Ultrasound-Induced Release of Nimodipine from Drug-Loaded Block Copolymer Micelles: In Vivo Analysis

As nanocarriers, commercially purchased Pluronic® F-127 block copolymers (BASF Corporation, NJ, USA) were used, without additional purification. Nimodipine powder was purchased from Sigma-Aldrich Chemical Company (MO, USA). Nimodipine-loaded block copolymers were produced using the direct dissolution method as previously described [15]. For this purpose, a defined amount of Pluronic® F-127 block copolymer (three different concentrations 5%, 10%, and 15%) and 2 mg nimodipine powder were added simultaneously to 10 mL deionized water and subsequently stirred at 100 rpm and room temperature 25 °C for 3 h. Afterwards, the precipitated nimodipine was separated from the micelle suspension by filtering through 1.0-µm filter membranes (pluriStrainer®, pluriSelect Life Science, Leipzig, Germany).

Particle size measurements, morphological examination, and assessment of the micelles’ stability were conducted by performing an in vitro study for preparation and characterization of nimodipine-loaded Pluronic® F-127 block copolymers [16]. The particle size measurements were performed using a transmission electron microscope (TEM) and nanoparticle tracking analysis (NTA, NanoSight® LM10-HS; Malvern Instruments, Worcestershire, UK; DZNE Göttingen, Germany). The average particle size was less than < 150 nm, which was increasing from 127 to 149 nm. The particle morphology was evaluated under TEM, where the nimodipine-loaded micelles presented with a spherical form and a smooth surface.

The nimodipine-loaded block copolymers show stability over time concerning their size and morphology, after the solutions were stored for 3 months at room temperature and at 4 °C with evaluation of size and morphology by TEM and NTA at defined time points (1 week, 1 month, and 3 months), respectively [16].

White fertilized eggs of the Bresse Gauloise chicken breed (Poulets-de-France, Hildesheim, Germany) were used. A fortnightly incubation was ensured by the incubator BSS 420, species number 8301/01 (Grumbach® Automatic System GmbH & Co.KG, Nürnberg, Germany). On day 0, the eggs were incubated at 37.7 °C with a constant humidity of 65% according to a previously described technique by Leighton et al. [17]. On day 4 of incubation, the shell was penetrated at the lower pole, in the area of the air chamber. The window was fixed and stabilized by applying Leukotape® (BSN medical, Hamburg, Germany), and then impregnated with a fine needle and fenestrated with scissors. Subsequently, the egg was closed with transparent film (Parafilm® Pechiney Plastic Packaging, OH, USA) until the next step. On day 14, the tape was removed, and the experiments were started after ensuring the survival of the chicken embryo.

The in vivo vasospasm induction in the CAM model is based on the application of ultrasound (continuous wave mode, maximum effective area of 5 cm², 3 MHz frequency, and an intensity of 1.0 W/cm) to the CAM vessels for 60 s. First, the fertilized, incubated eggs were stabilized in a cup. Thermostable ultrasound gel (Sonosid® Asid Bonz GmbH, Herrenberg, Germany) was applied to the CAM protected by Parafilm®. The ultrasound probe was applied directly to the gel (Fig. 1).

The nimodipine-loaded micelles were delivered directly to the CAM exposed through the egg-windowing by using an Eppendorf pipette (1 µL each). The ultrasound treatments were performed with a PHYSIOSON-Expert device (Physiomed ®, Paderborn, Germany). The ultrasound-controlled drug release was done with the same device used for vasospasm induction, but by applying different device settings: continuous wave mode, maximum effective area of 5 cm², 1 MHz frequency, and an intensity of 1.7 W/cm². The ultrasound treatment was applied with a duration of 60 s.

The visualization of vessel diameter changes was performed using a 5-MP HD-microscope camera (Leica® MC170 HD, Leica Camera AG, Wetzlar, Germany). The vessel diameter was determined as a quantitative measure at predefined time points using Fiji ImageJ [18]. The values were expressed as arbitrary units according to existing standards [19]. The distance of the ultrasound probe to the vessels remained unchanged in this experimental setting. During the ultrasound application, no microscopic images and monitoring of changes could be made due to the experimental setup. Accordingly, images were taken at 1-min intervals over a period of 20 min and the macroscopic changes were recorded. The measurements were taken on one vessel, reproducibly at the same location each time. The criteria used for qualitative microscopic evaluation were retraction of surrounding CAM and vessel bursting. Additionally, macroscopic examinations were performed by observing the embryonic behavior (alteration in embryonic motility).

Neither the addition of 1 µL deionized aqua nor the lower concentration (4 µg/mL) of nimodipine Carinopharm led to any microscopically visible and quantitatively measurable alterations in the vessel diameter. In contrast, after application of 1 µL nimodipine Carinopharm 0.2 mg/mL, an almost immediate bursting of the vessels with hemorrhage was observed (Fig. 2).

Neither a macroscopic change nor a measurable alteration in vessel diameter over the observation period of 20 min was detected when applying the nimodipine-loaded block copolymers to a CAM collective without vasospasm. This was the case in all evaluated solutions (5%, 10%, 15%) of the nimodipine-loaded block copolymers (Fig. 3). The ultrasound-induced drug release also did not lead to any quantifiable or microscopically evaluable changes in vessel diameter.

After vasospasm induction the mean vessel diameter changed to 57% (ranging from 44 to 61%) compared to the initial vessel diameter, which was set at 100%. Neither microscopically visible nor measurable alterations of the vasospastic vessels could be detected after application of the nimodipine-loaded block copolymer without drug release (Fig. 4). After ultrasound-induced drug release, the mean vessel diameter of vasospastic vessels increased to 89% (range 83–91%) of their baseline diameter, which was statistically significant (ANOVA, p = 0.0002). Images of the CAM vessels before and after vasospasm induction as well as before and after ultrasound-induced drug release are demonstrated in Fig. 5. The overall vasospasm resolution rate in the 5% solution was comparable to the vasospasm resolution rate in the 10% solution, when taking into consideration the higher vasospasm severity grade in the 5% solution group compared to the 10% solution group (Fig. 6).

A comprehensive analysis of undesirable effects of nanomaterials and determination of the toxicological profile is necessary before the application of the investigated approach in humans. In this study, we decided to use the CAM model for several reasons: (1) cost-effectiveness; (2) good vascularization; (3) easy to perform as well as repeatable, and (4) direct visualization of vessel changes. The CAM model also most closely resembles the complex conditions of a living organism [22, 23]. According to the 3R principle (reduce, refine, replace), the CAM model seems to be a suitable model for in vivo analysis before the initiation of animal studies [24]. The CAM model offers an uncomplicated setup that requires neither special surgical skills nor a complex surgical environment. After in vivo opening, the eggs do not require any further care as this is the case when using animal models. The low workload of incubating the eggs enables large numbers of test units. In particular, this experimental model allows optimization, standardization, and thus improvement of reproducibility of the in vivo experiments. Furthermore, the free access to the CAM allows a variety of different analyses such as optical evaluation by microscopy as well as further imaging and therapeutical techniques [25]. Here, it enabled us to perform a highly reproducible and comparable qualitative and quantitative evaluation of the CAM vessels. The absence of nociceptive nerves in the CAM and in the chicken embryo up to the 14th day makes the CAM model an ideal alternative to equivalent rodent models. The CAM model thus offers the best conditions to act as a mediator between in vitro and in vivo experiments.

Nanomaterials unveil a variety of potential applications, including their use as drug carriers [26]. As previously described, Pluronic® F-127 copolymers were here used as nanocarrier, which are widely available and considered to be non-toxic [15, 27]. Focused ultrasound emerges as a promising candidate for the active induction of drug delivery systems due to, e.g., its deep soft tissue penetration and its ability to focus its energy on the targeted location of desired drug release [28‐30]. Ultrasound can be used to trigger the release of encapsulated drug from copolymers. Although the mechanisms underlying ultrasound-induced drug release from block copolymers are not clear and have to be elucidated in future studies, there is some evidence that mechanical action of ultrasound is involved in this process [28]. Nevertheless, except for our study, no reports currently exist on the use of the CAM model in combination with ultrasound for in vivo vasospasm assessment. We were able to reproducibly induce vasospasm with a diameter reduction of approximately 50% by a short ultrasound application. Due to the overall small CAM vessels, we were not able to induce measurable diameter reduction of more than 50%, which we consider a limitation of the model. The findings of our study are, however, encouraging for further evaluation of the concept in animal models and can be seen as a solid basis for the planning of future experiments.

There are several limitations of the study that need to be acknowledged here. A general limitation of the CAM model is its short study period. The observation period is limited either by hatching of the embryos at day 21 or by termination of the setup at day 14 due to nociception. Furthermore, the allantoic sac gradually shrinks and loses its function after 19 days of egg hatching with a limited length of research window [31]. Another limitation is that chickens are not mammals. Ultimately, this leads to the consequence that the obtained results cannot be transferred to humans without restrictions [32]. The CAM model provides the possibility for evaluation of drug effects on vessels in a living organism but does not allow a comprehensive assessment of all drug effects. For these reasons, animal models remain irreplaceable whereby the CAM model should be considered a transition model from in vitro to in vivo experiments facilitating the generation of relevant insights, which are the prerequisite for an efficient planning of in vivo experiments involving animal models. Due to methodological limitations. our preliminary study does not provide answers on the question concerning the stability of the nanodrug within the cerebrospinal fluid space (CSF) and interactions with the CSF clearance, which is the objective of future in vivo experiments in a rat model. In this preliminary study, we were able to provide a proof of the concept for production of a nimodipine-loaded nanodrug with on-demand drug release by ultrasound application, which was the prerequisite to plan further experiments in an animal model. Another objective in future in vivo experiments is the dose determination of the nanodrug for an effective resolution of cerebral vasospasm in a subarachnoid hemorrhage rat model. Furthermore, the definition of the ultrasound parameters for application through the skull is also one of the objectives for future research alongside with experiments for assessment of antivasospastic as well as toxicity-related side effects of the nanodrug after intrathecal application.