| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
COMMENTARY |
a Istituto di Patologia Generale, Università di Firenze, Viale G. B. Morgagni 50, 50134 Firenze, Italy
A review by Gasparini et al. (1) that recently appeared in The FASEB Journal covered many aspects of the validity and perspectives of peripheral markers of Alzheimer's disease (AD). The authors have reaffirmed the role of peripheral tissues, especially skin fibroblasts, as either experimental models or easy sources of biological material to seek for the features that might complement the clinical diagnosis of the disease.
There is, however, a concern arising from the assumption that relevant markers in AD could be limited to (and searched for) alterations in amyloid precursor protein (APP), calcium homeostasis, oxidative metabolism, and transduction systems. Additional items could be of import and, particularly, the regulation of proteolysis should not be neglected due to the close relationships between AD and proteases as inferred by 1) the detection of different types of proteases (2, 3) and inhibitors of both cysteine (4, 5) and serine proteases (6, 7) in senile plaques; 2) intracellular trafficking and proteolytic cleavage of APP (8–10) as well as all the efforts made to identify the still elusive 
-secretase generating amyloidogenic APP fragments; 3) the intriguing occurrence of the serine protease inhibitor domain of the Kunitz type in the APP751 and APP770 isoforms (11–13); and 4) identification of cathepsin B as potential 
-secretase (14). Some of these topics have been covered in a conference that was specifically devoted to the role of proteases and protease inhibitors in AD pathogenesis (15).
Degenerative diseases have often been associated with proteolysis, being responsible for either the amplification of tissue injury by hydrolytic enzymes released from damaged cells or removal of necrotic material. Proteolysis, however, is receiving more and more attention for its role in the processing of polypetptide chains and conversion of precursor molecules into biologically active compounds which are known to be involved in a variety of physiopathological processes. On the other hand, proteases, whose activities depend exclusively on the sequence and conformation of their catalytic sites, may have multiple substrates; therefore, an imbalance of proteolysis could potentially affect several components and be seen at virtually every level of cellular compartments. Among these, the vescicular compartment, which includes endoplasmic reticulum, Golgi apparatus, and lysosomes and endosomes, deserves special mention, because of its complex role in transportation, remodeling, folding, and eventually degradation or recycling of proteins. In theory, any nascent protein chain directed to membranes, extracellular milieu, or that will be packaged in lysosomes could be differently assembled/activated (and have, therefore, distinct fates) depending on the extent and quality of proteolytic cleavage of polypeptides and occurrence of concurrent posttranslational modifications such as glycosylation, sulfation, and phosphorylation.
These general considerations on cellular proteolysis seem especially pertinent to AD since several lines of evidence suggest that the membrane spanning APP molecule might not be the only protein undergoing alterations in the brain (16–19) and the peripheral tissues of Alzheimer patients (1, 20). Regarding strictly AD fibroblasts, a panel of cellular constituents and processes have been proposed as ancillary pathological markers of disease including altered calcium homeostasis (21, 22), potassium channel dysfunction (23), altered glucose utilization and increase of lactate production (24), defective cell spreading and adhesiveness (22, 25), abnormal signal transduction (26, 27), and reduced expression and activity of protein kinase C (28, 29). On the basis of this synthetic list, AD fibroblasts should be truly regarded as abnormal cells. In this respect, it might be surprising that affected patients do not show gross structural and functional alterations of connective tissue and disturbances in wound healing.
Moreover, according to the recent report of De Strooper et al. (30), the presenilin (PS1 and PS2) mutations would alter the processing of APP by serving as activators or regulators of -secretase activity, according to a mechanism resembling that of the SREBP-cleavage activating protein (SCAP) in cholesterol metabolism. As Haass and Selkoe point out (31), however, it is likely that, in addition to APP, other proteins could be incorrectly targeted and abnormally transported through the endoplasmic reticulum and Golgi compartments, thus leading to enhanced ßA4 production. This might explain why AD cells were actually reported to express multiple alterations; moreover, it provides further support for the hypothesis that AD cells might suffer from a proteolytic imbalance due to changes in the amounts or activities of proteases, their incorrect targeting and final destinations, or inappropriate levels of inhibitors. It is worth recalling that the accumulation of a molecular variant of cystatin C, an extracellular thiol-protease inhibitor, is known to be the cause of Icelandic amyloidosis (32).
A clear marker of altered proteolysis in AD fibroblasts has not been demonstrated so far. Nevertheless, there is a general consensus that an imbalance of proteolysis may be involved in the development of the disease. Our data on the abnormal degradation of transketolase (TK) seem to support this hypothesis (33–35). TK is a cytosolic enzyme that has probably nothing to do with AD pathogenesis apart from the fact that it is abnormally and characteristically degraded during extraction of AD fibroblasts. Conversely, TK of fibroblasts from patients with other neurological diseases and control subjects remains unaltered. Further investigation on this early biochemical marker (also some of the asymptomatic relatives of AD patients were exhibiting clear TK alterations) led to the conclusion that lysosomal cathepsins, particularly cysteine proteinases, are involved and could be the more relevant targets to investigate for both pathogenetic and diagnostic purposes. An interesting feature of TK alterations is that they could be abolished by treating AD fibroblasts with cysteine proteinase inhibitors or the acidotropic agent NH4Cl, which impairs the function of the endolysosomal compartment (35). These findings suggest that whatever abnormal proteolysis occurs, it can be modulated and potentially corrected. Whether TK alterations correlate positively with the same proteolytic mechanisms generating ßA4 is still questionable. However, NH4Cl, chloroquine and monensin (36, 37), or more specifically, the cysteine proteinase inhibitor N-acetyl-Leu-Leu-norleucinal (ALLN) (38), have been shown to affect APP metabolism and prevent cells from excess of ßA4 formation. On the other hand, leupeptin was reported to cause the intracellular accumulation of C-terminal APP fragments; this effect, however, could not clearly be addressed to inhibition of either serine or cysteine proteases. On the whole, these data suggest that altered protein processing in the vescicular compartment by the deleterious effects of presenilin mutations could be counteracted in part by inhibitors of cysteine proteases. The role of cathepsins and cystatins in AD has been excellently reviewed by Bernstein et al. (39) and received further support from studies on cathepsin S (40).
Gasparini et al. (1), and many Alzheimer researchers, consider that a single molecular trait could not be sufficient for biochemical AD diagnosis which, in turn, could be more reliably accomplished by complementation of distinct cellular markers. This assumption, however, implies the dysfunction of a rather `nonspecific' process, which might give rise to a variety of phenotypical and molecular effects as the results of metabolic amplification through side pathways. In our view, the abnormal control of proteolysis in AD cells might be the potentially unifying concept from which a panel of apparently unrelated alterations involving calcium regulation, oxidative metabolism, transduction system, and APP processing could be derived. Why resulting cellular and molecular derangements are mild in the majority of peripheral tissues including fibroblasts and so severe to the brain is still to be devised. We have, however, learned from cellular pathology that elementary lesions of cells potentially can produce numerous secondary biochemical alterations that might not be apparently related to the basic pathology of the disease. In addition, the functional and morphologic features of most affected tissues are not always readily pathognomonic of the disease and are sometimes anatomically distant. Knowing this, we see that cultured skin fibroblasts could be a useful model in seeking pathologic hallmarks of AD, but we must pay special attention to dysregulation of intracellular proteolysis as one of the possible pathogenetic mechanisms of the disease.
ACKNOWLEDGMENTS
We are indebted to Prof. E. Bergamini for his friendly suggestions and support. Granted by MURST (ex 40%).
FOOTNOTES
1 Correspondence: E-mail: paoletti{at}ats.it 
REFERENCES
1-antitrypsin and 1-antichymotrypsin are in the lesions of Alzheimer disease. Neuroreport 3, 201–203[Medline]
-trace basic protein (cystatin C). Proc. Natl. Acad. Sci. USA 83, 2974–2978
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
