Pregnant mare’s serum gonadotropin (PMSG) (#HOR-272, ProSpec)

Molecular Grade Water (RNase-, DNase-, Proteinase-free) (#786-73C, G-Biosciences)

20x saline-sodium citrate (SSC) buffer (#R020, G- Biosciences)

Water Jacketed CO₂ incubator, 37°C, 5% CO₂ with HEPA filtration (model 3120, Thermo Fisher)

Four-to five-week-old female mice were injected with 5 IU of PMSG; 48 h later, they were injected with 5 IU of hCG, then immediately placed in a cage with a fertile stud male for mating. On the morning of the next day (between 9 and 10 a.m.), the presence of vaginal plugs was checked. Mated female mice were placed in a new cage separate from the males; this time point was considered embryonic day (E) 0.5. Preimplantation embryos—blastocysts (E3.5) were collected 3 days later.

Before the embryos were collected from the uterus of the pregnant female mice on the desired day, M2 medium was warmed to 37°C. Then, 4% PFA in PBS, 1% agarose in saline, and thawed acidic Tyrode’s solution were prepared. A glass Pasteur pipette was pulled using an open flame to draw a capillary end at the tip to manipulate the embryos. Pregnant female mice were sacrificed by CO₂ inhalation and cervical dislocation. Preimplantation mouse embryos (blastocysts) were collected by flushing the uterus with an M2 medium. The embryos were collected using the mouth-controlled, glass Pasteur pipette with a capillary end, transferred to a new dish, and further washed by moving them through two to three drops of fresh M2 media. A stereoscope dissection microscope was used to locate the embryos. The zona pellucida was removed from the blastocysts by briefly washing it in acidic Tyrode’s solution; as soon as the zona pellucida was dissolved, all embryos were transferred back to the M2 medium. Notably, the embryos were very sticky after removing the zona pellucida. The glass pipette was coated with 1% BSA before use to both reduce the adherence of the embryos to the glass surface and reduce the risk of damage to or loss of the embryos.

Next, 500 µL of PBA was added to one of the wells of the coated 4-well tissue culture dish, and 500 µL of 4% PFA was added to another well. The embryos were briefly washed with PBA before fixing them in 4% PFA for 20 min at room temperature (RT), then washed three times in PBX for 5 min. The fixed embryos were used to perform the IF followed by smRNA FISH assay.

Note: At this stage, the fixed embryos can be stored for up to 2 days at 4°C before proceeding to the next step. To store the embryos, cover the PBS containing the embryos with a layer of mineral oil to prevent evaporation.

Standard IF was performed (Saiz et al., 2016) with minor modifications. The desired number (e.g., 20 embryos) of fixed embryos were transferred to a coated (1% agarose) round bottom 96-well plate to perform all the steps involved in IF and smRNA FISH. Notably, the round bottom 96-well plate was coated with 1% agarose to prevent the embryos from sticking to the plastic, and the 96-well plate minimizes the use of reagents. The embryos were incubated in 1% Triton X-100 in PBS for 30 min on ice for permeabilization. Then, the embryos were washed three times in PBX for 5 min each at RT, blocked in blocking solution for 1 h at RT, and incubated in primary antibody diluted in blocking solution (1:200 for anti-Cdx2 and 1:100 for anti-Tead4) overnight at 4°C. The solution was covered with a layer of mineral oil to prevent evaporation. The next day, the embryos were washed in PBX three times for 5 min each at RT, then incubated in secondary antibody diluted in blocking solution (1:500, Alexa Fluor 488-Goat anti-mouse IgG) for 2 h at 4°C. The embryos were washed in PBX three times for 5 min each at RT and followed up with the smRNA FISH procedure.

smRNA FISH was performed using ACD RNAscope technology according to the manufacturer’s instructions, with modifications. smRNA FISH method was initiated by an additional fixation step using 4% PFA for 10 min at RT. Then, the embryos were rinsed three times in fresh PBX solution (0.1% Triton X-100 in PBS). Please note that the protease digestion step was skipped. The embryos were then incubated for 2 h at 40 ± 2°C in a probe-containing solution (RNAscope Positive Control Probe, RNAscope Negative Control Probe, lncRNAs Platr4 and Malat1 and mRNA Pou5f1 probes. PBS was used as a no probe control. Note: We recommend using (RNase-DNase-, Proteinase-free) molecular grade water and Ca₂ ⁺/Mg₂ ⁺-free PBS to dilute any buffers, which preserves both FISH and IF signals.

Due to the specific gravity of the probe-containing solution, embryos tend to float; they also become transparent, which makes it challenging to find them in the solution. Using the stereoscope dissection microscope, it is crucial that the embryos are at the bottom of the well, so they are fully submerged during the 2-h incubation process. After the probe hybridization, the embryos were washed three times for 5 min each in SSCT at RT. We did not use ACD wash buffer because it disintegrates the embryos. After washing the embryos, we performed a series of incubations with pre-amplifier and amplifier reagents (Amp 1, Amp 2, and Amp 3) and fluorophore at 40 ± 2°C for 15–30 min each, according to manufacturers’ protocols. After every incubation, the embryos were washed twice in SSCT for 5 min at RT. After the final wash, the embryos were stained with DAPI solution and left overnight at 4°C, and the solution was covered with a layer of mineral oil. The next day, the embryos were washed twice in PBS for 5 min at RT and then transferred to a MatTek dish containing a drop of PBS for imaging.

Whole-mount preimplantation mouse embryos were imaged using a Zeiss LSM 780 laser-scanning confocal microscope equipped with a GaAsP detector with laser lines at 405, 488, 561, and 633 nm. Z-stack images were acquired with a 63x/1.4 oil immersion objective lens and collected in 12-bit and the pixel size in the software was set to zoom 1.5 of 88 nm in x, y with a z-spacing of 1.0 µm with Zen 2011 SP7 software. Line averaging of two–four and pinhole size was set to one to 1.68 Airy Unit (AU) were used to balance a good SNR, optical sectioning resolution and time to collect 3D datasets, the GaAsP detectors gain were set to achieve strong signal without saturating, the digital offset/background was set to avoid any pixels reading zero intensity values. ZEN Microscopy software was used for image processing and analysis.

We present a step-by-step protocol to superovulate mice, collect preimplantation embryos, and fix them in PFA (Figure 1A), followed by optimized, combined IF and smRNA FISH techniques (Figure 1B) to detect proteins and lncRNAs in intact preimplantation mouse embryos. The workflow diagram in Figure 1 outlines all the steps in this protocol. We have also provided images of live blastocysts (Supplementary Figure S1)

To visualize protein and lncRNA in the same sample, we first optimized the hybridization conditions using no antibody control, no probe control (Figure 2A), and negative and positive control probes in mouse blastocysts (Figures 2B,C). We have used the ACD RNAscope proprietary probes for the Positive and Negative control samples. Each of these probes was designed based on NCBI sequence data. Probes consist of 20 oligonucleotide pairs and each pair has a double Z configuration: One side of the Z is complementary to the specific RNA transcript and the other side is complementary to the amplifier DNA sequence (Wang et al., 2012)

The Escherichia coli dihydrodipicolinate reductase (dapB) gene mRNA was used as the negative control to verify the non-specific signal, and our data showed that we achieved complete negative staining in our sample (Figure 2B). We have detected the positive control probe for the housekeeping gene RNA polymerase 2a (polr2a) mRNA (detection channel-1, C1) as red punctuate dots (Figure 2C). The expression level of Polr2a mRNA was low in the cells, which others have shown in mouse preimplantation embryos, cultured mammalian cells, and tissue (Wang et al., 2012; Xie et al., 2018).

To detect the localization of a lncRNA and protein in the same sample, we first performed smRNA FISH using an ACD RNAscope probe and reagents, followed by standard IF (smRNA FISH - IF). The sequential smRNA FISH and standard IF can be technically challenging because of RNA degradation. However, previously this approach has been used to detect RNA and protein in nonadherent mammalian cells (Arrigucci et al., 2017), cultured mouse neurons (Eliscovich et al., 2017), paraffin-embedded tissue sections (Kwon et al., 2020), zebrafish (Gross-Thebing et al., 2014) and the C. elegans germline (Yoon et al., 2016). Therefore, we performed sequential smRNA FISH and IF in preimplantation mouse embryos. We used an anti-Cdx2 antibody to detect Cdx2 protein in mouse blastocysts. The transcription factor Cdx2 is expressed in the TE lineage and is vital for TE differentiation (Meissner and Jaenisch, 2006). We also used a target probe set for the lncRNA Platr4, which is crucial for differentiation and expressed in both the TE and inner cell mass (Hazra et al., 2021). However, using this sequential combination of the two techniques, smRNA FISH followed by IF, we were unable to detect the lncRNA (Platr4) signals but did detect the protein (Cdx2) signal. We reasoned that this failure to detect the lncRNA signal could be due to 1) reagents (e.g., blocking buffer, crude antibody solution): used in IF were not RNase-free, 2) the order of performing IF and smRNA FISH because in cell culture systems when using these two methods simultaneously in some cases the procedures work when IF is done first and in other cases (different targets) it works when the RNA FISH is done first.

Next, we optimized a protocol where an RNase-free IF method was performed first, followed by smRNA FISH in whole-mount mouse blastocysts. We further modified the smRNA FISH method in four critical ways: 1) we used RNase-free reagents to perform IF, 2) we added an additional post-fixation step after incubation with the secondary antibody but before starting the probe hybridization for smRNA FISH to prevent the removal or displacement of the antibody, 3) we eliminated the protease digestion treatment step to prevent loss of protein signal. Interestingly, these optimized modifications allowed us to successfully detect both IF signals for the Cdx2 and Tead4 proteins and the smRNA FISH signals for the Platr4 and Malat1 lncRNAs (Figures 3A–D, Supplementary Figures S2A, B). In the preimplantation mouse embryo, Tead4 is required to establish the TE-specific transcriptional program and segregate the TE lineages from the ICM (Home et al., 2012). In addition, Tead4 is essential for the expression of TE-specific transcription factor Cdx2 (Home et al., 2012). Our results also have shown the TE-specific expression of the Cdx2 protein (Figures 3A,C), whereas Tead4 expression is present in both TE and ICM lineages (Figures 3B,D). We have also found that both lncRNAs Platr4 and Malat1 are expressed in the TE and ICM lineages (Figures 3A–D, Supplementary Figures S2A, B).

We further sought to detect multiple RNA transcripts simultaneously in the three-dimensional context of preimplantation mouse embryos. Therefore, we have used two different RNA transcripts (lncRNA and mRNA) smRNA FISH probes in combination with either Tead4 or Cdx2 protein (Figures 4A,B). For the visualization of multiple RNA transcripts, we have shown expression of either Platr4 or Malat1 lncRNAs and Pou5f1 mRNA, which are expressed in the ICM of mouse blastocysts (Ohta et al., 2008). In addition, we showed detection of two mRNA transcripts (Pou5f1 and Polr2a) in combination with Tead4 protein (Supplementary Figure S3). Thus, our combined methods successfully detected protein, lncRNA, and mRNA in the same sample, indicating that this method is robust and suitable for visualizing multiple biological molecules in the three-dimensional context of developing mouse embryos.