In Vitro Analysis of Microneedle Patches for Transdermal Drug Delivery: A Minimally Invasive Route to Systemic Circulation

Abstract

Nanoparticles offer great potential as drug delivery vehicles but conveniently administering them systemically at therapeutically relevant doses remains a challenge. To date, systemic delivery of nanoparticles is achieved through intravenous injection or oral administration; however, each of these approaches are limited by shortcomings (e.g., painful delivery, low bioavailability). In the present study, we explore the development and in vitro testing of alginate-based microneedle patches and their potential to deliver various nanoparticles. We demonstrate that therapeutics encapsulated in these MNPs effectively release from the polymer matrix, and that released therapeutics retain their bioactivity and can be taken up by cells of interest. Here, curcumin released from alginate MNPs retained its antimicrobial activity, and nanogels were uptaken by both cancerous and normal breast epithelial cells. When antibodies were conjugated to the nanogel surface, they remained functional for efficient ligand-based targeting. Finally, released extracellular vesicles were successfully uptaken by fibroblasts. Taken together, these results demonstrate the payload versatility of this system. Furthermore, we show that these microneedle patches can successfully penetrate the epidermis of skin ex vivo, allowing encapsulated nanoparticles to permeate into deeper tissue. Overall, these studies highlight the capacity of microneedle patches to release functional nanoparticles in vitro, suggesting the potential of this system to enhance the bioavailability of therapeutics in a minimally invasive and self-administered fashion.

1. INTRODUCTION

Traditional drug delivery methods such as oral and intravenous delivery have historically been the standard for achieving systemic circulation of drugs and nanoparticles¹; however, these administration methods for nanoparticles are challenging². For example, oral drug delivery may be preferred by patients due to its convenience, low cost, and ease of administration, but many nanoparticles that are delivered orally face reduced bioavailability due to premature degradation in the harsh gastrointestinal tract^1,3–5 making drug delivery rates and dosing difficult⁵. While intravenous delivery allows for faster onset of the therapeutic effect and precise dosing, the pain associated with this administration route – and the requirement for a healthcare professional to administer it – reduces patient compliance⁶. As a result, the investigation of transdermal drug delivery systems like microneedle patches (MNPs) present an opportunity to overcome the oral and intravenous limitations by delivering nanoparticles through the skin, facilitating treatment in a minimally invasive and potentially self-administered approach.

Nanoparticles have greatly expanded the potential of drug delivery by presenting an opportunity to modify the properties of drugs they encapsulate, including solubility, diffusivity, and half-life⁵. Nanoparticles increase the stability of the drugs within them, leading to enhanced bioavailability and improved therapeutic efficacy by protecting drugs from degradation⁷. Additionally, they offer the potential to co-deliver multiple drugs⁸ and target cells of interest, effectively reducing toxicity and side effects^7,9–11. Few drugs can successfully pass through the skin to achieve systemic circulation¹² due to the limited permeability of the lipophilic matrix in the stratum corneum¹³ to hydrophilic and high molecular weight drugs^12,14. In recent years, microneedle patches (MNPs) have emerged as a transdermal drug delivery system that serves to increase the skin’s permeability to drugs¹⁵ by directly bypassing the stratum corneum—the outermost layer of the skin. Using an array of hundreds of micron-sized needles, typically ranging from 150 μm - 1500 μm in length, MNPs temporarily disrupt the stratum corneum, creating channels to facilitate drug delivery into the bloodstream¹⁶. The temporary channels created by MNPs enable the drug to be delivered directly into the epidermis or upper dermis, facilitating uptake into the systemic circulation while bypassing the pain associated with needle injections.

While microneedles have been employed to successfully deliver small molecules, proteins, and nucleic acids, we have found few emerging examples of nanoparticle delivery from MNPs^17–20. Initially, we used a small molecule to develop and characterize our MNPs, followed by exploring the incorporation of a biologically-native nanoparticle, extracellular vesicles (EVs), followed by engineered nanoparticles, nanogels (NGs), and antibody-nanogel conjugates (ANCs), which are a highly versatile polymeric drug delivery platform that can carry, protect, target, and deliver encapsulated cargoes in a spatially and temporally controlled manner²¹. Our alginate MNP was designed to rapidly release functional nanoparticles in vitro and across the stratum corneum ex vivo, demonstrating the potential of this system to deliver a wide variety of nanoparticles that are poised to improve and optimize drug delivery for a multitude of diseases.

2. MATERIALS AND METHODS

2.1. Preparation of alginate solution for microneedle fabrication:

Medium viscosity alginate (Sigma Adrich, A2033) was used to study the respective release kinetics and in vitro bioactivities of the following therapeutics released from MNPs: curcumin (CUR), extracellular vesicles (EVs), nanogels (NGs), and antibody-nanogel conjugates (ANCs). Stock solutions of the therapeutics (also referred to as “free” therapeutic solutions) were prepared as follows:

Curcumin:

5 mg/mL of curcumin (Sigma Aldrich, C1386-5G) was dissolved in NaOH for 2 hours or until a homogenous was obtained.

Extracellular Vesicles:

EVs were isolated from bovine milk using differential centrifugation. Briefly, pasteurized milk was clarified by centrifuging twice at 2,500 x g to remove larger molecules. The resulting supernatant was treated with 500 mM EDTA to remove excess milk proteins. Following, supernatant was centrifuged at 12 k, 30 k, and 70 k x g for 60 minutes, and filtered through a 0.22 μm vacuum filter system. The filtrate was centrifuged at 100 k x g to pellet EVs and the remaining supernatant was removed. The final EV concentration (4.485 mg/mL) was determined using a microBCA assay per manufacturer’s instructions, and the EVs were resuspended in PBS to a working concentration of 4.84 μg/mL. The isolated EVs were characterized by Dynamic Light Scattering (DLS) (Figure S4).

NGs and ANCs:

NGs were prepared as previously described²². Briefly, stock solutions of methoxy terminated and tetrazine terminated PEG-PDS-BCP (polyethylene glycol–pyridyl disulfide ethyl methacrylate–block copolymer), abbreviated as OMe-BCP and Tz-BCP, were prepared in DMSO (100 mg/mL). Two types of NGs were prepared: DiI-encapsulated NGs and Cy5.5 labelled NGs. DiI-encapsulated NGs were prepared by mixing 10 μL each of OMe-BCP and Tz-BCP (1:1 ratio) to 10 μL of the DiI stock solution (20 mg/mL – 10% loading). To produce Cy5.5-labeled NGs, OMe-BCP, Tz-BCP, and Cy5.5-BCP were prepared in DMSO at a ratio of 4:5:1 respectively. To this polymer-dye mixture, 1 mL of PBS was added, and the solution was sonicated until clear. Each type of NGs were crosslinked using 20% of DTT (dithiothreitol) as a reducing agent for 3 hours, followed by dialyzing overnight, and final polymer concentration of NG was brought to 2 mg/mL. For the ANCs, trastuzumab antibody was modified with TCO-PEG4-NHS ester linker and then mixed with the NGs in a 1:2 weight ratio with respect to the polymer concentration in the NGs. This mixture was left to incubate at 4°C for 24 hours. The concentration of the ANC was calculated accordingly (1.68 mg/mL). The obtained NG and ANC were characterized by Dynamic Light Scattering (Figure S5).

The final concentrations of therapeutics in each alginate solution were: 5 mg/ml for curcumin; 4.84 μg/ml for EVs; 2 mg/ml for NGs; and 1.68 mg/ml for ANCs. Following the preparation of these solutions, each 2% w/v alginate was added, and the solutions were spun until homogeneous. As a negative control, blank alginate solutions were prepared by adding 2% w/v alginate to PBS; this solution was also used as the backing of the MNP.

2.2. Fabrication of Rapid Release Microneedles:

Following fabrication of alginate polymer solutions, approximately 50 mg of the alginate was placed in silicone MNP molds (Micropoint Technologies, ST-06). The final quantities of therapeutics in each MNP were: 0.25 mg for curcumin; 242 ng for EVs; 100 μg for NGs; and 84 μg for ANCs. The molds were then placed in flat bottom centrifuge tubes and centrifuged at 3000 x g for 15 minutes to allow the alginate solution to fill the needles. The MNPs were dried at room temperature before adding 50 mg of blank alginate to create the backing of the patch. The centrifugation and drying process was repeated, and the final microneedle arrays were removed from the molds using fine-tip tweezers and stored needle side up. The resulting MNPs are 8 mm by 8 mm and have an array of 15 × 15 needles (225 needles total). Each needle is 600 μm long with a base width of 200 μm. Between each needle base, there is a 500 μm gap.

2.3. Ex vivo Skin Penetration:

Samples used for ex vivo analyses were from carcasses used in animal studies that were approved by University of Massachusetts Amherst Institutional Animal Care and Use Committee. Ex vivo releases were performed on rat skin to assess the efficacy of microneedle penetration into the skin and to observe the permeation of therapeutics across the epidermis. Briefly, skin from rat cadavers was harvested and shaved. Fur was completely cleared from the treatment area using a depilatory cream (e.g. Nair). MNPs were then administered using a spring-loaded microneedle applicator device (Micropoint Technologies, SA-01; see Figure S1). After 60 minutes, the patch was removed, and the skin was imaged. Images were taken both top-down and cross-sectionally by slicing the skin perpendicular to the needle puncture points.

2.4. Release kinetics:

The release profile of each therapeutic was investigated in vitro. Fabricated MNPs were placed in a 2 mL solution of 1X PBS. At designated time points (1, 5, 10, 15, 30, 45, 60 minutes), 300 μL of the release medium was sampled and the solution was replenished with an equal amount of fresh PBS to maintain a constant volume. The 60-minute endpoint was chosen based on our observations of the system and release data that plateaued by that timepoint. The amount of drug released at each time interval was quantified using a spectrophotometer by comparing the absorbance of the samples to that of a standard curve.

2.5. Cell culture:

BT474 cells (epithelial cells from an invasive ductal breast carcinoma with a high expression of human epidermal growth factor receptor 2, HER2^high) were cultured in Dulbecco’s Modified Eagle Medium (DMEM, Gibco), supplemented with 10% fetal bovine serum (FBS) and 1% Penicillin/Streptomycin (Pen/Strep). Experiments were performed on BT474 cells from passages 10–14. MCF10A cells (non-tumorigenic breast epithelial cells, with low expression of HER2 (HER2^low)) were grown in a 1:1 mixture of Dulbecco’s Modified Eagle’s Medium (DMEM, Gibco) and Ham’s F-12 Nutrient Mixture (Gibco), supplemented with 5% FBS, 1% Pen/Strep, 10 mM HEPES (Gibco), 20 ng/mL insulin, 1 μg/mL hydrocortisone, 1 μg/mL EGF (human), and 200 ng/mL cholera enterotoxin. Experiments were performed on MCF10A cells on passage 13–14. BJ fibroblasts (healthy neonatal foreskin fibroblasts) were cultured in Eagle’s Minimum Essential Medium (EMEM, Gibco), supplemented with 10% FBS and 1% Pen/Strep. Experiments were performed on BJ fibroblasts on passage 5. All cells were incubated at 37 °C and 5% CO₂ with medium changes every 2–3 days.

2.6. Functional Effects of Released Therapeutic

2.6.1. Bacteria biofilm treatment:

DH5α competent E. coli bacterial cells were cultured in LB broth overnight. Bacterial cells were collected and plated at an optical density of 0.8. The resulting bacterial solution was plated in a 96 well plate and allowed to form biofilms (18–24 hours). Curcumin MNPs were fabricated prior, and a 60-minute in vitro release assay was performed in 1 mL PBS. Following, 100 μL of the releasate was plated on the biofilms, along with positive and negative controls: free curcumin at the same concentration as the MNP acted as a positive control, and a blank MNP and PBS (no treatment) acted as negative controls. Each treatment group was placed on the biofilms and incubated at room temperature for 1 hour. Immediately following treatment, 10 μL of alamarBlue was added to each well and incubated at room temperature for 1 hour to quantify metabolic activity of the bacteria post-treatment. Each treatment group was then replated in a black well plate, and fluorescence intensity was read using a spectrophotometer as a measure of cell viability. Additionally, brightfield images of the biofilms were taken using an EVOS fluorescence imaging microscope.

2.6.2. Nanoparticle Uptake and Fluorescence Imaging

EVs:

Fluorescence imaging was performed to observe uptake of fluorescently labeled EVs on fibroblasts. Two days prior to treatments, BJ fibroblasts were seeded at 10K cells/well in a 48-well plate. As previously mentioned, EVs were isolated in house from bovine milk through repeated differential centrifugation and filtration steps. EV MNPs (final concentration 4.84 μg/mL) were fabricated, and a 60-minute in vitro release assay was performed in 1 mL PBS. For labeled EVs, a CellTrace^™ CFSE Cell Proliferation Kit was used according to manufactures instructions. Following the in vitro release of EVs, 1 μl of CFSE was added to 300 μl aliquots of EVs and solutions were incubated for 2 hours with mixing every 30 minutes to ensure sufficient fluorescent labeling of released EVs. To remove excess CFSE dye, the labeled EV solution was transferred to Amicon Ultra 10K MWCO filter tubes and the solution was centrifuged for 14,000 x g for 20 min. Labeled EVs were collected in a clean tube and the total volume of filtrate was plated on the BJ fibroblasts, incubating for 24 hours to allow for EV uptake. CFSE fluorescence only occurs once internalized by the cell; to confirm the functionality of the CSFE dye, the dye was directly added to cells as a positive control. Unlabeled MNP-released EVs and PBS were used as negative controls. Cells were then washed with PBS, fixed in 4% paraformaldehyde (PFA) for 10 minutes, permeabilized with Triton-X for 5 minutes, and incubated with Phalloidin 647 for 1 hour. The nuclei were then stained with Hoechst 33342 (2 μg/mL). Finally, cells were imaged using an EVOS fluorescence imaging microscope.

NGs & ANCs:

Fluorescence imaging was performed to observe cellular uptake of DiI and Cy5.5-labeled NGs/ANCs released from MNPs. One day prior to treatments, BT474 (HER2^high) and MCF10A (HER2^low) cells were seeded at 10K cells/well in a 48 well plate. An in vitro release was performed by placing NG and ANC MNPs in 1 mL of cell media, allowing release for 60 minutes. Blank alginate MNPs acted as a negative control. Standard curves of the stock NG (2 mg/mL) and ANC (1.68 mg/mL) solutions were created in respective cell media and used to determine the concentration of the MNP-released NGs/ANCs after 60 minutes. The stock solutions were then diluted ensuring that positive controls and experimental groups received equal concentrations of NGs/ANCs. Cells were washed with PBS and then treated with each group. After a 4-hour incubation, treatments were removed, wells were washed 3 times with 1X PBS, and the nuclei were stained with Hoechst 33342 at 2 μg/mL. Cells were then imaged using a KEYENCE BZ-X800 fluorescence microscope.

2.6.3. Flow cytometry

Flow cytometry was performed to quantify cellular uptake of Cy5.5-labeled NGs/ANCs released from alginate MNPs. One day prior to treatments, BT474 and MCF10A cells were seeded at 100K cells/well in a 48 well plate. The procedure for in vitro release was repeated as stated above, incubating cells with treatments for both 1 and 4 hours to examine the time-dependency of uptake of HER2 targeting ANCs (via anti-HER2 monoclonal antibody, trastuzumab) on both HER2-overexpressing (BT474) and HER2-low expressing (MCF10A) breast cell lines, as compared to uptake of non-targeting NGs. Following treatment, the cells were washed and lifted with 100 μl trypsin/well and pelleted by centrifugation at 750 x g for 7 minutes and resuspended in 500 μl of FACS buffer (10% FBS in PBS). Flow cytometry was then performed using a Cytek Aurora flow cytometer. Live cells were first distinguished based on their forward/side scatter properties. Next, untreated cells were used to set gating regions to discern Cy5.5 positive from negative cells and distinguish any background variation. Cells treated with Cy5.5-labeled NG/ANC stock solutions were used as a reference for Cy5.5 positive gating.

2.7. Statistical Analysis

Data are represented as mean ± standard error of the mean (SEM) using GraphPad (Prism 10.0). For the determination of statistical significance, one-way or two-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test were used, as applicable. P-values < 0.05 were determined as significant and denoted with a single asterisk (*).

3. RESULTS

3.1. Therapeutic-loaded microneedle patches were successfully fabricated with a wide range of therapeutics.

MNPs were successfully fabricated for each therapeutic studied (Figure 1). Employing a two-cast approach with a blank backing layer, needles were consistently well-formed using silicone microneedle template molds. Macroscopic images show the fully formed microneedle arrays for each therapeutic (Figure 1B). We also used SEM imaging to show the square pyramidal architecture of the polymeric microneedles as formed by the molds (Figure 1C). Following removal from skin, we confirmed that the microneedles remain mostly intact during application (Figure S2)

Figure 1: Microneedle patches were loaded with therapeutics for transdermal drug delivery.

(A) Microneedle patch dimensions obtained via silicone 15×15 Mpatch microneedle array template. Created with Biorender.com. (B) Macroscopic images of microneedle arrays fabricated from alginate solutions loaded with therapeutics of interest. (C) SEM images of microneedle arrays fabricated with a blank alginate solution using micropoint silicone microneedle template. Scale bars 400 μm and 200 μm (left and right image, respectively).

3.2. MNP-released curcumin effectively reduced E. coli bacterial biofilms.

Prior to investigating the application of nanoparticles within our MNP delivery system, curcumin was employed as an accessible model to examine the functional effects of released therapeutics in vitro. Curcumin rapidly released from the patches over a period of 60 mins, with a max release = 53.4 ± 13.6% at 60 mins (Figure 2A). An initial burst release of the small molecule was observed due to the swelling of the alginate MNP in solution and hydrophobic nature of the drug; ~42% of the encapsulated curcumin was released from the polymer matrix after just one minute.

Figure 2: Curcumin released from alginate MNPs reduced E. coli biofilms.

(A) In vitro cumulative release profile for curcumin from alginate MNP over 60 mins (n=6). (B) Images of in vitro curcumin release over time. Red arrow shows alginate MNP after the curcumin has been released from the polymer matrix. (C) Ex vivo release of curcumin on rat skin (top-down), and (D) in cross-section. (E) alamarBlue was used to quantify relative bacterial metabolism following treatment of E. coli biofilms with free curcumin versus MNP-released curcumin (n=9; mean ± SEM; one-way ANOVA with Tukey’s post hoc test; * = p < 0.05).

Macroscopically, it was observed that the structure of the alginate MNP was preserved in solution after 60 minutes of release (Figure 2B). An ex vivo penetration test was performed on rat skin to demonstrate the ability of our MNP to achieve successful permeation of curcumin into the skin. The needles remain intact during penetration, facilitating delivery of curcumin across the epidermis, which was assessed via cross-sectional images of the skin along the microneedle puncture points (Figure 2C, D). To study the antimicrobial bioactivity and functional effects of MNP-released curcumin, we analyzed bacterial viability following the application of MNP-released curcumin to E. coli bacterial biofilms (Figure 2E). E. coli biofilms treated with curcumin successfully resulted in bacterial cell death and efficacy was not significantly different from free curcumin controls. With brightfield imaging, disruption of the biofilms can be observed with increasing concentrations of MNP-released curcumin, further confirming that the antimicrobial effects of curcumin are preserved throughout the process of MNP fabrication and release (Figure S3).

3.3. MNP-released EVs were endocytosed by BJ fibroblasts.

To analyze the bioactivity of MNP-released EVs (characterized by DLS in Figure S4) and their influence on individual cells, we performed a cellular uptake assay using fluorescently labeled EVs on BJ fibroblasts (Figure 3). An in vitro release assay using EV MNPs was performed and released EVs were labeled with CFSE dye (green – 448 nm excitation/530 nm fluorescence). This CFSE dye only fluoresces when it is intracellular, as shown by adding CSFE dye directly to cells (“CFSE Alone (+)”). To ensure that EVs were not auto fluorescent, unlabeled MNP-released EVs (EV MNP (−), i.e., no CFSE) acted as a negative control, along with a PBS (no treatment) group. To confirm the EVs were intracellular, the actin cytoskeleton was also labeled with Phalloidin. After a 24-hour incubation with the treatment groups, cells treated with labeled EVs released from MNPs were effectively uptaken by BJ fibroblasts (74.8 ± 14.9% uptake) as indicated by colocalization of CSFE with the actin cytoskeleton (positive controls with CFSE directly added = 100%). With successful loading, release, and uptake of a naturally occurring biological nanoparticle from our MNPs, we next explored the application of this technology for a therapeutically relevant synthetic nanoparticle.

Figure 3: EVs released from MNPs were effectively taken up by BJ fibroblasts.

(A) When CFSE is added directly to fibroblasts (CFSE Alone (+)), it fluoresces when intracellular. (B) When CFSE-labelled MNP-released EVs (EV MNP(+)) are added to fibroblasts, the CFSE signal colocalized with the nuclei and actin cytoskeleton, demonstrating successful uptake of released EVs. Unlabeled EVs and PBS did not show any fluorescence. Blue, cell nuclei. Green, CFSE. Magenta, actin filaments.

3.4. Nanogels and antibody-nanogel conjugates released from MNPs retained their bioactivity and functional effects and were uptaken by both cancerous and normal breast cells in vitro.

We next wanted to test our previously developed nanogels (NGs) and anti-HER2 antibody-nanogel conjugates (ANCs) to explore the potential of this system to deliver these nanoparticles for breast cancer treatment (see Figure S5 for characterization). An in vitro release assay using DiI-labeled NG alginate MNPs revealed that, unlike curcumin, NGs release gradually over 60 minutes, achieving a maximum release of 78.0 ± 12.1% (Figure 4A, B). Since the NGs are larger than curcumin particles and consist of a hydrophilic shell around a hydrophobic core, the burst release observed for curcumin is not observed for NGs. In addition, an ex vivo permeation test using alginate MNPs showed effective delivery of DiI-labeled NGs across the outermost layer of skin, which appears to adequately permeate into deeper epidermal skin layers (Figure 4C, D). Finally, we wanted to confirm the uptake of the nanogels/DiI in cells in vitro. We observed comparable uptake for free NG (i.e., non-MNP loaded) and MNP-released NGs (Fig. 4E, F), with a slight reduction observed for the MNP-released nanogels.

Figure 4: Nanogels were released in vitro and across the epidermis of rat skin ex vivo.

(A) In vitro release profile for cumulative release of NGs from alginate MNP over time (n=12). (B) Images of in vitro NG release over time. (C) Ex vivo release of NGs on rat skin (top-down). Scale bars 500 μm. (D) Ex vivo release of NGs on rat skin (cross-section). Scale bar 500 μm. (E) BT474 uptake of non-covalently bound DiI-NG (n=3; mean ± SEM; one-way ANOVA with Tukey’s post hoc test; * = p < 0.05). (F) Uptake of non-covalently bound DiI-NG. Scale bars 100 μm.

After demonstrating successful release of NGs both in vitro and ex vivo, and the uptake of dye, we sought to confirm the maintained bioactivity of these nanoparticles, including their capacity for cellular internalization, which is often required for the delivery of therapeutic payloads from these nanoparticles. Since the DiI is not covalently bound to the nanogels, we aimed to confirm that our previous observations were not due to an uptake of leached dye by switching to covalently labelled Cy5.5 NGs. We examined uptake of these nanogels in both cancerous and normal breast epithelial cell lines, namely BT474 (human ductal carcinoma) and MCF10A (human breast epithelial), respectively (Figure 5). Following a 4-hour incubation with MNP-released Cy5.5-labeled NGs, fluorescence imaging indicated that Cy5.5 and cell nuclei colocalization was present for all NG groups, with uptake appearing to be higher for free NG solutions on both cancerous (BT474) and normal (MCF10A) breast cells compared to MNP-released NGs. Flow cytometry confirmed these observations quantitatively: we found that the internalization of free NGs on BT474 cells was approximately 4 times higher than MNP-released NGs (Figure 5A); this discrepancy was not observed on MCF10A cells, with these cells having equivalent uptake between free and MNP-released NGs (Figure 5B). These results suggest that MNPs can be used to release functional NGs that are capable of uptake by cells of interest for effective drug delivery. Once we observed cellular internalization of NGs, we evaluated the capacity of our MNPs to release antibody conjugated (anti-HER2) nanogels (ANCs) on these breast epithelial cells that have varied levels of HER2 expression.

Figure 5: Fluorescence microscopy and flow cytometry revealed uptake of MNP-released NGs on both cancerous (BT474) and healthy (MCF10A) breast cell lines after a 4-hour incubation.

(A) BT474. (B) MCF10A. Top, cellular uptake of Cy5.5 NGs analyzed by fluorescence microscopy after 4-hour incubation. Scale bars 100 μm. Blue, cell nuclei stained with Hoechst 33342; Yellow, Cy5.5 NG. Bottom, uptake of MNP-released Cy5.5 NGs analyzed by flow cytometry (n=3; mean ± SEM; one-way ANOVA with Tukey’s post hoc test; * = p < 0.05).

The ANCs in this study were conjugated with trastuzumab, an anti-HER2 antibody that allows the ANCs to be preferentially internalized by HER2 overexpressing cells, thereby improving the specificity of cellular uptake to diseased cells²³. MCF10A cells have low levels of HER2 expression (HER2^low), making them a useful control when assessing the targeting effects of anti-HER2 ANCs post-MNP release. The previous study design was repeated using free and MNP-released ANCs, incubating treatments for 4 hours on both BT474 (HER2^high) and MCF10A (HER2^low) cells (Figure 6). Fluorescence microscopy revealed colocalization of Cy5.5-labeled ANCs and cell nuclei among all ANC groups, with uptake appearing to be higher on HER2^high cells versus MCF10A as expected. Flow cytometry quantitatively confirmed this, with uptake of MNP-released ANCs being more than 7 times higher on HER2^high cells versus HER2^low cells (36.2% vs. 4.7%) after 1 hour, and 5 times higher after 4 hours (34.4% vs. 6.9%). Collectively, these data indicate that the conjugated antibody is not compromised during MNP fabrication and that its release retains its bioactivity to selectively target HER2-overexpressing cells.

Figure 6: Fluorescence microscopy and flow cytometry revealed uptake of MNP-released ANCs on both cancerous (BT474) and healthy (MCF10A) breast cell lines after a 4-hour incubation.

(A) BT474 (B) MCF10A Top, cellular uptake of Cy5.5 NGs analyzed by fluorescence microscopy after 4-hour incubation. Scale bars 100 μm. Blue, cell nuclei stained with Hoechst 33342; Yellow, Cy5.5 NG. Bottom, uptake of MNP-released Cy5.5 NGs analyzed by flow cytometry (n=3; mean ± SEM; one-way ANOVA with Tukey’s post hoc test; * = p < 0.05).

Lastly, we replotted the data in Figures 5 and 6 (Figure S6) to compare the effect of antibody conjugation on uptake of the NGs and ANCs for both HER2^high cells and HER2^low cells. In HER2^high cells, the MNP-released ANCs did significantly better than the NG-MNP in the first hour (p < 0.001), and at 4 hours, while the free ANC were significantly better than the NGs at 4 hours. This was not observed for the HER^low MCF10A cells; in fact, at 4 hours, there was a statistically significant decrease for the ANCs versus the NGs. Overall, uptake was higher in the BT474 HER2^high cells for both the NGs and ANCs; however, the difference was much more pronounced for the ANCs.

4. DISCUSSION

The ultimate aim of this work is to overcome the current challenge(s) of systemic nanoparticle delivery when delivered orally (low bioavailability) or intravenously (pain, needle phobia) using microneedle patch technology. In this first study towards this aim, we demonstrated that alginate MNPs encapsulating nanoparticles can rapidly release nanoparticles that maintain their bioactivity in vitro and can deliver nanoparticles across the epidermis ex vivo. These MNPs could release extracellular vesicles, nanogels, and antibody-conjugated nanogels as well as the small molecule, curcumin. Releasing their payload in ~60 minutes makes these MNPs a promising approach for rapid and pain-free nanoparticle delivery through the skin. These MNPs were not crosslinked to allow for swelling and payload release over the 60-minute application time. As they had delivered their payload in 60 minutes and maintained their structure macroscopically, in vivo it is proposed that the microneedles be removed at that timepoint. In addition, imaging of MNPs pre- and post-application confirmed that the needles withstand the force of application to puncture the stratum corneum and deliver cargo across the epidermis.

The bioactivity of released therapeutics was confirmed in vitro using functional and cellular uptake assays. MNP-released curcumin retained its antimicrobial properties to eradicate E. coli bacterial biofilms, while flow cytometry showed that released NGs and ANCs were effectively uptaken by cancerous and healthy breast epithelial cells. Natural EVs are especially promising as biologically derived nanocarriers, offering low-immunogenicity and efficient uptake through natural endocytosis pathways²⁴. MNP-released labeled EVs were confirmed to be uptaken by fibroblasts by colocalization of CSFE – which is non-fluorescent until it enters viable cells – and the actin filaments of the cellular cytoskeleton. Therapeutically, EVs have been explored from a variety of natural cell sources and genetically engineered cells, as well as post-secretory modified EVs for a wide range of therapeutic effects (anti-inflammatory, tissue engineering, antibiotic²⁵). Although EVs have previously been incorporated into MNPs in prior studies²⁶, the majority of these systems do not use aqueous-based materials. Since EVs are highly sensitive any harsh processing such as high shear, temperature, and pH can compromise their membrane, degrading cellular uptake and cargo delivery.^27,28 Our fabrication process should help mitigate some of these concerns as it does not require the use of harsh solvents since alginate is water soluble, and it does not require high temperatures for fabrication. In future studies, we will explore the therapeutic effects of MNP-released EVs in vitro and in vivo, with a particular focus on their role in regenerative medicine.

In this work, we were particularly interested in the delivery of synthetic nanoparticles, namely nanogels²³, for developing more effective breast cancer treatments. Synthetic nanoparticles have been widely explored for their use in cancer treatment; however, their translation has not been smooth^29–31. While challenges remain once the nanoparticles reach the tumor site, increasing their bioavailability via enhanced delivery into systemic circulation greatly enhances the chance of success. Proponents of MNP technology assert that it provides a convenient way to overcome the requirement to survive and exit the gastrointestinal tract (compared with oral delivery)^32,33, and the pain and need for a healthcare professional (compared with intravenous injection)^34,35. MNPs have been explored recently for the delivery of anti-inflammatory²⁰, anti-cancer³⁶, and psoriasis-treating³⁷ nanoparticles. To our knowledge, nobody has used these MNPs to deliver nanogels or nanoparticles with conjugated antibodies. As drug delivery vehicles, NGs offer high stability and drug loading capacity^38,39 for both hydrophilic and hydrophobic drugs⁴⁰, and the ability to co-deliver a wide range of encapsulated cargo, including small molecule drugs, proteins, nucleic acids, and more^21,40.

We first tested the MNPs with DiI-loaded nanogels due to its rapid loading and its hydrophobic nature; many cancer drugs are hydrophobic. Our results demonstrated the successful release of DiI labelled nanogels, and their uptake by HER2^high cells. Since the DiI is not covalently conjugated to the nanogel, we could not confirm that the DiI had not independently left the MNPs and entered the cells. Since the Cy5.5 label is covalently conjugated to the NG surface, we next demonstrated colocalization of Cy5.5 with the cell nuclei to confirm the MNPs were capable of releasing intact nanogels that could participate in cell uptake. Finally, to confirm our MNPs were capable of releasing antibody conjugated nanogels (ANCs), we loaded our MNPs with Cy5.5 labeled anti-HER2 ANCs. These ANCs remained functional post-MNP fabrication and release to efficiently target HER2-overexpressing breast cancer cells. The active ligand targeting improved the efficiency of uptake in the HER2^high versus the HER2^low cells, a result that was particularly pronounced in the MNP groups. In the context of cancer, ANCs offer significant therapeutic advantages by using antibodies to enable active ligand-based targeting of diseased cells. Unlike NGs that rely on passive uptake into cells with no specificity, the targeting capability of ANCs proves to be effective by decreasing the concentration of drug required to produce a therapeutic effect²³. ANCs are especially attractive for use in cancer therapy due to the ability to conjugate antibodies specific to tumor-specific antigens to induce a targeted cytotoxic response in cancer cells and limit off-target effects.

We did observe a decrease in uptake for the MNP NG/ANCs groups versus the free NG/ANC groups in some of our studies despite matching concentrations post-release. We suspect that this may be due to incomplete dissolution of the alginate or aggregation of the nanoparticles within the alginate matrix during microneedle fabrication. Consistent with this, in Figure 4F, we observed some larger red dots suggesting a grouping of several nanoparticles. Nonetheless, there was still a robust signal from the MNP-released NGs, and from a translational perspective, in vivo testing would be used to adjust the dosing as necessary to account for this and ensure efficacy. Interestingly this difference between free and MNP patch groups was more pronounced for the experiments with BT474 cells, which may be due to differences in the cellular microenvironment or heterogeneity in how these clusters of particles are dealt with by specific cell types. Finally, while our EVs and nanogels demonstrated functionality post-release from MNPs – and other authors have processed EVs into microneedle patches successfully^26,41 – we cannot rule out that the decrease in efficacy versus free controls could relate to the drying and rehydration of these hydrophilic cargoes. Because the ANCs are freely dispersed in solution, they remain continuously available for cellular internalization, leading to increased uptake over time. This effect is consistent with typical nanoparticle kinetics, where uptake increases as a function of treatment duration until equilibrium or saturation is reached. For the MNP it was noted above that some incomplete alginate dissolution is possible; thus, in the MNP groups, more particles may be released from alginate over time contributing to this effect too.

Overall, this work offers a potential transdermal strategy to deliver nanoparticles systemically using MNPs, with the goal of overcoming challenges with conventional nanoparticle delivery methods and presenting an opportunity to apply this system to a variety of nanocarriers. To fully realize this, these patches will be tested in vivo in future work and the full biodistribution of their nanoparticle payload explored. While these MNPs can provide a facile rapid delivery of nanoparticles, which can be applied in certain applications, a sustained release profile over a longer period may be desirable in other applications. The future iterations of the device will examine the use of slow degrading polymers (e.g., PLGA) that release the nanoparticles over a longer period.

4. CONCLUSION

In this work we present the fabrication of microneedle patches capable of releasing a hydrophobic small molecule and a range of nanoparticles (EVs, nanogels, ANCs) for uptake by cells of interest in vitro. Additionally, we demonstrate transdermal delivery of these therapeutics ex vivo. For the first time, we show that ANCs released from alginate MNPs can be effectively internalized by both cancerous and normal breast cells and demonstrate that the HER2 antibody on MNP-released nanoparticles remains bioactive to successfully target HER2-overexpressing cells, enhancing uptake. This approach takes steps towards improving and optimizing delivery mechanisms for the treatment of a diverse range of nanoparticle-treatable diseases, including atherosclerosis, osteoarthritis, diabetes, bacterial infections, and cancers.

Supplementary Material

Supporting information includes additional experimental data, photographs of device setup and microneedle patches, dynamic light scattering characterization of nanoparticles, and replots of nanogel and ANC data to compare between experiments.