Nucleic acid-binding polymers as anti-inflammatory agents: reducing the danger of nuclear attack

Keywords::

• DNA
• innate immunity
• polymers
• RNA
• Toll-like receptors

As highlighted in recent studies on the anti-inflammatory effects of nucleic acid-binding polymers (NABPs) [1], the innate immune system involves a remarkably diverse set of molecules to signal danger and promote protective responses. In this context, danger refers to threats to the organism from tissue injury, whether induced by infection, physical or chemical trauma or conditions like ischemia. Key to these responses is rapid, widespread immune activation, including recruitment of cells to the site of injury. The molecules signaling danger display molecular patterns and can be exogenous products from infectious organisms or endogenous molecules from damaged cells. The exogenous stimuli are called pathogen-associated molecular patterns (PAMPs) while the endogenous stimuli are called damage-associated molecular patterns (DAMPs). Among DAMPs, both DNA and RNA can create important, albeit unexpected, signals in innate immunity. Indeed, nucleic acids may act as universal DAMPs since they are present in every cell [2,3].

Like PAMPs, DAMPs can stimulate receptors called pattern-recognition receptors. Pattern-recognition receptors include Toll-like receptors (TLRs) as well as non-TLR sensors [4]. Since PAMPs originate from foreign organisms, they can present truly unique structural patterns for receptor interactions. In contrast to PAMPs, DAMPs are normal constituents of the cell and are definitely not ‘foreign’. An important question therefore concerns the mechanism by which a normal cellular component can generate a new molecular pattern and transform into a potent immune trigger.

In contrast to preformed PAMPs, DAMPs require structural modifications to become a recognizable pattern. These modifications, which can occur in the setting of cell death, include degradation, denaturation, post-translational modification and redox reactions. While these events can alter the structure of a DAMP, the most important change relates to its location. Thus, as a cell dies, intracellular molecules can move from the inside of a cell to the outside and, in the extracellular space, display a new immunological identity. Since the function of DAMPs is to activate (or alarm) the immune system, the term alarmin can also be used since these molecules have the hallmarks of a powerful and noisy alarm signal [5].

DAMPs are highly diverse and encompass every biochemical class, ranging from small molecules, such as ATP and uric acid, to large molecules, such as proteins. Perhaps the most surprising DAMPs are nucleic acids, both DNA and RNA. Although dsRNA molecules such as poly(I:C) were long known as stimulators of interferon production, DNA appeared to be very different. Indeed, many studies showed that DNA (especially mammalian dsDNA in the B conformation) was essentially devoid of antigenic and immunogenic activity outside the setting of systemic lupus erythematosus. Systemic lupus erythematosus is a prototypic autoimmune disease characterized by the production of antibodies to DNA that form immune complexes and thus drive pathogenesis. These studies suggested that immune recognition of DNA was the feature of only a very disturbed immune system and that DNA was unique among macromolecules in its immunological inertness [6].

During the last two decades, the perspective on the immune activity of DNA has changed drastically, reflecting important studies on the mitogenic activity of bacterial DNA and the discovery of immunostimulatory DNA sequence motifs (CpG motifs); these motifs occur much more frequently in bacterial than mammalian DNA and can denote foreignness. Together, these studies showed that CpG DNA, whether a natural bacterial DNA or a synthetic oligonucleotide, can stimulate immune cells via the TLR9 receptor, which resides in an endosomal compartment. The internal location of this receptor has suggested a role in the recognition of DNA from intracellular infections, especially with viruses [7].

The immune machinery for recognition of DNA resembles that for RNA since ssRNA can trigger TLR7 while dsRNA can trigger TLR3. TLR3, 7 and 9 are all located on the inside of cells. In this regard, the intracellular location of nucleic acid sensors may protect the system from inadvertent stimulation by circulating nucleic acids, the probable residue of cell turnover or perpetually occurring cell death. In addition to their effects on TLRs, DNA and RNA can trigger other internal nucleic acid sensors, including DAI and members of the RIG family [7,8].

Since the receptors triggered by nucleic acids are intracellular, they should be shielded from any DNA or RNA in the blood. How then do nucleic acids released from dead and dying cells serve as DAMPs? The answer to this question came originally from studies in lupus indicating that the immune properties of DNA are, in fact, mutable and that even mammalian DNA, which is inactive by itself, can be transformed into an active moiety when bound to proteins. In the case of lupus, the transformative proteins were autoantibodies, which can form immune complexes to promote DNA uptake into cells and provide access to internal TLR and non-TLR sensors [9,10]. Further studies indicated that DNA can bind to other proteins, such as LL37, a cathelicidin, to allow cell entry and activation of macrophages and dendritic cells, key sentinels in innate immunity [11]. DNA can also bind to the nonhistone nuclear protein HMGB1 to form complexes with immunostimulatory properties. Interestingly, by itself, neither HMGB1 nor mammalian DNA has immunological activity. It is the complex that is stimulatory, perhaps by allowing DNA on the outside of the cell to get back inside [12,13].

Together, these observations suggest that, in the context of cell death, DNA, RNA and nuclear proteins exit dying cells, rise to high levels in the blood and form immunostimulatory complexes. In these complexes, otherwise inert molecules acquire activity as an ensemble; these considerations do not exclude the possibility that complex assembly occurs inside the cell because of nuclear–cytoplasmic mixing that can occur as the nuclear membrane breaks down during cell death. Since nuclear molecule release appears to be common with cell injury and death, DAMPs comprised of nucleic acids may act in a host of clinical conditions [14,15]. Evidence for this role comes from studies in animal models on the effects of TLR oligonucleotide antagonists as well as specific TLR knockouts in conditions that include lupus, acetaminophen toxicity, malaria and multiple sclerosis [16,17].

Since DAMPs can be pathogenic, strategies to reduce their expression and downstream effects on inflammation could be therapeutic in immune-mediated disease. While specific TLR antagonists have shown efficacy in disease models in mice, these agents could have adverse effects by blocking the response to nucleic acids arising during the course of intracellular infection. By contrast, agents that reduce the concentration of extracellular nucleic acids could represent a more attractive approach since it would leave the intracellular signaling pathways intact while reducing the drive from the extracellular DAMPs. Such agents would prevent the outside-to-inside translocation that is critical to activation by extracellular nucleic acids.

Among agents that could block DAMP activity, NABPs may serve as unique scavengers for extracellular nucleic acid because of their capacity for electrostatic interactions. NABPs originated in the field of nonviral gene delivery and were developed to condense nucleic acids into nanocomplexes or nanoparticles for the delivery of plasmids or siRNA into cells [18,19]. Extensive studies of the polymers have generated a wealth of data on the structure–property relationships of NABPs complexing with nucleic acids; these data are directly relevant to the task of capturing extracellular nucleic acids in the blood.

NABPs that can scavenge nucleic acids in vitro and in vivo include structures such as β-cyclodextrin-containing polycation, polyaminidoamine dendrimer, polyphosphoramidate and hexadimethrine bromide. These polycations demonstrate sufficient specificity and avidity to bind extracellular DNA or RNA and block their activity as indicated by in vitro and in vivo studies [1,20]. Importantly, in a murine model, simultaneous intravenous delivery of NABPs could, in a dose-dependent manner, reduce death induced by the administration of a CpG oligonucleotide or poly(I:C) to mice treated with galactosamine to sensitize them to the effects of TNF. This proof-of-principle study suggests an exciting application of NABPs as anti-inflammatory agents [1].

The inhibitory activity of the NABPs is notable since the resulting complexes with nucleic acids are inactive, whereas the complexes of DNA and RNA with other proteins (e.g., LL37 and HMGB1) can be immune stimulators. These results suggest that the stimulation by complexes may depend critically on the charge or 3D structure resulting from the interaction of DNA with proteins or polymers. Alternatively, the intracellular trafficking of the protein–DNA or polymer–DNA complexes may differ. Thus, even if taken up into cells, polymer–DNA complexes may fail to access internal nucleic acid sensors or undergo modification in the endolysosomal compartment to render the DNA immunologically inactive.

DAMP blockade by NABPs is a promising new approach to the therapy of immune-mediated diseases since NABPs can keep ‘dangerous’ DNA and RNA on the outside of cells and thereby prevent stimulation by internal nucleic acid receptors and the downstream effects of exuberant cytokine production. While much remains to be done to understand how NABPs bind extracellular nucleic acids and interdict their signaling, the studies performed thus far suggest a novel approach to block inflammation by using molecular scavengers to reduce the danger from the immune system’s nuclear attacks.

Financial & competing interests disclosure

This work was supported by a VA Merit Review grant, an Alliance for Lupus Research Grant, NIH grant AI093960 and a Wallace H Coulter Grant.The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.