In 1979, Alexander Rich, PhD, and his colleagues solved the structure of a self-complementary DNA hexamer and discovered that the molecule formed a left-handed double helix (Z-DNA). (Credit: Journal of Biological Chemistry).

Following a surge of enthusiasm at their discovery in the early 1980s and a subsequent crash of unrealistic expectations, the publication of three research articles in Nature on July 20, 2022, provide converging evidence that nucleic acids in uncustomary left-handed, Z configurations are key movers in the regulation of immune responses. The studies uncover mechanistic insights on how excessive Z-DNA/Z-RNA in the cell triggers cell death, controlling outcomes of host-pathogen interactions, autoimmune disease and immune responses to cancer.

“These papers represent an interesting turn-around for the field where the biological relevance of the left-handed Z-DNA and Z-RNA has long been questioned,” said Alan Herbert, PhD, founder, and CEO of InsideOutBio, a Boston-based immune-oncology company that is developing Z-DNA-based drugs to trigger immune responsiveness in resistant cancers. (Herbert was not involved in these Nature articles).

A cell must be constantly alert to internal and external threats. The presence of fragments of free-floating nucleic acids from the cell’s own genome, embedded quiescent viral sequences, or invading viruses are bona fide red alerts.

“Answers to the mystery of how a cell knows what is normal and what isn’t have come from left-field, literally,” said Herbert.

Z nucleic acids sensors

The staircase in the image on the left represents a left-handed helix and that on the right represents a right-handed helix (Credit: Nikolay/ YouTube)

If you imagine yourself climbing up a helix of double stranded DNA or RNA in a cell as if it were a spiral staircase, chances are you’ll find the handrail on your right. However, when humidity is low and salts are high, right-handed helices can flip into alternative left-handed, nuclease resistant, Z-DNA/Z-RNA configurations with a zig-zagging backbone in live cells. The biological function of Z forms of nucleic acids has been poorly understood, until now.

DNA and RNA in the Z configuration are detected by Zα domains found in two mammalian proteins: ZBP1 (Z-DNA binding protein 1), and ADAR1 (adenosine deaminase acting on RNA 1). When large amounts of Z-DNA/RNA are detected by these proteins, they trigger various cell death pathways—apoptosis (internally programmed), necroptosis (RIPK3-dependent, externally triggered), and pyroptosis (caspase-dependent, immune triggered).

Earlier studies showed, the RNA-editing enzyme ADAR1 keeps the accumulation of immune-triggering double-stranded RNA (dsRNA) in check by converting adenosine to inosine, to prevent the activation of innate immune RNA sensors, such as MDA5, OAS–RNAse L and PKR.

A naturally occurring point mutation in the Z-DNA binding domain of ADAR1 when paired with a loss-of-function mutation in the second allele of ADAR1 causes Aicardi–Goutières Syndrome (AGS)–a severe autoinflammatory disease marked by neuronal death in the striata of both hemispheres of the brain (bilateral striatal necrosis). In mice, while a complete loss of ADAR1 activity is embryonically lethal, AGS-like mutations cause autoinflammation.

ZBP1 recognizes not only foreign but also self-DNA/RNA from damaged or dying cells. When activated, ZBP1 can trigger cell death and transcription through the kinases RIPK1 and RIPK3, the protease caspase 8 and the type I interferon(IFN) family of cytokines.

“Excess Z-RNA can arise during viral infection. The sensing of this Z-RNA by ZBP1 provokes an inflammatory response that targets elimination of the virus by killing the cell that the virus has invaded,” Herbert explains. “However, this process is also capable of detecting Z-RNA that is produced by RNAs encoded in the host genome. So, the actions of ZBP1 must be checked so that a cell does not attack itself. ADAR1’s role is to recognizes Z-RNA and stop it from activating ZBP1.” Together the three papers unravel the mechanisms by which ADAR1 suppresses the lethal consequences of Z-RNA activated ZBP1 cell death.

Z nucleic acids and interferonopathies

Z-DNA/Z-RNA derived from pathogens and host cells can induce the production of type I IFNs that in turn stimulate an array of genes that play an essential role in inflammation, immunoregulation, tumor cells recognition, and T-cell responses. Interactions of ADAR1 and Z-RNA prevent IFN activation, but the underlying mechanism was unclear until now.

Manolis Pasparakis, PhD, professor at the University of Cologne.

One of the three papers published in Nature on Jul 20, 2022, (ADAR1 averts fatal typeI interferon induction by ZBP1) shows that when mice have a mutated allele of an IFN-inducible isoform of ADAR1, ZBP1 upregulates type I IFN activation, resulting in death soon after birth. Mutations in both ZBP1 and ADAR1 reduce IFN activation and the expression of IFN-stimulated genes, and prevent death in infancy, in mice. The study was led by Manolis Pasparakis, PhD, a professor at the University of Cologne.

Pasparakis’ team showed, the mutant ADAR1 results in abnormal editing and consequential accumulation of complementary Z-RNA reads derived from non-coding, mobile genetic elements (retroelements) in the mouse genome. These retroelements could potentially be a source of Z-RNAs that activate ZBP1.

The novelty of the mechanism by which ZBP1 upregulates IFN activation and severe pathology in the ADAR1 mutant mice is indicated by the fact that the process does not depend on any of the known downstream effectors of ZBP1 that trigger cell death–RIPK1, RIPK3, MLKL or caspase-8.

“Our findings identify suppression of endogenous Z-RNA formation by ADAR1 as a key mechanism preventing aberrant activation of pathogenic IFN responses by ZBP1,” Pasparakis and his colleagues noted.

Partners in pathology

Andrew Oberst, PhD, associate professor of immunology at the University of Washington.

An article in the same issue of Nature (ADAR1 mutations causes ZBP1 dependent immunopathology), corroborates the findings from the Pasparakis team’s and indicates that autoinflammatory pathology resulting from mutations in ADAR1’s Z-DNA binding domain depends on ZBP1 signaling. The study was led by Andrew Oberst, PhD, associate professor of immunology at the University of Washington.

When ADAR1 detects double-stranded left-handed nucleic acids, it edits these by changing their adenosine bases to inosine. This suppresses innate immune activation and resultant pathologies. Mutations in ADAR1’s Z-DNA binding domain, prevents it from binding Z-configurations, removing this suppression, and triggering severe autoinflammation.

Oberst and his team found that autoinflammatory pathologies caused by mutations in ADAR1’s Z-DNA binding domain can be fully rescued by eliminating ZBP1. However, they also found that eliminating ZBP1 does not fully reverse the downstream inflammatory program induced by the ADAR1 mutation.

Activation of ZBP1 can trigger several downstream effectors: RIPK1, RIPK3 or caspase8 and MLKL. The authors removed each of these downstream players to identify the mechanistic underpinnings of the functional interaction between ZBP1 and ADAR1.

They found loss of individual necroptotic signaling molecules, RIPK3 and MLKL, partially replicated the protective effects of removing ZBP1, but removing combinations of RIPK3 and caspase-8, caspase-8 and MLKL, or caspase-8 alone, exacerbated the ADAR1 mutant inflammatory pathology. The authors noted that the pathological effects of ADAR1 mutations could be mediated by apoptosis, necroptosis and inflammation, depending on the cell or tissue type and the levels of the components of the individual pathways.

“A surprising finding from our paper is that the pathogenic effects of ZBP1 in this setting seem to be independent of both apoptosis and necroptosis. Our findings suggest a possible role for RIPK1-dependent cytokine production and resulting inflammation downstream of ZBP1,” Oberst said, “Unfortunately, this likely means that blocking apoptosis or necroptosis using available drugs (for example, inhibitors of RIPK1, RIPK3 or the caspases), even in combination, would not ameliorate the pathology caused by ADAR1 mutation.”

Preventing spontaneous autoinflammation

Jonathan Maelfait, PhD, team leader at the VIB-UGent Center for Inflammation Research in Belgium.

Concurrent cellular and mouse model studies from the laboratory of Jonathan Maelfait, PhD, team-leader at the VIB-UGent Center for Inflammation Research in Belgium, that was also published in the same issue of Nature (ADAR1 prevents autoinflammation by suppressing spontaneous ZBP1 activation) adds to the mechanistic understanding of the ADAR1 and ZBP1 interplay by demonstrating that ADAR1 inhibits spontaneous activation of ZBP1.

Maelfait’s team demonstrates that the Zα domain of ADAR1 restricts ZBP1 activation and prevents ZBP1-induced apoptosis and nectroptosis. Their findings corroborate that ZBP1 plays an integral role in the mammalian double-stranded RNA response and in pathologies linked to Aicardi–Goutières syndrome.

Adding to earlier studies that have shown ZBP1 acts as a sensor for double-stranded viral RNA, this study shows ZBP1 acts as a sensor for double-stranded RNA formed by pairing inverted DNA repeat elements that are transcribed from the mouse genome. In ADAR1-deficient cells, the authors corroborate ZBP1 triggers apoptosis via caspase-8 and necroptosis via MLKL and in ADAR1 knockout mice, the authors show ZBP1 activation contributes to embryonic lethality. They also show ZBP1 mediates intestinal cell death and skin inflammation in the ADAR1 mutant mouse model.

In conclusion

“The findings are generally compatible,” said Oberst, comparing the findings of the three Nature papers. “It’s interesting that there are some differences in the pathology induced by the different ADAR1 mutation models used. These differences led to slightly different outcomes for the studies: for example, we observe a partial rescue of our ADAR1 mutant mice with RIPK3 ablation, while the Pasparakis group does not see this with their (different, and more severe) ADAR1 mutation model.”

“This suggests a really fine balance between ADAR1 and ZBP1, which can be shifted significantly by small changes in either protein or altered by small changes to the genetic background of the mice (as we show),” Oberst continues. “It’s probably also influenced by the microbiome and other aspects of mouse maintenance. Understanding and reconciling these differences, and figuring out how they apply to human disease, will be a topic for further study.”

Herbert, whose team at InsideOutBio is exploiting the role of Z-DNA in immune-regulation to therapeutically activate inflammatory cell-death in resistant cancer cells (as reported on GEN News), said, “These papers confirm if the interaction between ADAR1 and ZBP1 is unbalanced, inflammatory responses can arise in tissues in the absence of viruses, causing Mendelian diseases like Aicardi Goutières Syndrome with features similar to autoimmune disease like systemic lupus erythematosus. The strength of the three papers is that each comes to the same conclusion through work done in completely separate labs using mouse models that are independently derived.”

“Intriguingly, all the papers show how junk RNA that forms Z-RNA can be used to tag host RNAs in a way that allows ADAR1 to inhibit their activation of ZBP1 under normal conditions,” continues Herbert. “The mechanism shows how Nature can turn lemons into lemonade.”