A dual-design expandable colorectal stent for malignant colorectal obstruction: results of a multicenter study.

BACKGROUND AND STUDY AIMS: It is known that metal stent placement is safe, easy, and effective for the treatment of malignant colorectal obstruction, but these stents are associated with delayed complications of tumor ingrowth and stent migration. The aim of this study was to prospectively investigate the technical feasibility, clinical effectiveness, and safety of a dual-design colorectal stent (consisting of an outer stent and an inner bare nitinol stent) in patients with malignant colorectal obstruction. PATIENTS AND METHODS: Placement of the dual stent using a 4.5-mm stent delivery system was attempted in 151 patients with malignant colorectal obstruction, either before surgery (n = 50) or for palliation (n = 101). Multivariate logistic regression analysis was used to identify risk factors associated with complications. RESULTS: Stent placement was technically successful in 145/151 patients (96%). Of the patients who had a technically successful placement, bowel obstruction resolved within 2 days after stent placement in 48/50 (96%) of the patients in the bridge-to-surgery group and in 87/95 (92%) of the patients in the palliative group. Perforation occurred in 16 patients, incomplete stent expansion in eight patients, stent migration in four patients, tumor overgrowth in five patients, severe rectal pain in five patients, and bleeding in eight patients. Complete obstruction was the only significant risk factor for perforation (odds ratio 6.88, 95% CI 2.04-23.17, P = 0.002). In the palliative group, the median survival was 152.0 days and the mean survival was 263.8 days. CONCLUSIONS: The dual stent with a 4.5-mm stent delivery system is easy to insert, safe, and reasonably effective for the palliative treatment of malignant colorectal obstruction. However, a great deal of care is needed in its deployment because of the high rate of perforation.

Transcaval TIPS in patients with failed revision of occluded previous TIPS.

OBJECTIVE: To determine the feasibility of transcaval transjugular intrahepatic portosystemic shunt (TIPS) in patients with occluded previous TIPS. MATERIALS AND METHODS: Between February 1996 and December 2000 we performed five transcaval TIPS procedures in four patients with recurrent gastric cardiac variceal bleeding. All four had occluded TIPS, which was between the hepatic and portal vein. The interval between initial TIPS placement and revisional procedures with transcaval TIPS varied between three and 31 months; one patient underwent transcaval TIPS twice, with a 31-month interval. After revision of the occluded shunt failed, direct cavoportal puncture at the retrohepatic segment of the IVC was attempted. RESULTS: Transcaval TIPS placement was technically successful in all cases. In three, tractography revealed slight leakage of contrast materials into hepatic subcapsular or subdiaphragmatic pericaval space. There was no evidence of propagation of extravasated contrast materials through the retroperitoneal space or spillage into the peritoneal space. After the tract was dilated by a bare stent, no patient experienced trans-stent bleeding and no serious procedure-related complications occurred. After successful shunt creation, variceal bleeding ceased in all patients. CONCLUSION: Transcaval TIPS placement is an effective and safe alternative treatment in patients with occluded previous TIPS and no hepatic veins suitable for new TIPS.

cDNA cloning and gene mapping of a candidate human cell cycle checkpoint protein.

A family of proteins involved in cell cycle progression, DNA recombination, and the detection of DNA damage has been recently identified. One of the members of this family, human ATM, is defective in the cells of patients with ataxia telangiectasia and is involved in detection and response of cells to damaged DNA. Other members include Mei-41 (Drosophila melanogaster), Mec1p (Saccharomyces cerevisiae), and Rad3 (Schizosaccharomyces pombe), which are required for the S and G2/M checkpoints, as well as FRAP (Homo sapiens) and Torl/2p (S. cerevisiae), which are involved in a rapamycin-sensitive pathway leading to G1 cell cycle progression. We report here the cloning of a human cDNA encoding a protein with significant homology to members of this family. Three overlapping clones isolated from a Jurkat T-cell cDNA library revealed a 7.9-kb open reading frame encoding a protein that we have named FRP1 (FRAP-related protein) with 2644 amino acids and a predicted molecular mass of 301 kDa. Using fluorescence in situ hybridization and a full-length cDNA FRP1 clone, the FRP1 gene has been mapped to the chromosomal locus 3q22-q24. FRP1 is most closely related to three of the PIK-related kinase family members involved in checkpoint function--Mei-41, Mec1p, and Rad3--and as such may be the functional human counterpart of these proteins.

Control of p70 s6 kinase by kinase activity of FRAP in vivo.

When complexed with the intracellular protein FKBP12, rapamycin is a potent immunosuppressant and an inhibitor of a mitogen-stimulated signalling pathway that leads to activation of p70 S6 kinase (p70S6k) and cyclin-dependent kinases (CDKs). A recently cloned FKBP12-rapamycin-associated protein (FRAP/RAFT) is the likely mediator of these effects. Using FRAP variants that do not bind FKBP12-rapamycin, we demonstrate here that FRAP is a rapamycin-sensitive regulator of p70S6k in vivo and that the kinase activity of FRAP is required for this regulation. In addition, we show that FRAP autophosphorylates in vitro. Consistent with an essential role for FRAP kinase activity in vivo, autophosphorylation of FRAP is inhibited by FKBP12-rapamycin. Deletion studies indicate that the kinase activity of FRAP alone is not sufficient for control of p70S6k and that an amino-terminal domain in FRAP is also required.

A mammalian protein targeted by G1-arresting rapamycin-receptor complex.

The structurally related natural products rapamycin and FK506 bind to the same intracellular receptor, FKBP12, yet the resulting complexes interfere with distinct signalling pathways. FKBP12-rapamycin inhibits progression through the G1 phase of the cell cycle in osteosarcoma, liver and T cells as well as in yeast, and interferes with mitogenic signalling pathways that are involved in G1 progression, namely with activation of the protein p70S6k (refs 5, 11-13) and cyclin-dependent kinases. Here we isolate a mammalian FKBP-rapamycin-associated protein (FRAP) whose binding to structural variants of rapamycin complexed to FKBP12 correlates with the ability of these ligands to inhibit cell-cycle progression. Peptide sequences from purified bovine FRAP were used to isolate a human cDNA clone that is highly related to the DRR1/TOR1 and DRR2/TOR2 gene products from Saccharomyces cerevisiae. Although it has not been previously demonstrated that either of the DRR/TOR gene products can bind the FKBP-rapamycin complex directly, these yeast genes have been genetically linked to a rapamycin-sensitive pathway and are thought to encode lipid kinases.

1H and 15N assignments and secondary structure of the PI3K SH3 domain.

The sequential 1H and 15N assignments of the SH3 domain of human phosphatidyl inositol 3'-kinase (PI3K) were determined by a combination of homonuclear and heteronuclear NMR experiments. With the exception of several protons belonging to lysine and proline residues, all proton and proton-bearing amide nitrogen resonances were assigned. Based on the sequential nuclear Overhauser effects (NOEs), 3JNH-C alpha H coupling constants and locations of slowly exchanging amide protons, we determined that the secondary structures of the protein consists of six beta-strands, two beta-turns and four short helices. Additional long range NOEs indicate that these beta-strands form two antiparallel beta-sheets. The topology of secondary structural elements of the PI3K SH3 domain is similar to those of the SH3 domains from c-Src and alpha-spectrin, suggesting that the SH3 family has a common tertiary structural motif.

Structure of the PI3K SH3 domain and analysis of the SH3 family.

Src homology 3 (SH3) domains, which are found in many proteins involved in intracellular signal transduction, mediate specific protein-protein interactions. The three-dimensional structure of the SH3 domain in the p85 subunit of the phosphatidylinositol 3-kinase (PI3K) has been determined by multidimensional NMR methods. The molecule consists of four short helices, two beta turns, and two antiparallel beta sheets. The beta sheets are highly similar to corresponding regions in the SH3 domain of the tyrosine kinase Src, even though the sequence identity of the two domains is low. There is a unique 15 amino acid insert in PI3K that contains three short helices. There are substantial differences in the identity of the amino acids that make up the receptor site of SH3 domains. The results suggest that while the overall structures of the binding sites in the PI3K and Src SH3 domains are similar, their ligand binding properties may differ.

Solution structure of the SH3 domain of Src and identification of its ligand-binding site.

The Src homology 3 (SH3) region is a protein domain of 55 to 75 amino acids found in many cytoplasmic proteins, including those that participate in signal transduction pathways. The solution structure of the SH3 domain of the tyrosine kinase Src was determined by multidimensional nuclear magnetic resonance methods. The molecule is composed of two short three-stranded anti-parallel beta sheets packed together at approximately right angles. Studies of the SH3 domain bound to proline-rich peptide ligands revealed a hydrophobic binding site on the surface of the protein that is lined with the side chains of conserved aromatic amino acids.

Partitioning of fluorescent phospholipid probes between different bilayer environments. Estimation of the free energy of interlipid hydrogen bonding.

Fluorescence spectroscopy has been used to monitor the partitioning of a series of exchangeable neutral phospholipid probes, labeled with carbazole, indolyl or diphenylhexatrienyl moieties, between large unilamellar vesicles containing 1-palmitoyl-2-oleoylphosphatidylcholine (POPC), 1,2-dioleoyloxy-3-(trimethylammonio) propane (DOTAP) or N-hexadecyl-N-(9-octadecenyl)-N,N-dimethylammonium chloride (HODMA). Phosphatidylethanolamine (PE) probes desorb from POPC-containing vesicles at markedly slower rates than do phosphatidylcholine (PC) probes with the same acyl chains. The rate of probe desorption from such vesicles is progressively enhanced by successive N-methylations of the amino group but not by methylation of C-2 of the ethanolamine moiety, a modification that leaves unaltered the hydrogen-bonding capacity of the polar headgroup. By contrast, the rates of desorption of different probes (with the same acyl chains) from HODMA or from DOTAP vesicles are much more comparable and reflect no clear systematic influence of the headgroup hydrogen-bonding capacity. Equilibrium-partitioning measurements indicate that the relative affinities of different probes for PC-rich vesicles, in competition with HODMA or DOTAP vesicles, increase with increasing hydrogen-bonding capacity of the probe headgroup in the order PC less than N,N-dimethyl PE less than N-methyl PE less than PE approximately phosphatidyl-2-amino-1-propanol. From such partitioning data, we estimate that interlipid hydrogen-bonding interactions (in competition with lipid-water interactions) contribute roughly -300 cal mol-1 to the free energy of a PE molecule in a hydrated liquid-crystalline phospholipid bilayer; this free-energy contribution is somewhat smaller, but still significant, for N-mono- and dimethylated PE's.