Enhanced Immunostimulatory Effects of DNA-Encapsulated Peptide Hydrogels

Abstract

DNA that encodes tumor-specific antigens represents potential immunostimulatory agents. However, rapid enzymatic degradation and fragmentation of DNA during administration can result in limited vector expression and, consequently, poor efficacy. These challenges have necessitated the use of novel strategies for DNA delivery. Herein, we study the ability of cationic self-assembling peptide hydrogels to encapsulate plasmid DNA, and enhance its immunostimulatory potential in vivo. The effect of network charge on the gel’s ability to retain the DNA was assessed employing three gel-forming peptides that vary systematically in formal charge. The peptide HLT2, having a formal charge of +5 at neutral pH, was optimal in encapsulating microgram quantities of DNA with little effect on its rheological properties, allowing its effective syringe delivery in vivo. The plasmid, DNA(TA), encapsulated within these gels encodes for a melanoma-specific gp100 antigen fused to the alarmin protein adjuvant HMGN1. Implantation of DNA(TA)-loaded HLT2 gels into mice resulted in an acute inflammatory response with the presence of polymorphonuclear cells, which was followed by infiltrating macrophages. These cellular infiltrates aid in the processing of encapsulated DNA, promoting increased lymphoproliferation and producing an enhanced immune response mediated by CD4+/IFNγ+ expressing Th1 cells, and complemented by the formation of gp100-specific antibodies.

Introduction

DNA constructs, including plasmids and oligonucleotides, have been explored as potential therapeutics with applications in immunotherapy, vaccine platforms and gene therapy [1, 2]. As an immunostimulatory agent, DNA vectors that encode for antigenic protein are capable of inducing antigen-specific antibodies through humoral mechanisms, and eliciting T cell responses, following their expression in cells [3, 4]. Consequently, upon antigen expression, plasmids can serve as both effective vaccines and therapeutics, with potential to treat a variety of diseases [1, 2]. However, systemic administration of naked DNA, even at high doses, often results in poor efficacy due to its susceptibility to enzymatic degradation, as well as DNA fragmentation as a consequence of shear stress applied during injection [5, 6]. To enhance the efficacy of delivered DNA, and overcome these issues, techniques including “gene gun” administration [7], electroporation [2, 6], and nanoparticle delivery [8, 9], are currently being explored. While these methods offer some advantages, a significant disadvantage is that the delivered DNA does not stay localized to the tissue to which it is applied. Consequently, the vector can be rapidly cleared in vivo before sufficient expression of the antigen.

Alternatively, implantable soft hydrogel materials have the potential to encapsulate large quantities of DNA, and show prolonged retention of the vector at the site of injection, increasing the probability of a more sustained expression and, subsequently, an enhanced immune response [10–12]. In addition, the initial foreign body response, which is elicited by nearly every type of implanted material, results in significant inflammatory cell infiltration [13]. This creates a large local concentration of cells that can internalize and express the DNA. Expression by the host of cellular responders can provide a depot of antigenic protein for infiltrating antigen-presenting cells (APCs), such as macrophages, which are capable of processing the expressed protein and facilitating an adaptive immune response. Since the soft material is capable of being bioresorbed, the acute/chronic inflammation associated with the foreign body response will give way to subsequent tissue repair and remodeling. Such a material represents a traceless delivery vehicle for implanted DNA. The choice of gel composition to facilitate this approach is critical. Candidate materials are those that can directly encapsulate plasmid DNA and allow for its subsequent facile implantation in vivo. The gel must be capable of retaining the DNA locally within its network over a time period commensurate with the inflammatory response, while allowing for cellular uptake of the construct and its effective expression. Herein, we employ a class of hydrogels formed from self-assembling peptides to directly encapsulate and deliver plasmid DNA, Fig. 1A. We assessed the bioresportion profile of these DNA loaded gels, and determine their immunostimulatory potential in vivo.

Figure 1:

DNA encapsulation into peptide hydrogels and the proposed delivery of the vector to infiltrating cells following subcutaneous injection. (A) Direct encapsulation of plasmid DNA into hydrogel networks composed of self-assembled β-hairpin peptide fibrils. (B) Shear-thin injection of DNA-loaded gels into the flanks of mice and subsequent infiltration of inflammatory cells. Internalization, transcription and translation of the encapsulated pDNA by various cell types produce the adjuvant-antigen fusion protein HMGN1-gp100 to be taken up by antigen-presenting cells.

Table 1 shows the sequence of the parent gel-forming peptide, MAX1, which is comprised of two amphiphilic β-strands containing alternating hydrophobic and hydrophilic residues (lysine and valine) flanking a tetrapeptide type II’ β-turn [14–16]. In low temperature, low ionic strength aqueous solutions at pH 7.4, the peptide is freely soluble and remains unfolded due to side-chain electrostatic repulsions. A sol-gel phase transition can be triggered by increasing the ionic strength of the solution and warming to 37 °C. Increasing the ionic strength screens the lysine-borne charge of the peptide, and increasing the solution temperature facilitates the desolvation of hydrophobic residues. These environmental changes trigger the folding and assembly of the peptide into a nanofibrillar hydrogel network [16–20]. Importantly, the exterior of each fibril comprising the gel displays cationic lysine residues allowing for direct encapsulation of the anionic DNA when present during triggered hydrogelation.

Table 1: Sequence and formal charge of peptides utilized to prepare DNA loaded hydrogels.

Underlined residues represent design changes relative to the MAX1 sequence.

Peptide	Sequence	Net Charge
MAX1	VKVKVKVKVDPPTKVKVKVKV-NH2	+9
MAX8	VKVKVKVKVDPPTKVEVKVKV-NH2	+7
HLT2	VLTKVKTKVDPPTKVEVKVLV-NH2	+5

The primary sequence of MAX1 contains eight lysine residues, and along with its N-terminal amino group, is characterized by a formal charge of +9 at neutral pH. This electropositive character, although needed to sequester and retain the DNA in the network during delivery and during the initial foreign body response, may also inhibit dissociation of the peptide from the DNA, which ultimately must occur for the proper gene expression of the antigenic protein. In fact, earlier studies investigating the mass transport properties of similar hairpin peptide gels with respect to protein delivery showed that network charge significantly impacts the retention of protein within the gel [21, 22]. To determine if network electrostatics play an important role in the encapsulation and delivery of highly charged DNA, we also studied the activity of two additional peptides that differ in their overall charge state. The peptide MAX8 incorporates a glutamate residue at position 15 instead of lysine reducing its formal charge to +7 [23]. The third peptide, HLT2 contains the same glutamate, but also replaces two additional lysine residues at positions 2 and 19 with leucine affording a peptide having a +5 formal charge [24]. HLT2 also contains two threonine residues, replacing two of MAX1’s valine residues, to increase its hydrophilicity. The electropositive character of these three hydrogel networks vary systematically, which may influence not only their DNA binding and delivery aptitude, but also their bioresorptive properties.

Finally, a distinctive feature of these self-assembled hydrogels is that the non-covalent cross-links which define their networks eliminates the need for chemical cross-linking agents. With respect to DNA delivery, this is advantageous as it excludes the potential for chemical modification of the vector during gel formation. More importantly, these physical cross-links allow for shear-thin delivery of the material via syringe. Previous studies from the Pochan group have shown that the shear force applied from depressing the syringe plunger results in the thinning of the material only at the interface between the gel and the syringe-barrel. As a result, the gel flows as a plug through the syringe where the bulk of the material experiences little shear rate during delivery [25]. This is particularly valuable for delivery of DNA as it will likely reduce fragmentation of the vector during injection, as opposed to gene gun delivery. Once implanted, the gel can serve as a depot for the processing of the DNA by infiltrating cells as shown in Fig. 1. After administration, clearance of the peptide gel by the cellular responders will allow for internalization and expression of the vector, ultimately leading to presentation of the antigen protein to elicit a specific immune response (Fig. 1B).

We selected a model plasmid vector designed to express an antigen capable of stimulating a tumor-specific immune response. Specifically, this vector was designed to encode for a protein fusion composed of the melanoma-specific tumor antigen gp100, and the adjuvant HMGN1 (high-mobility group binding protein 1). The gp-100 protein mediates intracellular trafficking that regulates melanosome biogenesis. It has been shown to induce cell migration by interacting with chemokine (C-C motif) receptor 2 (CCR2), and stimulates CCR2 expressing monocytes, T cells and immature dendritic cells [26]. HMGN1 is an endogenous alarmin protein capable of mediating the recruitment and activation of APCs by interacting with toll-like receptor 4 to induce antigen-specific Th1 immune responses via innate and adaptive mechanisms [27]. As a result, the HMGN1-gp100 fusion may stimulate an immune response against tumors that express this antigen. Research in the Oppenheim lab has shown that this HMGN1-gp100 DNA vector can amplify cellular immune responses that protect mice against melanoma tumor challenge as compared to separate HMGN1 and gp100 DNA vectors [28]. Herein, we study the immunostimulatory potential of this plasmid encapsulated in self-assembled peptide gels. We demonstrate that all three peptide hydrogels are capable of directly encapsulating and retaining the HMGN1-gp100 DNA vector with minimal disruption of the materials’ rheological properties, thus allowing their subcutaneous implantation by syringe. In vivo studies showed that plasmid DNA loaded gels formed from the HLT2 peptide produced a significant humoral immune response to the gp100 antigen, while being bioresorbed over the course of treatment. Histological analysis showed that material implantation resulted in an acute inflammatory foreign body response characterized by significant infiltration of polymorphonuclear cells, followed by macrophages, which presumably facilitates an antigen-specific immune response.

1. Materials and Methods

2.1. Materials

Fmoc-protected amino acids were purchased from Novabiochem. PL-Rink resin was purchased from Polymer Laboratories. 1H-Benzotriazolium 1-[bis(dimethylamino) methylene]-5chloro-hexafluorophosphate (1-),3-oxide (HCTU) was obtained from Peptides International. Trifluoroacetic acid was obtained from Acros organics. 1,2-ethanedithiol was purchased from Fluka. Diethyl ether was purchased from Fisher Scientific. Tritiated thymidine [³H TdR] was purchased from Amersham Pharmacia. Cytokines were purchased from PeproTech (Rocky Hill, NJ). Heat-inactivated fetal calf serum was Hyclone purchased from Thermo. L-Glutamine and 100 U/ml penicillin and 100μg/ml streptomycin were purchased from Gibco, Tissue culture media and plastic ware was purchased from Corning Life Sciences. All other materials were purchased from Sigma/Aldrich

2.2. Peptide Synthesis and Purification

Peptides were synthesized on PL-Rink resin using an automated ABI 433A peptide synthesizer. Synthesis was carried out via Fmoc-based solid-phase peptide chemistry with HCTU activation. Dried resin-bound peptides were cleaved from the resin and simultaneously side-chain deprotected using a trifluoroacetic acid/thioanisole/1,2-ethanedithiol/anisole (90:5:3:2) cocktail for 2 h under argon atmosphere. Crude peptides were precipitated with cold diethyl ether and then lyophilized. Reverse-phase HPLC equipped with a semi-preparative Vydac C18 column was employed to purify the peptides. HPLC solvents consisted of solvent A (0.1% TFA in water) and solvent B (0.1% TFA in 9:1 acetonitrile/water). Linear gradients consisted of 0–100% solvent B over 100 min for MAX1 and MAX8 peptides. The HLT2 peptide was purified using a gradient of 0 – 20% solvent B over 10 min., followed by 20 – 100% over an additional 80 min. Peptide solutions were collected and then lyophilized. The purity of peptides was verified by analytical HPLC and electrospray ionization (ESI-positive mode) mass spectrometry.

2.3. Dynamic Oscillatory Rheology

Oscillatory rheology experiments were performed on an AR-G2 rheometer (TA Instruments) using 25 mm diameter stainless steel parallel plate geometry. For the rheological measurements, gels were prepared directly on the rheometer plate in the following manner. Peptide stock solutions were first prepared in glass vials by dissolving 4 mg of peptide in 200 μL of sterile, chilled water. To this solution, 200 μL of chilled BTP buffer (100 mM, pH 7.4) containing 300 mM NaCl was added. To prepare gels that directly encapsulate the DNA, 200 μL of chilled BTP buffer (100 mM, pH 7.4) containing 300 mM NaCl and different amounts of the respective DNA was added to the peptide solution. This results in 1.0 weight-percent (wt %) gel of a final total volume of 400 μL. Then, 300 μL of the resulting solution was quickly added to the rheometer plate, which was pre-equilibrated at 5°C. The parallel plate tool was then lowered to a gap height of 0.5 mm and the temperature was ramped linearly to 37 °C to initiate gelation. For the gelation experiment, a dynamic time sweep was performed to measure the storage (G′) modulus at a frequency of 6 rad/s and 0.2% strain as a function of time for 1 h.

2.4. DNA Release from Peptide Hydrogels

For the bulk release studies, MAX1, MAX8 or HLT2 peptide stock solutions were first prepared in glass vials by dissolving 4 mg of each peptide in 200 μL of sterile, chilled water. Then, 100 μL of each stock was added to a separate glass vial (n = 3). 100 μL of chilled, fresh saline containing DNA was added to the stock solution. Samples were carefully shaken to initiate hydrogelation, resulting in 1.0 wt% gels of a final total volume of 200 μL containing 7 μg of DNA. Samples were placed in an incubator at 37 °C for 1 h. All hydrogels were made in cylindrical glass vials and had only the top surface exposed for release. After 1 h, 1 mL of pH 7.4, 50 mM BTP, 150 mM NaCl buffer was added to the top of gel. At scheduled time points, the entire volume of buffer above the gel was removed and replaced with fresh buffer. DNA concentration was determined for each removed aliquot as a function of time. Each time point was performed in triplicate and experiments were carried out for 14 days. At the end of the release experiment gels were mechanically disrupted to recover any unreleased DNA which was not complexed with the hydrogel network. The concentration of DNA within the supernatant was determined via PicoGreen assay following the manufacturer’s instructions (Invitrogen), with results compared to a calibration curve to calculate total cumulative DNA release over time.

2.5. Immunization of mice

Six- to eight weeks old female wild-type C57BL/6 mice were purchased from Charles River (Frederick, MD) and housed under specific pathogen-free conditions with water and food given ad libitum. All experiments were approved by the National Cancer Institute at Frederick Animal Care and Use Committee and performed in accordance with the principles and procedures outlined in the Guide for Care and Use of Laboratory Animals (National Research Council, Washington, DC). These studies were performed under approved protocol ASP 13–294 (n ≥ 4). Plasmid DNA constructs were designed as described previously [28]. Briefly, the fusion gene for HMGN1-gp100 was cloned into the eukaryotic expression plasmid pcDNA 3.1/myc-His B (Life Technologies, Carlsbad, CA). Plasmid DNA was purified using EndoFree Plasmid Mega Kit (Qiagen, Germantown, MD) as per manufacturer’s instructions. 1 wt% MAX1, MAX8 or HLT2 peptide hydrogels were prepared with encapsulated HMGN1-gp100 expressing plasmid DNA (DNA(TA)) and syringe delivered subcutaneously to both the left and right dorsal flanks of mice. Gels prepared without DNA, or containing an empty plasmid vector (DNA(−)), were also implanted into the flanks of mice as controls. For initial experiments, 10 μg of DNA(TA) in 100 μL of gel was utilized for prime (day 1) and boost (day 8) treatments. Mice were euthanized 28 days after priming for analysis of the harvested tissues and remaining hydrogels. Subsequent experiments utilizing HLT2 peptide hydrogels were performed by injecting 50 μL of gel containing 20 μg of DNA(TA) in both the left and right dorsal flanks of mice as previously described. For these experiments, mice were primed (day 1) and boosted twice (day 8 and 15), with groups sacrificed on days 7 and 21 to harvest tissues and remaining hydrogels for further analysis. Finally, naked DNA was also administered to mice at equivalent concentration as controls, prepared as a bolus dose in saline or delivered via electroporation following previously reported protocols [29].

2.6. Histology and immunohistochemistry

Mouse tissues were fixed in 10% neutral buffered formalin, routinely processed, and embedded in paraffin before sectioning at 5 μm and stained with hematoxylin and eosin for histopathological examination (n = 4). For immunohistochemistry, 5 μm sections were deparaffinized into ethanol and endogenous peroxidase activity was blocked using 0.6% H₂O₂ in methanol. To stain samples for the presence of macrophages, antigen retrieval was carried out by microwaving slides in citrate buffer, followed by incubation with F4/80 (macrophage cell marker) biotin-conjugated antibodies at a 1:50 dilution for 60 minutes. A separate series of slides were stained for the presence of endothelial cells by microwaving sections in 10mM Tris, 1mM EDTA, pH 9.0 buffer for antigen retrieval, followed by blocking with 2% normal rabbit serum for 20 minutes and then incubated with anti-mouse CD34 antibodies (1:50). Secondary binding to CD34 was performed using the biotinylated rabbit anti-rat IgG, mouse adsorbed (Vector Laboratories). For both sets of slides, detection of cell surface signal was performed using the avidin-biotinylated enzyme complex (Vector Laboratories) with 3,3’-diaminobenzidine (Sigma) as a chromagen, and additionally counterstained with hematoxylin.

To stain samples for the presence of B cell and T cell lymphocytes, isolated tissue sections were microwaved in Citra solution (citrate buffer, BioGeneX), followed by incubation with biotin-conjugated CD3 (T cell marker) antibodies (1:50 dilution) for 60 minutes. Detection of CD3 binding was accomplished using the VECRASTAIN^® ABC-AP system (Vector Laboratories) with vector blue as a chromagen. Sections were co-stained via incubation with CD45R/B220 (B cell marker) antibodies (1:200 dilution) for 60 minutes, followed by antibody detection using the VECRASTAIN^® ABC system (Vector Laboratories) with biotin block and 3,3’-diaminobenzidine as a chromagen. All slides were photographed using an Olympus BX40 microscope equipped with an Q-image RGB camera, and images captured using Bioquant (Nashville TN) software.

2.7. T Cell Proliferation and Vital Assay

Single-cell suspensions of spleen and draining inguinal lymph node cells were plated in quadruplicate into wells of a round-bottom 96-well tissue culture plate (2×10⁶ cells mL⁻¹) and incubated for 5 days in complete medium (RPMI 1640, supplemented with 10% FBS and beta-mercaptoethanol) at 37°C in 5% CO₂. During the last 18 hours cells were pulsed with tritiated thymidine [³H TdR] at 5μCi/mL after which [³H TdR] incorporation was measured by beta scintillation counting (n = 4).

A tumor cell killing assay was performed by co-culturing B16F1 mouse melanoma cells with splenic or lymph node effector cells. B16F1 cells were cultured in Dulbecco’s Modified Eagle Medium (Corning, Manassas, VA) supplemented with 10% FBS (Hyclone, Logan, UT), 2mM L-glutamine, 25mM HEPES, 1x vitamin solution and 1x nonessential amino acids solution (Life Technologies). Target B16F1 cells were labeled with CellTrace™ Violet (CellTrace™ Violet Cell Proliferation Kit, Life Technologies) at 1×10⁶ cells per μL 5mM stock solution, and cultured together with effector spleen and draining inguinal lymph node cells in polypropylene tubes at effector:target ratios of 10:1 and 3:1, respectively. After incubating the co-culture for 5 hours at 37°C in 5% CO₂ cells were collected, washed and stained with propidium iodide (Sigma Aldrich, St. Louis, MO). Data was acquired on an LSR II flow cytometer (BD Biosciences, San Jose, CA) and the percentage of dead B16F1 cells was analyzed as CellTrace™ Violet/Propidium Iodide double positive cells with FlowJo software (TreeStar Inc., Ashland, OR).

2.8. Measurement of gp100-specific antibody titer

Antibodies were quantified using the ELISA method of Stenger et al. (n ≥ 4), with the modification that the antigen coated onto the wells of Nunc polystyrene microtiter plate was 4 μg/mL of recombinant murine gp100 [30]. The plates were developed using TMB (KPL) and the reaction stopped with acidified stop solution (KPL). The resulting optical density was read at 450 nm and the OD values plotted against serum dilution. The resulting data was analyzed using nonlinear regression of semi log data as performed by GraphPad Prism.

2. Results and Discussion

3.1. DNA Encapsulation and Retention in Peptide Hydrogels

The ability of each peptide gel to encapsulate and retain the HMGN1-gp100 DNA vector, defined hereafter as DNA(tumor-antigen) or DNA(TA), was assessed. DNA(TA) can be directly encapsulated by the addition of buffer which contains 40 – 800 μg/mL of DNA to a 2 wt% aqueous solution of peptide. Folding and self-assembly are immediately triggered resulting in 1 wt% hydrogels containing 20 – 400 μg/mL of DNA(TA). This method ensures complete loading of the plasmid directly into the peptide network. Dynamic oscillatory time-sweep rheological experiments showed that increasing amounts of DNA(TA) had little effect on the rate of gel formation and the mechanical rigidity of the resulting loaded gels, (Supplementary Fig. 1). This suggests that the presence of DNA during the folding and self-assembly of the peptides does not change the mechanism of hydrogelation, nor significantly disrupt the formation of the fibrillar network comprising the gels.

Once the DNA is encapsulated and the loaded-gel delivered, the vector should be effectively retained within the peptide gel network over a period of time commensurate with the rate of inflammatory cell infiltration. Acute inflammatory cells migrate to implanted foreign materials over the course of a few days, followed by infiltration of monocytes that takes place over the course of days to weeks [13]. Thus, the ability of each peptide gel to retain encapsulated DNA was measured in an ex vivo experiment over a two-week period, by submerging the loaded gels in physiologic buffer under sink conditions. Fig. 2 shows that after 14 days each gel is capable of retaining >90% of the loaded DNA(TA). This suggests that a 1 wt% fibrillar network composed of a peptide having a minimum formal charge of +5 is sufficient to retain high levels of the loaded DNA. Further, after the two-week incubation period, the gels were mechanically disrupted by repetitive shear-thinning to determine the possibility of DNA release during syringe delivery. It should be noted that the repetitive shear-thin/recovery cycles used in this experiment are much more aggressive than the conditions encountered by the gel during actual syringe delivery. At any rate, only 14% of the encapsulated DNA was released under these aggressive conditions, suggesting that the gel network will be capable of retaining the majority of its payload during its syringe-based implantation. This level of retention is similar to other polycationic materials used for gene therapy applications [31, 32].

Figure 2:

Encapsulated DNA is highly retained within peptide hydrogels. Percent of encapsulated HMGN1-gp100 plasmid (DNA(TA)) retained within 1 wt% MAX1, MAX8 or HLT2 hydrogels incubated in a 37 °C, pH 7.4, 50 mM BTP, 150 mM NaCl physiologic buffer over two weeks. Gels were mechanically disrupted (MD) at the end of the release experiment and total DNA retained quantified.

3.2. Injectable DNA-Loaded Peptide Hydrogels Induce Increased Ex Vivo Lymphoproliferation

Each of the DNA-loaded gels were assessed for their ability to promote lymphoproliferation, which serves as an indicator for the loaded materials’ ability to elicit an immune response. This provides an initial screen of the materials with respect to their immunostimulatory potential. Mice were primed with DNA(TA) loaded MAX1, MAX8 or HLT2 peptide hydrogel depots via subcutaneous syringe injection into the left and right dorsal flanks. After this priming dose, mice were boosted with the same formulation one week (day 8) later, and the lymphoproliferative response was assessed at day 28 (Fig. 3A). Stimulation of lymphocytes was measured by removing the inguinal lymph nodes of mice and measuring the proliferation of isolated lymphocytes using a thymidine incorporation assay. The data show that MAX8 and HLT2 hydrogels containing the DNA(TA) sequence produced an approximate 3 – 4 fold increase in ex vivo lymphocyte proliferation compared to both subcutaneous administration of naked DNA(TA) and control hydrogels without DNA. This suggests, and later experiments confirm, that for these two loaded gels, the DNA is being processed by cells present at the injection site and presented by APCs to lymphocytes, resulting in their activation. Interestingly, the MAX1 peptide gel containing encapsulated DNA(TA) showed no significant increase in cellular proliferation compared to the controls. This could be due to the inability of the DNA to be released from the highly cationic MAX1 peptide network, a necessary requirement for DNA expression. As of yet, we do not know the exact mechanism by which the DNA enters cells nor the identity of the cell types responsible for its expression. However, it is probable that a variety of cells present at the injection site participate, such as infiltrating inflammatory cells, endothelial and stromal cells, to name a few. Irrespective of cell type and internalization mechanism, if the DNA remains tightly bound to the highly cationic MAX1 peptide its intracellular expression will be inhibited. In fact, this phenomena has been reported for cell-penetrating peptide-DNA complexes [33–36], where the cationic charge of the peptide must be optimized to allow initial DNA binding and intracellular delivery, but also to permit disassociation of the DNA for its ultimate expression. In the case of the gels studied here, the lesser-charged peptides, MAX8 and HLT2, may form gels that strike a balance to allow both DNA retention in the bulk gel network, and dissociation when internalized by cells.

Figure 3:

Lymphoproliferative capacity and bioresorption of DNA-loaded gels. (A) Single-cell suspensions of the draining (inguinal) lymph nodes were prepared on day 28 after mice were primed with MAX1, MAX8 and HLT2 peptide hydrogels, with or without encapsulated DNA(TA), and boosted on day 8. Lymph nodes were removed and cells cultured for 5 days in complete medium and pulsed during the last 18 hours before measuring [³H TdR] incorporation. Cell proliferation is shown as mean cpm ± sem. (B & C) H&E stained sections of residual DNA(TA) loaded (B) MAX8 or (C) HLT2 peptide hydrogels 28days after implantation. MAX8 hydrogels were poorly degraded with large fragments of gel remaining (red). DNA-loaded HLT2 gels were nearly completely readsorbed on day 28, with extensive infiltration of mononuclear cells (purple). Histology images shown at 40x objective (scale bar = 50 μm).

Histology was used to assess the bioresorption potential of the MAX8 and HLT2 DNA-loaded gels. Fig. 3B shows that after 28 days, the MAX8 gel persists at the injection site. The large red fragments are H&E stained gel, where the apparent fragmentation of the gel is most likely due to the dehydration step of sample fixation and processing. In contrast, panel C shows that the HLT2 gel has been fully resorbed giving way to neotissue primarily consisting of collagen (stained pink). These images also show the possible presence of infiltrating cells. The representative image in panel B shows that the MAX8 gel contains few infiltrating cells, in contrast to the HLT2 gel shown in panel C. The fact that the HLT2 gel, loaded with DNA, showed a lymphoproliferative effect, the presence of a significant number of infiltrating cells, and a bioresorption rate conducive to multiple boosting regiments, warranted its further study.

3.3. Kinetics of Inflammatory Cell Infiltration and Bioresorption of DNA-Loaded HLT2 Peptide Hydrogels

The initial assessment of cellular infiltration into HLT2 gels shown in Fig. 3 was made at day 28. To gain a better understanding of the time-dependent migration of cells within the gel we analyzed implants at 7 and 21 days by histology. For these experiments, HLT2 gels were prepared with two-fold the amount of encapsulated HMGN1-gp100 encoding DNA(TA) in half the gel volume, compared to the formulations used in the previous experiments. This was done in order to deliver a higher concentration of the vector and elicit a greater immune response. Furthermore, mice were boosted twice weekly after the priming dose, and the immune response measured one week after the final boost.

Fig. 4, panels A & B, show that the HLT2 gel remains intact at the implant site 7 days after administration. Importantly, there is considerable infiltration of polymorphonuclear cells indicative of an acute inflammatory response at the injection site. Analogous experiments with gel only and gel with empty vector DNA(−) showed a similar response, Supplementary Fig. 2. This suggests that the peptide gels themselves elicit an initial foreign body response soon after subcutaneous administration. We next probed for the presence of macrophages, as well as B and T cell lymphocytes, in the implanted gels 7 days after administration by performing immunohistochemical staining. Panel C shows a paucity of the F4/80 mouse specific macrophage marker, which stains brown, in the gels indicating that macrophages are not yet present at day 7. This is consistent with the observation of limited bioresorption at day 7, which in part, is mediated by these cells. A similar lack of infiltrating macrophages was observed in the controls (gel alone and gel+DNA(−)), Supplementary Fig. 3, A & C. Likewise, at day 7 no significant presence of B or T cells were observed in the HLT2 gels with or without DNA (Supplementary Fig. 4).

Figure 4:

Early polymorphonuclear cell infiltration into DNA-loaded HLT2 peptide hydrogels. Histological sections from mice receiving DNA(TA) loaded HLT2 peptide gels removed 7 days after implantation into the flanks. H&E stained images are displayed in (A) 10x magnification (scale bar = 100 μm), with selected region magnified in (B) 40x (scale bar = 50 μm). Polymorphonuclear cells are stained purple, whereas the gel appears red. (C) F4/80 staining of DNA(TA) loaded HLT2 gel sections (10x magnification, scale bar = 100 μm), where cell nuclei appear blue and macrophages, if present, would stain brown.

At 21 days, infiltrating polymorphonuclear cells were replaced by mononuclear cells, as shown in Fig. 5, panels A and B. Here, histology images indicate that a large portion of the implanted gel has been resorbed 21 days after implantation, with the original site now being occupied by a considerable population of mononuclear leukocytes. Controls of gel alone, or gel encapsulating DNA(−), showed a similar cellular infiltration profile, Supplementary Fig. 5. Importantly, limited formation of collagen-associated scar tissue is observed in these images, indicating the implant is replaced by fibroblast and collagenous tissue following its resorption. Furthermore, at 21 days the majority of cells in the residual gel network stain positive for F4/80 (Fig. 5C) indicating the injection site is extensively populated by macrophages. The presence of macrophages was also observed for gel alone, or gel encapsulating DNA(−), Supplementary Fig. 3, B & D. These cells are responsible for both clearance of foreign debris (i.e. peptide gel) and, together with dendritic APCs, are critical to the processing and presentation of antigens to lymphocytes in order to initiate an adaptive immune response. This demonstrates that acute inflammation has given way to the chronic inflammation associated with the foreign body response, presumably facilitating the antigen-specific immune response observed in later experiments. We also probed for the presence of B and T cell lymphocytes within the implanted gels at the 21 day time point. Circulating B cells can be activated by direct binding of antigens at the implant site leading to one mechanism by which antigen-specific antibodies can be generated. In addition, T cell lymphocytes play multiple roles in cell-mediated humoral and cytotoxic immune activities. As was the case at day 7, B and T cells were not present to any significant amount at day 21, Supplementary Fig. 6. To investigate the influence of encapsulated DNA(TA) on monocyte-to-macrophage differentiation, the ratio of macrophage to total cell nuclei was calculated for mice receiving the gel implant alone and gels encapsulating either DNA(TA) or the empty control vector DNA(−). This was accomplished by measuring the brown to blue staining ratio from a series of images taken of tissue sections from mice 21 days after priming with the gels. The raw images of representative tissue sections are provided in Supplementary Fig. 3B, Supplementary Fig. 3D, and Fig. 5C for gel alone, gel+DNA(−) and gel+DNA(TA), respectively. All the samples showed the presence of macrophages. However, Fig. 5D shows that gel containing DNA(TA) have a >4 fold enhancement of macrophages with respect to total cell number compared to gel alone or gel containing the empty vector. This suggests that the DNA(TA) vector is enhancing macrophage development.

Figure 5:

Macrophage infiltration into HLT2 peptide hydrogels. (A-C) Histological sections from mice receiving DNA(TA) loaded HLT2 peptide gels removed 21 days after implantation into the flanks. H&E stained images are displayed at (A) 10x magnification (scale bar = 100 μm), with selected regions magnified to (B) 40x (scale bar = 50 μm). Results indicate a marked accumulation of mononuclear cells at the site of implantation 21 days post administration, with only residual fragments of gel remaining. (C) F4/80 staining of these gel sections confirms infiltrating cells are largely macrophages (10x magnification, scale bar = 100 μm). (D) Increase in the ratio of F4/80 to nuclei staining for DNA(TA) loaded HLT2 hydrogels 21 days after implantation compared to sections from mice receiving gels prepared with the control DNA(−) or no vector (** p < 0.01).

To determine if the gel implant is replaced by host cells following its bioresorption, the injection site from mice receiving DNA(TA)-loaded HLT2 gels was stained for the presence of endothelial cells 21 days after priming. These cells are critical to the expansion of vascular networks into implanted materials, which promotes the clearance and removal of the foreign debris and allow for restoration of the local tissue. Tissue sections from the injection site were treated with the endothelial cell-specific CD34 antibody, which stains brown. The image in Fig. 6 demonstrates that bioresportion of HLT2 gels leads to extensive replacement of the gel material by endothelial cells, in addition to the marked infiltration of monocytes observed in Fig. 5C. Moreover, implanted gels which did not contain loaded DNA (Supplementary Fig. 7A), or gels encapsulating the empty control vector DNA(−) (Supplementary Fig. 7B), all have extensive presence of endothelial cells at the site of injection. These results suggest that implanted HLT2 gels could become well vascularized over time, facilitating the observed infiltration of inflammatory cells into the gel network and subsequent material resorption.

Figure 6:

Presence of endothelial cells at injection site for DNA-loaded HLT2 peptide hydrogels. CD34 stained image from the DNA(TA) containing HLT2 gels removed 21 days after implantation showing extensive replacement of the gel mass by endothelial cells (brown) (10x magnification; scale bar = 100 μm).

3.4. Assessment of Adaptive Immune Response to DNA-Loaded Hydrogels

The ability of the DNA(TA)-loaded HLT2 gel to stimulate an adaptive response was investigated by measuring lymph node activity. First, lymph node size was approximated by counting the total population of cells isolated from the draining lymph nodes. Lymph nodes were isolated from mice receiving naked DNA, gel alone and gels containing DNA(−) or DNA(TA) after the priming and boost regiment. Fig. 7A shows that lymph nodes of mice receiving HLT2 gels that contain DNA(TA) are significantly larger than mice receiving DNA in saline or gel alone. On average, the lymph nodes of HLT2-DNA(TA) mice were larger than those receiving DNA(TA) via electroporation or the empty vector in gel. Lymphocyte activity was also assessed by measuring the ex vivo proliferation of stimulated cells isolated from lymphoid tissue. Fig. 7B shows a >3 fold enhancement in lymphoproliferation at day 21 after priming of mice with gel loaded with DNA(TA) as compared to DNA(−) or no DNA. Taken together, the data in panel A and B show that gel alone and DNA(TA) alone weakly stimulate lymph node cell proliferative activity. However, DNA(TA) encapsulated in the HLT2 gel significantly stimulates lymphoproliferation.

Figure 7:

Cellular and humoral immune response to DNA-loaded HLT2 peptide hydrogels. Mice were primed on day 1 and boosted twice (day 8 and 15) with the different formulations, followed by measuring immune response on day 21. (A) Total cell numbers from draining (inguinal) lymph nodes of mice treated with naked DNA(TA), administered as a bolus dose in saline (sa.) or via electroporation (el.), as well as HLT2 peptide hydrogels with and without encapsulated DNA (** p < 0.01, *** p < 0.001). (B) Ex vivo T cell proliferation of cultured single-cell suspensions from the draining (inguinal) lymph nodes. After removal lymphocytes were cultured for 5 days in complete medium and pulsed during the last 18 hours with tritiated thymidine [³H TdR] before assaying for radioactivity. Results shown as mean cpm ± sem (*** p < 0.001). (C) Percentage of CD4+/IFNγ+ cells detected from the draining (inguinal) lymph nodes of mice treated with DNA(TA) alone or encapsulated within HLT2 peptide hydrogels (** p < 0.01). (D) Increase in optical absorbance (λ = 450 nm) representing antibody titers specific for the tumor-antigen expression plasmid (gp100-specific) (* p < 0.05). Sera from mice were pooled 21 days after priming with naked DNA, or HLT2 peptide gel formulations encapsulating the control DNA(−) or treatment DNA(TA) vector, and subjected to a gp100-specific binding ELISA assay.

Next, the capacity of the DNA(TA)-loaded HLT2 hydrogels to induce a cytotoxic T cell-mediated antitumor response was assessed. Here, the tumor killing activity of activated lymphocytes was measured by co-incubation of B16F1 melanoma target cells with single-cell suspensions prepared from spleen and lymph node tissue isolated from mice receiving DNA(TA)-loaded gels (Fig. S8). These experiments showed no significant killing of B16F1 cells at both 7 and 21 days after priming, suggesting minimal activation of CD8+ T cells.

Due to the known ability of the alarmin portion of the HMGN1-gp100 fusion to induce Th1-mediated activity, the ability of the DNA(TA)-loaded gel to facilitate an immune response mediated by CD4+/IFNγ+ expressing Th1 cells was investigated. Fig. 7C shows that the percentage of CD4+/IFNγ+ cells isolated from the draining lymph nodes was considerably higher in mice receiving implants of HLT2 gels encapsulating the DNA(TA) vector compared to mice receiving naked DNA administered in saline or via electroporation. We next investigated whether this cellular response led to the production of gp100-specific antibodies detectable in mouse sera. Fig. 7D shows the total antigen-specific antibody levels in the serum at day 21 measured by ELISA. Results show that an increase in gp100-specific antibodies was present in the sera of mice immunized with the HLT2 gels incorporating the DNA(TA) vector, compared to gels without DNA or containing the empty control vector. It should be noted that we previously observed, through histological analysis, a limited presence of B cells into the implanted material at both 7 and 21 days. This indicates that the antigenic protein expressed at the injection site is being processed and presented to B cells located within the draining lymph tissue leading to the observed humoral response, rather than due to infiltration of circulating B cells into the implanted material. The magnitude of this humoral response was estimated by calculating antibody titers at three-weeks. Using a second order polynomial, constrained by a lower limit of 0.08 OD for the background of the assay, nonlinear regression analysis of our data showed the antibody titer for control animals receiving gel alone to be 7,077, while that of animals receiving HLT2 gels encapsulating the DNA(TA) vector to be 10,965 (p = 0.0001). The titer level generated by the HLT2-DNA(TA) system was compared to the activity of other peptide-based immunostimulatory materials recently reported. For example, peptide nanofibers condensed with DNA for HIV immunotherapy achieved four week titer levels in mice that were on the order of 10⁴ [37]. Although not utilized to deliver DNA, the Collier lab have designed clever self-assembling fibrillar systems capable of presenting ovalbumin [38, 39] and malaria-associated [40] antigens that generate similar titer values over the same time period. Taken together, the data indicates that the DNA-loaded HLT2 gel stimulates a humoral response mediated by CD4+/IFNγ+ expressing Th1 cells.

3. Conclusion

Antigen expressing DNA vectors represent potential immunostimulators capable of producing specific humoral and cytotoxic immune responses. However, delivery of DNA in vivo is complicated by rapid enzymatic degradation and mechanical fragmentation from high shear stresses applied during injection. Herein, we studied the immunostimulatory effects of peptide-based hydrogels utilized to encapsulate and deliver DNA by simple syringe injection. We show that hydrogels formed from cationic peptides are capable of encapsulating and retaining large quantities of DNA(TA), a plasmid that encodes a protein fusion composed of the melanoma-specific tumor antigen gp100 and the adjuvant HMGN1 (high-mobility group binding protein 1). The effect of network cationic charge on the gel’s bioresorptive properties and their ability to encapsulate and retain the DNA was assessed employing a family of three gel-forming peptides that systematically vary in their formal charge. The peptide HLT2, having a formal charge of +5 at neutral pH, is capable of encapsulating microgram quantities of DNA with little effect on its rheological properties, allowing its effective syringe-based delivery in vivo. The DNA-loaded HTL2 gel is resorbed over a time regime amenable to multiple boosting regiments. Injection of the DNA-loaded gel results in an initial foreign body response that promotes the infiltration of inflammatory cells, and thereby facilitates DNA expression and antigen processing. Histology showed that at day 7, the DNA-loaded gel was populated primarily by polymorphonuclear cells, followed by a significant infiltration of macrophages at day 21. At this time lymphoproliferative responses within the draining lymph nodes was significantly increased for mice receiving HLT2 gel implants loaded with the therapeutic DNA(TA) vector compared to controls. Immunologic experiments demonstrated that mice receiving DNA(TA)-loaded gels produced a cellular immune response mediated by CD4+/IFNγ+ expressing Th1 cells, complemented by the formation of gp100-specific antibodies. Based on these results, DNA-loaded self-assembled peptide hydrogels show promise as immunostimulatory materials. However, further studies are required to explore whether this approach is more effective for DNA delivery compared to gene gun and electroporation techniques. Nevertheless, with further refinement, nucleic acid-loaded peptide gels may be useful in vaccine development, gene therapy, and RNA silencing applications, to name a few. Future studies will be focused on formulations that include chemical and peptidic adjuvants to further control the immune response to these materials.