). The placenta
possesses a choline
transport mechanism,
which is essential
because a growing fetus
has to derive all its
metabolic choline
requirements from its
mother via the placenta
(74). It has both a
passive diffusion system
and an active
carriermediated process
of low affinity with a
Km of 350 lM choline
(75). Passive diffusion
is the least
physiologically relevant
type of choline
transport, considering
positivelycharged
choline molecules and
the blood concentration
of choline, which is in
the 10–50 lM range. More
relevant is the active
choline transport, as
present at the apical
side of
BeWo and JEG-3
epithelial cells of the
human placenta (76). The
uptake is Naž-independent
and saturable, with a Km
of 108 lM and 206 lM,
respectively. Several
cationic drugs,
including
diphenhydramine,
etilefrine, clonidine,
ranitidine, and
butylscopolamine
interact with this
transport system and
could be instrumental in
the drug’s transferring
from the maternal blood
to the fetal circulation
(76). All ingested
choline and free choline
generated by
phospholipid metabolism
enters the hepatic
circulation, making the
liver, where, as
discussed earlier, there
are very active
biochemical pathways for
choline metabolism, a
significant ‘‘sink’’ for
choline (Fig. 1). The
liver contains a
transport system
resembling the
‘‘low-affinity’’
facilitated diffusion,
but multiple systems are
also detectable under
different conditions
(77). Studies of rat
liver plasma membrane
vesicles have
demonstrated the
presence of a
basolateral transport
for both choline and
organic cations (77). It
seems that the murine
transporter organic
cation transporter (OCT)
1 is responsible for the
most of the hepatic
choline and cationic
drug circulation (78).
At present, it is
unknown whether a
choline-specific
transport is operating
in the liver or whether
different systems could
be present at different
membrane sites. Renal
tubular transport of
choline is of importance
because it maintains the
plasma choline
concentration within
relatively narrow limits
by employing both net
secretion and
reabsorption (79). When
choline is presented to
the kidney in excess of
a species-specific
threshold concentration,
it will be excreted from
the kidney into the
urine (80). If the
plasma concentration is
below this level,
choline will be
reabsorbed into the
kidney, not excreted
(80). Membrane transport
of choline is elevated
in renal failure in
erythrocytes and
cerebral tissue, but the
origins and clinical
importance of that are
unknown. (53). OCT1,
for example, is
localized in the
basolateral membrane of
renal proximal tubules,
in intestinal
enterocytes, and in the
sinusoidal membrane of
hepatocytes, where it
probably mediates the
efflux of many organic
cations, including
choline. The substrate
specificity of rat OCT1
is for more hydrophilic
(type 1) cations like
tetra-ethyl ammonium and
choline. Quinine,
quinidine, and cyanine
863, which have been
classified as type 2
cations, are not
eliminated with rat OCT1
(81). The secretion
(efflux) of organic
cations by renal
proximal tubule is
electrogenic,
facilitated diffusion,
and it is associated
with basolateral uptake
of type 1 organic
cations. These processes
are also believed to
support electroneutral
organic cation/organic
cation exchange. An
extensive presentation
of renal polyspecific
transport of organic
cations is described by
Wright and Dantzler
(82).
Choline Transporters
Different families of
proteins that have
either been demonstrated
or thought to play a
role in choline
transport have been
isolated in a variety of
species. The most
prominent are (i) the
high-affinity choline
transporter family (CHT),
(ii) the family of OCTs,
and (iii) the choline
transporter–like family
(CTL). They exhibit
varying affinities not
only for choline
transport but also for
transport of a number of
other organic cations
(Table 1). Studies of
the effect of various
organic cation
inhibitors revealed that
hemicholinium-3 is the
most effective inhibitor
of choline uptake.
Although CHT
transporters are mostly
important for choline
transport for
acetylcholine synthesis
rather than for lipid
synthesis, we want to
include a brief summary
of these high-affinity
transport systems to
complete the picture of
choline transport.
Comprehensive reviews of
CHT are presented by
Ferguson and Blakely
(83) and Sarter and
Parikh (84).
High Affinity Choline-Specific
Transporters
Cholinergic
neurotransmission plays
an important role in
regulation of many
physiological functions
(85–88). It is well
known that basal
forebrain cholinergic
neurons are involved in
learning and memory
processes, and that
hypoactivity of the
cholinergic system is
responsible for the
cognitive deficit in
Alzheimer’s disease (4).
Naždependent and
HC-3-sensitive choline
transport is very active
in presynaptic
terminals, and it is
mediated by a
highaffinity choline
transporter. After many
years of searching, this
transporter, named CHT1,
has recently been
successfully cloned,
first from
Caenorhabditis elegans
(cho-1) and rat (rCHT1;
Ref. 89), and
subsequently from humans
(hCHT1; Refs. 90, 91),
Limulus (LchCoT; Ref.
92), and mouse (mCHT1;
Ref. 93). Based on the
cDNA sequence, there is
a high degree of
conservation among CHT1
proteins, and their
closest structural
similarity is with the
Naž-dependent glucose
transporter SGLT (89).
hCHT1 is located on
chromosome 2q12. The
entire gene is &25 kb
long and contains 9
exons. hCHT1 protein is
made of 580 amino acids
with a predicted
molecular mass of 63 kDa
and could form 13
transmembrane domains
(90, 91, 94). CHT1 mRNA
expression is restricted
to the brain regions
rich in cholinergic
neurons, like the
forebrain, striatum,
brain stem, and spinal
cord (89–93). Recently,
however, CHT1
transcripts have been
also detected in
nonneuronal cells, but
the function of CHT1
outside of neurons is
not known (95–97). Now
it is clear that CHT1
provides choline for the
synthesis of
acetylcholine in all
cholinergic neurons (92–
94). Functional
expression of rat CHT1
in Xenopus oocytes or
COS-7 cells produces
typical high-affinity
choline uptake that is
identical with the
high-affinity choline
uptake in synaptosomes:
(i) Km for choline of 2
lM, (ii) inhibition by
CH-3 with Ki of 2–5 nM,
and (iii) Naž and Cl ion
dependence (94). Apart
from these cholinergic
transport
characteristics, the
regulation of CHT
expression is becoming
increasingly important.
It has been shown that
CHT1 could be regulated
by neuronal
depolarization (98–102),
second messengers
(103–105), and acute
drug treatments
(106–108). CHT1 contains
serine and threonine
residues for regulation
of function and surface
expression by protein
kinase C (PKC; Refs. 92,
94) and protein
phosphatase 1/2A
(PP1/PP2A; Ref. 92). It
seems a very common
phenomenon that PKC and
PP1/PP2A regulate
neurotransmitter
transporter proteins via
rapid changes in
transporter surface
expression and
phosphorylation (109).
Recently it was reported
that prostate apoptosis
response-4 (Par-4), a
leucine zipper protein
that plays an important
role in neuronal
dysfunction and cell
death in
neurodegenerative
disorders
such as Alzheimer’s
disease (88, 110–116),
inhibits CHT1- mediated
choline uptake activity
and reduces the cell
surface expression of
CHT1 by directly
interacting with CHT1
(88). The importance of
the biological function
of CHT1 has been further
confirmed by the fact
that disruption of CHT1
in mice is lethal to the
neonate (117).
Low-Affinity
Polyspecific Organic
Cation Transporters
The transport of
endogenous organic
cations, clinical drugs,
and exogenous toxic
cations is generally
mediated by the family
of OCTs also known as
the solute carrier 22
family of proteins
(118). This family is
comprised of five
members, including OCT1,
OCT2, OCT3 (119–126),
OCTN1 (127), and OCTN2
(128), and has been
subject to ongoing
intensive research since
the initial cloning of
OCT1 from a rat kidney
library in 1994 (119).
OCT1 is expressed in the
liver, kidney, and small
intestine in rodents
(119, 121, 129, 130);
however, in humans, it
is mainly expressed in
the liver (120). OCT2
has substrate
specificity similar to
that of OCT1, but its
expression is limited to
the kidney and specific
regions in the brain
(122, 131). OCT3 is
expressed most
abundantly in the
placenta and moderately
in the intestine, heart,
and brain (Table 1; Ref.
132). The functional
characterizations from
different species
strongly suggest that
OCT1, OCT2, and OCT3
facilitate polyspecific
cationic transport
(119–121, 128, 131–133).
Regarding their
physiological role in
choline transport,
however, the data are
less extensive and the
information varies in
different experiments.
In transiently
transfected human
embryonic kidney (HEK)
293 cells, choline was
transported exclusively
by OCT1, but not by OCT2
(134). When expressed in
Xenopus oocytes, both
OCT1 and OCT2 could
facilitate choline
uptake with low
affinity, that is, Km of
346 lM for rOCT1, 441 lM
for rOCT2, and 102 lM
for hOCT2 (135). OCT-1
and OCT2-mediated
choline uptake could be
inhibited by cimetidine
(136) and larger, more
hydrophobic cations like
the antiarrhythmics
quinine and quinidine
(81, 135, 137).
Disprocynium 24 and
decynium 22, the most
potent inhibitors
recognized so far,
inhibit OCT2 with Ki
values around 10 nM
(134, 138), and
corticosterone, another
potent inhibitor,
inhibits OCT2 with Ki
values around 200 nM
(137). Disprocynium 24,
decynium 22, and
corticosterone are less
effective with OCT1
(134, 137). The major
driving force for
choline transport via
OCT1 and OCT2 is the
membrane potential (135,
139). An in vivo study
indicates that OCT2
plays an active role in
brain choline
homeostasis and that
choline is not a
substrate for OCT3
(135). A most recent
summary of the OCT1 and
OCT2 family is shown in
Table 1. The genes of
OCT1 and OCT2 have been
mapped on chromosome
6q26 in human (138,
140). Rat and mouse
forms, rOCT1 and
mOCT1(Lx1), have been
localized on chromosome
1q11-q12 and chromosome
17A1, respectively (121,
141). The primary
structure of rOCT1 and a
tentative model of its
membrane topology
predict 12 transmembrane
domains (TMDs) and three
potential glycosylation
sites located in the
large hydrophilic loop
between the first and
second transmembrane
domains (reviewed in
reference 133). Some
data indicate the
existence of
substrate-binding
pockets in rOCT1 (142)
and rOCT2 (143). There
are four potential
protein kinase C
phosphorylation sites on
the large intracellular
loop between TMD6 and
TMD7 (133).
OCT1 knockout mice (OCT
/ mice) are healthy and
fertile (118). OCT /
mice have reduced
hepatic uptake and
intestinal excretion of
various organic cations,
and although OCT1, OCT2,
and OCT3 have
overlapping substrates,
the loss of OCT1 is not
compensated for by
upregulation of either
OCT2 or OCT3.
Interestingly, the
levels of radiolabeled
[14C] choline in the
livers of OCT / mice
were unaffected compared
to those in the livers
of wild-type mice,
suggesting that an
efficient choline
transport system other
than the OCT type exists
in the liver (118). A
deficiency in OCT2 and a
combined deficiency in
OCT1 and OCT2 had no
obvious effect on the
physiology of mice (144,
145), but in OCT1/2 /
mice, renal secretion of
TEA was completely
abolished (144), which
implies that OCT1 and
OCT2 together are
essential for renal
secretion of small
organic cations;
however, their role in
choline transport is not
known (144). OCTN1 and
OCTN2 have been
identified in human and
rat (127, 128, 146,
147). However, choline
is not a substrate of
OCTN1 and OCTN2 (128,
147). Choline-Specific
Transporter-Like
Proteins The choline
transporter–like
proteins of the CTL
family are comprised of
the five genes,
CTL1-CTL5, with CTL1
being the main member of
the family (148). In
2000, O’Regan et al.
(149) first cloned and
characterized Torpedo
marmorata CTL1 (tCTL1)
and rat CTL1 (rCTL1).
Subsequently the human
form was cloned by Dr.
Stockinger’s laboratory
(hCTL1/CDw92; Ref. 150),
and we have cloned the
mouse form, mCTL1 (151).
It is interesting that
the tCTL1 was isolated
by complementation of a
yeast strain defective
in choline transport and
that hCTL1 was cloned
from a cDNA expression
library using a
monoclonal antibody for
the human antigen CDw92.
CDw92 has a role in
immune cell function as
part of an
auto-regulatory
signaling loop that
controls expression and
maintenance of IL-10
production in dendritic
cells (150). The
function assigned to
CDw92 as a choline
transporter has been
based
on the cDNA sequence
similarity to rCTL1 and
other family orthologs.
Mouse CTL2 and CTL4 cDNA
have been initially
identified in the
Genbank libraries as
members of this family
(149); however, the
mouse form of CTL1 was
not detected, suggesting
that this form could be
rare or not present in
the mouse. We searched
for this rare form and
were able to
successfully clone the
mCTL1 cDNA from the
mouse embryonic
fibroblasts, to fully
characterize the mouse
gene that is localized
at chromosome 4B2. Thus,
we added a new member to
the CTL1 family of
proteins that is a
choline transporter
abundantly expressed in
the mouse skeletal
muscle, the function of
which would be in
transporting choline for
phospholipid synthesis
and/or choline recycling
at the neuromuscular
junctions. The protein
sequence
alignment of mouse, rat,
and human CTL1 is shown
in Figure 2. They are
all transmembrane
proteins containing 10
highly conserved TMDs,
possibly with
intracellular Nand
C-termini. Figure 3
shows that mCTL1 protein
contains three potential
N-glycosylation sites
(N134GSA137, N179ISC182,
N506STN509) and seven
protein kinase C
phosphorylation sites
(S12SK, T86HR, T151SK,
S209KE, S268 PK, S513AK,
S628RK), five of which
are distributed within
the N-terminal domain.
Additionally (not
shown), there is one
cysteine-rich region
(C473ARC476MLKSC
481IC483C484LWC487LEKC491)
and one
cadmium-resistance
transporter domain
between L220 and F341,
covering the membrane
domains 3, 4, and 5. The
hCTL1 gene is located at
chromosome 9q31.2, the
rCTL1 gene at chromosome
5q24, and the mCTL1 gene
at chromosome 4B2.
(Table 1). tCTL1 mRNA is
present throughout the
central nervous system
and along the electric
nerves (152); mCTL1 RNA
is expressed in testis,
brain, heart, and
skeletal muscle, but the
mCTL1 protein is
abundantly present only
in the skeletal muscle
(Fig. 4; Ref. 151).
hCTL1 protein is
expressed as two major
polypeptides of 50 and
23 kDa in a variety of
tissues, including
brain, heart, small
intestine, kidney,
liver, lung, skeletal
muscle, pancreas,
spleen, ovary, and
testis (Fig. 4; Ref.
153). Rat rCTL1a mRNA is
expressed in the ileum
and colon, whereas
rCTL1b mRNA is expressed
exclusively in the brain
(148, 149). Both
isoforms of rCTL1 were
recently functionally
expressed and identified
as an
intermediateaffinity,
Naž-independent choline
transport system in N18
neuroblastoma cells
(148) and in astrocytes
(154). Two different
splice variants have
been identified in rat
(rCTL1a and rCTL1b) and
humans (hCTL1a and
hCTL1b;
Refs. 148, 149, 153).
The hCTL1 gene spans
over 194 kb at
chromosome 9q31.2 and
consists of 17 exons and
16 introns (Fig. 5). The
last two exons, 16 and
17, are alternatively
spliced in the two
isoforms hCTL1a and
hCTL1b; hCTL1a is made
of exons 1–16, whereas
hCTL1b uses exons 1–15
and exon 17, resulting
in distinct 39 UTRs in
both isoforms (153). The
alternative splicing
produces proteins with
differing C-terminal
peptides, A651SGASSA657
peptide in hCTL1a and
L651KKR654 peptide in
CTL1b. Noteworthy, KKXX
is a short carboxy-terminal
signal that plays a
crucial role for the
localization in the
endoplasmic reticulum.
The
existence of KKXX motif
for hCTL1b would support
the identification of
the additional site for
choline transport at the
endoplastic reticulum
(148, 153, 155). Our
human tissue expression
profiles indicate that
the hCTL1a mRNA is
highly expressed in
placenta and brain and
the hCTL1b mRNA is most
abundant in the liver
(153). The significance
of both hCTL1 isoforms
has yet to be
discovered. The function
of CTL1 as a choline
transporter has recently
been disputed. CTL1 was
initially identified to
restore choline
transport in a triple
yeast mutant strain (ctr
ise URA3D), which is
deficient in both
choline transport and
neosynthesis under
selective conditions
(149, 156).
Furthermore,expression
of tCTL1 in Xenopus
oocytes increased
sodium-independent
high-affinity choline
uptake (156). However,
Zufferey et al. reported
that the loss of
PNS1, a CTL1 homolog in
yeast, does not impair
choline transport, and
that overexpression of
both PNS1 and tCTL1 in
transport-deficient
mutants did not restore
choline uptake in the
yeast strain (157). In
our laboratory, we
demonstrated a 2.5-fold
increase in choline
transport by
overexpression of mCTL1
in Cos-7
cells (151). We also
identified hCTL1 to be
the predominant choline
transporter in THP-1
cells and demonstrated
that choline uptake is
high in monocytic cells
and severely reduced
after PMA treatments as
a result of diminished
hCTL1 protein
trafficking to the
plasma membrane (Fig.
6). Our choline
transport studies using
[3H]-choline revealed
that choline uptake in
THP-1 cells was almost
completely inhibited by
HC-3, whereas
OCT-specific inhibitors
had no effect on choline
transport. The transport
was of intermediate
affinity for choline (Km
¼ 68 6 12 lM),
PMAdependent and
CTL1-related, because we
strongly excluded
(rCTL1a and rCTL1b) and
humans (hCTL1a and
hCTL1b; Refs. 148, 149,
153). The hCTL1 gene
spans over 194 kb at
chromosome 9q31.2 and
consists of 17 exons and
16 introns (Fig. 5). The
last two exons, 16 and
17, are alternatively
spliced in the two
isoforms hCTL1a and
hCTL1b; hCTL1a is made
of exons 1–16, whereas
hCTL1b uses exons 1–15
and exon 17, resulting
in distinct 39 UTRs in
both isoforms
(153). The alternative
splicing produces
proteins with differing
C-terminal peptides,
A651SGASSA657 peptide in
hCTL1a and L651KKR654
peptide in CTL1b.
Noteworthy, KKXX is a
short carboxy-terminal
signal that plays a
crucial role for the
localization in the
endoplasmic reticulum.
The existence of KKXX
motif for hCTL1b would
support the
identification of the
additional site for
choline transport at the
endoplastic reticulum
(148, 153, 155). Our
human tissue expression
profiles indicate that
the hCTL1a mRNA is
highly expressed in
placenta and brain and
the hCTL1b mRNA is most
abundant in the liver
(153). The significance
of both hCTL1 isoforms
has yet to be
discovered. The function
of CTL1 as a choline
transporter has recently
been disputed. CTL1 was
initially identified to
restore choline
transport in a triple
yeast mutant strain (ctr
ise URA3D), which is
deficient in both
choline transport and
neosynthesis under
selective conditions
(149, 156). Furthermore,
expression of tCTL1 in
Xenopus oocytes
increased
sodium-independent
high-affinity choline
uptake (156). However,
Zufferey et al. reported
that the loss of PNS1, a
CTL1 homolog in yeast,
does not impair choline
transport, and that
overexpression of both
PNS1 and tCTL1 in
transport-deficient
mutants did not restore
choline uptake in the
yeast strain (157). In
our laboratory, we
demonstrated a 2.5-fold
increase in choline
transport by
overexpression of mCTL1
in Cos-7 cells (151). We
also identified hCTL1 to
be the predominant
choline transporter in
THP-1 cells and
demonstrated that
choline uptake is high
in monocytic cells and
severely reduced after
PMA treatments as a
result of diminished
hCTL1 protein
trafficking to the
plasma membrane (Fig.
6). Our choline
transport studies using
[3H]-choline revealed
that choline uptake in
THP-1 cells was almost
completely inhibited by
HC-3, whereas
OCT-specific inhibitors
had no effect on choline
transport. The transport
was of intermediate
affinity for choline (Km
¼ 68 6 12 lM),
PMAdependent and
CTL1-related, because we
strongly excluded
Interestingly, we also
demonstrated that hCTL1
protein is overexpressed
in the skeletal muscle
of muscular dystrophy
patients (153). The CTL1
promoter contains a
binding site for the
MyoD transcription
factor, which typically
regulates muscle
myogenesis by
maintaining fast muscle
fiber phenotypes as well
as other regulatory
functions (162). The
remarkable regenerative
capacity of skeletal
muscle depends largely
on the number of
available satellite
cells and their
proliferative capacity
(163). An increased
macrophage accumulation
can also be observed in
the diseased or injured
muscle, with the extent
of macrophage
accumulation correlating
with protein clearance
following injury and the
efficiency of
regeneration (164). The
increase of hCTL1
protein in muscular
dystrophy is a sign that
the hCTL1 gene is
responsive to processes
common to degenerated
muscle, and it could
represent a target for
improving muscle
function.
Conclusions and
Prospects
With the identification
of three kinds of
choline transport
systems—facilitated
diffusion (64),
low-affinity
(68), and high-affinity
(65)—and the
identification of three
families of choline
transport proteins—CHT,
OCT, and CTL—it is now
possible to further
elucidate the role of
choline transport in
health and disease, and
to identify possible
practical applications
in the treatment of
disorders associated
with abnormal choline
metabolism. It is
crucial to further
investigate the
regulation of choline
transporters at the
protein and RNA level
and to clarify the role
that protein kinase C
may play, considering
the fact that all
choline transporters
contain highly conserved
potential PKC
phosphorylation sites.
The identification of
choline binding sites
and the precise
mechanism of choline
uptake of the different
transporters, as well as
the crystallization and
identification of their
tertiary structures,
will present a further
challenge in the future.
Printable
PDF Version
|
