Continuous low-intensity ultrasound attenuates IL-6 and TNFα-induced catabolic effects and repairs chondral fissures in bovine osteochondral explants

Bovine osteochondral explants (8 mm diameter, 1-2 mm cartilage thickness and 5 mm bone segment) from stifle joints were purchased from Articular Engineering (Northbrook, IL, USA). A full-thickness cylindrical incision in the cartilage segment of the osteochondral explant was created by a 4 mm biopsy punch (Integra™ Miltex®, USA). Following incision, the osteochondral explants were categorized as (a) Control (No cytokine or no-cLIUS stimulation) (b) cLIUS (c) IL-6 (d) cLIUS+IL-6 (e) TNFα (f) cLIUS+ TNFα with each study group consisting of at least 12–20 explants. The study design is schematically depicted in Fig. 1. The culture media for explants was DMEM-F12 supplemented with 10% FBS, 1× antibiotic-antimycotic solution (Gibco, USA) and 50 μg/ml ascorbic acid. IL-6 or TNFα (R&D Systems, Minneapolis, MN) was added to appropriate wells at 20 ng/ml concentration as IL-6 or TNFα dosage of 10–25 ng/ml demonstrated enhanced catabolic effects in chondrocyte cultures [16, 17]. All explants were placed in 6-well ultra-low attachment tissue culture plates (TCPs) and were either incubated in a CO₂ incubator (37 °C, 5% CO₂) or were cultured in the ultrasound-assisted bioreactor developed at UNL for 28 days. Media were replenished every 2 days. Sampling for outcome analyses was done on day 14 and day 28 of culture.

Bovine articular chondrocytes were isolated from stifle joints as described elsewhere [18] and cultured in DMEM-F12 supplemented with 10% FBS, 1× antibiotic-antimycotic solution (Gibco, USA) and 50 μg/ml ascorbic acid in 6 well TCPs for quantitative real time polymerase chain reaction (qRT-PCR) and scratch assays. The wells (n = 3 per group) were categorized as (a) Control (No cytokine or cLIUS treatment) (b) cLIUS (c) IL-6 (d) cLIUS+IL-6 (e) TNFα (f) cLIUS+ TNFα and incubated in a CO₂ incubator (37 °C, 5% CO₂). IL-6 or TNFα (20 ng/ml) was added to appropriate wells.

cLIUS (< 20 mW cm^− 2) treatment was conducted in a custom-designed ultrasound-assisted bioreactor developed and characterized at the Department of Chemical and Biomolecular Engineering, UNL, USA, with operating procedures described elsewhere [15, 19]. cLIUS was applied to explants at 5 MHz at a constant pressure amplitude of 14 kPa at input voltage of 2.5 Vpp (peak-to-peak voltage) for 20 min per application and 4 applications per day for a period of 28 days. In a separate experiment, explants were exposed to cLIUS at 2 MHz, a frequency outside the resonant bandwidth [13], for a period of 14 days (n = 6) in the cLIUS-assisted bioreactor at the constant pressure amplitude of 14 kPa (6.0 Vpp, 20 min per application,4 applications/day). For adherent chondrocyte cultures, cLIUS was applied using non-focused immersion transducers (Panametrics V300, 12.7 mm diameters, Panametrix, Waltham, MA, USA) at 5 MHz and 2.5 Vpp (or 14 kPa) for either 5 or 20 min.

Cell viability was assessed by Live/Dead Viability/Cytotoxicity kit (Molecular Probes, USA) in the osteochondral explants of each group (n = 3 per group per time point) on day 3 and 21 of culture. The explants were incubated with 6 μM calcein^AM and 4 μM ethidium homodimer-1 solution for 1 h at 37 °C under aseptic conditions and visualized by confocal microscopy (Olympus 1× 81) at 10× magnification (z step size = 5 μm, 30–40 slices per sample, ~ 150-200 μm deep) from the top or superficial zone of cartilage at the interfacial region.

Explants of each group (n = 6 per group) were typically fixed in 10% neutral buffered formalin and decalcified using a 1:1 ratio of 8% hydrochloric acid and 8% formic acid solution. The explants were cut in half and each half was paraffin embedded, sectioned and stained at the Tissue Science Facility, University of Nebraska Medical Center (Omaha, NE, USA) using standard protocol and were reviewed by Dr. Steven H. Hinrichs (Professor, Department of pathology and microbiology, UNMC, Omaha, NE, USA). Histological staining with Alcian Blue (pH 1) and Safranin O solution (Millipore, USA), and immunohistochemical staining with collagen II (ab34712, Abcam, USA) were performed on deparaffinized sections (4 μm) and visualized at 2× and 20× magnification. Normal bone tissue sections served as negative control and normal colon tissue served as positive control for collagen II staining. Fast green (Millipore, USA) was used as a counterstain for Safranin O staining. All sections were stained using the same staining solution and at the same time. Representative image of 18–36 stained interfacial regions obtained from each group (n = 6) was used.

Alcian blue stained explants were processed to quantify percent apposition by measuring the length of the bonded interfaces (no gaps) divided by the total length of the interface [20]. Apposition was defined as the region along the interface that was bonded and did not show any discernable gaps [21]. All measurements were performed using histological images at 2× magnification (n = 6) using ImageJ software.

Bovine articular chondrocytes were plated at an initial seeding density of 2 × 10⁵ cells/well. cLIUS was applied to appropriate wells for 5 min followed by homogenization with Trizol. RNA was then extracted using RNeasy Mini Kit (Qiagen, USA) as per the manufacturer’s protocol. Homogenates from 2 wells per group served as one replicate and 3 such replicates were used for analysis (n = 3). The qRT-PCR analysis was carried out on Realplex™ real-time PCR system (Eppendorf, USA) using TaqMan® RNA-to-CT™ 1-Step Kit (Life Technologies) as per manufacturer’s guidelines. TaqMan® probes and primers (Life Technologies, USA) used are as follows: GAPDH (Bt03210917_g1), MMP13 (Bt03214050_m1), ADAMTS4 (Bt03224697_m1), NF-κB (Bt03243457_m1), Collagen II (Bt03251861_m1) and TIMP1 (Bt03223720_m1). The expression of mRNA transcripts was normalized to GAPDH expression and relative expression levels were calculated using the 2^-ΔΔCt method.

Bovine articular chondrocytes were plated at a seeding density of 1 × 10⁵ cells per well (n = 3). Upon 90% confluency followed by serum starvation, a scratch representing a gap was created by a 200 μl pipette tip [11]. cLIUS was applied for 20 min to appropriate wells. Gap area was imaged at 0 h, 6 h, 12 h, 24 h, 48 h and 72 h following the creation of scratch by bright field inverted microscope at 5× magnification and the gap area covered by migrating cells was quantified by ImageJ.

To visualize the migration of cells in a 3D format, a 4 mm central chondral region was cored out of the osteochondral explant using a biopsy punch and filled with sterile 1% agarose hydrogel. The cartilage-hydrogel constructs were categorized as (a) Control (No cytokine or cLIUS treatment) (b) cLIUS (c) IL-6 (d) cLIUS+IL-6 (e) TNFα (f) cLIUS+ TNFα and cultured in DMEM-F12 supplemented with 10% FBS, 1× antibiotic-antimycotic solution and 50 μg/ml ascorbic acid. IL-6 or TNFα was added at a concentration of 20 ng/ml and cLIUS was applied at 14 kPa (5 MHz, 2.5 Vpp) for 20 min to appropriate wells. The explants (n = 3 per group) were incubated with 6 μM calcein^AM and 4 μM ethidium homodimer-1 solution (Live/Dead Viability/Cytotoxicity kit, Molecular Probes, USA) for 1 h at 37 °C on day 24 of culture. The central region of interest was imaged by confocal microscopy (Olympus 1× 81) at 10× magnification (5 μm Z-size, 30–40 slices). Transmitted light micrograph at 10× magnification was included to distinguish hydrogel region of the construct from the native cartilage.

The data are represented as mean ± 95% confidence interval where mean is calculated from 3 separate experiments, each conducted in triplicate per time point per study group. For gene expression and percent apposition, the statistical significance between the means of two groups with unequal variance was calculated by Welch’s test. For NF-κB gene expression, Welch’s analysis of variance (ANOVA) was used to compare control, TNFα and cLIUS+TNFα groups followed by post-hoc Games-Howell test for pair-wise comparisons. For scratch assays, two-way ANOVA with post-hoc Turkey’s test was used, and for cases where variance homogeneity was violated, Welch’s ANOVA followed by Games-Howell test was used for pair-wise comparisons. P values < 0.05 were considered significant and exact p values were indicated in the figures.

Cellular viability of the osteochondral explants in the vicinity of the incision was examined by live-dead assay and shown in Fig. 2. Cells surrounding the site of incision were predominantly live (green) interspersed with few dead (red) cells. On day 3 of culture, chondral fissure was noticeable as a distinct gap in all the explants (Fig. 2a-f). After 21 days in culture, explants that received cLIUS stimulation either in the presence or absence of IL-6 or TNFα displayed enhanced cellular infiltration into the fissure when compared to non-cLIUS-stimulated controls (Fig. 2j-l).

Explants were stained with alcian blue and shown in Fig. 3. Explants that received cLIUS at 5 MHz displayed closed fissures (Fig. 3d) on day 14 and a percent apposition of 58.04% (Fig. 4) was noted. In comparison, explants that received cLIUS at the non-resonant frequency of 2 MHz displayed intact gaps (Additional file 1: Figure S1) along with a percent apposition of 5.75% (Fig. 4). Percent apposition in non-cLIUS-stimulated explants at the end of 14 days was 3.32% (Fig. 4). Thus, subsequent experiments examining the effect of cytokine on the closure of fissures was performed at cLIUS at 5 MHz.

Explants that received cLIUS (5 MHz) in the presence or absence of IL-6 or TNFα displayed closed gaps on day 14 (Fig. 3d-f) and day 28 of culture (Fig. 3j-l). In contrast, non-cLIUS-stimulated explants and explants treated with either IL-6 or TNFα alone displayed intact (Fig. 3a, b) or partially narrowing fissures (Fig. 3c) on day 14, and closed fissures on day 28 of culture (Fig. 3g-i). The quantification of percent apposition corroborated the alcian blue staining results and are shown in Fig. 4. On day 14 in the absence of cytokine, the percent apposition was significantly (p = 0.027) increased from 3.32% in non-cLIUS-stimulated control to 58.04% under cLIUS. The inclusion of cytokines in the culture medium led to a higher percent apposition in non-cLIUS-stimulated controls. The percent apposition remained significantly elevated under cLIUS when compared to non-cLIUS-stimulated controls in the presence of IL-6 or TNFα. On day 28 in the absence of cytokine, the percent apposition was significantly higher in cLIUS-stimulated explants (p = 0.039) when compared to non-cLIUS-stimulated control. In contrast, no significant difference was noted between cLIUS and non-cLIUS-stimulated explants in the presence of either IL-6 or TNFα on day 28, corresponding to closed fissures observed in Fig. 3 (h, i, k, l).

After 28 days, explants exposed to IL-6 alone displayed weak alcian blue stain (Fig. 3h), indicating depletion of proteoglycans. In contrast, intense alcian blue stain was noted in IL-6-treated explants that received cLIUS stimulation (Fig. 3k). Further analysis by safranin O staining displayed noticeable depletion of proteoglycans in explants treated with either IL-6 or TNFα alone (Fig. 5b-c). cLIUS stimulation of IL-6 or TNFα-treated explants showed an intense and homogenous Safranin O stain (Fig. 5e-f). Similarly, after 28 days in culture, immunohistochemical staining of explants for collagen II exhibited weaker staining in non-cLIUS stimulated explants in the presence or absence of IL-6 or TNFα (Fig. 5g-i) when compared to cLIUS-stimulated explants (Fig. 5j-l).

Taken together, the results indicated that cLIUS at 5 MHz closed chondral fissures while preserving the proteoglycan and collagen distribution in the presence of IL-6 or TNFα; thus, demonstrating a chondroprotective effect.

To further ascertain the chondroprotective effect of cLIUS, the expression of select catabolic (MMP13, ADAMTS4) and anabolic genes (Collagen II and TIMP1) were evaluated in cultured chondrocytes by qRT-PCR and shown in Fig. 6. The expression of the transcription factor (NF-κB) in regulating the expression of catabolic genes was also assayed. In the absence of any cytokine insult, the exposure of chondrocytes to cLIUS reduced the gene expression of MMP13 (p = 0.037) and ADAMTS4 (p = 0.001) when compared to non-cLIUS-stimulated controls (indicated by rectangle in Fig. 6a, b). The inclusion of IL-6 or TNFα significantly elevated the gene expression of MMP13 (10.48 ± 7.43-fold in IL-6 and 3.55 ± 1.07-fold in TNFα) and ADAMTS4 (92.26 ± 61.10-fold in IL-6; 2.26 ± 0.46-fold in TNFα) in non-cLIUS-stimulated chondrocytes (indicated by open arrows in Fig. 6a, b). Whereas, cLIUS diminished the IL-6 or TNFα-induced upregulation of the catabolic markers (p = 0.038 in cLIUS+IL-6 versus IL-6, and p = 0.004 in cLIUS+TNFα versus TNFα for MMP13; p = 0.049 in cLIUS+IL-6 versus IL-6, and p = 0.036 in cLIUS+TNFα versus TNFα for ADAMTS4), and returned the gene expression to baseline levels observed under cLIUS-stimulation only (indicated by closed arrows in Fig. 6a, b). As anticipated, the exposure of chondrocytes to TNFα led to an increase in the gene expression of NF-κB (1.44 ± 0.03-fold in TNFα versus 1.01 ± 0.01-fold in control, p = 0.019) while no discernable differences in NF- κB gene expression were noted when chondrocytes were exposed to IL-6 (Fig. 6c). Under cLIUS, the gene expression of NF- κB was suppressed in chondrocytes exposed to TNFα (0.75 ± 0.13-fold in cLIUS+ TNFα versus 1.44 ± 0.03-fold in TNFα, p = 0.046) (Fig. 6c).

The gene expression of the cartilage-specific marker, collagen II, remained significantly elevated (5.85 ± 0.57-fold p = 0.004 versus non-cLIUS-stimulated control) under cLIUS stimulation irrespective of IL-6 (5.88 ± 0.95-fold in cLIUS+IL-6 versus 0.89 ± 0.07-fold in IL-6; p = 0.009) or TNFα treatment (5.01 ± 0.41-fold in cLIUS+TNFα versus 0.63 ± 0.11-fold in TNFα; p = 0.003) (Fig. 6d). Significant increases (3.15 ± 1.10-fold in cLIUS+TNFα versus 0.61 ± 0.30-fold in TNFα, p = 0.038) in the gene expression of TIMP1, an anabolic inhibitor of metalloproteinases, was observed when IL-6-treated chondrocytes were exposed to cLIUS (Fig. 6e). TNFα treatment had no significant effect on the gene expression of TIMP1 with or without cLIUS stimulation.

The results indicated that the catabolic genes were downregulated under cLIUS and this downregulation was sustained in the presence of pro-inflammatory cytokines IL-6 and TNFα. In addition, cLIUS stimulation demonstrated elevated levels of anabolic genes regardless of cytokine treatment.

Increasing the availability of cells for migration at the wound edge has been shown to improve integration of cartilage surfaces [21, 22]. Therefore, cellular migration under cLIUS in the presence of IL-6 or TNFα was studied in 2D and 3D formats. Migration of cells from the surrounding cartilage tissue into the cell-free central hydrogel core in cartilage-hydrogel constructs was visualized by live (green)/dead (red) staining and presented in Fig. 7 under different treatment conditions. On day 24, an enhanced cell migration to the hydrogel core was noted in all constructs that received cLIUS stimulation irrespective of IL-6 or TNFα treatment (Fig. 7d-f).

To better explain the results observed in 3D format in Fig. 7, a 2D scratch was employed to study the migration of adult chondrocytes under cLIUS stimulation in the presence or absence of IL-6 or TNFα and is shown in Fig. 8a. After 48 h of cLIUS stimulation, the migration of chondrocytes was significantly increased in the presence or absence of IL-6 or TNFα, as evidenced by 55.27 ± 5.07%, 78.94 ± 1.65%, and 38.96 ± 3.89% coverage of the scratch area in cLIUS, cLIUS+IL-6, and cLIUS+TNFα samples, respectively (indicated by black arrows in Fig. 8b-d). In contrast, the scratch area covered in non-cLIUS stimulated, and chondrocytes treated with IL-6 or TNFα alone was significantly reduced to 28.53 ± 4.94%, 22.62 ± 18.39% and 11.99 ± 2.67% respectively.