The Stanford Microarray Database: data access and quality assessment tools

The Stanford Microarray Database (SMD; http://genome-www.stanford.edu/microarray/) serves as a microarray research database for Stanford investigators and their collaborators. In addition, SMD functions as a resource for the entire scientific community, by making freely available all of its source code and providing full public access to data published by SMD users, along with many tools to explore and analyze those data. SMD currently provides public access to data from 3500 microarrays, including data from 85 publications, and this total is increasing rapidly. In this article, we describe some of SMD's newer tools for accessing public data, assessing data quality and for data analysis.     

Oxysterol-induced toxicity in R28 and ARPE-19 cells

Studies have shown an intimate relationship between cholesterol and retinal diseases; we examined the effects of cholesterol oxides on cultured cells. Using the rat retinal precursor cell line R28 and the human RPE cell line ARPE-19, we investigated the potential cytotoxicity of cholesterol oxides. Cultured R28 and ARPE-19 cells were treated with either 25-hydroxycholesterol and 7-ketocholesterol (0–50 µg/ml). Cell viability was determined by the WST-1 colorimetric assay. Production of reactive oxygen intermediate (ROI) was assessed by a fluorescent probe–based assay (2,7-dichlorodihydrofluorescein diacetate [H₂DCFDA]). To detect the presence of apoptosis, DNA fragmentation gel analysis and Hoescht nuclear staining were performed. Both cholesterol oxides tested were toxic in a time- and dose-dependent fashion to the two cell lines used in this study. Treatment of R28 cells with either 25-hydroxycholesterol or 7-ketocholesterol at a concentration of 25 µg/ml resulted in greater than 50% loss of cell viability after 24 h. ARPE-19 cells were slightly less affected, with a loss of cell viability of approximately 20% and 40% after 24 h-exposure of 25-hydroxycholesterol and 7-ketocholesterol, respectively. DNA fragmentation and chromatin condensation demonstrated apoptotic events occurring in 7-ketocholesterol–treated cells. The fluorescent assay for ROI production showed that after an hour of exposure to 7-ketocholesterol, R28 cells responded with increased levels of ROIs, whereas no immediate production of ROIs were detected with treated ARPE-19 cells. These in vitro findings provide evidence that cholesterol oxides can directly damage cultured retinal and RPE cells. The oxysterol-induced oxidative stress in these cells may be a factor in the pathology of retinal degenerative diseases.     

Formation of a DNA mismatch repair complex mediated by ATP

The mismatch repair proteins, MutS and MutL, interact in a DNA mismatch and ATP-dependent manner to activate downstream events in repair. Here, we assess the role of ATP binding and hydrolysis in mismatch recognition by MutS and the formation of a ternary complex involving MutS and MutL bound to a mismatched DNA. We show that ATP reduces the affinity of MutS for mismatched DNA and that the modulation of DNA binding affinity by nucleotide is even more pronounced for MutS E694A, a protein that binds ATP but is defective for ATP hydrolysis. Despite the ATP hydrolysis defect, E694A, like WT MutS, undergoes rapid, ATP-dependent dissociation from a DNA mismatch. Furthermore, MutS E694A retains the ability to interact with MutL on mismatched DNA. The recruitment of MutL to a mismatched DNA by MutS is also observed for two mutant MutL proteins, E29A, defective for ATP hydrolysis, and R266A, defective for DNA binding. These results suggest that ATP binding in the absence of hydrolysis is sufficient to trigger formation of a MutS sliding clamp. However, recruitment of MutL results in the formation of a dynamic ternary complex that we propose is the intermediate that signals subsequent repair steps requiring ATP hydrolysis.     

Tumor necrosis factor receptor-associated factor 2 (TRAF2)-deficient B lymphocytes reveal novel roles for TRAF2 in CD40 signaling

CD40 function is initiated by tumor necrosis factor (TNF) receptor-associated factor (TRAF) adapter proteins, which play important roles in signaling by numerous receptors. Characterizing roles of individual TRAFs has been hampered by limitations of available experimental models and the poor viability of most TRAF-deficient mice. Here, B cell lines made deficient in TRAF2 using a novel homologous recombination system reveal new roles for TRAF2. We demonstrate that TRAF2 participates in synergy between CD40 and B cell antigen receptor signals, and in CD40-mediated, TNF-dependent IgM production. We also find that TRAF2 participates in the degradation of TRAF3 associated with CD40 signaling, a role that may limit inhibitory actions of TRAF3. Finally, we show that TRAF2 and TRAF6 have overlapping functions in CD40-mediated NF-kappaB activation and CD80 up-regulation. These findings demonstrate previously unappreciated roles for TRAF2 in signaling by TNF receptor family members, using an approach that facilitates the analysis of genes critical to the viability of whole organisms.     

The inhibitor thiomandelic acid binds to both metal ions in metallo-beta-lactamase and induces positive cooperativity in metal binding

Thiomandelic acid is a simple, broad spectrum, and reasonably potent inhibitor of metallo-beta-lactamases, enzymes that mediate resistance to beta-lactam antibiotics. We report studies by NMR and perturbed angular correlation (PAC) spectroscopy of the mode of binding of the R and S enantiomers of thiomandelic acid, focusing on their interaction with the two metal ions in cadmium-substituted Bacillus cereus metallo-beta-lactamase. The 113Cd resonances are specifically assigned to the metals in the two individual sites on the protein by using 113Cd-edited 1H NMR spectra. Each enantiomer of thiomandelate produces large downfield shifts of both 113Cd resonances and changes in the PAC spectra, which indicate that they bind such that the thiol of the inhibitor bridges between the two metals. For R-thiomandelate, this is unambiguously confirmed by the observation of scalar coupling between Halpha of the inhibitor and both cadmium ions. The NMR and PAC spectra reveal that the two chiral forms of the inhibitor differ in the details of their coordination geometry. The complex with R-thiomandelate, but not that with the S-enantiomer, shows evidence in the PAC spectra of a dynamic process in the nanosecond time regime, the possible nature of which is discussed. The thiomandelate complex of the mononuclear enzyme can be detected only at low metal to enzyme stoichiometry; the relative populations of mononuclear and binuclear enzyme as a function of cadmium concentration provide clear evidence for positive cooperativity in metal ion binding in the presence of the inhibitor, in contrast to the negative cooperativity observed in the free enzyme.     

Gene and protein expression in the epididymis of infertile c-ros receptor tyrosine kinase-deficient mice

Transgenic male mice bearing inactive mutations of the receptor tyrosine kinase c-ros lack the initial segment of the epididymis and are infertile. Several techniques were applied to determine differences in gene expression in the epididymal caput of heterozygous fertile (HET) and infertile homozygous knockout (KO) males that may explain the infertility. Complementary DNA arrays, gene chips, Northern and Western blots, and immunohistochemistry indicated that some proteins were downregulated, including the initial segment/proximal caput-specific genes c-ros, cystatin-related epididymal-spermatogenic (CRES), and lipocalin mouse epididymal protein 17 (MEP17), whereas other caput-enriched genes (glutathione peroxidase 5, a disintegrin and metalloproteinase [ADAM7], bone morphogenetic proteins 7 and 8a, A-raf, CCAAT/enhancer binding protein beta, PEA3) were unchanged. Genes normally absent from the initial segment (gamma-glutamyltranspeptidase, prostaglandin D2 synthetase, alkaline phosphatase) were expressed in the undifferentiated proximal caput of the KO. More distally, lipocalin 2 (24p3), CRISP1 (formerly MEP7), PEBP (MEP9), and mE-RABP (MEP10) were unchanged in expression. Immunohistochemistry and Western blots confirmed the absence of CRES in epididymal tissue and fluid and the continued presence of CRES in spermatozoa of the KO mouse. The glutamate transporters EAAC1 (EAAT3) and EAAT5 were downregulated and upregulated, respectively. The genes of over 70 transporters, channels, and pores were detected in the caput epididymidis, but in the KO, only three were downregulated and six upregulated. The changes in these genes could affect sperm function by modifying the composition of epididymal fluid and explain the infertility of the KO males. These genes may be targets for a posttesticular contraceptive.     

Crystal structure and biochemical analysis of the MutS.ADP.beryllium fluoride complex suggests a conserved mechanism for ATP interactions in mismatch repair

During mismatch repair ATP binding and hydrolysis activities by the MutS family proteins are important for both mismatch recognition and for transducing mismatch recognition signals to downstream repair factors. Despite intensive efforts, a MutS.ATP.DNA complex has eluded crystallographic analysis. Searching for ATP analogs that strongly bound to Thermus aquaticus (Taq) MutS, we found that ADP.beryllium fluoride (ABF), acted as a strong inhibitor of several MutS family ATPases. Furthermore, ABF promoted the formation of a ternary complex containing the Saccharomyces cerevisiae MSH2.MSH6 and MLH1.PMS1 proteins bound to mismatch DNA but did not promote dissociation of MSH2.MSH6 from mismatch DNA. Crystallographic analysis of the Taq MutS.DNA.ABF complex indicated that although this complex was very similar to that of MutS.DNA.ADP, both ADP.Mg(2+) moieties in the MutS. DNA.ADP structure were replaced by ABF. Furthermore, a disordered region near the ATP-binding pocket in the MutS B subunit became traceable, whereas the equivalent region in the A subunit that interacts with the mismatched nucleotide remained disordered. Finally, the DNA binding domains of MutS together with the mismatched DNA were shifted upon binding of ABF. We hypothesize that the presence of ABF is communicated between the two MutS subunits through the contact between the ordered loop and Domain III in addition to the intra-subunit helical lever arm that links the ATPase and DNA binding domains.