Start codon disruption with CRISPR/Cas9 prevents murine Fuchs’ endothelial corneal dystrophy

  1. Hironori Uehara
  2. Xiaohui Zhang
  3. Felipe Pereira
  4. Siddharth Narendran
  5. Susie Choi
  6. Sai Bhuvanagiri
  7. Jinlu Liu
  8. Sangeetha Ravi Kumar
  9. Austin Bohner
  10. Lara Carroll
  11. Bonnie Archer
  12. Yue Zhang
  13. Wei Liu
  14. Guangping Gao
  15. Jayakrishna Ambati
  16. Albert S Jun
  17. Balamurali K Ambati

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

A missense mutation of collagen type VIII alpha 2 chain ( COL8A2 ) gene leads to early-onset Fuchs’ endothelial corneal dystrophy (FECD), which progressively impairs vision through the loss of corneal endothelial cells. We demonstrate that CRISPR/Cas9-based postnatal gene editing achieves structural and functional rescue in a mouse model of FECD. A single intraocular injection of an adenovirus encoding both the Cas9 gene and guide RNA (Ad-Cas9-Col8a2gRNA) efficiently knocked down mutant COL8A2 expression in corneal endothelial cells, prevented endothelial cell loss, and rescued corneal endothelium pumping function in adult Col8a2 mutant mice. There were no adverse sequelae on histology or electroretinography. Col8a2 start codon disruption represents a non-surgical strategy to prevent vision loss in early-onset FECD. As this demonstrates the ability of Ad-Cas9-gRNA to restore the phenotype in adult post-mitotic cells, this method may be widely applicable to adult-onset diseases, even in tissues affected with disorders of non-reproducing cells.

Article activity feed

  1. Version published to 10.7554/elife.55637 on eLife
  2. eLife

    ###Reviewer #2

    This is an interesting study demonstrating the use of CRISPR/Cas9 to prevent development of Fuchs' corneal dystrophy in a mouse model in which the human mutation (Q455K / Q455K) was knocked into the Col8a2 gene. This gene mutation has been previously shown to induce early-onset Fuchs' dystrophy in patients. This is an important observation with translational potential to treat a subpopulation of patients with Fuchs' dystrophy.

    In general, the data support the author's conclusion that Adenovirus-Cas9-gRNA restores the phenotype in adult post-mitotic cells.

    I have a two major questions/issues:

    1. The data presented in Fig. 3 are critical to the paper and show that Ad-Cas9-Col8a2gRNA treatment reduces expression of the Col8A2 protein in corneal endothelial cells. However, there is no quantitative assessment of the protein reduction other than the images presented from three cross sections. Since Fig 2a indicates the transduction of the corneal endothelial cells is not evenly distributed, some type of quantitative assessment is needed for Fig. 3, either measuring the antibody staining in numerous sections from several different corneas, or by western blot. This is necessary, even though there is quantitative assessment of the change in phenotype of the treated corneas (corneal endothelial cell density, morphology, and guttae-like lesion expression).

    2. To demonstrate that Ad-Cas9-Col8a2gRNA treatment rescued corneal endothelial cell function in the mutant mice, the authors developed an assay that measured the ability of endothelial cell pump function to reduce swelling of the stroma after the corneas were induced to swell by adding hypertonic solutions. While this assay does measure pump function, there is a more direct measure of mutant corneal endothelial cells. The investigators that created the Col8A2 (Q455K / Q455K) mutant mice demonstrated the mutation caused an activation of UPR (unfolded protein response) as shown by an increase in Grp78 and Grp153 in corneal endothelial cells. In my opinion, demonstrating rescue of this function in the mutant mice would have been significantly more impressive.

  3. eLife

    ###Reviewer #1

    The study by Uehara et al titled "Start codon disruption with CRISPR/Cas9 prevents murine Fuchs' endothelial corneal dystrophy" describes a strategy for resolving a dominant negative disease phenotype by CRISPR/Cas9 targeting of the start codon of the causative gene, Col8a2. The authors employ recombinant adenovirus packaging SpCas9 and a single gRNA targeting the start codon of the Col8a2 ORF. In vivo efficacy in wild type mice correlates with a qualitative reduction in COL8A2 expression in these mice by immunostaining. Using a mouse model homozygous for a causative mutation, Col8a2Q455K/Q455K, the authors show a significant reduction in disease pathology, qualitatively via tissue architecture and quantitatively by assaying corneal endothelial pump function. Off-target effects are modeled in vitro and identify several sites, but no significant concerns noted. Overall, the study provides proof-of-concept and feasibility of utilizing this approach, with significant possible outcomes for FECD. Significant concerns pertaining to cassette design, data analysis and additional experiments are highlighted below.

    1. The vector construct utilizes a ubiquitous promoter, Chicken beta actin (truncated) to drive Cas9 expression and a U6 promoter to drive guide RNA. It is unclear why the authors only see a qualitative effect on protein knockdown by immunostaining in the endothelium. Does Adenovirus not infect underlying stromal or epithelial cells? The presence/absence of Ad DNA in these other cells has not been evaluated.

    2. A correlation between expression of Cas9, gRNA and COL8A2 (protein and mRNA) would be important to establish in mice. This is especially critical to demonstrate in the disease model not only to correlate protein knockdown with restored function, but because the efficiency of Ad infection or gene editing could vary in diseased cells.

    3. The authors note that the indel frequency, determined by deep sequencing, appears inconsistent with the observed protein knockdown as determined by immunostaining of tissue sections. However, while the indel frequency is determined quantitatively (~20-25%), but the protein and mRNA levels are not quantified. Is the half-life of wt and mutant COL8A2 known? The authors also report an editing normalized indel rate of 102% in endothelial cells. While the hypothesis of gDNA contamination from non-targeted tissue is likely true (supported by experimental evidence from Supplemental Figure 2), the method used for correction is insufficient to be used to report a true, corrected indel frequency.

    4. Overall what is the minimum/threshold % of endothelial cells that need to be edited to restore function? This information will be critical in designing vector dose and altering promoter strength/specificity to reduce off-target effects. While the impact of vector dose on COL8A2 expression knockdown is assessed, data pertaining to off-target effects at different doses are not presented.

    5. Does overexpression of spCas9, gRNA and knockdown of COL8A2 affect the expression of other genes in the endothelium? The authors analyze the impact of Ad dosing at the inflammatory level, but consequences of control vs treatment vector on endothelial cell gene expression have not been evaluated (e.g., Yu et al., Nat Commun, 2017).

  4. eLife

    ###This manuscript is in revision at eLife

    The decision letter after peer review, sent to the authors on June 12, 2020, follows.

    Summary

    Repair of any genetic disease is of interest, and Uehara and colleagues have shown an improvement in corneal tissue architecture and function in a mouse model of Fuchs' Dystrophy using gene editing delivered by adenovirus. The current review raises a number of important points. A quantitative assessment of Col protein level relative to the expression of Cas9 and gRNA (Reviewer 1, point 2) would strengthen the data shown in Figure 3, as was also suggested by Reviewer 2 (point 1), and must be carried out. This would also help the argument presented by authors regarding genomic DNA contamination that was indirectly addressed by Sup. Fig 2. Although not required, it is recommended that the question of inflammation and/or effects on gene expression by the adenovirus be addressed more thoroughly, by sequencing or by a more thorough evaluation of gene expression changes. This is an issue as Adenovirus is known to incite pathological inflammatory effects. Finally, again not required by recommended, the authors are encouraged to assay for a correction of the UPR.

    Essential Revisions

    Please quantify the levels of collage protein. Please see the reviews for additional comments.

  5. Version published to 10.1101/2020.03.18.996728v1 on bioRxiv