Abstract
Gene regulatory networks describe interactions between transcription factors and regulatory DNA and can provide a systems-level view of tissue development. Here, we describe a protocol for isolation and interrogation of tissue-specific cis-regulatory elements during zebrafish embryonic development using low-input ATAC-seq. With the methods described, genome-wide assessments of regulatory DNA in small populations of developing tissues can be identified allowing the construction of gene regulatory networks.
Keywords: Assay for transposase-accessible chromatin using sequencing (ATAC-seq), Fluorescence activated cell sorting (FACS), Enhancer screening, Zebrafish
1. INTRODUCTION
Tissue development is both a highly dynamic and structured process that is achieved via tight regulation of gene expression. The study of gene regulation allows us to build gene networks by characterizing cis-regulatory elements, such as enhancers, and their effects on target gene expression. Tissues within the same organism utilize gene networks differentially during development to achieve unique tissue-level phenotypes1. This is accomplished largely by distinct activation of cis-regulatory elements. Identifying cis-regulatory elements controlling each molecular state can help to understand how these networks become dysregulated in instances of congenital birth defects2.
The zebrafish is an excellent model for developmental genetics due to the ease of transgenics and visualizing the embryo. However, until recently, the ability to study gene regulation on a tissue-specific level in zebrafish has been limited by available technologies. The ability to do chromatin genomics on specific tissues has now introduced exciting opportunities to further build developmental gene networks by identifying tissue-specific cis-regulatory elements. In particular, the Assay for Transposase-Accessible Chromatin with high-throughput sequencing (ATAC-seq) can identify potential enhancers by sequencing stretches of DNA with open or accessible chromatin genome-wide. ATAC-seq works by utilizing a hyperactive Tn5-transposase that has been preloaded with sequencing adapters that will simultaneously fragment and tag stretches of open chromatin3,4. These regions are then amplified using sample-specific indexed primers followed by sequencing. Sequencing reads are then mapped to the genome and enhancer regions are identified using downstream bioinformatics approaches.
Here, we demonstrate how to identify tissue-specific enhancers in zebrafish development by performing bulk ATAC-seq on transgenic, dissected zebrafish embryonic tissues with focus on neural crest development. We then demonstrate how to validate these enhancers by creating transgenic fish and performing site-directed mutagenesis screening. This protocol introduces a way to study tissue-specific gene regulation on a high-throughput scale in zebrafish development.
2. MATERIALS
2.1. Equipment
Qubit 4 Fluorometer.
Agilent 2100 Bioanalyzer Instrument.
BioRad C1000 Touch Thermal Cycler.
Eppendorf Thermomixer C.
Leica THUNDER Imager Model Organism Fluorescent Stereoscope.
Eppendorf Centrifuge 5810R.
Sutter Instruments Needle Puller Model P-2000.
Narishige Manipulator M-152.
ASI MPPI-3 Pressure Injector.
Eppendorf Centrifuge 5425R.
2.2. Software
2.3. Reagents (see Note 1)
2.3.1. Dissections and Dissociation
2.3.1a. Solutions (see Note 2):
HANKS Sorting Buffer: 1X HBSS, 0.25% BSA, 10 mM HEPES, pH 8, 10 mM MgCl2, Nuclease Free-Water. After preparation, make sure to sterile filter HANKS with a 50 mL syringe and 0.2 μm filter.
Cell Dissociation Mix10: 0.5 mM EDTA, 10 mg/mL B. lich. Protease (#P5380–1G), 125 U/mL RQ1 Dnase, 1 X DPBS.
20 X Egg Water: 0.6% Instant Ocean, 0.15% CaSO4 dihydrate, MilliQ Water.
10 X Tricaine: 0.4% Tricaine powder in 1X Egg Water. pH to 7.0.
1% BSA: 1g/100 mL of 1 X DPBS.
2.3.1b. Other Materials:
1. Promega RQ1 Dnase (#PRM6101).
2. 70 μm cell strainer.
6. 15 mL and 50 mL conical tubes.
7. 0.2 μm sterile filter.
8. 50 mL syringe.
9. 24 well plate.
10. Low Adhesion pipette tips.
11. Plastic petri dishes.
12. #55 and #5 Forceps.
2.3.2. FACS
2.3.2a. Solutions:
HANKS Elution Buffer: 1 X HBSS, 0.25% BSA, 10 mM HEPES, 5 mM EDTA, Nuclease-Free Water. After preparation, make sure to sterile filter HANKS Elution Buffer with a 50 mL syringe and 0.2 μm filter.
2.3.2b. Other Materials:
5 mL Round Bottom Polystyrene Tubes (“FACS Tubes”).
DAPI.
1.7 mL low-adhesion Eppendorf Tubes.
Trypan blue.
Hemocytometer.
2.3.3. Low Input ATAC
2.3.3a. Solutions/Chemicals:
Cold Lysis Buffer: 10 mM Tris-HCl, pH 7.4, 10 mM NaCl, 3 mM MgCl2, 0.1% IGEPAL CA-630, Nuclease Free H2O.
For each sample, prepare the following transposition mix: 5 μL of Nextera TD Buffer (Illumina FC-121–1030), 0.5 μL of Tn5 transposase (Illumina FC-121–1030), 4.5 μL Nuclease Free H2O (see Note 3).
80% Ethanol.
0.5 M EDTA.
2.3.3b. Other Materials:
2.3.4. Enhancer Screening
2.3.4a. Solutions/Chemicals:
Tol2 mRNA.
Rhodamine dextran.
2.3.4b. Other Materials:
3.5’ Drummond Glass Capillaries (#3–000-203-G/X).
EMS Micro Stage Micrometer (#60210–12PG).
20 μL Eppendorf Microloader Pipette Tips (#930001007).
3. Methods
3.1. Dissections and Dissociations (4 hours)
In order to obtain tissue-specific regulatory elements, we recommend choosing a transgenic line that labels your cell/tissue type of interest, dissecting the tissue of interest from your developmental time point, and dissociating dissected regions of interest for fluorescence activated cell sorting (Section 3.2) to obtain pure populations of your cells of interest. Below, we outline our protocol for dissections of axial-specific populations of neural crest cells from the transgenic line Tg(sox10:eGFP) (Fig. 1).
Figure 1. Flowchart of zebrafish tissue-specific bulk ATAC-seq protocol and analysis pipeline.
A) Diagram illustrating dissection of transgenic zebrafish embryonic tissue of interest. B) Schematic showing dissociation of specific zebrafish tissue into a single cell-suspension. Positive and negative cell populations are shown in green and yellow, respectively. C) Fluorescent activated cell sorting (FACS) to enrich for the cell population of interest. D) Tn5 transposase pre-loaded with sequencing adapters fragments and tags regions of open chromatin. These regions are then amplified and sequenced. E) Differentially accessible elements between two cell types are depicted in a volcano plot after Diffbind analysis. F) Candidate enhancer regions are injected into the zebrafish embryo at the one cell stage. Activity of these enhancers are screened using a fluorescent reporter. Created with BioRender.com
Coat all pipette tips and other plastics thoroughly in 1% BSA.
Place all plastics and solutions on ice.
Dechorionate the embryos with #5 forceps.
Anesthetize the embryos in 1 X Tricaine.
De-yolk the embryos (see Note 4) in 1 X Tricaine.
Transfer the de-yolked embryos to a plastic petri dish filled with pre-chilled 1 X DPBS + 1 X Tricaine mixture.
Dissect the tissue region of interest in 1 X DPBS + 1 X Tricaine mixture using #55 forceps (see Note 5) (Fig. 1A).
Transfer the dissected tissue to a 24-well plate on ice using a P20 Low Adhesion ART tip.
Remove the excess of 1 X DPBS from well and add 1 mL of Cell Dissociation Mix to each sample.
Triturate using a P1000 ART pipette tip every 2 minutes to facilitate dissociation. While the sample is dissociating, pre-cool the Eppendorf Centrifuge 5810R with a swinging bucket rotor to 4°C.
Keep each sample in Cell Dissociation Mix until completely dissociated (see Note 6).
Once dissociated, transfer solution from the 24-well plate to a pre-chilled 15 mL conical tube. Wash each used well with 2 mL of HANKS Sorting Buffer and add this to the same conical tube. Then, add 5 mL of fresh HANKS Sorting Buffer to the conical to halt the protease (Fig. 1B).
Spin down the samples at 4°C at 400 g for 10 minutes to pellet the dissociated cells.
Carefully remove the supernatant without disturbing the pellet and resuspend in 1 mL of fresh HANKS Sorting Buffer (Note 7).
Pipette the solution through a 70 μm cell strainer and into a pre-chilled 50 mL conical tube.
Spin down the samples at 4°C for 10 minutes at 400g with cell strainer in place.
Remove the supernatant and resuspend in 500 μL of HANKS Sorting Buffer + 25 μL RQ1 DNase. Transfer the sample to the appropriate pre-chilled FACS tube.
3.2. FACS (2 hours)
In this step, the cells are collected into Eppendorf tubes + HANKS Elution buffer and directly used for low input bulk ATAC-seq (Section 3.3). All embryonic samples have been tested using this protocol with a 100 μM FACS nozzle in a BD FACSAria Fusion (Fig.1C).
If possible, set up the sorter during dissociation, so that the cells do not sit on ice for long periods of time.
- To check for cell viability via FACS, cells were stained with DAPI (1:2000) in FACS tubes on ice. In addition to unstained cells, appropriate sorting controls should be prepared based on your transgenic fluorescent reporter. For example:
- Unstained, non-transgenic cells.
- DAPI-stained, non-transgenic cells.
- DAPI-stained, transgenic gating controls (Note 8).
- DAPI-stained, experimental cells.
Load the unstained, non-transgenic cells to gate for cell size and potential cell aggregates.
Load the DAPI-stained, non-transgenic cells to gate for cell viability. There should be a clear separation between DAPI-positive cells and unstained cells.
Load the DAPI-stained, transgenic gating controls to set laser voltages and gating for experimental cells.
Load the DAPI-stained, experimental cells into the sorter. Load 1.7 mL Eppendorf tubes prefilled with 500 μL of HANKS Elution Buffer in the collection tube holder for collecting pure populations of both transgenic reporter-positive and -negative cells. Aim to collect 500–5000 cells per sample for each replicate to proceed with the low input ATAC-seq protocol.
Verify cell viability post-FACS by staining with Trypan Blue and a hemocytometer.
3.3. Low Input ATAC Protocol (4 hours)
In this step, cells are lysed and regions of DNA with open chromatin are simultaneously fragmented and tagged with sequencing primers using hyperactive Tn5 transposase4 (Fig. 1D). DNA libraries are then amplified via PCR using barcoded primers and sequenced on a NovaSeq 6000 50PE Flow Cell SP.
After FACS, pellet the cells by centrifuging in a tabletop, refrigerated centrifuge (ie. Eppendorf Centrifuge 5425R) at 400 g for 15 minutes (4°C). Gently remove and discard the supernatant without disturbing the pellet.
Wash the pellet by adding 50 μL of chilled 1 X DPBS. Centrifuge at 400 g for 15 minutes (4°C) and discard supernatant as above. Repeat for a total of 2 washes.
Add 25 μL of cold lysis buffer and gently pipette up and down to resuspend the cell pellet.
Centrifuge the cells in lysis buffer at 500 g for 30 minutes (4°C). Gently remove and discard supernatant, and immediately continue to the next step for the transposition reaction.
Add 10 μL of the transposition reaction mix to each sample and gently pipette up and down to resuspend nuclei. Incubate at 37°C for 30 minutes (See Note 9).
Directly proceed to DNA purification using a QIAGEN PCR purification kit. Elute purified DNA in 11 μL of nuclease free H2O.
-
To amplify transposed DNA fragments, prepare the amplification reaction in PCR strip tubes (See Note 10):
11 μL of transposed DNA
2 μL of 25 μM Customized Nextera PCR Index Primer 1 (no Mx)
2 μL of 25 μM Customized Nextera PCR Index Primer 2 (Ad#)
15 μL of Q5 2X PCR master mix
-
Place amplification reactions in a thermocycler with the following program:
98°C; 30 seconds
98°C; 10 seconds
65°C; 75 seconds
Repeat steps 2–3, 12–14X (Note 11)
65°C 5 mins
Hold at 4 °C
During the thermocycler reaction, let the AMPure beads warm to room temperature for at least 30 minutes.
Vortex the AMPure beads and add 30 μL (1:1 ratio) to each sample. (Note 12).
Mix the AMPure beads with reaction thoroughly by pipetting up and down at least 10 times (Note 13).
Briefly spin down the samples to collect the liquid from the side of the tube.
Place the samples on the 10 X magnetic separation device and let sit for at least 5 minutes. The beads should all be pulled to the magnet, and there should be no beads visibly left in the supernatant.
With strip tubes still on the magnet, slowly remove supernatant by gently aspirating without disturbing the pellet.
Wash the pellet by adding 200 μL of 80% ethanol to each sample still on the magnet, being careful not to disturb the beads. Wait 30 seconds and then carefully remove supernatant without disrupting the pellet. The DNA fragments will remain bound to the beads during the washing process.
Repeat Step 15 once more for a total of 2 ethanol washes.
Briefly centrifuge the samples to collect and remove any excess of ethanol. Place the samples on the magnet and wait for 30 seconds for beads to re-pellet at the magnet. Aspirate all of the remaining ethanol with a pipette to get the pellet as dry as possible.
Leave the samples at room temperature for approximately 2 minutes to dry (see Note 14).
Once the pellet has dried, remove the samples from the magnet and add 17 μL of nuclease free water. Pipette the mix until it is fully resuspended.
Incubate for 2 minutes at room temperature to rehydrate the DNA.
Briefly centrifuge and place the samples back on the magnet for 1 minute. Do not dispose of the supernatant at this step!
Transfer the clear supernatant containing the purified DNA fragments from each tube to a nuclease-free, low adhesion tube. Label each tube with the sample information and store at −20°C until sequenced.
Check the concentration of DNA in samples using Qubit and distribution of fragment sizes on an Agilent Bioanalyzer or TapeStation.
Submit the ATAC libraries for sequencing. Libraries are sequenced on a NovaSeq 6000 50PE Flow Cell SP at a depth of 50 million reads per sample.
3.4. ATAC Analysis (~1 day)
After demultiplexing the sequenced libraries, the reads are mapped to the zebrafish genome and analyzed for differentially accessible regions between cell populations of interest and surrounding cells. Comparing accessible elements between different cell populations allows for a more refined analysis of putative enhancer regions that are tissue-specific. While there are many approaches to analyzing ATAC-seq data and many potential downstream analyses/questions, to obtain differentially accessible elements between two cell populations we follow the pipeline below:
After the initial quality checks using FastQC, the adapters are trimmed from sequencing reads using Cutadapt v2.8.
Trimmed, paired-end reads are then mapped to the zebrafish genome (danRer11) using Bowtie2.
Mitochondrial reads and PCR duplicates are removed, and peaks are called using Genrich.
After peak calling, Diffbind is used to identify differentially accessible peaks between transgenic-positive and transgenic-negative libraries to identify peaks enriched in cell type of interest (Fig. 1E). Enriched peaks are then annotated using ChIPseeker, and motif enrichment analysis is done using HOMER.
3.5. Enhancer Screening and Transcription Factor Binding Site Mutagenesis
Enhancer screening is a critical step in validating the activity of candidate enhancer regions in vivo. Below, we outline the basic workflow for initial enhancer screening and then for the creation of stable enhancer transgenic reporter lines. This protocol has been optimized for Tol2-mediated transgenesis11. Once the initial enhancer activity is validated, transcription factor binding sites within the enhancer regions are then mutagenized (scrambled or deleted) and then screened for changes in activity/reporter expression.
1. Once the enhancer regions of interest have been identified, the sequences can be amplified from genomic DNA or ordered from a vendor (ie. Twist Biosciences or IDT) for cloning into transgenesis vectors using Gibson Assembly11,12(Note 15).
3. Sequence the plasmids with presumed enhancer region to verify correct insertion before injection.
4. On the day of injections, prepare the injection solution on ice: 12.5 ng/μL Tol2 RNA, 12.5 ng/μL plasmid DNA, 0.5 μL of dye (Note 16), Nuclease-free H20 up to 5 μL.
5. Pull the injection needles by loading 3.5’ Drummond Glass Capillaries into the Sutter Instruments Needle Puller Model P-2000.
6. Backload the injection needle with 2 μL of the injection mix using a 20 μL Eppendorf Microloader Pipette Tips.
7. Break the needle tip and calibrate the injection needle with a stage micrometer, so that 2 nL of injection mix is expelled.
8. Collect the zebrafish embryos immediately after they are laid and inject into the single cell at the one cell stage (Note 17).
9. Use a fluorescent stereoscope to screen for positive enhancer expression at your time point of interest.
10. Raise the positive F0 embryos to adulthood.
11. Outcross each F0 fish with a wild-type fish and screen the embryos for positive enhancer expression/germline transmission.
12. Raise the positive embryos (F1s) as these will have the transgene stably integrated in the genome.
13. Enhancer sequences can be analyzed for transcription factor binding sites of interest using MEME Suite. New enhancers with mutagenized binding sites can be injected into the embryo to identify changes in enhancer activity and reporter expression.
Footnotes
All materials, reagents, and benchtops should be kept RNAse-free.
All solutions should be prepared fresh on the day of the experiment. All solutions and plastics should be kept on ice.
For less than 1000 cells, use 0.25 uL of Tn5 instead.
We use Rhodamine dextran dye but any dye that does not interfere with reporter expression works.
We find it easiest to de-yolk the embryos by stabilizing the anterior part of the yolk with one forceps and then using the other forceps to push the yolk away from the body in an anterior to posterior motion.
Depending on your tissue of interest, it might be helpful to dissect under a fluorescent stereoscope.
Depending on your tissue of interest, you may need a shorter or longer dissociation time. Typically, dissociation of embryonic zebrafish tissue lasts from 5 – 30 minutes. It is important to make sure that dissociation does not proceed for too long, as this might decrease cell integrity. However, it is also important to make sure that tissue is fully dissociated, otherwise your sample will not be able to be processed properly. The mixture should look homogenous and cloudy when the tissue is fully dissociated. Factors that can influence dissociation time include amount of tissue dissected, type of tissue dissected and age of the embryos dissected.
Depending on the tissue you are dissociating, the pellet might not be visible. If this is the case, be especially careful when removing supernatant.
These controls should be collected from the same transgenic lines at the same developmental timepoints to ensure proper gating.
Transposition reaction time may need to be altered to prevent over-tagmentation or under-tagmentation of sample.
Samples sequenced together need a unique index, so you can identify the sample when de-multiplexing sequencing data.
Higher concentrations of transposed DNA may need less PCR amplification cycles.
This ratio is used to remove any adapter dimers.
Beads are viscous. Pipette slowly.
Let the pellet dry long enough, so that it looks matte but not cracked.
We use a vector backbone that has Tol2 sites flanking our enhancer insert, driving the expression of a cfos minimal promoter which then drives the expression of the fluorescent reporter, GFP. Additionally, we use Gibson Cloning, but any preferred cloning method can be used.
It is critical to inject into the single cell to increase chances of integration of your injected DNA into the germline.
References
- 1.Davidson EH Emerging properties of animal gene regulatory networks. Nature 468, 911–920 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Martik ML & Bronner ME Regulatory Logic Underlying Diversification of the Neural Crest. Trends Genet 33, 715–727 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Buenrostro JD, Giresi PG, Zaba LC, Chang HY & Greenleaf WJ Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position. Nat Methods 10, 1213–1218 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Buenrostro JD, Wu B, Chang HY & Greenleaf WJ ATAC-seq: A Method for Assaying Chromatin Accessibility Genome-Wide. Curr Protoc Mol Biol 109, 21.29.1–9 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Martin M Cutadapt removes adapter sequences from high-throughput sequencing reads. Embnet J 17, 10–12 (2011). [Google Scholar]
- 6.Langmead B & Salzberg SL Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Heinz S et al. Simple Combinations of Lineage-Determining Transcription Factors Prime cis-Regulatory Elements Required for Macrophage and B Cell Identities. Mol Cell 38, 576–589 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Yu G, Wang L-G & He Q-Y ChIPseeker: an R/Bioconductor package for ChIP peak annotation, comparison and visualization. Bioinformatics 31, 2382–2383 (2015). [DOI] [PubMed] [Google Scholar]
- 9.Ross-Innes CS et al. Differential oestrogen receptor binding is associated with clinical outcome in breast cancer. Nature 481, 389–393 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Adam M, Potter AS & Potter SS Psychrophilic proteases dramatically reduce single cell RNA-seq artifacts: A molecular atlas of kidney development. Dev Camb Engl (2017) doi: 10.1242/dev.151142. [DOI] [PMC free article] [PubMed]
- 11.Kwan KM et al. The Tol2kit: A multisite gateway‐based construction kit for Tol2 transposon transgenesis constructs. Dev Dynam 236, 3088–3099 (2007). [DOI] [PubMed] [Google Scholar]
- 12.Parker HJ, Sauka-Spengler T, Bronner M & Elgar G A Reporter Assay in Lamprey Embryos Reveals Both Functional Conservation and Elaboration of Vertebrate Enhancers. PLoS ONE 9, e85492 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]