The Research Ethics Committee of Changzhou No. 2 People's Hospital (Changzhou, China) approved the resection of all specimens. Written informed consent was obtained by the patients or their relatives to obtain human intervertebral tissue at surgery. NP samples (n=22) were obtained from patients (33–78 years old; mean age, 53 years) who underwent disc resection surgery or spinal fusion to relieve lower back pain. In addition the NP specimens were grouped according to a grading system for IVD degeneration, which was based on preoperative magnetic resonance images (MRI) (51). No medications or anti-inflammatory drugs were used before surgery. In the present study, samples graded II–III (n=12) were designated the low degeneration group (used as a relatively normal control); samples graded IV–V (n=10) were designated the high degeneration group. All tissues were snap-frozen in liquid nitrogen at the time of surgical removal and stored at 80°C or were fixed in 10% neutral buffered formalin until use.

NP tissues were identified by their macroscopic morphology and were carefully separated from any obvious granulation tissue, cartilaginous endplates, or annulus fibrosus as previously described (52). Isolation of NP cells was performed as previously described (25). Briefly, cells were extracted from minced NP tissue specimens from 12 low graded II–III, who underwent posterior discectomy surgery of lumbar degenerative disease, by digesting in 2 U/ml protease in Dulbecco's modified Eagle's medium (DMEM)/F12 (Gibco; Thermo Fisher Scientific, Waltham, MA, USA) for 30 min at 37°C followed by 0.25 mg/ml type II collagenase (Gibco; Thermo Fisher Scientific) for 4 h at 37°C. The cell suspension was centrifuged at 800 × g for 5 min and NP cells were resuspended in DMEM/F12 (10% FBS, 100 U/ml penicillin, 100 µg/ml streptomycin and 1% L-glutamine). Subsequently, the NP cells were treated with 10 µg/ml of LPS (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany).

For miR-140 overexpression and knockdown, miR-140 mimics, miR-140 inhibitor and two scrambled miRNAs used as negative controls (mimics NC for miR-140 mimics and inhibitor-NC for miR-140 inhibitor, respectively), were purchased from Shanghai GeneChem, Co., Ltd. (Shanghai, China). The 20 nM miRNAs were transfected into NP cells using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific) following the manufacturer's instructions.

Total RNA was extracted from cultured NP cells and degenerative IVD tissues using TRIzol reagent (Invitrogen; Thermo Fisher Scientific) following the manufacturer's instructions. RNA molecules smaller than 200 nucleotides in size were purified using a mirVana miRNA isolation kit (Ambion, Austin, TX, USA) according to the manufacturer's instructions, as previously described (53). Complementary DNA (cDNA) was synthesized using an M-MLV kit (Life Technologies; Thermo Fisher Scientific). qRT-PCR was carried out using the Platinum SYBR-Green qPCR SuperMix-UDG kit (Life Technologies; Thermo Fisher Scientific) on an ABI Prism 7500 PCR system (Applied Biosystems; Thermo Fisher Scientific). Data were normalized to β-actin. qRT-PCR analysis of mature miR-140 was performed using a TaqMan MicroRNA Assay kit (Applied Biosystems; Thermo Fisher Scientific). Briefly, TaqMan-based real-time quantification of miR-140 included two steps: stem-loop RT and real-time PCR. Stem-loop RT primers bound at the 3′ portion of miR-140 molecules and were reverse transcribed with reverse transcriptase. Subsequenlty, the RT product was quantified using conventional TaqMan PCR that included miR-140-specific forward primer, reverse primer and a dye-labeled TaqMan probe. The purpose of the tailed forward primer at the 5′ end was to increase the melting temperature (Tm) of the miR-140 molecules. Data were normalized to U6 snRNA. All primers are listed in Table I. Each test was performed in triplicate. Relative expression was calculated using the 2^−∆∆Ct method (54).

Total protein was extracted from NP cells using a protein extraction kit and quantified using a BCA assay kit (Sigma-Aldrich; Merck KGaA). Following separation by SDS-PAGE, 60 µg of protein was transferred to a PVDF membrane and blocked for 1 h with TBS buffer (10 mM Tris-HCl, pH 7.5 and 150 mM NaCl) containing 5% skimmed milk. The membrane was then probed with primary antibody at 4°C overnight. The primary antibodies were TLR4 (1:500; cat. no. 13556; Abcam) and IκΒα (1:500; cat. no. 4814), p-IκΒα (1:500; cat. no. 9246), p65 (1:500; cat. no. 8242), and p-p65 (1:500; cat. no. 3039) (all were from Cell Signaling Technology, Beverly, MA, USA). Following three 5-min washes in TBS containing Tween-20 (TBST), the membrane was probed with secondary antibody, goat anti-rabbit IgG-HRP (1:10,000; cat. no. ab6721) and rabbit anti-mouse IgG H&L (HRP) (1:10,000; cat. no. ab6728) (all from Abcam) at room temperature for 2 h. After further washing in TBST, immunolabeling was detected using Pierce ECL Western Blotting Substrate (Pierce Biotechnology; Thermo Fisher Scientifc). β-actin was included as a control and densitometric analysis was performed to determine the relative expression of the target proteins.

The luciferase reporter assay was performed as previously described (39). The putative miR-140 binding sites in the 3′-UTR of TLR4 was predicted using TargetScan (http://www.targetscan.org); 3′-UTR cDNA fragments of TLR4 containing the putative wild-type or mutant miR-140 binding sites were amplified using the following primers: wild-type 3′-UTR of TLR4, forward, 5′-GTTTAAGACGTGCTTCAAATATCCA-3′ and reverse, 5′-TGATAAGACCAGGAAGCGGA-3′; mutant 3′-UTR of TLR4, forward, 5′-TCAAATACCATATTATGGTGTA-3′ (the bold text indicates the mutation in the 3′-UTR of TLR4) and reverse 5′-AAACTTCTGCTGCAACTCTCATT-3′. The amplified cDNA fragments were subcloned into a psiCHECK-2 vector (Promega, Madison, WI, USA) at the XhoI and NotI sites downstream of the luciferase gene. A total of 293 cell lines were co-transfected with the luciferase reporter systems and miR-140 mimics or mimics NC as indicated in the figure legends. Luciferase activity was detected 24 h after transfection using the Dual-Luciferase reporter assay system (Promega) following the manufacturer's instructions. Data were normalized to Renilla luciferase activity.

Tissues were fixed in 10% neutral buffered formalin, embedded in paraffin, dewaxed in xylene and dehydrated in alcohol. Endogenous peroxidase activity was blocked by hydrogen peroxide pretreatment for 15 min using avidin/biotin blocking kit (ab64212; Abcam). Paraffin-embedded, formalin-fixed tissues were immunostained with rabbit polyclonal primary antibody against human TLR4 (1:500; cat. no. ab13556), IL-1β (1:300; cat. no. ab9722), IL-6 (1:300; cat. no. ab6672), TNF-α (1:400; cat. no. ab6671), aggrecan (1:500; cat. no. ab36861) and collagen type II (1:500; cat. no. ab34712) (all from Abcam) overnight at 4°C. Rabbit IgG (cat. no. ab109489; Abcam) replaced the primary antibody as negative control (at an equal protein concentration) after washing the samples in phosphate-buffered saline (PBS). The second goat anti-rabbit HRP antibody (cat. no. ab80437; Abcam) was incubated for 30 min at room temperature. Then, the signal was amplified and visualized with diaminobenzidine-chromogen, followed by counterstaining with hematoxylin. TLR4, inflammation cytokines and ECM protein immunostaining were analyzed using intensity and distribution measurements as previously described (55,56) and the staining intensity was determined as follows: 0, no staining; 1, weak; 2, moderate and 3, strong. The staining distribution was determined as the percentage of positive cells (0, none; 1, <25%; 2, 25–75%; and 3, 75–100%). The expression of TLR4, IL-1β, IL-6, aggrecan and collagen II was evaluated by multiplying the intensity and distribution. Cells were further divided according to TLR4 expression into ‘low’ (TLR4 low) and ‘high’ (TLR4 high) groups, according to a cut-off point. The cut-off point for TLR4 expression was calculated using the X-tile software program (The Rimm Lab at Yale University (New Haven, CT, USA); http://www.tissuearray.org/rimmlab) as previously described (57).

For immunofluorescent staining, cells were first fixed in 4% PFA for 10 min at room temperature, and then incubated overnight in anti-aggrecan (1:100; cat. no. ab36861) or anti-collagen II (1:100; cat. no. ab34712) antibody (all from Abcam), respectively. The following day, cells were labeled with the corresponding goat anti-rabbit IgG H&L (Cy3^®) preadsorbed ab6939 (Abcam) for 2 h at 37°C, counterstained with DAPI and observed under an Olympus FluoView 2000 laser scanning confocal microscope (Olympus Corp., Tokyo, Japan).

Results are presented as the mean ± standard deviation (SD). Data analysis was performed using SPSS version 13.0 (SPSS, Inc., Chicago, IL, USA). Student's t-test or one-way ANOVA was applied to test differences between the groups. Pearson's Chi-square test was used to analyze the relationship between the TLR4 expression and the clinicopathological features. P<0.05 were considered to indicate a statistically significant result.

The expression of TLR4 in the clinical characteristics of patients with high and low-grade IVD degeneration are detailed in Table II. There was no significant difference between the level of TLR4 expression and age, sex, body mass index and disc level. However, TLR4 was expressed at a significantly higher level in the tissue of patients with high-grade IVD degeneration and at a significantly lower level in the tissue of patients with low-grade IVD degeneration. Relative expression of TLR4 and miR-140 was assessed by quantitative real-time PCR (qRT-PCR) in high- and low-grade (used as a relatively normal control) degenerative IVD tissue (Fig. 1A and B). TLR4 was found to be expressed at a higher level in grade IV–V tissues compared to grade II–III (P<0.01). In contrast, miR-140 was downregulated in grade IV–V tissues compared to grade II–III tissues (P<0.05). TLR4 was also detected by immunohistochemistry (IHC) in IVD degenerative tissues. Staining of 12 grade II–III tissues was less intense than that of 10 grade IV–V tissues indicating a high level of TLR4 in high-grade tissue (Fig. 1C). In addition, a higher level of the inflammatory cytokines IL-6, IL-1β and TNF-α was detected in high-grade tissues than in low-grade tissues (Fig. 1D-F). However, a lower level of aggrecan and collagen II expression was detected in high-grade tissues than in low-grade tissues (Fig. 1G and H). These results demonstrated that TLR4 and inflammatory cytokines were upregulated in the process of IVD tissue degeneration whereas miR-140 and ECM proteins were downregulated.

Subsequently, we assessed the expression of TLR4 and miR-140 in human NP cells stimulated with LPS. As expected, qRT-PCR analysis revealed that TLR4 expression was significantly increased in cells treated with a series of LPS concentrations (0.01–10 µg/ml) (Fig. 2A). The highest level of TLR4 expression was found in response to 1 µg/ml LPS (P<0.01 vs. the control). In contrast, the expression levels of miR-140 decreased significantly with increasing concentrations of LPS (Fig. 2B) with the lowest expression level in response to 10 µg/ml LPS (P<0.01 vs. the control). A time series of LPS (1 µg/ml) stimulation over 48 h revealed the highest levels of TLR4 expression were at 24 h (P<0.01 vs. the control), after which the levels began to decrease (Fig. 2C). Whereas, the expression of miR-140 was decreased (Fig. 2D) and continued to decrease with the lowest level at 48 h (P<0.01 vs. the control). These results indicated that the presence of LPS induced TLR4 activation which reached a peak at a specific level. In contrast, the continued presence of LPS inhibited the expression of miR-140.

NP cells were transfected with miR-140 mimics, miR-140 mimics-negative control (mimics NC), miR-140 inhibitor or miR-140 inhibitor-negative control (inhibitor NC) for 24 h. The expression of miR-140 was detected by qRT-PCR (Fig. 3A and B) and confirmed the activity of the miR-140 mimic and miR-140 inhibitor. Fig. 3C displays the putative miR-140 binding site in the TLR4 3′-UTR. A luciferase reporter plasmid containing wild-type or mutant TLR4 was co-transfected into NP cells with miR-140 mimics or mimics NC. Luciferase activity was determined at 24 h after transfection using a dual-luciferase assay and demonstrated as the relative luciferase activity normalized to Renilla activity. A significantly lower luciferase activity was found with the miR-140 mimic then with the wild-type TLR4 3′-UTR (P<0.01; Fig. 3D). We then assessed the expression and protein levels of TLR4 in NP cells transfected with the miR-140 mimic and miR-140 inhibitor. TLR4 was significantly downregulated with the miR-140 mimic (P<0.01) while it was upregulated with the miR-140 inhibitor (P<0.01; Fig. 3E). Similar results were obtained with western blotting (P<0.01; Fig. 3F). The upregulation of TLR4 in the absence of miR-140 or when the putative 3′-UTR binding site of miR-140 was mutated indicated that miR-140 may regulate TLR4.

The influence of miR-140 on the expression of inflammation cytokines was assessed in NP cells stimulated with LPS. The expression of TLR4, interleukin (IL)-1β, IL-6 and tumor necrosis factor (TNF)-α was significantly increased by LPS (P<0.01 vs. the control), however, these increased levels of expression were lower in cells co-transfected with miR-140 mimic. Furthermore, without LPS treatment, the expression of TLR4, IL-1β, IL-6 and TNF-α was also significantly lower in cells transfected with miR-140 mimic (P<0.01 vs. mimic NC) (Fig. 4A-D). The expression levels of the ECM components aggrecan and collagen II were also assessed in LPS-stimulated NP cells (Fig. 4E and F). Aggrecan and collagen II were both downregulated in response to LPS stimulation (P<0.05 vs. control for both), whereas the expression of aggrecan and collagen II was reversed by co-transfecting cells with miR-140 mimic. In addition, without LPS stimulation, the aggrecan and collagen II expression were also significantly increased when the cells were transfected with miR-140 mimics (P<0.01 vs. mimic NC). Aggrecan and collagen II expression was also visualized in NP cells by immunofluorescence staining (Fig. 4G). Nuclei were stained with DAPI and untransfected NP cells were used as a control. LPS stimulation clearly reduced the expression of aggrecan and collagen II in NP cells. However, in cells co-transfected with miR-140 mimic, aggrecan and collagen II levels were similar to levels in control cells under LPS treatment; however, without LPS stimulation, Collectively, these results imply that overexpression of miR-140 can inhibit the effect of TLR4 on IVD inflammation and degeneration induced by LPS stimulation.

Finally, we assessed the TLR4 expression and NF-κB activation in response to LPS stimulation in NP cells by measuring levels of IκBα, p-IκBα, p65 and p-p65 protein expression by western blotting (Fig. 5). Increased levels of TLR4, p-IκBα and p-p65 were found in NP cells after LPS stimulation (P<0.01 vs. the control) but these levels were lower in miR-140 mimic co-transfected cells for TLR4 (P<0.01 vs. mimic NC), p-IκBα/IκBα (P<0.05 vs. mimic NC) and p-p65/p65 (P<0.05 vs. mimic NC). These results indicated that miR-140 inhibited LPS-induced TLR4 expression and NF-κB activation. Furthermore, without LPS stimulation, overexpression of miR-140 can also inhibit TLR4 expression and inhibit NF-κB activation.