Proximity Ligation Assay Allows the Detection, Localization, and Quantification of Protein Arginine Methylation in Fixed Tissue

Arginine methylation by protein arginine methyltransferases (PRMTs) is an abundant post-translational modification (PTM) involved in numerous biological processes. PRMTs catalyze the transfer of methyl groups from the S-adenosyl methionine to arginine residues. The PRMT family comprises nine members classified according to the type of methylation they perform. All members perform monomethylation (MMA). Type 1 (PRMT1, 2, 3, 4, 6, and 8) PRMTs catalyze asymmetrical dimethylation (ADMA), whereas type 2 (PRMT5 and 9) catalyze symmetrical dimethylation (SDMA), and type 3 (PRMT7) only generate MMA¹. By methylating numerous substrates, the different PRMTs regulate a wide variety of important cellular processes such as DNA repair, transcriptional regulation, immune response, RNA processing, and signal transduction^2,3. This is particularly true for steroid hormone signaling, where PRMTs modify the activity of steroid receptors by methylating not only the receptors themselves but also their regulators or histones².

Arginine methylation is largely studied in cancer, as the majority of PRMTs were shown to be overexpressed in cancer in comparison with normal tissues, and their expression is often associated with poor prognosis^4,5. Detection of arginine methylation in vivo is essential in understanding cellular functions associated with this modification. This is conventionally achieved by conducting immunohistochemistry (IHC) with a specific antibody recognizing the methylated arginine residue. However, this method is very limited as it is based on the identification of the modified arginine residue and relies on the efficacy of the antibody used. In situ proximity ligation assay (PLA) was initially developed to study protein/protein interactions in fixed cells or tissues⁶. Interestingly, this technology can also be used to detect PTMs using a pan antibody against the modification of interest, as well as an antibody recognizing the targeted protein. Our team previously adapted this technique to study estrogen receptor alpha (ERα) methylation, using an anti-ERα antibody and an antibody specifically recognizing the methylation site on arginine 260⁷. Of note, this technique can be extended to antibodies recognizing a special type of methylation even when the methylated residue is unknown. Indeed, several companies supply pan antibodies specifically recognizing MMA, ADMA, or SDMA that can be successfully used to study protein methylation in vivo.

Here, as a proof-of-concept, we present a detailed analysis of GR methylation using SDMA antibody in human breast tumors from experimental design to data analysis.

Written informed consent was obtained from each patient. The study protocol was approved by the institutional ethics committee of the Cancer Research Center of Lyon.

1. Choice of the antibodies

1. Use primary antibodies validated by IHC or immunofluorescence (IF).
   ​NOTE: The primary antibodies selected for the study are crucial to the success of PLA and more particularly in fixed tissue. Using an antibody validated by IHC or IF will increase the success rate of the experiment. Optimization and specificity of antibodies are required before conducting experiments, ideally by performing control experiments in cells invalidated for the expression of the protein of interest and using an inhibitor specific to the enzymatic activity of the methyltransferase involved. Here the conditions have previously been optimized¹⁰.

2. Paraffin-embedded cell line pellet preparation

NOTE: The samples are embedded in a gel and then placed in an automated tissue processor for sample dehydration and paraffin embedding.

1. Prepare the cell pellet in a 1.5 mL microcentrifuge tube using trypsin dissociation.
   NOTE: Do not scrape to harvest cells.
2. Resuspend the cell pellet in 1.5 mL of phosphate buffered saline (PBS).
3. Centrifuge the cells at room temperature (RT) at 200 x g. Aspirate the PBS supernatant.
4. Resuspend the pellet in 1 mL of 4% formalin and store the tubes at 4 °C for 3 h.
5. Wash the cells in 1X Tris-buffered-saline (TBS).
   NOTE: Do not use a phosphate-based solution as it depolymerizes the gel.
6. Add 1 mL of TBS and homogenize for 30 s by vortexing. Centrifuge for 5 min at 200 x g and remove the supernatant.
7. Repeat steps 2.5 to 2.7 twice.
8. Perform inclusion of the tissue in the automated tissue processor following the specific program:
   Formalin         1 h      37 °C
   Distilled water 2 min  RT
   EtOH 70%    40 min  45 °C
   EtOH 80%    40 min  45 °C
   EtOH 95%    40 min  45 °C
   EtOH 100%  40 min  45 °C
   EtOH 100%  40 min  45 °C
   EtOH 100%  40 min  45 °C
   Xylene          40 min  45 °C
   Xylene          40 min  45 °C
   Xylene          40 min  45 °C
   Paraffin          1 h       60 °C
   Paraffin          1 h       60 °C
   Paraffin          1 h 30 min  60 °C

3. Tissue fixation and slide preparation

1. Fix the tissues in 4% formalin for 24 h before inclusion.
2. Cut 3 μm thick serial sections of the blocks containing tissues with a microtome.
3. Conduct deparaffinization with the autostainer by sequentially incubating slides in xylene (10 min) twice, 100% ethanol (5 min), 95% ethanol (5 min), and water (5 min).
4. Perform heat-induced epitope retrieval in 10 mM citrate buffer using a water bath at 98 °C for 40 min at the appropriate pH (usually 6-9).
   NOTE: The two antibodies have to work at the same pH for the method to be successful. Several tests were performed before establishing the optimal pH used herein. The antibodies used in this study are SDMA antibody and the GR antibody.
5. Let the slides cool for 20 min, then wash them in 1x PBS for 20 min.

4. IHC experiment

NOTE: IHC experiments are performed using the Discovery XT research instrument (Table of Materials) for automation and reproducibility.

1. To avoid peroxidase quenching, use the inhibitor included in the diaminobenzidine (DAB) kit.
2. Incubate the slides with the two primary antibodies diluted in the antibody diluent for 60 min.
3. Incubate the slides with the secondary anti-rabbit horse radish peroxidase (HRP) antibody for 16 min or the anti-mouse antibody for 30 min.
   NOTE: It is recommended to use a commercial diaminobenzidine (DAB) kit (Table of Materials) for IHC detection.
4. Apply hematoxylin (100 µL/slide for 8 min) and bluing (100 µL/slide for 4 min) to the samples to stain the nuclei.
5. Wash the slides in warm soapy water, then dehydrate using an autostainer and mount on an automated cover slipper using a mounting medium.

5. Proximity ligation assay reactions

NOTE: All of the reagents used are included in the PLA kits (Table of Materials). It is recommended to use 40 μL of reagent for 1 cm² of tissue. All of the incubations need to be performed in a humid environment to prevent excessive evaporation. Do not allow the sample to dry out, as this may lead to background noise. It is recommended to use 1x buffer A (in situ wash buffer, Table of Materials) for the washes in the jars. There should be a minimum volume of 70 mL in the jars when incubating samples under agitation. Samples should be kept at RT before use.

1. Peroxidase quenching
   1. Delimit the reaction area using a hydrophobic pen to prevent excessive evaporation of the solution during the experiment. Add one drop of hydrogen peroxide solution per 1 cm² of each sample. Incubate for 5 min at RT.
      NOTE: It is recommended to optimize the incubation time according to the samples/antibodies used.
2. Blocking
   1. Gently drip buffer A onto the samples and wash in buffer A at 50 rpm under agitation for 5 min at RT.
   2. Tap off the remaining buffer A and add one drop per 1 cm² of blocking solution (included in the PLA kit). Incubate for 30 min at 37 °C.
3. Primary antibody incubation
   1. Dilute the two primary antibodies at suitable concentrations in the antibody diluent.
   2. Remove the blocking solution from the slides and immediately add the primary antibody solution (included in the PLA kit). Incubate for 1 h in a humidity chamber at 37 °C.
4. PLA probe incubation
   1. Dilute the PLA probe PLUS (1:5) with PLA probe minus (1:5) in the antibody diluent.
      NOTE: It is recommended to use 8 µL of PLA probe minus stock (5x), 8 µL of PLA probe plus stock (5x), and 24 µL of antibody diluent for a 40 µL reaction.
   2. Gently drip buffer A onto the samples and wash in buffer A at 50 rpm for 5 min at RT.
   3. Tap off the remaining buffer A and apply the PLA probe solution (included in the PLA kit) to the samples. Incubate for 1 h at 37 °C.
5. Ligation
   1. Thaw the ligation buffer (included in the PLA kit) prior to the following steps.
   2. Gently drip buffer A onto the samples and wash in buffer A at 50 rpm for 5 min at RT.
   3. Dilute the 5x ligation buffer (1:5) (included in the PLA kit) in high-purity water. Add the ligase (1:40) (included in the PLA kit) immediately before adding it to the sample.
      NOTE: It is recommended to add 8 µL of the 5x ligation buffer to 31 µL of high purity water for a 40 µL reaction. Then add 1 µL of ligase.
   4. Tap off the remaining buffer A and apply the ligase solution to the samples. Incubate for 30 min at 37 °C.
6. Amplification
   1. Thaw the amplification buffer prior to the following steps.
   2. Gently drip buffer A onto the samples and wash in buffer A at 50 rpm for 2 min at RT.
   3. Dilute the 5x amplification buffer (1:5) (included in the PLA kit) in high-purity water. Add the polymerase (1:80) (included in the PLA kit) immediately before gently dripping it onto the sample.
      NOTE: It is recommended to add 8 µL of the 5x ligation buffer to 31.5 µL of high purity water for a 40 µL reaction. Then add 0.5 µL of ligase.
   4. Tap off the remaining buffer A and apply the amplification solution to the samples. Incubate for 2 h at 37 °C.
7. Detection
   1. Gently drip buffer A onto the samples and wash in buffer A at 50 rpm for 2 min at RT.
   2. Dilute the 5x detection bright-field buffer (1:5) (included in the PLA kit) in high-purity water.
      NOTE: It is recommended to add 8 µL of the 5x detection buffer to 32 µL of high purity water for a 40 µL reaction.
   3. Tap off the remaining buffer A and apply the detection solution to the samples. Incubate for 1 h at RT.
8. Substrate development
   1. Gently drip buffer A onto the samples and wash in buffer A at 50 rpm for 2 min at RT.
   2. Dilute the substrate solution by diluting the substrate reagents A (1:70), B (1:100), C (1:100), and D (1:50) (included in the PLA kit) in high purity water.
      NOTE: It is recommended to add 0.6 µL of substrate A, 0.4 µL of substrate B, 0.4 µL of substrate C, and 0.8 µL of substrate D in 37.8 µL of high purity water for a 40 µL reaction.
   3. Tap off the remaining buffer A and apply the amplification solution to the samples. Incubate for 10 min at RT.
9. Nuclear staining
   1. Gently drip distilled water onto the samples and wash in distilled water at 50 rpm for 2 min at RT.
   2. Tap off the remaining distilled water. Add one drop of nuclear stain (included in the PLA kit) to each 1 cm² sample and incubate for 2 min at RT.
   3. Rinse the slides under running tap water for 10 min to let the stain mature and obtain a blue coloration. Do not use standing tap water.
10. Dehydration
    1. Incubate the slides in a solution containing 96% ethanol twice for 2 min. Then incubate the slides in a solution containing 99.7% ethanol twice for 2 min.
    2. Incubate the slides in xylene for 10 min. Then transfer the slides to fresh xylene.
11. Mounting of the slides
    1. Use a minimum volume of non-aqueous mounting medium and apply a coverslip on top of the sample, ensuring no air bubbles get caught under the coverslips.
    2. Let the slides dry before analyzing on a bright-field microscope.

6. Imaging for localization

1. Acquire images using an upright microscope.
   NOTE: In this study, imaging of slides was performed under an upright bright-field microscope (Table of Materials). Images were acquired under identical conditions at 40x magnification for at least 10 randomly chosen fields of view in an automated manner. It is recommended to analyze a minimum of 500 cells per condition.

7. Analysis for quantification

NOTE: Quantification of samples was performed using ImageJ software⁸. FIJI, an ImageJ distribution including ImageJ and other pre-installed plugins, was used for the subsequent analyses^9,10.

1. To determine the number of dots:
   1. Open the image.
   2. Perform Color Deconvolution. For this, click on Image > Colors > Color Deconvolution.
   3. Select image: - (color_1) in the filename.
   4. Use the Find Maxima command to determine the number of dots (Process > Find Maxima).
2. To determine the number of cells:
   1. Open the image.
   2. Perform Color Deconvolution. For this, click on Image > Colors > Color Deconvolution.
   3. Select image: - (color_1) in the file name.
   4. Find a threshold that shows the maximum number of nuclei. For this, click on Image > Brightness/Contrast > Threshold.
   5. Fill the holes by clicking on Process > Binary > Fill Holes.
   6. Use the watershed command to divide connected components into separate ones. Click on Process > Binary > Watershed.
   7. Analyze particles to determine the number of nuclei by clicking on Analyze > Analyze Particles.

Using the procedure described above, it is possible to detect and quantify the methylation of a protein of interest. Here, we show the example of the methylation of GR by PRMT5. The antibodies and the experimental conditions for PLA were previously applied to cells¹⁰. Briefly, primary antibodies targeting GR and SDMA are recognized by proximity probes conjugated with complementary oligonucleotides. Then, the hybridization of a circular DNA probe occurs when the proteins are in close proximity. Subsequent amplification of this DNA is visualized by HRP under bright-field microscopy (Figure 1). The ideal conditions to detect GR and PRMT5 by IHC were first analyzed in several breast tumors (Figure 2). For detecting GR methylation, we used primary antibodies raised in different animal species recognizing GR (mouse) and pan SDMA (rabbit); a subsequent PLA experiment was performed as previously described, and images were acquired (Figure 3A). Quantification was performed to determine the number of dots per cell using ImageJ software (Figure 3B).

Figure 1. Schematic diagram of the PLA technique. (A) Primary antibodies raised in different animal species recognizing GR and pan SDMA are incubated on the slides containing the fixed tissues. PLA probes containing secondary antibodies then link to short complementary oligonucleotides recognizing primary antibodies. After annealing and ligation, the signal is amplified and detected by colorimetry. Each brown dot represents an arginine methylation event. (B) Images were acquired using a bright-field microscope. Please click here to view a larger version of this figure.

Figure 2: GR and PRMT5 expression in breast tumor samples. GR and PRMT5 expression were assessed by IHC using the GR antibody (1/200) and the PRMT5 antibody (1/400). Here are two examples of the staining we obtained. Tumor A shows a strong expression in the nuclei of tumor cells. Tumor B shows low level of GR and PRMT5 expression. Objective: 40x. Please click here to view a larger version of this figure.

Figure 3: GR methylation in breast tumor samples. (A) PLA was performed on the same tumors presented in Figure 2 using the GR antibody (1/75) and the SDMA antibody (1/100). Tumor A shows brown dots reflecting GR methylation, whereas tumor B shows no staining, meaning that GR is not methylated in this sample. Objective: 40x. (B) Quantification of the number of dots is performed using an algorithm in ImageJ. Please click here to view a larger version of this figure.

Arginine methylation, like other PTMs, contributes to the fine regulation of protein functions. However, its impact is underestimated due to the difficulty in assessing these modifications, primarily because of a lack of tools. This is particularly true when studying methylation in vivo, where the only way to measure arginine methylation is to possess specific antibodies recognizing the methylated residue of the protein of interest. This clearly constitutes a limitation as the methylated arginine residue must be known, and a specific effective antibody for the modification must be available for IHC application, which is rarely the case.

Here we present a way to circumvent these obstacles by using pan methyl antibodies raised against the different types of methylation (MMA, SDMA, and ADMA). We previously showed that coupling pan methyl antibodies to an antibody recognizing the protein of interest was effective for measuring methylation not only in fixed cells¹⁰ but also in fixed tissues.

However, even though this technology is powerful, there are some limitations that one should be aware of before starting the experiment. The choice of primary antibody is crucial for PLA in cells and even more so in tissues. To maximize the success rate, choose antibodies that are validated in IHC and respect the dilutions proposed. Optimization could first be conducted in cells that express the protein of interest or not, or after inhibition of the target and of the PRMT involved in the methylation (or inhibitor targeting their enzymatic activity). It is possible to fix cells in formalin and embed them in paraffin to conduct experiments in tissue. Indeed, fixation is also a very important step; although formalin-fixed tissues provide good results, Bouin and acidified formol alcohol are not compatible with PLA on tissue.

In addition, as most commercially available pan methyl antibodies arise from rabbits, this implies that the antibody raised against the protein of interest must come from a different species. Here, the GR antibody we used was produced in mice.

Another constraint of the method is that, as heat-induced epitope retrieval is to be conducted before starting PLA experiments, the two antibodies have to work at the same pH. If the optimal pH values are not compatible, it will be necessary to test a range of pH values for each antibody and choose the best pH for both antibodies.

In conclusion, the technique presented here offers a unique opportunity to localize within cells and quantify protein methylation in vivo. Interestingly, this technique can be extended to other PTMs as long as effective pan antibodies recognizing the desired modification are available.