The potential of CRISPR/Cas9 genome editing for the study and treatment of intervertebral disc pathologies

The CRISPR/Cas9 system has emerged as a powerful tool for mammalian genome engineering. In basic and translational intervertebral disc (IVD) research, this technique has remarkable potential to answer fundamental questions on pathway interactions, to simulate IVD pathologies, and to promote drug development. Furthermore, the precisely targeted CRISPR/Cas9 gene therapy holds promise for the effective and targeted treatment of degenerative disc disease and low back pain. In this perspective, we provide an overview of recent CRISPR/Cas9 advances stemming from/with transferability to IVD research, outline possible treatment approaches for degenerative disc disease, and discuss current limitations that may hinder clinical translation.


| DISC DEGENERATION: THE NEED FOR NOVEL TREATMENTS
Degeneration of the intervertebral disc (IVD) is an age-related process that is characterized by a catabolic shift, leading to matrix breakdown and-ultimately-structural failure. Apparent degenerative changes first occur in the nucleus pulposus (NP) and are associated with a shift from collagen type II to more fibrotic collagen type I as well as with a reduction in proteoglycans and a consequent loss in hydration and disc height. 1 However, the annulus fibrosus (AF) also undergoes degenerative changes as evidenced by disorganization of the lamellar structure, possibly leading to structural defects, such as clefts and tears. 2 The altered biomechanical status during degeneration contributes to the development of tissue damage through the creation of areas of peak stress, exposing the disintegrated tissue to hyperphysiological loading that it cannot withstand. 3 As the IVD possesses little regenerative capacity, and healing can only take place in the outer AF where nutrient supply is greatest, degeneration gradually progresses without treatment.
Although disc degeneration is a main contributor to back pain, only a subpopulation will become symptomatic and experience so-called degenerative disc disease (DDD) that is associated with increased expression of inflammatory molecules, including interleukins IL-1β, IL-8, and IL-6 and tumor necrosis factor (TNF)-α (reviewed in References [4][5][6][7]. At the moment, patients suffering from DDD are initially treated conservatively, that is with physiotherapy and analgesic medication, but may have to undergo discectomy if symptoms do not improve. Thus, current therapies only target symptoms but not the underlying molecular processes contributing to disc degeneration and pain development. Accordingly, major effort has been made to design novel, biologically targeted treatment options over the past years, with a focus on 2 approaches: On the one hand, regenerative therapies to counteract the degeneration process have been attempted but without compelling results thus far. The use of cellular therapies, for example, stem cell treatment, is negatively affected by the harsh microenvironment of the IVD that is characterized by high mechanical loads, inflammatory cytokines, hypoxia, low glucose levels, acidic pH, and high osmolarity. 8 The application of anabolic substances promoting the production of extracellular matrix (ECM) is hampered by the low cellularity within the IVD (4000 cells/mm 3 in the NP and 9000 cells/mm 3 in the AF) and, furthermore, by the fact that these few cells are metabolically not very active. 9,10 The outcome of injection of classical anabolic factors such as bone morphogenetic protein (BMP)-7, transforming growth factor (TGF)-β, or growth differentiation factor (GDF)-5 is compromised even more by the short half-life of these growth factors and their rapid diffusion out of the IVD. 8 On the other hand, numerous recent research activities have focused on the means to modulate inflammation in the IVD, mostly via inhibition of the inflammatory cascade (eg, biologics such as epigallocatechin gallate [EGCG], resveratrol, or piperine) [11][12][13] or by neutralization of inflammatory mediators (eg, infliximab, a TNF-α inhibitor 14 ). Although these molecular treatments constitute a novel means for molecular disease modulation, their success is likely not sustained, especially as repeated injection of therapeutics into the IVD is not desired. 15 Novel tools that would allow for safe genetic manipulation of resident cells to modulate the catabolic and inflammatory shift or of therapeutic cells to enhance their robustness, and thus allow them to better withstand the IVD environment, could help to circumvent the limitations of current therapeutic approaches. The CRISPR/Cas9 system is composed of the bacterial endonuclease Cas9, which can be directed to any DNA sequence by singleguide RNA (sgRNA). sgRNA is a short synthetic RNA composed of a scaffold sequence necessary for Cas9-binding, that is made of crRNA, loop, and trans-activating crRNA (tracrRNA), and a user-defined, variable 17-20 nucleotide spacer that defines the genomic target to be modified. 20 The short nucleotide spacer of sgRNA is complementary to the genomic DNA target sequence in close proximity to the protospacer adjacent motif (PAM) that, in case of Cas9 obtained from Streptococcus pyogenes, is a conserved NGG sequence. However, the PAM is not part of gRNA as it is only present downstream of the target DNA ( Figure 1). 21,22 The recognition of the target DNA is ensured FIGURE 1 Schematic representation of the CRISPR/Cas9 system. Single-guide RNA (sgRNA) consists of tracer RNA (trRNA); a loop; crispr RNA (crRNA); and protospacer sequence, which is homologous to the target DNA. wtCas9 possess 2 cleavage activities, HNH and RuvC. CRISPR/Cas9 editing tools consist of sgRNA guiding precisely the Cas9 enzyme to the DNA based on the homology between the protospacer motif and DNA. When the heteroduplex between sgRNA and target DNA is formed, Cas9 performs DNA cleavage in close proximity of the PAM sequence and introduces a double-strand DNA break by heteroduplex formation between the nucleotide spacer of sgRNA and the complementary strand of the target DNA, which is followed by Cas9-mediated DNA cleavage. 23 Typical wild-type Cas9 demonstrates double-stranded DNA cleavage activity provided by 2 domains, RuvC and HNH. 24 Compared to other components guiding the programmable nuclease to the targeted DNA locus, sgRNA design and synthesis are simple and cost effective. However, a particular concern of CRISPR/Cas9 can be its off-target activity as the sgRNA can still recognize sequences in the genome with a single-base mismatch, causing unwanted DSB and mutations. To mitigate this disadvantage, more precise sgRNA designs, synthetically engineered Cas9, or nickase-Cas9 (Cas9n) with D10A point mutation possessing only single-stranded DNA cleavage activity have been developed. [25][26][27] CRISPR/Cas9 has been successfully employed to induce single gene mutations, multiple mutations in one cell, 28 and to cleave highly methylated regions. 29 Furthermore, a full range of CRISPR/Cas9 library screening platforms, from genome-wide to pathway-specific, is being developed and used to reveal critical biological processes, regulatory genes in development, aging, or drug resistance. 25,30,31 As such, CRISPR/Cas9 represents a programmable, versatile, and efficient tool for editing virtually any gene. To date, this system has been exploited to reveal exact gene functions, uncover new drug targets, produce more accurate models of human diseases, and provide potential gene correction therapy. 32,33 CRISPR/Cas9-based techniques can be used not only to disrupt but also to repair and/or regulate gene expression ( Figure 2). To generate CRISPR/Cas9-mediated knockouts, RNA-guided Cas9 induces DSBs, commonly activating the nonhomologous end-joining (NHEJ) repair pathway. NHEJ produces small random insertions or deletions (indels), resulting in frameshift mutations and loss-of-function phenotypes. 34 CRISPR/Cas9-mediated gene editing is achieved in the presence of template DNA, when DSBs are repaired by so-called homology-directed repair (HDR) pathways, which act instead of NHEJ and provide precise insertion of donor DNA into the target site. Apart from site-specific DNA repair, HDR can aid in generating controlled gene knockouts and inserting marker sequences or resistance genes for further selection of cells with desired phenotypes. 35 CRISPR/Cas9-mediated transcriptional regulation of gene expression can be achieved by CRISPR interference (CRISPRi) and CRISPR activation (CRISPRa), including CRISPR/Cas9-mediated epigenetic modification of histones. These techniques utilize catalytically inactive RNA-guided Cas9 (so-called dead Cas9, dCas9), fused with transcriptional activators and repressors (VP64 and KRAB, respectively) 36,37 or with histone-modifying domains (eg, p300, LSD1) that can regulate transcription by altering chromatin structure. 38 These gRNA-dCas9 complexes can be designed to reversibly target specific regulatory sequences, act as a scaffold for various transcriptional factors, or directly interfere with transcription. 17,33 In addition, CRISPR technology (particularly CRISPR/Cas13) can be applied to edit RNA by targeting Cas13a protein to RNA, instead of DNA. 39 An overview of possible CRISPR/Cas-based techniques and their specifications is given in Table 1   Collagen II and aggrecan have recently been targeted by CRISPRa in tissue-derived stem cells (ASCs) to shift the cell phenotype toward an IVD phenotype and could also be applied directly in IVD cells. 42 In cartilage research, the Swarm rat chondrosarcoma (RCS) cell line is commonly used to study chondrocyte-specific phenotype due to its similarity to normal rat cartilage and ability to generate cell where a specific antibody is fused to a dCas9 coexpressed with a gRNA. 58 In IVD research, specifically in studies focusing on IVD oxygenation, enCHiP could, for example, be used to investigate the interaction between hypoxia-responsive elements (HREs) and hypoxiainducible factors such as HIF-1α. 47  CRISPR/Cas9 represents a major asset to generate in vivo disease models, such as transgenic mice. Compared to traditional methods, this approach is faster, easier, and more cost effective, and protocols consisting of the injection of a plasmid construct containing the Cas9 and gRNA in mouse zygotes are available. 71 Another approach is to replace the nucleus of an isolated oocyte with the nucleus of a geneedited somatic cell, 72 thus rendering the use of embryonic stem cells obsolete. In addition to single points, 73 large domains can also be mutated, 74 and multiple genes can be targeted simultaneously, with an efficiency of 95% for single mutants and 80% for double mutants. 28 More recently, in vivo genome editing in specific tissue has been demonstrated by synthetic short-lived gRNA-Cas9 RNP complexes. 75 As genetics is postulated to be a main contributor to disc degeneration and DDD, 76 in vivo models expressing Cas9 specifically in the IVD or delivery of short-lived RNP specifically to the IVD will be useful to better understand the role of highlighted candidate genes in disease progression. Such genes include the vitamin D receptor, aggrecan, type IX collagen, asporin, MMP3, IL-1, and IL-6 as polymorphisms in these genes may influence IVD degeneration mechanisms. 77 In vivo inducible systems based on dCas9 have been particularly useful in phenotypic and epigenetic studies in postnatal mammals and in cancer models, allowing the study of gene-level changes. 78,79 Specifically, induced epigenetic remodeling enabled the amelioration of disease symptoms in mice 80 93 Alternative splicing of multiple mRNAs is also involved in chondrogenic differentiation in response to hypoxia, but no such data exist for IVD cells. 94 Specific RNA editing by CRISPR may not only be used to correct false alternative splicing, but also mimic protective alleles, or guide differentiation of stem cells. CRISPR-based genome modification of IVD can be performed directly in vivo or indirectly ex vivo in therapeutic cells that are subsequently transplanted into the IVD. Although targeted CRISPR/Cas9-mediated transgene integration would be ideal, it is not yet completely feasible, and current delivery methods have to be improved.

| Ex vivo edited autologous IVD cells
Autologous IVD cells can be obtained from biopsies removed during surgeries. These cells are frequently affected by pre-existing degeneration and suffer from poor expansion rates in vitro. 95,96 Nevertheless, IVD cells subjected to conventional adenoviral gene delivery of TGFβ, BMP-2, BMP-7, or sex-determining region Y box 2 (SOX2) demonstrated the ability to restore the proteoglycan content and modulate the biological processes in vitro and in vivo, 97,98 suggesting that precise gene targeting in degenerated IVD cells is also possible. Recently, human articular chondrocytes with stable CRISPR/Cas9 knockout of IL1R1 were prepared in vitro and found to have superior properties over nonedited therapeutic cells, 99 with recent evidence that TNFR1 and IL1R1 can similarly be targeted via epigenome editing in human primary IVD cells. 53 This suggests that deletion or knockdown of IL1R1 in therapeutic cells may improve the outcome of cell therapies for patients suffering from joint diseases.

| Ex vivo edited stem cells
Adult stem cells, such as bone marrow mesenchymal stromal/stem cells (MSCs) or ASCs, can activate hallmarks of IVD regeneration. 100,101 Some of their advantages include high proliferation rates, and thus the possibility to expand cells with target modifications.
Implantation of adult stem cells in animal models has resulted in the restoration of an IVD-like phenotype, and promising outcomes were found in pilot clinical applications (phase I/III clinical trials). 102 However, the major drawback of stem cell-based therapies is the poor survival rate of implanted cells due to the pre-existing catabolic and inflammatory environment in the IVD. 103   Clinical translation of CRISPR/Cas9 for IVD pathologies holds promise as this will allow targeting of the underlying molecular processes contributing to disc degeneration and pain development and hence constitute a major improvement compared to current, symptom-driven therapies. While approaches for CRISPR/Cas9-based IVD regeneration may be challenging to translate to clinical practice due to unfavorable risk-benefit ratios 119 (especially when considering that degeneration itself does not always induce suffering or pain), treatments that specifically target disc-related pain have higher chances for translation. Both CRISPR/Cas9-based regulation of inflammatory processes within the IVD and modification of nociception in spinal nerves may have the highest potential for clinical application due to the high socioeconomic burden of LBP. 120 Furthermore, the means to enhance the mechanical stability of the AF through genome editing and thus prevent IVD herniation would be of great significance for the healthcare systems, 121

| CONCLUSION
While our understanding of the mechanisms of the CRISPR/Cas9 system and its application in the clinical setting are still developing, CRISPR/Cas9 has the potential to induce a paradigm shift in the study and treatment of human diseases, including DDD and LBP.
CRISPR/Cas9 systems provide a novel tool to improve the modeling of DDD and allow studying functions from a macroscopic (body) to a microscopic (cell) scale across all mammalian species. Modeling of IVD degeneration has always been a challenge and is a limitation to the field's progression that may be overcome by CRISPR/Cas9. In addition, gene editing, knockout, and endogenous gene expression control represent powerful tools to advance cell engineering in novel and efficient ways, hindered thus far by technological limitations, complexity of applications, and/or the expense/rapidity of development. This will have profound effects on both cell therapeutics and gene therapies for IVD degeneration application. However, the field has much to learn about the delivery and safety of these systems.
Despite these challenges, CRISPR/Cas9 represents a promising new tool that has the potential to break through technological barriers that have been impeding the field's progress and therefore change our thinking and treatment of IVD degeneration, DDD, and LBP.