Different hepatocytes express the cholesterol 7 alpha-hydroxylase gene during its circadian modulation in vivo.

Cholesterol 7 alpha-hydroxylase, the rate-limiting enzyme in bile salt synthesis from cholesterol is a P450 enzyme (CYP7A). Its expression and activity are regulated by bile salts, cholesterol, hormones and a circadian modulator. Here we define the hepatocytes contributing to the expression of the rat CYP7A gene during its in vivo circadian variation. The diurnal expression of the CYP7A messenger RNA (mRNA) was studied by in situ hybridization and correlated with the diurnal rate of CYP7A gene transcription and mRNA expression. At 10 AM, the time of lowest mRNA expression and gene transcription rate, only four to five hepatocytes, located close to the hepatic venules ("perivenular"), contained the CYP7A mRNA. At 10 PM, the time of highest mRNA expression and fastest in vitro transcription rate, approximately one half of the hepatocytes (still in a "perivenular" location) contained the cholesterol 7 alpha-hydroxylase mRNA. In addition, the measured half-life of the CYP7A mRNA was shorter at 10 AM than at 10 PM suggesting that posttranscriptional mechanisms also contributed to the observed circadian differences. Therefore, the basal transcription rate of the CYP7A gene is maintained by four to five "perivenular" hepatocytes. During the circadian variation, the rate of gene transcription increases in these "perivenular" hepatocytes, but in addition, there is recruitment of other more proximal hepatocytes to transcribe the gene. It is proposed here that the response of specific hepatocytes to the various modulators of CYP7A gene expression is dependent on the relative position of these hepatocytes within the liver cell plate.

Possible nizatidine-induced subfulminant hepatic failure.

Subfulminant hepatic failure has not been reported to occur with the histamine-2 receptor antagonists. We report a possible case of nizatidine-induced subfulminant hepatic failure leading to the eventual development of cirrhosis.

Novel control of the position-dependent expression of genes in hepatocytes. The GLUT-1 transporter.

The basal hepatocyte phenotype is conferred by the expression of liver-specific genes. In the adult liver, the basal hepatocyte phenotype is further modified by transcriptional and post-transcriptional regulation of genes which result in the appearance of specific proteins in selected hepatocytes. One of these proteins is the erythroid/brain or GLUT-1 glucose transporter. The GLUT-1 protein is detected in the plasma membrane of only one or two hepatocytes located at the end of the liver cell plate, contiguous to the hepatic venule. The objective of this study was to define the molecular mechanisms responsible for the restricted expression of the GLUT-1 protein in rat liver. Hepatocytes were isolated from either the proximal ("periportal") or the distal ("perivenular") half of the liver cell plate. The GLUT-1 mRNA as well as the GLUT-1 protein content and intracellular distribution were defined after subcellular fractionation of each hepatocyte population. In addition, the location of the GLUT-1 protein in liver tissue was determined by confocal microscopy. We propose that the GLUT-1 gene is transcribed and the mRNA is translated by both "periportal" and "perivenular" hepatocytes. However, insertion of the GLUT-1 protein into the plasma membrane occurs only in the last two hepatocytes contiguous to the hepatic venule. In other hepatocytes, the protein remains in a different cellular compartment characterized here as a "low density microsomal" fraction.

P450IIB gene expression in rat small intestine: cloning of intestinal P450IIB1 mRNA using the polymerase chain reaction and transcriptional regulation of induction.

Intestinal cytochromes P450 (P450) may function in the "first pass" metabolism of drugs, the detoxification of xenobiotics, or the activation of carcinogens. However, little is known about the expression of specific P450 genes in intestinal mucosa. We have previously shown that a P450 mRNA that is homologous to rat liver P450IIB1 (P450b) is expressed in rat small intestine and is inducible by phenobarbital, polyhalogenated biphenyls, and organochlorine pesticides. However, there are multiple highly homologous genes in the IIB subfamily and, therefore, studies using liver-derived cDNAs or oligonucleotides based on those cDNAs cannot definitively establish the identity of the intestinal mRNA(s). The polymerase chain reaction was used to enzymatically amplify cDNA synthesized from intestinal and hepatic RNA, and the amplified segments were identified by Southern blot analysis. These studies demonstrated that the amplified segment of the phenobarbital-inducible P450 mRNA in intestine was identical to this same segment of the hepatic P450b mRNA; furthermore, this analysis showed that P450e was not expressed in intestine. To definitively establish the identity of the intestinal mRNA, the full coding sequence of the P450b mRNA was cloned from intestinal and hepatic RNA and sequenced. The sequences of the intestinal and hepatic cDNA were identical and coded for P450b; the deduced protein sequence in the F344 rat differed in one amino acid from the reported sequence in Sprague-Dawley rats and, thus, represents a different allele of the same gene. An increment in intestinal P450b mRNA was detected as early as 1 hr following a single intraperitoneal injection of phenobarbital; this prompt rise in mRNA suggested that transcriptional activation may be the primary mechanism for induction. Nuclear run-on experiments were performed using nuclei isolated from intestinal mucosa 3 and 6 hr following treatment with phenobarbital. The rate of transcription of the P450IIB1 gene was increased approximately 6-fold 6 hr following phenobarbital; this was very similar to the increment in P450b mRNA as measured by quantitative dot blot analysis. Therefore, the predominant mechanism for the induction of P450b mRNA in intestine in response to phenobarbital was an increase in gene transcription. These studies indicate that the same member of the P450IIB subfamily, P450IIB1 or P450b, is expressed and inducible by similar mechanisms in small intestine and liver. Although putative P450b mRNA and apoprotein have been identified in lung and testes, the capacity for induction by phenobarbital, and presumably other xenobiotics, is unique to liver and intestine.(ABSTRACT TRUNCATED AT 400 WORDS)

Long-term maintenance of the adult pattern of liver-specific expression for P-450b, P-450e, albumin and alpha-fetoprotein genes in intrasplenically transplanted hepatocytes.

Hepatocytes isolated from livers of Fischer 344 rats and transplanted into the spleens of rats from the same strain survived for at least 15 mo in the absence of immunosuppressive drugs. Hepatocytes attached themselves only in the red pulp of the spleen, most commonly in clumps without a discernible structure. Throughout the 15-mo period, intrasplenically transplanted hepatocytes expressed cytochrome P-450b, P-450e and albumin messenger RNAs, whereas alpha-fetoprotein messenger RNA was not expressed. In addition, the relative expression of albumin and P-450 genes was similar to that in liver. For example, albumin messenger RNA was expressed to higher levels than P-450b or e messenger RNAs. Northern blots hybridized with oligonucleotides specific for P-450b or P-450e showed that, as in liver, both P-450b and P-450e genes were induced in response to phenobarbital. Quantitative slot-blot hybridizations performed at 15 days and 1, 6, and 15 mo after hepatocyte transplantation revealed that cytochrome P-450b and P-450e messenger RNAs were induced about 20- to 30-fold by a single dose of phenobarbital. This level of induction was also similar to that observed in liver. Hence, intrasplenically transplanted hepatocytes represent a unique system in which hepatocytes, cultured in an extrahepatic in vivo environment, maintain for at least 15 mo a pattern of expression for these four liver genes similar to that in the adult liver. Moreover, these studies suggest that neither the organization of liver into acini nor a specific zonal sinusoidal microenvironment is necessary for adult hepatocytes to respond to phenobarbital with induction of P-450b and P-450e genes.

Induction of P-450IIB genes within the rat liver acinus is not dependent on the chemical inducer or on the acinar organization.

P-450IIB genes (P-450b and P-450e in rat) are induced following treatment with phenobarbital predominantly in hepatocytes located in zones 2 and 3 of the liver acinus. The previous finding of phenobarbital-mediated induction of P-450IIB mRNAs and apoproteins in the same zonal hepatocytes suggests that this differential induction is most likely due to zonal differences in the activation of gene transcription. To determine whether differential P-450IIB gene transcription was dependent on the type of inducer (i.e., inducer specificity) or on the capacity of hepatocytes located in different acinar zones to respond to inducers (i.e. zonal specificity), the pattern of acinar induction was evaluated following treatment with inducers that have diverse physiochemical properties and inductive capacities. The zonal distribution of P-450b and P-450e mRNAs in liver was determined by in situ hybridization after the administration of either phenobarbital, polychlorinated biphenyls, chlordane, or chlorpromazine. Liver sections were hybridized with 3H-labeled RNA transcripts of a P-450e cDNA that recognizes sequences of both P-450b and P-450e mRNAs and the pattern of zonal mRNA induction was measured by quantitative image analysis. Each inducer increased P-450b,e mRNA levels predominantly in hepatocytes of zones 2 and 3 of the hepatic acinus. The P-450b and P-450e apoproteins were induced in the same zonal hepatocytes as the P-450b,e mRNAs as shown by immunofluorescence studies using monoclonal antibodies. Therefore, differential transcriptional induction of the P-450IIB genes in the liver acinus does not seem to be dependent on a specific chemical inducer, but rather is a characteristic capacity of hepatocytes located in different acinar zones. To determine whether the induction of P-450b and P-450e was dependent on the liver tissue organization, hepatocytes were isolated by collagenase perfusion of the liver and transplanted into the spleens of syngeneic rats. Induction of P-450b and P-450e mRNAs and apoproteins, assessed 1 month after transplantation, was evident with each of the chemical inducers. These data suggest that, once hepatocytes have attained the capacity to respond to inducers, the organization of hepatocytes into acini and the sinusoidal microenvironment of the liver are not required for hepatocytes to maintain the ability to respond to inducers with an increase in transcription of P-450IIB genes. Moreover, the splenic hepatocytes demonstrated a heterogeneous pattern of P-450b,e apoprotein induction; this raises the possibility that the acinar organization is also not required for the heterogenous expression of P-450IIB genes.

Demonstration by in situ hybridization of the zonal modulation of rat liver cytochrome P-450b and P-450e gene expression after phenobarbital.

The various physiological processes that constitute liver function are compartmentalized within the hepatic acinus. The molecular mechanisms modulating the development and maintenance of this hepatocyte heterogeneity have not been defined. The objective of this study was to determine whether transcriptional or posttranscriptional zonal modulation of cytochromes P-450b,e gene expression was responsible for the heterogeneous induction of the P-450 proteins, which is observed after phenobarbital (PB) administration. The exact localization in liver tissue of hepatocytes responding to PB with induction of either P-450b,e mRNA or proteins was established by in situ hybridization and by immunofluorescence, respectively. As demonstrated by quantitative assessment of autoradiographs of approximately 20 hepatocytes located between a terminal portal venule and a hepatic venule, PB induced the P-450b,e mRNA up to sixfold in the 12-15 hepatocytes located closer to the hepatic venules (zones 2 and 3). In contrast, there was only a twofold induction in the 4-6 hepatocytes surrounding the terminal portal venules (zone 1). Quantitative immunofluorescence using an MAb showed that the acinar distribution of PB-induced P-450b,e proteins was similar to that of the mRNA. This combined approach indicated that, most likely, an increased rate of transcription of cytochromes P-450b,e genes in hepatocytes of zones 2 and 3 concomitantly, with a relative lack of activation, or repression, of these genes in hepatocytes of zone 1, were responsible for the heterogeneous phenotype observed after PB administration. Therefore, modulation of gene expression among hepatocytes of the liver acinus is one mechanism by which the functional heterogeneity of hepatocytes is attained. Experiments in which the induction of cytochromes P-450b,e genes was studied after administration of either PB or para-hydroxyphenobarbital, a main hepatic metabolite of PB, suggested that the species involved in the inductive process is the parent PB molecule rather than para-hydroxyphenobarbital.

Expression of cytochrome P450b and P450e genes in small intestinal mucosa of rats following treatment with phenobarbital, polyhalogenated biphenyls, and organochlorine pesticides.

The expression of cytochromes P450b and P450e genes was studied in the small intestinal mucosa of rats using a cDNA which recognizes the mRNAs of both cytochromes as well as oligonucleotide probes which are able to differentiate between the two gene products. Animals were treated with oral and intraperitoneal doses of phenobarbital, gamma-chlordane, trans-non-achlor, and polychlorinated and polybrominated biphenyls. RNA was extracted from small intestinal mucosa and liver. After treatment with each of the compounds, P450b mRNA was markedly induced in small intestinal mucosa and in liver. The greatest degree of induction was found in mucosa of the proximal small intestine where P450b mRNA levels were 4-6-fold higher than levels found in the distal small intestine. This distribution of P450b mRNA was not dependent on the route of administration of inducers. In contrast, although P450e mRNA was induced in the liver after treatment, P450e mRNA in the small intestine did not increase in response to any of the administered inducers. The location of the P450b mRNA within the intestinal mucosa following treatment with inducers was studied by in situ hybridization; the message was induced predominantly in enterocytes located in intestinal villi. These data indicate that the P450b gene is induced in the small intestine following treatment with various xenobiotics and that this induction may be secondary to either transcriptional activation of the gene or to mRNA stabilization in enterocytes located in the villi of the intestinal mucosa. The differential induction of P450b versus P450e genes in the small intestine and liver indicates that the regulation of these closely linked genes is tissue-specific. Furthermore, the marked induction of P450b mRNA in response to the administered xenobiotics indicates that this isoenzyme may have an important biological role in the small intestinal metabolism of environmental toxicants and drugs.

Phenobarbital induction of cytochrome P-450 b,e genes is dependent on protein synthesis.

Phenobarbital induces liver cytochrome P-450 b,e proteins mainly by increasing the rate of transcription of these genes. The mechanism responsible for the phenobarbital increment in the rate of transcription of cytochrome P-450 b,e genes is unknown. The objective of this study was to assess whether active protein synthesis was needed for phenobarbital to induce the liver cytochrome P-450 b,e genes. Cycloheximide (2 mg per kg, i.p.) was administered 90 min prior to a single inductive dose of phenobarbital (80 mg per kg, i.p.) and mRNAS measured at 3, 6 and 12 hr by dot-blot hybridization. While phenobarbital increased cytochrome P-450 b,e mRNAs about 12-fold at 3 hr, this induction was abolished by cycloheximide. To define whether the absence of protein synthesis in hepatocytes inhibited the phenobarbital induction of cytochrome P-450 at the transcriptional level, in vitro transcription rates using isolated nuclei were measured. After phenobarbital administration, there was about a 20-fold increment in transcriptional rate of cytochrome P-450 b,e genes. This increment was abolished by prior injection of cycloheximide. It is proposed that either preexisting regulatory proteins or transacting factors dependent on active protein synthesis participate in the regulation of liver cytochrome P-450 b,e gene transcription after phenobarbital.

Cytochrome P-450 gene expression in the functional units of the fetal liver.

Hepatocytes of the right and left lobes of the fetal liver are surrounded by different microenvironments. The right and left lobes of the fetal liver are perfused by vascular systems carrying different concentrations of oxygen and constitute distinct functional units. The aim of this study was to assess the expression of the phenobarbital-inducible cytochrome P-450 b,e genes in hepatocytes of the right and left fetal liver lobes in mice. Northern-blot analysis using [32P]cDNAs and quantitative dot-blot hybridization were performed to assess the size and levels of these mRNAs in the right and left fetal liver lobes. In fetal mice, the levels of cytochrome P-450, b,e mRNAs were higher in the left than in the right fetal liver lobe. During the last days of gestation and in the immediate postnatal period, the levels of liver cytochrome P-450 b,e mRNAs increased predominantly in the left liver lobe. In contrast, the levels of albumin and alpha-fetoprotein mRNAs (genes studied to assess the specificity of these findings) were similar in each of functional units of the fetal liver. Phenobarbital induction of cytochromes P-450 b,e mRNAs was not observed in either of the fetal liver lobes. Postnatally, phenobarbital induced these cytochromes similarly in the right and left liver lobes. Therefore, the microenvironment surrounding fetal hepatocytes seems to influence the expression of the cytochrome P-450 b,e genes. This lobar heterogeneity of expression disappears as the pattern of adult liver circulation is attained.

Cytomegalovirus cholangitis in a homosexual man with acquired immune deficiency syndrome.

A 29-year old white homosexual man with acquired immune deficiency syndrome presented initially with right upper quadrant pain and progressive cholestasis. Diffuse mucosal irregularities were seen at endoscopic retrograde cholangiography. Histopathological examination of the gallbladder and wedge liver biopsy showed evidence of cytomegalovirus infection. A repeat endoscopic retrograde cholangiography for persistent symptoms of right upper quadrant pain and cholestasis showed progressive mucosal irregularities of the intra- and extrahepatic bile ducts consistent with progressive cholangitis. Subsequently the patient developed evidence of disseminated infection and died. Postmortem examination revealed histologic features of cytomegalovirus infection in lungs, pancreas, small bowel, adrenal glands, and liver. Immunohistochemical staining of liver confirmed the presence of cytomegalovirus infection of the biliary duct system.

Heterogeneous expression of phenobarbital-inducible cytochrome P-450 genes within the hepatic acinus in the rat.

Within the hepatic acinus, the functional unit of liver parenchyma, the induction of cytochrome P-450 protein by phenobarbital is manifested primarily in hepatocytes located closer to the hepatic venule, i.e., distal hepatocytes. The objective of this study was to determine the levels of cytochromes P-450b and P-450e mRNAs in populations of hepatocytes originating in the proximal or distal half of the liver acinus in the rat, as an approach to the elucidation of the mechanisms responsible for the heterogeneous zonal expression of cytochrome P-450 protein. The development of a new method to isolate hepatocytes originating from the proximal or distal half of the liver acinus enabled the measurement of total cytochrome P-450 content and of cytochromes P-450b and P-450e mRNAs in these hepatocytes. Levels of cytochromes P-450b and P-450e mRNAs were assessed in proximal and distal hepatocytes by Northern blot hybridization of poly(A+)RNA with a cDNA recognizing sequences of these two cytochromes. The kinetics of induction were defined by measuring these parameters after a single phenobarbital injection. Cytochrome P-450 mRNA levels reached maximum induction at 16 hr, returning to basal values by 48 hr. In contrast, total cytochrome P-450 microsomal protein content reached maximum induction after 33 hr. Hepatocytes of the distal half of the hepatic acinus responded to phenobarbital with higher levels of cytochromes P-450b and P-450e mRNAs than proximal hepatocytes. These results indicated that there is modulation of the expression of the cytochromes P-450b and P-450e genes within the hepatic acinus.

The isolation of functionally heterogeneous hepatocytes of the proximal and distal half of the liver acinus in the rat.

The objective of this study was to isolate hepatocytes of the proximal half (Zones 1 and 2) or distal half (Zones 2 and 3) of the liver acinus. The zonal origin of the isolated hepatocytes was recognized by: the presence in hepatocytes of a fluorescent marker, acridine orange, selectively delivered to either the proximal or the distal half of the acinus by in situ perfusion prior to cell isolation and the measurement of the induction of cytochrome P-450 by phenobarbital, an induction known to occur predominantly in the distal half of the acinus. Following the selective labeling of the acinus with acridine orange, livers were perfused with collagenase in either the portal to hepatic vein direction (anterograde) or in the retrograde direction. Hepatocytes isolated by either an anterograde or a retrograde perfusion were separated by centrifugation in a Percoll density gradient. This procedure isolated populations of proximal or distal hepatocytes, respectively, which were intact and 90% fluorescent. In an effort of assessing the heterogeneity of the separated proximal and distal hepatocytes, each population was further fractionated by centrifugal elutriation. This resulted in the arbitrary separation of proximal or distal hepatocytes into five fractions. Total cytochrome P-450 was determined spectrophotometrically in each of the fractions isolated from controls and after 3 days of the in vivo administration of phenobarbital. On the basis of the pattern of fluorescence in isolated hepatocytes and on the cytochrome P-450 inductive response to phenobarbital administration, it is proposed that: the anterograde or retrograde perfusion of the liver with collagenase separated hepatocytes predominantly of the proximal or distal half of the liver acinus, respectively and that hepatocytes of the distal half of the liver acinus responded to phenobarbital administration with the highest level of cytochrome P-450 induction, indicating that the isolated hepatocytes conserved the functional heterogeneity observed in vivo.

Hepatocytes of Zones 1 and 3 conjugate sulfobromophthalein with glutathione.

The capacity of hepatocytes of Zones 1 and 3 of the liver acinus to conjugate sulfobromophthalein (BSP) with glutathione and to secrete the conjugate into bile was studied. BSP was infused into the in situ perfused rat liver in concentrations of 0.01 and 0.05 mM. Perfusions were performed either in the portal to hepatic vein direction (forward perfusion) or in the hepatic vein to portal vein direction (retrograde perfusion). Hepatocytes contributing to the uptake, metabolism and biliary secretion of BSP were directly assessed qualitatively by light microscopy, and also semiquantitatively by microspectrophotometry. BSP was taken up predominantly by hepatocytes of Zone 1 during forward perfusion and by those of Zone 3 during the retrograde perfusion of BSP. The biliary products of BSP metabolism by each acinar zone were subsequently assessed. Hepatocytes of both Zones 1 and 3 of the acinus secreted BSP into bile in the form of BSP-glutathione, and BSP-glutathione conjugate represented about 78% of the total BSP secreted into bile by each zone. The rate of BSP biliary secretion by both zones was similar at a concentration of 0.01 mM BSP, while a slight decrease (15%) in BSP secretory rate by hepatocytes of zone 3 was observed at concentrations of BSP near biliary Tm. Liver perfusion with exogenous BSP-glutathione provided results similar to those obtained with BSP. In contrast, the perfusion of 3,6- dibromosulphthalein , a compound which is not conjugated by hepatocytes, resulted in similar biliary secretory rates regardless of the direction of perfusion. These results indicate that: (a) the capacity for glutathione conjugation of BSP is distributed among hepatocytes of all acinar zones; (b) near biliary Tm, the biliary secretory rate of BSP by hepatocytes of Zone 3 is slower than that of hepatocytes of zone 1, and (c) in vivo, all hepatocytes likely contribute to the uptake, metabolism and biliary secretion of BSP.

Albumin influences sulfobromophthalein transport by hepatocytes of each acinar zone.

To determine whether profiles of decreasing concentration were generated among hepatocytes of the liver acinus during the transport of sulfobromophthalein sodium (BSP), rat livers were perfused with various concentrations of this dye (10 microM to 1 mM) in the presence and absence of albumin. After steady-state conditions for the biliary secretion of BSP had been attained, pieces of liver were rapidly frozen. Following the alkalinization of cryostat-cut sections, the relative concentration of BSP in hepatocytes of each zone and the effect of albumin on this localization were quantitated by microspectrophotometry. The results showed that BSP, perfused in the absence of albumin, was efficiently extracted by the liver (95% on a single pass), generating distinct profiles of decreasing cellular concentration from zone 1 to zone 3 at every concentration of BSP. However, the addition of albumin to the perfusate greatly reduced the extraction of BSP from the sinusoidal compartment and resulted in the abolition of the differences in BSP content between hepatocytes of zone 1 and zone 3. These results represent a direct demonstration that, as predicted by mathematical modeling, binding of BSP to albumin indeed results in a more homogeneous distribution of BSP within the liver acinus. A simple and direct microspectrophotometric method is therefore available to follow the changes in the relative concentration of BSP among the hepatocytes of the various acinar zones.

Functional and anatomic heterogeneity in the liver acinus: impact on transport.

The site of solute exchange between blood and hepatocytes is the liver acinus, the structural and functional unit of hepatic parenchyma. Because the perfusion of acinar hepatocytes occurs in a sequential manner, differences in solute concentration between hepatocytes located at the inlet (zone 1) and at the outlet (zone 3) of the acinus are predictable. However, the distribution and transport of solutes by hepatocytes of each acinar zone also seem to be influenced by factors other than the sequential perfusion of blood. The concentration of receptors in hepatocytes at each zone, the presence of specific chemical groups on the incoming molecules, and the binding of the solutes to plasma protein are all factors that modify the interaction with, and the cellular concentration attained by, solutes in each acinar zone. Therefore, solute concentration in hepatocytes of each acinar zone may produce a profile of decreasing solute concentration from cells in zone 1 to those in zone 3, a reverse profile from zone 3 to zone 1, or a similar concentration in all hepatocytes.