Metabolism of amyloid precursor protein in COS cells transfected with a β-secretase candidate

Thimet oligopeptidase (TOP) is a thiol- andmetallo-dependent peptidase and has been shown to beone of the -secretase candidates. TOPexpressed in COS cells cleaved amyloid precursorprotein (APP) at the -secretase site, and wefound a proteolytic product of APP called secretedform of APP by -secretase (sAPP) in theconditioned media. Here we demonstrate thatsAPP was increased in conditioned media whenTOP was coexpressed in COS cells with APP and treatedwith an ADAM inhibitor SI-27. In addition, althoughTOP expressed in COS cell was localized at nuclei orGolgi apparatus, it exclusively colocalized at Golgiapparatus when APP was coexpressed with TOP.     

Glycosylphosphatidylinositol-anchored proteins play an important role in the biogenesis of the Alzheimer's amyloid beta-protein.

The Alzheimer's amyloid protein (Abeta) is released from the larger amyloid beta-protein precursor (APP) by unidentified enzymes referred to as beta- and gamma-secretase. beta-Secretase cleaves APP on the amino side of Abeta producing a large secreted derivative (sAPPbeta) and an Abeta-bearing C-terminal derivative that is subsequently cleaved by gamma-secretase to release Abeta. Alternative cleavage of the APP by alpha-secretase at Abeta16/17 releases the secreted derivative sAPPalpha. In yeast, alpha-secretase activity has been attributed to glycosylphosphatidylinositol (GPI)-anchored aspartyl proteases. To examine the role of GPI-anchored proteins, we specifically removed these proteins from the surface of mammalian cells using phosphatidylinositol-specific phospholipase C (PI-PLC). PI-PLC treatment of fetal guinea pig brain cultures substantially reduced the amount of Abeta40 and Abeta42 in the medium but had no effect on sAPPalpha. A mutant CHO cell line (gpi85), which lacks GPI-anchored proteins, secreted lower levels of Abeta40, Abeta42, and sAPPbeta than its parental line (GPI+). When this parental line was treated with PI-PLC, Abeta40, Abeta42, and sAPPbeta decreased to levels similar to those observed in the mutant line, and the mutant line was resistant to these effects of PI-PLC. These findings provide strong evidence that one or more GPI-anchored proteins play an important role in beta-secretase activity and Abeta secretion in mammalian cells. The cell-surface GPI-anchored protein(s) involved in Abeta biogenesis may be excellent therapeutic target(s) in Alzheimer's disease.     

Expression of human beta-secretase in the mouse brain increases the steady-state level of beta-amyloid

beta-Site APP-cleaving enzyme (BACE) initiates the processing of the amyloid precursor protein (APP) leading to the generation of beta-amyloid, the main component of Alzheimer's disease senile plaques. BACE (Asp2, memapsin 2) is a type I transmembrane aspartyl protease and is responsible for the beta-secretase cleavage of APP producing different endoproteolytic fragments referred to as the carboxy-terminal C99, C89 and the soluble ectodomain sAPPbeta. Here we describe two transgenic mouse lines expressing human BACE in the brain. Overexpression of BACE augments the amyloidogenic processing of APP as demonstrated by decreased levels of full-length APP and increased levels of C99 and C89 in vivo. In mice expressing huBACE in addition to human APP wild-type or carrying the Swedish mutation, the induction of APP processing characterized by elevated C99, C89 and sAPPbeta, results in increased brain levels of beta-amyloid peptides Abeta40 and Abeta42 at steady-state.     

cDNA Cloning and Molecular Characterization of Human Brain Metalloprotease MP100

Metalloprotease MP100 was originally isolated as a beta-secretase candidate from human brain using a beta-amyloid precursor protein (beta-APP)-derived p-nitroanilide (pNA) peptide substrate. Peptide sequences from purified MP100 were now found to resemble sequences reported for a puromycin-sensitive aminopeptidase (PSA) highly enriched in brain, and cDNA cloning revealed nearly complete homology of MP100 to PSA, with only a single bp difference resulting in an amino acid change at position 184. Another MP100 cDNA encoded a protein with a 36-amino acid deletion (positions 180-217) and a two-amino acid insertion after Val533. Purified recombinant human MP100 cleaved the original pNA substrate as well as a free beta-site-spanning amyloid beta (A beta) peptide (A beta(-10/+10)), generating A beta(1-10). The latter substrate, however, remained uncleaved, if N- and C-terminally blocked, and also purified beta-APP was not cleaved. Double immunoimaging revealed partial, patchy, colocalization of beta-APP and MP100 in doubly transfected human embryonic kidney cells (HEK cells) and in normal neuroblastoma cells, and both proteins could be coimmunoprecipitated from rat brain extracts, suggesting their close vicinity in vivo. Coexpression of MP100 and beta-APP695, however, did not boost A beta levels in HEK cells, although active enzyme was produced. Thus, MP100 does not exert true beta-secretase-like function in cells, although it may well act as a secondary exoprotease in a complex beta-APP/A beta metabolism.     

Sialylation enhances the secretion of neurotoxic amyloid-beta peptides

Alzheimer's disease (AD) is characterized by amyloid-beta peptide (Abeta) deposition in the brain. Abeta is produced by sequential cleavage of amyloid precursor protein (APP) by beta-secretase (BACE1: beta-site APP-cleaving enzyme 1) and gamma-secretase. Previously, we demonstrated that BACE1 also cleaves beta-galactoside alpha2,6-sialyltransferase (ST6Gal-I) and down-regulates its transferase activity. Here, we report that overexpression of ST6Gal-I in Neuro2a cells enhanced alpha2,6-sialylation of endogenous APP and increased the extracellular levels of its metabolites [Abeta by two-fold, soluble APPbeta (sAPPbeta) by three-fold and sAPPalpha by 2.5-fold). Sialylation-deficient mutant (Lec-2) cells secreted half as much Abeta as wild-type Chinese hamster ovary (CHO) cells. Furthermore, wild-type CHO cells showed enhanced secretion of the APP metabolites upon ST6Gal-I overexpression, whereas Lec-2 cells did not, indicating that the secretion enhancement requires sialylation of cellular protein(s). Secretion of metabolites from a mutant APP (APP-Asn467,496Ala) that lacked N-glycosylation sites was not enhanced upon ST6Gal-I overexpression, suggesting that the N-glycans on APP itself are required for the enhanced secretion. In the mouse brain, the amount of alpha2,6-sialylated APP appeared to be correlated with the sAPPbeta level. These results suggest that sialylation of APP promotes its metabolic turnover and could affect the pathology of AD.     

Assembly of a polymeric chain of SUMO1 on human topoisomerase I in vitro

Human (h) DNA topoisomerase I has been identified as a major SUMO1 target in camptothecin-treated cells. In response to TOP1-mediated DNA damage induced by camptothecin, multiple SUMO1 molecules are conjugated to the N-terminal domain of a single TOP1 molecule. To investigate the molecular mechanism of SUMO1 conjugation to TOP1, an in vitro system using purified SAE1/2, Ubc9, SUMO1, and TOP1 peptides was developed. Consistent with results from in vivo studies, multiple SUMO1 molecules were found to be conjugated to the N-terminal domain of a single TOP1 molecule. Systematic analysis has identified a single major SUMO1 conjugation site located between amino acid residues 110 and 125 that contains a single lysine residue at 117 (Lys-117). Using a short peptide spanning this region, we showed that a poly-SUMO1 chain was assembled in this peptide at Lys-117. Interestingly, a Ubc9-poly-SUMO1 intermediate had accumulated to a high level when the sumoylation assay was performed in the absence of hTOP1 substrate, suggesting a possibility that the poly-SUMO1 chain is formed on Ubc9 first and then transferred en bloc onto hTOP1. This is the first definitive demonstration of the assembly of a poly-SUMO1 chain on protein substrate. These results offer new insight into hTOP1 polysumoylation in response to TOP1-mediated DNA damage and may have general implications in protein polysumoylation.     

Intracellular copper deficiency increases amyloid-beta secretion by diverse mechanisms

In Alzheimer's disease there is abnormal brain copper distribution, with accumulation of copper in amyloid plaques and a deficiency of copper in neighbouring cells. Excess copper inhibits Abeta (amyloid beta-peptide) production, but the effects of deficiency have not yet been determined. We therefore studied the effects of modulating intracellular copper levels on the processing of APP (amyloid precursor protein) and the production of Abeta. Human fibroblasts genetically disposed to copper accumulation secreted higher levels of sAPP (soluble APP ectodomain)alpha into their medium, whereas fibroblasts genetically manipulated to be profoundly copper deficient secreted predominantly sAPPbeta and produced more amyloidogenic beta-cleaved APP C-termini (C99). The level of Abeta secreted from copper-deficient fibroblasts was however regulated and limited by alpha-secretase cleavage. APP can be processed by both alpha- and beta-secretase, as copper-deficient fibroblasts secreted sAPPbeta exclusively, but produced primarily alpha-cleaved APP C-terminal fragments (C83). Copper deficiency also markedly reduced the steady-state level of APP mRNA whereas the APP protein level remained constant, indicating that copper deficiency may accelerate APP translation. Copper deficiency in human neuroblastoma cells significantly increased the level of Abeta secretion, but did not affect the cleavage of APP. Therefore copper deficiency markedly alters APP metabolism and can elevate Abeta secretion by either influencing APP cleavage or by inhibiting its degradation, with the mechanism dependent on cell type. Overall our results suggest that correcting brain copper imbalance represents a relevant therapeutic target for Alzheimer's disease.     

Protein kinase C-dependent alpha-secretase competes with beta-secretase for cleavage of amyloid-beta precursor protein in the trans-golgi network

The release of amyloidogenic amyloid-beta peptide (Abeta) from amyloid-beta precursor protein (APP) requires cleavage by beta- and gamma-secretases. In contrast, alpha-secretase cleaves APP within the Abeta sequence and precludes amyloidogenesis. Regulated and unregulated alpha-secretase activities have been reported, and the fraction of cellular alpha-secretase activity regulated by protein kinase C (PKC) has been attributed to the ADAM (a disintegrin and metalloprotease) family members TACE and ADAM-10. Although unregulated alpha-secretase cleavage of APP has been shown to occur at the cell surface, we sought to identify the intracellular site of PKC-regulated alpha-secretase APP cleavage. To accomplish this, we measured levels of secreted ectodomains and C-terminal fragments of APP generated by alpha-secretase (sAPPalpha) (C83) versus beta-secretase (sAPPbeta) (C99) and secreted Abeta in cultured cells treated with PKC and inhibitors of TACE/ADAM-10. We found that PKC stimulation increased sAPPalpha but decreased sAPPbeta levels by altering the competition between alpha- versus beta-secretase for APP within the same organelle rather than by perturbing APP trafficking. Moreover, data implicating the trans-Golgi network (TGN) as a major site for beta-secretase activity prompted us to hypothesize that PKC-regulated alpha-secretase(s) also reside in this organelle. To test this hypothesis, we performed studies demonstrating proteolytically mature TACE intracellularly, and we also showed that regulated alpha-secretase APP cleavage occurs in the TGN using an APP mutant construct targeted specifically to the TGN. By detecting regulated alpha-secretase APP cleavage in the TGN by TACE/ADAM-10, we reveal ADAM activity in a novel location. Finally, the competition between TACE/ADAM-10 and beta-secretase for intracellular APP cleavage may represent a novel target for the discovery of new therapeutic agents to treat Alzheimer's disease.     

Alzheimer's beta-secretase cleaves a glycosyltransferase as a physiological substrate

Alzheimer's beta-secretase (BACE1) is a membrane-bound protease that cleaves the amyloid precursor protein (APP) in the trans-Golgi network, an initial step in the pathogenesis of Alzheimer's disease. Although BACE1 is distributed among various tissues including brain, its physiological substrate other than APP have not been identified. We have recently found that when BACE1 was overexpressed in COS cells together with alpha2,6-sialyltransferase (ST6Gal I), the secretion of ST6Gal I markedly increased, suggesting that BACE1 cleaves ST6Gal I as a physiological substrate. Thus BACE1 is the first identified protease that is responsible for the cleavage and secretion of glycosyltransferases.     

alpha- and beta-secretase: profound changes in Alzheimer's disease

The amyloid plaque, a neuropathological hallmark of Alzheimer's disease, is produced by the deposition of beta-amyloid (Abeta) peptide, which is cleaved from Amyloid Precursor Protein (APP) by the enzyme beta-secretase. Only small amounts of Abeta form in normal brain; more typically this is precluded by the processing of APP by alpha-secretase. Here, we describe a decrease in alpha-secretase (81% of normal) and a large increase in beta-secretase activity (185%) in sporadic Alzheimer's disease temporal cortex. Since alpha-secretase is present principally in neurons known to be vulnerable in Alzheimer's disease, and there is known competition between alpha- and beta-secretase for the substrate APP, it is significant that the majority of Alzheimer samples tested here were low in alpha-secretase. Eighty percent of Alzheimer brains examined had an increase in beta-secretase, a decrease in alpha-secretase, or both; which may account for the means by which the majority of people develop Alzheimer's disease.     

Alzheimer's β-secretase cleaves a glycosyltransferase as a physiological substrate

Alzheimer's beta-secretase (BACE1) is a membrane-bound protease that cleaves the amyloid precursor protein (APP) in the trans-Golgi network, an initial step in the pathogenesis of Alzheimer's disease. Although BACE1 is distributed among various tissues including brain, its physiological substrate other than APP have not been identified. We have recently found that when BACE1 was overexpressed in COS cells together with α2,6-sialyltransferase (ST6Gal I), the secretion of ST6Gal I markedly increased, suggesting that BACE1 cleaves ST6Gal I as a physiological substrate. Thus BACE1 is the first identified protease that is responsible for the cleavage and secretion of glycosyltransferases. Published in 2004. This revised version was published online in August 2006 with corrections to the Cover Date.     

Interferon gamma stimulates beta-secretase expression and sAPPbeta production in astrocytes

Neurons, but not astrocytes, are known as the major source of Abeta, because astrocytes express low levels of putative beta-secretase (BACE). Astrocytes near senile plaque cores show enhanced levels of BACE protein expression, however, suggesting that astrocytes can contribute to Abeta production under pathological conditions. To investigate factors that stimulate BACE protein expression in astrocytes, we tested the effects of interleukin-1beta (IL-1beta) and interferon-gamma (IFN-gamma) on BACE protein expression in U373MG astrocytoma cells and primary astrocyte cultures from Tg2576 mouse brains. BACE protein expression and sAPPbeta production were dramatically increased, without changes in holo APP levels, following IFN-gamma treatment in both cell types. AG490, which is a blocker of IFN-gamma-induced STAT signaling, decreased IFN-gamma-induced BACE protein expression and sAPPbeta production in a dose-dependent manner. These results show that astrocytes are capable of expressing BACE and producing sAPPbeta in response to certain stimulating factors, and IFN-gamma is one such factor.     

Amyloidogenic processing of the Alzheimer beta-amyloid precursor protein depends on lipid rafts

Formation of senile plaques containing the beta-amyloid peptide (A beta) derived from the amyloid precursor protein (APP) is an invariant feature of Alzheimer's disease (AD). APP is cleaved either by beta-secretase or by alpha-secretase to initiate amyloidogenic (release of A beta) or nonamyloidogenic processing of APP, respectively. A key to understanding AD is to unravel how access of these enzymes to APP is regulated. Here, we demonstrate that lipid rafts are critically involved in regulating A beta generation. Reducing cholesterol levels in N2a cells decreased A beta production. APP and the beta-site APP cleavage enzyme (BACE1) could be induced to copatch at the plasma membrane upon cross-linking with antibodies and to segregate away from nonraft markers. Antibody cross-linking dramatically increased production of A beta in a cholesterol-dependent manner. A beta generation was dependent on endocytosis and was reduced after expression of the dynamin mutant K44A and the Rab5 GTPase-activating protein, RN-tre. This inhibition could be overcome by antibody cross-linking. These observations suggest the existence of two APP pools. Although APP inside raft clusters seems to be cleaved by beta-secretase, APP outside rafts undergoes cleavage by alpha-secretase. Thus, access of alpha- and beta-secretase to APP, and therefore A beta generation, may be determined by dynamic interactions of APP with lipid rafts.     

A protease activity associated with acetylcholinesterase releases the membrane-bound form of the amyloid protein precursor of Alzheimer's disease.

Amyloid deposits in the brains of patients with Alzheimer's disease (AD) contain a protein (beta A4) which is abnormally cleaved from a larger transmembrane precursor protein (APP). APP is believed to be normally released from membranes by the action of a protease referred to as APP secretase. Amyloid deposits have also been shown to contain the enzyme acetylcholinesterase (AChE). In this study, a protease activity associated with AChE was found to possess APP secretase activity, stimulating the release of a soluble 100K form of APP from HeLa cells transfected with an APP cDNA. The AChE-associated protease was strongly and specifically inhibited by soluble APP (10 nM) isolated from human brain. The AChE-associated protease cleaved a synthetic beta A4 peptide at the predicted cleavage site. As AChE is decreased in AD, a deficiency of its associated protease might explain why APP is abnormally processed in AD.     

The novel beta-secretase inhibitor KMI-429 reduces amyloid beta peptide production in amyloid precursor protein transgenic and wild-type mice

Alzheimer's disease (AD) is a neurodegenerative disorder characterized by the accumulation of amyloid plaques and neurofibrillary tangles in the brain. The major component of the plaques, amyloid beta peptide (Abeta), is generated from amyloid precursor protein (APP) by beta- and gamma-secretase-mediated cleavage. Because beta-secretase/beta-site APP cleaving enzyme 1 (BACE1) knockout mice produce much less Abeta and grow normally, a beta-secretase inhibitor is thought to be one of the most attractive targets for the development of therapeutic interventions for AD without apparent side-effects. Here, we report the in vivo inhibitory effects of a novel beta-secretase inhibitor, KMI-429, a transition-state mimic, which effectively inhibits beta-secretase activity in cultured cells in a dose-dependent manner. We injected KMI-429 into the hippocampus of APP transgenic mice. KMI-429 significantly reduced Abeta production in vivo in the soluble fraction compared with vehicle, but the level of Abeta in the insoluble fraction was unaffected. In contrast, an intrahippocampal injection of KMI-429 in wild-type mice remarkably reduced Abeta production in both the soluble and insoluble fractions. Our results indicate that the beta-secretase inhibitor KMI-429 is a promising candidate for the treatment of AD.     

Kinetic properties of cathepsin D and BACE 1 indicate the need to search for additional beta-secretase candidate(s)

Many studies suggest that BACE 1 is the genuine beta-secretase; however, this is not undisputed. The wild-type (WT) beta-site of the amyloid precursor protein (APP) present in the worldwide population is cleaved very slowly (kcat/Km: approx. 50 m(-1) s(-1)), while proteases acting on relevant substrates are much more efficient (kcat/Km: 10(4)-10(6) m(-1) s(-1)). Knock-out of BACE 1 in mouse markedly reduces A beta formation. Nevertheless, studies in other systems show that knock-out experiments in rodents and corresponding genetic defects in human may reveal different phenotypes. Considering these issues, we searched for other beta-secretase candidate(s), identified cathepsin D, and evaluated properties of cathepsin D related to BACE 1 that were not examined previously. The kinetic constants (kcat, Km, kcat/Km) for cleaving peptides with beta-sites of the WT or the mutated Swedish families (SW) APP by human BACE 1 and cathepsin D were determined and found to be similar. Western blots reveal that in human brain cathepsin D is approximately 280-fold more abundant than BACE 1. Furthermore, pepstatin A strongly inhibits the cleavage of SW and WT peptides by both brain extracts and cathepsin D, but not by BACE 1. These findings indicate that beta-secretase activity observed in brain extracts is mainly due to cathepsin D. Nevertheless, as both BACE 1 and cathepsin D show poor activity towards the WT beta-site sequence, it is necessary to continue the search for additional beta-secretase candidate(s).     

Inhibition of cathepsin B reduces beta-amyloid production in regulated secretory vesicles of neuronal chromaffin cells: evidence for cathepsin B as a candidate beta-secretase of Alzheimer's disease

The regulated secretory pathway of neurons is the major source of extracellular A beta that accumulates in Alzheimer's disease (AD). Extracellular A beta secreted from that pathway is generated by beta-secretase processing of amyloid precursor protein (APP). Previously, cysteine protease activity was demonstrated as the major beta-secretase activity in regulated secretory vesicles of neuronal chromaffin cells. In this study, the representative cysteine protease activity in these secretory vesicles was purified and identified as cathepsin B by peptide sequencing. Immunoelectron microscopy demonstrated colocalization of cathepsin B with A beta in these vesicles. The selective cathepsin B inhibitor, CA074, blocked the conversion of endogenous APP to A beta in isolated regulated secretory vesicles. In chromaffin cells, CA074Me (a cell permeable form of CA074) reduced by about 50% the extracellular A beta released by the regulated secretory pathway, but CA074Me had no effect on A beta released by the constitutive pathway. Furthermore, CA074Me inhibited processing of APP into the COOH-terminal beta-secretase-like cleavage product. These results provide evidence for cathepsin B as a candidate beta-secretase in regulated secretory vesicles of neuronal chromaffin cells. These findings implicate cathepsin B as beta-secretase in the regulated secretory pathway of brain neurons, suggesting that inhibitors of cathepsin B may be considered as therapeutic agents to reduce A beta in AD.     

A novel serine protease highly expressed in the pancreas is expressed in various kinds of cancer cells

We have isolated a cDNA that encodes a novel serine protease, prosemin, from human brain. The cDNA of human prosemin is 1306 bp, encoding 317 amino acids. It showed significant homology with the sequence of a chromosome 16 cosmid clone (accession no. NT_037887.4). The prosemin gene contains six exons and five introns. The amino acid sequence of prosemin shows significant homology to prostasin, gamma-tryptase, and testisin (43%, 41%, and 38% identity, respectively), the genes of which are also located on chromosome 16. Northern hybridization showed that prosemin is expressed predominantly in the pancreas and weakly in the prostate and cerebellum. However, western blot and RT-PCR analyses showed that prosemin is expressed and secreted from various kinds of cancer cells, such as glioma, pancreas, prostate, and ovarian cell lines. Prosemin is secreted in the cystic fluid of clinical ovarian cancers. Furthermore, immunohistochemistry showed prosemin protein localized in the apical parts of ovarian carcinomas. Recombinant prosemin was expressed in COS cells and was purified by immunoaffinity chromatography. Recombinant prosemin preferentially cleaved benzyloxycarbonyl (Z)-His-Glu-Lys-methylcoumaryl amidide (MCA) and t-butyloxycarbonyl (Boc)-Gln-Ala-Arg-MCA. Our results suggest that prosemin is a novel serine protease of the chromosome 16 cluster that is highly expressed in the pancreas. The usefulness of this serine protease as a candidate tumor marker should be further examined.     

The cholesterol transporter ABCG1 modulates the subcellular distribution and proteolytic processing of beta-amyloid precursor protein

Although intracellular cholesterol levels are known to influence the proteolysis of beta-amyloid precursor protein (APP), the effect of specific genes that regulate cholesterol metabolism on APP processing remains poorly understood. The cholesterol transporter ABCG1 facilitates cholesterol efflux to HDL and is expressed in brain. Notably, the human ABCG1 gene maps to chromosome 21q22.3, and individuals with Down syndrome (DS) typically manifest with Alzheimer's disease (AD) neuropathology in their 30s. Here, we demonstrate that expression of ABCG1 enhances amyloid-beta protein (Abeta) production in transfected HEK cells in a manner that requires functional cholesterol transporter activity. ABCG1-expressing cells also exhibit increased secreted APP (sAPP)alpha and sAPPbeta secretion and display increased cell surface-associated APP. These results suggest that ABCG1 increases the availability of APP as a secretase substrate for both the amyloidogenic and nonamyloidogenic pathways. In vivo, ABCG1 mRNA levels are 2-fold more abundant in DS brain compared with age- and sex-matched normal controls. Finally, both Abeta and sAPPalpha levels are increased in DS cortex relative to normal controls. These findings suggest that altered cholesterol metabolism and APP trafficking mediated by ABCG1 may contribute to the accelerated onset of AD neuropathology in DS.     

A novel activation mechanism of caspase-activated DNase from Drosophila melanogaster

Caspase-activated DNase (CAD) is an enzyme that cleaves chromosomal DNA in apoptotic cells. Here, we identified a DNase in Drosophila Schneider cells that can be activated by caspase 3, and purified it as a complex of two subunits (p32 and p20). Using primers based on the amino acid sequence of the purified proteins, a cDNA coding for Drosophila CAD (dCAD) was cloned. The polypeptide encoded by the cDNA contained 450 amino acids with a calculated M(r) of 52,057, and showed significant homology with human and mouse CAD (22% identity). Mammalian CADs carry a nuclear localization signal at the C terminus. In contrast, dCAD lacked the corresponding sequence, and the purified dCAD did not cause DNA fragmentation in nuclei in a cell-free system. When dCAD was co-expressed in COS cells with Drosophila inhibitor of CAD (dICAD), a 52-kDa dCAD was produced as a heterotetrameric complex with dICAD. When the complex was treated with human caspase 3 or Drosophila caspase (drICE), the dICAD was cleaved, and released from dCAD. In addition, dCAD was also cleaved by these caspases, and behaved as a (p32)(2)(p20)(2) complex in gel filtration. When a Drosophila neuronal cell line was induced to apoptosis by treatment with a kinase inhibitor, both dCAD and dICAD were cleaved. These results indicated that unlike mammalian CAD, Drosophila CAD must be cleaved by caspases to be activated.