Slower clearance of human SAA1.5 in mice: implications for allele specific variation of SAA concentration in human

Characterization of reconstituted high-density lipoprotein particles formed by lipid interactions with human serum amyloid A

The acute-phase human protein serum amyloid A (SAA) is enriched in high-density lipoprotein (HDL) in patients with inflammatory diseases. Compared with normal HDL containing apolipoprotein A-I, which is the principal protein component, characteristics of acute-phase HDL containing SAA remain largely undefined. In the present study, we examined the physicochemical properties of reconstituted HDL (rHDL) particles formed by lipid interactions with SAA. Fluorescence and circular dichroism measurements revealed that although SAA was unstructured at physiological temperature, α-helix formation was induced upon binding to phospholipid vesicles. SAA also formed rHDL particles by solubilizing phospholipid vesicles through mechanisms that are common to other exchangeable apolipoproteins. Dynamic light scattering and nondenaturing gradient gel electrophoresis analyses of rHDL after gel filtration revealed particle sizes of approximately 10nm, and a discoidal shape was verified by transmission electron microscopy. Thermal denaturation experiments indicated that SAA molecules in rHDL retained α-helical conformations at 37°C, but were almost completely denatured around 60°C. Furthermore, trypsin digestion experiments showed that lipid binding rendered SAA molecules resistant to protein degradation. In humans, three major SAA1 isoforms (SAA1.1, 1.3, and 1.5) are known. Although these isoforms have different amino acids at residues 52 and 57, no major differences in physicochemical properties between rHDL particles resulting from lipid interactions with SAA isoforms have been found. The present data provide useful insights into the effects of SAA enrichment on the physicochemical properties of HDL.

Influence of polymorphism on glycosylation of serum amyloid a4 protein

Serum amyloid A4 (SAA4) is a constitutive apolipoprotein of high-density lipoprotein. It exhibits N-linked glycosylation in its second half. There are both glycosylated and nonglycosylated forms in plasma and the ratio of these two forms varies among individuals. This study was conducted to examine the influence of genetic polymorphism of SAA4 on its glycosylation status. In 55 healthy subjects, SAA4 polymorphism was analyzed by PCR combined direct sequencing and its glycosylation status was analyzed by immunoblotting. The results showed that the percentage of glycosylation in subjects with amino acid substitutions at positions 71 and/or 84 was significantly (P < 0.05) higher than that in subjects with the wild type. The polymorphism had no influence on the plasma concentration of SAA4. These findings suggest that the changes in protein structures alter the efficiency of glycosylation in the SAA4 molecule. The functional implication of this should be of interest.

Effect of amino acid variations in the central region of human serum amyloid A on the amyloidogenic properties

Human serum amyloid A (SAA) is a precursor protein of the amyloid fibrils that are responsible for AA amyloidosis. Of the four human SAA genotypes, SAA1 is most commonly associated with AA amyloidosis. Furthermore, SAA1 has three major isoforms (SAA1.1, 1.3, and 1.5) that differ by single amino acid variations at two sites in their 104-amino acid sequences. In the present study, we examined the effect of amino acid variations in human SAA1 isoforms on the amyloidogenic properties. All SAA1 isoforms adopted α-helix structures at 4°C, but were unstructured at 37°C. Heparin-induced amyloid fibril formation of SAA1 was observed at 37°C, as evidenced by the increased thioflavin T (ThT) fluorescence and β-sheet structure formation. Despite a comparable increase in ThT fluorescence, SAA1 molecules retained their α-helix structures at 4°C. At both temperatures, no essential differences in ThT fluorescence and secondary structures were observed among the SAA1 isoforms. However, the fibril morphologies appeared to differ; SAA1.1 formed long and curly fibrils, whereas SAA1.3 formed thin and straight fibrils. The peptides corresponding to the central regions of the SAA1 isoforms containing amino acid variations showed distinct amyloidogenicities, reflecting their direct effects on amyloid fibril formation. These findings may provide novel insights into the influence of amino acid variations in human SAA on the pathogenesis of AA amyloidosis.

Differential affinity of serum amyloid A1 isotypes for high-density lipoprotein

Serum amyloid A (SAA), a precursor of reactive amyloid deposits, is a multigene product. SAA1, predominant both as an amyloid precursor and in plasma, consists of three allelic variants (SAA1.1, SAA1.3, and SAA1.5). Several investigations have shown that the SAA1.3 allele is associated with susceptibility to AA-amyloidosis in Japanese, and the SAA1.5 allele is related with higher serum concentrations of SAA. However, these results have not been interpreted functionally. This study assessed the affinity of SAA isotypes for high-density lipoprotein (HDL), to which SAA binds in plasma. Using a surface plasmon resonance-based apparatus (BIAcore), the affinity between immobilized recombinant human SAAs and HDL was determined. The SAA concentration was measured in fractions after ultracentrifugation (d = 1.23) of sera from patients with rheumatoid arthritis, whose SAA1 genotypes were determined. In the BIAcore analysis, as the dissociation reaction under the conditions used was too rapid to fit the typical kinetic model, the steady-state affinity model was used. The affinity (kd) of SAA1.1, SAA1.3, and SAA1.5 for HDL was 1.4 x 10(-5), 1.8 x 10(-5), and 3.7 x 10(-6), respectively. rSAA1.5 showed significantly (p < 0.05) stronger affinity than the other two. The fraction of lipid-free SAA in serum was significantly (p < 0.001) lower in the patients with larger numbers of the 1.5 allele at the SAA1 locus. These results suggest that the relatively high affinity of SAA1.5 may cause the high serum concentration and may be related to the low susceptibility to amyloidosis.

Increased susceptibility of serum amyloid A 1.1 to degradation by MMP-1: potential explanation for higher risk of type AA amyloidosis

OBJECTIVE

Genetic polymorphisms in serum amyloid A (SAA) have been shown to substantially influence the risk of developing type AA amyloidosis. Recently, a role for MMP-1 has been suggested in the pathogenesis of AA amyloidosis. Therefore, we investigated if the SAA1 isotypes are differentially degraded by MMP-1.

METHODS

Degradation of different SAA isotypes by MMP-1 was assessed by immunoblotting. MALDI-TOF mass spectrometry was used to identify degradation fragments.

RESULTS

We found that SAA1.5 is more resistant to degradation by MMP-1 than SAA1.1. This difference is caused by the capacity of MMP-1 to cleave at the site of the polymorphism at position 57.

CONCLUSION

These results may explain the higher risk of amyloidosis in patients with a SAA1.1/1.1 genotype vs SAA1.5/1.5 or SAA1.1/1.5 genotype. In addition, the impaired degradation of SAA1.5 by MMP-1 could also explain the higher serum SAA concentrations in persons with a SAA1.5 genotype.

Measurement of serum amyloid A1 (SAA1), a major isotype of acute phase SAA

Serum amyloid A (SAA), a plasma precursor of reactive amyloid deposits, is a multigene product. SAA1 and SAA2, with primary structures that are 93% identical (98 of 104 amino acids), behave as acute phase proteins, as demonstrated by their increasing levels in plasma. Heretofore, it has been understood that SAA1 predominates and functions as an isotype in plasma. However, accurate measurements differentiating the two isotypes have not been reported. In this study, using monoclonal antibodies specific for SAA1, we developed an enzyme-linked immunosorbent assay (ELISA) for SAA1. The levels and ratios of SAA1 in total SAA (TSAA) were investigated in healthy subjects and patients with rheumatoid arthritis (RA). The SAA1/TSAA ratio was 74 +/- 12% and 77 +/- 12% in healthy subjects and RA patients, respectively. In RA patients, the ratios were not influenced by SAA1 genotype, which has been proposed to affect plasma SAA values. The kinetics of SAA1 in inflamed patients undergoing hemodialysis was found to be parallel with total SAA and C-reactive protein. Finally, this study confirmed that SAA1 is a major isotype of acute phase SAA and may determine total SAA values. This specific assay could be used in the evaluation of SAA behavior in several clinical conditions.