Plant and Animal microRNAs (miRNAs) and Their Potential for Inter-kingdom Communication

Abstract

microRNAs (miRNAs) comprise a class of ~18–25 nucleotide (nt) single-stranded non-coding RNAs (sncRNAs) that are the smallest known carriers of gene-encoded, post-transcriptional regulatory information in both plants and animals. There are many fundamental similarities between plant and animal miRNAs—the miRNAs of both kingdoms play essential roles in development, aging and disease, and the shaping of the transcriptome of many cell types. Both plant and animal miRNAs appear to predominantly exert their genetic and transcriptomic influences by regulating gene expression at the level of messenger RNA (mRNA) stability and/or translational inhibition. Certain miRNA species, such as miRNA-155, miRNA-168, and members of the miRNA-854 family may be expressed in both plants and animals, suggesting a common origin and functional selection of specific miRNAs over vast periods of evolution (for example, Arabidopsis thaliana-Homo sapiens divergence ~1.5 billion years). Although there is emerging evidence for cross-kingdom miRNA communication—that plant-enriched miRNAs may enter the diet and play physiological and/or pathophysiological roles in human health and disease—some research reports repudiate this possibility. This research paper highlights some recent, controversial, and remarkable findings in plant- and animal-based miRNA signaling research with emphasis on the intriguing possibility that dietary miRNAs and/or sncRNAs may have potential to contribute to both intra- and inter-kingdom signaling, and in doing so modulate molecular-genetic mechanisms associated with human health and disease.

Electronic supplementary material

The online version of this article (doi:10.1007/s10571-017-0547-4) contains supplementary material, which is available to authorized users.

Introduction: microRNAs (miRNAs) in Plants and Animals

First characterized in the plant and animal kingdoms about 16 years ago, microRNAs (miRNAs) as single-stranded non-coding RNAs (sncRNAs): (i) have established themselves as essential post-transcriptional regulators of messenger RNA (mRNA) speciation, abundance, and complexity; (ii) have been shown to play significant roles in health, development, aging and disease; and (iii) have been implicated in basic mechanisms involving multiple human pathologies, and prominently, in cancer, cardiovascular disease and age-related neurodegeneration (Lagos-Quintana et al. 2001; Lau et al. 2001; Lee and Ambros 2001; Reinhart et al. 2002; Millar and Waterhouse 2005; Lukiw 2007; Guo et al. 2010; Coruh et al. 2014; Zhao et al. 2015; Guo et al. 2016; Previdi et al. 2017; Wong et al. 2016; Alural et al. 2017; Ding et al. 2017; Djami-Tchatchou et al. 2017; Liang et al. 2017; Van Roosbroeck et al. 2017). All miRNAs derive from long double-stranded RNA (dsRNA) precursors generated by RNA Pol II or RNA Pol III that in eukaryotes are processed into smaller mature miRNAs, ~18–25 nt long, by nuclear RNaseIII or RNaseIII-like processing enzymes (Bartel 2009; Carthew and Sontheimer 2009). What is remarkable is that from bioinformatics considerations alone, miRNAs are probably the most highly selected and information-dense of all small RNA species (Lukiw 2012; Zhao et al. 2015; Hill and Lukiw 2016). For example, the integrated use of chip-based miRNA arrays and microfluidics, RNA sequencing technologies, applied bioinformatics, and miRNA analytical algorithms indicate that a 22 nucleotide (nt) sncRNA, the typical size of a plant or animal miRNA, consisting of just 4 different ribonucleotides—(adenine, A; cytosine, C; guanine, G; or uridine, U)—have the potential to generate just a few biologically useful miRNAs from well over 10 trillion possible sncRNA sequence combinations. The observation that, for example, there are only a very much smaller number—only about 2650 miRNAs of different sequence in the entire human organism, and only about 35–40 abundant miRNAs in the human central nervous system (CNS)—suggests an unusually high selection and evolutionary pressure to utilize only highly specific miRNA ribonucleotide sequences to function in biologically useful miRNA–mRNA interactions (Hill and Lukiw 2014, 2016; Fabris and Calin, 2016). This miRNA-mRNA recognition, based on base-pair complementarity, ultimately regulates transcriptional output and shapes patterns of global gene expression and the composition of the transcriptome in development, aging and disease (Lau et al. 2001; Bartel 2009; Hill et al. 2014; Hill and Lukiw 2014; Pogue et al. 2014; Zhao et al. 2015; Hill and Lukiw 2016; Previdi et al. 2017; Wong et al. 2016; Ding et al. 2017; Alural et al. 2017). Put another way, it is remarkable that only about 1 in 10 billion potential miRNA sequences have established useful functions in miRNA-mRNA-based gene regulation in all of human biology, and less than 1 in ~200 billion possible miRNA sequences have found a useful purpose in the miRNA-mRNA-based gene regulation of global gene expression patterns in the human CNS. Interestingly there are recent reports that miRNA or even anti-miRNA sequences may take the form of miRNA circles which because of their lack of free 3′ or 5′ termini are resistant to endonuclease, and hence extremely stable in structure and function even in highly acidic environments such as those encountered in the human gastrointestinal (GI) tract (Lukiw 2013; Zhao et al. 2016).

The following sections briefly highlight some recent and remarkable discoveries in the field of plant- and animal-based miRNA and genetic signaling research with emphasis on the molecular genetics of inter- and intra-kingdom signaling pathways, and how these RNA signals may have potential for influencing pathogenic mechanisms associated with both cancers and diseases involving inflammatory neurodegeneration.

Cross-kingdom communication: miRNA and human disease The first convincing evidence for miRNA involvement in human disease came from studies on their roles in the development of cancer (chronic lymphocytic leukemia; CLL; Calin et al. 2002; Mirzaei et al. 2017) and in age-related neurodegenerative disease (Alzheimer’s disease; AD; Lukiw 2007; Millan 2017; Recabarren and Alarcón 2017). Of emerging interest is the recently recognized endogenous and exogenous signaling potential for miRNAs in plants and our diet, and the intriguing possibility of ‘horizontal’ cross-kingdom communication in which orally ingested miRNAs may influence gene expression complexity in the host (Chen et al. 2012a, b; Vaucheret and Chupeau 2012; Zhang et al. 2016; Liang et al. 2012, 2013; Lukasik and Zielenkiewicz 2014; Budak and Akpinar 2015; Liang et al. 2015; Budak and Zhang 2017; Luo et al. 2017). To date, about 872 miRNAs, belonging to 42 families of plants, have been identified in ~71 individual plant species by genomic analysis, screening and RNA sequencing, and at least 325 miRNAs have been fully characterized in the angiosperm Arabidopsis thaliana (Zhang et al. 2006; Axtell et al. 2011; Budak and Zhang 2017; Luo et al. 2017; mirBASE release 21.0; http://www.mirbase.org/cgi-bin/mirna_summary.pl?org=ath; accessed 12 July 2017). Cloning, RNA sequencing, dietary miRNA transfer experiments and analysis, and resistance to the strong iodine-containing oxidizing agent periodate indicate that about 5% of all detectable miRNAs in the human serum are indeed plant miRNAs containing a 2′-O-methyl modified 3′ end that makes them resistant to oxidative degeneration (Hill and Lukiw 2016; Luo et al. 2017). In general, it appears that after plant and other dietary 2′-O-methyl-protected miRNAs pass through the GI tract, blood–brain barrier (BBB) and/or other physiological barriers, and are transported via blood serum and the circulatory system and delivered to specific physiological sub-compartments and tissue/organ systems, these miRNAs have real potential to modulate host gene expression (Vaucheret and Chupeau 2012; Zhang et al. 2016; Lukasik and Zielenkiewicz 2014; Liang et al. 2015; Luo et al. 2017). Although these ideas still remain controversial, that miRNAs in our diet derived from plants and animals suggests the interesting possibility that, perhaps remarkably, dietary foods are not only essential suppliers of nutrients but also carriers of highly specific post-transcriptional regulatory information (Kang et al. Kang et al. 2017; Moran et al. 2017). There are currently at least three highly illustrative examples for trans-species and trans-kingdom miRNA communication and these include miRNA-155, miRNA-168, and miRNA-854. Both published and unpublished work from our lab and other researchers have provided evidence that these miRNAs are: (i) active in, and common to, both plants and animals; and (ii) have each been shown to transit intra- and inter-kingdom biophysiological barriers with potential to modulate or regulate essential biological and potentially pathological processes across each major life group (Table 1).

Table 1.

Summary of similarities and differences between plant and animal small non-coding RNAs (sncRNAs): focus on microRNAs (miRNAs); for specific references see text

Reference	Plants	Animals
First report in the literature	2002 (Arabidopsis thaliana); (viroids; 1971)	2001 (Caenorhabditis elegans)
First report in disease	Arabidopsis infection by turnip mosaic virus alters miRNAs (2003)	Lymphocytic leukemia (2002); Alzheimer’s disease (2007)
Size (number of nucleotides)	18–25 nt	18–25 nt
miRNA biosynthesis from DNA	Generally RNA Pol II	Generally RNA Pol II; RNA Pol III
Location of biosynthesis/action	Nucleus	Nucleus/cytoplasm
Ancillary processing enzymes	Dicer-like (DCL)	Drosha, Dicer
Precursor processing	RNA Pol II, RNA Pol III; Dicer-like 1 (DCL1)	RNA Pol II, RNA Pol III; Dicer and Drosha
Stability; 3′ modification	Generally stable; 2′-O-methyl modified	Generally unstable (half-life of 1-3 h); free 2′ and 3′ hydroxyl
Periodate sensitivity	Resistant	Sensitive
Mechanism of target recognition	Ribonucleotide complementarity	Ribonucleotide complementarity
Mechanism of repression	mRNA cleavage	Translational repression
Location within the genome	Predominantly intergenic regions; sometimes clustered	Intergenic regions and introns; sometimes clustered
# miRNA genes (miRNAs) present	<200 (Arabidopsis thaliana)	2650 (Homo sapiens)
Location of miRNA binding motifs in mRNA	Predominantly in the open reading frame	3′ untranslated region (3′UTR) of multiple mRNAs
Number of miRNA binding sites within target mRNAs	Generally single	Multiple
miRNA-mRNA complementarity	Generally a perfect complementarity	Imperfect; seed sequences and variable flanking complementarity
miRNA biosynthesis from DNA	Generally RNA Pol II	Generally RNA Pol II; RNA Pol III
Location of biosynthesis/action	Nucleus	Nucleus/cytoplasm
Ancillary processing enzymes	Dicer-like (DCL)	Drosha, Dicer
Mechanism of repression	mRNA cleavage	mRNA targeting and translational repression
Function of known target mRNAs	Gene regulation during development, aging, disease	Gene regulation during development, aging, disease
Can be pathogenic	Yes	Yes

miRNA-155

An inducible, NF-kB-regulated miRNA-155a has been recently shown to specifically target the expression of Bcl2, c-fos, STAT3, and other transcription-regulating DNA-binding proteins that directly modify B cell activation and B-cell transmembrane receptor signaling in CLL (Mirzaei et al. 2017); similarly an up-regulated miRNA-155 in sporadic AD has been shown to down-regulate complement factor H (CFH), stimulate inflammatory signaling, and elicit an aberrant innate-immune response (Lukiw, 2007; Li et al. 2012; Millan 2017). Interestingly, miRNA-155 is associated with multiple cancer, autoimmune, and inflammatory pathways, and anti-miRNA-155 therapeutic strategies have recently shown efficacy in the clinical and investigational treatment of experimental autoimmune myocarditis (Yan et al. 2016), in models of acute myelogeneous leukemia (Liang et al. 2017), in various cancers of the lung (Van Roosbroeck et al. 2017) and in insect-borne transmissible infections such as malaria (Barker et al. 2017). Animal fat-sourced miRNA-155 has recently been shown to play a pivotal role in the development, priming, and activation of the immune system, and the activity of regulatory T cells, including the induction of pivotal epigenetic and immunoregulatory modifications which may be conducive to diseases involving progressive inflammatory degeneration associated with aging (Banerjee et al. 2012; Wagner et al. 2015; Barker et al. 2017; Liang et al. 2017; Yan et al. 2016). Widespread in the plant kingdom the resinous material produced by bees from various plants known as propolis exhibits numerous biological-regulatory properties, highlighting its immunomodulatory actions in part based on the regulation of human-specific miR-155 activities (Conti et al. 2016). Plants may therefore exert inter-kingdom effects not only directly via miRNA transfer but also by plant products directly targeting specific host miRNAs. It is remarkable that a homologous miRNA-155 species in both plants and animals and the intra- and inter-kingdom communication via this innate-immune miRNA may provide a novel network of links between dietary fat and inflammatory pathologies beyond the more well understood high fat-cholesterol (HF-C) and vascular disease connection (Alexandrov et al. 2005; Bagyinszky et al. 2017, Prasad 2017). Hence, specific plant-derived extracts obtained through dietary sources appear to have multiple effects on host miRNA-155 neurobiology, an inducible NF-kB-regulated miRNA known to be in high abundance in the human brain and retina that also appear to be directly involved in brain and retinal diseases that include blood–brain barrier dysfunction, AD, Down’s syndrome (DS) and age-related macular degeneration (AMD; Banerjee et al. 2012; Li et al. 2012; Hill et al. 2015; Barker et al. 2017).

miRNA-168a

As probably the most extensively studied of plant sncRNAs, miRNA-168 may represent the ‘archetypical plant miRNA’; at ~850 copies per plant cell it represents one of the most abundant miRNAs in plants and has a recognized capacity for trans-kingdom communication (Hu et al. 2015; Luo et al. 2017). For example, after feeding fresh maize to pigs for 7 days, at least 18 zea maize-derived 2′-O-methylated 3′ end-modified miRNAs (including zma-miRNA-168a-5p) could be detected in porcine serum, brain, heart, and other tissues; indeed both in vivo and in vitro experiments demonstrated that dietary maize miRNAs could cross GI tract barriers to enter the porcine bloodstream (Luo et al. 2017). In porcine cells, it was found that plant miRNAs are very likely to specifically target their endogenous porcine mRNA targets and influence gene expression in a fashion highly similar to that of mammalian miRNAs (Zhang et al. 2011; Hu et al. 2015; Luo et al. 2017). A fairly recent in silico study further indicates that a small family of plant miRNAs—including miRNA-168a—is also present within a population of mammalian breast milk exosomes, however, the significance is not well understood (Lukasik and Zielenkiewicz 2014; Melnik et al. 2014). Plant-derived miRNA-168 has (i) been further shown to regulate the expression of AGO1 mRNA that encodes a central component of the plant RNA silencing complex (RISC) that binds to the vast majority of plant miRNA undergoing processing (Vaucheret and Chupeau 2012; Luo et al. 2017); and (ii) displays a high degree of complementarity with exon 4 of the mammalian low density lipoprotein receptor adapter protein 1 (LDLRAP1) mRNA. LDLRAP1 mRNA encodes a liver-enriched clathrin-associated sorting protein (CLASP) that facilitates the removal of low density lipoprotein (LDL) from the human circulatory system, and appears to provide a dietary-sourced cardiovascular health benefit (Vaucheret and Chupeau 2012; Luo et al. 2017). Research work from our laboratories indicate that miRNA-168a is indeed abundant in several dietary plants and certain murine species who consume these plants as part of their diet also show detectable miRNA-168a levels in both their systemic circulation and CNS (unpublished observations 2017).

miRNA-854a

Certain miRNA primary sequences contain ‘ribonucleotide fingerprints’ conserved across multiple species and these RNA sequence fingerprints represent some of the most highly preserved nucleic acid sequences observed over the course of evolution (Hill et al. 2014; Pogue et al. 2014; Hill and Lukiw 2016). To cite just one important example, using deep RNA sequencing and genome-wide computational approaches to characterize miRNAs based on both nucleic acid sequence and structure alignments, the miRNA-854 family members have been shown to be abundantly expressed in Arabidopsis thaliana, Caenorhabditis elegans, Mus musculus, and Homo sapiens (Arteaga-Vázquez et al. 2006; Hill and Lukiw 2016; unpublished observations 2017). In these diverse species, the 21-nt G-rich miRNA-854 (miRNA-854a; 5′-GAUGAGGAUAGGGAGGAGGAG-3′) plays analogous functional roles in the common targeting of the oligouridylate binding protein 1b (UBP1b) mRNA 3′-untranslated region (3′–UTR) that encodes a member of a heterogeneous nuclear RNA (hnRNA) binding protein family. Hence a common origin for miRNA-854 as a trans-kingdom regulator of basal ribonucleic-specific uridine-mediated transcriptional mechanisms in both plants and animals has existed across vast periods of evolution (A. thaliana–H. sapiens divergence about 1.5 billion years; https://www.newscientist.com/article/dn17453-timeline-the-evolution-of-life/; https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1689654/; Hill et al. 2014; Pogue et al. 2014; Hill and Lukiw 2016). Moreover, the secondary structures of sncRNA precursors also appear to be significantly conserved across multiple miRNA species, and internal stems, loops, and mis-paired RNA ‘bulges’ are evident at equivalent positions in a significant number of pre-miRNA sequences (Arteaga-Vázquez et al. 2006; Axtell et al. 2012; Hill et al. 2014; Hill and Lukiw 2016). Interestingly, miRNA-854 has been found to be increased in abundance in AD, and one of its major targets, UBP1b mRNA, is decreased in sporadic AD brains. This may be related to the widely reported global transcriptional deficits observed in AD tissues when compared to age-and gender-matched controls (Lukiw 2012; Barbash and Sakmar 2017).

Controversy of Plants and Animals Sharing miRNAs

A wealth of recent findings and observations suggest highly complex and interactive pathways governing miRNA biogenesis and the miRNA-mediated regulation of gene expression in both plants and animals. However, the idea of the horizontal transmission of miRNA species between plants and animals, as in any novel and emerging investigative field remains a somewhat controversial research area. For example (i) recent investigations on plant miRNAs provides evidence that plant miRNAs may have evolved independently from other miRNAs (Igaz et al. 2012; Budak and Akpinar 2015; Budak and Zhang 2017); and (ii) other current studies that surveyed the presence and abundances of cross-species miRNAs (xenomiRs) using comprehensive computational analytical approaches indicated that xenomiRs probably originate from technical artifacts rather than dietary intake (Kang et al. 2017). The commonalities and differences of miRNA signaling pathways in various eukaryotes, their possible evolutionary origin, their functional roles in complex organisms and their proposed link to complexity and multicellularity in plants and animals have been recently reviewed in some detail (Zhao et al. 2015; Moran et al. 2017; Perge et al. 2017). Ongoing studies of miRNA activity in plants and animals will progressively expand our understanding of sncRNA biology and multicomponent and interactive miRNA-related signaling pathways, and are sure to open up new frontiers and research perspectives.

Concluding Remarks

Taken together, these current studies have broadened our view of cross-kingdom communication, and underscore the idea that the incidence and importance of genetic material and/or epigenetic information being transferred from one species to another via miRNAs and/or sncRNAs remains both controversial and an intriguing biological consideration. It is remarkable (i) that certain sncRNAs (as miRNAs) of plants and animals have persisted in their size, biogenesis, form, and function throughout many hundreds of millions of years of evolution as discrete information-carrying entities; (ii) that miRNAs have retained critical and essential regulatory roles in the modulation of gene expression, and may play significant pathophysiological roles in both plants and animals in health and disease; and (iii) that perhaps most importantly, our dietary intake may turn out to be not only the essential nutrient supplier for our bodies but may also be a carrier of important genetic regulatory information (Jiang et al. 2012; Lukiw 2012; Pirrò et al. 2016; Zhang et al. 2016; Perge et al. 2017). In fact it has been recently proposed that miRNAs may in part be the actual bioactive components of health-promoting medicinal plants (Xie et al. 2016).

It is our speculation that the reasons for the highly similar mode of action amongst plant and animal miRNAs is that these two similar 18–25 nucleotide (nt) systems probably arose from common ancient RNA components and evolved separately after the evolutionary divergence of plants and animals, that and their mechanistic similarities are an intriguing example of convergent evolution. While still controversial, these findings together support the ideas: (i) that our dietary habits may affect our physiological condition at a genetic level; (ii) that nutrigenomics may well be involved in the pathobiology of most prevalent age- and lifestyle-related human conditions such as cancer, cardiovascular disease and inflammatory neurodegeneration of the CNS; (iii) which raises the possibility that diet-derived miRNAs and sncRNAs may provide us with another means to deliver necessary therapeutics or nutrients to our bodies; (iv) that either animal- or plant-derived products and/or miRNAs may have an impact on human health and disease; and (v) that miRNAs may hold the key as potential targets for genetic manipulation to design stress tolerance in the nutrient intake of both animals and plants. Indeed, it may be such that plants may have evolved to utilize specific sncRNAs as critical post-transcriptional regulators of gene expression in an environmentally-linked RNA sequence-specific manner and these would have a rapidly evolving capacity to respond to the numerous stressors that they may face during their growth cycles—these include extreme temperatures, nutrient deprivation, competition with other species, drought, salinity, and heavy metal and other exposure to toxic species. Lastly, it should be mentioned that multiple species of plants contain a family of ~18–22 nt disease-causing sncRNAs known as viroids, and the recent findings that miRNA analogues and/or related sncRNAs may be used effectively to control miRNA- and/or viroid-based infections in plants (Daròs et al. 2006; Ding and Wang 2009; Ehrenreich and Purugganan 2008; Kasschau et al. 2003; Agius et al. 2012; Carbonell and Daròs 2017; Chellappan et al. 2005; Djami-Tchatchou et al. 2017; Shriram et al. 2016). Transmission of plant viroids to animals via dietary intake certainly occurs, but its relevance, if any, is not well understood. Importantly, because of the rather unique 2′-O-methyl modified 3′ end of plant miRNAs and their remarkable stability in the human GI tract, the influences of these and other sncRNAs on the GI tract microbiome is yet another research area worthy of further investigation for their potential roles in inter-kingdom communication that may contribute to human health and/or disease.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author Contributions

YZ, LC, and WJL discussed the genomic data and scientific implications of these ideas; WJL researched and wrote this paper; the authors are sincerely grateful to colleagues and collaborators for helpful discussions and for sharing unpublished data.