This paper describes a protocol for the rapid and efficient spatiotemporal monitoring of normal and aberrant cytosine methylation within intact zebrafish embryos.
Cytosine methylation is highly conserved across vertebrate species and, as a key driver of epigenetic programming and chromatin state, plays a critical role in early embryonic development. Enzymatic modifications drive active methylation and demethylation of cytosine into 5-methylcytosine (5-mC) and subsequent oxidation of 5-mC into 5-hydroxymethylcytosine, 5-formylcytosine, and 5-carboxylcytosine. Epigenetic reprogramming is a critical period during in utero development, and maternal exposure to chemicals has the potential to reprogram the epigenome within offspring. This can potentially cause adverse outcomes such as immediate phenotypic consequences, long-term effects on adult disease susceptibility, and transgenerational effects of inherited epigenetic marks. Although bisulfite-based sequencing enables investigators to interrogate cytosine methylation at base-pair resolution, sequencing-based approaches are cost-prohibitive and, as such, preclude the ability to monitor cytosine methylation across developmental stages, multiple concentrations per chemical, and replicate embryos per treatment. Due to the ease of automated in vivo imaging, genetic manipulations, rapid ex utero development time, and husbandry during embryogenesis, zebrafish embryos continue to be used as a physiologically intact model for uncovering xenobiotic-mediated pathways that contribute to adverse outcomes during early embryonic development. Therefore, using commercially available 5-mC-specific antibodies, we describe a cost-effective strategy for rapid and efficient spatiotemporal monitoring of cytosine methylation within individual, intact zebrafish embryos by leveraging whole-mount immunohistochemistry, automated high-content imaging, and efficient data processing using programming language prior to statistical analysis. To current knowledge, this method is the first to successfully detect and quantify 5-mC levels in situ within zebrafish embryos during early development. The method enables the detection of DNA methylation within the cell mass and also has the ability to detect cytosine methylation of yolk-localized maternal mRNAs during the maternal-to-zygotic transition. Overall, this method will be useful for the rapid identification of chemicals that have the potential to disrupt cytosine methylation in situ during epigenetic reprogramming.
Enzymatic modifications drive active methylation and demethylation of cytosine into 5-methylcytosine (5-mC) and subsequent oxidation of 5-mC into 5-hydroxymethylcytosine, 5-formylcytosine, and 5-carboxylcytosine1,2. Tris(1,3-dichloro-2-propyl) phosphate (TDCIPP) is a widely used flame retardant in the United States that has been previously demonstrated to alter the trajectory of cytosine methylation following early embryonic exposure from 0.75 hours post-fertilization (hpf) through early gastrulation (6 hpf)3,4,5,6,7,8. Within vertebrates, 5-mC and its modified derivatives are critical for regulating early embryonic development9. Fertilization of an embryo triggers demethylation of parental DNA, followed by maternal mRNA degradation, zygotic genome activation, and remethylation of the zygotic genome9. Biologically relevant processes that utilize cytosine methylation include histone modification, recruitment of transcriptional machinery, RNA methylation, epigenetic reprogramming, and determination of chromatin structure10,11. Cytosine methylation is also conserved among vertebrate species, underscoring the importance of understanding and investigating how aberrant cytosine methylation may affect the trajectory of an organism's development11. Furthermore, in utero development is sensitive to maternal exposure and has the potential to cause adverse outcomes such as immediate phenotypic consequences, long-term effects on adult disease susceptibility, and transgenerational effects of inherited epigenetic marks12,13,14.
Long stretches of cytosine-guanine pairs, or CpG islands, have been the primary foci of investigators that aim to characterize the dynamics of cytosine methylation across the genome15,16,17. Bisulfite-based strategies such as whole-genome bisulfite sequencing, reduced representation bisulfite sequencing, and bisulfite amplicon sequencing represent the gold standard for interrogating cytosine methylation at base-pair resolution. However, sequencing-based approaches are cost-prohibitive and, as such, preclude the ability to monitor cytosine methylation across developmental stages, multiple concentrations per chemical, and replicate embryos per treatment. In addition, sequencing-based approaches do not provide information about spatial localization, which is critical for understanding potentially affected cell types and areas within a developing embryo. Similarly, global DNA methylation assays such as methylation-dependent restriction analysis, 5-mC enzyme-linked immunoassays (ELISAs), and 5-methyl-2'-deoxycytidine (5-mC) liquid chromatography-mass spectrometry (LC-MS) rely on cell or tissue homogenates and, as such, preclude the ability to monitor the localization and magnitude of cytosine methylation over space and time within intact specimens12,18.
Due to the ease of automated in vivo imaging, genetic manipulations, rapid ex utero development time, and husbandry during embryogenesis, zebrafish embryos continue to be widely used as physiologically intact models to uncover xenobiotic-mediated pathways that contribute to adverse outcomes during early embryonic development. Therefore, using commercially available antibodies specific to 5-mC, the protocol below describes a cost-effective strategy for rapid and efficient spatiotemporal monitoring of cytosine methylation within individual, intact zebrafish embryos by leveraging whole-mount immunohistochemistry (IHC), automated high-content imaging, and efficient data processing using programming language prior to statistical analysis.
To current knowledge, this method is the first to monitor 5-mC within intact zebrafish embryos. The method enables the detection of DNA methylation within the cell mass and also has the ability to detect cytosine methylation of yolk-localized maternal mRNAs during the maternal-to-zygotic transition. Overall, this method will be useful for the rapid identification of chemicals that have the potential to disrupt cytosine methylation in situ during epigenetic reprogramming.
Adult breeders were handled and treated in accordance with an Institutional Animal Care and Use Committee (IACUC)-approved animal use protocol (#20180063) at the University of California, Riverside.
1. Zebrafish embryo collection and chemical exposure
2. Dechorionation of embryos
3. Immunohistochemistry using 5-mC-specific antibody
4. Automated imaging of embryos within 96-well plates
5. Data analysis
The overall aim of this protocol is to determine whether a treatment affects the relative abundance of 5-mC by assessing the total area and relative intensity of fluorescence within fixed and labeled zebrafish embryos. After completing the protocol, a fluorescence stereomicroscope can be used to first determine whether the whole-mount IHC was successful. When labeled embryos are observed under a FITC or GFP filter, a positive result is indicated by a positive FITC signal within the embryo, whereas a negative result is indicated by the absence of fluorescence within control embryos. Using a high-content screening system, these results can also be confirmed during image acquisition using a FITC filter. In addition, during data analysis, the custom module will identify and quantify the total area and integrated intensity of fluorescence. Representative images of successful results are shown in Figure 1, where acquired images were measured successfully by the custom module, and individual data points (shown as blue overlay) were acquired for both total area and integrated intensity.
During data extraction, the threshold stringency within the custom module is an additional variable that needs to be optimized to maximize the signal-to-noise ratio and increase the probability of detecting a significant treatment-specific difference in 5-mC abundance. During optimization, this final threshold provided the most significant separation between medians of control and treatment groups (e.g., the largest signal-to-noise ratio). Representative results at varying custom module thresholds that impact signal-to-noise ratios are presented in Figure 2.
Figure 1: A flow diagram providing a graphical representation of the protocol for exposure and in situ detection of 5-methylcytosine for 6 hpf zebrafish embryos. The direction of the flow diagram is provided with black arrows. Exposures occurred in replicates of four replicate dishes per treatment and 50 embryos per replicate dish. Abbreviations: cm = cell mass; ys = yolk sac. Please click here to view a larger version of this figure.
Figure 2: Optimization of the custom module. Panels A–D show optimization of the custom module by assessing the median and distribution of 5-mC-specific total area and integrated intensity within zebrafish embryos at 6 hpf. (A,B) Different thresholds tested are listed in the legend to the right (difference of 100 between each threshold) and are ordered by stringency, where 1500 and 2500 represent the least and most stringent thresholds tested, respectively. (C,D) Different thresholds tested are listed in the legend to the right (difference of 250 between each threshold) and ordered by stringency, where 1500 and 2500 represent the least and most stringent thresholds tested, respectively. The (*) in C,D denotes thresholds that are significantly different from vehicle-treated embryos (p < 0.05). For A,B, all thresholds tested were significant from the vehicle control. The x-axis denotes exposure to either vehicle (0.1% DMSO) or 0.78 µM TDCIPP (positive control). The y-axis denotes relative fluorescence. Panel A displays the total area of 5-mC detected within embryos as a function of treatment, whereas Panel B displays the integrated intensity of 5-mC within that same area. One embryo is represented by a single data point for a total of N = 96 for each treatment group. All exposures were performed in replicates of four dishes per treatment with 50 embryos per glass dish. Please click here to view a larger version of this figure.
Supplemental File 1: 5mC program code. Please click here to download this File.
During this protocol, there are a few steps that are critical. First, when dechorionating embryos, it is important to point the needle away from the tissue of the embryo/yolk sac/cell mass, as these portions of the developing embryo are very fragile and easy to puncture. Second, when transferring labeled embryos to individual wells, use a glass pipette to transfer embryos as they will adhere to a plastic pipette. Third, when performing whole-mount IHC, ensure that the plate is protected from light. Finally, after completing the whole-mount IHC protocol, allow the plate to incubate in 1x PBS at 4 °C overnight before imaging, as this will minimize autofluorescence that may interfere with imaging.
If there are embryos that have been severed during dechorionation or IHC, exclude these embryos from the remainder of the protocol. If no fluorescence is detected, this may be solved by incubating for longer (up to 16 h). Since this is a 5-mC-specific antibody, both DNA and RNA may be labeled. A general understanding of spatial localization is important.
There are some limitations in this method. First, it is a non-targeted technique, meaning it will not provide the exact quantity of 5-mC, but rather the relative abundance based on the total area and integrated intensity of fluorescence. In addition, this protocol has only been tested on embryos prior to segmentation, so if testing later stages of development, some additional optimization may be needed. Furthermore, since the method is IHC-based, it may be susceptible to non-specific binding; therefore, it is not certain to only be staining 5-mC localized to DNA, but may also be staining 5-mC localized to RNA as well. Lastly, optimization of the data analysis threshold may be needed depending on the chemical and stringency of the conditions that are preferred.
Overall, this method provides quick and cost-efficient detection of 5-mC across multiple stages of development and chemical concentrations3. It therefore provides an alternative to cost-prohibitive bisulfite sequencing-based approaches. By offering this protocol, investigators can use this method to quickly screen chemicals and assess how the abundance of 5-mC may be affected during early embryonic development. In addition, this method may be utilized as a prescreening tool to identify the concentration range, period of development, and/or window of sensitivity in which the chemical of interest affects 5-mC abundance. Alternatively, this same method may be utilized for a different biomarker and antibody, albeit further optimization is needed. By utilizing this method, an investigator can quickly, efficiently, and cost-effectively screen and identify a chemical that alters the relative abundance of 5-mC within zebrafish embryos prior to investing in labor-intensive bisulfite sequencing-based approaches. However, this method is zebrafish-specific, and further research is needed to determine whether 5-mC can be detected in situ within early embryos of other model organisms.
The authors have nothing to disclose.
Research support was provided by a UCR Graduate Division Fellowship to SAB, a NRSA T32 Training Program Fellowship (T32ES018827) to SAB, and a National Institutes of Health grant (R01ES027576) and USDA National Institute of Food and Agriculture Hatch Project (1009609) to DCV.
1.5-mL microcentrifuge tubes | Fisher Scientific | 540225 | |
10-µL glass microcapillary pipette | Fisher Scientific | 211762B | |
100-mm plastic Petri dish | Fisher Scientific | 08757100D | |
10x phosphate-buffered saline | Fisher Scientific | BP399500 | |
1-mL pipette | Fisher Scientific | 13690032 | |
250-mL Erlenmeyer flask | Fisher Scientific | FB501250 | |
5-mL pipette | Fisher Scientific | 13690033 | |
60-mm glass petri dishes with lids | Fisher Scientific | 08747A | |
96-well plate | Fisher Scientific | 720089 | |
AlexaFluor 488-conjugated goat anti-mouse IgG antibody | Fisher Scientific | A21121 | |
Bovine serum albumin | Fisher Scientific | BP67110 | |
DMSO | Fisher Scientific | BP2311 | |
Hotplate | Fisher Scientific | 1110016SH | |
In-tank breeding traps | Aquatic Habitats | N/A | This product is no longer available following acquisition of Aquatic Habitats by Pentair. Investigators can use standard off-system breeding tanks available from multiple vendors. |
ImageXpress Micro XLS Widefield High-Content Screening System | Molecular Devices | N/A | Any high-content screening system equipped with transmitted light and FITC filter will be suitable. |
Immunochemistry (IHC) basket | N/A | N/A | Manufactured in-house using microcentrifuge tubes with conical portion removed and bottom fitted with mesh, sized for 24- or 48-well plates. |
MetaXpress 6.0.3.1658 | Molecular Devices | N/A | Any software capable of quantifying total area and integrated intensity of fluorescence will be suitable. |
Microspatula | Fisher Scientific | 2140115 | |
Monoclonal mouse anti-5-mC antibody | Millipore Sigma | MABE146 | |
NaOH | Fisher Scientific | BP359-500 | |
Orbital shaker | Fisher Scientific | 50998290 | |
Parafilm | Fisher Scientific | 1337412 | |
Paraformaldehyde | Fisher Scientific | 18612139 | |
Plastic transfer pipette | Fisher Scientific | 1368050 | |
Rstudio | RStudio | N/A | RStudio is open-source software and can be downloaded at https://www.rstudio.com. |
Sheep serum | Millipore Sigma | S3772-5ML | |
Stereomicroscope | Leica | 10450103 | |
Temperature-controlled incubator | Fisher Scientific | PR505755L | |
Tween-20 | Fisher Scientific | P7949-500ML |