Constitutive Expression of Peroxisome Proliferator-Activated Receptor -Regulated Genes in Dwarf Mice

    CIIT Centers for Health Research, Research Triangle Park, North Carolina (A.J.S., A.L., R.A.S., J.C.C., C.S.)
    Department of Pharmacology, Physiology
    Therapeutics, University of North Dakota School of Medicine, Grand Forks, North Dakota (H.B.B.)
    National Cancer Institute and National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina (J.L., M.P.W.), Toxicology and Mycotoxin Research Unit, Agricultural Research Service, United States Department of Agriculture, Athens, Georgia (K.A.V.), Department of Biomedical Sciences, College of Osteopathic Medicine and Edison Biotechnology Institute, Ohio University, Athens, Ohio (J.J.K.)
    ToxicoGenomics, Chapel Hill, North Carolina (J.C.C.)

    Abstract

    Defects in growth hormone secretion or signaling in mice are associated with decreased body weights (dwarfism), increased longevity, increased resistance to stress, and decreases in factors that contribute to cardiovascular disease and cancer. Peroxisome proliferators (PP) alter a subset of these changes in wild-type mice through activation of the nuclear receptor family member PP-activated receptor  (PPAR). We tested the hypothesis that an overlap in the transcriptional programs between untreated dwarf mice and PP-treated wild-type mice underlies these similarities. Using transcript profiling, we observed a statistically significant overlap in the expression of genes differentially regulated in control Snell dwarf mice (Pit-1dw) compared with phenotypically normal heterozygote (+/dw) control mice and those altered by the PP 4-chloro-6-(2,3-xylidino)-2-pyrimidinyl)thioacetic acid (WY-14,643) in +/dw mice. The genes included those involved in - and -oxidation of fatty acids (Acox1, Cyp4a10, Cyp4a14) and those involved in stress responses (the chaperonin, T-complex protein1) and cardiovascular disease (fibrinogen). The levels of some of these gene products were also altered in other dwarf mouse models, including Ames, Little, and growth hormone receptor-null mice. The constitutive increases in PPAR-regulated genes may be partly caused by increased expression of PPAR mRNA and protein as observed in the livers of control Snell dwarf mice. These results indicate that some of the beneficial effects associated with the dwarf phenotype may be caused by constitutive activation of PPAR and regulated genes.

    The peroxisome proliferator-activated receptors (PPAR) are members of the nuclear receptor superfamily and are activated by a structurally diverse group of compounds. Many of these compounds increase the size and number of peroxisomes and are thus called peroxisome proliferators (PP). The three PPAR subtypes (, /, ) have unique tissue distributions and ligand-specificities. In rodents, PP elicit a predictable course of adaptive responses in the liver, including peroxisome proliferation, induction of lipid-metabolizing genes, and hepatomegaly (Corton et al., 2000). There is overwhelming evidence that PPAR mediates most if not all of these effects in the rodent liver (Lee et al., 1995; Klaunig et al., 2003).

    Growth hormone plays an essential role in maintaining cellular and tissue homeostasis. Many PP-regulated genes are under control of growth hormone. Hypophysectomization of female rats enhanced expression of PP-inducible proteins that was reversed by growth hormone infusion (Sugiyama et al., 1994). STAT5b, a growth hormone-inducible transcription factor, inhibited the ability of PPAR to activate PPAR-dependent reporter gene transcription by endogenous or xenobiotic PP in vitro (Zhou and Waxman, 1999a,b). STAT5b-null mice expressed higher levels of some PPAR-regulated gene products involved in lipid metabolism (Zhou et al., 2002). These studies provide evidence that growth hormone negatively regulates PPAR in the intact animal through activation of STAT5b.

    Strains of dwarf mice have lower circulating levels of growth hormone or defects in growth hormone signaling (Tatar et al., 2003). Snell dwarf mice carry a mutation in the Pit1 gene (Pit1dw) required for development of pituitary cell bodies, which produce growth hormone, prolactin, and thyrotropin. The levels of these hormones are virtually undetectable in these mice. Ames dwarf mice are homozygous for the df mutation on the Prophet of Pit-1 gene (Prop1df), a transcription factor controlling expression of Pit1. Ames mice have very low levels of circulating growth hormone, prolactin, and thyrotropin. "Little" mice carry a mutation in the growth hormone-releasing hormone receptor (Ghrhr) gene and mice homozygous for the Ghrhrlit mutation cannot respond to hypothalamic growth hormone-releasing hormone. Little mice have 5% of normal levels of circulating growth hormone. Inactivation of the growth hormone receptor/binding protein (Ghr) by homologous recombination results in a dwarf mouse unable to respond to growth hormone (Zhou et al., 1997). Dwarf mice live significantly longer than their heterozygous littermates maintained under the same conditions (Tatar et al., 2003).

    Dwarf mice share phenotypic similarities with PP-treated rodent models. Compared with their heterozygote counterparts, Ames dwarf mice had decreased circulating cholesterol, triglycerides (H. Brown-Borg, unpublished observations), insulin and glucose levels (Borg et al., 1995), whereas growth hormone increased triglyceride and cholesterol levels in wild-type mice (Marmary et al., 1999). PPAR agonists have been used clinically for many years to lower cholesterol and triglyceride levels in patients at risk of coronary heart disease (Corton et al., 2000). Snell dwarf mice had decreased incidences and severities of dimethyl-benzanthracene-induced skin papillomas (Bielschowsky and Bielschowsky, 1961) and sarcoma 180 (Rennels et al., 1965). Ames dwarf mice had decreased numbers and severity grade of spontaneous lung adenocarcinomas (Ikeno et al., 2003), and the Little dwarf mutation suppressed spontaneous liver tumors in the susceptible C3H/HeJ strain (Bugni et al., 2001). In contrast to the well known effects of strong PP on induction of liver cancer (Klaunig et al., 2003), PPAR agonists suppressed certain types of diethyl nitrosamine-induced preneoplastic foci in the rat liver (Cattley and Popp, 1989) as well as skin cancer in mice (Thuillier et al., 2000). PPAR-null mice had a higher incidence of spontaneous hepatocellular adenomas and carcinomas than wild-type mice (Howroyd et al., 2004). Snell dwarf mouse fibroblasts exhibited increased resistance to diverse physical and chemical stressors (Murakami et al., 2003). Likewise, either pretreatment with PPAR agonists in wild-type mice or an intact PPAR gene itself (compared with PPAR-null mice) protected the livers of mice from chemical or physical stress (Mehendale, 2000; Anderson et al., 2002, 2004). One explanation for this phenotypic overlap is that dwarf mice have an increased level of PPAR-regulated gene expression. Although there is clear evidence that STAT5b-null mice express higher levels of PPAR-regulated gene products (Zhou et al., 2002), the effect of mutations in other genes that control growth hormone secretion and signaling on PPAR-regulated gene expression is not known.

    We posed the hypothesis that the overlap in the phenotypic characteristics of PP-treated rodents and dwarf mice has at its basis an overlap in the transcriptional programs regulated by PPAR. We tested this hypothesis by examining transcript profiles in the livers of control and PP-treated wild-type and Snell dwarf mice. Our analysis of a number of transcriptional and post-transcriptional targets demonstrated that the overlap includes PPAR-dependent genes with functions in lipid metabolism, stress responses, and cardiovascular disease.

    Materials and Methods

    Animal Treatments. This study was conducted under federal guidelines for the use and care of laboratory animals and was approved by the Institutional Animal Care and Use Committees for CIIT Centers for Health Research. Male mice were used throughout these studies. Nine-week-old DW/J-Pit1dw/dwln/ln homozygous (dw/dw) Snell dwarf, DW/J-Pit1+/dwln/ln heterozygous (+/dw), C57BL/6J-Ghrhrlit homozygous (lit/lit) Little dwarf or heterozygous (+/lit) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Wild-type and PPAR-null mice on a SV129 background, 12 weeks of age, were from a colony established at CIIT. Control and treated mice were provided with NIH-07 rodent chow (Ziegler Brothers, Gardner, PA) and deionized filtered water ad libitum. Lighting was on a 12-h light/dark cycle. Snell dwarf and heterozygous mice were given a single gavage dose of WY-14,643 (ChemSyn Science Laboratories, Lenexa, KS) at a concentration of 50 mg/kg body weight in methylcellulose vehicle (0.1%) and sacrificed 12 h later. Snell, Little mice and their heterozygous control mice were given gavage doses (50 mg/kg body weight) each day for 3 days with sacrifice 24 h after the last dose. Control mice were dosed with methylcellulose (0.1%) vehicle alone. In separate experiments, wild-type and PPAR-null mice were fed a control diet or a diet containing WY-14,643 in the diet (500 ppm) for 1 week. At the designated time after treatment (12 h, 72 h, or 1 week), mice were deeply anesthetized by pentobarbital injection and killed by exsanguination. The livers were removed, rinsed with isotonic saline, snap-frozen in liquid nitrogen, and stored at eC70°C until analysis.

    Ames dwarf (Prop1df) mice, 16 weeks of age, and GHR-null mice, 12 to 16 weeks of age and corresponding groups of age-matched heterozygote (+/df) or wild-type mice were maintained at the University of North Dakota (Brown-Borg and Rakoczy, 2000) or Ohio University (Zhou et al., 1997) vivarium facilities. All procedures involving animals were reviewed and approved by the appropriate Institutional Animal Care and Use Committees.

    Analysis of Gene Expression by Gene Arrays. Because we were initially interested in effects of dwarf mutations on the ability of PPAR to regulate endpoints associated with liver cancer (Styles et al., 1990), we used the Atlas 1.2 Cancer arrays (BD Biosciences Clontech, Palo Alto, CA) containing 1178 genes with direct or indirect involvement in cancer. Total RNA was isolated by modification of guanidium isothiocyanate method using RNA Stat60 according to the manufacturer's instructions (Tel-Test, Friendswood, TX). 32P-labeled first-strand cDNA was generated by reverse-transcribing poly A+ RNA in the presence of oligo-dT primers. 32P-labeled cDNA was purified using ProbeQuant G-50 Micro Columns (Amersham Biosciences, Piscataway, NJ). Membranes were prehybridized with salmon testes DNA to block nonspecific binding of probe to the membrane. Labeled probes were denatured and incubated with Cot-1 DNA to block repetitive sequences and subsequently neutralized before adding to the membrane. Labeled probe from each of 12 samples (three mice in each of four groups) representing approximately 2 to 5 x 106 cpm was incubated with a separate but identical array overnight at 68°C. Membranes were heat-sealed in Seal-A-Meal bags and exposed to PhosphorImager autoradiography cassettes for 24 h. Macroarray imaging was performed using the Molecular Imaging System SI (Bio-Rad, Hercules, CA). Macroarray data analysis was performed using Atlas Image 1.5 software (BD Biosciences Clontech) to determine levels of alterations in gene expression. Local background-subtracted spot intensities (Int) were log2 transformed. The transformed data were then fitted to this following ANOVA-type model: Yijk = log2 (Intijk) = ij (k) + ik +ijk, where i represents the ith strain/treatment, j represents the jth replicate of a strain/treatment, and k represents the kth gene on the array. In the initial stage, trend effects (k) are removed by local regression (Kepler et al., 2002), which assumes that the expression of most genes is not changed from treatment to treatment. Genes with low "between-treatment variation" (i.e., variance of the gene across all replicates of all treatments) compared with the "within-treatment variation" (i.e., variance of the replicates of a single treatment) are selected iteratively to create a "not changed" population (i.e., high "within-treatment variation" genes). It is assumed that the "between-treatment effect" for genes in this population is approximately zero. The entire population is then normalized based on this selected subset and the overall treatment effect (ik) is calculated. The entire modeling process repeats (in cyclical fashion) until a stable, recurring set of "not changed" genes is obtained. For this ANOVA-type model, the interaction term (ij (k)) considers the "array effect" and "gene effect" to be related; thus, it is nonparametric. The threshold for significance was set at p < 0.005, and genes that exhibited a 1.5-fold or eC1.5-fold were reported as a log2 fold-change relative to the control. Regulated genes were visualized using CLUSTER and TreeView (Eisen et al., 1998). Spearman rank correlation test of the regulated genes was performed using SAS (ver. 6.12; SAS Institute, Research Triangle Park, NC).

    Real-Time RT-PCR Analysis. Expression levels of selected genes were quantified using real-time RT-PCR analysis. In brief, total RNA was extracted as described above, purified with RNeasy column and on-column DNase I digestion (QIAGEN, Valencia, CA). Purified total RNA was reverse-transcribed with MuLV reverse transcriptase and oligo-dT primers. The forward and reverse primers (Table 1) were designed using Primer Express software, v2.0 (Applied Biosystems, Foster City, CA). The SYBR green PCR master mix (Applied Biosystems) was used for real-time PCR analysis. The dissociation curve (melting curve) for each gene was performed to verify the quality of the primers. The standard curve for each gene was performed to quantify the gene expression. The expressions of genes were first normalized with 18 S ribosomal RNA gene of the same sample, and then the relative differences between control and treatment groups were calculated and expressed as relative increases setting the control as 100%.

    Analysis of Palmitoyl-CoA Oxidase Activity. Analysis of palmitoyl-CoA oxidase activity was determined using the procedures in Sausen et al. (1995). Liver from the left lobe was used to prepare 20% homogenates in 50 mM Tris-HCl and 154 mM KCl, pH 7.2. Homogenates were prepared on the day of enzyme assays by centrifugation at 2500g for 10 min. Palmitoyl-CoA oxidase activity was assayed by measurement of hydrogen peroxide production in the presence of 25 e palmitoyl-CoA. Enzyme activity was normalized per gram of protein.

    Western Analysis. Liver lysates were prepared in 250 mM sucrose, 10 mM Tris-HCl, pH 7.4, and 1 mM EDTA with protease inhibitors (0.2 mM phenylmethylsulfonyl fluoride, 0.1% aprotinin, 1 e/ml pepstatin, and 1 e/ml leupeptin). Fifty micrograms of whole-cell lysate was subjected to 12% SDS-polyacrylamide gel electrophoresis followed by transfer to nitrocellulose membranes. Immunoblots were developed using primary antibodies against acyl-CoA oxidase (ACO) (a gift from Dr. S. Alexson, Huddinge University Hospital, Huddinge, Sweden), MFP-I, MFP-II, and thiolase (gifts from Dr. T. Hashimoto, Japan), Cyp4a (BD Gentest, Waltham, MA), PPAR (Affinity Bioreagents, Inc., Golden, CO), T-complex protein1, and other chaperones and chaperonins (Stress Gen, Victoria, BC, Canada), and 1-antitrypsin and fibrinogen (Affinity Biologicals, South Bend, IN) followed by appropriate secondary antibodies conjugated with horseradish peroxidase (Santa Cruz Biotechnology, Santa Cruz, CA) in the presence of chemiluminescent substrate ECL (Amersham Biosciences, Piscataway, NJ). Cyp4a antibody probably recognizes multiple members of the Cyp4a family (Cyp4a10, Cyp4a12, Cyp4a14). MFP-I probably recognizes multiple forms of the MFP-I complex, including protein products encoded by Dci, Ech1, and Ehhadh. MFP-II was raised against the product of the gene Hsd17b4. Blots were quantitated densitometrically (BioRad Imaging Densitometer and Molecular Analyst software) after exposure of the membrane to X-ray film.

    Statistical Analysis of Data. Means and S.E.M. (n = 3eC4) for RT-PCR data were calculated by the Students' t test. The level of significance was set at p  0.05. The expression ratios between WY-14,643-treatment and control mice and between dwarf mice and heterozygote mice were used for comparisons. For Western data, a two-way ANOVA was used to determine statistical significance (p  0.05). Students' t tests were used for comparisons within genotype when significant interactions were detected between genotype and treatment. Spearman rank correlation test was performed using SAS.

    Results

    The pit-1dw Mutation Results in Constitutive Increases in a Subset of WY-14,643-Responsive Genes. We used transcript profiling to gain insight into the underlying gene expression patterns that may explain phenotypic similarities between dwarf mice and PP-treated heterozygote mice. Heterozygous (+/dw) and homozygous (dw/dw) Snell dwarf mice were treated with the PP WY-14,643 or methylcellulose carrier and sacrificed 12 h later. The livers were examined for gene expression using the Atlas mouse cancer 1.2 arrays containing 1178 genes. Genes with a p value 0.005 and 1.5-fold or eC1.5-fold were reported as a log2 fold-change relative to the respective control genes. Regulated genes were visualized using CLUSTER and TreeView programs (Eisen et al., 1998) using two-dimensional clustering. Figure 1A shows the 111 genes with significant changes in one or more groups. The regulated genes were divided into groups (AeCH) based on expression behavior differences. Although the Atlas mouse cancer arrays do not contain typical markers of PPAR activation such as genes involved in fatty acid -oxidation, the genes regulated by WY-14,643 in the +/dw mice versus control +/dw mice (Fig. 1A, groups BeCD, F, and G) included genes or their products regulated by PP in other studies. These genes were cathepsin B (Conway et al., 1989), adipose differentiation-related protein (Liu et al., 2003), as well as PPAR itself (Sterchele et al., 1996) (Table 2). WY-14,643 exposure in +/dw mice also led to increased expression of chaperones (Hsp86 and Hsp84) and chaperonin genes [chaperonin subunits 4 (), 6a (), 6b (), 7 ()]. We showed in recent studies that PP exposure increases expression of a large number of genes in wild-type mice that maintain the health of the proteome, including Hsp86 and chaperonin subunits (Anderson et al., 2004).

    Gene classes were determined by hierarchical clustering (Fig. 1A). Gene coordinates, NCBI numbers, and names are according to BD Biosciences Clontech.

    The dw mutation itself resulted in the altered regulation of a total of 41 genes found within groups A, C, G and H (Fig. 1A). Using GenBank numbers as unique identifiers, we compared our list of genes with the genes identified as different in 6-month-old dw/dw mice compared with wild-type control mice using a similar array platform (Dozmorov et al., 2002). Only four of the genes identified in the previous study (insulin receptor, interleukin 15, Cyp1a1, and insulin-like binding protein 1) were altered in one of the three comparisons in our study. None of these genes was significantly altered in the control dw/dw versus control +/dw mice in our study; however, insulin-like binding protein 1 (Igf1bp1) was up-regulated in control dw/dw versus control +/dw mice (2-fold; p = 0.034) but did not pass the statistical cut-off. Igf1bp1 was found by RT-PCR to be induced in control dw/dw mice (Table 3). The fact that there was no overlap in the two studies could be attributable to differences in animal husbandry and age of mice (9 weeks versus 6 months.).

    Data are mean ± S.E.M. of four individual animals.

    Hierarchical clustering of the compared groups showed that the control dw/dw pattern was more similar to that of the WY-14,643-treated +/dw pattern than that of the pattern of WY-14,643-treated dw/dw mice (Fig. 1A). We used Spearman correlation to determine the statistical significance of similarities between groups. The control dw/dw pattern was more similar to WY-14,643 +/dw pattern (0.216, p = 0.023) than the WY-14,643 dw/dw pattern (eC0.013; p = 0.892). This similarity was driven primarily by 16 of 40 genes (40%) similarly regulated in control dw/dw and WY-14,643-treated +/dw mice. These included 12 up-regulated (group C) and 4 down-regulated (group G) genes (Fig. 1B). The genes included down-regulation of receptors for interleukin-1 and interferon  (Fig. 1, group G) predicted to decrease responsiveness to these inflammatory mediators. These data indicate that control dw/dw mice constitutively express genes that are also regulated by WY-14,643 in +/dw mice.

    The WY-14,643 +/dw and WY-14,643 dw/dw patterns exhibited similarity that approached significance (0.179; p = 0.06). WY-14,643 treatment in dw/dw mice did little to further alter the expression of most of the overlapping genes regulated in dw/dw mice. However, two genes up-regulated in control dw/dw mice (epoxide hydrolase 1, microsomal and RAN) were further up-regulated by WY-14,643, whereas three genes up-regulated in control dw/dw mice [angiotensin II receptor, type 2; protein tyrosine phosphatase, receptor type, M; cell division cycle 2 homolog (Schizosaccharomyces pombe)-like 1] were down-regulated by WY-14,643, indicating more complex control by the Pit-1dw gene and WY-14,643.

    Given the overlap in the transcript profiles of control dw/dw and WY-14,643 in +/dw, we determined the mRNA expression of known PPAR-regulated genes in the livers of +/dw and dw/dw mice treated with WY-14,643 or carrier for 3 days. We chose a 3-day treatment period to facilitate comparison of mRNA and protein levels described in further detail below. Genes with known roles in fatty acid metabolism were induced by WY-14,643 in +/dw mice, as expected. The genes included Acox1 (acyl-CoA oxidase 1, peroxisomal), Cte1 (cytoplasmic thioesterase 1), Cyp4a10, Cyp4a14, Dci (dodecenoyl-CoA delta isomerase), Ech1 (enoyl CoA hydratase 1, Ehhadh (enoyl-CoA, hydratase/3-hydroxyacyl CoA dehydrogenase), Fabp4 (fatty acid binding protein 4), Mte1 (mitochondrial thioesterase 1), and Pex11a (peroxisomal biogenesis factor 11a) (Table 3). A number of these genes were constitutively up-regulated in control dw/dw mice compared with control +/dw mice including Acox1, Cyp4a10, Cyp4a14, Dci, Ech1, Ehhadh, Fabp4, and Pex11a; Acox1, Cyp4a10, Cyp4a14, Ech1, and Ehhadh and Fabp4 achieved significance. A subset of these genes was further increased in dwarf mice including Acox1, Cyp4a10, Cyp4a14, Dci, and Ehhadh; Acox1, Cyp4a10, Cyp4a14, Dci, and Ehhadh became significant. The remaining genes, except for Fabp4 and Pex11a, were increased by WY-14,643 in dwarf mice to about the same extent in dwarf and heterozygote mice.

    Genes involved in a number of other functional categories were examined. Insulin signaling pathways were altered in control dw/dw mice including dramatic down-regulation of Igf1 (insulin-like growth factor 1) and up-regulation of Igfbp1. WY-14,643 did not alter the expression of these genes. Genes that help maintain the health of the proteome (Cct3, Cct5, Hsp60, Hsp70) were up-regulated in control dw/dw mice versus control +/dw mice. Fibrinogen -chain (Fab) previously shown to be down-regulated by PP (Corton et al., 1998) and cyclin D1 (Ccnd1) were down-regulated in dw/dw mice. These results demonstrate that control dw/dw mice exhibit features of their transcriptional profiles similar to PP-treated heterozygote mice.

    We next examined the expression of gene products regulated by PP in a PPAR-dependent manner in the liver. ACO, the first rate-limiting enzyme in the fatty acid -oxidation pathway, is expressed as an inactive precursor (ACO-A) cleaved to active forms ACO-B and ACO-C. The antibody we used can detect ACO-C, but because expression is weak and sometimes variable, only ACO-A and ACO-B forms are shown in the following experiments. Levels of ACO protein expression were barely detectable in untreated +/dw mice (Fig. 2A). WY-14,643 treatment increased ACO-A and ACO-B levels in +/dw mice after 3 days of exposure. In the absence of WY-14,643 treatment, dw/dw mice express higher levels of ACO-B compared with untreated +/dw mice, but the increase was not significant (Fig. 2, A and B). After exposure to WY-14,643, ACO-A but not ACO-B levels were increased in the livers of dw/dw mice to the same extent as those in control animals. Palmitoyl-CoA oxidase activity, a measure of ACO activity, was increased in WY-14,643-trreated +/dw mice compared with control +/dw mice but did not reach statistical significance in dw/dw WY-14,643-treated versus control mice (Fig. 2C).

    Cyp4a protein levels were increased in WY-14,643-treated +/dw mice compared with control mice (Fig. 2, A and B). The control dw/dw mice constitutively express higher levels of Cyp4a protein compared with control treated +/dw mice. In dw/dw mice, treatment with WY-14,643 did not further increase Cyp4a expression above that observed in the untreated dw/dw control mice.

    Other gene products regulated by PPAR were also examined, including three proteins involved in peroxisomal fatty acid -oxidation [multifunctional protein-I (MFP-I), MFP-II, and thiolase] and cytosolic and mitochondrial thioesterases [CTE-1 and MTE-1]. Control dw/dw mice did not exhibit statistically significant increases in MFP-I, MFP-II, MTE-I, thiolase, and CTE-I (Fig. 2, D and E). Thus, not all PP-responsive gene products are regulated in a similar manner by the pit-1dw mutation as seen in the RT-PCR studies. MFP-II and thiolase exhibited significantly greater increases in expression after WY-14,643 exposure in dw/dw mice than in +/dw mice, indicating that dw/dw mice are more responsive to PP induction of these gene products. Although these changes in protein expression are generally consistent with the RT-PCR results, differences may reflect both transcriptional and posttranscriptional control mechanisms by Pitdw and WY-14,643.

    To begin to explain the constitutive activation of PPAR-regulated genes, we determined whether PPAR mRNA and protein levels were directly altered in dw/dw mice. Messenger RNA levels of PPAR were elevated in control dw/dw mice compared with the +/dw mice (Fig. 3). In parallel with the PPAR mRNA, PPAR protein levels were increased 2.3-fold compared with +/dw mice. No changes were observed after WY-14,643 exposure in any strain. These results indicate that the increased expression of PPAR gene targets in the livers of untreated Snell dwarf mice may be partly caused by increased expression of PPAR.

    Alteration of PPAR-Regulated Gene Products in Other Dwarf Mouse Models. We examined expression of PP-dependent gene products in other types of dwarf mice. In control homozygous Ames mice (df/df), ACO-B, and Cyp4a protein expression levels were increased compared with their heterozygote (+/df) littermates (Fig. 4A). Other proteins examined above were not grossly affected in the df/df mouse strain under these conditions, including MFP-I, thiolase, MTE-I, and CTE-I (data not shown).

    We examined the expression of PPAR-regulated gene products in "Little" mice that carry a mutation in the Ghrhr gene. Compared with the Snell and Ames dwarf mice, the control lit/lit mice did not exhibit constitutive increases in any of the proteins examined (Fig. 4, B and C). However, the lit/lit mice exhibited increased expression of ACO-A, ACO-B, MFP-II thiolase, and MTE-1 proteins after WY-14,643 exposure compared with WY-14,643-treated +/lit mice, but only MFP-II attained statistical significance because of significant interactions between genotype and treatment for the other proteins (Fig. 4, B and C).

    We examined fatty acid catabolism protein expression in the livers of control wild-type and GHR-null mice. Cyp4a and thiolase exhibited increased levels in untreated GHR-null mice compared with the untreated wild-type mice (Fig. 4D). Other fatty acid metabolism gene products were not appreciably altered in the GHR-null mice (data not shown). Taken together, these results indicate that in dwarf mice, expression of a subset of PPAR-regulated gene products is either constitutively up-regulated (Snell, Ames and GHR-null dwarf mice) or has increased responsiveness to a PP (Snell, Little dwarf mice), indicating that a growth hormone signaling pathway(s) plays a role in the regulation of several PPAR-dependent gene products.

    Altered Expression in Dwarf Mice of PPAR-Regulated Gene Products Involved in Stress Resistance and Cardiovascular Disease. Based on the results of the transcript profiles, we examined the expression of gene products involved in protein folding and stress resistance that were increased by WY-14,643. The large family of proteins involved in protein folding including chaperones and chaperonin proteins [also called T-complex protein or chaperonin-containing T-complex protein-1, (CCT)] were initially examined. We were particularly interested in those proteins that we had previously demonstrated to be altered by PP exposure in a PPAR-dependent manner, including Hsp86, Hsp70, and T-complex protein-1 subunits (Anderson et al., 2004). We examined the expression of these proteins in the livers of control and WY-14,643-treated +/dw and dw/dw mice. Although Hsp86 and Hsp70 proteins exhibited increased levels after WY-14,643 exposure, there were no detectable differences in expression between the strains (data not shown). The results are consistent with lack of changes in Hsp86 mRNA in dw/dw mice but do not reflect the increases in Hsp70 mRNA, indicating that additional mechanisms control protein expression. We examined expression of T-complex protein 1 family members to which antibodies were available (T-complex protein-1,-,-). The increase in T-complex protein-1 expression in the livers of wild-type SV129 mice was shown to be PPAR-dependent in that T-complex protein-1 was increased in wild-type but not PPAR-null mice after WY-14,643 exposure (Fig. 5A). T-complex protein-1 was increased in the control dw/dw strain but not the lit/lit strain compared with their respective heterozygote control mice (Fig. 5B). These results are consistent with the RT-PCR results showing a modest induction of the Cct5 gene. T-complex protein-1 expression was significantly increased after WY-14,643 treatment in both dw/dw and lit/lit dwarf strains as well as heterozygote strains (Fig. 5B). It is surprising that the expression of the Cct5 gene was not altered by WY-14,643 treatment in either strain. Like the fatty acid catabolism genes discussed above, T-complex protein-1 was increased to a greater extent in WY-14,643-treated dw/dw mice compared with WY-14,643-treated +/dw mice (Fig. 5B). T-complex protein-1 was also elevated in the livers of GHR-null mice compared with wild-type mice (Fig. 5C). No changes in the expression of other chaperones or chaperonins, including T-complex protein-1, T-complex protein-1, ERp72, Hsp25, Hsp60, Hsp65, and Hsp84 were noted in the livers or hearts from control and WY-14,643-treated Snell dwarf mice or in the livers from GHR-null mice (data not shown).

    We performed a preliminary examination of PPAR-regulated gene products in the kidneys of Snell dwarf mice from the same 3-day WY-14,643 study described above. Like the liver, Cyp4a protein levels were elevated in the kidneys of WY-14,643-treated +/dw and dw/dw mice (Fig. 5D). T-complex protein-1 exhibited increased expression in WY-14,643-treated +/dw mice compared with control +/dw mice (Fig. 5E). The expression of the chaperone Hsp25 was also increased in control dw/dw mice compared with control +/dw mice but not after WY-14,643-treatment in either strain (Fig. 5E). No changes in the expression of T-complex protein-1, T-complex protein-1, ERp72, Hsp60, Hsp65, and Hsp84 were noted in the kidneys from control and WY-14,643-treated Snell dwarf mice (data not shown). These results indicate that the chaperonin T-complex protein-1 and Hsp25 are constitutively elevated in the kidneys of Snell dwarf mice.

    PPAR regulates a large battery of genes expressed in the liver whose gene products play roles in atherosclerosis through lipid transport, inflammation, and clot formation. A number of these genes encode acute phase proteins (APP) that are elevated during times of acute or chronic infection. PP generally down-regulate the expression of these APP, possibly through the negative regulation by PPAR of transcription factors controlling APP gene expression (Corton et al., 2000). Given that the transcript profiling results predicted a general decrease in inflammatory responses in control dw/dw and WY-14,643-treated +/dw mice through decreases in receptors for interleukin 1 and interferon gamma, we examined the expression of APP in the livers of dwarf mice known to be regulated by WY-14,643 (Corton et al., 1998). All fibrinogen subunits (, , and ) were down-regulated in control dw/dw compared with control +/dw mice, although decreases in - and -fibrinogen did not reach statistical significance (Fig. 6A). Under these conditions WY-14,643 had no significant effect on fibrinogen levels in +/dw mice. Longer exposure times may be required to decrease fibrinogen gene and protein levels (Corton et al., 1998) and would explain why -fibrinogen, a gene on the Atlas Cancer 1.2 array, was not identified as down-regulated 12 h after WY-14,643 exposure. There was a significant decrease in the expression of - and -fibrinogen in WY-14,643-treated dw/dw mice compared with WY-14,643-treated +/dw mice. Another APP gene product, 1-antitrypsin, exhibited decreased expression in WY-14,643-treated +/dw compared with control +/dw mice (Fig. 6A). Decreased levels of 1-anti-trypsin in control and WY-14,643-treated dw/dw mice were not comparable with +/dw mice because of a significant interaction. -Fibrinogen levels exhibited decreases by WY-14,643 in both +/lit and lit/lit strains (Fig. 6B). Although levels of -fibrinogen were decreased in the control lit/lit compared with the control +/lit strain, a significant interaction between genotype and treatment prevented statistical comparison. Finally, GHR-null mice had decreased levels of -fibrinogen compared with wild-type control mice (Fig. 6C). These results indicate that some markers of inflammation and atherosclerosis are decreased in dwarf mice in a manner similar to PP exposure in wild-type mice.

    Discussion

    We posed the hypothesis that the overlap in the phenotypic effects of PP treatment in wild-type mice and control dwarf mice has as its basis an overlap in the transcriptional programs regulated by PPAR. We examined gene expression by transcript profiling in the livers of Snell (Pit-1dw) dwarf (dw/dw) and heterozygote (+/dw) mice treated with WY-14,643. Control dw/dw mice exhibited characteristics of PPAR activation. Of the 41 genes identified as differentially regulated in control dw/dw mice versus control +/dw mice, 41% were regulated similarly by WY-14,643 in +/dw mice. Many PPAR target genes and their protein products involved in fatty acid metabolism were increased in control Snell dwarf livers. The increase in PPAR-regulated gene products involved in lipid metabolism was found to various extents in the livers of three additional dwarf mouse models that have in common defects in growth hormone secretion and/or signaling. Ames dwarf mice exhibited increased expression of ACO and Cyp4a, whereas GHR-null mice exhibited increases in thiolase and Cyp4a. Little and Snell mice exhibited greater inductions of lipid metabolism gene products after WY-14,643 exposure compared with corresponding WY-14,643-treated heterozygote strains, demonstrating greater responsiveness to PP in these dwarf mice. The increases in lipid metabolism genes may be caused by increases in PPAR mRNA and protein as observed in the livers of control dw/dw mice. These results are consistent with two studies that showed an inverse relationship between growth hormone signaling and PPAR-regulated gene expression. Mice with defects in growth hormone-responsive STAT5b transcription factor also exhibited constitutive increases in PPAR-regulated gene expression (Zhou et al., 2002). On the other hand, a transcript profiling study in the livers of growth hormone-treated rats revealed a subset of PPAR-regulated genes that were down-regulated, including CYP4A3 as well as PPAR itself (Tollet-Egnell et al., 2001). Taken together, mice with defects in growth hormone secretion and signaling exhibit increased constitutive expression of a subset of lipid metabolizing gene products under control of PPAR.

    In addition to the lipid metabolism genes identified here, WY-14,643 and dwarf mutations alter additional gene products with links to cardiovascular disease. Fibrinogen and 1-anti-trypsin are acute phase proteins that are used as general indicators of inflammatory processes associated with atherogenesis (Barbier et al., 2002). An elevated plasma fibrinogen level is an independent risk factor for coronary heart disease, stroke, and peripheral vascular disease. APP genes are negatively regulated by PP in a PPAR-dependent manner (Corton et al., 1998, 2000). At least one of the fibrinogen family members was down-regulated in two dwarf mouse models (Snell, GHR-null). 1-Anti-trypsin was down-regulated in Snell mice. The basis for the down-regulation of the APP by PP may be the ability of PPAR to negatively interfere with the actions of the transcription factors nuclear factor-B, activator protein-1, nuclear factor of activated T cells, and CCAAT/enhancer-binding protein , which regulate pro-inflammatory genes (Barbier et al., 2002). Because PPAR is a common regulator of factors that impact atherosclerosis (Barbier et al., 2002), the induction of PPAR and regulated genes in dwarf mice may lead to decreases in not only circulating triglyceride and cholesterol levels (H. Brown-Borg, unpublished observations) through modulation of lipid metabolism genes but also negative regulation of inflammatory genes that contribute to cardiovascular disease.

    The chaperonin family of related T-complex protein-1 proteins interact with hydrophobic regions on nascent polypeptide chains, preventing aggregation and subsequent toxicity. Like the chaperone heat shock proteins, the chaperonins are induced after physical or chemical stress in which protein denaturation is a prominent feature (Yokota et al., 2000). Low-level exposure to different stressors leads to increased expression of chaperonins and chaperones that better protect the cell from further insults (Latchman, 2001). We found that the T-complex protein-1 protein was induced by both WY-14,643 and dwarf mutations. Increased expression was found in the livers of Snell and GHR-null mice. The T-complex protein-1 gene (Cct5) was also induced in control dw/dw mice but not by WY-14,643 in either strain. Although additional genes involved in protein folding were induced in control dw/dw mice including Cct3, Hsp60, and Hsp70, other genes were not, indicating that the dwarf mutations do not lead to a general increase in the expression of the protein-folding machinery. The livers of PP-treated wild-type mice in which many heat-shock proteins and T-complex protein-1 are induced are protected from different types of chemical insults (Mehendale, 2000), including oxidative stress (Anderson et al., 2004). The relationship between the expression of T-complex protein-1 and increased protection of Snell fibroblasts to different forms of stress (Murakami et al., 2003) needs to be examined.

    The basis of the constitutive activation of PPAR and PPAR-regulated gene products in dwarf mice may be the abolishment of growth hormone-responsive STAT5b interference with PPAR activity. The ability of STAT5b to negatively regulate PPAR required the ligand-independent activation function (AF-1) in the N terminus of PPAR but not the C-terminal ligand-dependent AF-2 (Zhou and Waxman, 1999a,b). These activation functions probably recognize overlapping sets of interacting coactivators and corepressors, resulting in regulation of distinct sets of genes. Negative interference of PPAR by STAT5b could occur by a number of mechanisms, including 1) interference with the ability of PPAR to bind a PPRE or interact with appropriate coactivators/corepressors or 2) competition for essential coactivators or other interacting proteins (Zhou and Waxman, 1999a,b). In our study, not all PP-dependent gene products were constitutively regulated by pit-1dw but required WY-14,643 exposure for induction in dwarf mice. Differences in pit-1 dependent (e.g., Acox1, Cyp4a10, Cyp4a14) and WY-14,643-dependent (e.g., Mte1) induction in Snell dwarf mice may reflect differences in the requirement for AF-1 versus AF-2 for gene activation. Only those genes that require the STAT5b-inhibitable AF-1 would be pit-1 dependent. Exposure of dwarf mice to WY-14,643 did not further increase the expression of some pit-1-dependent gene products (e.g., Fabp4), possibly because of the high intrinsic activity of AF-1 compared with AF-2.

    In addition to the interference of PPAR activity by STAT5b, PPAR can negatively interfere with the ability of STAT5b to activate growth hormone-responsive genes. The AF-1 of PPAR was required for negative interference of STAT5b, but how PPAR interferes with STAT5b function is not known (Shipley and Waxman, 2003). These studies imply that given the correct cellular context in which PPAR and STAT5b are both expressed, PP could act as dwarf mouse mimetics through inhibition of STAT5b. Thus, it may be possible not only that some of the beneficial effects of the dwarf genotype occur through activation of PPAR but also that activation of PPAR by PP in wild-type animals may negatively interfere with growth hormone signaling and lead to dwarf mouse-like effects. PP exposure leads to changes similar to those caused by defects in growth hormone signaling, including increases in the fatty acid-metabolizing genes identified here as well as down-regulation of growth hormone-regulated cytochrome P450 family members (Corton et al., 1998). Compounds that mimic the dwarf mouse phenotype would ideally target the PPAR AF-1. With this in mind, mitogen-activated protein kinase activates PPAR through the AF-1 activation domain important for insulin-dependent signaling (Juge-Aubry et al., 1999) and cardiac metabolic stress responses (Barger et al., 2001).

    PPAR may play roles in other models of stress resistance and/or longevity. Using comprehensive transcript profiling, PPAR was required for 20% of the gene expression changes in the liver after a 5-week caloric restriction (CR) and is required for CR to protect the liver from a lethal dose of thioacetamide, a hepatotoxic agent (Corton et al., 2004). As in the livers of Snell dwarf mice, the PPAR-dependent CR genes included those involved in fatty acid metabolism (i.e., Cyp4a10, Cyp4a14). With our findings of increased expression of PPAR-dependent genes in dwarf mice, alterations in PPAR target gene regulation seems to be common in many animal models of longevity and/or stress resistance. A dwarf mouse strain in which the PPAR gene is inactivated will be important to unequivocally establish PPAR's role in the beneficial effects associated with defects in growth hormone signaling.

    Acknowledgements

    We thank Dr. Andrzej Bartke for tissues, Dennis House for assistance in performing some of the statistics, the CIIT Animal Care and Necropsy and Histology Units for assistance in performing these studies, Drs. Alexson and Hashimoto for antibodies, Drs. Igor Dozmorov and Richard Miller for providing the list of genes regulated in dwarf mice, Sharlene Rakoczy for technical assistance, and Drs. Kevin Gaido and Rusty Thomas for critical review of the manuscript.

    doi:10.1124/mol.104.007278.

    1 Current address: Eli Lilly and Company, Greenfield, IN 46140.

    References

    Anderson SP, Howroyd P, Liu J, Qian X, Bahnemann R, Swanson C, Kwak MK, Kensler TW, and Corton JC (2004) The transcriptional response to a peroxisome proliferator-activated receptor  agonist includes increased expression of proteome maintenance genes. J Biol Chem 279: 52390eC52398.

    Anderson SP, Yoon L, Richard EB, Dunn CS, Cattley RC, and Corton JC (2002) Delayed liver regeneration in peroxisome proliferator-activated receptor-alpha-null mice. Hepatology 36: 544eC554.

    Barbier O, Torra IP, Duguay Y, Blanquart C, Fruchart JC, Glineur C, and Staels B (2002) Pleiotropic actions of peroxisome proliferator-activated receptors in lipid metabolism and atherosclerosis. Arterioscler Thromb Vasc Biol 22: 717eC726.

    Barger PM, Browning AC, Garner AN, and Kelly DP (2001) p38 mitogen-activated protein kinase activates peroxisome proliferator-activated receptor : a potential role in the cardiac metabolic stress response. J Biol Chem 276: 44495eC44501.

    Bielschowsky F and Bielschowsky M (1961) Carcinogenesis in the pituitary of dwarf mouse. The response to dimethylbenzanthracene applied to the skin. Br J Cancer 15: 257eC262.

    Borg KE, Brown-Borg HM, and Bartke A (1995) Assessment of the primary adrenal cortical and pancreatic hormone basal levels in relation to plasma glucose and age in the unstressed Ames dwarf mouse. Proc Soc Exp Biol Med 210: 126eC133.

    Brown-Borg HM and Rakoczy SG (2000) Catalase expression in delayed and premature aging mouse models. Exp Gerontol 35: 199eC212.

    Bugni JM, Poole TM, and Drinkwater NR (2001) The little mutation suppresses DEN-induced hepatocarcinogenesis in mice and abrogates genetic and hormonal modulation of susceptibility. Carcinogenesis 22: 1853eC1862.

    Cattley RC and Popp JA (1989) Differences between the promoting activities of the peroxisome proliferator WY-14,643 and phenobarbital in rat liver. Cancer Res 49: 3246eC3251.

    Conway JG, Tomaszewski KE, Olson MJ, Cattley RC, Marsman DS, and Popp JA (1989) Relationship of oxidative damage to the hepatocarcinogenicity of the peroxisome proliferators di(2-ethylhexyl)phthalate and Wy-14,643. Carcinogenesis 10: 513eC519.

    Corton JC, Fan LQ, Brown S, Anderson SP, Bocos C, Cattley RC, Mode A, and Gustafsson JA (1998) Down-regulation of cytochrome P450 2C family members and positive acute-phase response gene expression by peroxisome proliferator chemicals. Mol Pharmacol 54: 463eC473.

    Corton JC, Anderson SP, and Stauber A (2000) Central role of peroxisome proliferator-activated receptors in the actions of peroxisome proliferators. Annu Rev Pharmacol Toxicol 40: 491eC518.

    Corton JC, Apte U, Anderson SP, Limaye P, Yoon L, Latendresse J, Dunn C, Everitt JI, Voss KA, Swanson C, et al., (2004) Caloric restriction mimetics include agonists of lipid-activated nuclear receptors. J Biol Chem 279: 46204eC46212.

    Dozmorov I, Galecki A, Chang Y, Krzesicki R, Vergara M, and Miller RA. Gene expression profile of long-lived snell dwarf mice. J Gerontol A Biol Sci Med Sci 57: B99eCB1082.

    Eisen MB, Spellman PT, Brown PO, and Botstein D (1998) Cluster analysis and display of genome-wide expression patterns. Proc Natl Acad Sci USA 95: 14863eC14868.

    Howroyd P, Swanson C, Dunn C, Cattley RC, and Corton JC (2004) Decreased longevity and enhancement of age-dependent lesion in mice lacking the nuclear receptor peroxisome proliferator-activated receptor  (PPAR) Toxicol Pathol, in press.

    Ikeno Y, Bronson RT, Hubbard GB, Lee S, and Bartke A (2003) Delayed occurrence of fatal neoplastic diseases in Ames dwarf mice: correlation to extended longevity. J Gerontol A Biol Sci Med Sci 58: 291eC296.

    Juge-Aubry CE, Hammar E, Siegrist-Kaiser C, Pernin A, Takeshita A, Chin WW, Burger AG, and Meier CA (1999) Regulation of the transcriptional activity of the peroxisome proliferator-activated receptor  by phosphorylation of a ligand-independent trans-activating domain. J Biol Chem 274: 10505eC10510.

    Kepler TB, Crosby L, and Morgan KT (2002). Normalization and analysis of DNA microarray data by self-consistency and local regression. Genome Biol 3(7): RESEARCH0037.

    Klaunig JE, Babich MA, Baetcke KP, Cook JC, Corton JC, David RM, DeLuca JG, Lai DY, McKee RH, Peters JM, et al. (2003) PPARalpha agonist-induced rodent tumors: modes of action and human relevance. Crit Rev Toxicol 33: 655eC780.

    Latchman DS (2001) Heat shock proteins and cardiac protection. Cardiovasc Res 51: 637eC646.

    Lee SS, Pineau T, Drago J, Lee EJ, Owens JW, Kroetz DL, Fernandez-Salguero PM, Westphal H, and Gonzalez FJ (1995) Targeted disruption of the alpha isoform of the peroxisome proliferator-activated receptor gene in mice results in abolishment of the pleiotropic effects of peroxisome proliferators. Mol Cell Biol 15: 3012eC3022.

    Liu PC, Huber R, Stow MD, Schlingmann KL, Collier P, Liao B, Link J, Burn TC, Hollis G, Young PR, et al. (2003) Induction of endogenous genes by peroxisome proliferator activated receptor alpha ligands in a human kidney cell line and in vivo. J Steroid Biochem Mol Biol 85: 71eC79.

    Marmary Y, Parlow AF, Goldsmith CM, He X, Wellner RB, Satomura K, Kriete MF, Robey PG, Nieman LK, and Baum BJ (1999) Construction and in vivo efficacy of a replication-deficient recombinant adenovirus encoding murine growth hormone. Endocrinology 140: 260eC265.

    Mehendale HM (2000) PPAR-alpha: a key to the mechanism of hepatoprotection by clofibrate. Toxicol Sci 57: 187eC190.

    Murakami S, Salmon A and Miller RA (2003) Multiplex stress resistance in cells from long-lived dwarf mice. FASEB J 17: 1565eC1566.

    Rennels EG, Anigstein DM, and Anigstein L (1965) A cumulative study of the growth of sarcoma 180 in anterior pituitary dwarf mice. Tex Rep Biol Med 23: 776eC781.

    Sausen PJ, Lee DC, Rose ML, and Cattley RC (1995) Elevated 8-hydroxydeoxyguanosine in hepatic DNA of rats following exposure to peroxisome proliferators: relationship to mitochondrial alterations. Carcinogenesis 16: 1795eC1801.

    Shipley JM and Waxman DJ (2003) Down-regulation of STAT5b transcriptional activity by ligand-activated peroxisome proliferator-activated receptor (PPAR)  and PPAR. Mol Pharmacol 64: 355eC364.

    Sterchele PF, Sun H, Peterson RE, and Vanden Heuvel JP (1996) Regulation of peroxisome proliferator-activated receptor-alpha mRNA in rat liver. Arch Biochem Biophys 326: 281eC289.

    Styles JA, Kelly MD, Pritchard NR, and Foster JR (1990) Effects produced by the non-genotoxic hepatocarcinogen methylclofenapate in dwarf mice: peroxisome induction uncoupled from DNA synthesis and nuclearity changes. Carcinogenesis 11: 387eC391.

    Sugiyama H, Yamada J, and Suga T (1994) Effects of testosterone, hypophysectomy and growth hormone treatment on clofibrate induction of peroxisomal beta-oxidation in female rat liver. Biochem Pharmacol 47: 918eC921.

    Tatar M, Bartke A, and Antebi A (2003) The endocrine regulation of aging by insulin-like signals. Science (Wash DC) 299: 1346eC1351.

    Thuillier P, Anchiraico GJ, Nickel KP, Maldve RE, Gimenez-Conti I, Muga SJ, Liu KL, Fischer SM, and Belury MA (2000) Activators of peroxisome proliferator-activated receptor-alpha partially inhibit mouse skin tumor promotion. Mol Carcinog 29: 134eC142.

    Tollet-Egnell P, Flores-Morales A, Stahlberg N, Malek RL, Lee N, and Norstedt G (2001) Gene expression profile of the aging process in rat liver: normalizing effects of growth hormone replacement. Mol Endocrinol 15: 308eC318.

    Yokota SI, Yanagi H, Yura T, and Kubota H (2000) Cytosolic chaperonin-containing t-complex polypeptide 1 changes the content of a particular subunit species concomitant with substrate binding and folding activities during the cell cycle. Eur J Biochem 267: 1658eC1664.

    Zhou Y, Xu BC, Maheshwari HG, He L, Reed M, Lozykowski M, Okada S, Cataldo L, Coschigamo K, Wagner TE, Baumann G and Kopchick JJ (1997) A mammalian model for Laron syndrome produced by targeted disruption of the mouse growth hormone receptor/binding protein gene (the Laron mouse). Proc Natl Acad Sci USA 94: 3215eC3220.

    Zhou YC, Davey HW, McLachlan MJ, Xie T, and Waxman DJ (2002) Elevated basal expression of liver peroxisomal beta-oxidation enzymes and CYP4A microsomal fatty acid omega-hydroxylase in STAT5b(eC/eC) mice: cross-talk in vivo between peroxisome proliferator-activated receptor and signal transducer and activator of transcription signaling pathways. Toxicol Appl Pharmacol 182: 1eC10.

    Zhou YC and Waxman DJ (1999a) Cross-talk between janus kinase-signal transducer and activator of transcription (JAK-STAT) and peroxisome proliferator-activated receptor- (PPAR) signaling pathways. Growth hormone inhibition of pparalpha transcriptional activity mediated by stat5b. J Biol Chem 274: 2672eC2681.

    Zhou YC and Waxman DJ (1999b) STAT5b down-regulates peroxisome proliferator-activated receptor  transcription by inhibition of ligand-independent activation function region-1 trans-activation domain. J Biol Chem 274: 29874eC29882.