Transcriptional profiling and pathway analysis of monosodium iodoacetate-induced experimental osteoarthritis in rats: relevance to human disease.

OBJECTIVE: The objective of this study was to characterize the rat monosodium iodoacetate (MIA)-induced model for osteoarthritis (OA) and determine the translatability of this model to human disease. This was accomplished through pathway, network and system level comparisons of transcriptional profiles generated from animal and human disease cartilage. METHODS: An OA phenotype was induced in rat femorotibial joints following a single injection of 200mug MIA per knee joint for a period of 2 or 4 weeks. Lesion formation in the rat joints was confirmed by histology. Gene expression changes were measured using the Agilent rat whole genome microarrays. Cartilage was harvested from human knees and gene expression changes were measured using the Agilent human arrays. RESULTS: One thousand nine hundred and forty-three oligos were differentially expressed in the MIA model, of these, approximately two-thirds were up-regulated. In contrast, of the 2130 differentially expressed oligos in human disease tissue, approximately two-thirds were down-regulated. This dramatic difference was observed throughout each level of the comparison. The total overlap of genes modulated in the same direction between rat and human was less than 4%. Matrix degradation and inflammatory genes were differentially regulated to a much greater extent in MIA than human disease tissue. CONCLUSION: This study demonstrated, through multiple levels of analysis, that little transcriptional similarity exists between rat MIA and human OA derived cartilage. As disease modulatory activities for potential therapeutic agents often do not translate from animal models to human disease, this and like studies may provide a basis for understanding the discrepancies.

Reconstitution of human telomerase with the template RNA component hTR and the catalytic protein subunit hTRT.

The maintenance of chromosome termini, or telomeres, requires the action of the enzyme telomerase, as conventional DNA polymerases cannot fully replicate the ends of linear molecules. Telomerase is expressed and telomere length is maintained in human germ cells and the great majority of primary human tumours. However, telomerase is not detectable in most normal somatic cells; this corresponds to the gradual telomere loss observed with each cell division. It has been proposed that telomere erosion eventually signals entry into senescence or cell crisis and that activation of telomerase is usually required for immortal cell proliferation. In addition to the human telomerase RNA component (hTR; ref. 11), TR1/TLP1 (refs 12, 13), a protein that is homologous to the p80 protein associated with the Tetrahymena enzyme, has been identified in humans. More recently, the human telomerase reverse transcriptase (hTRT; refs 15, 16), which is homologous to the reverse transcriptase (RT)-like proteins associated with the Euplotes aediculatus (Ea_p123), Saccharomyces cerevisiae (Est2p) and Schizosaccharomyces pombe (5pTrt1) telomerases, has been reported to be a telomerase protein subunit. A catalytic function has been demonstrated for Est2p in the RT-like class but not for p80 or its homologues. We now report that in vitro transcription and translation of hTRT when co-synthesized or mixed with hTR reconstitutes telomerase activity that exhibits enzymatic properties like those of the native enzyme. Single amino-acid changes in conserved telomerase-specific and RT motifs reduce or abolish activity, providing direct evidence that hTRT is the catalytic protein component of telomerase. Normal human diploid cells transiently expressing hTRT possessed telomerase activity, demonstrating that hTRT is the limiting component necessary for restoration of telomerase activity in these cells. The ability to reconstitute telomerase permits further analysis of its biochemical and biological roles in cell aging and carcinogenesis.

Telomerase catalytic subunit homologs from fission yeast and human.

Catalytic protein subunits of telomerase from the ciliate Euplotes aediculatus and the yeast Saccharomyces cerevisiae contain reverse transcriptase motifs. Here the homologous genes from the fission yeast Schizosaccharomyces pombe and human are identified. Disruption of the S. pombe gene resulted in telomere shortening and senescence, and expression of mRNA from the human gene correlated with telomerase activity in cell lines. Sequence comparisons placed the telomerase proteins in the reverse transcriptase family but revealed hallmarks that distinguish them from retroviral and retrotransposon relatives. Thus, the proposed telomerase catalytic subunits are phylogenetically conserved and represent a deep branch in the evolution of reverse transcriptases.

The RNA component of human telomerase.

Eukaryotic chromosomes are capped with repetitive telomere sequences that protect the ends from damage and rearrangements. Telomere repeats are synthesized by telomerase, a ribonucleic acid (RNA)-protein complex. Here, the cloning of the RNA component of human telomerase, termed hTR, is described. The template region of hTR encompasses 11 nucleotides (5'-CUAACCCUAAC) complementary to the human telomere sequence (TTAGGG)n. Germline tissues and tumor cell lines expressed more hTR than normal somatic cells and tissues, which have no detectable telomerase activity. Human cell lines that expressed hTR mutated in the template region generated the predicted mutant telomerase activity. HeLa cells transfected with an antisense hTR lost telomeric DNA and began to die after 23 to 26 doublings. Thus, human telomerase is a critical enzyme for the long-term proliferation of immortal tumor cells.

Specific association of human telomerase activity with immortal cells and cancer.

Synthesis of DNA at chromosome ends by telomerase may be necessary for indefinite proliferation of human cells. A highly sensitive assay for measuring telomerase activity was developed. In cultured cells representing 18 different human tissues, 98 of 100 immortal and none of 22 mortal populations were positive for telomerase. Similarly, 90 of 101 biopsies representing 12 human tumor types and none of 50 normal somatic tissues were positive. Normal ovaries and testes were positive, but benign tumors such as fibroids were negative. Thus, telomerase appears to be stringently repressed in normal human somatic tissues but reactivated in cancer, where immortal cells are likely required to maintain tumor growth.

Exocrine pancreas transcription factor 1 binds to a bipartite enhancer element and activates transcription of acinar genes.

Exocrine pancreas (XP) enhancers, which contain a conserved core sequence, are active only in XP cells. A core enhancer-binding activity also appears to be restricted to XP nuclei. Here we describe the properties of a factor, purified approximately 100,000-fold from pancreas nuclei, which displays core enhancer-binding activity. It is not identical to previously characterized factors and is termed exocrine pancreas transcription factor 1 (XPF-1). In the highly purified preparation, only a single major protein of 60 kDa was detected by silver staining on sodium dodecyl sulfate-gels and by UV cross-linking. XPF-1 binds to the core enhancer of all tested XP genes and not to a mutant sequence which is inactive in vivo. High-affinity binding sites are bipartite. The results of competition binding and UV-cross-linking assays suggest that XPF-1 interacts with both motifs. XPF-1 selectively stimulates transcription of core enhancer templates in an in vitro transcription system. We hypothesize that XPF-1 plays a role in activation of the transcription of XP-specific genes.

A pancreatic exocrine cell factor and AP4 bind overlapping sites in the amylase 2A enhancer.

A factor found in pancreatic exocrine cell lines and pancreatic nuclei binds selectively to the alpha-amylase 2A transcriptional enhancer. Pancreatic exocrine cell extracts protect asymmetrically an unusually large, 35 base pair region from DNase I digestion in vitro, suggesting the involvement of a multimeric DNA binding complex. We show that this region of the enhancer contains a major affinity recognition sequence for the HeLa transcription factor AP4. A 4 base pair mutation in the enhancer sequence shown previously to abolish activity in vivo [Boulet, A. M., Erwin, C. R., & Rutter, W. J. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 3599-3603] abolishes AP4 binding in vitro and weakens but does not eliminate the binding of adjacent enhancer factors. Further, sequences similar to the AP4 binding site are found within a consensus sequence of most pancreatic exocrine genes (Boulet et al., 1986). We have identified three AP4 binding sites in the pancreatic elastase gene: one occurs in the consensus sequence of the enhancer. Thus, protein(s) with the binding selectivity of AP4 may play a role in the expression of the pancreatic exocrine gene family.

The chymotrypsin enhancer core. Specific factor binding and biological activity.

A 20-base pair (bp) conserved sequence present in the 5'-flanking regions of genes highly expressed in the exocrine pancreas forms part of the enhancers of the rat amylase 2A, chymotrypsin B, and elastase I genes. Factor(s) that interact with the conserved DNA sequence of the rat chymotrypsin B gene in vitro have been detected in extracts from acinar cells but not in three other cell lines tested. Transfection experiments suggest that the acinar cell factor(s) recognizing this enhancer core sequence are transcriptional activators. Multimers of a 28-bp sequence located by DNase I protection are capable of activating heterologous promoters in acinar cells. In vitro competition binding and methylation interference analyses indicate protein-DNA interactions at two distinct sites 10 base pairs apart on the DNA. This interaction is identical in extracts from cultured acinar cells as well as from whole pancreas tissue. The presence of two contiguous binding motifs on the same face of the DNA suggests that two (multiple) factors cooperate in the transcriptional regulation of this gene. We term these factors CACCTG pan-1 and TTTCCC pan-1. A factor with the binding specificity of the adenovirus major late transcription factor (MLTF) cross-reacts with the factor CACCTG pan-1 in vitro. However, the distribution of this factor in the various cells does not correlate with the activity of the enhancer core element in vivo. Further, conversion of the chymotrypsin sequence into a consensus MLTF site by three-point mutations abolished enhancer activity in acinar cells. Thus, the MLTF-like factor cannot substitute functionally for the factor CACCTG pan-1 and may act as an inhibitor of chymotrypsin enhancer function.

Nucleotide sequence and molecular genetic analysis of the vaccinia virus HindIII N/M region encoding the genes responsible for resistance to alpha-amanitin.

The genomic location of the gene(s) which provides vaccinia virus (VV) alpha-amanitin-resistant mutants with a drug-resistant phenotype have been mapped to the HindIII N/M region of the genome by the use of marker rescue techniques [E. C. Villarreal and D. E. Hruby (1986) J. Virol. 57, 65-70]. Nucleotide sequencing of a 2356-bp HindIII-Sau3A fragment of the vaccinia virus genome encompassing this region reveals the presence of two complete leftward-reading open reading frames (ORFs, N2 and M1) and two incomplete ORFs (N1 and M2). By computer analysis the N2 and M1 ORFs would be predicted to encode soluble VV polypeptides with molecular weights of approximately 20 and 48 kDa, respectively. The N2 and M1 ORFs have extremely A-T-rich 5'-proximal sequences, consistent with previous data regarding the location and A-T-richness of viral early promoters. Likewise, the consensus signal believed to be involved in terminating VV early gene transcription, TTTTTNT, was evident at the 3'-boundary of both the N2 and M1 ORFs suggesting that these genes may be VV early genes. The in vivo transcriptional activity, orientation, and limits of these putative transcriptional units were investigated by Northern blot, nuclease S1, and primer extension analysis. Both N2- and M1-specific transcripts were detected in the cytoplasm of VV-infected cells, suggesting that these loci are bonafide viral genes. Time-course nuclease S1 experiments revealed that the N2 gene was transcribed exclusively prior to VV DNA replication. In contrast, the M1 gene was transcribed throughout infection, although different start sites were used at early versus late times postinfection. These results are discussed in relation to the drug-resistant phenotype and future experiments to identify the viral gene product responsible.

Processing and secretion of nerve growth factor: expression in mammalian cells with a vaccinia virus vector.

To study posttranslational mechanisms for the control of nerve growth factor (NGF), we used a recombinant vaccinia virus vector to independently express the two major NGF transcripts in a variety of mammalian cell lines. The two major transcripts contain NGF (12.5 kilodaltons [kDa]) at the C-terminus and differ by alternative splicing of an N-terminal exon, so that the large precursor (34 kDa) had 67 amino acids upstream of an internal signal peptide and the smaller precursor (27 kDa) had this signal peptide at its N-terminus. In L929 cells, expression of either NGF transcript with the vaccinia virus vector gave rise to an apparently identical intracellular 35-kDa glycosylated precursor formed by cleavage of the primary gene product after the signal peptide. These cells also secreted biologically active NGF. To determine whether NGF processing is restricted by cell type, we infected a variety of mammalian cell lines with both recombinant viruses; all accumulated the same 35-kDa precursor and secreted NGF. Thus, many types of cells have the machinery to process and secrete NGF. However, NGF accumulated intracellularly (presumably in secretory granules) in cells with a regulated pathway of secretion (e.g., AtT-20 and HIT cells). In these cells, a membrane-permeable cyclic AMP analog, 8-bromo-cyclic AMP, stimulated NGF secretion. This suggests a mechanism for the regulation of NGF levels in which specific secretagogues, e.g., neurotransmitters, control NGF secretion.

Molecular dissection of cis-acting regulatory elements from 5'-proximal regions of a vaccinia virus late gene cluster.

Promoter elements responsible for directing the transcription of six tightly clustered vaccinia virus (VV) late genes (open reading frames [ORFs] D11, D12, D13, A1, A2, and A3) from the HindIII D/A region of the viral genome were identified within the upstream sequences proximal to each individual locus. These regions were identified as promoters by excising them from the VV genome, abutting them to the bacterial chloramphenicol acetyl transferase gene, and demonstrating their ability to drive expression of the reporter gene in transient-expression assays in an orientation-specific manner. To delineate the 5' boundary of the upstream elements, two of the VV late gene (A1 and D13) promoter: CAT constructs were subjected to deletion mutagenesis procedures. A series of 5' deletions of the ORF A1 promoter from -114 to -24 showed no reduction in promoter activity, whereas additional deletion of the sequences from -24 to +2 resulted in the complete loss of activity. Deletion of the ORF A1 fragment from -114 to -104 resulted in a 24% increase in activity, suggesting the presence of a negative regulatory region. In marked contrast to previous 5' deletion analyses which have identified VV late promoters as 20- to 30-base-pair cap-proximal sequences, 5' deletions to define the upstream boundary of the ORF D13 promoter identified two positive regulatory regions, the first between -235 and -170 and the second between -123 and -106. Background levels of chloramphenicol acetyltransferase expression were obtained with deletions past -88. Significantly, this places the ORF D13 regulatory regions within the upstream coding sequences of the ORF A1. A high-stringency computer search for homologies between VV late promoters that have been thus far characterized was carried out. Several potential consensus sequences were found just upstream from RNA start sites of temporally related promoter elements. Three major conclusions are drawn from these experiments. (i) The presence of promoters preceding each late ORF supports the hypothesis that each is expressed as an individual transcriptional unit. (ii) Promoter elements can be located within the coding portion of the upstream gene. (iii) Sequence homologies between temporally related promoter elements support the notion of kinetic subclasses of late genes.

Noncoordinate regulation of a vaccinia virus late gene cluster.

Identification of a tightly spaced and tandemly oriented late gene cluster within the central conserved region of the vaccinia virus genome suggested the possibility of coordinate regulation of the genes within this domain (S.L. Weinrich and D.E. Hruby, Nucleic Acids Res. 14:3003-3016, 1986). To test this hypothesis, the steady-state levels of transcripts derived from the individual late genes were examined. Cytoplasmic RNA was isolated from infected cells at hourly intervals throughout infection and was used in concert with 5' S1 nuclease mapping procedures to detect transcripts from specific late genes. Among the set of six closely linked late genes, marked differences were observed in both the levels of transcription and the kinetic patterns of expression, providing direct evidence for the existence of differentially regulated gene subsets within the late gene class. Furthermore, these experiments identified one of the genes (encoding a 33,000-molecular-weight polypeptide) as being expressed both early and late postinfection. Interestingly, although transcripts from the constitutively expressed gene were initiated at the same start sites throughout infection, a discrete terminus for these transcripts was detected only at early times. These data suggest that the lack of cis-acting termination signals is not the reason for the late gene transcript heterogeneity observed in vaccinia virus-infected cells.

A tandemly-oriented late gene cluster within the vaccinia virus genome.

The nucleotide sequence of a 5.1 kilobase-pair fragment from the central portion of the vaccinia virus genome has been determined. Within this region, five complete and two incomplete open reading frames (orfs) are tightly-clustered, tandemly-oriented, and read in the leftward direction. Late mRNA start sites for the five complete orfs and one incomplete orf were determined by S1 nuclease mapping. The two leftmost complete orfs correlated with late polypeptides of 65,000 and 32,000 molecular weight previously mapped to this region. When compared with each other and with sequences present in protein data banks, the five complete orfs showed no significant homology matches amongst themselves or any previously reported sequence. The six putative promoters were aligned with three previously sequenced late gene promoters. While all of the nine are A-T rich, the only apparent consensus sequence is TAA immediately preceeding the initiator ATG. Identification of this tandemly-oriented late gene cluster suggests local organization of the viral genome.

Transcriptional and translational analysis of the vaccinia virus late gene L65.

Among the products of vaccinia virus genes which are expressed late in infection is a major polypeptide (Mr, 65,000) designated L65. Pulse-chase analyses indicated that L65 is not subject to posttranslational cleavage as is the core polypeptide p4b which migrates to a similar position in sodium dodecyl sulfate-polyacrylamide gels. A polypeptide of 65,000 molecular weight produced in reticulocyte lysates programmed with viral mRNA isolated late in infection was identified as L65 by peptide mapping. L65 mRNA was purified by hybridization selection to restriction fragments of the viral genome and translated in vitro. This allowed the gene encoding L65 to be mapped to the rightmost 4.5 kilobase pairs of the HindIII D fragment. Transcriptional mapping of this region of the genome detected a late mRNA which was initiated at 450 base pairs to the right of the HindIII D-A junction, was transcribed in the leftward direction, and was terminated in the nondescript manner typical of vaccinia virus late mRNAs.