Influences of circulatory factors on intervertebral disc aging phenotype

Circulatory factors affect disc aggrecan gene expression

Disc matrix PG loss is largely due to destruction of aggrecan, the major PG structural constituent of the disc ECM. The aggrecan core proteins function in maintaining the disc NP’s hydration and osmotic pressure. Aggrecan is an important biomarker of disc health as its loss directly correlates to age-associated IDD [8]. As expected, old mice (O-O) were found to have less disc aggrecan immunofluorescence than young mice (Y-Y) (Figure 1A). Compared to young mice in the isochronic pairing (Y-Y), young mice in the heterochronic pairing (Y-O) exhibited 21% decrease in disc aggrecan immunofluorescence (p<0.001). On the other hand, old mice in the heterochronic pairing (O-Y) were found to have slightly higher disc aggrecan immunofluorescence (0.82 ± 0.03) than the old mice (O-O) in the isochronic pairing (0.76 ± 0.04), but the increase was not statistically significant. These aggrecan immunofluorescence findings are consistent with disc aggrecan mRNA expression (Figure 1B), demonstrating that exposure to old circulatory factors suppressed aggrecan expression while exposure to young circulatory factors enhanced aggrecan expression in mouse IVDs.

Circulatory factors affect disc aggrecanolysis and matrix metalloproteinase (MMP) gene expression

As expected, disc aggrecan fragmentation catalyzed by MMP and ADAMTS classes of MMPs was elevated in old mice (O-O) compared to young mice (Y-Y) (Figure 2). Likewise, young mice exposed to circulatory factors from old mice significantly increased both MMP- and ADAMTS-mediated disc aggrecanolysis (Y-Y versus Y-O). This finding is consistent with the enhanced mRNA and protein expression of MMP13 and ADAMTS4 in discs of Y-O mice compared to those in Y-Y mice (Figure 3). Disc MMP13 mRNA increased more than fourfold (Figure 3B) and its protein increased 85% (Figure 3A) in Y-O mice compared to Y-Y mice. Disc ADAMTS4 mRNA and protein also increased 20-30% (Figure 3C, 3D) in Y-O mice compared to Y-Y mice.

When old mice were exposed to young circulatory factors, disc aggrecanolysis descreased (O-Y versus O-O), but the differences between these two groups were not significant (Figure 2). This finding is also consistent with the reduced mRNA and protein expression of MMP13 (~30%, Figure 3A, 3B) and ADAMTS4 (15-30%, Figure 3C, 3D) in discs of O-Y mice compared to those in O-O mice. While these effects were seen in the heterochronic pairing for eight weeks, it should be noted that reduced aggrecan fragmentation was also seen in 4-week parabiosis (Supplementary Figure 2), but the effects were less pronounced, suggesting that prolonged exposure to young circulatory factors is needed to suppress aggrecanolysis in the aged disc.

Circulatory factors affect disc cellular senescence and SASP

Enhanced aggrecanolysis and catabolic gene expression observed in Y-O mice compared to Y-Y mice could be due to increased disc cellular senescence. This is because senescent disc cells exhibited perturbed matrix homeostasis, which serves as compelling evidence of the paracrine and endocrine effects of senescence-associated secretory phenotype (SASP) on inducing senescence and matrix catabolism in distal tissues. To test this, we assessed the level of disc cellular senescence by measuring three key markers of cellular senescence: p16^INK4a, p21^Cip1, and p53 [12]. As expected, expression of all three markers was elevated in the discs of old mice (O-O) mice, compared to young mice (Y-Y) (Figure 4). When exposed to old blood, the discs of heterochronically-paired young mice (Y-O) displayed a significant increase in p53 protein expression compared to isochronically paired young mice (Y-Y). Discs of heterochronically-paired young mice (Y-O) also trended toward an increase in p21 and p16^INK4a proteins compared to those in isochronically-paired young mice (Y-Y), although the changes were not statistically significant.

To further evaluate other senescence biomarkers, we measured and compared the mRNA expression levels of several key SASP factors in disc tissue, including TNFα, IL-1β, and IL6. Disc IL-1ß gene expression was significantly greater in isochronically-paired old mice (O-O), compared to isochronically-paired young mice (Y-Y), as expected (Figure 5). Unexpectedly, TNFα and IL-6 mRNA levels were slightly lower in old mice (O-O), compared to young mice (Y-Y). Exposure to old blood also unexpectedly decreased IL-1ß gene expression in disc tissue of heterochronically-paired young mice (Y-O), compared to Y-Y mice. No significant changes in disc TNFα or IL-6 expression were observed between the Y-O and Y-Y mice. These findings suggest that exposure to an old environment only had modest influences on disc cellular senescence, and the differential effects of old blood on expression of different SASP factors in young mice highlight the complex interaction between the disc tissue and its surrounding systemic environment.

Levels of p16^INK4a mRNA and p21^Cip1 and p53 proteins in the IVDs of old mice exposed to young blood (O-Y) were all lower than those measured in isochronically-paired old mice (O-O) (Figure 4). Based on these measurements, it was expected that expression of SASP factors would also be reduced in disc tissue of O-Y compared to O-O mice. However, only disc IL-1β mRNA was found to be lower in O-Y than O-O mice, whereas no significant changes were observed for TNFα and IL6 gene expression. These results suggest that exposure to young circulatory factors suppressed markers of senescence burden in disc tissue of aging mice, but not necessarily expression of the pro-inflammatory cytokines.

Circulatory factors affect disc histologic features

To determine whether blood circulatory factors affect disc degenerative changes in mice at the tissue level, we performed H&E histology on lumbar discs (L4-L5) to compare the gross morphological disc features among groups. We quantified the changes in several key disc histological features using a grading system previously established by Tam et al. for mouse discs [23] in which higher histologic scores correspond to greater degenerative changes. Histology grading was done in a blinded manner by three scorers on multiple H&E-stained disc sections. Compared to isochronically-paired young mice (Y-Y), isochronically-paired old mice (O-O) exhibited degenerative histological changes in their discs. These included increased AF clefts/fissures, loss of NP cellularity, loss of a well-defined AF/NP boundary, thinning and loss of NP matrix, and structural disorganization of AF lamellae (Figure 6). Most of these degenerative changes, with the exception of the AF/NP boundary parameter, were elevated in discs of young heterochronic parabionts (Y-O), compared to the young isochronic pairs (Y-Y) (Figure 6 and Table 1), suggesting that exposure to old blood accelerates the disc aging phenotype.

Some histological degenerative changes, particularly NP cellularity loss and fissures, were reduced in the disc tissue of old mice exposed to young blood (O-Y), as compared to isochronically-paired old mice (O-O) (Figure 2 and Table 1). However, while disc histological features tended to improve in old mice paired with young mice, the individual and composite scores did not differ significantly between O-O and O-Y mice. Together, these results suggest that the disc aging phenotype can be influenced by systemic factors.

Circulatory factors affect disc tissue aging in accelerated aging Ercc1^-/Δ mice

DNA repair-deficient Ercc1^-/Δ mice age rapidly and exhibit many age-related disorders, including accelerated disc aging due to persistent DNA damage. Past studies have reported on the health rejuvenation and extended lifespan of these mice upon systemic injection of muscle derived stem cells (MDSCs) [24]. Because exposure to young blood mitigates age-related IDD in natural aging mice, we investigated if similar effects occurred in Ercc1^-Δ/ mice with accelerated aging. Indeed, compared to isochronically-paired old Ercc1^-/Δ mice, Ercc1^-/Δ mice from a heterochronic pairing with young Wt mice for four weeks also exhibited improved histological features (Supplementary Figure 1A), greater aggrecan immunofluorescence in their disc NP (Supplementary Figure 1A, 1B) and higher total NP GAG content (Supplementary Figure 1C). Loss of laminB1 expression is a marker of cellular senescence [25]. Discs of Ercc1^-/Δ mice from the heterochronic pairing also contained higher levels of laminB1 in their AF and NP (Supplementary Figure 1D), suggesting reduced disc cellular senescence in these mice after being exposed to young blood. The results indicate that exposure to young systemic factors blunt age-associated IDD in natural as well as accelerating aging mice. The effects of systemic circulatory factors on a disc aging phenotype in mice are summarized in Figure 7.

Effects of old circulatory factors on young discs

Similar to previous results, the IVDs in old mice exhibited tissue level degeneration and perturbed proteoglycan matrix homeostasis, including elevated aggrecanolysis and aggrecan loss as typically seen in humans and other animals [8]. These degenerative changes, while absent in young mice, appeared in young mice sharing blood circulation with old mice for just ~eight weeks in our heterochronic parabiosis model. For instance, disc aggrecan immunofluorescence decreased while aggrecanolysis increased in Y-O mice compared to Y-Y mice. In addition, greater histological degenerative features, especially loss of NP cellularity and AF lamella disorganization, were found in the Y-O mice than Y-Y mice. It should be noted, however, that these disc degenerative features were less pronounced in four-week (Supplementary Figure 2) heterochronic parabiosis compared to pairing for eight weeks. Thus, prolonged exposure to old blood factors accelerate disc matrix PG imbalance and tissue degeneration in young mice. These findings are consistent with the previous blood infusion and heterochronic parabiosis studies that reported inhibitory and detrimental impacts of old blood on multiple tissues, including muscle, liver, and brain hippocampus [28].

Cellular senescence has recently emerged as a major driver of disc aging. Senescent disc cells have been demonstrated to acquire a phenotype that disrupts PG homeostasis [13, 29, 30]. Systemic clearance of cellular senescence also mitigates age-associated IDD in mice [12]. Thus, one possible mechanism of action of old blood on a young disc is the induction of disc cellular senescence. However, exposure to old blood neither significantly increased expression of the key markers of cellular senescence, e.g. p16^INK4a or p21^Cip1, nor induced expression of the key SASP factors such as IL-1β, IL-6, or TNFα. Hence, the negative effects of old blood on young disc must be mediated through mechanisms other than disc cells becoming senescent. This contrasts with the effects of heterochronic parabiosis on senescence observed in liver, kidney, and lung tissues where significant up or down regulation of senescence and SASP markers were significantly affected by circulating factors [31]. Part of this difference could be due to clearance of senescent cells by young immune cells in tissues other than the disc and/or to the lower concentrations of the key circulating SASP factors, which drive paracrine senescence, within the poorly vascularized disc. However, the discs of Y-O mice did have higher levels of ADAMTS4 and MMP13 proteins than the discs of Y-Y mice, but whether the sources of these proteins are exogenous or endogenous requires further study. While it is possible that these catabolic proteins were derived from old blood and entered the disc tissue from the systemic environment, increased mRNAs of ADAMTS4 and MMP13 in Y-O discs suggest that at least part of the increase was due to production by the resident disc cells. The specific humoral factors in old blood responsible for upregulating gene expression of disc ADAMTS4 and MMP13, as well as circulating factors responsible for mediating aggrecanolysis and promoting aggrecan loss in young disc remains to be resolved.

While the implications of this study suggest that circulating factors may exist that can be either restored or inhibited in old blood decelerate IDD and/or restore disc integrity, the identity of such factor(s) remain to be identified. Previous heterochronic parabiosis studies reported specific circulatory soluble factors that closely correlate with the effects on specific tissue. For example, the chemokine CCL11 (eotaxin) and ß2M were identified in old blood as able to impair neurogenesis in the brains of young heterochronic parabiont mice [32]. On the other hand, loss of normal cardiac function in an aging heart, leading to cardiac hypertrophy and diastolic heart failure, was reported to be partially due to the lack of certain circulating factors in old mice. One of these factors was identified as the growth differentiation factor 11 (GDF11), which was initially reported to be reduced in the old circulation, though this remains complex and controversial [20]. Clearly, the identification of the specific humoral factors or cells in the circulation that regulate the disc environmental niche, its resident cells, and matrix homeostasis will be important.

Effects of young circulatory factors on old mouse discs

Old mice exposed to young blood for a short (4 weeks, Supplementary Figure 2) or longer (8 weeks, Figures 1–3) duration in our heterochronic parabiosis showed minimal evidence of rejuvenation in their disc proteoglycan matrix health. Specifically, there was negligible improvement in disc aggrecan immunofluorescence and an insignificant decrease in MMP- and ADAMTS-mediated disc aggrecanolysis in O-Y mice, compared to O-O mice. Similarly, exposure to young blood did not significantly improve disc histologic features in the old heterochronic parabionts, apart from the loss of NP cellularity (Figure 6 and Table 1). It is interesting to note that while exposure to young blood significantly blunted the loss of NP cellularity in old mice, such effect did not translate to overall disc aggrecan content (Figure 1), possibly because NP cells in these old mice do not actively synthesize PG matrix or the rate of production was too slow to produce a measurable difference after eight weeks of parabiosis. Because blood circulation is not fully shared until 2-3 weeks after surgical pairing, a longer parabiosis duration, e.g. 12-16 weeks, might be necessary to increase beneficial effects of young blood on old discs.

The lack of rejuvenation of old discs by young blood could also be due to irreversible changes conferred by aggrecanolysis and the substantial loss of disc aggrecan in 18-month-old mice prior to being paired with the young mice. Aggrecanolysis, the proteolytic fragmentation of aggrecan, is irreversible and leads to the eventual loss of aggrecan from disc tissue [33]. Hence, performing the heterochronic parabiosis on younger mice (<18 months old) and for a longer duration might be necessary to prevent or delay this age-related degenerative change. Nevertheless, rejuvenation of cognitive function and synaptic plasticity by young blood has been reported in mice when they were heterochronically-paired at 18 months of age [34]. Additionally, young blood rejuvenates hepatic progenitor cells and skeletal muscle stem cells even in older mice (19-26 months) when they were first paired with young mice [35]. Circulating peripheral blood cells contributed to meniscus repair in wildtype rats sharing blood with transgenic rats [22]. Together with our findings, the beneficial impacts of heterochronic parabiosis are marginal for IVDs and neurogenesis, while more notable effects have been noted for liver, cardiac hypertrophy, fracture, and skeletal muscle repair—robust for muscle and modest for liver and meniscus. These findings underscore the complex and nonuniform nature of interactions between different organs and their systemic environment, as well as the nature of the studies, e.g. parabiosis duration, genetic strains used, tests performed, etc.

While young blood only had marginal effects on restoring old disc matrix homeostasis, it significantly reduced disc cellular senescence in the old mice. Levels of all three key protein markers of cellular senescence (p16, p21, and p53) were decreased in disc tissue of O-Y mice compared to O-O mice, a result similar to what we have observed in liver, kidney, and lung of the paired mice [31]. In addition, MMP13, and ADAMTS4 were also greatly reduced in O-Y compared to O-O mice. How young blood decreases these proteins in the IVDs of old mice is not understood, although dilution of old blood factors by young blood could be a mechanism if these SASP factors enter the disc from the systemic environment. Otherwise, if MMP13, and ADAMTS4 are produced by the resident disc cells, then their expressions in the niche of old disc tissue must be down regulated by some circulating factors from the young mice.

Effects of young blood on discs of accelerated aging mice

Interestingly, while exposure to young blood only marginally mitigated age-related IDD in natural aging mice, it significantly reduced disc degeneration in accelerated aging Ercc1^-/Δ mice. Ercc1^-/Δmice paired with young mice for only four weeks had about twice the level of disc NP aggrecan immunofluorescence and 40% greater NP GAG content compared to the isochronic paired Ercc1^-/Δ mice. The positive effects could be due to accelerated aging of Ercc1^-/Δ mice, which have a compressed average lifespan of 5-6 months, or about four times shorter than naturally aging mice [36]. Thus, a 4-week parabiosis of Ercc1^-/Δ mice would be equivalent to a 16-week parabiosis of natural aging mice, a sufficient duration for young blood to confer health benefit to disc tissue in these animals. The beneficial effects of young blood on Ercc1^-/Δ mice in our heterochronic parabiosis are consistent with previous findings demonstrating extension of lifespan and healthspan in these progeroid mice through multiple systemic interventions [24, 37–39]. These interventions include inhibition of IKK/NF-kB signaling, treatment with senolytics such as an HSP90 inhibitor or the combination of dastinbi and quercetin, and treatment with adult stem cells from young, but not old mice.

Unusual and unexpected findings

Our heterochronic parabiosis produced several unexpected results. First, there was no direct correlation between disc aggrecanolysis and the expression levels of two key matrix metalloproteinases (MMPs)—ADAMTS4 and MMP-13. Exposure of young mice to old blood greatly increased both disc aggrecanolysis and expression of ADAMTS4 and MMP-13, suggesting that these MMPs are responsible for disc aggrecanolysis. In contrast, exposure of old mice to young blood marginally reduced disc aggrecanolysis, but significantly decreased disc expression of ADAMTS4 and MMP-13. This suggests that either MMPs other than these are responsible for disc aggrecanolysis or substantial aggrecanolysis already happened in the old mice before pairing. Second, there is a disconnect between the disc aggrecan mRNA level and aggrecan protein level in O-Y mice compared to O-O mice. While disc aggrecan mRNA was greatly upregulated, its protein level (aggrecan immunofluorescence) was only slightly increased when old mice were paired with young mice (Figure 1). This observation suggests that young blood exposure primarily upregulates transcription of disc aggrecan but not the post-transcription steps such as translation, post-translation modifications, export to the ECM, etc. Third, expression of SASP factors is highly variable in disc tissue of old mice. For example, only IL-1β, not IL-6 and TNFα, were found to be expressed significantly more in the disc of 20-month old mice (O-O) compared to young mice (Y-Y). This is unexpected as all these three cytokines were found in elevated levels in degenerative and aged human discs [40–42]. Additionally, exposure of old mice to young blood did not affect their disc IL-1β gene expression while exposing young mice to old blood significantly increased their disc IL-1β gene expression. Altogether, these unexpected findings suggest that the effects of systemic circulatory factors on disc matrix homeostasis and cellular activities are highly complex.

It should be noted that biomechanical considerations cannot be completely ruled out as contributors to our results, and that this could be addressed in part via Sham controls in which heterochronic pairing are surgically done without shared blood circulation. However, this is unlikely to be a major confounder in our data for a few key reasons. First, our 4 and 18 month old mice were chosen strategically as this is an age in which the long bones are developed in young mice making the union more successful, whereas the older mice are not yet too advanced in age. Second, the body size of younger (~30-35g) and older (~40-45g) mice only vary slightly, based mostly on differences in adiposity. By the time animals were necropsied at 6 and 20 mo of age, respectively, these differences were even smaller. Second, we also did not typically observe any noticeable impact or dominance between parabionts on ambulation, as animals typically do not “drag” each other around, but instead become increasingly synchronized in terms of sleeping, eating and moving in unison within days of surgery. For these reasons, the notion of “young animals carrying around the old mice”, an observation we did not really observe in this model, as driving the disc phenotypes seemed unlikely.

Parabiosis of Wt mice

Young (Y, aged 4 months) and old (O, aged 18 months) wildtype C57BL/6 male mice were obtained from the NIA-aged rodent colony for parabiosis and three mouse pairs were generated: young-young (Y-Y), young-old (Y-O), and old-old (O-O). Parabiosis surgery was performed as described in the Einstein Health Span Core [43, 44]. In brief, an incision was made from lateral aspect of the elbow to the knee to separate the skin from the fascia layer, this allowed for fusion of the skin as well as part of the peritoneum, in order to increase the area for blood exchange. Mice were kept one pair per cage following surgery and provided with adequate food and water. Mice were sacrificed approximately eight weeks after surgical pairing by exsanguination under isoflurane anesthesia. Spines and tails were isolated from the pairings and separated into four groups (Young Isochronic (Y-Y), Young Heterochronic (Y-O), Old Isochronic (O-O), and Old Heterochronic (O-Y), and were immediately frozen using liquid nitrogen and stored at -80°C until study. Of note, O-Y refers to the old mouse from the heterochronic pair while Y-O refers to the young mouse from the heterochronic pair. Parabiosis and related procedures were approved by the Einstein Institutional Animal Care and Use Committee (IACUC).

Parabiosis of Ercc1^-/Δ mice

Ercc1^-/Δ mice exhibit aging symptoms by the age of 5 weeks. Therefore, older 8-week-old female Ercc1^-/Δ mice were surgically paired with younger 6-week-old female f1 Wt mice (C57BL/6:FVB/n) for 4 weeks before being sacrificed for spine isolation and analyses.

Histological stain

Isolated lumbar spines segments were decalcified and embedded in paraffin (Tissue Tek processor and Leica embedder, ID:17492, Rankin, Holly, MI). Seven-micrometer sections were stained with hematoxylin and eosin (H&E) by standard procedures [26], then photographed under 40-100x magnification (Nikon Eclipse Ts100, Tokyo, Japan). Changes in histological features were evaluated by three blinded scorers using the histological grading system established for mouse discs [23] from the four groups. Three lumbar sagittal sections from each mouse were scored for the following specific features in the NP and anulus fibrosus (AF) as specified by Tam et al [23]. Normal young disc tissue is defined as having: (1) high NP cellularity that is centralized as a single mass of vacuolated cells enclosed within a layer of matrix; (2) AF that is aligned in concentric lamellae without disruption to the structure; (3) an NP matrix structure in which the matrix mass is continuous and full without significant empty space; and (4) a well-defined AF/NP boundary whereby the AF and NP compartments are always distinct. Each feature was scored individually with a score of 0–14; whereby, 0 represented the young health state and 14 represented the maximum degenerative state. A composite score was also calculated by adding all the individual scores (Table 1).

Immunofluorescence

Mouse lumbar IVDs were isolated and fixed for two hours at 4°C in 2% paraformaldehyde. The tissues were cryoprotected with 30% sucrose in phosphate buffered saline (PBS, Cat.No.08118006, ThermoFisher Scientific, Pittsburgh, PA) overnight at 4°C, and then embedded in OCT (Cat.No. 4583, Tissue-Tek,Torrance, CA). Serial axial plane lumbar disc cryosections were cut at a thickness of 7 μm. The tissue sections were rehydrated in PBS, permeabilized, and blocked with 0.25% Triton X-100, 10% goat serum and 1% Bovine Serum Albumin (BSA, Cat.No.166099, ThermoFisher Scientific, Pittsburgh, PA) in PBS for 30 minutes at room temperature. Incubation with the specific primary antibodies (anti-Aggrecan, Cat.No. AB1031, Millipore; anti-MMP13, Cat.No. ab39012, Abcam, Cambridge, MA; anti-ADAMTS4, Cat. No. 185722, abcam, Cambridge, MA) were carried out overnight at 4°C following blocking. The sections were then incubated in a secondary antibody Cy3-conjugate goat anti-rabbit IgG (Cat.No.130233, Jackson Laboratory, West Grove, PA) solution for one hour according to the manufacturer’s protocol. Immunostained sections were imaged and analyzed under the Nikon A1 confocal laser microscope and NIS-elements microscopy imaging software.

1,9-Dimethylmethylene BLUE (DMB) colorimetric assay for sulfated Glycosaminoglycans (GAGs)

NP tissue was isolated and pooled from six lumbar IVDs of each mouse. The pooled tissue sample was digested using papain at 60°C for two hours. GAG content was measured in duplicate by DMB procedure using chondroitin-6-sulfate (Sigma, Milwaukee, WI, C-8529) as a standard. The DNA concentration of each sample was measured using the PicoGreen assay (Molecular Probes, Sunnyvale, CA) and used to normalize the GAG values. Average values from six reaction samples (two duplicates × three mice per group) were calculated +1 SE.

Western blot analysis

The cervical and thoracic discs from each mouse were used to assess aggrecan fragmentation utilizing a previously established method [45] using anti-aggrecan primary antibody (Cat.No. ab36861, Abcam, Cambridge, MA) and anti-rabbit-HRP secondary antibody (Cat.No. 31460, ThermoFisher Scientific, Pittsburgh, PA). Protein from the spine disc samples was extracted using 4M guanidinium chloride as reported previously [45]. All samples were measured for total protein concentration using a BCA Protein Assay Kit (ThermoFisher Scientific, Pittsburgh, PA) to ensure equal loading. Loading buffer was added to 30 μg protein and separated on 4-20% SDS-PAGE gels before transferred onto PVDF membranes (Bio-Rad, Hercules, CA). The blots were blocked using 5% non-fat dry milk (Bio-Rad, Hercules, CA) at room temperature, shaken for one hour, and then incubated overnight with primary antibodies using MMP13 (Cat.No.39012, Abcam, Cambridge, MA), ADAMTS4 (Cat.No.185722, Abcam, Cambridge, MA), p16 (Cat.No.189034, Abcam, Cambridge, MA), p21 (Cat.No. 109199, Abcam, Cambridge, MA), p53 (Cat.No.1C12, ThermoFisher Scientific, Pittsburgh, PA), and β-actin (Cat.No. PA1-183, ThermoFisher Scientific, Pittsburgh, PA). The following day, the blots were washed three times with PBS-T buffer then incubated with the anti-rabbit-HRP secondary antibody (Cat.No. 31460, 1:10000, ThermoFisher Scientific, Pittsburgh, PA) for one hour at room temperature. Following another three washes with PBS-T, the blots were viewed with the LiCoR Odyssey imager (LI-COR Biosciences, Lincoln, NE) to visualize and quantifiy the protein bands.

mRNA analysis

Total RNA from whole IVD tissue of each specimen’s lumbar vertebrae (L1-3) was isolated using TRIzol™ Reagent per manufacturer’s instructions (Cat. No. 15596026, ThermoFisher Scientific, Pittsburgh, PA). Real-time quantitative RT-PCR was run using iTaq™ Universal SYBR® Green One-Step Kit (Cat. No. 1725151, BIO-RAD, Hercules, CA) and Bio-Rad iCyclyer IQ5 detection system. Target gene expression was calculated by the comparative C_T method (ΔΔ C_T) and normalized to GAPDH mRNA expression. The housekeeping gene GAPDH was chosen because its mRNA expression in disc tissue remains unchanged in aging and degenerating disc [46]. PCR primers used in the study are provided in Table 2.

Statistical analysis

Shapiro-Wilk test was used to test for normality. Student's independent t test was used to analyze data that were found to be normal. For two-variable nonparametric data, the Mann-Whitney test was used. Analysis of variance (ANOVA) with Bonferroni correction for multiple comparison was used in cases of data with multiple experimental groups. Statistics were derived using GraphPad Prism from GraphPad Software (San Diego, CA). All graphs show mean values with error bars. Unless specified otherwise, p< 0.05 was considered significant.