**2. DNAzymes**

Among the gene-silencing technologies, Breaker and Joyce, in 1994, used an *in vitro* selection method to identify a special Dz from a random pool of single-stranded DNA to catalyze Pb2+ dependent cleavage of an RNA phosphodiester linkage [1]. Afterward, a number of Dzs were created with the capacity to catalyze many reactions, including the cleavage of DNA or RNA, the modification and ligation of DNA, and the metalation of porphyrin rings. However, because of the low efficiency of RNA cleavage, they are not widely used for biological applications except for 10-23 Dz [2]. The inherent catalytic RNA-cleaving property of Dzs has been used with different mRNA targets as *in vitro* diagnostic and analytical tools, as well as *in vivo* therapeutic agents.

### **2.1. The possible mechanisms and characteristics of DNAzymes**

Dzs of the 10-23 subtype are single-stranded DNA catalysts that comprise a central cationdependent catalytic core of around 15 deoxyribonucleotides [ggctagctacaacga], and two complementary binding arms of 6–12 nucleotides that are specific for each site along the target RNA transcript [3]. As diagrammed in Figure 1, the enzyme binds the substrate through Watson-Crick base pairing and cleaves a particular phosphodiester linkage located between an unpaired purine and paired pyrimidine in the RNA. This results in the formation of 5' and 3' products, which contain a 2', 3'-cyclic phosphate and 5'-hydroxyl terminus, respectively. Even though the 10-23 Dz can cleave any RY junction, the reactivity of each substrate dinu‐ cleotide compared in the same background sequence with the appropriately matched DNA‐ zyme is found to follow the scheme AU = GU > GC >> AC. Murray *et al*. found that when the target site core is an RC dinucleotide, the relatively poor activity could be enhanced up to 200 fold by substituting deoxyguanine with deoxyinosine, which could effectively reduce the strength of Watson-Crick pairing between bases flanking the cleavage site [4].

Due to the simple cleavage-site requirement, Dzs are capable of cleaving any particular mRNAs for multiple turnover by appropriately designing the sequence in the binding arms. Several features make Dzs attractive from a drug developmental viewpoint. For example, these are inexpensive to synthesize, and their small size allows specificity. Moreover, DNAzymes can be rendered more stable by structural modifications, such as phosphorothioate (PS) linkages, locked nucleic acids (LNAs), and 3'-3' inverted nucleotide end of the DNAzyme [5]. Enhanced biostability, low toxicity, affinity, and versatility suggest great promise for diag‐ nostic and therapeutic applications [6]. Limitations thus far in the development of DNAzymes as novel therapeutics have been delivery and biodistribution, which revolve around poor cellular uptake and stability. Delivery systems depend on the route of administration and the target site. Moreover, an ideal delivery system would facilitate rapid and efficient distribution to the site of action, stability, low toxicity, and efficacy.

**Figure 1.** Secondary structure of the 10-23 DNAzyme–substrate complexes. The 10-23 DNAzyme consists of two varia‐ ble binding arms, designated arm I and arm II, which flank a conserved 15 base unpaired motif that forms the catalytic core. The only requirement of the RNA substrate is for a core sequence containing an RY junction.

#### **2.2. DNAzymes delivery systems – Past to present**

current movements in these technologies, focusing mainly on Dzs and siRNAs, because these

Among the gene-silencing technologies, Breaker and Joyce, in 1994, used an *in vitro* selection method to identify a special Dz from a random pool of single-stranded DNA to catalyze Pb2+ dependent cleavage of an RNA phosphodiester linkage [1]. Afterward, a number of Dzs were created with the capacity to catalyze many reactions, including the cleavage of DNA or RNA, the modification and ligation of DNA, and the metalation of porphyrin rings. However, because of the low efficiency of RNA cleavage, they are not widely used for biological applications except for 10-23 Dz [2]. The inherent catalytic RNA-cleaving property of Dzs has been used with different mRNA targets as *in vitro* diagnostic and analytical tools, as well as

Dzs of the 10-23 subtype are single-stranded DNA catalysts that comprise a central cationdependent catalytic core of around 15 deoxyribonucleotides [ggctagctacaacga], and two complementary binding arms of 6–12 nucleotides that are specific for each site along the target RNA transcript [3]. As diagrammed in Figure 1, the enzyme binds the substrate through Watson-Crick base pairing and cleaves a particular phosphodiester linkage located between an unpaired purine and paired pyrimidine in the RNA. This results in the formation of 5' and 3' products, which contain a 2', 3'-cyclic phosphate and 5'-hydroxyl terminus, respectively. Even though the 10-23 Dz can cleave any RY junction, the reactivity of each substrate dinu‐ cleotide compared in the same background sequence with the appropriately matched DNA‐ zyme is found to follow the scheme AU = GU > GC >> AC. Murray *et al*. found that when the target site core is an RC dinucleotide, the relatively poor activity could be enhanced up to 200 fold by substituting deoxyguanine with deoxyinosine, which could effectively reduce the

Due to the simple cleavage-site requirement, Dzs are capable of cleaving any particular mRNAs for multiple turnover by appropriately designing the sequence in the binding arms. Several features make Dzs attractive from a drug developmental viewpoint. For example, these are inexpensive to synthesize, and their small size allows specificity. Moreover, DNAzymes can be rendered more stable by structural modifications, such as phosphorothioate (PS) linkages, locked nucleic acids (LNAs), and 3'-3' inverted nucleotide end of the DNAzyme [5]. Enhanced biostability, low toxicity, affinity, and versatility suggest great promise for diag‐ nostic and therapeutic applications [6]. Limitations thus far in the development of DNAzymes as novel therapeutics have been delivery and biodistribution, which revolve around poor cellular uptake and stability. Delivery systems depend on the route of administration and the target site. Moreover, an ideal delivery system would facilitate rapid and efficient distribution

are poised to play an integral role in antigene therapies in the future.

**2.1. The possible mechanisms and characteristics of DNAzymes**

strength of Watson-Crick pairing between bases flanking the cleavage site [4].

to the site of action, stability, low toxicity, and efficacy.

**2. DNAzymes**

138 Nucleic Acids - From Basic Aspects to Laboratory Tools

*in vivo* therapeutic agents.

As in all nucleic-acid-based reagents, efficient drug delivery systems (DDSs) to deliver the Dzs to targeting site are highly needed. Furthermore, by adopting DDSs, it could be helpful to solve the obstacles about DNAzymes' stability, biological effects, and toxicity. Several seminal studies have demonstrated that certain DNAzyme delivery systems can efficiently encapsulate DNAzymes and transfect them into cells without clear toxicity. The attempt first involved the microspheres of co-polymers poly (lactic acid) and poly (glycolic acid) (PLGA), which encapsulated the Dzs. PLGA microspheres are able to achieve biphasic release and sustained accumulation of the Dzs [7]. In a second delivery system, a chimeric aptamer–DNAzyme conjugate was generated for the first time using a nucleolin aptamer (NCL-APT) and survivin Dz (Sur\_Dz). This conjugate could be used as a specific gene-targeting therapy to kill the targeted cancer cells [8]. A third delivery system is developed and studied based on the cationic liposomal formulation technology. Li *et al.* reported the effect of a c-Jun targeted DNAzyme (Dz13) in a rabbit model of vein graft stenosis after autologous transplantation in a cationic liposomal formulation containing 1,2-dioleoyl-3-trimethylammonium propane (DOTAP)/1,2 dioleoyl- snglycero-3-phosphoethanolamine (DOPE). Dz13/DOTAP/DOPE allows sufficient uptake by the veins and reduces SMC (smooth muscle cell) proliferation and c-Jun protein expression *in vitro*. Meanwhile, a Phase I clinical trial has indicated that it is safe and well tolerated after local administration in skin cancer patients [9]. Finally, due to their low toxicity and no side effects, nanoparticulate systems have spread rapidly and could significantly enhance the efficacy of tumoricidal Dzs. Marquardt *et al.* found that c-Jun targeted Dz13 delivered in this manner is capable of enhanced skin penetration efficiency and cellular uptake with a high reduced degradation of Dz13 in *vitro* [10]. These results indicate that, with more suitable delivery approaches, the biological effects of Dzs would be further increased, and the Dzs could be applied to new subject areas.

#### **2.3. Application of DNAzymes** *in vivo* **and** *in vitro*

Increasing evidence indicates the efficacy and potency of DNAzymes *in vivo* and *in vitro* in a range of disease settings, allowing characterization of key pathogenic pathways and their potential use as therapeutic agents (Table 1). DNAzymes have been widely applied as a new interference strategy in the treatment of many conditions, including cancer, viral diseases, and vein graft stenosis. For instance, Dz13 targeting the transcription factor c-Jun has shown promise in experimental models of mice infected with H5N1 virus via reducing H5N1 influenza virus replication and decreasing expression of pro-inflammatory cytokines [11]. Furthermore, Dz13/DOTAP/DOPE reduces SMC proliferation and c-Jun protein expression *in vitro*, and inhibits neintima formation after end-to-side transplantation, which may potentially be useful to reduce graft failure [9]. Likewise, Cai *et al.* demonstrated that safe and welltolerated Dz13 could inhibit tumor growth and reduce lung nodule formation in a model of metastasis [12].


**Table 1.** *In vivo* and *in vitro* applications of 10-23 DNAzymes

As is well known, treatment resistance is one of the leading causes of tumor recurrence. We have recently evaluated Dz1 targeting latent membrane protein 1 (LMP1) in the setting of nasopharyngeal carcinoma model and demonstrated that injected intratumorally DZ1 with fuGENE 6 in nude mice inoculating LMP1-positive cells resulted in a significant inhibition of tumor growth and an enhanced radiosensitivity. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) showed that DZ1 reduces the angiogenesis and microvascular permeability [13]. Other studies have used DNAzymes to target the other key genes in cancer therapy. DNAzyme targeting the Bcl-XL gene significantly sensitized a panel of cancer cells to apoptosis and further to reverse the chemoresistant phenotype [23]. Due to a secondary mutation at T790M in the epidermal growth factor receptor (EGFR), most of nonsmall-cell lung cancer (NSCLC) patients will eventually develop resistance to tyrosine kinase inhibitors (TKIs) treatment. Allele-specific silencing of EGFR T790M expression and downstream signaling by DNAzyme DzT could suppress the growth of xenograft tumors derived from H1975TM/LR cells, indicating that DzT is capable of overcoming EGFR T790M mutant-based TKI resistance [25]. In a similar way, Kim *et al.* developed the DNAzyme that specifically targets the site of the point mutation (T315I), conferring imatinib resistance in BCR–ABL mRNA. Cleavage of T315Imutant ABL mRNA by DNAzyme could significantly induce apoptosis and inhibit prolifera‐ tion in imatinib-resistant BCR-ABL-positive cells [24].

#### **2.4. DNAzymes in clinical trials**

potential use as therapeutic agents (Table 1). DNAzymes have been widely applied as a new interference strategy in the treatment of many conditions, including cancer, viral diseases, and vein graft stenosis. For instance, Dz13 targeting the transcription factor c-Jun has shown promise in experimental models of mice infected with H5N1 virus via reducing H5N1 influenza virus replication and decreasing expression of pro-inflammatory cytokines [11]. Furthermore, Dz13/DOTAP/DOPE reduces SMC proliferation and c-Jun protein expression *in vitro*, and inhibits neintima formation after end-to-side transplantation, which may potentially be useful to reduce graft failure [9]. Likewise, Cai *et al.* demonstrated that safe and welltolerated Dz13 could inhibit tumor growth and reduce lung nodule formation in a model of

**Target Summary Description on Biological Effects**

·Inhibiting proliferation and metastasis

·Promoting apoptosis ·Enhancing radiosensitivity

Egr-1 ·Inhibiting proliferation and metastasis

MMP-9 ·Inhibiting invasion and metastasis

β-integrin ·Inhibiting invasion and metastasis

·Inhibiting proliferation

·Suppressing tumor growth

BCR-ABL T315I ·Overcoming imatinib resistance based on BCR-ABL T315I

TXNIP ·Attenuating oxidative stress, renal fibrosis, and collagen

IGF-II ·Inhibiting proliferation

survivin ·Inhibiting proliferation

VEGFR-1 ·Blocking angiogenesis

Bcl-XL ·Promoting apoptosis

**Table 1.** *In vivo* and *in vitro* applications of 10-23 DNAzymes

**(In Vitro and In Vivo)**

·Suppressing tumor growth [16]

·Suppressing tumor growth [17, 18]

·inducing caspase-dependent apoptosis [19]

·Promoting apoptosis [8]

·Blocking angiogenesis [20]

·Suppressing tumor growth [21]

·Enhancing Taxol chemosensitivity [23]

Mutation [24]

deposition [26]

DNMT1 ·Inhibiting proliferation [22]

·Restraining virus replication and host inflammation

EGFR T790M ·Overcoming EGFR T790M mutant-based TKI resistance [25]

**Refs.**

[13-15]

[9, 11, 12]

metastasis [12].

LMP1

140 Nucleic Acids - From Basic Aspects to Laboratory Tools

c-Jun

The favorable properties of 10-23 Dzs, such as their enhanced biological stability, negligible side effects, and lack of immunogenicity, have paved the way for Dzs to enter clinical trials [17]. Up to now, Dzs to three targets have been undergoing clinical trials and at least one of them has proved its therapeutic efficacy in Phase II trials (Table 2). These results further show the potential of Dzs therapeutic approach for the treatment of diseases and represent a major advance in this field.

As we have found that LMP1-targeted Dz1 could effectively inhibit the growth and enhance the radiosensitivity of NPC cells both *in vivo* and *in vitro*, we investigated the antitumor and radiosensitizing effects of Dz1 in NPC patients for the first time [27]. Being safe and well tolerated, a randomized and double-blind clinical study was conducted in 40 NPC patients, who received Dz1 or saline intratumorally in conjunction with radiation therapy. In a 3-month follow-up, compared with the saline control group, the mean tumor regression and undetect‐ able EBV-DNA copy number in the DZ1 group is significantly higher. Molecular imaging analysis found that Dz1 was tested to accelerate the decline of Ktrans, generally recognized as a marker of tumor blood flow and permeability [28].

The nuclear transcription factor c-Jun is preferentially expressed in a range of cancers. Dz13 cleaves at the G1311U junction in human c-jun mRNA and exerts its antitumor activity via induction of apoptosis, inhibition of angiogenesis, and the induction of adaptive immunity [11]. A phase I first-in-human trial is conducted to determine the safety and tolerability of Dz13 in nine patients with basal-cell carcinoma (BCC), who received a single intratumoral injected


**Table 2.** Clinical trials of DNAzymes in anti-diseases therapy

dose of Dz13 (10, 30, or 100 ìg) [29]. Followed-up over four weeks, c-Jun expression is reduced in all nine participants. Meanwhile, Dz13 could significantly promote apoptosis and stimulate inflammatory and adaptive immune responses in the tumors. Among the participants, five patients have a reduction in histological tumor depth. These results indicated that Dz13 possibly could represent a future treatment option for BCC prior to excision by surgery.

The transcription factor GATA-3 plays an important role in the regulation of Th2-mediated immune mechanisms such as in allergic bronchial asthma, and the DNAzyme hgd40 has been shown to specifically and selectively reduce expression of GATA-3 mRNA. Turowska *et al.* found that hgd40 is evenly distributed in inflamed asthmatic mouse lungs within minutes after single dose application, and could slowly eliminate from lung tissue with the goal to minimize accumulation and to ensure continued exposure for efficacy [32]. Safety pharmacology studies showed that with no observable adverse event, hgd40 has a highly favorable toxicity profile when administered by aerosol inhalation at the therapeutic doses [33]. With good safety and tolerability in the phase I program [31], a randomized, double-blind, placebo-controlled, multicenter clinical trial of hgd40 was conducted in patients with allergic asthma, who had biphasic early and late asthmatic responses after laboratory-based allergen provocation [30]. After each study drug administered by inhalation once daily for 28 days, hgd40 significantly attenuates both late and early asthmatic responses and improves lung function. Moreover, the Th2-regulated inflammatory responses are also attenuated.

These studies, taken together, further demonstrate the potential use of DNAzymes as genetargeting drugs. As Dzs are safe and well tolerated in humans, there is a good chance that we may witness the Dzs reaching the clinic in the near future.
