MicroRNA Amplification and Recognition through Locked-nucleic-acid In situ Hybridization as a Novel Detection and Quantification Method

Summary

The protocol describes a tandem immunofluorescence hybridization assay with formalin-fixed paraffin-embedded tissue to detect and quantify microRNA and multiplexed proteins simultaneously.

Abstract

When using traditional probe-based in situ hybridization (ISH) imaging detection methods, microRNAs (miRNAs) are difficult to capture due to their size, which ranges from 19 to 25 nucleotides. However, as miRNAs are key epigenetic regulators that contribute to normal physiology and numerous pathological conditions, understanding the spatiotemporal behavior of miRNAs is important to unravel their function at the cellular and subcellular levels. This creates a clear need for adequate detection and visualization techniques. Therefore, we created MicroRNA Amplification and Recognition through Locked-nucleic-acid in situ hybridizatioN (MARLIN), which was developed and optimized utilizing multiplex immunofluorescence staining, and locked nucleic acid (LNA) probes targeted to a miRNA. This method can quantify miRNA with multiplexed protein expression, which allows the quantification of intracellular and extracellular miRNA within the tumor microenvironment (TME) with high sensitivity and reproducibility. The combined use of miRNA detection and protein capture assays can also provide a comprehensive understanding of cellular functions and regulatory mechanisms within that TME.

Introduction

microRNAs
Although microRNAs (miRNA) are incredibly small, ranging 19-25 nucleotides in length, these single-stranded RNA molecules are potent regulators in gene expression, modulating messenger RNAs (mRNAs) post-transcriptionally through sequence-specific interactions¹. In essence, miRNAs act as negative regulators by binding to target mRNAs, either preventing their translation or promoting their degradation through mRNA cleavage, depending on the degree of complementarity, which ultimately leads to decreased protein output from the targeted gene. While miRNAs are basically non-coding RNA, approximately 40% of all miRNAs are still found within introns or exons of protein-coding genes². The functional purpose of miRNAs encompasses several key biological processes such as cell proliferation, angiogenesis, apoptosis, etc. Notably, the dysregulation of miRNAs has been implicated in tumorigenesis, as aberrant expression of specific miRNAs and miRNA-mediated gene regulation has been associated with oncogenic transformation in multiple cancer types³.

Since their discovery, miRNAs have been extensively researched in various diseases due to their tissue-specific expression and regulatory functions, with their dysregulation linked to many diseases, including neurodegenerative disorders, autoimmune disorders, and various cancers^4,5,6. The expression patterns of miRNAs can be correlated with cancer type, stage, and clinical variables, making miRNA profiling a valuable tool for cancer diagnosis and prognosis^4,5,6,7. The discovery of new miRNAs has also been accelerated by advancements in high-throughput sequencing technologies, enhancing the understanding of miRNA involvement in diverse biological processes⁸. The significance of miRNAs was underscored by the 2024 Nobel Prize in Medicine awarded to Victor Ambros and Gary Ruvkun for their discovery of the function of miRNA in posttranscriptional regulation⁹.

As valuable biomarkers for disease diagnosis and prognosis, the therapeutic potential of miRNAs has been explored, leading to the development of miRNA-targeting drugs. The therapeutic potential of miRNAs in cancer treatment lies in their ability to coordinately regulate modules of genes, potentially restoring normal cellular functions¹⁰. Previously, miRNA-based anticancer therapies have been developed either as monotherapy or in combination with existing targeted therapies to enhance treatment responses and increase cure rates¹¹. The dynamic spectrum of miRNAs has positioned them as essential resources for innovative cancer treatment strategies. In the context of therapeutic resistance, targeting miRNAs has shown promise in reducing recurrence and resistance to traditional cancer treatments, potentially improving patient survival by targeting cancer cells¹². By exploring the roles of miRNAs in modulating the activities of various cell types in the tumor microenvironment (TME), unique and clinically relevant cancer treatments can potentially be discovered.

Available assays to detect miRNAs
Currently, the most common method for miRNA measurement relies on extracted RNA for quantitative real-time PCR (q-RT-PCR) or Next Generation Sequencing (NGS), losing spatial context. There are companies that will provide TaqMan microRNA assays, which use target-specific stem-loop primers for cDNA synthesis, and SYBR Green dye-based chemistries with universal cDNA synthesis. These assays can offer high sensitivity and specificity and can be run on a range of q-RT-PCR instruments. Unfortunately, TaqMan assays cannot provide absolute quantification, which can lead to issues such as false positives and variability in results¹³. Also, despite their high sensitivity, the application of these assays in differing sample types can be limited by the need for extensive sample preparation and potential interference from other components in the sample.

Next-Generation Sequencing (NGS): miRNA-seq is increasingly popular for differential expression analysis across the entire miRNome of biofluids. A benefit of NGS technologies like miRNA-seq is that they allow for massive parallel analyses of the genome-wide expression of miRNAs, enabling the quantification of individual miRNAs, identification of sequence variations, and discovery of novel miRNAs¹⁴. A substantial drawback to sequencing-based methods, such as IlluminaTruSeq Small RNA Sample Preparation Kit, is the induction of quantitative biases, affecting the accuracy of differential expression analyses, due in large part to the homology and smaller read lengths of miRNAs¹⁴. Not only that, but the presence of isomiRs (variants that differ in sequence or length from the canonical miRNA sequence) and technical artifacts can complicate data interpretation, potentially leading to misdetections and confounding factors in biomarker assays¹⁴. Also, different tools for miRNA-seq data analysis, such as sRNAbench and miRDeep2, can yield varying results, affecting the consistency and reliability of novel miRNA discovery^15,16. Even something like microarray technology can be insufficient in most cases. Though it can provide a powerful, high-throughput platform for detecting miRNA or gene expression levels, making it suitable for genome-wide studies, the technology is still limited by the number of previously identified miRNAs and cannot directly verify miRNA-target interactions, which limits their utility in functional studies¹⁷.

The most important factor that can be stressed here when comparing all these available options is the lack of spatial context. They do not provide any context in relation to the tissue morphology or spatiotemporal distribution within the cells. Visualizing the spatiotemporal behavior of miRNAs is necessary to unravel their function at the cellular and subcellular levels. Presently, miRNAscope in situ Hybridization (ISH) assays are accessible, usually utilizing chromogenic detection methods. While these assays do allow for the visualization of miRNA with single-cell resolution, the results are lacking in quality, making it difficult to differentiate between the miRNA expression and the background of the sample. Generally, many of these produce low-quality assays, which are unreliable and usually require manual quantification.

Therefore, we developed MicroRNA Amplification and Recognition through Locked-nucleic-acid in situ hybridizatioN (MARLIN) using multiplex immunofluorescence staining (by Opal protein staining) and locked nucleic acid (LNA) probes targeted to a miRNA, within formalin-fixed paraffin-embedded (FFPE) tissues. We accomplished this by implementing a Tyramide Signal Amplification (TSA) system that is activated by an Anti-Digoxigenin (DIG) Horseradish Peroxidase (HRP) Conjugate antibody. The HRP acts as a catalyst to the TSA, while the Anti-DIG attaches to the LNA probe. This assay is of particular use in understanding the essential nature of miRs in post-transcriptomic modification with proteins, and potentially other RNAs simultaneously. Furthermore, due to the lack of a commercially available assay that can quantify both miRNA and protein expression, MARLIN provides an effective way to understand the in situ relationships of miRs and proteins within FFPE samples.

The initial results from the quantification of miR-181c-3p using MARLIN demonstrated the feasibility of parallel analysis of miRNAs and proteins within the native spatial context of human ovarian FFPE tissue. Beyond this example, based on the components of MARLIN, it is theoretically possible to expand to up to four proteins utilizing commercial fluorescent multiplexing fluorophores like Alexa Fluor, or other small molecule dyes¹⁸.

In addition, since this methodology employs variations on standard histology methods, it should be compatible with other mammalian FFPE tissues, given appropriate probe and antibody selection. This methodology preserves spatial architecture and allows the quantification of cellular and subcellular miRNA and protein colocalization. This approach addresses a critical gap in spatial biology by mapping miRNAs-protein co-expression patterns and spatial distribution, which can be directly contextualized within native tissue architectures. Here we present MARLIN with a detailed graphical overview in Figure 1. While the protocol presented is optimized for miR-181c-3p, other miRs are possible to study, given an appropriate probe design. Critical parameters for this method include: the assessment of tissue quality and the optimization of probe concentrations. It is recommended to assess RNA integrity of the tissue block using a DV200 metric by utilizing two 10 µm sections and extracting the RNA using a protocol that preserves all RNA fragment sizes. Samples with a DV200 score greater than 50 or better, and protocol concentrations, need to be systematically tested.

Protocol

This protocol uses formalin-fixed paraffin-embedded (FFPE) tissues of human ovarian tumor tissue sections from patients with HGSC. These samples were collected from previously untreated patients undergoing primary cytoreductive surgery for ovarian cancer. All samples and clinical data were obtained from the ovarian cancer repository of the Department of Gynecologic Oncology and Reproductive Medicine under protocols approved by the University of Texas MD Anderson's Institutional Review Board. Written informed consent from the patients was obtained by front desk personnel, and the studies were conducted in accordance with recognized ethical guidelines.

1. Tissue preparation

1. Bake all FFPE samples overnight at 60 °C and clean each piece of equipment and tools with RNAse-free water. Ensure the sections are around 5 µm thick.
   NOTE: This protocol has been optimized for ovarian tumors but can potentially be used with other tissue types after further testing.
2. Perform deparaffinization with three rounds of Xylene for 5 min each round, and 10 rounds of rehydration for 3 min each round (100% ethanol (EtOH) 2x, 95% EtOH 2x, 70% EtOH 2x, 50% EtOH 2x, 30% EtOH 2x). Wash with 1x tris-buffered saline (TBS) for 2 min, once.
   NOTE: This dewaxing step, although recommended, is optional and dependent on tissue type.
3. After deparaffinization and rehydration, apply a solution of 3% Hydrogen Peroxide (H₂O₂) for 10 min. Wash for 2 min with 1x TBS, 2x. Then, apply a solution of 3.7% Paraformaldehyde (PFA) for 10 min. Wash for 2 min with 1x TBS, 2x.
   NOTE: This step is optional and should only be considered for fatty tissues, or samples that are prone to detachment.
4. Proceed to the first round of antigen retrieval. This protocol uses the EZ-Retriever IR System. Following the system's instructions, microwave the samples while submerged in fresh EZ-AR1 buffer at 95 °C for 15 min. Leave to cool for 20 min.
5. Wash for 2 min with 1x TBS. Then for 2 min with 1x TBS-T (1x TBS with 0.1% Tween), 2x.

2. Protein application

1. Draw a barrier around each tissue section with a hydrophobic barrier pen.
2. Add 1x Antibody blocking solution (enough to cover the entire tissue area) and incubate for 30 min at room temperature in a humidity chamber (ambient and uncontrolled humidity).
3. Subsequently, dilute the first primary antibody with the same blocking solution, apply enough to cover the entire tissue area, and leave to incubate for 1 h at room temperature in a humidity chamber.
   NOTE: During the development of this protocol, Anti-CD8 was used at a dilution of 1:75.
4. Wash for 2 min in 1x TBS-T, 3x. Then, add the Anti-Ms + Rb HRP and leave the samples to incubate for 10 min at room temperature in a humidity chamber (ambient and uncontrolled humidity).
5. Wash for 2 min in 1x TBS-T, 3x. Dilute the first fluorophore with the amplification dilutant at 1:100 and incubate for 10 min at room temperature in the humidity chamber.
   NOTE: During the development of this protocol, a 690 nm emission fluorophore was used. From this point, it is recommended to use minimal lighting for the rest of the procedure after applying the first round of fluorophore.
6. Wash for 2 min in 1x TBS-T, 3x. Begin the second round of antibody staining with the second round of antigen retrieval in the microwave, by again submerging the samples in fresh EZ-AR1 buffer at 95 °C for 15 min, then cooling for 20 min.
7. Wash for 2 min with 1x TBS. Then wash for 2 min with 1x TBS-T, 2x.
8. Apply the 1x Antibody blocking solution again for 30 min in a humidity chamber at room temperature (ambient and uncontrolled humidity).
9. Dilute the second primary antibody with the same blocking solution, apply enough to cover the entire tissue area, and leave to incubate overnight at 4 °C in a humidity chamber for at least 16 h.
   NOTE: During the development of this protocol, Anti-EpCAM was used at a dilution of 1:200.
10. Begin the following day by washing for 2 min in 1x TBS-T, 3x. Then add the Anti-Ms + Rb HRP and leave the samples to incubate for 10 min at room temperature in a humidity chamber (ambient and uncontrolled humidity).
11. Wash for 2 min in 1x TBS-T, 3x. Dilute the second fluorophore with the amplification dilutant at 1:100 and incubate for 10 min at room temperature in the humidity chamber.
    NOTE: During the development of this protocol, a 620 nm emission fluorophore was used.

3. Probe application

1. Wash for 2 min in 1x TBS-T, 3x.
2. Start the third and final antigen retrieval with EZ-AR1 Elegance in the microwave at 107 °C for 15 min. Leave to cool for 20 min. Wash for 3 min in 1x TBS.
3. Apply RNAscope Protease Plus (enough to cover the entire tissue area) and keep at 40 °C for 30 min. Use an oven. Wash for 5 min in 1x TBS, 2x.
4. Perform probe preparation by denaturation at 90 °C for 4 min of the miRNA LNA probe. Then immediately dilute the probe using miRNA-ISH buffer. Vortex and spin the probe solution.
   NOTE: During the development of this protocol miR-181c-3p target probe was used at a dilution of 1:250.
5. Apply probe (enough to cover the entire tissue area) and place back in the oven for 1 h at 60 °C. Wash for 5 min in 5x Saline Sodium Citrate (SSC) buffer.
6. Perform five stringent washes in SSC prewarmed at 60 °C in the following order and dilutions for 5 min each round: one round of 5x SSC, two rounds of 1x SSC, and two rounds of 0.2x SSC.
7. Wash for 5 min in 0.2x SSC at room temperature. Incubate with blocking buffer for 30 min at room temperature in a humidified chamber. Use a combination of 1x TBS and 0.5% Blocking Reagent Powder, make sure to vortex and spin the blocking solution before application.
8. Apply an Anti-Digoxigenin (Mouse) HRP Conjugate (for these samples, a dilution of 1:500 was used) and incubate for 30 min at room temperature in a humidified chamber.
9. Wash for 5 min in 1x TBS-T, 3x Apply a recommended dilution of 1:50 of the Tyramide Signal Amplification Plus Cyanine 3 (TSA+Cy3) fluorophore for 10 min in a humidity chamber at room temperature (ambient and uncontrolled humidity). Make sure to vortex and spin the TSA+Cy3 solution before application.
10. Wash for 5 min in 1x TBS-T, 3x.

4. Nuclear staining and mounting

1. Apply 4',6-diamidino-2-phenylindole (DAPI) working solution (enough to cover the entire tissue area) for 5 min at room temperature in a humidified chamber (ambient and uncontrolled humidity).
   NOTE: For this protocol, a dilution of 1:1000 was used, with the working DAPI stock diluted in TBS.)
2. Wash for 5 min in 1x TBS-T, 3x. At completion, apply coverslips using mounting media and leave slides to dry for 10 min. Store slides at 4 °C in the dark between 24-72 h before imaging and up to three months after.

5. Imaging and analysis

1. For imaging, use a slide scanning microscope (preferred) at 40x magnification.
2. Perform subsequent quantitative analysis to enumerate puncta utilizing DAPI and protein counterstaining for cell segmentation and phenotyping.
   NOTE: During this protocol, we utilized imaging with a Phenoimager HT. Full slides scans were captured at a 40x magnification utilizing appropriate filters for the fluorophores present, and multiplexed images were then exported as OME-TIFF files. The scanned images were then uploaded into Visiopharm Image Analysis Software for quantitative analysis.

Results

When testing this protocol, it is important to have two serial sections of the same representative sample for accuracy and comparison. In this case, a High-Grade Ovarian Cancer (HGSC) sample was used to compare the results. A control slide (Figure 2A-B) that would only go through the miRNA staining and a test slide that would undergo both protein and miRNA staining (Figure 2C-J) were used.

Discussion

Though the importance of miRNAs is recognized and understood, observations of both normal development and disease states are currently hindered by the difficulty of studying miRNAs using standard methods. The scarcity of in situ methods led to the development of MARLIN (miRNA FISH assay). Given the homology and short length of miRNAs, it becomes a challenge to visualize their location and effect on post-transcriptomic regulation. This protocol development utilizes Locked Nucleic Acid Probes (LNA). LNAs are modif...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
10x TRIS-buffered Saline (TBS) pH 7.6	Thermo Fisher Scientific	J62662 K7	For intermittent washes between steps
1x Opal Antibody Diluent/Block	Akoya Biosciences	ARD1001EA	Protein blocking solution
1x Plus Automation Amplification Diluent	Akoya Biosciences	FP1609	Used to dilute Opal fluorophores
20x Concentrate Saline Sodium Citrate (SSC) Buffer	Sigma-Aldrich	S6639-1L	For miRNA washes after probe hybridization
4',6-diamidino-2-phenylindole (DAPI)	Thermo Fisher Scientific/ Invitrogen	EN62248 / 62248	For nuclear staining
ACD HybEZ II Hybridization System	ACD Biotechne	321711	Oven used to regulate temperature during miRNAscope protocol
Anti-CD8 (D8A8Y), Rabbit mAb	Cell Signaling Technologies	85336S	Example of primary antibody used to optimize this protocol
Anti-Digoxigenin (Mouse) HRP Conjugate	Revvity Health Sciences Inc	NEF832001EA	Secondary reagent used for miRNA detection
Anti-EpCAM (E6V8Y), Rabbit mAb	Cell Signaling Technologies	93790	Example of primary antibody used to optimize this protocol
Blocking Reagent Powder	Akoya Biosciences	FP1012	miRNA blocking solution
Ethyl Alcohol 100% (200 Proof)	Pharmco	111000200	For deparaffinization
EZ-AR1 Buffer	BioGeneX	HK521-XAK	Used for antigen retrieval of FFPE tissue sections at the protein step
EZ-AR1 Elegance Buffer	BioGeneX	HK546-XAK	Used for antigen retrieval of FFPE tissue sections at the miRNA probe step
EZ-Retriever IR System	BioGeneX	MW015-IR	Microwave used for antigen retrieval of FFPE tissue sections
Hydrogen Peroxide Solution 30%	Sigma-Aldrich	HX0640-5	For Tissue preparation. Helps to block endogenous peroxidase activity and to yield highly colored products
miR-181c-3p miRNA Probe	Qiagen	339112	Example of miRNA probe used to optimize this protocol
miRCURY LNA miRNA ISH Buffer Set (FFPE)	Qiagen	339450	Used to dilute miRNA probe
Opal 620	Akoya Biosciences	FP1495001KT	Fluorophore to identify primary antibody
Opal 690	Akoya Biosciences	FP1497001KT	Fluorophore to identify primary antibody
Opal Anti-Ms + Rb HRP	Akoya Biosciences	ARH1001EA	Secondary antibody used as a detection agent
Pierce 16% Formaldehyde (w/v), Methanol-free	Thermo Fisher Scientific	28908	To make 3.7% PFA for tissue preparation
ProLong Diamond Antifade	Thermo Fisher Scientific	P36970	Mounting media
RNAscope Protease Plus	ACD Biotechne	322331	To permeabilize tissue sections by partially digesting proteins & expose target miRNA molecules
RNAse-free water	Corning	46-000-CM	To clean and prep all equipment, and to use as a dilutant when necessary
Tween 20	Sigma-Aldrich	P1379-25ML	To make TBS-T
Tyramide Signal Amplification Plus Cyanine 3	Akoya Biosciences	NEL744001KT	Fluorophore to identify miRNA
U6 and Scramble miRNA Probes	Qiagen	339459	Optional control probes
Xylenes Histological Grade	Thermo Fisher Scientific	UN1307	For deparaffinization