Methods for In Situ Detection of DNA and RNA

Cross Reference to Related Application

This application claims priority to U.S. Provisional Application No. 62/860,366, filed June 12, 2019, the disclosure of which is incorporated by reference herein in its entirety.

Incorporation of Sequence Listing

This application includes a Sequence Listing which is being submitted in ASCII format via EFS- Web, named "DVC001PCT_ST25.txt,” which is 693 Bytes in size and created on June 11, 2020. The contents of the Sequence Listing are incorporated herein by reference in their entirety.

Background

RNA transcriptomics has attracted great interest due to its implication in many aspects. Its profiling in real tissue including mammalian tissue is especially important for understanding its function in real biological system. However, the complexity and various components included in tissue like lipids incur high autofluorescence and non-specific binding of probes such as fluorescent probes based on fluorescence in situ hybridization (FISH) techniques to detect the RNA within tissue. Thus, tissue clearing techniques to remove lipids and eliminate

autofluorescence are required when fluorescence imaging is used to profile single-cell transcriptomics in tissue for both high- abundance and low-abundance RNA.

Hydrogel based tissue clearing methods like CLARITY and Expansion Microscopy have been used for high performance RNA FISH imaging (Chen, Wassie et al. 2016, Sylwestrak, Rajasethupathy et al. 2016). However, currently, single molecule RNA FISH imaging with tissue clearing suffers from very slow hybridization speed, typically requiring overnight (Raj, van den Bogaard et al. 2008, Lubeck, Coskun et al. 2014, Chen, Wassie et al. 2016, Sylwestrak,

Rajasethupathy et al. 2016). MERFISH takes 18-36 hours for the oligo pool probes to bind to their RNA targets in cultured cells (Chen, Boettiger et al. 2015). MERFISH and seqFISH+ with primary tissues embedded in hydrogel even requires 40-60h for the probe-RNA hybridization to complete (Moffitt, Hao et al. 2016, Eng, Lawson et al. 2019).

For DNA FISH, existing methods typically require hours to overnight hybridization as well. For example, traditional DNA FISH methods using bacterial artificial chromosomes to generate probes require overnight hybridization to achieve sufficient signal. SureFISH and IQFISH requires 1-2 hours or longer for the hybridization to complete (Viale, Paterson et al. 2016, Sugita, Aoyama et al. 2017, Takeda, Kasai et al. 2017). Although using unnatural DNA probes such as peptide nucleic acids (PNA) can increase the hybridization speed significantly (15 min), the high cost of PNA probes limits their use in molecular diagnostics mainly to detecting ribosome RNAs in pathogens in blood culture (Kempf, Trebesius et al. 2000, Stender 2003,

Deck, Anderson et al. 2012, Frickmann, Zautner et al. 2017).

Current single molecule oligo DNA/RNA FISH technologies without signal

amplification can only achieve limited genomic resolution (typically >=lkb) when high detection efficiency (90-100%) is desired. For example, single molecule RNA FISH, which uses >=20 20- nt, single-stranded and single-labeled DNA oligonucleotide probes complementary to various regions of a target RNA (>=lkb), is optimized to achieve >90% detection efficiency (Raj, van den Bogaard et al. 2008, Lubeck and Cai 2012, Annaratone, Simonetti et al. 2017, Busse, Paroni et al. 2017). A FISH technology with better genomic resolution is required to detect short DNA and RNA sequences of <lkb. For higher genomic resolution of DNA and RNA FISH, decreasing background and also increasing the detection signal are both important. Several in-situ detection approaches have been suggested to increase genomic resolution by either decreasing background or increasing the detection signal but none of them can be used to detect short DNA and RNA sequences of <lkb reliably and also efficiently. Single molecule RNA FISH such as Stellaris FISH (LGC Biosearch Technologies) uses multiple copy of single dye labeled DNA oligos to detect single copy of RNA in situ at ~lkb genomic resolution but takes long time to implement (usually 4-20 hours) (Raj, van den Bogaard et al. 2008). Other methods such as RNAScope (Advanced Cell Diagnostics), BaseScope (Advanced Cell Diagnostics), PLISH, and in situ HCR, require signal amplification to detect DNA/RNA sequences of <lkb and also take multiple hours to complete the assay (Wang, Flanagan et al. 2012, Crosetto, Bienko et al. 2015, Shah, Lubeck et al. 2016, Nagendran, Riordan et al. 2018). Recently, tissue clearing was used for high

performance and multiplex RNA imaging in situ but due to the limitation of the probe design, limited genomic resolution (3-6 kb) was achieved (Chen, Boettiger et al. 2015, Moffitt, Hao et al. 2016).

Summary In one aspect, the present disclosure provides a method for nucleic acid FISH, comprising the following steps, in sequence: heat denaturing a sample which has been attached to a scaffold; embedding the (denatured) sample in a gel; anchoring nucleic acid targets in the sample to the gel; clearing non-targets from the sample; adding a first plurality of probes for at least one (or some) of nucleic acid targets; and detecting an optical signal from the first plurality of probes.

In the method, the denaturing temperature can be between 40°C and 80°C, e.g., between 45°C and 75°C, between 70°C and 75°C, between 50°C and 60°C, or between 60°C and 70°C. The scaffold can be a glass having a thickness of smaller than 1 mm, e.g., between 0.1 and 0.2 mm, or between 0.2 and 0.3 mm. In certain examples, the glass can have a thickness of smaller than 0.17 mm.

In some embodiments of the method, each of the first plurality of probes has at least one optical label, such as fluorescent dyes and/or nanoparticles.

In some embodiments, the method further comprises adding a plurality of intermediate nucleic acid probes to the sample after heat denaturation but before gel embedding.

In some embodiments, the method further comprises adding a plurality of intermediate nucleic acid probes to the sample after clearing non-targets from the sample but before adding the first plurality of probes.

In some embodiments of the method, the first plurality of probes comprise single strand oligonucleotides. In some other embodiments, the first plurality of probes comprise peptides. Yet in certain embodiments, the probes comprise peptide and nucleic acid hybrids.

In some embodiments, the method further comprises the following steps, in sequence: stripping the first plurality of probes from the sample after the detection, adding a second plurality of probes for at least one nucleic acid target in the stripped sample, and detecting an optical signal from the second plurality of probes. The stripping temperature can be 60°C or lower.

In some embodiments, the method further comprises the following steps, in sequence: stripping the first plurality of probes after the detection; heat redenaturing the sample; adding a second plurality of probes for at least one nucleic acid targets in the redenatured sample; and detecting the optical signal from the second plurality of probes. The stripping temperature can be 60°C or lower. In these embodiments, any of the first plurality of probes for the RNA target, the second plurality of probes for the second RNA target, and the second plurality of probes for the DNA target, can be shorter than 150 nt, or shorter than 100 nt, or shorter than 50 nt. Any of the first plurality of probes for the RNA target, the second plurality of probes for the second RNA target, and the second plurality of probes for the DNA target can include at least 2 different

oligonucleotides.

In another aspect, the present disclosure provides a method for DNA FISH, comprising the following steps in sequence: embedding a sample in a gel; anchoring DNA targets in the sample to the gel; clearing non-targets from the sample; heat denaturing the sample; adding a first plurality of probes with optical labels (e.g., fluorescent dyes or nanoparticles) to the sample; and detecting an optical signal from the first plurality of probes. The first plurality of probes can comprise nucleic acids or peptides, or peptide and nucleic acid hybrids. In some embodiments, the method can further include adding a plurality of intermediate nucleic acid probes to the sample after the heat denaturation but before adding the first plurality of probes.

In another aspect, the present disclosure provides a method for nucleic acid FISH, comprising the following steps, in sequence: denaturing a sample attached on a scaffold by heating; adding a first plurality of probes to the sample; detecting an optical signal from the first plurality of probes; removing the first plurality of probes; redenaturing the sample by heating; adding a second plurality of probes to the redenatured sample; and detecting an optical signal from the second plurality of probes. In some embodiments, the method further comprises the following steps, in sequence: removing the second plurality of probes from the sample; adding a third plurality of probes to the sample; and detecting an optical signal from the third plurality of probes. In certain embodiments, the method further comprises the following steps, in sequence: removing the second plurality of probes; redenaturing the sample by heating; adding a third plurality of probes to the sample; and detecting an optical signal from the third plurality of probes. The redenaturing temperature can be between 40°C and 70°C, e.g., below 60 °C but higher than 50 °C.

Brief Description of the Drawings

Figure 1: Four probe design schemes a) a set of single fluorophore labeled oligo probes for regular single molecule FISH. Each probe can be natural DNA oligo, RNA, LNA, or PNA oligo, etc. b) a set of multiple fluorophore labeled DNA probes. Each probe contains multiple fluorophores. Each probe may consist of a hybridization sequence only, or a hybridization sequence and 1-2 readout sequences at either or both ends of the hybridization sequence c) a set of hybrid probes. Each probe contains a hybridization sequence made of PNA or LNA and a readout sequence made of DNA or RNA. d) a Holliday Junction probe with multiple

fluorophores.

Figure 2: Temperature gradients for rapid and sequential DNA/RNA FISH.

Figure 3: Schemes of Rapid DNA/RNA FISH with hydrogel based tissue clearing.

Figure 4: Rapid RNA FISH in Cultured HeFa Cells with different denaturation temperature. (a)-(b) Confocal imaging of TFRC mRNA labeled Hela Cells with heat denaturation temperature at 90°C for 10 min. (a). Signals of TFRC mRNA with an enlarged region (b).

(c)-(d) Confocal imaging of TFRC mRNA labeled Hela Cells with heat denaturation temperature at 85°C for 10 min. (c). Signals of TFRC mRNA with an enlarged region (d).

(e)-(f) Confocal imaging of TFRC mRNA labeled Hela Cells with heat denaturation temperature at 80°C for 10 min. (e). Signals of TFRC mRNA with an enlarged region (f).

(g)-(h) Confocal imaging of TFRC mRNA labeled Hela Cells with heat denaturation temperature at 75°C for 10 min. (g). Signals of TFRC mRNA with an enlarged region (h).

(i)-(j) Confocal imaging of TFRC mRNA labeled Hela Cells with heat denaturation temperature at 70°C for 10 min. (i). Signals of TFRC mRNA with an enlarged region (j).

Figure 5: Rapid and sequential RNA FISH in cultured HeFa cells. The imaging of five rounds of RNA FISH were taken in the same area.

(a)-(b) Round-1: Confocal imaging of GAPDH mRNA labeled Hela Cells (a). Signals of GAPDH mRNA with an enlarged region (b).

(c)-(d) Round-2: Confocal imaging of NEAT1 FncRNA labeled Hela Cells (c). Signals of NEAT1 FncRNA with an enlarged region (d).

(e)-(f) Round-3: Confocal imaging of MAFAT1 FncRNA labeled Hela Cells (e). Signals of MATAT1 FncRNA with an enlarged region (f). (g)-(h) Round-4: Confocal imaging of TFRC mRNA labeled Hela Cells (g). Signals of TFRC mRNA with an enlarged region (h).

(i)-(j) Round-5: Confocal imaging of POLR2A mRNA labeled Hela Cells (i). Signals of POLR2A mRNA with an enlarged region (j).

(k)-(l) DAPI image (k) and bright-field image (1) of the same area.

Figure 6: Rapid RNA and DNA FISH Sequentially in Cultured HeLa Cells. The imaging of four rounds of RNA FISH and DNA FISH were taken in the same area.

(a)-(b) Round- 1: Confocal imaging of TFRC mRNA labeled Hela Cells (a). Signals of TFRC mRNA with an enlarged region (b).

(c)-(d) Round-2: Confocal imaging of GAPDH mRNA labeled Hela Cells (c). Signals of GAPDH mRNA with an enlarged region (d).

(e)-(f) Round-3: Confocal imaging of Telomere labeled Hela Cells (e). Signals of Telomere with an enlarged region (f).

(g)-(h) Round-4: Confocal imaging of Centromere labeled Hela Cells (g). Signals of Centromere with an enlarged region (h); (i)-(j) DAPI image (i) and bright-field image (j) of the same area.

Figure 7: Confocal imaging of GAPDH probe labeled MEF cells.

(a) Cells were processed with heat denaturation at 75°C, with tissue clearing, and with probe labeling for 2 hr.

(b) An enlarged region from (a).

(c) Cells were processed without heat denaturation, but with tissue clearing and probe labeling for 20 hr.

(d) An enlarged region from (c).

(e) Cell were processed without denaturation and without tissue clearing, but with probe labeling for 20 hr.

(f) An enlarged region from (e).

Figure 8: Confocal imaging of TFRC probe labeled mouse liver tissue. (a)-(b) Confocal imaging of TFRC probe labeled mouse liver tissue with tissue clearing and also a heat denaturation step at 75°C before tissue clearing. The hybridization step was done at 37°C for lh. (a) Signals of TFRC mRNA with an enlarged region in (b).

(c)-(d) Confocal imaging of TFRC probe labeled mouse liver tissue with tissue clearing but without heat denaturation before tissue clearing. The hybridization step was done at 37°C for lh, (c) Signals of TFRC with an enlarged region in (d).

(e)-(f) Confocal imaging of TFRC probe labeled mouse liver tissue with tissue clearing but without heat denaturation before tissue clearing. The hybridization step was done at 37°C for 16h. (e) Signals of TFRC with an enlarged region in (f).

(g)-(h) Confocal imaging of TFRC probe labeled mouse liver tissue with heat denaturation and without tissue clearing. The hybridization step was done at 37°C for 1 h. (g) Signals of TFRC with an enlarged region in (h).

Figure 9: Confocal imaging of POLR2A probe labeled mouse liver tissue.

(a)-(b) Confocal imaging of POLR2A probe labeled mouse liver tissue with tissue clearing and also a heat denaturation step at 75°C before tissue clearing. The hybridization step was done at 37°C for 0.5h. (a) Signals of POLR2A mRNA with an enlarged region in (b).

(c)-(d) Confocal imaging of POLR2A probe labeled mouse liver tissue with tissue clearing but without heat denaturation before tissue clearing. The hybridization step was done at 37°C for 0.5h. (c) Signals of POLR2A with an enlarged region in (d).

(e)-(f) Confocal imaging of POLR2A probe labeled mouse liver tissue with tissue clearing and with heat denaturation before tissue clearing. The hybridization step was done at 37°C for 16h. (e) Signals of POLR2A with an enlarged region in (f).

(g)-(h) Confocal imaging of POLR2A probe labeled mouse liver tissue with heat denaturation and without tissue clearing. The hybridization step was done at 37°C for 0.5 h. (g) Signals of POLR2A with an enlarged region in (h).

Figure 10: Different heat denaturation temperature for POLR2A in mouse liver primary tissue done with tissue clearing (a) 50°C heat denaturation for 10 min. (c) 60°C heat denaturation for 10 min. (e) 70°C heat denaturation for 10 min. (g) 80°C heat denaturation for 10 min. (b), (d), (f),

(h) are zoomed-in regions from (a), (c), (e), (g), respectively. Figure 11: Confocal imaging of POLR2A probe labeled mouse liver tissue with hydrogel based tissue clearing. Figure (a-b) was obtained with a heat denaturation step at 75°C for 5 min before tissue clearing. Figure (c-d) was obtained with a heat denaturation step at 75°C for 10 min before tissue clearing. Figure (e-f) was obtained with a heat denaturation step at 75°C for 20 min before tissue clearing. Figure (g-h) was obtained with a heat denaturation step at 75°C for 10 min after tissue clearing.

Figure 12: Confocal imaging of POLR2A labeled mouse liver tissue with heat denaturation at 75 °C before tissue clearing and different probe hybridization temperature.

(a) A hybridization step at room temperature (~25 °C) for 0.5h with a zoomed-in region in (b). (c) A hybridization step at 37 °C for 0.5h with a zoomed-in region in (d).

(e) A hybridization step at 47 °C for 0.5h with a zoomed-in region in (f).

Figure 13: Confocal imaging of rapid and sequential RNA FISH in mouse liver tissue. Images were taken from the same tissue section area.

(a) First round, POLR2A probes labeled tissue with a zoomed-in region in (b).

(c) Second round, Neatl labeled tissue with a zoomed-in region in (d). The brightest puncta are Neatl clusters.

(e) Third round, Ctnnbl labeled tissue with a zoomed-in region in (f).

(g) Fourth round, Notchl labeled tissue with a zoomed-in region in (h).

(i) Fifth round, TBP labeled tissue with a zoomed-in region in (j).

Figure 14: Confocal imaging of DNA FISH in mouse liver tissue with heat denaturation at 75°C after tissue clearing and with a hybridization step at 37°C. (a) Signals of Telomere with an enlarged region in (b). (c) Signals of Centromere with an enlarged region in (d).

Figure 15: Confocal imaging of rapid and sequential RNA FISH and DNA FISH in mouse liver tissue. Images were taken from the same tissue section area (a) First round, Neatl probes labeled tissue with a zoomed-in region in (b). Neatl clusters are shown as bright puncta in the cell nucleus. (c) Second round, Ctnnbl labeled tissue with a zoomed-in region in (d).

(e) Third round, Telomere labeled tissue with a zoomed-in region in (f).

(g) Fourth round, Centromere labeled tissue with a zoomed-in region in (h).

Figure 16: Confocal imaging of rapid and sequential DNA FISH in mouse liver tissue. Images were taken from the same tissue section area.

(a) First round, Telomere-Cy5 labeled tissue with a zoomed-in region in (b).

(c) Second round, Centromere-Cy3 labeled tissue with a zoomed-in region in (d).

(e) Third round, Centromere-Cy5 labeled tissue with a zoomed-in region in (f).

(g) Fourth round, Telomere-Cy3 labeled tissue with a zoomed-in region in (h).

Detailed Description of Certain Embodiments of the Invention

The present disclosure generally provides methods of in-situ DNA/RNA detection. In some aspects, the methods include: heat denaturing a sample attached to a scaffold; embedding the (denatured) sample in a gel; anchoring (or chemically linking) nucleic acid targets in the sample to the gel; clearing non-targets from the sample; adding a plurality of probes for at least one (or some) of nucleic acid targets; and detecting an optical signal(s) from the plurality of probes. Other aspects of the invention and various embodiments of the methods are described below in conjunction with accompanying drawings.

1. Oligonucleotide RNA and DNA FISH

Natural oligonucleotide probes such as DNA or RNA probes can be used for RNA and DNA FISH here (Figure la). For example, a DNA oligo probe can be used to detect a DNA or RNA target; an RNA oligo probe can be used to detect a DNA or RNA target. For one RNA or DNA target, typically, a plurality of oligo probes comprising at least 2 oligonucleotide probes are designed and synthesized. For multiplex RNA or DNA FISH, a number of oligo probes for one or multiple targets are hybridized with their targets in a sample at the same time. For sequential RNA or DNA FISH, a number of oligo probes (>=2) for one or multiple targets are used in each round of FISH. Probes are bound to nucleic acid targets in a sample directly or indirectly. At least one of a set of probes for a target is labeled by an optical label or multiple optical labels.

The probes are allowed to hybridize with the nucleic acids in the sample in suitable conditions and if target nucleic acids are present in the sample, the probes are expected to bind with the target nucleic acids. Excess probes are then removed. Optical signals from the probes left in the sample (including those probes that are bound to the target nucleic acids and probes that are non- specifically bound to the sample) are then detected by optical microscopy or spectroscopy. From the signals, the presence of the target RNA or DNA sequences or species in the sample can be determined. In addition, if using optical microscopy, the sequence information and the location of the nucleic acid targets can be determined at the same time.

Unnatural nucleic acid based oligo probes or oligonucleotide probes made of a mixture or hybrid of natural nucleic acids and unnatural nucleic acids can also be used for DNA and RNA FISH here (Figure la and lc). Unnatural nucleic acids include, but not limited to, Focked Nucleic Acids (UNA) and Xenonucleic Acids (XNA). They have higher binding affinity to DNA and RNA targets than natural DNA and RNA, which are useful for achieving higher genomic resolution of FISH imaging.

To label multiple fluorophores per oligonucleotide, KREATECH universal linkage system (Feica Biosystems) can be used (Figure lb). This method uses a platinum dye complex to form a stable adduct with the N7 position of guanine and, to a lesser extent, adenine bases in nucleic acids. In this way, every 15-20 nucleotides can be labeled with one fluorophore without obvious quenching.

Fabeling by Holliday junction probes: some variations on oligo design can be used for high resolution FISH here, such as using Holliday junction principle to design probes (Figure Id). One Holliday junction probe set consists of two probes. A set of probes for a DNA/RNA target consists of at least one Holliday junction probe set. Each probe set comprises: i) a first probe comprising a 5' overhang region and a region that hybridizes to the target nucleic acid at a first target site; ii) a second probe comprising a 3' overhang region and a region that hybridizes to the target nucleic acid at a second target site. Each probe set can be labeled with multiple dyes. The two probes for a probe set can be labeled with the same or different dyes. KREATECH labeling can be used for multi-dye conjugation. DNA, RNA or ENA oligos can be used as the hybridization sequences of the Holliday junction probes.

Scaffold for sample attachment: A solid scaffold (or sample support, substrate) can be used so that samples such as tissue and cells are attached onto the scaffold. The scaffold can be thin cover glasses or thick glass slides. Then probes are added onto the sample to hybridize with DNA or RNA targets on the scaffold. For rapid RNA FISH, typically thin cover glasses such as # 1.5 coverslips are required for efficient heat transfer and fast denaturation of RNAs in the sample. Cover slides which are much thicker (typically ~lcm) than cover glasses are not good for rapid RNA FISH. When doing heat denaturation through thick glass slides, oligo probes may not localize to the right RNA targets in the sample due to the much slower heating speed of the thick glass.

Detecting FISH signal without scaffold: samples can also be stained with nucleic acid probes without attaching onto a scaffold. For example, cell or tissue samples can be suspended in a solution. Probes with optical labels can hybridize with nucleic acid targets in suspended cells. Finally, the probe signals from suspended cells can be detected by various instruments such as flow cytometry.

2. Rapid and Sequential DNA/RNA FISH by 3-5 temperature gradients in cultured cells without tissue clearing:

It has been found that to improve the assay performance such as better hybridization efficiency and lower the non-specific binding, another two temperature gradients can be added: washing in a heated buffer and heat redenaturation after probe removal and before next-round of probe hybridization. For example, a heat redenaturation between 35-50 °C can be used for sequential RNA FISH; alternatively, a heat redenaturation between 50-60 °C can be used for sequential RNA FISH; alternatively, a heat redenaturation between 50-70 °C can be used for sequential DNA FISH; a heat redenaturation between 70-90 °C can be used for sequential DNA FISH. Therefore, for rapid and sequential DNA/RNA FISH , 3-5 temperature gradients can be used: heat denaturation (temperature- 1), hybridization in a heated buffer (temperature-2), washing in a heated buffer (temperature-3), heat removal of probes (temperature-4), heat redenaturation (temperature- 5) before next round of probe hybridization (Figure 2). Among these temperature gradients, the initial heat denaturation to a sample (temperature- 1) is required. For Temperature 2-5, 2-4 of them are selected and combined depending on three factors: the cell and tissue type, the locus of the DNA/RNA target in a cell, the intrinsic physical and chemical properties of the targeted DNA or RNA sequence such as the GC content. For example, depending on the targeted sequences, a mild heat redenaturation (35-70 °C) and short duration (2-10min) can be added after probe stripping during sequential DNA/RNA FISH; non-specific probe binding can be washed in a buffer at 37-50 °C after probe hybridization; for a mix of RNA FISH and DNA FISH together with the same sample, typically RNA FISH is done ahead of DNA FISH, and meanwhile heat redenaturation is added after RNA FISH and before the following DNA FISH.

3. Using peptide as labeling probes

Novel probe design of oligo FISH by peptide is disclosed here that achieves <1 kb genomic resolution of DNA and RNA FISH. Typically, Peptide Nucleic Acids (PNA), non- nucleic acid based oligos, are used here (Figure la and lc). PNAs are useful to image short DNA locus, small RNA transcripts, or even single exons of an RNA gene, as they have higher binding affinity towards natural nucleic acid targets than natural oligonucleotide probes. They can be used as probes to replace natural nucleic acid based oligo probes to increase the number of probes and fluorophores available for a DNA/RNA target. In other words, higher genomic resolution can be achieved with less and shorter probes. For example, an RNA target of 0.5kb long can be detected by 10-20 PNA probes with single dye on each peptide probe, while the same number of single-dye labeled oligonucleotide probes may not be able to detect the same target reliably; alternatively, 5-10 PNA probes with double dyes on each peptide probe can be used. Usually, at least two such probes are used to detect each nucleic acid target. Genomic resolution can be increased by shorter hybridization sequence per probe, less oligo probe per target, and more fluorophores per probe. For example, a scheme of probes with both a hybridization sequence and a readout sequence per probe can be used (Figure lc). Peptide nucleic acids (PNA) can be used to design the hybridization sequences as PNA have higher binding affinity for DNA and RNA targets than natural nucleic acids (Figure lc). Therefore, shorter oligo probes can be used. DNA oligos can be used to make readout sequences and label multiple fluorophores onto each readout sequences by the KREATECH universal linkage system. Further, PNA based hybridization sequences with a linker (such as azide or alkyne chemical group) at 5’ or 3’ end of each probe and DNA based readout sequences can be synthesized separately. Lastly, PNA oligos and DNA oligos can be conjugated together through site-specific conjugation chemistry such as azide-alkyne or tetrazine-vinyl based click chemistry. PNA oligos can also be used as the hybridization sequences of the Holliday junction probes.

The signal from peptide probes can be detected in the same approach used for oligonucleotide probes.

4. Rapid DNA and RNA FISH with Tissue Clearing

Tissue clearing technique can be combined with rapid RNA FISH, a fast and convenient RNA detection method, to achieve fast and accurate RNA detection in real tissue. While other techniques developed by the academia and industry in profiling RNA

transcriptomics in tissue normally require overnight to 3 days hybridization, the techniques of the present invention only require 1 hour or even less than 1 hour to complete the hybridization. A key step is to denature the tissue sample before the tissue clearing procedures, thus ensuring the fast probe hybridization after tissue clearing.

Not just for RNA imaging, the rapid FISH technology disclosed in the present invention can also be used to detect DNA sequences in cells and tissues such as a gene locus of chromatins and chromosomes in mammalian cells and tissues after tissue clearing. By adding the heat denaturation step after tissue clearing, rapid DNA FISH was successfully achieved in primary tissues. As tissue clearing with protein digestion can potentially increase the chromatin accessibility as well, the speed of DNA FISH could also be accelerated, comparing with that without tissue clearing.

Described below are certain aspects for the implementation of rapid DNA and RNA FISH with gel based tissue clearing according to embodiments of the present invention:

4.1. Labeling and sample preparation schemes: As shown in Figure 3, multiple schemes can be used for rapid RNA/DNA FISH and rapid and sequential RNA/DNA FISH. For rapid RNA FISH, typically heat denaturation is done before tissue clearing (Figure 3a and 3b). For direct labeling of probes in rapid RNA FISH, probes are hybridized to the RNA targets after tissue clearing (Figure 3a). For indirect labeling using intermediate probes such as primary probes (i.e. target binding probes) without optical labels, primary probes can be hybridized either before or after tissue clearing (Figure 3b). For rapid DNA FISH, probes are typically hybridized to the DNA targets after tissue clearing (Figure 3c). For rapid and sequential RNA and DNA FISH together, RNA FISH is done before DNA FISH to minimize RNA degradation and mis-localization of RNA targets (Figure 3d).

4.2. Gel based tissue clearing: tissue clearing was done in a gel based approach (Figure 3).

4.2.1. A variety of polymers can be used to make the gel such as a hydrogel made of polyacrylamide or agarose. For example, a polyacrylamide hydrogel can be polymerized from acrylamide and bis-acrylamide. Alternatively, the gel synthesis approach for Expansion Microscopy can be used. This modified recipe adds a third molecule sodium acrylate into the gel synthesis formula to turn the polyacrylamide hydrogel into an expandable gel. Many other polymers can be used in various embodiments that polymerize small molecules (gel subunits) to form a gel. These small molecules include but not limited to acrylic acid, ethylene glycol diacrylate, ethylene glycol dimetharcrylate, poly(ethylene glycol dimethacrylate).

4.2.2. Clearing non-targets in the sample: This step removes various biomolecules but not nucleic acids, such as proteins, lipids, and other cell/tissue components from the sample. As with other RNA FISH methods with tissue clearing, a protein digestion and lipid removal step is used after gel synthesis to decrease the non-specific binding of oligo probes. This step uses proteinase such as proteinase K or pepsin to digest protein molecules. Lipid removal reagents such as SDS (sodium dodecyl sulfate), LDS (lithium dodecyl sulfate). This step is also helpful for faster probe diffusion and probe binding to the target.

4.2.3. Anchoring probes: A small molecule linker can be used to link nucleic acids to the gel. For example, a small molecule linker called LabelX can be used to covalently link RNA molecules to the gel. It is synthesized from two building blocks, Label- It Amine (MimsBio) and 6-((Acry]oy])amino)hexanoic acid

(Acryloyl-X SE, here abbreviated AcX, Life Technologies). The former building block contains both an amine as well as an alkylating group that primarily reacts to the N7 of guanine, and the later contains an amine-reactive succtnamide ester and a polymerizable acrylamide moiety. Alternatively, an anchor probe that links the polyA tail of RNA molecules to the gel can be used. It contains a poly-dT group to hybridize with the polyA tail. To facilitate stronger binding, the poly-dT group can be synthesized partially or ail with nucleic acid derivatives such as locked nucleic acids (LNA).

4.3. Heat denaturation for RNA FISH in hydrogel based tissue clearing: Adding a heat

denaturation step before tissue clearing could speed up the probe hybridization process in samples done with tissue clearing significantly: the probe-target hybridization time was shortened from overnight or longer to 10-60 min.

4.3.1. Depending on the sample type and the locus of RNA species, the temperature of heat denaturation can be set between 40-80 °C. Adding a heat denaturation step at 70-75 °C for 5-10 min in a 50-90% formamide buffer before tissue clearing works best for mouse primary tissues such as liver and lung. Alternatively, heat denaturation can be done lower than 70°C but with extended heating time. For example, heating at 60-65 °C for 10-20 min or at 50-55 °C for 20-30min can also speed up the probe hybridization process by 5-10X. The opposite way doesn’t work: adding the same heat denaturation step after tissue clearing leads to non-specific probe binding and aggregation in the cellular nucleus. If using heat denaturation temperature >80 °C, RNA in tissue samples tends to degrade so that the RNA detection efficiency decreases and meanwhile non-specific binding in nucleus increases significantly. For heat sensitive biological samples, heating temperature and/or heating time may need to be reduced to avoid over-heating and non-specific binding.

4.3.2. The denaturing time is typically 5-10 min at 70-75 °C with a formamide

denaturation buffer. Denaturing too short results in low detection efficiency.

Denaturing too long results in non-specific probe binding and aggregation in the cellular nucleus.

4.3.3. 1-2 rounds of Heat denaturation: Typically, only one round of heat denaturation is used before tissue clearing on a biological sample. For some embodiments, if primary probes or target-binding probes are added after tissue clearing, a second round of heat denaturation can be added after tissue clearing but before primary probe hybridization. The heating temperature of the second round should be 10-20 °C lower than the first round to avoid over-heating and non-specific binding.

4.3.4. Instead of using formamide as the solvent of the denaturation buffer, using

ethylene carbonate as the denaturation buffer can significantly lower down the denaturation temperature to <70 °C which are important to decrease the non-specific binding of probes and increase signal to noise (or background) ratio of FISH imaging. For example, denaturation temperature can be set in-between 40-70 °C for 5-10 min; alternatively, denaturation temperature can be set between 60 and 65 °C for 5-10 min; alternatively, denaturation temperature can be set at 62 °C for 5-10 min, or 3-8 min, or 8-15 min; denaturation temperature can be set at 55 °C for 5-10 min; denaturation temperature can be set at 50 °C for 5-15 min;

4.4. Heat denaturation after tissue clearing for DNA FISH: For RNA FISH, heat denaturation is typically done at below 80 °C and even below 70 °C. However, for DNA FISH, higher denaturation temperature (80 °C or higher) is usually used without worrying about DNA degradation.

4.5. Heating transfer condition and scaffold: Usually, samples such as tissue and cells are attached onto solid scaffold such as thin cover glasses or thick glass slides. It is found that fast heating transfer rate is important for efficient and reliable heat denaturation.

This is essential for RNA FISH in tissue cleared sample. For RNA FISH in embodiments of the present invention, #1.5 coverslips (~0.17mm thick) are used instead of thick glass slides (1mm thick) for tissue attachment and heat denaturation to make sure heat transfer is efficient. Coverslips are put directly onto a heated dry bath without air separation when doing heat denaturation. When using thick glass slides, probes for RNA FISH are often mis-localized to the cell nucleus due to slower heat transfer rate.

4.6. Oligo Probes: In certain embodiments, natural nucleic acid based oligo nucleotides are used in the present invention. However, for higher genomic resolution and/or higher signal to noise (S/N) imaging, unnatural nucleic acid such as locked nucleic acid (LNA) or peptide such as peptide nucleic acid (PNA) can be used instead of natural nucleic acid as the backbone of oligo probes for hybridization. Alternatively, hybrid oligo probes which are made of peptide nucleic acids and natural nucleic acids are used. When combining these higher affinity probes with tissue clearing, better genomic resolution and better S/N can be achieved.

4.7. Timing for adding probes for probe-target hybridization: For traditional oligo FISH

schemes using probes labeled with dyes, probes are added into the sample for DNA or RNA target detection after heat denaturation and also after tissue clearing. For an oligo FISH scheme like MERFISH and seqFISH+ using primary probes and secondary probes, primary probes without dyes are added in-between heat denaturation and tissue clearing, and secondary probes are added after tissue clearing; alternatively, both primary probes and secondary probes are added after tissue clearing.

4.8. Probe length: In general, shorter probes allow for faster diffusion and target

hybridization. For example, probes of <150 nt are used; alternatively, probes of <100 nt are used; alternatively, probes of <50 nt are used; alternatively, probes are 20-40 nt each; alternatively, probes are 10-20 nt each. An oligo probe may contain the hybridization sequence only; alternatively, an oligo contains a hybridization sequence and a readout sequence at 5’ or 3’ end of the hybridization sequence; alternatively, an oligo contains a hybridization sequence and two readout sequences at both 5’ and 3’ end of the hybridization sequence; alternatively, a probe set of a DNA or RNA target contains some oligos with hybridization sequences only and some oligos with both hybridization sequences and readout sequences. The length of hybridization sequences varies. In some embodiments, the hybridization sequences are 10-20 nt; in other embodiments, they are 20-40 nt; in other embodiments, they are 30-50 nt.

4.9. Optical labeling:

4.9.1. Optical labels: a variety of optical labels can be used here, such as fluorescent dyes, nanoparticles, quantum dots, etc. Usually, the fluorescence intensity from optical labels are detected by detectors such as EMCCD, sCMOS camera, PMT. Alternatively, the fluorescence lifetime of optical labels is detected by a

spectrofluorometer or an imaging microscopy scheme. .9.2. Direct labeling: direct labeling here means hybridization probes are conjugated with optical labels such as fluorescent dyes before its hybridization with DNA or RNA target. .9.3. Indirect labeling: indirect labeling here means using intermediate probes without optical labels for hybridization. For example, primary probes without dyes

(intermediate probes) are used when hybridizing with the DNA or RNA target in a sample. After this, secondary probes with dyes are added to detect the probes without dyes. Multiple intermediate probes without dyes can be used and finally readout probes with dyes are added to recognize intermediate probes. For example, similar to MERFISH and seqFISH+, primary probes without dyes are hybridized with their target first after heat denaturation, then secondary probes with dyes are added to recognize the primary probes. Alternatively, dyes are conjugated with the probes with linkers through site-specific reaction such as click-chemistry or physical interaction such as biotin- strep tavidin conjugation after the probe-target

hybridization step. . Temperature gradients for rapid and sequential RNA (and/or DNA) FISH with tissue clearing: .10.1. For rapid sequential DNA/RNA FISH, 1-5 temperature gradients are used: heat denaturation (temperature- 1), hybridization in a heated buffer (temperature-2), washing in a heated buffer (temperature-3), heat removal of probes (temperature-4), heat redenaturation(temperature-5) before next round of probe hybridization.

Among these temperature gradients, heat denaturation (temperature- 1) is required. Temperature 2-5 are optional, depending on the gel synthesis conditions, sample types, RNA and DNA species to be detected. For example, for probe hybridization, alternatively, hybridization at 30-40 °C can be used. For the washing condition after probe hybridization, heated washing at 37-50 °C after probe hybridization can be used; alternatively, washing at room temperature without heating can be used. For probe stripping, heated stripping at 40-60 °C can be used to remove probes for subsequent round of FISH; alternatively, stripping in a high concentration of formamide buffer (>=30%) should be used without heating. Mild heat redenaturation (30-60 °C) can be added after heat removal of probes (probe stripping) and before next round of sequential FISH. For DNA FISH after RNA FISH, heat redenaturation at 50-90 °C is usually used after the very last round of sequential RNA FISH and before the first round of sequential DNA FISH. 4.10.2. To decrease non-specific binding, washing in a heated buffer (30-50 °C)

immediately after each round of probe hybridization can be added before imaging.

4.11. Hybridization Buffer composition: Depending on the tissue sample, the locus of the target and the detection efficiency desired, dextran sulfate can be removed or added into the hybridization buffer to further promote probe binding. For example, the

concentration can be 0, 1%, 5%, 10%, 15%, or 20%.

Examples

Example 1 Rapid and sequential RNA FISH in cultured cell lines without tissue clearing:

Experiment:

A rapid RNA FISH on cell line without tissue clearing was performed as follows:

1) Hela cells were seeded into #1.5 glass bottom 8-well chambers (Ibidi) for heat denaturation and hybridization.

2) Cells were fixed in 4% PFA for 12mins and then in 70% ethanol at RT (room

temperature) for 5min.

3) The cells were submitted to 75 °C denaturation in 80% formamide, 2X SSC for 10 min in a dry bath (D1302, Labnet Digital Dry Bath, dual block).

4) After denaturation, the cells were hybridized with 200 uL probe buffer containing 10% formamide, 2X SSC, 20% dextran sulfate and 100 nM RNA probes for lOmin at 37 °C.

5) After washing in 2X SSC at 50 °C, or 20% formamide in 2X SSC at 37 °C the sample was proceeded to imaging. 6) To strip off the probes, the cells were submitted to 80% formamide, 2X SSC at 50 °C for lmin.

7) The steps 4-7 (hybridization, washing and imaging, stripping) were then repeated for sequential RNA FISH.

Imaging Setup:

Fluorescence images were acquired on an inverted wide-field spinning disk confocal microscope (Ti-E; Nikon, Melville, NY) with a 100X 1.45 N.A. oil immersion objective. The custom-build epi-illumination optics (X-Cite XLED1; Lumen Dynamics, Mississauga, Ontario, Canada) provided excitation in DAPI, FITC, Cy3, and Cy5 channels. Andor Borealis CSU-W1 spinning disk confocal was connected to the left port of the microscope body. 4 lasers (100 mW at 405, 561, and 640 nm; 150 mW at 488 nm) were equipped to generate fluorescence excitation and emission in DAPI, FITC, Cy3, and Cy5 channels. The internal dichroic in the CSU-W was used for the incoming lasers. No additional excitation filters were used for spinning disk confocal imaging. 4 lasers were used and matched with 4 emission filters (Chroma): 405 nm laser with an emission filter at 450/50m; 488 nm laser with an emission filter at 525/50m, 561nm laser with an emission filter at 600/50m; 640nm laser with an emission filter at 700/75m. All emission filters were mounted into a motorized filter- wheel (Lambda 10-B; Sutter Instrument, Novato, CA). A motorized microscope stage (Applied Scientific Instrumentation (ASI), Eugene, OR) controls the xy and z translation of the sample. The images were recorded with a sCMOS camera (Andor Zyla 4.2 sCMOS camera). Images were typically acquired at 200ms exposure time if not clearly described separately. The microscope, light source, motorized stage, motorized filter wheel, and camera were controlled through custom configuration in an open- source software Micro-Manager (https:// www.micro-manager.org).

Image Analysis:

Images were first treated with a real-space bandpass filter that suppresses pixel noise and long-wavelength image variations. The bandpassed image were then applied with an algorithm to find local maxima to pixel level accuracy which provides a rough guess of the centers of fluorescent spots. To distinguish the fluorescent spots from background signal arising from the non-specific binding probes, the intensity threshold was required to be set up in this step where only spots with intensity four times of the mean intensity of an area without fluorescent spots were picked up. The picked local maxima were then further fitted by a

Gaussian-fitting algorithm to achieve intensity, spot number and sub-pixel resolution of locations of fluorescent spots, assuming that all reasonable fluorescent spots were confined within 10 pixels by 10 pixels region.

Probes:

Single molecule oligo pools for human RNA targets were ordered from LGC Biosearch Technologies:

Oligo pool for GAPDH: 30 oligos per probe set, 20 nt each oligo, one Quasar-670 dye per oligo.

Oligo pool for TFRC: 32 oligos per probe set, 20 nt each oligo, one Quasar-570 dye per oligo.

Oligo pool for MALAT1 : 48 oligos per probe set, 20 nt each oligo, one Quasar-670 dye per oligo.

Oligo pool for POLR2A: 32 oligos per probe set, 20 nt each oligo, one Quasar-570 dye per oligo.

Oligo pool for NEAT1 5’ Segment: 48 oligos per probe set, 20 nt each oligo, one Quasar-570 dye per oligo.

Results and Discussion:

The effect of different temperature of heat denaturation: Rapid RNA FISH with cultured Hela cells at different heat denaturation temperatures were performed. As shown in Figure 4, it is found that at 90 °C, the sample is over denatured and the RNA probes are nonspecifically binding to the nucleus of Hela cells. While at 80 °C-85 °C, the sample is partially over denatured where some specific binding could be observed together with bright regions within the nucleus. The optimal temperature range to perform heat denaturation in a formamide buffer was found to be 70 °C-75 °C with specific binding signals. 5 rounds of rapid sequential RNA FISH with cultured HeLa cells is shown in Figure 5. It is found that an additional washing step with a heated buffer (37-50 °C here) after probe hybridization during sequential rounds of RNA FISH were important to lower down the non specific binding of oligo probes in HeLa cells. The temperature of the heated washing buffer could be adjusted depending on the sample, the RNA species and the locus of the RNA species.

Example 2 Rapid and Sequential RNA and DNA FISH in cultured cell lines without tissue clearing

A rapid RNA and DNA FISH with cultured cells was performed as follows:

1) Hela cells were seeded into #1.5 glass bottom 8-well chambers (Ibidi) for heat denaturation and hybridization.

2) Cells were fixed in 4% PFA for 12mins and then in 70% ethanol at 4C for 1 h.

3) The cells were submitted to 75 °C denaturation in 80% formamide, 2X SSC for 10 min.

4) After denaturation, the cells were hybridized with 200 uL probe buffer containing 10% formamide, 2X SSC, 20% dextran sulfate and 100 nM RNA probes for lOmin at 37 °C.

5) After washing in 20% formamide, 2X SSC at 37 °C, the sample was proceeded to imaging.

6) To strip off the RNA probes, the cells were submitted to 80% formamide, 2X SSC at 37 °C for lmin.

7) The steps 4-6 (hybridization, washing and imaging, stripping) were then repeated for sequential RNA FISH.

8) To continue on DNA FISH, cells were redenatured in 80% formamide at 75 °C for 5mins. 9) After denaturation, the cells were hybridized with 200 uL probe buffer containing 10% formamide, 2X SSC, 20% dextran sulfate and 100 nM DNA probes for lOmin at 37 °C.

10) After washing in 20% formamide, 2X SSC at 37 °C, the sample was proceeded to imaging.

11) To strip off the DNA probes, the cells were submitted to 80% formamide, 2X SSC at 50 °C for lmin.

12) The steps 9-11 were then repeated for sequential DNA FISH.

Probes:

Single molecule oligo pools for RNA targets were ordered from LGC Biosearch Technologies:

Oligo pool for GAPDH: 30 oligos per probe set, 20 nt each oligo, one Quasar-670 dye per oligo.

Oligo pool for TFRC: 32 oligos per probe set, 20 nt each oligo, one Quasar-570 dye per oligo.

Telomere probes: CCCTAACCCTAACCCTAA (SEQ ID NO:l), 5’ labeled with Cy3 Centromere probes: ATTCGTTGGAAACGGGA (SEQ ID NO:2), 5’ labeled with Cy5

The imaging setup and image analysis process were the same as that used in Example- 1.

Results and Discussion:

See Figure 6 for representative results. For sequential RNA and DNA FISH, usually RNA FISH is done ahead of DNA FISH to minimize RNA degradation and sample over- denaturation, which are two big problems when using the reverse order. If sample is over heated, the probes for RNA targets are often mis-localized and form clusters in cells especially in the cell nucleus. For sequential RNA and DNA FISH, usually another round of heating is required in- between RNA FISH and DNA FISH. This can be achieved by heat renaturation above 60 °C or higher. Typically, 70-85 °C heat redenaturation for 5-10 min is used in-between RNA FISH and DNA FISH. Depending on the DNA targets, it is found that brief heat stripping the probes for the last RNA targets detected also works for some DNA targets such as telomere and centromere tested here when doing subsequent DNA FISH.

Example 3 Rapid RNA FISH in Cultured Cell Lines with Tissue Clearing and Direct Labeling

A rapid RNA FISH in MEF cell line with Tissue Clearing was performed as follows:

1) MEF cells were seeded on poly-lysine (PLL) coated #1.5 coverslips and cultured in DMEM supplemented with 10% FBS and 1% penicillin for at least 12 hrs.

2) After DMEM solution was removed, MEF cells were immediately fixed with 4% paraformaldehyde (PFA) for lOmins.

3) The fixed cells were washed with 2X SSC for 3 times and permeabilized with 70% Ethanol at 4°C for at least lhr.

4) The ethanol solution was then removed and the MEF cells were washed with 2X SSC for 3 times before submitting to heat denaturation at 75°C in 70% formamide (Sigma), 2X SSC for 10 min.

5) After denaturation, cells were washed twice with 2X SSC, once with 20 mM MOPS pH7.7, and incubated with LabelX diluted to a Label-IT Amine concentration of 0.02 mg/mL in 20mM MOPS pH 7.7 for 4 hr at 37 °C.

6) For gelling, a 50 uL of the gel solution consisting of 4% (vol/vol) of 19:1

acrylamide/bis-acrylamide, 60 mM Tris-HCl pH 8, 0.3 M NaCl, 0.2% (wt/vol) ammonium persulfate and 0.2% (vol/vol) TEMED was added to the surface of cells. A 18mm round coverslip was used to cover the gel droplet and form a thin layer. Then the gel was allowed to cast in a humidified chamber for 2 hr at 37 °C.

7) The sample was digested in a digestion buffer consisting of 0.8 M guanidine-HCl, 50 mM Tris-HCl pH 8, 1 mM EDTA and 0.5% (vol/vol) Triton X-100 supplemented with 1% (vol/vol) proteinase K overnight at 37 °C. 8) After removing the digestion buffer and washing in 2X SSC twice, the sample was incubated with 200 uL probe buffer containing 10% formamide, 2X saline-sodium citrate, 10% dextran sulfate and 100 nM RNA probes for 2 h at 37 °C.

9) After washing in 2X SSC buffer 2-3X, the sample was proceeded to imaging. A normal RNA FISH with tissue clearing was performed in the same steps above but with heat denaturation skipped and probe labeling time extended to 20h at 37 °C.

A normal RNA FISH without tissue clearing was performed below: Cells were attached, fixed and permeabilized in the same way as rapid RNA FISH with tissue clearing. After that, MEF cells were washed with 2X SSC for 3 times and then incubated with 200 ul probe buffer containing 10% formamide, 2X saline-sodium citrate, 10% dextran sulfate and 100 nM RNA probes for 20 h at 37 °C.

The imaging setup and image analysis process were the same as that used in Example- 1.

Results:

1. We compared the result obtained with conditions of denaturation at 75°C, tissue clearing and labeling for 2 h, specified in the above protocol, with two other conditions: 1) no denaturation, tissue clearing and labeling for 20 h, and 2) no denaturation, no tissue clearing but with probe labeling for 20 h. The confocal images are assembled in Figure 7. The intensities of spots from Figure 7(a)-(b) are 87.0+27.8 (A.U.), the average intensities of FISH spots from Figure 7(c)-(d) are 107.3+38.7 (A.U.) and the average intensities of FISH spots from Figure 7(e)-(f) are 115.7+41.8 (A.U.).

2. From Figure 7(c)-(f), we can tell that tissue clearing is necessary to remove the

nonspecific binding signals in the nucleus as shown in Figure 7(e)-(f). However, it takes around 20 h to achieve strong signal. Our technique of combining tissue clearing with heat denaturation reduces the labeling time greatly by 10X (from 20h to 2h) to reach about 80% of level of signals of 20h labeling without heat denaturation.

3. We used #1.5 coverslips (~0.17mm thick) to attach cultured cells and run heat

denaturation so that much faster heat transfer rate can be achieved than microscope slides of 1mm thick which will decrease non-specific disruption of cellular structures and reduce non-specific oligo binding.

4. Fower heat denaturation temperature can be used. Extended heating time may be required when heating at lower temperature. For example, heating at 65 °C for 15-20min, 45-55 °C for -30 min. Some cell lines may be very sensitive to heat, lower heating temperature and/or shorter heating time may give better performance.

5. We expect that after primary heat denaturation before tissue clearing, adding a secondary heating and also only a brief mild heating step after tissue clearing but before probe hybridization would further improve the hybridization speed and meanwhile lower down the non-specific binding rate.

Example 4 Rapid RNA FISH in Primary Tissue with Tissue Clearing and Direct Labeling Experiment:

A rapid RNA FISH on samples with Tissue Clearing was performed as follows:

1) Mouse Tissue Collection: A block of CD1 mouse liver tissue (ordered from Cell Biologies) was embedded with OCT and cryo-cut into 8 um sections. Sections were collected on poly-lysine (PLL, CultreX) coated grid coverslips (Gridded glass coverslips Grid-500, #1.5H (-170 um) D 263 Schott glass, ibidi).

2) Fixation and Permeabilization: The tissue section was immediately fixed with 4% paraformaldehyde (PFA) (Electron Microscopy Sciences) for lOmins and then treated with 70% Ethanol (Fisher BioReagents) for 1 h at RT.

3) Heat Denaturation: The tissue section was submitted to 70-75 °C denaturation in 80% formamide (Sigma) for 10 min in a dry bath (D1302, Fabnet Digital Dry Bath, dual block).

4) Tissue Clearing- 1: After denaturation, the tissue section was incubated with FabelX (synthesized from Acryloyl-X, SE; Thermo Fisher, and Fabel-IT Amine; Mirus Bio) that resulted in a Fabel-IT Amine concentration of 0.1 mg/mF for 1 h at 37 °C. 5) Tissue Clearing -2: For gelling LabelX treated tissue section, a 50 uL of the gel solution consisting of 4% (vol/vol) of 19:1 acrylamide/bis-acrylamide (Sigma), 60 mM Tris-HCl pH 8 (Invitrogen), 0.3 M NaCl (Alfa Aesar), 0.2% (wt/vol) ammonium persulfate (Sigma) and 0.2% (vol/vol) TEMED (Sigma) was added to the surface of the tissue section. A 18mm round coverslip was used to cover the gel droplet and form a thin layer. Then the gel was allowed to cast in a humidified chamber for 1.5 h at 37 °C.

6) Tissue Clearing-3: The sample was digested in a digestion buffer consisting of 1% SDS (Fisher BioReagents), 50 mM Tris-HCl pH 8, 1 mM EDTA (Invitrogen) and 0.5% (vol/vol) Triton X-100 (Fisher BioReagents) supplemented with 1% (vol/vol) proteinase K (PK; New England Biolabs) overnight at 37 °C.

7) Heat Hybridization: After washing in 2X saline-sodium citrate (SSC; Bio Basic Inc), the sample was incubated with 20 uL probe buffer containing 10% formamide, 2X saline-sodium citrate, 10% dextran sulfate (Sigma, >500,000) and 500 nM RNA probes for 0.5 h at 37 °C. Alternatively, higher concentration of dextran sulfate such as 20% can be used.

8) Imaging: After washing in 2X SSC buffer for 1 min at RT, the sample was

proceeded to imaging on a spinning disk confocal microscope with laser excitation (Nikon Ti microscope, Yokogawa CSU-W1 confocal scanner, a sCMOS camera Andor Zyla 4.2, lOOx oil objective). Each fluorescence image was acquired with 200ms exposure time.

The imaging setup and image analysis process were the same as that used in Example- 1.

Probes:

Single molecule oligo pools for RNA targets were ordered from LGC Biosearch Technologies:

Oligo pool for POLR2A: 32 oligos per probe set, 20 nt each oligo, one Quasar-570 dye per oligo; Oligo pool for TFRC: 47 oligos per probe set, 20 nt each oligo, one Quasar-570 dye per oligo.

Results and Discussion:

1. One notable strategy that was developed here was to bring the heat denaturation step (70- 75 °C and lOmin for the sample used here) before the tissue clearing steps 4-6 including LabelX incubation (step 4), gel synthesis (step 5) and proteinase K digestion (step 6), which substantially reduced the labeling time required (at least overnight) previously for efficient labeling in the gel.

2. The effect of tissue clearing: Comparing with the results of rapid RNA FISH without tissue clearing, it is found that tissue clearing could significantly decrease background and non-specific binding and improve the signal to background ratio of FISH images. As shown in Figures 8g-8h and 9g-9h, without tissue clearing, TFRC and POLR2A RNAs couldn’t be distinguished due to strong background and non-specific binding.

3. Efficient TFRC probe labeling by heat denaturation: the labeling efficiency of lh

hybridization with heat denaturation is much higher than the one without heat denaturation (negative control). In Figure 8a-8b, when TFRC probes are hybridized at 37 °C for 1 hour, the averaged intensity of TFRC is 30.8+9.6 (A.U.) with 627 dots detected. Figure 8c-8d show that the detection efficiency in undenatured samples is much lower than that in Figure 8a-8b, where only a few spots are detected. The labeling efficiency measured by the average intensity of all individual FISH spots in Figure 8a-8b is even a bit higher than that in overnight hybridization in a region of a similar size (16h, Figure 8e-8f), where the averaged intensity is 28.7+18.3 (A.U.) and 504 dots detected, which means lh hybridization with heat denaturation is sufficient for TFRC probe binding. The lower detection efficiency is possibly due to the RNA degradation after overnight hybridization.

4. Efficient POLR2A probe labeling by heat denaturation: the labeling efficiency of 0.5h hybridization with heat denaturation is much higher than the one without heat denaturation (negative control). In Figure 9a- 9b, the averaged intensity of POLR2A spots is 96.0+45.3(A.U.) with 1369 dots detected. While in Figure 9c-9d, where the same experimental conditions as in Figure 9a-9b but without heat denaturation were performed, only 13 dots are detected with averaged intensity 41.7±30.6(A.U.) in a region of similar size as in Figure 9a-9b. The average intensity of POLR2A in Figure 9c-9d is just 43% of that in Figure 9a-9b, while the number of FISH spots is around 1% of that in Figure 9a- 9b, indicating the labeling efficiency without heat denaturation is much lower than that with heat denaturation. Consistent with the observations of TFRC, the labeling efficiency of POR2A for 0.5h is even higher than that of overnight staining, in which the average intensity is 44.4+31.8(A.U.) and 773 dots are detected (Figure 9e-9f). Too long staining time just resulted in higher non-specific binding and more aggregates. The lower detection efficiency is partly induced by RNA degradation for longer hybridization and partly caused by higher background. This comparison together with that of TFRC in Figure 8 indicate the necessity and significance of heat denaturation in pursuit of fast labeling while still preserving a high labeling efficiency. The effect of different temperature of heat denaturation: Though heat denaturation is a way to efficiently label RNA probes, the optimal denaturation temperature range is quite narrow. As shown in Figure 10, the denaturation temperature at 50°C and 60°C is too low to effectively denature RNA. While the denaturation temperature at 80°C or higher is so high that RNA probes bind non- specific ally in the nucleus. When the denaturation temperature is 70°C, RNA transcripts are localized well (Figure lOe-lOf), which shows a similar pattern as in Figure 9a-9b with denaturation temperature at 75°C. Thus, the temperature range 70°C-75°C is an optimal range for the tissue sections and coverslips used here, which is consistent with the conclusion drawn from cultured cell lines as shown in Example 1. However, as heat denaturation performance is also closely related with the heat transfer rate of the scaffold used, for other types of glasses with different thickness and materials, the optimal denaturation temperature range may be different and need to be determined individually. The effect of duration and order of heat denaturation: Figure 1 la- 1 If show the heat denaturation before the tissue clearing for 5mins (Figure 11a- l ib), lOmins (Figure 11c- 1 Id), with the exactly same experimental conditions as in Figure 9a-9b) and 20mins (Figure 1 le- 1 If). It is found that short denaturation time for 5mins is not sufficient to denature all RNAs completely in the tissue section, which results in strong nonspecific binding. While samples with denaturation for 20mins is over denatured, left with only the nonspecific binding within the nuclei. Thus, denaturation at 75°C for lOmins is optimal for our experimental conditions. On the other hand, when heat denaturation after the tissue clearing steps (Figure 1 lg-1 lh) is performed, it is observed that heat denaturation before tissue clearing is crucial to preserve the RNA pattern as the heat denaturation after tissue clearing disrupts the RNA morphology and probes is localized in cell nucleus instead of binding to the right RNA targets. The possible reason could be the linker linking RNA to the hydrogel is destroyed during the heat denaturation after tissue clearing. Thus, heat denaturation before tissue clearing is the suitable order. The effect of the temperature of heat hybridization: As shown in Figure 12, heat hybridization was performed at RT (Figure 12a- 12b), 37°C (Figure 12c- 12d) and 47°C (Figure 12e-12f), respectively. It is found that at RT, most RNA probes bind

nonspecifically to the nucleus and very low specific signals are detected. In Figure 12c- 12d with the same experimental conditions as in Figure 9a- 9b and Figure 1 lc-1 Id, the average intensity of POLR2A spots is 87.2+34.0 with 1419 dots detected, similar to the statistics results in Figure 9a-9b. While in Figure 12e-12f, the average intensity of POLR2A spots is 78.3+31.8 with 1134 dots detected in a similar region as in Figure 12c- 12d. Thus, the intensity of POLR2A labeled at 47°C is around 10% lower than that labeled at 37°C and the number of molecules detected is also smaller at 47°C than that at 37°C. This comparison suggests that heat hybridization above RT is necessary, which is optimal at 37°C for our current experimental conditions. The performance of scaffold (or substrate) used for heating: It is found that the performance of heat denaturation is closely related with the materials and thickness of the scaffold/substrate used to attach the biological sample. Similar to rapid RNA FISH with cultured cells, rapid RNA FISH with gel-based tissue clearing here performs well with thin coverslips (#1.5) but not thick microscope glass slides (~lmm thick), the most widely used glass substrate for clinical pathology. When using glass slides of 1mm thick, the nucleic acid probes for RNA targets easily and even mostly aggregate in the cell nucleus. Example 5 Rapid RNA FISH in Primary Tissues without Tissue Clearing:

Experiment:

A rapid RNA FISH on samples with Tissue Clearing was performed as follows:

1) A block of CD1 mouse liver tissue (ordered from Cell Biologies) was embedded with OCT and cryo-cut into 8 um sections. Sections were collected on poly-lysine (PLL, CultreX) coated coverslips.

2) The tissue section was immediately fixed with 4% paraformaldehyde (PFA)

(Electron Microscopy Sciences) for lOmins and then treated with 70% Ethanol (Fisher BioReagents) for 1 h at RT.

3) The tissue section was submitted to 75 °C denaturation in 80% formamide (Sigma) for 10 min.

4) The sample was then hybridized with 20 uL probe buffer containing 10%

formamide, 2X SSC, 10% dextran sulfate (Sigma) and 500 nM RNA probes for 0.5 h at 30 °C.

5) After washing in pre-warmed 5% formamide, 2X SSC at 50 °C for 1 min, the sample was proceeded to imaging on a spinning disk confocal microscope with laser excitation (Nikon Ti microscope, Yokogawa CSU-W1 confocal scanner, a sCMOS camera Andor Zyla 4.2, lOOx oil objective). Each fluorescence image was acquired with 200ms exposure time.

The imaging setup and image analysis process were the same as that used in Example- 1. Results and Discussion: see Figure 8g, 8h, 9g, 9h. Discussion see Example 4.

Example 6. Rapid and Sequential RNA FISH with Tissue Clearing and Direct Labeling Experiment:

A rapid RNA FISH on samples with Tissue Clearing was performed as follows: 1) A block of CD1 mouse liver tissue (ordered from Cell Biologies) was embedded with OCT and cryo-cut into 8 um sections. Sections were collected on poly-lysine (PLL) coated grid coverslips (Gridded glass coverslips Grid-500, #1.5H (-170 um) D 263 Schott glass, ibidi).

2) The tissue section was immediately fixed with 4% paraformaldehyde (PFA) for lOmins and then treated with 70% Ethanol for 1 h at RT.

3) The tissue section was submitted to 70-75 °C denaturation in 80% formamide for 10 min.

4) After denaturation, the tissue section was incubated with LabelX that resulted in a Label-IT Amine concentration of 0.1 mg/mL for 1 h at 37 °C.

5) For gelling LabelX treated tissue section, a 50 uL of the gel solution consisting of 4% (vol/vol) of 19:1 acrylamide/bis-acrylamide, 60 mM Tris-HCl pH 8, 0.3 M NaCl, 0.2% (wt/vol) ammonium persulfate and 0.2% (vol/vol) TEMED was added to the surface of the tissue section. A 18mm round coverslip was used to cover the gel droplet and form a thin layer. Then the gel was allowed to cast in a humidified chamber for 1.5 h at 37 °C.

6) The sample was digested in a digestion buffer consisting of 1% SDS, 50 mM

Tris-HCl pH 8, 1 mM EDTA and 0.5% (vol/vol) Triton X-100 supplemented with 1% (vol/vol) proteinase K (PK) overnight at 37 °C.

7) After washing in 2X saline-sodium citrate (SSC), the sample was incubated with 20 uL probe buffer containing 10% formamide, 2X saline-sodium citrate, 20% dextran sulfate (Sigma, >500,000) and 500 nM RNA probes for 0.5 h at 37 °C.

8) After washing, the sample was proceeded to imaging on a spinning disk confocal microscope with laser excitation (Nikon Ti microscope, Yokogawa CSU-W 1 confocal scanner, a sCMOS camera Andor Zyla 4.2, lOOx oil objective). Each fluorescence image was acquired with 200ms exposure time.

9) The imaged sample was submitted to 37 °C stripping in 80% formamide for lmin and 2-3 times to strip off the RNA probes. 10) After washing in 2X saline-sodium citrate, the sample was incubated with 20 uL probe buffer containing 10% formamide, 2X saline-sodium citrate, 20% dextran sulfate (Sigma, >500,000) and 500 nM RNA probes for 0.5 h at 37 °C

11) After washing, the sample was proceeded to imaging on a spinning disk confocal microscope.

12) Subsequent sequential rounds of RNA FISH could be performed by repeating steps

(9)-(l l).

Probes:

Single molecule oligo pool probes for various mouse RNA targets (POLR2A, Neatl, Ctnnbl, Notchl, TBP) were ordered from LGC Biosearch Technologies. 20-48 oligo probes for each RNA species were designed and synthesized. Each oligo was labeled by one Quasar-670 dye.

The imaging setup and image analysis process were the same as that used in Example- 1.

Results and Discussion:

Representative results are shown in Figure 13. Significance of heat stripping: stripping at 37 °C with high concentration formamide (50-90%) is suitable for sequential RNA FISH in gels. Alternatively, higher temperature and lower concentration formamide could be used. However, when the stripping temperature was higher than 60 °C, oligo probes tended to bind to the nucleus and form puncta in gels. Therefore, a stripping buffer with moderate temperature (room temperature to 60 °C, typically at 37 °C) should be used.

Example 7. Rapid DNA FISH with Tissue Clearing:

Experiment:

A rapid DNA FISH on samples with Tissue Clearing was performed as follows:

1) A block of CD1 mouse liver tissue was embedded with OCT and cryo-cut into 8 um sections. Sections were collected on poly-lysine (PEE) coated coverslips (#1.5). 2) The tissue section was immediately fixed with 4% paraformaldehyde (PFA) for lOmins and then treated with 70% Ethanol for 1 h at RT.

3) The tissue section was then incubated with LabelX that resulted in a Label-IT

Amine concentration of 0.1 mg/mL for 1 h at 37 °C.

4) For gelling LabelX treated tissue section, a 50 uL of the gel solution consisting of 4% (vol/vol) of 19:1 acrylamide/bis-acrylamide, 60 mM Tris-HCl pH 8, 0.3 M NaCl, 0.2% (wt/vol) ammonium persulfate and 0.2% (vol/vol) TEMED was added to the surface of the tissue section. A 18mm round coverslip was used to cover the gel droplet and form a thin layer. Then the gel was allowed to cast in a humidified chamber for 1.5 h at 37 °C.

5) The sample was digested in a digestion buffer consisting of 1% SDS, 50 mM

Tris-HCl pH 8, 1 mM EDTA and 0.5% (vol/vol) Triton X-100 supplemented with 1% (vol/vol) proteinase K overnight at 37 °C.

6) The tissue embedded gel was submitted to 75 °C denaturation in 80% formamide for 10 min.

7) After washing in 2X SSC, the sample was incubated with 20 uL probe buffer

containing 10% formamide, 2X saline- sodium citrate, 10% dextran sulfate and 100 nM DNA probes for lOmins at 37 °C.

8) After washing, the sample was proceeded to imaging with the same setup in

Example 1.

Probes:

Probe sequences for the repetitive regions of telomere and centromere in mouse genomes were used here:

Telomere probes: CCCTAACCCTAACCCTAA (SEQ ID NO:l), 5’ labeled with Cy3 Centromere probes: ATTCGTTGGAAACGGGA (SEQ ID NO:2), 5’ labeled with Cy5 The imaging setup and image analysis process were the same as that used in Example- 1.

Results and Discussion: Representative FISH images are shown in Figure 14. One notable difference between the rapid DNA FISH with tissue clearing and rapid RNA FISH with tissue clearing is that the heat denaturation step (step (6)) is after the gel synthesis and protein digestion steps but right before the DNA probe hybridization step in DNA FISH to avoid chromosome DNA reannealing if doing heat denaturation before gel synthesis. 10 min hybridization is sufficient for complete hybridization. The hybridization efficiency is as high as that of overnight hybridization.

Example 8. Rapid RNA and DNA FISH Sequentially with Tissue Clearing

RNA and DNA FISH can be done sequentially in the following way when tissue clearing is done with a biospecimen: (1) heat denaturation; (2) hydrogel embedding; (3) RNA FISH; (4) heat stripping of probes or heat redenaturation; (5) DNA FISH.

A rapid RNA and DNA sequential FISH on samples with tissue clearing was performed as follows:

1) A block of mouse liver tissue was embedded with OCT and cryo-cut into 8 um

sections. Sections were collected on PLL coated coverslips (Gridded glass coverslips Grid-500, #1.5H (-170 um) D 263 Schott glass, ibidi).

2) The tissue section was immediately fixed with 4% PFA for lOmins and then treated with 70% Ethanol for 1 h at RT.

3) The tissue section was submitted to 75 °C denaturation in 80% formamide for 10 min in a dry bath.

4) After denaturation, the tissue section was incubated with LabelX that resulted in a Label-IT Amine concentration of 0.1 mg/mL for 1 h at 37 °C.

5) For gelling LabelX treated tissue section, a 50 uL of the gel solution consisting of 4% (vol/vol) of 19:1 acrylamide/bis-acrylamide, 60 mM Tris-HCl pH 8, 0.3 M NaCl, 0.2% (wt/vol) ammonium persulfate and 0.2% (vol/vol) TEMED was added to the surface of the tissue section. A 18mm round coverslip was used to cover the gel droplet and form a thin layer. Then the gel was allowed to cast in a humidified chamber for 1.5 h at 37 °C. 6) The sample was digested in a digestion buffer consisting of 1% SDS, 50 mM Tris-HCl pH 8, 1 mM EDTA and 0.5% (vol/vol) Triton X-100 supplemented with 1% (vol/vol) proteinase K overnight at 37 °C.

7) After washing in 2X saline-sodium citrate, the sample was incubated with 20 uL probe buffer containing 10% formamide, 2X saline-sodium citrate, 10% dextran sulfate and 500 nM RNA probes for 1 h at 37 °C.

8) After washing, the sample was proceeded to imaging.

9) The imaged sample was submitted to 50 °C stripping in 80% formamide for lmin to strip off the RNA probes. This step was critical as it not only removed probes but also further denatured chromatin DNA for the following DNA FISH. Alternatively, a heat redenaturation (75 °C, 5-10 min) step after probe stripping could be added to increase the accessibility of chromatin DNA for DNA FISH next step.

10) After washing in 2X SSC, the sample was incubated with 20 uF probe buffer

containing 10% formamide, 2X saline- sodium citrate, 10% dextran sulfate and 100 nM DNA probes for lOmins at 37 °C.

11) After washing, the sample was proceeded to imaging.

RNA FISH Probes:

Single molecule oligo pools for Neatl and Ctnnbl RNA targets were ordered from

FGC Biosearch Technologies. 20-48 oligo probes were designed for each RNA species.

DNA FISH Probes:

Probe sequences for the repetitive regions of telomere and centromere in mouse genomes were used here:

Telomere probes: CCCTAACCCTAACCCTAA (SEQ ID NO:l), 5’ labeled with Cy3

Centromere probes: ATTCGTTGGAAACGGGA (SEQ ID NO:2), 5’ labeled with Cy5

The imaging setup and image analysis process were the same as that used in Example- 1.

Results and Discussion: Representative results are shown in Figure 15. Because rapid RNA FISH is only suitable with heat denaturation before tissue clearing while heat redenaturation or heated probe stripping after tissue clearing is a necessary step to ensure the accessibility of DNA during DNA FISH. The order of performing RNA FISH first then DNA FISH is the straightforward and best way to detect RNA and DNA targets in the same tissue to minimize RNA degradation and probe non-specific binding.

Example 9: Rapid and Sequential DNA FISH with Tissue Clearing

Rapid and sequential DNA FISH can be done with hydrogel based tissue clearing. A rapid and sequential DNA FISH experiment with tissue clearing was performed as follows:

1) A block of mouse liver tissue was embedded with OCT and cryo-cut into 8 um sections. Sections were collected on PLL coated coverslips.

2) The tissue section was immediately fixed with 4% PFA for lOmins and then treated with 70% Ethanol for 1 h at RT.

3) The tissue section was then incubated with LabelX that resulted in a Label-IT

Amine concentration of 0.1 mg/mL for 1 h at 37 °C.

4) For gelling LabelX treated tissue section, a 50 uL of the gel solution consisting of 4% (vol/vol) of 19:1 acrylamide/bis-acrylamide, 60 mM Tris-HCl pH 8, 0.3 M NaCl, 0.2% (wt/vol) ammonium persulfate and 0.2% (vol/vol) TEMED was added to the surface of the tissue section. A 18mm round coverslip was used to cover the gel droplet and form a thin layer. Then the gel was allowed to cast in a humidified chamber for 1.5 h at 37 °C.

5) The sample was digested in a digestion buffer consisting of 1% SDS, 50 mM

Tris-HCl pH 8, 1 mM EDTA and 0.5% (vol/vol) Triton X-100 supplemented with 1% (vol/vol) proteinase K overnight at 37 °C. 6) The tissue embedded gel was submitted to 75 °C denaturation in 80% formamide for 10 min.

7) After washing in 2X SSC, the sample was incubated with 20 uL probe buffer

containing 10% formamide, 2X saline- sodium citrate, 10% dextran sulfate and 100 nM DNA probes for lOmins at 37 °C.

8) After washing in the hybridization buffer without probes 1-2 times, the sample was proceeded to imaging.

9) The imaged sample was submitted to 50 °C stripping in 80% formamide for lmin to strip off the DNA probes. Alternatively, a redenaturation step could be added after probe stripping here and before next round probe hybridization. The redenaturation temperature could be 50 °C - 100 °C with a duration of 1-10 min, depending on the DNA loci.

10) After washing in 2X SSC, the sample was incubated with 20 uL probe buffer

containing 10% formamide, 2X saline- sodium citrate, 10% dextran sulfate and 100 nM DNA probes for lOmins at 37 °C.

11) After washing to remove excessive probes, the sample was proceeded to imaging.

12) Subsequent sequential rounds of DNA FISH could be performed by repeating steps

(9)-(l l).

DNA FISH Probes:

Probe sequences for the repetitive regions of telomere and centromere in mouse genomes were used here:

Telomere probes: CCCTAACCCTAACCCTAA (SEQ ID NO:l), 5’ labeled with Cy3 CCCTAACCCTAACCCTAA (SEQ ID NO:l), 5’ labeled with Cy5

Centromere probes: ATTCGTTGGAAACGGGA (SEQ ID NO:2), 5’ labeled with Cy5 ATTCGTTGGAAACGGGA (SEQ ID NO:2), 5’ labeled with Cy3 The imaging setup and image analysis process were the same as that used in Example- 1. Results and Discussion:

Representative FISH images are shown in Figure 16. In Figure 16, images of different rounds of DNA FISH were taken in the same field of view but at different z position. Thus, the telomere signals in Figure 16b and Figure 16h look different and also centromere signals in Figure 16d and Figure 16f look different.

Example 10: Rapid and Sequential RNA FISH with Tissue Clearing and Primary + Secondary Probes

A rapid RNA FISH on samples with Tissue Clearing can be performed as follows:

1) A block of CD1 mouse liver tissue (ordered from Cell Biologies) is embedded with OCT and cryo-cut into 8 um sections. Sections are collected on poly-lysine (PLL, CultreX) coated coverslips (#1.5).

2) The tissue section is immediately fixed with 4% paraformaldehyde (PFA) (Electron Microscopy Sciences) for lOmins and then treated with 70% Ethanol (Fisher BioReagents) for 1 h at RT.

3) The tissue section is submitted to 50-80 °C denaturation in 80% formamide (Sigma) for 10 min.

4) After washing in 2X saline-sodium citrate (SSC; Bio Basic Inc), the sample is

incubated with 20 uL probe buffer containing 10% formamide, 2X saline-sodium citrate, 10% dextran sulfate (Sigma, >500,000) and 500 nM primary probes for 0.5- lh at 37 °C. Then the sample is washed in heated (40-50 °C ) hybridization buffer only 2-3X or in a high concentration formamide buffer (30-80%) at room temperature to remove non-specific bound probes and unbound probes.

5) After denaturation, the tissue section is incubated with LabelX (synthesized from Acryloyl-X, SE; Thermo Fisher, and Label-IT Amine; Mirus Bio) that resulted in a Label-IT Amine concentration of 0.1 mg/mL for 1 h at 37 °C. 6) For gelling LabelX treated tissue section, a 50 uL of the gel solution consisting of 4% (vol/vol) of 19:1 acrylamide/bis-acrylamide (Sigma), 60 mM Tris-HCl pH 8 (Invitrogen), 0.3 M NaCl (Alfa Aesar), 0.2% (wt/vol) ammonium persulfate (Sigma) and 0.2% (vol/vol) TEMED (Sigma) is added to the surface of the tissue section. A 18mm round coverslip is used to cover the gel droplet and form a thin layer. Then the gel is allowed to cast in a humidified chamber for 1.5 h at 37 °C.

7) The sample is digested in a digestion buffer consisting of 1% SDS (Fisher

BioReagents), 50 mM Tris-HCl pH 8, 1 mM EDTA (Invitrogen) and 0.5% (vol/vol) Triton X-100 (Fisher BioReagents) supplemented with 1% (vol/vol) proteinase K (PK; New England Biolabs) overnight at 37 °C.

8) After washing in 2X saline-sodium citrate (SSC; Bio Basic Inc), the sample is

incubated with 20 uL probe buffer containing 10% formamide, 2X saline-sodium citrate, 10% dextran sulfate (Sigma, >500,000) and 500 nM secondary probes for 0.5-1 h at 37 °C.

10) After washing, the sample is proceeded to imaging on a spinning disk confocal microscope with laser excitation (Nikon Ti microscope, Yokogawa CSU-W 1 confocal scanner, a sCMOS camera Andor Zyla 4.2, lOOx oil objective). Each fluorescence image can be acquired with 200ms exposure time.

Probe design and synthesis:

Primary-probe design. To obtain the hybridization sequences of each primary probe, 20- nucleotide (nt) sequences of each gene are extracted, first using the coding sequence region only. For genes that do not yield enough target sequences from the coding region, sequences from the untranslated regions are used. The masked genome and annotation from the University of California Santa Cruz (UCSC) can be used to look up the gene sequences. Probe sequences are required to have GC content within the range 45-65%. Any probe sequences that contain five or more consecutive bases of the same kind are dropped. Any genes that do not achieve a minimum number of 20 probes are dropped. A local BLAST query is run on each probe against the mouse transcriptome to ensure specificity. BLAST hits on any sequences other than the target gene with a 15-nt match are considered off targets. ENCODE RNA-seq data across different mouse samples are used to generate an off-target copy-number table. Any probe that hit an expected total off-target copy number exceeding 10,000 FPKM is dropped to remove housekeeping genes, ribosomal genes and very highly expressed genes. To minimize cross-hybridization between probe sets, a local BEAST database can be constructed from the probe sequences, and probes with hits of 14 nt or longer are removed by dropping the matched probe from the larger probe set.

Readout-probe design: 4 Readout probes of 20 nt in length are used here. The GC content of the probes and their sequence homology to mouse genome and transcriptome are carefully evaluated so that they have minimal non-specific binding to cellular RNA and DNA sequences. Each probe is labeled with one Cy5 dye at 5’ end.

Primary-probe synthesis: For each primary probe, the hybridization sequence is added with a 20 nt readout sequence at both ends which are complementary with the readout probes designed above. The 5’ and 3’ end of each primary probe have different readout sequences. The primary probes for the same RNA have the same set of readout sequences but different RNA species use different set of readout sequences to avoid cross-hybridization of readout probes between species The primary probes for RNA targets can be ordered from LGC Biosearch Technologies.

Oligo pool for Polr2A: 60 oligos per probe set, 20 nt each oligo;

Oligo pool for TFRC: 40 oligos per probe set, 20 nt each oligo.

Readout-probe synthesis. Single dye labeled readout probes of 20 nt in length are ordered from Integrated DNA Technologies.

The imaging setup and image analysis process were the same as that used in Example- 1.

Example 11. High Resolution RNA FISH by PNA in Cultured Cell Lines

5 mRNAs (one exon from each RNA) are chosen in mouse MEF cells to demonstrate high resolution RNA FISH. The 5 RNAs are GAPDH (high expression), TFRC (moderate expression), POFR2A (low expression), neurexin (abundant splicing isoforms in neuronal cells), and protocadherin (abundant splicing isoforms in neuronal cells). A set of hybridization probes is designed to target 200-1000 nt regions of each RNA above. Here, various length (10-15 amino acids) of PNA oligos can be tested. In one design, each PNA oligo had one dye at each end so that 2 same dyes is conjugated to an oligo. In the other design, each PNA oligo is conjugated with a readout sequence through azide and alkyne click chemistry. Each readout sequence is 30- 100 nt long and can be labeled with 2-6 fluorophores by KREATECH universal linkage system. Other high affinity oligos such as LNA can also be used to replace PNA as the choice of hybridization sequences on probes.

A Rapid RNA FISH can be done in cultured MEF cells. 8-well glass bottom culture chambers (Ibidi) can be used for sample imaging. Cells are fixed in 4% paraformaldehyde (PFA; Electron Microscopy Sciences) at room temperature for 5-10 min and then permeabilized in pure ethanol (Fisher BioReagents) for 5-10 min. Then the sample is processed by the following steps: (1) thorough heat denaturation in a dry bath in 80% formamide (Sigma, 2X SSC) buffer at 70- SOD for 10 min; (2) addition of PNA oligonucleotide probes with a hybridization time of 30 min to lhour in a hybridization buffer of 10% formamide, 10% dextran sulfate (Sigma,

MW>500,000) and 2X SSC; (3) washing by 2X SSC buffer 2-3 times; (3) Imaging the sample. The probe concentration should be 100-200 nM during hybridization. Imagine setup is the same as that in Example 1.

Evaluating detection efficiency: An open-source software (ImageJ) and customized software can be used to analyze data. Individual fluorescent spots can be identified and counted by a Gaussian fitting algorithm. Using normal single molecule oligo FISH as the gold standard, RNA detection efficiency can be determined by the colocalization analysis of these two FISH assays on the same genes and tissues.

Example 12. High Resolution RNA FISH by PNA in Primary Tissues:

1) Decreasing background signal by tissue clearing: Primary tissues usually have much higher auto-fluorescence than cultured cells, resulting from extracellular matrixes, intracellular lipids, proteins and other small molecules. Moreover, neuronal cells typically have higher background than other cell types due to the intracellular accumulation of lipofuscins. In addition, cellular proteins also contribute to a lot of non specific binding of DNA probes when doing RNA FISH. Thus, tissue clearing to remove lipids, proteins and intracellular fluorophores is necessary to achieve high resolution RNA imaging. To do tissue clearing, the hydrogel-embedding approach can be used here. This method embeds tissue in polyacrylamide hydrogel, anchors RNA molecules to the gel matrix with cross-linking reagents, and then uses protease and detergents to remove non-RNA components such as proteins and lipids. Two different gels can be used for tissue clearing: acrylamide without sodium acrylate and acrylamide plus sodium acrylate. Compared to the former, the latter not only provides a gel network to support tissue, but also expands tissue homogenously so that high expression mRNAs can be visualized individually. The former gel can be used to image low expression genes such as Polr2A and the latter to image high expression genes such as GAPDH.

2) CD1 mouse liver tissue is chosen to demonstrate high resolution RNA FISH. Mouse tissue is frozen into OCT-embedded tissue blocks and cut into lOum thick tissue sections. Tissue sections are attached onto #1.5 coverslips coated with poly-lysine. Then tissues are fixed in 4% paraformaldehyde 5-10 min and permeabilized with pure ethanol at room temperature 5-15 min or 70% ethanol at room temperature for 30 min.

3) PNA Probes of various length (10-15 amino acid each probe) for TFRC and Polr2A RNA can be designed. For TFRC probes, each PNA oligo has one dye at each end so that 2 dyes are conjugated to an oligo. For Polr2A probes, each PNA oligo is conjugated with a readout sequence through azide and alkyne click chemistry. Each readout sequence is 30- 100 nt long and labeled with 3-10 fluorophores by KREATECH universal linkage system.

4) Tissue clearing, RNA FISH process, imaging and image analysis are in the way as

regular and rapid RNA FISH in Example 4.

While the invention has been disclosed with reference to specific embodiments, it is apparent that other embodiments and variations of this invention may be devised by others skilled in the art without departing from the true spirit and scope of the invention. The appended claims are intended to be construed to include all such embodiments and equivalent variations.