The amyloid-␤ (A␤) peptide, which likely plays a key role in Alzheimer disease, is derived from the amyloid-␤ precursor protein (APP) through consecutive proteolytic cleavages by ␤-site APP-cleaving enzyme and ␥-secretase. Unexpectedly ␥-secretase inhibitors can increase the secretion of A␤ peptides under some circumstances. This "A␤ rise" phenomenon, the same inhibitor causing an increase in A␤ at low concentrations but inhibition at higher concentrations, has been widely observed. Here we show that the A␤ rise depends on the ␤-secretase-derived C-terminal fragment of APP (␤CTF) or C99 levels with low levels causing rises. In contrast, the N-terminally truncated form of A␤, known as "p3," formed by ␣-secretase cleavage, did not exhibit a rise. In addition to the A␤ rise, low ␤CTF or C99 expression decreased ␥-secretase inhibitor potency. This "potency shift" may be explained by the relatively high enzyme to substrate ratio under conditions of low substrate because increased concentrations of inhibitor would be necessary to affect substrate turnover. Consistent with this hypothesis, ␥-secretase inhibitor radioligand occupancy studies showed that a high level of occupancy was correlated with inhibition of A␤ under conditions of low substrate expression. The A␤ rise was also observed in rat brain after dosing with the ␥-secretase inhibitor BMS-299897. The A␤ rise and potency shift are therefore relevant factors in the development of ␥-secretase inhibitors and can be evaluated using appropriate choices of animal and cell culture models. Hypothetical mechanisms for the A␤ rise, including the "incomplete processing" and endocytic models, are discussed.Evidence suggests that the amyloid-␤ (A␤) 9 peptide plays a key role in Alzheimer disease. A␤ is generated by proteolytic processing of the amyloid-␤ precursor protein (APP) through consecutive cleavages by the ␤-site APP-cleaving enzyme (BACE) and ␥-secretase. APP is cleaved by BACE to form a ␤-secretase-derived C-terminal fragment of APP (␤CTF), which undergoes further cleavage by ␥-secretase to form A␤. In addition, APP is cleaved by ␣-secretase to form ␣CTF, which undergoes ␥-secretase cleavage to produce an N-terminally truncated form of A␤ known as "p3." Using the conventional amino acid numbering of A␤, BACE cleavage leads to A␤ peptides with N-terminal ends at positions 1 and 11, whereas ␣-secretase leads to p3 peptides with an N-terminal end at position 17. Cleavage of ␤CTF and ␣CTF by ␥-secretase produces a mixture of different C termini in the resulting A␤ and p3 peptides. For example, the predominant ␥-secretase cleavage of ␤CTFs at position 40 produces A␤-(1-40) and A␤-(11-40), whereas other ␥-secretase cleavage sites produce a range of less abundant A␤ peptides, such as the disease-associated A␤-(1-42) (1, 2).

Pyrimidine dimer formation in response to UV radiation is governed by the thymine content of the potential dimer and the two flanking nucleotides. An enzymatic activity can be purified from Micrococcus luteus that cleaves the N-glycosyl bond between the 5' pyrimidine of a dimer and the corresponding sugar without rupture of a phosphodiester bond. We propose that strand scission at a dimer site by the M. luteus enzyme requires two activities, a pyrimidine dimer DNA-glycosylase and an apyrimidinic/apurinic endonuclease.

Using a high-throughput screening strategy, a series of 1-aryl-4,5-dihydro-1H-pyrazolo[3,4-d]pyrimidin-4-ones was identified that inhibit the cyclin-dependent kinase (CDK) 4/cyclin D1 complex-mediated phosphorylation of a protein substrate with IC(50)s in the low micromolar range. On the basis of preliminary structure-activity relationships (SAR), a model was proposed in which these inhibitors occupy the ATP-binding site of the enzyme, forming critical hydrogen bonds to the same residue (Val96) to which the amino group in ATP is presumed to bind. X-ray diffraction studies on a later derivative bound to CDK2 support this binding mode. Iterative cycles of synthesis and screening lead to a novel series of potent, CDK2-selective 6-(arylmethyl)pyrazolopyrimidinones. Placement of a hydrogen-bond donor in the meta-position on the 6-arylmethyl group resulted in approximately 100-fold increases in CDK4 affinity, giving ligands that were equipotent inhibitors of CDK4 and CDK2. These compounds exhibit antiproliferative effects in the NCI HCT116 and other cell lines. The potency of these antiproliferative effects is enhanced in anilide derivatives and translates into tumor growth inhibition in a mouse xenograft model.

The double-stranded replicative form (RF) DNA of the autonomous parvovirus H-1 can be isolated from infected cells in a covalent complex with protein. The protein is present on most or all of the RF DNA, including actively replicating molecules, and is associated with the 5'-terminal endonuclease Hae III fragments of both the viral and complementary strands of RF. The size of the protein is estimated to be 60,000-70,000 daltons from its effect on buoyant density of DNA. DNA with covalently bound protein has not been found in H-1 virions.Since the description of plasmid relaxation complexes (1,2) there have been reports of covalent complexes between protein and nucleic acid in several RNA and DNA viruses. The protein has been found at the 5' end of the nucleic acid molecules in virions (3-13) and, in a few cases, also associated with the intracellular replicative forms of the nucleic acids (3, 9, 11).The studies described here were carried out with the autonomous parvovirus H-1 (14), which has a linear singlestranded DNA genome of 1.6 X 106 daltons with hairpin duplexes at both the 5' and 3' ends (14-16). Replication of viral DNA in the infected cell proceeds through a double-stranded intermediate, replicative form (RF), only one strand of which is packaged into the virion. We report here the presence of protein covalently associated with intracellular H-I RF DNA. The RF was found in the form of a complex of nucleic acid with a capsid-like structure which, after treatment with strong protein-denaturing conditions, left protein covalently bound to RF DNA. The covalently bound protein was not found in virions, in contrast to the results with other viral DNA-protein complexes. Some of the features of the covalent protein-DNA complexes are described here. MATERIALS AND METHODSPreparation of Labeled H-1 Viral Intermediates. Crude stocks of unlabeled H-1 virus were prepared from infected newborn hamsters by a procedure suggested by Solon Rhode (personal communication), which included homogenization in sodium dodecyl sulfate (NaDodSO4) and centrifugation through a 30% sucrose pad. NB cells (human embryonic kidney cells transformed by simian virus 40), grown as monolayers in modified Eagle's medium with 10% calf serum, were infected with H-1 virus (15 infectious units per cell) at a cell density of 4 X 106/10-cm plate (14, 17). The cells were labeled 18-20 hr after infection with [3H]dThd at 10 ,uCi/ml or [14C]dThd at 0.5 ,uCi/ml (1 Ci = 3.7 X 10l°becquerels), and intracellular viral DNA intermediates were separated from host DNA by NaDodSO4 lysis/NaCl precipitation (18).Samples requiring Pronase treatment were kept at 60°C for 3 hr in 1% NaDodSO4/30 mM Tris-HCI/20 mM EDTA, pH 8, with three additions of Pronase (CB grade, Calbiochem) at 1 mg/ml followed by incubation for 16 hr at 370C with additional Pronase at 2 mg/ml. Where indicated, this was followed by extraction with phenol and precipitation with ethanol.Isopycnic Centrifugation. Neutral CsCl solutions for equilibrium density centrifugation contained 50 mM ...

The MMP-11 proteinase, also known as stromelysin-3, probably plays an important role in human cancer because MMP-11 is frequently overexpressed in human tumors and MMP-11 levels affect tumorogenesis in mice. Unlike other MMPs, however, human MMP-11 does not cleave extracellular matrix proteins, such as collagen, laminin, fibronectin, and elastin. To help identify physiologic MMP-11 substrates, a phage display library was