Primary tissues obtained from patients following total knee arthroplasty provide an experimental model for osteoarthritis research with maximal clinical translatability. This protocol describes how to identify, process, and isolate RNA from seven unique knee tissues to support mechanistic investigation in human osteoarthritis.
Osteoarthritis (OA) is a chronic and degenerative joint disease most often affecting the knee. As there is currently no cure, total knee arthroplasty (TKA) is a common surgical intervention. Experiments using primary human OA tissues obtained from TKA provide the capability to investigate disease mechanisms ex vivo. While OA was previously thought to impact mainly the cartilage, it is now known to impact multiple tissues in the joint. This protocol describes patient selection, sample processing, tissue homogenization, RNA extraction, and quality control (based on RNA purity, integrity, and yield) from each of seven unique tissues to support disease mechanism investigation in the knee joint. With informed consent, samples were obtained from patients undergoing TKA for OA. Tissues were dissected, washed, and stored within 4 h of surgery by flash freezing for RNA or formalin fixation for histology. Collected tissues included articular cartilage, subchondral bone, meniscus, infrapatellar fat pad, anterior cruciate ligament, synovium, and vastus medialis oblique muscle. RNA extraction protocols were tested for each tissue type. The most significant modification involved the method of disintegration used for low-cell, high-matrix, hard tissues (considered as cartilage, bone, and meniscus) versus relatively high-cell, low-matrix, soft tissues (considered as fat pad, ligament, synovium, and muscle). It was found that pulverization was appropriate for hard tissues, and homogenization was appropriate for soft tissues. A proclivity for some subjects to yield higher RNA integrity number (RIN) values than other subjects consistently across multiple tissues was observed, suggesting that underlying factors such as disease severity may impact RNA quality. The ability to isolate high-quality RNA from primary human OA tissues provides a physiologically relevant model for sophisticated gene expression experiments, including sequencing, that can lead to clinical insights that are more readily translated to patients.
The knee is the largest synovial joint in the human body, comprising the tibiofemoral joint between the tibia and the femur and the patellofemoral joint between the patella and the femur1. The bones in the knee are lined with articular cartilage and supported by various connective tissues, including menisci, fat, ligaments, and muscle, and a synovial membrane encapsulates the whole joint to create a synovial fluid-filled cavity1,2,3 (Figure 1). A healthy knee functions as a mobile hinge joint that allows frictionless motion in the frontal plane1,3. Under pathological conditions, movement can become restricted and painful. The most common degenerative knee joint disease is osteoarthritis (OA)4. A variety of risk factors are known to predispose to OA development, including older age, obesity, female sex, joint trauma, and genetics, among others5,6. There are currently an estimated 14 million people in the USA with symptomatic knee OA, with the prevalence increasing due to rising population age and rates of obesity7,8. Initially considered to be a disease of the cartilage, OA is now understood as a disease of the whole joint9. Commonly observed pathological changes in OA include articular cartilage erosion, osteophyte formation, subchondral bone thickening, and inflammation of the synovium9,10. Since there is no known cure for OA, treatments primarily focus on symptom (e.g., pain) management11,12, and once OA has progressed to end-stage, joint replacement surgery is often indicated13.
Joint replacement surgeries can either be partial or total knee replacements, with total knee arthroplasty (TKA) including replacing the entire tibiofemoral articulation and the patellofemoral joint. As of 2020, approximately 1 million TKAs are performed in the USA each year14. During TKA, an orthopedic surgeon resects the upper portion of the tibial plateau and the lower femoral condyles (Figure 2A, 2B) to be fitted with prosthetic implants. Sometimes misinterpreted by patients, in a TKA, only 8-10 mm is resected from the end of each bone, which is subsequently capped or resurfaced, with metal. An interposed polyethylene liner forms the bearing surface (i.e., padding) between the two metal implants. In addition, several soft tissue components of the joint are fully or partially excised to achieve proper joint balance. Among these tissues are the medial and lateral menisci (Figure 2C), infrapatellar fat pad (Figure 2D), anterior cruciate ligament (ACL; Figure 2E), synovium (Figure 2F), and vastus medialis oblique muscle (VMO; Figure 2G)15. Though TKAs are generally successful for OA treatment, around 20% of patients report reoccurrence of pain post-surgery16. Along with the high cost and relative invasiveness of the procedure, these limitations point to the need for further research to identify alternative treatments to mitigate the progression of OA.
To explore disease mechanisms in OA that may present new avenues for therapeutic intervention, experimental systems, including cells, tissue explants, and animal models can be used. Cells are typically cultured in monolayer and are derived from primary human or animal tissues (e.g., chondrocytes isolated from cartilage) or immortalized cells (e.g., ATDC517 and CHON-00118). While cells can be useful for manipulating experimental variables in a controlled culture environment, they do not capture conditions of the natural joint which are known to impact cell phenotypes19. To better recapitulate the complex cascade of chemical, mechanical, and cell-to-cell communication underlying OA, an alternative is found in primary human or animal tissue samples, whether used fresh or cultured ex vivo as explants, to preserve tissue structure and the cell microenvironment20. In order to study the joint in vivo, small (e.g., mouse21) and large (e.g., horse22) animal models for OA (e.g., through surgical induction, genetic alteration, or aging) are also useful. However, translation from these models to human disease can be limited by anatomical, physiological, and metabolic differences, among others23. Considering the advantages and disadvantages of experimental systems, the key strengths of being species-specific and maintaining the extracellular niche offered by the primary human OA tissues maximize the translational potential of research findings.
Primary human OA tissues can be readily obtained following TKA, making the high frequency of TKAs a valuable resource for research. Among potential experimental applications are gene expression and histological analyses. To realize the potential of primary human OA tissues for these research approaches and others, outlined are the following key considerations. First, the use of patient specimens is subject to ethical regulation, and protocols must meet Institutional Review Board (IRB) approvals24. Second, the inherent heterogeneity of human primary diseased tissues and the influence of variables such as age and sex, among others, create the need for careful patient selection (i.e., application of eligibility criteria) and data interpretation. Third, the unique biological properties of different tissues in the joint (e.g., low cellularity of cartilage and meniscus25) can present challenges during experiments (e.g., isolating high quality and quantity of RNA). This report addresses these considerations and presents a protocol for patient selection, sample processing, tissue homogenization, RNA extraction, and quality control (i.e., assessment of RNA purity and integrity; Figure 3) to encourage the use of primary human OA tissues in the research community.
This study protocol was approved and followed institutional guidelines set by the Henry Ford Health System Institutional Review Board (IRB #13995).
1. Patient selection
2. Sample processing (for RNA)
NOTE: Perform all tissue processing in a class II biosafety cabinet and follow sterile techniques. Always wear appropriate PPE (nitrile gloves, lab coat, safety goggles) when processing human samples. Several bone fragments are produced during TKA, a large amount of bone/cartilage will potentially be available for dissection. Due to disease progression, articular cartilage degeneration may be more severe on some bone portions, which can be factored into experimental design. Only tissues that mandate electrocautery for resection have thermal edge damage, and a concerted surgical effort is made to procure most tissues with a scalpel to minimize damage. Resected tissues must be kept hydrated at all times with sterile PBS.
3. Sample processing (for histology)
CAUTION: Formalin is a hazardous chemical, only use in a chemical fume hood.
4. Tissue homogenization
CAUTION: The protocol uses the hazardous chemical phenol. Work with phenol must be performed in a chemical fume hood.
NOTE: Thoroughly clean all equipment and surfaces to be used with 70% ethanol (soak for a minimum of 10 min), followed by RNase decontaminant (soak for a minimum of 10 min), rinse with DEPC-treated water, wipe with a clean, lint-free paper towel, and then respray or soak with 70% ethanol.
5. RNA extraction from tissues
CAUTION: This protocol uses hazardous chemicals such as phenol, chloroform, and isopropanol. Perform all the work in a chemical fume hood.
NOTE: Equipment and reagents are reserved for RNA work only and must be of proper chemical grade for molecular applications (i.e., sterile, nuclease-free). This protocol succeeds both Parts 4.1 and 4.2 tissue homogenization protocols. Thoroughly clean all equipment and surfaces to be used with 70% ethanol (soak for a minimum of 10 min), followed by RNase decontaminant (soak for a minimum of 10 min). Rinse with DEPC-treated water, wipe the residual liquid with a clean, lint-free paper towel, and then respray or soak with 70% ethanol.
6. Quality control
Seven unique human knee joint tissues are available for collection from patients undergoing TKA for OA (Figure 1). In this protocol, each of these tissues were identified and processed within 4 h of surgical removal (Figure 2). Following the steps outlined in Figure 3, portions of each tissue were formalin-fixed for histological assessment (Figure 4), while other portions were flash-frozen for RNA isolation. Separating hard tissues from soft tissues by method of disintegration (pulverization versus homogenization, respectively), RNA of high integrity and purity was extracted from each tissue type, with representative results shown in Table 1 (High quality columns). Notably, some subjects yielded lower quality RNA across multiple tissues (Table 1, Low quality columns), suggesting that despite an optimized method, external factors (e.g., disease severity) may be impacting RNA quality across tissue types.
Figure 1: Schematic of the human knee joint showing a lateral cross-section. Each of the seven labeled tissues are collected during TKA and used for research purposes as described in this protocol. VMO = vastus medialis oblique muscle. Image accessed and modified from OpenStax College under a creative commons license26. Please click here to view a larger version of this figure.
Figure 2: Representative gross images for each of the seven tissues obtained from patients undergoing TKA. (A) Anterior femoral bone cut with arrow pointing to articular cartilage to be collected. Cartilage is identified by a whitish layer found on the surface of the bone. (B) Anterior femoral bone cut with arrow pointing to subchondral bone to be collected. (C) Meniscus. Avoid collecting burnt sections caused by electrocauterization during TKA. (D) Infrapatellar fat pad (yellow in color). (E) Anterior cruciate ligament (white, fibrous, spongy tissue). (F) Synovium. One side will appear light colored and fibrous, often containing fat tissue, while the opposite side will appear pinkish and less fibrous. The pink side of the membrane contains the synovial lining. (G) Vastus medialis oblique muscle (red). This may often be the smallest tissue portion and may contain some fat tissue. Scale bar = 2 cm. Please click here to view a larger version of this figure.
Figure 3: Overview of steps taken to collect primary human OA tissues for histology and RNA. This protocol describes patient selection, sample processing for histology and RNA, tissue homogenization for hard tissues and soft tissues, RNA extraction, and quality control for seven primary human knee OA tissues. Please click here to view a larger version of this figure.
Figure 4: Histological sections for each of the seven tissues obtained from patients undergoing TKA. Hematoxylin and eosin-stained sections are shown at 6x magnification in panels A–F with inset magnified to 40x in panels A'–F' and A". (A, A') Articular cartilage; (A, A") subchondral bone; (B, B') meniscus; (C, C') infrapatellar fat pad; (D, D') anterior cruciate ligament; (E, E') synovium; (F, F') vastus medialis oblique muscle. Scale bars = 400 µm for A-F and 50 µm for A'-F' and A". Please click here to view a larger version of this figure.
Tissue | N | RIN | [RNA] (ng/µL) in 25 µL | A260:A280 | A260:A230 | ||||||||
High quality | Low quality | High quality | Low quality | High quality | Low quality | High quality | Low quality | High quality | Low quality | ||||
"Hard tissues" | |||||||||||||
Articular Cartilage | 10 | 4 | 7.0 ± 0.8 | 1.3 ± 1.2 | 135 (36-243) |
46 (26-78) |
1.80 ± 0.09 | 1.47 ± 0.34 | 1.40 ± 0.36 | 0.45 ± 0.26 | |||
Subchondral Bone | 10 | 4 | 7.8 ± 0.6 | 3.6 ± 1.1 | 514 (181-1586) |
342 (122-769) |
1.96 ± 0.05 | 1.90 ± 0.17 | 1.72 ± 0.27 | 1.12 ± 0.62 | |||
Meniscus | 10 | 4 | 7.5 ± 0.6 | 2.4 ± 0.3 | 242 (38-1629) |
95 (60-318) |
1.91 ± 0.05 | 1.71 ± 0.15 | 1.41 ± 0.47 | 0.77 ± 0.59 | |||
"Soft tissues" | |||||||||||||
Infrapatellar Fat Pad | 10 | 3 | 8.5 ± 0.6 | 6.1 ± 1.4 | 1905 (668-5100) |
1151 (381-2306) |
1.99 ± 0.01 | 2.02 ± 0.03 | 2.09 ± 0.17 | 1.47 ± 0.71 | |||
Anterior Cruciate Ligament | 8 | 3 | 7.4 ± 0.4 | 5.2 ± 1.9 | 1836 (613-8456) |
727 (97-1479) |
1.97 ± 0.03 | 1.91 ± 0.18 | 1.88 ± 0.20 | 1.35 ± 0.70 | |||
Synovium | 9 | 3 | 8.5 ± 0.6 | 6.2 ± 3.1 | 2239 (401-4100) |
1897 (902-3366) |
1.99 ± 0.04 | 2.00 ± 0.03 | 2.05 ± 0.11 | 1.91 ± 0.20 | |||
Vastus Medialis Oblique | 9 | 3 | 8.5 ± 0.6 | 8.4 ± 0.7 | 1002 (377-1715) |
1097 (308-2138) |
1.96 ± 0.03 | 1.99 ± 0.03 | 1.82 ± 0.11 | 1.62 ± 0.27 |
Table 1. Quality and quantity of RNA isolated from OA tissues collected from TKA patients. RIN = RNA Integrity Number. Data presented as mean ± standard deviation for RIN, A260:A280 and A260:A230 ratios, and mean (range) for RNA concentration values. High quality samples consist of patients from which all tissue types yielded RNA with RIN > 6 (n = 8-10). Low quality samples consist of patients from which multiple tissue types yielded RNA with RIN < 6 (n = 3-4).
The protocol presented has proved successful for collecting seven primary human OA tissues for RNA extraction (Table 1) and histological processing (Figure 4). Prior to collecting patient samples, it is necessary to establish an IRB-approved protocol, ideally in collaboration with a surgeon or surgical team. Applying a standardized protocol for specimen collection (e.g., resection from consistent in situ locations) is essential for maximizing experimental reproducibility. Tissue samples should be transported to the lab in sterile containers and processed within 4 h of the surgery to avoid degradation. During tissue dissection and processing, all tissues are kept hydrated in sterile PBS and are rinsed in fresh, sterile PBS to remove potential surface contaminants such as biofluids and other unwanted debris prior to being flash-frozen for RNA extraction or formalin-fixed for histology. A useful application of histological analysis is confirmation of the tissue types and the disease severity since these can be distinguished by cell number, distribution, and morphology, among other factors observable by standard stains such as hematoxylin and eosin (Figure 4).
Primary human OA tissues can present challenges for extracting RNA of sufficient quantity and quality, as defined by purity and integrity27. RNA quantity is a function of the overall cellularity of the tissue, and in the knee joint, there are low-cell, high-matrix tissues such as the cartilage, bone, and meniscus, and relatively high-cell, low-matrix tissues such as the fat pad, ACL, synovium, and VMO. For example, both articular hyaline cartilage and meniscal fibrocartilage28 are characterized by low cellularity, with the extracellular matrix containing varying quantities of collagens, proteoglycans, and other glycoproteins28,29. Having fewer cells results in less RNA per volume of tissue (reducing quantity) and having more protein results in co-purification with RNA (reducing purity)25,30. RNA purity can be determined by spectrophotometry where A260:A280 and A260:A230 values of <1.5 reflect the presence of organic contaminants (e.g., protein) and values of ~2.0 reflect pure RNA31. RNA integrity reflects the level of degradation, whether caused by experimental conditions (i.e., shearing forces) or by enzymatic digestion (e.g., nucleases), and is often determined by electrophoretic analysis. An RNA Integrity Number (RIN) of 1 reflects degraded RNA and a RIN of 10 reflects intact RNA31,32. For RNA-sequencing, a minimum RIN of 7 is often recommended33,34,35. Data presented in Table 1 reveals that these A260:A280, A260:A230, and RIN thresholds were met across all tissues from the patient samples in the High quality RNA group compared to patient samples in the Low quality RNA group, except for some A260:A230 values, which may reflect protein contamination of RNA in the low-cell, high-matrix tissues. While there are many factors that could be contributing to the quality of RNA isolated from a given patient sample, among them may be the level of disease severity. The diseased nature of OA tissues suggests degradative processes are occurring through increased levels of enzymes that can digest tissues, but also RNA, thereby reducing quality.
This protocol for RNA extraction aims to maximize RNA quantity and quality from primary human OA tissues. The most critical step pertained to whether the tissues were disintegrated by being pulverized or homogenized, and this was found to correlate with the tissue cellularity and matrix composition. Initially, all seven tissues were subjected to the same protocol where tissues were first pulverized by mortar and pestle using liquid nitrogen, then transferred to acid-guanidinium-phenol solution, and further homogenized using a hand-held tissue homogenizer. This method produced favorable RNA yield, purity, and integrity for the fat pad, ACL, synovium, and VMO (collectively soft tissues; also, relatively high-cell, low-matrix), but unfavorable results for cartilage, bone, and meniscus (collectively hard tissues; also, low-cell, high-matrix). Based on these observations, the seven tissues were divided into two groups for further protocol refinement. It was observed that the additional homogenization had minimal effect on further disintegrating the hard tissues after they had been pulverized into a fine powder. Conversely, dissociation of the soft tissues was successfully achieved with homogenization alone and did not require pulverization. Therefore, the homogenization of the hard tissues and pulverization of the soft tissues was eliminated. This was beneficial for minimizing shearing forces, processing time, and temperature fluctuation, all of which can improve RNA integrity. Two rounds of phenol/chloroform phase separation for all seven tissues were performed, as this has been reported to improve RNA purity without reducing yield31.
A potential limitation of this protocol is the batch effect that may arise from separating the tissues into two groups if the experimental design requires comparison among all tissues. The use of pulverization versus homogenization methods may alter technical (e.g., processing time) and environmental (e.g., temperature fluctuations) conditions that can introduce variability36. A second limitation is the potential inconsistency in identifying, dissecting, and orienting (for histological sectioning) the tissues within and across subjects. A third limitation is our inability to confirm potential correlations between patient disease severity and RNA quality in the current report. A fourth limitation is the lack of availability of healthy control tissues for comparison. Though control samples may be available from cadavers, these are less readily available than OA tissues from TKA. An experimental strategy to circumvent this is to use each subject as their own control, whether making comparisons across tissues or within tissues comparing treatment to control or lesioned to preserved areas. Finally, using tissue explants for RNA extraction does not permit gene expression analysis of the individual cell types that comprise the tissues (e.g., synovial fibroblasts versus synovial macrophages37).
Despite the noted limitations, primary human OA tissues are a valuable resource for research, offering advantages over other experimental systems for OA, including preserving the cell niche23. However, primary human OA tissues may be underutilized in research due to logistic or technical challenges. This protocol describes patient selection, sample processing, tissue homogenization, RNA extraction, and quality control to support the use of the samples obtained from TKA. Following sample processing, several experimental approaches can be pursued, including gene expression and histology, among others. Most relevant to the rapidly evolving omics field is the ability to isolate sufficient quantities of high-quality RNA for applications such as RNA-sequencing38,39. Molecular profiles can be compared within and across tissues from subjects with specific disease phenotypes (e.g., based on age, sex, and other OA risk factors). Insights gained may inform new therapeutic avenues that can be more readily translated back to the OA patient population.
The authors have nothing to disclose.
The authors thank the study participants who made this research possible and dedicate this report to new scientists in the osteoarthritis field.
1.5 mL microcentrifuge tubes | Eppendorf | 05 402 | Sterile, nuclease-free. Reserved for RNA work only. |
10% Formalin | Cardinal Health | C4320-101 | Store in chemical cabinet when not in use. |
100% Chloroform (Molecular Biology Grade) | Fisher Scientific | ICN19400290 | Sterile, nuclease-free. Reserved for RNA work only, store in chemical cabinet when not in use. |
100% Ethanol (Molecular Biology Grade) | Fisher Scientific | BP2818500 | Sterile, nuclease-free. Reserved for RNA work only, when diluting use DEPC/nuclease-free water. |
100% Isopropanol (Molecular Biology Grade) | Fisher Scientific | AC327272500 | Sterile, nuclease-free. Reserved for RNA work only, store in chemical cabinet when not in use. |
100% Reagent Alcohol | Cardinal Health | C4305 | Diluted to 70% with dH2O for cleaning purposes. |
15 cm sterile culture dishes | Thermo Scientific | 12-556-003 | Sterile, nuclease-free. |
15 mL polypropylene (Falcon) tubes | Fisher Scientific | 14 959 53A | Sterile, nuclease-free. |
2 mL cryovials (externally threaded) | Fisher Scientific | 10 500 26 | Sterile, nuclease-free. |
5 mL round-bottom tubes | Corning | 352052 | Sterile, nuclease-free. Reserved for RNA work only. |
50 mL polypropylene (Falcon) tubes | Fisher Scientific | 12 565 271 | Sterile, nuclease-free. |
Bioanalyzer | Agilent | G2939BA | For RNA integrity measurement. |
Biosafety Cabinet | General lab equipment | ||
Bone Cutters | Fisher Scientific | 08 990 | Sterilized with 70% EtOH. |
Chemical Fume Hood | General lab equipment | ||
Disposable Scalpels (No.10) | Thermo Scientific | 3120032 | Sterile, nuclease-free. |
EDTA | Life Technologies | 15-576-028 | 10% solution with dH2O. |
Forceps | Any vendor | Sterilized with 70% EtOH. | |
Glycoblue Coprecipitant | Fisher Scientific | AM9516 | Reserved for RNA work only, store at -20 °C. |
Kimwipes | Fisher Scientific | 06-666 | |
Liquid Nitrogen | Any vendor | ||
Liquid Nitrogen Dewar | General lab equipment | ||
Mortar and Pestle | Any vendor | Reserved for RNA work only, sterilzed per protocol. | |
Nanodrop Spectrophotometer | Thermo Scientific | ND-2000 | For RNA purity and yield measurements. |
Nuclease-free/DEPC-treated water | Fisher Scientific | Sterile, nuclease-free. Reserved for RNA work only. | |
PBS (Sterile) | Gibco | 20 012 050 | Sterile, nuclease-free. |
Pipettes (2 µL, 20 µL, 200 µL, 1000 µL) & tips | Any vendor | Sterile, nuclease-free. | |
Plasma/Serum Advanced miRNA kit | Qiagen | 217204 | |
Refrigerated Centrifuge 5810R | Eppendorf | 22625101 | |
RNAlater | Thermo Scientific | 50 197 8158 | Sterile, nuclease-free. |
RNAse Away/RNAseZap | Fisher Scientific | 7002 |
|
Spatula (semimicro size) | Any vendor | Reserved for RNA work only. | |
Tissue homogenizer | Pro Scientific | 01-01200 | Reserved for RNA work only, sterilzed per protocol. |
TRIzol Reagent | Fisher Scientific | 15 596 026 | Sterile, nuclease-free. Reserved for RNA work only. |