Regulation of cytosolic free calcium concentration
by extracellular nucleotides in human hepatocytes
CHRISTOF SCHÖFL,1 MARTIN PONCZEK,1 THILO MADER,1 MARK WARING,1
HEIKE BENECKE,1 ALEXANDER VON ZUR MÜHLEN,1 HEIKO MIX,2 MARKUS CORNBERG,2
KLAUS H. W. BÖKER,2 MICHAEL P. MANNS,2 AND SIEGFRIED WAGNER2
1Departments of Clinical Endocrinology and 2Gastroenterology and Hepatology, Medizinische
Hochschule Hannover, 30623 Hannover, Germany
adenosine triphosphate; uridine triphosphate; nucleotide receptor; intracellular calcium; human liver
EXTRACELLULAR ATP AND related compounds have significant biological actions on many tissues and cell types,
including hepatocytes (3–5, 11, 13, 18). ATP and other
nucleotides have been shown to act on specific P2
receptors coupled to the phosphatidylinositol (PI)-Ca21
signaling pathway and to increase cytosolic free Ca21
concentration ([Ca21]i ) in hepatocytes from several
species (7, 9, 10, 12, 19, 25, 31, 41). An increase in
G164
[Ca21]i is one of the key signals for the regulation of
hepatic enzymes, such as glycogen phosphorylase, the
rate-limiting enzyme for glycogenolysis, and extracellular nucleotides have been reported (6, 20–23) to activate glycogen phosphorylase and to stimulate glycogenolysis, most likely by binding to specific P2 receptors in
both rat and human hepatocytes. Furthermore, extracellular ATP has been suggested to be involved in
canalicular contraction of hepatocyte doublets (25) and
in intercellular communication between hepatocytes
and between hepatocytes and bile duct cells in the rat
liver (37). This clearly suggests a potential role of
extracellular nucleotides in the control of human hepatocyte functions. Several P2 receptors exist that are
classified as ligand-gated cationic channels or P2X
receptors and G protein-coupled P2Y receptors (3, 5, 11,
13, 14). On the basis of the relative potencies of various
ATP analogs and nucleotides, P2Y and P2X receptors
can be further subclassified and up to seven P2X and
P2Y receptor subtypes have been cloned so far (3, 14,
24). P2Y receptors coupled to the PI-Ca21 signaling
pathway have been described on hepatocytes from
various species (7, 9, 10, 12, 19, 25, 31, 41), and P2X
receptors have been shown on guinea pig hepatocytes
(7). Because differences between species in the regulation of hepatic functions exist (8, 23, 38), we characterized the P2 receptors expressed on human hepatocytes
and investigated the signal transduction mechanisms
involved. Because P2 receptors are coupled to an increase in [Ca21]i, we measured [Ca21]i in single fura
2-loaded primary human hepatocytes and in the differentiated human hepatoma cell lines Hep G2 (19, 20)
and HuH-7 (30). A single cell approach was chosen
because this circumvents several problems of population measurements in primary cells, including contamination with other cell types that are known to express
P2 receptors (2, 17, 35, 43).
MATERIALS AND METHODS
Materials. Fura 2-AM and Pluronic F-127 were purchased
from Molecular Probes (Eugene, OR). Nifedipine was generously provided by Bayer (Leverkusen, Germany). Williams’
medium E, DMEM, FCS, and antibiotics were from Life
Technologies (Berlin, Germany). Thapsigargin was from Calbiochem (Bad Soden, Germany), and collagenase type H and
hexokinase were from Boehringer (Mannheim, Germany). All
other substances were from Sigma Chemical (Munich, Germany). Stock solutions were prepared in water or as follows.
Thapsigargin (5 mM) was prepared in DMSO and nifedipine
(10 mM) in ethanol.
Because ADP and UDP preparations from commercial
sources might be contaminated with the respective nucleotide
0193-1857/99 $5.00 Copyright r 1999 the American Physiological Society
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.1 on June 18, 2017
Schöfl, Christof, Martin Ponczek, Thilo Mader, Mark
Waring, Heike Benecke, Alexander von zur Mühlen,
Heiko Mix, Markus Cornberg, Klaus H. W. Böker, Michael P. Manns, and Siegfried Wagner. Regulation of
cytosolic free calcium concentration by extracellular nucleotides in human hepatocytes. Am. J. Physiol. 276 (Gastrointest. Liver Physiol. 39): G164–G172, 1999.—The effects of
extracellular ATP and other nucleotides on the cytosolic free
Ca21 concentration ([Ca21]i ) have been studied in single
primary human hepatocytes and in human Hep G2 and
HuH-7 hepatoma cells. ATP, adenosine 58-O-(3-thiotriphosphate) (ATPgS), and UTP caused a concentration-dependent
biphasic increase in [Ca21]i with an initial peak followed by a
small sustained plateau in most cells. In some cells, however,
repetitive Ca21 transients were observed. The rank order of
potency was ATP $ UTP . ATPgS, and complete crossdesensitization of the Ca21 responses occurred between ATP
and UTP. The initial transient peak in [Ca21]i was resistant to
extracellular Ca21 depletion, which demonstrates mobilization of internal Ca21 by inositol 1,4,5-trisphosphate whose
formation was enhanced by ATP and UTP. In contrast, the
sustained plateau phase required influx of external Ca21.
Ca21 influx occurs most likely through a capacitative Ca21
entry mechanism, which was shown to exist in these cells by
experiments performed with thapsigargin. On the molecular
level, specific mRNA coding for the human P2Y1, P2Y2, P2Y4,
and P2Y6 receptors could be detected by RT-PCR in Hep G2
and HuH-7 cells. However, ADP and UDP, which are agonists
for P2Y1 and P2Y6 receptors, respectively, caused no changes
in [Ca21]i, demonstrating that these receptors are not expressed at a functional level. Likewise, a,b-methylene-ATP,
b,g-methylene-ATP, AMP, and adenosine were inactive in
elevating [Ca21]i, suggesting that the ATP-induced increase
in [Ca21]i was not caused by activation of P2X or P1 receptors.
Thus, on the basis of the pharmacological profile of the
nucleotide-induced Ca21-responses, extracellular ATP and
UTP increase [Ca21]i by activating P2Y2 and possibly P2Y4
receptors coupled to the Ca21-phosphatidylinositol signaling
cascade in human hepatocytes. This suggests that extracellular nucleotides from various sources may contribute to the
regulation of human liver cell functions.
P2Y RECEPTORS ON HUMAN HEPATOCYTES
added with 200 µl of 10 mM EGTA (pH 7.0). The samples were
neutralized by adding 600 µl of a 1:1 (vol/vol) mixture of
1,1,2-trichlorotrifluoroethane and tri-n-octylamine, followed
by vigorous mixing and centrifugation. A 800-µl portion of the
upper phase was removed, and intracellular inositol 1,4,5trisphosphate (IP3 ) was evaluated using a commercial radioreceptor assay (Amersham, Braunschweig, Germany).
RNA isolation. RNA was extracted from confluent cultures
of Hep G2, HuH-7, and MG-63 cells using a commercial RNA
isolation kit (QIAGEN RNeasy, Hilden, Germany). Isolation
was performed according to the manufacturer’s guidelines.
RT-PCR for evaluation of P2Y subtype mRNA expression.
Total RNA (2.5 µg) was used as a template for first-strand
cDNA synthesis in a 20-µl reaction volume containing the
following reagents: 5 nmol dNTP (dATP, dTTP, dCTP, and
dGTP), 10 pmol oligo(dT)12–18 (Pharmacia, Freiburg, Germany), 20 units rRNasin (Promega, Heidelberg, Germany),
200 units Superscript RTase (Life Technologies, Berlin, Germany), 50 mM Tris · HCl (pH 8.3), 75 mM KCl, 3 mM MgCl2,
and 10 mM dithiothreitol in diethyl pyrocarbonate-treated
distilled and deionized water. The reaction was incubated at
42°C for 30 min and terminated by heating at 90°C for 5 min.
PCR reactions were carried out in a 50-µl reaction volume
containing the following reagents: 4 µl of cDNA preparation,
20 mM Tris · HCl (pH 8.0), 50 mM KCl, 1.5 mM MgCl2, 0.2 mM
of dATP, dGTP, dCTP, and dTTP, 2.5 U of Taq DNA polymerase (Life Technologies), and 0.5 µM of sense and antisense
primer (Pharmacia). PCR was performed on a RoboCycler
Gradient 96 (Stratagene, Heidelberg, Germany), using the
following conditions for denaturation, annealing, and extension (35 cycles): 94°C for 1 min, 62°C for 1 min, and 72°C for 2
min, followed by 72°C for 10 min. PCR products were
separated by electrophoresis on a 1.8% agarose gel stained
with ethidium bromide. Specificity of the amplified fragments
was confirmed by sequencing the PCR products with the T7
sequencing kit (Pharmacia), according to the manufacturer’s
suggested protocol. To verify that the amplified products were
from mRNA and not genomic DNA contamination, we performed negative controls by omitting the RT. In the absence of
RT, no products were observed. The primer oligonucleotides
of the different P2Y genes were selected from published cDNA
sequences (Ref. 29; Table 1). Glyceraldehyde-3-phosphate
dehydrogenase was used as a housekeeping gene (sense,
58-GGT-CGG-AGT-CAA-CGG-ATT-TGG-TCG-38; antisense, 58CCT-CCG-ACG-CCT-GCT-TCA-CCA-C-38), yielding a 782-bp
product.
Statistics. Unless representative tracings are shown,
values are means 6 SE. Statistical analysis was performed
using Student’s t-test for unpaired or paired data. EC50
values were calculated using GraphPad Prism software
(San Diego, CA).
Table 1. Primers used to identify human
P2Y receptor subtypes
P2Y
Subtype
P2Y1
P2Y2
P2Y4
P2Y6
Sequence of Primer Pairs
58-TGTGGTGTACCCCCTCAAGTCCC-38 (S)
58-ATCCGTAACAGCCCAGAATCAGCA-38 (AS)
58-CCAGGCCCCCGTGCTCTACTTTG-38 (S)
58-CATGTTGATGGCGTTGAGGGTGTG-38 (AS)
58-CGTCTTCTCGCCTCCGCTCTCT-38 (S)
58-GCCCTGCACTCATCCCCTTTTCT-38 (AS)
58-CCGCTGAACATCTGTGTC-38 (S)
58-AGAGCCATGCCATAGGGC-38 (AS)
Length,
bp
260
367
433
464
S, sense; AS, antisense. The length specifies the size of the PCR
product. Information from Ref. 29.
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.1 on June 18, 2017
triphosphates, we preincubated stock solutions of ADP and
UDP (1 mM) with or without hexokinase (10 U/ml) and 10
mM glucose for 1 h at 37°C as described previously (28, 32),
and hexokinase (1 U/ml) was added to the medium in
experiments with nucleotide diphosphates. This protocol has
been shown to effectively convert any contaminating nucleotide triphosphate to the respective nucleotide diphosphate
and to prevent possible conversion of nucleotide diphosphates
to nucleotide triphosphates by membranous nucleoside diphosphokinase (27, 32).
Isolation of primary human hepatocytes. Pieces of liver
tissue were obtained from macroscopically normal regions of
five lobectomies performed for unilocular metastases at the
Department of Abdominal and Transplantation Surgery, Medical School Hannover. Cut vessels on the surface of the liver
tissue were cannulated, and hepatocytes were prepared by
collagenase perfusion, using a protocol described previously
for the isolation of rat hepatocytes by Seglen (40). The
viability of human hepatocytes obtained was 70–85% as
indicated by trypan blue exclusion. Isolated hepatocytes were
suspended in Williams’ medium E supplemented with 10%
FCS (vol/vol), plated on glass coverslips coated with poly-Llysine, and incubated for 4–24 h at 37°C in 5% CO2 (vol/vol)
and 95% air (vol/vol) to allow attachment of the cells. [Ca21]i
measurements were done within 4–24 h after cell isolation.
Cell culture. HuH-7 and Hep G2 cells were grown in
DMEM medium supplemented with 10% FCS (vol/vol), 100
units of penicillin/ml, and 100 µg streptomycin/ml at 37°C in
5% CO2 (vol/vol) and 95% air (vol/vol). For Ca21 measurements, cells were seeded on glass coverslips and used after
2–3 days. Human osteoblastic MG-63 cells were cultured in
RPMI 1640 medium supplemented with 10% FCS (vol/vol),
100 units of penicillin/ml, and 100 µg streptomycin/ml at
37°C in 5% CO2 (vol/vol) and 95% air (vol/vol).
Measurement of [Ca21]i. Primary hepatocytes and Hep G2
or HuH-7 cells attached on glass coverslips were loaded with
5 µM fura 2-AM for 30 min at 37°C in medium containing 130
mM NaCl, 4.7 mM KCl, 1.2 mM KH2PO4, 1.2 mM MgSO4, 1.5
mM CaCl2, 10 mM glucose, 20 mM HEPES, 2% BSA (wt/vol),
and 0.1% Pluronic F-127 (wt/vol), gassed with 100% O2
(vol/vol), pH 7.4. After loading, the coverslips were mounted
in a temperature-controlled superfusion chamber (37°C; Intracel, Royston Herts, United Kingdom) and placed on the stage
of a Zeiss Axiovert IM 135 microscope equipped with a 340
Achrostigmat oil immersion objective (Zeiss, Jena, Germany).
The chamber was superfused at a flow rate of 0.8 ml/min
using a similar medium as described above but with 0.1%
BSA (wt/vol) and without Pluronic F-127. Ca21 measurements were done on cells that were morphologically clearly
identified as hepatocytes and of healthy appearance (round in
shape, no membrane blebs). Fura 2 fluorescence from a single
cell was recorded with a dual wavelength excitation spectrofluorometer system (Deltascan 4000, Photon Technology Instruments, Wedel, Germany), and [Ca21]i was calculated as
previously described (16, 39).
Measurement of inositol 1,4,5-trisphosphate. Hep G2 and
HuH-7 cells grown to confluency in petri dishes (100 3 20
mm) were preincubated for 30 min in medium containing 130
mM NaCl, 4.7 mM KCl, 1.2 mM KH2PO4, 1.2 mM MgSO4, 1.5
mM CaCl2, 10 mM glucose, 20 mM HEPES, 0.1% BSA
(wt/vol), and 10 mM LiCl, gassed with 100% O2 (vol/vol), pH
7.4, at 37°C. Cells were washed and incubated for 20 s with or
without the respective test agents in 1 ml of medium. The
incubation period was stopped by adding 1 ml of 10% ice-cold
HClO4, and the cells were put on ice for 20 min. The cells were
detached and centrifuged, and 800 µl of the supernatant were
G165
G166
P2Y RECEPTORS ON HUMAN HEPATOCYTES
RESULTS
Fig. 1. Effect of extracellular ATP and UTP on intracellular Ca21 concentration ([Ca21]i ) in single primary human
hepatocytes. In most cells ATP and UTP caused a biphasic increase in [Ca21]i and complete cross-desensitization
occurred between ATP and UTP (A and B). In 20–30% of cells, repetitive Ca21 transients were observed in response
to ATP or UTP (C and D). Bars indicate the presence of the respective agents in the superfusion medium.
Representative tracings are shown from 4 to 13 cells.
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.1 on June 18, 2017
Effects of extracellular ATP and UTP on [Ca21]i in
primary human hepatocytes. In single primary human
hepatocytes, resting [Ca21]i was 225 6 17 nM (n 5 40).
Stimulation with ATP at low micromolar concentrations caused a rapid increase in [Ca21]i with a large
initial peak followed by a sustained plateau in most
cells (Fig. 1A). The peak increase in [Ca21]i was dependent on the concentration of extracellular ATP (see Fig.
3A). The estimated EC50 for the ATP-induced increase
in [Ca21]i was ,1.8 µM with a threshold concentration
,1 µM and a maximal effective concentration of 10–30
µM ATP. In some cells, however, repetitive Ca21 transients were observed in response to ATP (Fig. 1C). The
pyrimidine nucleotide UTP elicited a biphasic increase
in [Ca21]i or repetitive Ca21 transients that was indistinguishable from that seen after ATP stimulation (Fig. 1,
B and D). UTP (30 µM) increased [Ca21]i by 223 6 12
nM (n 5 3), which was similar to the effect of ATP (30
µM) in the same preparation (189 6 15 nM; n 5 3).
Pretreatment of cells with maximal concentrations of
ATP or UTP (30 µM) abolished the Ca21 response to
supramaximal concentrations of UTP or ATP (100 µM),
as shown in Fig. 1, A and B, suggesting a common P2
receptor present on primary human hepatocytes. The
nucleotide-induced Ca21 responses were observed as
early as 4 h after the preparation of primary hepatocytes, and no increase in the magnitude of the [Ca21]i
changes occurred during 24 h of short-term culture (not
shown). Furthermore, in primary hepatocytes the ATPor UTP-induced increases in [Ca21]i were of a magnitude similar to those in the hepatoma cell lines (see
below). Together, these data indicate that the nucleotideinduced changes in [Ca21]i in the primary cells reflect
P2 receptor expression in vivo rather than P2 receptor
upregulation during short-term culture as has been
reported in rat salivary gland cells (44).
Effects of extracellular nucleotides on [Ca21]i in single
Hep G2 and HuH-7 cells. Basal [Ca21]i was 226 6 10
nM (n 5 50) and 139 6 4 nM (n 5 50) in the
differentiated human hepatoma cell lines Hep G2 and
HuH-7, respectively. As occurred in primary human
hepatocytes, extracellular ATP caused a concentrationdependent increase in [Ca21]i with a large initial peak
followed by a sustained plateau in the vast majority of
Hep G2 cells (Fig. 2A), while in most HuH-7 cells only a
single transient increase in [Ca21]i was observed (Fig.
2B). In some cells of both cell lines, however, ATPinduced repetitive Ca21 transients were seen (not
shown). The P2 receptor was further characterized
P2Y RECEPTORS ON HUMAN HEPATOCYTES
G167
pharmacologically by investigating the effects on [Ca21]i
of UTP, adenosine 58-O-(3-thiotriphosphate) (ATPgS),
ADP, UDP, AMP, adenosine, a,b-methylene-ATP, and
b,g-methylene-ATP. As in primary hepatocytes, UTP
elicited an increase in [Ca21]i that was indistinguishable from that seen after ATP stimulation and complete
cross-desensitization occurred between both nucleotides in Hep G2 and HuH-7 cells, respectively (Fig. 2, C
and D). Likewise, ATPgS, which is resistant to hydrolysis, caused an increase in [Ca21]i, although this compound was less potent than ATP or UTP (Fig. 3, B and
C). The order of potency of the nucleotides for their
effects on [Ca21]i was ATP $ UTP . ATPgS (Fig. 3, B
and C, Table 2). ADP and UDP at higher concentrations
(.10 µM) also increased [Ca21]i in a concentrationdependent fashion (Fig. 3, B and C). Because ADP and
UDP preparations from commercial sources might be
contaminated with the respective nucleotide triphosphates, stock solutions of ADP and UDP were preincubated with or without (controls) 10 U/ml hexokinase
and 10 mM glucose for 1 h at 37°C to convert any
contaminating nucleotide triphosphate to the respective nucleotide diphosphate (28, 32). Furthermore,
hexokinase (1 U/ml) was added to the medium in
experiments with nucleotide diphosphates to prevent
possible conversion of nucleotide diphosphates to nucleotide triphosphates by membranous nucleoside diphosphokinase (32). As depicted in Fig. 2, E and F, pretreatment of nucleotide diphosphates with hexokinase
abolished the increase in [Ca21]i, whereas UDP or ADP
treated in the same way but without hexokinase increased [Ca21]i in both cell lines (n 5 7 each). This
demonstrates that the nucleotide diphosphate-induced
increase in [Ca21]i was caused by contaminating ATP or
UTP rather than by ADP or UDP itself. AMP, adenosine, a,b-methylene-ATP, and b,g-methylene-ATP at
concentrations as high as 500 µM caused no changes in
[Ca21]i in either Hep G2 or HuH-7 cells (n 5 4–8; not
shown).
Mechanisms of nucleotide-induced Ca21-signals in
human hepatocytes. In Ca21-free medium (2.5 mM
EGTA), the initial peak in [Ca21]i in response to extracellular ATP or UTP (100 µM) was mainly preserved
and amounted to ,60% of the peak increase in the
presence of extracellular Ca21 in primary hepatocytes
and Hep G2 and HuH-7 cells (Fig. 4, A and C). This
indicates that the initial increase in [Ca21]i is caused by
mobilization of Ca21 from intracellular pools most
likely by IP3 whose formation was determined in Hep
G2 and HuH-7 cells to exclude errors arising from
contamination of primary human hepatocytes with
other cell types. Stimulation of Hep G2 and HuH-7 cells
with ATP (100 µM) or UTP (100 µM) for 20 s increased
cellular IP3 levels by 25 6 5 and 30 6 8%, respectively,
above that of control levels (n 5 6 each, P , 0.05). In
contrast, ATP or UTP caused no plateau increase in
[Ca21]i in the absence of extracellular Ca21, and withdrawal of extracellular Ca21 by adding EGTA (2.5 mM)
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.1 on June 18, 2017
Fig. 2. Effect of extracellular nucleotides on
[Ca21]i in single Hep G2 and HuH-7 cells.
Effect of ATP on [Ca21]i in Hep G2 (A) and
HuH-7 cells (B) is shown. Complete crossdesensitization occurred between ATP and UTP
in Hep G2 (C) and HuH-7 cells (D). Pretreatment of commercial preparations of ADP (E) or
UDP (F) with hexokinase (ADP-HK or UDPHK, respectively) as described in MATERIALS
AND METHODS abolished the increase in [Ca21]i
caused by stimulating Hep G2 cells with untreated nucleotide diphosphate. Identical results were obtained in HuH-7 cells (not shown).
Bars indicate the presence of the respective
agents in the superfusion medium. Representative tracings are shown from 6 to 21 cells.
G168
P2Y RECEPTORS ON HUMAN HEPATOCYTES
rapidly abolished the sustained plateau in primary
hepatocytes and Hep G2 and HuH-7 cells (Fig. 4, A and
C). This indicates that the sustained plateau increase
in [Ca21]i requires Ca21 influx from the extracellular
space. Ca21 influx could occur through voltage-sensitive (VSCC) or voltage-insensitive Ca21 channels
(VICC). To test whether VSCC are present on human
hepatocytes and whether they participate in the cytosolic Ca21 response induced by the nucleotides, we
depolarized cells with high K1 (45 mM) and investigated the effects of the VSCC blockers nifedipine or
verapamil on the sustained plateau. High K1 (45 mM)
caused no changes in [Ca21]i, and nifedipine (10 µM) or
verapamil (50 µM) had no effects on the ATP (30 or 100
µM)-induced sustained plateau increase in [Ca21]i in
primary hepatocytes and Hep G2 and HuH-7 cells (n 5
4 each; not shown). This clearly demonstrates that
human hepatocytes are nonexcitable cells and do not
express VSCC. In nonexcitable cells, depletion of the
IP3-sensitive Ca21 store causes an influx of Ca21 across
the plasma membrane to the cytosol, termed capacitative Ca21 entry, that is thought to be the basis for
Table 2. EC50 values for effects of various extracellular
nucleotides on [Ca21 ]i in Hep G2 and HuH-7 cells
EC50 , µM
Agonist
Hep G2
HuH-7
ATP
UTP
ATPgS
0.9 6 0.4
1.4 6 0.6
6.1 6 1.2
0.7 6 0.2
1.7 6 0.7
7.2 6 2.8
Values are means 6 SE. [Ca21 ]i , intracellular Ca21 concentration.
ATPgS, adenosine 58-O-(3-thiotriphosphate).
sustained Ca21 responses in these cells (34). A major
tool used to study capacitative Ca21 entry is the microsomal Ca21-ATPase inhibitor thapsigargin, which releases intracellular Ca21 by preventing reuptake into
the IP3-sensitive pool (42). In the presence of extracellular Ca21, thapsigargin (2 µM) caused a sustained
increase in [Ca21]i, whereas in Ca21-free medium the
increase in [Ca21]i was transient in primary hepatocytes and Hep G2 and HuH-7 cells, respectively, (Fig. 4,
B and D, Table 3). Thus thapsigargin depletes intracellular Ca21 stores and stimulates influx of extracellular
Ca21, which is consistent with capacitative Ca21 entry
being operational in human hepatocytes. After intracellular Ca21 stores were discharged by successive stimulation with increasing concentrations of ATP in Ca21free medium, thapsigargin was still capable of evoking
further release of intracellular Ca21 (Fig. 4C). The
thapsigargin-induced increase in [Ca21]i, however, was
greatly diminished and amounted to 30–50% of the
Ca21 response seen in Hep G2 or HuH-7 cells that had
not been pretreated with ATP in Ca21-free medium (n 5
7 each). In contrast, depletion of intracellular Ca21
stores by thapsigargin abolished the increase in [Ca21]i
in response to high concentrations of ATP (Fig. 4D).
This demonstrates that there is a significant overlap
between the ATP- and thapsigargin-releasable intracellular Ca21 stores with the latter being larger than the
ATP-sensitive store.
Expression of P2Y receptor subtypes in Hep G2 and
HuH-7 cells. The expression of specific P2Y1, P2Y2,
P2Y4, and P2Y6 receptor mRNA was further investigated in Hep G2 and HuH-7 cells to avoid problems
arising from contamination of primary hepatocytes
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.1 on June 18, 2017
Fig. 3. Dose-response curves for extracellular nucleotides on [Ca21]i in primary hepatocytes and Hep G2 and HuH-7
cells. A: dose-response curve of ATP-induced increase in [Ca21]i in primary hepatocytes. B and C: dose-response
curves for various nucleotides on [Ca21]i in Hep G2 (B) and HuH-7 cells (C). [Ca21]i denotes peak increase above
basal. ADP and UDP, untreated commercial preparation; ADP 1 HK and UDP 1 HK, ADP and UDP pretreated with
hexokinase and glucose as described in MATERIALS AND METHODS. ATPgS, adenosine 58-O-(3-thiotriphosphate). Data
points represent means 6 SE from 3 to 21 cells of at least 3 different preparations.
P2Y RECEPTORS ON HUMAN HEPATOCYTES
G169
with other cell types. Human P2Y receptor subtypespecific primer pairs were used as described in MATERIALS AND METHODS, and mRNA from human MG-63 cells,
which have been previously shown to express P2Y1,
P2Y2, P2Y4, and P2Y6 receptors (29), served as a
positive control. PCR amplification of cDNA derived
from RNA extracts of Hep G2 and HuH-7 cells and
sequencing of the PCR products showed that all P2Y
Table 3. Effect of thapsigargin on [Ca21 ]i in primary
hepatocytes and Hep G2 and HuH-7 cells in the
presence or absence of external Ca21
2 µM Thapsigargin
1.5 mM Ca21
pHep
Hep G2
HuH-7
Ca21 free
D[Ca21 ]i
peak, nM
n
D[Ca21 ]i
plateau, nM
n
D[Ca21 ]i
peak, nM
n
81 6 19
64 6 14
86 6 14
9
7
5
26 6 6
37 6 8
43 6 9
9
7
5
35 6 9
38 6 7
25 6 8
3
9
3
Values are means 6 SE; n 5 no. of experiments. Ca21 free, medium
with EGTA (2.5 mM); D[Ca21 ]i peak, peak Ca21 above basal; D[Ca21 ]i
plateau, [Ca21 ]i above basal determined 5 min after addition of
thapsigargin. pHep, primary hepatocytes.
receptor subtypes were expressed in both cell lines
(Fig. 5).
DISCUSSION
In this study, we report for the first time that
extracellular ATP at low micromolar concentrations
elevates [Ca21]i in primary human hepatocytes and in
two differentiated human hepatoma cell lines, Hep G2
and HuH-7. The concentration range over which ATP
increased [Ca21]i is similar to the dissociation constant
value for binding of ATP to human liver plasma membranes (21), indicating the expression of specific P2
receptors on human hepatocytes. P2 receptors are
subdivided into two main groups based on their mode of
signal transduction. The P2X receptors are members of
the transmitter ion channel receptors, whereas the P2Y
receptor subtypes are members of the G proteincoupled receptor superfamily, which are further subclassified pharmacologically by their differing responses to
nucleotide analogs in various tissues (3, 5, 11, 13, 14).
We characterized the P2 receptor expressed on human
hepatocytes by investigating the effects on [Ca21]i of
UTP, ATPgS, ADP, UDP, a,b-methylene-ATP, b,gmethylene-ATP, AMP, and adenosine. UTP and ATPgS
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.1 on June 18, 2017
Fig. 4. A: effect of extracellular ATP on [Ca21]i in the absence of extracellular Ca21 in single primary hepatocytes. B:
in the presence of extracellular Ca21 the endoplasmic Ca21-ATPase inhibitor thapsigargin caused a sustained
increase in [Ca21]i, which was abolished by withdrawal of extracellular Ca21 in primary hepatocytes. C: effect of
depletion of ATP-sensitive Ca21 stores on the thapsigargin-induced increase in [Ca21]i in Ca21-free medium in Hep
G2 cells. D: effect of thapsigargin on [Ca21]i in Ca21-free medium and on ATP-induced mobilization of internal Ca21
in Hep G2 cells. Bars indicate the presence of the respective agents in the superfusion medium. Representative
tracings are shown from 3 to 9 cells.
G170
P2Y RECEPTORS ON HUMAN HEPATOCYTES
elicited an increase in [Ca21]i that was indistinguishable from that seen after ATP stimulation, and complete cross-desensitization occurred between UTP and
ATP in primary hepatocytes and in Hep G2 and HuH-7
cells. The order of potency of the nucleotides for their
effects on [Ca21]i was ATP $ UTP . ATPgS. ADP and
UDP at higher concentrations (.10 µM) also increased
[Ca21]i in a concentration-dependent fashion. Because
commercial ADP and UDP preparations might be contaminated with the respective nucleotide triphosphates, stock solutions of ADP and UDP were preincubated with hexokinase and glucose to convert any
contaminating nucleotide triphosphate to the respective nucleotide diphosphate and hexokinase was added
to the medium in experiments with nucleotide diphosphates to prevent possible conversion of nucleotide
diphosphates to nucleotide triphosphates by membranous nucleoside diphosphokinase (28, 32). The latter
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.1 on June 18, 2017
Fig. 5. RT-PCR analysis of P2Y receptor subtype expression in the
human hepatoma cell lines Hep G2 and HuH-7. Primer pairs specific
for the amplification of P2Y1, P2Y2, P2Y4, and P2Y6 receptor subtypes
were used as previously described (28). Control amplifications were
done using glyceraldehyde-3-phosphate dehydrogenase (GAPDH)specific primers as a housekeeping gene. In Hep G2 (A) and HuH-7
cells (B), P2Y1, P2Y2, P2Y4, and P2Y6 receptor-specific transcripts
were detected. C: in osteoblastic MG-63 cells, which served as
positive controls, all P2Y receptor subtypes are expressed as described previously (28).
point, however, appears to be of minor importance in
the present study as the cells were continuously perfused with fresh medium. ADP or UDP treated in such
a way, however, caused no changes in [Ca21]i, indicating
that the observed increases in [Ca21]i after stimulation
with commercial ADP or UDP were produced by contaminating ATP or UTP rather than by ADP or UDP
itself. The P2X receptor agonists a,b-methylene-ATP
and b,g-methylene-ATP at concentrations as high as
500 µM caused no changes in [Ca21]i in either Hep G2
or HuH-7 cells. This contrasts with a recent report (7)
demonstrating the expression of P2X receptors in guinea
pig hepatocytes, which could be activated by ATP and
a,b-methylene-ATP. Because not all the P2X receptor
subtypes are equally sensitive to a,b-methylene-ATP or
b,g-methylene-ATP (3, 5, 14), we cannot rule out with
certainty that a specific P2X receptor subtype insensitive to a,b-methylene-ATP or b,g-methylene-ATP is
expressed. No evidence, however, could be found for an
ATP-gated cationic channel or P2X receptor on human
hepatocytes, since the shape and time course of the
nucleotide-induced changes in [Ca21]i were identical
and complete cross-desensitization of the Ca21 changes
occurred between ATP and UTP with no further increase in [Ca21]i in response to high concentrations of
ATP after maximal UTP stimulation. Likewise, AMP
and adenosine had no effects on [Ca21]i in Hep G2 and
HuH-7 cells, which indicates that the ATP-induced
increase in [Ca21]i was not caused by P1 receptors.
According to a recent classification of P2 receptors,
this pharmacological profile is characteristic of P2Y2
receptors present on human hepatocytes (3, 5, 11, 13,
14). This is consistent with the expression of mRNA
coding for the cloned P2Y2 receptor in human liver
tissue (33). However, liver tissue such as isolated
primary hepatocytes could be contaminated with cell
types other than hepatocytes. As the nucleotideinduced Ca21 responses were virtually identical in
primary hepatocytes and in the hepatoma cell lines, the
expression of P2Y receptor subtype-specific mRNA was
investigated in Hep G2 and HuH-7 cells. Because
recent evidence suggests (24) that the P2Y receptors
designated as P2Y5 and P2Y7 do not belong to the
family of nucleotide receptors, we used primer pairs for
the human P2Y1, P2Y2, P2Y4, and P2Y6 receptors. In
accordance with the pharmacological characterization
at the functional level, P2Y2 receptor-specific mRNA
could be detected by RT-PCR in both Hep G2 and
HuH-7 cells. However, specific mRNA for the other
three P2Y receptor subtypes could also be found in both
cell lines. Because ADP and UDP, which are the most
potent agonists for P2Y1 and P2Y6 receptors, respectively, caused no changes in [Ca21]i, this strongly
suggests that these receptor subtypes are not expressed at a functional level. This points to posttranscriptional processes that appear to be of prime importance for the regulation of functional P2Y receptor
protein expression. Thus caution is needed when specific P2Y mRNA expression is provided to suggest
functional receptor expression. Because P2Y2 and P2Y4
receptors are only sensitive to ATP and UTP, albeit with
P2Y RECEPTORS ON HUMAN HEPATOCYTES
mobilizing agonists in a variety of nonexcitable cells
(34).
The range of concentrations over which ATP increased [Ca21]i is similar to the concentrations required
for ATP to stimulate glycogen phosphorylase activity in
human hepatocytes, which is the rate-limiting step in
liver glycogenolysis (20). This clearly offers a functional
role for extracellular ATP in human liver metabolism.
In addition, activation of the PI-Ca21 signaling cascade
and an increase in [Ca21]i, which is a major intracellular signal in many cells, may be involved in the
regulation of other hepatic functions such as protein
synthesis or gene expression (15). Furthermore, nucleotides such as ATP have been recently shown (37) to be
involved in intercellular signal propagation between
hepatocytes and between hepatocytes and bile duct
cells in the rat liver. Given the pronounced effects of
ATP or UTP on [Ca21]i, the regulated release of both
nucleotides, e.g., from sympathetic nerve endings, where
ATP is coreleased with norepinephrine, from purinergic
nerves in the autonomic nervous system, from hepatocytes by cell volume-controlled autocrine or paracrine
secretion, or from platelets and damaged cells during
inflammation, could prove to be physiologically important in the control of human hepatic functions (5, 11,
18, 36).
We thank Dr. N. Schütz, University of Würzburg, for the supply of
MG-63 cells.
This work was supported by Deutsche Forschungsgemeinschaft
Grants Scho 466/1-3, Wa 757/1-1, SFB 280, and SFB 265.
Address for reprint requests: C. Schöfl, Abteilung Klinische Endokrinologie, Medizinische Hochschule Hannover, 30623 Hannover,
Germany.
Received 19 December 1997; accepted in final form 14 October 1998.
REFERENCES
1. Aden, D. P., A. Fogel, S. Plotkin, I. Damjanov, and B. B.
Knowles. Controlled synthesis of HBsAg in a differentiated
human liver carcinoma-derived cell line. Nature 282: 615–616,
1979.
2. Athari, A., K. Hänecke, and K. Jungermann. Prostaglandin
F2a and D2 release from primary Ito cell cultures after stimulation with noradrenaline and ATP but not adenosine. Hepatology
20: 142–148, 1994.
3. Boarder, M. R., and S. M. O. Hourani. The regulation of
vascular function by P2 receptors: multiple sites and multiple
receptors. Trends Pharmacol. Sci. 19: 99–107, 1998.
4. Boarder, M. R., G. A. Weisman, J. T. Turner, and G. F.
Wilkinson. G protein-coupled P2 purinoceptors: from molecular
biology to functional responses. Trends Pharmacol. Sci. 16:
133–139, 1995.
5. Burnstock, G. The past, present and future of purine nucleotides as signalling molecules. Neuropharmacology 36: 1121–
1139, 1997.
6. Buxton, D. B., S. M. Robertso, and M. S. Olson. Stimulation
of glycogenolysis by adenine nucleotides in the perfused rat liver.
Biochem. J. 237: 773–780, 1986.
7. Capiod, T. ATP-activated cation currents in single guinea-pig
hepatocytes. J. Physiol. (Lond.) 507: 795–805, 1998.
8. Capiod, T., J. Noel, L. Combettes, and M. Claret. Cyclic
AMP-evoked oscillations of intracellular [Ca21]i in guinea-pig
hepatocytes. Biochem. J. 275: 277–280, 1991.
9. Charest, R., P. F. Blackmore, and J. H. Exton. Characterization of responses of isolated rat hepatocytes to ATP and ADP. J.
Biol. Chem. 260: 15789–15794, 1985.
10. Creba, J. A., C. P. Downes, P. T. Hawkins, G. Brewster, R. H.
Michell, and C. J. Kirk. Rapid breakdown of phosphatidylinosi-
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.1 on June 18, 2017
different potencies, demonstration of functional coexpression of both receptors in the same cell is hard to
achieve. Therefore, the possibility exists that functional P2Y4 receptors are coexpressed with P2Y2 receptors and contribute to the nucleotide-induced changes
in [Ca21]i in the human hepatoma cell lines.
P2Y receptor subtypes belong to the G proteincoupled receptor superfamily (3, 5, 11, 13, 14), and the
mechanisms involved in the generation of the nucleotide-induced Ca21-signals in human hepatocytes were
investigated in the present study. ATP and the other
nucleotides caused a biphasic increase in [Ca21]i with
an initial peak followed by a sustained plateau in the
presence of extracellular Ca21 in the majority of cells.
In some cells, however, repetitive Ca21 transients were
observed as has been reported from single rat hepatocytes at lower nucleotide concentrations (12). In the
absence of extracellular Ca21 only a transient increase
in [Ca21]i was seen. This indicates mobilization of Ca21
from intracellular stores during the initial peak, most
likely mediated by IP3 whose formation was enhanced
after stimulation of Hep G2 and HuH-7 cells with ATP
or UTP. The nucleotide-stimulated increase in cellular
IP3 levels in the hepatoma cells was similar to that
previously reported from rat hepatocytes (9, 22). In
addition to mobilization of intracellular Ca21, influx of
Ca21 from the extracellular space is required as demonstrated by the absence of the sustained plateau increase in [Ca21]i in Ca21-free medium. Because the
plateau increase in [Ca21]i was insensitive to blockers
of VSCC of the L-type, Ca21 influx does not occur
through L-type Ca21 channels. The lack of effect on
[Ca21]i of membrane depolarization by high extracellular K1 further indicates that VSCC channels are not
expressed on human hepatocytes. Other routes of Ca21
influx involve VICC, and Ca21 influx across the plasma
membrane to the cytosol is stimulated by depletion of
the IP3-sensitive Ca21 store. This mechanism has been
termed capacitative Ca21 entry and is thought to be the
basis for sustained Ca21 responses in nonexcitable cells
(34). Consistent with capacitative Ca21 entry being
operational in human hepatocytes, thapsigargin, which
depletes intracellular Ca21 stores without formation of
IP3, caused a sustained increase in [Ca21]i in the
presence of extracellular Ca21, whereas in Ca21-free
medium the increase in [Ca21]i was transient. Because
the thapsigargin- and the ATP-releasable Ca21 stores
overlap, ATP-induced mobilization of internal Ca21
might also activate capacitative Ca21 entry. Thus activation of phospholipase C and concomitant formation of
IP3 could explain the mobilization of Ca21 from intracellular stores by ATP and UTP, which is in agreement
with a P2Y2 and possibly P2Y4 receptor coupled to the
PI-Ca21 signaling cascade as has been demonstrated in
other tissues (3, 5, 11, 13, 14). Depletion of the IP3sensitive Ca21 store might then activate influx of
extracellular Ca21 across the plasma membrane via a
capacitative Ca21 entry mechanism shown to exist in
human hepatocytes. This might be the basis for the
sustained increase in [Ca21]i observed in response to
ATP or UTP as has been shown for other Ca21-
G171
G172
11.
12.
13.
14.
15.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
tol 4-phosphate and phosphatidylinositol 4,5-bisphosphate in rat
hepatocytes stimulated by vasopressin and other Ca21-mobilizing hormones. Biochem. J. 212: 733–747, 1983.
Dalziel, H. H., and D. P. Westfall. Receptors for adenine
nucleotides and nucleosides: subclassification, distribution, and
molecular characterization. Pharmacol. Rev. 46: 450–466, 1994.
Dixon, C. J., N. M. Woods, K. S. R. Cuthbertson, and P. H.
Cobbold. Evidence for two Ca21-mobilizing purinoceptors on rat
hepatocytes. Biochem. J. 269: 499–502, 1990.
Dubyak, G. R., and C. El Moatassim. Signal transduction via
P2-purinergic receptors for extracellular ATP and other nucleotides. Am. J. Physiol. 265 (Cell Physiol. 34): C577–C606, 1993.
Fredholm, B. B., M. P. Abbracchio, G. Burnstock, G. R.
Dubyak, T. K. Harden, K. A. Jacobson, U. Schwabe, and M.
Williams. Towards a revised nomenclature for P1 and P2
receptors. Trends Pharmacol. Sci. 18: 79–82, 1997.
Gardemann, A., G. P. Püschel, and K. Jungermann. Nervous
control of liver metabolism and hemodynamics. Eur. J. Biochem.
207: 399–411, 1992.
Grynkiewicz, G., M. Poenie, and R. Y. Tsien. A new generation of Ca21 indicators with greatly improved fluorescence properties. J. Biol. Chem. 260: 3440–3450, 1985.
Hashimoto, N., T. Watanabe, Y. Shiratori, Y. Ikeda, H. Kato,
K. Han, H. Yamada, G. Toda, and K. Kurokawa. Prostanoid
secretion by rat hepatic sinusoidal endothelial cells and its
regulation by exogenous adenosine triphosphate. Hepatology 21:
1713–1718, 1995.
Häussinger, D. Regulation of hepatic metabolism by extracellular nucleotides and eicosanoids. J. Hepatol. 8: 259–266, 1989.
Irving, H. R., and J. H. Exton. Phosphatidylcholine breakdown
in rat liver plasma membranes: roles of guanine nucleotides and
P2-purinergic agonists. J. Biol. Chem. 262: 3440–3443, 1987.
Keppens, S., and H. De Wulf. Characterization of the liver
P2-purinoceptor involved in the activation of glycogen phosphorylase. Biochem. J. 240: 367–371, 1986.
Keppens, S., A. Vandekerckhove, and H. De Wulf. Characterization of purinoceptors present on human liver plasma membranes. FEBS Lett. 248: 137–140, 1989.
Keppens, S., A. Vandekerckhove, and H. De Wulf. Extracellular ATP and UTP exert similar effects on rat isolated hepatocytes. Br. J. Pharmacol. 105: 475–479, 1992.
Keppens, S., A. Vandekerckhove, H. Moshage, S. H. Yap, R.
Aerts, and H. De Wulf. Regulation of glycogen phosphorylase
activity in isolated human hepatocytes. Hepatology 17: 610–614,
1993.
Khakh, B. S., and C. Kennedy. Adenosine and ATP: progress in
their receptors’ structures and functions. Trends Pharmacol. Sci.
19: 39–41, 1998.
Kitamura, T., U. Brauneis, Z. Gatmaitan, and I. M. Arias.
Extracellular ATP, intracellular calcium and canalicular contraction in rat hepatocyte doublets. Hepatology 14: 640–647, 1991.
Knowles, B. B., C. C. Howe, and D. P. Aden. Human hepatocellular carcinoma cell lines secrete the major plasma proteins and
hepatitis B surface antigen. Science 209: 497–499, 1980.
Lazarowski, E. R., W. C. Watt, M. J. Stutts, H. A. Brown,
R. C. Boucher, and T. K. Harden. Enzymatic synthesis of
UTPgS, a potent hydrolysis resistant agonist of P2u-purinoceptors. Br. J. Pharmacol. 117: 203–209, 1996.
Leon, C., B. Hechler, C. Vial, C. Leray, J.-P. Cazenave, and
C. Gachet. The P2Y1 receptor is an ADP receptor antagonized by
ATP and expressed in platelets and megakaryoblastic cells.
FEBS Lett. 403: 26–30, 1997.
29. Maier, R., A. Glatz, J. Mosbacher, and G. Bilbe. Cloning of
P2Y6 cDNAs and identification of a pseudogene: comparison of
P2Y receptor subtype expression in bone and brain tissues.
Biochem. Biophys. Res. Commun. 237: 298–302, 1997.
30. Nakabayashi, H., K. Taketa, K. Miyano, T. Yamane, and J.
Sato. Growth of human hepatoma cell lines with differentiated
functions in chemically defined medium. Cancer Res. 42: 3858–
3863, 1982.
31. Nathanson, M. H., and K. Mariwalla. Characterization and
function of ATP receptors on hepatocytes from the little skate
Raja erinacea. Am. J. Physiol. 270 (Regulatory Integrative Comp.
Physiol. 39): R561–R570, 1996.
32. Nicholas, R. A., W. C. Watt, E. R. Lazarowski, Q. Li, and
T. K. Harden. Uridine nucleotide selectivity of three phospholipase C-activating P2 receptors: identification of a UDP-selective,
a UTP-selective, and an ATP- and UTP-specific receptor. Mol.
Pharmacol. 50: 224–229, 1996.
33. Parr, C. E., D. M. Sullivan, A. M. Paradiso, E. R. Lazarowski, R. C. Boucher, and J. T. Turner. Cloning and
expression of a human P2u nucleotide receptor, a target for cystic
fibrosis pharmacotherapy. Proc. Natl. Acad. Sci. USA 91: 3275–
3279, 1994.
34. Putney, J. W., Jr., and G. S. T. Bird. The inositol phosphatecalcium signaling system in non-excitable cells. Endocr. Rev. 14:
610–631, 1993.
35. Ralevic, V., R. T. Mathie, B. Alexander, and G. Burnstock.
Characterization of P2x- and P2y-purinoceptors in the rabbit
hepatic arterial vasculature. Br. J. Pharmacol. 103: 1108–1113,
1991.
36. Roman, R. M., A. P. Feranchak, D. K. Guenette, and J. G.
Fitz. Regulated release of ATP from human hepatocytes plays a
key role in cell volume homeostasis (Abstract). Gastroenterology
114: A1329, 1998.
37. Schlosser, S. F., A. D. Burgstahler, and M. H. Nathanson.
Isolated rat hepatocytes can signal to other hepatocytes and bile
duct cells by release of nucleotides. Proc. Natl. Acad. Sci. USA 93:
9948–9953, 1996.
38. Schöfl, C., A. Sanchez-Bueno, G. Brabant, P. H. Cobbold,
and K. S. R. Cuthbertson. Frequency and amplitude enhancement of calcium transients by cyclic AMP in hepatocytes. Biochem. J. 271: 799–802, 1991.
39. Schöfl, C., P. Schulte, L. Rössig, A. von zur Mühlen, and G.
Brabant. Vasopressin induces frequency-modulated repetitive
calcium transients in single insulin-secreting HIT cells. Mol.
Cell. Endocrinol. 108: 185–192, 1995.
40. Seglen, P. O. Preparation of isolated rat liver cells. Methods Cell
Biol. 13: 29–83, 1976.
41. Sistare, F. D., R. A. Picking, and R. C. Haynes. Sensitivity of
the response of cytosolic calcium in quin-2-loaded rat hepatocytes to glucagon, adenine nucleosides, and adenine nucleotides.
J. Biol. Chem. 260: 12744–12747, 1985.
42. Takemura, H., A. R. Hughes, O. Thastrup, and J. W. Putney,
Jr. Activation of calcium entry by the tumor promoter thapsigargin in parotid acinar cells. J. Biol. Chem. 264: 12266–12271,
1989.
43. Takemura, S., N. Kawada, K. Hirohashi, H. Kinoshita, and
M. Inoue. Nucleotide receptors in hepatic stellate cells of the rat.
FEBS Lett. 354: 53–56, 1994.
44. Turner, J. T., G. A. Weisman, and J. M. Camden. Upregulation of P2Y2 nucleotide receptors in rat salivary gland cells
during short-term culture. Am. J. Physiol. 273 (Cell Physiol. 42):
C1100–C1107, 1997.
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.1 on June 18, 2017
16.
P2Y RECEPTORS ON HUMAN HEPATOCYTES