1.Novel DNA Binding Properties Of The RNA-binding Protein TIAR 2.Pharmacokinetics Studies Of Felodipine In Healthy Chinese Volunteers

TIA-1 related protein (TIAR) is a high-affinity RNA-binding protein that promotes apoptosis. It possesses three RNA recognition motifs (RRMs) and shares 80% amino acid homology with TIA-1. Both proteins have a predilection for binding uridylate (U)-rich sequences. Two relevant binding targets for TIAR and TIA-1 have been identified: the 3'-untranslated regions (3'-UTRs) of certain mRNAs including interleukin (IL)-8, human matrix metalloproteinase (MMP)-13, cyclooxygenase-2, and tumor necrosis factor (TNF)- , and the U-rich sequence near 5' splice sites of several pre-mRNAs, such as fibroblast growth receptor 2, Fas, human calcitonin/CGRP and TIAR itself. When binding to the 3'-UTR, such as with TNF- or MMP-13, these proteins have been linked to translational silencing possibly by sequestering the mRNA away from the translational machinery to cytoplasmic stress granules. By binding to pre-mRNAs, on the other hand, TIAR and TIA-1 modulate splicing. In certain cell systems, TIA-1 may promote apoptosis by producing the alternatively spliced, membrane-bound isoform of Fas responsible for relaying the apoptotic signal. Here, we observed selective and potent inhibition of TIAR–RNP complex formation with IL-8 and VEGF 3'-untranslated regions (3'-UTRs) using thymidine-rich deoxyoligonucleotide (ODN) sequences derived from the VEFG 3'-UTR. By ultraviolet crosslinking and electrophoretic mobility shift assays, we get the results as follows:1. VEGF antisense ODNs block TIAR binding to IL-8 and VEGF 3'-UTR We incubated the probe/ODN mixture with nuclear extract from HS578t breast cancer cells, followed by UV crosslinking and electrophoresis. We observed that one of the VEGF-specific ODNs (VEGF-AS1) effectively inhibited the formation of RNP complexes in the 42–45 kDa size range, with both the IL-8 and VEGF 3'-UTR RNA probes The other RNP complexes, which include HuR and KSRP as identified, were not competed by VEGF-AS1. To confirm that this RNP complex included TIAR, we carried out the UV-crosslinking competition assay with the IL-8 3'-UTR probe and then immunoprecipitated the extract with a TIAR-specific antibody. RNP complexes in the 42–45 kDa size range were immunoprecipitated with the anti-TIAR antibody and the complex formation was blocked in the presence of VEGF AS1 but not AS2 ODNs. We next confirmed this inhibitory effect using an ELISA-based RNA-binding assay. Potent inhibition of RNA binding was observed with TIAR but not HuR2. TIAR binds directly to thymidine-rich ssDNAWith TIAR, we observed a strong crosslinked band with VEGF-AS1 but not with VEGF-AS2 ODN. We performed the same assay with recombinant HuR and observed binding only with the IL-8 3'-UTR riboprobe To delineate sequences important for binding, we divided the VEGF-AS1 ODN into 5' and 3' fragments of equal length (VEGF-AS1a and VEGF-AS1b). We observed a strong band with VEGF-AS1b versus a weak one with VEGF-AS1a. The most striking sequence difference between the two ODNs was the presence of a 12 nt stretch of thymidines in VEGF-AS1b. As an alternative analysis of DNA binding, we performed EMSA with the VEGF-AS1b ODN and found a prominent shifted complex.. No complexes were observed with the negative control ODN (VEGF-AS2), and the shifted complex could readily be competed by unlabeled VEGF-AS1b. Addition of the anti-TIAR antibody, but not control IgG, obliterated the shifted complex and also produced a supershifted complex.We next analyzed more precisely the sequence requirements for optimal inhibition of RNA binding to TIAR. A competitor ODN was added to the reaction simultaneously with the IL-8 3'-UTR riboprobe prior to UV crosslinking. there was potent inhibition with VEGF-AS1 but not with VEGF-AS2. VEGF-S2, blocked binding to the RNA probe as effectively as VEGF-AS1. ODNs rich in cytidines or cytidines/guanidines (C+, GC+) produced little or no inhibition of binding. The 3' end possessed sequences necessary for complete inhibition of binding (VEGF-AS1b) versus no inhibition with the 5' end (VEGF-AS1a). When the thymidine stretch was truncated to 10 residues, there was some decrease in binding inhibition (VEGF-AS1b.1). As the thymidines were further truncated, there was progressive reduction in inhibition (VEGF-AS1b.3), with a complete reversal at seven residues (VEGF-AS1b.2). To confirm that this binding inhibition was limited to the TIAR–RNP complex, we performed UV crosslinking of HS578t nuclear extracts in the presence of VEGF-AS1a and VEGF-AS1b at 10-fold excess of the IL-8 RNA probe. There was complete obliteration of the TIAR complex with VEGF-AS1b, similar to the full-length VEGF-AS1 ODN. No significant changes were observed with HuR, KSRP or other unidentified RNP complexes, indicating the specificity of inhibition. As with the UV crosslinking of recombinant TIAR, VEGF-AS1a, which lacked any extended thymidine stretches, did not inhibit TIAR–RNP complex formation.We evaluated whether the DNA-binding capacity of TIAR was limited to ssDNA. We utilized the negative control, VEGF-AS2, and its compliment, VEGF-S2. We pre-hybridized the labeled VEGF-S2 ODN with varying amounts of unlabeled complementary ODN (VEGF-AS2), and then UV crosslinked the mixture to TIAR. As expected, VEGF-S2 bound to TIAR. When the probe was hybridized to VEGF-AS2, even at 2-fold excess, there was near complete obliteration of binding. However, when a non-complementary ODN was added (C+), there was continued binding even at 20-fold excess (middle panel). No appreciable binding was observed with radiolabeled VEGF-AS2, alone or when hybridized to VEGF-S2. We confirmed these observations using EMSA. Both assays indicated that the TIAR binds only to ssDNA.3. A DNA-binding locus is present outside of the RNA-binding siteWe next determined whether the DNA-binding capacity of TIAR was solely related to the U-rich RNA-binding site. To address this question, we performed a series of competition assays utilizing IL-8 3'-UTR and VEGF-AS1 probes. We analyzed binding to TIAR in the presence of four different unlabeled competitors, VEGF-AS1, VEGF-AS1b, rVEGF-AS1 and an IL-8 3'-UTR in vitro transcript. TIAR bound to IL-8 3'-UTR, and the binding was competed by all four competitors. Interestingly, the DNA competitors blocked binding at lower concentrations than either the IL-8 3'-UTR transcript or rVEGF-AS1, suggesting that the DNA molecule had a higher affinity for TIAR. On the other hand, neither RNA competitor could compete with binding to VEGF-AS1 even at 50-fold excess. As an alternative approach to test this possibility, we pre-crosslinked TIAR with unlabeled rVEGF-AS1 or VEGF-AS1b ODN, and then tested the protein for binding to the DNA or RNA probe by EMSA. When TIAR is pre-crosslinked with unlabeled VEGF-AS1, there is a loss of shifted complex with both the DNA probe and the RNA probe. On the other hand, when the protein is pre-crosslinked with RNA, a shifted complex is still produced with the DNA probe similar to the control. The RNA probe, however, fails to produce any substantial shift. Thus, both the UV crosslinking and EMSA data strongly suggest the presence of an additional DNA-binding site(s) outside of the RNA-binding locus.4. TIAR binding to DNA is displaced by active transcriptionOne potential role for the dual binding capacity of TIAR would be to shuttle between DNA and RNA molecules, potentially even bridging them. One of the main loci where ssDNA and RNA interface is that of active transcription. Splicing, for example, which is one of the defined roles of TIAR and TIA-1 with respect to RNA binding, is known to occur concurrently with transcription at the same locus. One possible scenario would be displacement of DNA binding by active transcription. To test this possibility in vitro, we synthesized an ODN that contained a single-stranded TIAR binding site and a T7 promoter site. We chose most of the VEGF-AS1b sequence for the TIAR binding locus. The ODN was also designed such that when rCTP was not added to the reaction, the transcript would stall prior to the TIAR binding site. We first tested whether this single-stranded template (duplexed at the T7 promoter site) could direct transcription. A full-length transcript was formed with rCTP in the transcription reaction. When rCTP was not added, a truncated transcript was formed. Thus, the partially duplexed ODN was an adequate template for RNA transcription. With the full-length transcript, we were able to demonstrate binding to TIAR by UV crosslinking. The shortened transcript, which is relatively U-rich, produced a faint band suggestive of binding as well. We next radiolabeled the template strand of this partially duplexed ODN and pre-incubated it with TIAR prior to transcription. T7 RNA polymerase was then added to the reaction for various time points up to 45 min. The reaction was subjected to UV crosslinking and then immunoprecipitated with the anti-TIAR antibody. We observed a progressive loss of TIAR binding to the ODN over the 45 min incubation period. When no polymerase was added, TIAR binding remained strong at 45 min. When polymerase was added in the absence of rCTP, TIAR also remained bound to the ODN similar to the control. The intensification of crosslinked TIAR in the controls compared with the time just prior to adding polymerase (0 min) is likely due to the increased time for labeled ODN to bind TIAR. These results indicate that active transcription through the TIAR binding site was necessary for displacement of the protein from its DNA-binding site. To determine the concentration of felodipine by liquid chromatography/tandem mass spectrometry and study its pharmacokinetics in healthy Chinese volunteers.1. The LC/MS/MS method of felodipine in biological samplesThe validation results demonstrated that this method had satisfactory precision and accuracy across the calibration range. The procedure involves minimal sample preparation and is well suited to clinical studies of the drug involving large numbers of generated samples.2. Pharmacokinetics studies of felodipine in healthy Chinese volunteersWe selected 30 healthy male volunteers to enrol in a clinical pharmacokinetic study. The Regional Ethics Committee approved the clinical protocol and all subjects gave written informed consent. All volunteers passed the medical history, physical examination, electrocardiagrams (ECG), and routine laboratory tests to be in good health.After a single oral administration of 5 mg felodipine, the compounds were extracted from the matrix using a simple liquid-liquid extract procedure and analysed by high performance liquid chromatography/tandem mass spectrometry. The MS/MS method was optimised using a triple quadrupole mass spectrometer in multiple reaction monitoring (MRM) mode, using electrospray ion source with positive ion detection. Diphenhydramine was employed as the internal standard.The method was validated over the concentration range of 0.05-10.00 ng/mL felodipine in human plasma. The lower limit of quantitation was 0.05 ng/mL. Its main pharmacokinetic parameters were: the mean Cmax was 1.67±0.84 ng/mL occurring at 3.93±2.49 h. The mean plasma elimination half life was 23.08±9.48 h. MRT was 19.36±5.27 h, AUC0-t was 25.38±12.19 ng·h·mL-1,AUC0-∞was 29.94±14.39 ng·h·mL-1。3. ConclusionA highly sensitive, selective and rapid method for the determination of felodipine in human plasma is reported using HPLC/MS/MS. The sensitivity was sufficient to pharmacokinetic studies in human plasma after oral administration of felodipine. The method allows high sample throughput due to the short run time and relatively simple sample preparation procedure.