molecules Article Development of a Hybrid Bioinorganic Nanobiocatalyst: Remarkable Impact of the Immobilization Conditions on Activity and Stability of β-Galactosidase Luigi Tavernini 1 , Oscar Romero 1,2, * , Carla Aburto 1 , Fernando López-Gallego 3,4 , Andrés Illanes 1 and Lorena Wilson 1, * 1 Escuela de Ingeniería Bioquímica, Pontificia Universidad Católica de Valparaíso, Avenida Brasil 2085, Valparaíso 2362803, Chile; Luigi.tavernini@pucv.cl (L.T.); carla.aburto@pucv.cl (C.A.); andres.illanes@pucv.cl (A.I.) 2 Bioprocess Engineering and Applied Biocatalysis Group, Department of Chemical Biological and Environmental Engineering, Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain 3 CIC biomaGUNE, Basque Research and Technology Alliance (BRTA), Paseo de Miramón 182, 20014 Donostia-San Sebastián, Spain; flopez.ikerbasque@cicbiomagune.es 4 Ikerbasque, Basque Foundation for Science, 48013 Bilbao, Spain * Correspondence: oscarenrique.romero@uab.cat (O.R.); lorena.wilson@pucv.cl (L.W.) Abstract: Hybrid bioinorganic biocatalysts have received much attention due to their simple syn- thesis, high efficiency, and structural features that favor enzyme activity and stability. The present work introduces a biomineralization strategy for the formation of hybrid nanocrystals from β-   galactosidase. The effects of the immobilization conditions were studied, identifying the important effect of metal ions and pH on the immobilization yield and the recovered activity. For a deeper Citation: Tavernini, L.; Romero, O.; understanding of the biomineralization process, an in silico study was carried out to identify the Aburto, C.; López-Gallego, F.; Illanes, A.; Wilson, L. Development of a ion binding sites at the different conditions. The selected β-galactosidase nanocrystals showed high Hybrid Bioinorganic Nanobiocatalyst: specific activity (35,000 IU/g biocatalyst) and remarkable thermal stability with a half-life 11 times Remarkable Impact of the higher than the soluble enzyme. The nanobiocatalyst was successfully tested for the synthesis of Immobilization Conditions on Activity galacto-oligosaccharides, achieving an outstanding performance, showing no signs of diffusional and Stability of β-Galactosidase. limitations. Thus, a new, simple, biocompatible and inexpensive nanobiocatalyst was produced with Molecules 2021, 26, 4152. https:// high enzyme recovery (82%), exhibiting high specific activity and high stability, with promising doi.org/10.3390/molecules26144152 industrial applications. Academic Editor: Teofil Jesionowski Keywords: β-galactosidase; nanoimmobilization; biomineralization Received: 7 June 2021 Accepted: 5 July 2021 Published: 8 July 2021 1. Introduction Publisher’s Note: MDPI stays neutral The need to improve the economic and catalytic performance of enzymes for industrial with regard to jurisdictional claims in application has driven the development of numerous immobilization strategies aimed at published maps and institutional affil- enhancing biocatalyst stability (either thermal or mechanical) to re-use the enzymes in iations. multiple operational cycles [1]. In addition, recent advances in enzyme immobilization not only allow their reuse, but also improve and modulate their catalytic properties [2,3]. Nevertheless, immobilization often impacts negatively on enzyme activity by causing inac- tivation due to the harsh immobilization conditions, and by adding inert mass in the form Copyright: © 2021 by the authors. of a carrier [4]. Consequently, much effort has been spent in optimizing the immobilization Licensee MDPI, Basel, Switzerland. protocols to reduce these negative effects and making the advantages of immobilization This article is an open access article prevail. In this direction, during the last decades there has been growing interest in the distributed under the terms and synthesis and use of nanostructured materials for enzyme immobilization. Nanostructured conditions of the Creative Commons materials display a large surface-to-volume ratio and highly adjustable surface characteris- Attribution (CC BY) license (https:// tics, allowing an increased enzyme loading per unit mass, and immobilization conditions creativecommons.org/licenses/by/ tailored for the specific requirements of the proteins [5]. The nanobiocatalysts thus devel- 4.0/). oped exhibit improved immobilization efficiency, high stability, easy handling and recovery, Molecules 2021, 26, 4152. https://doi.org/10.3390/molecules26144152 https://www.mdpi.com/journal/molecules Molecules 2021, 26, 4152 2 of 14 as well as reduced mass transfer resistance for substrates, facilitating biotechnological ap- plications [6]. In recent years, Ge et al. [7,8] introduced an innovative immobilization strategy based on the biomineralization of metal phosphates with proteins, resulting in what has been termed hybrid organic-inorganic nanoflowers, or simply nanoflowers, due to the flower-resembling nanostructure of these biocatalysts [9]. Hybrid nanoflowers have drawn the interest of researchers owing to the large surface reaction area offered by their petal-like nanostructures, increasing immobilization efficiency and substrate access, as well as stability and selectivity. Another outstanding characteristic is their very simple, eco-friendly, and cost-efficient synthesis [10–12]. The biomineralization process is carried out under mild reaction conditions at room temperature in aqueous phosphate buffer, harnessing the strong affinity of enzymes for metal ions which bind to the protein backbone via coordination interactions with reactive aminoacidic groups, forming nucleation points distributed across the protein surface [7,13,14]. Many factors have been reported to affect the formation and structure of the nanoflow- ers, including the concentration of reagents (enzymes, phosphate, and metal ions), pH, temperature, and biomineralization time [15]. However, to date little is known about how these factors affect the catalytic properties of the biocatalyst, such as expressed activity and stability [12,16]. Therefore, immobilization conditions must be thoroughly investi- gated to determine those yielding the optimal biocatalytic performance. This is crucial for enzymes for industrial use, where having a biocatalyst with good catalytic perfor- mance must be balanced with the economic viability of the process. This is the case of β-galactosidase (β-gal), a well-known enzyme with wide applications in the production of lactose-free products or prebiotics, such as galacto-oligosaccharide (GOS), lactulose and other syrups [17–20]. The need to improve the economic and catalytic performance of β-galactosidases for industrial application has driven the development of numerous immobilization strategies [21]. Despite the growing interest, the immobilization of β- galactosidase as bioinorganic nanoflowers has received little attention, and not much emphasis has been placed on the immobilization conditions, nor on its application in synthetic reactions [22–24]. The aim of this work was to evaluate the influence of the immobilization condition on the catalytic properties of the nanobiocatalyst, focusing on the effects of metal ions, salt/protein ratio, biomineralization time and pH. Furthermore, in order to get insight into the biomineralization process, a computational study was carried out to identify the potential nucleation sites of the enzyme at different immobilization conditions. Finally, the nanobiocatalyst was assessed in terms of its expressed activity, thermostability and its performance in the synthesis of GOS. 2. Results and Discussion 2.1. Metal Ion Screening A screening was conducted with Cu2+ , Mg2+ , Zn2+ , Co2+ , Fe2+ , Ca2+ to determine those ions suitable for the immobilization of β-gal as hybrid bioinorganic nanocrystals. The results in terms of immobilization yield (IY) led to the selection of Ca2+ for the following research stages, given the substantial advantage shown by this ion over the other ions assayed, obtaining an IY of 29.4%, which exceeds by five times the IY of the next best ion (Cu2+ ) (Table S1, Supplementary Materials). 2.2. Biomineralization of β-Galactosidase with Calcium Once the calcium ion was selected, the study of biomineralization at pH 7.4 was conducted using CaCl2 /protein ratios of 115, 230, 290, 345, 400 and 460 mmolCaCl2 /gprotein , and contact times of 30, 60 and 120 min. The results obtained in terms of immobiliza- tion yield (IY) and specific activity (SA) (Figure 1) showed an optimum in the range from 290 to 345 mmolCaCl2 /gprotein in the case of IY, and around 290 mmolCaCl2 /gprotein in the case of SA, regardless of the biomineralization time evaluated. Maximum IY was 44% at 345 mmolCaCl2 /gprotein , which is very close to the value of 43% obtained at Molecules 2021, 26, x FOR PEER REVIEW 3 of 13 Molecules 2021, 26, 4152 3 of 14 regardless of the biomineralization time evaluated. Maximum IY was 44% at 345 mmol- 290/g CaCl2 mmol protein,CaCl which 2 /gprotein is verywith a negligible close difference to the value of 43% between obtained30 atand 290 60 minCaCl2 mmol of biomineraliza- /gprotein with a tion. A peak negligible of 21,300between difference IU/gbiocatalyst 30 andwas 60 achieved unambiguously atA290 min of biomineralization. mmol peak CaCl2 /g of 21,300 IU/g bio- , protein again showing no significant difference between 30 and 60 min of biomineralization. catalyst was achieved unambiguously at 290 mmolCaCl2/gprotein, again showing no significant Over- all IY and SA difference slightly between 30decreased and 60 minat aofbiomineralization biomineralization.time of 120IY Overall min, andinSA comparison slightly de- to 30 and 60 min. creased at a biomineralization time of 120 min, in comparison to 30 and 60 min. Figure Figure 1. 1. Immobilization Immobilization yield yield (IY) (IY) (a) (a) and and specific specific activity activity (SA) (SA) (b) (b) of of calcium-phosphate calcium-phosphate nanocrystals nanocrystals ofof β-galactosidase β-galactosidase produced at pH 7.4 at CaCl2/protein ratios (R) of 115, 230, 290, 345, 400 and 460 mmolCaCl2/gprotein for 30 min, 60 min, and produced at pH 7.4 at CaCl2 /protein ratios (R) of 115, 230, 290, 345, 400 and 460 mmolCaCl2 /gprotein for 30 min, 60 min, and 120 min of biomineralization time. Data are presented as mean ± margin of error, n = 2. Dashed lines represent the model 120 min of biomineralization time. Data are presented as mean ± margin of error, n = 2. Dashed lines represent the model for each data set according to Equation S1 (Supplementary Materials). for each data set according to Equation (S1) (Supplementary Materials). Remarkably, IY and SA followed the same trend, and hence exhibited the same opti- Remarkably, IY and SA followed the same trend, and hence exhibited the same mum, unlike most other immobilization methods, particularly the support-based ones, in optimum, unlike most other immobilization methods, particularly the support-based ones, which enzyme load per unit mass of carrier usually results in a clear compromise between in which enzyme load per unit mass of carrier usually results in a clear compromise IY and SA. A very close trend to the one shown in Figure 1 can be observed for protein between IY and SA. A very close trend to the one shown in Figure 1 can be observed immobilization (Figure S1, Supplementary for protein immobilization Materials) being (Figure S1, Supplementary the optimum Materials) beingprotein immo- the optimum bilization yield obtained between 290 and 345 mmol CaCl2/gprotein (62% in nanocrystals pro- protein immobilization yield obtained between 290 and 345 mmolCaCl2 /gprotein (62% in duced after 60 nanocrystals min of biomineralization, produced and 50% in nanocrystals after 60 min of biomineralization, and 50% inproduced nanocrystalsafter both 30 produced and after120bothmin30 of and biomineralization). Diffusional limitations 120 min of biomineralization). Diffusional may be playing limitations may a role on the be playing activity drop in the CaCl 2/protein range beyond the optimum (around 290 mmolCaCl2/gpro- a role on the activity drop in the CaCl2 /protein range beyond the optimum (around tein), although, as seen in Figure S1, protein immobilization followed a similar trend to the 290 mmolCaCl2 /gprotein ), although, as seen in Figure S1, protein immobilization followed SA and IYtrend a similar in Figure to the1,SAshowing and IYthat at higher in Figure CaCl2/protein 1, showing that at ratios higherlessCaClprotein was being 2 /protein ratios biomineralized. This is possibly due to the formation of more naked less protein was being biomineralized. This is possibly due to the formation of more naked calcium-phosphate salt at the expense ofsalt calcium-phosphate theatprotein-calcium-phosphate complex. On the other the expense of the protein-calcium-phosphate hand, the complex. Ondet- the rimental other hand, effect theobserved detrimental at 120 min effect of biomineralization observed may be caused bymay at 120 min of biomineralization modifications be caused of bythe enzyme’s structure modifications leadingstructure of the enzyme’s to activity loss, or leading due to increasing to activity loss, or duediffusional re- to increasing strictions brought about by more compact and complex structures diffusional restrictions brought about by more compact and complex structures hindering hindering substrate ac- cess. It hasaccess. substrate been It widely has beenreported widelythat the structure reported that the complexity of nanoflowers structure complexity increase of nanoflowers with incubation increase time [11,15]. with incubation time [11,15]. The The results results were were arranged arranged into into a full 32 factorial experimental design to simplify analysis analysis and and subjected subjected to to an an analysis analysis of variance (ANOVA) to assess the statistical signifi- signif- icance cance of of the the variables variables CaCl CaCl22/protein /protein ratio ratioand andbiomineralization biomineralization time time (see (see Supplementary Supplementary Materials). Materials). Considering Consideringaa level level of of significance significance ofof αα == 0.05, 0.05, both both variables, variables, as as well well as their combination, exerted aa statistically combination, exerted statisticallysignificant significanteffect effectononIYIY and and SASA of of thethe β-gal β-gal nanocrys- nanocrystals, tals, validating validating the previous the previous observations observations drawndrawn fromfrom Figure Figure 1. 1. 2.3. Effect 2.3. Effect of of pH pH on on the the Biomineralization Biomineralization of of β-galactosidase β-Galactosidase Enzyme immobilization Enzyme immobilization as as hybrid hybrid bioinorganic bioinorganic nanoflowers nanoflowers has has been been shown shown toto be be critically affected by pH. Despite this, few examples of biomineralization at different critically affected by pH. Despite this, few examples of biomineralization at different pHs pHs Molecules 2021, 26, x FOR PEER REVIEW 4 of 13 Molecules 2021, 26, 4152 4 of 14 are described in the literature [15], motivating the present study of the effect of pH on are described in the literature [15], motivating the present study of the effect of pH on biomineralization. biomineralization. The effect of pH on β-gal biomineralization was firstly evaluated by means of pro- spectiveThe effect carried assays of pH onout atbiomineralization β-gal 290 mmolCaCl2/gwas firstly evaluated by means of prospec- protein for 60 min. The results show a linear tive assays carried out at 290 mmolCaCl2 /gprotein for 60 min. The results show a linear correlation in the pH range from 6 to 9 for both IY and SA (Figure 2a,b). Beyond pH 9.0, a correlation in the pH range from 6 to 9 for both IY and SA (Figure 2a,b). Beyond pH 9.0, sudden drop in both parameters was observed, while at pH below 6.0 no activity was a sudden drop in both parameters was observed, while at pH below 6.0 no activity was recovered. recovered.Quite Quiteremarkably, IYincreases remarkably, IY increases from from 0 at0pH at 6.0 pHto6.0 72%toat72% pH at 9.0,pH 9.0, demonstrat- demonstrating ingthe thestrong strong effect of pH on the biomineralization effect of pH on the biomineralization process. process. Figure 2. Immobilization Figure 2. Immobilization yield (IY) yield (a)(a)and (IY) andspecific specific activity (SA)(b) activity (SA) (b)ofof calcium-phosphate calcium-phosphate β-galactosidase β-galactosidase nanocrystals nanocrystals produced at a at produced CaCl 2/protein a CaCl ratio of 290 mmolCaCl2/gprotein for 60 min at different pHs. Immobilization yield (IY) (c) and 2 /protein ratio of 290 mmolCaCl2 /gprotein for 60 min at different pHs. Immobilization yield (IY) (c) and specific activity specific (SA)(SA) activity (d) (d) of calcium-phosphate of calcium-phosphateβ-galactosidase nanocrystalsproduced β-galactosidase nanocrystals produced at pH at pH 9.0 9.0 formin for 60 60 min at different at different /protein CaCl2CaCl ratios. Data are presented as mean ± margin of error, n = 2. Dashed lines in (c) and (d) represent 2 /protein ratios. Data are presented as mean ± margin of error, n = 2. Dashed lines in (c) and (d) represent the model the model for each data set according to Equation S2 (Supplementary Materials). for each data set according to Equation (S2) (Supplementary Materials). The The effectof effect of pH pH on onthetheformation formationof nanoflowers is presumably of nanoflowers due to changes is presumably in the due to changes in theionization ionization state of the state enzyme’s of the reactive enzyme’s groups reactive as well groups asaswell to changes in the distribution as to changes in the distribu- equilibrium of the phosphate species [13,25]. Several authors have suggested that an tion equilibrium of the phosphate species [13,25]. Several authors have suggested that an increase of the net negative charge of the enzyme molecules at higher pH had a positive increase of the net negative charge of the enzyme molecules at higher pH had a positive effect on the binding affinity of metal ions, promoting biomineralization by increasing effect on the points nucleation binding affinity [26]. of metal Furthermore, ions, promoting previous works havebiomineralization by increasing reported that an increase in nucleation pH yielded points [26]. Furthermore, less ordered previous and less compact worksstructures nanoflower have reported [26,27].that Thisan increase could favorin pH yielded less ordered the diffusion and less of substrates andcompact productsnanoflower structures in and out from [26,27]. This the biocatalyst, could favor explaining the the diffusion of substrates and products in and out from the biocatalyst, explaining the in- increase in specific activity at higher pHs. crease in specific activity at higher pHs. These results led to the formulation of a new experimental run at pH 9.0, considering the same CaCl2/protein ratio range assayed at pH 7.4 but considering only 60 min of bio- mineralization. As control, we incubated the soluble enzyme at the biomineralization pH in order to detect potential activity loss. As a result, we observed that the enzyme Molecules 2021, 26, 4152 5 of 14 These results led to the formulation of a new experimental run at pH 9.0, consider- ing the same CaCl2 /protein ratio range assayed at pH 7.4 but considering only 60 min of biomineralization. As control, we incubated the soluble enzyme at the biomineral- ization pH in order to detect potential activity loss. As a result, we observed that the enzyme remained fully active for up to 3 h of incubation in PBS at pH 9. Figure 2c,d displays the behavior of IY and SA in relation to the CaCl2 /protein ratio as a result of the biomineralization of β-gal with calcium-PBS at pH 9.0 for 60 min. As was the case for the biomineralization at pH 7.4, production of the β-gal nanocrystals at pH 9.0 exhibited a quadratic correlation of both IY and SA with respect to the CaCl2 /protein ratio, with a clear maximum in the vicinity of 290 mmolCaCl2 /g protein. The maximum values of SA and IY were 35,000 IU/gbiocatalyst and 81.7% respectively, showing a remarkable improvement over previous results at pH 7.4, confirming the successful optimization in the biocatalyst preparation. These data were modeled according to the quadratic expression (model and parameters are shown in Supplementary Information). ANOVA testing showed that the regression model is highly significant, supporting the critical effect of both pH and CaCl2 /protein ratio on the biomineralization of β-gal. 2.4. Thermal Stability of the β-Galactosidase Nanocrystals The thermal stability of selected β-galactosidase nanocrystals was evaluated under nonreactive conditions 100 mM PBS pH 7.0 at 50 ◦ C. Thermal stability of free β-gal was also carried out for comparative purposes. To do so, the stability tests were performed for the biocatalysts made at pH 7.4 and 9.0, considering 115, 290, and 460 mmolCaCl2 /gprotein , as these CaCl2 /protein ratios. As can be seen from Figure 3a,b, nanocrystals in general showed higher stability than soluble β-gal, the latter being almost completely inactivated after 100 h of incubation, while the β-gal nanocrystals made at pH 7.4 and pH 9.0 retained over 20% and 40% of activity, respectively, after the same incubation time. Another relevant finding derived from the inactivation kinetics of the β-gal nanocrystals was that stability increased proportionally to the CaCl2 /protein ratio. This could be explained by the fact that an increase in CaCl2 /protein ratio generates more compact and complex structures (Figure 3), creating a protective effect on the enzyme structure and the activity thereof. Parameters of the mathematical models for the inactivation kinetics are shown in Sup. Information. The ANOVA and coefficients of determination prove the statistical significance of the models, and the parameters half-life time, SF and CP were determined therefrom. Figure 3c,d show the half-life times and the catalytic potential of different β-gal bioinorganic nanocrystals compared to the free enzyme. β-gal nanocrystals synthesized at pH 9.0 exhibit overall higher half-life times than the soluble enzyme and the nanocrystals produced at pH 7.4, respectively. A SF of 14.7 was achieved at the most stable condition (pH 9.0, 460 mmolCaCl2 /gprotein and 60 min of biomineralization), and the best immobi- lization condition showed a SF of 11.6 (pH 9.0, 290 mmolCaCl2 /gprotein , 60 min). These results are substantially better than those reported with other immobilization methods, in Table 1 a comparison of the results of different methods of immobilization for A. oryzae β-galactosidase are shown. The developed biocatalyst exhibited both high SA and high stability as opposed to other works reviewed that usually yield a trade-off between these two parameters (Table 1). ization at pH 7.4, production of the β-gal nanocrystals at pH 9.0 exhibited a quadratic correlation of both IY and SA with respect to the CaCl2/protein ratio, with a clear maxi- mum in the vicinity of 290 mmolCaCl2/g protein. The maximum values of SA and IY were 35,000 IU/gbiocatalyst and 81.7% respectively, showing a remarkable improvement over pre- Molecules 2021, 26, 4152 vious results at pH 7.4, confirming the successful optimization in the biocatalyst prepara- 6 of 14 tion. These data were modeled according to the quadratic expression (model and param- eters are shown in Sup. Information). ANOVA testing showed that the regression model Table 1. Comparative resultsis highly significant,ofsupporting of immobilization the critical Aspergillus oryzae effect ofby β-galactosidase both pH and different CaCland methods 2/protein ratio on thermal the biomineralization stability of the resulting biocatalysts. of β-gal. Immobilization Immobilization 2.4. Thermal Stability of the Specific Activity β-galactosidase Nanocrystals Stability Support/Precipitant Reference Methods Yield [%] [IU/g] 1 Factor The thermal stability of selected β-galactosidase nanocrystals was evaluated under Biomineralization - 81.7 35,000 11.6 This work nonreactive conditions 100 mM PBS pH 7.0 at 50 °C. Thermal stability of free β-gal was Glyoxyl-agarose also carried out for comparative 85 purposes. To do 2500so, the stability1.5 [28] tests were performed for Amino-glyoxyl-agarose 57 7700 2.3 [28] the biocatalysts Carboxy-glyoxyl-agarose made at pH 7.4 13.14 and 9.0, considering780 115, 290, and1.3 460 mmolCaCl [28] 2/g pro- tein, as these CaCl2/protein ratios. Chelate-glyoxyl-agarose 5.13 As can be seen from Figure 3a,b, 455 1.4nanocrystals in general [28] showed higher stability than 54 Chitosan-epichlorohydrin soluble β-gal, the 2951 latter being almost 1.8completely [29] inactivated Covalent bonding Amino-glyoxyl-agarose after 100 h of incubation, while 40 the β-gal nanocrystals 2294 made at pH 2.57.4 and pH 9.0 [29]retained Chitosan-glutaraldehyde 70.5 3070 3.9 [30] over 20% and 40% of activity, respectively, after the same incubation time. Another rele- Alginate-cellulose beads 70 n.r. 3.4 [31] vant finding derived from the Chitosan-glutaraldehyde inactivation kinetics 18.4 n.r. of the β-gal nanocrystals 1.2 was [32]that sta- bility fibers-PEI Cotton increased proportionally 80 to the CaCl2/protein n.r. ratio. This could 10.3 be explained[33] by the Sodium fact thatalginate an increase in CaCln.r. 2/protein ratio generatesn.r. more compact2.8 [34] struc- and complex tures -(Figure 3), creating a 100 protective effect 64,000 on the enzyme structure1.1 and the [35] activity thereof.- Parameters of the mathematical 90 models 44,560 2.3 kinetics are for the inactivation [35] shown in CLEAs - 13.5 Not reported 1.2 Sup. Information. The ANOVA and coefficients of determination prove the statistical sig- [32] - 30 15,000 1.3 [36] nificance of the models, and the parameters half-life time, SF and CP were determined 1 Activity determined using o-NPG as substrate. n.r.: not reported. therefrom. Figure 3. Inactivation kinetics at 50 ◦ C and pH 7.0 of free β-galactosidase (β-gal) and β-galactosidase nanocrystals produced at pH 7.4 (a), and at pH 9.0 (b) after 60 min of mineralization time at CaCl2 /protein ratios of 115 mmolCaCl2 /gprotein (R1), 290 mmolCaCl2 /gprotein (R3), and 460 mmolCaCl2 /gprotein (R6). Data are presented as mean ± margin of error, n = 2. The dashed lines represent the models. Half-life time (c) and catalytic potential (CP) (d) of free β-galactosidase (β-gal) and β-galactosidase nanocrystals produced after 60 min of mineralization time at pH 7.4 (pH 7.4, blue column) and pH 9.0 (pH 9.0, yellow column) at different CaCl2 /protein ratios (R). The catalytic potential was evaluated at 30% residual activity. Molecules 2021, 26, 4152 7 of 14 Molecules 2021, 26, x FOR PEER REVIEW 7 of 13 The catalytic potential (CP) denotes the total catalytic capacity of the biocatalyst along its lifespan and allows the merging of activity of the immobilized biocatalysts a server for predicting metal ion binding sites (MIB at http://bioinfo.cmu.edu.tw/MIB/ (ac- and their stability into one compounded parameter [37]. Figure 3d shows the CP of cessed on 11/06/2020)) [38]. Based on this methodology, potential sites of nucleation have different biocatalysts made under different conditions and compares them with the free been previously determined for different bioinorganic composites [13,39]. A great number enzyme. 2+As shown in all previous results, the β-gal nanocrystals produced at pH 9.0 of Ca nucleation sites were predicted for β-gal, which may explain the fast and efficient performed markedly better than those made at pH 7.4. The highest CP was obtained at formation of the bioinorganic composites. The predicted binding sites for β-gal are shown 290 mmolCaCl2 /gprotein with a value of 2.868 × 103 IU·h/gbiocatalyst , as compared to the in Figure 4 and the Table S2 (Supplementary Materials), and as expected many of them 245.3 × 103 IU·h/gbiocatalyst of the free enzyme, highlighting the importance of optimization involve amino acids usually found at calcium binding sites such as: Asp, Glu and Asn. It of the immobilization conditions (see Figure 3d). is also observed that no binding sites are located near the catalytic pocket, which supports 2.5. the high Prediction In Silico recoveredofenzyme activity Metal-Ion Bindingupon the biomineralization process. Regions Then, we were interested in understanding the effect of pH over the predicted nucle- In order to get insight into the biomineralization process, potential nucleation sites ation points. To that aim, we determined the protonation states of each residue of β-gal at for Ca2+ were identified based on the structural information of β-gal (PDB ID: 4IUG) and both pH 7.4 and 9.0 using the PlayMolecule repository [40]. The same nucleation sites a server for predicting metal ion binding sites (MIB at http://bioinfo.cmu.edu.tw/MIB/ were identified at both pHs, with no significant difference in their binding scores (Table (accessed on 11 June 2020)) [38]. Based on this methodology, potential sites of nucleation S2, Supplementary Materials). This is probably because none of the amino acids involved have been previously determined for different bioinorganic composites [13,39]. A great in calcium binding undergo changes in their protonation state between pH 7.4 and 9.0 number of Ca2+ nucleation sites were predicted for β-gal, which may explain the fast and [38,40], suggesting that the difference in biomineralization at varying pHs may rather be efficient formation of the bioinorganic composites. The predicted binding sites for β-gal are due shown to changes in Figurein4 the anddistribution the Table S2equilibrium of phosphate (Supplementary species Materials), (Hexpected and as 2PO4−; HPO42−). many Crystallization of them involve and precipitation amino acids usually foundofatcalcium calcium phosphates binding sitesissuch a complex as: Asp,phenomenon Glu and Asn.driven by both It is also thermodynamic observed and kinetic that no binding interactions sites are and located near theaffected bypocket, catalytic variables such as which composition, supersaturation, temperature, pH, ionic strength supports the high recovered enzyme activity upon the biomineralization process. and stoichiometric ratios [41]. Figure 4. Predicted nucleation sites of calcium ions on β-galactosidase using MIB server (green) and Figure 4. Predicted nucleation sites of calcium ions on β-galactosidase using MIB server (green) and active site of the enzyme (red). active site of the enzyme (red). Theweeffect Then, wereof pH is prevalent interested on the solution in understanding speciation the effect of pH overandthe the net charge predicted of the nucle- ation points. To that aim, we determined the protonation states of each residue of β-galions, mineral surface. In alkaline solutions, the mineral surface is negatively charged and suchpH at both as 7.4 H+ and and Ca 2+, bind to the surface forming complex structures [41]. Calcium phos- 9.0 using the PlayMolecule repository [40]. The same nucleation sites werephate speciation identified at bothhas been pHs, determined with to involve no significant no less difference than in their elevenscores binding variants, theS2, (Table most Supplementary Materials). This is probably because none of the amino acids involved in in- prominent of which are monocalcium phosphate, dicalcium phosphate, and apatites cluding calcium octacalcium binding undergophosphate, changes in amorphous calcium their protonation phosphate, state between pH hydroxyapatite and cal- 7.4 and 9.0 [38,40], cium deficient apatite [42]. Interactions with 3D-folded biomacromolecules suggesting that the difference in biomineralization at varying pHs may rather be due such asto β-gal add more complexity to the phenomenon, and thorough research changes in the distribution equilibrium of phosphate species (H2 PO4 ; HPO4 ). − must be 2 −conducted to elucidate the mechanisms involved, a task beyond the scope of this Crystallization and precipitation of calcium phosphates is a complex phenomenonwork. driven by both thermodynamic and kinetic interactions and affected by variables such Molecules 2021, 26, 4152 8 of 14 as composition, supersaturation, temperature, pH, ionic strength and stoichiometric ra- tios [41]. The effect of pH is prevalent on the solution speciation and the net charge of the mineral surface. In alkaline solutions, the mineral surface is negatively charged and ions, such as H+ and Ca2+ , bind to the surface forming complex structures [41]. Calcium phosphate speciation has been determined to involve no less than eleven variants, the most prominent of which are monocalcium phosphate, dicalcium phosphate, and apatites including octacalcium phosphate, amorphous calcium phosphate, hydroxyapatite and calcium deficient apatite [42]. Interactions with 3D-folded biomacromolecules such as Molecules 2021, 26, x FOR PEER REVIEW 8 of 13 β-gal add more complexity to the phenomenon, and thorough research must be conducted to elucidate the mechanisms involved, a task beyond the scope of this work. Characterization of 2.6. Characterization of the the β-galactosidase Nanocrystals β-galactosidase Nanocrystals SEM imagery SEM imagery was wastaken takenfor forsamples samplesofofthe theβ-gal β-gal nanocrystals nanocrystals made made at at pHpH9.09.0 forfor 60 60 min min andandat at different different CaCl CaCl 2 /protein 2/protein ratios ratios (Figure5).5).Most (Figure Moststudies studiesdescribe describe aa three-step process: in a first first stage, stage, metal metalionsionsbind bindtotoproteins proteinsvia viacoordination coordination groups groups forming formingnucle- nu- ation points for the metal-phosphate crystals; in a second growth cleation points for the metal-phosphate crystals; in a second growth stage, the protein- stage, the protein-metal- phosphate complexes metal-phosphate aggregate complexes into larger aggregate intopetal-like agglomerates, larger petal-like which finally agglomerates, whichdevelop finally into full multilayered flower-like structures in a third stage of anisotropic develop into full multilayered flower-like structures in a third stage of anisotropic growth growth [9,43]. Many Many [9,43]. studies have looked studies have lookedinto the intokinetics of the the kinetics of process, the process,confirming confirming thethe presence of presence irregular of irregularnanocrystals nanocrystals atat very veryearly earlystages stages(0.1 (0.1hhtoto22h),h),followed followed by by flower-mimicking flower-mimicking nanostructures with nanostructures with increasing increasingincubation incubationtimetime(3(3hhtoto2424h),h),yielding yieldingcompact compact nanoflowers nanoflow- ers after about 24 h, with the time-frames varying in each particular case [44,45]. Figure after about 24 h, with the time-frames varying in each particular case [44,45]. Figure 55 shows a development process as a function shows a development process as a function of the CaCl of the CaCl /protein ratio. As in the kinetics 2 2/protein ratio. As in the kinetics of the process, Figure 5a–c shows the effect of calcium/protein ratio of the process, Figure 5a–c shows the effect of calcium/protein ratio on the morphology of on the morphology of the bioinorganic the bioinorganic composites. composites. While While aa ratio ratio of of 115 115 mmol mmolCaCl /gprotein CaCl22/gprotein forms forms nanoparticles, nanoparticles, a a much higher ratio (460 mmol CaCl /g protein ) forms flower-like much higher ratio (460 mmolCaCl2/gprotein) forms flower-like structures. Rectangular micro- 2 structures. Rectangular microcrystal crystal of approximately of approximately 2 × 102 µm× 10were µm were foundfound at intermediate at intermediate ratio,ratio, indicating indicating thatthat the the CaCl2 /protein clearly influences the hierarchical assembly of bioinorganic nanocrystals. CaCl2/protein clearly influences the hierarchical assembly of bioinorganic nanocrystals. Other structures, beside nanoflowers, have been reported for calcium-based nanocrystals Other structures, beside nanoflowers, have been reported for calcium-based nanocrystals as those found in this work [46]. The stability tests in Figure 3 suggest that the higher as those found in this work [46]. The stability tests in Figure 3 suggest that the higher structural complexity associated with higher CaCl2 /protein ratios also plays a role in structural complexity associated with higher CaCl2/protein ratios also plays a role in en- enhancing stability, possibly due to bigger and more robust structures exerting a protective hancing stability, possibly due to bigger and more robust structures exerting a protective effect on the enzyme. effect on the enzyme. Figure 5. SEM photography of the calcium-phosphate β-galactosidase nanocrystals produced at pH 9.0 for 60 min at Figure 5. SEM photography of the calcium-phosphate β-galactosidase nanocrystals produced at pH 9.0 for 60 min at CaCl2/protein ratios of 115 mmolCaCl2/gprotein (a), 290 mmolCaCl2/gprotein (b), and 460 mmolCaCl2/gprotein (c). CaCl2 /protein ratios of 115 mmolCaCl2 /gprotein (a), 290 mmolCaCl2 /gprotein (b), and 460 mmolCaCl2 /gprotein (c). 2.7. Synthesis of Galacto-Oligosaccharides 2.7. Synthesis of Galacto-Oligosaccharides The bioinorganic β-gal nanocrystals were tested as catalysts in the synthesis of GOS. The bioinorganic β-gal nanocrystals were tested as catalysts in the synthesis of GOS. The results in terms of percentage of total saccharides can be seen in Figure 6. GOS of The results in terms of percentage of total saccharides can be seen in Figure 6. GOS varying length, namely GOS-3, GOS-4 and GOS-5 were considered as total GOS. In Figure of varying length, namely GOS-3, GOS-4 and GOS-5 were considered as total GOS. In 6, the profiles of the hydrolysis products glucose and galactose are also represented, and Figure 6, the profiles of the hydrolysis products glucose and galactose are also represented, the remaining lactose. The kinetics of GOS synthesis with free β-gal (Figure 6a) and with β-gal nanocrystals (Figure 6b) mirror each other almost exactly, both reaching a maximum GOS yield (30%) at about 40 min. These results agreed with those obtained by Vera et al. (2012) at the same conditions using the same soluble β-gal from A. oryzae. GOS conversion yields obtained with β-gal from A. oryzae using other immobilization methods usually Molecules 2021, 26, 4152 9 of 14 and the remaining lactose. The kinetics of GOS synthesis with free β-gal (Figure 6a) and Molecules 2021, 26, x FOR PEER REVIEW β-gal nanocrystals (Figure 6b) mirror each other almost exactly, both reaching with a 9 of 13 maximum GOS yield (30%) at about 40 min. These results agreed with those obtained by Vera et al. (2012) at the same conditions using the same soluble β-gal from A. oryzae. GOS conversion yields obtained with β-gal from A. oryzae using other immobilization reduces productivity, therefore compromising the competitiveness of these biocatalysts methods usually range from 15 to 27% [47], which highlights the good performance of the and sets out serious concerns about their potential use as catalysts for industrial processes. biocatalysts herein presented. These are remarkable results since immobilized enzymes Theshow usually β-galpoorer nanocrystals showed performance thannothe signs free of diffusional enzymes limitations due to diffusionaland no negative limitations [48]. impact This supports the notion that nanoflowers, or nanocrystals, are not subjected toofsignificant on productivity. Furthermore, considering the high catalytic potential the β-gal nanocrystals, mass transfer as reflecteddue limitations on atoSF value their of reactive large 11.6, it isarea, safesuggesting to conclude that that thethe biocatalyst vast majority developed of the immobilized enzymes are highly exposed to the bulk of the reaction. Thesynthe- has promising prospects for GOS synthesis, and possibly for lactulose lower sis and lactose performance ofhydrolysis, immobilized asbiocatalysts well. Additionally, thetobiocompatibility in relation of calcium the soluble enzymes opensthat is a feature up multiple applications for reduces productivity, this type therefore of biocatalystthe compromising in competitiveness the areas of functional foods of these and bio- biocatalysts medicine [49,50]. and sets out serious concerns about their potential use as catalysts for industrial processes. 6. Kinetics Figure 6. Kineticsofofsynthesis synthesisofofgalacto-oligosaccharides galacto-oligosaccharides with with freefree (a) and (a) and bioinorganic bioinorganic nanocrystals nanocrystals (b) of(b) of β-galacto- β-galactosidase. sidase. Synthesis Synthesis conditions: conditions: 100 IU/g100 IU/glactose, lactose 50% w/w in citrate-phosphate buffer pH 4.5 ◦and 50 °C. Data are pre- lactose , lactose 50% w/w in citrate-phosphate buffer pH 4.5 and 50 C. Data are presented as sented±asmargin mean mean ±ofmargin error, nof= error, 2. n = 2. 3. Materials andnanocrystals The β-gal Methods showed no signs of diffusional limitations and no negative impact on productivity. Furthermore, considering the high catalytic potential of the β-gal 3.1. Chemicals and Reagents nanocrystals, as reflected on a SF value of 11.6, it is safe to conclude that the biocatalyst Commercial β-galactosidase Enzeco Fungal Lactase from A. oryzae (β-gal) (EC developed has promising prospects for GOS synthesis, and possibly for lactulose synthesis 3.2.1.23) with a specific activity (SA) of 152,361 ± 4755 IU/gprotein was supplied by Enzyme and lactose hydrolysis, as well. Additionally, the biocompatibility of calcium opens up Development Corporation, EDC (New York, NY, USA). All chemicals were purchased multiple applications for this type of biocatalyst in the areas of functional foods and from Sigma Aldrich (St. Louis, MO, USA) at the highest purity available. biomedicine [49,50]. 3.2. Hydrolyticand 3. Materials Activity of β-galactosidase Methods One international 3.1. Chemicals unit of hydrolytic activity of β-galactosidase (IU) was defined as and Reagents the amount Commercial β-galactosidase1Enzeco of enzyme releasing µmol of o-NPLactase Fungal per minfrom as a A. result of the oryzae hydrolysis (β-gal) of o- (EC 3.2.1.23) NPG with aatspecific pH 7.0 activity and 40 °C. (SA)Molar extinction of 152,361 coefficient ± 4755 of o-NP IU/gprotein was was 2.21 mM supplied −1 cm at 420 by Enzyme −1 De- nm. velopment Corporation, EDC (New York, NY, USA). All chemicals were purchased from Sigma Aldrich (St. Louis, MO, USA) at the highest purity available. 3.3. Biomineralization of β-galactosidase 3.2. Hydrolytic Activity of β-Galactosidase The biomineralization procedure was based on the method reported by Escobar et al. [13] where 400 µL of aunit One international metal salt solution of hydrolytic (CuSO activity , MgSO4, ZnCl2(IU) of 4β-galactosidase , CoCl was 2, FeSO 4, CaCl defined 2) as the were added under magnetic stirring to 5 mL of a 1 mg/mL enzyme solution amount of enzyme releasing 1 µmol of o-NP per min as a result of the hydrolysis of o-NPGin phosphate- buffered at pH 7.0 saline ◦ C. Molar and 40(PBS) madeextinction from 50 mM sodiumofphosphate coefficient 150 − and mM o-NP was 2.21 mM −1 at 420 1 cmNaCl at room nm. temperature. Biomineralization time (min, t), pH of the PBS solution, and metal/protein ratio (mmol of metal salt/g of contacted protein, R) were established as required by the experimental conditions under study (details in Supplementary Materials). The resulting biomineral suspension was subjected to centrifugation at 10,000 rpm for 15 min and then rinsed and resuspended in 5 mL of 25 mM potassium phosphate buffer pH 7.0, repeating the process two further times. The β-gal nanocrystals were then stored at 4 °C in rinsing buffer. Molecules 2021, 26, 4152 10 of 14 3.3. Biomineralization of β-Galactosidase The biomineralization procedure was based on the method reported by Escobar et al. [13] where 400 µL of a metal salt solution (CuSO4 , MgSO4 , ZnCl2 , CoCl2 , FeSO4 , CaCl2 ) were added under magnetic stirring to 5 mL of a 1 mg/mL enzyme solution in phosphate-buffered saline (PBS) made from 50 mM sodium phosphate and 150 mM NaCl at room temperature. Biomineralization time (min, t), pH of the PBS solution, and metal/protein ratio (mmol of metal salt/g of contacted protein, R) were established as required by the experimental conditions under study (details in Supplementary Materials). The resulting biomineral suspension was subjected to centrifugation at 10,000 rpm for 15 min and then rinsed and resuspended in 5 mL of 25 mM potassium phosphate buffer pH 7.0, repeating the process two further times. The β-gal nanocrystals were then stored at 4 ◦ C in rinsing buffer. All results were evaluated in terms of specific activity (SA) and immobilization yield (IY) both defined according to: SA = A E /Mb ·100 (1) IY = A E /A I ·100 (2) where AE is the total activity expressed by the enzyme or immobilized biocatalyst (IU), Mb is the dry weight of the biocatalyst (g), and AI is the initial activity subjected to immobilization (IU). SA was expressed in international units of hydrolytic activity per g of protein in the case of free β-gal (IU/gprotein ), and in international units of hydrolytic activity per g of biocatalyst in the case of the β-gal bioinorganic nanocrystals (IU/gbiocatalyst ). 3.4. Thermal Stability of the β-Galactosidase Bioinorganic Nanocrystals A selection of the bioinorganic β-gal nanocrystals developed and free β-gal were subjected to thermal inactivation at 50 ◦ C in 100 mM PBS pH 7.0. The resulting kinetics of inactivation were modeled according to a series biphasic inactivation mechanism without residual activity [37]. The model equations are described in Supporting Information. Additionally, half-life time (t1/2 ), stabilization factor (SF) and catalytic potential (CP) were determined for the β-gal nanocrystals from the thermal inactivation kinetics. The description and equations for these parameters is presented in Supplementary Materials. 3.5. In Silico Prediction of Calcium Binding Sites The possible sites of nucleation of β-gal (PDB: 4IUG) were determined using the fragment transformation method available as a public server at MIB (http://bioinfo.cmu. edu.tw/MIB/introduction (accessed on 11 June 2020)) [38]. The protonation states of each residue at pH 7.4 and 9.0 was determined using the ProteinPrepare at PlayMolecule repository [40]. Both protonation states were evaluated in MIB server for the prediction of metal binding sites for Ca2+ . 3.6. Characterization of the β-Galactosidase Bioinorganic Nanocrystals The calcium-phosphate β-gal nanocrystals were analyzed using scanning electron microscopy (SEM) using a HITACHI SU 3500 instrument. The samples were previously washed with distilled water to remove unspecific crystals from the phosphate buffer and then sputtered with gold to minimize surface charging. 3.7. Synthesis of Galacto-Oligosaccharides The developed immobilized biocatalyst was tested in the synthesis of GOS along with the free β-gal for comparative purposes. The procedure and reaction conditions were based on the work by Vera et al. [51]. The reaction assays were conducted in duplicates in 150 mL Erlenmeyer flasks at 50 ◦ C under magnetic stirring. Lactose monohydrate was added at 50% w/w (12.5 g) in 100 mM citrate-phosphate buffer pH 4.5 dissolving it by heating at a temperature close to 100 ◦ C. The reaction was initiated once the free β-gal solution or Molecules 2021, 26, 4152 11 of 14 β-gal biocatalyst was added at a ratio of 100 IU/glactose . The products of the reactions were analyzed by HPLC in a Jasco RI 2031 delivery system coupled with a refractive index detector using a BP-100 Ca2+ column (300 mm × 7.8 mm) (Benson Polymeric, Reno, NV, USA) and milli-Q water as mobile phase [51]. 4. Conclusions This study reports the successful immobilization of A. oryzae β-galactosidase as hybrid bioinorganic nanocrystals. Calcium was determined to be the most adequate metal ion for β-gal biomineralization, outperforming copper by five times in terms of immobilization yield, whereas almost negligible results were obtained with the other ions tested. Condi- tions such as metal/protein ratio and pH proved to be highly relevant. CaCl2 /protein ratio showed an optimum of both immobilization yield and specific activity in the vicinity of 290 mmolCaCl2 /gprotein . The metal/protein ratio was also influential on the hierarchical assembling of the nanocrystals with larger, more complex and more robust structures at higher metal/protein ratios, as confirmed by SEM and the stability tests. Particularly signif- icant was the effect of pH, allowing an increase in the immobilization yield from 0 to 82% in the pH range from 6.0 to 9.0, clearly demonstrating the high importance of optimization of the immobilization conditions in the development of this types of nanobiocatalyst. The in silico analysis showed no correlation between the predicted nucleation points and shifts in the ionization state of critical reactive groups as a consequence of pH change. Although such correlation cannot be completely dismissed, the effect of pH might be rather related to variations of calcium phosphate speciation. The developed biocatalyst showed a specific activity of 35,000 IU/gbiocatalyst , and a stability factor of 11.6 (at 50 ◦ C), surpassing the results reported to date for other immobi- lized β-gal in which a compromise usually occurs between both parameters. Furthermore, the bioinorganic β-gal nanocrystals were successfully tested for the synthesis of GOS per- forming as well as the free enzymes in terms of yield and volumetric productivity, showing therefore, no signs of diffusional limitations. These features make the β-gal nanocrystals highly competitive biocatalysts for the synthesis of GOS and for other important industrial and biomedical applications. In short, β-galactosidase immobilization as bioinorganic nanocrystals was shown to be a remarkably simple cost-efficient methodology that allows high retention of enzyme activity, producing a biocatalyst with high specific activity, high stability, and optimal performance of GOS synthesis from lactose. Furthermore, because calcium is safe and compatible as a food and pharmaceutical additive, multiple applications open up for this biocatalyst in the production of prebiotics and drug-delivery. Supplementary Materials: Figure S1. Protein immobilization yield (IYP ) of calcium-phosphate nanocrystals of β-galactosidase produced at pH 7.4 for biomineralization times of 30 min (T30), 60 min (T60), and 120 min (T120), and salt/protein ratios (R) of 115, 230, 290, 345, 400 and 460 mmolCaCl2 /gprotein ; Table S1. Immobilization yield of β-galactosidase bioinorganic nanocrystals using several metal ions. Biomineralization conditions: PBS pH 7.4, 230 mmolsalt /gprotein , for 2 h at room temperature; Table S2. Binding residues and scores predicted for Ca+2 by the MIB server; Table S3. Results of the 32 full factorial experimental design for the biomineralization of β-galacto- sidase with calcium-phosphate. Variables: time (t), and salt/protein ratio (R). Response parame-ters: specific activity (SA) and immobilization yield (IY); Table S4. ANOVA test of the 32 full factorial experimental design for the biomineralization of β-galactosidase with calcium-phosphate. Variables: time (t), and salt/protein ratio (R). Response pa-rameter: specific activity (SA); Table S5. ANOVA test of the 32 full factorial experimental design for the biomineralization of β-galactosidase with calcium-phosphate. Variables: time (t), and salt/protein ratio (R). Response pa-rameter: immo- bilization yield (IY); Table S6. Parameters of the models developed for the immobilization yield (IY) and specific activ-ity (SA) of the biomineralization of β-galactosidase with calcium-phosphate according to Equation (S4); Table S7. Parameters, p-values, and coefficients of determination (R2 ) of the models developed for the immobilization yield (IY) and specific activity (SA) resulting from the biomineralization of β-galactosidase at pH 9.0 for 60 min. The models are represented by Molecules 2021, 26, 4152 12 of 14 Equation (S5); Table S8. Model parameters, p-values, and coefficients of determination of the kinetic models of inactivation of β-galactosidase nanocrystals produced at pH 7.4 and pH 9.0 for 60 min of bio-mineralization at calcium/protein ratios of 115 mmolCaCl2 /gprotein (R1), 290 mmolCaCl2 /gprotein (R3), and 460 mmolCaCl2 /gprotein (R6). The models correspond to the biphasic inactivation mech- anism as represented by Equation (S6). Equation (S1): Multiple linear regression; Equation (S2): Regression models for the experimental designs. Author Contributions: L.T.: investigation, writing—original draft; O.R.: investigation, visualization, writing—review and editing. C.A.: investigation; F.L.-G.: conceptualization; writing—review and editing; A.I.: writing—review and editing; L.W.: funding acquisition, supervision, writing—review and editing. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by FONDECYT-Chile (grant number 1190709). The authors thanks María Villarroel and FONDEQUIP-CONICYT number 140079 for technical support in SEM analysis. LT acknowledges funding from PUCV for doctoral fellowship and FLG acknowledges the funding of IKERBASQUE. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Conflicts of Interest: The authors declare no conflict of interest. Sample Availability: Samples of the compounds are not available from the authors. References 1. Basso, A.; Serban, S. Industrial applications of immobilized Enzymes—A review. Mol. Catal. 2019, 479, 110607. [CrossRef] 2. Cowan, D.A.; Fernandez-Lafuente, R. Enhancing the functional properties of thermophilic enzymes by chemical modification and immobilization. Enzym. Microb. Technol. 2011, 49, 326–346. [CrossRef] 3. Godoy, C.A.; Romero, O.; De Las Rivas, B.; Mateo, C.; Fernandez-Lorente, G.; Guisan, J.M.; Palomo, J.M. Changes on enantioselec- tivity of a genetically modified thermophilic lipase by site-directed oriented immobilization. J. Mol. Catal. B Enzym. 2013, 87, 121–127. [CrossRef] 4. Sheldon, R.A. Cleas, combi-cleas and ‘smart’ magnetic cleas: Biocatalysis in a bio-based economy. Catalysts 2019, 9, 261. [CrossRef] 5. Misson, M.; Zhang, H.; Jin, B. Nanobiocatalyst advancements and bioprocessing applications. J. R. Soc. Interface 2015, 12, 20140891. [CrossRef] 6. Cipolatti, E.P.; Valério, A.; Henriques, R.O.; Moritz, D.E.; Ninow, J.L.; Freire, D.M.G.; Manoel, E.A.; Fernandez-Lafuente, R.; De Oliveira, D. Nanomaterials for biocatalyst immobilization-state of the art and future trends. RSC Adv. 2016, 6, 104675–104692. [CrossRef] 7. Ge, J.; Lei, J.; Zare, R.N. Protein-inorganic hybrid nanoflowers. Nat. Nanotechnol. 2012, 7, 428–432. [CrossRef] [PubMed] 8. Wu, X.; Hou, M.; Ge, J. Metal-organic frameworks and inorganic nanoflowers: A type of emerging inorganic crystal nanocarrier for enzyme immobilization. Catal. Sci. Technol. 2015, 5, 5077–5085. [CrossRef] 9. Lee, S.W.; Cheon, S.A.; Kim, M.I.; Park, T.J. Organic-inorganic hybrid nanoflowers: Types, characteristics, and future prospects. J. Nanobiotechnol. 2015, 13. [CrossRef] [PubMed] 10. Al Zoubi, W.; Kamil, M.P.; Fatimah, S.; Nisa, N.; Ko, Y.G. Recent advances in hybrid organic-inorganic materials with spatial architecture for state-of-the-art applications. Prog. Mater. Sci. 2020, 112, 100663. [CrossRef] 11. Bilal, M.; Asgher, M.; Shah, S.Z.H.; Iqbal, H.M.N. Engineering enzyme-coupled hybrid nanoflowers: The quest for optimum performance to meet biocatalytic challenges and opportunities. Int. J. Biol. Macromol. 2019, 135, 677–690. [CrossRef] 12. Shende, P.; Kasture, P.; Gaud, R.S. Nanoflowers: The future trend of nanotechnology for multi-applications. Artif. Cells Nanomed. Biotechnol. 2018, 46, 413–422. [CrossRef] 13. Escobar, S.; Velasco-Lozano, S.; Lu, C.H.; Lin, Y.F.; Mesa, M.; Bernal, C.; López-Gallego, F. Understanding the functional properties of bio-inorganic nanoflowers as biocatalysts by deciphering the metal-binding sites of enzymes. J. Mater. Chem. B 2017, 5, 4478–4486. [CrossRef] 14. Zhang, B.; Li, P.; Zhang, H.; Fan, L.; Wang, H.; Li, X.; Tian, L.; Ali, N.; Ali, Z.; Zhang, Q. Papain/Zn3 (PO4 )2 hybrid nanoflower: Preparation, characterization and its enhanced catalytic activity as an immobilized enzyme. RSC Adv. 2016, 6, 46702–46710. [CrossRef] 15. Cui, J.; Jia, S. Organic–inorganic hybrid nanoflowers: A novel host platform for immobilizing biomolecules. Coord. Chem. Rev. 2017, 352, 249–263. [CrossRef] 16. Liu, Y.; Ji, X.; He, Z. Organic-inorganic nanoflowers: From design strategy to biomedical applications. Nanoscale 2019, 11, 17179–17194. [CrossRef] [PubMed] Molecules 2021, 26, 4152 13 of 14 17. Araya, E.; Urrutia, P.; Romero, O.; Illanes, A.; Wilson, L. Design of combined crosslinked enzyme aggregates (combi-CLEAs) of β-galactosidase and glucose isomerase for the one-pot production of fructose syrup from lactose. Food Chem. 2019, 288, 102–107. [CrossRef] 18. Misson, M.; Dai, S.; Jin, B.; Chen, B.H.; Zhang, H. Manipulation of nanofiber-based β-galactosidase nanoenvironment for enhancement of galacto-oligosaccharide production. J. Biotechnol. 2016, 222, 56–64. [CrossRef] 19. Banjanac, K.; Carević, M.; Ćorović, M.; Milivojević, A.; Prlainović, N.; Marinković, A.; Bezbradica, D. Novel β-galactosidase nanobiocatalyst systems for application in the synthesis of bioactive galactosides. RSC Adv. 2016, 6, 97216–97225. [CrossRef] 20. Serey, M.; Vera, C.; Guerrero, C.; Illanes, A. Immobilization of Aspergillus oryzae β-galactosidase in cation functionalized agarose matrix and its application in the synthesis of lactulose. Int. J. Biol. Macromol. 2021, 167, 1564–1574. [CrossRef] 21. Ureta, M.M.; Martins, G.N.; Figueira, O.; Pires, P.F.; Castilho, P.C.; Gomez-Zavaglia, A. Recent advances in β-galactosidase and fructosyltransferase immobilization technology. Crit. Rev. Food Sci. Nutr. 2020. [CrossRef] 22. Liu, Y.; Wang, B.; Ji, X.; He, Z. Self-assembled protein-enzyme nanoflower-based fluorescent sensing for protein biomarker. Anal. Bioanal. Chem. 2018, 410, 7591–7598. [CrossRef] [PubMed] 23. Talens-Perales, D.; Fabra, M.J.; Martínez-Argente, L.; Marín-Navarro, J.; Polaina, J. Recyclable thermophilic hybrid protein- inorganic nanoflowers for the hydrolysis of milk lactose. Int. J. Biol. Macromol. 2020, 151, 602–608. [CrossRef] [PubMed] 24. Fabra, M.J.; Talens-Perales, D.; Roman-Sarmiento, A.; López-Rubio, A.; Polaina, J. Effect of biopolymer matrices on lactose hydrolysis by enzymatically active hydrogel and aerogels loaded with β-galactosidase nanoflowers. Food Hydrocoll. 2021, 111, 106220. [CrossRef] 25. Lei, Z.; Gao, C.; Chen, L.; He, Y.; Ma, W.; Lin, Z. Recent advances in biomolecule immobilization based on self-assembly: Organic-inorganic hybrid nanoflowers and metal-organic frameworks as novel substrates. J. Mater. Chem. B 2018, 6, 1581–1594. [CrossRef] [PubMed] 26. Somturk, B.; Yilmaz, I.; Altinkaynak, C.; Karatepe, A.; Özdemir, N.; Ocsoy, I. Synthesis of urease hybrid nanoflowers and their enhanced catalytic properties. Enzym. Microb. Technol. 2016, 86, 134–142. [CrossRef] 27. Li, Y.; Fei, X.; Liang, L.; Tian, J.; Xu, L.; Wang, X.; Wang, Y. The influence of synthesis conditions on enzymatic activity of enzyme-inorganic hybrid nanoflowers. J. Mol. Catal. B Enzym. 2016, 133, 92–97. [CrossRef] 28. Guerrero, C.; Vera, C.; Serna, N.; Illanes, A. Immobilization of Aspergillus oryzae B-galactosidase in an agarose matrix functionalized by four different methods and application to the synthesis of lactulose. Bioresour. Technol. 2017, 232, 53–63. [CrossRef] 29. Urrutia, P.; Bernal, C.; Wilson, L.; Illanes, A. Use of chitosan heterofunctionality for enzyme immobilization: β-galactosidase immobilization for galacto-oligosaccharide synthesis. Int. J. Biol. Macromol. 2018, 116, 182–193. [CrossRef] 30. Huerta, L.M.; Vera, C.; Guerrero, C.; Wilson, L.; Illanes, A. Synthesis of galacto-oligosaccharides at very high lactose concentrations with immobilized β-galactosidases from Aspergillus oryzae. Process Biochem. 2011, 46, 245–252. [CrossRef] 31. Ansari, S.; Husain, Q. Immobilization of Aspergillus oryzae β galactosidase on concanavalin A-layered calcium alginate-cellulose beads and its application in lactose hydrolysis in continuous spiral bed reactors. Pol. J. Chem. Technol. 2011, 13, 15–20. [CrossRef] 32. Gaur, R.; Pant, H.; Jain, R.; Khare, S.K. Galacto-oligosaccharide synthesis by immobilized Aspergillus oryzae β-galactosidase. Food Chem. 2006, 97, 426–430. [CrossRef] 33. Albayrak, N.; Yang, S.T. Immobilization of β-galactosidase on fibrous matrix by polyethyleneimine for production of galacto- oligosaccharides from lactose. Biotechnol. Prog. 2002, 18, 240–251. [CrossRef] [PubMed] 34. Souza, C.J.F.; Garcia-Rojas, E.E.; Souza, C.S.F.; Vriesmann, L.C.; Vicente, J.; de Carvalho, M.G.; Petkowicz, C.L.O.; Favaro-Trindade, C.S. Immobilization of β-galactosidase by complexation: Effect of interaction on the properties of the enzyme. Int. J. Biol. Macromol. 2019, 122, 594–602. [CrossRef] [PubMed] 35. Guerrero, C.; Aburto, C.; Súarez, S.; Vera, C.; Illanes, A. Improvements in the production of Aspergillus oryzae β-galactosidase crosslinked aggregates and their use in repeated-batch synthesis of lactulose. Int. J. Biol. Macromol. 2020, 142, 452–462. [CrossRef] 36. Guerrero, C.; Vera, C.; Araya, E.; Conejeros, R.; Illanes, A. Repeated-batch operation for the synthesis of lactulose with β- galactosidase immobilized by aggregation and crosslinking. Bioresour. Technol. 2015, 190, 122–131. [CrossRef] [PubMed] 37. Illanes, A.; Wilson, L. Parameters for the evaluation of immobilized enzymes under process conditions. Methods Mol. Biol. 2020, 2100, 65–81. 38. Lin, Y.F.; Cheng, C.W.; Shih, C.S.; Hwang, J.K.; Yu, C.S.; Lu, C.H. MIB: Metal ion-binding site prediction and docking server. J. Chem. Inf. Model. 2016, 56, 2287–2291. [CrossRef] [PubMed] 39. Aires, A.; Llarena, I.; Moller, M.; Castro-Smirnov, J.; Cabanillas-Gonzalez, J.; Cortajarena, A.L. A simple approach to design proteins for the sustainable synthesis of metal nanoclusters. Angew. Chem. Int. Ed. 2019, 58, 6214–6219. [CrossRef] 40. Martínez-Rosell, G.; Giorgino, T.; De Fabritiis, G. PlayMolecule ProteinPrepare: A web application for protein preparation for molecular dynamics simulations. J. Chem. Inf. Model. 2017, 57, 1511–1516. [CrossRef] 41. Orme, C.A.; Giocondi, J.L. Model systems for formation and dissolution of calcium phosphate minerals. Handb. Biominer.: Biol. Asp. Struct. Form. 2008, 2, 135–157. 42. Tung, M.S. Calcium phosphates: Structure, composition, solubility, and stability. In Calcium Phosphates in Biological and Industrial Systems; Amjad, Z., Ed.; Springer: Boston, MA, USA, 1998; pp. 1–19. [CrossRef] 43. Kim, K.H.; Jeong, J.M.; Lee, S.J.; Choi, B.G.; Lee, K.G. Protein-directed assembly of cobalt phosphate hybrid nanoflowers. J. Colloid Interface Sci. 2016, 484, 44–50. [CrossRef] [PubMed] Molecules 2021, 26, 4152 14 of 14 44. Sun, J.; Ge, J.; Liu, W.; Lan, M.; Zhang, H.; Wang, P.; Wang, Y.; Niu, Z. Multi-enzyme co-embedded organic-inorganic hybrid nanoflowers: Synthesis and application as a colorimetric sensor. Nanoscale 2014, 6, 255–262. [CrossRef] [PubMed] 45. Zhu, X.; Huang, J.; Liu, J.; Zhang, H.; Jiang, J.; Yu, R. A dual enzyme-inorganic hybrid nanoflower incorporated microfluidic paper- based analytic device (µPAD) biosensor for sensitive visualized detection of glucose. Nanoscale 2017, 9, 5658–5663. [CrossRef] [PubMed] 46. Wang, L.B.; Wang, Y.C.; He, R.; Zhuang, A.; Wang, X.; Zeng, J.; Hou, J.G. A new nanobiocatalytic system based on allosteric effect with dramatically enhanced enzymatic performance. J. Am. Chem. Soc. 2013, 135, 1272–1275. [CrossRef] 47. Vera, C.; Córdova, A.; Aburto, C.; Guerrero, C.; Suárez, S.; Illanes, A. Synthesis and purification of galacto-oligosaccharides: State of the art. World J. Microbiol. Biotechnol. 2016, 32, 197. [CrossRef] [PubMed] 48. Guerrero, C.; Aburto, C.; Suárez, S.; Vera, C.; Illanes, A. Effect of the type of immobilization of β-galactosidase on the yield and selectivity of synthesis of transgalactosylated oligosaccharides. Biocatal. Agric. Biotechnol. 2018, 16, 353–363. [CrossRef] 49. Hoyos, P.; Bavaro, T.; Perona, A.; Rumbero, A.; Tengattini, S.; Terreni, M.; Hernaíz, M.J. Highly efficient and sustainable synthesis of neoglycoproteins using galactosidases. ACS Sustain. Chem. Eng. 2020, 8, 6282–6292. [CrossRef] 50. Zhang, Y.; Zhong, Q. Freeze-dried capsules prepared from emulsions with encapsulated lactase as a potential delivery system to control lactose hydrolysis in milk. Food Chem. 2018, 241, 397–402. [CrossRef] 51. Vera, C.; Guerrero, C.; Conejeros, R.; Illanes, A. Synthesis of galacto-oligosaccharides by β-galactosidase from Aspergillus oryzae using partially dissolved and supersaturated solution of lactose. Enzym. Microb. Technol. 2012, 50, 188–194. [CrossRef]