Transfer of C-terminal residues of human apolipoprotein A-I to insect apolipophorin III creates a two-domain chimeric protein with enhanced lipid binding activity.

Apolipophorin III (apoLp-III) is an insect apolipoprotein (18kDa) that comprises a single five-helix bundle domain. In contrast, human apolipoprotein A-I (apoA-I) is a 28kDa two-domain protein: an α-helical N-terminal domain (residues 1-189) and a less structured C-terminal domain (residues 190-243). To better understand the apolipoprotein domain organization, a novel chimeric protein was engineered by attaching residues 179 to 243 of apoA-I to the C-terminal end of apoLp-III. The apoLp-III/apoA-I chimera was successfully expressed and purified in E. coli. Western blot analysis and mass spectrometry confirmed the presence of the C-terminal domain of apoA-I within the chimera. While parent apoLp-III did not self-associate, the chimera formed oligomers similar to apoA-I. The chimera displayed a lower α-helical content, but the stability remained similar compared to apoLp-III, consistent with the addition of a less structured domain. The chimera was able to solubilize phospholipid vesicles at a significantly higher rate compared to apoLp-III, approaching that of apoA-I. The chimera was more effective in protecting phospholipase C-treated low density lipoprotein from aggregation compared to apoLp-III. In addition, binding interaction of the chimera with phosphatidylglycerol vesicles and lipopolysaccharides was considerably improved compared to apoLp-III. Thus, addition of the C-terminal domain of apoA-I to apoLp-III created a two-domain protein, with self-association, lipid and lipopolysaccharide binding properties similar to apoA-I. The apoA-I like behavior of the chimera indicate that these properties are independent from residues residing in the N-terminal domain of apoA-I, and that they can be transferred from apoA-I to apoLp-III.

Characterization of secondary structure and lipid binding behavior of N-terminal saposin like subdomain of human Wnt3a.

Wnt signaling is essential for embryonic development and adult homeostasis in multicellular organisms. A conserved feature among Wnt family proteins is the presence of two structural domains. Within the N-terminal (NT) domain there exists a motif that is superimposable upon saposin-like protein (SAPLIP) family members. SAPLIPs are found in plants, microbes and animals and possess lipid surface seeking activity. To investigate the function of the Wnt3a saposin-like subdomain (SLD), recombinant SLD was studied in isolation. Bacterial expression of this Wnt fragment was achieved only when the core SLD included 82 NT residues of Wnt3a (NT-SLD). Unlike SAPLIPs, NT-SLD required the presence of detergent to achieve solubility at neutral pH. Deletion of two hairpin loop extensions present in NT-SLD, but not other SAPLIPs, had no effect on the solubility properties of NT-SLD. Far UV circular dichroism spectroscopy of NT-SLD yielded 50-60% α-helix secondary structure. Limited proteolysis of isolated NT-SLD in buffer and detergent micelles showed no differences in cleavage kinetics. Unlike prototypical saposins, NT-SLD exhibited weak membrane-binding affinity and lacked cell lytic activity. In cell-based canonical Wnt signaling assays, NT-SLD was unable to induce stabilization of ß-catenin or modulate the extent of ß-catenin stabilization induced by full-length Wnt3a. Taken together, the results indicate neighboring structural elements within full-length Wnt3a affect SLD conformational stability. Moreover, SLD function(s) in Wnt proteins appear to have evolved away from those commonly attributed to SAPLIP family members.

A facile method for isolation of recombinant human apolipoprotein A-I from E. coli.

Apolipoprotein (apo) A-I is the major protein component of high-density lipoprotein (HDL) and plays key roles in the Reverse Cholesterol Transport pathway. In the past decade, reconstituted HDL (rHDL) has been employed as a therapeutic agent for treatment of atherosclerosis. The ability of rHDL to promote cholesterol efflux from peripheral cells has been documented to reduce the size of atherosclerotic plaque lesions. However, development of apoA-I rHDL-based therapeutics for human use requires a cost effective process to generate an apoA-I product that meets "Good Manufacturing Practice" standards. Methods available for production and isolation of unmodified recombinant human apoA-I at scale are cumbersome, laborious and complex. To overcome this obstacle, a streamlined two-step procedure has been devised for isolation of recombinant untagged human apoA-I from E. coli that takes advantage of its ability to re-fold to a native conformation following denaturation. Heat treatment of a sonicated E. coli supernatant fraction induced precipitation of a large proportion of host cell proteins (HCP), yielding apoA-I as the major soluble protein. Reversed-phase HPLC of this material permitted recovery of apoA-I largely free of HCP and endotoxin. Purified apoA-I possessed α-helix secondary structure, formed rHDL upon incubation with phospholipid and efficiently promoted cholesterol efflux from cholesterol loaded J774 macrophages.

Apolipophorin III interaction with phosphatidylglycerol and lipopolysaccharide: A potential mechanism for antimicrobial activity.

Apolipophorin III (apoLp-III) is a model insect apolipoprotein to study structure-function relationships of exchangeable apolipoproteins. The protein associates with lipoproteins to aid in the transport of neutral lipids, and also interacts with the bacterial membrane. To better understand a potential role as an antimicrobial protein, the binding interaction of apoLp-III from Locust migratoria and Galleria mellonella with phosphatidylglycerol and lipopolysaccharides was analyzed. ApoLp-III from either species induced a robust release of calcein from phosphatidylglycerol vesicles, but was ineffective for phosphatidylcholine vesicles with comparable side-chain architecture. Acetylation of L. migratoria apoLp-III lysine residues greatly reduced the calcein release from phosphatidylglycerol vesicles, indicating a critical role of lysine side-chains in phosphatidylglycerol vesicles interaction. Isothermal calorimetry provided Kd values of 0.26 µM (L. migratoria) and 0.50 µM (G. mellonella) for binding to dimyristoylphosphatidylglycerol vesicles, which is an order of magnitude stronger compared to zwitterionic vesicles. A strong preference of apoLp-III for dimyristoylphosphatidylglycerol vesicles was also observed with differential scanning calorimetry with a concentration dependent shift in the lipid phase transition temperature. Native PAGE analysis showed that LPS binding was significantly weaker for L. migratoria apoLp-III compared to G. mellonella apoLp-III. This difference was confirmed by fluorescence titration analysis of L. migratoria apoLp-III, which also indicated that acetylation of the apolipoprotein did not affect LPS binding. Taken together, the results indicate that apoLp-III phosphatidylglycerol interaction may follow a detergent model with an important electrostatic binding component. Since lipopolysaccharide binding was not affected by neutralization of apoLp-III lysine-side chains, the binding interaction may be distinctly different from that of phosphatidylglycerol.