| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online September 2, 2004 as doi:10.1096/fj.03-1436fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,,2
* Departments of Biochemistry,
Pathology and Molecular Medicine,
The Syl and Molly Apps Research Center, Kingston General Hospital and the
Department of Chemistry, Queen’s University, Kingston, Ontario Canada
1Correspondence: Department of Pathology and Molecular Medicine, Queen’s University, Kingston, Ontario, Canada K7L 3N6. E-mail 3jba1{at}post.queensu.ca
SPECIFIC AIMS
We sought to develop a cell culture model to study the role of heparan sulfate (HS) in amyloidogenesis, a disease in which >20 different polypeptides, including Aß in Alzheimer’s disease and serum amyloid A (SAA) in chronic inflammatory diseases (AA-amyloidosis), refold and assemble to form fibrillar tissue deposits called amyloid. The effect of specifically inhibiting HS:SAA interactions on AA-amyloidogenesis was evaluated as a potentially novel treatment strategy for amyloidoses.
PRINCIPAL FINDINGS
1. J774A.1 cells mediated the conversion of native high density lipoprotein-associated SAA (HDL-SAA) into AA-amyloid deposits
The monocytic cell line J774A.1 when incubated for 7 days with HDL-SAA (0.3 mg/mL) produced significant quantities of amyloid as detected by Congo red (CR) staining and thioflavin T fluorescence (Th-T) (Fig. 1
). Maximum amyloid load (Fig. 1A
) was dependent on pretreatment with amyloid-enhancing factor (AEF), an extract of amyloid-laden tissue containing amyloid protofibrils that are believed to "seed" the process of fibrillogenesis. In its absence, much less amyloid was produced (Fig. 1B
, light green spots; D). The kinetics of amyloid deposition in the cell culture and an experimentally inducible murine model of AA-amyloidosis was monitored for 7 days post-AEF administration by Th-T fluorescence analysis (Fig. 1D
). Both appeared to follow the same kinetics, but based on total protein content, the amyloid load was 
54-fold greater in cell culture than in mouse spleens. The absence of AEF in cell culture delayed by 5 days the production of detectible amyloid (Fig. 1D
, open circles).
|
2. SAA isoform preference and fibril processing are identical to that in vivo
Mice, like humans, express two major isoforms of SAA (SAA1.1 and SAA2.1) during an acute-phase response, but in mice only SAA1.1 is deposited as AA-amyloid. To determine which of the two major SAA isoforms was being converted into AA-amyloid, cells were incubated as described above with AEF and purified SAA1.1 or SAA2.1 at 50 µg/mL (equivalent to their respective concentrations in HDL-SAA at 0.3 mg/mL), HDL-SAA, or HDL reconstituted with one or the other purified SAA isoform. Analysis by Th-T fluorescence showed that only SAA1.1 produced amyloid but in much reduced quantity (9% of HDL-SAA), whereas no amyloid was detected with SAA2.1. When purified SAA1.1 or 2.1 were reassociated with HDL, the resulting amyloid load with the reconstituted HDL-SAA1.1 was close to the amount assayed for native HDL-SAA. Reconstituted HDL-SAA2.1 produced no amyloid. AA-amyloid fibrils in vivo are normally composed of a set of peptides spanning the amino two-thirds of SAA1.1. Western blotting analysis using anti-SAA antibody showed that the proteolytic fragmentation of SAA1.1 appeared to be identical between amyloid-laden cell culture and mice spleens.
3. HS, but not chondroitin or dermatan sulfate, co-deposit with amyloid
Of the six major glycosaminoglycans (GAGs)—dermatan sulfate, chondroitin sulfate, keratan sulfate, heparin, HS, and hyauronan acid—only HS has been shown to be a universal component of amyloids. Hence we determined which GAG, if any, was associated with the cell culture amyloid. Intense staining of cell culture amyloid was observed with sulfated Alcian blue (SAB), indicating a high content in sulfated-GAGs. Amyloid deposits in which HS was eliminated enzymatically did not stain with SAB, although the residual amyloid deposits could still be discerned. Specific digestion of chondroitin and dermatan sulfate showed no change in staining intensity.
4. Heparin and polyvinylsulfonate inhibit amyloid deposition
Both native heparin (Mr 12,000) and low molecular weight heparin (Mr 3000) could inhibit amyloidogenesis in cell culture, with only the latter being well tolerated by the cells at higher concentrations. Chondroitin sulfate was unable to affect amyloid deposition at any concentration tested. Polyvinylsulfonate (PVS), a low molecular weight anionic polymer containing structural features similar to sulfated GAGs, could achieve 50% and 100% inhibition (IC50 and IC100) at 0.5 and 9 µM, respectively. The anti-amyloid property of PVS has been demonstrated in vivo.
5. A synthetic peptide corresponding to SAA1.1’s HS binding site is highly anti-amyloidogenic
We previously identified a heparin/HS binding site on the carboxyl-terminal end of SAA1.1 (77-ADQEANRHGRSGKDPNYYRPPGLPAKY-103) and postulated that an interaction between this 27-mer sequence and HS promotes SAA1.1 fibrillogenesis. To test this, we attempted to block SAA1.1:HS interaction during amyloidogenesis by incubating the cells with increasing concentrations of a synthetic 27-mer (Fig. 2
). Subsequent quantitation of the amyloid loads revealed that the 27-mer was a profound inhibitor of amyloidogenesis, with an IC50 of 0.02 µM, 25-fold lower than for PVS. This effect was sequence specific. Scrambling 27-mer’s sequence (PLPAQGKPGPDHYARNDSYAKNRYERG) or replacing residues R83, H84, and R86 with A, which destroys HS binding, caused a complete loss of inhibitory activity. The 27-mer did not interfere with HDL-SAA binding to cells, suggesting that the amyloidogenic pathway was being affected specifically. Unlike the 27-mer, both heparin and PVS prevented HDL-SAA binding to cells, which is likely responsible for their anti-amyloid activities.
|
CONCLUSIONS
Mononuclear phagocytes (macrophage, microglia, Kupffer cells) have long been suspected of being directly involved in amyloidogenesis. They are often found in close proximity to amyloid deposits as in AA-amyloid, ß-2-microglobulin amyloid (associated with chronic renal dialysis), and Aß amyloid, prion amyloid (i.e., Scrapies). Aß and AA-amyloid fibrils have also been observed in cytoplasmic vesicles of macrophage, although it was unclear if fibril assembly or degradation was taking place. However, a transformed peritoneal-macrophage cell line (IC-21 cells) has been reported to produce AA-amyloid when cultured for up to 2 wk with AEF and bacterially expressed delipidated SAA. We have also observed that selective reduction of splenic macrophage in mice by treatment with clodronate (dichloromethylene diphosphonate) greatly impeded AA-amyloidogenesis (unpublished data). Thus, we chose a well-characterized monocytic cell line to develop a physiologically relevant model for amyloidosis that would allow us to investigate the mechanism of amyloidogenesis in greater detail and aid in identifying effective amyloid inhibitors.
SAA as a pure, delipidated protein poses significant experimental difficulties because it is extremely hydrophobic and prone to denaturation and aggregation in solution. To overcome this problem, cells were incubated with SAA in its native state/conformation associated with HDL. We found that native HDL-SAA or reconstituted HDL-SAA1.1 produced 11-fold more amyloid than delipidated SAA. This suggested that the preamyloidogenic conformation of SAA is critical for refolding and assembly into fibrils and that a lipid microenvironment was necessary for this process. There is evidence that these factors both play a role in the development of other types of amyloid as well. The degree of denaturation of TTR (transthyretin; familial amyloidotic polyneuropathy) and insulin greatly influences the efficiency of their fibrillogenesis with partially denatured proteins, generating appreciably more fibrils than more extensively denatured intermediates. Fibrillogenesis of Aß and prion protein occurs optimally on bilayer lipid surfaces or in cholesterol-rich microdomains.
AA-amyloid deposition in cell culture follows the same nucleation-dependent kinetics observed in mouse spleens, suggesting that the major steps in AA-amyloidogenesis, including SAA dissociation from HDL, refolding, assembly, proteolytic processing, and the co-deposition of HS with AA-amyloid, could all be mediated by monocytic cells. The rate of amyloid deposition relative to total protein was 54-fold greater in cell culture than in the spleens, again indicating that the important factors required for amyloidogenesis were contained within this cell culture system. Heparin and the sulfonate PVS have been reported to prevent amyloid formation in mice and were effective in culture. Both blocked HDL-SAA binding to macrophage probably by competing for cell surface HS binding sites. The enzymatic removal of cell surface HS has been shown to prevent HDL and HDL-SAA binding to macrophage.
More direct evidence that SAA:HS interactions play a critical role in AA-amyloidogenesis was provided by a peptide (27-mer) corresponding to the HS binding site of SAA 1.1. It was found to be a potent inhibitor of amyloidogenesis with an IC50 of 20 nM, which is 25-fold lower than that of PVS (IC50=500 nM). Surprisingly, unlike heparin and PVS, the 27-mer did not appear to interfere with normal binding of HDL-SAA to macrophages, for which there are up to 55,000 binding sites/cell. Thus, it seems likely that the 27-mer’s inhibitory activity targets the amyloidogenic pathway downstream of the cell surface binding event (Fig. 3
).
|
Other reports have described the ability of short peptides to inhibit fibrillogenesis in vitro for Aß and IAPP (islet amyloid precursor polypeptide forms amyloid in 95% of type II diabetics). In a cell-free system, an Aß peptide (residues 16-20) at 100 µM has been shown to associate with Aß1-40 and prevent fibril assembly. Short IAPP peptides (residues 20-25 and 24-29) at a 10-fold molar excess (100 µM) over IAPP could also reduce the amyloid loads in vitro by 80–85%. In our assay, a similar level of inhibition could be achieved with 1400-fold less 27-mer (70 nM), which is 60-fold less than the SAA1.1 concentration (4.2 µM) used to generate AA-amyloid in culture.
The concentrations of IAPP and Aß used in these studies ranged from 10 to 100 µM, which is far in excess of their normal physiological concentrations. Even in AD brains, Aß remains in the nanomolar range. In fact, it has been reported that little or no Aß fibrillogenesis can take place below 14 µM. However, HS is known to promote fibrillogenesis of Aß and IAPP in vitro, and plaque-rich tissue extracts that contain HS could induce fibril assembly of nM concentrations of Aß. Our data clearly demonstrate that amyloid polypeptide:HS interactions are fundamental to amyloidogenesis, and suggest that peptides corresponding to the amyloid precursor HS binding site may block this interaction and be an effective strategy in preventing these devastating amyloid-based diseases.
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.03-1436fje;
This article has been cited by other articles:
|
|
 |
|
  I. J. Morten, W. S. Gosal, S. E. Radford, and E. W. Hewitt Investigation into the Role of Macrophages in the Formation and Degradation of beta2-Microglobulin Amyloid Fibrils J. Biol. Chem., October 5, 2007; 282(40): 29691 - 29700. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-P. Li, M. L. E. Galvis, F. Gong, X. Zhang, E. Zcharia, S. Metzger, I. Vlodavsky, R. Kisilevsky, and U. Lindahl In vivo fragmentation of heparan sulfate by heparanase overexpression renders mice resistant to amyloid protein A amyloidosis PNAS, May 3, 2005; 102(18): 6473 - 6477. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Barret, L. Forestier, J.-P. Deslys, R. Julien, and P. F. Gallet Glycosylation-related Gene Expression in Prion Diseases: PrPSc ACCUMULATION IN SCRAPIE INFECTED GT1 CELLS DEPENDS ON {beta}-1,4-LINKED GalNAc-4-SO4 HYPOSULFATION J. Biol. Chem., March 18, 2005; 280(11): 10516 - 10523. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
