**3. Results and Discussion**

In our previous study, five poly(εCL-*ran*-δCL) copolymers, herein named ε87-δ13, ε81-δ19, ε75-δ25, ε61-δ39 and ε45-δ55, and the two respective homopolymers PεCL and P*δ*CL, herein referred to as ε100-δ0 and ε0-δ100, were synthesized exhibiting a constant HHB [17]. It was demonstrated that the HHB of the bulk polymers correlated with the HHB of the corresponding NPs when particles were prepared in THF using a polymer concentration of 1 mg mL−<sup>1</sup> [17]. In the present study, the particle formation of the ε100-δ0 and ε0-δ100 was investigated over a wider range of polymer concentrations in THF ranging from 0.25 to 10 mg mL-1 using an automated pipetting robot that was adapted for the HTnanoprecipitation [25]. Particles were formulated without surfactant as well as with PVA of different concentrations (0.25 to 1% (*w/v*)). Previous studies revealed that PVA of less than 0.5% (*w/v*) generated stable drug-loaded PLGA NPs, and it could be demonstrated that even concentrations of up to 5% (*w/v*) were generally non-toxic in vitro [26]. ε100-δ0 and ε0-δ100 homopolymers both formed NPs up to the highest tested polymer concentration of 10 mg mL−<sup>1</sup> when PVA was used as a surfactant (SI, Figure S1). Even the lowest tested PVA concentration of 0.25% (*w/v*) was sufficient to obtain stable particle dispersions and ε100-δ0 NPs with a size of 150 to 300 nm and ε0-δ100 NPs with a particle size of 120 to 280 nm with PDI < 0.3. However, ε0-δ100 NPs prepared without surfactant failed to produce stable NP dispersions above concentrations of 0.5 mg mL−<sup>1</sup> as indicated by a strong aggregation of the particles. This is not surprising since ε0-δ100 is above its glass transition temperature at room temperature, which could disturb the particle formation in the absence of a stabilizer. It is well-known that several factors influence the final NP properties, including the polymer concentration, the solvent used to dissolve the polymer and the type and the concentration of the surfactant [26–28]. THF was demonstrated to be a suitable solvent in the HT-screening, resulting in stable particle formation within a broad polymer concentration range when PVA was used as a surfactant. Hence, it was selected as solvent for the subsequently performed BRP-187 encapsulation experiments. All other formulation parameters for the preparation of PCL[BRP-187] NP were adapted from our previous study that described the encapsulation of BRP-187 into PLGA NPs [21]. The first batch nanoprecipitation with the drug and a polymer concentration of 5 mg mL−<sup>1</sup> in THF yielded large particles with a diameter (dH) of 400 to 600 nm with high LC values (SI, Table S2). However, the particles revealed significant aggregation after centrifugation and lyophilization, as indicated by the higher PDI values of 0.3 to 0.6. Hence, the initial polymer concentration was reduced to 2.5 mg mL−<sup>1</sup> to optimize the dispersion stability and to decrease the particle size [28]. Particles within a size range of 200 to 260 nm and PDI values below 0.3 were obtained for all PCLs using a polymer concentration of 2.5 mg mL−<sup>1</sup>

(Table 1, SI Table S4). It was further observed that empty NPs were approximately 30 to 50 nm smaller compared to the BRP-187-loaded NPs (SI, Table S3). The particle size of the empty NPs increased by approximately 40 to 80 nm when NPs were lyophilized and subsequently reconstituted in water (SI, Table S3). Similar tendencies were also observed for the PCL[BRP-187] NPs, although here the difference in size was on average only about 30 to 50 nm (Table 1), presumably caused by the strong affinity of the hydrophobic drug with the polymer matrix [29]. The particles were also investigated via SEM (Figure 1), which revealed individual or clustered particle populations within the particle size range as indicated by DLS measurements.

**Table 1.** Overview of PCL[BRP-187] NP properties prepared in THF using a polymer concentration of 2.5 mg mL−<sup>1</sup> .


d<sup>H</sup> represents the intensity-weighted distribution (n ≥ 4 batches) and zeta-potential (ZP) (n = 3 ELS measurements) \* Amorphous or near amorphous polymers with glass transition temperature T<sup>g</sup> below 37 ◦C [17]. <sup>a</sup> Bulk degree of crystallinity as determined by wide-angle X-ray scattering (WAXS) at room temperature. <sup>b</sup> NPs measured after purification. <sup>c</sup> NPs measured after lyophilization and subsequent resuspension in water. <sup>d</sup> Yield = (mass of NPs recovered – mass of found PVA)/(mass of polymer + mass of drug) in the formulation <sup>×</sup> 100. <sup>e</sup> Determined by UV-VIS spectroscopy at λ = 316 nm (n = 4) and calculated using LC = (mass of drug recovered)/(mass of particle recovered) × 100.

**Figure 1.** SEM micrographs of PCL[BRP-187] particles consisting of the homo- or copolymers with a varying composition. Scale bar = 1 µm.

The average LC of the PCL[BRP-187] NPs was between 1.4 and 1.9% for ε100-δ0 and the poly(εCL-*ran*-δCL) copolymers (Table 1) and similar to the LC values of PLGA NPs encapsulating the same drug [21]. The only exception was the ε0-δ100 homopolymer with an LC of 3.2%, probably due to its almost liquified state at room temperature. This resulted in a viscous dispersion with emulsion-like properties in which the drug was apparently entrapped during the purification process.

In general, the yield of both empty and drug-loaded PCL NPs decreased with increasing molar fraction of *δ*CL (Figure 2A). In other words, NP yield increased with the degree of crystallinity of the polyester materials. Amorphous materials are frequently utilized as excipients in pharmaceutical formulations since they are known to increase the dissolution

rate of insoluble drugs and to enhance their bioavailability [30]. However, their major disadvantage is seen in the fact that they exhibit high energy states at a molecular level and thus are prone to physical instabilities. In particular, such tendencies were observed with the NPs of the amorphous PδCL homopolymer, which displayed a higher polydispersity and the lowest yield. In technical terms, the low yield of the copolymers with a higher fraction of δCL could have resulted from their near-molten state at room temperature causing them to sediment at a lower rate due to their lower density. Thus, after 60 min of centrifugation, a lower amount of the NPs was recovered.

**Figure 2.** Influence of the δCL fraction on the yield of drug-loaded PCL NPs (**A**), influence of the polymer crystallinity on the residual PVA content of drug-loaded PCL NPs (**B**), apparent degradation represented by the normalized relative count rate (%) after 20 h plotted against the δCL fraction of the copolymers (**C**) and influence of polymer crystallinity on the efficiency of drug-loaded PCL NPs to inhibit 5-LO product formation (**D**). Black-circled data points represent PCL polymers with a degree of crystallinity below 10% and a Tg < 37 ◦C.

Furthermore, it was observed that the residual amount of PVA in the drug-loaded NPs was higher compared to the empty NPs for all PCL copolymers (Table 1 and SI, Table S3). As mentioned before, such differences are typically a result of strong drug–polymer interactions [31], and in this case, the interactions of the BRP-187 with the chains of PVA polymer. Moreover, the residual PVA content was noticeably higher for less crystalline copolymers with a higher δCL fraction and highest for the particles consisting of the PδCL homopolymer (Figure 2B). Apparently, the surfactant molecules tended to stick to the surface or were even incorporated into the particles formed from amorphous polyesters that are above their glass transition temperature during formulation. As soon as the materials were semicrystalline and below Tm, the degree of crystallinity did not influence the amount of residual PVA anymore. Besides providing dispersion stability, surfactants also influence the degradation rate of NPs since they adsorb at the surface of the particles forming a layer that protects from enzymatic hydrolysis to some degree [32]. Additional characterization experiments of the PCL[BRP-187] NPs were performed to investigate the degradation kinetics as well as the biological evaluation of the NP efficiency to inhibit the drug targets in vitro.
