Background
Tau was initially identified as a microtubule-associated protein that subsequently was found to be the main component of neurofibrillary tangles and other aggregated forms of tau in several neurodegenerative diseases, referred to as tauopathies [
1‐
4]. These diseases include several neurodegenerative disorders such as progressive supranuclear palsy, corticobasal degeneration, Pick’s disease, and certain forms of frontotemporal dementia. Alzheimer’s disease, the most common cause of dementia, is a secondary tauopathy that is accompanied by the presence of amyloid-β pathology. Within the central nervous system (CNS), tau is predominately expressed in neurons and maintained at higher concentrations within axons to promote microtubule assembly and stability [
5]. The regulation of tau binding to microtubules is mediated by the phosphorylation of serine and threonine residues at sites immediately adjacent to or within the microtubule-binding domain (MBD) [
2]. Phosphorylation within or around the MBD alters the conformation of the MBD decreasing its affinity for microtubule binding and liberating tau [
6] . The accumulation of hyperphosphorylated, aggregated tau is present in insoluble paired helical filaments and other structures that result in tau inclusions.
Emerging evidence now suggests the trans-cellular propagation of tau pathology mediates the progression of tauopathy in these diseases [
7‐
9]. Therefore, immunotherapeutic approaches that target extracellular tau could be a way to decrease the spread of tau pathology. In line with this notion, recent studies have demonstrated immunotherapy with different anti-tau antibodies prevented the accumulation of pathological tau in mouse models of tauopathies
10–17. Active vaccination studies in tau transgenic mice targeting aberrant phosphorylated tau protein have also been shown to reduce tau-associate pathology and improve behavioral abnormalities [
10,
11]. Similar to active immunotherapies, passive immunotherapy before or soon after the onset of tauopathy with anti-tau monoclonal antibodies (mABs) also provided some beneficial effects in transgenic tau mice [
12‐
16].
In former studies, we developed and assessed an anti-tau mAB (HJ8.5) that blocked cellular tau seeding in a biosensor assay [
17‐
19]. In addition, passive immunization with the HJ8.5 anti-tau mAB in transgenic mice expressing human P301S tau (P301S-tg) decreased tau-associated pathology, slowed disease progression, and improved cognitive deficits [
17‐
19]. In additional immunotherapy studies, we also generated anti-tau single-chain variable fragments (scFvs) to evaluate the necessity of the Fc domain, as a potential safety concern in general for immunotherapies is the development of adverse effects due to an inflammatory response by IgG activation of FcγRs on microglia [
20,
21]. The expression of the secreted anti-tau scFvs by adeno-associated virus (AAV) – mediated gene transfer into the CNS led to a significant decrease in the accumulation of pathological tau in aged P301S-tg mice, likely by sequestering extracellular tau [
20]. These studies validate the efficacy of anti-tau scFvs and suggest that Fc-dependent microglial mediated tau clearance is likely not mandatory for anti-tau antibodies to elicit neuroprotective effects. Although these studies support immunotherapeutic approaches with either a secreted anti-tau scFv or a conventional antibody being sufficient to delay tau pathogenesis, the majority of pathological tau remains intracellular in the cytosol, making that location of tau likely inaccessible to both cellularly secreted scFvs and full-length antibodies administered intraperitoneally, subcutaneously, intravenously, or intracerebroventricularly. Thus, targeting intracellular tau with anti-tau scFvs expressed intracellularly in the cytosol (intrabodies) may be more efficacious in preventing tauopathy progression and even in removing existing pathological forms of intracellular tau.
Advances in recombinant antibody technology have provided the ability to express intrabodies intracellularly in the cytosol that function by impairing protein-protein interactions or neutralizing protein function based on the epitope targeted. However, the neutralizing mechanisms of conventional intrabodies limit their efficacy when targeting pathological proteins such as tau, in which protein interactors that are detrimental remain unknown and identifying intrabodies that neutralize the aggregation of intracellular tau is somewhat challenging. To overcome this potential obstacle and limitation, we have engineered chimeric anti-tau intrabodies fused to ubiquitin harboring distinct mutations that make the intrabody prone for either proteasomal or lysosomal degradation with the goal of targeting intracellular tau for degradation. We hypothesized that expression of the modified tau-degrading intrabodies in neurons would reduce intracellular tau protein levels leading to a greater neuroprotective effect relative to conventional immunotherapeutic approaches for tauopathies.
In this study, we show the expression of chimeric anti-tau intrabodies significantly reduces tau protein levels in primary neuronal cultures expressing P301S human tau (h-tau) relative to the conventional anti-tau intrabody containing no modifications. By utilizing AAV2/9-mediated gene transfer, we expressed our anti-tau intrabodies or controls in P301S-tg mice prior to overt tau pathology. Intriguingly, despite the ability for both chimerically fused anti-tau intrabodies to decrease P301S h-tau in primary neurons, the expression of anti-tau intrabodies selective for proteasomal mediated degradation in aged P301S-tg mice displayed effects on preventing the accumulation of pathological tau. In contrast, expression of the anti-tau intrabody that targets lysosomal-mediated degradation or the conventional anti-tau intrabody were both ineffective in preventing tauopathy. Additionally, we expressed our anti-tau intrabodies or controls in aged P301S-tg mice after the onset of tauopathy by AAV-mediated gene transfer. Similar to the effects we observed with the expression of anti-tau intrabodies prior to overt tauopathy, expressing the anti-tau intrabody selective for proteasomal mediated degradation displayed effects on decreasing pathological tau in aged P301S-tg mice.
Methods
Animal model of Tauopathy
The mouse model utilized for the present study were transgenic mice expressing human P301S tau (P301S-tg) purchased from The Jackson Laboratory and bred on a B6C3 background
16. P301S mice develop widespread tau-associated pathology that is age dependent primarily in the hippocampus, parts of the neocortex, amygdala, piriform cortex, and entorhinal cortex. In this study, male age matched P301S mice littermates were analyzed due to the high variability in tau-associated pathology in aged P301S female mice [
14,
15]. All experiments were conducted under the institutional guidelines and were approved by the Institutional Animal Care and Use Committee at Washington University School of Medicine. At the time of endpoint, mice were anaesthetized before cardiac perfusion with PBS/heparin followed by brain dissection and immersion in 4%PFA. Following 48 h, the PFA was exchanged with 30% sucrose and the brain was incubated for an additional 48 h before freezing with 2-Methylbutane on dry ice and stored at − 80 °C before sectioned on a microtome (30 μm thickness). Floating sections were kept in 24-well plates with 30% ethylene glycol, 15% sucrose and 0.2 M sodium phosphate at − 20 °C until used.
AAV injections
All intrabodies were cloned into the AAV vector under the control of the chicken-β-actin (CBA) promoter. All AAV used in these studies were produced at the Hope Center Viral Vectors Core at Washington University School of Medicine. Eight month-old P301S mice were anesthetized with isofluorane (4% induction and 2.0% maintenance) followed by bupivacaine (2 mg/kg) administration subcutaneously prior to the incision on the scalp. Once anesthetized, the hair above the skull was shaved and swabbed with Betadine for sterilization. The animal was then placed onto a small animal Kopf stereotaxic apparatus equipped with an anesthetic mask. The skin was resected just posterior to the eyes to the base of the skull and the skull is cleaned with sterile artificial cerebrospinal fluid. The virus was injected unilaterally into the hippocampus at coordinates bregma: − 2.5 mm posterior, 2.0 mm horizontal from midline (right), at a depth of 2.2 mm utilizing a Hamilton syringe (1702RNR) at 0.5 μl/minute for a final volume of 2.5 μl. The Hamilton syringe was then left for an additional five minutes and gently withdrawn.
Characterization of intrabodies in 293 t (HEK cells)
Prior to plating HEK cells, plates were coated with 10 μg ml− 1 poly-L-lysine (PLL, Sigma, P2636) 10 min followed by brief washes with water. The HEK293t cells were plated in DMEM supplemented with 10% FBS at a density of 60 K cells/well. Following 24 h HEK293t cells were transiently transfected with Lipofectamine according to the guidelines of the manufacturer (ThermoFisher). The concentrations of transfected intrabody expressing plasmids ranged from 0.1μg, 0.5μg, 1.0 μg, and 2.0μg pDNA together with co-transfection of 0.2μg pDNA of a tau expressing plasmid. Following 72 h post-transfection the HEK293t whole cell lysates were collected in standard lysis buffer (50 mM Tris-buffer, 150 mM NaCl, 1 mM EDTA, 1% triton and a cocktail of protease and phosphatase inhibitors) and 5μg of total protein lysate was subjected to SDS-PAGE immunoblotting analysis. For proteasomal inhibition, at 72 h post-transfection the cells were treated with 50 μm MG132 (Sigma Aldrich M7449) for 90 min followed by collection in standard lysis buffer. For lysosomal inhibition, at 72 h post-transfection the cells were treated with 500 μm bafilomycin (Sigma Aldrich B1793) for 6 h followed by collection in standard lysis buffer. Afterwards, to detect both the expression of each intrabody, controls, and the relative tau protein levels, immunoblots were subjected to anti-HA conjugated HRP (clone 3F10, Roche; 1:10,000) and human tau-specific antibody anti-HJ8.5 previously characterized14.
Primary neuronal cultures expressing human tau and the intrabodies
We performed neuronal cultures from E17 wild-type mice, described previously27. For immunofluorescence analysis of the relative levels of human tau protein, primary cortical neurons were infected with AAV2/8-synapsin-P301S tau virus for 6 h on ice. Cells were then pelleted, washed, and plated onto 24-well tissue culture plates glass cover slips pre-coated with 10 μg ml− 1 poly-L-lysine (PLL, Sigma, P2636) at a density of 150,000 cells per well in neurobasal medium. Following 72 h post-transduction with AAV-human P301S tau, the primary neuronal cultures were infected with AAV2/9-CBA encoding the various anti-tau intrabodies and cultured for an additional seven days. For analysis, primary neuronal cultures were fixed in 4% PFA, 4% sucrose for 10 min followed by permeabilization with 0.3% triton in PBS for 10 min and stained with human specific tau antibody Ht7B (Thermo Fisher, MN1000B).
For measurements of the relative human P301S tau protein in primary neuronal cultures following the expression of the various anti-tau intrabodies, primary cortical neurons were infected with AAV2/8-synapsin-P301S tau virus for 6 h on ice. Cells were then pelleted, washed, and plated onto 6-well tissue culture plates pre-coated with 10 μg ml− 1 poly-L-lysine (PLL, Sigma, P2636) at a density of 400,000 cells per well in neurobasal medium. Following 72 h post-transduction with AAV-human P301S tau, the primary neuronal cultures were infected with AAV2/9-CBA encoding the various anti-tau intrabodies and cultured for an additional seven days. For analysis, 200 μl of standard RIPA buffer was added to each well for collection of total neuronal protein lysates followed by sonication. For human tau-specific measurements in neuronal lysates, a Sandwich-ELISA was used. Plates were coated with Tau5 (gift from L. Binder, Northwestern University, Chicago, Illinois, USA) and blocked with 4% BSA in PBS. Samples were diluted in sample buffer and as a detector antibody biotinylated anti-htau (clone HT7, Thermo Fisher Scientific MN1000B) followed by streptavidin-HRP (65R-S104PHRP; Fitzgerald) was used. All ELISAs were developed with super slow TMB substrate solution (Sigma) before reading on a plate reader at 650 nm.
Immunohistochemistry
Utilizing the free-floating method, we subjected unilaterally injected AAVs encoding intrabody-K48R, intrabody-K63, conventional intrabody, or AAV-control injected mice brain sections to pathological tau marker anti-AT8 (phospho-tau Ser202, Thr205, Thermo Scientific, MN1020B) and for detection of the various intrabodies anti-HA (Vector Laboratories and Bethyl Laboratories, Inc.). Brain sections (30 μm) were picked that displayed a clear structured cross-section of the hippocampus, piriform, and entorhinal cortex regions. For quenching auto fluorescence, the sections were incubatated in sudan black (40 mg in 40 ml of 70% EtOH) for 10 min followed by washing with 0.03% PBS-Tween. Nonspecific binding was blocked by PBS 3% BSA, 3% normal goat serum 0.1% Triton X-100. Brain sections were then incubated with primary antibodies, rabbit anti-HA (1:1000), and anti-AT8 (1:500) overnight at 4C. Thereafter, the sections were washed three times in PBS for 10 min each followed by incubation with anti-streptavidin conjugated 568 (Molecular Probes 1:500), and anti-rabbit conjugated 488 (Molecular Probes 1:500) for 2 h at room temperature in the dark. Sections were then washed three times in PBS for 20 min each and mounted with ProLong Gold anti-fade reagent. Images were taken with an epi-fluorescence microscope at 4× magnification and quantified using MetaMorph as previously reported27. For analysis of p-tau (anti-AT8), the dentate gyrus, granule cell layer, mossy fibers, CA1, CA2, CA3 regions were traced from a minimum of 2 sections (300 μm) apart from each other per mouse using MetaMorph. To determine the percent area covered per region, we quantified the average ratio of p-tau (AT8) positive area from the contralateral to ipsilateral side.
To measure total tau levels brain sections from mice injected unilaterally with AAVs encoding intrabody-K48R, intrabody-K63, conventional intrabody, or AAV-control injected were subjected to immunohistochemistry with human anti-HT7 (Thermo Fischer, MN1000B). Free floating whole brain sections cut at 30 μm were incubated in .3% H2O2 in TBS for 10 min at room temperature to quench endogenous peroxidase. Sections were blocked in 3% milk .25% Triton X-100. Sections were incubated with primary antibody anti-HT7(1:500) for 1 h at room temperature. ABC Elite solution (1:400) was prepared from VECTASTAIN Elite ABC system (Vector Labs) and sections were incubated in solution for 1 h at room temperature. Staining was developed for 5 min with DAB plus .01% Nickel and .0005% H2O2. Sections were then dipped in ethanol and Xylene to clear any excess DAB and mounted with Cytoseal. Images were captured using the Olympus Nanozoomer 2.0-HT (Hammatsu). Images were then exported to ImageJ and for analysis of total tau, the dentate gyrus, granule cell layer, mossy fibers, Ca1, Ca2, Ca3, striatum radiatum, and hippocampal fissure were manually traced from 2 sections per mouse. To determine the percent area covered per region, we quantified the average ratio of tau positive area from contralateral to ipsilateral side.
Fluorescence in-situ hybridization
For generation of the RNA probe, the anti-tau scFV plasmid was cloned into the pCR II-TOPO vector using the TOPO™ TA Cloning™ Kit, Dual Promoter (ThermoFisher 450,640). The plasmid was then digested with Not1 for 1 h and subsequently purified. The linearized DNA was then transcribed into RNA in the presence of DIG-UTP using the DIG RNA Labeling Kit (SP6/T7) using standard protocol procedures (Sigma-Aldrich (11,175,025,910 Roche).
Mounted sections were incubated with 1 mg/mL pepsin in 0.2 M HCl at 37 °C for 7 min for antigen retrieval followed by three washes of 5 min in PBS. The sections were then incubated twice in 0.1 M phosphate buffer (PB) for 10 min, twice in 0.75% glycine in 0.1 M PB for 15 min, and once in 0.3% Triton X-100 in 0.1 M PB for 20 min, followed by a 5 min wash in PB. For cell permeabilization the sections were incubated for 30 min at 37 °C in 1 M Tris-Cl 0.5 M EDTA buffer containing 0.5 μg/mL proteinase K followed by a single wash in 0.75% glycine for 5 min and one wash in 5X SSC for 5 min. Following the proteinase K digestion, the sections were incubated in hybridization buffer (5X SSC buffer containing 50% formamide, 1x Denhardts, 10 mg/ml salmon sperm DNA, 12.5 mg E. coli tRNA) for 2 h at room temperature. The RNA probe was diluted (1uL/100uL) in hybridization buffer, heated at 80 °C for 5 min prior to applying to the sections which were placed in a vertical chamber humidified with 5X SSC in 50% formamide overnight at 65 °C. The next day, the slides were submerged in pre-warmed 5X SSC for 5 min at 65 °C followed by 3 washes in 0.2% SSC for 30 min each at 65 °C. Sections were then blocked with 0.25% PBS-Triton X-100 5% normal goat serum for 30 min at room temperature followed by incubating with the anti-phospho-tau mAB (AT8 1:500) in 3% BSA-PBS .1% triton overnight at 4 °C. Following three consecutive washes in PBS for 10 min. Fluorescently labeled secondary antibodies were diluted 1:500 in 3% BSA-PBS and applied to the sections for 2 h at room temperature. After three 20 min washes with PBS, sections were coverslipped with Prolong Gold with DAPI (Invitrogen).
Statistical analysis
Blinding and randomization was performed on all analysis. All graphs represent means ± SEM. Statistical analysis was performed with GraphPad Prism 5.01 using one-way ANOVA with Tukey’s multiple comparison.
Discussion
By utilizing ubiquitin, we have engineered anti-tau intrabodies that are prone to target tau for either the proteasome or lysosomal-degradation to identify the most proficient mechanism for reducing tauopathy. We demonstrate an anti-tau intrabody chimerically fused to ubiquitin harboring a mutation prone for proteasome degradation significantly reduced tauopathy in aged P301S-tg mice prior to overt tauopathy (early disease) and after substantial pathological tau deposition (late disease) relative to the conventional intrabody and controls. Follow up studies should aim at evaluating the global expression of the tau-degrading intrabodies and their efficacy in preventing brain atrophy and behavioral abnormalities of aged P301S-tg mice.
Intriguingly, despite all three intrabodies displaying the ability to reduce tau in HEK293t cells and primary neuronal cultures, the chimeric intrabody fused to ubiquitin harboring a mutation prone for lysosome-mediated degradation failed to prevent tauopathy in aged P301S mice after the onset of disease. One possibility is that the decreased efficacy seen with the K48R mutation may stem from the fact that polyubiquitination at K63 has been shown to display a variety of cellular functions in addition to lysosomal degradation that includes DNA repair and internalization of plasma membrane proteins [
27‐
29]. Thus, the ubiquitin harboring a K48R mutation in vivo may be less prone for lysosomal degradation or perhaps the pathophysiological events leading to tauopathy decreases the efficiency of this particular intrabody’s ability to allow for targeted lysosomal degradation [
30,
31]. Alternatively, the anti-tau intrabody
-K48R is rapidly degraded before binding to tau protein leading to a decrease in efficacy. Future studies should aim at evaluating whether targeting other tau mutants for proteasomal degradation also display a higher efficacy relative to the lysosomal degradation with our tau degrading intrabodies.
Bi-functional intrabodies directed toward the proteasome targeting other proteinopathies (e.g. Huntingtin and TAR DNA-binding protein 43) have been proposed in previous studies by fusing a PEST domain to certain scFv [
32,
33]. Proteins with PEST domains display a short half-life, thus intrabodies fused to a PEST domain mimic a short half-life and are rapidly degraded. Although, conceptually PEST-intrabodies decreased the targeted pathological protein in these prior studies, the effects were demonstrated only in cellular culture models by the co-expression of the intrabody with the pathological protein [
32,
33]. In the current study, we evaluated a disconnect between the ability for a chimeric intrabody fused to ubiquitin harboring a mutation prone for lysosome-mediate degradation to reduce tau protein levels in primary neuronal cultures/HEK293t cells and its inability to reduce tauopathy in vivo in aged P301S-tg mice. This disconnect illustrates that although cell culture models are important for rational design, the complexity of in vivo models do not always translate in vivo. Therefore, the therapeutic potential or the extent to which PEST-intrabodies prevent proteinopathies in vivo in aging mammalian models of proteinopathies remains largely unknown. Moreover, the efficiency for PEST-intrabodies to reduce the proteinopathy after the onset of disease also remains a critical question that is unanswered and is important for a thorough evaluation.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.