Establishment of immunohistochemical double staining on formalin fixed paraffin embedded tissue
samples with Pax 5, PD1, PDL1 and PDL2.
Written by
Justina Damjanovic Vesterlund
Supervised by Patrik Micke, docent at the department of immunology, genetics and pathology; clinical and experimental pathology
and
Tor Halle, MSc. Biomedicine, clinical pathology, FoUU, UAS.
2 Abstract
Recently the immune cell marker proteins Pax 5, PD1, PDL1 and PDL2 have become in focus as biomarkers for cancer immunotherapy. The
identification of the pattern of expression might be important to guide immune modulatory treatment options and in non-‐small-‐cell-‐luch-‐
carcinoma (NSCLC), melanoma and renal cancer. Pax 5 is a nuclear protein expressed by B-‐lymfocytes. PD1 is a membrane receptor protein on T-‐
lymphocytes that interacts with its ligands PD-‐L1 and PD-‐L2 expressed by antigen presenting cells, dentritic cells and cancer cells.
The aim of the study was to establish double immunohistochemical staining protocols, on normal positive tissue, to visualize the location of two
different proteins at the same time in a tissue sample. Staining optimization regarding antibody concentration, incubation times, antigen retrieval,
chromogen incubation was performed. Double stainings for the antibody pairs Pax 5 and PDL1 and PD1 and PDL1 were established successfully.
However, the staining of Pax 5/ PDL2, and Pax5/ PD1, could not be established.
Both successfully established staining protocols will be tested on a larger selection of cancer tissue. This might in the future help in clinical
diagnostics and therapeutics.
3 Table of contents
Introduction ...4
Immune system ... 4
T-‐cells ...4
B-‐cells...5
PD1... 5
Pax 5 ... 6
Immunohistochemistry ... 7
Aim... 9
Material and methods... 10
FFPE tissue preparation and staining...10
Pre-‐ treatment-‐ antigen retrieval...11
Staining...12
Dehydration...14
Antibody and Ancillary ...15
Results... 17
Optimization of single staining ...17
PD1 ...18
PDL1 ...18
PDL2 ...19
Pax 5 ...20
Double staining...21
Pax 5 and PDL1 double staining...21
PD1 and PDL1 double staining...23
Pax 5 and PD1 double staining...24
Discussion ... 24
Acknowledgement... 26
Annex... 27
References... 29
4
Introduction
Immune system
The immune system consists of the primary and secondary lymphoid organs. The primary organs produce the cells that will develop into the immune system (bone marrow and thymus) and the secondary is the organs where the immune reaction will occur (lymph nodes, spleen, tonsils, special locations in the lungs and in the mucus of the gut).
The cells of the immune system are divided in cells involved in the innate immune response and cells of the acquired immune response, with some overlap. The acquired immune system consists of cells including B and T lymphocytes and plasma cells. T-‐ cells are further divided into helper T cells and cytotoxic T cells. Cell of the innate immune system comprise different types of phagocytic cells. 2
T-‐cells
These cells mature in the thymus and are thereby called T-‐cells. The T cells react to cell-‐
bound antigens presented by specific molecules. The specific molecules are produced by genes encoded in the major histocompatibility complex (MHC) presented on antigen presenting cells (APC). There are two different classes of the MHC, class MHC I and MHC II. All nucleated cells express the class I MHC molecules, MHC II molecules are only expressed on APCs (dendritic cells, macrophages and B cells), epithelial reticular cells of the thymus and endothelial cells. The human variant of the mouse MHC is the human leukocyte antigens (HLAs). T-‐helper cells and regulatory T-‐ cells can interact with other cell of the immunesystem and modify the response. Cytotoxic T cells (also called effector T-‐ cells) can destroy target cells. T-‐ memory cells will remember antigens and improve the efficiency of the immune system.
The T-‐ cells express different kinds of proteins on their surface thereby determining the type of T-‐ cell it is. These marker antigens are called cluster of differentiation, CD.
Therefore a T-‐ cell expressing CD4 (helper T-‐ cell) can be differentiated from a T-‐ cell presenting CD8 (killer T-‐cell). The receptor on the T-‐ cell responsible for antigen recognition presented by the MHC-‐antigen fragment complex is called the T-‐ cell receptor (TCR). The receptor and some other cofactors stabilize the binding of antigen presenting cells to the T-‐ cell. When the TCR recognizes an antigen bound in the MHC either the CD4 or CD8 is recruited. CD4 is associated with MHC II and CD8 with MHC I.
The CD4 expressing helper T-‐ cells recognize antigens associated with MHC II and CD8
5 expressing cytolytic T-‐ cells recognize antigens associated with MHC I. T cells shall not react to self-‐ antigens, but only to foreign antigen fragments bound. The lack of response is called self-‐tolerance. 2
B-‐cells
They will mature in the bone marrow and after antigen contact will differentiate to plama cells or B-‐memory cells. Plama cells produce different kinds of antibodies of different classes such as IgM, IgG, IgA, IgD and IgE. When the B-‐cells are activated they proliferate and produce antibodies (immunoglobulins) that will bind the antigen targets and neutralize them or lead to the destruction of the target by enzymes or macrophages.
Macrophages are called “eater cells” and they will engulf the target cells. They develop from monocytes in the blood. They can also recognize antigens and will upon activation produce cytokines that will cause a local inflammation. 1 2
PD1
Programmed cell death 1 gene, PD1, is a part of the immunoglobulin superfamily. It was first described in mice cells to induce apoptosis . The protein has two hydrophobic stretches, one at the N-‐terminus and one at position 168-‐196, revealing a domain spanning the membrane as an alpha helix. 4 3
PD1 is expressed during B cell differentiation and is expressed in normal lymphoid tissue like appendix, tonsil and lymph node
predominantly in T-‐cells. 5 PD1 is a membrane protein, which works as a suppressor of the immune system upon the binding of its ligand PDL1 that can be expressed by dentritic cells or also cancer cells. CD8+ T-‐cell cells are the ones that are thought to be mostly affected by the PD1 checkpoint pathway, since PD1 is often present on CD8+ T-‐ cells. Figure 1 shows the interaction PD1 and PDL1 presenting cells have and the effect the interaction has on the immune response. PD1 is involved in cancer
development and treatment. The cancer cells express PDL1 and PD1 to evade the immune response and thereby disguising themselves as harmless cells. PD1 pathway is
Figure 1 Picture of the interaction and effect of
PD1 and PDL1. MHC-‐ major histocompability comples presents antigen to the T-‐ cell through TCR-‐ t-‐ cell receptor, immune response is then triggered. But by having the PD1/ PDL1 interaction the reaction is inhibited.
6 apparently most active within the tumour itself. The cancer specific cells appear to target the CD8+ cytotoxic T-‐ cells. 6 The PD1 has a cytoplasmic immunoreceptor
tyrosine-‐ based inhibitory motif which together with a immunoreceptor tyrosine-‐ based switch motif will affect signaling pathways downstream upon ligand binding. Activation of PD1 results in cytokine downregulation, inhibition of proliferation and induction of apoptosis in CD8+ presenting T-‐ cells. 7 It is now known that PD1 checkpoint pathway has an influence in some cancer forms and the blockage of this pathway, is effective on cancer forms that are resistant to other traditional treatments, such as non small cell lung cancer (NSCLC), advanced melanoma and renal cell carcinoma. Anti PD1 drugs called nivolumab and pembrolizumab have been approved for clinical trials by the FDA in the United States for treatment against these cancer forms since the response in patients is good and the extended overall survival is also higher. It is also expected for other cancer forms to get the anti-‐PD1 treatment approved, such as Hodgkin lymphoma, bladder cancer and head and neck cancer. Responses to treatment can be different in individuals with the same type of cancer, consequently proper characterization of each tumour is important, to understand mechanisms of action each checkpoint inhibitor has.
A vide variety of biomarkers should be screened for on each tumour microenvironment to understand the mechanisms going on and the right checkpoint proteins to attack and treat. Establishing immunohistochemical (IHC) stains and reliable staining protocols for detecting biomarkers such as PD1 and PDL1 are necessary since the anti PD1 and anti PDL1 drugs already have been approved for different cancer forms in the U.S. Issues with creating standardized, reliable and therefore comparable results needs to be taken into consideration. 6 PDL1 and PDL2 proteins are ligands for PD1. PDL2 is not very well invesitgated, but also identified as a ligand of PD1. These proteins are presented on antigen presenting cells (dendritic cells and macrophages). 8
Pax 5
Pax 5 gene codes for a transcription factor called B-‐ cell lineage specific activation protein, BSAP. It is located on 9 p13 in the human genome. This transcription factor was detected as mammalian homolog to the sea urchin protein TSAP (tissue specific
activator protein), in the late 1980s. It was confirmed to have TSAP homology by recognizing the four binding sites for TSAP. It was detected in nearly all B-‐ cell
development stages but not in fully developed plasma cells. 9 It was later confirmed by experiments that Pax 5 knock out results in arrested B-‐ cell development at pre B-‐ cell
7 (in fetal livers) and pro B-‐ cell (in bone marrow) stages. The hypothesis that Pax 5
downregulation is a prerequisite for B-‐ cell terminal differentiation into plasma cells is supported by the fact that Pax 5 is not expressed in fully differentiated cell lines
(plasmacytoma and myeloma cell lines) of murine and human origin. Experimental B-‐
cell differentiation has also been obtained by repression of Pax 5 in splenic B-‐ cells.
Some cancers circumvent the terminal differentiation and thereby keep the cells in constant proliferation by rearranging the gene location, thus probably placing the gene expression under different regulatory elements and enhancers. 1011 The Pax 5 gene is involved in a number of different cell developmental and differentiating genes that can induce oncogenic behavior such as p53, telomerase reverse transcriptase, death-‐
domain associated protein Daxx and tumour-‐ suppressor retinoblastoma protein, for instance. All these proteins are involved in oncogenesis. 12 Pax 5 protein BSAP will further be referred to as Pax 5.
Immunohistochemistry
Immunohistochemistry is a technique that is used all over the world for diagnostics and research on formalin fixed paraffin embedded, FFPE, tissue samples. The technique to preserve tissue in formalin has been used since the late 1800s. Formaldehyde was discovered in 1859, by Butlerov and formalin fixation was discovered and developed by Ferdinand Blum in the later 1800s and has then been the method of choice for fixation of tissue samples. 13 How the actual fixation is achieved is not yet fully understood but extensive research has been done to elucidate the phenomenon. It is known that formalin chemically alters the sample in ways like methylol group additions, cross-‐
linking between amino, imido, amido, guanidyl, hydroxil-‐ and aromatic rings of proteins.
Intra-‐ and inter-‐ molecular cross-‐ links has also been detected. 14 Proper formalin fixation is very important for further analyzes of the tissue sample. It is very important that the formalin fully penetrates the tissue sample to create a good fixation. 15 After fixation the water in the tissue is replaced by alcohol and then all the alcohol is cleared by xylene and the sample is soaked in paraffin and then embedded in paraffin. The tissue is sectioned and samples are put on glass slides for analysis. The diagnostic standard staining method is hematoxylin (dyes the nucleus of the cells blue) and eosin (dyes the cytosol of the cells red) to easily make the different bigge cell components, such as endoplasmic retuculum, golgi and mitocondria, visible. More detailed diagnostic methods can be used such as immunohistochemistry (IHC), immunoelectron microscopy
8 (IEM) in situ hybridization (ISH), flow cytometry (FCM), imaging flow cytometry (IMS) and other techniques. This report will focus on IHC.
The tissue sample will get a pretreatment, to retrieve the antigen or epitope. Epitope or antigen is the general term for a part of a protein recognized by an antibody. 16 The purpose of the pretreatment is to “unmask” the tissue and reverse the crosslinks and other alternations made in the fixation process. In 1991 Shi et al presented a new method of antigen retrieval or de-‐ masking, heat induced antigen retrieval, HIAR (also called heat induced epitope retrieval, HIER). The HIAR process revolutionized IHC and resulted in more antigens could be detected, increased sensitivity, lowering the
concentration of antibodies for good detection and thereby lowering costs. The fact that proteins exposed to heat will denature and precipitate is well known. 17 In HIAR, energy added by the heat aids to break covalent bonds formed in the fixation process. 1819 HIAR will not affect the primary structure of the protein (protein chain) but the secondary (alpha helix or sheet structure) will be partially altered and the teriary (global folding of the protein) will be totally lost. 18 The use of high pH about pH 9-‐10) or low pH (about pH 6) in combination with high temperature is often used. The pH is thought to affect the refolding of the denatured protein in the cooling process and depending on the amino acid composition different parts of the protein will be exposed with different pH.
20Therefore different pH of the pretreatment can affect the antibody binding and
staining results. In IHC the detection of antigen is done by antibody visualisation.
Antibodies have a high affinity towards the target antigen (differing from antibody to antibody) and can be directed towards any protein or peptide (antigen). Antibodies are
Figure 2 Schematic picture of indirect detection method in IHC. Signal amplification is made by attaching many
enzymes to a polymer conjugated to the secondary antibody. Substrate is added (not shown in the picture) and the enzymes will convert it into a colored product.
9 therefore used in detection of a wide variety of proteins. The antibody used to detect the antigen is called primary antibody. The primary antibody can be directly labeled and the target protein can be detected. Another detection method includes an unlabeled primary antibody, then a labeled secondary antibody that will detect the primary antibody. A probe (Denoted as such by the company, no further information given.) The secondary antibody will then detect both rabbit primary antibody and the probe. The secondary antibody is conjugated to a polymer which has a number enzymes conjugated to it, thereby giving signal amplification. 21 22 In this stydy the indirect detection method was used. The detection method is schematicly described in Figure 2. In IHC enzymatic labels (enzymes conjugated to the antibody) are common, such as horseradish peroxidase (HRP) and alkaline phosphatase (AP). HRP is an enzyme that will convert non-‐colored chromogens to colored substances. HRP is often used together with chromogens such as DAB (diaminobenzidine), Deep space black, TMB. HRP (40 kDa) is stable, small, has a high turnover and therefore gives a rather strong signal compared to AP (140 kDa). 23 The tissue sample is then counterstained with hematoxylin for nuclear blue staining.
Hematoxylin is a positively charged molecule that will react with negatively charged molecules in the tissue, such as RNA and DNA. The nucleus of each cell will therefore be stained blue and other negatively charged compounds in the tissue will also be stained.
24
Since immunohistochemistry is the method of choice to visualise specific proteins in the tissue environment, this technique is particular useful to describe the expression of the previously described immune markers in human cancer.
Aim
The aim of the study was to establish double staining protocols for combinations of Pax 5, PD1, PDL1 and PDL2 with regards to antigen retrieval methods, antibody
concentration and positive tissue selection. The stainings was performed on normal tissue samples, which expresses the combinations of the proteins that were to be detected. In this case all proteins were expressed lymphatic tissue.
10
Material and methods
FFPE tissue preparation and staining
The preparation of the tissue samples is important for preservation and good staining.
The following text will guide step by step in the process. Formalin fixation, paraffin embedding preparation steps and staining are explained in the following text. All steps were done in the order they are presented in the text.
Tissue fixation: The fresh unprocessed tissue is formalin fixated for a least 24 h in 4%
formalin.
Dehydration: The tissue is then dehydrated in the machine: VIP-‐ Vacuum infiltration tissue processor, Sakura Tissue-‐ Tek VIP 6, model VIP-‐6-‐E2. Protocol for dehydration as following (protocol used at the department of Clinical Pathology, UAS):
1. Formalin 20 min 2. H2O 5 min 3. 70 % EtOH 60 min 4. 95 % EtOH 60 min 5. 95 % EtOH 60 min 6. Abs. EtOH 60 min 7. Abs. EtOH 60 min 8. Xylene 60 min 9. Xylene 60 min 10. Xylene 60 min 11. Paraffin 45 min 12. Paraffin 60 min 13. Paraffin 105 min 14. Paraffin 60 min
Embedding: Tissue is embedded in paraffin and let to cool on a freeze block.
Sectioning: Tissue is sectioned at 3µm on waterfall microtome, Microm HM 355S, let to stretch in room temperature water and then let to stretch a little bit further in 40-‐ 50 oC water. The tissue section is then placed on a glass slide, Thermo scientific superfrost plus. The slides are dried in 58 oC for at least 15 min to get rid of all water residuals.
Deparaffination/ Rehydration: Before staining the tissue on the glass slide is
deparaffinized and rehydrated for the antibody detection method to work, since the antibodies are water soluable.
Table 1 Deparaffination protocol is performed according to the protocol below. Glass slides are let to stand in the solutions in the order in the protocol.
Solvent Time (min)
Xylene 3
Xylene 3
Xylene 3
Absolute pure ethanol 3
Absolute pure ethanol 3
95 % ethanol 3
95 % ethanol 3
70 % ethanol 3
dH2O 1
Pre-‐ treatment-‐ antigen retrieval
Antigen retrieval is performed to reverse the crosslinks formed in the formalin fixation and thereby demasking the antigen.
• Samples are loaded in the pressure cooker (decloaking chamber-‐ Biomedical, DC2008NTL) in either Dako Citrate buffer pH 6 (low pH pretreatment) or Dako TE buffert pH 9 (high pH pretreatment). The program used is 98 OC for 49 minutes then, the system will depressurize and be finished when the temperature is 74 OC.
• Sample is then taken out from the decloaking chamber and let to cool down at room temperature for 15 min, then washed with Dako wash buffer (0,05 mol/L Tris/HCl, 0,15 mol/L NaCl, 0,05 % Tween 20, pH 7,6), 4 x 5 min and then loaded on the racks for the IntelliPATH FLX, automated slide staining system, Biocare Medical.
• Enzymatic digestion: Proteinase K (serine protease) 5 min incubation on slide, rinse with Dako wash buffer. In the case of enzymatic digestion, the serine protease will process the tissue by removing some of the crosslinks formed in the fixation process, thereby aiding in unmasking the antigen. 25
Staining
Staining is performed in the machine IntelliPATH, where solutions are dispensed onto the tissue glass slides in order given in table 2. The incubation times for the primary antibodies are given in Table 6. The protocols given are template protocols where the parameters are altered according to what is given in the tables. All single stainings are made according to the protocol in Table 2. All double stainings are made according to Table 3, parameters open for alterations are noted with asterix and explained below each table. Sequential staining tried on one combination of two proteins of same antibody origin, PD1 and Pax 5, was stained according to Table 4.
Table 2 Template protocol for single staining. Steps are performed in the order in the table.
Activity Volume
( µL) Incubation time
(min) Wash Number of
washes
Pretreatment ∗ -‐ -‐ -‐
Peroxidase
blocking 300 5 Buffer** 1
Primary
antibody 300 Depending on Ab Buffer** 1
Mouse probe
secondary 300 15 Buffer** 1
HRP conjugate Secondary antibody
300 30 Buffer** 2
HRP chromogen 300 Denpending on
chromogen Buffer** 1
Wash 300 1 dH2O
Haematoxylin
counter stain 600 5 Buffer** 1
Wash 300 3 dH2O
*Pretreatment done outside the machine. ** Biocare medical TBS automation wash buffer pH 7.7 (containing TBS, tris buffered saline, surfactant Tween 20 and preservative Proclin 950)
13
Table 3 This is a template protocol for double staining used in intelliPATH automatic stainer machine. The steps in the machine were performed in the order in the table.
Protocol step Volume
(µL) Incubation time
(min) Wash Number of
washes
Pretreatment* -‐ -‐ -‐ -‐
Peroxidize
blocking 300 5 Buffer**** 1
1st primary Ab 300 Depending on Ab Buffer**** 1 2nd primary Ab 300 Depending on Ab Buffer **** 1 MACH 2 **
Secondary Ab 300 30 Buffer**** 2
DAB chromogen 300 5 Buffer**** 1
WR chromogen 300 10 *** Buffer**** 1
Wash dH2O 300 1 -‐ -‐
Haematoxylin
counter stain 600 5 Buffer**** 1
Wash dH2O 300 3 -‐ -‐
*Done outside the machine. ** Depending on wanted results either MACH 2 double stain 1 or double stain 2. *** In noted slides 20 min. **** Biocare medical TBS automation wash buffer pH 7.7 (containing TBS, tris buffered saline, surfactant Tween 20 and preservative Proclin 950)
14 Table 4 This is the protocol for sequential double staining and it is used when primary antibodies have the same origin. The staining was performed in the intelliPATH automatic stainer and the staining steps were performed as given in the table below.
Activity Volume
(µL) Incubation
time (min) Wash Number of
washes
Pretreatment* -‐ -‐ -‐ -‐
Peroxidize
blocking 300 5 Buffer** 1
1st primary Ab
(PD1) 300 75 Buffer** 1
Mouse probe 300 15 Buffer** 1
HRP conjugate
Secondary Ab 300 30 Buffer** 2
Chromogen
DAB 300 5 Buffer** 1
Wash 300 1 dH2O 1
2nd primary
Ab (Pax 5) 300 20 Buffer** 1
Mouse probe 300 15 Buffer** 1
HRP conjugate
Secondary Ab 300 30 Buffer** 2
Chromogen deep space
black 300 10 Buffer** 1
Wash 300 1 dH2O 1
Hematoxylin
counter stain 600 5 Buffer** 1
Wash 300 3 dH2O 1
*done outside the machine ** Biocare medical TBS automation wash buffer pH 7.7 (containing TBS, tris buffered saline, surfactant Tween 20 and preservative Proclin 950)
Dehydration
Dehydration is performed when the staining is done, to preserve the results of the staining. All dehydration is performed in the order given in Table 5, except for double stainings when chromogen warp red is used, then the dehydration procedure is given after Table 5.
15
Table 5 The tissue sample is dehydrated after staining for preservation according to following dehydration protocol.
The tissue slide is let to stain the solvent in the given order.
When using chromogen Warp red the following rapid dehydration protocol is used to avoid chromogen fading:
• Rinse in tap water
• Dip 10 times in 3 different sets of absolute pure EtOH
• Clear the slides by dipping 10 times in two sets of Xylene.
Slides are then mounted with cover glass using Pertex mounting medium, Histolab.
Slides are let to dry for at least 15 minutes in 58 oC and then evaluated in light microscope.
Antibody and Ancillary
Information about the incubation time and origin of the primary antibodies are given in Table 6. Secondary antibodies product names, how they work and wich chromogen they can be combined with is given in Table 7. In formation about the chromogens and notes on altered incubation time is given in Table 8. Extended information about the primary antibodies is given in Table 9.
Table 6 Antibody: incubation information.
Antibody Standard
incubation time
(min) Origin Notes
Pax5 20 Mouse
PD1 90 Mouse
PDL1 30 Rabbit
PDL2 30 Rabbit 45 min inc. time on
noted slides
Solvent Time (min)
70 % EtOH 3
95 % EtOH 3
Absolute pure EtOH 3
Absolute pure EtOH 3
Xylene 3
Xylene 3
Xylene 3
16
Table 7 Information about secondary antibodies and enhancers.
Product Specifications Related chromogen
MACH1 probe
Secondary antibody Mouse antibody probe -‐
HRP polymer
Secondary antibody Detects mouse probe and
rabbit antibodies DAB, Deep space black
MACH 2 double stain 1 Mixture of two different secondaryantibodies
HRP-‐ conjugated secondary Ab, detects rabbit primary Ab.
AP-‐ conjugated secondary Ab, detects mouse primary Ab.
HRP-‐ DAB and deep space black
AP-‐ Warp red
MACH 2 double stain 2 Mixture of two different secondaryantibodies
HRP-‐ conjugated secondary Ab, detects mouse primary Ab.
AP-‐ conjugated secondary Ab, detects rabbit primary Ab.
HRP-‐ DAB and deep space black
AP-‐ Warp red
Table 8 Chromogen information about incubation time and alterations made in the experiments.
Chromogen Resulting color Incubation time (min) Notes
DAB Brown to black 5 Color depending on
staining intensity
Warp Red Red 10
Extended
incubation time of 20 min on noted slides.
Deep space black Black 10
Table 9 General information about the primary antibodies.
Antibody Normally expressed in tissue
Cell localization
(active protein) Type Conc.
(mg/mL) Clone Company Product nr
PD1 tonsil, appendix membrane mouse 1 monoclonal Abcam ab52587
PDL1 Tonsil, appendix membrane,
extracellular Rabbit 1,016 polyclonal Cell Signaling 13684 PDL2 Spleen, tonsil, appendix membrane, extra
cellular rabbit 1 polyclonal Sigma Aldrich HPA013411
Pax5 tonsil, appendix nuclear mouse 0,157 monoclonal Dako M7307
17
Results
Experiments started out with new antibodies and had not been optimized on the
instruments used in this experiment. Suggestions on concentrations and pretreatments were given by some of the manufacturers but not all. Since different proteins were to be detected in the same sample, research on the Human Protein Atlas web page gave indications where the proteins are mainly expressed and how the staining looked for some antibodies against the protein.26 The antibodies used there were not from the same manufacturer used in the experiments in this report. Positive tissue is normal tissue where the protein is naturally expressed, in this case lymphatic tissue (tonsil, spleen, appendix used). The proteins are supposed to be expressed in different parts of the cells in the lymphatic tissue and also in different cells. Pax 5 is a B-‐ cell marker and is epressed in the cell nucleus so there are round areas in the tissue where these B-‐ cells are located. These areas are expected to be stained in the cell nucleus for Pax 5. PD1 are mostly expected to be expressed on T-‐ cells in the cell membrane. PDL1 and PDL2 are expected to be expressed on antigen presenting cells, such as squamous epithelial cells, and they are also located in the cell membrane, pertuding extracellulary.
Optimization of single staining
Single staining optimization was performed regarding concentration, positive tissue, pre-‐ treatment and incubation times of the primary Ab. The following antibodies were optimized: Pax 5, PD1, PDL1 and PDL2. The conditions and the results, as well as the slide numbers of the single stains are shown in annex 1.
The single stains were analyzed and appropriate conditions were then used for double staining (annex 2). General information about the antibodies is shown in table 9.
18 PD1
Antibody dilutions 1:4000, 1:1000 and 1:500 were used. The tissue chosen was tonsil A and samples were both run in low and high pH pretreatment pressure cooker according to protocol. Incubation time was set to 75 minutes. Dilution 1:4000 and 1:1000 of the antibody was too diluted, but 1:500 showed weak staining results, low pH better than high. Continued tonsil A staining on low pH pressure cooker pretreatment, dilution 1:200, 1:100 and 1:50. Dilution 1:50 showed the best results. Last staining optimizing PD1 a new tonsil called tonsil B, slide 4.1, was tested with the previous settings and the results were good, Figure 3 A and B. PD1 is showing on cell membranes in normal lymphatic tissue in T-‐ cells and developing B-‐ cells.
PDL1
Antibody dilutions 1:4000, 1:1000 and 1:500, were used on tissue tonsil A. Pretreatment with both high and low pH in pressure cooker, according to protocol, was used.
Incubation time was set to 30 minutes. No detection of staining in high or low pH pretreated 1:4000 or 1:1000 diluted antibody stained tissue. Both high and low pH pretreated tissue slides at 1:500 dilution showed very weak staining results. Low and high pH pretreated Tonsil A was then stained with antibody concentration
corresponding to dilution 1:200, 1:100 and 1:50. Low pH pretreated samples at dilutions 1:200 and 1:100 yielded weak and diffuse staining patterns. Low pH
pretreated sample showed moderately staining results. High pH pretreated samples at dilution 1:200 and 1:100 yielded weak staining, but 1:50 showed moderate staining with slightly stronger staining than low pH pretreated 1:50 stained slide. A new tonsil,
Figure 3. PD1 single staining. A Overview. Figure showing the optimized staining results for PD1. Dilution 1:400,
low pH pressure cooker pretreatment and 75 minutes incubation time. The staining is done with DAB chromogen. B.
Slide 4.1 Close up of the staining.
19 tonsil B was selected and stained according to the best result obtained on tonsil A. Slide 1.2, tonsil B, high pH pretreated, 1:50 diluted PDL1 antibody showed strong staining results, Figure 4. PDL1 is expected to be expressed in antigen presenting cells in lymphatic tissue, such as tonsil in this case. In this case squamous epithelial cells show staining.
Figure 4. PDL1 single staining A. Slide 1.2 Overview of the optimized staining of PDL1. Dilution 1:50, high pH
pressure cooker pretreatment, 30 minutes incubation time, tonsil B.. B. Slide 1.2 Close up of the DAB staining. PDL1 shows in cell membrane and cytoplasma.
PDL2
PDL2 staining started out at dilution 1:200, 1:100 and 1:50 in low pH pretreatment pressure cooker according to protocol with 30 minutes incubation time on both spleen and tonsil A tissue. Spleen stained slides were all stained good, almost too strong and thereby spilling over to the surrounding tissue at dilution 1:50. Because other antibodies had tonsil as reference tissue, the analysis of spleen slides were not continued. Tonsil A slides showed weak staining at dilutions 1:200 and 1:100, but moderate staining at 1:50.
Slide 2.3, tonsil B low pH pretreated 1:50 diluted slide with extended incubation time 45 minutes showed the best staining results, Figure 5 A and B. PDL2 is expected to show positive staining in normal lymphatic tissue, antigen presenting cells. However the staining was not regarded as specific, because all cell types showed some staining signals. Therefore the antibody was excluded from further analysis.
Figure 5. PDL2 single staining A. Slide 2.3 Overview of PDL2 optimized single staining. Dilution 1:50, low pH pressure
cooker pretreatment and 45 minutes incubation time on tonsil B. B. Slide 2.3 Close up of the DAB stained PDL2.
.
20
Pax 5
Pax 5 staining optimization was performed on tissue tonsil A, at dilutions 1:50, 1:30 and 1:15 in low and high pH pressure cooker pretreatment according to protocol in table 2.
The staining was very strong in all slides. Slides pretreated in high pH showed stronger staining than in low pH pretreated slides. Further staining in high pH with dilutions at 1:200, 1:100 was performed and showed equally strong staining as with higher
concentrations. Further optimization on tonsil B at dilutions 1:400 and 1:200 in both high and low pH pretreatment was performed. Both low and high pH
pretreatment at dilution 1:400 resulted in strong staining, but the staining was better defined and slightly stronger at high pH pretreatment (slide 5.5) than at low pH
pretreatment. On a different tonsil tissue, tonsil B, an equal or almost better staining was obtained with the higher dilution 1:400 than with dilution 1:50 on tonsil A, Figure 6. Pax 5 is expected to be present in lymphatic tissue in the early to late B-‐ cells, predominantly located in the germinal centra (where B-‐ cells develop), appearing as round areas in the lymphtic tissue, Figure 6 A and C. The optimized staining was high pH pretreatment, dilution 1:400, on tonsil B.
Figure 6. Pax 5 single staining. A. Slide 5.5 Overview of the optimized Pax 5 single staining. Dilution 1:400, high pH
pressure cooker pretreatment, 20 min incubation time on tonsil B. B. Slide 5.5 Close up of the DAB stained, nuclear localized Pax 5. C. Slide 5.1.1 Overview of DAB stained Pax 5 in tonsil A at dilution 1:50 in low pH pressure cooker pretreatment. D. Slide 5.1.1 Close up of picture C.
21 Double staining
Pax 5 and PDL1 double staining
For the double staining the best single stainings were compared and a compromise protocol was created.
The double staining according to the protocol in table 3 was performed with three different tissue samples on the same slide, tonsil D, appendix and tonsil A. A
pretreatment with low pH and incubation times Pax 5 20 minutes, PDL1, 30minutes was used. MACH 2 double stain 1 was used, which is rabbit-‐ HRP conjugate and mouse-‐ AP conjugate. Pax 5 (rabbit origin Ab) was stained with warp red (AP-‐ chromogen)
resulting in a red colored staining, and PDL1 with DAB (HRP-‐ chromogen) resulting in a brown staining. PDL1 commonly show membrane and some cytoplasmic localization and Pax 5 has a nuclear localization. The PDL1 staining, in both appendix and tonsil D, was good and well defined but the Pax 5 warp red stain was slightly weaker, Figure A and B. The staining on tonsil A was negative for both Pax 5 and PDL1, maybe because of lack of representative tissue. Subsequently, Pax 5 and PDL1 were evaluated on tonsil D with four different pretreatments to compare the impact of different antigen retrieval,
Figure . Slides were incubated at concentrations for Pax 5 at 1:400 and PDL1 at 1:50 with following pretreatment alternatives: low and high pH pressure cooker, no pretreatment and enzymatic pretreatment. Without pretreatment or with enzymatic pretreatment the tissue did not demonstrate any staining. Low pH pressure cooker yielded moderate DAB PDL1 staining but no visible Pax 5 staining. High pH yielded strong and distinct DAB PDL1 staining and fairly moderate warp red Pax 5 staining. Consequently, further optimization was done for the wrap-‐red Pax 5 protocol. Since hematoxylin is predominant nuclear stain of blue color, the antibody visualization of the nuclear protein Pax 5 with warp red might be difficult to detect. Thus, tonsil D was stained with two different protocols, one with a prolonged warp red incubation time extended from 10 to 20 minutes. The other protocol, with a switch of the applied dyes, i.e. now DAB was used for the nuclear protein PAX5 and warp red for the membrane protein PD-‐L1. This should facilitate to distinguish colors on the hematoxylin counterstained tissue sections.
Slide 6.9 gave the best staining, with the extended incubation time resulted in a more distinct warp red Pax 5 stain and a distinct PDL1 DAB staining, Figure C and D. The reverse staining gave strong Pax 5 DAB staining and a moderate PDL1 warp red staining,
Figure E and F.
22
Figure 8. Pax 5 (red) and PDL1 (brown) double staining with different antigen retrieval protocol. A. Slide 6.2 Low pH pressure cooker pretreatment. B. Slide 6.1 High pH pressure cooker pretreatment. C. Slide 6.3 No pretreatment. D.
Slide 6.4 Enzymatic pretreatment.
Figure 7 Pax 5 and PDL1 double staining. A. Slide 6.1 Displays the best stained Pax 5 and PDL1. Pax 5 is stained with
chromogen warp red and PDL1 is stained with DAB. B. Slide 6.1 Close up on the stain in picture A. Stained PDL1 (brown) is located in the membrane and Pax 5 in the nucleus. C. Slide 6.9 Double staining of Pax 5 and PDL1 with all the same parameters as in 6A, but the incubation time for warp red was prolonged to 20 min. D Slide 6.9 Close up of C. E. Slide 6.1.1 Reverse staining was also performed. Pax 5 is therefore stained with DAB and PDL1 is stained with warp red. F. Slide 6.1.1 Close up of the reverse staining.
23
PD1 and PDL1 double staining
Double staining was carried out on a slide with three different tissue samples, tonsil D, appendix and tonsil A. PD1 primary antibody is a mouse antibody and PDL1 is a rabbit primary antibody. MACH 2 double stain 1 was used, with DAB and warp red chromogens.
PD1 was stained with warp red and PDL1 with DAB. The PDL1 DAB staining of both appendix and tonsil D was strong and the warp red stained PD1 was a slightly weak.
PD1 and PDL1 double stain was then carried out on two appendix tissue samples, one with low pH and the other in high pH pressure cooker pretreatment, the incubation times of the primary antibodies not altered, but the warp red chromogen incubation time was prolonged from 10 to 20 minutes. The low pH pretreated sample showed moderate PD1 warp red staining but weak PDL1 DAB staining, Figure 9 A-‐ C. The high pH pretreated sample showed strong PDL1 DAB staining and moderate PD1 warp red staining, Figure 9 D-‐ F. Two TMA, tissue micro array, samples were also stained. The TMA was constructed to work as positive control for PDL1. Both TMA slides were run in low pH pressure cooker pretreatment, standard incubation times for the primary antibodies.
One slide was incubated with warp red chromogen for the extended time 20 minutes and the other with standard incubation time. The TMA slide stained with the standard chromogen incubation time showed moderate PDL1 DAB staining and weak warp red PD1 staining. The TMA slide with the extended warp red incubation time showed strong DAB PDL1 staining and moderate warp red PD1 staining.
Figure 7 PD1 and PDL1 double staining. A. Slide 6.1.2 PD1 and PDL1 double stained appendix. PDL1 is stained with DAB and
PD1 is stained with warp red. Low pH pressure cooker pretreatment. B. Slide 6.1.2 Close up of picture 7A. C. Slide 6.1.2 Close up of another part of the appendix showing cells that express PDL1 to a higher extent. D. Slide 6.1.3 PD1 and PDL1 double stain with a high pH pressure cooker pretreatment. E. Slide 6.1.3 Close up of the staining. F. Slide 6.1.3 Close up of the same region of the appendix as in 7C.
