Anti-tau intrabodies: From anti-tau immunoglobulins to the development of functional scFv intrabodies

Abstract

Over the last decade, there has been a growing interest in intrabodies and their therapeutic potential. Intrabodies are antibody fragments that are expressed inside a cell to target intracellular antigens. In the context of intracellular protein misfolding and aggregation, such as tau pathology in Alzheimer's disease, intrabodies have become an interesting approach as there is the possibility to target early stages of aggregation. As such, we engineered three anti-tau monoclonal antibodies into single-chain variable fragments for cytoplasmic expression and activity: PT51, PT77, and hTau21. Due to the reducing environment of the cytoplasm, single-chain variable fragment (scFv) aggregation is commonly observed. Therefore, we also performed complementarity-determining region (CDR) grafting into three different stable frameworks to rescue solubility and intracellular binding. All three scFvs retained binding to tau after cytoplasmic expression in HEK293 cells, in at least one of the frameworks. Subsequently, we show their capacity to interfere with either mouse or mutant human tau aggregation in two different primary mouse neuron models and organotypic hippocampal slice cultures. Collectively, our work extends the current knowledge on intracellular tau targeting with intrabodies, providing three scFv intrabodies that can be used as immunological tools to target tau inside cells.

Keywords: CDR grafting; aggregation; intrabodies; scFv; tau.

Graphical abstract

Figure 1

Properties of the selected mAbs and characterization of derived scFvs expressed in the periplasm of E. coli and through the secretory pathway of HEK293 cells (A) Summary table with Fab (or mAb) affinity against human tau and tau paired helical filaments (PHFs) from AD brains; efficacy in immunoprecipitating (IP) tau with aggregation capabilities from brain homogenates of P301S mice and AD patients. Adapted from Vandermeeren et al. (B) scFvs were expressed in the periplasm of E. coli and cleared cell lysates were used for characterization. Expression levels were determined by western blotting and arbitrary unit (AU) concentrations were determined based on western blotting quantification. Images are representative of two independent experiments. (C and D) All lysates were tested for binding against recombinant human tau on ELISA, starting from 1 AU of scFv. Detection in western blotting and ELISA was done with an anti-HA tag HRP-labeled antibody. (E and F) ePHF coating was used for phospho-specific scFv PT77. Results are shown as mean of two independent experiments. (G) scFvs were expressed as secreted protein from HEK293 cells and culture medium was used for characterization. Expression levels were determined by western blotting. (H and I) All samples were tested against recombinant human tau on ELISA, starting from undiluted culture medium. (J and K) ePHF coating was used for phospho-specific scFv PT77. Results are shown as mean ± SD of three independent experiments. Detection is done with an Anti-FLAG-tag HRP-labeled antibody.

Figure 2

Characterization of scFvs expressed in the cytoplasm of HEK293 cells (A) scFvs were expressed in the cytoplasm of HEK293 cells and cleared cell lysates were used for characterization. Intrabody presence in the lysates was determined by western blotting. (B and C) Cell lysates were tested on ELISA in serial dilution starting at 10 μg of total protein against recombinant human tau. (D and E) ePHF coating was used for phospho-specific scFv PT77. Results are shown as mean of two independent experiments. Detection is done with an Anti-FLAG-tag HRP-labeled antibody.

Figure 3

Intrabody solubility in the cytoplasm before and after CDR grafting (A) Schematic representation of the CDR grafting strategy. The CDRs from each chain of the original scFvs (represented in blue) are transferred to a new framework (represented in brown). Framework amino acids identified as important for binding are transferred as well (represented by blue stripes). Additionally, versions where cysteines (indicated by C) are replaced by the amino acid combination Val-Ala (indicated by V and A, respectively) were also designed. Designed with biorender.com. (B) Immunocytochemistry evaluation of intrabody solubility in the cytoplasm. (B1–B6) Original intrabody sequences. (B7–B9) Intrabodies designed in the VL-VH orientation. (B10–B22) CDR-grafted versions, with and without disulfide bonds (SS⁻). Images representative of three independent experiments with two replicates each. Scale bar, 25 μm. SS⁻, scFv without the cysteines that participate in disulfide bonds.

Figure 4

Evaluation of intrabody tau binding after CDR grafting into different frameworks Intrabodies were expressed in HEK293 cells and cell lysates were tested in a serial dilution starting at 1:3 dilution. Detection is done with an Anti-FLAG-tag HRP-labeled antibody. Results are represented as mean ± SD of three independent experiments.

Figure 5

Evaluation of intrabody tau binding in the cytoplasm of HEK293 cells (A) When the intrabody is co-expressed with tau-NLS, it can be translocated to the nucleus only if it is capable of binding tau in the cytoplasmic environment. Co-expression with human α-synuclein-NLS is used as negative control. (B) Tau-NLS phosphorylated at S199/S202 was detected in cell lysates using a sandwich MSD assay with PT77 as capture antibody. Images are representative of three independent experiments with two biological replicates each. Scale bar, 50 μM.

Figure 6

Effect of intrabodies on AD-seed-mediated aggregation Endogenous mouse tau aggregation was measured on cell lysates with sandwich MSD assays. Results are shown as percentage of condition without intrabody expression after normalization to total mouse α-synuclein levels, represented as mean ± SD of three independent experiments. Statistical analysis done with the fitted mixed-effect model with Dunnett correction for multiple comparisons (∗∗p ≤ 0.01). Western blot images representative of three independent experiments.

Figure 7

Effect of intrabodies on hTau-P301L aggregation upon K18-P301L seeds addition hTau-P301L aggregates were measured on cell lysates using sandwich MSD assays. Results are shown as percentage of condition without intrabody expression after normalization to total mouse α-synuclein levels, represented as mean ± SD of three independent experiments. Expression levels were detected with western blotting using undiluted lysates. β-Actin was used as loading control. Statistical analysis done with the fitted mixed-effect model with Dunnett correction for multiple comparisons (∗p ≤ 0.5; ∗∗p ≤ 0.01). Western blot images representative of three independent experiments.

Figure 8

Evaluation of the effect of anti-tau scFv intrabodies in OHSC (A) Aggregated and phospho-aggregated tau levels are shown as percentage of the AU of slices treated with K18 and no intrabody (K18 alone) with mean ± SD. Statistical analysis was done with the fitted mixed-effect model with Dunnett’s correction for multiple comparisons (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001). Note that, for the AT8/AT8 assay, no values could be interpolated for scFv intrabody PT77. Thus, the values were set to the lowest limit of quantification. (B) scFv intrabody expression levels were evaluated with Anti-FLAG staining on western blotting. (C) Tau aggregates isolated from the brain of patients with AD were incubated with increasing amount of mAb PT77 prior to the measurements.