**4. Discussion**

Numerous studies have reported the beneficial properties of soy isoflavones (SIF). However, its application in the food industry and pharmaceuticals are limited owing to its low solubility in water and poor bioavailability. Goat milk whey protein has excellent emulsifying and foaming properties [6,7], which can be used to produce bioactive compounds-loaded whey protein to expand their applications in the production of functional food products as well as for designing new drug delivery systems [37]. In this study, we prepared SIF nanoparticles using polymerized goa<sup>t</sup> milk whey protein (PGWP) as wall material.

The encapsulation e fficiency is the index used to indicate how many SIF were loaded in the nanoparticles. The encapsulation e fficiencies of SIF in the nanoparticles were higher than that obtained from gel beads of soybean hull polysaccharides by Wang et al. [5], which may be attributed to the di fferences in methods and conditions, and the result indicated that it was feasible to encapsulate SIF using PGWP as wall material. Particle size and zeta potential are important properties to provide valuable information on micro-encapsulated compounds regarding the formation of stable formulations. The particle size of the nanoparticles ranged from 135 nm to 155 nm due to the concentrations of SIF in PGWP-SIF. The values were similar to previous report by Khan et al. [11], who prepared whey protein isolate-DIM (3,3-Diindolylmethane) nanoparticles to a mean particle size of 96–157 nm. In the study, concomitant observations were obtained from transmission electron microscope (TEM) images and dynamic light scattering (DLS) analyses, the microspheres were homogeneously spherical in shape. However, compared with the dynamic light scattering data, the diameter of particles appeared smaller in the TEM images, which may be due to the di fferent principles of the two analytical methods. In addition, all the nanoparticles had values of zeta potential below −30 mV, which suggested that the samples seemed to be stabilized [24]. The nanoparticles with higher surface charge aggregated less, which suggested that they were somewhat more stable. These results suggested that PGWP could be considered as a promising encapsulation agen<sup>t</sup> for the incorporation of bioactive compounds such as SIF.

Fourier transform infrared spectroscopy (FT-IR) analysis can be used to investigate the potential interactions between SIF and PGWP. FT-IR spectra of the SIF showed characteristic peaks were related to the stretching vibration of the aromatic ring and aromatic ketone [5]. The characteristic absorption peaks of SIF in nanoparticles spectrum increased from 1623.77 cm<sup>−</sup><sup>1</sup> and 1516.31 cm<sup>−</sup><sup>1</sup> to 1655.15 cm<sup>−</sup><sup>1</sup> and 1541.95 cm<sup>−</sup>1, respectively, which suggested that the successful formation of PGWP-SIF nanoparticles. The results were consistent with the report described by Wang et al. [38], who investigated the formation of complexes between spiral dextrin sub-fraction and soy isoflavones. After interacting with SIF, the amide I band of PGWP changed, which may be due to hydrophobic interactions between the aromatic ring of SIF and the hydrophobic amino acids of proteins [39]. The amide I band was the most useful for infrared spectroscopic analysis of the secondary structure of proteins [40]. From Figure 4B, it could be observed that the α-helix and β-sheet contents for PGWP decreased after adding SIF, suggesting that the interactions between PGWP and SIF could lead to the alteration of PGWP secondary structure. It is presumable that SIF were non-covalently grafted onto PGWP, resulting in the partly unfolded protein, which may lead to the exposure of buried amino acids and promote hydrophobic interactions [41]. These results indicated that the secondary structure of goa<sup>t</sup> milk whey protein changed after interacting with SIF, and the interaction between PGWP and SIF was probably through hydrogen bonds or hydrophobic interactions.

To confirm that the conformation of PGWP was changed after interacting with SIF, we studied the fluorescence spectrum. Fluorescence spectrum was widely used to study mechanisms of the interactions between proteins and small molecules and the microenvironment changes of proteins. The main components of whey proteins are β-lactoglobulin (β-LG) and α-lactalbumin ( α-LA). Each β-LG molecule has two tryptophan residues and four tyrosine residues, while four tryptophan residues are found in α-LA molecule [42]. The fluorescence of tyrosine and tryptophan was excited by di fferent wavelengths to analyze the structural changes of proteins. Therefore, the inherent fluorescence is a useful approach to study the structural transition and binding properties of protein. Fluorescence experiments proved that SIF quenched PGWP fluorescence strongly in static mode. The nature of the interaction forces between SIF and PGWP can be partially unveiled by studying the thermodynamic parameters of the system, and the results suggested that the main interaction force was hydrophobic interaction [33]. Synchronous fluorescence confirmed that SIF a ffected the conformation of PGWP by interacting with its tyrosine and tryptophan residues. In addition, tryptophan residues had a stronger fluorescence quenching effect than that of tyrosine residues, which indicated that tryptophan residues played an important role in fluorescence quenching. Combining the fluorescence spectrum and the synchronous fluorescence spectrum, the amino acids involved in the reaction were tryptophan and tyrosine, and it can be inferred that tryptophan and tyrosine residues were involved in the hydrophobic interaction. To further confirm that SIF were encapsulated inside the nanoparticles, we studied the thermal properties of SIF before and after encapsulation. The characteristic peaks of SIF disappeared in that of PGWP-SIF, which may be attributed to that SIF were entrapped in PGWP nanoparticles. These results were in agreemen<sup>t</sup> with the report of Wang et al. [38], who observed that the disappearance of the endotherm of soy isoflavones at 184.54 ◦C in the thermogram of the complexes may be due to the formation of spiral dextrin sub-fraction/soy isoflavones complexes. This work provided some comprehensive understanding about the interactions between SIF and PGWP, and these characteristics made PGWP a novel wall material for the encapsulation of SIF. However, this research has some limitations, the antioxidant activity and in vitro release behavior of the nanoparticles will be carried out in the next study, and molecular modeling study will be performed to predict the precise binding sites of soy isoflavones in goa<sup>t</sup> milk whey protein. Our study indicated that the polymerized goa<sup>t</sup> milk whey protein can be considered as a promising carrier for encapsulating bioactive compounds. The results may be helpful in expanding the industrial application of soy isoflavones in functional foods. The study laid the foundation for further research into the interaction between goa<sup>t</sup> milk whey protein and soy isoflavones.
