Presented here is a protocol for generating formalin-fixed, paraffin-embedded cell pellet controls for immunohistochemistry.
Positive and negative controls with known expression of target proteins are essential for the development of immunohistochemistry (IHC) assays. While tissue controls are beneficial for well-characterized proteins with defined tissue and cellular expression patterns, they are less suitable for the initial development of IHC assays for novel, poorly characterized, or ubiquitously expressed proteins. Alternatively, due to their standardized nature, cell pellets, including cancer cell lines with defined protein or transcript expression levels (e.g., high, medium, and low expression), transfected over-expressing cell lines, or cell lines with genes deleted through cell engineering technologies like CRISPR, can serve as valuable controls, especially for the initial antibody characterization and selection. In order for these cell pellets to be used in the development of IHC assays for formalin-fixed, paraffin-embedded tissues, they need to be processed and embedded in a manner that recapitulates the procedures used for tissue processing. This protocol describes a process for creating and processing formalin-fixed, paraffin-embedded cell pellet controls that can be used for IHC method developments.
Immunohistochemistry (IHC) is one of the most commonly used assays in investigative and diagnostic pathology. Depending on the context and assay, IHC is used to assist in cancer diagnosis1,2, predict treatment response3,4, identify pathogens5, characterize cell types in diseased tissues6, and study biologic pathways and tissue responses7,8. In all situations, the basic principle of an IHC assay is that an antibody binds specifically to a target of interest, most commonly a protein, and this binding event is subsequently visualized in a tissue section9. However, one of the greatest challenges of any IHC assay is assuring the antibodies are specifically detecting the target of interest10. Antibody specificity is a challenge in most immunoassays, but immunohistochemistry offers unique challenges in that there are no secondary measures, such as molecular weight, to differentiate between specific and nonspecific labeling. This is particularly troublesome when evaluating poorly characterized or ubiquitously expressed targets that lack well-defined cellular localization patterns. Therefore, robust controls that can help characterize binding specificity are critical when developing a new IHC assay10.
For targets that are well-defined with characteristic cellular expression patterns, tissue controls are frequently utilized in IHC method developments. Based on a wealth of prior data, one can determine if the antibody is labeling the tissue, the cell, and the subcellular compartment in which it is known to be expressed, and that it is not labeling tissue components where it should not be present11. However, tissue controls are of limited use for poorly characterized, novel targets without known expression patterns or for proteins that are widely expressed and lack distinct expression patterns. In both of these scenarios, the lack of a well-defined expression pattern makes it impossible to discriminate specific from non-specific labeling in tissues. In these situations, cell pellets offer a valuable alternative IHC control. Cell pellet controls can include: cancer or other cell lines that have endogenous or intrinsic/non-induced expression levels of the protein of interest, and whose protein expression can be characterized by western blotting, flow cytometric analyses, or extrapolated from transcriptional profiling; engineered cell lines that either over-express the protein of interest or have the encoding gene of interest deleted; or cells that have been treated under specific conditions to induce protein expression or signaling events of interest (e.g., phosphorylation)10,12. Well-characterized protein expression levels in cell lines also allow for evaluation of the sensitivity of an assay by using a panel of cell lines with high, medium, low, and absent protein expression. Additionally, engineered cell pellets can be valuable species specific controls for veterinary species, for which there may be limited characterization or available tissue controls13. While cell pellets have their limitations, such as the limited proteome present in cell lines that will not reflect the diverse proteome in tissues, they serve as suitable controls for confirming that the antibody can detect the target of interest as well as ruling out indiscriminate binding by the primary antibody, secondary antibody, or other regents in the assay10.
Most tissues in diagnostic and investigative pathology are fixed in neutral buffered formalin, dehydrated in a series of alcohols, cleared in xylene, and processed and embedded in paraffin wax. Formalin fixation cross-links proteins, and fixation and each additional step in tissue processing can directly affect proteins and the ability of antibodies to detect them9,14. Therefore, it is important that any controls used in an IHC assay undergo the same fixation, tissue processing, and embedding procedures. This article describes the unique considerations to process and embed cultured cells to serve as controls for developing IHC assays in formalin-fixed paraffin embedded tissues, with the methodology primarily focusing on the handling and processing of the cell pellet in a histology laboratory.
1. Cell pellet preparation
2. Fixation of cell pellet
3. Trimming and processing of cell pellet
4. Embedding of cell pellet
5. Sectioning of cell pellet
Following the addition of the agarose-based gel, the cell pellet should form a solid gelatinous mass amenable to handling (Figure 1B). Once embedded, the pellet will have a consistency similar to a solid tissue, and should be relatively easy to routinely section on a microtome. Once cell pellets are embedded, they can be used in histology and IHC experiments in an identical manner as formalin-fixed, paraffin-embedded tissues. This includes the use of heat-induced epitope or enzymatic antigen retrieval with indirect chromogenic detection methods (e.g., using a secondary antibody with horseradish peroxidase and diaminobenzidine detection as shown in Figure 2 and Figure 3) using commercial IHC platforms9. Histologically, the cells should be evenly dispersed throughout the section with minimal cell clumping, although adherent cells may maintain their cell-cell interactions in the pellet. This dispersion allows for distinction between individual cells, although this distinction will be significantly influenced by cell size and relative density within the gel pellet. The nucleus, cytoplasm, and cell membrane are more distinct and easier to visualize upon dispersion (Figure 2). In Figure 2, three cell lines are immunolabeled for the TEAD transcription factor. Immunolabeling is visualized with the brown diaminobenzidine (DAB) chromogen. The cell lines have varying levels of expression of the TEAD transcription factor, ranging from no expression (Figure 2A) to weak expression (Figure 2B), to strong expression (Figure 2C). In this example, labeling is observed in the nucleus, as would be expected from a transcription factor, and is absent from the cytoplasm and cell membrane, which are visible but lack labeling (Figure 2). Cell pellets can be incorporated into cell pellet microarrays (Figure 3) on standard 23 mm x 75 mm x 1 mm slides. Incorporating the cell pellets in a microarray allows for the evaluation of controls with varying expression levels in the same slide. In this example, PEG10 deficient mouse embryonic stem cells (left) serve as a negative control and 293T cells that over-express PEG10 (right, brown immunolabeling) serve as a positive control in the microarray (Figure 3). There is no variation in the tissue and embedding procedures for cell pellets that will be included in microarrays, and microarrays can be generated using methods similar to those used for tissues19.
Figure 1: Cell pellet processing. (A) Cells are pelleted in 50 mL conical tubes and mixed with the gel. (B) Once solidified, pellets are serially sliced to fit into a 26 mm x 26 mm x 5 mm tissue cassette. (C) Cell pellets are wrapped in biopsy paper prior to being placed in the tissue processor. (D) Up to three cell pellets can be serially collected on a single 25 mm x 75 mm glass histology slide. Please click here to view a larger version of this figure.
Figure 2: Cell pellets with different levels of protein or transcript expression can be used to evaluate the dynamic range of the assay. Cell pellets are immunolabeled for the TEAD transcription factor and demonstrate varying levels of TEAD expression.(A) DAUDI cells lack TEAD expression and have no immunolabeling in the cytoplasm or nucleus. Blue stain is hematoxylin counterstain of the nucleus. (B) 293T cells demonstrate weak nuclear labeling (brown diaminobenzidine (DAB) chromogen) of the TEAD transcription factor. (C) Detroit 562 cells strongly express TEAD as demonstrated by the intense brown labeling in the nucleus. Labeling within the nucleus, but not the cytoplasm (off-white to gray region around the brown nucleus), demonstrates the appropriate labeling of the immunohistochemistry assay. Note that in all three cell lines, the cells are dispersed, allowing for the visualization of individual cells, including their nuclear and cytoplasmic morphologies. Scale bar = 25 µm. Please click here to view a larger version of this figure.
Figure 3: Cell pellet microarray created by using multiple cell pellets with varying protein expression levels allows for the simultaneous assessment of controls in a single slide. PEG10 deficient mouse embryonic stem cells (left) and 293T cells over-expressing PEG10 (right) are immunolabeled for PEG10. While strong labeling (brown DAB chromogen) is observed in 293T cells over-expressing PEG10, no labeling is apparent in PEG10 deficient cells. Blue represents hematoxylin counterstain. Scale bar = 200 µm. Please click here to view a larger version of this figure.
This protocol describes a methodology to generate formalin-fixed, paraffin-embedded cell pellets that can be used as controls for downstream immunohistochemistry and in situ hybridization studies. The histology methodologies described in this protocol are applicable to a diverse range of cancer and primary cell lines, and primarily adapts routine histology techniques to produce these pellets17,18. When processed and embedded, the pellets can be used in a similar manner to tissues. This includes using them with heat-induced and enzymatic antigen retrieval protocols for immunohistochemistry experiments. One aim of the methodologies used in this protocol is to preserve the morphology and antigenicity of the cells throughout the process. Therefore, EDTA, which is relatively gentle in terms of changes to both the morphology and antigenicity, is used for detaching the cells. This is not to say that other approaches, such as physical disruption, are not viable; however, any approach to detach cells should ensure that the cells are not injured in the process. The second aim of this protocol is to fix and process the cells in a manner similar to tissues, using the same fixative, fixative ratio, processing schedule (fixation, dehydration, clearing, and paraffin infiltration), and embedding techniques, so they can serve as comparable controls for downstream assays. Hence, the fixation and processing approaches used to generate cell pellets should mimic the approaches used for tissues.
The protocol described in this report is not adapted to handle cells with infectious agents, such as cells infected with pathogenic bacteria or viruses, as cells are detached, collected, and centrifuged before fixation. Investigators might consider protocols to fix the cells before detachment, but this would require further optimization to collect the cells without disturbing the cell morphology. Furthermore, fixation times and handling conditions for cells with infectious agents require additional considerations based on the infectious agent and associated institutional biosafety protocols.
Formalin-fixed, paraffin-embedded cell pellets are uniquely advantageous in having well-defined protein expression levels10. While cancer and endogenous cell lines offer a selection of cells with varying levels of protein expression, genetic engineering technologies enable scientists to model the protein expression, through over-expressing proteins of interest and the use of CRISPR technologies to excise or insert encoding genes of interest20,21. The disadvantage of over-expressed proteins in cell lines is that they are poor measures for the sensitivity of an assay as they may not represent endogenous protein levels22. In contrast, both endogenous cell lines and cancer cell lines can better represent endogenous expression levels, and CRISPR-mediated deletion of the encoding gene in a cognate line can serve as a corresponding negative control. Furthermore, endogenous cell lines or cancer cell lines with varying levels of protein expression are ideal for titration experiments to pick final antibody dilutions and to best understand the sensitivity of an assay (Figure 2). The decision surrounding what cell lines to use should be based on individual experimental needs and will often utilize a combination of approaches.
In addition to evaluating whether an antibody can detect the protein of interest, a panel of cell lines can be used to define an antibody's specificity. For instance, a panel of cell lines that individually express a family of closely related proteins can be used to test whether an antibody is specific for an individual protein or whether it also detects other closely related proteins. More elaborate controls may involve using cell lines that express, either through CRISPR knock-in or through over-expression, proteins with point mutations that cannot undergo specific signaling events that are being detected (e.g., mutation in specific phosphorylation sites when evaluating a phospho-specific antibody). While these are more complex approaches, they may be necessary to confirm that the antibody used only labels under specific circumstances10.
It is important to note that genetic manipulations of cell lines might not generate a homogenous cell population. For instance, transfection efficiency in overexpressing cell lines is typically not 100% and some cells might not over-express the protein of interest. Inclusion of FLAG or a related tag in the transfection that can be detected with standard methods can be helpful to assess the transfection efficiency of the cell line23. This can be helpful both to determine if transfection was successful and rule out a lack of detection due to the protein expression, and to serve as a reference for the expected proportion of cells that should express the protein of interest.
In situ hybridization (ISH) can also be a beneficial tool to characterize the target gene expression in cell pellets and inform IHC method development10. When screening antibodies, it can be informative to know when the transcript, and thus the potential protein, can be detected. Additionally, cell pellet ISH screening can be beneficial for ISH method development. While specificity is less frequently an issue for ISH assays, it is still important to have appropriate controls and similar considerations for the development and utilization of controls that can be applied to ISH studies.
Tissue microarrays are created by removing cores from donor blocks and transferring these cores, often in a grid pattern, into a recipient paraffin block. In the end, the recipient block contains a spectrum of samples in a single block, allowing for all of the samples in the block to go through identical IHC procedures and allowing for the direct comparison of multiple samples on the same slide24,25. Since cell pellets are relatively uniform populations, they can be accurately represented by 1 mm cores, making them ideal candidates for inclusion in similar arrays. Using a cell pellet array, it is possible to include cell pellets with varying levels of expression, cell pellets that uniquely express related proteins, and cell pellets that express orthologs of the protein of interest from different species in a single slide (Figure 3). This allows all the cell pellets to be simultaneously evaluated under uniform conditions in a rapid manner, while minimizing the use of reagents25.
A minimum starting cell pellet volume of 2 mL gathered from eight T175 flasks is recommended for this protocol. This volume allows for the production and archiving of multiple cell pellet blocks, so that cell pellet controls can be standardized over longer periods of time and multiple cell pellet arrays can be created from a given pellet. Lower cell pellet volumes can be used when dealing with primary patient-derived cell lines, slow-growing cell lines, or when conditions limit the volume of the sample. Of course, lower starting volumes will potentiate the negative impact from any cell loss during fixation and preparation of the pellet and will limit the material available for downstream processes. It is especially important to take care when removing fixative after centrifugation, to minimize any associated loss. For lower sample volumes, a 1.5 mL, capped centrifuge tube can be used to pellet the cells. These tubes can be longitudinally bisected with a blade for processing.
Similar to tissue fixation, cell fixation within this pellet is critical for downstream IHC assessments. In order to achieve complete fixation, we use at least a 10:1 formalin to cell pellet ratio and invert the conical tube to maintain the cells in suspension. If the cells are not in suspension at the time of fixation, there is a risk of incomplete or inadequate fixation, which will affect downstream immunolabeling. Often this manifests as strong labeling in the periphery and loss of labeling in the center of the pellet or labeling of variable intensity throughout the pellet.
This protocol uses Histogel, which is primarily composed of hydroxyethyl agarose, to both bind the cell pellet and to enable the cells to be evenly distributed throughout the cell pellet. Without it, cells become compacted and lose their cytomorphologic details. These cytomorphologic details are often important during antibody screening, as they provide additional information regarding whether the antibody is labeling in the appropriate subcellular compartment (e.g., the nucleus, cytoplasm, or plasma membrane). In contrast, too much gel can result in lower cell densities within the pellet, leading to the cells being widely distributed, and reducing the cell numbers per section.
This protocol can be routinely used to develop IHC controls. The materials needed to create these pellets are common in investigative biology and histology laboratories, and the methods are straightforward and easy to adapt. While cell pellets have limitations as IHC controls, they serve as great tools for initial antibody screening and complement other tissue controls.
The authors have nothing to disclose.
We would like to acknowledge the collaboration of our colleagues in the Genentech's Research organization, and especially the Pathology core (P-core) laboratories who contributed to the development of these methods over the years.
10% Neutral Buffered Formalin | VWR | 16004-128 | |
50 mL Conical Tube | Becton Dickinson Labware | #0747-1886 | |
70% Ethanol | Koptec | V1401 | |
95% Ethanol | Koptec | V1101 | |
Biopsy Wraps | Surgipath Medical Industries, Inc | #01090 | |
Costar Stripette serological pipette 10mL | Corning | CLS4101 | |
Flex 100 | Epredia | 8101 | |
Flex 95 | Epredia | 8201 | |
Histogel | Thermo Scientific | #R904012 | |
Leica Automated Rotary Microtome | Leica | RM2255 | |
Micro Spatula, rounded and tapered ends | Tedd Pella | #13510 | |
NanoZoomer 2.0 HT whole slide imager | Hamamatsu | ||
Paraplast Tissue Infiltration/Embedding Paraffin | Surgipath | 39601006 | |
Pipette Controller | CAPP | PA-100 | |
Reagent Alcohol | Epredia | 9111 | |
Sterling Probe 5” 2mm Tip with Eye | Roboz Surgical Instrument Co., Inc | #RS-9522 | |
Superfrost Plus positively charged microscope slides | Thermo Scientific | 6776214 | |
Tissue cassettes; PrintMate Slotted Cassette | Epredia | B851120WH | |
TMA Tissue Grand Master | 3DHistech LTD | ||
Xylenes | VWR | 89370-088 |