Journal Club

Alzheimer’s Disease (AD) accounts for a large number of dementia cases resulting in impaired memory, thinking, and behavior. Risk factors for AD include age and family history, but unfortunately there is not yet a definitive way to predict if an individual will develop the disease. There are reference biomarkers that can indicate a higher risk of developing AD, such as APOE genotype. Carriers of the APOE4 allele, present in ~20% of the population, are at increased risk for AD. Cerebrospinal fluid (CSF) is a body fluid found in the brain and spine that cushions and protects the brain from injury. CSF protein biomarkers, such as Aβ42, tau and phospho-tau, are important in screening for brain disease, but these reference markers often lack the sensitivity and specificity necessary for clinical utility.

Extracellular RNA, specifically microRNA (miRNA), has been found in CSF and may serve as a useful resource for improved AD biomarkers. In a recently published study, the Saugstad lab from Oregon Health and Science University examined CSF from a large group of living donors to identify unique miRNA biomarkers enriched in AD patients. In the study, miRNA expression levels from 50 AD and 49 control subjects were assessed using TaqMan Low Density Arrays containing probes for 754 validated miRNAs. Each miRNA was given a “Multitest Score” combining the results of four statistical tests, and miRNAs that passed two or more of the tests were considered for further analyses.

Two statistical tests, log-rank and logistic regression, were used to identify candidates that were twice as likely to be associated with AD status as not. The other tests were two variants of random forest classifier, CART and CHAID, designed to select biomarker candidates able to reliably distinguish AD from non-AD status when grouped with random subsets of other miRNAs. 36 miRNA biomarker candidates were identified by at least two of these analyses. The researchers found that linear combinations of subsets of miRNA, and the addition of ApoE genotyping status, further increased the sensitivity and specificity of AD detection (Figure 1).

Figure 1. CSF miRNA biomarkers and APOE genotype predict AD status better together. AUC – Area Under the Curve; higher AUC indicates higher predictive power.

Reprinted with permission from IOS Press.

This study shows the potential use of miRNAs isolated from CSF as AD biomarkers. The stringent statistical analyses and large sample size together provided strength to these initial studies. These 36 candidate biomarkers are currently being tested in further validation studies in CSF from a new group of 120 donors, which will also include APOE genotyping and Aβ42 and tau protein levels. Ultimately, a combination of miRNA CSF biomarkers with existing reference biomarkers (APOE, Aβ42, tau) may provide a specific and sensitive tool for the diagnosis of AD in the clinic.

Citation:
MicroRNAs in Human Cerebrospinal Fluid as Biomarkers for Alzheimer’s Disease
Lusardi T, Phillips J, Wiedrick, J, Harrington C, Lind B, Lapidus J, Quinn J, Saugstad J. Journal of Alzheimer’s Disease (2017) 55: 1223-1233. doi: 10.3233/JAD-160835

Multiple sclerosis (MS) is an autoimmune demyelinating disease of the central nervous system (CNS). Currently, magnetic resonance imaging (MRI) is the most commonly used method to diagnose and monitor MS, but there is a poor correlation between MRI disease measures and clinical disability or disease progression in MS. MRI is also an expensive tool that might carry potential risks due to brain accumulation of contrast material (Kanda et al., 2015). In the last few years, a lot of effort has been invested in the identification of biomarkers for MS; however, to date, few of these findings have proven clinically useful. Thus, there is a strong unmet clinical need for objective body fluid biomarkers to assist in early diagnosis, predicting long-term prognosis, monitoring treatment response, and predicting potential adverse effects in MS.

Circulating miRNAs have been detected in several body fluids (Cortez et al., 2011) where they are highly stable as they are resistant to circulating ribonucleases (Mitchell et al., 2008). Their stability, along with the development of sensitive methods for their detection and quantification (Guerau-de-Arellano M. et al., 2012), makes them ideal candidates for biomarkers. We previously reported changes in circulating plasma miRNAs in MS patients (Gandhi R. et al., 2013). In a new study, our group investigated serum miRNAs as biomarkers in MS as part of an NCATS-funded UH2 initiative. We found that several serum miRNAs were differentially expressed in MS, were associated with disease stage, and correlated with disability.

Study Design (Figure 1): Serum from 296 participants including patients with MS, other neurologic diseases (Alzheimer’s disease and amyotrophic lateral sclerosis), inflammatory diseases (rheumatoid arthritis and asthma), and healthy controls (HC) were tested. miRNA profiles were determined using LNA (locked nucleic acid) based qPCR. MS patients were categorized according to disease stage and disability. In the discovery phase, 652 miRNAs were measured from the serum of 26 MS patients and 20 healthy controls. Those miRNAs from the discovery set that were significantly differentially expressed (p <0.05) in cases vs controls were validated using qPCR in 58 MS patients and 30 healthy controls.

Note: Results in the current study were normalized to the four most stably expressed miRNA across all the subjects. We agree with other blogs posted on exRNA.org suggesting that there is an immediate need to identify reference miRNA/exRNA that could be used for data normalization.

Figure 2: Differentially expressed circulating miRNAs as biomarkers in Multiple Sclerosis (MS). Up to top five miRNAs with p<0.05 are represented for each group comparison; a) MS, b) relapsing remitting MS (RRMS) and secondary progressive (SPMS) compared to the healthy control (HC), c) RRMS vs. SPMS, and d) the correlation of miRNA with the expanded disease severity scale (EDSS).

Results: We found 7 miRNAs (p<0.05 in both discovery phase and validation) that differentiate MS patients from healthy controls; miR-320a up-regulation was the most significantly changing serum miRNA in MS patients. We found 8 miRNAs that differentiated relapsing-remitting MS (RRMS) from HC. Among these, miR-484 up-regulation in RRMS patients showed the strongest association. When comparing secondary progressive MS (SPMS) patients to HC, 34 miRNAs significantly differentiated between the groups in both phases, with miR-320a up-regulation showing the strongest link. We also identified two miRNAs linked to disease progression, with miR-27a-3p being the most significant. Ten miRNAs correlated with degree of disability according to the Kurtzke Expanded Disability Status Scale (EDSS), of which miR-199a-5p had the strongest correlation with disability. Of the 15 unique miRNAs we identified in the different group comparisons, 12 have previously been reported to be associated with MS, but not in serum. Kegg Pathway Analysis showed that significant and differentially expressed miRNAs target important immune functions and are related to the maintenance of neuronal homeostasis. For example, miR-27a-3p, the strongest miRNA to distinguish RRMS from SPMS and progressive MS (PMS) (up-regulated in the relapsing form as compared to the progressive forms) shows a strong link to both the neurotrophin signaling pathway and the T cell receptor signaling pathway. Other studies have shown that miR-27a-3p targets multiple proteins of intracellular signaling networks that regulate the activity of NF-κB and MAPKs 6. As a consequence, miR-27a inhibits differentiation of Th1 and Th17 cells and promotes the accumulation of Tr1 and Treg cells (Min S. et al., 2012). It has also been shown that miRa-27-3p is up-regulated in MS active brain lesions and that the level of miR-27a-3p in CSF is reduced in patients with dementia due to Alzheimer’s disease (AD) (Frigerio C.S. et al., 2013). Of all the miRNAs, miR-486-5p was identified in the largest number of comparisons. It correlates with EDSS and is up-regulated in MS compared to HC, to other neurological diseases, and to other inflammatory diseases. This particular miRNA was found to be associated with TGF-beta signaling pathways and is a known tumor suppressor (Oh H.K. et al., 2011). miR-320a has been previously described to be highly expressed in B cells of MS patients and was suggested to contribute to increased blood-brain barrier permeability due to regulation of MMP-9 (Aung L.L. et al., 2015). Pathway analysis links this miRNA to cell-to-cell adhesion pathways, another indication that it may be linked to blood-brain barrier permeability.

The current study is the most comprehensive evaluation to date of the role of serum miRNAs as biomarkers in MS, with the largest sample size and employing two independent cohort designs. One limitation of our study is that participant subject samples were collected from a single MS center. Further external validation of our results will require investigating samples from patients at other centers. We are currently performing such multicenter studies, which may also increase the power of our results. A second limitation of our study is the relatively small number of participants who contributed to each group comparison. Future work will require larger sample sizes to ensure that we have sufficient power to detect miRNAs with smaller effect sizes. Although miRNAs have been studied in cells and the CNS of MS patients, ours is the first comprehensive investigation of serum miRNAs.

Conclusions: Our findings identify circulating serum miRNAs (Figure 2) as potential biomarkers to diagnose and monitor disease status in MS. These findings are now being tested using patient samples obtained from other international MS centers. We are now investigating the role of miRNA as biomarkers for disease prognosis and treatment response in MS.

Acknowledgements: This study is a highly collaborative project, and I thank my whole team at the Ann Romney Center for Neurologic Diseases & MS Center for their contribution. The grant TR000890 is supported by the NIH Common Fund, through the Office of Strategic Coordination / Office of the NIH Director.

Cell culture is a staple of modern biology, and Fetal Bovine Serum (FBS) is an essential component of many cell culture protocols. A specific use for FBS is to supply nutrients to cells and to stimulate their growth. Another role of FBS in cell culture research is to represent the complexities and functionality of endogenous biological environments; however, precisely this complexity has long been a potential confounding factor for researchers. For example, cytokines in FBS can lead to the stimulation of cells, thus producing unintended experimental environments. Despite these disadvantages, FBS retains a prominent role in modern cell culture, with estimated sales as high as 700,000 liters per year. Because of its ubiquity in cell culture research, it is critical to investigate how the components of FBS may be influencing experiments and downstream analysis.

Variability and uncertainty in the composition of FBS is especially problematic for studies that evaluate cellular secretions. For example, to successfully determine the array of RNA secreted by cultured cells, we need to know the extent to which the medium is contaminated by exogenous RNA. Additionally, extracellular RNA (exRNA) is not only found distributed freely throughout the liquid medium, but it is also often found packaged inside of extracellular vesicles (EVs) or lipoprotein complexes. Therefore, in a paper released online yesterday, Wei et al. evaluated how the RNA composition of FBS might be confounding research.

The authors first evaluated exogenous RNA contamination. They grew cultures of a cell type known not to express a particular RNA, then evaluated the presence of that RNA in the culture media. If that RNA was found, its origin was probably the media itself. For example, the authors demonstrated that miR-122, a liver-specific miRNA, is present in media from cultured glioma cells, suggesting that its source is likely FBS itself. They then attempted to deplete RNA from FBS via ultracentrifugation, but despite a 24 hour spin at 100,000g, about 75% of total RNA remained in the supernatant. This result has also been found by researchers attempting to deplete FBS of RNA-containing EVs and emphasizes the difficulty of producing media truly free from contaminating RNA.

These results led the authors to ask whether existing studies have wrongly attributed the presence of exRNA to a particular experimental procedure or cell type, when it should be recognized as a component of the FBS in the cell culture media. To answer this question, the authors first broadly profiled the RNA composition of FBS using RNA sequencing. They determined that between 9% and 22% of FBS RNA mapped to the human genome, depending on the stringency of the mapping algorithm and FBS preparation. They also checked for the presence of bovine-specific RNA in existing human cell culture exRNA datasets, finding levels as high as 17%, with samples from exosomes (a type of EV) containing particularly high levels. Finally, they demonstrated experimentally that bovine-specific transcripts are taken up into cells, interfering not only with exRNA analysis but also with intracellular RNA studies.

Moving forward, a significant remaining issue is deciding how to treat conserved RNA known to be present in both FBS and the cell line under study. Switching from FBS to purely chemically defined media can help with this problem, but it is not possible for all cell types and experimental conditions. Alternatively, a quantitative analysis of the chemical composition of the media might make it possible to estimate which RNAs are secreted by the cells of interest by filtering out known FBS RNAs from the total RNA pool.

This research cautions us to be careful in the design and interpretation of experiments to identify extracellular RNAs that use FBS in culture media. The paper, Fetal Bovine Serum RNA Interferes with the Cell Culture derived Extracellular RNA, released in Scientific Reports yesterday, is authored by Zhiyun Wei, Arsen O. Batagov, David R. F. Carter, and Anna M. Krichevsky.

Extracellular vesicles (EVs) play an important role in cell-to-cell communication. Recently, EVs have been shown to be involved in immune modulation, tumor biology, and tissue regeneration. The mechanisms of action of EVs are associated with their ability to stimulate target cells directly and to transfer proteins, biologically active lipids, and nucleic acids to the target cells. In fact, mRNAs, long non-coding RNAs (lncRNAs), and microRNAs (miRNAs) can be compartmentalized into EVs, escape enzymatic degradation, and be delivered to target cells. This horizontal transfer of extracellular RNAs carried by EVs can induce epigenetic alterations in recipient cells. The result is a change of phenotype or even a genetic and functional reprogramming of the recipient cells. Furthermore, EVs carry a selection of miRNAs different from the miRNAs most expressed in the cells of origin. However, little is known about the mechanisms of miRNA enrichment in EVs.

Argonaute2

Alix

We hypothesized a possible interaction between the Alix and Argonaute 2 (Ago2) proteins. The resulting complex may have a role in miRNA transport into EVs. Ago2 is a reasonable candidate to play a role in miRNA packaging within multivesicular bodies during EV biogenesis because of its central role in miRNA maturation. We observed that Ago2, as well as several other ribonucloproteins involved in RNA storage and stability, is expressed in EVs derived from adult human liver stem-like cells (HLSCs). Cells which express mesenchymal and embryonic markers.
 

Alix is a multifunctional protein commonly used as a marker of EVs. It is an accessory protein of the Endosomal Sorting Complex Required for Transport (ESCRT), and several studies indicate that ESCRT is involved in the biogenesis of EVs.

We observed that HLSC-derived EVs express both Alix and Ago2. Co-immunoprecipitation (Co-IP) experiments with Alix or Ago2 antibody showed that the two proteins are associated. We also found that the miRNAs enriched in HLSC-EVs precipitate with the Alix – Ago2 complex. After the incubation of HLSC-EVs with human endothelial cells, we observed that miRNAs from HLSC-EVs are transferred to these cells.
 

After the silencing of Alix expression in HLSCs, we observed the absence of both Alix and Ago2 proteins in EVs derived from the knockdown HLSCs and a strong reduction in the number of miRNAs normally enriched in HLSC-EVs. On the other hand, EV size, surface expression of CD63 and Tsg101, and the number of released EVs were not affected. After incubation with endothelial cells, EVs derived from Alix-knockdown HLSC do not transfer miRNAs to cells.

Alix is known to be involved in endocytic membrane trafficking and cytoskeletal remodeling. It is also associated with the ESCRT machinery, which participates in processes of vesiculation and cargo sorting, including multivesicular body biogenesis. Our data suggest that Alix binds Ago2 and drives it into EVs together with the associated miRNAs.
 

This might be a general mechanism of miRNA transport into EVs, common to other cell types. Enrichment of a selected set of miRNAs might also depend on the affinity of miRNAs for carrier proteins such as Ago2.

Source: Iavello A, Frech VS, Gai C, Deregibus MC, Quesenberry PJ, Camussi G. Role of Alix in miRNA packaging during extracellular vesicle biogenesis. Int J Mol Med. 2016 37:958-966. doi: 10.3892/ijmm.2016.2488. PMID: 26935291.

The potential of extracellular vesicle (EV) RNA as biomarkers of disease is increasingly being recognized. Circulating extracellular RNAs can potentially indicate the presence of disease without the need for an invasive biopsy of diseased tissue. A recent study by Yuan et al (Scientific Reports, Jan 2016) performed a systematic analysis of circulating exRNA in plasma obtained from 50 healthy individuals and 142 cancer patients. The authors conducted the largest RNA sequencing study reported to date for profiling circulating extracellular RNA (exRNA) species in order to provide useful insights into their baseline expression level. High-throughput multivariate statistical analysis identified a set of RNA candidates that were associated with age, sex, and cancer type.

The study directly addressed a major challenge in the field of exRNA, namely the lack of a baseline reference to accurately determine RNA abundance. Currently qRT-PCR is the most common method used to quantitate gene expression, but it relies on well-established reference controls for normalization. However, information on reference controls has not been established for exRNA. Control RNAs used for normalization in qRT-PCR experiments of cellular RNA cannot be reliably used for exRNA. Exogenous spike-in RNAs such as those derived from species like C. elegans cannot be used to normalize across biological or pathological states. Here the authors surveyed a pool of almost 200 exRNA profiles to identify a few potential reference candidates for exRNA quantification. Most candidate were miRNAs like miR-99a-5p. The study also provided convincing evidence for the presence in plasma of species of RNA other than miRNA, such as piwiRNA (see figure).

The dataset generated by this study is available in the exRNA Atlas for other researchers to explore. Future studies will be needed to validate these findings; however, with the current study we have taken a big leap towards the goal of determining the biomarker potential of exRNA for human diseases.
 
 

RNA species detected in the plasma of 142 cancer patients and 50 healthy controls
Source: Scientific Reports / CC BY 4.0

The ability to perform small RNA sequencing on extracellular RNA samples allows us to measure, at least in a semi-quantitative manner, the abundance of miRNAs, piRNAs, fragments of tRNAs, long non-coding RNAs, and coding RNAs. One of the most significant advantages of RNA-Seq technology is that it can detect and measure any RNA that is present, whether or not it is a known sequence. Nonetheless, there are important unanswered questions about the accuracy of RNA-Seq and the optimal approach for processing the data obtained. All RNA-seq experiments are subject to sources of systematic variation such as library size, transcript length, and G-C content (Dillies et al., 2013). Small RNA-seq experiments are further impacted by the highly non-normal distribution of expression of different small RNAs, particularly miRNAs. Often a few miRNAs account for a very large fraction of total reads while the vast majority of miRNAs each contribute a small percentage of reads. Moreover, sample input amounts of extracellular RNA are often extremely limited, increasing the potential for both sampling error and experimental bias.

To address these issues, a number of normalization methods have been developed which can be assigned three basic categories: 1) scaling; 2) normalizing in order to achieve similar data distributions; and 3) Reads Per Kilobase Mapped (RPKM). There appears to be a lack of consensus regarding the optimal normalization method, but it is known that different methods can result in different results in downstream analysis, particularly differential expression analysis (DE).

Two studies highlighted here, Garmire and Subramaniam, 2012 and Dillies et al., 2013, do not agree on the best way to do normalization for small RNA-seq, but point to a number of methods analyzed and elucidate some of the key problems. We review these studies briefly here and point out the importance of rigorous comparisons of normalization methods for the future of differential comparisons of data sets.

The first class of methods, scaling methods, involves application of a standard linear mathematical operation to each sample. Scaling generally means changing the size of something, and normalizing by simply dividing by the total read number (going from read numbers to read fractions) is probably the simplest kind of scaling. Scaling approaches to normalization include global scaling, Lowess, and the Trimmed Mean Method (TMM) (Robinson and Oshlack, 2010). These approaches each use a different method to calculate a linear scaling factor. Global scaling uses a factor that is based on the difference in the means of the data sets to be compared. Lowess normalizes based on a multiple-regression model. TMM determines a scaling factor, which is the weighted trimmed mean of log expression. This factor is calculated after double trimming values at the two extremes based on log-intensity ratios (M-values) and log-intensity averages (A-values). According to Garmire and Subramaniam, variability in estimated RNA content may be even more pronounced for microRNA-seq datasets after application of this method.

Another general approach to normalization is to preserve aspects of the distribution of the data among different data sets. As with scaling, there are a number of different approaches to achieving matched data distributions, including quantile, Variance stabilization (VSN), the invariant method (INV), and DESeq. Quantile normalization has been extensively used for microarray data, with the goal of making the distribution of expression levels across samples similar. Conditional quantile normalization is a modification of quantile normalization that combines robust generalized regression and quantile normalization (Hansen, Irizarry, and Wu, 2012). The goal of the VSN method is to make the distribution of variance across different levels of expression similar. In INV normalization, a set of invariant miRNAs are selected, which are then used with one of the other methods (such as Lowess or VSN). In DESeq, a scaling factor for each sample in the dataset is obtained by computing the median of the ratios of each gene in one sample over the geometric mean of that gene across all samples. The same scaling factor is then applied to the read counts for all of the genes in that sample. RPKM is a method that has been widely used for long RNAseq datasets. RPKM is performed on each sample separately, and consists of taking the ratio of the number of counts for a given gene and the product of the total counts for all genes and the mature transcript length for that gene. Methods such as DEseq and TMM rely on the assumption that most genes are not differentially expressed (Dillies et al., 2013), which may not hold true for all microRNA-seq data sets.

Garmire and Subramaniam compared a number of normalization methods applied to mammalian microRNA-Seq data using two publicly available datasets that were chosen by the authors due to the availability of matched PCR data. They assessed the performance of the normalization methods by calculating the mean square error (MSE) and the Kolmogorv-Smirnov statistic (which is a measure of the difference of two distributions), as well as comparison with PCR data on the same samples and inspection of the results of differential expression. The authors show that Lowess, quantile and VSN normalization resulted in a smaller MSE, while TMM and VSN produced a higher MSE. Similarly, TMM and INV resulted in a larger K-S statistic, indicating a bigger change in the distribution. Quantile and Lowess normalization also had the best concordance with qPCR data. The authors thus concluded that Lowess and quantile normalization performed better than other methods studied. The primary limitations of this study include the fact that it did not incorporate strategies to compare the results of the normalization method to any “gold standard” for which the real distribution was known, and that the number of data sets used for the analysis was very small. In addition, other methods have been implemented since the publication of their study.

Dillies et al. published another comparison of normalization methods for RNAseq data in about the same time frame. Most of the datasets used were long RNAseq data, but a murine microRNA dataset was included. The seven methods compared in this study included TC (total counts), UQ (Upper Quartile), Median, DEseq, TMM, quantile and RPKM. TC, UQ and median are scaling approaches that are quite similar, involving the calculation of a scaling factor based on either the ratio of total counts (TC), upper quartile of counts (UQ) or median of counts. The authors assessed these normalization methods on real as well as simulated data. In their analysis, it is apparent that when boxplots of counts before and after normalization are assessed, the differences are most apparent in conditions where there are large differences in library size between samples. These differences do not improve after TC and RPKM normalization of the microRNA-seq data. Other features of microRNA-seq data that influence the results of the normalization according to the authors are the presence of high-count genes and a large number of 0 counts. Quantile normalization increases the intra-condition variability in the murine miRNA data. Based on the results of differential expression in this study, it is apparent that the differences in the final results depend on the normalization method and not on the model chosen to assess for differential expression (DESeq or TSPM). In the simulated data, which contains high count genes and may resemble microRNA seq data more closely than the other datasets, only DESeq and TMM were successful in achieving a low false positive rate and high power. The authors conclude that DESeq and TMM are the methods of choice based on their ability to perform in the presence of different library sizes and composition.

The differences in the conclusions from these two studies are likely due to the different normalization methods and specific datasets each used, and on the parameters they used to evaluate the performance of the normalization methods. Since the Garmire and Subramanian paper looks only at miRNA data and the Dillies et al. paper looks at a mix of data, including miRNA and mRNA data, it is difficult to compare the results. We suggest that the additional complexities of compiling mRNA levels from multiple read sequences may confuse the latter analysis. However, both studies agree that the use of different normalization approaches can result in significant differences in downstream differential expression results. One of the key weaknesses of both of these papers is the lack of data from “gold standard” datasets for which the quantities of the different RNAs were definitively known.

For this reason, it would be valuable to develop the tools necessary for rigorous comparison of the available normalization methods. Such tools may include the generation of standardized data sets, such as small RNA libraries constructed from purely synthetic miRNAs, for which the content can be completely controlled, or RNAs from biological specimens that contain synthetic miRNAs spiked in at controlled concentrations. Corresponding qPCR results, used with calibration curves, should be generated for such datasets, which will serve as an orthogonal measurement technology for developing and evaluating normalization methods. Overall, this important topic needs careful attention for the establishment of reference exRNA profiles, and for the realization of the full potential of the powerful technology of high throughput RNA-seq.

* * *

Note that one of us recently presented a web seminar on a related topic, Understanding and using small RNA-seq, that is available for viewing. Also, members of the ERCC will be jointly presenting a workshop on Data Normalization Challenges and Solutions as part of the CHI conference Extracellular RNA in Drug and Diagnostic Development in Cambridge, MA, 3-6 April 2016. See the Event page for more details.

Heart failure and sudden cardiac death are end-stage manifestations of coronary heart disease worldwide. With an aging population and improvements in therapy for coronary revascularization during myocardial infarction, the number of patients proceeding to advanced heart failure is growing. The role of extracellular RNAs (exRNAs) in the field of cardiovascular medicine has been growing in parallel, as diagnostic, prognostic, and potentially even mechanistically and physiologically relevant biomarkers.

Our interest in this field grew out of an initial foray into the world of advanced heart failure patients referred for biventricular pacemaker therapy. Ventricular “dyssynchrony” due to delayed activation of the left ventricular (LV) lateral wall is present in nearly 50% of patients with symptomatic advanced heart failure (HF) and reduces effective LV function. Cardiac resynchronization therapy (CRT) is a type of pacemaker that can reduce dyssynchrony and mitigates adverse cardiac remodeling and improves prognosis in up to 65-70% of individuals. Unfortunately, even in patients who meet criteria for CRT, 30% or more do not derive hemodynamic or clinical benefit, and efforts to define clinical, image-based, plasma or electrocardiographic biomarkers to predict responsive patients have not been very successful. Given the role of miRNAs in regulating gene networks relevant to cardiovascular diseases (e.g., fibrosis, arrhythmia), we became interested in investigating ex-RNAs as potential markers of CRT responsiveness.

To identify circulating miRNAs that predict response to CRT, we assessed plasma miRNA profiles prior to CRT in patients with advanced HF and dyssynchrony (HF_DYS) with or without subsequent echocardiographic improvement after CRT. In this group, we discovered a set of miRNAs disregulated in responders, but focused after statistical efforts on microRNA-30d, which was a strong, independent marker of risk in adjusted logistic and linear regression. Baseline concentrations of miRNA-30d were associated with CRT response.

Clinical research identifies a novel ex-RNA biomarker of disease. (A) Logistic regression for CRT responsiveness; (B) Linear regression of changes in LV ejection fraction post-CRT with miR-30d concentration; (C) Logistic probability function of responsiveness by miR-30d; (D) ROC curve of miR-30d versus QRS duration. (E) Changes in miR-30d at 6 months after CRT.

Given that CRT responsiveness may derive from “fixing” the electrical delays between the lateral and septal wall of the ventricle, we then found that miR-30d is more highly expressed in the lateral wall of the LV in a canine model of dyssynchrony. Mechanistically, miR-30d was synthesized and released by cardiomyocytes (CM) in response to increased mechanical stress, and mediated CM hypertrophy with adaptive features, including protection against TNF-α-induced CM apoptosis.

To further explore the functional role of miR-30d, we identified miR-30d targets that are dynamically regulated in the canine model of HF_DYS. Using a LIMMA (linear models for microarray data) approach, genes whose expression was inversely correlated to miR-30d expression (i.e. genes which were down-regulated in the lateral wall of dyssynchronous dogs where miR-30d levels were highest compared to the septal wall) were identified, and we screened these and other putative miR-30d targets using the Tarbase and TargetScan prediction algorithms. Subsequently, using the IPA (Ingenuity Pathways Analysis) approach we generated predicted functional pathways associated with the observed changes in miR-30d expression. Several of these genes affect multiple biologically relevant pathways in the heart, specifically, LIMS1, PPP1R14c, MAK3K13, and JAK1. In addition, several other miR-30d targets with known roles in cardiac hypertrophy were identified using Tarbase. Intriguingly, one of these, mitogen associated protein kinase 4 (MAP4K4), a downstream mediator of tumor necrosis factor-alpha (TNF-α) has recently been identified as a target of miR-30d in pancreatic tissue 1 and has been shown to play a key role in TNF-mediated inflammatory processes. Because TNF-α not only plays an important role in pathogenesis of cardiac remodeling, but also is differentially expressed in the lateral and septal walls in the HF_DYS canine model 2, we deemed MAP4K4 a potentially important target of miR-30d in HF_DYS.

To validate candidates identified through bioinformatic analyses, we assessed the ability of miR-30d overexpression in CMs to effectively silence the mRNAs for the putative targets. Transient transfection of CMs with miR-30d caused a significant decrease in the mRNA levels of LIMS1, MAP4K4, and PPP1R14c relative to scramble transfected cells. Given the involvement of TNF-α signaling in the pathophysiology of HF_DYS as described above, we focused on the interaction between miR-30d and MAP4K4. MAP4K4 protein was significantly decreased in the lateral wall in HF_DYS compared to lateral wall of control animals, correlating inversely with miR-30d levels in these two regions. Conversely, in the CRT lateral wall (where wall stress presumably decreased with resynchronization), miR-30d levels fell compared to HF_DYS, and MAP4K4 protein level returned to control levels, suggesting that MAP4K4 was an in vivo target of miR-30d in these models. Treatment of scramble transfected CMs with TNF-α led to a robust increase in MAP4K4 mRNA, which was markedly attenuated by miR-30d overexpression. miR-30d transfection inhibited TNF-α mediated apoptosis in CMs, suggesting that miR-30d may be protective against the maladaptive effects of TNF-α. miR-30d transfection also prevented TNF-α-mediated increase in molecular markers associated with pathological hypertrophy and fibrosis (ANP, TIMP1 and β-MHC).

Finally, we measured levels of high sensitivity troponin T, a marker of myocardial injury, in our patients, demonstrating an inverse association between miR-30d and troponin TnT (Spearman r=-0.51, p=0.001), suggesting that higher levels of miR-30d may be cardioprotective in human HF patients.

These experiments provide ‘in vivo’ supportive evidence for our hypothesis that miR-30d may be protective against inflammation (TNF-α) induced cardiomyocyte injury, thereby promoting cardiomyocyte survival and favorably influencing the degree of cardiac remodeling in response to CRT. These results support the hypothesis that clinically useful ex-RNA biomarkers may be functionally implicated in disease pathogenesis.

As part of the exRNA consortium, our group hopes to extend these results to a population of patients at risk for LV remodeling (post-MI), to determine whether this approach (biomarker discovery through clinical research; validation through basic research; and reapplication for disease diagnostics and physiology through translational research) is feasible in developing novel biomarkers of human cardiovascular disease.

Extracellular vesicles, such as exosomes and microvesicles, have recently been discovered to contain different types of regulatory RNA such as long non-coding RNA (lncRNA).   The lncRNA are a group of diverse non-coding genes that are being increasingly implicated in biological processes, but they remain poorly characterized and understood.   Their role in cancers such as hepatocellular cancer (HCC) has been increasingly recognized. We have discovered that some lncRNA that contain highly conserved sequences, termed ultraconserved RNA (ucRNA), are transcribed and altered in HCC.   We examined the role of extracellular vesicles in mediating intercellular transfer of these lncRNA and influencing tumor cell behavior.  We found that lncRNA could be isolated in extracellular vesicles secreted from cultured HCC cells and then internalized by other cell types.  By profiling RNA expression in these extracellular vesicles and their donor cells, we identified and then cloned a ucRNA highly expressed and enriched in extracellular vesicles. This novel lncRNA, termed TUC339, was found to modulate tumor cell growth and adhesion. Suppression of TUC339 with siRNA reduced HCC cell proliferation, clonogenic growth and anchorage independent growth. Inversely, enforced expression of TUC339 increased cell proliferation, thus representing a functional mechanism by which intercellular transfer of TUC339 can promote HCC growth and spread. We conclude that these findings support the existence of selective mechanisms for lncRNA export from cells and implicate the transfer of lncRNA via extracellular vesicles as a mechanism by which tumor cells can modulate their local cellular environment.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3807642/

Kogure T, Yan IK, Lin WL, Patel T. (2013) Extracellular Vesicle-Mediated Transfer of a Novel Long Noncoding RNA TUC339: A Mechanism of Intercellular Signaling in Human Hepatocellular Cancer. Genes Cancer  4(7-8), 261-72.