Probe Design: Sequences and Templates
1) Label probes (detection probes conjugated to fluorophores)
Label probe / Sequence (5’--3’)DP5- Cy5 / ATGCTGCTGCTGTACTACG
Y1- A488 / AGAGGCACTTGAGCTAGAAT
X1- Cy3 / GGGAGACTGATTACACGATA
X4 - Texas red / AGAACCTCCATTAACATCGC
DP4- Pacific blue 405 / TGCGTCTATTTAGTGGAGCC
● Order Label probes from IDT as 100 nmole oligos with 5’ conjugation of fluorophore and HPLC purification.
● Use Cy5 for the most important target gene. If you use Cy3 and Texas red readouts in the same sample, image optimally to avoid bleedthrough (e.g., tunable collection ranges, deconvolution).
● Pacific blue conjugated oligos are not available through IDT. Buy Pacific Blue 405 NHS ester (Thermo Fisher P10163) and 5’ amino modified oligo (DP4) from IDT. Set up conjugation reaction. In our experience, PB-405 signals are weaker and a stock may lose its potency over time compared to Alexa and Cy dye signals (so use PB-405 for your highest abundance target).
2) Circle components:
CC- Connector Circle; CB - Common Bridge (Use the same bridge for all CC's)
CB2-31AC (this is the Common bridge) / TCAACTCGACGTATAACATAACGACGTAAGTCC / Sequence
CC05-DP5 (works with Cy5) / TGTTAGCGCTAACAAAATGCTGCTGCTGTACTACGAACAACAATACACATGTTACGACGT
CC02-Y1 (works with A488) / TAGACGTACGAGATTAAAACATAGAGGCACTTGAGCTAGAATCATATAACAGGTTGTTCT
CC03-X1 (works with Cy3) / TCG AAC GAA TAA CAA AGG GAG ACT GAT TAC ACG ATA ACA ACA ATA CAC ACA ATA CGT CGG
CC04-X4 (works with Texas Red) / TCGCCTACTAAACAAAGAACCTCCATTAACATCGCAACAACAATACACACGCACTTATGT
CC6-DP4 ( works with DP4-PB405) / TTCCTGACCTAACAAAACATGCGTCTATTTAGTGGAGCCACATAATTATAACCTGGCTAT
Order CC and CB from IDT as 100 nmole oligos with HPLC purification and phosphorylate them by setting up a Polynucleotide Kinase (PNK) reaction. (This is essential for ligating CC and CB to form circles). Alternatively, you can order phosphorylated oligos and skip having to perform the PNK reaction.
PNK reaction: (Have this completed before you start ISH)
T4 PNK reaction to phosphorylate DNA oligos (CC and CB)
1X / Component5.0 uL / 10X T4 PNK buffer
5.0 uL / 10mM ATP
1.0 uL / T4 PNK enzyme (NEB- M0201)
15.0 uL / 20uM DNA oligo (CC or CB)
24.0 uL / DEPC-treated H2O
Incubate at 37C for 30 min, 65C for 20 min, then hold at 4C for 5 minutes.
Add 2.0ul 0.5M EDTA to each reaction and store at -20C
3) Hybridization Probe (H probe) Design
● Use NCBI nucleotide (https://www.ncbi.nlm.nih.gov/nucleotide/) to locate the mRNA sequence of the gene of interest. Copy and paste the entire sequence or region of interest into a Word document.
● Use Primer3 ( to find candidate H probes. Copy and paste the sequence into the search window of Primer3, select “pick hybridization probe” command (de-select the options to pick left and right primers), set the number of probes to return (normally no less than 100).
● Once the list of possible probes has been acquired, go through the list and find a probe that ends with either AG or TA. Go back to the Word document and locate the probe sequence. Pick a 40 bp sequence that includes 19 bases upstream AG or TA and 19 bases downstream AG [upstream seq (19 bases) +AG or TA (2 bases) + downstream stream seq (19 bases) = total 40 bases]. Each H probe is 40bp long, with AG (or TA) being in bp positions 20 and 21.
● Use or similar webpages to acquire the reverse complement of the probes. Paste the retrieved sequence into an Excel sheet.
● BLAST sequences to ensure that they don’t share significant complementarity to other (non-target) genes.
● Separate the sequence into 2 halves (the first 20bp is the right H probe and the last 20bp is the left H probe).
● Use Oligo Calculator (http://biotools.nubic.northwestern.edu/OligoCalc.html) to check that the Tm of each probe is within the range of 45 ̊ C to 65 ̊ C. Also check for and exclude probes with potential hairpin formations, self-annealing, and 3’ complementarity.
● If a probe’s Tm is within the optimal range and there are no potential structural issues, the left and the right H probe sequences can be added to a suitable template (refer table below) that works with the desired fluorophore.
*When designing probes for multiple genes to be used on the same sample, it is imperative to decide the color of the fluorophore for each in advance to ensure they are in distinct channels.
Template for designing:
Fluorphore / Compatible CC / Design Templates for H Probes (2X, 3X, 4X, 5X, 6X) - / Design Templates for H Probes (2X, 3X, 4X, 5X, 6X) -For Cy5 / CC05-DP5 + CB2-31AC / HL5X-Gene-Position / TAGCGCTAACAACTTACGTCGTTATG:LLLLL
HR5X-Gene-Position / RRRRR:TTATACGTCGAGTTGAACGTCGTAACA
For A488 / CC2-Y1 + CB2-31AC / HL2X-Gene-Position / TCGTACGTCTAACTTACGTCGTTATG:LLLLL
HR2X-Gene-Position / RRRRR:TTATACGTCGAGTTGAAGAACAACCTG
For Cy3 / CC03-X1 + CB2-31 AC / HL3X-Gene-Position / TATTCGTTCGAACTTACGTCGTTATG:LLLLL
HR3X-Gene-Position / RRRRR:TTATACGTCGAGTTGACCGACGTATTG
For Texas Red / CC04-X4 + CB2-31AC / HL4X-Gene-Position / TTAGTAGGCGAACTTACGTCGTTATG:LLLLL
HR4X-Gene-Position / RRRRR:TTATACGTCGAGTTGAACATAAGTGCG
For PB405 / CC06-DP4 + CB2-31AC / HL6X-Gene-Position / TAGGTCAGGAAACTTACGTCGTTATG:LLLLL
HR6X-Gene-Position / RRRRR:TTATACGTCGAGTTGAATAGCCAGGTT
Example for designing H probes:
Selection Criteria for H Probes: Target Tm = 60C, Ideal Length = 20, Junction = AG or TA, Uniqueness of Probe Sequences
Mouse Fgf10 (NCBI reference- NM_008002.4)
fwd seq FGF10 - start position : 965 / GAGAAGAACGGCAAGGTCAGCGGGACCAAGAATGAAGACcorresponding rev complementary / GTCTTCATTCTTGGTCCCGCTGACCTTGCCGTTCTTCTC
For Cy5 / probe sequence (mRNA binding region) / Complete H probe sequence
mmHL5X-FGF10-965 / TGACCTTGCCGTTCTTCTC / TAGCGCTAACAACTTACGTCGTTATG: TGACCTTGCCGTTCTTCTC
mmHR 5X-FGF10-965 / GTCTTCATTCTTGGTCCCGC / GTCTTCATTCTTGGTCCCGC:TTATACGTCGAGTTGAACGTCGTAACA
For A488 / probe sequence (mRNA binding region) / Complete H probe sequence
mmHL2X-FGF10-965 / TGACCTTGCCGTTCTTCTC / TCGTACGTCTAACTTACGTCGTTATG:TGACCTTGCCGTTCTTCTC
mmHR 2X-FGF10-965 / GTCTTCATTCTTGGTCCCGC / GTCTTCATTCTTGGTCCCGC:TTATACGTCGAGTTGAAGAACAACCTG
Each H probe has
● a region that binds to the corresponding Connector Circle /CC (marked in green)
● a region that binds to the common bridge (CB2-31AC) (marked above in magenta)
● a region (20bp) that binds to the target RNA (marked above in blue)
Order the complete H probe sequence from IDT as 25nmole oligo with standard desalting.
Sample preparation
Growing cells on poly-Lysine coated coverslips:
Preparing poly-Lysine coated coverslips:
1. Wash coverslips in 70% EtOH briefly and transfer into sterile plates (that have also been pre-washed with 70% EtOH).
2. Rinse once with TC-compatible H2O (Pyrogen free).
3. Dilute Poly-Lysine solution (Sigma P4707) 1:50 by volume into DEPC treated H2O.
4. Incubate poly-Lysine solution over coverslips for 20 mins in hood.
5. Rinse 3X with TC H2O.
6. Aspirate off all remaining H2O and allow to air dry 10 min in hood.
Splitting cells growing in flask:
1. Aspirate off media and discard.
2. For a medium flask, wash 1X in 5mL PBS/EDTA solution. (Alternatively, you can
use trypsin).
3. Cover cells in 1mL PBS/EDTA solution, then incubate at 37C in incubator for 2-3 min
(no more than 5 min).
4. Gently tap cells and ensure they have detached into solution (cells should be visibly
floating).
5. Pre-dilute at least 1:10 final into fresh pre-warmed media
6. Perform final dilution (1:20 etc.) with fresh media and then gently pipet onto petri plate
with coverslips. Grow cells till they reach desired confluency.
● When your cells are ready, rinse coverslips in 1X PBS. Fix in 3.7% formaldehyde with 0.1% DEPC at RT for 20 min.
● Antigen retrieval - Pre-treat coverslips with 10 mM Citrate pH 6.0 + 0.05% LDS at 70C for 30 min, followed by RT drying (5 min).
● Dehydrate in Ethanol series - 50%, 75%, 90%, 100%.
● Apply seal chambers (GBL621505 Sigma). Proceed to probe Hybridisation.
For Tissues (on Superfrost glass slides):
● Prepare sections from OCT cryo blocks or paraffin blocks of PFA fixed tissue.To make cyro blocks- we collect perfused mouse lung samples, fix in 4% PFA at 4C overnight, transfer tissue to 30% sucrose solution at 4C overnight followed by OCT embedding and store blocks at -80C
● For OCT blocks: Cut 20um sections and collect on slides. Air dry for 5-10 min then post fix sections in 3.7% formaldehyde with 0.1% DEPC at RT for 20 min. Proceed to Antigen retrieval step…
● For FFPE samples : Cut 10 um paraffin sections and collect on slides. Air dry for 30 minutes then bake at 60C for 1 hr; Deparaffinize- Immerse in Xylene/Histoclear 5 min; Xylene/Histoclear 5 min; 100% Ethanol-5 min, 100% ethanol – 5 min. Proceed to Antigen retrieval step.
Antigen retrieval - Pre-treat slides with 10 mM Citrate pH 6.0 + 0.05% LDS at 70C
for 30 min, followed by RT drying. Remove excess moisture using Kimwipe.
● Apply seal chambers (GBL621505 Sigma)
For Tissues (on poly-Lysine coated coverslips):
Poly-L-lysine solution (Sigma P8920) diluted 0.1 % (w/v) in DEPC treated H2O
Incubate coverslips with this solution for 5 min, remove excess and air dry slides for 10min or until all the moisture has evaporated. Store slides protected from dust.
In situ hybridization
The example below shows how to perform PLISH for 1 target gene with 1 set of probes designed to work in Cy5
1) Hybridization of H probes:
Hybridization buffer: 1M NaTCA, 5 mM EDTA, 50 mM Tris pH 7.4, 0.2mg/mL Heparin.
Use H probes at 100 nM final concentration each. If using more than 5 probe sets, the pooled concentration should not exceed 1000 nM for any single target gene. Note, if co-staining for multiple genes, apply this same concentration rule for each (maximum 1000 nM).
1x / Reaction component49 uL / Hybridization buffer
0.5uL / 10 uM HL5X-gene-position
0.5uL / 10 uM HR5X-gene-position
Incubate at 37C in a humid chamber for 2 hours.
Wash 4 x 5 mins in Hybridization buffer at RT (20mins total).
2) Circularization Reaction:
1X / Reaction Component1.0 uL / phosphorylated CB2-31AC (PNK reaction mix)
1.0 uL / phosphorylated CC05-DP5 (PNK reaction mix)
1.0 uL / BSA (20 mg/mL)
0.5 uL / Heparin (20 mg/mL)
5.0 uL / 10X Ligase Buffer with 10 mM ATP
4.17 uL / 3 M NaCl
0.5 uL / 0.5% Tween-20
36.83 uL / DEPC-treated H2O
Incubate at 37C in humid chamber for 60 min.
Wash 5 min in PBST (PBS with 0.05% Tween-20) at RT.
3) Circularization Ligation Mix:
1X / Reaction Component0.25 uL / T4 DNA Ligase (NEB- M0202T)
1.0 uL / BSA (20 mg/mL)
0.5 uL / RNaseOUT
5.0 uL / 10X Ligase Buffer with 10 mM ATP
4.17 uL / 3 M NaCl
0.5 uL / 0.5% Tween-20
38.58 uL / DEPC-treated H2O
Add Ligation Mix and incubate at 37C in humid chamber for 2h.
Wash 2 x 5 min in Hybridization buffer at RT.
Wash 2 min in 1X phi 29 polymerase buffer (comes with kit)
4) RCA Reaction:
1X / Reaction Component5.0 uL / Phi 29 polymerase (Nxgen phi 29 polymerase from Lucigen- 30221)
5.0 uL / 10X phi 29 polymerase buffer
5.0 uL / 50% Glycerol
1.25 uL / 4 x 10mM dNTP mix
1.0 uL / BSA(20 mg/mL)
0.5 uL / RNaseOUT
32.25 uL / DEPC-treated H2O
Incubate at 37C in humid chamber overnight.
Wash 2 x 5 min in Label Probe Hybridization buffer (2x SSC/20% Formamide in DEPC-treated water)
5) Label Probe Hybridization:
Final concentration of Label probes - 100nM each
1X / Reaction component49 uL / Label probe hyb buffer(2x SSC/20% Formamide in DEPC treated water)
0.5 uL / 10uM (working) DP5-Cy5 (Detection oligo or Label probe).
0.5 uL / Heparin (20 mg/ml)
Incubate at 37C in humidity chamber for 30-60 mins.
(If not proceeding to IHC, wash samples in PBST (PBS with 0.05% Tween-20) 2 x 5 min followed by 5 min of DAPI in PBS (stock – 1mg/ml ; working 1:1000 in PBS). Mount in Vectashield – Anti fade mounting medium-H1000.)
Immunohistochemistry (post ISH):
*Add EDTA to all reagents (5mM final)
Remove ISH samples from 37C. Wash 3 x 5 min in PBST (PBS with 0.05% Tween-20) at RT.
Add Blocking serum and incubate for 1h at RT (with rocking).
Blocking serum: 50ul/ml (5%) Serum, 1ul/ml (0.1%) Triton X-100 and 0.03g/ml (3%) BSA in PBS. Mix thoroughly.
Remove blocking. Add Primary Ab and incubate at RT for 2h (with rocking). Primary Ab: 1:500 (or other appropriate dilution) in blocking solution.
Wash 4 x 5 min in PBST at RT. Add Secondary Ab and incubate at RT for 1h (with rocking). Secondary Ab: 1:500 (or other appropriate dilution) in blocking solution.
Wash 4 x 5 min in PBST at RT.
Wash 5 min in PBS including DAPI.
Mount in Vectashield -Anti fade mounting medium-H1000.
Data Analysis
Collect data using a Confocal microscope. Align images if required.
Use FIJI (or other software of your choice) to:
- split image into individual channels
- generate maximum intensity projections
- save/export images as TIF files
CellProfiler analysis:
Adding Images and assigning names:
Drag and Drop images (MIP of individual channels) to Cell profiler.
Click “NamesandTypes” under Input modules. Choose option “Image matching rules” under Assign a name to. Assign appropriate names to images.
Run Pipeline:
If using a previous pipeline, go to File, choose Import → Pipeline from file → Open.
If you want to create your own pipeline, go to Edit → Add Module (you can choose from a list of available options).
PLISH pipeline:
Identify primary objects: Nuclei
- Enhance or Suppress Features: This applies a filter to the DAPI image to remove very small objects. Label object as “FilteredBlue”. Select operation “Suppress”.
- Identify Primary objects:
Use “FilteredBlue” as input image. Name primary objects to be identified → “Nuclei”.
We have used ”11-26” pixel units as typical diameter of objects. The threshold strategy is “Automatic”.
Method to distinguish clumped objects is “Laplacian of Gaussian”.
Method to draw dividing lines between clumped objects is “Propagate”. Manually enter the size of Smoothing filter - in this case we set the size as “30”.
This module first thresholds the filtered DAPI image using the maximum correlation algorithm to set the threshold value. It then uses a Laplacian of Gaussian transformation with a characteristic size parameter (automatic or user-defined) to identify pixels that (i) are local intensity maxima, and (ii) where the increase in intensity gets steep closer to the pixel with a gradient that exceeds a threshold value (automatic or user-defined). These points are used as nuclei "seeds". The module applies a propagation algorithm to expand the seeds into full nuclei with defined edges, and smooths the edges over a user-defined length scale. The module then fuses putative nuclei that are closer together than a lower threshold (either user defined or estimated automatically). The module then applies a closing transformation to the putative nuclei to remove holes. Finally, it optionally removes putative nuclei with areas that are too big or too small, and optionally removes putative nuclei that touch the edge of the image.
The IdentifyPrimaryObjects module is configured NOT to eliminate nuclei based on edge/size criteria. The questionable nuclei must be retained for the IdentifySecondaryObjects module, to prevent them from being incorporated into the boundaries of a neighboring cell. These nuclei assignments are also used in the OverlayOutlines module.
- MeasureObjectSizeShape module computes the geometrical properties of all nuclei, including the questionable ones. This information is used later in FilterObjects, and to provide the downstream Matlab/R analysis with information that can be used for cell-type assignments.
- Save images: saves nuclei objects as TIF files.
Identify Secondary Objects: Module to assign the boundaries of cells.
5. IdentifySecondaryObjects :
Select input image → “DAPI”. Select input objects → “Nuclei” . name the objects to be identified → Cells. Select method to identify Secondary objects → “DistanceN”. Number of pixels to expand Primary objects → “2”. Discard Secondary objects touching border of Image → “No”
This module expands the boundaries of the unfiltered nuclei by two pixels so that the resulting "cell" assignments will include the peri-nuclear region. It then applies a closing transformation to the putative cells to remove holes. It optionally eliminates cells whose boundaries contact the edge of the image.
6. IdentifySecondaryObjects :
Select input image → “DAPI”. Select input objects → “Nuclei” . name the objects to be identified → Cells. Select method to identify Secondary objects → “DistanceN”
Number of pixels to expand Primary objects → “2”
Discard Secondary objects touching border of Image → “Yes” . Discard associated Primary objects → “No”.
IdentifySecondaryObject is applied twice, once with elimination of cells in contact with the edge of the image, and once without. The double execution is necessary because of a conceptual bug in CellProfiler's nuclei/cell indexing scheme. The first instance is used to identify non-edge cells (CellsNotEdge) for the OverlayOutlines module. The second instance is used by Matlab/R to correlate nuclear and cell indices, and also for the OverlayOutlines module.
7. Save Images: SaveImages saves the cell_object.tif file that records the location of all cell assignments.
Filter cell assignments:
8. Filter Objects:
Name output objects → NucsSized. Select Object to filter → “Nuclei”. Select Filtering mode → “Measurements” . Select Measurement to filter by → Category→ “AreaShape”. Measurement → “Area”
Filtering method → “Limits”.
In our analysis we used a Minimum value - “95” and Maximum Value → “531”. Select additional objects to relabel “CellsNotEdge”. Name the relabelled objects “CellsNotEdgeGoodNuc”
The first instance of FilterObjects takes the cells that do not contact the edge of the image (CellNotEdge), and eliminates the ones that derive from nuclei with areas outside the specified bounds (computed above in MeasureObjectSizeShape). These doubly filtered cells (CellNotEdgeGoodNuc) are used in the OverlayOutlines module.
9. Filter Objects:
Name output→ “CellsInside”. Select Object to filter → “Cells”. Select Filtering mode → “Image or mask folder”
The second instance of FilterObjects flags edge-contacting cells (those eliminated from CellsInside) for the downstream Matlab analysis. CellsInside would appear redundant with CellNotEdge, but the extra filter is required because of a conceptual bug in CellProfiler's nuclei/cell indexing scheme.
10. OverlayOutlines :
Select image to display outlines → “DAPI”. Name output image → “OrigOverlay”
Load outlines from an image or object → “Objects”
Select outline colour → Purple
Select objects to display → “Cells”
Load outlines from an image or object → “Objects”
Select outline colour → Red
Select objects to display → “CellNotEdgeGoodNuc”
Load outlines from an image or object → “Objects”
Select outline colour → Green
Select objects to display → “Nuclei”
The OverlayOutlines module outlines the rejected cells in purple, the accepted cells in red, and all of the nuclei in green --- placed on top of the DAPI image. The output can be used to judge the quality of the nuclei/cell assignments.
Thresholding Images:
11. Apply Threshold : for SPC
Select input image → “SPC”
Name output image → “SPCt”
Threshold strategy → Automatic
12. Apply Threshold : for SCGB
13. Apply Threshold : for AGER
14. Apply Threshold : for FTL1
The ApplyThreshold modules use CellProfiler's default maximum correlation thresholding algorithm to threshold the ISH data images.
Measure intensity of ISH signals
15. Measure Object intensity :
Select an image to measure → “ SPCt”
Select an image to measure → “AGERt ”
Select an image to measure → “ FTL1t”
Select an image to measure → “ SCGB1A1t”
Select objects to measure → “Cells”
The MeasureObjectIntensity records a variety of different metrics related to intensity information integrated over the assigned boundaries of each cell, both accepted and rejected. Data from rejected cells is removed in the downstream Matlab analysis
16. Export to Spreadsheet: Add Filename prefix → “MyExpt_Image1”
Use the Mean intensity values recorded in the output csv file. You can save these values in a new csv file (Meanint.csv)
R analysis - Generate heatmap
Open R studio. Import dataset ---> From CSV (Install updates if it prompts any). Browse → Choose “Meanint.csv”. Now the file will be added to R.
MI < - read.csv(“Meanint.csv”)
logMI < - log10(MI+0.1)