Resurrecting DNAzymes as Sequence-Specific Therapeutics

See allHide authors and affiliations

Science Translational Medicine  20 Jun 2012:
Vol. 4, Issue 139, pp. 139fs20
DOI: 10.1126/scitranslmed.3004080


In a mouse model of skin cancer, intratumoral injection of a sequence-specific mRNA-cleaving DNA enzyme caused potent inhibition of tumor growth and unusually benign pharmacodynamic profiles.

In the early 1980s, RNAs were shown to have catalytic functions that cleave the phosphodiester backbones of RNAs (1, 2), and the concept of using nucleic acid–based sequence-specific agents to regulate gene expression by cleaving and functionally destroying mRNAs was born. The excitement in the field centered around the ability of these small catalytic RNAs to cleave mRNAs in a site- or sequence-specific manner mediated by the base pairing of RNA catalysts with their target RNAs. The smallest RNA molecules with site-specific cleavage capabilities are the hammerhead ribozymes, which are derived from the self-cleaving domain of plant RNA viroids and virusoids. This self-cleaving motif was adapted for application in trans, wherein the ribozyme could be engineered to base pair with mRNAs and catalyze strand cleavage of the target, followed by recycling of the ribozyme, making these agents true catalytic enzymes (3). For more than a decade, hammerhead ribozymes were the favored RNA-cleaving platform for targeted therapeutic applications that alter biological processes—and disease states—by interfering with the synthesis of selected proteins. In this issue of Science Translational Medicine, Cai et al. (4) demonstrate that catalyticDNA—a DNAzyme (Fig. 1)—inhibits synthesis of the cancer-related transcription factor c-Jun and tumor growth in a mouse model of skin cancer.

Fig. 1.

All in the delivery? Shown is the mechanism of DNAzyme-mediated cleavage of c-jun mRNA. The cationic lipid–catalytic DNA complex was introduced directly into basal or squamous cell tumors in mice. A 10-23 version of the Santoro and Joyce DNAzyme is depicted interacting with the c-jun mRNA via Watson-Crick base pairing. A purine (R = G or A) at the beginning of the DNAzyme catalytic core pairs with a pyrimidine (Y = C or U) in the target mRNA, and cleavage takes place between the pyrimidine and purine in the target sequence. The cleavage results in a 2′-3′ cyclic phosphate and 5′ OH on the target. Once cleaved, the mRNA is degraded by cellular ribonucleases. Inhibition of c-jun expression triggers a cascade of events leading to apoptosis of the cancer cell.


On the basis of the chemistry behind ribozyme-based cleavage, scientists believed that DNA could not carry out such a catalytic function because it lacks the 2ʹ hydroxyl group present in RNA. Then, in the mid-1990s, Gerald Joyce and colleagues used an in vitro selection methodology that employed short single-stranded DNAs to identify DNA motifs that could effect site-specific RNA strand scission (5). These scientists used an ingenious selection scheme and 10 rounds of selection to enrich for RNA-cleaving DNA motifs that displayed high catalytic turnover of mRNA. Two catalytic motifs emerged from this selection scheme: One cleaves RNA between nucleotides G and A, and the other, which is more generally useful, cleaves between any pyrimidine-purine base-pair combination.

This pioneering research established DNA as a true catalytic enzyme, and, in fact, the kinetic parameters of the DNAzymes on RNA targets were superior to those of their RNA counterparts. Clear advantages of the DNAzyme versus other RNA-targeting agents are its ease of chemical synthesis, broad target recognition properties, and high catalytic turnover. Since the initial description of DNAzymes in 1997, numerous applications of these catalytic molecules have been tested, in particular as anticancer therapeutics (6), but no clinical applications have emerged.

Despite the potential broad-based utility of DNAzymes, they have not caught on as therapeutic agents. This deficit results, in part, from the fact that, like all other nucleic acid drugs, delivery has been a challenge for DNAzyme therapeutic applications (7). Through careful choices of a selective therapeutic target—mRNA that encodes the cell proliferation–related transcription factor c-Jun—and a disease setting in which the drug can be applied directly—injection into tumors in mouse models of two skin cancers—Cai et al. (4) have resurrected the potential clinical utility of DNAzymes.

In the new work, intratumoral injection of a c-jun mRNA–targeted DNAzyme that had been complexed with a DOTAP/DOPE {N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium methyl-sulfate}–based lipid carrier had profound inhibitory effects on skin cancer cell proliferation and metastases in murine models of both basal and squamous cell carcinomas. The authors also showed that injection of the c-jun–targeting DNAzyme in immunocompetent animals triggered additional antitumor immune responses, resulting in even more profound tumor inhibition. Doses as high as 100 micrograms per mouse were well tolerated, and at this dose, cessation of tumor growth was observed even when DNAzyme treatment was suspended.

In providing proof of principle for nucleic acid therapeutics, it is necessary to demonstrate not only efficacy but also specificity. Indeed, Cai et al. demonstrated by both RNA and protein analyses that the DNAzyme reduced concentrations of c-jun mRNA and, consequently, its corresponding protein in Western (immune) blot and immunohistochemical assays. A mutant version of the ribozyme with an intact catalytic core and scrambled binding arms had no tumor inhibitory effects nor did it affect the concentrations of c-Jun protein or encoding mRNA. A point mutation in the catalytic core that abolished cleavage activity but still allowed the DNAzyme to base pair with its complementary mRNA did not result in c-Jun down-regulation or inhibition of tumor progression. Taken together, these essential controls validated the efficacy and specificity of the DNAzyme.

Inhibition of c-Jun synthesis had multiple downstream effects. These included reduction in the expression of proliferating cell nuclear antigen, a DNA polymerase accessory factor; cyclin-dependent kinase 4, which is required for cell cycle progression; fibroblast growth factor–2; and tissue remodeling and metastasis-related proteins matrix metalloproteinase–2 (MMP-2), MMP-9, and vascular endothelial growth factor–A, as well as elevation in expression of the tumor suppressor p53 and p53-inducible proteins p21 (CIP1/WAF1) and apoptosis-related caspases 3, 8, and 9. The mutated DNAzyme with scrambled binding arms and the DOTAP/DOPE vehicle alone had no effects on the production of any of these proteins.

In the world of nucleic acid therapeutics, efficacy is often tempered by unpredicted and undesirable toxicities and side effects. Cai et al. gave careful attention to many possible nonspecific effects that might account for off-target modulation of amounts of c-Jun mRNA and protein (5). The authors first demonstrated that the DNAzyme was not activating the DNA CpG receptor Toll-like receptor 9 (TLR9), which functions in activation of innate immunity. Cai et al. then went on to perform large-scale systemic pharmacodistribution studies in cynomologous monkeys, minipigs, and rodents and observed relatively rapid clearance of the DOTAP/DOPE–complexed DNAzymes from blood and little or no accumulation in any of the major organs of these animals. Moreover, the authors did not detect inhibition by the anti–c-jun DNAzyme in 70 separate assays that assessed a variety of biological reactions associated with important pharmacological indicators, such as blood cell counts, clotting factor levels, and other hematological enzymatic activities. Thus, these investigators have provided a strong set of analyses that validate the safety of their lead DNAzyme for human clinical trials.

With respect to nucleic acid therapeutics, the inert nature of the c-jun DNAzyme with respect to toxicities is an obvious advantage over other nucleic acid–based drugs, which have been shown to trigger thrombocytopenia, complement activation (phosphorothioate antisense DNA oligos), and TLR activation with subsequent type I interferon responses (small interfering RNAs) (810). One possible explanation for the benign pharmacodynamics is that the c-jun DNAzyme has no base or backbone modifications and lacks CpG stimulatory motifs. The use of intratumoral injection in the context of a lipid-based carrier that itself is benign may be the keys to resurrecting DNAzymes as safe and effective nucleic acid therapeutic agents. It will be of great interest to follow the next steps in the development of the c-jun targeted DNAzyme for the treatment of skin cancers in human clinical trials.

References and Notes

  1. Funding: NIH grants AI29329 and AI42552. Competing interests: The author is the chair of the scientific advisory board of Dicerna Pharmaceuticals, an RNAi company targeting cancer, and a consultant for Calando Pharmaceuticals, an RNAi delivery company.
View Abstract

Stay Connected to Science Translational Medicine

Navigate This Article