Search

P38α/β has been described as a tumor-suppressor controlling cell cycle checkpoints and senescence in epithelial malignancies. However, p38α/β also regulates other cellular processes. Here, we describe a role of p38α/β as a regulator of acute lymphoblastic leukemia (ALL) proliferation and survival in experimental ALL models. We also report first evidence that p38α/β phosphorylation is associated with the occurrence of relapses in TEL-AML1-positive leukemia. First, in vitroexperiments show that p38α/β signaling is induced in a cyclical manner upon initiation of proliferation and remains activated during log-phase of cell growth. Next, we provide evidence that growth-permissive signals in the bone marrow activate p38α/β in a novel avian ALL model, in which therapeutic targeting can be tested. We further demonstrate that p38α/β inhibition by small molecules can suppress leukemic expansion and prolong survival of mice bearing ALL cell lines and primary cells. Knockdown of p38α strongly delays leukemogenesis in mice xenografted with cell lines. Finally, we show that in xenografted TEL-AML1 patients, ex vivop38α/β phosphorylation is associated with an inferior long-term relapse-free survival. We propose p38α/β as a mediator of proliferation and survival in ALL and show first preclinical evidence for p38α/β inhibition as an adjunct approach to conventional therapies.

We discovered that a shift between the state of tumorigenicity and dormancy in human carcinoma (HEp3) is attained through regulation of the balance between two classical mitogen-activated protein kinase (MAPK)-signaling pathways, the mitogenic extracellular regulated kinase (ERK) and the apoptotic/growth suppressive stress-activated protein kinase 2 (p38(MAPK)), and that urokinase plasminogen activator receptor (uPAR) is an important regulator of these events. This is a novel function for uPAR whereby, when expressed at high level, it enters into frequent, activating interactions with the alpha5beta1-integrin, which facilitates the formation of insoluble fibronectin (FN) fibrils. Activation of alpha5beta1-integrin by uPAR generates persistently high level of active ERK necessary for tumor growth in vivo. Our results show that ERK activation is generated through a convergence of two pathways: a positive signal through uPAR-activated alpha5beta1, which activates ERK, and a signal generated by the presence of FN fibrils that suppresses p38 activity. When fibrils are removed or their assembly is blocked, p38 activity increases. Low uPAR derivatives of HEp3 cells, which are growth arrested (dormant) in vivo, have a high p38/ERK activity ratio, but in spite of a similar level of alpha5beta1-integrin, they do not assemble FN fibrils. However, when p38 activity is inhibited by pharmacological (SB203580) or genetic (dominant negative-p38) approaches, their ERK becomes activated, uPAR is overexpressed, alpha5beta1-integrins are activated, and dormancy is interrupted. Restoration of these properties in dormant cells can be mimicked by a direct re-expression of uPAR through transfection with a uPAR-coding plasmid. We conclude that overexpression of uPAR and its interaction with the integrin are responsible for generating two feedback loops; one increases the ERK activity that feeds back by increasing the expression of uPAR. The second loop, through the presence of FN fibrils, suppresses p38 activity, further increasing ERK activity. Together these results indicate that uPAR and its interaction with the integrin should be considered important targets for induction of tumor dormancy.

Familial British dementia (FBD) is a disorder characterized by the presence of amyloid deposits in cerebral blood vessels and brain parenchyma coexisting with neurofibrillary tangles in limbic areas. The amyloid subunit (ABri) is a 4 kDa fragment of a 266 amino acid type II single-spanning transmembrane precursor protein encoded by the BRIgene located on chromosome 13. In FBD patients, a single base substitution at the stop codon of this gene generates a larger 277-residue precursor (ABriPP-277). Proteolytic processing by a furin-like enzyme at the C-terminus of the elongated precursor generates the 34 amino acid ABri that undergoes rapid aggregation and fibrillization. ABri is structually unrelated to all known amyloids including A?, the main component of the amyloid lesions in Alzheimer's disease (AD), indicating that cerebral deposition of amyloid molecules other than A? can trigger similar neuropathological changes leading to neuronal loss and dementia. These data support the concept that amyloid deposition in the vascular wall and brain parenchyma is of primary importance in the initiation of neurogeneration.

Cerebral amyloid angiopathy is commonly associated with normal aging and Alzheimer's disease and it is also the principal feature of hereditary cerebral hemorrhage with amyloidosis Dutch type, a familial condition associated to a point mutation G to C at codon 693 of the amyloid beta (Abeta) precursor protein gene resulting in a Glu to Gln substitution at position 22 of the Abeta (E22Q). The patients carrying the AbetaE22Q variant usually present with lobar cerebral hemorrhages before 50 years of age. A different mutation described in several members of three Italian kindred who presented with recurrent hemorrhagic strokes late in life, between 60 and 70 years of age, also associated with extensive cerebrovascular amyloid deposition has been found at the same position 22, this time resulting in a Glu to Lys substitution (E22K). We have compared the secondary structure, aggregation, and fibrillization properties of the two Abeta40 variants and the wild type peptide. Using flow cytometry analysis after staining with propidium iodide and annexin V, we also evaluated the cytotoxic effects of the peptides on human cerebral endothelial cells in culture. Under the conditions tested, the E22Q peptide exhibited the highest content of beta-sheet conformation and the fastest aggregation/fibrillization properties. The Dutch variant also induced apoptosis of cerebral endothelial cells at a concentration of 25 micrometer, whereas the wild type Abeta and the E22K mutant had no effect. The data suggest that different amino acids at position 22 confer distinct structural properties to the peptides that appear to influence the onset and aggressiveness of the disease rather than the phenotype.

Familial British dementia (FBD) is an early-onset autosomal dominant disorder characterized by progressive cognitive impairment, spasticity, and cerebellar ataxia. Hippocampal neurofibrillar degeneration and widespread parenchymal and vascular amyloid deposits are the main neuropathological lesions. Amyloid fibrils are composed of a novel 34 amino acid subunit (ABri) with no sequence identity to any known amyloid molecule. The peptide derives from a larger precursor protein codified by a single gene BRIon chromosome 13. Affected family members have a single base substitution at the stop codon of the BRIgene that generates a longer open-reading frame resulting in a larger precursor protein. The release of the 34 C-terminal amino acids from the mutated precursor originates the ABri amyloid subunit. Our discovery of a new amyloid associated with the development of dementia supports the concept that amyloid peptides may be of primary importance in the initiation of neurodegeneration.

Widespread cerebral deposition of a 40-43-amino acid peptide called the amyloid beta-protein (Abeta) in the form of amyloid fibrils is one of the most prominent neuropathologic features of Alzheimer's disease. Numerous studies suggest that Abeta is toxic to neurons by free radical-mediated mechanisms. We have previously reported that melatonin prevents oxidative stress and death of neurons exposed to Abeta. In the process of screening indole compounds for neuroprotection against Abeta, potent neuroprotective properties were uncovered for an endogenous related species, indole-3-propionic acid (IPA). This compound has previously been identified in the plasma and cerebrospinal fluid of humans, but its functions are not known. IPA completely protected primary neurons and neuroblastoma cells against oxidative damage and death caused by exposure to Abeta, by inhibition of superoxide dismutase, or by treatment with hydrogen peroxide. In kinetic competition experiments using free radical-trapping agents, the capacity of IPA to scavenge hydroxyl radicals exceeded that of melatonin, an indoleamine considered to be the most potent naturally occurring scavenger of free radicals. In contrast with other antioxidants, IPA was not converted to reactive intermediates with pro-oxidant activity. These findings may have therapeutic applications in a broad range of clinical situations.

Natural resistance of humans to the cattle pathogen Trypanosoma brucei brucei has been attributed to the presence in human serum of nonimmune factors that lyse the parasite. Normal human serum contains two trypanosome lytic factors (TLFs). TLF1 is a 500-kDa lipoprotein, which is reported to contain apolipoprotein A-I (apoA-I), haptoglobin-related protein (Hpr), hemoglobin, paraoxonase, and apoA-II, whereas TLF2 is a larger, poorly characterized particle. We report here a new immunoaffinity-based purification procedure for TLF2 and TLF1, as well as further characterization of the components of each purified TLF. Immunoaffinity-purified TLF1 has a specific activity 10-fold higher than that of TLF1 purified by previously described methods. Moreover, we find that TLF1 is a lipoprotein particle that contains mainly apoA-I and Hpr, trace amounts of paraoxonase, apoA-II, and haptoglobin, but no detectable hemoglobin. Characterization of TLF2 reveals that it is a 1,000-kDa protein complex containing mainly immunoglobulin M, apoA-I, and Hpr but less than 1% detectable lipid.

Abstract: Most contemporary progress in Alzheimer's disease (AD) stems from the study of a 42–43 amino acid peptide, called the amyloid beta protein (Aβ), as the main neuropathologic marker of the disorder. It has been demonstrated that Aβ has neurotoxic properties and that such effects are mediated by free‐radicals. Exposure of neuronal cells to Aβ results in a spectrum of oxidative lesions that are profoundly harmful to neuronal homeostasis. We had previously shown that Aβ25 35 induces oxidative damage to mitochondrial DNA (mtDNA) and that this modality of injury is prevented by melatonin. Because Aβ25‐35 does not occur in AD and because the mode of toxicity by Aβ25‐35 may be different from that of Aβ1‐42 (the physiologically relevant form of Aβ), we extended our initial observations to determine whether oxidative damage to mtDNA could also be induced by Aβ1‐42 and whether this type of injury is prevented by melatonin. Exposure of human neuroblastoma cells to Aβ1‐42 resulted in marked oxidative damage to mtDNA as determined by a quantitative polymerase chain reaction method. Addition of melatonin to cell cultures along with Aβ completely prevented the damage. This study supports previous findings with Aβ25‐35, including a causative role for Aβ in the mitochondrial oxidative lesions present in AD brains. Most important, the data confirms the neuroprotective role of melatonin in Aβ‐mediated oxidative injury. Because melatonin also inhibits amyloid aggregation, lacks toxicity, and efficiently crosses the blood‐brain barrier, this hormone appears superior to other available antioxidants as a candidate for pharmacologic intervention in AD.

Gerstmann‐Sträussler‐Scheinker (GSS) disease is a familial neurological disorder pathologically characterized by amyloid deposition in the cerebrum and cerebellum. The GSS amyloid is immunoreactive to antisera raised against the hamster prion protein (PrP) 27–30. This is a proteinase K‐resistant glycoprotein of 27–30 kd that is derived from an abnormal isoform of a neuronal glycoprotein of 33–35 kd designated PrPSc and is a molecular marker of amyloid fibrils isolated from animals with scrapie and humans with related disorders. We have purified and characterized proteins extracted from amyloid plaque cores isolated from two patients of the Indiana kindred of GSS disease. We found that the major component of GSS amyloid is an 11 kd degradation product of PrP, whose N‐terminus corresponds to the glycine residue at position 58 of the amino acid sequence deduced from the human PrP cDNA. In addition, amyloid fractions contained larger PrP fragments with apparently intact N‐termini and amyloid P component. These findings suggest that the disease process leads to proteolytic cleavage of PrP, generating an amyloidogenic peptide that polymerizes into insoluble fibrils. The N‐terminal cleavage of PrP in GSS disease occurs at a tryptophan‐glycine peptide bond identical to that cleaved by proteinase K in vitro to generate PrP 27–30 from hamster PrPSc at codon 90. Since no mutations of the structural PrP gene have been found in the Indiana family of GSS disease, it is conceivable that factors other than the primary structure of PrP play a crucial role in the process of amyloid formation and the development of clinical neurologic dysfunction.

We synthesized polypeptides corresponding to sequences encoded by normal and mutant alleles in the regions of codon 178 (Asp-->Asn) and codon 200 (Glu-->Lys) of the chromosome 20 amyloid gene that have been linked to familial Creutzfeldt-Jakob disease. Peptide suspensions from both regions spontaneously formed amyloid fibrils with different morphological characteristics and aggregation tendencies. Fibrillar arrays were denser and more profuse in mutant than in normal peptide suspensions and were even more marked when the homologous mutant and normal peptides were mixed together. Preparations from the region of codon 200 were in all cases more fibrillogenic than corresponding peptides from the region of codon 178. These in vitro observations support the hypothesis that amino acid changes from pathogenic single-allele point mutations in Creutzfeldt-Jakob disease may nucleate the in vivo folding behavior of the normal host protein to favor formation of insoluble amyloid fibrils.

The amyloid fibrils of Alzheimer's neuritic plaques and cerebral blood vessels are mainly composed of aggregated forms of a 39 to 44 amino acids peptide, named amyloid beta (Aβ). A similar although soluble form of Aβ (sAβ) has been identified in plasma, cerebrospinal fluid and cell culture supernatants, indicating that it is produced under physiologic conditions. We report here that sAβ in normal human plasma is associated with lipoprotein particles, in particular to the HDL₃ and VHDL fractions where it is complexed to ApoJ and, to a lesser extent, to ApoAl. This was assessed by immunoprecipitation experiments of purified plasma lipoproteins and lipoprotein-depleted plasma and confirmed by means of amino acid sequence analysis. Moreover, biotinylated synthetic peptide Aβ_1-40 was traced in normal human plasma in in vitro experiments. As in the case of sAβ, biotinylated Aβ_1-40 was specifically recovered in the HDL₃ and VHDL fractions. This data together with the previous demonstration that Aβ_1-40 is taken up into the brain via a specific mechanism and possibly as an A_1-40-ApoJ complex indicate a role for HDL₃- and VHDL-containing ApoJ in the transport of the peptide in circulation and suggest their involvement in the delivery of sAβ across the blood-brain barrier.Copyright 1994, 1999 Academic Press, Inc.

The simultaneous presence of intracellular neurofibrillary tangles (NFT) and extracellular senile plaques in Alzheimer's disease (AD) suggests that the two lesions could be synergistically interrelated. However, although the main protein components of NFT and senile plaques, τ (tau) and amyloid β-protein, respectively, are well characterized, the molecular mechanisms responsible for their deposition in lesions are unknown. We demonstrate, using four independent techniques, that τ directly interacts with a conformation-dependent domain of the amyloid β-protein precursor (βPP) encompassing residues βPP714–723. The putative τ–binding domain includes βPP717mutation sites that are associated with familial forms of AD. Our findings strongly suggest that NFT and senile plaques, often thought of as independent structures, may play a role in each other's formation during the pathogenesis of AD.

The origin of amyloid β (Aβ) deposited in brain and cerebral blood vessels of patients with Alzheimer′s Disease (AD) is not known. In this study, we tested whether soluble Aβ (sAβ) can cross the blood-brain barrier (BBB). An in vivovascular brain perfusion model and capillary depletion technique in guinea-pigs were used to determine cerebral capillary sequestration and blood-brain transport of a synthetic peptide identical with residues 1-40 (SP-40) of Aβ. A saturable, specific binding of SP-40 has been demonstrated at the luminal side of the BBB, with the Kdof 25 ± 2 nM, and Bmaxof 188 ± 11 fmol/min/g of isolated microvessels. A specific transcellular BBB transport of SP-40 into brain parenchyma exhibited the Kmof 49 ± 10 nM, and Vmaxof 111 ± 19 fmol/min/g of capillary depleted brain. We concluded that the BBB has the capability to control the cerebrovascular sequestration and blood-to-brain transport of circulating sAβ. Hence, sAβ can contribute to both cerebrovascular and parenchymal amyloid formation.

Apolipoprotein E (apo E) is associated with Alzheimer's β-amyloid (Aβ) in senile plaques. Aβ is now known to be a normal soluble peptide (sAβ) found in the cerebrospinal fluid (CSF) and other biological fluids. We have used synthetic Aβ peptides bound to affinity membranes in order to determine whether apo E or any other amyloid associated protein will bind to these membranes, when they are bathed in CSF. Under these conditions apo E, as well as another apolipoprotein, apolipoprotein J (Apo J), bound to the membranes. Using ELISA and ligand binding studies, we found a high avidity binding of Aβ peptides to apo E. This suggests that apo E, as well as other related proteins may bind not only amyloid Aβ but also sAβ. This interaction may be critical in amyloid formation.Copyright 1993, 1999 Academic Press

A gamma-trace variant protein is the major constituent of the amyloid fibrils in patients from Iceland with hereditary cerebral hemorrhage with amyloidosis. The protein consists of 110 residues and is similar to human urinary gamma-trace basic protein (or cystatin C) beginning at its 11th amino-terminal residue. It has an amino acid substitution (glutamine for leucine) at position 58 (position 68 in gamma-trace numbering), which is near the proposed active site of related proteins--namely, cysteine protease inhibitors and kininogens. It is postulated that a point mutation has occurred, leading to the production of an unusual protein that is abnormally degraded, bound, and/or precipitated. Alternatively, gamma-trace basic protein may be genetically polymorphic, and the variant described here may represent an as-yet-undiscovered isotype or an allelic form that is linked to, but not responsible for, the deposition disease. Our data on the structure of a gamma-trace variant protein suggests that its gene expresses a polyprotein precursor in which active peptides are flanked by basic amino acid residues that permit cleavage to liberate small internal peptides. It is likely that the nucleotide sequence coding for Arg-Xaa and Lys-Xaa repeated several times in the molecule may function as alternative splicing sites for mRNA processing.

Amyloid β (Aβ) is a fibrillar component in Alzheimer's disease amyloid deposits and a soluble peptide (sAβ) normally present in body fluids. We have recently reported that the blood-brain barrier (BBB) has a capability to control cerebrovascular sequestration and transport of circulating sAβ. In this study, we examined whether two circulating amyloid-associated proteins shown to bind sAβ, apolipoproteins J (apo J) and E (apo E), can cross the BBB alone and/or complexed to a synthetic peptide homologous to a major form of sAβ, sAβ_1-40. Brain perfusion experiments in guinea pigs showed significant uptake of both apo J and sAβ_1-40-apo J complexes. In contrast, blood-brain transport of sAβ_1-40-apo E was negligible, while apo E had a limited access across the BBB, indicating that the apo E found within the brain is produced locally. It is concluded that sAβ_1-40 binding to apo J and apo E results in significant (> 100-fold) difference in brain uptake of their respective complexes. We hypothesize that in normal brain apo J facilitates sAβ transport.Copyright 1994, 1999 Academic Press, Inc.

We amplified DNA encoding the 3′ 109 codons of Alzheimer's-disease amyloid precursor protein (APP) inclusive of the beta protein (A beta) and cytoplasmic domains from cDNA using oligonucleotide primers designed to facilitate cloning into the T7 expression vector pT7Ad23K13. We also modified this construct to generate recombinant molecules incorporating two recently described APP mutants by site-directed mutagenesis. Both native C109 (deletion construct inclusive of the C-terminal 109 residues of APP) and constructs with a single mutation at codon 642 (T-->G, resulting in a substitution of glycine for valine) or a double mutation at codons 595 (G-->T, substituting asparagine for lysine) and 596 (A-->C, substituting leucine for methionine) were expressed in Escherichia coli to levels of 5-20% of total bacterial protein after induction. The major constituent of expressed C109 protein had an apparent molecular mass of 16-18 kDa by SDS/PAGE and appeared to be the full-length construct by size and N-terminal microsequencing. Also present was a 4-5 kDa species that co-purified with C109, constituting only approximately 1% of expressed protein, which was revealed by Western-blot analysis with antibodies specific for A beta epitopes and after biotinylation of purified recombinant C109. This fragment shared N-terminal sequence with, and appeared to arise by proteolysis of, full-length C109 in biosynthetic labelling experiments. C109 spontaneously precipitated after dialysis against NaCl or water, and with prolonged (> 20 weeks) standing was found by electron microscopy to contain a minor (< 5%) fibrillar component that was reactive with antibodies to a C-terminal epitope of APP. Recombinant C109 appears to duplicate some of the biochemical and physicochemical properties of C-terminal A beta-inclusive fragments of APP that have been found in transfected cells, brain cortex and cerebral microvessels.

The amyloid fibrils deposited in Alzheimer's neuritic plaque cores and cerebral blood vessels are mainly composed of aggregated forms of a unique peptide, 39-42 amino acids long, named amyloid beta (A beta). A similar, although soluble, A beta (‘sA beta’) has been identified in cerebrospinal fluid, plasma and cell supernatants, indicating that it is normally produced by proteolytic processing of its precursor protein, amyloid precursor protein (APP). Using direct binding experiments we have isolated and characterized an 80 kDa circulating protein that specifically interacts with a synthetic peptide identical with A beta. The protein was unmistakably identified as SP-40,40 or ApoJ, a cytolytic inhibitor and lipid carrier, by means of amino acid sequence and immunoreactivity with specific antibodies. Immunoprecipitation with anti-SP-40,40 retrieved soluble A beta from cerebrospinal fluid, indicating that the interaction occurs in vivo.

Amyloid beta (A beta), the major constituent of the fibrils composing senile plaques and vascular amyloid deposits in Alzheimer's disease (AD) and related disorders, is a 39-42-residue self-aggregating degradation peptide of a larger multidomain membrane glycoprotein designated amyloid precursor protein (APP). An array of biological functions has been assigned to different APP domains, including growth regulation, neurotoxicity, inhibitory activity of serine proteinases and promotion of cell-cell and cell-matrix interactions. A beta is generated through an as-yet-unknown catabolic pathway that by-passes or inhibits the cleavage of APP within the A beta sequence. We have identified a 16 kDa intermediate APP C-terminal fragment containing A beta in leptomeningeal vessels of aged normal individuals and AD patients by means of its immunoreactivity with a panel of four different anti-(APP C-terminal) antibodies, indicating a different pathway of APP processing. Previous studies have indicated that the APP C-terminal domain is the most likely to be involved in cell-matrix interactions. A 109-amino-acid construct C109 with a sequence analogous to the C-terminal of APP (positions 587-695 of APP695), similar in length and immunoreactivity to the 16 kDa fragment, was found to promote cell adhesion. By use of synthetic peptides, this activity was initially located to the extracellular 28 residues of A beta. Inhibition studies demonstrated that the sequence RHDS (amino acids 5-8 of A beta, corresponding to residues 601-604 of APP695 was responsible for the adhesion-promoting activity. The interaction is dependent on bivalent cations and can be blocked either by the tetrapeptides RHDS and RGDS or by an anti-(beta 1 integrin) antibody. Thus, through integrin-like surface receptors, APP or its derivative proteolytic fragments containing the sequence RHDS may modulate cell-cell or cell-matrix interactions.

Two synthetic peptides with sequences identical with those of fragments of the extracellular domain of the Alzheimer's-disease amyloid precursor protein (APP) were used to raise antibodies. SP28 comprises positions 597-624 of the APP695 isoform, whereas SP41 extends towards the N-terminus (amino acids 584-624) and contains the entire SP28 peptide. Using e.l.i.s.a. and inhibition experiments we identified the two beta-turn-containing segments 602-607 and 617-624 as the epitopes recognized by anti-SP41 and anti-SP28 respectively. Both antibodies immunolabelled amyloid lesions in brains from Alzheimer's-disease patients and patients with related disorders, whereas they were unreactive in control brains. However, when probed on immunoblots, anti-SP28 failed to detect full-length APP from baculovirus-infected Sf9 cells, and anti-SP41 reacted weakly compared with other anti-APP antisera. The data suggest that these antibodies are directed to conformational epitopes not existent in the native molecules but present after alternative APP processing.

Familial amyloidosis, Finnish type (FAF), is an inherited form of systemic amyloidosis clinically characterized by cranial neuropathy and lattice corneal dystrophy. We have demonstrated that the protein subunit isolated from amyloid fibrils shows considerable sequence identity with gelsolin, an actin-binding protein. We have purified the amyloid subunit from a second case and further analysed different fractions from the previous one. Sequence analysis shows that, in both cases, the amyloid subunit starts at position 173 of the mature molecule; it has a heterogeneous N-terminus and contains one amino acid substitution, namely asparagine for aspartic acid, at position 15 (gelsolin residue 187), that is due to a guanine-to-adenine transversion corresponding to nucleotide-654 of human plasma gelsolin cDNA. The substitution maps in a fragment with actin-binding activity and is located in a repetitive motif highly conserved among species. Thus FAF is the first human disease known to be caused by an internal abnormal degradation of a gelsolin variant. We designate this variant of gelsolin-associated amyloidosis ‘Agel Asn-187’.

It is generally postulated that the amyloid beta protein (Abeta) plays a central role in the progressive neurodegeneration observed in Alzheimer's disease. Important pathologic properties of this protein, such as neurotoxicity and resistance to proteolytic degradation, depend on the ability of Abeta to form beta-sheet structures or amyloid fibrils. We report that melatonin, a hormone recently found to protect neurons against Abeta toxicity, interacts with Abeta1-40 and Abeta1-42 and inhibits the progressive formation of beta-sheets and amyloid fibrils. These interactions between melatonin and the amyloid peptides were demonstrated by circular dichroism and electron microscopy for Abeta1-40 and Abeta1-42 and by nuclear magnetic resonance spectroscopy for Abeta1-40. Inhibition of beta-sheets and fibrils could not be accomplished in control experiments when a free radical scavenger or a melatonin analog were substituted for melatonin under otherwise identical conditions. In sharp contrast with conventional anti-oxidants and available anti-amyloidogenic compounds, melatonin crosses the blood-brain barrier, is relatively devoid of toxicity, and constitutes a potential new therapeutic agent in Alzheimer's disease.

A soluble form of Alzheimer disease amyloid beta-protein (sA beta) is transported in the blood and cerebrospinal fluid mainly complexed with apolipoprotein J (apoJ). Using a well-characterized in situ perfused guinea pig brain model, we recently obtained preliminary evidence that apoJ facilitates transport of sA beta (1-40)-apoJ complexes across the blood-brain barrier and the blood-cerebrospinal fluid barrier, but the mechanisms remain poorly understood. In the present study, we examined the transport process in greater detail and investigated the possible role of glycoprotein 330 (gp330)/megalin, a receptor for multiple ligands, including apoJ. High-affinity transport systems with a Km of 0.2 and 0.5 nM were demonstrated for apoJ at the blood-brain barrier and the choroid epithelium in vivo, suggesting a specific receptor-mediated mechanism. The sA beta (1-40)-apoJ complex shared the same transport mechanism and exhibited 2.4- to 10.2-fold higher affinity than apoJ itself. Binding to microvessels, transport into brain parenchyma, and choroidal uptake of both apoJ and sA beta (1-40)-apoJ complexes were markedly inhibited (74-99%) in the presence of a monoclonal antibody to gp330/megalin and were virtually abolished by perfusion with the receptor-associated protein, which blocks binding of all known ligands to gp330. Western blot analysis of cerebral microvessels with the monoclonal antibody to gp330 revealed a protein with a mass identical to that in extracts of kidney membranes enriched with gp330/megalin, but in much lower concentration. The findings suggest that gp330/megalin mediates cellular uptake and transport of apoJ and sA beta (1-40)-apoJ complex at the cerebral vascular endothelium and choroid epithelium.

A prominent feature of the life cycle of intracellular parasites is the profound morphological changes they undergo during development in the vertebrate and invertebrate hosts. In eukaryotic cells, most cytoplasmic proteins are degraded in proteasomes. Here, we show that the transformation in axenic medium of trypomastigotes of Trypanosoma cruzi into amastigote-like organisms, and the intracellular development of the parasite from amastigotes into trypomastigotes, are prevented by lactacystin, or by a peptide aldehyde that inhibits proteasome function. Clasto-lactacystin, an inactive analogue of lactacystin, and cell-permeant peptide aldehyde inhibitors of T. cruzi cysteine proteinases have no effect. We have also identified the 20S proteasomes from T. cruzi as a target of lactacystin in vivo. Our results document the essential role of proteasomes in the stage-specific transformation of a protozoan.

Cystatin C (gamma-trace) was found to be a constitutively secreted protein of isolated human monocytes and mouse peritoneal macrophages, as well as the histiocytic lymphoma cell lines U937, P388D.1, and J774. This proteinase inhibitor is not uniquely secreted by monocytes/macrophages, but was also identified in the conditioned media from several primary cells, including brain cells, and diverse established cell lines. In vitro treatment of resident mouse peritoneal macrophages with either LPS or IFN-gamma caused a downregulation in cystatin C secretion. Elaboration of this protein was also diminished by macrophages that had been stimulated by thioglycollate in vivo, and treatment of these cells with LPS led to further decline. It is suggested that, under some inflammatory conditions, downregulation of cystatin C may contribute to tissue pathology.

Cystatin C is an inhibitor of lysosomal cysteine proteases and consists of 120 amino acids. A variant of cystatin C lacking the first NH2-terminal residues and having one amino acid substitution at position 68 forms amyloid deposits mainly in the walls of brain arteries, causing fatal strokes in Icelandic patients with familial cerebral hemorrhage secondary to a form of an autosomal dominant amyloidosis. To understand the molecular basis of the genetic defect, the gene encoding cystatin C was isolated from genomic DNA libraries made from normal tissue and the brain of an Icelandic patient with hereditary cerebral hemorrhage with amyloidosis (HCHWA-I). The data indicate that the cystatin C gene encodes a polypeptide of 146 amino acids, of which the first 26 correspond to a secretory peptide signal sequence. The gene contains two intervening sequences that interrupt the coding region at amino acids 55 and 93. Comparison with genes encoding salivary cystatins and kininogen proteins show sequence homology and conservation of exon-intron structure. Except for a mutation in the second exon (CAG instead of CTG in the normal gene, resulting in the substitution of glutamine for a leucine residue), the gene cloned from the brain of the Icelandic patient is identical to the normal cystatin C gene. Thus, HCHWA-I is the first familial type of amyloidosis related to a point mutation in a gene encoding for an inhibitor. The mutation in the structural gene encoding cystatin C appears to be the primary defect in this inherited disorder causing amyloid fibril formation and accumulation followed by cerebral hemorrhage.

The main component of Alzheimer's amyloid deposits, A beta, has been found also as a soluble (sA beta) normal constituent of biological fluids and cell culture supernatants. Whether or not sA beta is the immediate precursor of A beta, it is clear that peptides with the same amino acid sequence can have both fibrillar and non-fibrillar conformations. The interconversion mechanism from one form to another is presently under intensive investigation. We have previously described that (i) a synthetic peptide A beta 1-40 immobilized on affinity matrices was able to retrieve apolipoprotein J (apoJ) from plasma and cerebrospinal fluid; and (ii) the interaction of sA beta with apoJ occurs in vivo, as demonstrated by the ability of anti-apoJ to co-precipitate sA beta from normal cerebrospinal fluid. We have characterized the binding between A beta 1-40 and apoJ and found that the interaction is saturable, specific, and reversible. The dissociation constant of 2 x 10(-9) M is indicative of high affinity binding. The stoichiometry of the reaction is 1:1; apoJ has five times more affinity for fresh A beta 1-40 than for the aggregated peptide. Competitive inhibition studies carried out with apolipoprotein E (isoforms E2, E3, and E4), transthyretin, vitronectin, and alpha 1-antichymotrypsin indicate that the complex apoJ.A beta 1-40 cannot be dissociated by any of these competitors at physiologic concentrations. The data strongly suggest that apoJ plays an important role as a carrier protein for sA beta.

Hydropathic anticomplementarity of amino acids indicates that peptides derived from complementary DNA strands may form amphiphilic structures and bind one another. By using this concept, we have found that the antisense peptide Ser-Tyr-Asp-Leu complementary to the segment Gln-Ile-Val-Ala-Gly (residues 55-59) in cystatin C (an inhibitor of cysteine proteases) is located at positions 611-614 of the beta chain of human C4, the fourth component of complement. Here we describe and characterize the specific interaction between cystatin C and C4 by ligand affinity chromatography and ELISA. Interaction between the two native proteins was mimicked on replacement of one of them with the corresponding sense-antisense peptide coupled to a carrier protein, and the binding was inhibited by these synthetic peptides in solution. Through the interaction with C4, cystatin C may play a regulatory role in complement activation that might be of particular importance at tissue sites where both proteins are produced by macrophages.

Familial amyloidosis, Finnish type (FAF), is an autosomal dominant form of familial amyloid polyneuropathy. The novel amyloid fibril protein found in these patients is a degradation fragment of gelsolin, an actin-binding protein. We found a mutation (adenine for guanine) at nucleotide 654 of the gelsolin gene in genomic DNA isolated from five FAF patients. This site is polymorphic since the normal allele was also present in all the patients tested. This mutation was not found in two unaffected family members and 11 normal controls. The A for G transition causes an amino acid substitution (asparagine for aspartic acid) that was found at position 15 of the amyloid protein. The mutation and consequent amino acid substitution may lead to the development of FAF.