Genome Engineering for Personalized Arthritis Therapeutics

Abstract

Arthritis represents a family of complex joint pathologies responsible for the majority of musculoskeletal conditions. Nearly all diseases within this family, including osteoarthritis, rheumatoid arthritis, and juvenile idiopathic arthritis, are chronic conditions with few or no disease-modifying therapeutics available. Advances in genome engineering technology, most recently with CRISPR-Cas9, have revolutionized our ability to interrogate and validate genetic and epigenetic elements associated with chronic diseases such as arthritis. These technologies, together with cell reprogramming methods, including the use of induced pluripotent stem cells, provide a platform for human disease modeling. We summarize new evidence from genome-wide association studies and genomics that substantiates a genetic basis for arthritis pathogenesis. We also review the potential contributions of genome engineering in the development of new arthritis therapeutics.

Keywords: Arthritis; GWAS; autoimmune; drug screening; genome engineering; iPSC; inflammation.

Key Figure, Figure 1. Functional analysis of candidate arthritis causal variants in a human iPSC arthritis modeling platform

Shown, is a schematic of arthritis modeling by editing candidate variants in human iPSCs and assaying the changes in induced cartilage tissue after administering disease-associated stimuli (i.e. inflammatory cytokines, mechanical loading). Candidate disease-modifying therapeutics might then be assayed to potentially reverse deficiencies in cells harboring genetic variants.

Fig 2. Description of eQTLs and identification of candidate variants using CRISPR-Cas9 screening technology

A) Illustration of cis- and trans-expression quantitative trait loci (eQTL). i) Transcription factor A and B (TF-A/TF-B) bind to enhancer elements upstream of Gene X to promote transcription. ii) A polymorphism in the promoter sequence of Gene X inhibits binding of TF-A, reducing transcription of Gene X (cis-eQTL). iii) A polymorphism in the coding region of TF-A prevents TF-A binding to Gene X enhancer element, attenuating expression (trans-eQTL). B) Non-coding CRISPR screen to identify enhancer elements of Gene ×. gRNAs are saturated in non-coding regions flanking and within the open reading frame (ORF). The cell line of interest is transduced with a pool of lentivirus encoding the DNA library. Cells are sorted by expression of Gene X or phenotype of interest and gRNAs within positive and negative populations are sequenced to determine enrichment of gRNAs. Enriched gRNAs bind to candidate enhancers. GWAS-SNPs within these regions can be edited for functional validation as illustrated in Key Figure, Figure 1.

Fig 3. Repair Mechanisms Following Targeted Double Strand Break by Site-Specific Nucleases

A) Non-homologous end joining (NHEJ) can result in repair of dsDNA with random insertions or deletions that may cause frameshift mutations, resulting in premature termination of translation. B) Introduction of a homologous donor sequence (ssODN shown) can facilitate precise editing through homologous recombination repair pathways. C) Use of two nucleases can create specific deletions, allowing the removal of entire regions of the genome.

Fig 4. Common Genome Engineering Tools

A) Zinc Finger Proteins (ZFPs) and Transcription Activator-Like Effector (TALE) modules are engineered to bind a desired sequence, and then fused to the FokI endonuclease to generate targeted double strand breaks. In the CRISPR system, a custom designed 20 base pair (bp) guide RNA (gRNA) recruits Cas9 endonuclease to induce double strand breaks. B) ZFPs and TALEs fused to transcriptional activators and epigenetic modifiers are shown. In the CRISPR system, these domains are fused to a deactivated Cas9 (dCas9)