Aggrecan and small leucine-rich proteoglycan fragments correlate with Toll-like receptor-2 mediated inflammation in painful degenerative disc disease

Abstract

Background

Painful intervertebral disc degeneration is a leading cause of chronic low back pain. Proteolytic cleavage fragments of extracellular matrix components, particularly aggrecan and small leucine-rich proteoglycans (SLRPs), may act as endogenous danger signals activating inflammatory pathways. To determine whether proteolytic fragments of aggrecan and SLRPs correlate with disc degeneration severity and Toll-like receptor-2 (TLR-2) mediated inflammation.

Methods

Human disc tissues were analysed from 20 non-degenerated cadaveric controls (Thompson Grades 1–2) and 35 patients with painful degeneration (Pfirrmann Grades 3–5). Western blotting assessed fragmentation of aggrecan and SLRPs (decorin, biglycan, lumican, fibromodulin, chondroadherin). Immunofluorescence localized these molecules in disc sections. Disc cells were cultured under four conditions unstimulated controls, TLR-2 agonist Pam2CSK4-stimulated controls, cells extracted from degenerated discs, and those treated with the TLR-2 antagonist MMG-11. Cytokine profiles were determined using antibody array.

Results

Fragmented peptides of aggrecan and SLRPs (22–45 kDa) were predominantly detected in Pfirrmann Grades 4 and 5 discs. TLR-2 expression was significantly higher in degenerated disc cells versus controls (P < 0.001), further upregulated by Pam2CSK4 and attenuated by MMG-11. Cytokine analysis revealed marked pro-inflammatory shifts in patient discs (interleukin [IL]-6 ↑1.37×, IL-8 ↑1.30×, IL-1β ↑1.25×), while control discs exhibited an anabolic profile with elevated expressions of growth factors (TGF-β, EGF, VEGF).

Conclusions

Aggrecan and SLRP fragments were observed alongside TLR-2 mediated inflammatory responses in advanced disc degeneration, suggesting a potential association with tissue catabolism. Targeting TLR-2 signaling may warrant further investigations as a potential therapeutic strategy for painful disc disease.

INTRODUCTION

Human intervertebral discs (IVDs) are resilient, load-bearing pads of white fibrocartilage located between adjacent vertebral bodies [1]. They facilitate spinal motion allowing flexion, extension, as well as medial and lateral rotation, while resisting static, compressive, and shear forces [2]. IVDs are elliptical in shape (Fig. 1), bordered above and below by the hyaline cartilage endplate that separates it from the vertebral bodies [3]. Each disc has three integrated parts: the centrally placed gel-like nucleus pulposus (np) region, surrounded by the adjacent fibrillar inner annulus fibrous (iaf), and the lamellated collagen dense outer annulus fibrosus (oaf) region. In healthy adults, the centrally placed viscous nucleus pulpous region is made up of aggrecan, a large hydrophilic proteoglycan molecule that maintains the internal swelling pressure within the disc and allows it to function as a hydrostatic shock absorber [4]. The inner annulus serves as a transitional zone surrounding the central np, composed predominantly of fibrillar type II collagen that counteracts shear forces [5]. The outer annulus is the most peripheral disc tissue region comprising approximately of 20–25 concentric lamellae of dense type I collagen fibers that constrains the swelling pressure from np. In non-degenerate discs, only the outermost two to three layers of the collagen lamellae are innervated making it one of the largest avascular and aneural structure in the human body [6–8].

Fig. 1.

High-resolution superior view of cadaveric non-degenerated Grade 1 IVDs illustrating the lamellated oaf (black arrow) encasing the centrally located, gel-like np (red arrow). The inner annulus fibrosus (green arrow) serves as a transitional zone between the oaf and the np. IVDs: intervertebral discs, np: nucleus pulposus.

The np and inner annulus fibrosus regions are populated by rounded, chondrocyte-like cells [4]. In moderate to severely degenerated discs, these cells often form clusters of varying sizes [9]. In contrast, the oaf of non-degenerated IVDs shows flattened, fibroblast-like cells aligned parallel to the collagen lamellae, with no evidence of cell clustering [9,10]. Throughout life, disc cells are exposed to diverse mechanical stimuli and they reside in a relatively nutrient-poor microenvironment contributing to sparse cell density, promoting accumulation of metabolic byproducts such as lactate [11]. As a result, IVD tissue has limited capacity for adaptive remodelling or intrinsic healing. Cells in the np and inner annulus are surrounded by a deeply staining pericellular matrix, composed of fine fibrillar collagen and non-collagenous molecules [12] which includes aggrecan and small leucine-rich proteoglycans (SLRPs). These molecules are the key components of the IVD matrix playing an essential role in the maintenance of tissue structure and functions.

SLRPs in IVDs are low molecular weight matrix proteins characterized by a central core protein made of multiple leucine-rich repeats, with covalently attached glycosaminoglycan (GAG) side chains [13,14]. SLRPs are comprised of eighteen members, classified into five distinct classes based on their gene location, number of leucine-rich repeats, and the presence or absence of sulphated GAG chains. SLRP protein members such as decorin (DCN), biglycan, chondroadherin (CHAD), fibromodulin (FMOD), and lumican (LUM) are found within the non-degenerate IVD matrix and they play multifaceted regulatory roles in promoting cellular proliferation, facilitating adhesion of the matrix molecules that control the diameter of the collagen fibrils, and modulate cellular responses to growth factors, chemokines, and cytokines [15–17]. In contrast, aggrecan is a large proteoglycan composed of three globular domains, G1, G2, and G3. Hyaluronan binds to the G1 domain via the link proteins, the G2 domain attach to numerous GAG sidechains, while the G3 domain interacts with the cell adhesion molecules [18]. Both aggrecan and SLRPs can engage with toll-like receptors (TLRs), integrins, collagens, fibronectin, fibrillin, and other matrix components to maintain the disc structure [19]. During accelerated disc degeneration, lesions of articular cartilage and knee meniscus, SLRPs and aggrecan tend to undergo proteolytic cleavage and fragment into smaller precursor forms due to combined action of increased mechanical stress, injury, elevated synthesis of matrix-degrading enzymes, and a synthesis of pro-inflammatory factors reflecting a common pathological endpoint in cartilaginous, cancerous, and musculoskeletal tissues [20]. In degenerated discs and articular cartilages, fragments of aggrecan and SLRPs can activate TLRs- 2 and 4 [21,22] that trigger innate sterile inflammation mechanism in the absence of blood vessels. These changes impair tissue repair and progressive degeneration promotes infiltration of inflammatory cells, including even the ingrowth of blood vessels and nerves [23,24] that directly contribute to pain.

Cell surface TLRs in IVDs respond to both endogenous and exogenous signals by activating inflammatory pathways such as the NF-κB (nuclear factor kappa-light-chain-enhancer of activated B cells) and mitogen-activated protein kinases (MAPKs) leading to the production of pain mediators such as prostaglandins, nitric oxide, and phospholipase A2 [25,26]. These molecules can induce nociceptive changes even in distant structures like the dorsal root ganglion, contributing to pain perception. Recent studies have shown that TLR-2 can be activated by fibronectin fragments, which act as a danger signal and promote inflammation that exacerbates pain [27]. Similarly, fragmented biglycan functions as a damage-associated molecular pattern (DAMP) recognition molecule [28], mimicking the behaviour of pathogen-associated molecular patterns (PAMPs) such as the bacterial lipopolysaccharides, flagellin, and viral RNA to trigger TLR-mediated inflammatory responses [22]. These changes indicate that fragmentation of aggrecan and various SLRPs leads to increased synthesis of nerve growth factors (NGF) [7,23] allowing ingrowths of blood vessels and nerves into the discs, a key feature of degeneration, inflammation, and pain.

Thus, in this investigation, the authors hypothesize that as IVD degeneration progresses, the proteolytic breakdown of aggrecan, DCN, biglycan, LUM, and CHAD increases with the severity of disc degeneration grades from 1 to 5. These peptide fragments may disrupt cellular homeostasis by upregulating the expression of TLR-2, thereby enhancing secretion of inflammatory and nociceptive mediators in the affected patients. Thus, the aim of the study was to demonstrate the expression levels of TLR-2 receptors and proteolytic fragments of aggrecan and SLRPs in non-degenerated control versus patient disc tissues exhibiting varying grades of degeneration.

MATERIALS AND METHODS

1. Sample size

The human disc collection protocol was approved by the Institutional Review Board (IRB/IEC/2021-22) prior to the commencement of the study. Twenty non-degenerated IVD tissues were harvested from donated human cadavers for medical teaching within 4 hours of postmortem to minimize autolytic and proteolytic changes. Cadavers with a history of spinal tumors, severe infection, or genetic degenerative disease were excluded from this study. Immediately after retrieval, discs were rinsed in sterile phosphate-buffered saline (PBS) to remove blood and debris, snap-frozen in liquid nitrogen, and stored at –80°C until further analysis. Protease inhibitors were added during the subsequent tissue homogenization process to prevent further degradation. These samples were designated as the control group and evaluated macroscopically according to Thompson grading criteria [29]; only non-degenerated discs classified as Thompson Grade 1 or Grade 2 were included as controls as these grades represent the most unaltered stages of disc morphology, characterized by an intact annulus fibrosus, a well-hydrated np, and a uniform cartilaginous endplate without fissures and any gross morphological changes. In parallel, after informed consent, 35 degenerated disc tissues were obtained from patients undergoing spinal surgeries for pathological conditions such advanced disc degeneration, herniation, and protrusion following ethical guidelines and regulations. The surgically excised specimens were primarily comprised of np, with varying amounts of annulus fibrosus, cartilaginous endplate, and occasionally bone, consistent with previous published reports [24]. Because the np is proteoglycan-rich, it readily extrudes from the pressurised disc environment following rupture, leading to rapid tissue collapse and leaving behind a collagen-rich, proteoglycan-depleted scaffold.

All surgical specimens were collected within 30 minutes of excision and processed under standardized protocols (sterile saline rinse, cold-chain maintenance, and rapid fixation) to reduce mechanical or surgical artifacts. Each of the patient discs were additionally graded prior to surgery for assessing degeneration severities using the MRI-based Pfirrmann classification [30]. To reduce potential confounding factors, patient and control samples were matched for age and sex (Table 1).

Table 1.

Summary data for human discs control and patient groups

Disc detail	Non-degenerated cadaveric controls disc	Painful degenerate & herniated disc
Total number	20	35
Age (yr)	Mean 59 (range 46–72)	Mean 55 (range 42–67)
Spinal level (n)	L1–L2 (4), L2–L3 (4), L3–L4 (4), L4–L5 (4), L5–S1 (4)	L1–L2 (7), L2–L3 (7), L3–L4 (7), L4–L5 (7), L5–S1 (7)
Sex (n)	Male (10), Female (10)	Male (17), Female (18)
Grade	Mean 1.5 (range 1–2)	Mean 4 (range 3–4)
Pain (mo)	No pain	Mean 17 (range 5–60)

2. Experimental design

This study involved a comparative analysis of 35 surgically excised IVD tissues from patients exhibiting degeneration grades 3, 4, and 5, as classified by the Pfirrmann system, and twenty non-degenerate cadaveric controls with grades 1 and 2 were procured in situ without visible signs of degeneration. All procedures adhered to a standardized protocol detailed in the subsequent sections, ensuring uniformity in sample handling, grading, and analytical methods. Qualitative assessment of SLRPs and aggrecan fragments was performed using Western blotting, while immunofluorescence analysis was employed to visualize their spatial distribution in relation to disc degeneration. Additionally, TLR-2 immunopositive cells were quantified across three selected fields of view (FOV) under 20× magnification.

3. Protein extraction and digestion

Following an established method [15–17], disc tissue rinsed in PBS were finely diced and proteins were extracted at 4°C under continuous agitation for 72 hours using 15 volumes of 4 M guanidine hydrochloride (Sigma-Aldrich), 50 mM sodium acetate and 10 mM EDTA (Sigma-Aldrich). The pH of the solution was adjusted to 5.8, and protease inhibitor (Sigma-Aldrich) was added before introducing the diced facet samples into the extraction buffer. The extract was then separated from the tissue by centrifugation at 13,000 g, and 8 uL aliquots were prepared for SDS-PAGE following precipitation with 9 volumes of 95% ethanol. Precipitated proteins recovered by centrifugation were further washed with 90% ethanol, and re-dissolved in 25 µL of 50 mM sodium acetate (Sigma-Aldrich) with a pH of 7.4. This solution was digested with keratanase (Ambsio LLC) at 1 milliunits/25µL of extract for 6 hours, and further digested overnight with chondroitinase ABC (refers to the enzyme’s ability to degrade chondroitin sulfate A, dermatan sulfate B, and chondroitin sulfate C; Ambsio LLC).

4. Western blotting

Lamelli buffer containing 2x sodium dodecyl sulphate (Sigma-Aldrich), 2-mercaptoethanol (Sigma-Aldrich), and milliQ double filtered water was added directly to the disc sample solution obtained after protein extractions. The sample solution was heated in a water bath at 95°C for 10 minutes before being fractionated on 4%–12% polyacrylamide gels at 94 V (constant voltage) for 1 hour 40 minutes. Aggrecan and SLRP proteins were transferred to nitrocellulose membranes (0.45 µm) by electroblotting using transfer buffer supplemented with 20% methanol at 100 V (constant voltage) for 1 hour. The membrane was blocked with 3% skim milk powder in 0.01M Tris-HCL (Sigma-Aldrich), 0.15 M NaCl (Sigma-Aldrich), and 0.1% Tris-buffered saline with Tween 20 (TBST) (Sigma-Aldrich) with a pH of 7.4. Affinity purified monoclonal anti-carboxy terminal antibodies to SLRPs biglycan, LUM, CHAD, DCN, FMOD, and aggrecan (ABCAM) were applied to the blots at 1:1000 dilutions in TBST buffer with 1% bovine serum albumin (ABCAM) for 2 hours, followed by TBST washes. Secondary goat anti-rabbit conjugated antibody to horseradish peroxidase (Thermo Fisher Scientific) were applied to the blots at 1:1,000 dilutions, followed by TBST washes. Protein bands were visualised and developed by chemiluminescence using enhanced chemiluminescence (ECL) detection reagent mixture (GE Healthcare). Blots were also analysed omitting the primary antibody to check for IgG species present in the tissue extracts that cross reacted with the conjugated secondary antibody.

5. Immunofluorescence to assess the organization and distribution of SLRPs and aggrecan

Disc samples were snap-frozen in liquid isobutane and stored at –80°C. Prior to sectioning, tissues were equilibrated in a cryostat at –20°C, embedded in optimal cutting temperature compound, and sectioned at 12 µm using a Leica CM3050S cryostat. Sections were fixed in ice-cold acetone and incubated with appropriate serum to block non-specific binding. Primary monoclonal antibodies targeting SLRPs (DCN, CHAD, FMOD, LUM, biglycan) and aggrecan were applied at 1:100 dilution and incubated overnight. Immunofluorescent labeling was achieved using Alexa Fluor-conjugated secondary antibodies (515 and 594; Thermo Fisher Scientific) at 1:200 dilution. Sudan Black was used to quench autofluorescence, and nuclei were counterstained with DAPI (Vector Laboratories). Negative controls were prepared by omitting the primary antibody to assess background staining. Sections from each specimen were examined using a Leica fluorescence microscope, and immunopositive structures were analysed by a blinded reviewer using ImageJ software (ImageJ).

6. Cell culture and protein antibody array

Disc tissue from both the control and patient samples were rinsed with PBS supplemented with 50 μg/mL gentamycin, 0.5 μg/mL amphotericin B (Thermo Fisher Scientific) for 5 minutes. The tissue was then minced and subjected to enzymatic digestion using pronase followed by a collagenase application (Thermo Fisher Scientific). Post-digestion, the cell suspensions were filtered, washed in serum-free medium, and the supernatant was recovered by centrifugation. The isolated cells were then cultured in Dulbecco’s modified eagle medium with 4.5 g/L glucose (Sigma-Aldrich), 25 μg/mL gentamycin (Life Technologies), 2 mM Glutamax (Thermo Fisher Scientific) and 10% fetal bovine serum (Sigma-Aldrich) with 50 μg/mL ascorbic acid, and 10% fetal bovine serum (Sigma-Aldrich) on a volume per weight basis (10 × v/w), maintained at 37°C, 5% CO₂. Monolayer cultures were expanded for four passages until reaching 100% confluency, sufficient for the identification of TLR-2 receptors. Control and patient disc cells were treated with TLR-2 agonist PamCSK4 and antagonist MMG-11 at 0.5 µg/mL dilutions (InvivoGen) for 36 hours. At the end of the defined timepoint, culture media from the unstimulated control disc cells and patient discs were collected and analyzed in triplicate for protein array antibody profiling.

7. Statistical analysis

TLR-2 expression was quantified in disc cells under four experimental conditions: (1) non-degenerated cadaveric control disc cells, (2) non-degenerated control stimulated with the TLR-2 agonist Pam2CSK4, (3) surgically excised degenerated patient disc cells (P4 and P5), and (4) degenerated patient discs treated with the TLR-2 antagonist MMG-11. For each condition, cell counts were performed across three independent FOVs. As the data did not meet normality assumptions and sample size was limited, non-parametric Kruskal–Wallis tests were used to compare TLR-2 expression values among the four groups. Results were presented as mean ± standard deviation with 95% confidence intervals. Post-hoc pairwise comparisons were performed using Dunn’s test with Bonferroni correction to identify significant differences between specific conditions. Statistical significance was set at P < 0.001. All analyses were conducted using IBM SPSS software (version 28; IBM Co.). See Tables 1 and 2 for detailed summary statistics.

Table 2.

TLR-2 expression compared among four experimental conditions

TLR-2 expressions/mm3	Unstimulated control disc cells (n = 20)	PamCSK4 stimulated control disc cells (n = 20)	P4, P5 patient disc cells (n = 35)	P4, P5 disc cells stimulated with TLR-2 antagonist MMG-11 (n = 35)
1st FOV	10.24 (4.99)*	21.53 (9.13)	39.11 (15.22)*	26.67 (9.46)
2nd FOV	12.31 (5.77)*	37.91 (8.15)	45.12 (14.41)*	24.17 (10.94)
3rd FOV	17.04 (7.81)*	29.16 (11.0)	42.64 (15.61)*	21.45 (12.91)

(1) Unstimulated non-degenerated control disc cells; (2) Control cells stimulated with Pam2CSK4; (3) Degenerated patient disc cells (P4–P5); (4) Degenerated discs treated with TLR-2 antagonist MMG-11. Each group was analysed separately (n = number of samples). TLR-2 positive cell counts per mm³ are reported as mean (SD) with 95% CI. Control: 12.50 (2.10), 95% CI [11.52–13.48]; Pam2CSK4: 18.30 (3.40), 95% CI [16.71–19.89]; Patient: 35.70 (5.60), 95% CI [33.78–37.62]; MMG-11: 28.40 (4.80), 95% CI [26.75–30.05].

TLR-2: Toll-like receptor-2, P4: Pfirrmann Grade 4, P5: Pfirrmann Grade 5, FOV: field of view, SD: standard deviation, CI: confidence interval.

Significant differences between groups are indicated with significance levels denoted *(P < 0.001).

RESULTS

Intraoperative observations revealed distinct structural variations across different grades of disc degeneration. Grade 5 discs were completely extruded through the posterior annulus fibrosus, showing a total loss of continuity with the remaining disc structure. Grade 3 and 4 discs were also extruded but retained partial structural continuity with the intact disc tissue. MRI confirmed that these displaced disc tissues were in close proximity to the adjacent nerve roots. In contrast, Grade 1 and 2 discs obtained from cadaveric controls exhibited no signs of degeneration. These control specimens displayed well-defined regions of the np, inner annulus, and oaf (Fig. 1).

1. Western blots

Protein extracts from disc tissues of both patients and controls were analysed qualitatively to detect the core proteins and fragments of the aggrecan and SLRPs. GAPDH was used as an internal loading control, appearing at 36 kDa. To accurately determine the molecular weights of the SLRP core proteins and their fragments, the covalently attached GAGs side chains, including keratan sulphate, dermatan sulphate, and chondroitin sulphate were enzymatically removed using chondroitinase ABC treatment. In 4%–12% gel, lane 1 represented GAPDH (G) as internal loading control. Lane 2 was loaded with extracts from non-degenerated age-matched control (C) Grades 1 and 2 discs. Lanes 3–5 were disc extracts from patients with Pfirrmann Grades 3, 4, and 5 (P3, P4, and P5), representing increasing severity of degeneration. This setup enabled the visualization and comparison of intact aggrecan and SLRP core proteins and their proteolytic fragments across various stages of degeneration (Fig. 2).

Fig. 2.

Proteins were resolved on 4%–12% SDS-PAGE gels. Lane 1 shows GAPDH (G) as the internal loading control (C) (36 kDa). Lane 2 contains extracts from non-degenerated, age-matched control discs (Grades 1–2). Lanes 3–5 represent patient disc extracts with P3, P4, and P5, indicating increasing severity of degeneration. DCN core protein (45 kDa) exhibited fragmentation at 36 kDa and 22 kDa (A). FMOD (40 kDa) showed a consistent 22 kDa fragment in degenerated discs P4, P5 (B). LUM (45 kDa) displayed a 22 kDa fragments in P4 disc extract (C). Biglycan appeared as a doublet at 40 kDa, with weakly expressed fragments at 36 kDa (D). CHAD core protein appeared at 36 kDa, while its fragments appeared at 30 kDa in P4 and P5 discs (E). Aggrecan core protein was detected at 200–250 kDa, with fragmented peptides ranging from 35 to 22 kDa (F). Images presented in the figures are representative of various samples analysed. P3: Pfirrmann Grade 3, P4: Pfirrmann Grade 4, P5: Pfirrmann Grade 5, DCN: decorin, FMOD: fibromodulin, LUM: lumican, CHAD: chondroadherin.

2. DCN, FMOD, and LUM

Western blot analysis revealed distinct fragmented bands for the SLRPs DCN, LUM, and FMOD appearing below their respective core proteins at approximately 36 and 22 kDa (Fig. 2A–C). The core protein forms of all three SLRPs were consistently detected at 42–45 kDa. Notably, fragmented bands were predominantly present in P4 and P5 patient disc samples. Grade 1 and 2 discs from the controls and Grade 3 patient discs showed no detectable fragments, suggesting limited proteolysis at these stages.

3. Biglycan and CHAD

In both control and surgically removed P4 and P5 degenerated disc samples, as indicated in the figure, the biglycan core protein was detected as a prominent doublet band between 42 to 48 kDa, while biglycan fragments were present at 36 kDa (Fig. 2D). CHAD core protein appeared at 36 kDa, while its fragments appeared at 30 kDa in P4 & P5 discs (Fig. 2E).

4. Aggrecan

Fragmented aggrecan was detected within the molecular weight range of 35 to 22 KDa, while the core protein of aggrecan appeared as a 200–250 KDa band (Fig. 2F). In P4 disc samples aggrecan core protein showed degradation at the G1 globular domain mediated by the action of matrix metalloproteases and aggrecanase resulting in a visible fragment of 36 KDa. In P5 disc samples aggrecanase enzyme activity was confirmed by the presence of a distinct fragmented band at 45 KDa (Fig. 2F).

5. Immunofluorescence analysis to assess the organization and spatial distribution of SLRPs and aggrecan in disc tissue sections

Surgically excised P4 and P5 disc tissue revealed distinct formation of large cell clusters compared to controls which were comprised mainly of rounded cells forming small to medium sized clusters (Fig. 3; Fig. 4A, B). High resolution image analysis of patient disc tissue showed significant reduction and dispersed localisation of primary antibody staining for CHAD, LUM, biglycan, DCN, FMOD, and aggrecan, indicating disruption in the spatial distribution within the disc tissue (Fig. 4C, D). In contrast, control discs, demonstrated, well-organised alignment and precise distribution of SLRPs and aggrecan molecules (Fig. 4A, B). Since the primary objective of immunofluorescence analysis was to visualize cytoplasmic and matrix-bound localization patterns, quantitative intensity measurements were not performed.

Fig. 3.

Immunofluorescence analysis of control cadaveric disc tissue illustrating the spatial localization of SLRPs (A) DCN, (B) LUM, (C) CHAD, (D) biglycan, (E) FMOD, and (F) aggrecan. Distinct red and green fluorescence signals for DCN, LUM, and CHAD were concentrated around disc cells within the pericellular matrix. In contrast, FMOD, biglycan, and aggrecan exhibit a more diffuse distribution across both the pericellular and territorial matrix regions. Scale bar 100 μm. Images presented in the figures are representative of various samples analysed. SLRPs: small leucine-rich proteoglycans, DCN: decorin, LUM: lumican, CHAD: chondroadherin, FMOD: fibromodulin.

Fig. 4.

High-resolution double immunofluorescence staining of cadaveric control and patient (P5) disc tissues. The upper panel displays the control disc, showing localized red fluorescence for DCN (A) and CHAD (B) within the cell cytoplasm and pericellular matrix. Green fluorescence highlights pericellular expression of biglycan (A) and FMOD (B). In contrast, the lower panel illustrates the patient disc tissue, where cytoplasmic staining is diminished and red fluorescence appears dispersed within the territorial matrix, indicating altered localization of the SLRPs: DCN, biglycan, FMOD, and CHAD (C, D). Scale bar 50 μm. Images presented in the figures are representative of various samples analysed. P5: Pfirrmann Grade 5, DCN: decorin, CHAD: chondroadherin, FMOD: fibromodulin, SLRPs: small leucine-rich proteoglycans.

6. TLR-2 expression analysis

Significant differences in TLR-2 expression were observed between control and severely degenerated P4 and P5 disc cells (P < 0.001), as determined through positive cell counts in culture conditions (Table 2). To further validate these findings, TLR-2 expression in control disc cells was assessed following stimulation with a TLR-2 agonist Pam2CSK4, while P4 and P5 patient disc cells were treated with a TLR-2 antagonist/inhibitor MMG-11. Unstimulated control disc cells (without agonist treatment) exhibited low levels of TLR-2-positive cells, along with reduced expression of inflammatory markers and nociceptive growth factors, as confirmed by protein antibody array analysis of the culture medium (Fig. 5). In contrast, stimulation of control disc cells with a TLR-2 agonist resulted in elevated TLR-2 receptor expression. Similarly, P4 and P5 patient disc cells showed elevated TLR-2 receptor expression, which was attenuated upon treatment with a TLR-2 antagonist (MMG-11) (Fig. 5).

Fig. 5.

TLR-2 expression in control and patient disc cells under cell culture conditions, visualized by immunofluorescence. TLR-2-positive cells exhibited red cytoplasmic staining, while nuclei were counterstained blue. (A) Control disc cells showed low TLR-2 expression level. (B) Upon stimulation with the TLR-2 agonist Pam2CSK4, control disc cells demonstrated increased TLR-2 levels. (C) Disc cells from patient P5 group also exhibited elevated TLR-2 expression, and treatment with the TLR-2 antagonist MMG-11 attenuated TLR-2 expression in P5 disc cells (D). Scale bar 50 μm. Images presented in the figures are representative of various samples analysed. TLR-2: Toll-like receptor-2, P5: Pfirrmann Grade 5.

7. Secreted profile of inflammatory cytokines

Culture medium collected from control and patient-derived disc cells were semi-quantitatively analysed with Ray Biotech protein antibody array to assess the secreted profile of inflammatory cytokines, and growth factors. As shown in Fig. 6, disc cells from patients exhibited significant upregulation in inflammatory markers with actual fold-change values: interleukin (IL)-6 (1.37×), IL-8 (1.30×), IL-1β (1.25×), IL-1α (1.23×), GRO-α (1.22×), MIG (1.19×), G-CSF (1.21×), SCF (1.16×), and M-CSF (1.15×). Anti-inflammatory mediators were downregulated: IL-13 (0.81×), IL-1 (0.82×), IL-5 (0.84×), MCP-1 (0.85×), TGF-β (0.78×), EGF (0.80×), and vascular endothelial growth factor (VEGF) (0.81×). These values indicate a clear shift toward a pro-inflammatory and catabolic state in patient discs compared to controls exhibiting an anabolic microenvironment characterized by elevated growth factors synthesis (TGF-β, EGF, VEGF) and negligible pro-inflammatory cytokine expression.

Fig. 6.

Cytokine expression changes in patient IVD cells compared to controls. Pro-inflammatory cytokines (red) such as IL-6, IL-8, IL-1β, IL-1α, GRO-α, MIG, inflammatory growth factors such as G-CSF, SCF, and M-CSF were upregulated in patient disc, indicating an inflamed microenvironment. Anti-inflammatory regulatory cytokines (green) such as IL-13, IL-1, IL-5, MCP-1, and TGF-β, were downregulated in control discs, including growth factors such as EGF and VEGF. IVD: intervertebral disc, IL: interleukin, VEGF: vascular endothelial growth factor.

DISCUSSION

This study presents the first direct comparison between TLR-2 receptors, aggrecan, and SLRP proteolytic fragments in age-matched non-degenerated pain-free human IVD tissues and those exhibiting painful degeneration across increasing Pfirrmann Grades. The findings reveal that moderate to severely degenerated discs (P3, P4) are associated with distinct structural and molecular changes, particularly within the inner annulus and oaf, when compared to control non-degenerated Grade 1 and 2 discs, in situ. Analysis of specific SLRPs and aggrecan fragments demonstrates that they emerge predominantly during advanced stages of Pfirrmann disc degeneration (Grades 4 and 5). While the presence of aggrecan and SLRP intact core proteins in IVDs is well-established, this study is, to the authors’ knowledge, the first to report the detection of fragmented peptides of DCN, LUM, and FMOD in human discs with P4 and P5 degeneration grades. Although age-related degenerative changes in the distribution of SLRPs and aggrecan within the disc matrix have been previously described [20], limited data exist on their fragments, particularly in increasing grades of degeneration and pain along with an appropriate control. Cleaved core proteins and their fragments show an association with TLR-2 expression linking disc degradation to inflammation and pain [19,20,26]. Disc cells derived from P4 and P5 tissues exhibit increased secretions of cytokines and catabolic mediators which may be associated with inflammatory processes observed in advanced disc degeneration. In contrast, control disc cells expressed anabolic mediators and adhesion molecules. This differential expression, as demonstrated by cytokine array analysis, provides insights into the molecular alterations that may be involved in painful disc degeneration.

Abnormal mechanical loading and injury in IVDs trigger a phenotypic shift in disc cells from a normal phenotype to a hypertrophic one, characterised by large cell clusters and excessive secretion of inflammatory cytokines such as IL-1 [31,32]. Previous studies [15,23] indicate that this interaction is accompanied by increased synthesis of matrix-degrading enzymes including MMPs (–1, –3, –5, –8, –13) and ADAMTS (–4, –5), along with elevated expression of VEGF and NGF. These catabolic mediators disrupt matrix homeostasis accelerating aggrecan degradation, a key proteoglycan that interacts with collagen fibers and forms complexes with SLRPs [14,32]. The current study indicates that these proteolytic fragments of aggrecan are associated with advanced disc degeneration and are predominantly observed in P4 discs.

DCN, a class I SLRP, is pericellular in IVDs, and regulates chondrocyte cell mechanotransduction pathways in articular cartilage influencing matrix micromechanics [33,34]. In adult and ageing discs, multiple forms of DCN are reported across the annulus fibrosus, np, and cartilage endplate. Young discs predominantly contain glycanated, GAG-rich DCN in the annulus fibrosus and cartilage endplate, while these forms are scarce in the nucleus pulpous. In aged discs, glycanated DCN is minimal or absent making its GAG side chain a marker of age-associated degeneration [35]. Dermatan sulphate is the main GAG side chain attached to DCN in the annulus fibrosus, whereas chondroitin sulphate predominates in the cartilage endplate. DCN undergoes post-translational modifications in diseases such as cancer [36] and osteoarthritis (OA) [37], and are cleaved by the matrix degrading enzymes. The present findings align with the previous reports, showing DCN fragments appear in advanced stages of disc degeneration.

Biglycan, like DCN, is a class I SLRP localized in the territorial matrix of IVDs; both biglycan and DCN are associated with tissue remodelling and inflammatory signalling. Soluble biglycan fragments, detected in OA and rheumatoid arthritis, correlate with disease progression and inflammasome activation via TLR-2, TLR-4, and purinergic receptor pathways [25,38,39]. The authors’ findings align with these observations, showing biglycan fragments in advanced disc degeneration, particularly near annular tears in an experimental animal model, while being absent in age-matched controls [17]. This association suggests that disc cells in degenerative conditions secrete proteases and cytokines that modify biglycan, releasing intact, soluble, and fragmented peptides indicative of persistent inflammation [21]. The presence of biglycan fragments alongside DAMP receptors (TLR-2, TLR-4, P2X7/4) further supports its role in autoinflammatory and autoimmune processes contributing to cartilage and disc degeneration [40,41].

LUM, a class II SLRP, interacts with fibrillar collagens to maintain disc structure and is associated with TLR-2, TLR-4 mediated inflammatory signaling [42]. In OA, LUM expression decreases by over 50%, while its fragments are elevated in plasma, serum, and synovial fluid, indicating dysregulation in cartilage tissue [43]. TLR-4-expressing chondrocytes are associated with enhanced release of diverse DAMP molecules resulting from matrix proteolysis, thereby implicating LUM in the activation of innate immune pathways [44]. In severely degenerated P5 discs, LUM fragments were detected, indicating their depletion during degeneration, likely attributable to their small molecular size (< 22 kDa), which may facilitate their escape from the matrix in conjunction with collagen fibril disorientation [45].

FMOD, a class II SLRP, associates with tissue homeostasis and inflammatory regulation in articular cartilage disorders, where FMOD correlates with impaired cell adhesion and inflammatory damage [39,46]. Its interaction with complement components C1q and factor H links FMOD fragments to immune modulation, autoimmunity, and cytokine activation [47,48]. In OA, N-terminal FMOD fragments associate with IL-1 induction via C1q, while intact FMOD regulates complement activation [48]. In IVDs, FMOD expression shows an inverse association with methylation during progressive degeneration [49]. Bioinformatic analyses suggest that FMOD modulates cytoskeletal organization and integrin-mediated signaling through interactions with type I collagen, thereby promoting cell MIG and proliferation in fibrotic discs. Conversely, FMOD loss or silencing is associated with increased apoptosis and enhanced autophagy, whereas exogenous FMOD supplementation attenuates autophagic flux [46,47].

CHAD, a class IV SLRP, is localized within the pericellular matrix of chondrocytes and mediates adhesion to type II collagen via integrin α2β1, thereby facilitating cell–matrix communication and collagen fibril organization [16]. CHAD fragments have been associated with adolescent idiopathic scoliosis and disc degeneration, whereas they are reportedly absent in healthy tissues [15,50]. Its spatial distribution and C-terminal binding sequence correlate with cell adhesion and suggest potential interactions with heparan sulphate proteoglycans [16]. Experimental models deficient in CHAD exhibit delayed load-bearing capacity, highlighting its role in maintaining collagen network integrity. Fragmentation of SLRPs, including CHAD, is also associated with altered osmotic properties, diminished cartilage stiffness, and variability in collagen fibril diameter, underscoring their critical contribution to disc homeostasis and the pathogenesis of disc degeneration [15,18].

Aggrecan fragmentation and depletion are associated with impaired osmotic swelling capacity, which compromises the hydraulic permeability of the np. These alterations have been linked to a marked reduction in the effective indentation modulus of the superficial articular cartilage zone and increased variability in collagen fibril diameters within disc tissue [15,18]. Interestingly, the middle and deep cartilage zones exhibit no significant changes, with aggrecan concentrations remaining relatively high in these regions [15,18]. Collectively, current and previous findings highlight a complex interplay between SLRPs and aggrecan synthesis, emphasizing their yet to be fully explored roles in preserving IVD homeostasis, including cell adhesion, proliferation, survival, and the progression of disc pathology.

TLR-2–mediated inflammatory responses may be initiated by SLRP and aggrecan fragments, disrupting cell–matrix interactions in conditions such as OA, cancer, diabetes, skin disorders, tendinopathies, and, as demonstrated in the present study, with increasing grades of disc degeneration [24,25]. Elevated TLR expression and peptide fragmentation correlate with a catabolic, pro-inflammatory microenvironment that promotes cytokine release and the activity of matrix-degrading enzymes [25]. Matrix metalloproteinases (MMPs 1, 2, 3, 5, 7, 10, 13) contribute to ECM cleavage, generating fragments that activate TLR-2 in disc cells, thereby triggering inflammation, neovascularization, and nerve ingrowth in normally avascular discs [4,23,26]. Furthermore, gene expression analyses reveal upregulation of TLRs (1, 2, 4, 6, 9, 10) in association with disc degeneration, OA, and certain cancers [14,19,26].

The qualitative and quantitative observations from the present study suggest a potential link between ECM fragmentation and activation of innate immune pathways via TLR-2 signaling. This association may contribute to a catabolic and pro-inflammatory microenvironment, promoting cytokine release, neovascularization, and nerve ingrowth within the normally avascular disc tissue. These findings are consistent with previous reports in OA and other connective tissue disorders, indicating shared molecular mechanisms underlying chronic musculoskeletal pain. Western blot and immunofluorescence analyses were employed as qualitative assays to confirm the presence and localization of SLRPs, aggrecan, and their fragments; therefore, no densitometric measurements, fragment-to-core ratios, or statistical analyses were performed. Interpretation of these results is limited by factors such as the use of cadaver-derived non-degenerate discs as controls, potential post-mortem changes, grading variability, and regional heterogeneity in tissue sampling; however, these were minimized through standardized protocols as described in the MATERIALS AND METHODS section. Future investigations incorporating fresh surgical controls, real-time imaging, and in vitro models are warranted to validate TLR-2 activation and elucidate mechanistic associations.

In conclusion, the authors’ findings support the proposed hypothesis that disc degeneration is associated with increased TLR-2 expression and shows the presence of fragmented peptides of DCN, biglycan, LUM, FMOD, CHAD, and aggrecan. These fragments progressively increase with higher P4 and P5 and are absent in non-degenerate controls. Future research should thus focus on validating these observations through controlled mechanistic studies and exploring clinical strategies to modulate TLR-2 signaling as a potential approach to reduce inflammation and pain in degenerative disc disease. Additionally, longitudinal studies combined with advanced proteomic techniques are warranted to elucidate the temporal dynamics of matrix fragmentation and receptor activation, paving the way for precision diagnostics and targeted therapeutic interventions.