Dependence of the Ripeness Stage on the Antioxidant and Antimicrobial Properties of Walnut (Juglans regia L.) Green Husk Extracts from Industrial By-Products

Walnut green husk (WGH) is a waste generated by the walnut (Juglans regia L.) harvest industry. It represents a natural source of polyphenols, compounds with antioxidant and antimicrobial activities, but their activity could be dependent on the ripeness stage of the raw material. In this study, the effect of the different ripeness stages—open (OH) and closed (CH) husks—on the antioxidant and antimicrobial properties of WGH extracts were analyzed, emphasizing the influence of the extracts in inhibiting Escherichia coli growth. The ripeness stage of WGH significantly affected the antioxidant activity of the extracts. This was attributed to the different polyphenol profiles related to the mechanical stress when the husk opened compared to the closed sample. The antimicrobial activity showed inhibition of E. coli growth. OH-extracts at 96 µg/mL caused the lowest specific growth rate (µmax = 0.003 h−1) and the greatest inhibition percentage (I = 93%) compared to CH-extract (µmax = 0.01 h−1; I = 69%). The obtained results showed the potential of the walnut green husk, principally open husk, as an economical source of antioxidant and antimicrobial agents with potential use in the food industry.


Introduction
The walnut industry has had a steady growth worldwide, in terms of the volumes traded and the surface area dedicated to this crop. Chile ranks second among the leading exporters of walnuts worldwide [1]. However, this industry generates a large amount of waste that mainly corresponds to the walnut green husk (exocarp and mesocarp), equivalent to approximately 20% of the total walnut production [2]. Walnut green husk is a waste produced during the harvest season and currently has few uses. One of them is as a fertilizer because it is a tissue rich in organic matter. However, it is also used as a coloring for woods or a replacement for Chinese ink [3].
Several studies have shown that walnut green husk is rich in polyphenolic compounds with an antimicrobial and antioxidant capacity [3][4][5]. Therefore, walnut green husks represent a natural source of active compounds (polyphenols) and are ideal for extracting exposure of the mesocarp to the environmental conditions while it was in the tree because the husk is opening (Figure 1a) [19].  44.700 a ± 0.018 44.800 a ± 0.027 Non-nitrogen extract 18.918 a ± 0.051 17.243 a ± 1.806 OH: ripe walnuts green husk (open husk); CH: unripe walnuts green husk (closed husk). * Different letters (a, b) indicate significant differences (p < 0.05) between samples.

Antioxidant Capacity
Studies in the literature had previously reported that walnut green husks showed antioxidant capacity, which was attributed to the presence of polyphenolic compounds [3][4][5]. Nevertheless, this property has been scarcely studied comparing different ripeness stages and/or among fruits/vegetables of habitual consumption. This comparison is necessary to evaluate the technological viability and value-added of walnut green husk. Therefore, antioxidant capacity was compared between ripeness stages (OH and CH) and with fruits of known antioxidant activity (INTA database), using the ORAC methodology [20].
The obtained results showed the higher antioxidant capacity of the OH sample (47.94 µmol Trolox/g sample dry weight). This value was similar to the antioxidant capacities of dried fig (45.7 µmol Trolox/g sample dry weight), princess grape (43.12 µmol Trolox/g sample dry weight), and fresh coconut (39.48 µmol Trolox/g sample dry weight). In comparison, the antioxidant capacity of the CH sample (34.97 µmol Trolox/g sample dry weight) was comparable to an edranol avocado (36.55 µmol Trolox/g sample dry weight) and roasted peanuts (28.43 µmol Trolox/g sample dry weight) [21]. These results are in agreement with the work of Laroze et al. [22]. They indicated that the external parts of fruits, vegetables, and other lignocellulosic parts such as skins, husks, and seeds have a higher content of antioxidant phenolic compounds than the corresponding internal fractions (pulp of fruits). The difference in antioxidant activity between the ripeness stages could be attributed to different phenolic compound profiles (Table 2).

Antioxidant Capacity
Studies in the literature had previously reported that walnut green husks showed antioxidant capacity, which was attributed to the presence of polyphenolic compounds [3][4][5]. Nevertheless, this property has been scarcely studied comparing different ripeness stages and/or among fruits/vegetables of habitual consumption. This comparison is necessary to evaluate the technological viability and value-added of walnut green husk. Therefore, antioxidant capacity was compared between ripeness stages (OH and CH) and with fruits of known antioxidant activity (INTA database), using the ORAC methodology [20].
The obtained results showed the higher antioxidant capacity of the OH sample (47.94 µmol Trolox/g sample dry weight). This value was similar to the antioxidant capacities of dried fig (45.7 µmol Trolox/g sample dry weight), princess grape (43.12 µmol Trolox/g sample dry weight), and fresh coconut (39.48 µmol Trolox/g sample dry weight). In comparison, the antioxidant capacity of the CH sample (34.97 µmol Trolox/g sample dry weight) was comparable to an edranol avocado (36.55 µmol Trolox/g sample dry weight) and roasted peanuts (28.43 µmol Trolox/g sample dry weight) [21]. These results are in agreement with the work of Laroze et al. [22]. They indicated that the external parts of fruits, vegetables, and other lignocellulosic parts such as skins, husks, and seeds have a higher content of antioxidant phenolic compounds than the corresponding internal fractions (pulp of fruits). The difference in antioxidant activity between the ripeness stages could be attributed to different phenolic compound profiles (Table 2). These results indicate that these by-products, which are currently industrial waste, have antioxidant activity comparable to regular products. Therefore, it is of interest to use the polyphenols extracted from walnut green husk for a technological purpose based on the current trend of a circular economy.

Identification of Phenolic Compound Profiles
The control extraction methodology extracted polyphenolic compounds (point 3.4), were identified by reversed-phase high-performance liquid chromatography (HPLC-RP). The polyphenol presence was compared to 12 synthetic polyphenol standards previously found by other researchers in walnut green husk [23][24][25] (Table 2). Nine polyphenols were identified in the OH sample and seven were observed in the CH sample (Table 2). However, polyphenol profile differences were observed, comparable to those found in the literature, probably due to different solvents used in the extractions. Therefore, solvent and extraction conditions are critical parameters to extract other polyphenols not previously identified, because compounds could have different polarities.
Gallic acid was the polyphenol identified with the highest intensity on both samples (data not shown). Nevertheless, a greater area (4978 mUA/min) was observed for the OH sample compared to the CH sample (3585 mUA/min); this could be related to a higher concentration of this polyphenol in that sample. The higher antioxidant capacity of the OH sample may be related to the higher concentration of this phenolic acid, which several authors have previously described in the walnut green husk [23][24][25].
On the other hand, differences in the composition of extracts were observed because the OH sample presented caffeic acid and polydatin, unlike CH, which could be related to the observed antioxidant capacity differences [26,27]. Caffeic acid is a hydroxycinnamic acid found in plants as an intermediate for lignin (cell wall polymer) biosynthesis. Lignification of plant tissue occurs in response to various environmental factors such as mechanical stress [28], which could occur while a green husk is opened. Meanwhile, polydatin is a stilbene polyphenol, a precursor of resveratrol, and a natural phenol produced in the response of mechanical lesion [29], also attributed to the husk opening (OH).
Therefore, the differences previously observed for antioxidant activity due to the ripeness stage were attributed to different polyphenol profiles due to the presence of more identified antioxidant polyphenols on the OH sample, related to the mechanical stress of husk opening and the highest peak area of gallic acid polyphenol.
It is also essential to consider that polyphenols detected on ripped walnut green husks (OH sample) showed antioxidant properties that may benefit human health [30,31] (Ramassamy, 2006;Vita, 2005). These antioxidant polyphenols were gallic acid, caffeic acid, protocatechuic acid, catechin, ferulic acid, quercetin, and polydatin. It indicates that walnut green husk could be a new source to obtain antioxidant polyphenols. Considering their benefits to human health and high instability and degradation [6], their protection is essential during extraction and storage.
Results revealed that the OH sample showed the highest antioxidant capacity. Therefore, it was selected for optimization through ultrasound-assisted extraction.

Optimization of the Extraction Process of the Walnut Green Husk
Considering the different process extractions was used in the literature for walnut green husk, there is no standardized extraction process for this the sub-product [3,[8][9][10][11]. In this study, a central composite design was used to optimize the extraction process for this raw material to obtain the maximum antioxidant capacity of the extract (Table 3). The responses studied using the design were the total polyphenolic content (TPC) and antioxidant capacity measured by ORAC, DPPH, and FRAP. These responses were maximized through a desirability function. The desirability function considered all the studied parameters to obtain an optimal condition that maximized its values, i.e., the highest amount of total polyphenol and the most significant antioxidant capacity. Figure 2a shows the optimal condition for the studied levels, indicating an optimum value of desirability of 0.87 (Figure 2b), which corresponds to an optimal value of the solid/solvent factor of 1:25 (w/v) and an optimal value of ethanol/water solvent ratio of 75: The design generated a nonlinear, quadratic model represented by the following equation: The response (y) corresponds to the dependent variables associated with each factor/level combination, a 0 to a 5 are the regression coefficients of the respective variables, and x 1 , x 2 corresponds to the two independent variables. The terms x 1 x 2 and x i 2 represent the interaction and the quadratic terms, respectively. and antioxidant capacity measured by ORAC, DPPH, and FRAP. These responses were maximized through a desirability function. The desirability function considered all the studied parameters to obtain an optimal condition that maximized its values, i.e., the highest amount of total polyphenol and the most significant antioxidant capacity. Figure 2a shows the optimal condition for the studied levels, indicating an optimum value of desirability of 0.87 (Figure 2b), which corresponds to an optimal value of the solid/solvent factor of 1:25 (w/v) and an optimal value of ethanol/water solvent ratio of 75:25 (v/v).  The predicted response values from the desirability function were 100 mg GAE/g sample dry weight (TPC), 191 mg Trolox/g sample dry weight (DPPH), 123 mg FeSO 4 /g sample dry weight (FRAP), and 158 µmol Trolox/g sample dry weight (ORAC).
To validate the model, three independent runs at the determined optimized conditions were performed and compared to the predicted values. Results showed that although higher values (approximately 10%) of antioxidant capacities were observed, no significant differences (p < 0.05) were obtained between predicted and performed values. The coefficient of variation (CV) was lower than 10% and the coefficient of determination (R 2 ) between the experimental values of each evaluated factor and the predicted values suggested by the desirability function was higher than 0.95, validating the model. It is important to note that the values of polyphenol content and antioxidant capacity obtained using the optimized conditions from this study were higher than those values described previously in the literature [3,10,11]. Therefore, using a desirability function maximizes several parameters, allowing them to obtain satisfactorily and validated optimized conditions for the extraction process using walnut green husk as the polyphenol source.
The optimized extracts showed different phenolic compositions from that obtained by the control extraction. It was attributed to the solvent polarity used because the hydroethanolic solvent used in this study was more polar than the standard extraction solvent (acetone-water-acetic acid). Therefore, apolar polyphenols were identified when standard extraction [32] was used, as compared to the optimized extractions of this study.

Evaluation of the Antioxidant and Antimicrobial Capacity of the Optimized Extracts of Walnut Green Husk 2.3.1. Antioxidant Capacity
In this study, the antioxidant capacities of the optimized extracts samples (OH and CH) were measured through a test based on hydrogen atom transfer (HAT) as ORAC methodology, and through a test based on electron transfer (ET) as DPPH and FRAP methods. Antioxidant activity and total polyphenol content were compared between the control and the optimized extraction of the two samples (OH and CH) (Figure 3).
The obtained results indicated that the highest antioxidant capacity of the optimized samples corresponded to the OH sample (DPPH: 202 ± 1.2 mg Trolox/g sample dry weight, FRAP: 141 ± 3.5 mg FeSO 4 /g sample dry weight, and ORAC: 171 ± 2.4 µmol Trolox/g sample dry weight). These results are attributed to the highest amount of total polyphenol content (TPC) extracted (OH: 129 ± 0.5 mg GAE/g sample dry weight compared to CH: 105 ± 2.3 mg GAE/g sample dry weight). Furthermore, these values confirm the previous characterization results, where the ripeness stage of the walnut green husk was directly related to its antioxidant capacity.  The obtained results indicated that the highest antioxidant capacity of the optimized samples corresponded to the OH sample (DPPH: 202 ± 1.2 mg Trolox/g sample dry weight, FRAP: 141 ± 3.5 mg FeSO4/g sample dry weight, and ORAC: 171 ± 2.4 µmol Trolox/g sample dry weight). These results are attributed to the highest amount of total polyphenol content (TPC) extracted (OH: 129 ± 0.5 mg GAE/g sample dry weight compared to CH: 105 ± 2.3 mg GAE/g sample dry weight). Furthermore, these values confirm the previous characterization results, where the ripeness stage of the walnut green husk was directly related to its antioxidant capacity.
Optimized and control extractions were compared for the two analyzed samples (OH and CH). Figure 3 shows that optimized extracts containing a TPC increased 2.5-fold for both samples. The increase in the antioxidant capacity was 3.6-, 3.9-, and 2.2-fold for ORAC, DPPH, and FRAP, respectively, compared to the control extraction. Therefore, Optimized and control extractions were compared for the two analyzed samples (OH and CH). Figure 3 shows that optimized extracts containing a TPC increased 2.5-fold for both samples. The increase in the antioxidant capacity was 3.6-, 3.9-, and 2.2-fold for ORAC, DPPH, and FRAP, respectively, compared to the control extraction. Therefore, these results confirm the increase in these responses by the optimization design performed.
Therefore, optimized walnut green husk extracts had a higher antioxidant capacity than the control extraction, and could thus be used for technological purposes such as a potential natural additive in replacing synthetic additives.

Antimicrobial Capacity
Walnut green husk extracts showed the highest antimicrobial activity against Gram (+) bacteria [3,10], but the effect of extracts on the bacteria growth kinetic parameters on Gram (−) bacteria has seldom been studied. Therefore, the effect of the concentration of the walnut green husk optimized extracts (OH and CH) on the kinetic growth parameters was studied on Escherichia coli, a reference Gram (−) bacterium. Additionally, considering the potential application as active additives, E. coli was selected because it is the principal pathogenic bacteria on foods, transmitted through the consumption of contaminated fresh food, due to hygienic deficiencies in processing, and causes intestinal diseases in humans [33].
Kinetic growth parameters were affected by the increasing concentration of the optimized extracts (OH and CH), reflected in the decrease in the specific growth rate (µ max ), obtained by Equation (2), and the increase in the inhibition percentage (I%), calculated using Equation (3) ( Table 4). OH-extracts at 96 µg/mL showed the lowest µ max and the greatest I% (µ max = 0.003 h −1 and I = 93%) compared to CH extract (µ max = 0.01 h −1 and I = 69%). Therefore, OH-extracts showed the greatest antibacterial activity (Figure 4), which can be attributed to the significant differences (p < 0.05) in TPC between both samples (OH = 128.6 ± 0.3 mg GAE/g sample dry weight; CH = 104.52 ± 2.3 mg GAE/g sample dry weight). This effect follows previous results because phenolic compounds have been demonstrated to exhibit antibacterial activity [34][35][36][37]. Table 4. Comparison of the kinetic parameters of growth of Escherichia coli by adding extracts of the walnut green husks. The polyphenol compounds found in both extracts, such as gallic acid, protocatechuic acid, ferulic acid, and juglone (Table 2), have previously been reported to have antimicrobial activity against E. coli [23,38,39]. It is also essential to consider that the juglone area obtained by HPLC-RP was higher in the OH sample (32.8 mUA/min) than in the CH sample (27.4 mUA/min). It was determined that this polyphenol is involved in the walnut pathogenic defense mechanism; when the green husk is open, it is more prone to pathogen attack [40,41]. Therefore, polyphenol's presence on both extracts inhibits the growth of E. coli by this antimicrobial activity.

Extract (µg/mL) Kinetic Parameters I (%) RMS (%)
Although Fernandez-Agulló et al. [10] indicated that the walnut green husk did not inhibit the growth of E. coli by agar diffusion, in this work, the antimicrobial effect of walnut green husk extracts against this Gram (−) bacterium was demonstrated. It was attributed to the different walnut ripeness stages, variety, and extraction process conditions. The polyphenol compounds found in both extracts, such as gallic acid, protocatechuic acid, ferulic acid, and juglone (Table 2), have previously been reported to have antimicrobial activity against E. coli [23,38,39]. It is also essential to consider that the juglone area obtained by HPLC-RP was higher in the OH sample (32.8 mUA/min) than in the CH sample (27.4 mUA/min). It was determined that this polyphenol is involved in the walnut pathogenic defense mechanism; when the green husk is open, it is more prone to pathogen attack [40,41]. Therefore, polyphenol's presence on both extracts inhibits the growth of E. coli by this antimicrobial activity.
Although Fernandez-Agulló et al. [10] indicated that the walnut green husk did not inhibit the growth of E. coli by agar diffusion, in this work, the antimicrobial effect of walnut green husk extracts against this Gram (−) bacterium was demonstrated. It was attributed to the different walnut ripeness stages, variety, and extraction process conditions.

Samples
Walnut green husks were obtained from a walnut tree cultivation (Juglans regia L.), Chandler variety, in April 2019 in Cuncumen, Province of San Antonio, V Region, Chile. The orchard has a planting density of 6 × 3 m 2 . Mature trees were used in full production (6 years) to ensure homogeneity. Random sampling was carried out (10 kg) from the center of 10 rows, avoiding the tree tips because they are the most exposed to the environment and pests. Two types of samples were collected: (i) ripe walnut green husk (open husk (OH)): removed manually from the walnut (open stage); and (ii) unripe walnut green husk (closed husk (CH)): removed mechanically by a vibration machine (closed stage) (Figure 1).

Walnut Green Husk Drying
The collected walnut green husks were dried in a forced-air oven (ZenithLab, DHG-9053 A, Jiangsu, China) at 40 • C for 48 h until a constant weight was obtained. Then, dried husks were crushed and ground in Thermomix equipment (Vorwerk, Wuppertal, Germany) and stored at room temperature in glass bottles covered with aluminum foil to protect samples from light.

Proximal Characterization
Proximal analysis was performed on the two representative samples of dried walnut green husks: OH and CH. Moisture, protein, lipid, ash, crude fiber, and non-nitrogen extract contents were determined according to the Official Association of Analytical Chemistry methodologies.

Control Extraction
A control extraction was performed to compare the antioxidant capacity of the walnut green husks with the fruits and vegetables database of the Institute of Nutrition and Food Technology (INTA) of the Universidad de Chile. Extraction with a solvent mix of acetone/water/acetic acid (Merck, Darmstadt, Germany) (70/29.5/0.5% v/v) was performed in an ultrasound bath (Daihan, Power sonic 405, Kyonggi-Do, Korea) at 44 kHz frequency for 90 min. Subsequent maceration for 72 h at room temperature, using a solid-to-liquid ratio of 1:25 (g/mL), was developed according to the methodology described by Wu et al. [20]. The extraction was carried out in 1.5 mL amber microtubes, and the temperature in the ultrasound bath was regulated at 25 • C. Finally, the extracts were centrifuged (Hermle, Z306, Wehingen, Germany) under refrigeration conditions (0-5 • C), for 5 min at 7000 rpm, and the liquid phase was collected and stored at 5 • C until later analysis.

Ultrasound-Assisted Extractions
Ultrasound-assisted extractions (UAEs) (Sonics Materials, VCX 500, Newtown, CT, USA) were used to obtain the extracts of the walnut green husks (4 g) by using a highintensity 3/8-inch tip probe (600 W, 20 kHz, 25 • C, 40 min) and ethanol (Merck, Darmstadt, Germany) or an ethanol-water mixture (75:25) as the solvent. UAE was carried out in a 500 mL beaker immersed in a cold-water bath to control the temperature. The extracts were vacuum-filtered using Whatman N 1 filter paper, and then the solvent was evaporated under low-pressure conditions in a rotary evaporator (Buchi R-100, Flawil, Switzerland) at 40 • C.

Determination of Total Phenolic Content
A colorimetric assay estimated the total phenolic content (TPC) in the extracts based on procedures described by Singleton et al. [42] with some modifications. Briefly, 0.1 mL of extract of known concentration was added to a 10 mL volumetric flask with 4.9 mL of distilled water and 0.5 mL of Folin-Ciocalteu reagent (Merck, Darmstadt, Germany), followed by 1.7 mL of Na 2 CO 3 (20% w/v, Merck, Darmstadt, Germany) addition. Then, distilled water was added until it reached 10 mL. The reactive mixture was allowed to stand for 2 h in darkness. As an indicator of TPC, the formation of blue color was quantified at 740 nm using a UV-vis spectrophotometer (Shimadzu UVmini-1240, Kyoto, Japan). Gallic acid (Merck, Darmstadt, Germany) was used for constructing the standard curve (2.5 to 125 µg/mL). Results were expressed as milligrams of gallic acid equivalents/g sample dry weight (mgGA/g dw). All assays were performed in triplicate.

Antioxidant Activity
Extracts were diluted to obtain a lineal concentration on the standard curve, being used a different dilution for each extract, depending on the experimental design.

DPPH Radical Scavenging Assay
The effect of scavenging 2,2-diphenyl-1picrylhydrazyl (DPPH) was determined according to the method reported by Brand-Williams et al. [43]. Diluted known concentrations (50 µL) were mixed with 2.75 mL of a methanolic solution containing the DPPH radical (Sigma-Aldrich, St. Louis, MO, USA) (1 mM). The mixture was stirred and left in the dark for 30 min, and subsequently its absorbance at 517 nm was measured using a UV-vis spectrophotometer (Shimadzu UVmini-1240, Kyoto, Japan). The standard curve was constructed using Trolox (Sigma-Aldrich, St. Louis, MO, USA) (0 to 800 µM), and the results were expressed as mg Trolox/g sample dry weight. All assays were performed in triplicate.

Ferric Reducing Antioxidant Power (FRAP)
FRAP assay was performed according to Benzie and Strain [44]. FRAP is based on the ability to reduce yellow ferric tripyridyltriazine complex (Fe (III)-TPTZ) to blue ferrous complex (Fe (II)-TPTZ) by electron-donating phenolic compounds in an acidic medium, which is measured as an absorbance change of ferrous TPTZ complex. The FRAP reagent was prepared by mixing acetate buffer (300 mM, pH 3.6), TPTZ solution (10 mM TPTZ in 40 mM HCl), and FeCl 3 ·6H 2 O (20 mM in water solution) (Sigma-Aldrich, St. Louis, MO, USA) in a 10:1:1 (v/v) ratio. For a 25 µL volume of the known concentration extract, 175 µL FRAP reagent was added to microplate wells, and the mixture was allowed to stand in the dark for 30 min at 37 • C. The absorbance of the samples was measured at a wavelength of 595 nm and compared to a blank, using a microplate reader (Thermo Fisher, Multiskan Go, Vantaa, Finland). A calibration curve was performed using standard ferrous sulfate solutions (0.5 to 1.4 mmol/L). All measurements were made in triplicate, and the results were expressed in mg FeSO 4 /g sample dry weight.

Oxygen Radical Absorbance Capacity (ORAC)
This methodology determines the level of protection afforded by the antioxidant from the extract of the walnut green husks against oxidative damage caused by radicals generated by the aerobic thermal decomposition of AAPH (2,2 -azo-bis(2-amidinopropane) dihydrochloride) (Sigma-Aldrich, St. Louis, MO, USA), using fluorescein as the probe molecule [45]. Fluorescence intensities were measured at 485/528 nm of excitation/emission wavelength (96-well white polystyrene microplate) for one-minute intervals in a microplatereader (Bio-Teck Instruments, Synergy HT, Winooski, VT, USA), using Gen 5 software.
In each well of the microplate, 150 µL of a fluorescein solution (40 nM) in phosphate buffer pH 7.4 was prepared. Then, 25 µL of the diluted extract at known concentrations were added to be incubated at 40 • C for 7 min in the microplate reader. After this time, 25 µL of APPH solution was added, resulting in a final volume of 200 µL in each plate. The blank was made using a phosphate-buffered solution at pH 7.4, instead of the sample, according to the methodology described by Pilaquinga et al. [46].
Fluorescence decay curves were normalized (F/F 0 ), and the area under the curve (AUC) was calculated for both the samples and Trolox (Sigma-Aldrich, St. Louis, MO, USA), which was used as standard. After obtaining the Trolox calibration curve, the AUC of the samples was interpolated to finally express the results as µmol Trolox equivalent/g sample dry weight. Sample measurements were performed in triplicate for each extract.

Experimental Design
A central composite design was applied to optimize the total polyphenolic content and the antioxidant capacity of the walnut green husks extracts after UAE. This design allows the study of two factors: solid/solvent ratio (g/mL) and ethanol/water ratio (v/v) at three levels, generating 10 experimental runs. It considers central points that validate the design if the standard deviation between them is lower than 5% [47]. Independent of that, three replicates of the design were performed, and the mean data with the corresponding standard deviations are reported. Table 5 shows the variables and levels of the design. The low, medium, and high levels were defined, considering that the solvent volume increased from 1:10 to 1:30, and it was selected according to the preliminary studies. The response variables analyzed in the design were total polyphenol content (mg GAE/g sample dry weight) and antioxidant activity through three methods, DPPH (mg Trolox/g sample dry weight), FRAP (mg FeSO 4 /g sample dry weight), and ORAC (µmol Trolox/g sample dry weight). The experimental runs were randomized. TPC: total phenolic content; DPPH radical scavenging activity; FRAP: ferric reducing antioxidant power; ORAC: oxygen radical absorbance capacity; GAE: gallic acid equivalents; dw: dry weight.
The response variables were maximized with a desirability function, i.e., the highest concentration of total polyphenols and the most significant antioxidant capacity. The desirability function method was applied, which helped to determine the combination of experimental factors that simultaneously optimized multiple responses. Model validation was performed with an additional set of three independent trials using the optimized design conditions.

Kinetics of Bacterial Growth
The kinetic growth of Escherichia coli ATCC 25,922 (Institute of Public Health, Chile) in the presence of extracts from walnut green husk was determined, following the methodology described by Celis-Cofré et al. [48]. A mixture (200 µL) containing: concentrated Mueller Hinton broth (Biokar Diagnostics, Beauvais, France), fresh bacteria at a concentration of 1 × 10 6 UFC mL −1 , and extracts from walnut green husk at different concentrations (16; 32; 48; 64; 80; 96 µg/mL) were added to a microplate (Bottger, Chicopee, MA, USA). Then, samples were incubated at 37 • C for 24 h on a microplate reader (Thermo Fisher, Multiskan Go, Vantaa, Finland), which recorded the absorbance of the sample at 625 nm every 1 h.

Kinetic Model
The modified Gompertz model (Equation (2)) was used to determine the effect of the concentration of the optimized extracts of the walnut green husks on the growth kinetics of a pathogenic bacteria, such as Escherichia coli [49].
where D is the microbial growth quantified in absorbance values at 625 nm (UA), D 0 is the absorbance value at time zero (UA), A s is the asymptotic value of the maximum absorbance value (adimensional), µ max is the maximum growth rate (h −1 ), λ is the lag phase (h), t the time (h), and e is the corresponding 2.718 number.