An animal model study on the gene expression profile of meniscal degeneration

Meniscal degeneration is a very common condition in elderly individuals, but the underlying mechanisms of its occurrence are not completely clear. This study examines the molecular mechanisms of meniscal degeneration. The anterior cruciate ligament (ACL) and lateral collateral ligament (LCL) of the right rear limbs of seven Wuzhishan mini-pigs were resected (meniscal degeneration group), and the left rear legs were sham-operated (control group). After 6 months, samples were taken for gene chip analysis, including differentially expressed gene (DEG) analysis, gene ontology (GO) analysis, clustering analysis, and pathway analysis. The selected 12 DEGs were validated by real time reverse transcription-polymerase chain reaction (RT-PCR). The two groups showed specific and highly clustered DEGs. A total of 893 DEGs were found, in which 537 are upregulated, and 356 are downregulated. The GO analysis showed that the significantly affected biological processes include nitric oxide metabolic process, male sex differentiation, and mesenchymal morphogenesis, the significantly affected cellular components include the endoplasmic reticulum membrane, and the significantly affected molecular functions include transition metal ion binding and iron ion binding. The pathway analysis showed that the significantly affected pathways include type II diabetes mellitus, inflammatory mediator regulation of TRP channels, and AMPK signaling pathway. The results of RT-PCR indicate that the microarray data accurately reflects the gene expression patterns. These findings indicate that several molecular mechanisms are involved in the development of meniscal degeneration, thus improving our understanding of meniscal degeneration and provide molecular therapeutic targets in the future.

Preparation procedure of meniscal tissue. The animals were sacrificed at 26 weeks after surgery by acute massive exsanguination. The menisci were isolated using a standardized preparation technique. After the knee joint was opened from the front, the joint capsule and the ligaments were cut along the joint line near the side of the femur to separate the tibia from the femur. The medial menisci were carefully removed in one piece from the tibia. They were washed with phosphate-buffered saline (PBS) and cut into soybean-sized pieces. The preparation was carried out without touching the surfaces of the menisci or the hyaline articular cartilage to prevent contamination and destruction of the meniscal surface and deep structures. The samples were taken from the red-white zone of the meniscal pars intermedia, followed by placing in a cryopreservation tube for immediate freezing and storing in liquid nitrogen. The sampling was performed immediately after the death of the mini-pigs and was completed within 10 min. All equipment was treated with hydrogen peroxide to remove eventual RNAse.
RNA extraction and assessment. After DNase digestion, the total RNA from the meniscal tissue was extracted using an E.Z.N.A Total RNA Kit II (Omega Bio-Tek Inc., Norcross, GA, USA) according to the manufacturer's instructions. The quantity and quality of RNA were measured using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). RNA integrity was assessed by standard denaturing agarose gel electrophoresis. Agilent microarray study. RNA labeling and array hybridization were performed according to the Agilent One-Color Microarray-Based Gene Expression Analysis protocol (Agilent Technologies, Santa Clara, CA, USA). Briefly, the total RNA from each sample was linearly amplified and labeled with Cy3-UTP using the Agilent Quick Amp One-Color labeling kit (Agilent Technologies, Santa Clara, CA, USA). The Labeled cRNAs were purified using the RNeasy Mini Kit (Qiagen, Venlo, The Netherlands) according to the manufacturer's instructions. The concentration and specific activity of the labeled cRNAs (pmol Cy3/μg cRNA) were measured using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). After that, 1 µg of each labeled cRNA was fragmented by adding 11 μl of 10 × blocking agent and 2.2 μl of 25 × fragmentation buffer and heating the mixture at 60 °C for 30 min. Finally, 55 μl of 2 × GE hybridization buffer was added to dilute the labeled cRNA. Then, 100 μl of the hybridization solution was dispensed into the gasket slide and assembled to the gene expression microarray slide. The slides were incubated for 17 h at 65 °C in a hybridization oven (Agilent Technologies, Santa Clara, CA, USA). The RNA was hybridized to the Whole Pig Genome Microarray (Agilent Technologies, Santa Clara, CA, USA). The hybridized arrays were washed, fixed, and scanned using a DNA Microarray Scanner (G2505C, Agilent Technologies, Santa Clara, CA, USA) 12 .
Microarray data analyses. The raw gene expression data were extracted from the array images using the Feature Extraction software (version 11.0.1.1, Agilent Technologies, Santa Clara, CA, USA). Quantile normalization and subsequent data processing were performed with the GeneSpring GX v12.1 software package (Agilent Technologies, Santa Clara, CA, USA). After quantile normalization of the raw data, the genes from at least seven out of the 14 samples had flags detected ("All Targets Value") and were chosen for further data analysis. Statistically significant differentially expressed genes (DEGs) between the two groups were identified through volcano plot filtering, using a threshold of fold-change > 2.0 and P-value < 0.05 based on the Student t-test in the GeneSpring software). DEGs between the two samples were identified through fold change filtering. Unsupervised hierarchical clustering was performed using the R scripts to compare gene expression profiles among the www.nature.com/scientificreports/ samples of the same group. Gene ontology (GO, http://geneo ntolo gy.org) analysis and pathway analysis were performed with the standard enrichment computation method to associate the DEGs with GO categories and pathways. The GO categories comprised of three structured networks (biological processes, cellular components, and molecular function) of defined terms to describe the gene functions. To perform the GO analysis of differential genes, top GO was used to infer the molecular function they are involved in 12 . The KEGG pathway analysis was performed to infer the pathways the DEGs are involved in [13][14][15] . All the microarray data are MIAME compliant, and the raw data are available through the GEO database with the accession number GSE145402.
Unsupervised hierarchical clustering. The unsupervised cluster analysis of DEGs obtained after comparison of the two groups demonstrate the relationship as well as the difference between the groups. The results show that the two groups are well separated in terms of the DEGs (Fig. 2). All samples are accurately clustered into the same cluster of the corresponding category, indicating the reliability of the results of chip analysis, and that the gene expression pattern of the experimental group in comparison to the control group is consistent.
Results of the GO enrichment analysis. The DEGs obtained from the above analysis were classified into biological processes, cellular components, and molecular functions according to the functional relevance of the genes. The top ten biological processes with upregulated gene enrichment scores included the cellular response to hormone stimulus, response to hormones, cellular response to endogenous stimulus, C21-steroid hormone metabolic process, neuropeptide signaling pathway, negative regulation of reactive oxygen species metabolic process, regulation of nitric oxide biosynthetic process, response to endogenous stimulus, nitric oxide biosynthetic process, and nitric oxide metabolic process (Table 3 and Fig. 3A). The top ten biological processes with down-regulated gene enrichment scores included sex determination, muscle fiber development, mesenchymal morphogenesis, negative regulation of the reproductive process, male gonad development, primary male sexual characteristics development, striped muscle cell development, male sex differentiation, muscle cell development, and reproductive structure development (Table 4 and Fig. 3B).
The cellular components involved with upregulated DEGs included transcription factor complex, an intrinsic component of membrane, cell-substrate adherens junction, and cell-substrate junction (Table 5 and Fig. 4A). Cellular components involved with down-regulated DEGs included sarcomere, contractile fiber part, myofibril, contractile fiber, cell body, integral component of the plasma membrane, endoplasmic reticulum membrane, nuclear outer membrane-endoplasmic reticulum membrane network, endoplasmic reticulum subcompartment, intrinsic component of the plasma membrane, plasma membrane part, and endoplasmic reticulum part (Table 6 and Fig. 4B).
The top ten molecular functions with upregulated gene enrichment scores included transcription regulatory region sequence-specific DNA binding, sequence-specific double-stranded DNA binding, virus receptor activity, hijacked molecular function, transition metal ion binding, iron ion binding, double-stranded DNA binding, regulatory region nucleic acid binding, transcription regulatory region DNA binding, and RNA polymerase II regulatory region sequence-specific DNA binding (Table 7 and Fig. 5A). The molecular functions involved with downregulated DEGs included transcription factor activity, RNA polymerase II distal enhancer sequence-specific binding, horizon activity, enhancer-binding, DNA binding transcription factor activity, transcription regulator activity, and neuropeptide binding ( www.nature.com/scientificreports/ Results of the pathway analysis. According to the DEGs, a total of 36 signaling pathways with differential regulation were found. Among these, the top ten pathways included type II diabetes mellitus, taste transduction, prolactin signaling pathway, longevity regulating pathway, ovarian steroidogenesis, neuroactive ligandreceptor interaction, inflammatory mediator regulation of TRP channels, pantothenate and CoA biosynthesis, AMPK signaling pathway, and bile secretion. The down-regulated signaling pathways included thyroid hormone synthesis, cocaine addiction, metabolism of xenobiotics by cytochrome P450, legionellosis, glycerolipid metabo-  www.nature.com/scientificreports/ lism, chemical carcinogenesis, amphetamine addiction, acute myeloid leukemia, adherens junction, Salmonella infection, protein digestion and absorption, and hematopoietic cell lineage (Tables 9 and 10   www.nature.com/scientificreports/  www.nature.com/scientificreports/ inflammatory mediator regulation of TRP channels pathway (ADCY4 and TRPA1), and AMPK signaling pathway (ADIPOQ and SLC2A4). The results are shown in Table 11 and Fig. 7. The results indicate that the microarray data accurately reflects the gene expression patterns.

Discussion
The insufficiency of the ACL can lead to meniscal damage 16 . A previous study of meniscal degeneration models in mini-pigs showed that removing the ACL leads to meniscal degeneration changes as observed by histology 11 . In addition, the real-time quantitative PCR analysis showed that COL1A1 is highly expressed in the degenerative meniscus 11 . In this study, the meniscus degeneration model was constructed by removing the ACL and LCL, based on the Pond-Nuki model 10 . The function of the meniscus is mainly to increase knee stability, absorb shocks, and conduct loads. When other stabilizing structures of the knee joint (such as the ligaments) are impaired, the contribution of the meniscus is increased, leading to stress and wear. This stress is mainly manifested in the horizontal shear plane and is the main cause of meniscal tear after ACL rupture. In addition, after LCL resection, the knee joint shows a loose lateral stable structure, causing varus change of the knee and shifting the rear limb force line inward. More stress is then taken by the medial meniscus, which can reach 60-80% of the BW 17 . The www.nature.com/scientificreports/ www.nature.com/scientificreports/ advantage of the mini-pig model over smaller animals is that the degree of the load is close to that of the human characteristics, but the disadvantages include longer modeling time and higher animal facility requirements 5 .
The results of the DEG cluster analysis showed that all samples could be accurately aggregated into the right grouping, indicating that the experimental intervention induced the DEGs. This is consistent with other studies that showed that ACL resection induces morphological, histological, and genetic changes in the meniscus 11,18 . A study in humans showed that changes in DEGs occur in ligaments after injury and that such changes might be observed in other musculoskeletal tissues 19 .
Meniscal stress is associated with inflammation. One of the most important mediators in inflammation is nitric oxide, wherein it upregulates meniscus matrix catabolism and pro-inflammatory gene expression, resulting in increased meniscus glycosaminoglycan release and meniscal impairment 20,21 . Nitric oxide can reduce autophagy by down-regulating the JNK signaling pathway, thereby affecting meniscal repair and causing meniscal degeneration 22 . Hyaluronic acid, selenium, and interleukin-10 delay or improve meniscus degeneration due to the inhibition of nitric oxide [23][24][25] . This study showed that the upregulated biological processes included regulation of nitric oxide biosynthesis process, nitric oxide biosynthetic process, and nitric oxide metabolic process, indicating that nitric oxide plays an important role in meniscal degeneration.
There is a clear sex difference in knee osteoarthritis and meniscus degeneration [26][27][28] , suggesting the role of sex hormones in meniscal degeneration. Although our study used male Wuzhishan mini-pigs, many male-associated biological processes were down-regulated in the DEG analysis, such as sex determination, negative regulation of the reproductive process, male gonadal development, primary male sexual characteristics development, male sex differentiation, and reproductive structure development. This suggests that the main feature of male characteristics are down-regulated by meniscal degeneration, indicating that androgens might play a role in promoting meniscus damage repair, while its decreased expression level promotes meniscal degeneration. Secondly, sex hormones might indirectly cause meniscal degeneration by affecting muscle strength, walking endurance, and balance ability 29 . Indeed, androgens promote the development of the skeletal muscle system 30,31 . When this dynamic stabilization mechanism decreases, abnormal stress on the knee meniscus increases, causing meniscal degeneration through nitric oxide, as discussed above.
Calcification is common in meniscal degeneration [32][33][34] . Mechanical stress and inflammatory factors promote the overexpression of ANKH and ENPP1 in meniscal fibrochondrocytes, leading to increased intracellular calcium concentration or enhanced calcium signaling that affects the cell function, impairing the meniscal repair capacity [35][36][37] . Phosphocitrate can inhibit the proliferation of meniscal cells and calcification, delaying and reversing meniscal degeneration and osteoarthritis 38 . In this study, upregulated transition metal ion binding was observed, and calcium is a transition metal. Upregulated iron ion binding was also observed, and mice with hereditary hemochromatosis are more prone to knee osteoarthritis and meniscal degeneration due to iron overload 39 .
In the pathway analysis, this study found that the upregulated pathways with the highest enrichment scores were type II diabetes mellitus, inflammatory mediator regulation of TRP channels, and AMPK signaling pathway. Insulin-like growth factors (IGFs) play an extremely important role in normal physiological activities, including musculoskeletal system development and meniscal repair 40 . Impaired insulin secretion leads to declination in the use of glucose by the cells, which in turn stimulates the AMPK signaling pathway, leading to enhanced catabolism and inhibited anabolism. The incidence of meniscal degeneration or osteoarthritis in patients with type II diabetes mellitus is higher than in normal individuals 41 . Damaged joints cause a series of reactions due to pro-inflammatory factors (such as IL-1 and TNF-α) and activate the MAPK pathways, such as increased MMPs secretion, ADAMTS-4 enhanced activity, increased chemokine concentration, and increased nitric oxide synthesis, among others 21,42,43 .
Those results are supported by a study in humans that showed that meniscal degeneration involves the biological processes of immune response, inflammatory response, biomineral formation, and cell proliferation 44 . www.nature.com/scientificreports/ Despite the differences between humans and mini-pigs, the results of the involved pathways and genes are consistent between the two species. There are some limitations to the present study. First, the experiment used the contralateral leg as a control group. The biomechanical characteristics of the contralateral leg also change during walking because of the affected leg, which in turn can cause some changes in gene expression. Therefore, it cannot be considered an ideal control group. Second, meniscal tissue is composed of a variety of cells; therefore, the DEGs are, in fact, the response of the overall tissue to modeling. The role of each cell constituent of the meniscus in degeneration is difficult to define clearly. Further research should be conducted through cell isolation culture or single-cell gene sequencing 44,45 . Third, the results obtained from animal experiments might differ from humans, suggesting only possible mechanisms for human meniscal degeneration.

Conclusion
Wuzhishan mini-pigs can be used as animal models for studying meniscus degeneration through the resection of the ACL and LCL. The present study provides some insight into the molecular mechanisms underlying meniscal degeneration. They might also reveal potential targets for meniscal degeneration treatment and early diagnosis. www.nature.com/scientificreports/ www.nature.com/scientificreports/  www.nature.com/scientificreports/  www.nature.com/scientificreports/