Lentiviral Vectors Expressing Chimeric NEDD4 Ubiquitin Ligases: An Innovative Approach for Interfering with Alpha-Synuclein Accumulation

Abstract

One of the main pathological features of Parkinson's disease (PD) is a diffuse accumulation of alpha-synuclein (aS) aggregates in neurons. The NEDD4 E3 Ub ligase promotes aS degradation by the endosomal-lysosomal route. Interestingly, NEDD4, as well as being a small molecule able to trigger its functions, is protective against human aS toxicity in evolutionary distant models. While pharmacological activation of E3 enzymes is not easy to achieve, their flexibility and the lack of "consensus" motifs for Ub-conjugation allow the development of engineered Ub-ligases, able to target proteins of interest. We developed lentiviral vectors, encoding well-characterized anti-human aS scFvs fused in frame to the NEDD4 catalytic domain (ubiquibodies), in order to target ubiquitinate aS. We demonstrate that, while all generated ubiquibodies bind to and ubiquitinate aS, the one directed against the non-amyloid component (NAC) of aS (Nac32HECT) affects aS's intracellular levels. Furthermore, Nac32HECT expression partially rescues aS's overexpression or mutation toxicity in neural stem cells. Overall, our data suggest that ubiquibodies, and Nac32HECT in particular, represent a valid platform for interfering with the effects of aS's accumulation and aggregation in neurons.

Keywords: NEDD4; Parkinson’s disease; alpha-synuclein; ubiquibodies.

Figure 1

Schematic representation of the developed ubiquibodies, along with the map of the third-generation self-inactivating LV adopted in this study.

Figure 2

The developed LVs efficiently express ubiquibodies and their controls. All generated LVs (2.5 μg) were transfected in 293T cells. Forty-eight hours post-transfection, a western blot with an HA-specific antibody, recognizing the different ubiquibodies, along with an anti-tubulin antibody, as loading control, was performed. C- stands for untransfected cells; / indicates an empty lane; arrows and asterisks point to tubulin and ubiquibodies/NEDD4/HECT domain, respectively, and as indicated.

Figure 3

RLPs lead to the expression of the different ubiquibodies in 293T cells. Cells were transduced with RLPs, expressing the following constructs, as indicated: NEDD4 WT; NEDD4 C/S; Nac32HECT WT; Nac32HECT C/S; d10HECT WT; d10HECT C/S; d5eHECT WT; d5eHECT C/S. Forty-eight hours post-transduction, cells were harvested and proteins were run on an SDS-PAGE gel, followed by western blot with an HA antibody, recognizing NEDD4 (~120 kDa bands) as well as the different ubiquibodies (~70 kDa bands), along with an anti-tubulin antibody (~55 kDa bands) as loading control.

Figure 4

Ubiquibodies specifically interact with aS in murine dopaminergic cells. (a) MN9Dsyn cells (4 × 10⁴) were transduced with RLPs expressing similar amount of ubiquibodies and 24 h later, aS expression was induced by treatment with DOX (1 µg/mL). The following day, cells were fixed with 4% v/v PFA and stained with an anti-HA and an anti-aS antibody, followed by PLA probes minus and plus ligation and amplification. Two negative controls were adopted: (i) cells were incubated with the aS antibody alone (panel C, third line, right panel); (ii) cells were incubated with aS and GS antibodies (C (GS), last line, right panel). Cells were analyzed by confocal microscope. In blue—(DAPI): nuclei; in red—PLA spots. Name of ubiquibodies expressed in the different samples are reported. Shown scale bar corresponds to 20 µm. (b) One-way ANOVA (GraphPad Prism version 8.0.0 for Windows, GraphPad Software, Inc, San Diego, CA, USA) statistical analysis was performed by adopting 6 images per each sample. Results are expressed as mean ± SD taken in 6 different regions of the cell culture, from two separate experiments. All the ubiquibody samples have p < 0.0001 with respect to the negative control (C-).

Figure 5

Ubiquibodies expression leads to aS ubiquitination. 293T cells (1.5 × 10⁶) were transfected with LVs expressing ubiquibodies (8 µg) each in the WT or C/S form, along with a plasmid expressing HA-ubiquitin (Ub-HA, 2 µg). Cells were also transfected with LV encoding WT NEDD4 as a control. Twenty-four hours later, cells were transduced with RLPs expressing aS WT fused in frame with a flag tag (aS WT-Flag). In order to recover aS from the cell lysates, 48 h post-transduction, cells were lysed and proteins were subjected to a flag-immunoprecipitation, followed by an anti-HA, to recognize ubiquitinated forms of aS (upper panel), or an anti-aS western blot (lower panel) to control for aS recovery from the cell lysates. Star points to a band corresponding in size to mono-ubiquitinated aS. Results were replicated 3 times in independent experiments with similar results.

Figure 6

Nac32HECT expression does not affect ubiquitination levels of FIV Gag, a physiological target of NEDD4-like Ub-ligases. 293T cells were co-transfected with constructs expressing FIV Gag-STOP (1.5 µg) and HA-tagged Ub (Ub-HA, 2µg), along with LVs encoding either NEDD4 or Nac32HECT (6.5 µg), as indicated. Purified VLPs were subjected to SDS-PAGE, followed by western blotting analysis, as reported. [Ub]_n-Gag stands for ubiquitinated forms of FIV Gag. Results were replicated 3 times in independent experiments with similar results.

Figure 7

WT Nac32HECT reduces aS intracellular levels. Cells were transduced with RLPs expressing either GFP- aS WT (panel a) or aS WT (panel b) to assess by western blot the presence of a single band of the expected size in the sample harvested from cells expressing the fusion protein. Next, 293T cells were transduced with RLPs expressing either GFP alone (panel c) or GFP-aS WT (panel d) and, the day after, with RLPs expressing comparable amounts of WT or C/S Nac32HECT (black and white, respectively). Cells were harvested and green fluorescence derived from GFP was measured by cytofluorimetric analysis, before (pre-treatment) and after (post-treatment) treatment with ubiquibodies. Specifically, the geometric mean of green fluorescence (GFP Mean) was measured as reported in the y axis. 10,000 events were analyzed. Results are expressed as mean ± SD taken from three independent experiments of the cell culture.

Figure 8

Nac32HECT overexpression affects aS intracellular amounts. Two µg of a plasmid expressing aS (pHM6-alphasynuclein-WT) were transfected in the neuroblastoma SH-SY5Y cell line alone or in combination with 2 µg of LVs encoding for either Nac32HECT WT or Nac32HECT C/S. Twenty-four hours later, cells were fixed with PFA (4% v/v), permeabilized with Triton (0.5% v/v) and incubated with an anti-aS (Abcam) antibody, followed by an anti-rabbit rhodamine secondary antibody. Nuclei were marked with DRAQ5™ (ThermoFisher Scientific). The glasses were mounted in glycerol-PBS media and analyzed by confocal microscopy, by maintaining the same laser setting for all the acquisitions. Confocal imagines are reported in (panel a), while (panel b) shows the statistical analysis performed by adopting five z-stack images per each sample from a single experiment (t test, *** p < 0.0001 and ** p < 0.001). Scale bar in panels “a” corresponds to 50 µm.

Figure 9

Intracellular amounts of endogenous aS are affected by Nac32HECT expression in neuroblastoma-derived cells. SH-SY5Y cells were transduced with RLPs expressing Nac32HECT WT or Nac32HECT C/S. Twenty-four hours later, cells were incubated in fresh medium containing 20 μg/mL cycloheximide (Chx), harvested at 6, 8, and 10 h post-treatment, as indicated and lysed. A western blot was performed by adopting an anti-GAPDH and an anti-aS antibody. Values reported in the graphs below each western blot represent the ratio of the pixel intensity of the target protein band to that of the respective loading control, with the ratio calculated for 6 h post-treatment set to 1. Quantification of band intensities was conducted by densitometry analysis (ImageJ software; version number 1.51j).

Figure 10

WT Nac32HECT ubiquibody rescues PD NCS differentiation into DPs. NSCs #49 and #52 were transduced with RLPs expressing similar amounts of WT and C/S Nac32HECT ubiquibodies and cells were differentiated into DPs. Next, (a) a proliferation assay was performed by adopting the commercial kit CyQUANT™ NF Cell Proliferation Assay (ThermoFisher Scientific), along with (b) an evaluation of FOXA2 positive cells by FACS analysis. Statistical differences were determined by p-values of * p < 0.0332; ** p < 0.002; *** p < 0.0002.