Skip to main content
Download PDF

Responsive porous microneedles with riboflavin ocular microinjection capability for facilitating corneal crosslinking

Journal of Nanobiotechnology volume 22, Article number: 588 (2024) Cite this article

Abstract

Riboflavin-5-phosphate (riboflavin) is the most commonly used photosensitizer in corneal crosslinking (CXL); while its efficient delivery into the stroma through the corneal epithelial barrier is challenging. In this paper, we presented novel responsive porous microneedles with ocular microinjection capability to deliver riboflavin controllably inside the cornea to facilitate CXL. The microneedle patch was composed of Poly (N-isopropyl acrylamide) (PNIPAM), graphene oxide (GO), and riboflavin-loaded gelatin. After penetrating the cornea by the stiff and porous gelatin needle tip, the photothermal-responsive characteristic of the PNIPAM/GO hydrogel middle layer could realize the contraction of the gel under the stimulation of near-infrared light, which subsequently could control the release of riboflavin from the backing layer into the cornea stromal site both in vitro and in vivo. Based on the microneedles system, we have demonstrated that this microinjection technique exhibited superior riboflavin delivery capacity and treatment efficacy to the conventional epithelial-on protocol in a rabbit keratoconus model, with benefits including minimal invasiveness and precise administering. Thus, we believe the responsive porous microneedles with riboflavin ocular microinjection capability are promising for clinical corneal crosslinking without epithelial debridement.

Introduction

Keratoconus (KC) is a cornea ectasia disorder, characterized by progressive thinning and steeping of the cornea, which can lead to impaired visual acuity [1,2,3]. Corneal crosslinking (CXL) is now widely adopted as an effective approach with riboflavin-5-phosphate (riboflavin) as a photosensitizer to slow or even halt the deterioration of KC as well as eliminate the need for further keratoplasty [4, 5]. Riboflavin can absorb the ultraviolet (UV) light and excite into both singlet and triplet-excited riboflavin, facilitating the intermolecular and intramolecular molecular chain reactions as a photochemical process. CXL can increase collagen stiffness through the formation of cross-links among collagen monomers mediated by riboflavin under the irradiation of UV, thus maintaining corneal biomechanical stability and structural rigidity. The current standard clinical protocol for the CXL is the Dresden protocol, which needs to debride the epithelium prior to the riboflavin administration due to the hydrophobic corneal epithelial barrier with tight junction and the hydrophilic property of riboflavin [6, 7]. However, the epithelium-off strategy may result in persistent epithelial defect, corneal haze, and ocular infections postoperatively [8,9,10]. Epithelium-on approaches have been developed including the utilization of chemical enhancers to loosen the epithelial tight junctions, iontophoresis, and femtosecond laser combined with phonophoresis, etc [11,12,13]. Nevertheless, these protocols suffer from limitations, such as low effectiveness, hyperthermia-related complications, and a lack of long-term outcome [14]. Thus, a new ocular riboflavin delivery method is yet to be explored to facilitate corneal collagen crosslinking.

Herein, we developed a novel responsive porous microneedle (MN) for targeted and highly effective ocular riboflavin delivery, as schemed in Fig. 1. Microneedles are a new drug delivery system with micron-sized needles and a supporting base, featured by targeted and localized encapsulation and delivery of drugs, peptides, proteins, nucleic acids, nanoparticles, and so on [15,16,17,18]. Microneedles enable a minimally invasive penetration of the stratum corneum, establishing channels for precise transdermal delivery to the targeted local sites, which have shown promising therapeutic efficacy in various diseases [19,20,21,22,23,24]. This inherent characteristic of microneedles forms the basis of a platform for delivering compounds such as riboflavin. However, previously reported microneedles usually involve a slow dissolution process and lack specially designed structures, which can hinder the rapid and efficient delivery of compounds like riboflavin during corneal crosslinking [25]. Therefore, a new microneedle system that can tackle these challenges and provide a proficient platform for riboflavin delivery is still highly anticipated.

In this paper, we present distinctive responsive porous microneedles with riboflavin ocular microinjection capability for facilitating corneal crosslinking. The microneedle system was constructed with three distinct hierarchical structural components, including a freeze-drying derived stiff and porous gelatin needle tip with sufficient mechanical strength for corneal penetration, a photothermal-responsive Poly (N-Isopropyl acrylamide)/graphene oxide (PNIPAM/GO) gel in the middle that for pumping the liquid to the porous needle tip when contracting, and a riboflavin-loaded gelatin substrate working as a drug reservoir for the continuous supply of riboflavin into the system. As the microneedles were high biocompatibility, their treated cornea can recover within 24 h without affecting its structural integrity. In addition, due to the excellent photothermal effect of PNIPAM/GO, the release of riboflavin from the microneedle system could be effectively controlled and enhanced by using near-infrared light. Based on these features, we have demonstrated in a rabbit KC model that the responsive porous microneedles could offer superior riboflavin delivery capacity and treatment efficacy to the conventional protocol with minimal invasiveness and precise administering. These results indicated that our responsive porous microneedles with ocular microinjection capability are potentially valuable as a minimally invasive method for clinic drug delivery.

Fig. 1
figure 1

Schematic diagram of the use of responsive porous microneedles with riboflavin ocular microinjection capability for corneal collagen crosslinking. (a) Comprised of N-isopropyl acrylamide (NIPAM), graphene oxide (GO), and riboflavin-loaded gelatin, these microneedles enable precise delivery of riboflavin into the cornea. A photothermal-responsive hydrogel regulates riboflavin release upon near-infrared light stimulation, facilitating targeted and controlled delivery. (b) This microinjection technique demonstrates superior riboflavin delivery and treatment efficacy compared to conventional methods, with minimal invasiveness and precise administration

Full size image

Results and discussion

In a typical microneedle fabrication experiment, an array of 11 × 11 PDMS microneedle array template was employed for the fabrication of the Gelatin/PNIPAM/GO microneedle patch. Specifically, the solution composed of 20%(w/v), 30%, and 40% gelatin was added into a PDMS mold, followed by depositing the solution into the microcavities via a low-speed centrifuge and removing the excessive gelatin by scraping the mold, as shown in Fig. 2a. The microneedle patch was then frozen at -80 ℃ overnight and placed in the freeze dryer for 24 h for solidification, creating pores in the tip. Subsequently, a mixture of PNIPAM and GO was added to the mold, and excessive bubbles were removed through a low-speed centrifuge. The mixture was then polymerized under UV light on ice. Finally, a riboflavin-loaded 10% gelation was added to the mold as a supporting layer and the tips were detached. The microneedle size was about 400 μm in height, 250 μm in width, and 450 μm in tip inter-distance on a 10*10 mm2 support base (Fig. 2b-c). To clearly distinguish hierarchical structural components, fluorescence images of the microneedle tip presented the composition of the microneedle, with fluorescein (FITC)-labeled gelatin as the needle top and Rhodamine red-labeled PNIPAM/GO as the needle bottom (Fig. 2d). A distinctly sharp tip characterized by a porous structure was observed under the scanning electron microscope (Fig. 2e-g). The lyophilization criteria were further evaluated by varying the gelation concentration, in order to fabricate the porous tips of the microneedle patch with desired properties. With the increase of gelatin concentration, the porous structure became denser, providing a better biomechanical property and easiness for molding. Moreover, uniform microstructure could hardly be achieved when lowering the gelatin concentration (Figure S1, Supporting Information). Thus, a concentration of 20–40% was adopted to fabricate the porous microneedle and conduct the subsequent experiments.

Fig. 2
figure 2

Schematics of fabrication and morphology of the porous microneedle patch: a) Schematic diagram of the microneedle molding process; b-c) Digital and enlarged images of the microneedle patch; d) Fluorescence images of microneedle tip fabricated by fluorescein isothiocyanate (FITC)-labeled gelatin (top) and Rhodamine Red-labeled Poly (N-isopropyl acrylamide)/ graphene oxide (PNIPAM/GO) (bottom); e-g) Scanning electron microscope (SEM) image of the microneedle patch; Scale bars: 400 μm in (b), 200 μm in (c) and (d), 50 μm in (e) and (f), and 10 μm in (g)

Full size image

The photothermal properties of the PNIPAM/GO hydrogel patch were then studied under irradiation of a near-infrared (NIR) laser at 808 nm. We first characterized the structure of PNIPAM and PNIPAM/GO by scanning electron microscope (SEM) (Fig. 3a-b). The porous structure with interconnected pores was observed, providing a pathway for solvent uptake or drug release. As shown in Fig. 3c&S2, the shrinkage of hydrogel was observed after 3 min of NIR irradiation. The swelling ratio was also plotted as shown in Figure S3, indicating that the hydrogel deswelled when the temperature rose. Photothermal images of microneedle patches containing varying GO concentrations, including 0, 250, 500, 750, and 1000 µg/mL were captured (Fig. 3d). The temperature of all groups rose to the plateaus phase after irradiation for about 150s, while the environment presented indistinctive changes, as shown in the heating curve (Fig. 3e). Furthermore, heating curves of MNs loaded with 500 µg/mL GO were recorded under different irradiation power densities (0.25, 0.5, 0.75, 1, and 1.25 W/cm2) for 5 min (Fig. 3f) [26]. As the NIR power increased, the heating rate and the peak temperature significantly rose. GO also presented good photothermal stability, which was demonstrated by irradiation cycles (Fig. 3g, h). Considering the lower critical solution temperature of the PNIPAM and melting temperature of gelatin, GO concentration of 500 µg/mL and NIR power of 0.75 W/cm2 were determined as an optimized condition for subsequent application.

Fig. 3
figure 3

Characterization of the photothermal effect of Poly (N-isopropyl acrylamide)/ graphene oxide (PNIPAM/GO) patch. (a) Scanning electron microscope (SEM) image of PNIPAM (b) SEM image of PNIPAM/GO. (c) Digital photos of PNIPAM/GO hydrogel shrinkage after near-infrared (NIR) irradiation. (d) Photothermal images (e) and heating curves of microneedles incorporated with GO with different concentrations (n = 3). (f) Heating curves of microneedles incorporated with GO under different irradiation power densities (n = 3). (g) Photothermal stability of GO. (h) Temperature variation of primary heating and cooling of GO. Scale bar: 20 μm in (a) and (b); 5 mm in (c)

Full size image

Taking advantage of the photothermal conversion ability of GO and the thermoresponsive behavior of PNIPAM, the riboflavin release was controlled by repeated NIR laser irradiation, as demonstrated in Fig. 4a. The microneedles could continuously pump the liquid to the porous needle tip when contracting under NIR. The hydrogel could then revert to its initial state once NIR irradiation is discontinued. Meanwhile, the melted riboflavin-loaded gelation was then refilled to the hydrogel pores, which were then pumped to the cornea stroma upon NIR irradiation reactivated (Fig. 4b). This pumping cycle process ensured continuous and rapid riboflavin delivery. The in vitro study was performed to analyze the riboflavin delivery efficacy with NIR on and NIR off. The riboflavin concentration of the cornea was higher when the NIR was used to facilitate the pumping cycle. Moreover, an in vitro drug release study of the patches in agarose was performed to visualize the drug dispersal upon repeated irradiation (Fig. 4d). The riboflavin quickly dispersed under NIR laser irradiation compared with NIR off. The in vitro study also demonstrated that microneedles quickly released riboflavin during the first 30 min under NIR irradiation (Figure S4). An axial compression test then assessed the mechanical strength of microneedles. The mechanical strength of microneedles became much stiffer as the gelatin concentration increased (Fig. 4e-g). The HE staining also validated that the cornea could be penetrated when the gelation concentration reached as high as 40% (Fig. 4h, S5). Based on the results described above, 40% gelatin was selected to be the ideal concentration to conduct the subsequent experiments. To better visualize the penetration ability, the microneedles loaded with riboflavin were placed onto the 3% agarose and compressed with thumb. The microneedle presented superior penetration ability as well as the delivery efficacy (Figure S6). Then the efficacy of riboflavin delivery was further confirmed in vivo. The microneedle was applied to the rat cornea surface after anesthesia and irradiation cycle for 10 min before crosslinking under UV light (Fig. 4i). Conversely, the riboflavin was dipped onto the cornea surface without epithelial debridement as control. The cornea stroma was more compact than the control group after crosslinking, indicating riboflavin reached the satisfactory concentration for crosslinking (Fig. 4j & S7).

Fig. 4
figure 4

(a) Schematic diagram of the near-infrared (NIR)-controlled drug release procedure of microneedle patch; (b) Representative digital images capturing riboflavin release with NIR on and off; (c) Comparison of riboflavin infiltration in cornea between NIR on group and off group; (d) Time-dependent release pattern of riboflavin with NIR (upper) and without NIR (lower) in 0.4% agarose; (e) Illustration for axial mechanical force detection for microneedles; (f) Displacement-force curve of microneedle with different concentrations; (g) Comparison of the needle force with different concentrations; (h) HE staining of in vitro cornea following microneedle insertion; (i) The illustration of the application of microneedles to the rat cornea in vivo; (j) The compact structure of cornea following crosslinking. Scale bar: 200 μm in (b), 50 μm in (c), (d), (h) and (j). Data are presented as mean ± SD (n = 3), ** indicates p < 0.01, *** indicates p < 0.001, compared with the control group

Full size image

To evaluate the biocompatibility and biosafety of the microneedle patch, human corneal epithelial cells (HCEC), and human keratocytes (HK) were co-cultured with the leaching liquid from microneedle patches with PNIPAM and GO concentrations from 0, 250, 500, 750 to 1000 µg/ml, respectively. As shown in Fig. 5a, the live/dead staining of 72 h showed no significant difference in cell proliferation and morphology among groups, which was also consistent with the cell counting kit-8 (CCK-8) assay (Fig. 5b-c, S8). The penetration of microneedles through the cornea in vivo was assessed by a slit lamp in a rabbit model(Fig. 5d). We performed fluorescein staining to evaluate the corneal epithelium defect. The results indicated that the cornea epithelium was interrupted, confirming penetration of the microneedle into the stromal layer. The epithelium soon recovered within 24 h, rarely resulting in the resistant corneal epithelium defect. The endothelium also demonstrated similar structures and density of cells before and after microneedle insertion indicated by in vivo corneal confocal microscopy imaging (Fig. 5e). The epithelium did not show obvious apoptosis 24 h after microneedle insertion using terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) test (Fig. 5f).In vitro hemolysis test also confirmed that PNIPAM, gelatin, and GO which were used in microneedle fabrication showed good biocompatibility (Figure S9). Besides, in vivo biocompatibility of microneedles was evaluated by subcutaneous embedding. The HE staining results of the heart, liver, spleen, lung, and kidney indicated that the microneedle would not cause histological changes in the main organs (Figure S10). Thus, the corneal biocompatibility and minimal invasiveness of the microneedles were validated.

Fig. 5
figure 5

The biocompatibility evaluation of the microneedle patch in vitro and in vivo: a) Live/dead staining of human keratocyte and human corneal epithelial cells after incubation for 72 h (n = 3); b-c) CCK8 assay of human keratocyte and human corneal epithelial cells of 24 h, 48 h and 72 h (n = 6); d) Slit lamp images after microneedle insertion and its recovery following 24 h(i-iii indicated the microneedle (MN) penetrates the epithelial, iv-vi indicated the epithelial recovered in 24 h) (n = 3); e) In vivo corneal confocal images of the endothelium of cornea before and after microneedle insertion (n = 3); f)The TUNEL test of the cornea after 24 h insertion (n = 3); Scale bars: 2 mm in (d), 100 μm in (a) and (f), 50 μm in (e). Data are presented as mean ± SD

Full size image

To test the practical therapeutic effects of KC, type II collagenase digestion was used to establish the rabbit corneal ectasia model, as shown in Figure S11 [27]. Four representative refractive maps of Pentacam and SD-OCT images (4 weeks postoperatively) were obtained (Figure S12a). The treatment led to increased cornea curvature, corneal thinning, and biomechanical weakening (Figure S12b) [28]. Moreover, the collagen content and corneal thickness were significantly reduced after collagenase treatment (Figure S12c, d). The timeline and evaluation procedure of the animal model were presented in Fig. 6a. The rabbits were divided into three groups based on riboflavin delivery methods. These included the epi-on group, in which the riboflavin was dripped onto the cornea without epithelial removal; the epi-off group, which followed the conventional approach involving riboflavin application after corneal epithelium removal, and the microneedle group. Specifically, we repeatedly irradiated the microneedles with NIR after microneedle insertion to facilitate sustained riboflavin delivery to the corneal stroma for 10 min. Continuous UVA with 9 mW cm− 2 power density irradiated the cornea for 8 min, covering a 9 mm facula [29]. Considering all thickness of the rabbit ectasia cornea was below 400 μm, we reduced the irradiation power to a maximum of 4.32 J to protect the endothelium. The cornea of the epi-off group became significantly swollen and opacity 1 day postoperatively, and the central cornea was still defective, accompanied by severe conjunctival congestion. The cornea remained transparent with mild edema in epi-on and microneedle groups. One week later, the corneal edema returned to the preoperative level, and the corneal transparency was maintained. The complete re-epithelialization of the defective area was noted in the epi-off group (Fig. 6b). Consequently, the SD-OCT also indicated significant corneal edema soon after CXL, which returned to the preoperative (prep-op) level one week later (Fig. 6c, e). The maximum K value measured by Pentacam was reduced by approximately 2.1D one week following the crosslinking, indicating corneal flattening(Fig. 6d, f). To further evaluate the corneal biomechanical characteristics of CXL effect, stress-strain measurements were carried out. Specifically, the corneal strips measuring 1 cm*3 mm were fixed in the stretching apparatus. A pre-cycling tensile strength of 0.5 N was applied, with a tensile displacement of 1 mm at a rate of 2 mm/min for three cycles. Subsequently, each cornea was stretched to 10% deformation immediately at a loading rate of 2 mm/min. The stress-strain curves were plotted in Fig. 6g, h. The stress and Young’s modulus of the group were higher than the epi-on group and comparable with the epi-off group [30, 31]. Thus, riboflavin-loaded microneedles as the delivery system demonstrated superior efficacy in crosslinking procedures.

Fig. 6
figure 6

Timeline and results of crosslinking evaluation in rabbit cornea ectasia model. (a) The operation procedures of corneal ectasia model establishment and in vivo crosslinking (CXL) assessments; (b) The slit lamp examination of crosslinking post 1 day and 1 week; (c) SD-OCT of the cornea using three riboflavin delivery protocols; (d) Representative Pentacam images of the cornea using three riboflavin delivery protocols before and post 1 week; e-f) The biomechanical property of corneal strips; g) Central corneal thickness comparison before and post 1 day and 1 week; h) Maximum corneal curvature comparison before and post 1 week. Data are presented as mean ± SD (n = 5 for each group), *indicates p < 0.05, ** indicates p < 0.01, **** indicates p < 0.001, compared with epi-on group or baseline. pre-op: preoperative, MN: microneedle

Full size image

Cornea structural changes induced by CXL usually induce tissue matrix remodeling. To further verify the histological effects of CXL with different RF delivery methods, we performed HE staining as well as picrosirus red staining (Fig. 7a). Corneal epithelial defects in the UV-irradiated area were observed in the epi-off group 1 day postoperatively. The corneal epithelium was nearly completely recovered, while there were only 2 ~ 3 layers of corneal epithelial cells on day 7, indicating that the healing process elongated via epi-off delivery. The epithelial cells were flattened and rounded, slightly larger than those of the other two groups. The collagen fibers of the stroma were edematous and thickened in both epi-off and microneedle groups, which were also recovered on day 7. However, collagen fibers were disorganized, with a marked decrease of stromal cells in the epi-off group, further verified by the corneal confocal microscope and the nucleus labeled by DAPI was almost absent in immunofluorescent staining (Fig. 7b, c). The corneal epithelial cells became thinner and were not tightly adherent to the underlying stroma in the microneedle group. This might be caused by the thermo-effect of the NIR irradiation, and some epithelial cells might be exfoliated when rinsing the eyes due to the sticky gelatin, while it did not cause whole-thickness epithelial defect as the epi-off group. The collagen fibers in the stroma became compact and tightly alignment in both epi-off and MN groups on day 7, indicating the crosslinking efficacy. A significant rise in TUNEL-positive cells was observed in the epi-off group. The corresponding quantitative data of cornea collagen volume and TUNEL-positive cells were present in Fig. 7d, e. The comparison of histological changes indicated that the microneedle delivery system could inject the riboflavin into the stroma minimally invasively, which further tightened the collagen fiber connections and enhanced biomechanical properties of the cornea.

Fig. 7
figure 7

Histological and morphologic presentations of cornea postoperatively. a) H&E staining and Picrosirius red staining, b) TUNEL test of the cornea, c) Confocal microscopic examinations of the corneal anterior and posterior stroma, and endothelium, d-e) Quantification of collagen volume and apoptosis of cornea following crosslinking. Scale bars: 50 μm in (a) and (c). Data are presented as mean ± SD (n = 5 for each group), *** indicates p < 0.005, compared with epi-on group. MN: microneedle

Full size image

Conclusion

In conclusion, we have developed a responsive porous microneedle with riboflavin ocular microinjection capability for topical ocular drug delivery. The three distinct hierarchical structural components of the microneedle system enabled sustained riboflavin delivery into corneal stroma through the microchannels. The photothermal-responsive PNIPAM/GO could further control and accelerate the riboflavin release, effectively addressing the limitations of the slow dissolution process and the lack of specially designed structures of previous microneedle technologies. Compared to the conventional crosslinking protocol, the most distinctive advantage of our system is minimal invasiveness and precise administering. The microneedle delivery system also presented an outstanding performance in crosslinking surgery, including enhancement of corneal biomechanical properties, acceleration of postoperative recovery, and densification of collagen fibers. Thus, we believe that the responsive porous microneedle with riboflavin ocular microinjection capability is potentially valuable as a minimally invasive method for clinical drug delivery.

Methods

Materials

Gelatin and NIPAM were purchased from Sigma-Aldrich and riboflavin 5-monophosphate (riboflavin) was provided by Macklin (Shanghai, P. R. China). GO was purchased from XFNANO (Nanjing, China).

Microneedle Fabrication

Responsive porous microneedles were developed using a micro-molding approach. Initially, varying amounts of gelatin (2, 3, and 4 g) were dissolved in 10 mL of pure water to generate a gelatin solution. After complete dissolution at 50 °C with magnetic stirring, the gelatin was cast into a polydimethylsiloxane mold (11 × 11 array in 1 cm2 area). The excess solution was removed by scraping the template before transferring it to a refrigerator (-80 °C) overnight for freezing. Subsequently, the microneedle tips were freeze-dried for 24 h. A mixture of PNIPAM and GO was then added into the mold via low-speed centrifugation and consolidation under UV light on ice [32]. Finally, the microneedle patches were slowly detached from the PDMS mold, with gelatin added as the backing layer. For riboflavin-loaded microneedle, 0.01 g of riboflavin was dissolved in 10 mL of pure water and thoroughly mixed, with a concentration was 0.1%. Then, 1 g of gelatin was added to the riboflavin solution to produce a riboflavin-loaded gelatin aqueous solution serving as the backing layer. Then, we observed the microneedle morphology under a stereo microscope (Jiangnan). The microstructures of the microneedle were captured by a SEM (JSM-IT2000, Japan).

Photothermal responses Assessment

The 808 nm NIR emitter was positioned directly above the microneedles. The temperature fluctuations were measured by infrared thermal images captured by a thermal imager (FLIR, E5xt) [26]. To assess the effects of GO concentrations, the GO concentrations were set among 0, 250, 500, 750, and 1000 µg mL− 1, while the NIR power was fixed at 0.75 W. Similarly, to investigate the impact of different NIR power levels, powers ranging from 0.25 to 1.25 W were employed, with GO concentration at 500 µg mL− 1.

Microneedle mechanical property tests

A universal mechanical testing system (Instron) was used to perform the mechanical property test. In detail, the microneedles were put on the bottom fixed station, with the tips facing upwards. Then, the mobile station was slowly lowered towards the microneedles at a speed of 0.2 mm/min. The alterations in force and pressure-displacement curves were then recorded and plotted in real time throughout the process.

In vitro biocompatibility tests

Human corneal epithelial cells (HCEC) and human keratocytes (HK) were incubated in Dulbecco’s modified eagle medium (DMEM)/high glucose and 10% fetal bovine serum (Gibco). The cells were digested and seeded on a 96-well plate. Leaching liquid was prepared by immersing the gelatin microneedle with or without different concentrations of GO in the DMEM media for 24 h. The media was replaced by leaching liquors after 24 h of incubation. CCK8 test was applied to evaluate the cell viability of HCECs and HKs at 24 h, 48 h, and 72 h, respectively. Live/dead staining was also used to visualize the cell viability. Briefly, the cultured cells were rinsed with Phosphate Buffer Saline (PBS) and treated with 100 µL of 10µM Calcein acetoxymethyl ester (Calcein-AM) and propidium Iodide (PI) solution at 37 °C for 10–15 min. The solution was discarded and PBS was added to immerse the cells. The fluorescence staining of cells was then observed under an inverted fluorescence microscope (Nikon, Japan). For hemolysis testing, fresh whole blood harvested from rabbits was centrifuged at 3,000 rpm for 5 min and the supernatant was removed and refilled with PBS. The procedures were repeated 3 times until the supernatant was clear. Then we resuspended the pellet with 10 ml PBS. We added 200 µl suspension in 800 µl H2O as a positive control; and another 200 µl in 800 µl PBS as a negative control. Then suspension 200 µl were added in leaching liquid of PNIPAM, GO and gelatin, respectively. The mixtures were incubated at 37℃ for 4 h and centrifuged at 10,000 g for 5 min. The absorptance at 577 nm was measured by the micro-plate reader. The hemolysis rate was calculated as (Absorptance of samples- Absorptance of negative control)/ (Absorptance of postive control - Absorptance of negative control)*100%.

In vitro drug release profile

The microneedle tips were submerged in 5 mL PBS at 37 °C to evaluate the riboflavin release profiles. For the NIR-on group, the microneedle was irradiated under NIR on for 150s, followed by a NIR off lasting 150s which constitutes one cycle. At predetermined time points (3, 5, 10, 15, 30 min, 1 and 2 h), 100µL solution was withdrawn, with an additional 100µL of PBS added. The quantification of released riboflavin was assessed by a microplate reader at 446 nm wavelength (Infinite 200Pro, Tecan, Swiss). To further visualize the dispersal of riboflavin, the microneedle patch was inserted into the agarose, and corresponding fluorescent images of drug release at 0, 5, 10, and 20 min were captured.

In vivo cornea penetration and biocompatibility assessment

The slit lamp microscopic examination and fluorescein sodium staining were performed to assess the epithelial penetration capability of the microneedle [33]. Briefly, the microneedles were applied to the cornea by thumb pressure for 1 min. Then the cornea epithelial integrity was observed with the slit lamp. Moreover, the in vivo corneal confocal microscopy was performed before and after the microneedle insertion to compare the integrity and morphology of endothelium.

Rabbit cornea ectasia model establishment

5–6 months New Zealand white rabbits were utilized in the study. The experimental procedures were approved by the Ethical Committee of Eye and ENT Hospital, Fudan University (IACUC-DWZX-2022-017). Briefly, after the exclusion of ocular disease preoperatively, rabbits were anesthetized with 0.2 ml/kg xylazine hydrochloride intramuscularly. Proparacaine hydrochloride eye drops (Alcon) were then topically administered. A 4 mm cotton piece fully soaked in 5% type II collagenase was applied onto the central cornea after epithelial removal for 20 min [27]. The cornea was rinsed with 0.9% sodium chloride after removing the cotton piece. The rabbit cornea was examined 2 and 4 weeks postoperatively. Corneal topography was measured by Pentacam (Oculus, Germany). The anterior segment was evaluated by spectral domain optical coherence tomography (SD-OCT, Optovue, USA).

In vivo corneal crosslinking procedures

These rabbits were randomly divided into three groups with three distinct riboflavin delivery methods [34].For epi-on group, as the control group, the riboflavin was topically administered onto the cornea. For epi-off group, as the positive control group, topical anesthetic eye drops were applied to the operative eye 2–3 times, one drop each time. Remove the corneal epithelium with an epithelial scraper. Then 0.1% riboflavin was applied topically or used for soaking for 5 min until adequate riboflavin penetration through the cornea. In the microneedle group, as the treated group, the microneedle was applied to the cornea and irradiated with NIR repeatedly for 10 min. After topical riboflavin-loaded microneedles application and repeated NIR irradiation for 10 min, then the corneas were irradiated with continuous UV light. The efficacy of crosslinking was further evaluated by Pentacam and AS-OCT as described above on day 1 and day 7.

Corneal biomechanical test

In vitro biomechanical characteristics were assessed by a universal mechanical testing system (Instron) [34]. Briefly, corneas were freshly harvested from the rabbits and transferred to the laboratory for analysis. 10 × 3 mm corneal strips were fixed onto the stretching equipment. After 3 cycles of pre-stretching the cornea with 0.5 N force at a 2 mm/min rate, the cornea was stretched to 10% deformation with the same loading rate.

Corneal histopathological examinations

After sacrificing the rabbits, the harvested cornea was soon fixed with 4% paraformaldehyde (Biosharp) for 24 h. Then the cornea tissue was embedded in paraffin after dehydration with graded ethanol. For hematoxylin & eosin staining, the slides were gently immersed in hematoxylin for approximately 1–2 min and rinsed off the hematoxylin with water. Then, merge the slides with 1% acid-alcohol, and 0.3% ammonia water for seconds, followed by dehydration with 85% and 95% ethanol for 5 min. The slides were then incubated in eosin for 2 min and dehydrated with ethanol. A bright-field microscope (Leica) was used to capture tissue images. Picrosirius red staining (PSR) was performed according to the manufacturer’s instructions (Solarbio). Briefly, stain the section with iron hematoxylin for 5 min, and wash it with distilled water. Then drip the sirius red staining solution on the slide and dye for 15 min, and rinse with running water. The slides were observed under a polarized light microscope (Leica).

TUNEL staining

In Situ Cell Death Detection Kit (Servicebio, China) was used for TUNEL assay following the manufacturer’s instructions. Each section was immersed with 100µL Proteinase K at 37 °C for 20 min and rinsed with PBS 3 times. Then 50 µL equilibration buffer was used to cover the tissue at room temperature for 10 min, followed by incubation with 56 µL TdT incubation buffer at 37 °C for 1 h. Antifade mounting medium with DAPI was covered on the tissue. The slides were then observed under a fluorescence microscope and apoptotic cells appeared with green fluorescence due to the labeled DNA fragments.

Statistical analysis

All experiments were conducted at least in triplicate independently. The results are expressed as means ± standard deviation. The T-tests, one-way ANOVA and graphs were performed with GraphPad Prism (v9.4.1, California USA) where available. Statistical significance was defined as p < 0.05.

Data availability

No datasets were generated or analysed during the current study.

References

  1. Santodomingo-Rubido J, Carracedo G, Suzaki A, Villa-Collar C, Vincent SJ, Wolffsohn JS. Keratoconus: an updated review. Cont Lens Anterior Eye. 2022;45:101559.

    Article  PubMed  Google Scholar 

  2. Tan DT, Dart JK, Holland EJ, Kinoshita S. Corneal transplantation. Lancet. 2012;379:1749–61.

    Article  PubMed  Google Scholar 

  3. Borchert GA, Tan J. Munson’s sign in Keratoconus. N Engl J Med. 2023;389:939.

    Article  PubMed  Google Scholar 

  4. Wollensak G, Spoerl E, Seiler T. Riboflavin/ultraviolet-a-induced collagen crosslinking for the treatment of keratoconus. Am J Ophthalmol. 2003;135:620–7.

    Article  CAS  PubMed  Google Scholar 

  5. Hashemi H, Seyedian MA, Miraftab M, Fotouhi A, Asgari S. Corneal collagen cross-linking with riboflavin and ultraviolet a irradiation for keratoconus: long-term results. Ophthalmology. 2013;120:1515–20.

    Article  PubMed  Google Scholar 

  6. Lombardo M, Giannini D, Lombardo G, Serrao S. Randomized controlled trial comparing Transepithelial corneal cross-linking using iontophoresis with the Dresden Protocol in Progressive Keratoconus. Ophthalmology. 2017;124:804–12.

    Article  PubMed  Google Scholar 

  7. Wisse RPL, Simons RWP, van der Vossen MJB, Muijzer MB, Soeters N, Nuijts R, Godefrooij DA. Clinical evaluation and validation of the Dutch crosslinking for Keratoconus score. JAMA Ophthalmol. 2019;137:610–6.

    Article  PubMed  Google Scholar 

  8. Shalchi Z, Wang X, Nanavaty MA. Safety and efficacy of epithelium removal and transepithelial corneal collagen crosslinking for keratoconus. Eye. 2015;29:15–29.

    Article  CAS  PubMed  Google Scholar 

  9. Raiskup F, Hoyer A, Spoerl E. Permanent corneal haze after riboflavin-UVA-induced cross-linking in keratoconus. J Refract Surg. 2009;25:S824–828.

    Article  PubMed  Google Scholar 

  10. Khoo P, Cabrera-Aguas M, Watson SL. Microbial Keratitis after corneal collagen cross-linking for corneal ectasia. Asia-Pacific J Ophthalmol. 2021;10:355–9.

    Article  Google Scholar 

  11. Aytekin E, Öztürk N, Vural İ, Polat HK, Çakmak HB, Çalış S, Pehlivan SB. Design of ocular drug delivery platforms and in vitro - in vivo evaluation of riboflavin to the cornea by non-interventional (epi-on) technique for keratoconus treatment. J Control Release. 2020;324:238–49.

    Article  CAS  PubMed  Google Scholar 

  12. Celik Buyuktepe T, Ucakhan OO. Long-term visual, refractive, tomographic and aberrometric outcomes of corneal collagen crosslinking (CXL) with or without hypoosmolar riboflavin solution in the treatment of progressive keratoconus patients with thin corneas. Graefes Arch Clin Exp Ophthalmol. 2022;260:1225–35.

    Article  CAS  PubMed  Google Scholar 

  13. Mastropasqua L, Nubile M, Calienno R, Mattei PA, Pedrotti E, Salgari N, Mastropasqua R, Lanzini M. Corneal cross-linking: intrastromal riboflavin concentration in iontophoresis-assisted imbibition versus traditional and transepithelial techniques. Am J Ophthalmol. 2014;157:623–e630621.

    Article  CAS  PubMed  Google Scholar 

  14. Shajari M, Kolb CM, Agha B, Steinwender G, Müller M, Herrmann E, Schmack I, Mayer WJ, Kohnen T. Comparison of standard and accelerated corneal cross-linking for the treatment of keratoconus: a meta-analysis. Acta Ophthalmol. 2019;97:e22–35.

    Article  PubMed  Google Scholar 

  15. Chi J, Sun L, Cai L, Fan L, Shao C, Shang L, Zhao Y. Chinese herb microneedle patch for wound healing. Bioactive Mater. 2021;6:3507–14.

    Article  CAS  Google Scholar 

  16. Lin F, Zhuang Y, Xiang L, Ye T, Wang Z, Wu L, Liu Y, Deng L, Cui W. Localization of lesion cells and targeted Mitochondria Via embedded Hydrogel Microsphere using heat transfer Microneedles. Adv Funct Mater. 2023;33:2212730.

    Article  CAS  Google Scholar 

  17. Pacheco C, Baião A, Ding T, Cui W, Sarmento B. Recent advances in long-acting drug delivery systems for anticancer drug. Adv Drug Deliv Rev. 2023;194:114724.

    Article  CAS  PubMed  Google Scholar 

  18. Chang H, Chew SWT, Zheng M, Lio DCS, Wiraja C, Mei Y, Ning X, Cui M, Than A, Shi P, et al. Cryomicroneedles for transdermal cell delivery. Nat Biomed Eng. 2021;5:1008–18.

    Article  CAS  PubMed  Google Scholar 

  19. Zhang X, Wang Y, Chi J, Zhao Y. Smart Microneedles for Therapy and Diagnosis. Research 2020, 2020.

  20. Liu Y, Zhao X, Zhao C, Zhang H, Zhao Y. Responsive porous Microcarriers with Controllable Oxygen Delivery for Wound Healing. Small. 2019;15:1901254.

    Article  Google Scholar 

  21. Yang J, Zhang H, Hu T, Xu C, Jiang L, Shrike Zhang Y, Xie M. Recent advances of microneedles used towards stimuli-responsive drug delivery, disease theranostics, and bioinspired applications. Chem Eng J. 2021;426:130561.

    Article  CAS  Google Scholar 

  22. Lyu S, Dong Z, Xu X, Bei H-P, Yuen H-Y, James Cheung C-W, Wong M-S, He Y, Zhao X. Going below and beyond the surface: Microneedle structure, materials, drugs, fabrication, and applications for wound healing and tissue regeneration. Bioactive Mater. 2023;27:303–26.

    Article  Google Scholar 

  23. Zhu Z, Wang J, Pei X, Chen J, Wei X, Liu Y, Xia P, Wan Q, Gu Z, He Y. Blue-ringed octopus-inspired microneedle patch for robust tissue surface adhesion and active injection drug delivery. Sci Adv. 2023;9:eadh2213.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Hu S, Zhu M, Xing H, Xue Y, Li J, Wang Z, Zhu Z, Fang M, Li Z, Xu J, et al. Thread-structural microneedles loaded with engineered exosomes for annulus fibrosus repair by regulating mitophagy recovery and extracellular matrix homeostasis. Bioactive Mater. 2024;37:1–13.

    Article  CAS  Google Scholar 

  25. Lyu S, Dong Z, Xu X, Bei HP, Yuen HY, James Cheung CW, Wong MS, He Y, Zhao X. Going below and beyond the surface: Microneedle structure, materials, drugs, fabrication, and applications for wound healing and tissue regeneration. Bioact Mater. 2023;27:303–26.

    PubMed  PubMed Central  Google Scholar 

  26. Kong B, Liu R, Shan J, Li M, Zhou X, Zhao Y. Frozen reinforced microneedles loaded with NIR-photothermal nanozyme for keratitis treatment. Nano Today. 2023;52:102000.

    Article  CAS  Google Scholar 

  27. Wei J, He R, Wang X, Song Y, Yao J, Liu X, Yang X, Chen W, Li X. The corneal ectasia model of rabbit: a Validity and Stability Study. Bioeng (Basel) 2023, 10.

  28. Huo Y, Chen X, Cao H, Li J, Hou J, Wang Y. Biomechanical properties analysis of forme fruste keratoconus and subclinical keratoconus. Graefes Arch Clin Exp Ophthalmol 2022.

  29. Raiskup F, Spoerl E. Corneal crosslinking with riboflavin and ultraviolet A. I. Principles. Ocul Surf. 2013;11:65–74.

    Article  PubMed  Google Scholar 

  30. Chong J, Dupps WJ Jr. Corneal biomechanics: measurement and structural correlations. Exp Eye Res. 2021;205:108508.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Blackburn BJ, Rollins AM, Dupps WJ Jr. Biomechanics of Ophthalmic Crosslinking. Transl Vis Sci Technol. 2021;10:8.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Li L, Chen Z, Shao C, Sun L, Sun L, Zhao Y. Graphene Hybrid Anisotropic Structural Color Film for cardiomyocytes’ monitoring. Adv Funct Mater. 2020;30:1906353.

    Article  CAS  Google Scholar 

  33. Shi H, Zhou J, Wang Y, Zhu Y, Lin D, Lei L, Vakal S, Wang J, Li X. A Rapid corneal Healing Microneedle for efficient ocular drug delivery. Small. 2022;18:e2104657.

    Article  PubMed  Google Scholar 

  34. Yang M, Xu W, Chen Z, Chen M, Zhang X, He H, Wu Y, Chen X, Zhang T, Yan M, et al. Engineering Hibiscus-Like Riboflavin/ZIF-8 Microsphere composites to enhance Transepithelial corneal cross-linking. Adv Mater. 2022;34:e2109865.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

NA.

Funding

This research was supported by the National Key Research and Development Program of China (2023YFA0915000(04), the National Natural Science Foundation of China (Grant 82371091 and 82301251), and the Natural Science Foundation of Shanghai (Grant No.23ZR1409200).

Author information

Authors and Affiliations

  1. Eye Institute and Department of Ophthalmology, Eye & ENT Hospital, Fudan University, Shanghai, 200031, China

    Xiaojun Hu, Yunzhe Wang, Yuanjin Zhao, Meiyan Li & Xingtao Zhou

  2. Department of Rheumatology and Immunology, State Key Laboratory of Bioelectronics, School of Biological Science and Medical Engineering, Nanjing Drum Tower Hospital, Southeast University, Nanjing, 210096, China

    Bin Kong & Yuanjin Zhao

Authors
  1. Xiaojun Hu
    View author publications

    Search author on:PubMed Google Scholar

  2. Bin Kong
    View author publications

    Search author on:PubMed Google Scholar

  3. Yunzhe Wang
    View author publications

    Search author on:PubMed Google Scholar

  4. Yuanjin Zhao
    View author publications

    Search author on:PubMed Google Scholar

  5. Meiyan Li
    View author publications

    Search author on:PubMed Google Scholar

  6. Xingtao Zhou
    View author publications

    Search author on:PubMed Google Scholar

Contributions

Y.J.Z. conceived the idea and designed the experiment; M.Y.L and X.T.Z provided funding support. X.J.H. and Y.Z.W conducted experiments, data analysis and wrote the manuscript; B.K assited with cell culture, revised the paper, and checked the data.All authors reviewed the manuscript.

Corresponding authors

Correspondence to Yuanjin Zhao, Meiyan Li or Xingtao Zhou.

Ethics declarations

Ethics approval and consent to participate

All animal experiments were conducted in accordance with the principles and were approved by the Ethical Committee of Eye and ENT Hospital, Fudan University (IACUC-DWZX-2022-017).

Consent for publication

All authors agree to submit for publication in “Journal of Nanobiotechnology”.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Hu, X., Kong, B., Wang, Y. et al. Responsive porous microneedles with riboflavin ocular microinjection capability for facilitating corneal crosslinking. J Nanobiotechnol 22, 588 (2024). https://doi.org/10.1186/s12951-024-02851-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12951-024-02851-4

Keywords

Journal of Nanobiotechnology

ISSN: 1477-3155

Contact us