Search

The 4 kDa beta-amyloid peptide that forms the amyloid fibrils in the brain parenchyma of Alzheimer's disease patients is derived from the larger integral membrane protein, the amyloid precursor protein. In the nonamyloidogenic pathway, alpha-secretase cleaves the amyloid precursor protein within the beta-amyloid domain, releasing an extracellular portion and thereby preventing deposition of the intact amyloidogenic peptide. The release of the amyloid precursor protein from both SH-SY5Y and IMR-32 neuronal cells by alpha-secretase was blocked by batimastat and other related synthetic hydroxamic acid-based zinc metalloprotease inhibitors, but not by the structurally unrelated zinc metalloprotease inhibitors enalaprilat and phosphoramidon. Batimastat inhibited the release of the amyloid precursor protein from both cell lines with an I50 value of 3 microM. Removal of the thienothiomethyl substituent adjacent to the hydroxamic acid moiety or the substitution of the P2' substituent decreased the inhibitory potency of batimastat toward alpha-secretase. In the SH-SY5Y cells, both the basal and the carbachol-stimulated release of the amyloid precursor protein were blocked by batimastat. In contrast, neither the level of full-length amyloid precursor protein nor its cleavage by beta-secretase were inhibited by any of the zinc metalloprotease inhibitors examined. In transfected IMR-32 cells, the release of both the amyloid precursor protein and angiotensin converting enzyme was inhibited by batimastat, marimastat, and BB2116 with I50 values in the low micromolar range, while batimastat and BB2116 inhibited the release of both proteins from HUVECs. The profile of inhibition of alpha-secretase by batimastat and structurally related compounds is identical with that observed with the angiotensin converting enzyme secretase suggesting that the two are closely related zinc metalloproteases.

Mammalian angiotensin-converting enzyme (ACE; EC 3.4.15.1) is one of several proteins that exist in both membrane-bound and soluble forms as a result of a post-translational proteolytic processing event. For ACE we have previously identified a metalloprotease (secretase) responsible for this proteolytic cleavage. The effect of a range of structurally related zinc metalloprotease inhibitors on the activity of the secretase has been examined. Batimastat (BB94) was the most potent inhibitor of the secretase in pig kidney microvillar membranes, displaying an IC50 of 0.47 microM, whereas TAPI-2 was slightly less potent (IC50 18 microM). Removal of the thienothiomethyl substituent adjacent to the hydroxamic acid moiety or the substitution of the P2' substituent decreased the inhibitory potency of batimastat towards the secretase. Several other non-hydroxamate-based collagenase inhibitors were without inhibitory effect on the secretase, indicating that ACE secretase is a novel zinc metalloprotease that is realted to, but distinct from, the matrix metalloproteases. The full-length amphipathic form of ACE was labelled selectively with 3-trifluoromethyl-3-(m-[125I]iodophenyl)diazirine in the membrane-spanning hydrophobic region. Although trypsin was able to cleave the hydrophobic anchoring domain from the bulk of the protein, there was no cleavage of full-length ACE by a Triton X-100-solubilized pig kidney secretase preparation when the substrate was in detergent solution. In contrast, the Triton X-100-solubilized secretase preparation released ACE from pig intestinal microvillar membranes, which lack endogenous secretase activity, and cleaved the purified amphipathic form of ACE when it was incorporated into artificial lipid vesicles. Thus the secretase has an absolute requirement for its substrate to be inserted in a lipid bilayer, a factor that might have implications for the development of cell-free assays for other membrane protein secretases. ACE secretase could be solubilized from the membrane with Triton-X-100 and CHAPS, but not with n-octyl beta-D-glucopyranoside. Furthermore trypsin could release the secretase from the membrane, implying that like its substrate, ACE, it too is a stalked integral membrane protein.

The recently identified Alzheimer's disease-associated presenilin 1 and 2 (PS1 and PS2) genes encode two homologous multi membrane-spanning proteins. Rabbit antibodies to the N-terminal domain of PS1 detected PS1 in human neuroblastoma SH-SY5Y wild type and PS1 transfectants (SY5Y-PS1) as well as in mouse P19, in CHO-K1 and CHO-APP770 transfected cells, in rat cerebellar granule and hippocampal neurons, and astrocytes. Immunoblotting detected full-length protein of 50 kDa, and a major presumptive cleavage product of 30 kDa. The immunofluorescence pattern resembled labeling of the endoplasmic reticulum-Golgi intermediate compartment (ERGIC) marker protein ERGIC-53. PS1 distribution showed slight condensation after brefeldin A and more marked condensation after incubation of cells at 16 degrees C, characteristic of the ERGIC compartment. Double labeling showed colocalization of ERGIC-53 with PS1 in the SY5Y-PS1 cells. PS1 labeling of SY5Y-PS1 and P19 cells showed overlap of the cis-Golgi marker p210 and colocalization with p210 after brefeldin A which causes redistribution of p210 to the ERGIC. Expression of PS1 did not change in level or cellular distribution during development of neurons in culture. Double labeling for the amyloid precursor protein (APP) and PS1 on SY5Y-PS1 cells and CHO-APP770 cells showed some overlap under control conditions. These results indicate that PS1 is a resident protein of the ERGIC and could be involved in trafficking of proteins, including APP, between the ER and Golgi compartments.

A diverse range of membrane proteins of Type 1 or Type II topology also occur as a circulating, soluble form. These soluble forms are often derived from the membrane form by proteolysis by a group of enzymes referred to collectively as 'secretases' or 'sheddases'. The cleavage generally occurs close to the extracellular face of the membrane, releasing physiologically active protein. This secretion process also provides a mechanism for down-regulating the protein at the cell surface. Examples of such post-translational proteolysis are seen in the Alzheimer's amyloid precursor protein, the vasoregulatory enzyme angiotensin converting enzyme, transforming growth factor-alpha, the tumour necrosis factor ligand and receptor superfamilies, certain cytokine receptors, and others. Since the proteins concerned are involved in pathophysiological processes such as neurodegeneration, apoptosis, oncogenesis and inflammation, the secretases could provide novel therapeutic targets. Recent characterization of these individual secretases has revealed common features, particularly sensitivity to certain metalloprotease inhibitors and upregulation of activity by phorbol esters. It is therefore likely that a closely related family of metallosecretases controls the surface expression of multiple integral membrane proteins. Current knowledge of the various secretases are compared in this Review, and strategies for cell-free assays of such proteases are outlined as a prelude to their ultimate purification and cloning.

Presenilin-1 (PS-1) has been identified as the protein encoded by the chromosome 14 locus that, when mutated, leads to familial Alzheimer's disease (FAD). Using PS-1 transfected SHSY5Y neuroblastoma cells, we have demonstrated by immunodetection, using polyclonal antibodies, that PS-1 is processed to give two fragments: an N-terminal 28 kDa fragment, and a C-terminal 18 kDa fragment. In a number of non-transfected cell types, most PS-1 is detected as the cleaved products. The molecular weights of the PS-1 cleavage products suggest that the cleavage point will most probably be within a region of the hydrophilic loop domain coded for by either exon 8 or 9 of the PS-1 gene. The clustering of FAD mutations within exon 8 strongly suggests that it encodes a key functional domain. It seems likely that the cleavage of PS-1 is crucial to some aspect of its functionality. An understanding of this process will give insights into the pathology of AD, and may offer new opportunities for therapeutic intervention.

The expression of the presenilin 1 (PS-1) gene has been investigated by in situ hybridization in early onset familial Alzheimer's disease (FAD), late onset Alzheimer's disease (AD) and normal control brain. Mutations in this gene are responsible for chromosome 14-linked FAD. We have found that presenilin 1 mRNA is present throughout the human brain with a distribution consistent with both a glial and neuronal localization. The in situ hybridization pattern was similar for the controls, the early onset FAD cases and the late onset AD cases. However, one of the two forms of the mRNA for PS-1, the long form (which contains a sequence encoding a four amino acid (VRSQ) insert at its 5' end) was significantly reduced in early onset FAD brain compared with late onset AD. We suggest that this long transcript may alter the normal pathway for processing of amyloid precursor protein, the protein which appears to be central in the pathogenesis of AD.

Recent studies suggest that mutations in the presenilin 1 gene, which encodes a polypeptide predicted to be a multispanning membrane protein, are responsible for the majority of cases of early onset, autosomal dominant Alzheimer's disease. Here we describe a further mutation in the presenilin 1 gene (R269G) in a family with early onset Alzheimer's disease. This mutation is in exon 8 which appears to be a favoured region for pathogenic mutations. In the presenilin protein the region coded for by this exon is likely to comprise a domain located on the membrane surface. We discuss the likely effects of the exon 8 mutations on the structure of the exon and in the pathogenesis of the disease.

Missense mutations in the presenilin-1 (PS-1) and presenilin-2 (PS-2) genes have been shown to be causes of autosomal dominant Alzheimer's disease (the AD3 and AD4 loci, respectively). Alternative splicing has previously been reported in the PS-1 gene. In this study, elucidation of intron/exon boundary sequences revealed that PS-2 is encoded by 10 coding exons. In addition, PS-2 cDNA cloning and RT-PCR using RNA from a variety of normal tissues revealed the presence of alternatively spliced products. These products included species with in frame omissions of exon 8 and simultaneous omissions of exons 3 and 4.

The presenilin 1 gene has recently been identified as the locus on chromosome 14 which is responsible for a large proportion of early onset, autosomal dominantly inherited Alzheimer's disease (AD). We have elucidated the intron/exon structure of the gene and designed intronic primers to enable direct sequencing of the entire coding region (10 exons) of the presenilin gene in a large number of families. This strategy has enabled us to find a further two novel mutations in the gene. We discuss the distribution of mutations and the proportions of autosomal dominant AD with a mean age of onset below 60 years caused by mutations in this gene.

Approximately 75% of AD patients have an onset of the disease after the age of 60 years, and 60% of AD patients have no family history of the disease. Some cases of EOAD are clearly inherited in an autosomal-dominant manner. The beta APP gene on chromosome 21, the PS-1 gene on chromosome 14, and the PS-2 gene on chromosome 1 have all been characterized as genes in which mutations lead to familial EOAD. For LOAD, the work on ApoE indicates that the epsilon 4 allele is a risk factor for developing AD. However, 35-50% of all AD patients do not have an epsilon 4 allele. Other loci contributing to LOAD remain to be mapped and characterized. As in other complex disorders, these additional loci may involve genetic interactions with the known AD loci. Identification of all susceptibility loci for AD is a major goal in resolving the pathogenesis of AD.

A straightforward in vivo model of collagen degradation is described that can be used to measure the effects of different classes of proteinase inhibitors. Air pouches, formed subcutaneously in the dorsal thoracic region of rats, were inflamed 6 to 8 days later by injecting lambda-type carrageenan. 14C-Collagen was injected into the air pouches either 1 day before or 1 day after lambda-carrageenan-induced inflammation: in the latter case, the inflammatory exudate fluid was drained from the air pouches immediately prior to administering 14C-collagen. Ninety percent of the 14C-collagen was degraded and cleared within 3 days from pre-inflamed air pouches, but degradation was much slower from the post-inflamed or non-inflamed air pouches. Proteinase inhibitors injected simultaneously with the 14C-collagen, and again 6 hr later, reduced the extent of 14C-collagen degradation from air pouches measured after 24 hr. Forty-two percent of the degradation of 14C-collagen could be inhibited by a mixture of enzyme inhibitors (leupeptin, alpha 1-anti-proteinase, aprotinin, and pepstatin) injected together with 1,10 phenanthroline, the zinc metalloenzyme inhibitor. The 1,10 phenanthroline alone caused a 33% inhibition of 14C-collagen degradation, and the inhibitor mixture given alone inhibited 14C-collagen loss by 25%. Approximately 60% of the degradation of 14C-collagen in this model was mediated by mechanisms resistant to this combination of proteinase inhibitors, which may indicate the significant involvement of non-enzymic modalities, or degradation in intracellular compartments inaccessible to extracellular agents.

A simple in vivo model of collagen degradation has been developed, and the effects of various agents have been tested. Type I collagen was prepared from rat skin and acetylated with either [3H]- or [14C] acetic anhydride. The radiolabelled collagen was added to sterile cotton buds and incubated at 37 degrees C to allow the collagen to form native fibrils that were firmly adsorbed to the cotton matrix. After subcutaneous implantation of the collagen-gelled cotton buds into rats, the radiolabelled collagen was progressively removed over a period of weeks by an infiltrating granuloma. Of the agents that were administered directly into the cotton buds using subcutaneously implanted osmotic mini-pumps, only the synthetic collagenase inhibitors CI-A (containing a hydroxamate moiety as a zinc ligand) and CI-C (containing a thiol moiety as a zinc ligand) were able to prevent the removal of collagen: their efficacy correlated with the level of collagenase inhibitory activity assayed in the exudate fluid sequestered within the cotton bud granuloma. Of the agents that were administered systemically, including anti-inflammatory drugs and other compounds used as therapies for arthritis, only hydrocortisone was able to inhibit the removal of radiolabelled collagen. These results suggest that, in this model, interstitial collagenase, a member of the matrix metalloproteinase family, comprised the major degradative pathway for collagen. The collagen-gelled cotton bud model is a useful test system for delineating those processes that result in collagen catabolism. In addition, the model can be used for testing agents, including those of limited or unknown systemic bioavailability, in order to discover novel therapeutic agents for preventing collagen degradation in connective tissue diseases such as arthritis.

The synthesis of a series of N-phosphonalkyl dipeptides 6 is described. Syntheses were devised that allowed the preparation of single diastereoisomers and the assignment of stereochemistry. The compounds were evaluated in vitro for their ability to inhibit the degradation of radiolabeled collagen by purified human lung fibroblast collagenase. Several of the compounds were potent collagenase inhibitors and were at least 10-fold more potent than their corresponding N-carboxyalkyl analogues. Activity was lost when the phosphonic acid group P(O)(OH)2 was replaced by the phosphinic acid groups P(O)(H)(OH) and P(O)(Me)(OH). At the P1 position, (R)- or (S)-alkyl groups, especially ethyl and methyl (e.g., 12a,b, 52a,b, and 53a,b), or an (R)-phenethyl moiety (55a) conferred high potency (IC50 values in the range 0.23-0.47 microM). (S)-Stereochemistry was preferred for the P1' isobutyl side chain. Structure-activity relationships were also investigated at the P2' site, and interestingly, compounds with basic side chains, such as the guanidine 57a, were equipotent with more lipophilic compounds, such as 52a. As with other series of collagenase inhibitors, potency was enhanced by introducing bicyclic aromatic P2' substituents. The most potent phosphonic acid of the series was the bicyclic aromatic P2' tryptophan analogue 59a (IC50 0.05 microM).

The synthesis of a series of thiol-containing, modified dipeptide inhibitors (8) of human collagenase, which incorporate various carboxylic acid derivatives at the presumed P1 position, beta to the thiol group, is described. The compounds were evaluated, in vitro, for their ability to inhibit the degradation of rat skin type 1 collagen by purified human lung fibroblast collagenase, and structure-activity relationship studies are described. Optimum potency (IC50 values in the nanomolar range) was achieved by incorporating methyl (compounds 43a, 56a, and 57ab) or benzyl esters (44a) at the P1 position. Small amides were also accommodated (e.g. primary amide 47a), but in general, increasing the size of the P1 amide substituent lowered potency. PheNHMe, TrpNHMe, and Tyr(Me)NHMe substituents were found to be approximately equipotent P2'-residues. The results of testing all four diastereoisomers 56a-d of the compound with (S)-TrpNHMe at the P2' position indicated that the S,S,S diastereoisomer 56a possessed highest potency (IC50 2.5 nM) and that the second most potent diastereoisomer was 56d (IC50 12 nM) with the R,R,S configuration. It appeared that the orientation of the P1' and the thiol-bearing centers to each other is a more critical influence on potency than any absolute stereochemical requirements. It is suggested that the high potency of the beta-mercapto carboxylic acid derivatives may be a consequence of bidentate coordination of the thiol and carbonyl groups to the active-site zinc ion in the collagenase enzyme.