Discogenic cell transplantation directly from a cryopreserved state in an induced intervertebral disc degeneration canine model

A multitude of studies has indicated the potential of cell therapy as a method for intervertebral disc (IVD) regeneration. Transplantation of a variety of cells has been assessed and shown capable of deterring the rate of degeneration in animal models and in human clinical trials. In this study, a novel approach using human discogenic nucleus pulposus cells directly from their cryopreserved state was assessed. In an established canine disc degeneration model, the degeneration process was evaluated in IVDs receiving precultured discogenic cells, thawed‐only discogenic cells, and a saline sham injection after induction of degeneration. Degeneration progression was followed over time by the evaluation of the disc height index (DHI). Finally, after 12 weeks, the manipulated and control discs were explanted, histologically stained, and scored. Treated discs demonstrated retained DHI values for all treatment options. Histologic evaluations demonstrated significant improvement of matrix features compared to the sham. Moreover, thawed‐only cells function at least as well as precultured discogenic cells. In short, cell transplantation of human discogenic cells directly from their cryopreserved state can arrest disc height degeneration and maintain histological matrix features in a canine disc degeneration model. The presented work demonstrates the potential of an off‐the‐shelf cell therapy product to treat degenerative disc disease.


| INTRODUCTION
Low back and neck pain are the largest causes of disability worldwide and constitute a crucial global health issue. Approximately 5% to 10% of the estimated 632 million low back pain patients globally will advance to a chronic condition. 1,2 Together, these compelling numbers engender a critical social-economic burden on society. For example, within the USA, 100 billion USD is spent annually on low back pain associated costs. 3,4 Currently, treatment options are limited and fail to restore or halt further advancement of the underlying pathology. 5 Degeneration of the intervertebral disc (IVD) is widely considered to be a predominant cause of low back pain. IVD degeneration is hallmarked by a dysregulation in extracellular matrix (ECM) homeostasis. 6,7 The exact origin of IVD degeneration remains to be elucidated; however, the nucleus pulposus (NP) is believed to be the place of onset. 8,9 Progression of IVD degeneration is characterized by a decline in proteoglycan production, an increase in matrix degenerative proteins, and a switch from type II collagen to type I collagen production. 6,7 Moreover, NP cells undergo senescence and dedifferentiation toward a more fibrotic phenotype. 10 The NP disorganization and height loss causes incorrect loading of adhering tendons promoting reorganization of tendon ECM to thicker and stiffer structures, further disrupting the mechanical features along the spine. 11 Despite the well-established pathway of the degeneration pathology, a curative treatment remains nonexistent. One hallmark of IVD degeneration that offers a promising therapeutic approach is the decrease in progenitor cell populations 12 and overall NP cell populations. 10,12,13 Promising developments in cell transplantations offer opportunities to replenish the degenerating IVD with new and active cell populations. Different cell types have been explored as a source for IVD regeneration (reviewed in Sakai and Schol 14 ), including mesenchymal stem cells (MSCs), 15,16 chondrocytes, 17 and IVD-derived cells. 18,19 These in-human studies showed positive effects with the introduction of new cell populations, by halting or limiting the degenerative process. However, selecting the right cell source is crucial for producing optimal results, as was exhibited by the first in-human IVD cell therapy study. 20 In this study of Haufe and Mork, hematopoietic stem cells were injected into the IVD of 10 patients, which led to no observed improvement in pain reporting or radiographic findings 1-year posttransplantation. 20 22 Liu et al. applied genetically modified NP cells, designed to overexpress connective tissue growth factor or tissue inhibitor of metalloproteinases which after transplantation resulted in maintained disc features and increased expression of ECM components. 23 The first in-human assessment by Meisel et al. applied autologous NP cells to limit degeneration onset from disc herniation treated by discectomy, which resulted in improvement in low back pain and disc hydration, contradicting the result from Hegewald et al. 18,21,24 In a later in-human trial by Mochida et al., applied transplantation of autologous NP cells reactivated by MSC coculture in IVDs adjacent to a fused IVD resulted in a lack of degeneration progression in the treated discs. 19 Overall, these studies demonstrate the safety and potential of the transplantation of IVD-derived cells as a treatment of IVD degeneration.
In order to enhance clinical translatability and reduce overall costs of a potential cell therapy, the product would ideally comprise an off-the-shelf (OTS) treatment. It has been shown that discogenic cells can be reactivated postcryopreservation; 22,25 however, this would still require additional culture of the discogenic cells prior to application. Here, we describe a novel OTS strategy for the transplantation of discogenic cells. 26 The cells derived from human NP tissue were cultured for 4 to 6 weeks, and then frozen for long-term storage in the vapor-phase of liquid nitrogen. This study aims to assess the applicability of transplanting human discogenic cells directly from their cryopreserved storage condition as a treatment for IVD degeneration. In a canine disc degeneration model, safety and efficacy of this OTS approach were assessed and compared to the treatment of identical discogenic cells precultured for 2 weeks prior to delivery.
Radiographic and histological images were used to assess their regenerative capacity.

| Disc degeneration model
This study was conducted in accordance with protocols approved by the Tokai University School of Medicine committee for safe animal experimentation (162031). Ten female chondrodystrophic beagle dogs (Kitayama Lab Co. Ltd., Nagano, Japan) were obtained at 12 months of age with an average weight of 8.5 AE 0.46 kg. Canine subjects were all in good conditions and plain X-ray imaging confirmed the absence of lumbar disease. By intramuscular injection of 0.4 mg/kg midazolam (Astellas Pharma Inc., Tokyo, Japan), 0.02 mg/kg medetomidine (Kyoritsu Seiyaka Corp., Tokyo, Japan), and 0.4 mg/kg butorphanol tartrate (Meiji Seika Pharma Inc., Tokyo, Japan), canine subjects were rapidly sedated followed by continuous anesthetization of 2.5% isoflurane inhalation (Pfizer, New York City, New York, USA). Lumbar discs L5/6, L4/5, and L3/4 were exposed using a left anterior retroperitoneal approach. Under a fluoroscopy-guided approach, an 18-gauge needle with stopper was inserted through the annulus fibrosus (AF) into the NP. Through suction with a 10-mL syringe, approximately 25 mg wet weight NP tissue was aspirated. The induced degeneration was allowed to progress for 2 weeks prior to treatment. 27

| Cell culture
Human adult IVD tissue was procured from a single organ donor

| Cell transplantation
Two weeks following NP aspiration, canine recipients were rapidly

| Hematology screening
Prior to NP aspiration and 4 weeks after cell transplantation, approximately 6 mL of blood was obtained and evaluated by the Health Science Research Institute, Inc. (Yokohama, Japan). A hematology, coagulation, and clinical chemistry assessment was performed for each dog.

| Radiological assessment
Plain lumbar X-ray images were obtained from all canines 2 weeks prior, during, and 4, 8, and 12 weeks after cell transplantation. Images were taken by fluoroscopic imaging intensifier DHF-105CX (Hitachi, Tokyo, Japan) with 80 kV, 2 mA at a distance of 100 cm. Images were uploaded in OsiriX-lite (Pixmeo SARL, Bernex, Switzerland) and IVD height and vertebral height were converted to disc height index (DHI) ( Figure 1) according to published protocols. 28 In short, from a medial plane at 3 points, the height of involved IVD and neighboring vertebrae was determined. DHI was determined as 2 × ((C1 + C2 + C3)/

| Tissue explantation
Twelve weeks after cell transplantation, canines were rapidly sedated as previously described, followed by an excess intravenous injection of 300 mL 50 mg/mL P0776 pentobarbital sodium salt (Tokyo Chemical Industry Co., Tokyo, Japan). The lumbar sections of the spine were excised and both manipulated and control IVDs were macroscopically detached. Excess muscle, nerve, and fat tissue were manually removed. Tissue explants were transferred and stored at 4 C in 4% paraformaldehyde for 7 days, followed by 4 C decalcification solution A (Wako, Tokyo, Japan) for 7 days. Additionally, an array of organs was explanted from 2 randomly assigned dogs from both high-and low-dose recipients and macroscopically examined for tumor formation or other apparent abnormalities (Table S1). ing was used instead of alcian blue/Picrosirius Red staining to assess proteoglycan matrix abundance. 29 Finally, tissue explants (Table S1) were similarly stained by HE to screen for histological abnormalities.

| Immunohistochemistry
The presence of reminiscent human cells was evaluated by fluorescent immunohistochemistry staining targeting human leukocyte antigen (HLA) complexes of human discogenic cells. Sections were

| Statistics
Significance for nontemporal assessments was determined by ordinary 1-way analysis of variance (ANOVA) followed by Tukey's multiple comparisons. Temporal assessments were determined by 2-way ANOVA and corrected by Tukey's multiple comparisons. Differences of P < .05 are considered statistically significant.

| Disc degeneration model
Radiographic images, taken 2 weeks prior to transplantation, did not display any form of IVD associated diseases.

| General findings
No direct injection-related complications or surgical-complications were observed, with the exception of 1 dog requiring additional stitching 1-week posttransplantation due to the opening of the suture. The canines showed a slight decrease in average body weight 4-week posttransplantation; however, body weight showed a nonrelevant increase throughout the rest of the study ( Figure S1).

| Hematology
Blood values, obtained 4 weeks after transplantation, indicated a statistically significant (P = .041) decrease in serum phosphorus levels (

| IVD gross observations
After paraformaldehyde fixation and decalcification, nonmanipulated control discs maintained a white-glossy appearance where the gellike NP structure remained distinguishable from the AF (Figure 2).
Sham control samples lost their glossy appearance and the differentiation between NP and AF was diminished to completely lost. The CUL-LOW treated disc resulted in discoloration of most discs, with 1 disc presenting apparent necrotic zones in the adjoining vertebrae ( Figure S2). Moreover, a distinctive AF-NP border was diminished in

| Histology
HE-stained IVD sections confirmed the distinction of NP and AF in the L6/7 control discs as observed by macroscopic examination (Figure 3). Sham-treated discs displayed a loss of AF-NP differentiation, ranging from a loss of a distinguishable border to a uniformly fibrous-like structure. Treated discs presented mixed outcomes, ranging from an unaffected appearance to a complete transformation to a fibrotic structure. Similar to the macroscopic observation, the OTS-LOW treated disc appears to maintain a distinctive NP core and layered AF structure, as observed by HE staining. Notably, however, the intensity of the Safranin-O from the high-dose group (include sham and control) was much more profound compared to the low-dose group, despite concurrent staining of all samples. We hypothesize that these differences are due to different batches of fixation and decalcification reagents. Therefore, all comparisons were made relative to the internal control discs. OTS-LOW Safranin-O intensity was Additionally, an array of tissue biopsies (Table S1), including IVD biopsies, was explanted, which revealed no tumor formation or apparent abnormalities (data not shown).

| DHI assessment
Following injury, radiographic images revealed a significant decrease in IVD height as signified by %DHI values (Figure 6). Two weeks after NP aspiration, the average DHI of aspirated discs was reduced by 14.2% AE 6.6%. The sham control displayed a continuous decrease in DHI, with a final reduction of 21.9% AE 6.2% at week 12. As expected, the nonmanipulated control group did not experience significant changes to DHI throughout the study. Transplantation of both CUL-LOW and CUL-HIGH showed a trend of diminished %DHI reduction compared to the sham control. OTS-HIGH demonstrated more potent inhibition of %DHI reduction than the CUL groups.
OTS-LOW treatment had the best outcome, with a significantly (P = .0107) higher %DHI compared to the sham control, and showed a trend of slight DHI recovery.

| DISCUSSION
The primary goal for this study was aimed at demonstrating the overall safety of the direct use of a cryopreserved discogenic cell

| Disc height index
Disc height measurements revealed a successful induction of disc degeneration following NP aspiration. Previous investigations 22,27,33 utilized the same canine model and found that transplantation of other cell types could limit or halt the rate of DHI reduction, but were unable to demonstrate restoration of the disc height. In contrast, the current study showed that one of the conditions, OTS-LOW, displayed a trend of improved DHI and was statistically higher than   Disc height index (DHI) assessment. A, Average DHI determined for all discs prior and 2 weeks after degeneration induction. "Control" group represents nondegeneration L6/7 discs and "degeneration" are the collected L3/4, L4/5, and L5/6 DHI values. B, Average DHI tracked over 12 weeks. All values are set as a fraction of the DHI measurements prior to manipulation and subsequently placed relative to the internal control. Weeks represent the time after cell transplantation. Significance is determined by 2-way analysis of variance (ANOVA) and Tukey's test. Significance is signified compared to the healthy control unless stated otherwise. *P ≤ .05, **P ≤ .01, ***P ≤ .005, ****P ≤ .001. Graph demonstrates mean values (AESD)

| OTS treatment
Cell transplantation has gained significant momentum and initial cell products have reached the clinics. 34 A critical component for most cell transplantation products is the use of cryopreservation, and therefore, the effects of cryopreservation have been well studied. 35 The effect of cryopreservation on MSCs 36,37 and hematopoietic stem cells, 38,39 related to the potential as cell therapy products, overall demonstrates a minimal effect on cell viability and bioactivity. For example, cryopreservation of adipose tissue-derived MSCs did not result in any alterations in morphology, growth rate, karyotype, or marker expression. As well, there was no change in the capacity to react to stimuli or differentiation potential. 36 Similar results are observed with hematopoietic stem cells. 40 Moreover, previous work from Tanaka et al. 25 demonstrated that cryopreserved NP cells possessed similar ECM production potential and proliferation rates after an in vitro coculture system compared to nonfrozen NP cells. Subsequently, Nukaga et al. 22 showed that implantation of cryopreserved and reactivated NP cells resulted in similar regenerative effects in a canine model. Although both studies demonstrated that NP cells preserve their cell activity after cryopreservation, they both applied a reactivation coculture system to boost cell activity postcryopreservation. In the current study, cells were either transplanted directly from a frozen state or transplanted after 2 weeks of culture. Our OTS strategy outperformed identical cells transplanted after 2 weeks of preculturing, confirming that cryopreservation has limited effect on cell potency, similar to observations from previous studies. 22,25 The superiority of the OTS treatment is likely a consequence of a loss of potency of the precultured cells, which could result in changes to the original discogenic cell phenotype. A variety of studies have indicated a change of NP cell characteristics with extended culture, which might explain the enhanced outcomes for the OTS cells. 12,41,42 In particular considering no stimulating factors were supplemented to the media. However, additional growth factor supplementation and additional culture time would significantly increase production costs and scheduling challenges between cell transplantations. Therefore, the direct application of cryopreserved, OTS discogenic cells poses a potentially promising and novel approach to the treatment of IVD degeneration.
Finally, the cell density of the transplantation has been shown to be an important factor for the outcome of the therapy. 33  dose OTS discogenic cells demonstrated a significant and strong reduction in DHI loss and attained histological scores close to nondegenerating discs. Transplantation of directly thawed cells was superior to precultured cells in maintaining IVD characteristics. In conclusion, OTS discogenic cell transplantation appears to be a safe and effective treatment against induced chondrodystrophic canine disc degeneration, suggesting the use of cell transplantation populations directly from their cryopreserved state to be a viable transplantation strategy for human use.