Here, we describe a method that allows for the decalcification of freshly obtained bone tissues and the preservation of high-quality RNA. A method is also illustrated for sectioning Formalin Fixed Paraffin Embedded (FFPE) samples of non-demineralized bones to obtain good quality results if fresh tissues are not available or cannot be collected.
Understanding the relationship between the cells and their location within each tissue is critical to uncover the biological processes associated with normal development and disease pathology. Spatial transcriptomics is a powerful method that enables the analysis of the whole transcriptome within tissue samples, thus providing information about the cellular gene expression and the histological context in which the cells reside. While this method has been extensively utilized for many soft tissues, its application for the analyses of hard tissues such as bone has been challenging. The major challenge resides in the inability to preserve good quality RNA and tissue morphology while processing the hard tissue samples for sectioning. Therefore, a method is described here to process freshly obtained bone tissue samples to effectively generate spatial transcriptomics data. The method allows for the decalcification of the samples, granting successful tissue sections with preserved morphological details while avoiding RNA degradation. In addition, detailed guidelines are provided for samples that were previously paraffin-embedded, without demineralization, such as samples collected from tissue banks. Using these guidelines, high-quality spatial transcriptomics data generated from tissue bank samples of primary tumor and lung metastasis of bone osteosarcoma are shown.
Bone is a specialized connective tissue comprised mainly of fibers of collagen type 1 and inorganic salts1. As a result, bone is incredibly strong and stiff while being, at the same time, light and trauma-resistant. The great strength of bone derives from its mineral content. In fact, for any given increase in the percentage of mineral content, stiffness increases by five-fold2. Consequently, investigators face significant problems when they analyze, by means of histological sectioning, the biology of a bone specimen.
Undecalcified bone histology is feasible and sometimes required, depending on the type of investigation (e.g., to study the micro-architecture of bone); it is, however, very challenging, especially if the specimens are large. In these cases, tissue processing for histological purposes requires several modifications of the standard protocols and techniques3. In general, to perform common histological evaluations, bone tissues are decalcified right after fixation, a process that may require a few days to several weeks, depending on the size of the tissue and the decalcifying agent utilized4. Decalcified sections are often used for the examination of bone marrow, the diagnosis of tumors, etc. There are three main types of decalcifying agents: strong acids (e.g., nitric acid, hydrochloric acid), weak acids (e.g., formic acid), and chelating agents (e.g., ethylenediaminetetracetic acid or EDTA)5. Strong acids can decalcify bone very rapidly, but they can damage the tissues; weak acids are very common and suitable for diagnostic procedures; chelating agents are by far the most used and appropriate for research application since, in this case, the demineralization process is slow and gentle, allowing for retention of high-quality morphology and preservation of gene and protein information, as required by many procedures (e.g., in situ hybridization, immunostaining). However, when the whole transcriptome needs to be preserved, such as for gene expression analyses, even a slow and gentle demineralization may be detrimental. Therefore, better approaches and methods are needed when the morphological analysis of the tissues needs to be paired with gene expression analyses of the cells.
Thanks to recent improvements in single-cell RNA sequencing (scRNA-seq) and spatial transcriptomics, it is now possible to study the gene expression of a tissue specimen even when formalin fixation paraffin embedding (FFPE) was used to store the tissue samples6,7,8. This opportunity has unlocked access to a larger number of samples, such as those stored in tissue banks worldwide. If scRNA-seq is to be employed, RNA integrity is the most important requirement; however, in the case of spatial transcriptomics of FFPE samples, both high-quality tissue sections and high-quality RNA are necessary to visualize the gene expression within the histological context of each tissue section. While this has been easily achieved with soft tissues, the same cannot be said for hard tissues like bone. In fact, to the best of our knowledge, no study using spatial transcriptomics has ever been performed on FFPE bone samples. This is because of the lack of protocols that can effectively process FFPE bone tissues while preserving their RNA content. Here, a method to process and decalcify freshly obtained bone tissue samples while avoiding RNA degradation is provided first. Then, recognizing the need for transcriptomics analysis of the FFPE samples collected in tissue banks worldwide, developed guidelines to properly handle FFPE samples of non-demineralized bones are also presented.
All animal procedures described below were approved in compliance with the Guide for the Care and Use of Laboratory Animals at the University of Pittsburgh School of Dental Medicine.
1. Method to prepare FFPE blocks of bone tissue samples that require demineralization
2. Guidelines for sectioning un-demineralized FFPE samples
NOTE: The following guidelines provide valuable tips and instructions useful to greatly improve the quality of the tissue sections while avoiding RNA degradation especially in cases of small undecalcified bone specimens. These guidelines may also be applied to decalcified bone specimens when available. For larger bone tissue samples, demineralization should be performed to obtain better quality sections; in such cases, the method reported above should be followed, and parameters (timing, volumes of solutions, etc.) should be adjusted accordingly. The following guidelines are suitable either when FFPE bone tissues have already been processed (e.g., FFPE block collected from tissue banks or other labs) or when the collected sections will be utilized to perform any type of analysis that requires good quality RNA, high-quality tissue sections, or both.
The method presented here describes how to process freshly isolated bones to obtain demineralized FFPE samples that can be easily sectioned with a microtome while preserving the RNA integrity (Figure 1). The method has been successfully employed on murine femurs but can be followed for other bone tissue samples of similar dimensions, or it can be adapted for larger bone specimens (e.g., human samples) by increasing all the parameters (timing, volumes of solutions, etc.).
Figure 1: Schematic representation of the protocol. Schematic diagram of the method to obtain FFPE blocks of decalcified bone tissues with preserved RNA integrity (points 1 and 2). Schematic diagram of the guidelines to section FFPE blocks (point 3). Please click here to view a larger version of this figure.
To validate the correct timing of decalcification, a time course was performed in which undecalcified femurs, femurs decalcified for 3 h (for which EDTA was changed every 30 min for a total of 6 times), and femurs decalcified for 24 h (for which EDTA was changed every 30 min for a total of 10 times and then left in EDTA overnight) were compared (Figure 2). The obtained FFPE blocks were then sectioned according to the above guidelines, and the histological quality and RNA integrity of the obtained sections were verified. To do so, the structural integrity and the morphology of tissue sections were evaluated by means of hematoxylin and eosin (H&E) staining followed by microscopic inspection (Figure 2A,C,E), while the RNA quality was assessed by evaluating the RNA fragment distribution value with size higher than 200 nucleotides (200 nt)(known as DV200 score)4 (Figure 2B,D,F). H&E images showed that undecalcified and 3 h decalcified femur sections presented with several fractures, holes, and damages (Figure 2A,C,), while sections of femurs decalcified for 24 h displayed good histological quality (Figure 2E). All the samples presented DV200 scores higher than 50%, which is considered the minimum value to perform scRNA-seq or spatial transcriptomic analyses4. Longer incubation times with daily changes of EDTA, at 4 °C, with milder agitation in smaller containers, were also tested and are not recommended since, using these conditions, the RNA integrity of the samples declines dramatically (Figure 2H). Therefore, the time of incubation was decreased to 24 h while the frequency, agitation, and volumes of decalcification were increased to boost decalcification.
Figure 2. Sections and RNA integrity quality control (QC) of 8-week-old mouse femurs after decalcification. Femurs from 8-week mice were freshly dissected, fixed, and decalcified with 20% EDTA pH 8.0. at different time points, embedded in paraffin using the described method, and sectioned following the reported guidelines. Histological QC of obtained sections was then performed by means of H&E staining, while RNA integrity was assessed by evaluating the RNA Fragment Distribution Value with a size larger than 200 nucleotides (200 nt)(DV200). (A) H&E staining showing 8 week-old mouse femur sections after no decalcification. The right image shows a higher magnification of the boxed area. (B) RNA QC of 8 week-old mouse femur sections after no decalcification. (C) H&E staining showing 8 week-old mouse femur sections after 3 h of decalcification. The right image shows a higher magnification of the boxed area. (D) RNA QC of 8 week-old mouse femur sections after 3 h of decalcification. (E) H&E staining showing 8 week-old mouse femur sections after 24 h of decalcification. The right image shows a higher magnification of the boxed area. (F) RNA QC of 8 week-old mouse femur sections after 24 h of decalcification. (G) H&E staining showing 8 weeks-old mouse femur sections after 72 h of decalcification. The right image shows a higher magnification of the boxed area. (H) RNA QC of 8 week-old mouse femur sections after 72 h of decalcification. Abbreviations: H&E = hematoxylin and eosin; DV200 (%) = % of fragment distribution value > 200 nt; nt = nucleotides. Please click here to view a larger version of this figure.
To test the efficacy of the reported guidelines to handle non-demineralized FFPE samples, FFPE blocks of non-demineralized human primary osteosarcoma and lung osteosarcoma metastases obtained from our Musculoskeletal Oncology Tumor Registry and Tissue Bank (MOTOR) were collected, and spatial transcriptomic analysis was performed (Figure 3). Before proceeding with spatial transcriptomic analysis, FFPE blocks were re-embedded in paraffin to obtain a smooth starting surface. Then, the RNA and histological quality of the sections were evaluated (data not shown). Hydration steps were decisive since primary osteosarcoma specimens could not be sectioned without proper hydration. By contrast, lung metastases did not require any hydration steps. After sectioning, tissues were placed on a spatial transcriptomic slide (Figure 3A), stained with H&E (Figure 3D), and spatial transcriptomic analysis was performed according to manufacturer specifications (see Table of Materials). Obtained cDNA libraries were sequenced, and processed data were visualized with Space Ranger for quality control5 (Figure 3B). Space Ranger output showed very high scores (nearly 100%) for valid barcodes, valid unique molecular identifiers (UMIs), Q30 bases, and median of detected genes per spot (between 1700 and 5000), demonstrating robustness and solidity of the obtained data (Figure 3B). By means of unbiased graph-based cluster analysis, 12 major clusters were identified, including clusters of osteogenic, immune, epithelial, and endothelial cells, as well as adipocytes (Figure 3E). Of note, the boundaries of the clusters overlapped with the edges of the histological regions identified by the pathologist (Figure 3D). Additional sections of the same samples were also stained with Goldner's Trichrome to visualize mineralized areas (Figure 3C).
Figure 3. Spatial transcriptomic analysis of two pairs of undecalcified primary osteosarcoma (OS) and lung metastases. FFPE blocks of two matching pairs of non-demineralized human primary osteosarcoma and lung metastasis were collected from our MOTOR tissue bank. Samples were sectioned using the reported guidelines and were placed onto a Visium Spatial Gene Expression slide for FFPE samples to perform spatial transcriptomic analysis. (A) Visium Spatial Gene Expression slide for FFPE samples with the attached sections. (B) Space Ranger output showing common parameters used to assess the quality of the obtained data for all samples. (C) Goldner's Trichrome stain showing localization of mineralized bone tissues and osteoid. (D) H&E stain showing the pathologist's annotation. (E) Cluster analysis showing localization of the tissue residing cell populations. Abbreviations: FFPE = formalin-fixed and paraffin-embedded; MOCs = malignant osteogenic cells. Please click here to view a larger version of this figure.
Here, a detailed method is provided to prepare FFPE blocks of decalcified bones and preserve RNA integrity for sequencing (i.e., next-generation sequencing (NGS)) or for other RNA-related techniques (i.e., in situ hybridization, quantitative reverse transcription polymerase chain reaction (qRT-PCR), etc.).
The method utilizes EDTA to decalcify bone tissue samples; the EDTA incubation allows for slow but fine demineralization of the samples, thus preserving the histological features of the tissue at their best. Usually, bone decalcification is performed using 10% EDTA pH 7.0 for at least 2 weeks by refreshing the solution every other day4. Long times of decalcification, however, significantly contribute to RNA degradation6,7,8. To overcome this problem, EDTA concentration was increased to 20%, the pH was raised to 8.0, and the frequency of solution replacement was increased to every 30 min. These conditions, in combination with the large amount of solution to which the tissue is exposed, the vigorous agitation, and the high container/tissue volume ratio, maximized decalcification and reduced treatment times, allowing for a gentle but rapid demineralization. Of note, it is important to mention that when bone tissue samples are to be used for transcriptome sequencing, it may not be necessary to completely decalcify them. In this case, the optimal proportion between enough demineralization and good RNA preservation is to be found. Another important thing to mention is that no matter how samples are treated if they have been fixed, their RNA will be fragmented. The method reported here allows for retainment of the RNA fragments with a size higher than 200 nucleotides (%DV200) (Figure 2F), which indicates high RNA quality9,11.
This method presents certain limitations. First, as mice were the only source of bone tissue samples, the method has been validated only on murine bones, which are relatively small. Moreover, it is not known whether this method can be used for mice older or younger than 8 weeks since it was tested on 8-week-old mice only. We chose 8-week-old mice because they are skeletally mature. As aging modifies the mineral content of bones, changes to this approach may have to be implemented.
Additionally, guidelines are reported to help improve the quality of the tissue sections when FFPE samples of small undecalcified bone specimens are utilized, like, for example, when they are directly collected from tissue banks. In these situations, nothing can be done if RNA is degraded, but section quality can be improved. On the other hand, for samples with good RNA content, such as those shown here, spatial transcriptomic analysis is possible if the provided guidelines are followed. Of note, in primary osteosarcoma specimens, bone structure can be altered by the presence of the tumor mass, making the actual samples weaker and brittle, which may facilitate the sectioning. Therefore, Goldner's Trichrome staining was utilized to assess whether this was the case since mineralized and non-mineralized areas can be identified with this staining technique (Figure 3C).
In conclusion, the reported method represents an easily applicable method to decalcify bone tissue samples and obtain FFPE sections that may effectively be used for NGS or other techniques that require RNA integrity. In addition, guidelines are provided for sectioning samples that may not have been demineralized prior to paraffin embedding, like samples obtained from tissue banks. Given the rapid development of spatial transcriptomics analyses, the proposed method and guidelines may be very helpful in bone research in general and in bone cancer research as well.
The authors have nothing to disclose.
This work was supported by funds from the Pittsburgh Cure Sarcoma (PCS) and the Osteosarcoma Institute (OSI).
Advanced orbital shaker | VWR | 76683-470 | Use to keep tissues under agitation during incubation as reported in the method instructions. |
Camel Hair Brushes | Ted Pella | 11859 | Use to handle FFPE sections as reported in the guidelines. |
Dual Index Kit TS Set A 96 rxns | 10X Genomics | PN-1000251 | Use to perform spatial transcriptomics. |
Ethanol 200 Proof | Decon Labs Inc | 2701 | Use to perform tissue dehydration as reported in the method instructions. |
Ethylenediaminetetraacetic Acid, Disodium Salt Dihydrate (EDTA) | Thermo Fisher Scientific | S312-500 | Use to prepare EDTA 20% pH 8.0. |
Fisherbrand Curved Medium Point General Purpose Forceps | Fisher Scientific | 16-100-110 | Use to handle FFPE sections as reported in the guidelines. |
Fisherbrand Fine Precision Probe | Fisher Scientific | 12-000-153 | Use to handle FFPE sections as reported in the guidelines. |
Fisherbrand Superfrost Plus Microscope Slides | Fisher Scientific | 12-550-15 | Use to attach sectioned scrolls as reported in the guidelines. |
High profile diamond microtome blades | CL Sturkey | D554DD | Use to section FFPE blocks as reported in the guidelines. |
Novaseq 150PE | Novogene | N/A | Sequencer. |
Paraformaldehyde (PFA) 32% Aqueous Solution EM Grade | Electron Microscopy Sciences | 15714-S | Dilute to final concentration of 4% with 1x PBS to perform tissue fixation. |
Phosphate buffered saline (PBS) | Thermo Fisher Scientific | 10010-049 | Ready to use. Use to dilute PFA and to perform washes as reported in the method instructions. |
Premiere Tissue Floating Bath | Fisher Scientific | A84600061 | Use to remove wrinkles from FFPE sections as reported in the guidelines. |
RNase AWAY Surface Decontaminant | Thermo Fisher Scientific | 7002 | Use to clean all surfaces as reported in the method instructions. |
RNeasy DSP FFPE Kit | Qiagen | 73604 | Use to isolate RNA from FFPE sections once they have been generated as reported in the guidelines. |
Semi-Automated Rotary Microtome | Leica Biosystems | RM2245 | Use to section FFPE blocks as reported in the guidelines. |
Sodium hydroxide | Millipore Sigma | S8045-500 | Prepare 10 N solution by slowly dissolving 400 g in 1 liter of Milli-Q water. |
Space Ranger | 10X Genomics | 2.0.1 | Use to process sequencing data output . |
Surgipath Paraplast | Leica Biosystems | 39601006 | Use to perform tissue infliltration and embedding as reported in the method instructions. |
Visium Accessory Kit | 10X Genomics | PN-1000194 | Use to perform spatial transcriptomic experiments. |
Visium Human Transcriptome Probe Kit Small | 10X Genomics | PN-1000363 | Use to perform spatial transcriptomic experiments. |
Visium Spatial Gene Expression Slide Kit 4 rxns | 10X Genomics | PN-1000188 | Use to place the sections if performing spatial transcriptomic experiments. |
Xylene | Leica Biosystems | 3803665 | Use to perform tissue clearing as reported in the method instructions. |