|
|
||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online December 4, 2003 as doi:10.1096/fj.03-0730fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

,
,||,2
* Departments of Pathology,
Psychiatry and
Pharmacology, New York University School of Medicine, New York, New York, USA;
Indiana Alzheimer Disease Center, Indiana University School of Medicine, Indianapolis, Indiana, USA; and
|| Nathan Kline Institute, New York University School of Medicine, Orangeburg, New York, USA
2Correspondence: Nathan Kline Institute, New York University School of Medicine, 140 Old Orangeburg Rd., Orangeburg, NY 10962, USA. E-mail: elevy{at}nki.rfmh.org
SPECIFIC AIMS
Amyloid peptides deposited in familial British and Danish neurodegenerative disorders (FBD and FDD, respectively) are generated by proteolytic processing of mutant forms of the precursor protein BRI2, ABriPP, and ADanPP. We have investigated endoproteolytic processing, subcellular localization, and normal biological function of wild-type and mutant forms of BRI2.
PRINCIPAL FINDINGS
1. Proteolytic processing of BRI2
Tissue culture cell lines were transiently transfected with amino-terminal myc epitope-tagged BRI2(wild-type British cDNA), BRI2-B (British variant of BRI2), or BRI2-D (Danish variant of BRI2) cDNAs. Western blot and metabolic labeling analyses with anti-myc antibody of cell lysates of N2a cells transfected with BRI2 revealed three bands of
44 kDa, 41 kDa, 17 kDa and
45 kDa, 41 kDa, and 17 kDa in BRI2-B or BRI2-D transfected cells (Fig. 1
A). These bands correspond to full-length BRI2 and two amino-terminal fragments. The 41 kDa fragment is the amino-terminal product of a described cleavage between amino acid 243 and 244. This cleavage results in the production of carboxyl-terminal
3 kDa wild-type peptide or
4 kDa ABri or ADan peptides (British or Danish amyloid peptides). The 17 kDa band was not recognized using antibodies specific for the carboxyl terminus of ABriPP and ADanPP proteins (Fig. 1B
). This fragment is the product of an additional cleavage in BRI2 amino-terminal to the site, resulting in production of the 41 kDa fragment. A similar processing pattern was observed in cell lines HEK293, NIH3T3, CHO-K1, HepG2, LoVo, and SK-N-SH.
|
2. The temporal profile of BRI2 expression, processing, and secretion
Immunoprecipitation with anti-myc antibody of cell lysate proteins labeled with 35S-cysteine showed that cleavage of the full-length protein occurs in 10 min of labeling to produce a 41 kDa fragment; a 17 kDa product is produced within 30 min of labeling. Both processing products accumulate intracellularly thereafter. These products were not detected in culture media. The 4 kDa fragment precipitated with antibodies to the carboxyl terminus of either ABriPP or ADanPP is seen in cell lysates in 10 min of labeling and is secreted into the cultured medium in
1 h of chase. We were unable to detect bands of 27 or 28 kDa, suggesting that the 17 kDa peptide is produced by further proteolytic cleavage of the 41 kDa peptide.
3. Cellular localization of BRI2 processing
Metabolic labeling experiments in the absence or presence of inhibitors were performed to determine the intracellular location of BRI2 processing. Brefeldin A causes accumulation of full-length ABriPP and inhibition of production of the amino-terminal processing products of 41 and 17 kDa (Fig. 1A
) and of the carboxyl-terminal 4 kDa ABri (Fig. 1B, C
). Brefeldin A blocks protein transport from the endoplasmic reticulum and causes redistribution of Golgi components to the endoplasmic reticulum. Monensin, which inhibits transport from the medial-Golgi to trans-Golgi, does not affect ABriPP processing but inhibits secretion of ABri (Fig. 1B, C
). The furin inhibitor decanoyl-Arg-Val-Lys-Arg-chloromethyl ketone does not affect BRI2 processing (Fig. 1A-C
). These results demonstrate that both proteolytic processing reactions of BRI2 occur in the cis- or medial-Golgi apparatus independent of the presence of the protease furin.
4. Intracellular transport of ABri via a regulated secretory pathway into secretory vesicles in neuronal cells
Experiments for the intracellular localization of BRI2 were carried out using indirect immunofluorescence. Staining of N2a cells transiently transfected with BRI2, BRI2-B, or BRI2-D cDNAs with an anti-amino-terminal antibody revealed cytoplasmic staining. Staining with antibodies to the carboxyl terminus showed cytoplasmic staining and punctate staining along the cell process that colocalized with carboxypeptidase E staining. These immunofluorescence microscopic observations demonstrate that full-length and amino-terminal fragments of wild-type and mutant forms of BRI2 are restricted to the cell soma and that only the carboxyl-terminal fragment is secreted via the regulated secretory pathway.
5. Expression of BRI2 promotes neurite outgrowth in N2a cells
N2a cells that do not express BRI2 mRNA were stably transfected with BRI2, BRI2-B, BRI2-D cDNA, or vector alone. Cells expressing any BRI2 cDNA had longer neurites than N2a cells transfected with vector alone. Measurements of neurite lengths showed a significant induction of elongation of neurites (Fig. 2
).
|
CONCLUSIONS AND SIGNIFICANCE
Proteolytic processing of BRI2 plays a crucial role in the pathological progress of FBD and FDD. We have studied the processing, subcellular localization, and secretion of processing products of BRI2 using cultured cell lines. Two cleavage sites were identified resulting in 1) an intracellular 41 kDa amino-terminal peptide and the corresponding 3 to 4 kDa carboxyl-terminal fragment secreted into the cultured media and 2) a novel 17 kDa amino-terminal fragment. A similar pattern was observed in a variety of cell lines. Using intracellular transport inhibitors and immunofluorescence microscopy, we demonstrated that both cleavages occur in cis- or medial-Golgi; whereas the amino-terminal fragments accumulate intracellularly, carboxyl-terminal processing products are secreted via a regulated secretory pathway. Finally, we demonstrated induction of neurite outgrowth in neuronal cells by BRI2 expression.
It has been demonstrated that furin cleaves the APLP1 epitope-tagged BRI2 and the associated variants ABriPP and ADanPP between amino acids 243 and 244 and that this proteolytic activity was not present in furin-deficient RPE.40, a CHO derived cell line. Our data indicate that the proteolytic processing occurs in LoVo cells that lack the endogenous furin processing activity. We demonstrate that the proteolytic processing of BRI2, ABriPP, and ADanPP occurs in the cis- or medial-Golgi and not in the trans-Golgi network where furin activity is predominantly localized. The proteolytic activity was not inhibited by adding a furin inhibitor or by adding monensin, an inhibitor of the transport from the medial-Golgi to the trans-Golgi. Thus although furin is able to cleave BRI2, a proteolytic enzyme other than furin is involved in cleavage of the carboxyl terminus of BRI2. This enzyme exhibits a ubiquitous cell-type expression, including in N2a cells that do not express BRI2 mRNA.
It was recently shown that ABri contains a disulfide bond. A form of ABri was identified in serum of FBD patients using immunoprecipitation, mass spectrometry, and Western blot analysis. A peptide of a mass consistent with ABri peptide with oxidized cysteine residues was found. The results suggested that some molecules might contain a single intrachain disulfide bond. ABri peptide with the same molecular mass was identified in conditioned medium of N2a cells transfected with BRI2-B cDNA. Disulfide-bonded loops in proproteins and neuropeptides were shown to be essential for sorting of peptides from the trans-Golgi network to a regulated secretory pathway. Immunofluorescence staining of N2a cells transfected with BRI2-B cDNA revealed colocalization of ABri epitopes with carboxypeptidase E in the axons of N2a cells, suggesting transport of ABriPP carboxyl-terminal fragments via a regulated secretory pathway. Cys5 and Cys22 in Bri may be playing a role in Bri sorting into the regulated secretory pathway (Fig. 3
). These two cysteine residues are conserved in all three BRI isoforms.
|
Although the neuropathology of FBD and FDD is similar to that of Alzheimers disease (AD), several differences exist. Both exhibit vascular and parenchymal amyloid, preamyloid deposition, and neurofibrillary tangles in neurodegenerating neurons. In AD, parenchymal Aß deposits are mainly present in the hippocampus and cerebral cortex, and Aß vascular lesions are predominantly seen in leptomeningeal and cortical vessels. Degenerating neurons containing neurofibrillary tangles are frequent in limbic areas and the neocortex. In FBD, severe amyloid angiopathy with perivascular plaque formation occurs throughout the central nervous system. Systemic deposition of amyloid was observed in FBD patients. The presence of soluble amyloid peptides in plasma is similar in FBD and AD, although plasma levels of soluble ABri are higher than soluble Aß. Immunohistochemical analysis of FBD and FDD brain sections demonstrated the presence of proteins of the complement system and their proinflammatory activation products associated with AD lesions. Similar to AD, the chronic inflammatory response generated by the amyloid peptides in vivo might play a role in the pathogenesis of FBD and FDD. Despite the similarities in the neurodegenerative disorders, the sequence and structure of the secreted peptides that form amyloid fibrils (ABri in FBD and ADan in FDD) are unrelated to those of Aß in AD. Thus, amyloid formation and the resultant neurodegeneration do not depend on the primary structure of the peptide or on cellular transport and production of the peptide. This may indicate that a still unknown factor/s induce the same pathological pathways leading to similar disorders.
Despite increasing evidence for a pathogenic role for ABriPP and ADanPP in FBD and FDD, respectively, the physiological function of the protein has remained unknown. Significant amino acid sequence identity exists between the carboxyl-terminal hydrophobic domains of the various BRI2 isoforms. The three proteins have a conserved sequence (KR) before the predicted cleavage site, suggesting a bioactivity for the secreted peptides. We have shown that stable expression of BRI2 in N2a cells that do not express endogenous BRI2 mRNA stimulate neurite outgrowth; thus, the BRI2 protein may have a role in differentiation of neuronal cells.
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.03-0730fje ![]()
This article has been cited by other articles:
![]() |
J. Kim, V. M. Miller, Y. Levites, K. J. West, C. W. Zwizinski, B. D. Moore, F. J. Troendle, M. Bann, C. Verbeeck, R. W. Price, et al. BRI2 (ITM2b) Inhibits A{beta} Deposition In Vivo J. Neurosci., June 4, 2008; 28(23): 6030 - 6036. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Martin, R. Fluhrer, K. Reiss, E. Kremmer, P. Saftig, and C. Haass Regulated Intramembrane Proteolysis of Bri2 (Itm2b) by ADAM10 and SPPL2a/SPPL2b J. Biol. Chem., January 18, 2008; 283(3): 1644 - 1652. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Fotinopoulou, M. Tsachaki, M. Vlavaki, A. Poulopoulos, A. Rostagno, B. Frangione, J. Ghiso, and S. Efthimiopoulos BRI2 Interacts with Amyloid Precursor Protein (APP) and Regulates Amyloid {beta} (A{beta}) Production J. Biol. Chem., September 2, 2005; 280(35): 30768 - 30772. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |