Single-chain fragment variable passive immunotherapies for neurodegenerative diseases

Abstract

Accumulation of misfolded proteins has been implicated in a variety of neurodegenerative diseases including prion diseases, Alzheimer's disease (AD), Parkinson's disease (PD), and Huntington's disease (HD). In the past decade, single-chain fragment variable (scFv) -based immunotherapies have been developed to target abnormal proteins or various forms of protein aggregates including Aβ, SNCA, Htt, and PrP proteins. The scFvs are produced by fusing the variable regions of the antibody heavy and light chains, creating a much smaller protein with unaltered specificity. Because of its small size and relative ease of production, scFvs are promising diagnostic and therapeutic reagents for protein misfolded diseases. Studies have demonstrated the efficacy and safety of scFvs in preventing amyloid protein aggregation in preclinical models. Herein, we discuss recent developments of these immunotherapeutics. We review efforts of our group and others using scFv in neurodegenerative disease models. We illustrate the advantages of scFvs, including engineering to enhance misfolded conformer specificity and subcellular targeting to optimize therapeutic action.

Figure 1

The process of amyloid formation. The first step of the process can be a conversion from the native α-helix rich soluble cellular protein into the pathogenic β-sheet rich isoforms, in which they are able to self-assemble through a variety of subsequent nucleation and growth steps to form oligomer, protofibril, and eventually amyloid fibril aggregates. Oligomer and protofibril are putative toxic species that drive neuronal dysfunction. We hypothesize the therapeutic targets are monomer, oligomer and protofibril.

Figure 2

Illustration of single-chain fragment variable (scFv) structure and recombinant adeno-associated virus (rAAV) delivery of scFv. Engineered scFv lacking the constant regions of IgG is a much smaller molecule that still retains the antigen binding affinity. scFvs can be packaged into rAAV and efficiently delivered in vivo for therapeutic purposes. Ag: Antigen; V: variable region; C: constant region; L: light chain; H: heavy chain.

Figure 3

Outcomes of recombinant adeno-associated virus (rAAV) delivered single-chain fragment variable (scFv) therapy for prion disease in transmissible spongiform encephalopathies (TSE) mouse model. (a) Clinical rating data show that the disease onset was delayed in mice administered rAAV2-scFvD18; (b) Gehan-Wilcoxon test data show that the incubation period of prion disease was delayed in mice treated with scFv3:3 and scFvD18; (c) Histoblot analyses of brain sections show decreased levels of proteinase K-resistant PrP^sc in scFvD18 and scFv3:3 treated mice; (d) Immunoblots of brain homogenates (left panel) show a decreased level of proteinase K-resistant PrP^sc in scFvD18 and scFv3:3 treated mice. Quantitative measurements (right panel) of the immunoblots indicate that the decrease in scFvD18 group is significant when compared with the Phe control (marked with “*”, p < 0.004, t-test). PK: proteinase K; “−”: no PK digestion; “+”: PK digested. Figures reproduced from Wuertzer et al., 2008 [39].

Figure 4

Assessment of recombinant adeno-associated virus (rAAV) delivered single-chain fragment variable (scFv) therapy for Alzheimer’s disease (AD) in 3xTg AD mouse model. Hippocampal injections of rAAV1-Phe-scFv (Phe-scFv), rAAV-Aβ-scFv (Aβ-scFv), or no injection control (None) into three-month-old 3xTg-AD mice led to no change in hippocampal soluble Aβ. (a) Levels but significant decreases in insoluble Aβ; (b) Levels in the Aβ-scFv group; (c) The levels of soluble oligomeric Aβ in the Aβ-scFv treated mice was significantly decreased compared to mice with no injection; (d,e,f) Congo red staining showed reduction of amyloid plaques in Aβ-scFv injected mice; (g) Quantitative measurements of amyloid plagues. (*p < 0.05; **p < 0.01; One-way ANOVA with Bonferroni’s multiple comparison post-hoc test.); (h) Morris Water Maze test showed improved spatial learning in Aβ-scFv treated mice when compared to the Phe-scFv treated mice; (i) The total number of trials that Phe-scFv injected mice failed to meet the learning criteria was significantly higher than that of the Aβ-scFv injected mice (p = 0.0041, Fisher’s exact test). Figures reproduced from Ryan et al., 2010 [69].

Figure 5

Differential cellular location of single-chain fragment variable (scFv) that targets specific pathogenic antigen in different neurodegenerative diseases. Recombinant adeno-associated virus (rAAV)-scFv can be engineered to localized to desired subcellular regions (intrabodies), or as a secreted peptide to target membrane-bound or extracellular proteins.