Formation of immunoglobulin light chain amyloid oligomers in primary cutaneous nodular amyloidosis

BACKGROUND

Primary cutaneous nodular amyloidosis (PCNA) is thought to be a plasma cell dyscrasia. The amyloid deposits are found in the dermis and subcutis, and they contain clonal immunoglobulin light chains, produced by a local proliferation of plasma cells. New insights into amyloid diseases have revealed that the pathology is due more to the presence of small, misfolded protein species termed oligomers than to the deposition of fibrillar material.

OBJECTIVES

To demonstrate the presence of amyloid oligomers in PCNA and to provide evidence that cutaneous amyloid diseases share a common pathogenic pathway similar to other amyloid diseases.

METHODS

Immunohistochemical staining with conformation-specific and sequence-specific antibodies was used to localize different amyloid species of light chain immunoglobulins in a case of PCNA. Additionally, in vitro characterization of immunoglobulin oligomers and fibrils was performed to determine, through toxicity studies in a human keratinocyte cell line, which amyloidogenic form of the immunoglobulin is toxic in PCNA.

RESULTS

Amyloid oligomers were identified in PCNA. Oligomers were mainly formed by lambda light chain immunoglobulins, and kappa light chain oligomers were detected in lesser amounts. Amyloid species were detected intra- and extracellularly. In addition, amyloid oligomers and fibrils, derived from unknown protein sources, were detected. This finding suggests that immunoglobulin amyloids can act as seeds capable of inducing the aggregation of heterogeneous proteins in the skin. Furthermore, cytotoxicity studies demonstrated that immunoglobulin oligomers, but not monomers or fibrils, are toxic to human keratinocytes.

CONCLUSIONS

These data indicate that PCNA has common pathways with other amyloid diseases with respect to protein misfolding and pathogenesis. Immunoglobulin oligomers may prove to be targets for the treatment of PCNA.

Isolation, structural, and functional characterization of an apoptosis-inducing L-amino acid oxidase from leaf-nosed viper (Eristocophis macmahoni) snake venom

The enzyme L-amino acid oxidase (LAO) from the leaf-nosed viper (Eristocophis macmahoni) snake venom was purified to homogeneity in a single step using high performance liquid chromatography on a Nucleosil 7C18 reverse phase column. The molecular mass of the purified enzyme was 58734.0 Da, as determined by matrix-assisted laser desorption/ionization mass spectrometry. The N-terminal amino acid sequence (ADDKNPLEEAFREADYEVFLEIAKNGL) and the chemical composition of the purified LNV-LAO shows close structural homology with other L-amino acid oxidases isolated from different snake venoms. The secondary structural contents analysis of LAO, established by means of circular dichroism, revealed ca. 49% alpha-helix, 19% beta-sheet, 10% beta-turn, and 22% random coil structure. The purified LNV-LAO not only retained its specific enzymatic activity (73.46 U/mg), determined against L-leucine as a substrate, but also exhibited potent haemolytic (1-10 microg/ml), edema- (MED 4.8 microg/ml) and human platelet aggregation-inducing (ED50 33 microg/ml) properties. Unlike other haemorrhagic snake venom L-amino acid oxidases, the LNV-LAO does not produce haemorrhage. In addition to these local effects, the purified LNV-LAO showed apoptosis-inducing activity in the MM6 cell culture assay. After 18 h treatment with 25-100 microg/ml of LAO, the typical DNA fragmentation pattern of apoptotic cells was observed by means of fluorescent microscopy and agarose gel electrophoresis.

Oxygen transport proteins: III. Structural studies of the scorpion (Buthus sindicus) hemocyanin, partial primary structure of its subunit Bsin1

The hemocyanin (Hc) from Buthus sindicus, studied in the native state, demonstrated to be an aggregate of eight different types of subunits arranged in four cubic hexamers. Both, the 'top' and the 'side' views of the native molecule have been identified from the negatively stained specimens using transmission electron microscopy. Out of these, eight different polypeptide chains, the partial primary structure (68%) of a subunit Bsin1 (Mr = 72422.7 Da) was established using a combination of automated Edman degradation and mass spectrometry. A multiple sequence alignment with other closely related cheliceratan Hc subunits revealed average identities of ca. 60%. Most of the structurally important residues, i.e. copper and calcium-binding ligands, as well as the residues involved in the presumed oxygen entrance pathway, proved to be strictly conserved in Bsin1. Sequence variations have been observed around the functionally important chloride-binding site, not only for the B. sindicus subunit Bsin1, but also for the subunit Aaus-6 of the scorpion A. australis and the subunit Ecal-a from the spider Eurypelma californicum Hcs. Deviation in the primary structure related to the chloride-binding site suggest that the effect of chloride ions may vary in different hemocyanins. Furthermore, the secondary structural contents of the Hc subunit Bsin1 were determined by circular dichroism revealing ca. 33% alpha-helix, 18%, beta-sheet, 19% beta-turn, and 30% random coil composition. These values are in good agreement with the crystal structure of the closely related Hc subunit Lpol-II from horseshoe crab L. polyphemus. Electron microscopic studies of the purified Hc subunit under native conditions revealed that Bsin1 has self aggregation properties. Results of these studies are discussed.

Rational design, conformational studies and bioactivity of highly potent conformationally constrained calcitonin analogues

Calcitonin is known for its hypocalcaemic effect and the inhibition of bone resorption, and is used therapeutically for the treatment of osteoporosis and Paget's disease. Our studies on the conformational features of human calcitonin (hCt) bioactivity have led to the conformationally constrained hCt analogue cyclo17,21-[Asp17, Lys21]hCt (1), which had a 5-10 times higher in vivo hypocalcaemic potency than hCt [Kapurniotu, A. & Taylor, J.W. (1995) J. Med. Chem. 38, 836-847]. We hypothesized that a stabilized, possibly type I beta turn/beta sheet conformation between residues 17 and 21 could play a crucial role in hCt bioactivity. Here, we designed, synthesized and studied the conformation and bioactivity of 19-member to 17-member ring-size analogues of 1 with the structure cyclo17,21-[Asp17,XX21]hCt with XX = Orn (2), Dab (3) and Dap (4), of the control peptide [Asp17,Orn21]hCt (5), and of the 19-member cyclo17,21-[Glu17,Dab21]hCt (6). Analyses of the far-UV CD spectra indicated increased type I beta turn and antiparallel beta sheet content in the bicyclic analogues compared with hCt. In the in vivo hypocalcaemic assay, cyclo17,21-[Asp17,Orn21]hCt (2) was found to have a 400-fold higher potency than hCt and was fourfold more potent than salmon calcitonin (sCt), which has been the most potent known Ct. Analogue 3 had a 30-fold higher potency than hCt, whereas the highly constrained analogue 4 was as potent as hCt. Bioactivity was not enhanced for the nonbridged compound [Asp17, Orn21]hCt (5), whereas cyclo17,21-[Glu17,Dab21]hCt (6) showed the same bioactivity as 1. This study identifies 2 as exhibiting the highest in vivo potency among currently known Cts, while it differs in only one amino acid residue from hCt, strongly suggesting that the introduced constraint may have served in 'freezing' hCt in a bioactive conformation. Our findings provide evidence for the first time that a beta turn/beta sheet conformation in region 17-21 of hCt and the topological features of the side chain of Asn17 are strongly associated with in vivo bioactivity, and offer a novel lead structure for a hCt-based drug for the treatment of osteoporosis and other bone-disorder-related diseases.

Conformational transitions of islet amyloid polypeptide (IAPP) in amyloid formation in vitro

Amyloid aggregates have been recognized to be a pathological hallmark of several fatal diseases, including Alzheimer's disease, the prion-related diseases, and type II diabetes. Pancreatic amyloidosis is characterized by the deposition of amyloid consisting of islet amyloid polypeptide (IAPP). We followed the steps preceding IAPP insolubilization and amyloid formation in vitro using a variety of biochemical methods, including a filtration assay, far and near-UV circular dichroism (CD) spectropolarimetry, 1-anilino-8-naphthalenesulfonic acid (ANS) binding, and atomic force (AFM) and electron (EM) microscopy. IAPP insolubilization and amyloid formation followed kinetics that were consistent with the nucleation-dependent polymerization mechanism. Nucleation of IAPP amyloid formation with traces of preformed fibrils induced a rapid conformational transition into beta-sheets that subsequently aggregated into insoluble amyloid fibrils. Transition proceeded via a molten globule-like conformeric state with large contents of secondary structure, fluctuating tertiary and quaternary aromatic interactions, and strongly solvent-exposed hydrophobic patches. In the temperature denaturation pathway at 5 microM peptide, we found that this state was mostly populated at about 45 degrees C, and either aggregated rapidly into amyloid by prolonged exposure to this temperature, or melted into denaturated but still structured IAPP, when heated further to 65 degrees C. The state at 45 degrees C was also found to be populated at 4.25 M GdnHCl at 25 degrees C during GdnHCl-induced equilibrium denaturation, and was stable in solution for several hours before aggregating into amyloid fibrils. Our studies suggested that this amyloidogenic state was a self-associated form of an aggregation-prone, partially folded state of IAPP. We propose that this partially folded population and its self-associated forms are in a concentration-dependent equilibrium with a non-amyloidogenic IAPP conformer and may act as early, soluble precursors of beta-sheet and amyloid formation. Our findings on the molecular mechanism of IAPP amyloid formation in vitro should assist in gaining insight into the pathogenesis and inhibition of pancreatic amyloidosis and other amyloid-related diseases.

Purification and primary structure of low molecular mass peptides from scorpion (Buthus sindicus) venom

The primary structures of four low molecular mass peptides (Bs 6, 8, 10 and 14) from scorpion Buthus sindicus were elucidated via combination of Edman degradation and matrix-assisted laser desorption ionization mass spectrometry. Bs 8 and 14 are cysteine-rich, thermostable peptides composed of 35-36 residues with molecular weights of 3.7 and 3.4 kDa, respectively. These peptides show close sequence homologies (55-78%) with other scorpion chlorotoxin-like short-chain neurotoxins (SCNs) containing four intramolecular disulfide bridges. Despite the sequence variation between these two peptides (37% heterogeneity) their general structural organization is very similar as shown by their clearly related circular dichroism spectra. Furthermore, Bs6 is a minor component, composed of 38 residues (4.1 kDa) containing six half-cystine residues and having close sequence identities (40-80%) with charybdotoxin-like SCNs containing three disulfide bridges. The non-cysteinic, bacic and thermolabile Bs10 is composed of 34 amino acid residues (3.7 kDa), and belongs to a new class of peptides, with no sequence resemblance to any other so far reported sequence isolated from scorpions. Surprisingly, Bs10 shows some limited sequence analogy with oocyte zinc finger proteins. Results of these studies are discussed with respect to their structural similarities within the scorpion LCNs, SCNs and other biologically active peptides.