A GATA3 Targeting Nucleic Acid Nanocapsule for In Vivo Gene Regulation in Asthma

Abstract

Allergic asthma is one of the leading chronic lung diseases of both children and adults worldwide, resulting in significant morbidity and mortality in affected individuals. Many patients have severe asthma, which is refractory to treatment, illustrating the need for the development of new therapeutics for this disease. Herein, we describe the use of a peptide cross-linked nucleic acid nanocapsule (NAN) for the delivery of a GATA3-specific DNAzyme to immune cells, with demonstration of modulated transcriptional activity and behavior of those cells. The NAN, built from peptide cross-linked surfactants, is chemically designed to degrade under inflammation conditions releasing individual DNAzyme-surfactant conjugates in response to proteolytic enzymes. Using the NAN, GATA3 DNAzymes were delivered efficiently to human peripheral blood mononuclear cells, with clear evidence of uptake by CD4+ helper T cells without the need for harsh transfection agents. Knockdown of GATA3 was achieved in vitro using human Jurkat T cells, which express GATA3 under homeostatic conditions. Additionally, mice treated with DNAzyme-NANs during house dust mite (HDM)-induced asthma developed less severe allergic lung inflammation than HDM-only control mice, as measured by pulmonary eosinophilia. This study suggests that peptide cross-linked GATA3 DNAzyme-NANs may have the potential to decrease the severity of asthma symptoms in human patients, and development of this technology for human use warrants further investigation.

Graphical Abstract

Asthma is a chronic inflammatory condition affecting the airways, which is estimated to affect as many as 300 million individuals worldwide. The disease features repeated bouts of rapid onset airway obstruction, bronchial hyperreactivity, and mucous hypersecretion, often as a result of exposure to specific external stimuli.^1,2 Asthma cases are often categorized based on the inciting stimuli and can be classified as either atopic (allergic-type) or non-atopic.³ Patients with non-atopic asthma traditionally experience attacks upon exposure to stimuli including cold air, stress, or tobacco smoke, but do not show evidence of immune conversion against the stimuli.¹ In contrast, atopic patients often respond to recognized aeroallergens, including antigens from common house dust mites (HDMs), animal dander, and pollen, and produce serologically evident immune conversion against the inciting cause, characterized by measurable antistimulus IgE. This asthma subtype is notable due to the sensitization period prior to onset of asthma, during which the immune system is primed for a highly inflammatory response upon re-exposure to the stimulus.²

Atopic asthma is a disease driven largely by inappropriate host responses to these external stimuli and is often thought of as a classical example of a Th2-type helper T cell-mediated disorder, with hallmarks of a type I hypersensitivity.⁴ In patients with atopic asthma, primary exposure during sensitization induces an immune response characterized by the formation of memory B and T cells specific to the inciting antigen. Upon secondary exposure to the inciting stimulus, often by inhalation of the antigen in the case of asthma, the host experiences antigen-specific inflammatory responses characterized by memory cell reactivation and antibody binding to the antigen. IL-4 and IL-5 expressed by Th2 cells are found at high levels in bronchoalveolar lavage fluid from atopic patients and contribute to lymphocyte activation and IgE class switch recombination as well as eosinophil activation and mucous hypersecretion, respectively. Recruited eosinophils respond against the allergen in an IgE-assisted manner, often at the luminal surface of the lungs, leading to the release of high concentrations of inflammatory molecules and driving the symptomology and pathology of the disease.² Expression of Th2-derived cytokines including IL-4 and IL-5 is dependent on the activity of the Th2 master transcriptional regulator GATA3.⁵ Specific inhibition of GATA3 during the sensitization phase of OVA-induced asthma has been shown to modify and ameliorate disease course in research animals,^6,7 and the utilization of DNAzymes for this purpose has been reviewed by Garn and Renz.⁸ However, many therapeutic nucleic acids, including DNAzymes, experience stability issues and often require transfection reagents for entry into cells, which can come with associated costs to the treated host.⁹ Additionally, many therapeutic nucleic acids are often heavily chemically modified to increase stability in cellular environments.^10,11 However, this may result in decreased activity, as many chemical modifications alter the folding and recognition properties of nucleic acids.^12,13

Nucleic acid nanocapsules (NANs) are surfactant-based cross-linked nanocapsules that can undergo degradation in the presence of specific enzymes, depending on the cross-linker utilized in the NAN core. Recent studies have shown NANs to be a promising delivery vehicle for therapeutic nucleic acid ligands, including DNAzymes¹⁴ and siRNA.¹⁵ Previously, our group has shown that a DNAzyme-functionalized NAN could achieve up to 60% knockdown of GATA3 mRNA in MCF-7 cells without the use of toxic transfection agents.¹⁴ These initial studies on the therapeutic potential of the DNAzyme-NAN were carried out with NANs assembled with ester cross-linkages that could degrade in response to esterases present in cellular endosomes. We have also shown that replacing the ester cross-linker of the NAN with an enzyme-specific peptide substrate results in increased specificity of NAN degradation by proteolytic enzymes in vitro.¹⁶ Herein, we sought to combine the efficacy of DNAzyme-functionalized NANs with the increased degradation specificity that the peptide substrate cross-linker offers to achieve disease-specific gene regulation in vivo (see Figure S1 for structures of cross-linkers). For these studies, we chose a peptide substrate specific for the matrix metalloproteinase-9 (MMP9) enzyme, as MMP9 has been shown to be upregulated in the airways of asthma patients.¹⁷ We hypothesized that the increase in MMP9 enzyme during airway inflammation should lead to the rapid disassembly of the NAN in the lungs, resulting in a disease-state-specific knockdown of GATA3 by DNAzyme-surfactant conjugates. Herein, we show the successful application of our laboratory’s recently developed peptide cross-linked NAN platform to deliver a chemically unmodified GATA3-specific DNAzyme^18,19 for the treatment of asthma in a HDM mouse model, without the use nucleic acid transfection agents or common chemical modifications to the DNAzyme. The results indicate the translational potential of the nucleic acid nanocapsule as an oligonucleotide delivery platform and highlights the importance of enzyme-specific degradation in designing biochemically compatible nanomedicines.

RESULTS AND DISCUSSION

Stepwise Assembly and Characterization of DNAzyme-NANs.

To build the peptide cross-linked DNAzyme-NAN construct, surfactant molecules are self-assembled into micelles as previously reported, with the exception that for this study, the cross-linker is built from protease-specific amino acids.²⁰

Specifically, the resulting micelle structure is cross-linked with a cysteine modified peptide substrate for the enzyme MMP9 (CGPLGLAGGERDGC) via a UV-mediated thiol−yne reaction between the terminal thiol groups of the peptide and the alkyne groups of the surfactant, forming a surface cross-linked micelle (SCM) (Figure 1).¹⁶ After cross-linking, a thiolated DNA anchor sequence was added to the surface of the micelle structure (see Table S1 for DNA sequences) via a second UV-mediated thiol−yne reaction to form the nucleic acid nanocapsules (NANs). NANs were further functionalized with a chemically unmodified (i.e., no phosphorothioate modifications) GATA3 DNAzyme through an enzymatic ligation to the DNA anchor presented on the NAN surface using our standard T4 DNA ligation protocol for attaching oligonucleotides.²⁰ Previous studies have shown that both the initial assembly of the NAN and the ligation of DNAzyme to the surface are highly efficient, resulting in nanocapsules that are densely coated with therapeutic nucleic acids with 2 DNAzymes per surfactant.¹⁴ These qualities make the particle behave similar to a spherical nucleic acid (SNA) structure—a class of materials which are known to exhibit high cellular uptake efficiencies and greater stabilization for chemically unmodified oligonucleotide ligands at their surface.^21,22

Figure 1.

Stepwise assembly of a peptide cross-linked GATA3 DNAzyme NAN. A multi-alkyne presenting surfactant is assembled in water, followed by a UV cross-linking step to attach thiolated peptide cross-linkers to the surface (see Figures S1) using 365 nm light. Next, a thiolated DNA anchor sequence (Table S1) is attached to the surface of the peptide SCM using a UV light catalyzed thiol−yne click reaction. Lastly, a GATA3 DNAzyme is ligated to the surface of the NAN using T4 DNA ligase.

Dynamic light scattering (DLS) and ζ potential measurements were used to characterize the NAN construct at each step of assembly (Figure 2A). SCMs, the NANs core, were shown to exhibit an average size of 23 nm and an average surface charge of +35 mV, due to the presence of the positively charged headgroup presented by the surfactants on the surface. After the addition of the DNA anchor to the SCMs exterior, the newly formed NANs are approximately 26 nm in size with a surface charge of roughly −33 mV. Finally, the ligation of the DNAzyme to the surface of the peptide cross-linked NAN results in a shift in size to approximately 49 nm as seen by DLS (Figure 2B, top) and observed by TEM (Figure 2B, bottom). The successful ligation of a dye-labeled DNAzyme to the NAN’s surface was also visualized using agarose gel electrophoresis as confirmation of the DNAzyme-NAN synthesis (Figure S2). Also, as similar surfactants to ours have been found to have an aggregation number close to 60,²³ we expect that we have close to 120 DNAzymes per NAN, as we have experimentally validated that we have on average 2 DNAzymes per surfactant in our structure.¹⁴ This density of DNAzymes is an important property of our structure particularly for cellular uptake and stability, as oligonucleotide density has been shown to play a key role in the uptake of DNA coated nanomaterials.²⁴

Figure 2.

GATA3 NAN characterization. (A) DLS and ζ measurements of peptide SCMs prior to DNA anchor attachment (top) and peptide cross-linked NANs post-DNA attachment (bottom). (B) DLS of DNAzyme-NANs after ligation of a GATA3 DNAzyme (top) and TEM analysis (bottom) showing DNAzyme-NANs stained with uranyl acetate (inset shows higher magnification image). (C) Schematic indicating the design of the ligation cassette used to assemble the DNAzyme NANs.

Effect of NAN Degradability on DNAzyme Activity.

After assembly and characterization of the peptide cross-linked DNAzyme-NANs (pep Dz-NANs), it was first of interest to investigate the knockdown of GATA3 mRNA in cells. Previous studies on the therapeutic potential of DNAzyme-NANs have shown that the release of individual DNAzyme-surfactant conjugates upon NAN degradation contributes to the endosomal escape and therapeutic efficacy of the DNAzyme.¹⁴

Importantly these DNAzyme-NANs were assembled using an ester cross-linkage. As the peptide cross-linker for this study is significantly longer than the original diazido ester cross-linker design and is composed of amino acid residues, it was important to investigate whether the pep Dz-NANs could achieve mRNA knockdown intracellularly. Therefore, it was important to verify the activity of the DNAzyme using the new MMP9 peptide cross-linked NANs. First, an in vitro activity assay was run using the active DNAzyme ligated to the MMP9 NANs. Its activity was monitored using an 8% denaturing polyacrylamide gel electrophoresis (PAGE) gel where the cleavage of a short 19mer truncated GATA3 mRNA sequence labeled with Cy3 was monitored to determine the relative activity of the DNAzyme. As a negative control, the active DNAzyme was compared to a GATA3 DNAzyme with mutations to either its flanking region, required for sequence-specific recognition of GATA3 mRNA, or to its catalytic loop, the region of the sequence needed for mRNA cleavage and turnover. As seen in the polyacrylamide gel assay in Figure S3, the GATA3 DNAzyme remains active in cleaving the mRNA target sequence in vitro when ligated to the NANs, and the mutated DNAzyme NANs are inactive. When compared in cell culture for knockdown activity in Jurkat T cells, it is shown that the mutant DNAzyme NANs are less effective in knocking down the expression of GATA3 mRNA than the active GATA3 DNAzyme NAN. The lack of consistent knockdown by the mutated DNAzymes, particularly by mutated DNAzymes 2 and 3, which are inactive due to loss of function of their catalytic loop, suggests the DNAzyme ligated to the NAN is cleaving the mRNA catalytically rather than through a knockdown mechanism due solely to an antisense effect or recruitment of RNase H. Using this information, we then investigated the effect the peptide-cross-linked NAN degradability had on GATA3 mRNA knockdown in vitro, wherein MCF-7 cells were treated with 250 nM pep-Dz-NANs for 4 h. We utilized MCF-7 cells for their high expression of GATA3 which would make it easy for us to monitor the effects of our pep-Dz-NANs. Relative mRNA expression levels were measured through qPCR and compared to untreated cells, wherein pep Dz-NANs achieved approximately 50% knockdown of GATA3 mRNA levels (Figure 3C). As a control, the peptide cross-linked NANs ability to knockdown GATA3 mRNA was compared to that of PEG cross-linked NANs, as the PEG cross-linker does not allow for enzymatic degradation (Figure 3A). Thus, the PEG cross-linked DNAzyme-NANs should remain intact when internalized into cells. The PEG DNAzyme-NANs showed approximately 30% knockdown of GATA3 mRNA relative to untreated cells (Figure 3C), less than that of the peptide cross-linked DNAzyme-NANs, knock down that likely resulted from the activity of DNAzymes that had been cleaved free of the NAN but not fully degraded themselves. To further investigate the role of MMP9 and cellular proteases on the stability of the peptide NANs, we treated them with MMP9 enzyme and 10% FBS for 24 h and observed the release of the DNAzyme as seen by 3% agarose gel electrophoresis (Figure 3B). These results indicate the importance of an enzyme cleavable cross-linker for achieving gene regulation. These results also reinforce the critical role of the surfactant in achieving gene regulation, as it has been shown to play a potentially important role in endosomal escape by assisting the DNAzyme’s efficacy in accessing cytosolic mRNA targets.¹⁴

Figure 3.

Enzymatic degradation and mRNA knockdown by a peptide cross-linked GATA3 DNAzyme NAN. (A) Schematic of a PEG cross-linked NAN indicating no degradation by enzymes compared to the degradation that is possible when the MMP9 peptide cross-linked NANs are exposed to cellular proteases. (B) 3% Agarose gel electrophoresis analysis of MMP9 peptide cross-linked NANs ligated with TYE 665 dye post-treatment with MMP9 enzyme and cellular proteases (10% FBS). (C) GATA3 mRNA knockdown assay showing the differences in silencing efficiency for a PEG cross-linked NAN vs MMP9 peptide cross-linked NANs (*p < 0.05 by Mann−Whitney U Test). (D) Dose response of GATA3 NAN treatment in MCF-7 cells (*p < 0.05 by Kruskal−Wallis Test).

Analysis of NAN Uptake by Flow Cytometry and Confocal Microscopy.

After confirming the activity of the pep Dz-NANs in vitro, it was of interest to analyze the extent to which human immune cells could internalize NANs, as these are the anticipated targets for GATA3 suppression in the treatment of asthma. For these studies, we utilized both fluorescence-assisted cell sorting (FACS) and confocal microscopy to track the uptake and localization of the pep Dz-NANs. Peptide cross-linked NANs were functionalized with TYE665-labeled DNAzymes to allow for the fluorescent tracking. Figure S4 indicates that the TYE 665 labeled DNAzyme alone with no transfection agent can not be taken up into cells, but that when ligated to the peptide, NANs are readily taken up into cells. Human peripheral blood mononuclear cells (PBMCs) were incubated with 250 nM TYE665 DNAzyme-NANs for 1 h, stained with a panel of fluorescently conjugated antibodies, and fixed with 4% formalin. For flow cytometry, treated cells were read using a BD FACSAria II Cell Sorter. All cell types showed evidence of uptake of NANs within the incubation time, but human peripheral blood monocytes (CD14+CD19− population) and human helper T cells (CD3+CD4+ population) appeared to have the greatest amount of uptake. Breakdown of NAN uptake by individual populations of cells can be found in Figure S5.

Approximately 97% of the monocyte population (Figure 4A, blue histograms) and 23% of CD4+ helper T cells (Figure 4A, green histograms) internalized NANs during the incubation period. In contrast, almost no CD8+ T cells (Figure 4A, red histograms) internalized the NANs, and <1% of the other probed cell types internalized NAN during the incubation (data not shown). For confocal microscopy, fixed cells were placed on a microscope slide and imaged using a Leica SP8 microscope. Fluorescence from the TYE665 DNAzyme on the NAN surface was visualized in both CD4+ T cells and monocytes, as seen by the colocalization of the DNAzyme signal with cell-specific antibodies (Figure 4B). Additional confocal images of helper T cells and all labeled immune cells can be found in Figures S6 and S7, respectively. These data suggest that NANs are capable of being internalized by human immune cells and, importantly, are capable of being internalized by human CD4+ helper T cells. It is important to note that although we aim to target T cells for GATA-3 reduction, there are numerous other cells, including bronchial epithelial cells that can intercept the NANs and result in knockdown as reported recently for the DNAzyme HGD40.²⁵ For our studies here, we focused on assessing knockdown in T cell cells in order to determine if the DNz-NAN construct can be internalized into this target population of cells.

Figure 4.

Uptake and colocalization of GATA3 DNAzyme NANs in human primary cells. (A) TYE665-labeled pep Dz-NAN uptake by various cell types. NAN internalization by CD14+ monocytes (blue), CD4+ helper T cells (green), and CD8+ T cells (red) is shown. Data are a stacked histogram representing three separate trials. (B) Representative confocal microscopy images of TYE665-labeled pep Dz-NAN internalization by CD14+ monocytes (left) and CD4+ helper T cell (right). TYE665-tagged pep Dz-NANs (red channel) are visible on the inside of the plasma membrane of both cell types. Scale bars are 20 μm.

NAN-Mediated GATA3 Knockdown in Jurkat T Cells In Vitro.

In order to determine the potential translational value of pep Dz-NANs to modulate allergic GATA3 expression in human cells, we decided to investigate their in vitro effect within immortalized human Th2-like Jurkat T cells. To this end, we incubated Jurkat T cells with 250 nM pep Dz-NANs for various durations of time and analyzed GATA3 expression changes via qRT-PCR. Pep Dz-NANs were found to efficiently induce the downregulation of GATA3 expression in Jurkat cells (Figure 5A). We also verified that this response was specific to the GATA3 DNAzyme by comparing its activity to a mutated DNAzyme (Figure S3B). GATA3 expression began to decrease as early as 4 h postexposure, with the greatest degree of downregulation evident at 24 h postexposure in comparison to untreated controls. By 48 h, GATA3 expression appeared to begin recovering toward levels seen in untreated control cells (Figure S8). The significant downregulation evident at 24 h postexposure, coupled with the trend toward restoration of baseline expression by 48 h, served as justification for the daily administration of pep Dz-NANs in our mouse model of HDM-induced asthma.

Figure 5.

GATA3 peptide DNAzyme NAN knockdown in human T cells and evaluation of in vivo efficacy. (A) Knockdown of GATA3 in Jurkat T cells by pep Dz-NAN. Percent expression is presented as percentage compared to untreated controls (dotted line). Data are shown as mean plus or minus SEM. ***P < 0.001 by Kruskil−Wallis test. (B) Eosinophil counts (relative %) from bronchoalveolar lavage fluid of mice treated with various concentrations of pep-Dz-NAN in a mouse model of HDM-induced allergic airway disease. Values shown are relative percentages of eosinophils present in slides made from collected bronchoalveolar lavage fluid. Each point represents the count from one animal. Statistical comparisons were completed using one-way ANOVA with Tukey’s posthoc. Significance was called for p < 0.05 (* <0.05, ** <0.01, ***<0.001, **** <0.0001). (C) Schematic illustrating in vivo analysis of peptide DNAzyme-NANs. (1) NANs are intranasally introduced to a HDM mouse model where (2) areas of inflammation within the lungs release MMP9 enzyme that can degrade the peptide DNAzyme NAN into individual DNAzyme surfactant conjugates for (3) for intracellular gene regulation of GATA3.

In Vivo Analysis of NAN Efficacy.

To identify the potential utility of the pep Dz-NANs in vivo, we investigated their ability to ameliorate allergic airway disease in a mouse model of HDM-induced asthma. Eighty 8-week-old male and female (n = 40 each) C57BL6 mice were purchased from the Jackson laboratory and housed in the vivarium facility at the University of Connecticut. After a 1-week acclimation period, mice were inoculated once daily, 5 days per week, for 5 weeks via the intranasal route. Study animals were enrolled in the following groups (n = 10 each, split evenly between males and females): PBS only, HDM only, 250 nM NAN only, HDM + 25 nM NAN, HDM + 125 nM NAN, HDM + 250 nM NAN, HDM + 1250 nM NAN, and HDM + 2500 nM NAN. Pep-Dz-NANs were delivered on a daily basis alongside HDM antigen based on the findings of the GATA3 knockdown study in Jurkat T cells, as described above. NANs without the DNAzyme but which present the inactive DNA anchor sequence were used as a control.

After the 5-week period concluded, mice were left for 48 h to allow for resolution of any acute inflammation and then humanely euthanized for bronchoalveolar lavage (BAL). Collected cells were spun down onto microscope slides, stained, and counted to identify relative percentages of Eosinophils. Overall, mice administered HDM antigen plus pep Dz-NANs exhibited lower levels of eosinophilia in BAL fluid as compared to HDM controls (Figure 5B). Notably, when eosinophil counts were broken down by sex, male mice who received NAN treatment maintained significantly different eosinophil counts in comparison to HDM only controls, whereas females only approached significance at the highest doses, suggesting a possible sex difference in the efficacy of the GATA3 DNAzyme treatments (data not shown). In addition, the NAN only sample bearing no DNAzyme did not result in an increase in eosinophils and caused no visible increase in inflammation, with eosinophil counts comparable to those of animals administered PBS control inoculations. This represents an important step in developing a safe and biodegradable nanocarrier for delivering therapeutic nucleic acids. These data suggest that, viewed globally, pep Dz-NAN treatments during the sensitization phase are capable of ameliorating eosinophil infiltration in HDM-induced allergic airway disease in mice and that potential sex differences in treatment efficacy warrant further investigation.

In this study, we showed the efficacy of peptide cross-linked GATA3-DNAzyme-functionalized nucleic acid nanocapsules (pep Dz-NANs) for the modification of cellular transcriptional activity and the amelioration of allergic airway disease symptoms in a mouse model of HDM-induced asthma. We found that pep Dz-NANs were readily taken up by human peripheral blood mononuclear cells, including both monocytes and CD4+ T cells. Notably, the transfection rates achieved with pep Dz-NANs were attained without the use of potentially cytotoxic transfection agents, which provides the additional benefit to these constructs of preventing damage to the target cell. Pep Dz-NANs were also found to significantly downregulate GATA3 expression in human Jurkat T cells by 24 h postexposure, demonstrating that robust GATA3 knockdown is achievable within Th2-like cells. Additionally, we determined that pep Dz-NANs administered to mice during HDM-induced allergic airway disease were able to significantly reduce pulmonary eosinophilia, a hallmark of allergic airway disease.

The GATA-3-specific DNAzyme utilized in this study, hgd40, was isolated via an in vitro selection in 2008 by Sel et al.¹⁹ Previous in vivo studies have shown that hgd40 exhibits limited toxicity after inhalation²⁶ and has favorable biodistribution and pharmacokinetic properties in allergic airway mouse models.²⁷ Due to its promising potential, hgd40 has also been investigated as the active ingredient in a drug developed by Sterna Biologicals, SB010. Clinical trials have shown improved lung function with limited off target effects in human patients, although long-term safety and efficacy of SB010 have not yet been determined.²⁸ Despite these encouraging results, the DNAzyme sequence utilized in previous studies required chemical modification with a 3′-inverted thymidine to increase nuclease resistance and enhance stability of the nucleic acid. In the NAN system presented here, chemical modifications were not necessary, as the stability of the DNAzyme is provided by the sterics generated by arranging the DNA at the surface of the nanocapsule—a property observed by nucleic acid functionalized nanostructures.²⁴ The incorporation of chemical modifications to the DNAzyme NAN platform would enable further stabilization, but the effectiveness of the native DNAzyme using the NAN highlights the importance of its structural design for achieving gene regulation highlighted here in the case of an allergic airway disease model.

At present, many asthmatic patients are treated using bronchodilators and other inhaled treatments intended to quell inflammation and relieve airway constriction, lessening their symptoms and discomfort. However, it is estimated that as much as 10% of asthmatic patients may suffer from asthma that is resistant to traditional steroid treatment options. These individuals often experience chronic, severe asthma with little to no relief from symptoms and additionally represent a substantial burden on healthcare systems.^29,30 For these patients, alternative therapeutic options, especially those that work at the cellular level and directly target the transcriptional controllers of asthma and the associated immune responses, may provide benefit. GATA3-NANs may be capable of providing benefit to patients experiencing asthma symptoms that are resistant to current treatment options and would be capable of doing so without the addition of potentially harmful agents often used for transcriptional modification of living cells. One potential limitation of this system is a lack of specific targeting of the pep Dz-NANs toward GATA3-expressing type 2 helper T cells. However, Th2 cells are not the sole GATA3-expressing cells involved in allergic asthma. GATA3 is also expressed in eosinophils,³¹ mast cells,³² and bronchial epithelial cells,³³ and these cells play important roles in the early stage asthmatic response through the secretion of cytokines and inflammatory proteins.²⁵ Knockdown of GATA3 in these cells needs to be evaluated further in vitro and in vivo, but decreased expression of GATA3 in these cell types would be expected to contribute to a dampening of the global asthmatic response. While the pep Dz-NANs did prove efficacious in mouse models of asthma by showing a decrease in the relative percent abundance of eosinophils, mice experience some different symptomology (namely, a lack of wheezing, limited plasma exudation, and modified eosinophil degranulation characteristics).^34,35 Demonstrated efficacy in mouse models of asthma shown in this study warrants future investigation to confirm the translational potential of these findings.

CONCLUSIONS

In summary, we have shown that a peptide cross-linked DNAzyme-NAN (pep Dz-NAN) can achieve GATA3 mRNA cleavage in MCF-7 cells and significant GATA3 knockdown in Jurkat T cells. Additionally, pep Dz-NANs undergo cellular uptake by primary immune cells, as seen through flow cytometry and confocal microscopy. Importantly, we have demonstrated that pep Dz-NANs are capable of regulating eosinophil levels in vivo, using a HDM-associated allergic airway disease mouse model. This indicates the utility of the pep Dz-NANs for delivering therapeutically active oligonucleotides for in vivo applications, and in particular, that they can be used for applications involving inflammation, both due to their enzyme-specific breakdown and target-specific knockdown capabilities.

METHODS

SCM and NAN Synthesis (Peptide Cross-Linked).

One mg of trialkyl modified surfactant (0.005 mmol) was dissolved in 233.4 μL of Millipore water. The solution was stirred at room temperature for 30 min. 15.6 μL of an 80 mM MMP9 peptide substrate stock (final concentration 5 mM) (sequence: CGPLGLAGGERDGC) was added to the solution to obtain a final volume of 250 μL, and the mixture was stirred in a Rhyonet reactor for 30 min. The resulting SCMs were characterized by DLS and ζ potential measurements. To prepare the NANs, 100 μM SCMs, 200 μM thiolated DNA, and 20 μM 2-hydroxy-4-(2-hydroxyethoxy)-2-methylpropiophenone (DHEMPP) were mixed in a total volume of 500 μL. The solution was placed in a Rhyonet reactor for 30 min and was purified via a Sephadex G-25 NAP-10 column. Fractions were analyzed by DLS and ζ potential.

SCM and NAN Synthesis (PEG Cross-Linked).

One mg of trialkyl modified surfactant (0.005 mmol) was dissolved in 242 μL of Millipore water, and the resulting solution was stirred at room temperature for 30 min. Five μL of a 25 mg/mL sodium ascorbate solution (0.00125 mmol), 2.5 μL of a 25 mM THPTA-Cu complex (0.000125 mmol), and 0.5 μL of a diazido PEG cross-linker (0.006 mmol) were added to the solution to obtain a final volume of 250 μL. The solution was stirred at room temperature for 3.5 h, and the resulting SCMs were purified via a Sephadex G-25 NAP-10 column. Fractions were analyzed by DLS and ζ potential measurements. To prepare the NANs, 100 μM SCMs, 200 μM thiolated DNA, and 20 μM DHEMPP were mixed in a total volume of 500 μL. The solution was placed in a Rhyonet reactor for 30 min and was purified via a Sephadex G-25 NAP-10 column. Fractions were analyzed by DLS and ζ potential.

Ligation of GATA3 DNAzyme to NAN.

Ten μM NANs were combined with 20 μM GATA3 DNAzyme and 40 μM GATA3 DNAzyme bridge in water to a final volume of 250 μL. The solution was heated at 70 °C for 10 min and cooled to room temperature. Five mM ATP, 10 μL of 1 U/μL T4 DNA ligase, and 1× ligase buffer were mixed in water to a final volume of 250 μL. The resulting solution was added to the DNA/NAN mixture and was placed on a 25 °C heat block for 2 h. Ligase was heat inactivated at 65 °C for 10 min, and the ligation product was purified via a Sephadex G-25 NAP-10 column. Fractions were analyzed by DLS measurements. Ligation of the TYE-665 dye labeled GATA3 DNAzyme to a NAN followed identical ligation conditions except that 5 μM NANs were combined with 10 μM TYE-665 labeled DNAzyme and 20 μM DNAzyme bridge in water to a final volume of 250 μL. Ligation fractions were analyzed by 3% agarose gel electrophoresis run at 120 V for 45 min. Gel was scanned at 670 nm.

Enzyme Degradation Assay.

TYE-665 labeled DNAzyme was ligated to MMP9 cross-linked NANs following ligation conditions as mentioned above except that 5 μM NANs were combined with 5 μM TYE-665 labeled DNAzyme (1:1 ligation) and 10 μM DNAzyme bridge. Ligation fractions were analyzed by 3% agarose gel electrophoresis run at 150 V for 35 min. Gel was scanned at 670 nm.

For the enzyme responsive cleavage assay, 2 μM of TYE-665 labeled DNAzyme NANs were treated with MMP9 enzyme (human, recombinant E. coli, Enzo, 0.6 μg) in the presence of enzyme buffer (25 mM HEPES and 10 mM CaCl₂) at a total volume of 100 μL. The solution was stirred at 37 °C, 550 rpm. Two μM NANs were treated with 10% FBS in PBS and stirred under the same conditions. Aliquots of these samples were taken at different time points (1,4, 18, and 24 h). Both these samples (enzyme treated and FBS treated) were compared to untreated NANs by 3% agarose gel electrophoresis run at 150 V for 35 min. The final concentration of NANs in the gel was 500 nM. Gels were scanned at 670 nm.

PAGE Gel Analysis of DNAzyme Activity.

GATA-3 mRNA truncate cleavage efficiency of free DNAzymes (active GATA-3 and mutated DNAzymes (1−3) containing mutations to either the DNAzymes flanking region or its catalytic loop, see sequences in Table S1) and those of NANs ligated with each of these DNAzymes were evaluated using PAGE analysis. This gel assays include a Cy3 labeled mRNA truncate, free active Dz + mRNA, mutated Dz 1 + mRNA, active Dz NAN + mRNA, mutated Dz 1 NAN + mRNA, mutated Dz 2 + mRNA, mutated Dz 2 NAN + mRNA, mutated Dz 3 + mRNA, and mutated Dz 3 NAN + mRNA.

A solution of 0.5 μM mRNA truncate was prepared with 5 μM of respective free Dz/NAN bound Dz in a total volume of 20 μL. The reaction mixtures were stirred at 37 °C, 550 rpm for 4 h. The cleavage reaction was carried out in the presence of 10 mM MgCl₂ and 100 mM NaCl. The final concentrations of particles in the gel were 0.25 μM of mRNA truncate and 2.5 μM free Dz/Dz NANs, and the gel electrophoresis run was run at 350 V for 35 min. The gel was scanned at 520 nm.

GATA3 mRNA Knockdown in MCF-7 Cells.

Confluent MCF-7 cells were treated with either 250 nM PEG-cross-linked DNAzyme-NANs or 250 nM peptide-cross-linked DNAzyme-NANs in OptiMEM for 4 h at 37 °C and 5% CO₂. After incubation, total cellular RNA was isolated using a Qiagen RNeasy Plus Mini Kit (Qiagen, Hilden, Germany, catalog no. 74134), following the protocol provided by the manufacturer. Reverse transcription was performed using BioRad iScript Reverse Transcription Supermix (BioRad, Hercules, CA, catalog no. 1708840), and RT-qPCR was performed on a BioRad CFX Connect Real-Time PCR Detection System, using iTaq Universal SYBR Green Supermix (catalog no. 1725121). Data are representative of 7 technical replicates. All data is normalized to GAPDH as a housekeeping gene. Statistical analysis was conducted by Mann−Whitney U Test with significance called for p < 0.05. For the dose-dependent studies, MCF-7 cells were cultured in DMEM (supplemented with 10% FBS) at 37 °C and 5% CO₂. Confluent cells were plated in a 6-well plate at a density of 100,000 cells/mL, a day before the experiment. Except the untreated (control) well, each well was treated with varying concentrations of GATA-3 NANs in OptiMEM (125 nM, 250 nM and 1250 nM) for 4 h. Cells were lysed, and RNA extraction was performed using Qiagen RNeasy Plus Mini Kit. The isolated RNA templates were reverse transcribed using iScript Supermix (Bio-Rad) followed by real-time PCR with respective primers and iTaq SYBR Green Supermix. mRNA expression was calculated using the data analysis tool present in Bio-Rad CFX Connect Real-Time PCR Detection System. The GATA-3 expression was normalized to the expression of the house keeping gene GAPDH.

GATA3 mRNA Knockdown in Jurkat T Cells.

Jurkat T cells were kindly donated to our group by Dr. Andrew Wiemer of the University of Connecticut School of Pharmacy. Cells were thawed using fresh, prewarmed RPMI 1640 media supplemented with l-glutamine and 10% FBS, and then established as a continuous culture via growth at 37 °C in 5% CO₂. Pep Dz-NANs were added to test wells at a concentration of 250 nM and incubated for 4, 24, and 48 h. RNA isolation was performed using a Qiagen RNeasy Plus Mini Kit (Qiagen, Hilden, Germany, catalog no. 74134) following the protocol provided by manufacturer. Reverse transcription was performed using BioRad iScript Reverse Transcription Supermix (BioRad, Hercules, CA, catalog no. 1708840) per manufacturer instructions. RT-qPCR was performed on a BioRad CFX Connect Real-Time PCR Detection System, using iTaq Universal SYBR Green Supermix (catalog no. 1725121) per manufacturer instructions. Data are representative of 3 technical replicates per time point. All data are normalized to GAPDH as a housekeeping gene, with GAPDH mRNA extracted from controls cultured for the same period of time. For the dose−response studies in Jurkat cells, each test well was treated with varying concentrations of GATA-3 NANs (125 nM, 250 nM and 1250 nM). For studies involving mutated DNAzymes, 250 nM of active GATA-3 was used and compared to 250 nM mutated DNAzymes 1, 2, and 3. The respective volumes of NAN were directly added to the test wells containing Jurkat cells suspended in RPMI media (no OptiMEM was used). Data were presented as percent expression compared to untreated controls, with percentage values analyzed for significance using a Kruskil−Wallis test with Dunn’s posthoc test. Data were considered significant for p < 0.05.

Mice.

Forty-six male and 46 female 8-week old C57BL/6 mice were purchased from Jackson Laboratories (Jackson Laboratories, Bar Harbor, ME). Mice were housed in the vivarium at the University of Connecticut for the duration of the study (Animal Protocol A17−034). Mice were acclimated to the vivarium for 1 week after arrival before the start of the study.

In Vivo Efficacy Study.

All methods described here were done so with prior approval by UConn’s IBC and IACUC review boards in accordance with local and national guidelines. Eighty total mice were divided into 8 groups (n = 10) with equal numbers of males and females. Mice were given 50 μL intranasal inoculations via micropipette 5 days per week for 5 weeks, as previously described.^34,35 Animals that received the HDM antigen were given a 1:1 mixture of lyophilized Dermatophagoides farinae and Dermatophagoides pteronyssinus antigen resuspended in sterile PBS to a concentration of 1 μg/uL. This suspension was then diluted with either 25 μL of PBS or pep Dz-NAN mixture for a final volume of 50 μL per inoculum. Animals receiving HDM plus NAN were given 25 μL of the HDM mixture plus 25 μL of appropriate NAN concentration. The groups were as follows: group 1, PBS only; group 2, HDM only; group 3, HDM + 25 nM pep Dz-NAN; group 4 HDM + 125 nM pep Dz-NAN; group 5 HDM + 250 nM pep Dz-NAN; group 6, HDM + 1250 nM pep Dz-NAN; group 7, HDM + 2500 nM pep Dz-NAN; and group 8, 250 nM pep Dz-NAN only. At the end of the 5-week period, mice were humanely euthanized by CO₂ asphyxiation and cervical dislocation, and BAL was performed. Whole lungs were then collected in 10% formalin for histological analysis. BAL fluid was spun onto microscope slides using a Cytospin 4 cytocentrifuge (ThermoFisher, Waltham MA), and stained for leukocyte differential counting using the ThermoScientific Shandon Kwik-Diff Stain Kit (ThermoFisher, Waltham MA). Slides were analyzed using a microscope, differential counting was performed, and cell subsets were presented as percent of total. Statistical analysis was performed via one-way ANOVA with Tukey’s posthoc test. Statistical significance was called for p < 0.05.

Cell Distribution Flow Cytometry.

Human PBMCs were purchased from ATCC (catalog no. PCS-800-011). One million cells were aliquoted into each sample group and resuspended in FACS buffer. Groups receiving nanoparticles were incubated with 250 nM TYE665-labeled pep Dz-NANs for 1 h at 37 °C in an incubator with light shaking. Cells were then washed with FACS and moved into the staining protocol. Cells were stained at 4 °C in the dark, washed with FACS buffer, and fixed in 4% PFA. Groups were stained with the following cocktail of monoclonal antibodies: CD19 PE-Cy7, CD14 PE, CD3 AF700, CD4 BV510, CD8 FITC, and CD56 BV421. Nanoparticle (NAN) groups were incubated with TYE665-labeled peptide cross-linked DNAzyme-NANs. NAN control groups received peptide cross-linked DNAzyme-NANs without TYE665. All samples were read on a BD LSRFortessa X-20 (BD Biosciences, Franklin Lakes, NJ), and data were analyzed in the FlowJo software (BD Biosciences, Franklin Lakes, NJ). For T cell subset analysis, cells were treated, stained, and analyzed per the above protocol, but using the following panel of monoclonal antibodies: CD3 AF700, CD4 BV510, CD8 FITC, CTLA-4 PE-Cy7, GATA3 PE, and pep Dz-NANs with or without TYE665. All samples were run with appropriate single stain and FMO controls, with doublets excluded from final analysis by gating.

Confocal Microscopy.

Human PBMCs were purchased from ATCC (catalog no. PCS-800-011). Cells were thawed and diluted in cold RPMI 1640 supplemented with l-glutamine and 10% FBS. One million cells were aliquoted into separate tubes and incubated with 250 nM TYE665-labeled pep Dz-NANs for 1 h at 37 °C and 5% CO₂. After incubation, cells were centrifuged at 1500 rpm for 5 min at 4 °C. Supernatant was removed, and cells were resuspended in 1 mL cold FACS buffer for blocking and staining. Cells were blocked with Human BD Fc Block for 15 min on ice. Cells were then stained with the following monoclonal antibodies: CD19 PE-Cy7, CD14 PE, CD3 AF700, CD4 BV510, CD8 FITC, and CD56 BV421. Cells were washed with FACS buffer and fixed with 4% PFA. After fixation, cells were washed again and resuspended in 2.5 mL cold FACS buffer. Twenty-five μL of fixed PBMC mixture was placed on a microscope slide, covered with a coverslip, and stored at 4 °C until imaging. Confocal microscopy images were taken using a Leica SP8 microscope.

Statistical Analysis.

Statistical analysis was done via the GraphPad Prism 8 software (GraphPad Software, San Diego CA). Relevant statistical analyses are listed in the methods subsection for each respective experiment. In all cases, data were considered significant for p < 0.05, with on-graph significance denoted as follows: *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.