Genetic knockout of Cathepsin D using zinc-finger nucleases delivered by AAV vectors

Abstract

Genetic engineering is known as a powerful technique for basic research and clinical applications. Recent progress in development of zinc-finger nucleases (ZFNs), which combine the DNA cleavage ability of Fok1 restriction enzyme with highly specific recognition properties of zinc-finger motifs, allows to improve efficiency and to broaden the field of use of genome editing. Here, we demonstrate our initial results in generating novel tools for Cathepsin D gene knockout in neurons based on ZFNs technology and mediated by adeno-associated virus (AAV) vectors. Pairs of AAV-ZFNs were produced and demonstrated the robust expression of nucleases in neuronal cell culture. Observed toxicity most likely was associated with heterodimerization but not homodimerization of ZFNs; cytotoxicity was greatly reduced when ZFN were provided at lower concentrations. Future studies evaluating efficiency of Ctsd knockout, off-target effects on molecular level and long-term outcomes in vivo can be performed.

Introduction

Zinc finger nucleases (ZFNs) are engineered restriction enzymes which are employed as effective and versatile tools for targeted genome editing. Each ZFN consists of two functional domains: Fok 1 restriction enzyme for DNA cleavage and zinc fingers (ZFs) for specific DNA binding (Figure 1). After heterodimerization of a ZFNs pair in an inverted orientation, Fok 1 can induce double-strand breaks (DSBs) between the DNA sequences, specified by ZFs motifs. Assemblies of ZFs recognize DNA in a modular fashion, where a single ZF protein interacts with a single triplet of nucleotides, thereby allowing highly specific targeting of any DNA sequence in a complex genome (Palpant NJ, 2013). DSBs dramatically increase efficiency of gene targeting by stimulating two evolutionary conserved repair mechanisms. First is homologous recombination which underlies targeted gene replacement between existing sequence and designed donor DNA. Second, non-homologous end-joining, is a rapid but error-prone mode of DNA repair, which provides targeted mutagenesis by small insertions, deletions, substitutions etc (Caroll D, 2011).

ZFNs gene targeting was successfully applied in numerous model organisms including Drosophila melanogaster, Danio rerio, Caenorhabditis elegans, Xenopus, Arabidopsis, rodents etc (Palpant NJ, 2013) and various human cell lines and primary cells, such as T lymphocytes, mesenchymal stromal cells, embryonic stem cells, induced pluripotent stem cells etc (Händel EM, 2012). Gene editing frequencies up to 50% were reported (Lombardo A, 2007). But frequencies of desired gene modifications vary between different cell types and developmental stages, and depend on the method of ZFNs delivery (Caroll D, 2011), which stipulates a necessity to understand the biology and to assess an efficiency of gene targeting in every system under study.

illustration not visible in this excerpt

Figure 1. Schematic structure of ZFN pair bound to the target DNA (modified from http://pnabio.com/pna/ZFN.htm# )

Genetic engineering mediated by ZFNs is a promising approach to alleviate genetic diseases. Thus, ZFN knockout of CCR5 is applied in current HIV/AIDS clinical trial. Moreover, ZFNs technology allows developing procedures for gene manipulation in non-traditional model organisms (Caroll D, 2011). Injections of ZFNs mRNA into mouse embryo enable fast generation of genetically-engineered mouse models, which creates an alternative to more time-consuming and expensive conventional gene targeting in embryonic stem cells (Sung YH). However, genetic engineering with ZFNs raises certain ethical and safety issues. ZFNs experiments require careful assessment of the off-target effects and finding a balance between ZFN-associated toxicity and essential level of nuclease activity (Händel EM, 2012).

In current project we are aiming to generate on the basis of ZFNs technology a novel tool for Cathepsin D gene (CTSD) knockout. Cathepsin D is the major lysosomal aspartic protease involved in regulation of numerous proteolytic pathways (Qiao L, 2008). Deficiency of this enzyme leads to the congenital form of human neuronal ceroid lipofuscinosis (NCL) – a devastating neurodegenerative disorder, which manifests clinically with extreme neuronal loss leading to microcephaly, status epilepticus, respiratory insufficiency, astrogliosis and microglia activation, and death within hours to weeks after birth (Siintola E, 2006). CTSDnull mice develop normally until P14, but afterwards acquire CNS and intestinal pathologies followed by premature death at postnatal day 26 + 1 (Shevtsova Z, 2010). Cathepsin D is likely to be involved in degradation of ɑ-synuclein (Qiao L, 2008; Sevlever D, 2008) and Tau (Khurana V, 2010). CTSD knockout models demonstrate extensive aggregation of mentioned above proteins and prominent neurotoxicity, while overexpression of the protease has a protective effect against ɑ–synuclein–induced neuronal loss (Qiao L, 2008). Moreover, the functions of cathepsin D are not restricted only to its enzymatic activity in lysosomes. Thus, it is involved in apoptotic pathways and acts as a mitogen, inducing progression and metastasis of certain tumors, e.g. breast cancer, and is considered to be a marker for poor prognosis. Procathepsin D is heavily secreted from various cancer cells; it supports tumor microenvironment and induces angiogenesis (Benes P, 2008; Masson O, 2010). Despite high interest of research community to Cathepsin D, many of its’ specific functions and substrates are yet to be discovered.

The goal of our study was to bring together ZFNs technology with adeno-associated virus (AAV) delivery system for CTSD knockout in neurons. We generated the vectors with different levels of ZFNs expression and evaluated their toxicity due to homo- and heterodimerization in neuronal cell culture. This work underlies further ex vivo and in vivo assessments of AAV-mediated ZFNs tools for CTSD ablation and their implementation in Cathepsin D research.

Materials and methods

Molecular cloning

Cloning procedure was simulated with the SECentral software. The constructs encoding ZFN1 and ZFN2, mur CTSD _pZFN1 and mur CTSD _pZFN2 (Figure 2), were produced by Sigma (Sigma-Aldrich, USA). As a backbone for subcloning we used the plasmid AAV-6P-NoTB-SEWB (Figure 3) (kindly provided by Dr. S.Kügler).

5-10 μg of plasmid DNA were used in each restriction digest reaction. Appropriate restriction enzymes and buffers (New England BioLabs, USA) were added to DNA and incubated for 1-2 h at the specified by manufacturer temperature. DNA fragments were separated on preparative 1% agarose gel. Desired bands were cut out and subjected to the gel extraction according to the QIAGEN kit protocol. Eluted DNA was precipitated with 3M sodium acetate and 100% ethanol in order to increase concentration and purity of the samples. Concentration of DNA was determined with analytical gel against λ-HindIII-standard. For ligation a vector and an insert DNA were added in 1:3 molar ratio and mixed with the T4 DNA ligase and T4 ligation buffer. Ligation reaction was performed for 20 min at room temperature. The product of ligation was electroporated into electrocompetent E.coli cells. Transformed cells were incubated in SOC++ medium (2% bactotryptone, 0.5% yeast extract, 10 mM NaCl, 10 mM KCl, 20 mM MgCl2 and 2 mM glucose) for 45 min at 37°C and then plated on LB agar plates containing ampicillin (100μg/ml) for the selection of the clones. DNA plasmid extractions were performed using the QIAGEN Plasmid Mini- and Megaprep kits according to the protocol of the manufacturer. Subsequent control digestion with SmaI -enzyme was used to confirm a ligation of proper DNA fragments (Figure 4B).

5-10 μg of plasmid DNA were used in each restriction digest reaction. Appropriate restriction enzymes and buffers (New England BioLabs, USA) were added to DNA and incubated for 1-2 h at the specified by manufacturer temperature. DNA fragments were separated on preparative 1% agarose gel. Desired bands were cut out and subjected to the gel extraction according to the QIAGEN kit protocol. Eluted DNA was precipitated with 3M sodium acetate and 100% ethanol in order to increase concentration and purity of the samples. Concentration of DNA was determined with analytical gel against λ-HindIII-standard. For ligation a vector and an insert DNA were added in 1:3 molar ratio and mixed with the T4 DNA ligase and T4 ligation buffer. Ligation reaction was performed for 20 min at room temperature. The product of ligation was electroporated into electrocompetent E.coli cells. Transformed cells were incubated in SOC++ medium (2% bactotryptone, 0.5% yeast extract, 10 mM NaCl, 10 mM KCl, 20 mM MgCl2 and 2 mM glucose) for 45 min at 37°C and then plated on LB agar plates containing ampicillin (100μg/ml) for the selection of the clones. DNA plasmid extractions were performed using the QIAGEN Plasmid Mini- and Megaprep kits according to the protocol of the manufacturer. Subsequent control digestion with SmaI -enzyme was used to confirm a ligation of proper DNA fragments (Figure 4B).

[...]