Protocol to differentially quantify spatially resolved viral protein-cellular protein interactions via proximity ligation assays

Summary

The spatial organization of viral and cellular proteins shapes signal transduction. Here, we present a protocol to quantify differential protein-protein interactions by measuring their spatial association using proximity ligation assays (PLAs). We describe steps for seeding, transfection, proximity labeling, and confocal microscopy imaging and provide procedures for quantitative analysis. Our approach complements co-immunoprecipitation-based interactome data by enabling in situ quantification of differential binding between cellular and viral interaction partners.

For complete details on the use and execution of this protocol, please refer to Klute et al.¹

Graphical abstract

Highlights

• •Assess the frequency of close proximity events between viral and cellular proteins
• •Quantify relative differences in interactions of protein variants in situ
• •Detailed suggestions and macros for (semi-)automated PLA quantification
• •Rapid, flexible system suitable for analyzing virtually any protein-protein interaction

Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics.

The spatial organization of viral and cellular proteins shapes signal transduction. Here, we present a protocol to quantify differential protein-protein interactions by measuring their spatial association using proximity ligation assays (PLAs). We describe steps for seeding, transfection, proximity labeling, and confocal microscopy imaging and provide procedures for quantitative analysis. Our approach complements co-immunoprecipitation-based interactome data by enabling in situ quantification of differential binding between cellular and viral interaction partners.

Before you begin

To promote their replication, viruses express proteins that manipulate their cellular environment including the cell-intrinsic innate immune system.^2,3,4,5 Many of these strategies involve binding of a viral protein to a cellular factor at a specific site within a cell. Understanding the spatial component of viral protein–cellular protein interactions (PPIs) is thus essential to understand the molecular details of virus-host interplay. During viral evolution or when comparing activity of viral strains and family members, binding strengths may vary. It is thus also important not only to assess whether a viral protein binds, but also whether it may have gained or lost binding efficiency. For example, during the recent COVID-19 pandemic, the severe acute respiratory coronavirus 2 (SARS-CoV-2) evolved so-called variants of concern that are characterized by increased viral fitness and immune evasion, usually conferred by distinct mutations that e.g., altered the binding properties of the viral proteins.^1,6,7,8,9 Discovery of interaction partners and differential analyses of their binding strength is often accomplished by purification of the cellular partner from cell lysates and identifying the co-purifying proteins for example by quantitative mass spectrometry or Western Blotting.¹⁰ Unfortunately, approaches assessing PPIs starting from a cell lysate disregards the spatial localization, and it can be sometimes tricky to retain the modifications on proteins. Of note, especially viral proteins often surgically interfere with cellular signaling cascades by interacting with cellular proteins at specific locations or in distinct activation modes. Thus, interaction analyses of viral proteins should ideally be in situ, retaining the spatial organization and preserving the activation status of the cellular interaction partner. There are several techniques available capable of achieving this. For example, pulldowns extended by chemical crosslinking or proximity labeling (BioID).¹¹ However, these techniques can be difficult to set up and require specialized analysis expertise to perform robustly. Another complementary approach to lysate-based (co-)pulldown assays, is immunofluorescence staining. They have the intrinsic advantage of retaining the spatial organization of proteins within a cell. In its simplest form, the co-localization of the fluorescent signal from two proteins can be used as a readout. However, if the localization of at least one of the partners is not very distinct, or the interaction is transient or rare, co-localization will not be very helpful. A technique based on immunofluorescence microscopy, that requires relatively simple technical setup but is able to show even rare interactions are proximity ligation assays (PLAs).¹² It was developed in 2002 by a team led by Sven Fredriksson and Ulf Landegren. Initially designed to detect platelet-derived growth factor (PDGF), PLA was later adapted to detect and visualize proteins, protein complexes, and protein modifications in situ (within cells) and has since become a popular tool for studying protein-protein interactions with high sensitivity.¹² In 2006, PLA was further adapted for in situ detection, allowing for direct observation of individual protein complexes in cells by using antibodies covalently linked to oligonucleotide arms, a technology later commercialized by Olink Bioscience. Later technological developments include the establishment of the Duolink PLA technology (Sigma-Aldrich). This protocol requires a pair of antibodies from two different species which detect the putative partners.¹³ Both of these primary antibodies will be stained with respective secondary antibodies, PLA probes, that are marked with DNA oligonucleotides. In case these nucleotides are within 40 nm of each other in situ Rolling-Circle-Amplification amplifies the oligonucleotides, allowing them to be stained by fluorescent oligonucleotide probes (Figure 1).

Figure 1.

Schematic depiction of the principle of the PLA assay

The viral protein is overexpressed in HeLa cells. Putative viral and cellular protein interaction partners are first stained with primary antibodies originated from different species. Species-specific secondary antibodies conjugated to oligonucleotides, termed PLUS and MINUS PLA probe, are added and bind selectively to their corresponding primary antibodies. If the two target proteins are in a proximity below 40 nm to each other, the oligonucleotides carried by the PLA probes can hybridize and form a circular DNA molecule which serves as a template for rolling circle amplification. Fluorescence-labeled complementary oligonucleotide probes bind to the amplified DNA product, generating a fluorescent signal, termed PLA signal, that can be detected and quantified.

In this protocol we sought to combine the spatial proximity information provided by a proximity ligation assay with the possibility to quantify relative complex formation/interaction strength between an overexpressed viral protein and an endogenous cellular protein, essentially creating a protocol for spatial-informed differential interaction analyses. Our method is based on and uses the in situ Duolink PLA technology®. In principle this protocol can be applied to study any protein-protein interaction and define relative differences in binding efficiency in situ.

Before you begin, there are several steps to consider.

Innovation

Proximity Ligation assays have been around since (at least) 2002 and have been applied to multiple biological problems.^{1,12,14,15,16} While a plethora of different protocols exist to study protein-protein interactions or protein-nucleic acid interactions, as well as their localization, the focus is often less on gauging differences in binding strength between mutants. Our protocol was optimized to complement co-immunoprecipitations or other protein-protein interaction analyses between overexpressed viral and endogenous proteins with differential spatial in situ analyses that can be accurately quantified to reflect the strength of the interaction.

Choice of cell model

Timing: 2–4 days

The choice of the right cell line is crucial if the close proximity between an endogenous and an overexpressed protein is assessed. First, it is necessary to ensure the target endogenous protein is expressed in the cell line of choice. Secondly, the cells need to be readily transfectable to achieve high transfection efficiency, which is also discussed in the section ‘Optimization of transfection’. We generally use HeLa cells as they are readily transfectable, have a convenient flat morphology for imaging, and readily attach to the cover slips. The latter is important during the many incubation and washing steps of the PLA protocol.

• 1.Browse the literature and use tools like the human protein atlas¹⁷ (https://www.proteinatlas.org/) to ensure that the targeted endogenous protein of choice is expressed in the cell line of choice.
• 2.We advise to perform immunofluorescence staining of the target protein before continuing with the actual PLA to also ensure that the antibody works and is specific.
• 3.
  In addition, the cellular protein of choice may be depleted e.g., by siRNA, to determine antibody specificity.

  • a.
    Seed 3.5x10⁴ HeLa cells (or another cell line of your choice) on coverslips in 24-well plate in 500 μL DMEM supplemented with 10% FBS, 2 mM L-glutamine and 13 μM gentamicin (DMEMxxx).
  • b.
    Transfect the respective siRNA using Lipofectamine RNAiMAX according to the manufacturer’s protocol (https://documents.thermofisher.com/TFS-Assets/LSG/manuals/Lipofectamine_RNAiMAX_Reag_protocol.pdf). The siRNA is mixed with Lipofectamine RNAiMAX and Opti-MEM, incubated for 10 min and carefully added to the cell monolayer in a dropwise manner.
  • c.
    Fix the cells as described in the detailed protocol below.
  • d.
    Incubate the fixed, permeabilized and blocked cells with the antibody in different concentrations for 2 h at 37°C (or 16 h at 4°C).

    Note: These steps are described in the step-by-step method details.

  • e.
    Wash cells 3 times with DPBS and incubate with a fluorophore-conjugated secondary antibody directed against the species of the primary antibody for 2 h at 37°C.
  • f.
    Wash cells 3 times with DPBS, one time with ultrapure water and mount the cover slips as described in detail below.
  • g.
    Image the mounted coverslips within 1–2 days.

Antibody dilution

Timing: 4 days

The right balance between signal and background is essential for a proper PLA. Thus, the antibody concentration targeting the overexpressed protein/protein-tag should be carefully titered using immunofluorescence staining (Figure 2A). Additionally, the background signal is checked by comparing to a vector control and the protein of choice transfected plasmids (Figure 2A). We recommend starting with a 400-fold dilution of the antibody.

• 4.
  Determining the right dilution of the antibody targeting the overexpressed protein or its tag.

  • a.
    Seed 3.5x10⁴ HeLa cells (or another cell line of your choice) on coverslips in 24-well plate in 500 μL DMEMxxx.
  • b.
    Transfect 0.5 μg expression vector per well.
  • c.
    Perform the transfection, cell harvesting and fixation according to the procedure described in detail below.

    Note: These steps are described in the step-by-step method details.

  • d.
    Wash cells 3 times with DPBS and incubate with a fluorophore-conjugated secondary antibody directed against the species of the primary antibody for 2 h at 37°C.
  • e.
    Wash cells 3 times with DPBS, one time with ultrapure water and mount the cover slips as described in detail below.
  • f.
    Image the mounted coverslips within 1–2 days.

    CRITICAL: Ensure the primary antibodies targeting the endogenous protein and the overexpressed protein of interest are raised in different species, which is necessary for the PLA protocol. Use e.g. an antibody raised in mouse to target the endogenous protein and an antibody raised in rabbit to target the overexpressed protein.

Figure 2.

Preparation for PLA by optimization of antibody concentration and transfection efficiency

(A) Varying antibody concentrations stain the overexpressed protein/-tag differentially well. This test is done to find the highest dilution to observe strong staining while limiting the amount of antibody used. Overexpression of SARS-CoV-2 E protein with a StrepII-tag. green, StrepII-AF488; blue, DAPI. Scale bar: 20 μm.

(B) Optimization of the transfection using different transfection reagents, here Lipofectamine2000 or Lipofectamine3000, which affects the expression of the (viral) protein. Overexpression of SARS-CoV-2 E protein with a StrepII-tag. green, StrepII-AF488; blue, DAPI. Scale bar: 50 μm.

Optimization of transfection

Timing: 4 days

Expression levels of the target protein in the cell and hence the availability of that protein to be in close proximity to the targeted endogenous protein. To achieve optimal transfection efficiency, it is advisable to test the transfection in the cell line of choice (Figure 2B).

• 5.
  Optimizing the transfection efficiency of the viral protein of choice.

  • a.
    Seed 3.5x10⁴ HeLa cells (or another cell line of your choice) on coverslips in 24-well plate in 500 μL DMEMxxx.
  • b.
    Transfect 0.5 μg per well of the expression vector constructs using Lipofectamine 2000 or 3000 or any transfection reagent that is recommended for the cell line of choice.

    Note: These steps are described in the step-by-step method details.

  • c.
    Harvest the cells as described in detail below or in the section ‘Antibody dilution and specificity’.

    CRITICAL: All overexpressed proteins and their variants should exhibit similar transfection efficiency to ensure correct read-out. Therefore, it is necessary to pre-test the expression vector constructs using the same cell line and transfection agents as in the actual PLA experiment.

Assessment of antibody and PLA specificity

Timing: 3–4 days

To assess and minimize the background/noise levels, perform controls before or at the same time as every PLA analysis. To this end, transfect control vector plasmids or protein expression vector plasmids. Then incubate samples with an antibody against the endogenous protein and one against the overexpressed protein/-tag. Here we use anti-ABCG2 as the antibody against the endogenous target and anti-StrepII as the antibody against the overexpressed protein/-tag. Test the antibodies individually and together. Quantify the resulting images and calculate statistics of the control to the actual analysis to ensure PLA dots are specific to the close proximity of the targeted proteins and not due to background noise.

• 6.
  Perform the PLA as described below in the step-by-step method section for single antibody or vector transfected controls.

  • a.
    Seed cells of choice and transfect one day later with the empty vector control and the WT protein of interest for 30 h.
  • b.
    Fix cells and incubate with either individual antibodies or combined antibodies; e.g., only anti-StrepII, only anti-ABCG2 or anti-StrepII and anti-ABCG2 together.
  • c.
    Perform PLA staining as described in the manufacturers guidelines (https://www.sigmaaldrich.com/DE/en/technical-documents/protocol/protein-biology/protein-and-nucleic-acid-interactions/duolink-fluorescence-user-manual?srsltid=AfmBOoq1PUtjj5g2uQs83EXMTdilwf31D1ik5otmgGnMiUhOz2m2RtvF) and as described below.

    Note: These steps are described in the step-by-step method details.

  • d.
    Image PLA samples and quantify using e.g., ImageJ.

    Note: Quantification details are described in the step-by-step method details.

    CRITICAL: The number of PLA dots resulting from the close proximity of the overexpressed protein or its variants to the endogenous protein of choice need to be statistically significantly higher than the single antibody or vector control-transfected sample (See section expected outcomes, Figure 9). Depending on the primary antibody, the expected PLA signal should range between 0 and 3 PLA dots in an ideal negative or single antibody control setting. PLA dot >5 can still be considered ok if the PLA count in the samples is significantly higher.

Figure 9.

Expected outcome of the PLA analysis of an overexpressed viral protein and an endogenous protein

(A) Representative images of PLA protocol performed in cells either transfected with a vector control or a vector encoding for StrepII-tagged SARS-CoV-2 E T9 (wildtype) or the T9I variant. Cells are incubated with the individual antibodies targeted against StrepII-tag or the endogenous target ABCG2. Red, PLA; blue, DAPI. Scale bar: 20 μm (B) Quantification of the immunofluorescence images of (A). Lines represent the mean of N = 19-70 (individual cells) ± SEM.

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Monoclonal mouse anti-ABCG2 Antibody (BXP-21) (PLA 1:100)	Santa Cruz Biotechnology	Cat#sc-58222; RRID: AB_630828
Monoclonal mouse anti-StrepII-tag Antibody (517) (PLA 1:400)	Novus Biologicals	Cat#NBP2-43735
Polyclonal rabbit anti-StrepII-tag Antibody (WB 1:2,000; PLA 1:450)	Abcam	Cat#ab76949; RRID: AB_1524455
Chemicals, peptides, and recombinant proteins
Bovine serum albumin (BSA)	KPL	Cat#5140-0006
Dulbecco’s Modified Eagle Medium (DMEM)	Gibco	Cat#41965039
Dulbecco’s Phosphate Buffered Saline (DPBS)	Gibco	Cat#14190-094
Duolink In Situ Detection Reagents FarRed (https://www.sigmaaldrich.com/DE/en/product/sigma/duo92013?srsltid=AfmBOooTtvjisVXRPAWk-Z_xevsRtj29ZTbRrs50eMht82TA6BUh6w4_)	Sigma-Aldrich	Cat#DUO92013
Duolink In Situ Mounting Medium with DAPI (https://www.sigmaaldrich.com/DE/de/product/sigma/duo82040?icid=sharepdp-clipboard-copy-productdetailpage)	Sigma-Aldrich	Cat#DUO82040
Duolink In Situ PLA Probe Anti-Mouse MINUS (https://www.sigmaaldrich.com/DE/de/product/sigma/duo92004?icid=sharepdp-clipboard-copy-productdetailpage)	Sigma-Aldrich	Cat#DUO92004
Duolink In Situ PLA Probe Anti-Rabbit PLUS (https://www.sigmaaldrich.com/DE/de/product/sigma/duo92002?icid=sharepdp-clipboard-copy-productdetailpage)	Sigma-Aldrich	Cat#DUO92002
Duolink In Situ Wash Buffers (A and B) (https://www.sigmaaldrich.com/DE/de/product/sigma/duo82049?icid=sharepdp-clipboard-copy-productdetailpage)	Sigma-Aldrich	Cat#DUO82049
Fetal bovine serum (FBS)	Capricorn	Cat#FBS-LE-12A
Gentamicin	PAN-Biotech	Cat#15710-049
L-glutamine	PAN-Biotech	Cat#P04-80100
Lipofectamine 2000 Transfection Reagent (https://www.thermofisher.com/order/catalog/product/11668019)	Invitrogen	Cat#11668019
Lipofectamine 3000 Transfection Reagent (https://www.thermofisher.com/order/catalog/product/de/en/L3000008)	Invitrogen	Cat#L3000008
Lipofectamine RNAiMax Transfection Reagent (https://www.thermofisher.com/order/catalog/product/13778150?SID=srch-srp-13778150)	Invitrogen	Cat#13778150
Opti-MEM reduced serum media (Opti-MEM)	Gibco	Cat#31985047
Paraformaldehyde (PFA)	Santa Cruz Biotechnology	Cat#sc-281692
Penicillin-Streptomycin	PANBiotech	Cat#P06-07100
Triton X-100	Sigma-Aldrich	Cat#T8787
0.05% Trypsin-0.02% EDTA	PAN Biotech	Cat#P10-023100
Trypan Blue Stain 0.4%	Invitrogen	Cat#T10282
UltraPure distilled water	Invitrogen	Cat#10977-049
Critical commercial assays
CellMask Plasma Membrane Stain	Invitrogen	Cat#C37608
MemBrite Fix 543/560 Cell Surface Staining Kit	Biotium	Cat#30094
Zenon Human IgG Alexa Fluor 488 labeling kit	Invitrogen	Cat#Z25402
Zenon Human IgG Alexa Fluor 594 labeling kit	Invitrogen	Cat#Z25407
Zenon Mouse IgG1 Alexa Fluor 488 labeling kit	Invitrogen	Cat#Z25002
Zenon Mouse IgG1 Alexa Fluor 568 labeling kit	Invitrogen	Cat#Z25006
Zenon Rabbit IgG Alexa Fluor 488 labeling kit	Invitrogen	Cat#Z25302
Zenon Rabbit IgG Alexa Fluor 568 labeling kit	Invitrogen	Cat#Z25306
Experimental models: Cell lines
Human: HeLa	ATCC	Cat#CCL-2; RRID: CVCL_0030
Recombinant DNA
Plasmid: pLVX-EF1alpha-empty	Konstantin Sparrer, Ulm University Medical Center, Ulm, Germany	Klute et al.1
Plasmid: pLVX-EF1alpha-SARS-CoV-2-E−2xStrep-IRES-Puro	Nevan Krogan and David Gordon, University of California San Francisco, San Francisco, USA	Gordon et al.18
Plasmid: pLVX-EF1alpha-SARS-CoV-2-E T9I-2xStrep-IRES-Puro	Konstantin Sparrer, Ulm University Medical Center, Ulm, Germany	Klute et al.1
Software and algorithms
Adobe Illustrator 2025	Adobe	https://www.adobe.com/products/illustrator.html
GraphPad Prism 10	GraphPad	https://www.graphpad.com
ImageJ (Fiji)	N/A	https://imagej.net/software/fiji
Imaris 10.1.1	Oxford Instruments	https://imaris.oxinst.com/
ZEN 2010	Zeiss	https://www.zeiss.com
Other
Cell culture plate 24-well, standard, flat base	Sarstedt	Cat#83.3922
Coverslips	VWR	Cat#6310150
Delicate task wipe	Klimtech Science	Cat#7557
Humid chamber – plastic container
Microscope slides	VWR	Cat#631-1551
Nail polish (transparent)	–	–
Vacusip	Integra Biosciences	Cat#159000
Zeiss LSM980	Zeiss	https://www.zeiss.com

Materials and equipment

Buffer and media

Fixation buffer: 46.25 mL 4% (v/v) Paraformaldehyde (PFA) (final concentration 3.7% (v/v)) and 3.75 mL Dulbecco’s Phosphate Buffered Saline (DPBS).

Note: Storage at 4°C for minimum 1 year.

CRITICAL: PFA is toxic. It is advised to work in a chemical fume hood and wear proper protection such as gloves, lab coat and goggles.

Permeabilization buffer: 250 μL Triton X-100 (final concentration 0.5% (v/v)) and 49.75 mL DPBS.

Note: Store at RT for up to 1 year.

CRITICAL: Triton-X-100 can cause skin and eye irritation. It is recommended to wear protection such as gloves, lab coat and goggles.

Blocking buffer: 2.5 mL Bovine serum albumin (BSA) (final concentration 5% (v/v)) and 47.5 mL DPBS.

Note: Store at 4°C for up to 1 year.

PLA wash buffer A: 1 pouch of Duolink In Situ Wash Buffer A diluted in 1000 mL deionized water. Stir until dissolved.

Note: Store at 2°C–8°C for maximum 2 weeks. Warm up to RT before use.

PLA wash buffer B: 3.62 g Duolink In Situ Wash Buffer B diluted in 100 mL deionized water. Stir until dissolved.

Note: Store at 2°C–8°C for maximum 2 weeks. Warm up to RT before use.

DMEMxxx medium

Reagent	Final concentration	Amount
Dulbecco’s Modified Eagle Medium (DMEM)	1x	500 mL
Fetal bovine serum (FBS) (heat-inactivated)	10% (v/v)	50 mL
L-glutamine	2 mM	5 mL
Optional: gentamicin	13 μM	325 μL
Optional: penicillin- streptamycin Solution	100 U/mL 100 μg/mL	5 mL
Total	N/A	N/A

All reagents are used under sterile conditions. For heat-inactivation of FBS, incubate it for 30 min at 56°C in a water bath. We typically discourage the use of antibiotics like gentamicin or penicillin-streptomycin in mammalian cell culture to prevent bacterial contaminations.

Note: Store the medium maximum for 3 weeks at 4°C or up to 1 week at RT.

Step-by-step method details

Cell seeding and transfection

Timing: 2 days

During this step HeLa cells are seeded on ⌀ 13 mm coverslips in a 24-well plate. Cells are transiently transfected 24 h post seeding with an expression vector construct encoding the viral protein of interest.

CRITICAL: The cells used for seeding have to be free of contamination, possess 80%–90% confluency in a T75 or T175 cell culture flask and exhibit a viability of over 90%.

Note: It is recommended to use cells with a passage between 5 and 20 post-thawing (for HeLa cells, if other cell types are used, this may vary).

• 1.
  Seeding HeLa cells on coverslips.

  • a.
    Remove the medium from the cell culture flask and rinse the cell layer once with 8–10 mL DPBS.
  • b.
    Remove DPBS and incubate cells with 3 mL 0.05% Trypsin-0.02% EDTA for 5–7 min at 37°C in a humified 5% CO₂-containing incubator.
  • c.
    Verify under an inverted phase-contrast microscope that the cell layer is detached.
  • d.
    Add 7 mL DMEMxxx medium and aspirate cells by gently pipetting.
  • e.
    Transfer the cell solution to a 50 mL falcon.

    Note: Proceed quickly after cell detachment to preserve cell viability.

  • f.
    Count the living cells by using dye exclusion (1:1 mix of cell solution with Trypan Blue Stain). Alternatively, other dyes, e.g., Erythrosin B, can be used to assess cell viability.
  • g.
    Adjust the cell concentration and volume with DMEMxxx for seeding.

    Note: Prepare at least 10% extra of the cell seeding solution to ensure every well receives the same number of cells.

  • h.
    Add one coverslip per well in a cell culture plate 24-well (standard, flat base) using a tweezer.
  • i.
    Seed 5x10⁴ cells in 500 μL DMEMxxx for a single well of a 24-well plate.
  • j.
    Press down the coverslip carefully with a sterile pipet tip, to ensure that it is not floating.

    Note: Use tweezer or a tip to keep the coverslip on the bottom of the well during seeding. This process allows the attachment of the cells only at the top side of the coverslip.

  • k.
    Incubate the plate at least overnight at 37°C in in a humified 5% CO₂-containing incubator.

    CRITICAL: It is recommended to monitor the cell growth before transfecting using an inverted phase-contrast microscope. Cell confluency should be around 60% to maximize the transfection efficiency.

    Note: Pre-warm Opti-MEM to 37°C before use to enhance the transfection efficiency and prevent thermal stress of the cells. Lipofectamine 3000 Transfection Reagent (P3000 reagent and Lipofectamine 3000 reagent) should be at 22°C, but use quickly and then stored again at 4°C.

    Note: Prepare 10% extra volume of the DNA-Opti-MEM-P3000 and Lipofectamine 3000-Opti-MEM mix to ensure every well receives the same amount of transfection mix.

• 2.
  Transfection of HeLa cells using Lipofectamine 3000 Transfection Reagent. Alternatively, Lipofectamine 2000 can be used (see also before you begin section).

  • a.
    Prepare a tube with 0.5 μg DNA construct, 25 μL Opti-MEM and 1 μL P3000 reagent for one well.
  • b.
    Prepare a second tube with 25 μL Opti-MEM and 1.5 μL Lipofectamine 3000 reagent for one well.
  • c.
    Add the solution of the second tube to the first and vortex.
  • d.
    Incubate the mixture for 15 min at 22°C.
  • e.
    Carefully add 50 μL of the mixture dropwise to the HeLa cells in one well, without disturbing the monolayer.
  • f.
    Incubate the plate at 37°C o.n. in a humified 5% CO₂-containing incubator.

Optional: Perform a medium change 6 h post transfection or after 16 h incubation with 500 μL fresh DMEMxxx to reduce cell cytotoxic effects of the transfection reagent. We highly recommend this step.

Cell fixation, permeabilization, and blocking

Timing: 2 h

This step fixes the structure the HeLa cell at a specific timepoint and allows the staining of the proteins of interests with antibodies and PLA probes in later steps.

CRITICAL: Prevent that the cells dry out at any point. Work fast between the steps of removing and adding liquid to the well.

Note: After fixation, incubation time with DPBS in washing steps can be prolonged for further minutes if necessary.

Note: Add liquid carefully to the wells by pipetting the liquid on the side of the well. It is possible to use a Multistepper pipette when pipetting slowly.

Note: All incubation steps below are advised to occur on a shaker. However, if there is none available, it is also possible, but not advised, to do without shaking.

Note: For visualization of the cell morphology, the plasma membrane can be optionally counterstained. Use the dyes prior to fixation, as any fixation procedure alters the integrity of the membrane and/or dye. Follow the dye manufacturers protocol (e.g. MemBrite; https://biotium.com/wp-content/uploads/2018/05/PI-MemBrite-Fix.pdf or CellMask; https://documents.thermofisher.com/TFS-Assets%2FLSG%2Fmanuals%2FCellMask_Plasma_Membrane_Stains_PI.pdf).

• 3.
  Fixation, permeabilization and background blocking of the HeLa cells on the coverslips.

  • a.
    Aspirate supernatant, then carefully wash the cells once with 0.5 mL cold DPBS per well.
  • b.
    Aspirate DPBS and incubate the cells with 0.5 mL fixation buffer (3.7% PFA in DPBS) per well for 10 min at 22°C.
  • c.
    Remove the fixation buffer.

    • i.
      Discard PFA accordingly to the safety standards.
    • ii.
      Wash the cells three times by adding 0.5 mL cold DPBS per well.
    • iii.
      Incubate for 5 min at 22°C.
    • iv.
      Aspirate the liquid.
  • d.
    Permeabilize the cells using 0.5 mL permeabilization buffer (0.5% Triton-X-100 in DPBS) per well for 7 min at 22°C.
  • e.
    Remove the permeabilization buffer.

    • i.
      Wash the cells three times by adding 0.5 mL cold DPBS per well.
    • ii.
      Incubate for 5 min at 22°C.
    • iii.
      Aspirate the liquid.
  • f.
    Incubate the cells with 0.5 mL blocking buffer (5% BSA in DPBS) per well for 1 h at 22°C.

Primary antibody staining

Timing: 16 h

During this step primary antibodies bind to the endogenous protein and the tag of the overexpressed viral proteins of interest within the HeLa cells.

CRITICAL: Choose primary antibodies raised in different species for the two target proteins (see section before you begin).

CRITICAL: If only limited amounts of the primary antibodies are available use a humid chamber for the staining process. To make your own humid chamber, prepare the following components in a container with a lid (bottom to top): Wet tissues which cover the whole area, one layer of parafilm, drops of 20 μL of the primary antibody solution, each coverslip with cells facing to one primary antibody drop (coverslip has to be turned upside down). It is recommended to make labels on the parafilm next to each primary antibody drop to differentiate between the coverslips and to ensure a distance of at least 2.5 cm between the drops. The container may be sealed with the lid.

CRITICAL: Keep the cover slip face down when using staining drops, face up in the 24-well plate to not damage the cells.

• 4.
  Primary antibody staining of the endogenous proteins of interest.

  • a.
    Prepare 200 μL of the antibody solution.

    • i.
      Dilute the primary antibody for the endogenous protein in blocking buffer (5% BSA in DPBS). Use standard antibody concentrations (1:100 for endogenous proteins) or custom dilution according to previous experiments, see Section before you begin.

      Note: Prepare 10% extra volume of the primary antibody solution to ensure that there is sufficient antibody solution for all wells.

  • b.
    Remove the blocking buffer from the 24-wells and add 200 μL of the primary antibody solution per well.
  • c.
    Incubate the cells with the primary antibody solution at 4°C on a shaker o.n.

    CRITICAL: Double check to ensure the cover slip is completely covered in sufficient amount of liquid, use the liquid drop staining in the humid chamber method alternatively if this cannot be ensured (see above).

• 5.
  Primary antibody staining of the overexpressed protein of interest.

  • a.
    Prepare a 10x solution of 20 μL of primary antibody diluted in blocking buffer (5% BSA in DPBS) per well. Use standard antibody concentrations (final concentration 1:400 for overexpressed proteins) or custom dilution according to previous experiments, see Section before you begin.

    Note: Prepare 10% extra volume of the primary antibody solution to ensure that there is sufficient antibody solution for all wells.

  • b.
    Add 20 μL of the 10x antibody solution for the overexpressed protein on top of the antibody solution staining the endogenous protein, so that the final dilution of the antibody staining the overexpressed protein is 1:400 or the custom dilution in the well.
  • c.
    Incubate for 1.5 h at 37°C.
  • d.
    Wash the cells three times by adding 0.5 mL cold DPBS per well for 5 min each and removing the liquid afterwards by pipetting.

    Note: To prevent drying out, keep the cells in 0.5 mL cold DPBS until the next step is prepared.

PLA probe staining and ligation

Timing: 2.5 h

In this step, species-specific secondary antibodies conjugated to oligonucleotides, referred to as PLA probes, bind to the primary antibodies that are attached to the target proteins. If the target proteins in close proximity (within 40 nm) the oligonucleotides of each primer probe hybridize and form a circular DNA molecule.

CRITICAL: Use a tweezer to transport each coverslip from the 24-well plate to the humid chamber. It is recommended to only touch the edge of the coverslip by the tweezer to prevent interruptions of the cell layer. Pay attention which side of the coverslips contain the cells. In the plate, cells are seeded on the top of the coverslip. In the humid chamber the coverslip has to be turned around, so the cells directly touch the droplet.

• 6.
  PLA probe staining.

  • a.
    Prepare 20 μL PLA probes working solution per coverslip as described in Table 1.
  • b.
    In a humid chamber (preparation see Primary antibody staining section) incubate the coverslips with the PLA probes working solution for 1 h at 37°C in a > 90% humified incubator.

    Note: We recommend a distance of at least 2.5 cm between the drops on the parafilm and a labeling besides the drops to ensure the correct identification of the coverslips.

    CRITICAL: The cover slips need to face down on the stain drop.

  • c.
    Return the coverslips back to the wells of the 24-well plate with the cells facing up again.
  • d.
    Wash the wells twice with 0.5 mL PLA wash buffer A for 2 min at 22°C under gentle shaking.
• 7.
  Ligation of the DNA oligonucleotides.

  • a.
    Prepare 20 μL Ligation working solution per coverslip as described in Table 2.
  • b.
    Pipette 20 μL drops of the Ligation working solution on a fresh parafilm in the humid chamber and add the coverslips with cells facing to the drops.
  • c.
    Incubate the coverslips in the humid chamber for 30 min at 37°C in a >90% humified incubator.
  • d.
    Return the coverslips back to the wells of the 24-well plate with the cells facing up again.
  • e.
    Wash the wells twice with 0.5 mL PLA wash buffer A for 2 min at 22°C under gentle shaking.

CRITICAL: Proceed immediately with the next step.

Table 1.

PLA probes working solution

Reagent	Amount
PLA probe Anti-mouse MINUS stock solution	4 μL
PLA probe Anti-rabbit PLUS stock solution	4 μL
Blocking buffer	12 μL
Total	20 μL

Prepare fresh before use. Alternatively, the combination of PLA probe anti-rabbit MINUS stock solution and PLA probe anti-mouse PLUS stock solution can be used.

Table 2.

Ligation working solution

Reagent	Final concentration	Amount
5x Ligation stock solution	1x	4 μL
Ligase	1 U/μL	0.5 μL
UltraPure distilled water	1x	15.5 μL
Total	N/A	20 μL

Prepare fresh before use. The 5x ligation stock solution and ligase are part of the Duolink In Situ Detection Reagents FarRed kit. Keep the ligase on ice.

DNA amplification, overexpression staining, and finalization

Timing: 5 h

In this step the oligonucleotides on the PLA primer probes are amplified. The amplified DNA is stained with a fluorescent probe for detection in the far-red wavelength spectrum. The optional staining of the overexpressed protein will highlight the cells expressing the viral protein. The coverslips will be mounted and simultaneously the nucleus counterstained with DAPI.

CRITICAL: Use from here on dim light for working to prevent degradation of the fluorescent signals.

• 8.
  Amplification of the circular DNA molecules.

  • a.
    Prepare 20 μL Amplification working solution per coverslip as described in Table 3.
  • b.
    Pipette 20 μL drops of the ligation working solution on a fresh parafilm in the humid chamber and add the coverslips with cells facing to the drops.
  • c.
    Incubate the coverslips in the human chamber for 1 h 40 min at 37°C in a > 90% humified incubator.
  • d.
    Return the coverslips back to the wells of the 24-well plate with the cells facing up.
  • e.
    Wash the wells twice with 0.5 mL PLA wash buffer B for 10 min each at RT under gentle shaking.
• 9.
  Optional: Staining of the overexpressed protein of interest.

  • a.
    Wash the wells twice with 0.5 mL DPBS for 2 min at 22°C under gentle shaking.
  • b.
    Prepare a pre-conjugated antibody solution to stain the overexpressed protein within the cell by using the Zenon labeling kit according to the manufacturer’s instruction (https://www.thermofisher.com/de/de/home/references/protocols/proteins-expression-isolation-and-analysis/antibody-protocol/mouse-igg-labeling.html).

    Note: The Zenon labeling kit is available for diverse species and fluorophores, e.g. mouse, rabbit or human.

  • c.
    Add 140–150μL pre-conjugated antibody solution per well and incubate it for 1.5 h at 37°C in the 24-well plate covered in aluminum foil OR incubate the coverslips on a 20 μL droplet of the pre-conjugated antibody solution in the humid chamber for 1.5 h at 37°C.
  • d.
    Wash the wells twice with 0.5 mL DPBS for 2 min at 22°C and afterwards once with 0.05 mL PLA wash buffer B for 10 min at 22°C under gentle shaking.

    Note: We recommend co-staining the overexpression of the viral protein, to quantify successfully transfected cells only. However, if the overexpression of the viral protein and its variants was normalized before (see in the section before you begin), it is also possible to leave this step out.

• 10.
  Finalization and mounting of the cover slip.

  • a.
    Aspirate wash buffer B and add 0.5 mL 0.01x PLA wash buffer B shortly before the next step.
  • b.
    Put the coverslip with the cells facing up on a delicate task wipe to dry the coverslip.
  • c.
    Add a drop of Duolink In Situ Mounting Medium with DAPI on an object slide and carefully mount the coverslip with the cells showing downwards to the mounting medium drop.

    Note: Hold the coverslip at the edge with the tweezer and try to avoid bubbles during the transfer on the drop of mounting medium with DAPI. We recommend to lay the coverslip on the edge of the drop down and slowly lower the rest of the coverslip.

  • d.
    Seal the edges of the coverslip on the object slide with transparent nail polish.
  • e.
    Store the slides in the dark at 4°C (or −20°C for long-term storage) until use.

    Note: It is recommended to image the object slides within the following 2 days after finalization.

Table 3.

Amplification working solution

Reagent	Final concentration	Amount
5x Amplification far red stock solution	1x	4 μL
Polymerase	10 U/μL	0.5 μL
UltraPure distilled water	1x	15.5 μL
Total	N/A	20 μL

Prepare fresh before use. The 5x amplification Far Red stock solution and polymerase are part of the Duolink In Situ Detection Reagents FarRed kit. Keep the polymerase on ice.

Imaging the PLA slides

Timing: 1–6 h

This section describes the imaging of the fixed and mounted cells on object slides using laser scanning confocal microscope. For best results we recommend using a 63x objective with immersion oil and to perform a tile scan together with imaging in z-stacks.

CRITICAL: Take images of at least 50–100 individual cells or 15–20 image tiles (depending on the cell density). All images used for quantification of samples and controls should be taken on the same day to ensure similar exposure.

• 11.
  Preparing the microscope.

  • a.
    Set-up the laser intensity and gain so PLA dots are visible but not oversaturated. This also accounts for the DAPI channel as well as the optionally co-stained overexpressed viral protein.

    Note: Image different parts of the cover slips, not only one specific region. In general, the cells should not be too dense at the imaging positions.

  • b.
    Set-up z-stacks of the parts of the cover slips that are to be imaged of 0.3–0.5 μm z-stacks with 4-5 z-stacks in total (Figure 3).

    Note: Imaging in z-stacks ensures visualization of all PLA dots independent on their localization in the vertical plane. Z-stack projections shows all PLA dots in one 2D image (Figure 3).

  • c.
    If your microscope supports the tile scan option, set this up alongside the z-stack to increase time efficiency of acquiring the images.
  • d.
    Save the images in the original microscope file and proceed to the image analysis.

Figure 3.

Importance of imaging in z-stacks for PLA analysis

The intensity and visibility of the PLA dots can be dependent on the imaged z-stack. Z-stack projections of combined z-stack slices will enhance the visibility of the PLAT dots. PLA, red; DAPI blue, Scale Bar: 20 μm.

Semi-automated quantification of PLA dots per individual cell using ImageJ

Timing: 1–6 h

To assess the number of PLA dots per sample, ImageJ/Fiji analysis of the immunofluorescence images can be used. In this step, the images are loaded into ImageJ together with ImageJ Macros to simplify the image analysis, which will combine Z-stacks, set thresholds, despeckle images and count PLA dots Figures 4 and 5).

• 12.
  Load images into ImageJ.

  • a.
    Drag and drop the output file from your LSM into ImageJ/Fiji.^19,20 This will open a window called Bio-Formats Import Options. Select View stack with ‘Hyperstack’, Dataset organization ‘Open files individually’ and Color mode ‘Composite’. Press OK. Then select import all images.

Note: Use the files directly from microscope software (e.g. Zeiss Zen software) to retain all information including the meta information.

• 13.
  Combine the individual Z-stacks to an overlay images, split and close some channels.

  • a.
    Open Plugins > Macros > Startup Macros. The Macro Manager Window will open.
  • b.
    Copy the Code below in the Macro window.

    // Get list of all open images

    titles = getList("image.titles");

    // Process each tile one by one

    for (i = 0; i < titles.length; i++) {

     selectWindow(titles[i]);

     title = titles[i];

     print("Projecting: " + title);

     // Check number of slices (to ensure it’s a stack)

     slices = nSlices;

     if (slices > 1) {

      // Create Max Intensity projection

      run("Z Project...", "projection=[Max Intensity]");

      // Rename the projected image for clarity

      rename("MAX_" + title);

     } else {

      // If there’s only 1 slice, just duplicate

      run("Duplicate...", "title=MAX_" + title);

     }

     // Optional: Close the original stack to save memory

     close(title);

    }
  • c.
    Press Run.

    Note: This command will combine the z-stacks of into one overlay image for each tile individually.

  • d.
    Load the macro below into a new macro window. In the example here DAPI is channel 3 and Cell Mask is channel 4. Therefore, C3 and C4 are closed within the command close (). In this example I use channel 1, which is the staining of the overexpression as guidelines for drawing the ROIs later.

    Note: This macro will split the channels into individual windows and closes all channels that are not necessary for PLA dot quantification, e.g. the DAPI channel.

    titles = getList("image.titles");

    for (i = 0; i < titles.length; i++) {

     selectWindow(titles[i]);

     title = titles[i];

     print("Splitting channels for: " + title);

     // Split the 4 channels

     run("Split Channels");

     // Define channel names based on standard Fiji naming

     c1 = "C1-" + title;

     c2 = "C2-" + title;

     c3 = "C3-" + title; // DAPI

     c4 = "C4-" + title; // CellMask

     // Close DAPI (C3)

     if (isOpen(c3)) {

     close(c3);

     print("Closed DAPI channel: " + c3);

     }

     // Close CellMask (C4)

     if (isOpen(c4)) {

     close(c4);

     print("Closed CellMask channel: " + c4);

     }

     // Optional: close the merged image to save memory

     if (isOpen(title)) {

     close(title);

     }

    }
  • e.
    Press Run.

    Note: Depending on which channel is selected for proper drawing of ROIs, do not close this channel during this step. Modify the macro, so this channel is not closed.

• 14.
  Set Threshold and despeckle images.

  Note: To analyze all samples uniformly, one threshold needs to be set and used for all images (Figure 4). Use one image e.g. positive control, that shows PLA dots to set the threshold.

  • a.
    Open threshold window by choosing Image > Adjust > Threshold…
  • b.
    Load the macro below into the Macro Window.
  • c.
    Add the values from your test image into the code: minThresh (e.g., 6) and maxThresh (e.g., 255).
  • d.
    Ensure that the PLA channel is marked as the one in which the threshold is changed. Here channel 2 (c2) is the PLA channel.

    minThresh = 5;

    maxThresh = 255;

    titles = getList("image.titles");

    for (i = 0; i < titles.length; i++) {

     selectWindow(titles[i]);

     title = titles[i];

     if (startsWith(title, "C2-")) {

     print("Applying manual threshold to: " + title);

     setThreshold(minThresh, maxThresh);

     setOption("BlackBackground", true);

     run("Convert to Mask");

     }

    }
  • e.
    Press Run.

    Note: This command will set the threshold in all PLA images and these images are turned into binary images.

  • f.
    To remove any unspecific individual pixels, which are significantly smaller than the PLA dots, the images are despeckled to visually see the PLA dots better (Figure 4). For this, copy the code below into a new macro window.

    Note: Ensure that the PLA channel is marked as the one which is despeckled. Here channel 2 (c2) is the PLA channel.

    titles = getList("image.titles");

    for (i = 0; i < titles.length; i++) {

     selectWindow(titles[i]);

     title = titles[i];

     // Only apply to Channel 2 images

     if (startsWith(title, "C2-")) {

     print("Processing Channel 2: " + title);

     run("Despeckle");

     }

    }
  • g.
    Press Run – the images are despeckled and (ideally) only PLA dots are left (Figure 4).

    CRITICAL: It is important to set the same threshold for all control and sample images, to ensure that all images are treated and normalized the same way.

• 15.
  Drawing Region of Interests (ROIs) and count PLA dots within.

  • a.
    This part has to be done tile by tile or image by image.
  • b.
    Open the ROI manager: Analyze > Tools > ROI Manager…
  • c.
    Pick one tile and the channel in which the cell membrane or the cell boarders are clearly visible.

    Note: To precisely circle cells it is recommended to either use the channel used to define the overexpression efficiency, or phase contrast or a cell membrane stain.

  • d.
    Circle one cell using the Freehand selection tool and add to the ROI manager by pressing ‘t’ or clicking on add in the ROI manager (Figure 5).
  • e.
    Select the corresponding PLA image.
  • f.
    Copy the code below into a new macro window.
  • g.
    Run the macro.

    Note: This macro will open a summary window. The ‘count’ section indicates the number of PLA dots (Figure 5; red box).

    // Clear the main Results table only (the final output)

    run("Clear Results");

    // Number of ROIs

    n = roiManager("count");

    for (i = 0; i < n; i++) {

     // Select ROI

     roiManager("select", i);

     // Run Analyze Particles for this ROI and APPEND summary output

     // NOTE: remove "clear" so the Summary table does not reset

     run("Analyze Particles...", "size=0-Infinity summarize");

     // After Analyze Particles, the Summary table now has a new row.

     // You can label the row by adding the ROI number to Results.

     setResult("ROI", i, i + 1);

     updateResults();

    }
  • h.
    Repeat step 5 for all tiles of one sample. Before you start a new tile, the ROIs from the image before need to be deleted from the ROI manager.

    Note: Using this macro will append the results from the particle count of each tile to the summary table. Transfer the count (Figure 5) to Excel or GraphPad Prism to plot the data and perform statistical analysis.

Figure 4.

Preparation of PLA images for PLA dot quantification wit ImageJ – Threshold and despeckling

Using an exemplary PLA picture to set a threshold. This threshold is applied to all images to generate binary files. These images are then despeckled to remove any unspecific pixels left. red; PLA dots; white, PLA dot images as binary files. Scale bar: 20 μm.

Figure 5.

Quantification of PLA dots per cell

ROIs are defined using images of the co-staining of the overexpressed protein/-tag and saved in the ROI manager in ImageJ. These ROIs are applied to the binary PLA files and within these ROIs the particles are automatically counted using the analyze particle function of ImageJ. The count in the summary table (red box) indicates the number of PLA dots per ROI. Green, StrepII-tag-AF488; white, PLA dot images as binary files. Scale bar: 20 μm.

Batch/bulk analysis of PLA using Imaris

Timing: 15–30 min

This section describes the batch analysis of PLA signals using the 3D image analysis software Imaris. PLA dots are quantified using the spots model. To normalize on the number of cells, nuclear surfaces are generated, providing the number of nuclei per image. The number of PLA dots per image is divided by the number of nuclei per image.

Note: This quantification alternative counts dots in one image or tile, and does not normalize on transfection efficiency. It is, however, faster and more convenient as well as accurate if a high and uniform transfection efficiency is achieved.

• 16.
  Image segmentation of nuclei:

  • a.
    Import image into the Imaris Arena (Figure 6A, red box), select all images and right-click to “Convert to Native Imaris File Format” (Figure 6A, red box).
  • b.
    Double-click an image to open the Surpass 3D View and click on the surface icon on the upper left (Figure 6B, red box) to create nuclear surfaces.
  • c.
    Use the following creation parameters (see below) to create nuclear surfaces.

    • i.
      DAPI is used as the source channel.
    • ii.
      Values for Thresholds and Filters for Seed Points and Surfaces must be optimized for each dataset.
    • iii.
      Thresholds and filters should be defined based on the intensity histogram and through visual comparison of the segmentation results with the raw image.

      [Algorithm]

       Enable Region Of Interest = false

       Enable Region Growing = true

       Enable Tracking = false

       Enable Classify = false

       Enable Shortest Distance = false

      [Segmentation Setup]

       Source Channel Index = 3

       Enable Smooth = true

       Surface Grain Size = 0,132 μm

       Enable Eliminate Background = false

      [Threshold]

       Active Threshold = true

       Enable Automatic Threshold = false

       Manual Threshold Value = 1.36465

       Active Threshold B = false

       Region Growing Estimated Diameter = 4,00 μm

       Region Growing Morphological Split = true

      [Filter Seed Points]

       "Quality" above 0,573

      [Filter Surfaces]

       "Number of Voxels Img=1" above 8,95e4
  • d.
    After generating the nuclear surfaces, open the creation tab and click “Store Parameters for Batch” (Figure 7, red box). Select Arena and Favorite Creation Parameters.
• 17.
  Spots model for PLA dots.

  • a.
    In the Surpass 3D View, click the spots icon.
  • b.
    Use the following parameters.

    • i.
      Select the PLA channel as the source channel.
    • ii.
      Switch to slice view and use the line measurement tool (right side of the window) to estimate spot diameter.
    • iii.
      XY Diameter and Filters for Spots must be optimized for each dataset.
    • iv.
      Thresholds and filters should be defined based on the intensity histogram and through visual comparison of the segmentation results with the raw image.

      [Algorithm]

       Enable Region Of Interest = false

       Enable Region Growing = false

       Enable Tracking = false

       Enable Classify = false

       Enable Region Growing = false

       Enable Shortest Distance = false

      [Source Channel]

       Source Channel Index = 2

       Estimated Diameter = 0,658 μm

       Background Subtraction = false

      [Filter Spots]

       "Quality" above 7,70
  • c.
    After spot creation, open the creation tab and click “Store Parameters for Batch”.

    • i.
      Select Arena and Favorite Creation Parameters.
• 18.
  Batch processing of nuclear surfaces and spots.

  • a.
    Return to Arena. Right-click the surfaces icon and select “Run Batch” (Figure 8A, red box).
  • b.
    After Imaris completes batch processing for nuclear surfaces, repeat “Run Batch” for the spots model.
  • c.
    Save all data for surfaces and spots from the Folder Batch History (Figure 8B, red box) as an Excel file.

    Note: The counts of nuclei and PLA spots for each are located in the “Overall” tab of the resulting file.

  • d.
    For each image, divide the number of PLA spots by the number of nuclei.

Figure 6.

Image segmentation of nuclei in Imaris

(A) Import images into Imaris and click convert into Imaris File Format (red box).

(B) Click on the surface icon (red box) in the surpass window to start creation of a nuclear surface.

Figure 7.

Store surface parameters for batch processing

After generation of the surface based on an exemplary image, click on the magic wand icon to store parameters for batch (red box). Select Arena and Favorite Creation Parameters and click OK.

Figure 8.

Batch analysis of PLA dots and nuclei

(A) Generated nuclear surface and spots model are applied to all images by right-click on the respective icon and click on run batch (red box).

(B) All results generated based on spots model and surface can be exported as an Excel sheet from the Folder Batch History.

Expected outcomes

The expected outcome are immunofluorescence images (Figure 9A) summarized into a scatter plot (Figure 9B) showing the quantitative impact of the variance in close proximity of an overexpressed (viral) protein with an endogenous protein in comparison to the close proximity of variants of the overexpressed protein with an endogenous protein.¹ The data shows that the envelope (E) protein derived from the SARS-CoV-2 variant of concern Omicron (E T9I) displays significantly more PLA dots with ABCG2 than the E protein derived from an early 2020 SARS-CoV-2 strain. This indicates stronger in situ binding of E T9I to ABCG2.

Quantification and statistical analysis

PLA dots were counted as described above, counting each cell individually using ImageJ/Fiji (Figure 10A) or performed bulk analysis using the Imaris software (Figure 10B). Statistical analyses were performed using GraphPad PRISM 10 (GraphPad Software). Outliers were identified using the ROUT method (Q=1%). Statistical analyses were done using student’s t-tests with Welch’s correction. Data is shown as either amount of PLA dots in individual cells or per tiles/cells ±SEM with lines representing the mean.

Figure 10.

Comparison of PLA quantification by ImageJ and Imaris

(A) Quantification of the images shown in Figure 9A using semi-automated counting in ImageJ. Only cells positively staining for StrepII-tag overexpression were included in the PLA dot quantification of cells transfected with constructs encoding for SARS-CoV-2 E T9 or E T9I. Lines represent the mean of N = 19-70 (individual cells) ± SEM. Cells in the vector control were not gated for StrepII expression.

(B) Quantification of the images shown in Figure 9A using bulk analysis with Imaris. PLA dots and DAPI-stained nuclei were automatically counted. Lines represent the mean of N = 19–24 (individual tiles) ± SEM after Outlier identification using ROUT method in GraphPad Prism. Student’s t test with Welch’s correction. ∗∗∗, p < 0.001.

Limitations

While certainly a powerful method, the proximity ligation assay as described here certainly has limitation. A major issue is the availability and suitability of the antibodies against the cellular interaction partners of choice. This problem could be overcome by overexpressing tagged construct, which may however also come with the additional limitation that overexpressed cellular proteins may show non-endogenous localizations, thus occluding spatial analyses. As opposed to in vitro binding assays, PLA does not yield any absolute quantification of binding strength thus making it almost impossible to compare samples that do not use the same set of antibodies. Furthermore, PLA does not provide information of the exact proximity or direct vs indirect nature of the interaction. It just gauges that two proteins are within the range of 40 nm which could very well be a complex. On the other hand, complexes which are too large (i.e., >40 nm) may not be detected in PLA, although an interaction may be indicated by co-immunoprecipitation. As for most biological assays, the choice of cell system is crucial. While primary cells are often preferred, the assay works similarly in cell lines. However, it needs to be considered that the localization of cellular proteins may differ in different cell types/tissues and that some interaction partners may not be sufficiently expressed.

Finally, from a technical point of view, it is a limitation that we suggest to analyze the samples as fast as possible after fixation, to not lose or diffuse the spatial information. In addition, this method can certainly be considered a low-throughput method, and only a limited number of samples may be processed at the same time to avoid handling mistakes and ensure the quality of the samples.

Troubleshooting

Problem 1

The images show uncommonly high PLA dots in the vector control and outside of the cells (e.g., >3 PLA dots in vector transfected cells) (see Imaging the PLA slides).

Potential solution

• •This effect might be due to varying freeze-thaw cycles of the Duolink PLA reagents (related to PLA probe staining and ligation). We observed higher background levels when using ‘fresh’ PLA reagents (Figure 11A). When comparing single antibody controls and the actual PLA, ensure you use the identical PLA kit that underwent the same amount of freeze-thaw cycles. If the PLA dot quantification in the samples is higher than in the controls using identical conditions, the experiment may be used.
• •Increase the last washing step with washing buffer B after the amplification (see section DNA amplification, overexpression staining and finalization). Usually the longer this washing step, the less background the PLA images will have.
• •Lowering the antibody concentration will result in less unspecific PLA dots (related to Antibody dilution). Similar to titration of the antibody in classical immunofluorescence staining of the overexpressed protein, it is also possible to perform this antibody titration directly with the PLA itself (Figure 11B).

Figure 11.

Potential background noise of PLA dots

(A) The amount of freeze-thaw cycles of the PLA kit influences the number of PLA dots. The more often the kit was thawed, the weaker the PLA will work, which also means that the background noise in the vector transfected cells will reduce. Red, PLA; blue, DAPI. Scale bar: 20 μm.

(B) Dilution series of anti-StrepII antibody influences the amount of PLA dots, and can prevent increased background staining. Red, PLA; blue, DAPI. Scale bar: 10 μm.

Problem 2

The viral wildtype (WT) protein and the variants show different expression efficiencies (related to Optimization of transfection).

Potential solution

• •Titrate the different expression plasmids and establish transfection efficiencies for each using immunofluorescence analysis. Adjust the amount of plasmid transfected to achieve the same expression efficiency for all constructs used.
• •Ensure the WT protein and the variants are all in the same expression vector. Using different expression vectors can influence the transfection efficiency.
• •Co-stain the PLA samples for expression and quantify the expression level per cell. Normalize the number of PLA dots to the expression level. However, this method will only provide information on the difference between WT protein and variants, and lacks the comparison to the vector control sample, which indicates specificity. In addition, the fluorescence staining may not be linear.

Problem 3

The PLA dots are very small and barely visible (see Imaging the PLA slides).

Potential solution

• •When taking images using the LSM it is highly recommended to do z-stacks (see section Imaging the PLA slides). This usually ensures that the PLA dots have a certain size and it will capture PLA dots in multiple images overlaying their signal.
• •Ensure that the Duolink® PLA kit is not expired and the protocol was followed through accurately (related to PLA probe staining and ligation and DNA amplification, overexpression staining and finalization).
• •We observed less efficient PLA when the cell monolayer is too dense (related to Cell seeding and transfection). Ensure the cells have an ideal confluency of 50%–60% at the time of transfection and not more than 80% at the time of fixation.

Problem 4

No PLA dots are visible or the cells are full of PLA dots (see Imaging the PLA slides).

Potential solution

• •Ensure that the Duolink PLA kit is not expired and the protocol was followed through accurately (steps of PLA probe staining and ligation and DNA amplification, overexpression staining and finalization).
• •Check the staining by the individual antibodies in single staining experiments to ensure that the proteins are properly recognized and that the antibody does not result in high background staining (see section Assessment of antibody and PLA specificity).
• •Check the expression levels of the overexpressed viral protein using e.g., Western Blotting (see section Optimization of transfection).

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Konstantin Maria Johannes Sparrer (konstantin.sparrer@dzne.de).

Technical contact

Technical questions on executing this protocol should be directed to and will be answered by the technical contact, Meta Volcic (meta.volcic@uni-ulm.de).

Materials availability

This study did not generate new unique reagents.

Data and code availability

• •Original/source data for Figures 2, 3, 4, 5, 6, 7, 8, 9, 10, and 11 are available from the corresponding author on request.
• •The published article includes all code generated or analyzed during this study.

Acknowledgments

We thank Jana-Romana Fischer and Birgit Ott for excellent technical assistance. K.M.J.S. acknowledges funding by the Federal Ministry of Research, Technology and Space of Germany (BMFTR; IMMUNOMOD-01KI2014) and the German Research Foundation (DFG; CRC1279, SP 1600/7-1, SP 1600/9-1, SP 1600/13-1). M.V. is supported by the German Research Foundation (DFG; KI 548/21-1). H.H. and Z.E. are part of the International Graduate School for Molecular Medicine, Ulm (IGradU). Z.E. is supported by a scholarship of the Friedrich Ebert Stiftung. We thank the CRC1506 (Prof. Hartmut Geiger, Ulm University Medical Center) for access to the Imaris software.

Author contributions

H.H. performed the experimental work. H.H. and Z.E. performed image analysis. H.H., S.K., and Z.E. arranged the figures and provided comments to the manuscript. K.M.J.S. wrote the initial draft and conceived the study. H.H., S.K., Z.E., M.V., and K.M.J.S. planned the experiments and wrote the manuscript. All authors reviewed and approved the manuscript.

Declaration of interests

The authors declare no competing interests.