Accepted Manuscript

Title: Notochordal cell-derived conditioned medium (NCCM) protects human

nucleus pulposus cells from stress-induced apoptosis

Author: Arne Mehrkens, Ajay Matta, M. Zia Karim, Sarah Kim, Michael G.
Fehlings, Stefan Schaeren, W. Mark Erwin

PII: S1529-9430(17)30003-7
DOI: http://dx.doi.org/doi: 10.1016/j.spinee.2017.01.003
Reference: SPINEE 57232

To appear in: The Spine Journal

Received date: 11-8-2016

Revised date: 8-12-2016
Accepted date: 5-1-2017

Please cite this article as: Arne Mehrkens, Ajay Matta, M. Zia Karim, Sarah Kim, Michael G.
Fehlings, Stefan Schaeren, W. Mark Erwin, Notochordal cell-derived conditioned medium
(NCCM) protects human nucleus pulposus cells from stress-induced apoptosis, The Spine Journal
(2017), http://dx.doi.org/doi: 10.1016/j.spinee.2017.01.003.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service
to our customers we are providing this early version of the manuscript. The manuscript will
undergo copyediting, typesetting, and review of the resulting proof before it is published in its
final form. Please note that during the production process errors may be discovered which could
affect the content, and all legal disclaimers that apply to the journal pertain.
 1 Title: Notochordal cell-derived conditioned medium (NCCM) protects human nucleus

2 pulposus cells from stress-induced apoptosis

3 Authors:

4 Arne Mehrkens§1,2, email: arne.mehrkens@usb.ch

5 Ajay Matta§1, email: amatta@uhnresearch.ca

6 M. Zia Karim§1, email: mkarim@uhnresearch.ca

7 Sarah Kim1, email: sarah.js.kim@gmail.com

8 Michael G. Fehlings1,3, email: Michael.Fehlings@uhn.ca

9 Stefan Schaeren2, email: stefan.schaeren@usb.ch

10 W. Mark Erwin*1,3,4, email: mark.erwin@utoronto.ca

11 §Authors contributed equally to the manuscript.

12 *Corresponding and senior author

13 Affiliations:

1
14 Krembil Research Institute, Toronto Western Hospital, 60 Leonard Avenue, Toronto, ON, M5T

15 2S8, Canada

2
16 Spine Surgery, University Hospital Basel, Spitalstr. 21, CH-4031 Basel, Switzerland

3
17 Division of Neurosurgery and Spine Program, University of Toronto, Toronto Western Hospital,

18 399 Bathurst Street, Toronto, ON, M5T 2S8, Canada

1
Page 1 of 24
 4
1 Division of Orthopaedic Surgery, University of Toronto, Toronto Western Hospital, 399

2 Bathurst Street, Toronto, ON, M5T 2S8, Canada

3 Corresponding Author: W. Mark Erwin DC, PhD, Krembil Research Institute, Toronto

4 Western Hospital, 60 Leonard Avenue, Toronto, ON, M5T 2S8,Canada.

5 Email: mark.erwin@utoronto.ca; Tel: 416-603-5800 ext 3308

7 ABSTRACT

9 Background Context: Degenerative Disc Disease (DDD) remains without an effective therapy

10 and presents a costly burden to society.

11 Purpose: Based upon prior reports concerning the effects of notochordal cell conditioned

12 medium (NCCM) upon disc cells, we performed a proof of principle study to determine whether

13 NCCM could reduce cytotoxic stress induced apoptosis in human disc nucleus pulpous (NP)

14 cells.

15 Study Design/Setting: 'in vitro' fundamental/basic science study

16 Methods: NP cells derived from 15 patients undergoing spinal surgery were treated with

17 interleukin 1β (IL-1β) and Fas ligand (FasL) or etoposide in the presence of NCCM. We

18 determined pro or anti-apoptotic events, using activated Caspase assays and determined genomic

19 regulation of apoptosis using PCR-arrays validated using western blotting methods. We

20 interrogated cellular apoptotic regulation using JC-1dye and flow cytometry and performed

21 enzyme linked immunosorbent assays (ELISA) to evaluate NP inflammatory cytokine secretion.

22 Disclosures: The authors declare no conflicts of interest and no financial disclosures related to

23 this work

2
Page 2 of 24
 1 Results: NCCM inhibits cytotoxic stress induced caspase—9 and -3/7 activity and maintains the

2 mitochondrial membrane potential in human NP cells, thereby suppressing the intrinsic apoptotic

3 pathway. Gene expression analysis revealed the X-linked inhibitor of apoptosis protein (XIAP)

4 as a key player responsible for evading etoposide induced apoptosis in the presence of NCCM

5 and we verified these data using western blotting. ELISA results revealed distinct differences in

6 interleukin-6 and interleukin-8 secretion by NP cells in response to etoposide in the presence of

7 NCCM.

8 Conclusions: Here we demonstrate for the first time that NCCM reduces cytotoxic stress

9 induced apoptosis in human NP cells. Soluble factors present in NCCM could be harnessed for

10 the development of novel therapeutics for treatment of DDD.

11

12 Key Words: Notochordal cell, human nucleus pulposus, apoptosis, cell proliferation,

13 degenerative disc disease

14

15 BACKGROUND

16

17 Progressive degeneration of the intervertebral disc (IVD) is a common cause of spinal

18 pain, neurological impairment and disability and represents a costly burden to the healthcare

19 system [1-3]. The hallmark of intervertebral (IVD) degenerative disc disease (DDD) is the

20 failure to maintain homeostatic regulation of the IVD leading to progressive cell death within the

21 nucleus pulposus (NP) [4-6]. Interestingly, the persistence of notochordal cells within the NP of

22 certain animals such as the non-chondrodystrophic canine (NCD) subspecies (mongrel dog) is

23 thought to contribute uniquely to resistance to DDD [7, 8]. This is in dramatic contrast to

3
Page 3 of 24
 1 naturally occurring DDD in the chondrodystrophic canine subspecies (CD) such as beagles,

2 dachshunds, and humans wherein the degenerative disc nucleus pulposus is largely acellular or

3 sparsely populated with chondrocyte like cells (CLCs) [7, 9]. Degeneration of the IVD is

4 associated with degradation of the extracellular matrix (ECM) and increased death of nucleus

5 pulposus (NP) cells [10, 11]. NP cells are referred to as type II apoptotic cells and undergo

6 apoptosis via two independent but associated pathways: the intrinsic mitochondrial pathway and

7 the extrinsic Fas receptor dependent pathway [12, 13]. The intrinsic mitochondrial-dependent

8 apoptotic pathway is activated by various cellular stress such as inflammation or genotoxic stress

9 leading to the loss of mitochondrial membrane integrity and subsequent release of cytochrome C,

10 apoptosis inducing factor (AIF), endonuclease (Endo G), and second mitochondria - activator of

11 caspase (Smac) into the cytoplasm. Stress induced apoptosis is the major cause of cell death in

12 the degenerative disc NP owing to its inflammatory microenvironment rich in pro-inflammatory

13 cytokines including IL-1β, IL-6, IL-8, IL-10, IL-17α, TNFα and IFN [14]. The release of

14 cytochrome C into the cytoplasm initiates a cascade of events whereby the initiator pro-caspase-

15 9 binds to apoptotic protease activating factor-1 (Apaf-1) that forms the apoptosome, which in

16 turn activates the final executioner caspase-3 leading to apoptotic cell death [15, 16]. In contrast,

17 in type 1 cells, the members of the tumor necrosis factor (TNF) - receptor superfamily classically

18 induce binding of Fas ligand (FasL) to the Fas receptor leading to activation of caspase-8 and

19 ultimately caspase-3/7 via direct proteolytic processing. However in some cells (such as IVD

20 NP cells) a form of cross talk occurs whereby the mitochondrial pathway is activated subsequent

21 to Fas receptor activation owing to cleavage of the BH-3 only protein ‘BID’ resulting in

22 BAX/BAK-dependent cytochrome c mitochondria release, apoptosome formation, capase-9

23 activation and downstream caspase -3/-7 activation and cell death [17]. In addition, the

24 increased expression of cell death ligand, FasL and its receptors has also been reported in human

4
Page 4 of 24
 1 degenerative disc NP [18]. Thus, the loss of NP cells due to apoptosis through multiple pathways

2 plays a critical role in the pathology of human DDD. Therefore, identification of factors that

3 have the ability to rescue stress-induced apoptosis in NP cells could provide an important

4 therapeutic strategy to treat DDD.

5 In the search for such biologically active molecular therapeutics, it has been reported

6 that notochordal cell (NC)-rich canine nucleus pulpous-derived notochordal cell conditioned

7 medium (NCCM) conferred protection to bovine NP cells exposed to the pro-inflammatory

8 cytokine, IL-1β in combination with FasL, a cell death promoting ligand [19]. This unique

9 phenomenon has led to a number of studies whereby the in vitro effects of NCCM have been

10 evaluated for its pro-anabolic, anti-catabolic, and anti-apoptotic effects upon NP cells [8, 19-22].

11 As a next step following this previous work, we elected to assess the effects of NCCM on human

12 disc nucleus pulpous cells collected from patients undergoing spine surgery with respect to anti-

13 apoptotic effects and a mechanistic determination of the signaling pathways involved. Here for

14 the first time we report that NCCM can suppress cytotoxic stress induced apoptosis in human NP

15 cells and identified the mechanisms through which these effects are mediated.

16

17 MATERIALS AND METHODS

18

19 Generation of Notochordal Cell Conditioned Medium (NCCM). We used

20 Advanced Dulbecco’s Modified Eagle Medium/F-12 (Advanced DMEM/F-12, Gibco Life

21 Technologies, Grand Island, NY, USA) supplemented with penicillin/streptomycin (10000

22 units penicillin and 10mg streptomycin/mL (SIGMA-ALDRICH, St. Louis, MO, USA)),

23 200mM L-Glutamine (Gibco Life Technologies, Grand Island, NY, USA) as basal medium

24 for all tissue culture experiments. For most experiments, this basal medium was

5
Page 5 of 24
 1 supplemented with fetal bovine serum (FBS; Gibco) to a concentration of 1%, 2% or 8%

2 FBS. These various culture media are referred to as ADMEM, ADMEM 1%, ADMEM 2%,

3 and ADMEM 8% (see supplemental Figure 1 for a summary of the experimental plan).

4 We developed conditioned medium from the notochordal cell rich non-

5 chondrodystrophic (NCD) canine nucleus pulposus using established protocols [19, 23].

6 Briefly, 10 NCD-canines (5 male and 5 female, age between 12 and 18 months) were

7 euthanized humanely as described earlier [21]. The thoracic and lumbar spines were

8 removed and the nucleus pulposus (NP) from each motion segment was recovered under

9 aseptic conditions. As described earlier, 2 - 3 NPs were placed in a 40μm cell strainer within

10 wells of a 6-well tissue culture plate supplemented with 6 ml of ADMEM 2% under hypoxic

11 conditions (3.5%O2, humidit , 7C). The medium from each well was collected every

12 24h, pooled, centrifuged at 500xg, filtered using 0.2μm filters (Corning, USA) and

13 subsequently stored at -80°C until further use. Only medium collected from day 3 until day

14 7 was used for the experiments. This conditioned medium is hereafter referred to as NCCM.

15 Patient Samples and Nucleus Pulposus (NP) Cell Culture. This study was

16 approved by the local research ethics board. Following informed consent, we collected

17 intervertebral disc NP samples from 15 donors (8 females and 7 males) undergoing surgery for

18 microdiscectomy or discectomy and fusion (Table 1). Patients undergoing surgery for

19 trauma, malignancy or infection were excluded. The tissue samples were collected in culture

20 medium, immediately following surgery, and washed with phosphate buffered saline (PBS,

21 pH = 7.2, 1X) under sterile conditions. Any tissue not resembling the nucleus pulposus

22 macroscopically (under 2x magnification) was removed. The remaining tissue was

23 sequentially digested with 0.2% Pronase (Roche Diagnostics, Mannheim, Germany) for 90

24 minutes followed by 0.015% collagenase type II (Life Technologies, CA, USA) for 20h as

6
Page 6 of 24
 1 described earlier [21, 23]. The digested material was filtered using a 70μm cell strainer and

2 human NP cells were cultured in ADMEM 8% under hypoxic conditions (3.5% O2 and 5%

3 CO2) at 370C.

4 Treatment Conditions and Induction of Apoptosis. We cultured all human NP

5 cells (P2) for 24 h in ADMEM 8% for 24h. Following serum starvation, cells were either

6 cultured in ADMEM 2% or NCCM. We induced apoptosis using a combination of 10

7 ng/mL recombinant human IL-1β (PeproTech, Rock Hill, NJ, USA) plus 100 ng/mL of

8 recombinant human Fas-ligand (FasL) (PeproTech, Rocky Hill, NJ, USA) [19, 24] or

9 300µM of the known cytotoxic agent etoposide (SIGMA-ALDRICH, St. Louis, MO, USA)

10 used in chemotherapy. Overall, six different treatment conditions were evaluated (ADMEM

11 2%, ADMEM 2% plus IL-1β and FasL and ADMEM 2% plus etoposide as well as NCCM,

12 NCCM plus IL-1β and FasL and NCCM plus etoposide – experimental design see

13 supplemental Fig 1).

14 Caspase Activity Assays for Evaluating Apoptosis. We plated human NP cells (p2)

15 in 96-well culture plates and treated as described above. After performing a number of pilot

16 studies whereby we evaluated apoptosis at 12, 24, 36, and 48 hours, we chose to use 24h as the

17 experimental period since this time point appeared to be optimal (Data not shown). At the end

18 of the treatment we determined activated caspase-3/7, caspase-9 and caspase-8 activity

19 using the Caspase-Glo® assay kit reagents (substrate and buffer) following the

20 manufacturer’s instructions (Promega, Madison, WI, USA). With each of the experiments

21 performed in quadruplicate and repeated at least twice.

22 Mitoprobe JC-1 Dye Assay. We used the Mitoprobe JC-1 dye assay using

23 immunofluorescence microscopy and flow cytometry in order to evaluate any alteration in

24 mitochondrial membrane potential in human NP cells (Life Technologies, NY, USA).

7
Page 7 of 24
 1 For immunofluorescence, 5x104 cells were plated over glass coverslips in 24-well plates and

2 treated as described for 24 h. Images were captured using a U2000 Nikon inverted fluorescence

3 microscope with appropriate filters.

4 Quantitative Reverse Transcriptase Polymerase Chain Reaction (qRT-PCR). At

5 the end of the 24 h treatment after induction of cell death, we used the RT² Profiler™ Human

6 Apoptosis PCR Array (384 well format containing 84 genes associated with apoptosis) to

7 evaluate the expression of apoptotic-related genes in donor #H1 and compared NCCM

8 treatments with no treatment controls (NTC) using real time PCR. Data analysis including

9 calculation of Ct values, fold changes and p-values was performed using software available

10 online (https://www.qiagen.com/ca). Fold-Regulation represents fold-change results in a

11 biologically meaningful way. Fold-change values greater or less than 1.5 were considered as an

12 up-regulation or down-regulation for further analysis.

13 Protein Detection. In order to confirm increased gene expression of XIAP, we

14 performed western blotting on human NP cells derived from 3 different patients that were treated

15 as described above. Equal amount of whole cell lysates (30μg) were prepared and incubated with

16 rabbit polyclonal anti-XIAP (ab2541, Abcam Inc., Canada, 1:200 dilution) or mouse monoclonal

17 β-actin antibody (ab8226, Abcam Inc., Canada,1:1000 dilution) at 4oC overnight. The blots were

18 then incubated with either rabbit / mouse HRP-conjugated anti-IgG secondary antibodies

19 (BioRad, CA). Protein bands were detected by the enhanced chemiluminescence method

20 (BioRad, CA) on Kodak Hyperfilm.

21 Cytokine Profile. We used the multianalyte cytokine profiling ELISA kit (Qiagen, ML,

22 USA) to detect the levels of 12 cytokines including IL-1α, IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10,

23 IL-12, IL-17α, IFN, TNFα and GM-CSF in media obtained from each treatment as described

8
Page 8 of 24
 1 above to determine if and which cytokines are secreted by NP cells under the aforementioned

2 treatment conditions. Each experiment was performed in triplicate and data has been reported as

3 mean ±S.D.

4 Statistical Analysis. All data are expressed as means ± SD for apoptosis assays

5 and ELISA. The significance of differences was analyzed using paired Student’s t-test.

6 Statistical analysis was performed using the Graphpad prism. A p-value < 0.05 was defined as

7 statistically significant for all tests.

8 A more detailed description of the methods can be found in materials and methods supplement.

10 RESULTS

11

12 NCCM Rescues Human Nucleus Pulposus Cells from Etoposide-mediated

13 Apoptosis.

14 Treatment of human NP cells with a combination of IL-1β (10ng/mL) and FasL (100 ng/mL)

15 failed to induce any significant caspase-3/7 activity within 24h in human NP cells (supplemental

16 Fig. 2). However, a significant increase in caspase-3/7 activity was observed in 14 out of the 15

17 (93%) cases following treatment with etoposide ( 00 μM) as earl as in 24h (Fig. 1).

18 NP cells treated with NCCM showed a significant reduction in etoposide induced

19 caspase-3/7 activity (p < 0.05) in 7 of the 14 cases that responded to treatment with etoposide

20 alone (Fig. 1). Hereafter, we refer to donor cells that exhibited suppressed activated caspase-3/7

21 activity after treatment with NCCM as ‘responders’ and those that did not as ‘non-responders’.

22 In 3 of 7 cases that demonstrated reduced activated caspase -3/7 we also observed a statistically

23 significant decrease in caspase-9 activity (Fig 2.). Two additional cases showed a high degree of

24 reduced activated caspase-9 activity but did not achieve statistical significance suggesting that

9
Page 9 of 24
 1 the components within NCCM can target the intrinsic mitochondrial pathway to rescue stress-

2 induced apoptosis in human NP cells.

3 The mitochondrial apoptotic pathway is dependent upon damage/alterations in the

4 mitochondrial permeability transition pore (MPTP) potential and release of cytochrome C into

5 the cytoplasm. Flow cytometry revealed that in 4 of the 6 cases analyzed (H1, H2, H7, H8), there

6 was a significant alteration in MPTP, as revealed by increased percentage (5-10%) of cells

7 showing green fluorescence in comparison to no treatment controls (1-2%). Notably, in the 4

8 cases mentioned above, the percentage of human NP cells showing green fluorescence

9 significantly decreased on treatment with etoposide in presence of NCCM (1-2%) as compared

10 to etoposide alone (p < 0.01, Fig. 3).

11 We verified these data using immunocytochemistry and the Mitoprobe JC-1dye. This

12 assay yields green fluorescence when the MPTP is compromised and red when the MPTP is

13 preserved. Human NP cells treated with etoposide alone showed increased green fluorescence,

14 with no treatment control cells showing exclusively red fluorescence. We show an example of a

15 responder case (H2) with such green fluorescence when treated with etoposide but when NCCM

16 was added to etoposide red fluorescence is detected indicating preservation of the MPTP. In

17 contrast, a non-responder (H10) depicts green fluorescence under both conditions indicating a

18 loss of the MPTP (Fig. 4) confirming our flow cytometry data.

19 NCCM Increases XIAP Expression in NP Cells. Statistical analysis of 84 apoptosis-

20 related genes in human NP cells treated with etoposide alone as compared to normal treatment

21 controls (NTCs) revealed significant alterations in the expression of 25 genes (12 upregulated

22 and 13 downregulated) (p < 0.05, Suppl. Fig. 3a, Suppl. Table 1a). On the other hand, human

23 NP cells treated with etoposide in the presence of NCCM revealed alterations in expression of

24 only 10 genes (6 upregulated and 4 downregulated) as compared to etoposide treatment only (Fig

10
Page 10 of 24
 1 5a, Suppl. Fig. 3b, Suppl. Table 1b). Interestingly, mRNA levels of an important anti-apoptosis

2 protein, XIAP, were higher in human NP cells treated with etoposide in presence of NCCM as

3 compared to etoposide treatment only (Fig. 5a, Suppl. Table 1b). A complete list of the analyzed

4 genes and results can be found in supplemental table 2.

5 We confirmed the increased expression of XIAP in human NP samples rescued from

6 etoposide induced apoptosis by NCCM using Western blots (Fig. 5b). Of note, no change in

7 XIAP expression was observed in human NP samples that did not respond to NCCM (i.e. donor

8 H8, Fig. 5b). These findings demonstrate the role of NCCM in interfering with apoptotic

9 signaling at two independent steps in human NP cells.

10 Cytotoxic Stress Induces Alterations in Cytokine Profile of human NP Cells. We

11 demonstrated that NP cells secreted measureable levels of IL-6 and IL-8 with no change in other

12 cytokines or measurements barely within the detectable level using ELISA (Fig. 6). Among the

13 10 cases analyzed (5 responders and 5 non-responders), 50% of the samples (1 responder and 4

14 non-responders) demonstrated higher levels of IL-6 secreted by NP cells when treated with

15 etoposide alone or a combination of etoposide and NCCM as compared to no treatment controls

16 (Fig. 7a). On the other hand, the NP cells secreted higher levels of IL-8 in all but one of the ten

17 cases (H2), irrespective of the treatment with etoposide alone or a combination of etoposide plus

18 NCCM. Interestingly, the NP cells in four of the ten cases (all of them responders) secreted a

19 significant increase in the levels of IL8 when treated with etoposide plus NCCM as compared to

20 etoposide alone (Fig. 7b).

21

22

23 DISCUSSION

24

11
Page 11 of 24
 1 The characteristic phenotype of the degenerative disc is one where under the influence of

2 inflammation and cytotoxic stress, NP cells fail to maintain a normal homeostatic tissue

3 microenvironment leading to the development of a fibrocartilaginous, degenerative extracellular

4 matrix accompanied by the loss of viable cells [19, 24-26]. Currently there are no disease

5 modifying, bioactive therapies available that can surmount the cytotoxic and pro-inflammatory

6 stresses leading to DDD, thus emphasizing the need for novel restorative agents. Previously,

7 using in vitro methods, it has been demonstrated that IL-1β+Fas ligand-induced bovine NP cell

8 apoptosis was rescued by NCCM and theorized that components of NCCM must influence the

9 intrinsic caspase system since we observed suppression of activated caspase -9 and -3/-7 in these

10 studies [19]. Here we showed that NCCM confers anti-apoptotic effects upon a sub-group of

11 human NP cells obtained at the time of spinal surgery, treated with the known cytotoxic agent

12 etoposide: a chemotherapeutic drug that causes cellular apoptosis by inducing double strand

13 breaks in deoxyribonucleic acid (DNA) [27]. Etoposide activates p53-dependent signaling that

14 leads to cell cycle arrest and induction of apoptosis via the intrinsic, mitochondrial pathway. In

15 keeping with prior investigation using bovine NP cells, our results demonstrated that in certain

16 human donors, NCCM is capable of acting within the intrinsic pathway by lowering activated

17 caspase -9 activity at least in part, by preserving the mitochondrial membrane potential.

18 Since we were interested in determining the potential lynchpins of anti-apoptotic

19 regulation in NP cells treated with NCCM, we performed preliminary genomic regulation

20 experiments by examining genes involved in cell death under etoposide +/- NCCM. Under these

21 conditions we observed upregulation of anti-apoptotic and down regulation of pro-apoptotic

22 regulatory genes in the presence of NCCM. Notably, in the presence of etoposide, we

23 determined that NCCM upregulated the genomic expression of an important member of the

24 inhibitory apoptotic protein family (IAPs) the X-linked inhibitor of apoptotic protein (XIAP).

12
Page 12 of 24
 1 We validated these findings at the protein level using Western blotting methods and

2 demonstrated that in responder cells, NCCM increased XIAP protein levels helping NP cells to

3 evade etoposide induced apoptosis.

4 The IAP’s contain an amino acid motif known as the baculoviral inhibitor of apoptosis

5 repeat (BIR) domains capable of binding to and inhibiting the activation of several caspases [28].

6 The IAP gene family members are the master regulators and most robust effectors of anti-

7 apoptotic signaling, capable of blocking both the mitochondrial and death receptor apoptotic

8 pathways [29]. However, as evidence of the tight regulation of this pathway, in some cases pro-

9 apoptotic proteins such as Smac/DIABLO may be released from mitochondria under the

10 influence of cytotoxic stress and bind to XIAP, thereby inhibiting its anti-apoptotic activity [29,

11 30]. Furthermore, depending upon the activity of specific Bcl-2 family related proteins, cells

12 under various conditions may either abort or commit to the apoptotic pathway indicating that

13 decisions regarding cell death are not binary but rather can be regulated by downstream

14 inhibitory or feedback dis-inhibitory molecules or other intersecting pathways [16]. We have

15 demonstrated that NCCM can alter the expression of several of these apoptotic related proteins in

16 human NP cells leading to suppressed apoptosis within the otherwise complex and tightly

17 regulated apoptotic process (Fig. 9). However, further investigation is required in order to better

18 understand the genomic/molecular mechanisms whereby the bioactive components of NCCM

19 suppress stress-induced apoptosis.

20 We assume that NP cellular responses are likely influenced by the pro-inflammatory

21 microenvironment of a degenerative NP that in addition to the disease phenotype may in turn be

22 influenced by the surgical procedure, microenvironment and heterogeneous cell type obtained at

23 the time of surgery. Senescent cells produce diminished and impaired extracellular matrix

13
Page 13 of 24
 1 (ECM) molecules, promote senescence of neighboring cells and the increased secretion of pro-

2 inflammator c tokines such as TNF-α, IL-1 α/β, IL-6, IL-17, IL-8, IL-2, IL-4, IL-10, IFN-γ,

3 chemokines, prostaglandin (PGE)2, and COX-2; the net result drives further degeneration [14,

4 31]. In this respect, we observed that cytotoxic stress had a strikingly different effect upon NP

5 cells obtained from responders vs non-responders in that responders secreted barely detectable

6 levels of IL-6 that was only marginally increased by the addition of NCCM. However, NP cells

7 obtained from non-responders treated with NCCM and etoposide secreted markedly higher levels

8 of IL-6 as compared to responder cells.

9 At least one reason for the difference in response to NCCM by responder vs non-

10 responders might be the degree to which non-responder cells were influenced by the

11 degenerative NP milieu and the process of senescence. It is known that senescent cells become

12 largely unresponsive to growth factors however, they can remain viable for extended periods.

13 Senescent cells are also known to shift from a matrix synthesis to matrix degrading phenotype

14 that could explain in part, why some cells did not respond to NCCM treatment and for their

15 difference in secretion of the pro-inflammatory cytokines IL-6 and -8 [31, 32]. Furthermore, it

16 is possible that stressed cells and/or cells undergoing senescence may well affect the expression

17 of surface receptors responsive to the essential factors contained within NCCM further

18 diminishing the cell’s responsiveness to NCCM treatment. Additionall , it is important to

19 remember that NCCM is a raw, conditioned medium that may not contain the necessary and

20 sufficient ‘dose’ of the essential molecules necessar to sufficientl activate NP cells.

21 Interestingly, in osteoarthritis (OA), elevated serum levels of IL-6 are strongly associated

22 with more severe disease on radiographic imaging and in degenerative discs, elevated levels of

23 IL-6 and IL-8 have been correlated with increased back pain [33-35]. Furthermore, in

14
Page 14 of 24
 1 osteoarthritic joints where cellular senescence is a key feature, elevated levels of IL-6 and IL-8

2 lead to decreased synthesis of extracellular matrix molecules including proteoglycans and

3 collagens that in conjunction with IL-1β and TNFα contribute to breakdown of the ECM [33,

4 36]. It is important to note that in addition to their influence upon ECM molecules, inflammation

5 and pain, IL-6 and IL-8 also play an important role in influencing cellular fate under stress and

6 that these effects are condition and cell-dependent. In our study NCCM treatment did not

7 suppress the secretion of stress induced IL-6 or IL-8 in any of the treatment conditions.

8 However, we determined that responder cells secreted much lower amounts of IL-6 as compared

9 to cells that did not respond to NCCM-possibly because responder cells were more sensitive to

10 growth factors ostensibly contained within the NCCM. It could be that patients’ IVD NPs where

11 cellular senescence dominates, NP cells respond as OA chondrocytes do; by the increased

12 secretion of pro-inflammatory cytokines that may in turn render the cells insensitive to growth

13 factors/mitogens. It is possible therefore, that the differential expression of IL-6 and IL-8 by NP

14 cells could prove to be important biomarkers for patients for whom a biologic therapy could be

15 contemplated. Additionally, several studies have shown the role of genetic factors in the

16 development and progression of DDD, thus genomic regulation (as with XIAP expression in

17 cells under stress) may play a role with respect to biologic therapy [37, 38].

18 This preliminary study offers promising results with respect to marshalling the anti-

19 apoptotic/anti-degenerative effects of notochordal cell-secreted molecules as a potential

20 biological therapy for the treatment of degenerative disc disease. In an effort to translate these

21 preliminary observations, it is vital to determine the identity and biology of the necessary and

22 sufficient bioactive molecules present within NCCM. Furthermore, delineation of a robust

23 predictive biomarker would be of tremendous value to determine the appropriate patient for such

24 biological therapy.

15
Page 15 of 24
 1 Currently, the use of provocative discography is the most reliable (and controversially

2 discussed) method of determining discogenic pain. If a suitable biological agent could

3 ameliorate DDD, the application of such an agent at the time of discography might be a prudent

4 approach. To this end, future studies might shed some light upon the potential biomarker use of

5 IL-6/IL-8 in the form of post discectomy assay of NP cytokine secretion. In the event that the

6 cytokine expression of surgically excised NP cells (such as IL-6/IL-8) display a predictive effect

7 of biologic treatment, a subsequent percutaneous deliver of such ‘rescue’ molecules could

8 enable the development of a novel, minimally invasive, biological therapy for DDD.

10 CONCLUSIONS

11

12 Our results demonstrate that under the influence of a cytotoxic drug such as etoposide,

13 conditioned medium generated by the notochordal cell-rich IVD NP has the ability to confer an

14 anti-apoptotic effect upon human NP cells. These results support our hypothesis that notochordal

15 secreted factors that help in the maintenance of a healthy, hydrophilic nucleus pulposus (once

16 defined) could be harnessed in a novel, molecular therapy for patients suffering from DDD.

17 DECLARATIONS

18 Ethics approval and consent to participate

19 All animals were obtained in collaboration with a licensed animal facility and all

20 practices were in accordance with the animal care policies and ethics approval board of

21 the Toronto Western Hospital (Re 11-0790-AE).

16
Page 16 of 24
 1 Written informed consent was obtained from the donating patients in accordance with the

2 policies of the Toronto Western Hospital on the use of samples of human tissue for research

3 purposes. The consent form is held b the authors (cop ) and in the patients’ clinical notes

4 (original) and is available for review by the Editor-in-Chief.

5 Consent for publication

6 Written informed consent was obtained from the patients for publication of their individual

7 details in this manuscript. The consent form is held b the authors (cop ) and in the patients’

8 clinical notes (original) and is available for review by the Editor-in-Chief.

9 Availability of data and material

10 All data are presented with this manuscript

11 Competing interests

12 All authors declare that they have no conflict of interest

13

14 Funding

15 This work was supported by a NASS Young Investigator Grant award.

16 Arne Mehrkens was supported by an AO Spine Europe research fellowship, Swiss orthopedics,

17 Freie Akademische Gesellschaft Basel and Bangerter-Rhyner-Stifung

18 W. Mark Erwin gratefully acknowledges partial salary support from the Canadian Chiropractic

19 Research Foundation and CMCC.

20 Authors' contributions

17
Page 17 of 24
 1 Ar.M., A.M., M.Z.K, and W.M.E. designed, performed experiments, analyzed data, wrote and

2 edited the manuscript. Ar.M., A.M., M.Z.K., and S.K. performed experiments. M.G.F. and S.S.

3 assisted with manuscript editing. W.M.E. provided all the reagents, financial support and

4 infrastructure for performing the experiments in addition to the NASS Young Investigator Award

5 to Ar.M. All authors read and approved the final manuscript.

6 Acknowledgements

7 We are thankful to Drs. Steven Lewis, Mohammad Shamji, Taufik Valiante and Michael

8 Fehlings for providing the human clinical samples used in this study.

10

11 LIST OF ABBREVIATIONS

12 Advanced DMEM/F-12 - Advanced Dulbecco’s Modified Eagle Medium/F-12

13 AIF - apoptosis inducing factor

14 Apaf-1 - activating factor-1

15 BIR - baculoviral inhibitor of apoptosis repeat

16 CLCs - chondrocyte like cells

17 DDD - Degenerative Disc Disease

18 DNA - deoxyribonucleic acid

19 DISC - death-inducing signal complex

20 ECM - extracellular matrix

21 ELISA - Enzyme linked immunosorbent assays

22 Endo G - endonuclease

23 FasL - Fas ligand

18
Page 18 of 24
 1 FBS - fetal bovine serum

2 IAP - inhibitory apoptotic protein

3 IL-1β - interleukin 1β

4 IVD - intervertebral disc

5 MPTP - mitochondrial permeability transition pore

6 NC - notochordal cell

7 NCD - non-chondrodystrophic

8 NCCM - notochordal cell conditioned medium

9 NP - nucleus pulpous

10 NTC - no treatment controls

11 PVDF - polyvinylidenedifluoride

12 qRT-PCR - quantitative reverse transcriptase polymerase chain reaction

13 RT - room temperature

14 SDS-PAGE - sodium dodecyl sulphate-polyacrylamide gels

15 Smac - second mitochondria - activator of caspase

16 TNF - tumor necrosis factor

17 TTBS - Tris-buffer saline

18 X-linked inhibitor of apoptosis protein (XIAP)

19

20

21

22

23 REFERENCES

24 1. Global burden of disease study 2013 collaborators: Global, regional, and national incidence,
25 prevalance, and years lived with disability for 301 acute and chronic diseases and injuries in

19
Page 19 of 24
 1 188 countries, 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013.
2 Lancet 2015, 386:743-800.
3 2. Hoy D, March L, Brooks P, Blyth F, Woolf A, Bain C, Williams G, Smith E, Vos T, Barendregt J et al:
4 The global burden of low back pain: estimates from the Global Burden of Disease 2010 study.
5 Annals of the Rheumatic Diseases 2014.
6 3. Waterman BR, Belmonst PJ Jr, Schoenfeld AJ: Low back pain in the United States: incidence
7 and risk factors for presentation in the emergency setting. Spine Journal 2012, 12:63-70.
8 4. Rajasekaran S, Babu J Naresh, Arun R, Armstrong B Roy Wilson, Shetty Ajoy Prasad,
9 Subramaniam M: A study of diffusion in human lumbar discs: A serial magnetic resonance
10 imaging study documenting the influence of the endplate on diffusion in normal and
11 degenerate discs. Spine 2004, 29(23):2654-2667.
12 5. Rajasekaran S, Venkatadass K, Naresh BJ: Pharmacological enhancement of disc diffusion and
13 differentiation of healthy, ageing and degenerated discs. Eur Spine Journal 2008, 17(626-643).
14 6. Urban JP, Smith S, Fairbank JC: Nutrition of the intervertebral disc. Spine 2004, 1;29(23):2700-
15 2709.
16 7. Hans-Jorgen H: A Pathologic-Anatomical Study on Disc Degeneration in Dog. Acta Orthopaedica
17 Scandinavica 1952, Suppl(11):1-117.
18 8. Aguiar DJ, Johnson SL, Oegema TR: Notochordal cells interact with nucleus pulposus cells:
19 regulation of proteoglycan synthesis. Experimental Cell Research 1999, 246:129-137.
20 9. Braund KG: Morphological studies of the canine intervertebral disc. The assignment of the
21 beagle to the achondroplastic classification. Research in Veterinary Science 1975, 9:167-172.
22 10. Adams MA, Roughley PJ: What is intervertebral disc degeneration and what causes it? Spine
23 2006, 31:2151-2161.
24 11. Buckwalter JA: Aging and Degeneration of the Human Intervertebral Disc. Spine 1995,
25 20(11):1307-1314.
26 12. Ozoren Nesrin, El-Deiry Wafik S: Defining Characteristics of Types 1 and II Apoptotic Cells in
27 Response to TRAIL. Neoplasia 2002, 4(6):551-557.
28 13. Krammer Peter H: CD95's deadly mission in the immune system. Nature 2000, 12
29 October(407):789-795.
30 14. Risbud MV, Shapiro IM: Role of cytokines in intervertebral disc degeneration: pain and disc
31 content. Nature Reviews Rheumatology 2014, 10:44-56.
32 15. Gewies A: Introduction to Apoptosis. In: ApoReview. 2003: 1-26.
33 16. Elmore Susan: Apoptosis: A Review of Programmed Cell Death. Toxicologic Pathology 2007,
34 35:495-516.
35 17. Cullen SP, Martin SJ: Caspase activation pathways: some recent progress. Cell Death and
36 Differentiation 2009, 16:935-938.
37 18. Yamamoto J, Maeno Koichiro, Takada Toru, Kakutuari K, Yurube Takashi, Zhang Z, Hirata Hiroaki,
38 Kurakawa T, Sakai D, Mochida J et al: Fas ligand plays an important role for the production of
39 pro-inflammatory cytokines in intervertebral disc nucleus pulposus cells. Journal of
40 Orthopaedic Research 2013, 31(4):608-615.
41 19. Erwin WM, Islam D, Inman RD, Fehlings M, Tsui FW: Notochordal cells protect nucleus pulposus
42 cells from degradation and apoptosis: implications for the mechanisms of intervertebral disc
43 degeneration. Arthritis Research and Therapy 2011, 13(6).
44 20. Erwin WM, Ashman K, Inman RD: The notochord cell as a key to intervertebral disc
45 homeostasis. In: 51rst Orthopaedic Research Society. Washington DC; 2005.
46 21. Erwin WM, Ashman K, O'Donnell P, Inman RD: Nucleus pulposus notochord cells secrete
47 connective tissue growth factor and upregulate proteoglycan expression by intervertebral disc
48 chondrocytes. Arthritis and Rheumatism 2006, 54(12):3859-3867.

20
Page 20 of 24
 1 22. Mehrkens A, Karim MZ, Kim S, Fehlings M, Erwin WM: Canine Notochordal Cell-Secreted
2 Factors Protect Murine and Human Nucleus Pulposus Cells from Apoptosis by Inhibition of
3 Activated Caspase-9 and Caspase -3/7. Evidence Based Spine Care Journal 2013, 4:154-156.
4 23. Erwin WM, Inman RD: Notochord cells regulate intervertebral disc chondrocyte proteoglycan
5 production and cell proliferation. Spine 2006, 31(10):1094-1099.
6 24. Cui LY, SL L, Ding Y, Huang DS, Ma RF, Huang WG, Hu BS, Pan QH: IL-1beta sensitizes rat
7 intervertebral disc cells to Fas ligand mediated apoptosis in vitro. Acta Pharmacologica Sinica
8 2007, 28:1671-1676.
9 25. Hoyland JA, Le Maitre CA, Freemont AJ: Investigation of the role of IL-1 and TNF in matrix
10 degradation in the intervertebral disc. Rheumatology 2008, 47(6):809-814.
11 26. Shamji MF, Setton LA, Jarvis W, So S, Chen J, Jing L, Bullock R, Isaccs RE, Brown C, Richardson WJ:
12 Pro-inflammatory cytokine expression profile in degenerative and herniated human
13 intervertebral disc tissues. Arthritis and Rheumatism 2010, 62(7):1974-1982.
14 27. Jakson JK, Higo T, Hunter WL, Burt HM: Topoisomerase inhibitors as anti-arthritic agents.
15 Inflammation Research 2008, 57:126-134.
16 28. Eckelman Brendan P, Salvesen Guy S, Scott Fiona L: Human inhibitor of apoptosis proteins: why
17 XIAP is the black sheep of the family. EMBO reports 2006, 7(10):988-994.
18 29. Holcik M: Translational Upregulation of the X-Linked Inhibitor of Apoptosis. Annals of the New
19 York Academy of Sciences 2003, 1010(1):249-258.
20 30. Stehlik C, de Martin R, Kumabashiri I, Schmid JA, Binder BR, Lipp J: Nuclear factor (NF)-kappa B-
21 regualted X-chromosome-linked IAP gene expression protects endothelial cells from tumor
22 necrosis factor alpha-induced apoptosis. Journal of Experimental Medicine 1998, 188(1):211-
23 216.
24 31. Feng Chencheng, Liu Huan, Yang Minghui, Zhang Yang, Zhou Yue: Disc cell senescence in
25 intervertebral disc degeneration: Causes and molecular pathways. Cell Cycle 2016,
26 15(13):1674-1684.
27 32. Kim Ki-Won, Chung Ha-Na, Ha Kee-Yong, Lee Jun-Seok, Young-Yul K: Senescence mechanims of
28 nucleus pulposus chondrocytes in human intervertebral discs. The Spine Journal 2009, 9(658-
29 666).
30 33. Burke JG, Watson RWG, McCormack D, Dowling FE, Walsh MG, JM F: Intervertebral discs which
31 cause low back pain secrete high levels of proinflammatory mediators. Journal of Bone and
32 Joint Surgery 2002, 84-B(2 March):196-201.
33 34. Wojdasiewicz Piotr, Poniatowski Lukasz, Dariusa S: The role of inflammatory and anti-
34 inflammatory cytokines in the pathogenesis of osteoarthritis. Mediators of Inflammation 2014,
35 2014(Article ID 561459).
36 35. Kepler Christopher K, Dessislava Z, Florian Dibra, Yadla Sanjay, Vaccaro Alexander R, Risbud
37 Makarand V, Albert Todd J, Anderson David Greg: Expression and relationship of
38 proinflammatory chemokine RANTES/CCL5 and cytokine IL-1B in painful human intervertebral
39 discs. Spine 2013, 38(11):873-880.
40 36. Fernandes Julio C, Martel-Pelletier Johanne, Pelletier Jean-Pierre: The role of cytokines in
41 osteoarthritis pathophysiology. Biorheology 2002, 39:237-246.
42 37. Eskola PJ, Lemmela S, Kjaer P, Solovieva S, Mannikko M, Tommerup N, Lind-Thomsen A,
43 Husqafvel-Pursiainen K, Cheung KM, Chang D et al: Genetic association studies in lumbar disc
44 degeneration: a systematic review. PLoS ONE 2012, 7(11).
45 38. Kao PY, Chan D, Samartzis D, Sham PC, YQ S: Genetics of lumbar disc degeneration: technology,
46 study designs, and risk factors. Orthopaedic Clinics of North America 2011, Oct; 42(4):479-486.

47

21
Page 21 of 24
 1 Figure Legends:

2 Figure 1: Caspase 3/7 Activity Assay

3 Histograms representing caspase-3/7 activity in human NP cells (#H1-H15) in response to

4 treatment with etoposide alone or in presence of NCCM after 24h as compared to no treatment

5 controls (NTC). Each bar presents mean+SD. ••p-value < 0.05 (NTC vs. etoposide), *p-value <

6 0.05 (etoposide vs. etoposide+NCCM).

7 Figure 2: Caspase 9 Activity Assay

8 Representative bar graphs showing caspase-9 activity in human NP cells (#H1-H9) in response

9 to treatment with etoposide alone or in presence of NCCM after 24h as compared to no treatment

10 controls (NTC). Each bar presents mean+SD. ••p-value < 0.05 (NTC vs. etoposide), *p-value <

11 0.05 (etoposide vs. etoposide+NCCM).

12 Figure 3: JC-1 Dye Assay

13 Representative bar graphs showing histograms showing % cells with alterations in

14 mitochondrial membrane potential in (i) no treatment controls (NTC), (ii) treatment with

15 etoposide and (iii) etoposide in presence of NCCM after 24h as determined by mitoprobe JC-1

16 dye assay using flow cytometry. *p-value < 0.05

17 Figure 4: Mitochondrial Membrane Potential Restoration

18 Representative immunofluorescence images of a responder (i H2) and a non-responder (ii H10)

19 showing green fluorescence in human NP cells treated with etoposide for 24h (left column, a)

20 and the restoration of mitochondrial membrane potential as represented by red fluorescence in

22
Page 22 of 24
 1 responder H2 (white arrows, top right) vs. no significant restoration in non-responder H10

2 (bottom right) after treatment with etoposide in presence of NCCM.

3 Figure 5

4 (a): Gene Expression Analysis : NCCM+Etoposide vs. Etoposide Only

5 Bar graphs showing fold changes in the gene expression of representative genes that are

6 significantly up- or downregulated on treatment with etoposide in presence of NCCM as

7 compared to etoposide alone for 24h. p-value < 0.05

8 (b): XIAP Expression

9 Western blots showing alterations in expression of XIAP in response to treatment with control

10 medium (ADMEM), NCCM, etoposide only and etoposide in presence of NCCM in human NP

11 cells, with a strong XIAP expression in presence of NCCM for responders H1 and H6 vs. very

12 weak/no expression in non-responder H8.

13 Figure 6: Cytokine Level in Culture Medium

14 Bar graph showing levels of inflammatory cytokine expression in culture medium of human NP

15 cells in representative samples from a responder (H1, (a)) and a non-responder (H12, (b)) after

16 treatment with ADMEM (no treatment control NTC, blue), NCCM (red), etoposide (green) and

17 etoposide plus NCCM (purple) for 24h.

18 Figure 7: Levels of IL-6 and IL-8 in Culture Medium

19 Bar graphs showing levels of IL-6 (a) and IL-8 (b) in the culture medium of human NP cells of

20 responders (H1, 2, 4, 6, 7) and non-responders (H9, 10, 12,13,15) treated with ADMEM (no

23
Page 23 of 24
 1 treatment control NTC, blue), NCCM (red), etoposide (green) and etoposide plus NCCM

2 (purple) for 24h.

3 Figure 8: Model of Possible Mechanisms of Action

4 A hypothetical model illustrating possible mechanism of action of NCCM in inhibiting

5 etoposide-induced apoptosis in human NP cells. Treatment with NCCM restores the

6 mitochondrial membrane potential, thereby suppressing the release of cytochrome c into the

7 cytosol, thus inhibiting the activation of intrinsic mitochondrial pathway. In addition, treatment

8 with NCCM induced expression of anti-apoptosis protein XIAP involved in the activation of

9 caspase-3/7 downstream of caspase-9 in human NP cells.

10

11 Supplemental Figure Legends:

12

13 Supplemental Figure 1: Cell culture and experimental setup

14 Supplemental Figure 2: Capsase-3/7 Activity Assay IL-1β + FasL at 24h

15 Histograms representing caspase-3/7 activity in human NP cells in response to treatment with a

16 combination of IL-1β (10ng/mL) and FasL (100 ng/mL) after 24h as compared to no treatment

17 controls (NTC). No significant increase of caspase-3/7 activity could be observed.

18 Supplemental Figure 3: Differential Gene Expression

19 Scatter plots showing differential expression of genes involved in apoptosis in human NP cells

20 (#H1) (a) treated with etoposide alone (b) etoposide in presence of NCCM for 24h. (red:

21 upregulated; green: downregulated; black: unchanged).

24
Page 24 of 24