Therapeutic CRISPR epigenome editing of inflammatory receptors in the intervertebral disc

Abstract

Low back pain (LBP) ranks among the leading causes of disability worldwide and generates a tremendous socioeconomic cost. Disc degeneration, a leading contributor to LBP, can be characterized by the breakdown of the extracellular matrix of the intervertebral disc (IVD), disc height loss, and inflammation. The inflammatory cytokine tumor necrosis factor α (TNF-α) has multiple signaling pathways, including proinflammatory signaling through tumor necrosis factor receptor 1 superfamily, member 1a (TNFR1 or TNFRSF1A), and has been implicated as a primary mediator of disc degeneration. We tested our ability to regulate the TNFR1 signaling pathway in vivo, utilizing CRISPR epigenome editing to slow the progression of disc degeneration in rats. Sprague-Dawley rats were treated with TNF-α and CRISPR interference (CRISPRi)-based epigenome-editing therapeutics targeting TNFR1, showing decreased behavioral pain in a disc degeneration model. Surprisingly, while treatment with the vectors alone was therapeutic, the TNF-α injection became therapeutic after TNFR1 modulation. These results suggest direct inflammatory receptor modulation as a potent strategy for treating disc degeneration.

Keywords: CRISPR; TNFR1; degenerative disc disease; gene therapy; inflammatory cytokine signaling; intervertebral disc; low back pain.

Graphical abstract

Figure 1

Biodistribution of LV vector expression is sequestered within target IVDs following intradiscal injection of rat caudal IVDs, and in vivo, CRISPR-based epigenome editing of TNFR1 in rat caudal IVDs robustly downregulates TNFR1 expression while maintaining healthy IVD morphology (A) Schematic of caudal luciferase experiments. Rat caudal IVDs were injected with luciferase-expressing LV vectors, and luciferase expression was measured after 1 week. (B) Representative c-arm fluoroscope image injecting LV into the rat caudal IVD. (C) Ex vivo bioluminescence images of luciferase-expressing LV vector injected (top left) and control (top right) rat tail segments demonstrated LV expression sequestered within the target disc with no expression in major organs (bottom). Red: high LV expression; blue: low LV expression; clear: no LV expression. (D) Quantification of bioluminescent signal intensity; n = 5 animals. ∗p < 0.05 compared to control disc. (E) Fold change of TNFR1 expression among NT and gRNAs screened in rat NP cells and quantified by quantitative real-time PCR. gRNA2 was used for all subsequent experiments. (F) Schematic of TNFR1 in vivo downregulation experiments. Rats received intradiscal injection of PBS, NT, or TNFR1 epigenome-editing LV vectors. Three weeks following injections, animals were sacrificed, and IVD histology, degeneration grading, and IHC for TNFR1 were performed. (G) IHC staining of TNFR1 expression (brown) in rat NP cells in representative IVD images from animals in the PBS (top), NT epigenome-editing (center), and TNFR1 epigenome-editing (TNFR1, bottom) CRISPR LV vector injection groups. (H) Quantification of TNFR1 IHC staining intensity from IHC images from animals in the PBS, NT, and TNFR1 epigenome-editing groups (TNFR1). ∗p < 0.05 compared to PBS treatment group. (I) Images of H&E (top) and Alcian blue (bottom) staining of histological sections of IVDs from animals in the naive, PBS injection, NT epigenome-editing, and TNFR1 epigenome-editing LV vector injection groups. (J) Degeneration scoring of histology images from animals in the naive, PBS, NT epigenome-editing, and TNFR1 epigenome-editing groups demonstrated no significant differences between treatment groups. Images scored by blinded observers.

Figure 2

In vivo, CRISPR-based epigenome editing of TNFR1 expression in rat caudal IVDs slows small needle puncture-induced disc degeneration and maintains TNFR1 downregulation for 3 months post-injection (A) A schematic of the study design: 3-month-old male Sprague-Dawley rats received an injection of PBS, NT epigenome-editing, or TNFR1 (TNFR1) epigenome editing LV vectors, or no injection (Naive). Three months post-injection, animals were sacrificed, and histology, IVD degeneration scoring, and IHC for TNFR1 were performed on IVDs from animals in each treatment group (n = 6 animals per group). (B) IHC staining of TNFR1 expression (brown) in NP cells in representative IVD images acquired from animals in the naive (left), PBS-injected (center left), NT epigenome-editing LV vector injected (center right), and TNFR1 epigenome-editing LV vector injected treatment groups (right). (C) Quantification of TNFR1 staining intensity. ∗p < 0.05 compared to naive (no injection) group. (D) Images of H&E (top) and Alcian blue (bottom) staining of histological sections of IVDs from animals in the naive, PBS injection, NT, and TNFR1 epigenome-editing LV vector injection treatment groups. (E) Degeneration scoring (0–10) of histological images of IVDs from animals in the naive, PBS injection, NT epigenome-editing LV vector injected, and TNFR1 epigenome-editing LV vector injected treatment groups. ∗p < 0.05 compared to naive (no injection) group.

Figure 3

In vivo, CRISPR-based epigenome editing of TNFR1 expression in rat caudal IVDs slows the progression of disc degeneration and reduces disc height loss in an AP model of disc degeneration (A) Schematic of study design: 3-month-old male Sprague-Dawley rats received caudal intradiscal injections of PBS, NT, or TNFR1 epigenome-editing LV vectors (TNFR1). Three weeks post-injection, PBS-injected animals received either sham or AP, while all NT and TNFR1 received AP. Four weeks following AP, disc height images were obtained, animals were sacrificed, and TNFR1 IHC, histology, and degeneration scoring of IVDs were conducted; n = 6 animals per group. (B) Quantification of TNFR1 IHC staining intensity of NP cells in IVDs from animals in the PBS+Sham, PBS+AP, NT+AP, and TNFR1+AP treatment groups. ∗p < 0.05 compared to sham. (C) DHI calculated from before/after AP fluoroscopic images of rat caudal IVDs of animals in the PBS+Sham, PBS+AP, NT+AP, and TNFR1+AP treatment groups. ∗p < 0.05 compared to PBS+Sham group. (D) IHC images targeting TNFR1 expression with a toluidine blue counterstain in the PBS+Sham, PBS+AP, NT+AP, and TNFR1+AP treatment groups. (E) Images of H&E- (top) and Alcian blue (bottom)-stained histology sections of rat caudal IVDs from animals in the PBS+Sham, PBS+AP, NT+AP, and TNFR1+AP treatment groups. (F and G) IVD degeneration scores of (F) total IVDs and (G) by disc region; AF, NP, AF and NP border (AF/NP), NP cellularity (NP Cell), NP matrix, and endplate (EP). ∗p < 0.05 compared to PBS+Sham group. (H and I) Spearman correlation modeling of (H) IVD degeneration score and (I) DHI as a function of TNFR1 staining intensity.

Figure 4

In vivo, CRISPR epigenome editing of TNFR1 expression in rat caudal IVDs reduced expression of TNF-α, and IL-1β in AP model of disc degeneration (A) Images of IHC staining for TNF-α expression (brown) in rat NP cells from IVDs in the PBS+Sham (top left), PBS+AP (top right), NT+AP (bottom left), and TNFR1+AP (bottom right) treatment groups. (B) Percentage of TNF-α immunopositive NP cells. ∗p < 0.05 compared to PBS+Sham. (C) Images of IHC staining for IL-1β expression (brown) in rat NP cells in the PBS+Sham (top left), PBS+AP (top right), NT+AP (bottom left), and TNFR1+AP (bottom right) treatment groups. (D) Percentage of IL-1β immunopositive NP cells. ∗p < 0.05 compared to PBS+Sham. (E) Images of IHC staining for IL-6 expression (brown) in rat NP cells in the PBS+Sham (top left), PBS+AP (top right), NT+AP (bottom left), and TNFR1+AP (bottom right) treatment groups. (F) Percentage of IL-6 immunopositive NP cells. ∗p < 0.05 compared to PBS+Sham.

Figure 5

In vivo, CRISPR-based epigenome editing in rat caudal IVDs prevents disc degeneration and disc height loss in an inflammatory disc degeneration model (A) Schematic of study design: 3-month-old male Sprague-Dawley rats received a caudal intradiscal injection of PBS, NT, or TNFR1 epigenome-editing LV vectors (TNFR1). Three weeks post-injection, PBS-injected animals received either sham or AP, while all NT and TNFR1-injected animals received TNF-α injection. Four weeks following AP, disc height images were obtained, animals were sacrificed, and TNFR1 IHC, histology, and degeneration scoring of IVDs were conducted; n = 6 animals per group. (B) Quantification of TNFR1 IHC staining intensity of IVDs from animals in the PBS+Sham, PBS+AP, NT+TNF-α, and TNFR1+TNF-α treatment groups. ∗p < 0.05 compared to PBS+Sham. (C) Normalized DHI calculated from radiographs of caudal IVDs obtained before and after TNF-α injections for animals in the PBS+Sham, PBS+AP, NT+TNF-α, and TNFR1+ TNF-α treatment groups. ∗p < 0.05 compared to PBS+Sham treatment group. (D) IHC images targeting TNFR1 expression with a toluidine blue counterstain in the PBS+Sham, PBS+AP, NT+AP, and TNFR1+AP treatment groups. (E) Representative images of H&E- (top) and Alcian blue (bottom)-stained histology sections of IVDs from animals in the PBS+Sham, PBS+AP, NT+ TNF-α, and TNFR1+TNF-α treatment groups. (F and G) IVD degeneration scores of (F) total IVDs and (G) by IVD region; AF, NP, AF and NP border (AF/NP), NP cellularity, NP matrix, endplate (EP). Scored by blinded observers. ∗p < 0.05 compared to PBS+Sham. (H and I) Spearman correlation modeling of (H) IVD degeneration score and (I) DHI as a function of TNFR1 staining intensity.

Figure 6

In vivo, CRISPR epigenome editing of TNFR1 expression in rat caudal IVDs reduced the expression of TNF-α and IL-1β in inflammation model of disc degeneration (A) Images of IHC staining for TNF-α expression (brown) in rat NP cells from IVDs in the PBS+Sham (top left), PBS+AP (top right), NT+TNF-α (bottom left), and TNFR1+TNF-α (bottom right) treatment groups. (B) Percentage of TNF-α immunopositive NP cells; n = 6 animals per group. ∗p < 0.05 compared to PBS+Sham. (C) Images of IHC staining for IL-1β expression (brown) in rat NP cells in the PBS+Sham (top left), PBS+AP (top right), NT+ TNF-α (bottom left), and TNFR1+ TNF-α (bottom right) treatment groups. (D) Percentage of IL-1β immunopositive NP cells; n = 6 animals per group. ∗p < 0.05 compared to PBS+Sham. (E) Images of IHC staining for IL-6 expression (brown) in rat NP cells in the PBS+Sham (top left), PBS+AP (top right), NT+ TNF-α (bottom left), and TNFR1+ TNF-α (bottom right) treatment groups. (F) Percentage of IL-6 immunopositive NP cells; n = 6 animals per group. ∗p < 0.05 compared to PBS+Sham.

Figure 7

In vivo, CRISPR-based epigenome editing of TNFR1 and injection of TNF-α in rat caudal IVDs prevents disc degeneration and disc height loss more than editing of TNFR1 alone, and TNFR2 signaling is a significantly altered metabolic pathway in in vitro RNA-seq of human NP cells (A) Representative images of H&E- (top) and Alcian blue (bottom)-stained histology sections of IVDs from animals in the PBS+Sham, PBS+AP, TNFR1, and TNFR1+TNF-α treatment groups. (B) IVD degeneration scores of total IVD. (C) Normalized DHI calculated from radiographs of caudal IVDs obtained before and after TNF-α injections for animals in the PBS+Sham, PBS+AP, NT+TNF-α, and TNFR1+TNF-α treatment groups. ∗p < 0.05 compared to PBS+Sham treatment group. (D) Gene expression changes of NT+TNF-α vs. NT groups. Red dots denote significant (p < 0.05) gene expression changes. (E) Significantly altered metabolic pathways of interest from the comparison in (D) using the BioCarta 2016 database. (F) Gene expression changes of TNFR1-edited + TNF-α vs. TNFR1-edited groups. Red dots denote significant (p < 0.05) gene expression changes. (G) Significantly altered metabolic pathways of interest from the comparison in (F) using the BioCarta 2016 database. (H) Normalized ratio of TNFR2/TNFR1 RNA-seq counts in NT, TNFR1-edited, NT+TNF-α, and TNFR1+TNF-α groups in in vitro human NP cells. (I) Gene expression changes of TNFAIP3 and caspases of interest in NT+TNF-α vs. NT groups. (J) Gene expression changes of TNFAIP3 and caspases of interest in TNFR1-edited + TNF-α vs. TNFR1-edited groups.

Figure 8

In vivo, CRISPR epigenome editing of TNFR1 expression in rat lumbar IVDs reduced thermal and mechanical pain sensitivity in the AP model of disc degeneration (A) Schematic of practice dye injections. Lumbar injections were performed, and the discs were immediately harvested for subsequent imaging. (B) A representative c-arm fluoroscope image of lentivirus delivery to the IVD using a novel approach. (C) Examples of harvested IVDs from practice injections showing successful dye injection in the NP of IVD. (D) Schematic of lumbar luminescence imaging experiments. (E) Delivery of luciferase-expressing LV vectors to the lumbar IVD showed expression in target IVDs (left, n = 5) and in the innervating DRG (right, n = 5). (F) Schematic of CRISPR epigenome editing of TNFR1 experiments with behavioral assays: 3-month-old male Sprague-Dawley rats received lumbar intradiscal injections of PBS (n = 7) or TNFR1 epigenome-editing LV vectors (TNFR1, n = 7). Two weeks post-injection, all animals received AP. Von Frey and Hargreaves behavioral assays were performed 3 days post-AP and then weekly for 4 weeks. (G) Normalized paw withdrawal threshold (PWT) means of PBS and TNFR1-edited groups using Von Frey behavioral assay. (H) Normalized paw withdrawal latency (PWL) means of PBS and TNFR1-edited groups using Hargreaves behavioral assay. Error bars are SEMs. “A” and “B” denote statistically significant differences between groups. The asterisks denote statistically significant differences at individual time points, α = 0.05.