Developing novel therapeutic strategies for Rett syndrome

Ross, Paul (2016) Developing novel therapeutic strategies for Rett syndrome. PhD thesis, University of Glasgow.

Abstract

Rett syndrome (RTT) is a rare paediatric disorder of females that leads to lifelong cognitive, motor, and respiratory impairment . In the vast majority of cases the disorder is caused by de novo mutations in the X-linked gene MECP2. There are currently no treatments, but genetic studies in mice have shown that the disease is reversible, even after the onset of symptoms. Since there remains a fundamental lack of knowledge about the downstream pathways involved in gene function, current therapeutic efforts are focused on targeting the disease at the gene level, mainly using viral based delivery of Mecp2 gene products. Recent work in mouse models has shown that exogenous delivery of a wild-type (WT) copy of the Mecp2 gene can lead to significant improvements in RTT-like symptoms, but significant challenges remain, both in the delivery of gene constructs to target cells, and in maintaining gene transcription within physiological tolerance.
The work in this thesis explores an alternative therapeutic approach, using newly developed genome editing technology. There were two major aims of this thesis. The first aim was to examine the role of the peripheral tissues in the development of the RTT phenotype. In order to develop correctly targeted therapies it is crucial to know what tissues are most relevant to disease development. It has been widely assumed that the major RTT symptoms can be explained solely by an absence of MeCP2 from cells in the nervous system. However, this was based on mouse studies in which only a few gross aspects of the disorder were examined. In this thesis a newly created peripheral knock-out (KO) mouse model, in which Mecp2 transcription is silenced in peripheral tissues but selectively reactivated in the cells of the nervous system, was comprehensively phenotyped in order to determine the role of peripheral MeCP2 in RTT. The second major aim of this thesis was to develop a novel strategy for Mecp2 mutation repair, using recently developed genome editing tools. Based on the results from the peripheral KO phenotyping, this strategy was designed to overcome the particular challenges associated with genome editing in the nervous system, and involved the insertion of a therapeutic construct directly into a non-coding region of the Mecp2 gene using TALEN and CRISPR. This construct was designed to splice to upstream Mecp2 exons in order to replace downstream mutated exons in the final mRNA transcript.
To generate the peripheral KO model (stop-cre), mice in which Mecp2 transcription was globally silenced by a cre-excisable stop cassette (stop-y) were crossed with a nestin-cre mouse line to selectively reactivate gene transcription in the nervous system. Southern blot analysis of tissues showed reactivation in a large number of cells (91.9%) in whole brain samples. Reactivation was particularly high in the cerebellum which showed 96.4% efficiency. Robust silencing was shown in peripheral tissues with only very small levels of reactivation in liver (0.9%), spleen (0.5%), skeletal muscle (1.2%) and heart (7.4%). Higher levels were seen in lung (14.3%) and kidney (24.4%) tissue. Peripheral KO mice did not show the early death phenotype seen in global KO mice and showed only very subtle RTT symptoms when examined using a well-established RTT scoring system. These mice also did not display any of the gait, balance, or respiratory dysfunction typical of RTT mouse models. However, peripheral KO mice did show a reduction in activity levels and exercise capacity across a number of tests. In the open-field, spontaneous activity levels were significantly reduced compared with WT (total distance moved = 3523 cm ± 215 SEM vs 4242 cm ± 167 in WT), on the accelerating rotarod, latency to fall was significantly reduced (168 s ± 14.9 vs 243.5 s ± 11.5 in WT) and on an inclined accelerating treadmill, the time lasted before exhaustion was markedly reduced (8.7 min ± 1.6 vs 16.5 min ± 1.3 in WT). In addition, peripheral KO mice also displayed the biomechanical abnormalities of bone seen in global KO mice, including reduced cortical stiffness and hardness.
The genome editing mutation repair strategy developed in this thesis required a non-coding target region free of repetitive sequence to be identified upstream of exons 3 and 4, where most of the disease causing mutations occur. A suitable 900 bp region of unique sequence was identified 1.6 – 0.7 kb upstream of the beginning of exon 3. To design TALEN pairs targeting this region, the Cornell University TALEN design tool was used to identify 100s of possible TALEN pairs, which were then filtered based on best practice-TALEN design and for the presence of unique restriction sites at the break site. Four pairs remained after filtering and these were assembled using a two stage cloning process based on the Golden Gate method. The efficiency of each pair was assessed using a restriction digest based assay and an online tool (TIDE) which relied on the decomposition of Sanger sequencing traces. The results showed a range of cutting efficiencies from 2.1% of cells (TALEN # 63) to 42.9% of cells (TALEN # 333). A CRISPR design tool was used to generate CRISPR guide target sequences. The four guides with the lowest predicted off-target effects were selected, synthesised as complementary oligonucleotides, and cloned upstream of a guide RNA scaffold in a CRISPR guide expression plasmid. Cutting efficiency was assessed using TIDE, and the results again showed a range of efficiencies from 60.4% (B52) to 22% (A65).
To assess if the best performing TALEN and CRISPR-Cas9 constructs could successfully target exogenous DNA into intron 2 of the Mecp2 gene, a repair construct was designed. This contained WT sequence for exons 3 and 4 of Mecp2 in a minigene format as well as appropriate splice elements and an mCherry fluorescent tag for easy detection. The repair construct was synthesised and cloned into a mammalian expression plasmid, with flanking regions, containing either TALEN or CRISPR target sites, inserted at either end of the construct. For unknown reasons the repair construct was toxic to bacterial cells when flanked by the CRISPR target sites and this version could therefore not be used for transfection experiments. The TALEN repair construct was transfected into cells along with TALEN pair # 333 and successful insertion was assessed using a PCR-based assay. Results showed that the repair construct was successfully inserted into non-coding genomic DNA in the correct location, and that a mutant version of the TALEN construct, designed to increase specificity, led to an increase in the levels of insertion in the correct orientation. To assess if this led to the production of corrected protein, cells were examined for mCherry protein expression using flow cytometry. Results showed that there was no significant increase in mCherry expression in transfected cells, suggesting the repair construct did not successfully splice to upstream Mecp2 exons.
In summary, the results from this thesis show that RTT is primarily a disorder of the nervous system, and that this should therefore be the main target of new therapies. However, they also show that an absence of MeCP2 from the peripheral tissues leads to a markedly reduced exercise capacity, and is also likely to be the primary cause for the bone dysfunction seen in RTT patients and mouse models. In this thesis, a novel therapeutic strategy for RTT was developed using genome editing tools. A number of TALEN and CRISPR constructs were designed that could successfully target specific non-coding regions of the Mecp2 gene, and were shown to enable the insertion of an exogenous DNA repair construct into the genome. However, further flow cytometry analysis showed that this did not lead to the expected protein repair suggesting further work is required on the design of the repair construct to enable splicing to endogenous Mecp2 exons. Overall, the results show that genome editing has a potential role in the treatment of genetic disorders like RTT, but that further work is required to enable successful repair of disease causing mutations.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: Rett Syndrome, CRISPR, genome editing, TALEN, MeCP2, peripheral KO, gene therapy.
• Subjects: Q Science > QH Natural history > QH301 Biology
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Psychology & Neuroscience
• Funder's Name: Biotechnology and Biological Sciences Research Council (BBSRC)
• Supervisor's Name: Cobb, Dr. Stuart and Bailey, Dr. Mark
• Date of Award: 2016
• Depositing User: Mr Paul Ross
• Unique ID: glathesis:2016-7306
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 12 May 2016 09:49
• Last Modified: 23 May 2016 10:34
• URI: https://theses.gla.ac.uk/id/eprint/7306