Vol. 278, Issue 4, G542-G550, April 2000
Differential lobular induction in rat liver of glutathione
S-transferase A1/A2 by phenobarbital
Niazy
Selim1,
Gene D.
Branum1,
Xia
Liu2,
Richard
Whalen2, and
Thomas D.
Boyer2
Departments of 1 Surgery and
2 Medicine, Emory University School of
Medicine, Atlanta, Georgia 30322
 |
ABSTRACT |
Phenobarbital and
other xenobiotics induce drug-metabolizing enzymes, including
glutathione S-transferase A1/A2 (rGSTA1/A2). We examined the
mechanism of induction of rGSTA1/A2 in rat livers after phenobarbital
treatment. The induction of rGSTA1/A2 was not uniform across the
hepatic lobule; steady-state transcript levels were threefold higher in
perivenous hepatocytes relative to periportal hepatocytes when examined
by in situ hybridization 12 h after a single dose of phenobarbital.
Administration of a second dose of phenobarbital 12 or 24 h after the
first dose did not equalize the induction of rGSTA1/A2 across the
lobule. The transcriptional activity of the rGSTA1/A2 gene was
increased 3.5- to 5.5-fold in whole liver by phenobarbital, but
activities were the same in enriched periportal and perivenous
subpopulations of hepatocytes from phenobarbital-treated animals. The
half-life of rGSTA1/A2 mRNA in control animals was 3.6 h, whereas it
was 10.2 h in phenobarbital-treated animals. We conclude that
phenobarbital induces rGSTA1/A2 expression by increasing
transcriptional activity across the lobule but induction of rGSTA1/A2
is greater in perivenous hepatocytes due to localized stabilization of
mRNA transcripts.
enzyme; regulation; half-life; mRNA; detoxication
| |
INTRODUCTION |
THE LIVER LOBULE IS a three-dimensional structure that
is perfused by both portal venous and hepatic arterial blood. The blood moves down the hepatic sinusoid, and hepatocytes near the entry of the
vessels (periportal area or zone 1) are exposed to a different environment than are cells located near the exit site for the blood at
the terminal hepatic veins (perivenous area or zone 3) (10,
12). Hepatocytes arise from the same precursor cells, and many of their
functions, such as the synthesis of albumin, are common to all
hepatocytes within the hepatic lobule (6). However, other functions are
regionally distributed. For example, carbohydrate and ammonia
metabolism, bile salt transport, and drug metabolism vary across the
lobule. These differences appear to reflect differential gene
expression in the various hepatocyte subpopulations within the hepatic
lobule (10). However, in at least one instance posttranscriptional
events account for the perivenous localization of one gene product, the
GLUT-1 glucose transporter (3).
Hepatocytes also respond differently to inducing agents depending on
their location within the lobule. For example, treatment of rats with
phenobarbital leads to an increase, which is greatest in perivenous
hepatocytes, in the transcript levels of cytochrome P-450 CYP2B1 and CYP2B2 (7). Differential
induction is observed whether the cells are in the liver or
transplanted into the spleen, suggesting that the variable response to
the inducer is intrinsic to the hepatocytes and is not dependent on
their location within the hepatic acinus (17). The differential
induction of CYP by phenobarbital is thought to be mediated
transcriptionally, with increased rates of transcription in zone
3 and a lack of induction in zone 1 (30).
The glutathione S-transferases (GSTs) are a family of
detoxication enzymes found in the cytosol of most cells. Like the CYP enzymes, the GSTs can be induced by a variety of agents, including phenobarbital, 3-methylcholanthrene, and products of oxidant stress (5,
11, 18, 27). The induction appears to be largely transcriptional and is
mediated by response elements in the 5'-flanking sequence of the
genes (11, 18, 27). Increased expression of rGSTA2 in perivenous
hepatocytes is found after treatment of animals with
3-methylcholanthrene (19). In the latter study, an enhancer site in the
5'-flanking sequence of the GST gene mediates the inducible
expression. Ischemia-reperfusion injury to the liver is also
associated with an increase in rGSTA1/A2 transcripts in the perivenous
hepatocytes of the hepatic lobule and is, at least in part,
transcriptionally mediated (5). All of these studies suggest that
hepatocytes in the perivenous region of the hepatic lobule respond
differently to inducing agents compared with periportal hepatocytes.
Most investigators have assumed that these differences are
transcriptionally mediated; however, transcriptional activity in the
different hepatocyte populations has not been determined.
In the current study, we systematically examined the lobular induction
of rGSTA1/A2 by phenobarbital. We defined the time course of induction
in both periportal and perivenous hepatocytes by in situ hybridization.
We then determined whether multiple doses of phenobarbital altered the
response of the hepatocytes to the inducing agent. We obtained
periportal and perivenous enriched subpopulations of hepatocytes and
determined the transcriptional activities in the two subpopulations
after treatment of the animals with phenobarbital. Finally, we
determined the effect of phenobarbital on rGSTA1/A2 mRNA half-life. The
results of these studies demonstrate that phenobarbital has effects on
rGSTA1/A2 mRNA at both transcriptional and posttranscriptional levels.
| |
METHODS |
All chemicals used were of analytical grade and, unless otherwise
noted, were purchased from Sigma Chemical (St. Louis, MO) or Fisher
Scientific (Pittsburgh, PA). Restriction enzymes were from Promega
(Madison, WI) or New England Biolabs (Beverly, MA), random primer DNA
labeling reagents were from Amersham (Arlington Heights, IL), and
radiolabeled nucleotides ([32P]dCTP,
[32P]UTP and [35S]UTP)
were from NEN Life Science (Boston, MA). Sodium phenobarbital was from
Elkins-Sinn (Cherry Hill, NJ). Adult male (250-350 g), pathogen-free, Sprague-Dawley rats were from Harlan (Prattville, AL),
and the experimental protocols received prior approval from the Emory
University Institutional Animal Care and Use Committee.
Phenobarbital-treated animals.
Phenobarbital (8 mg/100 g body wt) or 0.9% saline for control animals
was injected intraperitoneally, and liver samples from the animals were
obtained at various times after injection as described below. One group
of animals received a second injection of phenobarbital 12 h after the
initial injection and was killed 12 h later, whereas a second group
received a second injection 24 h after the first and was killed 24 h later.
Tissue collection and mRNA half-life determinations.
Livers were collected at laparotomy at prescribed time intervals using
pentobarbital anesthesia. At harvest, a small piece of liver was
removed and embedded in optimum cutting temperature compound (Miles,
Elkhart, IN), frozen immediately at 
80°C, and used for in
situ hybridization studies. The remaining liver was perfused with
saline buffer, and portions were then frozen in liquid nitrogen,
pulverized, and homogenized in ice-cold guanidine thiocyanate solution
(0.5 g liver/5 ml); total RNA was extracted as described previously
(5). Animals to be used for determination of mRNA half-life received an
intravenous injection of 
-amanitin (50 µg/100 g body wt) and
intraperitoneal actinomycin D (150 µg/100 g body wt) 6 h after
injection of phenobarbital or saline as described previously (14). The
animals were killed at various times after injection, and livers were
removed and processed as described above.
Preparations enriched in periportal (zone 1) and perivenous
(zone 3) hepatocytes were obtained from the same liver by
modifications of a previously described method (25). In brief, the
portal vein and caudal vena cava were catheterized and livers were
perfused with perfusion buffer (per liter: 115 mmol NaCl, 5.6 mmol KCl, 1.2 mmol MgCl2, 1.2 mmol NaH2PO4,
25 mmol NaHCO3, and 1 g glucose, pH 7.4, at 20°C) for 5 min at a flow rate of 20 ml/min. The buffer was gassed continuously
with O2/CO2 (19:1). The circulation to different lobes of the liver was sequentially blocked, and the liver
was perfused first in a prograde direction and then in a retrograde
direction with digitonin (4 mg/ml) until periportal and perivenous
blanching were seen, respectively. Digitonin was washed out of the
liver by perfusion in the opposite direction with perfusion buffer
before each change in perfusion direction. The livers were then
perfused with collagenase, and periportal and perivenous enriched
hepatocyte preparations were obtained from the same liver. The cells
were centrifuged using a Percoll gradient to separate hepatocytes from
cellular debris, including nuclei released from lysed cells. Glutamine
synthetase and alanine transaminase activities were measured in each
preparation to ensure that the populations were enriched (25).
Hepatocytes were snap frozen and stored at 80°C for the
preparation of nuclei for nuclear runoff assays as described below.
Northern blot analysis.
Total RNA from lobes of treated or control animals at each time
point were isolated separately. RNA (20 µg) from each sample was
denatured in 2.2 M formaldehyde/50% formamide and electrophoresed on
1% (wt/vol) agarose gels containing 2.2 M formaldehyde. The quality of
the RNA was judged after staining the gels with ethidium bromide. The
RNA was transferred to nylon membranes (NEN Life Science) by capillary
blotting in 10× standard saline citrate (SSC; 1× SSC is
0.15 M NaCl, 0.015 M sodium citrate; pH 7.0) overnight. The RNA was
linked to the membrane in a UV cross-linker (Stratagene, La Jolla, CA),
dried at 80°C for 60 min, and stored at room temperature until used
for hybridization.
The cDNAs for rGSTA2 (pGTB 38), rGSTA3 (pGTB 42), and quinone reductase
were gifts from Dr. C. Pickett (20, 24). The cDNAs for ribosomal
protein S14 (pCS14) and glyceraldehyde-3-phosphate dehydrogenase were
from the American Type Culture Collection (ATCC). The cDNA for mouse
albumin was from Dr. E. T. Morgan (Emory University, Atlanta, GA). A
DNA fragment (300 bp) was purified from a Bgl II/EcoR I
digest of the rGSTA2 plasmid that hybridized with rGSTA1 and rGSTA2
transcripts, the predominant alpha class isozymes in rat liver (20).
These transcripts run as a single band on formaldehyde-agarose gels,
and no other bands were detected. To acknowledge that the probe
hybridizes with both rGSTA1 and A2 transcripts, we have used rGSTA1/A2
when referring to transcript levels.
RNA blots were prehybridized for 4 h at 60°C in hybridization
solution (7% SDS, 1% BSA, 10% polyethylene glycol, 1.25 mM EDTA, 0.25 M NaCl, 0.125 M sodium phosphate buffer; pH
6.50). Radiolabeled cDNA probes were added and incubated overnight.
Blots were washed three times at 65°C with 1× SSC for 30 min
and exposed to film. The resulting autoradiograms were scanned in a
densitometer as described previously (15). Blots were stripped in
boiling 1% SDS/0.1× SSC for reprobing. Normalizing the GST
transcript signal densities to those of pCS14 controlled for intrablot
variability in RNA loading.
Nuclear runoff assays.
Assays were performed as described previously (15). In brief, nuclei
were isolated from whole liver or from periportal or perivenous
enriched hepatocyte preparations from control and phenobarbital-treated animals. The nuclei were isolated in a sucrose gradient, resuspended, and snap frozen in a glycerol buffer (75 mM HEPES, pH 7.5, 60 mM KCl,
15 mM NaCl, 0.1 mM EDTA, 0.1 mM EGTA, and 40% glycerol). Nuclei were
used less than 30 days after isolation. After thawing on
ice, ~107 nuclei (~250 µg DNA) were added to
transcription buffer that contained [32P]UTP.
The mixture was incubated at 26°C for 45 min. The reaction was
stopped, and labeled RNA was extracted by the addition of TRI Reagent
(Molecular Research Center, Cincinnati, OH). Cloned DNA plasmids (5 µg) were linearized, denatured, and transferred onto nitrocellulose
filters, which were then baked at 80°C for 2 h. Aliquots of labeled
RNA (1 × 107 cpm) were hybridized with the
cDNA-containing filters at 60°C for 60 h in 2 ml of hybridization
buffer. The blots were washed twice for 20 min each in 2× SSC at
60°C, treated with RNase A (10 µg/ml) for 30 min at 37°C,
washed a third time in 1× SSC for 15 min at 37°C, and then
exposed to film. The resulting bands on autoradiographs were quantified
by densitometry.
35S-labeled RNA probe synthesis for in situ
hybridization.
Sense (control) and antisense RNA probes labeled with
[35S]UTP for in situ hybridization were
synthesized on linearized rGSTA2 plasmid templates with riboprobe
transcription reagents (Promega) (16). A cDNA fragment (530 bp) from a
Pst I digest of the rGSTA2 plasmid was subcloned into the
Pst I site of pBluescript II KS+ plasmid
(Stratagene), and the orientation of the cDNA was
determined by sequencing the resulting construct. To synthesize an
antisense RNA probe specific only for rGSTA1 and A2 transcripts, the
plasmid was linearized with Bgl II (sequence was same as that
used for Northern blots), which, after in vitro transcription, resulted in a 377-bp RNA probe. The sense probe was synthesized after the plasmid was linearized with EcoR I, which resulted in a 615-bp RNA probe. The lengths of the RNA probes were confirmed by
autoradiography after PAGE.
Liver sections for in situ hybridization.
Liver sections were prepared and hybridizations were performed as
described previously (16). Frozen pieces of liver were sectioned (6 µm) in a cryostat, and sections were mounted on slides (Fisher
Superfrost/Plus) and immediately placed on dry ice; slides were stored
at 80°C over desiccant until used (<3 days). Sections were
thawed and then fixed with 4% paraformaldehyde in phosphate buffer (pH
7.4) for 10 min at 4°C. The sections were washed in 0.5× SSC
for 5 min and treated with proteinase K (1 µg/ml) in 0.5 M NaCl-0.01
M Tris buffer (pH 8.0) for 5 min at room temperature. The slides were
washed in 0.5× SSC and placed in air-tight hybridization boxes
containing filter paper saturated with buffer (4× SSC-50% formamide). Prehybridization solution (100 µl of 0.01 M
dithiothreitol, 0.3 M NaCl, 0.02 M Tris, 0.05 M EDTA, 1×
Denhardt's solution, 10% dextran sulfate, and 50% formamide, pH 8.0)
was placed on each section for 3 h at 42°C. Radiolabeled RNA probe
(2 µl at 300,000 cpm/µl in 10 mM Tris-1 mM EDTA, pH 8.0 ) and yeast
tRNA (1 µl of 2.5 mg/ml in water) was added to 17 µl of
hybridization buffer, placed on each section, and incubated overnight
at 55°C. The slides were washed and treated with RNase A (20 µg/ml in 0.5 M NaCl-0.01 M Tris, pH 8.0) and then washed again.
Finally, the tissue sections were dehydrated in a graded alcohol series
containing 0.3 M sodium acetate and dried under vacuum in a desiccator.
The slides were dipped in Kodak Type NTB2 emulsion (diluted 1:1 with water) at 42°C and left to dry in the dark for 2 h. The
emulsion-coated slides were sealed in slide boxes with desiccant at
4°C for 4 days of exposure and then were developed. A liver section
from a saline-treated animal and a corresponding section from a
phenobarbital-treated animal were placed together on each slide, and
treatment groups and their controls were all run in the same batch to
control for possible variations in thickness of the emulsions between
slides and batches. The slides were developed in Kodak D19 developer for 3 min, rinsed in water for 20 s, fixed in Kodak Fixer for 3 min,
and washed in water three times for 5 min each. Sections were
counterstained with hematoxylin and eosin (28, 29).
Western blots.
Cytosols from livers of phenobarbital-treated and
saline-treated rats were prepared, and Western blots were made as
previously described (16). Protein concentrations were determined by
the Bio-Rad protein assay, with BSA as the standard. Cytosolic proteins (5 µg) from individual animals were loaded onto SDS gels for
electrophoresis. After electrophoresis, the proteins were
electroblotted onto Immobilon-P membranes (Millipore, Bedford, MA). The
primary antiserum used for GST detection was rabbit anti-rat
GSTA1/A2/A3 antiserum (MED 26 YA, 1:3,000; Biotrin, Dublin, Ireland).
Bound primary antibodies were detected with enhanced chemiluminescence
reagents (Amersham or NEN Life Science). The densities of
the resultant bands were quantitated with a densitometer. Although the
primary antibody reacts with rGSTA1, A2, and A3, complete separation of
rGSTA3 from A1 and A2 was accomplished by electrophoresis (4).
Therefore, the densities reported reflect changes in rGSTA1 and A2
protein levels only.
Data analysis.
Northern and Western blot analyses were performed on individual
animals. Values are means ± SE of at least three animals unless otherwise stated. Samples from control and phenobarbital-treated animals were run on the same blots and compared with each other. In
situ hybridization sections were photographed, and fields were digitized and analyzed by computer image analysis software (Matrox Inspector, version 1.7, Matrox, QC, Canada). Briefly, high-power fields
(×119) were analyzed by dividing the lobules into equal thirds to
represent zone 1 (periportal), zone 2, and zone
3 (perivenous). The total area of exposed silver grains was then
measured in a constant-size sample area of each zone and averaged from
three different lobules from each of at least three different
phenobarbital- or saline-treated animals (nine lobules per time
period). Data were analyzed statistically by two-way ANOVA or
Student's t-test where appropriate. Statistical significance
was achieved when P 
0.05.
| |
RESULTS |
Effect of phenobarbital on mRNA, protein levels, and transcriptional
activity of rGSTA1/A2.
Previous studies have shown that a single dose of phenobarbital causes
an increase in the mRNA and protein levels of rGSTA1/A2 in rat liver
(20). In addition, transcriptional activity increases by 2 h and
remains elevated for at least 16 h after the administration of
phenobarbital (8). We established that a similar series of events
occurred in the present study. After a single dose of phenobarbital,
liver rGSTA1/A2 mRNA levels increased 2.5 ± 0.24-fold (P < 0.01; n = 3) at 24 h. Protein levels of rGSTA1/A2 also
increased 1.2 ± 0.09-fold (P < 0.02; n = 4) at 24 h. Transcriptional activity of rGSTA1/A2 was measured by
nuclear runoff assay at 12, 24, and 48 h after phenobarbital treatment
and increased 3.7-, 4.6-, and 5.5-fold, respectively, in treated
animals compared with control animals (n = 2 at each
time point; Fig. 1).
Transcriptional activity for albumin was unaffected by treatment with
phenobarbital, and differences were 0.49-, 0.99-, and 0.94-fold
relative to untreated controls at 12, 24, and 48 h, respectively.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 1.
Nuclear runoff assays from nuclei of whole livers from control and
phenobarbital-treated animals. Hepatocyte nuclei were isolated from
saline-treated animals (A) and from phenobarbital-treated
animals after 12 (B), 24 (C), or 48 h (D) to
measure transcriptional activity by nuclear runoff assay. Hybridization
probes were linearized plasmids containing cloned genes for rat liver
glutathione S-transferase A1/A2 (rGSTA1/A2), albumin,
glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and ribosomal protein
S14 (latter two as positive controls) plus the parent plasmids lacking
genes (negative controls). 1, pBluescript (negative control for
rGSTA1/A2); 2, rGSTA1/A2; 3, pBR322 (negative control
for GAPDH and albumin); 4, albumin; 5, GAPDH;
6, pGEM4Z (negative control for S14); 7, S14.
|
|
Effect of phenobarbital on lobular distribution of rGSTA1/A2.
The induction of rGSTA1/A2 transcripts by 3-methylcholanthrene is
greatest in the perivenous region of the hepatic lobule (19). We
examined the effect of phenobarbital on lobular induction of rGSTA1/A2
by in situ hybridization. The area covered by silver grains in the
lobules of saline-treated animals using the sense probe was three- to
fivefold less than the area covered when the antisense probe was used.
In addition, with either probe there was no significant intralobular
variability (Fig. 2). After treatment with
phenobarbital, the silver grain area seen with the sense probe did not
change compared with that of saline-treated animals [e.g., saline
vs. phenobarbital (48 h): zone 1 = 185 ± 17 vs. 152 ± 9;
zone 2 = 143 ± 11 vs. 140 ± 11; zone 3 = 147 ± 8 vs. 136 ± 10; means ± SE; not significant]. In
contrast, when the antisense probe was used there was an increase in
rGSTA1/A2 transcripts (Fig. 2). The earliest increase in transcript
levels was seen in zones 2 and 3 3 h after
phenobarbital treatment, and increases were present in all three zones
12 h after phenobarbital treatment. The increase in transcript levels
was greatest in zone 3 and least in zone 1 (Figs.
3 and 4). The
time course of induction in zones 1 and 3 relative to
control animals is shown in Fig. 5. The
greatest increase in transcript levels was seen in zone 3 at 12 h (11-fold compared with saline-treated animals), which then
decreased to 9-fold at 48 h. The relative induction in zone 1 peaked at a 3.8-fold increase at 24 h.
View larger version (166K):
[in this window]
[in a new window]
|
Fig. 2.
In situ hybridization using sense and antisense probes. In situ
hybridization was performed on liver sections obtained from animals
treated for 48 h with saline or phenobarbital. A:
saline-treated animal (sense); B:
phenobarbital-treated animal (sense); C: saline-treated
animal (antisense); D: phenobarbital-treated animal
(antisense). Reflective silver grains localize rGSTA1/A2 expression.
Sections were 6-µm thick (×119). CV, central vein.
|
|
View larger version (149K):
[in this window]
[in a new window]
|
Fig. 3.
Lobular induction of rGSTA1/A2 by phenobarbital. A: control
animal 24 h after injection of saline. B: experimental animal
24 h after injection of phenobarbital. Reflective silver grains
localize rGSTA1/A2 message. Sections were 6-µm thick (×119). c,
Central vein; p, portal vein.
|
|
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 4.
Effect of a single dose of phenobarbital on lobular distribution of
rGSTA1/A2. Liver sections were subjected to in situ
hybridization and were then photographed as described in
METHODS. Liver lobule was divided into 3 equal regions
(zone 1, A; zone 2, B;
zone 3, C), and average area of silver
grains/field in each region was determined by computer image analysis.
Values are means ± SE of 3 lobules from 3 different animals.
* P 0.001 vs. same zone in saline-treated animals
(ANOVA);  P 0.05 vs. zone 1 in
phenobarbital-treated animals; § P 0.02 vs.
zone 2 in phenobarbital-treated animals.
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 5.
Relative transcript levels in zones 1 and 3 of
phenobarbital- vs. saline-treated animals. Mean areas of silver grains
representing rGSTA1/A2 transcripts in these zones were obtained for
each time point, and relative magnitude (area in treated/area in
controls) of induction was calculated.
|
|
The slower rate of increase of transcript levels in zone 1 relative to zone 3 suggested that a delay in response or
insensitivity to phenobarbital might account for the observed
differences. We administered a second dose of phenobarbital 12 or 24 h
after the first and then performed in situ hybridization either 12 or
24 h later. Twelve hours after the second dose of phenobarbital, mRNA
levels increased 1.7-fold, whereas 24 h after the second dose of
rGSTA1/A2 mRNA levels increased 2.6-fold. The increase in transcripts
was greatest in zone 3 and least in zone 1. In fact,
zone 1 hepatocytes showed relatively less induction following two doses of phenobarbital than they did following one dose (Fig. 6).
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 6.
Effect of two doses of phenobarbital on lobular distribution of
rGSTA1/A2. See Fig. 4 for details. Animals received a second dose of
phenobarbital 12 or 24 h after first dose and then were killed either
12 or 24 h later. Values are means ± SE of 3 lobules from 3 different
animals. * P 0.001 vs. same region in saline-treated
animals; P 0.01 vs. zone 1 in
phenobarbital-treated animals. A: zone 1;
B: zone 2; C: zone 3.
|
|
Transcriptional activity in periportal and perivenous hepatocytes
after treatment with phenobarbital.
We questioned whether differences in transcriptional activity accounted
for the higher transcript levels of rGSTA1/A2 in perivenous vs. periportal hepatocytes after treatment with phenobarbital. Fractions enriched in either periportal or perivenous hepatocytes were
used to perform nuclear runoff assays. Transcriptional activity in the
periportal hepatocytes was similar to that in perivenous hepatocyte
preparations at 3 h or 12 h after administration of phenobarbital with
all cDNA probes used. No increase in transcriptional activity in
perivenous hepatocytes relative to periportal hepatocytes for rGSTA1/A2
was observed (Table 1).
Effect of phenobarbital on mRNA half-life of rGSTA1/A2.
Since the increase in transcript levels in zone 3 relative to
zone 1 after treatment with phenobarbital was not due to a
difference in transcriptional activity, we questioned whether
phenobarbital might have affected the posttranslational stability of
rGSTA1/A2 transcripts. Animals were treated either with phenobarbital
or saline and 6 h later given -amanitin and actinomycin D to block transcription. The half-life of rGSTA1/A2 transcripts in saline-treated animals was 3.6 h, whereas the half-life in phenobarbital-treated animals was 10.2 h (Fig. 7).
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 7.
mRNA half-life of rGSTA1/A2 6 h after injection of phenobarbital or
saline. mRNA half-life was determined as described in
METHODS. Values are means ± SE of 3 different animals.
Lines were drawn by linear regression analysis, and slopes were
significantly different (P < 0.05).
|
|
| |
DISCUSSION |
The induction of detoxication enzymes by xenobiotics has been the focus
of extensive investigation for many years. Of particular interest has
been the finding that much of the induction was due to regulation of
rates of transcription through response elements in the
5'-flanking sequences of the enzyme genes. The rGSTA2 gene has
been extensively studied in this regard, and its 5'-flanking sequence contains regulatory elements that are responsive to
xenobiotics such as phenobarbital and 3-methylcholanthrene (11, 18).
Studies in animals suggest that all of the increase in steady-state
levels of GST mRNA after treatment with phenobarbital can be accounted for by an increase in rates of transcription (8, 19, 20). However,
studies in cultured cells suggest that the effects of phenobarbital may
be more complex with both transcriptional and posttranscriptional
events accounting for the increase in rGSTA1/A2 mRNA levels
(26).
In the current study, phenobarbital was shown to increase rGSTA1/A2
transcript levels by increasing the transcriptional activity of the
gene in rat liver. This increase in transcriptional activity appeared
to account for the panlobular induction of rGSTA1/A2; however, it does
not account for the greater increase in rGSTA1/A2 transcripts that
developed in zones 2 and 3. The overall increase in
transcriptional activity was similar to the increase in transcript levels in zone 1. However, the increase in transcript levels in zone 3 was much greater than that in zone 1, despite
the fact that rates of transcription in periportal and perivenous
hepatocytes were similar. Thus posttranscriptional events rather than
an increase in rates of transcription must account for the greater
induction of rGSTA1/A2 in perivenous relative to periportal hepatocytes.
Posttranscriptional events have a significant impact on mRNA levels and
are important in gene regulation. Changes in mRNA half-lives occur in
response to alterations in cell growth, viral infections, and exposure
to toxins (2, 22). For example, changes in gene expression during liver
regeneration are mediated frequently by changes in mRNA stability and
not by alterations in rates of transcription (13). Similarly, when the
mechanism behind the increase in collagen synthesis by hepatic stellate cells was examined, an increase in mRNA stability accounted for most of
the increase in collagen mRNA (23). In the current investigation, the
mRNA half-life of rGSTA1/A2 increased more than threefold in the
phenobarbital-treated animals compared with untreated animals. Previous
studies in cultured cells also suggest that phenobarbital stabilizes
GST message (26). We have not measured rGSTA1/A2 mRNA half-lives in
periportal vs. perivenous hepatocytes for technical reasons, but it is
reasonable to assume that they are different in the absence of
differences in rates of transcription. A three- to fourfold increase in
rates of transcription in concert with a threefold increase in mRNA
half-life could have led to the ~11-fold increase in mRNA levels that
was observed in the perivenous hepatocytes.
Changes in mRNA stability are commonly mediated at the
3'-untranslated (3'-UTR) region of the mRNA. Deadenylation,
binding of proteins to the poly(A)+ tail, or
endonucleolytic cleavage within the coding region are mechanisms that
may affect mRNA stability (2, 22). The binding of specific proteins to
the poly(A)+ tail or to regions rich in
adenosine or cytosine in the 3'-UTR stabilizes mRNA (2, 22).
Treatment of mice with the inducing agents pyrazole or
3-methylcholanthrene leads to induction of specific proteins that bind
to the 3'-UTR of CYP2a5 and CYP1a2 mRNA (9, 21). These proteins
are thought to stabilize the mRNA and therefore lead to an increase in
steady-state transcript levels. Phenobarbital may also induce a protein
that binds to the 3'-UTR of rGSTA1/A2 mRNA, decreasing rates of
degradation of the message. If perivenous hepatocytes produce more of
these proteins in response to phenobarbital and 3-methylcholanthrene than periportal hepatocytes, then transcript levels will be higher in
perivenous relative to periportal hepatocytes. Identification of these
proteins will be required before this hypothesis can be established
with certainty.
In the current study we questioned whether the induction by
phenobarbital of rGSTA1/A2 in periportal hepatocytes was delayed relative to the induction in perivenous hepatocytes. We therefore administered a second dose of phenobarbital to determine if a further
induction of transcript levels above that induced by a single dose
could occur in periportal hepatocytes. However, the mRNA induction in
perivenous hepatocytes was always greater than that in periportal cells
whether one or two doses of phenobarbital were administered. These
results suggest that the difference in rGSTA1/A2 induction between
perivenous and periportal hepatocytes are intrinsic to the cells and do
not reflect their position within the lobule or how the phenobarbital
is administered. This conclusion is consistent with previous work in
which increased expression of CYP was greater in perivenous
hepatocytes. Transplanting the cells into another location (the spleen)
did not affect the induction of the CYP enzyme by phenobarbital, which
supports the idea that the difference in periportal and perivenous
hepatocytes is intrinsic to the different cell populations (1, 7,
17, 30). However, in contrast to previous investigators, we
believe posttranscriptional and not transcriptional events account for
the difference.
In conclusion, treatment of rats with phenobarbital causes a nonuniform
increase in transcript levels of rGSTA1/A2 in the hepatic lobule. The
greatest increase in rGSTA1/A2 transcript levels was found in
perivenous hepatocytes and exceeded that which could be accounted for
by an increase in the rate of transcription, because transcriptional
activity was uniform across the hepatic lobule. Phenobarbital also
increased the half-life of the rGSTA1/A2 message. We believe that this
increase in message half-life accounts for the higher transcript levels
in perivenous relative to periportal hepatocytes. Further studies are
required to define exactly how phenobarbital stabilizes rGSTA1/A2
message differentially in perivenous and periportal hepatocytes.
| |
ACKNOWLEDGEMENTS |
We are grateful to L. T. Tucker for secretarial assistance and to
Drs. E. T. Morgan and J. N. Wilcox. This work was supported in part by
National Institutes of Health Grant GM-31555 (to T. D. Boyer).
| |
FOOTNOTES |
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement"
in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Address for reprint requests and other correspondence: T. D. Boyer,
Emory Univ. School of Medicine, 2101 Woodruff Memorial Bldg., 1639 Pierce Dr., Atlanta, GA 30322 (E-mail: tboyer{at}emory.edu).
Received 26 April 1999; accepted in final form 4 November 1999.
| |
REFERENCES |
1.
Bars, RG,
Bell DR,
Elcombe CR,
Oinonen T,
Jalava T,
and
Lindros KO.
Zone-specific inducibility of cytochrome P450 2B1/2 is retained in isolated perivenous hepatocytes.
Biochem J
282:
635-638,
1992.
2.
Beelman, CA,
and
Parker R.
Degradation of mRNA in eukaryotes.
Cell
81:
179-183,
1995[ISI][Medline].
3.
Bilir, BM,
Gong TWL,
Kwasiborski V,
Shen CS,
Fillmore CS,
Berkowitz CM,
and
Gumucio JJ.
Novel control of the position-dependent expression of genes in hepatocytes: the GLUT-1 transporter.
J Biol Chem
268:
19776-19784,
1993[Abstract/Free Full Text].
4.
Boyer, TD,
Kenney WC,
and
Zakim D.
Structural, functional and hybridization studies of the glutathione S-transferases of rat liver.
Biochem Pharmacol
32:
1843-1850,
1983[ISI][Medline].
5.
Branum, GD,
Selim N,
Liu X,
Whalen R,
and
Boyer TD.
Ischaemia and reperfusion injury of rat liver increases expression of glutathione S-transferase A1/A2 in zone 3 of the hepatic lobule.
Biochem J
330:
73-79,
1998.
6.
Cereghini, S,
Raymondjean M,
Alejandro JC,
Herbomel P,
and
Yaniv M.
Factors involved in control of tissue-specific expression of albumin gene.
Cell
50:
627-638,
1987[ISI][Medline].
7.
Chianale, J,
Dvorak C,
May M,
and
Gumucio JJ.
Heterogeneous expression of phenobarbital-inducible cytochrome P-450 genes within the hepatic acinus in the rat.
Hepatology
6:
945-951,
1986[ISI][Medline].
8.
Ding, VD-H,
and
Pickett CB.
Transcriptional regulation of rat liver glutathione S-transferase genes by phenobarbital and 3-methylcholanthrene.
Arch Biochem Biophys
240:
553-559,
1985[ISI][Medline].
9.
Geneste, O,
Raffalli F,
and
Lang MA.
Identification and characterization of a 44 kDa protein that binds specifically to the 3'-untranslated region of CYP2a5 mRNA: inducibility, subcellular distribution and possible role in mRNa stabilization.
Biochem J
313:
1029-1037,
1996.
10.
Gumucio, J.
Hepatocyte heterogeneity: the coming of age from the description of a biological curiosity to a partial understanding of its physiological meaning and regulation.
J Hepatol
9:
154-160,
1989.
11.
Hayes, JD,
and
Pulford DJ.
The glutathione S-transferase supergene family: regulation of GST and the contribution of the isoenzymes to cancer chemoprotection and drug resistance.
Crit Rev Biochem Mol Biol
30:
445-600,
1995[ISI][Medline].
12.
Jones, L.
Hepatology: A Textbook of Liver Disease. Philadelphia: W. B. Saunders, 1996, p. 3-32.
13.
Kren, BT,
and
Steer CJ.
Posttranscriptional regulation of gene expression in liver regeneration: role of mRNA stability.
FASEB J
10:
559-573,
1996[Abstract].
14.
Kren, BT,
Trembley JH,
and
Steer CJ.
Alterations in mRNA stability during rat liver regeneration.
Am J Physiol Gastrointest Liver Physiol
270:
G763-G777,
1996[Abstract/Free Full Text].
15.
Lee, SJ,
and
Boyer TD.
The effect of hepatic regeneration on the expression of glutathione S-transferases.
Biochem J
293:
137-142,
1993.
16.
Lee, SJ,
Friedman SL,
Whalen R,
and
Boyer TD.
Cellular sources of glutathione S-transferase P in primary cultured rat hepatocytes: localization by in situ hybridization.
Biochem J
299:
79-83,
1994.
17.
Maganto, P,
Traber PG,
Rusnell C,
Dobbins WO,
Keren D,
and
Gumucio JJ.
Long-term maintenance of the adult pattern of liver-specific expression for P-450b, P-450e, albumin and -fetoprotein genes in intrasplenically transplanted hepatocytes.
Hepatology
11:
585-592,
1990[ISI][Medline].
18.
Mannervik, B,
and
Danielson UH.
Glutathione transferases
structure and catalytic activity.
Crit Rev Biochem Mol Biol
23:
283-337,
1988.
19.
Paulson, KE,
Darnell JE, Jr,
Rushmore T,
and
Pickett CB.
Analysis of the upstream elements of the xenobiotic compound-inducible and positionally regulated glutathione S-transferase Ya gene.
Mol Cell Biol
10:
1841-1852,
1990[Abstract/Free Full Text].
20.
Pickett, CB,
Telakowski-Hopkins CA,
Ding GJ-F,
Argenbright L,
and
Lu AYH
Rat liver glutathione S-transferases. Complete nucleotide sequence of a glutathione S-transferase mRNA and the regulation of the Ya, Yb, and Yc mRNAs by 3-methylcholanthrene and phenobarbital.
J Biol Chem
259:
5182-5188,
1984[Abstract/Free Full Text].
21.
Raffalli-Mathieu, F,
Geneste O,
and
Lang MA.
Characterization of two nuclear proteins that interact with cytochrome P-450 1A2 mRNA. Regulation of RNA binding and possible role in the expression of the Cyp1a2 gene.
Eur J Biochem
245:
17-24,
1997[ISI][Medline].
22.
Ross, J.
mRNA stability in mammalian cells.
Microbiol Rev
59:
423-450,
1995[Abstract/Free Full Text].
23.
Stefanovic, B,
Hellerbrand C,
Holcik M,
Briendl M,
Liebhaber SA,
and
Brenner DA.
Posttranscriptional regulation of collagen 1(I) mRNA in hepatic stellate cells.
Mol Cell Biol
17:
5201-5209,
1997[Abstract].
24.
Telakowski-Hopkins, CA,
Rodkey JA,
Bennett CD,
Lu AYH,
and
Pickett CB.
Rat liver glutathione S-transferases. Construction of a cDNA clone complementary to a Yc mRNA and prediction of the complete amino acid sequence of a Yc subunit.
J Biol Chem
260:
5820-5825,
1985[Abstract/Free Full Text].
25.
Tordjmann, T,
Berthon B,
Lardeux B,
Moreau A,
Jacquemin E,
Combettes L,
Feldmann G,
and
Claret M.
An improved digitonin-collagenase perfusion technique for the isolation of periportal and perivenous hepatocytes from a single rat liver: physiological implications for lobular heterogeneity.
Hepatology
26:
1592-1599,
1997[ISI][Medline].
26.
Vandenberghe, Y,
Tee L,
Morel F,
Rogiers V,
Guillouzo A,
and
Yeoh G.
Regulation of glutathione S-transferase gene expression by phenobarbital in cultured adult rat hepatocytes.
FEBS Lett
284:
103-108,
1991[ISI][Medline].
27.
Whalen, R
and Boyer TD. Human glutathione S-transferases.
Semin Liver Dis
18:
345-358,
1998[ISI][Medline].
28.
Wilcox, JN,
Smith KM,
Schwartz SM,
and
Gordon D.
Localization of tissue factor in the normal vessel wall and in the atherosclerotic plaque.
Proc Natl Acad Sci USA
86:
2839-2843,
1989[Abstract/Free Full Text].
29.
Wilcox, JN,
Smith KM,
Williams LT,
Schwartz SM,
and
Gordon D.
Platelet-derived growth factor mRNA detection in human atherosclerotic plaques by in situ hybridization.
J Clin Invest
82:
1134-1143,
1988.
30.
Wojcik, E
, Dvorak C, Chianale J, Traber PG, Keren D, and Gumucio JJ. Demonstration by in situ hybridization of the zonal modulation of rat liver cytochrome P-450b and P-450e gene expression after phenobarbital.
J Clin Invest
82:
658-666,
1988.
Am J Physiol Gastrointest Liver Physiol 278(4):G542-G550
0193-1857/00 $5.00
Copyright © 2000 the American Physiological Society
TOP
ABSTRACT
INTRODUCTION
