Double-stranded RNA and Toll-like receptor activation: a novel mechanism for blood pressure regulation

Abstract

Toll-like receptors (TLRs), such as TLR4 and 9, recognize pathogen-associated molecular pattern (PAMPs) and danger-associated molecular patterns (DAMPs) and are associated with increased blood pressure (BP). TLR3, residing in the endosomal compartment, is activated by viral double-stranded RNA (dsRNA) leading to activation of TIR receptor domain-containing adaptor inducing IFN-β (TRIF) dependent pathway. Besides foreign pathogens, the immune system responds to endogenous markers of cellular damage such as mitochondrial dsRNA (mtdsRNA). New evidence has shown a link between dsRNA and increased BP. Moreover, TLR3 activation during pregnancy was demonstrated to develop preeclampsia-like symptoms in both rats and mice. Hence, we hypothesize that the dsRNA derived from viral nucleic acids or cellular damage (mtdsRNA) will increase the inflammatory state through activation of TLR3, contributing to vascular dysfunction and increased BP. Therefore, inhibition of TLR3 could be a therapeutic target for the treatment of hypertension with potential improvement in vascular reactivity and consequently, a decrease in BP.

Introduction

The World Health Organization estimates that about 600 million people have arterial hypertension, with an overall increase of 60% of cases by 2025. In addition, hypertension contributes to 7.1 million annual deaths [1]. Despite extensive studies on hypertension and the vast therapeutic arsenal available to treat this disease, in most cases, this health condition is still classified as idiopathic and it is one of the key risk factors for cardiovascular disease (CVD) development. Hypertension is generally attributed to organ–target damage, including the vasculature, the kidney, the heart and the central nervous system [2].

The immune system activation has also been recognized as a hallmark characteristic of hypertension, where both the innate and acquired immune response has been implicated in the pathophysiology of this disease [2,3]. Toll-like receptors (TLRs) are a family of pattern recognition receptors (PRRs), from the innate immune system, which classically mediate and control the activation and progression of adaptive immunity. TLRs can recognize and respond to a unique repertoire of distinct molecules referred to as pathogen-associated molecular patterns (PAMPs) and danger-associated molecular patterns (DAMPs) [4].

For many years, the classical understanding of the immune response relied on its discriminate “self” from “non-self”. However, “The danger model” appears to show that the immune system is more concerned with things that do “damage” to the organism since, for example, CVD, such as hypertension, are associated with inflammation and activation of the immune system even in the absence of unequivocal infection [5]. Several studies have demonstrated the role of innate immune system cells in the pathogenesis of hypertension, with tissue infiltration by monocytes, macrophages, granulocytes, neutrophils and dendritic cells being part of the inflammatory response in different hypertension models [6–10].

Activation of TLR by pathogens or an excessive number of self-molecules induces gene expression of proteins involved in the immune system response. TLRs recognize and initiate inflammatory responses to numerous and varied DAMPs and the expression/activity of many TLR is increased in hypertension [5]. Whereas PAMPs for different TLRs are known, DAMPs are not well characterized [3,5].

The involvement of several types of TLRs in hypertension has been described to several receptors. TLR4 is one of the TLRs where a large body of evidence indicates that this receptor favor blood pressure (BP) elevation. For example, in spontaneously hypertensive rats (SHR), mesenteric resistance arteries displayed increased TLR4 expression and treatment with an anti-TLR4 antibody reduced mean arterial pressure (MAP) and the contractility to noradrenaline as well as TLR4 protein expression and serum levels of interleukin (IL)-6. These data suggest that TLR4 activation contributes to increased BP, low-grade inflammation and plays a role in the impaired vascular reactivity observed in SHR [66]. Further, TLR4 expression is augmented in the cardiovascular system in a model of hypertension induced by NG-nitro-L-arginine methyl ester (L-NAME) [11]. All experimental models of hypertension where augmented levels of TLR4 were described display end-organ damage, which may be directly linked to over-activation of these TLRs [12]. Shreds of evidence also show that hypertension-related damage may be blunted by anti-TLR4 treatment [13,14] or if conducted in TLR4 knockout (KO) mice [15,16].

TLR9 represents another member of the TLRs playing a distinct role in BP regulation. A TLR9 agonist was able to increase arterial BP and to cause endothelial dysfunction in rats with a normotensive background [10]. Further, inhibition of TLR9 with the lysosomotropic agent chloroquine (CQ) was associated with suppression of BP, decreased counts of circulating T cells and vascular infiltrating leukocytes in SHR, when CQ was administered during the pre-hypertensive phase [67]. Interestingly, besides TLR9, TLR3 is also present in the endosome and the action of CQ could also be TLR3-dependent. To more comprehensive information on TLR4 and 9 in hypertension, see the reviews [5,12,17–20].

For other TLRs, evidence to link their input to BP regulation is not so clear. For example, a role for TLR3 has been recognized in preeclampsia (PE), in a context where the immune system from the mother is overactivated [21]. Of importance, TLR9 has also been recognized to contribute to vascular dysfunction observed during maternal hypertension [22]. The role of TLR2 involved in hypertension is limited to inflammation in the renal system, where TLR2-activated nuclear factor-kB (NF-kB) signaling is significantly correlated with renal ischemia/reperfusion injury [23].

TLRs activate downstream adaptor molecules turning on specific signaling cascades, including NF-kB, interferon (IFN) response factors (IRFs) and mitogen-activated protein kinases [24,25], among others. All TLRs depend on the MyD88-dependent pathway with the exception of TLR3, which exclusively uses the receptor domain-containing adaptor inducing IFN-β (TRIF)-dependent pathway to induce expression of proinflammatory cytokines [24] and TLR4 can be activated by MyD88-independent and -dependent pathways [26].

TLR pathways may be triggered by exogenous or endogenous ligands. In this sense, self-proteins and endogenous nucleic acids such as biglycan (TLR2), messenger (m) ribonucleic acid (RNA) and double-stranded (ds) RNA (TLR3), fibrinogen (TLR4), RNA (TLR7), human cardiac myosin (TLR8) and desoxyribonucleic acid (DNA) (TLR9)] have been shown to cause the initiation and/or perpetuation of autoimmune reactions, including overproduction of cytokines, leading to tissue damage [25,27,28], contributing to the development of some pathological conditions, including hypertension [29–31].

TLRs may be expressed in the cell membrane, or alternatively, be intracellularly located, as is the case of TLR3, TLR7, TLR8 and TLR9. Specifically, to the TLRs intracellularly located, it is mandatory that their ligands should cross the cellular membrane or alternatively, they need to be intracellularly produced in order to activate these receptors. Nucleic acids are among the molecules that are recognized by TLRs since they can be produced both intracellularly or across cellular membranes inside endosomal vesicles [25,32].

Initially, TLRs were considered as the sensors of nucleic acids of viral origin. For example, TLR7 and TLR8 recognize single strand (ss) RNA viruses and synthetic oligoribonucleotides (ORNs), while TLR9 detects DNA containing unmethylated 5′C-phosphate-G3’ (CpG)-rich DNA. TLR3 detects dsRNA, which constitutes not only the genome of dsRNA viruses but also the intermediates produced during the replication of ssRNA and DNA virus sequences [33,34]. However, evidence suggests that they could have other functions in the cells being able to recognize ‘self’ RNAs or DNA released from another cell [28,35].

Double-stranded RNA

RNA molecules in cells are primarily found in the form of ssRNA since they are transcribed from the DNA template in this format. The ssRNA, however, often forms secondary structures that encompass segments of ds regions [36]. The term dsRNA refers to high molecular weight RNA in the “A” form, having the following properties: (1) a base composition expected for an RNA duplex composed of two complementary, antiparallel strands stabilized by hydrogen bonds and hydrophobic interactions; (2) a molar absorbance lower than in ssRNA; (3) an absolute hypochromic remarkably larger than in ssRNA; (4) temperature transition profiles of a cooperative type; and (5) a sedimentation rate that differs from ssRNA [37]. dsRNA is a signal for gene-specific silencing of expression in a number of organisms [38] and it is a key activator of the innate immune response against viral infections (dsRNA of more than 30-bp length) [39]. Some of these secondary structures play a critical role in their biological function [36]. For example, dsRNA regions are present in the precursors of microRNAs, small interfering RNA (siRNAs), mRNA, transfer RNA (tRNA), as well as in the genome of RNA viruses that can be released into cells upon infection. The family of proteins responsible for processing dsRNA is called double-stranded RNA-binding proteins (dsRBP). Various dsRNAs serve as cargoes, activators and substrates of dsRBPs in many biological pathways [40].

The history of extracellular dsRNA begins with reports of a strong systemic response induced by viral replication intermediates, which was initially termed a “viral toxin” [34]. In 1956, Rich and Davies announced in the Journal of the American Chemical Society that single strands of RNA can “hybridize”, joining together to form a dsRNA [41]. By the late 1990s, it was discovered that exposing cells to dsRNA results in gene silencing (“co-suppression” in plants, “quelling” in fungi, and “RNA interference” -RNAi- in nematodes) through destruction of mRNAs containing similar sequences [42]. Further, animal models reinforced that viral dsRNA is able to mediate both local and systemic effects and ultimately contribute to viral pathogenesis [34].

Similar to other constituents such as ssRNA, unmethylated CpG DNA, bacterial lipopolysaccharide (LPS), lipoproteins and flagellin, dsRNA can act as a PAMP/DAMP leading to induction of type I IFN. Once dsRNA is present within the extracellular milieu, it can bind various cell surfaces receptors, such as scavenger receptors, Raftlin, and CD14, and then be internalized via clathrin-dependent endocytosis. Once endocytosed, dsRNA can be recognized by TLR3 within the acidic environment of the lysosome leading to subsequent signaling via the adaptor protein TRIF [43–45]. Moreover, cytoplasmic RNA helicase RIG-I (retinoic acid-inducible gene I) and melanoma differentiation-associated gene-5 (MDA5) recognize dsRNA through SID, a transmembrane protein required to transport extracellular dsRNA into the cytoplasm (Figure 1) [45,46]. While dsRNAs longer than 30 bp are candidates to induce innate immune responses through TLR3, RIG-I most strongly recognizes short dsRNA (~10 bp) and MDA5 binds to long dsRNA (1–7 kp) [47–49].

Figure 1. dsRNA–TLR3 signaling cascade.

dsRNA is internalized by cell surface receptors via endocytosis and then recognized by TLR3 present in the endosome leading to subsequent signaling via the adaptor protein TRIF. Further, dsRNA can also be recognized by cytoplasmic RNA helicase RIG-I (retinoic acid-inducible gene I) and melanoma differentiation-associated gene-5 (MDA5) into the cytoplasm through SID transporter. Abbreviations: dsRNA, double-stranded RNA; TLR3, toll-like receptor 3; RIG-I, a retinoic acid-inducible gene I; MDA5, melanoma differentiation-associated gene. Adapted from Nguyen et al. [45] (permission obtained under license number 4757830360815).

Besides being stable within the host cell, dsRNA also shows stability when released into the extracellular milieu. Its origin can be exogenous (through replication by-product of the viruses) or endogenous (being released from activated or damaged cells) [35] and similar to the compounds mentioned above, dsRNA will target TLRs. Although TLRs are important to invading pathogens and dead or necrotic tissue, they have been also implicated in the pathogenesis of autoimmune diseases [50]. Nucleic acids binding to TLRs 7 and 9 have been connected to both human and mouse models of systemic lupus erythematosus [50–52], whereas TLR3 expressed in astroglia and microglia is involved in the development of neuro infection, brain ischemic injury, aberrant brain development and neurodegeneration (either by viral or cellular origin) [35,53]. Altogether, the literature provides evidence that dsRNA may contribute to pathological conditions through the activation of TLRs.

High blood pressure and dsRNA

Studies have shown that endogenous ligands such as Ang II, ADMA, C-reactive protein, fibrinogen, fibronectin, high-mobility group box 1, heat-shock protein 60, hyaluronan, oxidized phospholipids, uric acid, fatty acids, surfactant protein-A, heparin sulphate and serum amyloid A are associated with TLRs and hypertension [see reviews [17,54] for details].

Most of the knowledge relating the contribution of dsRNA to pathological conditions comes from viral infection. Indeed, several viral infections are associated with hypertension or an increase of BP, including human herpesvirus 8, cytomegalovirus and human immunodeficiency virus 1 in primary pulmonary hypertension [55–58]. However, the mechanisms underlying viral infection and their contribution to hypertension are not fully understood. dsRNA may be the link, since TLR3 recognizes dsRNA, which is produced by most viruses at the replication stage.

Participation of dsRNA on arterial hypertension

Essential, primary or idiopathic hypertension is defined as high BP in which secondary causes are not present and remain a major modifiable risk factor for CVD [59].

Treatment with TLR3 agonist and viral mimetic polyinosinic:polycytidylic acid (Poly I:C) impaired endothelium-dependent vasodilation, increased vascular production of reactive oxygen species, and reduced reendothelialization after carotid artery injury in wild-type mice, which was blunted in animals KO to TLR3 [60].

It has been shown that dsRNA targets TLR3 associated with the adaptor protein TRIF, to increase IFN-α, and this pathway was recently described to be at least partially involved in the development of Ang II-induced hypertension and cardiac hypertrophy [61]. The authors showed that the TLR3-mediated TRIF pathway, but not the TLR4-TRIF pathway, is necessary for sustained Ang II-induced hypertension. They believe that whereas TRIF directly associates with the TLR3 C-terminal, its association with TLR4 is mediated by another adaptor protein, the TRIF-related adaptor molecule (TRAM), and therefore, generating different outcomes.

The evidence linking TLR3–TRIF and BP rely on the production of type I IFNs, favoring hypertension, due to activation of vascular injury, a process involving endothelial dysfunction, extracellular matrix degradation, vascular proliferation, inflammatory cell infiltration, among others [62].

Although the evidence is still preliminary, other TLR3/TRIF-alternative pathways may also be involved in dsRNA-induced hypertension. In this regard, it was demonstrated that the involvement of TLR3/MDA5/RIG-I on the development of glomerulonephritis may contribute to the development of systemic hypertension [63,64]. Additionally, recent evidence shows mitochondrial dsRNA (mtdsRNA) as a novel DAMP. mtdsRNA is formed in circumstances under which the mitochondrial RNA processing machinery is compromised, leading to massive accumulation of dsRNA in mitochondria, leading this organelle to act as a source of self-nucleic acids, initiating an IFN response via the cytoplasmic receptor MDA5 [65,66]. Moreover, several studies provide evidence of hypertension-induced mitochondrial structural and functional abnormalities, including alterations of biogenesis and dynamics, renin–angiotensin–aldosterone system (RAAS)-induced mitochondrial damage, reactive oxygen species (ROS) overproduction, apoptosis and mitochondrial DNA mutations [67–72]. Therefore, despite the link between damaged mitochondria and hypertensive heart disease, a cause–effect relationship remains to be established.

Participation of dsRNA on pulmonary hypertension

Pulmonary hypertension is a hemodynamic, pathophysiological disorder defined by elevated MAP ≥25 mmHg, measured by right heart catheterization [73]. Notwithstanding, the effects of TLR3 activation may be controversial in the same pathology. A recent study demonstrated that Poly I:C increases TLR3 expression and reduces established pulmonary hypertension, suggesting a protective role involving the induction of IL-10. On the other hand, TLR3−/− mice failed to develop spontaneous pulmonary hypertension [74].

Participation of dsRNA to induce preeclampsia

Currently, the American College of Obstetrics and Gynecology’s practice guidelines define PE as the presence of hypertension, with or without proteinuria, occurring after 20 weeks of gestation in a previously normotensive patient [75].

Corroborating the idea that dsRNAs may play a role in increasing BP, a study showed that self dsRNA, released from excessive necrosis, led to inflammatory and autoimmune responses and, potentially, these effects could be linked to PE [34].

Maternal immune system modulation via TLR3 activation led to the development of PE like symptoms in both rats and mice. TLR3 is expressed in both human and rodent placentas and the excessive stimulation of TLR3 in rodents, using Poly I:C, led to the development of the characteristic markers of PE, including hypertension, endothelial dysfunction and proteinuria [76–80]. Yet, these classic events related to PE, including endothelial dysfunction, are attenuated in aortas from pregnant mice lacking TLR3 either in the maternal cells (placenta still has TLR3) or in the placental cells (maternal cells still expressed TLR3) and completely abolished when pregnant mice were lacking TLR3 in both maternal systemic and placental cells. In, addition, the complete lack of TLR3 in pregnant mice demonstrated resistance to splenomegaly and increases in pro-inflammatory immune cells induced by TLR3 activation [81].

In PE, women presented increased expression of proteins and mRNA levels of TLR3, TLR7 and TLR8 in the placenta. Additionally, activation of these receptors in human cytotrophoblasts significantly increased their protein levels in the placenta at 6 h of exposure to the agonists (Poly I:C, R-837 and CLO97, respectively). Similarly, treatment of pregnant mice with TLR3, TLR7 or TLR8 agonists (intraperitoneal injections) led to a significant increase in expression and mRNA for TLR3/7/8 in the placenta and an increase in systolic blood pressure at gestational day 17 compared with vehicle-treated controls. Further, several pro-inflammatory chemokines, cytokines, receptor/ligands and transcription factors were increased in mice treated with a TLR3/7/8 agonist during pregnancy [21].

As mentioned above, activation of TLR3 via synthetic dsRNA Poly I:C was able to induce PE like-symptoms in pregnant rodents, while no changes were seen in the blood pressure or proteinuria of virgin animals [76–80]. However, there is no evidence demonstrating whether the chronic treatment with the TLR3 agonist would lead to an increase in blood pressure and consequently, hypertension since the PE model used an acute treatment with Poly I:C.

On another hand, expression of RIG-1, an intracellular receptor shown to be activated by small sequences of dsRNA, was found diminished in the syncytiotrophoblast of placental villi, decidual cells and endothelium of villous capillaries from preeclamptic women [82]. In this particular study, the authors did not analyze which endogenous binder, nor the pathways, that could be leading to RIG-1 reduced activation. Of importance, RIG-1 is one of the important signaling molecules that promote activation of stem cell and tissue regeneration, a process that is impaired in PE.

dsRNA antagonism: a way to control hypertension?

Pharmacological and molecular tools based on dsRNA have been extensively reported in the literature to activate, or alternatively to inhibit TLR3.

The synthetic analog of dsRNA, Poly I:C, has been used to generate experimental models associated with exacerbated inflammation [83,84]. Poly I:C was also recognized as a possible candidate to work as an immunostimulant for vaccines directed against intracellular pathogens and cancer [85]. Furthermore, synthetic dsRNA suffers from rapid degradation. Another important issue to be considered is the length of the dsRNA, which may elicit undesirable immune stimulation and disorders.

Some molecules have been used to inhibit the Poly I:C-evoked actions. For example, Emodin (3-methyl-1, 6, 8-trihydroxyanthraquinone), a compound which can be found in Polygoni multiflori radix, was shown to inhibit the activation of RAW 264.7 mouse macrophages [86]. The development of several small molecules probes, based on the Emodin structure, resulted in the synthesis of several competitive dsRNA molecules targeting TLR3, in a highly specific manner [87]. Among them, the authors highlighted that a modification from the molecule T5626448, named by them as compound 4a, was recognized in their study as the most potent TLR3 antagonist, being able to repress TLR3-induced downstream pathway.

The inhibition of dsRNA/TLR3 complex by these inhibitors was later evaluated in the immune system, and also, in the vascular system. When superior mesenteric arteries were incubated with poly (I:C), for 24 h, the relaxation response evoked by sodium nitroprusside, a nitric oxide doner, was blunted, demonstrating a role for dsRNA/TLR3 to induce vascular dysfunction [88]. Interestingly, this effect was prevented by the TLR3 inhibitor.

The inflammation elicited by dsRNA in response to an acute respiratory viral infection increases chemokine production, infiltration of neutrophils and natural killer cells through CXCR2 and CCR5 activation, respectively, into the lungs. Interesting, CXCR2 and CCR5 inhibitors attenuate different components of the response to dsRNA, which may provide a therapeutic benefit to patients with chronic obstructive pulmonary disease.

Indeed, vascular functional impairment following exposition to dsRNA provides a new insight to study pathological conditions where the vasculature is dysfunctional, including hypertension. The vasculature is in charge to regulate vascular tone and blood flow, being one of the most important regulators for the BP control [89]. Several cellular and extracellular events, including remodeling, stiffening, fibrosis, among others, are related to inflammation-process to some degree [90], and somehow, could systemically provide mtdsRNA.

Therefore, taken all this evidence together, it is possible that the dsRNA derived from viral nucleic acids or cellular damage (mtdsRNA) will increase the inflammatory state through activation of TLR3, contributing to vascular dysfunction, resulting in augmented vascular resistance, and chronically, contributing to increment in the BP level. Therefore, inhibition of TLR3, or its downstream proteins, could represent a therapeutic target for the treatment of arterial hypertension, with potential improvement in vascular reactivity and consequently, a decrease in BP levels (Figure 2).

Figure 2. The schematic hypothesis of the proposed mechanisms of TLR3-inhibition to prevent hypertension.

dsRNA derived from viral nucleic acids or cellular damage trigs TLR3 leading to the production of inflammatory cytokines and IFN, contributing to vascular dysfunction and increased blood pressure. Inhibition of TLR3 could potentially improve vascular reactivity and consequently, decreases blood pressure. Abbreviations: dsRNA, double-stranded RNA; IFN, interferon; mtdsRNA, mitochondrial dsRNA; TLR3, toll-like receptor 3. Key references: [13,61,91].

Conclusion

CVD is the main cause of death on the planet, where hypertension is one of the major symptoms. Despite the treatment of hypertension’s symptoms, the cause of the development of this disease is not fully understood. Additionally, many patients are resistant to the medicines available for treatment, reinforcing that it is important to comprehend well the etiology of hypertension. Although this review brings to the light some new possibilities that could help identify the etiology of hypertension, it also stresses some possibilities that should be further investigated. For example, it should be further evaluated whether TLR activation participates in BP homeostasis in physiological conditions. Additionally, considering that both lifestyle and environmental factors may contribute to the genesis of hypertension through epigenetic-dependent mechanisms, it would be interesting to verify whether TLRs are targets for differential gene activation, resulting in higher BP levels. Yet, considering the robust body of evidence linking TLR activation by DAMPs and PAMPs resulting in blood pressure increments, it will be interesting to learn whether accumulation of these products during the life-time will favor pre-hypertension, or whether longer exposure to elevated levels of DAMPs and PAMPs lead to hypertension. These and other possibilities may provide insight on the development of new therapies, as well as how these and other intriguing research questions will drive our efforts in the future.

Abbreviations

Ang

Angiotensin

CpG

unmethylated 5′C-phosphate-G3’

CVD

cardiovascular disease

CQ

chloroquine

DAMP

danger-associated molecular pattern

DNA

desoxiribonucleic acid

ds

double-stranded

dsRBP

double stranded RNA binding proteins

IFN

interferon

IRF

interferon response factor

IL

interleukin

KO

knockout

L-NAME

NG-nitro-L-arginine methyl ester

LPS

lipopolysaccharide

m

messenger

MAP

mean arterial pressure

MDA5

melanoma differentiation associated gene-5

mtdsRNA

mitochondrial dsRNA

NF-kB

nuclear factor-kB

ORN

oligoribonucleotide

PAMP

pathogen-associated molecular pattern

PE

preeclampsia

Poly I:C

polyinosinic:polycytidylic acid

PRR

pattern recognition receptor

ROS

reactive oxygen species

RAAS

renin–angiotensin–aldosterone system

RIG-I

retinoic acid inducible gene I

RNA

ribonucleic acid

SHR

spontaneously hypertensive rat

siRNA

small interfering RNA

ss

single-strand

TLR

toll-like receptor

TRAM

TRIF-related adaptor molecule

TRIF

receptor domain-containing adaptor inducing IFN-β

tRNA

transfer RNA