Microneedles (MNs) have emerged as a simple technology that overcame several obstacles of the conventional drug delivery systems towards better drug delivery efficiency. Besides drug delivery, MNs have been evaluated for several other purposes over the last few years, including diagnosis, vaccination, and cosmetology. Fundamentally, MNs are arrays of micron-sized projections that pierce the biological barriers in a minimally invasive manner (Yang et al., 2019). MNs have been used successfully to deliver medications via various administration routes: buccal (Caffarel-Salvador et al., 2021), oral (Prausnitz et al., 2019), intramyocardial (Shi et al., 2020), ocular (Park et al., 2022), and most commonly the transdermal route (Shan et al., 2022, Chen et al., 2019). Clinically, a high number of MNs are in trials at different stages and some already have been marketed for vaccination and cosmetic purposes (Ingrole et al., 2021). Several categories of MNs exist, each with its strengths and limitations e.g. solid, hollow, coated, hydrogel-forming, and dissolving MNs. The latter has always received special attention owing to its unique properties; made from biocompatible materials that dissolve/degrade upon application into the body, by the effect of the tissue fluids, to deliver a wide variety of active pharmaceutical ingredients (APIs) (Yang et al., 2019). Yet, one of its biggest limitations in the drug delivery area is the low loading capacity of the APIs (Avcil and Çelik, 2021).

The solid state of the loaded API (i.e. amorphous or crystalline) is one of the factors that affect its behavior inside different drug delivery systems. Several studies have focused on the impact of the solid state on the release profile from different drug delivery carriers e.g. nanoparticles (Gautam et al., 2021, Dong et al., 2019) and microparticles (Maudens et al., 2018, Liu et al., 2016). However, only sparse studies have focused on its effect on loading into these carriers and showed a superiority of the APIs’ crystalline state with regard to both loading and stability (Zhang and Bodmeier, 2022, Martin et al., 2020). Concerning MNs, the crystallinity of the loaded APIs has primarily been employed to achieve controlled/sustained release profiles of the loaded compounds (He et al., 2018, Lee et al., 2021, Wang et al., 2023, Tekko et al., 2020). The effect of the APIs’ solid state on their loading into MNs is yet to be investigated. Previous studies have shown that crystalline forms of the APIs undergo precipitation in MN tips during manufacturing (Wang et al., 2023, He et al., 2018). This behavior could be harnessed not only to enhance the loading of the APIs but also to ameliorate the overall drug delivery efficacy of the MNs. Crystallization and crystal size modification in general can be achieved via two approaches, namely top-down and bottom-up technologies. Top-down size reduction can be accomplished through various methods such as milling, pulsed laser ablation, and high-pressure homogenization while the solvent anti-solvent precipitation, supercritical fluid precipitation, and solvent removal-based techniques can be used to achieve the bottom-up approach (Sinha et al., 2013, Eerdenbrugh et al., 2008b).

Given the preference of the API crystalline state for loading, this study focuses on exploring the effect of the crystal size of crystalline compounds on their loading into MNs. The natural polyphenolic compound phloretin (Ph) was used as a model drug for this purpose. Ph is a dihydrochalcone abundantly found in apples, pears, and strawberries. Through targeting different molecular mechanisms and signaling pathways, Ph possesses a vast pharmacological potential with many interesting properties such as antimicrobial (Barreca et al., 2014, Liu et al., 2021), anti-inflammatory (Chang et al., 2012, Lin et al., 2014), immunosuppressant (Lin et al., 2014, Lu et al., 2009), anti-cancer (Ma et al., 2016, Xu et al., 2018, Kim et al., 2022), antidiabetic (Schulze et al., 2014, Shen et al., 2020, Shen et al., 2017), cardioprotective (Ying et al., 2019, Stangl et al., 2005), hepatoprotective (Ebadollahi et al., 2011, Lu et al., 2017), and neuroprotective (Barreca et al., 2017, Liu et al., 2015) activities. Besides, Ph has a penetration-enhancing effect on other administered compounds through modulating the fluidity of the biological membranes (Valenta et al., 2001, Auner et al., 2005a, Auner et al., 2005b). Yet, the therapeutic potential of this compound is hindered by its low oral bioavailability (<9%), given its low hydrophilicity and absorption (Yuan Zhao et al., 2020), and its short half-life following both oral and intravenous administration (Crespy et al., 2001, Yuan et al. 2020). Ph was selected in this study for its crystalline nature and strong hydrophobicity (Log P = 3.5) (Auner et al., 2005a, Auner et al., 2005b), which minimizes the risk of solubilization in the milling medium and MN matrix during manufacturing. On the other side, hydrophilic compounds could be loaded as well in a solid state inside dissolving MNs, through the proper selection of the solvent and/or the MN-forming matrix. Moreover, dissolving MNs can be prepared by loading the API crystals directly in the molds, eliminating the need for an MN matrix (Sartawi et al., 2022), or by loading the API crystals inside a hot melt of the MN matrix, eliminating the need for a solvent (Donnelly et al., 2009, Chu and Prausnitz, 2011).

Herein, we first investigated the effect of the crystal size of crystalline compounds, for example Ph, on their loading into dissolving MNs. Ph was loaded into MNs both in its original crystalline form and after crystal size reduction. Crystal size reduction was accomplished via the top-down wet bead milling technique for different periods to obtain crystals with different sizes, in the micro and nano range, which were subsequently loaded into dissolving MNs. The reduction in the size and the morphology of the obtained crystals were confirmed by laser diffraction size measurements and microscopic imaging. The potential impact of the size reduction technique on the integrity of Ph was investigated as well. Additionally, the difference in the sedimentation behavior of these crystals was studied, both in water and in the MN-forming matrix. The manufactured MNs were subjected to an array of characterizations to investigate the state of Ph inside the MNs and assess the effect of its crystal size on the loading, integrity, sharpness, and mechanical strength of the MNs. Eventually, our size-dependent loading hypothesis was further examined by applying the bottom-up approach to modify the crystal size of the milled nano-sized Ph crystals and evaluating its effect on loading into MNs. Collectively, our study describes the relationship between crystal size and MN loading and highlights its importance in providing a potential solution to overcome one of the biggest limitations in MN technology, namely the limited loading capacity, paving the way towards the clinical translation of MNs.