Development of comprehensive functional genomic screens to identify novel mediators of osteoarthritis.

OBJECTIVE: The aim of this study was to develop high-throughput assays for the analysis of major chondrocyte functions that are important in osteoarthritis (OA) pathogenesis and methods for high-level gene expression and analysis in primary human chondrocytes. METHODS: In the first approach, complementary DNA (cDNA) libraries were constructed from OA cartilage RNA and full-length clones were selected. These cDNAs were transferred into a retroviral vector using Gateway Technology. Full-length clones were over-expressed in human articular chondrocytes (HAC) by retroviral-mediated gene transfer. The induction of OA-associated markers, including aggrecanase-1 (Agg-1), matrix metalloproteinase-13 (MMP-13), inducible nitric oxide synthase (iNOS), cyclooxygenase-2 (COX-2), collagen IIA and collagen X was measured by quantitative real-time polymerase chain reaction (QPCR). Induction of a marker gene was verified by independent isolation of 2-3 clones per gene, re-transfection followed by QPCR as well as nucleotide sequencing. In the second approach, whole cDNA libraries were transduced into chondrocytes and screened for chondrocyte cluster formation in three-dimensional agarose cultures. RESULTS: Using green fluorescent protein (eGFP) as a marker gene, it was shown that the retroviral method has a transduction efficiency of >90%. A total of 40 verified hits were identified in the QPCR screen. The first set of 19 hits coordinately induced iNOS, COX-2, Agg-1 and MMP-13. The most potent of these genes were the tyrosine kinases Axl and Tyro-3, receptor interacting kinase-2 (RIPK2), tumor necrosis factor receptor 1A (TNFR1A), fibroblast growth factor (FGF) and its receptor FGFR, MUS81 endonuclease and Sentrin/SUMO-specific protease 3. The second set of seven hits induced both Agg-1 and MMP-13 but none of the other markers. Five of these seven genes regulate the phosphoinositide-3-kinase pathway. The most potently induced OA marker was iNOS. This marker was induced 20-500 fold by seven genes. Collagen IIA was also induced by seven genes, the most potent being transforming growth factor beta (TGFbeta)-stimulated protein TSC22, vascular endothelial growth factor (VEGF) and splicing factor 3a. This screening assay did not identify inducers of collagen X. The second chondrocyte cluster formation screen identified 14 verified hits. Most of the genes inducing cluster formation were kinases. Additional genes had not been previously known to regulate chondrocyte cluster formation or any other chondrocyte function. CONCLUSIONS: The methods developed in this study can be applied to screen for genes capable of inducing an OA-like phenotype in chondrocytes on a genome-wide scale and identify novel mediators of OA pathogenesis. Thus, coordinated functional genomic approaches can be used to delineate key genes and pathways activated in complex human diseases such as OA.

HIV protease genotype and viral sensitivity to HIV protease inhibitors following saquinavir therapy.

OBJECTIVE: To examine the relationship between HIV protease genotype and altered protease inhibitor sensitivity of isolates from patients after therapy with saquinavir (SQV) in its hard gelatin formulation. DESIGN: Forty-one post-therapy isolates and corresponding baseline samples were obtained from 37 patients in four different clinical trials after therapy with SQV for 16-147 weeks. Post-therapy isolates were selected on the basis of preliminary sequence or drug sensitivity data. RESULTS: Fifteen out of 17 isolates without detectable Val-48 or Met-90 mutations retained sensitivity to SQV. (The remaining isolates showed only a marginal increase in median inhibitory concentration.) In addition, three out of 15 isolates with Met-90 retained sensitivity to all other protease inhibitors tested (indinavir, ritonavir, amprenavir, nelfinavir). Of the isolates showing reduced sensitivity to SQV, six out of 22 retained sensitivity to all other protease inhibitors, whereas only four out of 22 showed broad cross-resistance to all protease inhibitors tested. The reduction in sensitivity correlated closely with the presence of Val-48 or Met-90. Subsequent accessory substitutions were also linked to reduced sensitivity. However, significant linkage was observed only between mutations at residues 48 and 82 and between those at residues 82 and 74. CONCLUSIONS: Recruitment of Val-48/Met-90 mutations was not found to be synonymous with cross-resistance. Indeed, the majority of isolates with these mutations retained sensitivity to at least one protease inhibitor (Val-48, 86%; Met-90, 77%). The recruitment of accessory mutations may occur only after the selection of key resistance mutations. Furthermore, Met-90 was found to be a poor marker of cross-resistance in SQV-treated patients.

Complementary classification approaches for protein sequences.

We have studied five methods of protein classification and have applied them to the 768 groups of related proteins in the PROSITE catalog. Four of these methods are based on searching a database of blocks, and the other uses the frequently occurring motifs found in the protein families combined with a fingerprint technique. Our experimental results show that the block-based methods perform well when taking into account the probability of amino acids occurring in a block. Furthermore, the five methods give information that is complementary to each other. Thus, using the five methods together, one can obtain high confidence classifications (if the results agree) or suggest alternative hypotheses (if the results disagree). We also list those proteins whose current families documented in the PROSITE catalog differ from those suggested by our results. There are remarkably few of them, which is a testimony to the quality of PROSITE.

Discovering active motifs in sets of related protein sequences and using them for classification.

We describe a method for discovering active motifs in a set of related protein sequences. The method is an automatic two step process: (1) find candidate motifs in a small sample of the sequences; (2) test whether these motifs are approximately present in all the sequences. To reduce the running time, we develop two optimization heuristics based on statistical estimation and pattern matching techniques. Experimental results obtained by running these algorithms on generated data and functionally related proteins demonstrate the good performance of the presented method compared with visual method of O'Farrell and Leopold. By combining the discovered motifs with an existing fingerprint technique, we develop a protein classifier. When we apply the classifier to the 698 groups of related proteins in the PROSITE catalog, it gives information that is complementary to the BLOCKS protein classifier of Henikoff and Henikoff. Thus, using our classifier in conjunction with theirs, one can obtain high confidence classifications (if BLOCKS and our classifier agree) or suggest a new hypothesis (if the two disagree).