Analysis of early changes in the articular cartilage transcriptisome in the rat meniscal tear model of osteoarthritis: pathway comparisons with the rat anterior cruciate transection model and with human osteoarthritic cartilage.

OBJECTIVE: The purpose of this study was to use microarray technology to: (1) understand the early molecular events underlying the damage of articular cartilage initiated by this surgical procedure, and (2) determine whether these changes mimic those that are occurring in human osteoarthritic (OA) cartilage. DESIGN: Cartilage was harvested from both medial and lateral sides of the tibial plateaus and femoral condyles of both meniscal tear (MT) and sham surgery groups on days 3, 7 and 21 post-surgery. mRNA prepared from these rat cartilage samples was used for microarray analysis. RESULTS: Statistical analysis identified 475 genes that were differentially expressed between the sham and MT groups, at one or more of the time points that were analyzed. By integrating these genes with OA-related genes reported previously in a rat OA model and in human OA array studies, we identified 20 commonly changed genes. Six out of these 20 genes (Col5A1, Col6A2, INHBA, LTBP2, NBL1 and SERPINA1) were differentially expressed in two animal models and in human OA. Pathway analysis identified some key features of OA pathology, namely cartilage extracellular matrix remodeling, angiogenesis, and chondrocyte cell death that were recapitulated in the animal models. The rat models suggested increased inflammation and cholesterol metabolic pathways may play important role in early cartilage degeneration. CONCLUSION: We identified a large number of differentially expressed genes in the articular cartilage of the MT model. While there was lack of overall identity in cartilage gene expression between the rat models and human OA, several key biological processes were recapitulated in the rat MT OA model.

Data-driven analysis approach for biomarker discovery using molecular-profiling technologies.

High-throughput molecular-profiling technologies provide rapid, efficient and systematic approaches to search for biomarkers. Supervised learning algorithms are naturally suited to analyse a large amount of data generated using these technologies in biomarker discovery efforts. The study demonstrates with two examples a data-driven analysis approach to analysis of large complicated datasets collected in high-throughput technologies in the context of biomarker discovery. The approach consists of two analytic steps: an initial unsupervised analysis to obtain accurate knowledge about sample clustering, followed by a second supervised analysis to identify a small set of putative biomarkers for further experimental characterization. By comparing the most widely applied clustering algorithms using a leukaemia DNA microarray dataset, it was established that principal component analysis-assisted projections of samples from a high-dimensional molecular feature space into a few low dimensional subspaces provides a more effective and accurate way to explore visually and identify data structures that confirm intended experimental effects based on expected group membership. A supervised analysis method, shrunken centroid algorithm, was chosen to take knowledge of sample clustering gained or confirmed by the first step of the analysis to identify a small set of molecules as candidate biomarkers for further experimentation. The approach was applied to two molecular-profiling studies. In the first study, PCA-assisted analysis of DNA microarray data revealed that discrete data structures exist in rat liver gene expression and correlated with blood clinical chemistry and liver pathological damage in response to a chemical toxicant diethylhexylphthalate, a peroxisome-proliferator-activator receptor agonist. Sixteen genes were then identified by shrunken centroid algorithm as the best candidate biomarkers for liver damage. Functional annotations of these genes revealed roles in acute phase response, lipid and fatty acid metabolism and they are functionally relevant to the observed toxicities. In the second study, 26 urine ions identified from a GC/MS spectrum, two of which were glucose fragment ions included as positive controls, showed robust changes with the development of diabetes in Zucker diabetic fatty rats. Further experiments are needed to define their chemical identities and establish functional relevancy to disease development.

Differential effects of estrogen and raloxifene on messenger RNA and matrix metalloproteinase 2 activity in the rat uterus.

A detailed analysis of the differential effects of estrogen (E) compared to raloxifene (Ral), a selective estrogen receptor modulator (SERM), following estrogen receptor (ER) binding in gynecological tissues was conducted using gene microarrays, Northern blot analysis, and matrix metalloproteinase (MMP) 2 activity studies. We profiled gene expression in the uterus following acute (1 day) and prolonged daily (5 wk) treatment of E and Ral in ovariectomized rats. Estrogen regulated twice as many genes as Ral, largely those associated with catalysis and metabolism, whereas Ral induced genes associated with cell death and negative cell regulation. Follow-up studies confirmed that genes associated with matrix integrity were differentially regulated by Ral and E at various time points in uterine and vaginal tissues. Additional experiments were conducted to determine the levels of MMP2 activity in uterus explants from ovariectomized rats following 2 wk of treatment with E, Ral, or one of two additional SERMs: lasofoxifene, and levormeloxifene. Both E and lasofoxifene stimulated uterine MMP2 activity to a level twofold that of Ral, whereas levormeloxifene elevated MMP2 activity to a level 12-fold that of Ral. These data show that one of the significant differences between E and Ral signaling in the uterus is the regulation of genes and proteins associated with matrix integrity. This may be a potential key difference between the action of SERMs in the uterus of postmenopausal women.

Regulation of the promoters for the human bone morphogenetic protein 2 and 4 genes.

The bone morphogenetic proteins 2 and 4 are known to be important in bone formation and are expressed in both the developing and adult mammalian bone. Understanding the regulation of these genes in osteoblasts may yield methods by which we can control expression to induce bone formation. We have isolated and characterized the human BMP-2 and BMP-4 promoters and report substantially more upstream sequence information than that which has been published. Human osteoblasts were found to have a single transcript initiation site that is conserved across species, rather than multiple start sites, as has previously been reported (Feng, J.Q., Harris, M.A., Ghosh-Choudhury, N., Feng, M., Mundy, G.R., Harris, S.E., 1994. Structure and sequence of mouse morphogenetic protein-2 gene (BMP-2): comparison of the structures and promoter regions of BMP-2 and BMP-4 genes. Biochim. Biophys. Acta 1218, 221-224; Heller, L.C., Li, Y., Abrams, K.L., Rogers, M.B., 1999. Transcriptional regulation of the Bmp2 gene. J. Biol. Chem. 274, 1394-1400; Sugiura, T., 1999. Cloning and functional characterization of the 5'-flanking region of the human bone morphogenetic protein-2 gene. Biochem. J. 338, 433-440). A series of promoter deletions for both human BMP-2 and BMP-4 fused to the luciferase reporter gene were analyzed thoroughly in human and murine osteoblastic cell lines. Several compounds and growth factors that stimulate general or osteogenic pathways were used to treat cells transfected with the promoter constructs. Retinoic acid compounds and the phorbol ester, PMA were found to stimulate BMP-2 and, to a lesser degree, BMP-4. The combination of all trans-RA and PMA caused a synergistic increase in BMP-2 promoter activity and endogenous mRNA. The RA stimulation appears to be an indirect effect on the BMP-2 promoter, as the most highly conserved RRE in the BMP-2 promoter was unable to functionally bind or compete for protein binding. Potential binding sites in both promoters for the bone-specific transcription factor, Cbfa-1, were found to specifically bind Cbfa-1 protein in osteoblast nuclear extracts; however, deletion of these sites did not significantly affect transcriptional activity of the promoters in osteoblasts. These data thus present new sequence and regulatory information for the human BMP-2 and BMP-4 promoters and clarify the human BMP-2 gene transcriptional start site in osteoblasts.

Decreased bone mass and bone elasticity in mice lacking the transforming growth factor-beta1 gene.

Transforming growth factor-beta1 (TGF-beta1) knockout (TGF-beta1(-/-)) mice were used to investigate the role of TGF-beta1 in postnatal bone development. Volumetric bone mineral density (BMD) and mineral content (BMC) in these mice and in their normal (TGF-beta1(+/+)) and heterozygous (TGF-beta1(+/-)) littermates were analyzed by quantitative computed tomography (pQCT). Analysis of the proximal tibial metaphysis showed a significant decrease in the BMC of the TGF-beta1(-/-) mice compared to TGF-beta1(+/+) or TGF-beta1(+/-) mice; however, no significant difference was observed in BMD between the groups of mice. pQCT analysis of the tibial midshaft diaphysis showed no difference in the BMD or BMC of cortical bone between the groups. Histomorphometry revealed no significant difference in trabecular connectivity or in trabecular bone volume, number, or thickness. However, the width of the tibial growth plate and the longitudinal growth rate were significantly decreased in the TGF-beta1(-/-) mice, resulting in shorter tibia. Acoustic velocity measurements showed significant differences between the groups of mice with an apparent dosage effect of TGF-beta1 expression on the anisotropic properties of the bone. These data show that longitudinal growth and total mineral content are affected in mice lacking TGF-beta1, as well as the elastic properties of the bone, consistent with an important role for TGF-beta1 in bone modeling and bone quality.

H-ras 61st codon activation in archival proliferative hepatic lesions isolated from female B6C3F1 mice exposed to the leukotriene D4-antagonist, LY171883.

LY171883, a peroxisome proliferator and leukotriene D4-antagonist, induced a statistically significant increase in the number of hepatic lesions in B6C3F1 female mice in a 2 year oncogenicity study at dietary doses of 0.0225% and 0.075%. The mutation frequency and spectrum of the 61st codon of H-ras was determined for 64 independent, archived lesions from the LY171883 2 year oncogenicity study using the polymerase chain reaction (PCR), allele specific oligo hybridization (ASO) and DNA sequencing. Results showed 41 (64%) of these lesions had mutations at the 61st codon (16/21 hepatocellular carcinomas, 4/10 hepatocellular adenomas, 19/26 focal hepatocellular hyperplasias and 2/7 focal hepatocellular atypia). These mutations consisted of 18 C-A transversions, 16 A-G transitions and seven A-T transversions. Compared to the mutation frequency for spontaneously occurring archival B6C3F1 hepatic lesions (41%), the frequency of LY171883 lesions (64%) was significantly higher (P < 0.01). The frequencies of H-ras 61st codon mutations among the LY171883 lesion types (hepatocellular carcinomas 76%, hepatocellular adenomas 40%, focal hepatocellular hyperplasias 73% and hepatocellular atypia 29%) were also significantly different (P = 0.035). In contrast, spontaneous lesions showed no statistical difference in the frequencies of mutation among lesion types (P > 0.5). The mutation spectrum of the LY171883 lesions was not significantly different from the spontaneous spectra. It may be concluded that based on the similarity in mutation spectrum and the increase in mutation frequency, LY171883 may selectively promote spontaneous hepatic lesions containing H-ras 61st codon mutations. In addition, the difference in mutation frequency among lesion types does not support a linear progression of all LY171883 lesions through focal atypia, focal hepatocellular hyperplasias, hepatocellular adenomas and hepatocellular carcinomas.

Localization of the xanthine guanine phosphoribosyl transferase gene (gpt) of E. coli in AS52 metaphase cells by fluorescence in situ hybridization.

The purpose of this study was to localize the xanthine guanine phosphoribosyl transferase gene (gpt) to a specific chromosome to investigate its proposed autosomal location in the AS52 cell line. AS52 cells are hgprt-deficient Chinese hamster ovary (CHO) cells which carry a single functional copy of the E. coli gpt gene. Fluorescence in situ hybridization (FISH) and digoxigenin-labeled probes, as small as 673 bp, were used in an attempt to localize the 456 bp gpt gene to a specific chromosome. Chi-square analysis of 13 metaphases showed significant labeling on autosomal chromosomes 6 or 7, which are indistinguishable without further banding analysis. Furthermore, a majority of the signals were on the q arm, proximal to the centromere. The data collected supports incorporation of the gpt gene into an acrocentric autosome of the AS52 cell line.

Expression of TRPM-2 during involution and regeneration of the rat liver.

Increased message levels of testosterone-repressed prostate message-2 (TRPM-2) have been associated with programmed cell death in many tissues. To study its involvement in the apoptotic elimination of hepatocytes during liver involution and regeneration, levels of TRPM-2 message were evaluated in situ and by the ribonuclease protection assay. Although significant increases in apoptotic bodies were observed in rats 96 h following treatment with lead nitrate and ethylene dibromide, an increase in TRPM-2 message was not detected. Therefore, the expression of TRPM-2 mRNA may be a poor indicator of the extent to which apoptosis occurs during liver involution.

Temporal changes in the mutant frequency and mutation spectra of the 61st codon of the H-ras oncogene following exposure of B6C3F1 mice to N-nitrosodiethylamine (DEN).

Hepatocellular tumors were induced in 15 day old male B6C3F1 mice following a single exposure to N-nitrosodiethylamine (DEN; 5 mg/kg, i.p.). Tumors were collected at 38 and 65 weeks to compare the frequencies and types of mutations in the 61st codon of the H-ras oncogene. The 61st codon was amplified using the polymerase chain reaction (PCR). Allele-specific oligonucleotide (ASO) probes were used to determine the frequency and types of mutations present in these tumors. Forty-nine nodular hepatic lesions were obtained from seven animals at the 38 week timepoint. Five of these samples (10%) had mutations at the 61st codon with one CAA-AAA, one CAA-CGA and three CAA-CTA. Thirty-six nodular hepatic lesions were obtained from six animals at the 65 week timepoint. Ten of these samples (28%) had mutations at the 61st codon with one CAA-AAA, five CAA-CGA and four CAA-CTA. These data indicate that DEN-induced mutations at the 61st codon of the mouse H-ras oncogene (i) are an infrequent event, (ii) have different frequencies at the 38 and 65 week timepoints and (iii) are different from the types of mutations seen in spontaneous lesions.

Genetic alterations in the 61st codon of the H-ras oncogene isolated from archival sections of hepatic hyperplasias, adenomas and carcinomas in control groups of B6C3F1 mouse bioassay studies conducted from 1979 to 1986.

In order to better understand the molecular events in murine hepatocarcinogenesis, the frequency and types of mutations in the murine H-ras proto-oncogene isolated from 184 independent, spontaneously occurring hepatic lesions were determined. Hepatocellular foci, hyperplasias, adenomas and carcinomas were obtained from archival samples of control male (134 samples) and female (50 samples) B6C3F1 mice used in oncogenicity studies that were conducted at Lilly Research Laboratories from 1979 to 1986. The 61st codon region of the H-ras oncogene from these sections was amplified using the polymerase chain reaction. Mutation frequencies were determined by restriction fragment length polymorphism analysis. The types of mutations were characterized by allele-specific oligonucleotide hybridization and confirmed by DNA sequencing. Forty-two per cent of the carcinomas, 44% of the adenomas, 42% of the hyperplasias and 29% of the foci contained mutations at the 61 codon. The mutation spectra for the carcinomas, adenomas and hyperplasias consisted of mostly CAA-AAA transversions, followed by CAA-CGA transitions, followed by CAA-CTA transversions. These results demonstrate that: (i) the frequency of spontaneous mutations in the H-ras 61st codon is equivalent in murine hyperplasias, adenomas and carcinomas, and (ii) sex was not a determining factor in either the mutation frequency or mutation spectrum for the spontaneous lesions. If these lesions represent successive stages in the carcinogenic process, then these results suggest that mutations in the 61st codon of H-ras are early events in spontaneous murine hepatocarcinogenesis.

Interleukin-3 dependent mitogenesis in murine cells involves a predominant non-protein kinase C (pKC) dependent pathway for c-myc transcription. Role of a myc expression vector in rescuing pKC dependent mitogenesis.

The signaling pathways used by interleukin-3 (IL-3) and by active phorbol ester (12-0-tetradecanoyl phorbol-13-acetate, TPA) to stimulate mitogenesis in the growth factor dependent myeloid cell line FDC-P1 were studied by 'reporter' analysis of nuclear proto-oncogene expression. These studies revealed that IL-3 strongly stimulated c-myc expression by a transcriptional mechanism but IL-3 poorly stimulated c-jun expression, a measure of protein kinase C dependent signals. On the other hand, the protein kinase C agonist, TPA, strongly activated c-jun expression but poorly promoted expression (transcription) of c-myc in FDC-P1. These findings appeared to correlate with the poor mitogenic capacity of TPA for FDC-P1. However, stable transfection of FDC-P1 with a c-myc expression vector driven by a human methallothionein IIA promoter containing the TPA responsive element (TRE), led to a cell clone, FDMT myc.A1, in which TPA mediated selective transcription of the transfected TRE driven c-myc vector and down-regulated expression of the endogenous c-myc gene. IL-3 selectively failed to stimulate expression of the TRE driven c-myc vector in FDMT myc.A1. Augmented TPA dependent vector derived c-myc expression was accompanied by enhanced mitogenesis of the cell line FDMT myc.A1 compared with FDC-P1. In addition, TPA mediated expression of the transfected c-myc gene in FDMT myc.A1 was accompanied by augmented transcription of c-jun and c-fos in response to TPA. These studies show the importance of a non-protein kinase C dependent pathway for IL-3 mediated c-myc transcription. However, these studies reveal that protein kinase C mediated pathways can be promitogenic, especially when complemented by unregulated c-myc expression (in this case driven by an alternative, TRE containing promoter).