Myxomatous mitral valve degeneration is the most common heart disease in dogs, accounting for approximately 75% of all dogs with heart disease.1 It is a genetic disorder, but a causative mutation has not been identified.2 As mitral valve regurgitation progresses, there is structural and functional remodeling of the heart, which leads to volume overload and ultimately to CHF. Congestive heart failure is a devastating clinical syndrome characterized by severe cardiomegaly and clinical signs such as increased respiratory rate during sleep. A diagnosis of CHF is made on the basis of a combination of clinical signs, results of physical examination and diagnostic imaging, and response to diuretic treatment. It is often challenging to distinguish dogs with CHF from those in a preclinical stage of MMVD or with noncardiogenic respiratory signs. Because an accurate diagnosis of CHF is crucial to determine appropriate treatment and improve the long-term prognosis, blood-based biomarkers capable of detecting CHF have been evaluated extensively.3
Myocardial wall stress in response to volume overload increases the production of natriuretic peptides such as NT-proBNP from cardiomyocytes.4 Damage to cardiomyocytes causes dissociation of troponin from actin and allows leakage of cardiac troponin I into the extracellular space where it enters the circulation.5 Several factors could potentially account for the current limited usefulness of cardiac peptide biomarkers, including improper collection techniques, extended storage at room temperature, or multiple freeze-thaw cycles. Therefore, if proper handling techniques are not used, there may be degradation of these peptide biomarkers, which would thus yield inaccurate results.6 Furthermore, the effect of systemic diseases on the production or excretion of cardiac troponin I and NT-proBNP often results in false-positive or false-negative results for the diagnosis of CHF. Therefore, additional biomarkers that are specific to the stage of heart disease and released in proportion to the stage of heart disease are needed to determine the stage of MMVD, enhance the accuracy of diagnosis, and reliably predict the prognosis of dogs with CHF secondary to MMVD.
MicroRNAs are recognized as crucial regulators of gene expression in heart development and cardiac disorders.7 MicroRNAs are short (approx 22 nucleotides in length) noncoding RNAs that are transcribed from the genome. The biological function of these small RNAs is to negatively regulate gene expression through mRNA degradation or translational inhibition.8 One microRNA may regulate several target mRNAs and have substantial effects on downstream gene expression networks. The human genome is believed to encode approximately 1,000 microRNAs, and ≥ 60% of human protein-coding genes are regulated by these microRNAs.7–9 Recent studies7,9 have revealed that microRNAs are highly resistant to endogenous ribonuclease activity and that circulating microRNAs can be detected in the blood, which suggests that they may serve as excellent circulating molecular biomarkers for various cardiac disorders. Plasma miR-1 expression is significantly upregulated in human patients with acute myocardial failure, compared with results for healthy control subjects.10 A positive correlation exists between plasma miR-1 and troponin concentrations, which suggests that microRNAs can serve as superior biomarkers for the early detection of CHF.10 Furthermore, significant upregulation of circulating miR-133 has also been identified in human patients with symptomatic heart failure.11 It has been determined that circulating miR-423 concentrations are a potential prognostic indicator of CHF.12,13 In that study,12 the plasma concentration of miR-423 could be used to distinguished between CHF and noncardiogenic respiratory signs.
Next-generation sequencing technologies have advantages in terms of cost-effectiveness, sequencing speed, high resolution, and accuracy for genomic analyses, compared with microarray platforms. Expression profiles of circulating microRNAs via NGS technologies have not been reported for dogs with CHF secondary to MMVD,14 although microRNA characterization has been conducted with microarrays and real-time qRT-PCR assays.15,16 The objective of the study reported here was to elucidate the microRNA signaling network and characterize distinct expression profiles of circulating microRNAs through genome-wide sequencing of dogs with CHF secondary to MMVD.
Materials and Methods
Animals
Two groups of client-owned dogs (clinically normal vs CHF) were included in the study. On the basis of previous studies17,18 conducted by the use of NGS technologies, it was deemed necessary to include at least 7 dogs in each group to provide sufficient statistical power to detect differences between groups. Owner consent was obtained for the use of each dog. The study was approved by the Institutional Animal Care and Use Committee of Auburn University.
Inclusion criteria for dogs in the clinically normal group (American College of Veterinary Internal Medicine Stage A) included no echocardiographic evidence of heart disease, no clinical signs, no abnormalities on results of a CBC and biochemical analyses, and no history of medical treatment within the preceding 6 months. Dogs with acute-onset CHF (American College of Veterinary Internal Medicine Stage C) were enrolled on the basis of the following criteria: respiratory signs of tachypnea or dyspnea; radiographic evidence of cardiogenic pulmonary edema; echocardiographic evidence of a thickened mitral valve, severe mitral valve regurgitation, and marked enlargement of the left atrium; and no other systemic diseases as determined on the basis of results of a CBC and biochemical analyses. Dogs with concurrent cardiac disease other than MMVD, hemodynamically important cardiac arrhythmias, or evidence of other clinically relevant systemic disease were excluded. Dogs also were excluded if they had a history of CHF or had received drugs as treatment for cardiac disease. Congestive heart failure was defined as pulmonary edema that required furosemide administration. Pulmonary edema was diagnosed on the basis of clinical signs (cough, tachypnea, dyspnea, and orthopnea) and radiographic appearance (interstitial or alveolar pulmonary pattern). The left atrial size was considered to be severely enlarged when the left atrium-to-aortic root ratio was > 1.9. Mitral valve regurgitation was recorded as severe if color Doppler ultrasonographic mapping of the regurgitant jet revealed filling of > 50% of the area of the left atrium with concurrent severe enlargement of the left atrium.
Isolation of microRNAs
A blood sample (3 mL) was collected from each dog into EDTA-containing tubes via jugular venipuncture. For dogs with CHF, the blood sample was collected before administration of drugs for the treatment of acute CHF to avoid possible interactions between the drugs and the microRNA signaling network.
Each blood sample was centrifuged (1,500 × g for 15 minutes) within 10 minutes after sample collection. An aliquot (200 μL) of supernatant was used for isolation of microRNAs with a commercial kit,a with procedures conducted in accordance with the manufacturer directions. A synthetic housekeeping geneb (3 μL; 1.6 × 108 copies/μL) was added during the microRNA isolation process to serve as a synthetic control sample for normalization.
MicroRNA libraries and NGS
Extracted microRNAs were added into a small RNA library preparation set.c Total RNA (100 ng) containing microRNAs was subjected to reverse transcription (1 hour at 50°C) with a commercial kit.d The PCR assay amplification (15 cycles) was performed by use of a master mix.e The amplified PCR assay product was purified.f Yield and concentration of the prepared libraries were assessed.g Accurate quantification was assessed prior to use in sequencing applications.h Each library was diluted to a final concentration of 12.5nM and pooled in equimolar ratios prior to clustering. Cluster generation was performed,i and single-end NGSj was conducted to generate 15 million sequences/sample. Clustered flow cells were sequenced (56 cycles). Principal component analysis was performed to confirm there were distinct expression patterns between the clinically normal and CHF groups.
qRT-PCR assay
Expression of microRNAs that were selected from the NGS data and that are known to be associated with heart failure in humans was analyzed via a qRT-PCR assay to validate the NGS data. Total RNA (100 ng) isolated from each dog was subjected to reverse transcription.k A commercially available kitl and custom-designed primers that were validated and quality-controlled for canine-specific microRNAs were used to measure relative expression of plasma microRNAs. Thermocyclerm conditions were 95°C for 15 minutes, followed by 40 cycles of 94°C for 15 seconds, 55°C for 30 seconds, and 70°C for 30 seconds; a melting curve analysis was performed on the product. Each microRNA expression was normalized against results for the housekeeping gene. All experiments were performed in triplicate. Relative expression was calculated by the comparative threshold cycle method (ΔΔCt method), and the fold change relative to control samples was calculated.
Statistical analysis
A Shapiro-Wilk normality test was used to determine normality of the data distribution. Comparisons between the 2 groups (clinically normal vs CHF) were performed with a Student t test or Mann-Whitney U test, depending on normality of the data distribution, followed by post hoc testing with a Bonferroni multiple test.n Significance was set at values of P < 0.05.
Results
Animals
Nine clinically normal dogs were enrolled. Mean age of the dogs was 11.2 years (range, 8.2 to 13.8 years), and mean body weight was 5.8 kg (range, 3.9 to 8.1 kg). The group consisted of 3 Pomeranians, 2 Poodles, 2 Yorkshire Terriers, 1 Maltese, and 1 Dachshund. Eight dogs with CHF secondary to MMVD (confirmed on the basis of clinical signs and results of thoracic radiography and echocardiography) met the inclusion criteria for the CHF group (Supplementary Table S1, available at http://avmajournals.avma.org/doi/suppl/10.2460/ajvr.79.2.163). Mean age of the dogs was 9.7 years (range, 7.8 to 12.2 years), and mean body weight was 5.6 kg (range, 3.9 to 6.1 kg). The group consisted of 3 Maltese, 2 Chihuahuas, 1 Yorkshire Terrier, 1 Pomeranian, and 1 Shih Tzu.
Genome-wide sequencing of circulating microRNAs
Genome-wide sequencing of circulating microRNAs in canine plasma generated approximately 15 million single-end sequences/sample, and a total of 326 microRNAs were identified. Presence of abundantly expressed microRNAs, a median microRNA length of 22 bp, and a characteristic distribution of sequence frequency over sequence length suggested a successful enrichment of mature microRNAs for all libraries (Supplementary Figure S1, available at http://avmajournals.avma.org/doi/suppl/10.2460/ajvr.79.2.163).
Principal component 1 explained 47% of the differential expression of microRNAs between clinically normal dogs and dogs with CHF (Supplementary Figure S2, available at http://avmajournals.avma.org/doi/suppl/10.2460/ajvr.79.2.163). Principal component 2 revealed minimal variability (9%) of microRNA expression profiles within each group. Hierarchical clustering analysis provided distinct expression patterns of circulating microRNAs between clinically normal dogs and dogs with CHF (Supplementary Figure S3, available at http://avmajournals.avma.org/doi/suppl/10.2460/ajvr.79.2.163), which was in agreement with results of the principal component analysis.
A volcano plot was created to allow visual evaluation of the magnitude of differential expression for circulating microRNAs between the 2 groups (Figure 1). A difference in expression, with P < 0.05 and a minimal fold change of 2.5 (upregulation) or −2.5 (downregulation), was considered to be significant. Among the 326 microRNAs, 5 canine-specific microRNAs (miR-133, miR-1, miR-139, cfa-let-7e, and miR-125a) were significantly upregulated in plasma of dogs with CHF (range of fold increase, 2.6 to 23.1), compared with results for the clinically normal dogs (Table 1). Eighty-eight microRNAs were downregulated (range of fold decrease, −3.0 to −23.1; Supplementary Table S2, available at http://avmajournals.avma.org/doi/suppl/10.2460/ajvr/79.2.163).
Volcano plot of differentially expressed circulating microRNAs for 8 dogs with CHF, compared with results for 9 clinically normal dogs. The vertical dashed lines indicate the expression thresholds for fold changes of −2.5 and 2.5 in dogs with CHF, compared with results for the clinically normal dogs. The horizontal dashed line represents P = 0.05. Points in the bottom portion of the graph (gray circles) represent microRNAs that were not significantly upregulated or downregulated in dogs with CHF, whereas points in the top right and top left sectors represent microRNAs that are significantly upregulated (red circles) and downregulated (blue circles) in dogs with CHF.
Citation: American Journal of Veterinary Research 79, 2; 10.2460/ajvr.79.2.163
Volcano plot of differentially expressed circulating microRNAs for 8 dogs with CHF, compared with results for 9 clinically normal dogs. The vertical dashed lines indicate the expression thresholds for fold changes of −2.5 and 2.5 in dogs with CHF, compared with results for the clinically normal dogs. The horizontal dashed line represents P = 0.05. Points in the bottom portion of the graph (gray circles) represent microRNAs that were not significantly upregulated or downregulated in dogs with CHF, whereas points in the top right and top left sectors represent microRNAs that are significantly upregulated (red circles) and downregulated (blue circles) in dogs with CHF.
Citation: American Journal of Veterinary Research 79, 2; 10.2460/ajvr.79.2.163
Volcano plot of differentially expressed circulating microRNAs for 8 dogs with CHF, compared with results for 9 clinically normal dogs. The vertical dashed lines indicate the expression thresholds for fold changes of −2.5 and 2.5 in dogs with CHF, compared with results for the clinically normal dogs. The horizontal dashed line represents P = 0.05. Points in the bottom portion of the graph (gray circles) represent microRNAs that were not significantly upregulated or downregulated in dogs with CHF, whereas points in the top right and top left sectors represent microRNAs that are significantly upregulated (red circles) and downregulated (blue circles) in dogs with CHF.
Citation: American Journal of Veterinary Research 79, 2; 10.2460/ajvr.79.2.163
Differentially expressed circulating microRNAs as determined via NGS analysis of dogs with CHF secondary to MMVD.
| MicroRNA | Chromosome | Corrected P value | Fold change* |
|---|---|---|---|
| cfa-miR-133 | 7 | 0.001 | 23.1 |
| cfa-miR-1 | 7 | 0.007 | 12.3 |
| cfa-miR-139 | 21 | 0.010 | 5.2 |
| cfa-let-7e | 1 | 0.023 | 3.0 |
| cfa-miR-125a | 1 | 0.008 | 2.6 |
| cfa-miR-142 | 9 | < 0.001 | –17.3 |
| cfa-miR-128 | 23 | < 0.001 | –9.5 |
| cfa-miR-30c | 12 | < 0.001 | –8.7 |
| cfa-miR-423 | 9 | < 0.001 | –3.5 |
Positive values indicate upregulation, and negative values indicate downregulation.
Evaluation of microRNA expression by use of qRT-PCR assays
An NGS analysis provides direct counts of the absolute number of copies of each transcript as a measure of expression abundance, is species independent, has high sensitivity for low-abundance transcripts, and has excellent reproducibility. Evaluation of circulating microRNA expression profiles derived from results of NGS analysis was conducted with qRT-PCR assays. Relative expression of the 5 upregulated microRNAs was analyzed. The qRT-PCR assay results confirmed upregulation of circulating miR-133, miR-1, cfa-let-7e, and miR-125a in dogs with CHF (Figure 2). However, qRT-PCR assay results for miR-139 were not confirmatory because of high interindividual variability.
Mean ± SD expression (results of qRT-PCR assay) for 4 upregulated (on the basis of NGS analysis) microRNAs (miR-133 [A], miR-1 [B], cfa-let-7e [C], and miR-125a [D]) in plasma of dogs with CHF, compared with expression in plasma of clinically normal (CN) dogs. Notice that the scale on the y-axis differs among panels. *†‡Value differs significantly (*P = 0.04, †P = 0.008, and ‡P = 0.03) from the value for the clinically normal dogs.
Citation: American Journal of Veterinary Research 79, 2; 10.2460/ajvr.79.2.163
Mean ± SD expression (results of qRT-PCR assay) for 4 upregulated (on the basis of NGS analysis) microRNAs (miR-133 [A], miR-1 [B], cfa-let-7e [C], and miR-125a [D]) in plasma of dogs with CHF, compared with expression in plasma of clinically normal (CN) dogs. Notice that the scale on the y-axis differs among panels. *†‡Value differs significantly (*P = 0.04, †P = 0.008, and ‡P = 0.03) from the value for the clinically normal dogs.
Citation: American Journal of Veterinary Research 79, 2; 10.2460/ajvr.79.2.163
Mean ± SD expression (results of qRT-PCR assay) for 4 upregulated (on the basis of NGS analysis) microRNAs (miR-133 [A], miR-1 [B], cfa-let-7e [C], and miR-125a [D]) in plasma of dogs with CHF, compared with expression in plasma of clinically normal (CN) dogs. Notice that the scale on the y-axis differs among panels. *†‡Value differs significantly (*P = 0.04, †P = 0.008, and ‡P = 0.03) from the value for the clinically normal dogs.
Citation: American Journal of Veterinary Research 79, 2; 10.2460/ajvr.79.2.163
A computational algorithmo was used to predict putative target genes of downregulated microRNAs.20 Gene ontology analysis of these target genes indicated that downregulated microRNAs were involved in the processes of extracellular matrix signaling, endothelial-to-mesenchymal transition, and cardiac hypertrophy; phosphodiesterase activity; and TGF-β pathways. Four microRNAs (miR-30c, miR-128, miR-142, and miR-423) were selected from the 88 downregulated microRNAs as potential molecular biomarkers of CHF on the basis of results of gene ontology analysis and bioinformatics analysis and information in previous publications.12,13 These 4 microRNAs were found to have significantly lower expression in plasma of dogs with CHF, compared with expression for the clinically normal dogs, on the basis of results of qRT-PCR assays (Figure 3). Results of qRT-PCR assays were consistent with NGS data and supported the specificity of circulating microRNA expression profiles in dogs with CHF secondary to MMVD that were generated by use of NGS analysis.
Mean ± SD expression (results of qRT-PCR assays) for 4 downregulated (on the basis of NGS analysis) microRNAs (miR-142 [A], miR-30c [B], miR-128 [C], and miR-423 [D]) in plasma of dogs with CHF, compared with expression in plasma of clinically normal (CN) dogs. Notice that the scale on the y-axis differs among panels. *†‡Value differs significantly (*P < 0.001, †P = 0.005, and ‡P = 0.008) from the value for the clinically normal dogs.
Citation: American Journal of Veterinary Research 79, 2; 10.2460/ajvr.79.2.163
Mean ± SD expression (results of qRT-PCR assays) for 4 downregulated (on the basis of NGS analysis) microRNAs (miR-142 [A], miR-30c [B], miR-128 [C], and miR-423 [D]) in plasma of dogs with CHF, compared with expression in plasma of clinically normal (CN) dogs. Notice that the scale on the y-axis differs among panels. *†‡Value differs significantly (*P < 0.001, †P = 0.005, and ‡P = 0.008) from the value for the clinically normal dogs.
Citation: American Journal of Veterinary Research 79, 2; 10.2460/ajvr.79.2.163
Mean ± SD expression (results of qRT-PCR assays) for 4 downregulated (on the basis of NGS analysis) microRNAs (miR-142 [A], miR-30c [B], miR-128 [C], and miR-423 [D]) in plasma of dogs with CHF, compared with expression in plasma of clinically normal (CN) dogs. Notice that the scale on the y-axis differs among panels. *†‡Value differs significantly (*P < 0.001, †P = 0.005, and ‡P = 0.008) from the value for the clinically normal dogs.
Citation: American Journal of Veterinary Research 79, 2; 10.2460/ajvr.79.2.163
Discussion
Next-generation sequencing is a technique commonly used to evaluate microRNA expression profiles for their utility as molecular biomarkers of heart disease. Next-generation sequencing analysis as a high-throughput sequencing technique is not limited by a predefined number of probe design, cross-hybridization, or platform background issues.14 Furthermore, NGS provides a direct count of the absolute number of transcripts for expression abundance, is species independent, and has high sensitivity toward low-abundance microRNAs, excellent reproducibility, and multiplexing potential whereby it can amplify > 1 target by including > 1 pair of primers in the reaction. The qRT-PCR assay data of 9 microRNAs in Dachshunds with MMVD have been determined by use of amplification primers designed for human miR-302 sequences.15 Relative quantification of canine circulating microRNAs via microRNA PCR assay arrays has been conducted for dogs with MMVD, and microRNA-302 was proposed to be a regulator of TGF-α.16 To our knowledge, expression profiles of canine circulating microRNAs determined by use of NGS have not been reported for dogs with CHF secondary to MMVD. The study reported here indicated that the NGS technique successfully sequenced and quantified 326 mature microRNAs that were enriched in canine plasma, and the circulating microRNA panel of CHF was differentially expressed only in dogs with cardiogenic CHF.
Signaling implicated in MMVD includes serotonin, TGF-α, and components of the heart valve developmental pathways (eg, bone morphogenic protein and Wnt).21 High serotonin concentrations in the mitral valve and blood are pathological features of MMVD in dogs.22 Serotonin initiates TGF-α signaling, which in turn causes differentiation and proliferation of valvular interstitial cells and alters the valvular extracellular matrix. However, little is known about molecular mechanisms through which the serotonin gene becomes upregulated in MMVD. The overall complexity of molecular remodeling in MMVD suggests the involvement of additional regulatory mechanisms that modulate gene expression in the signaling networks. Expression profiles of circulating microRNAs may help researchers better understand the upstream molecular pathways associated with the development of CHF that cannot be explained on the basis of the currently accepted signaling mechanisms.
The microRNAs play an important role in heart development, cardiac hypertrophy, drug metabolism, and pathogenesis of cardiac disease.23 As upstream molecular events, microRNA signaling pathways mediate downstream transcriptional changes in CHF of humans. Thus, it can be speculated that microRNAs may serve as molecular biomarkers for CHF in dogs, and altered expression of microRNAs may characterize the onset of CHF. It has been reported that miR-133 is highly abundant in the ventricular myocardium of dogs.24 It has antifibrotic effects and regulates extracellular matrix proteins through the modulation of collagen expression; collagen type I is a target gene of miR-133, as determined on the basis of luciferase assays.25,26 Investigators of 1 study27 detected a decrease of miR-133 expression and an increase of collagen accumulation in the myocardium harvested from dogs with dilated cardiomyopathy experimentally induced by ventricular tachypacing. This molecular evidence of an increase in extracellular matrix synthesis can be explained, in part, by the downregulated miR-133 expression because of the reciprocal relationship between microRNA and mRNA expression. The aforementioned study27 revealed that cardiac fibroblasts play an important role for a greater reduction of miR-133 expression in left atrial tissues than in left ventricular tissues. On the other hand, volume overload in dogs with MMVD initiates collagen degradation and myocyte stretch as a result of matrix metalloproteinase and chymase released from mast cells.28 It is important to mention that an increase of miR-133 expression was detected in dogs with CHF secondary to MMVD in the present study. Therefore, findings of the present study may suggest that different types of heart disease are characterized by distinct microRNA expression profiles, and circulating microRNA expression profiles may not necessarily correlate with tissue-selective microRNA signaling pathways. An understanding of the molecular mechanisms responsible for gene regulation involved in extracellular matrix remodeling will possibly result in interventions to arrest cardiac progression. Further studies are needed to evaluate therapeutic potentials of the regulation of microRNA expression by microRNA mimics or microRNA antagonists to delay the onset of CHF.
Amounts of plasma miR-1 and miR-133 are increased 300-fold and 70-fold, respectively, in humans with acute heart failure.23 Interestingly, substantially increased plasma miR-1 concentrations return to physiologically normal concentrations with medical management of heart failure, which may suggest that changes in plasma miR-1 concentrations can be used as a biomarker for therapeutic monitoring of heart failure.10 Furthermore, overexpression of miR-1 can induce substantial apoptosis of cardiomyocytes, which may contribute to progression to CHF.29 Cardiac-specific miR-1 leads to developmental arrest as a result of premature differentiation of cardiomyocytes and their early withdrawal from the cell cycle.30 Exogenous miR-1 promotes arrhythmogenesis, whereas inhibition of endogenous miR-1 alleviates arrhythmogenesis. These results provide insight into the molecular mechanisms and confirm that miR-1 may play a role in the development of CHF.
Genome-wide analysis of the interactions between microRNA and mRNA has revealed that miR-128 targets downstream genes, phosphodiesterase 3, and TGF-β.13,31 In addition to the downregulation of miR-128 in dogs with CHF in the present study, it could be speculated that the miR-128 target genes would have increased expression. Overexpression of TGF-β can contribute to the valvular pathogenesis of dogs with MMVD.20 Inhibition of phosphodiesterase 3 activity by pimobendan has provided clinical and prognostic benefits for dogs with CHF.32 Investigating the molecular interaction and expression of miR-128 and phosphodiesterase 3 in preclinical dogs with MMVD is warranted to determine pharmacogenetic justification for the inhibition of phosphodiesterase 3.
Circulating miR-423-5p concentrations are increased in humans with acute heart failure caused by cardiomyopathy, and those concentrations could be used to distinguish patients with heart failure from healthy control subjects and patients with noncardiogenic respiratory distress.12,33 That study12 also found a correlation between plasma concentrations of miR-423-5p and NT-proBNP. However, it is important to mention that plasma miR-423 expression in the present study was downregulated in dogs with CHF caused by MMVD. This discrepancy in microRNA expression may indicate that the amounts of circulating microRNAs may differ depending on the type and stage of heart disease and myocardial wall stress.
MicroRNAs play an important role in mediating transcriptional changes observed in humans with CHF. This raises the possibility that microRNAs may potentially serve as molecular biomarkers for CHF in dogs and that altered expression of microRNAs may characterize different stages of heart disease. A distinct and validated panel of circulating microRNAs may serve as a predictor for the risk of CHF in dogs. However, it is possible that the expression profile of microRNAs identified in the present study may have been specific for small-breed dogs and MMVD. Effects of various types and stages of heart disease, biological variability, medical treatment, or other confounding cardiopulmonary risk factors on microRNA expression profiles will need to be investigated to overcome diagnostic limitations on the use of circulating microRNA concentrations in dogs. In the study reported here, microRNA expression patterns were distinct in dogs with CHF, compared with the expression patterns for healthy dogs. However, before clinical application can be considered, additional studies are needed to assess the predictive value with regard to identifying the canine patients that are likely to develop CHF by comparing microRNA profiles at different stages of heart disease.
Quantification of circulating microRNAs can be challenging because of the lack of proper endogenous control microRNAs for normalization. Expression of several endogenous circulating microRNAs can change as a result of cardiovascular disease or other risk factors.34 Preliminary experiments conducted by our research group with SNORD95 and RNU6-2 as endogenous circulating microRNAs failed to reveal accuracy in quantitative profiles (unpublished data). Caenorhabditis elegans miR-39 lacks sequence homology with canine microRNAs.35 An alternative approach with C elegans miR-39 as a control sample has successfully yielded microRNA expression profiles consistent with NGS data.35,36 Absolute quantification via the NGS platform has yielded significant differential expression of 93 microRNAs, with custom-designed primers specific for canine microRNAs for qRT-PCR assays. Small sample size and biological variability in microRNA copy numbers for each dog could have affected significant differences in the study reported here.
The use of NGS enabled comprehensive analysis of sequencing and quantification of microRNAs as an epigenetic regulator of heart disease. Data for the present study suggested that microRNA expression profiles were distinct for dogs with CHF, compared with the expression profiles for clinically normal dogs. Further validation of circulating microRNA expression profiles may lead to the discovery of novel diagnostic biomarkers for CHF in dogs. Circulating microRNAs may help guide treatment and lead to improved outcomes in dogs with heart disease.
Acknowledgments
Supported by a grant from the Animal Health and Disease Research Program of the College of Veterinary Medicine at Auburn University.
The authors declare that there were no conflicts of interest.
The authors thank Nripesh Prasad for assistance with the NGS analysis.
ABBREVIATIONS
| CHF | Congestive heart failure |
| MMVD | Myxomatous mitral valve degeneration |
| NGS | Next-generation sequencing |
| NT-proBNP | N-terminal pro B–type natriuretic peptide |
| qRT-PCR | Quantitative reverse transcription PCR |
| TGF | Transforming growth factor |
Footnotes
miRNANeasy plasma kit, Qiagen, Germantown, Md.
Caenorhabditis elegans miRNA-39, Qiagen, Germantown, Md.
Illumina, New England BioLabs Inc, Ipswich, Mass.
SuperScript III RT, Life Technologies Corp, Carlsbad, Calif.
LongAmp Taq 2X master mix, Illumina Inc, San Diego, Calif.
QIAquick PCR purification kit, Qiagen, Germantown, Md.
Qubit 2.0 fluorometer and bioanalyzer, Agilent Technologies Inc, Santa Clara, Calif.
Library quantification kit, KAPA Biosystems Inc, Wilmington, Mass.
Truseq single read cluster kit, Illumina Inc, San Diego, Calif.
Illumina HiSeq2000, Illumina Inc, San Diego, Calif.
miScript II RT kit, Qiagen, Germantown, Md.
miScript SYBR Green PCR kit, Qiagen, Germantown, Md.
CFX 96 thermocycler, Bio-Rad Laboratories Inc, Hercules, Calif.
GraphPad Prism, version 5.0, GraphPad Software Inc, La Jolla, Calif.
miRNATarBase2016, National Chiao Tung University, Hsinchu, Taiwan.
References
1. Detweiler DK, Patterson DF. The prevalence and types of cardiovascular disease in dogs. Ann NY Acad Sci 1965;127:481–516.
2. Madsen MB, Olsen LH, Haggstrom J, et al. Identification of 2 loci associated with development of myxomatous mitral valve disease in Cavalier King Charles Spaniels. J Hered 2011;102:S62–S67.
3. Smith KF, Quinn RL, Rahilly LJ. Biomarkers for differentiation of causes of respiratory distress in dogs and cats. J Vet Emerg Crit Care (San Antonio) 2015;25:311–329.
4. Boswood A. Biomarkers in cardiovascular disease: beyond natriuretic peptides. J Vet Cardiol 2009;11:S23–S32.
5. Langhorn R, Willesen JL. Cardiac troponins in dogs and cats. J Vet Intern Med 2016;30:36–50.
6. Oyama MA. Using cardiac biomarkers in veterinary practice. Vet Clin North Am Small Anim Pract 2013;43:1261–1272.
7. Devaux Y, Creemers EE, Boon RA, et al. Circular RNAs in heart failure. Eur J Heart Fail 2017;19:701–709.
8. Wong LL, Wang J, Liew OW, et al. MicroRNA and heart failure. Int J Mol Sci 2016;17:52–58.
9. Li M, Zhang J. Circulating MicroRNAs: potential and emerging biomarkers for diagnosis of cardiovascular and cerebrovascular diseases. BioMed Res Int 2015;7:30–35.
10. Ai J, Zhang R, Li Y, et al. Circulating microRNA-1 as a potential novel biomarker for acute myocardial infarction. Biochem Biophys Res Commun 2010;391:73–77.
11. Kuwabara Y, Ono K, Horie T, et al. Increased microRNA-1 and microRNA-133a levels in serum of patients with cardiovascular disease indicate myocardial damage. Circ Cardiovasc Genet 2011;4:446–454.
12. Tijsen AJ, Creemers EE, Moerland PD, et al. MiRNA423-5p as a circulating biomarker for heart failure. Circ Res 2010;106:1035–1039.
13. Masri S, Liu Z, Phung S, et al. The role of microRNA-128a in regulating TGFbeta signaling in letrozole-resistant breast cancer cells. Breast Cancer Res Treat 2010;124:89–99.
14. Creighton CJ, Reid JG, Gunaratne PH. Expression profiling of microRNAs by deep sequencing. Brief Bioinform 2009;10:490–497.
15. Hulanicka M, Garncarz M, Parzeniecka-Jaworska M, et al. Plasma miRNAs as potential biomarkers of chronic degenerative valvular disease in Dachshunds. BMC Vet Res 2014;10:205–212.
16. Li Q, Freeman LM, Rush JE, et al. Expression profiling of circulating microRNAs in canine myxomatous mitral valve disease. Int J Mol Sci 2015;16:14098–14108.
17. Grimes JA, Prasad N, Levy S, et al. A comparison of microRNA expression profiles from splenic hemangiosarcoma, splenic nodular hyperplasia, and normal spleens of dogs. BMC Vet Res 2016;12:272–284.
18. Zhao FR, Su S, Zhou DH, et al. Comparative analysis of microRNAs from the lungs and trachea of dogs (Canis familiaris) infected with canine influenza virus. Infect Genet Evol 2014;21:367–374.
19. Atkins C, Bonagura J, Ettinger S, et al. Guidelines for the diagnosis and treatment of canine valvular heart disease. J Vet Intern Med 2009; 23:1142–1150.
20. Chou CH, Chang NW, Shrestha S, et al. miRTarBase 2016: updates to the experimentally validated miRNA-target interactions database. Nucleic Acids Res 2016;44:D239–D247.
21. Orton EC, Lacerda CM, MacLea HB. Signaling pathways in mitral valve degeneration. J Vet Cardiol 2012;14:7–17.
22. Oyama MA, Levy RJ. Insights into serotonin signaling mechanisms associated with canine degenerative mitral valve disease. J Vet Intern Med 2010;24:27–36.
23. Xu J, Zhao J, Evan G, et al. Circulating microRNAs: novel biomarkers for cardiovascular diseases. J Mol Med 2012;90:865–875.
24. Vacchi-Suzzi C, Hahne F, Scheubel P, et al. Heart structure-specific transcriptomic atlas reveals conserved microRNA-mRNA interactions. PLoS One 2013;8:44–52.
25. Castoldi G, Di Gioia CR, Bombardi C, et al. MiRNA-133a regulates collagen 1A1: potential role of miRNA-133a in myocardial fibrosis in angiotensin II-dependent hypertension. J Cell Physiol 2012;227:850–856.
26. Duisters RF, Tijsen AJ, Schroen B, et al. MiRNA-133 and miRNA-30 regulate connective tissue growth factor: implications for a role of microRNAs in myocardial matrix remodeling. Circ Res 2009;104:170–178.
27. Chen Y, Wakili R, Xiao J, et al. Detailed characterization of microRNA changes in a canine heart failure model: relationship to arrhythmogenic structural remodeling. J Mol Cell Cardiol 2014;77:113–124.
28. Stewart JA Jr, Wei CC, Brower GL, et al. Cardiac mast cell- and chymase-mediated matrix metalloproteinase activity and left ventricular remodeling in mitral regurgitation in the dog. J Mol Cell Cardiol 2003;35:311–319.
29. Xu C, Lu Y, Pan Z, et al. The muscle-specific microRNAs miRNA-1 and miRNA-133 produce opposing effects on apoptosis by targeting HSP60, HSP70 and caspase-9 in cardiomyocytes. (Erratum published in J Cell Sci 2011;124:3187.) J Cell Sci 2007;120:3045–3052.
30. Wang Z, Lu Y, Yang B. MicroRNAs and atrial fibrillation: new fundamentals. Cardiovasc Res 2011;89:710–721.
31. Balakrishnan I, Yang X, Brown J, et al. Genome-wide analysis of miRNA-mRNA interactions in marrow stromal cells. Stem Cells 2014;32:662–673.
32. Häggstrom J, Boswood A, O'Grady M, et al. Effect of pimobendan or benazepril hydrochloride on survival times in dogs with congestive heart failure caused by naturally occurring myxomatous mitral valve disease. J Vet Intern Med 2008;22:1124–1135.
33. Fan KL, Zhang HF, Shen J, et al. Circulating microRNAs levels in Chinese heart failure patients caused by dilated cardiomyopathy. Indian Heart J 2013; 65:12–16.
34. Gupta SK, Bang C, Thum T. Circulating microRNAs as biomarkers and potential paracrine mediators of cardiovascular disease. Circ Cardiovasc Genet 2010;3:484–488.
35. Moldovan L, Batte KE, Trgovcich J, et al. Methodological challenges in utilizing miRNAs as circulating biomarkers. J Cell Mol Med 2014;18:371–390.
36. Corsten MF, Dennert R, Jochems S, et al. Circulating microRNA-208b and microRNA-499 reflect myocardial damage in cardiovascular disease. Circ Cardiovasc Genet 2010;3:499–506.
