Date Published: June 6, 2019
Publisher: Public Library of Science
Author(s): Akio Suzuki, Takeshi Yamasaki, Rie Hasebe, Motohiro Horiuchi, Jörg Tatzelt.
Anti-prion protein (PrP) monoclonal antibody 132, which recognizes mouse PrP amino acids 119–127, enables us to reliably detect abnormal isoform prion protein (PrPSc) in cells or frozen tissue sections by immunofluorescence assay, although treatment with guanidinium salts is a prerequisite. Despite the benefit of this mAb, the mechanism of PrPSc-specific detection remains unclear. Therefore, to address this mechanism, we analyzed the reactivities of mono- and bivalent mAb 132 to recombinant mouse PrP (rMoPrP) by enzyme-linked immunosorbent assay (ELISA) and surface plasmon resonance (SPR). In ELISA, binding of the monovalent form was significantly weaker than that of the bivalent form, indicating that bivalent binding confers a higher binding stability to mAb 132. Compared with other anti-PrP mAbs tested, the reactivity of bivalent mAb 132 was easily affected by a decrease in antigen concentration. The binding kinetics of mAb 132 assessed by SPR were consistent with the results of ELISA. The dissociation constant of the monovalent form was approximately 260 times higher than that of the bivalent form, suggesting that monovalent binding is less stable than bivalent binding. Furthermore, the amount of mAb 132 that bound to rMoPrP decreased if the antigen density was too low to allow bivalent binding. If two cellular PrP (PrPC) are close enough to allow bivalent binding, mAb 132 binds to PrPC. These results indicate that weak monovalent binding to monomeric PrPC diminishes PrPC signals to background level, whereas after exposure of the epitope, mAb 132 binds stably to oligomeric PrPSc in a bivalent manner.
Prion diseases are fatal neurodegenerative diseases in animals and humans, including bovine spongiform encephalopathy, scrapie in sheep and goats, and Creutzfeldt-Jakob disease in humans . The causative agents of prion diseases, prions, are mainly composed of an abnormal isoform of prion protein (PrPSc), which is generated from a host-encoded cellular isoform of prion protein (PrPC) by certain post-translational modifications including conformational transformation. After prion infection, production of PrPSc commences in the central nervous system during a long latency period, and eventually neuronal cell death is caused by production of PrPSc in neurons itself or an alteration of the neural niche as a response to PrPSc formation and accumulation. A critical question in elucidating the pathogenesis of prion diseases is where and how PrPSc accumulates in the central nervous system as disease progress.
Although mAb 132 is a pan-PrP antibody, the use of this mAb facilitates reliable PrPSc-specific staining in cells or tissue sections pretreated with a chaotropic reagent [18–20,32]. However, the mechanism of PrPSc-specific detection by mAb 132 is largely unclear. To investigate this mechanism, we analyzed the reactivity and binding kinetics of mono- and bivalent mAb 132 using ELISA and SPR. The monovalent binding of rFab-132 was significantly weaker than that of IgG-132, whereas that of rFab of mAbs 31C6 and 44B1 was as efficient as authentic IgG (Fig 2A and 2B). These results indicate that bivalent binding requires efficient binding of mAb 132. Consistent with the results of ELISA (Figs 2 and 3D), analyses of binding kinetics by SPR showed that antigen density influences mAb 132 binding (Table 2). Crystallographic analysis of IgG revealed that the distance between two Fab domains of intact mouse IgG1 is around 11.8 nm . The number of rMoPrP molecules in a circle of 11.8 nm in diameter can be estimated as 6.6 and 3.7 molecules in case of 100 ng/well rMoPrP in ELISA (Fig 2) and 1.71 ng/mm2 rMoPrP in SPR (Table 2), respectively. Thus, theoretically, the rMoPrP density in these conditions appears high enough to allow bivalent binding of mAb 132 . On the other hand, for 12.5 ng/well rMoPrP, the lowest concentration that mAb 132 could bind in ELISA (Fig 2B), and 0.14 ng/mm2 rMoPrP in SPR, the number of rMoPrP molecules can be estimated as 0.9 and 0.3, respectively, in a circle of 11.8 nm in diameter. Thus, it is expected that monovalent binding of IgG to rMoPrP occurred mainly under these conditions (Fig 9A). The estimated numbers of rMoPrP molecules are consistent with the idea that bivalent binding confers a higher binding stability to mAb 132 (Fig 9B). Indeed, compared with IgG-132, binding of rFab-132 to rMoPrP in ELISA was not so efficient even at the highest antigen concentration (100 ng/well rMoPrP) (Figs 2 and 3D). This trend was not observed with other pan-PrP mAbs, 31C6 and 44B1, and their rFab fragments. Furthermore, the amount of IgG-132 that bound to the sensorchip with 0.14 ng/mm2 rMoPrP was only 20% of that bound to the sensorchip with 1.71 ng/mm2 rMoPrP. Taken together, these results indicate that monovalent mAb 132 binding is less stable than bivalent mAb 132 binding.