Hepatic Gene Expression and Lipid Homeostasis in C57Bl/6 Mice Exposed to Hydrazine or Acetylhydrazine

Victoria E. Richards*, Binh Chau*, M. Randy White{dagger} and Charlene A. McQueen*,1

* Department of Pharmacology and Toxicology, University of Arizona, Tucson, Arizona 85721; and {dagger} PRI, Mt. Vernon Pathology, Evansville, Indiana 47721

Received May 12, 2004; accepted July 23, 2004


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Hydrazine (HD) and acetylhydrazine (AcHD) are metabolites of the antituberculosis drug isoniazid (INH) that have been implicated in INH-induced liver damage. The hepatotoxicity of AcHD and HD were compared in adult male C57Bl/6J mice by evaluating hepatic histopathology, plasma biochemistry, and hepatic gene expression. By all measures, HD had significantly greater effects than AcHD. There was no evidence of liver damage following exposure to AcHD (300 mg/kg, po). However, HD at this dose caused marked hepatic necrosis, macrovesicular degeneration, and steatosis. Lipid accumulation was initiated 2 h after HD exposure, with hepatic macrovesicular degeneration evident after 4 h, and severe necrosis by 36 h. Gene expression profiles were compared 24 h following 100 mg/kg po of HD or AcHD. HD changed the hepatic expression of more genes than AcHD, particularly lipid synthesis, transport, and metabolism genes that may be involved in steatosis. Hepatic expression of genes regulated by peroxisome proliferator activated receptors (PPAR) and sterol regulatory element binding protein (SREBP) transcription factors was increased only by HD. The hepatotoxicty and hepatic gene expression profile of HD, but not AcHD, indicate that exposure to HD initiates a process whereby the production and intracellular transport of hepatic lipids is favored over the removal of fatty acids and their metabolites.

Key Words: hydrazine; acetylhydrazine; cDNA microarray; steatosis; hepatotoxicity; peroxisome proliferator-activated receptor.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Human exposure to hydrazine (HD) and HD derivatives arises from environmental contamination, occupation, and medicinal use. HD is an intermediate in chemical syntheses, an aerospace propellant, and is found in tobacco smoke (Choudhary and Hansen, 1998Go). Additionally, HD is used as an unregulated supplemental therapy for advanced cancer. Exposure most frequently occurs via approved therapeutic use of HD derivatives, such as isoniazid (INH), a tuberculostatic drug, and hydralazine, an antihypertensive agent.

Hepatotoxicity is a major adverse effect of both INH and HD (Mitchell et al., 1975Go; Stuart and Grayson, 1999Go; Timbrell and Wright, 1984Go). Although 0.8–23% of individuals on INH therapy report some form of hepatic dysfunction ranging from elevated liver enzymes to hepatic failure and even death (Stuart and Grayson, 1999Go), the hepatotoxic product(s) have not been definitively elucidated. In humans, the primary route of INH metabolism is acetylation via hepatic N-acetyltransferase 2 (NAT2), using acetyl Coenzyme A (acetyl CoA) as the cofactor (Weber, 1987Go). There is genetic variation in the rates at which individuals acetylate INH, but no conclusive association has been made between acetylator status and INH-related hepatotoxicity (Mitchell et al., 1975Go; Timbrell and Wright, 1984Go). INH and acetyl-INH can be hydrolyzed by an amidohydrolase or amidase to generate HD (Sarich et al., 1999Go). HD is a substrate for NAT, forming AcHD, with a second acetylation yielding diacetylHD (DiAcHD), a reported detoxification product (Lauterburg et al., 1985Go; Preece et al., 1991Go). AcHD can be generated by the hydrolysis of acetyl-INH, then further hydrolyzed to indirectly form HD (Sarich et al., 1999Go). Cytochrome P450/FMO (flavin-containing monooxygenase)-mediated oxidation reactions generate reactive intermediates from AcHD and HD (Jenner and Timbrell, 1995Go).

Exposure of humans or animals to hydrazines can result in hepatic necrosis and steatosis (fatty liver) by unknown mechanisms (Lauterburg et al., 1985Go; Mitchell et al., 1975Go; Scales and Timbrell, 1982Go). AcHD and HD have been implicated as the hepatotoxic species. The P450 isoforms that convert AcHD to reactive products are generally present in low levels; consequently, AcHD only causes hepatic necrosis in rats following phenobarbital induction of P450 (Bahri et al., 1981Go). In contrast, exposure of naive animals to HD induces significant hepatic changes. HD hepatotoxicity in rodents includes increased hepatic triacylglycerol content (Lamb and Banks, 1979Go; Timbrell et al., 1982Go), elevated rates of free fatty acid (FFA) transport to the liver (Amenta and Dominguez, 1965Go; Trout, 1965Go), and inhibition of triglyceride secretion from the liver (Clark et al., 1970Go). There is also evidence of significant lipid accumulation within periportal and midzonal hepatocytes (Amenta and Dominguez, 1965Go). In addition to lipid changes, exposure to HD causes oxidative stress (Hussain and Fraizer, 2002Go), glutathione reduction (Jenner and Timbrell, 1995Go; Timbrell et al., 1982Go), as well as the production of reactive oxygen species and free radicals, including lipid peroxides (Walubo et al. 1998Go). In order to further elucidate the mechanisms by which hydrazines induce hepatic injury, changes in gene expression and hepatic pathology were evaluated in the C57Bl/6J mouse exposed to AcHD or HD. The hypothesis being tested was that HD produces greater liver damage than the acetylated derivative AcHD.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Animals. Male C57Bl/6J mice (approximately 8–10 weeks of age) were obtained from Jackson Laboratories, Inc. (Bar Harbor, ME). The animals were housed in cages with sawdust bedding and had free access to food (Teklad rodent laboratory chow, Madison, WI) and water. Animals were maintained on 12-h light/dark cycle and acclimated to this environment for at least 7 days before dosing. HD sulfate (99.7% purity) was purchased from Sigma Chemical Co. (St. Louis, MO), and AcHD (acetylhydrazide, 90% purity) from Aldrich (Milwaukee, WI). The test compounds were dissolved in saline. Mice were given a single oral dose of HD or AcHD (0, 100, or 300 mg/kg).

Hepatotoxicity. Hepatotoxicity was assessed by measuring plasma alanine aminotransferase (ALT) activity and by histopathology. After exposure to the test compounds or vehicle, animals were euthanized with CO2. Blood (0.8–1 ml) was drawn from the inferior vena cava, collected in a heparinized syringe, then transferred to a heparinized Eppendorf tube to isolate plasma. Plasma levels of ALT were determined using a diagnostic kit (Sigma Diagnostics, St. Louis, MO). The results are expressed as mean ± SEM with a minimum of three animals per group. Comparisons between multiple groups were conducted with a one-way analysis of variance followed by a post hoc Student-Neuman-Keuls test.

For histological assessment, livers were removed, and portions either fixed in 10% neutral-buffered formalin and embedded in paraffin or embedded in OCT media and slowly frozen at – 80°C. For histological examination, 5 µ-thick sections were cut and then stained with hematoxylin-eosin (H&E) or Oil Red O. To assess morphological changes, digital photomicrographs of the H&E stained sections were obtained. A magnification of 40x was selected for optimal vacuole visualization. To quantify lipid-droplets digital photomicrographs of the Oil Red O, stained sections were subject to image analysis (ImagePro Plus, Version 4.5.0.29, Media Cybernetics, Silver Spring, MD). The range of color for the lipid droplets (red) was selected based on the color cube method. Average numbers of droplets were determined in three liver sections (three fields/section). There were three animals per time point in treatment and control groups. A magnification of 10x was chosen for best coverage of lipid droplets. Mean ± SEM was calculated for each group. Comparisons between multiple groups were conducted with a one-way analysis of variance followed by a post hoc Dunn test. Data were considered significantly different at p < 0.05.

Gene expression. Livers were removed from C57Bl/6 mice (n = 3) 24 h after administration of either HD or AcHD (100 mg/kg, po) or saline, frozen in liquid nitrogen, and stored at – 80°C. Either total RNA (Trizol, Invitrogen, Carlsbad, CA) or PolyA+ RNA (Midi kit, Qiagen, Valencia, CA) was isolated according to manufacturer's recommendations. The livers were analyzed individually by the Arizona Cancer Center (ACC)/Southwest Environmental Health Sciences Center (SWEHSC) Microarray Facility using a mouse cDNA microarray slide with 5184 genes plus expressed sequences tags (EST). Each cDNA is a single spot on the array. The gene sequences used to create the microarrays were cDNA clones produced by the IMAGE consortium. Each clone has been sequence-validated and provided by Research Genetics, Inc. The RNA from vehicle and treated animals was labeled with Cy3 or Cy5-labeled dCTP, respectively, in a first strand cDNA (reverse transcription) reaction. The fluorescently labeled probes were purified, mixed, and cohybridized to the microarrays. Results were captured electronically on an Axon Instruments GenePix 4000 laser. Numerical and visual outputs of the raw microarray hybridization results were obtained with Axon Instruments' GenePix software. The gene expression intensities were first corrected with the local background, followed by intensity-dependent per-spot and per-chip normalization. The intensity of each gene was divided by its control channel value in each sample then divided by the 50th percentile of all measurements in that sample to correct for effects which arise from variation in the microarray technology, such as fluorescent dye balance, print quality, and scanner settings. The corrected and normalized data were interpreted as signal channel (Cy5) divided by the control channel (Cy3) using GeneSpring software Version 6.1 (Silicon Genetics, Redwood City, CA). Normalization was followed by restricting the gene list to show changes in expression of two-fold up or down in all three animals. Expression ratios for each animal were used to determine mean ± SEM for each group. Sequences were annotated with UniGene. Functional annotations were done using UniGene and Entrez-Gene systems through the NCBI webpage (http://www.ncbi.nlm.nih.gov/). Hierarchical clustering was performed using the relative gene expression ratios (Cy5/Cy3) to examine the relatedness among the expression pattern in the restricted dataset. The clustering algorithm calculated the standard correlation for each gene with every other gene in the set; the standard correlation measured the angular separation of expression vectors for two genes around zero. The minimum separation distance used was 0.001; the separation ratio, which determines how large the correlation difference between groups of clustered genes, was set at 0.5. The highest correlation was then determined, and the two genes were paired, generating an average of their expression profiles. The new composite gene was compared with all the other unpaired genes, with a finite number of iterations performed until all the genes had been paired. Numerical values assigned to nodes represent the stability of the cluster, with values below 0.001 or above 0.5 reflecting poor reliability.

Real-time quantitative PCR (RT-PCR). The alterations in expression ratios of selected genes were verified by quantitative RT-PCR. RNA from treated and control livers used for microarray analysis was isolated using an Absolutely RNA RT-PCR Miniprep Kit (Stratagene, LaJolla, CA). First strand cDNA was generated with 2 µg RNA using the PE-Biosystems Kit (Applied Biosystems, Foster City, CA). The genes of interest were amplified using 5 ml cDNA with SYBR-PCR Master Mix (Applied Biosystems, Foster City, CA) and 1 mM of each amplification primer (Table 1). Primer pairs were selected using GenBank or published literature. The temperatures, times, and number of amplification cycles were determined based on the gene being verified. The fluorescence resulting from SYBR Green binding to double-stranded DNA was obtained for each annealing cycle (Cepheid Systems, Sunnyvale, CA). Fluorescence (y-axis) was plotted against cycle number (x-axis) to generate the amplification curve. Threshold cycle (Ct) was calculated from the rate of change of the slope of the amplification curve (second derivative). The number of cycles needed to reach the log phase of amplification depends on the initial number of transcripts for a particular gene. Melting curves were generated by the Cepheid Systems software. Standard curves were generated for each gene. Data are presented as the expression ratio (treated:exposed) and compared to the ratios obtained in the microarrays. Results were normalized to the histone 3.3 gene to standardize for sample variation (McQueen and Chau, 2003Go).


View this table:
[in this window]
[in a new window]
 
TABLE 1 Primer Sequences Used for Quantitative RT-PCRa

 

    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The hepatotoxicity of HD and AcHD was characterized in C57Bl/6J mice. No significant changes in plasma ALT activity (Fig. 1), or liver morphology (Fig. 2), were seen with AcHD, even at 300 mg/kg, the highest dose tested. In contrast, HD induced a dose-dependent hepatic injury. Exposure to 100 mg HD/kg resulted in no biologically significant liver damage (Figs. 1 and 2) similar to a previous report (Sasaki et al., 1998Go). However, 300 mg HD/kg caused a ten-fold increase in ALT levels (Fig. 1) and histologic lesions, including hepatic necrosis and fatty infiltration (Fig. 2). This hepatic fatty infiltration was observed as large, generally spherical to oval nonstaining cytoplasmic vacuoles, known as macrovesicular degeneration. Oil Red O staining of these hepatocytes confirmed that the vacuoles contained neutral lipids (cholesterol and triglycerides) (Fig. 2). Further study at 300 mg HD/kg showed that the lipids also increased in a time-dependent manner. Minimal macrovesicular degeneration was evident after 4 h exposure to 300 mg HD/kg, and marked by 16 h (Fig. 3). After 24 h exposure to this dose, the presence of cytoplasmic vacuoles diminished, replaced by diffuse necrosis. Quantitation of lipid droplets showed a statistically significant increase at 2 h (Fig. 4). From 2 to 8 h, the accumulation of lipids continued to increase, then remained relatively constant between 8 and 36 h.



View larger version (22K):
[in this window]
[in a new window]
 
FIG. 1. Plasma ALT activity 24 h after exposure to HD or AcHD. Male C57Bl/6J mice were given an oral dose of either HD or AcHD. Blood was collected, centrifuged, and assayed for alanine aminotransferase. The results are presented as units/ml, mean ± SEM (n = 3). Comparisons between multiple groups were conducted with a one-way analysis of variance followed by a post hoc Student-Neuman-Keuls test. *p < 0.05 compared to control.

 


View larger version (75K):
[in this window]
[in a new window]
 
FIG. 2. Histological assessment of livers taken from C57Bl/6J mice 24 h after oral exposure to HD and AcHD. Photomicrographs of hematoxylin and eosin-stained paraffin sections (A: 1–5; 40x magnification) and of Oil Red O-stained frozen sections (B: 1–5; 10x magnification). (1) Saline; (2) 100 mg/kg HD; (3) 300 mg/kg HD; (4) 100 mg/kg AcHD; (5) 300 mg/kg AcHD. Arrows indicate lipid-filled intracytoplasmic vesicles. Staining with Oil Red O, a neutral lipid-specific dye, reveals the accumulation of triglyceride and cholesterol as red droplets; the nuclei are stained blue.

 


View larger version (161K):
[in this window]
[in a new window]
 
FIG. 3. Histological assessment of the time course of HD-mediated lipid accumulation in the C57Bl/6J liver. Adult mice were given a single oral dose of 300 mg/kg HD, sacrificed, and livers collected over a period of 72 h. (A: 1–12, hematoxylin and eosin-stained liver sections; B: 1–12, Oil Red O-stained liver sections). (1) Saline; (2) 0 h; (3) 2 h; (4) 4 h; (5) 6 h; (6) 8 h; (7) 10 h; (8) 12 h; (9) 16 h; (10) 24 h; (11) 36 h; (12) 72 h. There was minimal macrovesicular degeneration after 4 h exposure and marked by 16 h. From 2 to 8 h, the accumulation of lipids increased, then remained relatively constant between 8 and 36 h. After 24 h the cytoplasmic vacuoles diminished and were replaced by diffuse necrosis.

 


View larger version (19K):
[in this window]
[in a new window]
 
FIG. 4. Hepatic lipid droplets in C57Bl/6J mice after a time course of exposure to 300 mg HD/kg. Three animals were examined at each time point from the control and HD-treated groups. Separate sections of liver were analyzed for lipid droplets using computer software described in the materials and methods section. Droplets were counted and are presented as means ± SEM. Comparisons were made with a one-way analysis of variance followed by a posthoc Dunn test. Data were considered significantly different at *p < 0.05 compared to saline.

 
Changes in hepatic gene expression were analyzed following exposure to a dose with no visible changes in hepatic morphology. Separate hybridizations using individual livers (n =3) were carried out for both HD and AcHD treatments at a subtoxic dose of 100 mg/kg, po. The resulting data were considered significant if the expression ratio showed a two-fold change giving ratios greater than 2 or less than 0.5 in all three animals. There were compound-specific differences in the expression of hepatic genes following exposure to HD compared to AcHD. With AcHD, 64 known genes were changed, with 5 over-expressed and 59 under-expressed. With HD, there were 147 known genes with altered expression, 57 being over-expressed and 90 under-expressed. Nineteen known genes were shared by HD and AcHD. Several cytochrome P450 genes (Cyp2a5, 2b9, 2c40, 2d9, 2e1, 2j6, 3a25, and 4b1) were included on the array, but neither compound caused significant changes in these genes. Expression ratios of several genes obtained with microarray analysis were verified by quantitative RT-PCR (Table 2). These results generally correlated well with microarray results.


View this table:
[in this window]
[in a new window]
 
TABLE 2 Quantitative RT-PCR Analysis of Hepatic mRNA Following Hydrazine Exposure

 
To further examine patterns of gene expression after HD and AcHD exposure, a hierarchical clustering algorithm was applied to the restricted datasets. The output was a dendrogram, a tree representation of the data whose "leaves" are input patterns and whose "nonleaf" nodes represent a hierarchy of groupings. Clusters were defined based on the general molecular functions and biological processes of the gene products. Clusters of the genes and ESTs altered by AcHD were anchored by seven major nodes (A–G), enabling classification into functional categories including oxidative stress response, immune signaling, and structural remodeling (Fig. 5A). Likewise, HD gene expression changes were linked by seven major nodes (A–G) and classified into categories including necrosis, stress response, and lipid peroxidation (Fig. 5B). Dendrogram analysis of the common genes and ESTs between AcHD and HD revealed three major nodes (A–C) and were defined as plasma membrane stability, Ca2+ binding/immune signaling, and differentiation (Fig. 6). The information derived from the clusters was used to generate tables of genes and their functions, along with expression ratios for AcHD (Table 3), HD (Table 4), and both treatment groups (Table 5).



View larger version (38K):
[in this window]
[in a new window]
 
FIG. 5. Hierarchical cluster analysis of C57Bl/6J mouse liver 24 h after exposure to AcHD or HD. Cluster analysis was performed on the genes and ESTs changed by AcHD (A) or HD (B) as described in the text as reflecting a two-fold increase (red) or decrease (green). Each column represents an individual liver. To categorize the changes in hepatic genes in response to exposure, a standard correlation was performed on the data set, which measured the angular separation of expression vectors for two genes around zero. Numbers represent the stability of the cluster, as described in the text. Clustering was performed and visualized with GeneSpring software (Silicon Genetics, Redwood City, CA). Database mining for gene product function was conducted with UniGene and Entrez-Gene systems (NCBI), with clusters defined based on the functions of the resident genes.

 


View larger version (22K):
[in this window]
[in a new window]
 
FIG. 6. Hierarchical cluster analysis of the common genes changed after HD and AcHD in C57Bl/6J mouse liver 24 h after exposure. Cluster analysis was performed on the genes and ESTs changed by HD and AcHD described in the text as reflecting a two-fold increase (red) or decrease (green). To categorize the changes in hepatic genes in response to exposure, a standard correlation was performed on the data set, which measured the angular separation of expression vectors for two genes around zero. Clustering was performed and visualized with GeneSpring software (Silicon Genetics, Redwood City, CA). Database mining for gene product function was conducted with UniGene and Entrez-Gene systems (NCBI), with clusters defined based on the functions of the resident genes.

 

View this table:
[in this window]
[in a new window]
 
TABLE 3 Hepatic Gene Expression Analysis of Genes Following Exposure to Acetylhydrazine

 

View this table:
[in this window]
[in a new window]
 
TABLE 4 Hepatic Gene Expression Analysis of Genes Following Exposure to Hydrazine

 

View this table:
[in this window]
[in a new window]
 
TABLE 5 Common Expression Changes in Genes After Hydrazine and Acetylhydrazine Exposure

 
AcHD, but not HD, induced a decrease in expression of Mapk10 (MAP kinase 10) (0.44 ± 0.03), c-Myc (myelocytomatosis oncogene) (0.36 ± 0.06), as well as Hrasls (H-Ras like suppressor) (0.26 ± 0.02) (Table 3). These three genes are members of cluster E defined as involving regeneration and proliferation (Fig. 5a). The expression of Mt1 (metallothionein 1) was increased after AcHD exposure (15.60 ± 3.81) and sclustered with Cldn8 (claudin 8) (0.17 ± 0.07) and Cklsf2a (chemokine-like factor super family 2A) (0.36 ± 0.07), suggesting a response to oxidative stress (Table 3, Fig. 5a).

Genes altered by HD included several involved in necrosis, lipid peroxidation/fatty acid (FA) synthesis, transport, and metabolism, as well as inflammation and morphological changes (Table 4). Since HD caused hepatic steatosis, genes associated with lipid homeostatsis were of particular interest. Steatosis can result from impaired FA oxidation, increased mobilization of FA coupled with increased triglyceride synthesis, or decreased secretion of triglyceride-rich lipoproteins. Several lipid synthesis, transport, and metabolism genes were upregulated in response to HD exposure in the mouse liver and clustered together (Fig. 5b). There were increases in levels of Agpat3 (1-acylglycerol-3-phosphate O-acyltransferase 3), necessary for triglyceride biosynthesis (3.49 ± 0.21), Mvd (6.68 ± 0.51), necessary for cholesterol biosynthesis, and Cyp51 (7.09 ± 0.46), an enzyme required in the synthesis of cholesterol from lanosterol. These genes clustered with Apoa5 (apolipoprotein A5) (3.17 ± 0.19), which binds triglycerides. Hmgcs2, a key intermediate in acetyl CoA metabolism, was increased (3.41 ± 0.05), as well as the expression of Ldlr (low density lipoprotein receptor) (3.04 ± 0.20), which mediates the transfer of LDL, and Apoa4 (apolipoprotein A4) (13.62 ± 0.71), a regulator of cholesterol absorption.

Stress responses induced by HD were indicated by changes in several genes (Table 4). The greater than two-fold increase in the level of genes such as Hspb8 (heat shock 27 kDA protein 8) (2.63 ± 0.29), Gstt2 (glutathione S-transferase, theta 2) (3.34 ± 0.57), Atox1 (antioxidant protein 1) (2.67 ± 0.24), Cbr1 (carbonyl reductase 1) (3.12 ± 0.31), and Rdh11 (retinol dehydrogenase) (3.05 ± 0.45) support a response to oxidative stress (Fig. 5b). Protective measures against lipid accumulation and potential oxidative damage include FA oxidation and metabolism. In response to HD exposure, the levels of expression of FA oxidation genes, such as Cpt2 (carnitine palmitoyltransferase 2), a mitochondrial CoA transfer protein (2.32 ± 0.12), and Cyp4a14 (Cytochrome P450 4a14), involved in the microsomal {omega}-FA oxidation pathway, (20.1 ± 5.06) were increased.

There was also evidence of HD-induced changes in expression of genes associated with necrosis. In response to HD exposure, there was upregulation of proteases such as Ctsc (cathepsin C), Cpn (carboxypeptidase N), Tgm2 (transglutaminase 2), Mmp24 (matrix metalloproteinase 24), and Psmc3 (proteasome 26S subunit, ATPase), and downregulation of protease inhibitors Serpina12 (serine protease inhibitor) and Expi (extracellular proteinase inhibitor) (Table 4). Moreover, the levels of several ion transporters, such as Slc6a15 (0.26 ± 0.05), Slc12a1 (0.32 ± 0.06), Slc6a2 (0.08 ± 0.04), and Slc5a8 (0.21 ± 0.10) were decreased.

Genes whose expression was changed by both HD and AcHD included several involved in cholesterol biosynthesis, immune response, and signal transduction (Table 5). There was a cluster with increased expression of Cyp51 (cytochrome P450, 51), Mvd (mevalonate (disphospho) decarboxylase), and Pcks9 (proprotein convertase substilisin/kexin type 9), genes involved in cholesterol biosynthesis and metabolism (Fig. 6). In this cluster, genes with decreased expression included Ccm1 (cerebral cavernous malformations 1), encoding a small GTPase regulator, Mip (major intrinsic protein eye fiber 1), encoding a water channel, Cd28 (Cd28 antigen), a T-cell surface glycoprotein, and Rcn (reticulocalbin), a Ca2+-binding protein of the endoplasmic reticulum.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Both HD and AcHD are implicated as hepatotoxic products of INH biotransformation (Bahri et al., 1981Go; Lauterburg et al., 1985Go; Mitchell et al., 1975Go). The hypothesis that HD induced greater liver damage than AcHD was tested in C57Bl/6J mice. By all parameters, AcHD had significantly fewer effects than the same dose of HD. A single orally administered dose of AcHD did not result in liver damage (Fig. 2). In previous rodent studies where AcHD produced hepatotoxicity, induction of cytochrome P450-mediated bioactivation was necessary (Timbrell et al., 1980Go). Neither AcHD nor HD caused significant changes in the expression of the P450 genes on the microarray. Of these, only Cyp2e1 is associated with the biotransformation of hydrazines. The hepatic expression of Cyp2e1 was unchanged by 100 mg/kg of either compound, results similar to that seen in rats (Runge-Morris et al., 1996Go). In contrast to the lack of detectable hepatic damage with AcHD, 300 mg HD/kg resulted in hepatic macrovesicular degeneration, necrosis, and hepatic lipid accumulation (Fig. 2). HD-induced hepatotoxicity was progressive with time, starting with lipid accumulation at 2 h and macrovesicular degeneration evident at 4 h, with severe necrosis after 36 h (Figs. 3 and 4). It appears that the sequence of events involves increases in hepatic lipids that precede cellular damage. This is consistent with previous studies in rats showing HD, but not AcHD, results in a mobilization of fatty acids (FA) from extrahepatic sources and hepatic lipid accumulation (Clark, et al., 1970Go; Sarich et al., 1999; Trout, 1965Go).

Comparison of changes in hepatic gene expression induced by AcHD and HD provides insight into a mechanism leading to steatosis and necrosis. Investigation of such alterations requires that expression patterns associated with primary changes resulting from the chemical be separated from those occurring as a response to cell damage. This was accomplished by using doses that did not include pathological changes (i.e., 100 mg/kg). Comparison of the expression profiles revealed genes that were common to the two treatment groups (Table 5). Both compounds resulted in the increased expression of genes influencing membrane stability (Fig. 6). The downregulation of Rcn and Mip would increase cell adhesion and membrane integrity. This was coordinated with the upregulation of Cyp51, Mvd, and Pcks9, genes involved in cholesterol synthesis and metabolism. Properties of cell membranes can be changed by modulation of their cholesterol content. Increased expression of these genes would contribute to the maintenance of cholesterol-rich membrane microdomain stability. Although cholesterol biosynthesis might contribute to hepatic lipid accumulation, this seems unlikely, since only HD induced that pathology, but both compounds caused changes in expression of this set of genes.

There were also chemical-specific patterns of gene expression. The generation of reactive products and hepatic necrosis are seen with AcHD (Bahri et al., 1981Go; Jenner and Timbrell, 1995Go). Such damage can trigger stress responses such as the significant upregulation of Mt1 induced by AcHD as well as changes in cell proliferation. Cell division is controlled by growth factors and integrins that stimulate the cell cycle via several pathways including MAP kinase and Ras (Danen and Yamada, 2001Go). Decreased expression of the Ras-related signal transducers Vav2, Hrasls, and Mapk10 following exposure to a dose of AcHD that does not produce visible damage suggests that these are early hepatic responses to deal with oxidative stress and prevent cell injury.

HD induces steatosis and uniquely influences the expression of genes responsible for lipid transport, synthesis, and metabolism. HD causes an increase in hepatic free fatty acids and an accumulation of triglycerides (TG) (Clark et al., 1970Go; Trout, 1965). Normally, hepatic fatty acids are oxidized to form ketone bodies that are exported to extrahepatic tissue and serve as a source of fuel. Alternatively, they are esterified into triglycerides that are stored or are secreted as very low-density lipoproteins (VLDL) consisting of TG, cholesterol (CHO), and apolipoprotein B. As hepatic lipids begin to accumulate, lipid catabolism via mitochondrial ß- and microsomal {omega}-oxidations as well as the synthesis of triglycerides (TG) is increased. The microarray data 24 h after a subtoxic dose show HD-mediated alterations in genes associated with lipid export in a cluster including Apoa4, Ldlr, and C1qtnf3 (Table 2). These three genes are involved in regulating high-density lipoprotein (HDL) and low-density lipoprotein (LDL) levels. Ldlr, critical in controlling hepatic cholesterol homeostasis, mediates the hepatic uptake of LDL (Lee et al., 2003Go). Clqtnf3 is homologous to adiponectin, a substance that is associated with increased TG and decreased HDL (Hulthe et al., 2003Go), whereas the transfer of cholesterol esters between HDL and LDL within the hepatocyte is reportedly mediated by Apoa4 (Stan et al., 2003Go). HDL also shuttles tissue-derived cholesterol from the periphery to the liver (Lee et al., 2003Go). HD-induced accumulation of TG would increase the requirement for CHO and apolipoprotein B in order to assemble VLDL. The observed upregulation of Ldlr and Apoa4 and the downregulation of Clqtn3 would elevate hepatic CHO and the potential for lipid export via VLDL.

Lipids not targeted for secretion are stored as esterified FA (triglycerides and cholesterol esters) within the liver. These stored neutral lipids become substrates for oxidation, necessary for the production of acetyl CoA and CO2. FA oxidation occurs mainly in mitochondria and peroxisomes (ß-oxidation) and, to a lesser extent, microsomes ({omega}-oxidation). The coordination of these three oxidation systems is carried out by peroxisome proliferator-activated receptor {alpha} (PPAR{alpha}), a member of the nuclear hormone receptor superfamily. Other PPAR isoforms are ß/{delta} and {gamma}. The {alpha} and {gamma} isoforms are preferentially activated by FA, particularly arachidonic acid and its metabolites, and hypolipidemic drugs, such as fibrates and thiazolidinediones (Forman et al., 1997Go). PPAR{alpha} (and PPAR{delta}) have been shown to induce FA catabolism in metabolically active tissues, such as liver and muscle, while PPAR{gamma} is necessary for lipid storage and differentiation of adipocytes (Lee, 2003Go). Once activated by ligand binding, PPARs heterodimerize with RXR and induce the transcription of genes that contain a PPRE (peroxisome proliferator response element) within their promoters. Exposure to HD affected PPAR{alpha}- and PPAR{gamma}-regulated genes. HD altered the expression of three PPAR{gamma}-regulated genes, with decreased expression of Retn and increased expression of Inhbe and Ldlr (Table 4) (Akiyama, et al., 2002Go; Kobayashi, et al., 2003Go; Patel et al., 2003Go). This pattern of changes in expression is consistent with the observation that PPAR{gamma} is activated in steatotic liver (Gavrilova et al., 2003Go).

HD causes an increase in free fatty acids, physiologic ligands of PPARa (Clark et al., 1970Go; Clarke et al., 1999Go). Expression ratios of Cpt2, Hmgcs2, Apoa5, and Cyp4a14, genes regulated by activation of PPARa, were increased by HD (Table 4). Cpt2 and Hmgcs2 are important in the regulation of hepatic FA oxidation and ketogenesis (Madsen et al., 1999Go; Le May et al., 2000Go). Cpt2, located in the inner mitochondrial membrane, generates substrates (palmitoyl CoA) for ß-oxidation, whereas Hmgcs2 controls the conversion of acetyl CoA produced by ß-oxidation to ketones (acetoacetate). Hence, the induction of both Cpt2 and Hmgcs2 is a means of removing excess hepatic FA. In addition to mitochondrial ß-oxidation, lipid catabolism is mediated via microsomal {omega}-oxidation catalyzed by Cyp4a14. This is generally a relatively minor oxidative pathway, but it is induced under conditions of steatosis (Reddy, 2001Go). Upregulation of Cyp4a14 suggests that the ß-oxidation pathway is inadequate to deal with the lipid accumulation induced by HD, and a greater capacity for {omega}-oxidation is needed to increase the breakdown of hepatic lipids. Similarly, an increase in Apoa5 helps protect against lipid overload by transporting lipids (mostly triglyceride) and enhancing the breakdown of triglyceride-rich lipoproteins (van der Vliet et al., 2001Go; Vu-Dac et al., 2003Go).

There were also several PPAR{alpha}-regulated genes whose expression was not altered by HD, including Ehhadh (enoyl-Coenzyme A, hydratase/3 hydroxyacyl Coenzyme A dehydrogenase, 1.36 ± 0.11), Acox2 (acyl-Coenzyme A oxidase 2, 1.73 ± 0.16), Apoc1 (apoliprotein c1, (0.55 ± 0.02), Fasn (fatty acid synthetase, 1.54 ± 0.10) and Lpl (lipoprotein lipase, 0.53 ± 0.06). The first three are part of the peroxisomal ß-oxidation pathway and are induced under conditions leading to peroxisomal proliferation (Rao and Ready, 2001Go). HD has not been reported to increase hepatic peroxisomes. Fasn is involved in lipogenesis while Lpl hydrolyzes lipoproteins, processes that would not help in removal of hepatic lipids.

The coordination of lipid sensors is a strategy to address disturbances in lipid homeostasis. HD altered the expression of a number of lipid-related genes that are regulated by the transcription factor SREBP (sterol regulatory element binding protein). When cellular cholesterol levels drop, the membrane-bound precursor of SREBP undergoes proteolytic cleavage before entering the nucleus to direct the regulation of genes containing an SRE (sterol regulatory element) within their promoters (Horton, 2002Go). Cellular cholesterol and FA homeostasis are regulated by three SREBP isoforms, SREBP-1a, SREBP-1c (FA and glucose/insulin metabolism), and SREBP-2 (cholesterol synthesis). The expression of a set of SREBP-regulated genes was increased by HD (Hmgcs2, Ldlr, Cyp51, Mvd, Agpat3, Pcsk9, and Rdh11) (Table 4). Upregulation of these genes that are involved in cholesterol and triglyceride synthesis is a response that would help to reduce hepatic lipid accumulation (Horton, 2002Go; Kasus-Jacobi et al., 2003Go; Maxwell et al., 2003Go; Rawson, 2003Go).

Steatosis is associated with a state of oxidative stress and necrosis. FA {omega}-oxidation can generate H2O2 and potentially toxic dicarboxylic acids (Rao and Reddy, 2001Go). Peroxidation of accumulated lipids leads to the formation of toxic reactive aldehyde by-products and downstream effects such as impaired membrane integrity, mitochondrial and sarcoplasmic reticulum dysfunction, and altered calcium homeostasis. Additionally, HD can be converted to reactive products, causing oxidative stress and hepatotoxicity independent of steatosis (Hussain and Frazier, 2002Go). After exposure to HD, the expression of hepatic genes involved in oxidative stress response such as those encoding molecular chaperones (Hspb8 and Dnajb10) and electrophile scavengers (Gstt2) was increased (Table 4).

In summary, these results show that AcHD does not produce the same hepatic pathophysiology or gene expression changes seen with HD. The data confirm that HD causes lipids to accumulate in the liver. The data further indicate that this steatosis triggers PPAR and SREBP transcriptional responses. The hepatoxicity and hepatic gene expression profiles observed after exposure to HD are indicative of a process whereby the production and intracellular transport of hepatic lipids is favored over the removal of FA or FA metabolites. Although the extent to which each of these contributes to the hepatotoxicity of HD is not yet clear, the current work provides significant insights into the mechanism of HD-mediated steatosis.


    ACKNOWLEDGMENTS
 
The authors would like to thank Samuel H. Yalkowsky, Ph.D., for the use of the digital imaging system. These studies used the facilities of the Southwest Environmental Health Sciences Center (ES06694) and the Arizona Cancer Center (CA 23074). This work was funded in part by ES10047 (CAM). V. E. Richards was partially supported by ES07091.


    NOTES
 

1 To whom correspondence should be addressed at College of Pharmacy, The University of Arizona, 1703 E. Mabel, Tucson, AZ 85721. Fax: (520) 626-5420. E-mail: mcqueen{at}pharmacy.arizona.edu.


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Akiyama, T. E., Sakai, S., Lambert, G., Nicol, C. J., Matsusue, K., Pimprale, S., Lee, Y.-H., Ricote, M., Glass, C. K., Brewer, H. B., Jr., et al. (2002). Conditional disruption of the peroxisome proliferator-activated receptor {gamma} gene in mice results in lowered expression of Abca1, Abcg1, and apoE in macrophages and reduced cholesterol efflux. Mol. Cell. Biol. 22(8), 2607–2619.[Abstract/Free Full Text]

Amenta, J. S., and Dominguez, A. M. (1965). Fatty acid flux and triglyceride secretion in the hydrazine-induced fatty liver. Exp. Mol. Pathol. 4, 282–302.[CrossRef][ISI]

Bahri, A. K., Chiang, C. S., and Timbrell, J. A. (1981). Factors affecting acetylhydrazine hepatotoxicity. Adv. Exp. Med. Biol. 136, 1015–1065.

Choudhary, G., and Hansen, H. (1998). Human health perspective on environmental exposure to hydrazine: A review. Chemosphere 37, 801–843.[CrossRef][ISI][Medline]

Clark, D. A., Leeder, L. G., Foulds, E. L., and Trout, D. L. (1970). Changes in lipids of rat liver after hydrazine injection. Biochem. Pharmacol. 19, 1743–1752.[CrossRef][ISI][Medline]

Clarke, S. D., Thuillier, P., Baillie, R. A., and Sha, X. (1999). Peroxisome proliferator-activated receptors: A family of lipid activated transcription factors. Amer. J. Clin. Nutr. 70, 566–571.[Abstract/Free Full Text]

Danen, E. H. J., and Yamada, K. M. (2001). Fibronectin, integrins and growth control. J. Cell. Physiol. 189, 1–13.[CrossRef][ISI][Medline]

Forman, B. M., Chen, J., and Evans, R. M. (1997). Hypolipidemic drugs, polyunsaturated fatty acids, and eicosanoids are ligands for peroxisome proilferator-activated receptors {alpha} and {delta}. Proc. Natl. Acad. Sci. U.S.A. 94, 4312–4317.[Abstract/Free Full Text]

Gavrilova, O., Haluzik, M., Matsusue, K, Cutson, J. J., Johnson, L., Dietz, K. R., Nicol, C. J., Vinson, C., Gonzalez, F. J., and Retiman, M. L. (2003). Liver peroxisome proliferator-activated receptor {gamma} contributes to hepatic steatosis, triglyceride clearance, and regulation of body fat mass. J. Biol. Chem. 278(36), 34268–34276.[Abstract/Free Full Text]

Horton, J. D. (2002). Sterol regulatory element-binding proteins: Transcriptional activators of lipid synthesis. Biochem. Soc. Trans. 30, 1091–1095.[ISI][Medline]

Hulthe, J., Hulten, L. M., and Fagerberg, B. (2003). Low adipocyte-derived plasma protein adiponectin concentrations are associated with the metabolic syndrome and small dense low-density lipoprotein particles: Atherosclerosis and insulin resistance study. Metabolism 52, 1612–1614.[CrossRef][ISI][Medline]

Hussain, S. M., and Frazier J. M. (2002). Cellular toxicity of hydrazine in primary rat hepatocytes. Toxicol. Sci. 69, 424–432.[Abstract/Free Full Text]

Jenner, A. M., and Timbrell, J. A. (1995). In vitro microsomal metabolism of hydrazine. Xenobiotica. 25, 599–609.[ISI][Medline]

Kasus-Jacobi, A., Ou, J., Bashmakov, Y. K., Shelton, J. M., Richardson, J. A., Goldstein, J. L., and Brown, M. S. (2003). Characterization of mouse short-chain aldehyde reductase (SCALD), and enzyme regulated by sterol regulatory element-binding proteins. J. Biol. Chem. 278, 32380–32389.[Abstract/Free Full Text]

Kobayashi, T., Niimi, S., Kawanishi, T., Fukuoka, M., and Hayakawa, T. (2003). Changes in peroxisome proliferator-activated receptor gamma-regulated gene expression and inhibin/activin-follistatin system gene expression in rat testis after an administration of di-n-butyl phthalate. Toxicol. Lett. 138, 215–225.[CrossRef][ISI][Medline]

Lamb, R. G., and Banks, W. L. (1979). Effect of HD exposure on hepatic triacylglycerol biosynthesis. Biochim. Biophys. Acta. 574, 440–447.[ISI][Medline]

Lauterburg, B. H., Smith, C. V., Todd, E. L., and Mitchell, J. R. (1985). Pharmacokinetics of the toxic HD metabolites formed from isoniazid in humans. J. Pharmacol. Exp. Ther. 235, 566–570.[Abstract]

Lee, C.-H., Olson, P., and Evans, R. M. (2003). Minireview: Lipid metabolism, metabolic diseases, and peroxisome proliferator-activated receptors. Endocrinology 144(6), 2201–2207.[Abstract/Free Full Text]

Le May, C., Pineay, T., Bigot, K., Kohl, C., Girard, J., and Pérgorier, J.-P. (2000). Reduced hepatic fatty acid oxidation in fasting PPAR{alpha} null mice is due to impaired mitochondrial hydroxymethylglutaryl-CoA synthase gene expression. FEBS Lett. 475, 163–166.[CrossRef][ISI][Medline]

Madsen, L., Garras, A., Asins, G., Serra, D., Hegardt, F. G., and Berge, R. K. (1999). Mictochondrial 3-hydroxy-3-methylglutaryl coenzyme A synthase and carnitine palmitoyltransferase II as potential control sites for ketogenesis during mitochondrion and peroxisome proliferation. Biochem. Pharmacol. 57, 1011–1019.[CrossRef][ISI][Medline]

Maxwell, K. N., Soccio, R. E., Duncan, E. M., Sehayek, E., and Breslow, J. L. (2003). Novel putative SREBP and LXR target genes identified by microarray analysis in liver of cholesterol-fed mice. J. Lipid. Res. 44, 2109–2119.[Abstract/Free Full Text]

McQueen, C. A., and Chau, B. (2003). Neonatal ontogeny of murine arylamine N-acetyltransferases: Implications for arylamine genotoxicity. Toxicol. Sci. 73, 279–286.[Abstract/Free Full Text]

Mitchell, J. R., Thorgeirsson, U. P., Black, M., Timbrell, J. A., Snodgrass, W. R., Potter, W. Z., Jollow, H. R., and Keiser, H. R. (1975). Increased incidence of isoniazid hepatitis in rapid acetylators: Possible relation to hydrazine metabolites. Clin. Pharmacol. Ther. 18, 70–79.[ISI][Medline]

Patel, L., Buckels, A. C., Kinghorn, I. J., Murdock, P. R., Holbrook, J. D., Plumpton, C., Macphee, C. H., and Smith, S. A. (2003). Resistin is expressed in human macrophages and directly regulated by PPARg activators. Biochem. Biophys. Res. Commun. 300, 472–476.[CrossRef][ISI][Medline]

Preece, N. E., Nicholson, J. K., and Timbrell, J. A. (1991). Identification of novel hydrazine metabolites by 15N-NMR. Biochem. Pharmacol. 41, 1319–1324.[CrossRef][ISI][Medline]

Rao, M. S., and Reddy, J. K. (2001). Peroxisomal beta-oxidation and steatohepatitis. Semin. Liver Dis. 21, 43–55.[CrossRef][ISI][Medline]

Rawson, R. (2003) The SREBP pathway—insights fron insigs and insects. Nature Rev. Mol. Cell Biol. 4, 631–640.[CrossRef][ISI][Medline]

Reddy, J. K. (2001) Nonalcoholic steatosis and steatohepatitis III: Peroxisomal ß-oxidation, PPAR{alpha}, and steatohepatitis. Am. J. Physiol. Gastrointest. Liver Physiol. G1333–G1339.

Runge-Morris, M., Feng, Y., Zangar, R. C., and Novak, R. F. (1996) Effects of hydrazine, phenylezine, and hydralazine treatment on rat hepatic and renal drug-metabolizing enzyme expression. Drug Metab. Dispo. 24, 734–737.[Abstract]

Sarich, T. C., Adams, S. P., Petricca, G., and Wright, J. M. (1999). Inhibition of isoniazid-induced hepatotoxicity in rabbits by pretreatment with an amidase inhibitor. J. Pharmacol. Exp. Ther. 289, 695–702.[Abstract/Free Full Text]

Sasaki, Y. F., Saga, A., Akasaka, M., Ishibashi, S., Yoshida, K., Su, Y. Q., Matsusaka, N., and Tsuda, S. (1998). Organ-specific genotoxicity of the potent rodent colon carcinogen 1,2-dimethylhydrazine and three hydrazine derivatives: Difference between intraperitoneal and oral administration. Mutat. Res. 415, 1–12.[ISI][Medline]

Scales, M. D., and Timbrell, J. A. (1982). Studies on hydrazine hepatotoxicity. 1. Pathological findings. J. Toxicol. Environ. Health 10, 941–953.[ISI][Medline]

Stan, S., Delvin, E., Lambert, M., and Seidman, E. (2003) Apo A-IV: An update on regulation and physiologic functions. Biochim. Biophys. Acta 1631, 177–187.[ISI][Medline]

Stuart, R. L., and Grayson, M. L. (1999). A review of isoniazid-related hepatotoxicity during chemoprophylaxis. Aust. N.Z. J. Med. 29, 362–367.[ISI][Medline]

Timbrell, J. A., Mitchell, J. R., Snodgrass, W. R., and Nelson, S. D. (1980). Isoniazid hepatotoxicity: The relationship between covalent binding and metabolism in vivo. J. Pharmacol. Exp. Ther. 213, 364–369.[Abstract]

Timbrell, J. A., Scales, M. D., and Streeter, A. J. (1982). Studies on hydrazine hepatotoxicity. 2. Biochemical findings. J. Toxicol. Environ. Health. 10, 955–968.[ISI][Medline]

Timbrell, J. A., and Wright, J. M. (1984). Urinary metabolic profile of isoniazid in patients developing isoniazid related liver damage. Hum. Toxicol. 3, 485–495.[ISI][Medline]

Trout, D. L. (1965). Effect of hydrazine on plasma free fatty acid transport. Biochem. Pharmacol. 14, 813–821.[CrossRef][ISI][Medline]

van der Vliet, H. N., Sammel, M. G., Leegwater, A. C. J, Levels, J. H. M, Reitsma, P. H., Boers, W., and Chamuleau, A. F. M. (2001). Apolipoprotein A-V: A novel apolipoprotein associated with an early phase of liver regeneration. J.Biol. Chem. 276, 44512–44520.[Abstract/Free Full Text]

Vu-Dac, N., Gervois, P., Jakel, H., Nowak, N., Baugé, E., Dehondt, H., Staels, B., Pennacchio, L. A., Rubin, E. M., Fruchart-Najib, J., et al. (2003). Apolipoprotein A5, a crucial determinant of plasma triglyceride levels, is highly responsive to PPAR{alpha} activators. J. Biol. Chem. 278, 17982–17985.[Abstract/Free Full Text]

Walubo, A., Smith, P., and Folb, P. I. (1998). The role of oxygen free radicals in isoniazid-induced hepatotoxicity. Methods Find. Exp. Clin. Pharmacol. 20, 649–655.[ISI][Medline]

Weber, W. W. (1987). The Acetylator Genes and Drug Response, pp. 1–248. Oxford Press, New York.





This Article
Abstract
FREE Full Text (PDF)
All Versions of this Article:
82/1/318    most recent
kfh232v1
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Disclaimer
Request Permissions
Google Scholar
Articles by Richards, V. E.
Articles by McQueen, C. A.
PubMed
PubMed Citation
Articles by Richards, V. E.
Articles by McQueen, C. A.