Extracellular circulating miRNAs as stress-related signature to search and rescue dogs

Study approval was provided by the Research Ethics Committee of the University of Perugia (report n.2018-21 of 11/12/2018) according to Italian Ministry of Health legislation18. All methods were carried out following relevant guidelines and regulations and the study was carried out in compliance with the ARRIVE guidelines. Informed consent is not required as no human subjects were included in the study.

Animal enrolment

The dogs enrolled in the study were tested physically and behaviorally18. The day before SAR trials, all dogs were tested for routine health control by analyzing some blood biochemical parameters. The experiment included the dogs refer to the respective table (SM 1. SAR dog characteristics). Dog-handler teams were all specialized in the avalanche, surface, and rubble search at least by specific and increasing challenging training courses at the GdF Dog Breeding and Training Centre18 (Castiglione del Lago, Perugia, Italy) and the SAR Alpine School (Passo Rolle, Trento, Italy). All SAR dogs lived and worked with their handlers all year round to strengthen the dog-handler relationship, one of the most essential requirements for the success of a rescue operation.

SAR trial

The simulated SAR trial required dogs to find a target odor (a hidden operator and his breath, simulating a missing person) on a rubble field (30 × 35 m), within a maximum time of 15 min. Before starting SAR trials dog handlers ignored where the operator was hidden. During the search, both dog and the handler were allowed to enter the field area. Simulated SAR performance included some potential stress factors, such as the presence of people working on the rubble area, shouting and making loud noises with hammer and shovel, and sources of smoke to simulate a real post-earthquake scenery. The SAR trial is considered successful when the handler raises his arm to communicate that his dog has signaled the target odor. If the dog does not find the hidden operator within the assigned time, the SAR trial is considered a failure. To safeguard the welfare of the SAR dogs, during all trials, the authors observed if the dogs showed signs of stress or anxiety according to the codes described in our previous studies19,20. The results of these observations were only used for practical monitoring of the level of stress during the field search and interrupting it if necessary.

Sample collections time and experimental flowchart

Time points

All dog measures were evaluated in two experimental time points: T0—basal value at rest, immediately before the SAR trial; T1—immediately after the SAR trial. The points T0 and T1 allowed a comparison before and after the SAR trial.

Experimental flowchart

The research plan was divided into two experimental blocks; NGS SAR Trial and QPCR SAR Cluster, and both experimental blocks included T0 and T1 sampling times. NGS SAR Trial comprehended the 1st SAR trial, while QPCR SAR Cluster comprehended four SAR trials: 1st, 2nd, 3rd, and 4th. The four trials were held in different sessions, with different GdF dog-handler teams but the same experimental. The four trials included 22 dogs distributed respectively: 6 in the 1st, 6 in the 2nd, 6 in the 3rd, and 4 in the 4th Trial (see SM 1. SAR dog characteristics).

NGS SAR Trial investigation was planned to evaluate dog serum miRNome and identify differently expressed ecmiRNAs between T0 and T1. QPCR SAR Cluster validated in large sample size, including the 1st, 2nd, 3rd, and 4th trials, ecmiRNAs with different expression profiles between T0 and T1 (according to the data obtained from NGS SAR Trial) (Fig. 1).

Figure 1
figure 1

Research flowchart. The upper part of figure (a) shows the six dogs included in the NGS SAR Trial. The serum of six dogs was collected at two experimental time points (T0 and T1), RNA was extracted and processed for next-generation sequencing miRNome. The figure below (b) shows the second part of the research plan referred to as QPCR SAR Cluster. This part of the study includes blood dog sampling at T0 and T1 of the 2nd, 3rd, and 4th SAR trials. Serum RNA was extracted, reverse transcribed, and only the microRNAs differently expressed, detected in NGS SAR Trial, were validated by qPCR. The figure is created by G. Guelfi using PowerPoint 2021, Microsoft Corporation, USA.

The day before the SAR, blood parameters were evaluated in all dogs included in the research. At experimental times, T0 and T1, heart rate, rectal temperature, serum cortisol level, and differential levels of serum microRNA expression were evaluated. The performance of the SAR dogs was examined throughout the SAR operation.

Physiological parameter

The GdF veterinarian monitored heart rate (HR) in beats per minute with a stethoscope (3 M LITTMANN, Classic II SE, Milano, Italia) and measured rectal body temperature with a digital thermometer (Reckitt Benckiser SPA, MB Termo 7126500, Milano, Italia) at T0 and T1, in each SAR dog belonging to First SAR Trial, and QPCR SAR Cluster.

Blood sampling and serum RNA purification

Blood sampling was taken following the routine health check protocol in the Gdf training program. During the GdF veterinarian procedures, the handler asked the dog to stand and stay still for 1 min while gently manipulating and distracting it. Next, the handler asked the dog to sit; simultaneously, the veterinarian collected the 3 mL of blood sample via the radial vein into Vacuette Z Serum Sep Clot Activator (GREINER BIO-ONE). After centrifugation (2000 × g, 10 min), serum was obtained and stored at − 80 °C until use. Hemolysis was controlled in all serum samples to prevent the release of microRNA contained in the blood cells altered ecmiRNA profile (SM 2. Hemolysis assessment during sample preparations). Each serum sample was separated into two collection tubes: one tube (300 µL) for analyzing biochemical parameters and one (200 µL) to explore differential miRNA expression profiles. Total RNA, including ecmiRNAs, was extracted from 200 µL of serum using the miRNeasy Serum/Plasma Kit (QIAGEN CLC bio, Aarhus, Denmark) according to the manufacturer’s instructions, with an elution volume of 14 µL (SM 3. RNA extraction and Spike-in for qPCR validations) and then stored at − 80 °C until use. MiRNA concentration was assessed using the Qubit Fluorometer 4 and the Qubit microRNA Assay Kit (THERMOFISHER SCIENTIFIC, Kandel, Germany).

Spike-in to monitor RNA extraction, cDNA, and qPCR technical quality

Spike-in was used to perform quality control (QC) of ecmiRNA from biofluid samples. The QC assessment is essential because it enables obtaining sensitive and reliable microRNA data from low RNA content samples. During sample preparation, the spike-in control oligonucleotide was added to observe the sample QC of the NGS SAR Trial and the QPCR SAR Cluster (Fig. 2). To assess the technical reproducibility and linearity of the mapped NGS reads, in the samples targeted for NGS, 1 µl of 52 QIAseq miRNA Library QC Spike-in (QIAGEN CLC bio, Aarhus, Denmark) was added during extraction, as suggested by QIAGEN NGS Services proprietary protocol. Whereas, in the QPCR SAR Cluster samples, the spike-ins UniSp2 and UniSp4 were added during the RNA extraction phase to assess the efficiency and yield of the RNA isolation as recommended by the manufacturer. UniSp6 spike-in was included in reverse-transcription reaction to monitor cDNA synthesis performance and check the presence of PCR inhibitors.

Figure 2
figure 2

Quality control (QC) workflow. The NGS QC (left panel) was performed through a panel of 52 Spike-ins. The 52 RNA spike-in mix was added during RNA isolation to identify contamination and measure and validate assay parameters. In QPCR QC (right panel), 2 spike-in controls were added during RNA extraction to monitor the yield of RNA isolation, and 1 spike-in control was added to cDNA synthesis to monitor the efficiency of RT reactions. In addition, 3 spike-ins control PCR amplification. Spike-in synthetic oligonucleotides fragments were amplified and quantified by QPCR thanks to the respective primer pair. The figure is created by G. Guelfi using PowerPoint 2021, Microsoft Corporation, USA.

MiRNA library preparation

EcmiRNA sequencing experiments and data analysis were conducted by QIAGEN Genomic Services (Hilden, Germany). Library preparation is the first step of next-generation sequencing. Once libraries were prepared, the following workflow step will be high-throughput sequencing. The RNA used in this step was extracted (as described in SM 1) by adding to the serum 1 μL of a 52 miRNA Library QC Spike-In mix (QIAGEN CLC bio, Aarhus, Denmark) (SM 4. NGS serum miRNA Spike-in). The library preparation was done using the QIAseq miRNA Library Kit. A total of 5 µl RNA was converted into miRNA NGS libraries. Adapters containing UMIs (Unique Molecular Identifiers) were ligated to the RNA. Then RNA was converted to cDNA. The cDNA was amplified using qPCR (22 cycles), and during the qPCR, indices were added. After qPCR, the samples were purified. Library preparation was quality controlled using capillary electrophoresis (Agilent DNA 1000 Chip). Based on the quality of the inserts and the concentration measurements, the libraries were pooled in equimolar ratios. The library pools were quantified using qPCR. The library pool was then sequenced on a NextSeq (Illumina) to obtain 19 M 1 × 75 bp reads. Raw data was de-multiplexed, and FASTQ files for each sample were generated using the bcl2fastq (Illumina).

Next-generation sequencing

NGS Data analysis was performed via QIAGEN CLC Genomics Server v20.0.4 (Hilden, Germany). The miRNA-seq counts were normalized by a TMM (trimmed mean of M values) method21 to calculate the effective library sizes, which were then used as part of the per-sample normalization. Adapters containing Unique Molecular Identifiers (UMIs) were ligated to the RNA, and the RNA was then converted to cDNA. The 12 nt UMIs between the linker sequence and Illumina adapter sequence were extracted from each sample-demultiplexed sequencing reads, and the corresponding UMI annotated the reads. After removing all artificial sequences, we discarded reads with < 15 nt or > 55 nt. Reads with the same or 1 nt differences in their UMIs were grouped as UMI grouped reads, and reads from each UMI group were merged to produce a consensus sequence representing a single molecule from the starting RNA sample. These consensus reads were then aligned to miRBase v22 with a maximum mismatch of 2 nt, and the unmapped reads were subsequently mapped to the human genome GRCh38 (ENSEMBL).

Quantitative validation of ecmiRNAs by RT-qPCR

RT-qPCR reactions were performed to validate the expression profiling of selected miRNAs following the NGS results filtered with max group means more than 10, FDR p-value less than 0.05, and standard p-value significance at < 0.05.

A first-strand cDNA synthesis reaction was conducted using miRCURY LNA RT Kit (QIAGEN CLC bio, Aarhus, Denmark). Purified RNA (10 ng) was reverse-transcribed, including artificial RNA Spike-in control UniSp6. RT reaction was carried out according to the manufacturer’s guidelines (SM 5. Reverse transcription reaction). QPCR amplifications were executed utilizing miRCURY LNA SYBR Green PCR Kits (QIAGEN CLC bio, Aarhus, Denmark), in a final volume of 20 μL according to the manufacturer’s recommendations (SM 6. QPCR ecmiRNA validations). QPCR reaction was performed with 3 μL cDNA (diluted 1:50), and 9 specific pair primers (QIAGEN CLC bio, Aarhus, Denmark): 3 kinds of spike-in control (SM 7. Spike-ins Quality control); 4 kinds of potentials endogenous control (EC) of serum samples (SM 8. Selection of potential EC miRNAs); and 2 target miRNAs. The 9 pair primers are listed in Table 1. The amplification was performed in the StepOnePlus real-Time PCR system (APPLIED BIOSYSTEMS, California, USA). Three technical replicates were performed for each biological sample, and the average Cq (Quantification cycle according to the MIQE guidelines22) the averaged values of triplicate Cq were calculated; no-template controls were included in the analysis check contamination. The StepOnePlus Real-Time PCR software plotted the fluorescence intensity against the number of cycles and provided the Cq value using the automatic method. The 2^-ΔCq method was used to calculate the relative expression of the target miRNAs23.

Table 1 List of primers used in qPCR.

Analysis of reference miRNA stability

MiRNA qPCR analysis requires data normalization with the best EC to minimize data variation that can mask or exaggerate biological changes. Among the analyzed EC miRNAs: miR-320, miR-148a, miR-24, miR-23a, the best ECs were estimated thanks to four algorithms: GeNorm, Normfinder, BestKeeper, and the comparative method ΔCt (Cq is the recent nomenclature of Ct); RefFinder integration tool was then used to compare and integrate the four algorithms results.

Statistical analysis

RT-qPCR reactions were performed to validate the expression profiling of selected miRNAs following the NGS results filtered with max group means higher than 10, FDR p-value less than 0.05, and standard p-value significance at < 0.05. The quantification and statistical tests used were described in the figure legends and the methods section. No data were excluded from our studies. GraphPad Prism 8 (GraphPad Software Inc.) was used to plot all of the graphs and calculate statistical significance using a two-tailed Student’s t-test.

Read more here: Source link