**Biotechnological Potential of** *Araucaria angustifolia* **Pine Nuts Extract and the Cysteine Protease Inhibitor AaCI-2S**

**Roberto Carlos Sallai 1,2,**† **, Bruno Ramos Salu 1,**† **, Rosemeire Aparecida Silva-Lucca <sup>3</sup> , Flávio Lopes Alves <sup>1</sup> , Thiago Henrique Napoleão 4 , Patrícia Maria Guedes Paiva <sup>4</sup> , Rodrigo da Silva Ferreira <sup>1</sup> , Misako Uemura Sampaio <sup>1</sup> and Maria Luiza Vilela Oliva 1,\***


Received: 23 October 2020; Accepted: 25 November 2020; Published: 30 November 2020 -

**Abstract:** Protease inhibitors are involved in the regulation of endogenous cysteine proteases during seed development and play a defensive role because of their ability to inhibit exogenous proteases such as those present in the digestive tracts of insects. *Araucaria angustifolia* seeds, which can be used in human and animal feed, were investigated for their potential for the development of agricultural biotechnology and in the field of human health. In the pine nuts extract, which blocked the activities of cysteine proteases, it was detected potent insecticidal activity against termites (*Nasutitermes corniger*) belonging to the most abundant termite genus in tropical regions. The cysteine inhibitor (AaCI-2S) was purified by ion-exchange, size exclusion, and reversed-phase chromatography. Its functional and structural stability was confirmed by spectroscopic and circular dichroism studies, and by detection of inhibitory activity at different temperatures and pH values. Besides having activity on cysteine proteases from *C. maculatus* digestive tract, AaCI-2S inhibited papain, bromelain, ficin, and cathepsin L and impaired cell proliferation in gastric and prostate cancer cell lines. These properties qualify *A. angustifolia* seeds as a protein source with value properties of natural insecticide and to contain a protease inhibitor with the potential to be a bioactive molecule on different cancer cells.

**Keywords:** *Araucaria angustifolia*; bioactive compounds; cysteine protease inhibitor; functional food; insecticide; plant extracts; termites; tumor cells; pine nuts; urban pest

#### **1. Introduction**

*Araucaria angustifolia* is a native gymnosperm of the greatest economic and biological importance in Brazil. It withstood the rigors of the natural selection process for hundreds of millions of years as the planet underwent intense geological and climate change. Because of its wide distribution in Parana state, this species is its state symbol and known for Paraná pine. Uncontrolled logging and the expansion of new agricultural areas harmed the forests harboring these trees to such a critical point that *A. angustifolia* is now on the official list of endangered species of the Brazilian flora and the red list of the International Union for Conservation of Nature [1]. *Araucaria* seeds have a remarkable structure, whose development is controlled directly or indirectly by changes in gene expression patterns and is an interesting biological model for cellular organization studies, protein accumulation, and differential gene expression. In addition to starch (36.28%), proteins (3.57%), lipids (1.26%), carbohydrates (2.43%), and minerals provide high nutritional value [2]. The reddish-brown peel and the thin film are rich in polyphenolic compounds with antioxidant properties that, when transferred to the edible part during the cooking process, make it a very healthy food product. In forests, the pine nuts of Araucária are a key food for various vertebrates such as agouti, squirrels, monkeys, rodents, and various species of birds [3].

The overall seed characteristics and a variety of diverse structures have led researchers to seek new substances with anti-tumor activity and therapeutic effects on various diseases [4,5]. A few studies have been performed with *A. angustifolia*, and lectins having anti-inflammatory, antibacterial, and antidepressant action on the central nervous system were isolated [6,7].

Plants synthesize numerous proteins that contribute to the protection against attack by microorganisms (fungi and bacteria) and/or invertebrates (insects and nematodes). In most cases, the biological role of these proteins is assigned based on their in vitro activity, as is the case with lectins and enzyme inhibitors. In other cases, their role is confirmed by more direct analysis such as the incorporation of these in artificial diets used in insect feeding, their incorporation into culture media for microorganism culture, or even through the expression of these proteins in transgenic plants [8,9].

Inhibitors of serine proteases have been known when Kunitz and Bowman in 1946, and Birk in 1963 purified and characterized trypsin inhibitors from soybean seeds [10]. Since then, inhibitors have been isolated mainly from reserve organs such as seeds and tubers. Cysteine protease inhibitors or cystatins are reversible protease inhibitors of the papain family and related proteases (e.g., cathepsin B, L, ficin, and bromelain). In plants, orizacystatin from rice seeds was the first inhibitor of cysteine proteases considered a cystatin [11]. Numerous biological functions attributed to phytocystatins have recently been reviewed. It is assumed that they may play a regulatory role in all physiological processes involving cysteine proteases. More systematic studies have led to very promising results, especially the use of such inhibitors as instruments for the study of protease involvement in pathophysiological processes [12,13].

Phytocystatins have been identified in a large variety of monocotyledons such as rice, corn, maize, barley, and sugarcane, and dicotyledons such as beans, potatoes, avocados, kiwis, and nuts [8,14,15]. Fewer reports exist on the purification of cysteine protease inhibitors. In this work, we describe the biotechnological potential of *Araucaria angustifolia* pine nuts on phytopathogenic organisms, extending structural and functional characterization of a cysteine protease inhibitor toxic for human tumor cell lines improving the qualification of the nuts as a functional food.

#### **2. Results**

To minimize possible proteolysis, the extract was heated at 60 ◦C for 15 min, and the inhibitory activity of the cysteine proteases papain, cruzain, and human L-cathepsin was preserved. No inhibitory activity was detected on cathepsin B or serine proteases, such as trypsin, human plasma kallikrein, porcine pancreatic elastase, or human neutrophil elastase.

#### *2.1. E*ff*ect of Extract on Adult Insects*

Protease inhibitors have demonstrated insecticidal activity by interfering with digestion, which leads to poor nutrient absorption and decreased amino acid bioavailability [16]. Thus, we investigate the protective effect of pine nuts on adult termites.

The extract exhibited termiticidal activity on *N. corniger* workers at all tested concentrations (Figure 1a). All the workers died after 10 days in the treatments with extract while 100% mortality in negative control was reached only until the twentieth day. No significant differences (*p* > 0.05)

were detected between the effects of the concentrations tested. Regarding the effect on soldiers, the extract was also able to kill the insects at all tested concentrations (Figure 1b); in negative controls, 100% mortality occurred only on the seventeenth while in the treatment at 1.0 mg/mL, for example, all insets had died on the fifth day. *Plants* **2020**, *9*, x FOR PEER REVIEW 3 of 21 mortality occurred only on the seventeenth while in the treatment at 1.0 mg/mL, for example, all insets had died on the fifth day.

**Figure 1.** Effect of *A. angustifolia* seed extract on the survival of *Nasutitermes corniger* workers (**a**,**c**) and soldiers (**b**,**d**) during 20 days. Saline solution (0.15 M NaCl) was used in negative control. Each point represents the mean ± standard deviations of three repetitions. (\* *p* < 0.05, \*\* *p* < 0.005, \*\*\* *p* < 0.0001; **Figure 1.** Effect of *A. angustifolia* seed extract on the survival of *Nasutitermes corniger* workers (**a**,**c**) and soldiers (**b**,**d**) during 20 days. Saline solution (0.15 M NaCl) was used in negative control. Each point represents the mean ± standard deviations of three repetitions. (\* *p* < 0.05, \*\* *p* < 0.005, \*\*\* *p* < 0.0001; one way-ANOVA, follow Tukey's multiple comparison test).

#### *2.2. Purification of the Cysteine Protease Inhibitor AaCI-2S*

the preparation.

one way-ANOVA, follow Tukey's multiple comparison test).

*2.2. Purification of the Cysteine Protease Inhibitor AaCI-2S*  The characterization of the cysteine protease inhibitor became the focus of the present study because it is not much studied in gymnosperms, like the inhibitors of serine proteases. The acetonefractionated proteins from the saline extract (Figure 2a, line 1) were separated by chromatography using a DEAE-Sephadex anion exchange resin followed by the cation exchange chromatography in SP-Sephadex. In the DEAE-Sephadex anion exchange resin, the inhibitor did not bind under the buffer at pH 8. Even with the change in ionic strength and pH parameters, the chromatographic profile was not modified, and most of the inhibitory activity was detected in the non-bonded material eluted with the column equilibration buffer. Using the same buffering conditions, the inhibitor also did not bind to the cationic resin SP-Sephadex. The papain inhibitory activity was detected after the chromatographies were dialyzed, lyophilized, and loaded on a Superdex 30 column in an ÄKTA purifier system (GE Life Sciences, USA). Figure 2b shows the protein profile and the location of the inhibitory activity indicated by the second peak. Fractions with inhibitory activity were pooled and analyzed by SDS-PAGE and reverse-phase chromatography. The estimated molecular mass of the inhibitor was around 18 kDa (Figure 2a, lane 2), and under reducing conditions, it showed a unique band of approximately 9 kDa (Figure 2a, lane 4). Reverse-phase chromatography onto a C-18 column The characterization of the cysteine protease inhibitor became the focus of the present study because it is not much studied in gymnosperms, like the inhibitors of serine proteases. The acetone-fractionated proteins from the saline extract (Figure 2a, line 1) were separated by chromatography using a DEAE-Sephadex anion exchange resin followed by the cation exchange chromatography in SP-Sephadex. In the DEAE-Sephadex anion exchange resin, the inhibitor did not bind under the buffer at pH 8.Even with the change in ionic strength and pH parameters, the chromatographic profile was not modified, and most of the inhibitory activity was detected in the non-bonded material eluted with the column equilibration buffer. Using the same buffering conditions, the inhibitor also did not bind to the cationic resin SP-Sephadex. The papain inhibitory activity was detected after the chromatographies were dialyzed, lyophilized, and loaded on a Superdex 30 column in an ÄKTA purifier system (GE Life Sciences, USA). Figure 2b shows the protein profile and the location of the inhibitory activity indicated by the second peak. Fractions with inhibitory activity were pooled and analyzed by SDS-PAGE and reverse-phase chromatography. The estimated molecular mass of the inhibitor was around 18 kDa (Figure 2a, lane 2), and under reducing conditions, it showed a unique band of approximately 9 kDa (Figure 2a, lane 4). Reverse-phase chromatography onto a C-18 column in an HPLC system (Figure 2c)exhibited the presence of a single major peak, indicating the purity of the preparation.

in an HPLC system (Figure 2c) exhibited the presence of a single major peak, indicating the purity of

**Figure 2.** Purification profile of the AaCI-2S inhibitor. (**a**) SDS-polyacrylamide gel electrophoresis (15%). Lane 1, Brazilian pine saline extract (100 µg); Lane 2, non-reduced AaCI-2S (20 µg); Lane 3, molecular mass markers; Lane 4, AaCI-2S (10 µg) under reducing conditions. (**b**) Superdex 30 column equilibrated with 0.05 M Tris-HCl buffer (pH 8.0) containing 0.15 M NaCl at a flow rate of 0.5 mL/min. Absorbance at 280 nm is indicated in red and the inhibitory activity on papain in blue. The arrow indicates the fractions pooled. Sample: protein (2 mg A280) after ion-exchange chromatography. (**c**) Reverse-phase chromatography Vydac C-18 column. The proteins were eluted with an acetonitrile gradient in 0.1% TFA. *2.3. Secondary Structure Estimation and Intrinsic Fluorescence Emission of AaCI-2S*  **Figure 2.** Purification profile of the AaCI-2S inhibitor. (**a**) SDS-polyacrylamide gel electrophoresis (15%). Lane 1, Brazilian pine saline extract (100 µg); Lane 2, non-reduced AaCI-2S (20 µg); Lane 3, molecular mass markers; Lane 4, AaCI-2S (10 µg) under reducing conditions. (**b**) Superdex 30 column equilibrated with 0.05 M Tris-HCl buffer (pH 8.0) containing 0.15 M NaCl at a flow rate of 0.5 mL/min. Absorbance at 280 nm is indicated in red and the inhibitory activity on papain in blue. The arrow indicates the fractions pooled. Sample: protein (2 mg A280) after ion-exchange chromatography. (**c**) Reverse-phase chromatography Vydac C-18 column. The proteins were eluted with an acetonitrile gradient in 0.1% TFA.

#### Circular dichroism (CD) spectroscopy was used to characterize the secondary structure of the *2.3. Secondary Structure Estimation and Intrinsic Fluorescence Emission of AaCI-2S*

inhibitory molecule. The spectrum displayed two negative bands, one at 208 and another at 222 nm, a positive band at 192 nm (Figure 3a), and its deconvolution estimated 58% of α-helix, 12% of β-turns, 8% of β-sheets, and 22% of disordered structures. The cluster analysis indicated that AaCI-2S belonged to the α+β class of proteins, which presented a more pronounced band at 208 nm than the one at 222 nm [17], the typical secondary structure of members of the prolamin superfamily [18], similar to napin [19] and 2S albumin isolated from melon *Momordica charantia* [20]. The emission fluorescence measurements of the AaCI-2S aromatic amino acids with excitation wavelengths at 280 nm (blue line) and 295 nm (red line) are shown in Figure 3b. The recorded spectra were very different in both the form and location of the emission peak. The intrinsic fluorescence Circular dichroism (CD) spectroscopy was used to characterize the secondary structure of the inhibitory molecule. The spectrum displayed two negative bands, one at 208 and another at 222 nm, a positive band at 192 nm (Figure 3a), and its deconvolution estimated 58% of α-helix, 12% of β-turns, 8% of β-sheets, and 22% of disordered structures. The cluster analysis indicated that AaCI-2S belonged to the α+β class of proteins, which presented a more pronounced band at 208 nm than the one at 222 nm [17], the typical secondary structure of members of the prolamin superfamily [18], similar to napin [19] and 2S albumin isolated from melon *Momordica charantia* [20]. *Plants* **2020**, *9*, x FOR PEER REVIEW 5 of 21

analysis exhibited that with an excitation at 295 nm, the emission peak occurred at 341 nm, which is

**Figure 3.** Spectroscopic characteristics of the AaCI-2S. Samples contained a 3 µM concentration of inhibitor, in PBA buffer 10 mM, pH 7.0. (**a**) Far UV-CD spectrum was recorded using a 1 mm cell path length cylindrical cuvette with an average of 8 scans, at 25 °C. The CDPro program was used to estimate the AaCI-2S secondary structure. (**b**) Fluorescence emission spectra of AaCI-2S. The samples were excited at 280 nm and 295 nm, and the fluorescence emission was monitored in the 290–450 and 305–450 nm ranges, respectively, at 25 °C. *2.4. Effects of pH and Temperature on the Activity and Structure of the AaCI-2S Inhibitor*  **Figure 3.** Spectroscopic characteristics of the AaCI-2S. Samples contained a 3 µM concentration of inhibitor, in PBA buffer 10 mM, pH 7.0. (**a**) Far UV-CD spectrum was recorded using a 1 mm cell path length cylindrical cuvette with an average of 8 scans, at 25 ◦C. The CDPro program was used to estimate the AaCI-2S secondary structure. (**b**) Fluorescence emission spectra of AaCI-2S. The samples were excited at 280 nm and 295 nm, and the fluorescence emission was monitored in the 290–450 and 305–450 nm ranges, respectively, at 25 ◦C.

intensity. However, as these residues are already partially exposed to the solvent, the ANS extrinsic probe was used to monitor the global conformational changes in the tertiary structure of the inhibitor. This dye has a low fluorescence quantum yield in aqueous environments because it binds preferentially to the hydrophobic sites, promoting a pronounced increase in the fluorescence intensity and a blue shift of the emission peak. Figure 5c showed that significant changes in the fluorescent properties of the probe occur only in acidic environments since the emission peak maximum shifts toward shorter wavelengths (from 510 nm to 478 nm) and the fluorescence intensity increases up to seven-fold, at pH 2, suggesting the exposure of hydrophobic regions at this pH, which was previously inaccessible to the probe. The thermal stability of the secondary structure of the inhibitor can be monitored by the decrease in the CD bands at 208 and 222 nm (Figure 5d). Partial loss of structure was observed after treatment at 100 °C, wherein the inhibitor lost 20% of its activity within 2 h (Figure 4b), but it did not disappear completely even after 3 or 4 h of incubation (Figure

4c).

AaCI-2S was stable over a wide pH range (Figure 4a) and temperature (Figure 4b). Its stability

The emission fluorescence measurements of the AaCI-2S aromatic amino acids with excitation wavelengths at 280 nm (blue line) and 295 nm (red line) are shown in Figure 3b. The recorded spectra were very different in both the form and location of the emission peak. The intrinsic fluorescence analysis exhibited that with an excitation at 295 nm, the emission peak occurred at 341 nm, which is a characteristic profile of tryptophan class II residues partially exposed to solvent, as in sunflower [21] and buckwheat *Fagopyrum esculentum* 2S albumins [22].

#### *2.4. E*ff*ects of pH and Temperature on the Activity and Structure of the AaCI-2S Inhibitor*

AaCI-2S was stable over a wide pH range (Figure 4a) and temperature (Figure 4b). Its stability was confirmed by CD spectroscopy since no modifications of its secondary structure were observed in the pH range of 2 to 10 (Figure 5a). The results of the intrinsic fluorescence emission of this inhibitor revealed that the microenvironment of Trp residues also did not undergo significant changes in this pH range (Figure 5b), maintaining the emission peak at around 341 nm and subtle variations in intensity. However, as these residues are already partially exposed to the solvent, the ANS extrinsic probe was used to monitor the global conformational changes in the tertiary structure of the inhibitor. This dye has a low fluorescence quantum yield in aqueous environments because it binds preferentially to the hydrophobic sites, promoting a pronounced increase in the fluorescence intensity and a blue shift of the emission peak. Figure 5c showed that significant changes in the fluorescent properties of the probe occur only in acidic environments since the emission peak maximum shifts toward shorter wavelengths (from 510 nm to 478 nm) and the fluorescence intensity increases up to seven-fold, at pH 2, suggesting the exposure of hydrophobic regions at this pH, which was previously inaccessible to the probe. The thermal stability of the secondary structure of the inhibitor can be monitored by the decrease in the CD bands at 208 and 222 nm (Figure 5d). Partial loss of structure was observed after treatment at 100 ◦C, wherein the inhibitor lost 20% of its activity within 2 h (Figure 4b), but it did not disappear completely even after 3 or 4 h of incubation (Figure 4c). *Plants* **2020**, *9*, x FOR PEER REVIEW 6 of 21

**Figure 4.** Effects of pH and temperature on the activity of the AaCI-2S inhibitor. (**a**) Functional stability at different pH values. The inhibitor samples were pre-incubated in solutions with different pH values for 30 min, neutralized to an initial pH (8.0), and the inhibitory activity on papain assay was measured. (**b**) Functional stability at different temperatures. The inhibitor was heated at different temperatures for 30 min. (**c**) Functional stability at 100 °C for up to 4 h. After boiling, the inhibitory activity on papain was measured. ((**b**,**c**) after different pretreatment temperatures, the samples were cooled down at room temperature for 30 min before the inhibitory assays). (\*\*\* *p* < 0.0001, one-way ANOVA, follow Tukey's multiple comparison test). **Figure 4.** Effects of pH and temperature on the activity of the AaCI-2S inhibitor. (**a**) Functional stability at different pH values. The inhibitor samples were pre-incubated in solutions with different pH values for 30 min, neutralized to an initial pH (8.0), and the inhibitory activity on papain assay was measured. (**b**) Functional stability at different temperatures. The inhibitor was heated at different temperatures for 30 min. (**c**) Functional stability at 100 ◦C for up to 4 h. After boiling, the inhibitory activity on papain was measured. ((**b**,**c**) after different pretreatment temperatures, the samples were cooled down at room temperature for 30 min before the inhibitory assays). (\*\*\* *p* < 0.0001, one-way ANOVA, follow Tukey's multiple comparison test).

*Plants* **2020**, *9*, x FOR PEER REVIEW 7 of 21

**Figure 5.** Effects of pH and temperature on AaCI-2S conformation. For pH dependence assays, the inhibitor (4 µM) was incubated in 10 mM PBA buffer for 30 min, at 25 °C: (**a**) Far-UV CD spectra, (**b**) Tryptophan fluorescence spectra of the AaCI-2S at different pH values (blue line, pH 2.0; black line, pH 7.0 and red line, pH 10.0) (**c**) ANS fluorescence spectra in the absence (cyan line) and presence of AaCI-2S as a function of pH. Spectra were taken 30 min after the addition of ANS probe to the protein samples. (**d**) Temperature effects on CD spectra of AaCI-2S (4 µM), in 10 mM PBA buffer, pH 7.0. Before measurements, samples were incubated at their respective temperatures for 30 min and then **Figure 5.** Effects of pH and temperature on AaCI-2S conformation. For pH dependence assays, the inhibitor (4 µM) was incubated in 10 mM PBA buffer for 30 min, at 25 ◦C: (**a**) Far-UV CD spectra, (**b**) Tryptophan fluorescence spectra of the AaCI-2S at different pH values (blue line, pH 2.0; black line, pH 7.0 and red line, pH 10.0) (**c**) ANS fluorescence spectra in the absence (cyan line) and presence of AaCI-2S as a function of pH. Spectra were taken 30 min after the addition of ANS probe to the protein samples. (**d**) Temperature effects on CD spectra of AaCI-2S (4 µM), in 10 mM PBA buffer, pH 7.0. Before measurements, samples were incubated at their respective temperatures for 30 min and then cooled to 25 ◦C.

#### cooled to 25 °C. *2.5. Inhibitor Sequence*

Knowledgebase under the accession number C0HLT8.

*2.5. Inhibitor Sequence*  The amino acid sequence of the inhibitor when compared with other protein sequences in the UniProt Knowledgebase database reveals similarities with 2S albumins of conifers and angiosperms. Because of the inhibitory activity on cysteine proteases and the structural similarity with 2S-albumin, the isolated *A. angustifolia* inhibitor was named AaCI-2S. Figure 6 showed the multiple alignments of AaCl-2S with conifer and angiosperm 2S albumin. The highest scores were obtained with conifers *Pinus strobus*, *Picea glauca,* and *Pseudotsuga menziesii* and with the angiosperms *Corylus avellana* (hazel) and *Anacardium occidentale* (cashew tree). A comparison of the AaCl-2S sequence was also performed using the BLAST program with those deposited in the MEROPS database of peptidases and their inhibitors. The search revealed a similar identity with a family of proteins whose inhibitory activity has not yet been demonstrated, denominated in the database by "Family I6 unassigned peptidase inhibitor homolog." The protein sequence data reported in this paper will appear in the UniProt The amino acid sequence of the inhibitor when compared with other protein sequences in the UniProt Knowledgebase database reveals similarities with 2S albumins of conifers and angiosperms. Because of the inhibitory activity on cysteine proteases and the structural similarity with 2S-albumin, the isolated *A. angustifolia* inhibitor was named AaCI-2S. Figure 6 showed the multiple alignments of AaCl-2S with conifer and angiosperm 2S albumin. The highest scores were obtained with conifers *Pinus strobus*, *Picea glauca*, and *Pseudotsuga menziesii* and with the angiosperms *Corylus avellana* (hazel) and *Anacardium occidentale* (cashew tree). A comparison of the AaCl-2S sequence was also performed using the BLAST program with those deposited in the MEROPS database of peptidases and their inhibitors. The search revealed a similar identity with a family of proteins whose inhibitory activity has not yet been demonstrated, denominated in the database by "Family I6 unassigned peptidase inhibitor homolog." The protein sequence data reported in this paper will appear in the UniProt Knowledgebase under the accession number C0HLT8.


*Plants* **2020**, *9*, x FOR PEER REVIEW 8 of 21

**Figure 6.** Comparison between the amino acid sequence obtained from AaCI-2S with precursors of 2S reserve proteins from angiosperms and gymnosperms. The sequences were aligned using ClustaW2. Similar strings include D0PWG2\_CORAV: hazelnut (*Corylus avellana*); B6EU55\_BEREX: Brazil nut (*Bertollethia excelsa*); Q8L694\_MOMCH: São Caetano melon (*Momordica charantia*), B9SA28\_RICCO: castor bean (*Ricinus communis);* Q8H2B8\_ANAOC: cashew nut (*Anacardium occidentale*); Q40997\_PINST: pinus (*Pinus strobus*); O81412\_PICGL: spruce (*Picea glauca*); and O64931PSEMZ: Douglas fir (*Pseudotsuga menziesii*). Spaces (-) were introduced to maintain alignment. The eight conserved cysteine residues are indicated numerically in blue. The conserved hydrophobic residues are in red, and the arrows indicate a region rich in arginine and glutamic acid residues. *2.6. Biological Properties of AaCl-2S*  The inhibition curves of some cysteine proteases by AaCl-2S was reported in Figure 7. The **Figure 6.** Comparison between the amino acid sequence obtained from AaCI-2S with precursors of 2S reserve proteins from angiosperms and gymnosperms. The sequences were aligned using ClustaW2. Similar strings include D0PWG2\_CORAV: hazelnut (*Corylus avellana*); B6EU55\_BEREX: Brazil nut (*Bertollethia excelsa*); Q8L694\_MOMCH: São Caetano melon (*Momordica charantia*), B9SA28\_RICCO: castor bean (*Ricinus communis*); Q8H2B8\_ANAOC: cashew nut (*Anacardium occidentale*); Q40997\_PINST: pinus (*Pinus strobus*); O81412\_PICGL: spruce (*Picea glauca*); and O64931PSEMZ: Douglas fir (*Pseudotsuga menziesii*). Spaces (-) were introduced to maintain alignment. The eight conserved cysteine residues are indicated numerically in blue. The conserved hydrophobic residues are in red, and the arrows indicate a region rich in arginine and glutamic acid residues.

#### of papain inhibition (Kiapp = 0.2 nM), which reflects a greater affinity for cathepsin L. AaCI-2S *2.6. Biological Properties of AaCl-2S*

family, such as ficin (Kiapp = 1.1 nM) and bromelain (Kiapp = 8.4 nM). No inhibitory activity was detected on serine proteases (trypsin, human plasma kallikrein, and elastase). The inhibition curves of some cysteine proteases by AaCl-2S was reported in Figure 7. The calculated Kiapp of the inhibition of cathepsin L (Kiapp = 0.01 nM) was about 20 times lower than that of papain inhibition (Kiapp = 0.2 nM), which reflects a greater affinity for cathepsin L. AaCI-2S presented no inhibitory activity on cathepsin B, but it also inhibited other enzymes in the papain family, such as ficin (Kiapp = 1.1 nM) and bromelain (Kiapp = 8.4 nM). No inhibitory activity was detected on serine proteases (trypsin, human plasma kallikrein, and elastase). *Plants* **2020**, *9*, x FOR PEER REVIEW 9 of 21

calculated Kiapp of the inhibition of cathepsin L (Kiapp = 0.01 nM) was about 20 times lower than that

presented no inhibitory activity on cathepsin B, but it also inhibited other enzymes in the papain

**Figure 7.** AaCI-2S inhibitory properties. Increasing the concentration of *Araucaria angustifolia* cysteine protease inhibitor was incubated with (**a**) papain (2 nM), (**b**) ficin, and (**c**) bromelain for 20 min, at 40 °C in 0.1 M Na2PO4 buffer (pH 6.3) containing 0.4 M NaCl, 0.01 M EDTA, and 8 mM DTT, the enzymatic activities were determined by the hydrolysis of Z-Phe-Arg-pNan (5 mM). *2.7. Effect of AaCI-2S on Predatory Insect Enzymes*  **Figure 7.** AaCI-2S inhibitory properties. Increasing the concentration of *Araucaria angustifolia* cysteine protease inhibitor was incubated with (**a**) papain (2 nM), (**b**) ficin, and (**c**) bromelain for 20 min, at 40 ◦C in 0.1 M Na2PO<sup>4</sup> buffer (pH 6.3) containing 0.4 M NaCl, 0.01 M EDTA, and 8 mM DTT, the enzymatic activities were determined by the hydrolysis of Z-Phe-Arg-pNan (5 mM).

proteolytic activity in the midgut extract of these larvae. Figure 8a indicated the rapid interaction of the inhibitor added to the incubation medium after 40 min and the resulting decrease in proteolytic activity on the colorimetric substrate Z-Phe-Arg-pNan. Figure 8b showed the inhibitory effect of increasing the concentrations of AaCl-2S on the residual activity of cysteine proteases present in the

midgut of larvae.

Enzymes of the class of cysteine proteases are the major proteolytic enzymes of coleopteran

#### *2.7. E*ff*ect of AaCI-2S on Predatory Insect Enzymes*

Enzymes of the class of cysteine proteases are the major proteolytic enzymes of coleopteran larvae. One of the physiological reasons for the presence of proteins with inhibitory activity in plant seeds is their involvement in the mechanism of seed protection against predatory insects. For this reason, we used larvae from the cowpea bruchid, *Callosobruchus maculatus,* a predator of string bean seeds *Vigna unguiculata*, as a model to investigate whether purified AaCI-2S would decrease the proteolytic activity in the midgut extract of these larvae. Figure 8a indicated the rapid interaction of the inhibitor added to the incubation medium after 40 min and the resulting decrease in proteolytic activity on the colorimetric substrate Z-Phe-Arg-pNan. Figure 8b showed the inhibitory effect of increasing the concentrations of AaCl-2S on the residual activity of cysteine proteases present in the midgut of larvae. *Plants* **2020**, *9*, x FOR PEER REVIEW 10 of 21

**Figure 8.** Action of AaCI-2S on the proteolytic activity of *Callosobruchus maculatus* larvae. **(a**) The blue line indicates the increase of proteolytic activity on Z-Phe-Arg-pNan. of the medium intestinal extract containing 43 µg of total proteins. The arrow indicates the time of the addition of 10 µg of the inhibitor to the incubation medium. The pink line indicates a decrease in proteolytic activity. (**b**) Inhibition of the proteolytic activity extracted from the intestine of *Callosobruchus maculatus*. A medium intestinal extract containing 11.5 µg of proteins was preincubated at 37 °C for 10 min with increasing concentrations of AaCl-2S in 0.1 M Na2PO4 buffer at pH 6.3, 0.4 M NaCl, 01 M, and 8 mM DTT. **Figure 8.** Action of AaCI-2S on the proteolytic activity of *Callosobruchus maculatus* larvae. (**a**) The blue line indicates the increase of proteolytic activity on Z-Phe-Arg-pNan. of the medium intestinal extract containing 43 µg of total proteins. The arrow indicates the time of the addition of 10 µg of the inhibitor to the incubation medium. The pink line indicates a decrease in proteolytic activity. (**b**) Inhibition of the proteolytic activity extracted from the intestine of *Callosobruchus maculatus*. A medium intestinal extract containing 11.5 µg of proteins was preincubated at 37 ◦C for 10 min with increasing concentrations of AaCl-2S in 0.1 M Na2PO<sup>4</sup> buffer at pH 6.3, 0.4 M NaCl, 01 M, and 8 mM DTT. Residual activity was determined by the hydrolysis of Z-Phe-Arg-pNan (5 mM).

#### *2.8. Investigation on the Antitumor Activity of AaCI-2S*

*2.8. Investigation on the Antitumor Activity of AaCI-2S*  Pine nuts are used by man as a functional food and, since cysteine proteases are involved in several types of tumors, we were interested in investigating the effect of AaCI-2S on tumor cells, where cathepsin L is recognized to play an important role, as in the models of gastric cancer and Pine nuts are used by man as a functional food and, since cysteine proteases are involved in several types of tumors, we were interested in investigating the effect of AaCI-2S on tumor cells, where cathepsin L is recognized to play an important role, as in the models of gastric cancer and prostate cancer.

Residual activity was determined by the hydrolysis of Z-Phe-Arg-pNan (5 mM).

prostate cancer. The effects of the inhibitor on the proliferation of prostate cancer cells (DU-145 and PC3), gastric The effects of the inhibitor on the proliferation of prostate cancer cells (DU-145 and PC3), gastric cancer (Hs746T), and non-tumor human fibroblasts were illustrated in Figure 9. The inhibitor

cancer (Hs746T), and non-tumor human fibroblasts were illustrated in Figure 9. The inhibitor did not

did not affect fibroblast (a) proliferation, while it inhibited the proliferation of both prostate cancer cell lines PC3 (b), DU-145 cells (c), and of the Hs746T cells (d). *Plants* **2020**, *9*, x FOR PEER REVIEW 11 of 21

**Figure 9.** Effects of the inhibitor on the proliferation of prostate cancer cells, gastric cancer, and human fibroblasts. Effect of AaCl-2S on the proliferation of (**a**) fibroblasts, (**b**) PC3, (**c**) DU145, and (**d**) Hs746T cells. Cells were pre-incubated with increasing concentrations of AaCl-2S for 15 min at room **Figure 9.** Effects of the inhibitor on the proliferation of prostate cancer cells, gastric cancer, and human fibroblasts. Effect of AaCl-2S on the proliferation of (**a**) fibroblasts, (**b**) PC3, (**c**) DU145, and (**d**) Hs746T cells. Cells were pre-incubated with increasing concentrations of AaCl-2S for 15 min at room temperature and analyzed at different incubation times (\* *p* < 0.05, unpaired *t*-test).

temperature and analyzed at different incubation times (\* *p* < 0.05, unpaired t-test).

#### **3. Discussion**

**3. Discussion**  In angiosperms, a considerable fraction of seed proteins includes inhibitors of serine proteases; however, to date, not many protease inhibitors have been purified and characterized in gymnosperms [23]. This was also confirmed by our investigation since, in the present study, the saline extract of *A. angustifolia* seeds did not inhibit trypsin or other serine proteases. A trypsin inhibition has been detected in the embryo tissues only after sample concentration with acetone precipitation [24]. In contrast, the saline extract inhibited two cysteine proteases, papain, and the enzyme cruzain In angiosperms, a considerable fraction of seed proteins includes inhibitors of serine proteases; however, to date, not many protease inhibitors have been purified and characterized in gymnosperms [23]. This was also confirmed by our investigation since, in the present study, the saline extract of *A. angustifolia* seeds did not inhibit trypsin or other serine proteases. A trypsin inhibition has been detected in the embryo tissues only after sample concentration with acetone precipitation [24]. In contrast, the saline extract inhibited two cysteine proteases, papain, and the enzyme cruzain (a recombinant form of the cysteine protease cruzipain from *Trypanosoma cruzi*) [25].

(a recombinant form of the cysteine protease cruzipain from *Trypanosoma cruzi*) [25]. To our knowledge, there are no reports on the effects of cysteine protease inhibitors against termites. However, the deleterious effects found in termites have been attributed to lectins and serine protease inhibitors [16,26–28] and this might not be the case of pine nuts. Although further studies are needed, the description of the termiticidal potential of the seeds is relevant, as it shows that nature has selected an alternative to its century-old forest protection against this pest. This property can be exploited commercially as an alternative for the use of Araucária other than its wood, thus protecting To our knowledge, there are no reports on the effects of cysteine protease inhibitors against termites. However, the deleterious effects found in termites have been attributed to lectins and serine protease inhibitors [16,26–28] and this might not be the case of pine nuts. Although further studies are needed, the description of the termiticidal potential of the seeds is relevant, as it shows that nature has selected an alternative to its century-old forest protection against this pest. This property can be exploited commercially as an alternative for the use of Araucária other than its wood, thus protecting the forest, a world heritage site.

the forest, a world heritage site. The presence of papain inhibitors has been reported in seeds of some gymnosperms such as *Pinus maritima*, *Picea pungens*, and *A. angustifolia* [29]; however, purification and characterization of the inhibitory activity were not achieved. We did not find any purified gymnosperm phytocystatin The presence of papain inhibitors has been reported in seeds of some gymnosperms such as *Pinus maritima*, *Picea pungens*, and *A. angustifolia* [29]; however, purification and characterization of the inhibitory activity were not achieved. We did not find any purified gymnosperm phytocystatin in the protein databases, but approximately 200 sequences have been identified through comparative

The concentration of the inhibitor, determined by titration with papain, was 17 mg/kg. The first cysteine protease inhibitor (oryzacystatin) identified in rice during the 1980s occurs at a concentration

in the protein databases, but approximately 200 sequences have been identified through comparative

genomic analysis [30]. These analyses have been largely useful for information on the conservation and evolution of proteolytic enzymes and their inhibitors [31].

The concentration of the inhibitor, determined by titration with papain, was 17 mg/kg. The first cysteine protease inhibitor (oryzacystatin) identified in rice during the 1980s occurs at a concentration of 2–3 mg/kg [32] and its structural and functional characterization was investigated only after its recombinant form was reported [33]. In contrast, the concentration of trypsin inhibitors is rather high in legumes [34], for example, the concentration in *Enterolobium contortisiliquum* is approximately 5600 mg/kg [35].

The fact that the inhibitor does not bind to ion exchange resins is an alternative and effective method for use in the processing of large amounts of extract through the batchwise system. Using this strategy, most of the proteins with a molecular weight above 25 kDa were eliminated, thereby favoring size exclusion chromatography in Superdex 30, in which the inhibitory activity was detected only in the second peak. Structurally, the protein database sequences displayed high similarity with conserved proteins of the 2S albumin family and no sequence homology with the typical phytocystatins or with other plant inhibitors of cysteine proteases as the described inhibitor of *B. bauhinioides*, BbCI, which also differs from phytocystatins [36].

From the earliest studies with protease inhibitors in plants, their potential role as a reserve protein has been postulated. The similarity with the 2S albumins of the gymnosperms *Picea glauca*, *Pseudotsuga menziesii* [37], and *Pinus strobus* was 64%, 55%, and 52.6%, respectively. Similarities of 62.5% and 67.5% were also observed with some 2S angiosperm albumins such as *Corylus hazelnut* [38] and *Anacardium occidentale* (cashew nuts) [39], respectively. Thus, based on similarities, sequence alignment, and arrangement of conserved cysteine residues, we can conclude that the *Araucaria angustifolia* inhibitor is a reserve protein of the 2S albumins class which justified the denomination adopted in this work (AaCI-2S). The AaCI-2S also displayed a high content of arginine residues, which is a common feature in conifer reserve proteins. Furthermore, 8 cysteine residues and the hydrophobic residues flanking cysteine residues are conserved positions in relation to the 2S gymnosperm and angiosperm albumins as in the sequence of 2S albumin from *Pseudotsuga menziesii* [37]. Additionally, AaCI-2S displays a molecular mass of 18 kDa and two identical polypeptide chains linked by disulfide bonds similar to the storage protein 2S albumin. Although the reserve function of proteins is usually assigned to 2S albumins, other biological activities have been described such as inhibiting fungal growth [20], hemagglutinating activity [40], the inhibitory activity of trypsin [41], and in the case of AaCI-2S, the inhibitory activity of cysteine proteases.

Many larvae of insects of the order Coleoptera are predators of seeds and the presence of cysteine proteases as digestive enzymes [42] leads to the hypothesis of a possible exogenous protective role of inhibitor of this class of enzymes. Numerous studies evidencing the in vitro and in vivo inhibition of the digestive proteases of these larvae [43,44] and on the growth of fungi [45] support this role. The possible action of the inhibitor to act as a defense protein was confirmed by the rapid and effective inhibition of the cysteine protease activity present in the insect intestine suggesting that this inhibitor may be employed in the functional study of these enzymes. Naturally, these insects do not use the seeds of *Araucaria*, but the result is interesting in the sense of indicating how proteins can be strategic in the composition of compounds involved in the protection of seeds in general.

The presence of protease inhibitors in legume seeds and cereals added to epidemiological studies that identify legumes as potential protective agents in reducing the incidence of some types of cancer in the vegetarian population have stimulated a series of studies involving inhibitory proteases influencing tumor promotion in vivo and in vitro [46–48].

In addition to the inhibition of cruzain and papain, AaCI-2S seems to be selective regarding the two human cathepsins tested, since cathepsin L was inhibited while cathepsin B activity was not altered. Both are medically relevant targets because they are involved in many physiological and pathological processes such as apoptosis, inflammation, and cancer [49]. Among other functions, cathepsins (mainly cathepsins B and L) are involved in extracellular matrix degradation, facilitating the growth, invasion, and metastasis of tumor cells [50,51]. AaCl-2S can interfere with the cellular proliferation of the two cell lines of prostate cancer and shows a more effective inhibitory effect on the proliferation of gastric tumor cells. Increased activity of cathepsins B and L and the reduction of secreted endogenous cystatins have been observed in prostate cancer cell lines PC3 and DU145. The invasive ability of these cell lines was partially inhibited by E-64, a synthetic inhibitor of cysteine proteases [52]. It is worth mentioning that the non-interference in non-tumorigenic cells demonstrate the inhibitory selectivity in cancer cell lines. As the seeds of *Araucaria* are used as food, their antiproliferative effect is of nutritional significance for future studies and provides important information regarding the benefits of including the pine nut in our diet.

The overall yield of the inhibitor by the purification process was low (12%), thus obtaining large amounts of inhibitor is difficult. Notably, this loss is not due to thermal stability, since the inhibitor spectra following treatment at different temperatures exhibited a structure that was quite resistant to thermal denaturation, displaying small conformational changes after heating up to 80 ◦C. Only the treatment at 100 ◦C caused a modification of the inhibitor structure with a partial loss of structure. These results are consistent with other observations regarding the thermostability of many members of the prolamin superfamily, such as the presence of intramolecular disulfide bonds, which have been implicated in this thermostability [53]. The maintenance of the inhibitory activity against papain after heating and exposure to extreme pH values indicates that the inhibitor is functionally stable. Many other protease inhibitors purified from plant seeds exhibit high stability at various temperatures [54], but we were surprised that the inhibitor isolated from the *Araucaria* seed maintained its activity even after a 60 min treatment at 100 ◦C and may implicate in the qualification of pine nuts as a functional food since they are normally consumed roasted or cooked.

#### **4. Material and Methods**

#### *4.1. Plant Material*

*Araucaria angustifolia* (Bertol.) O. Kuntze plant material was deposited in the Herbarium of Universidade Estadual da Bahia, UEDB, identified as HUESB 12431. The studies were conducted in accordance with Brazilian legislation (license no. 02/2014, process 02000.003472/2005– 62 Ministério do Meio Ambiente, Coordenação Geral de Autorização de uso da Flora e Floresta, SCEN).

The seeds were purchased from the city of Campos do Jordão—SP from the natural occurrence of *A. angustifolia* located in Campos do Jordão State Park.

#### *4.2. Experimental Reagents*

Bovine serum albumin, fibronectin, human neutrophil elastase (EC 3.4.2.37), and porcine pancreas elastase (EC 3.4.21.7) were purchased from Calbiochem®—(EMD Chemicals Inc., Port Wentworth, GA, USA). Bovine trypsin (EC 3.4.21.4), papain (EC 3.4.22.2), bromelain (EC 3.4.22.32), and ficin (3.4.22.3) were obtained from Sigma-Aldrich (Co., St. Louis, MI, USA). Kallikrein (human plasma) (EC 3.4.21.34) was purified according to Oliva [55]. Cruzain, cathepsin B, and L were provided by Prof. Dr. Luís Juliano Neto, Department of Biophysics, UNIFESP. Chromogenic substrates derived from p-nitroanilide (Bz-Arg-pNan, HD-Pro-Phe-Arg-pNan, Suc-Phe-pNan, HD-Val-Leu-Lys-pNan, HD-Phe-L-Pip-L -Arg-pNan, MeO-Suc-Ala-Ala-Pro-Val-pNan, N-Suc-Ala-Ala-Pro-Phe-pNan), and the fluorimetric aminomethyl coumarin substrate Z-Phe-Arg AMC were obtained from Calbiochem® (EMD Chemicals Inc., USA), and starch from Sigma-Aldrich Co. (Saint Louis, MO, USA).

DEAE–Sephadex® A-50; SP-Sephadex® C-50 e Superdex® 30 (GE Healthcare, Chicago, IL, USA)—Biogel® P30 (Bio-Rad Laboratories, Hercules, CA, USA)—C18 column *Protein & Peptide (Vydac*® *Ultrasphere*—Brea, CA, USA)—Column *Aquapore*® *RP 300 C* (Varian, Palo Alto, CA, USA). Dinitrosalicylic acid (ADNS), ammonium persulfate, MTT salt, toluidine blue dye, E-64, and 1-anilinonaphthalene-8-sulfonic acid (ANS) probe were obtained from Sigma-Aldrich Co. (USA). Coomassie Brilliant Blue R-250 was obtained from Bio-Rad Laboratories (USA). Fetal bovine serum LB

broth, cell culture media, RPMI 1640, DMEM, TEMED, and dithiothreitol from Gibco Invitrogen Co. (Waltham, MA, USA). Acrylamide and N, N, and methylene bisacrylamide were obtained from Serva (Heidelberg, Germany). Molecular weight standards were obtained from Fermentas Inc. (Burlington, ON, Canada) and Bio-Rad Laboratories (USA).

#### *4.3. Cells*

The PC3 cell line of prostate adenocarcinoma and the cell line HsT46T of gastric adenocarcinoma were provided by Prof. Dr. Barbara Mayer, from the Klinikum Groβhadern Surgery Department, University of Munich, Germany. The DU145 prostate adenocarcinoma cell line was provided by Prof. Dr. Heloisa Selistre de Araújo, from the Department of Physiological Sciences at the Federal University of São Carlos. The human lineage of fibroblasts, obtained from cells of the amniotic fluid, was provided by Prof. Dr. Leny Toma, from the Biochemistry Department at the Federal University of São Paulo.

#### *4.4. Protein Extraction and Fractionation*

The seeds were ground in a blender with a 0.15 M NaCl solution, at a 10% (*w*/*v*) density, heated at 60 ◦C for 30 min, cooled in an ice bath for 30 min, stirred at room temperature for 20 min, filtered with cotton and gauze, and centrifuged at 6000× *g* for 15 min at 4 ◦C. Proteins were estimated spectrophotometrically (A280) as well as by Bradford (1976) [56] assay using bovine serum albumin as the standard.

#### *4.5. Inhibitory Activity*

The protein extract of plant seeds as well as the purified inhibitor was tested on proteases. The p-nitroaniline released as a hydrolysis product was measured at 405 nm using a SpectraCount spectrophotometer. In the case of the fluorogenic substrate Z-Phe-Arg-AMC, the Hitachi F-2000 spectrofluorometer was used with excitation and emission wavelengths of 380 and 460 nm, respectively. Different concentrations of the inhibitor solutions were added to the appropriate volumes of activated enzymes in a 100-µL volume of buffer. The volume was topped to 230 µL with a 0.15 M NaCl solution and the mixture was pre-incubated at 37–40 ◦C for 10 min before the addition of the substrate. The reaction proceeded for 20–30 min at 37–40 ◦C and was stopped by the addition of 40% acetic acid (*v*/*v*). The absorbance obtained in the absence of the extracts was considered as 100% of enzymatic activity, and the inhibition was expressed as the reduction of enzyme activity percentage [57].

The concentration of active papain was determined by titration with the synthetic inhibitor E-64 according to Zucker et al. [58]. Once the inhibitory activity of papain was determined, the purified protein was titrated and used to determine the dissociation constants of the enzyme/inhibitor complexes (Kiapp) with other cysteine proteases. The determinations were performed following the model suggested by Morrison adapted to an enzymatic kinetics program for computer graphics, and the numerical value was calculated using the GraFit program [59].

#### *4.6. Evaluation of the Extract E*ff*ects on Adult Insect Survival*

The insecticidal activity of the extract was evaluated by a bioassay based on the method described by Kang [60]. Each assay consisted of a Petri dish (90 × 15 mm) with the bottom covered by a paper filter. Disks (4 cm in diameter) of paper filter were impregnated with 200 µL of extract (0.2, 0.4, 0.8, or 1.0 mg/mL). In the negative control, 200 µL of 0.15 M NaCl was added to the disks. A total of 20 active termites (at a worker-to-soldier ratio of 4:1) were transferred to each dish, which was maintained at 28 ◦C in the dark. Insect survival was evaluated daily until the death of all insects. The bioassays were carried out in quadruplicate for each tested concentration, and the survival rates (%) were calculated.

#### *4.7. Inhibitor Purification by Ion-Exchange Chromatography (Batchwise)*

The proteins from seed extract were precipitated by the slow addition of ice-cold acetone to a final concentration of 80% (*v*/*v*). After a sedimentation period (30 min), part of the acetone was sucked out with a rubber cannula, and the remaining fraction was centrifuged at 3000× *g* for 15 min at 4 ◦C. The acetone-precipitated proteins were spread on Petri dishes, dried at 24 ◦C, and then frozen until use. The acetone-precipitated protein was dissolved in water (1 g/10 mL), centrifuged at 3000× *g* at 4 ◦C for 15 min, and the conductivity values of the solutions were adjusted by 0.05 M Tris-HCl (pH 8.0) buffer using DEAE–Sephadex®, ion-exchange chromatography. The protein was added to the resins and stirred for 30 min. The mixture was filtered through a funnel with a porous plate, and the resin was washed with an equilibration buffer to remove the non-adsorbed proteins. Elution was performed using 0.15 M or 0.3 M NaCl solution in the equilibration buffer. The non-adsorbed fraction containing papain inhibitory activity was mixed with the SP-Sephadex resin and the procedure was repeated as described above. Papain inhibitory activity was also detected in the non-adsorbed fraction.

#### *4.8. Molecular Exclusion Chromatography*

The non-retained ion exchanging fraction was dialyzed in water using a 10 kDa cut off membrane, lyophilized and dissolved in a 0.05 M Tris-HCl buffer (pH 8.0) containing 0.15 M of NaCl, centrifuged at 10,000× *g* for 5 min at room temperature, and loaded onto a Superdex 30 column in a 0.05 M Tris-HCl buffer (pH 8.0) containing a 0.15 M NaCl (equilibrium buffer) under a 0.5 mL/min flow rate in the ÄKTA purifier system (GE Healthcare). The molecular mass of the inhibitor was estimated by standardizing the column with ferritin (440 kDa), SBTI (20 kDa), cytochrome C (12.4 kDa), and aprotinin (6.5 kDa). The protein profile was monitored by absorbance at 280 nm, and fractions (1 mL) with inhibitory activity were pooled, dialyzed, and lyophilized.

#### *4.9. Reverse-Phase Chromatography on an HPLC System*

The protein from Superdex 30 chromatography was further purified by C18 reverse phase (Protein & Peptide, 4.6 mm × 14 cm) in an HPLC system (Model SCL-6A—Shimadzu), equilibrated with a 0.1% trifluoroacetic acid (TFA) solution in water (Solvent A). The elution was performed on a gradient of 0.1% TFA in water with 90% acetonitrile (Solvent B) under a constant flow rate of 0.7 mL/min [61].

#### *4.10. Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis*

Denaturing electrophoresis was performed according to the method described by Laemmli [62] on a polyacrylamide gel (15%) in the presence of SDS. The samples were treated with a reducing agent in dithiothreitol-containing sample buffer (200 mg/mL) and heated for 10 min at 100 ◦C. The proteins were visualized by staining with Coomassie blue R250 solution.

#### *4.11. Estimation of Secondary Structure by Circular Dichroism (CD) Spectroscopy*

The far-ultraviolet (UV) CD spectra of the purified inhibitor (an average of 8 scans) were recorded on a J-810 (Jasco Corporation, Tokyo, Japan) spectropolarimeter within the range of 190 to 250 nm in a 1-mm optical path cylindrical quartz cuvette at 25 ◦C. The inhibitor (3 µM) was dissolved in 10 mM PBA buffer (pH 7.0) and its CD spectra were expressed as molar ellipticity [θ]. The estimated calculation of the secondary structure fractions was performed using the CDPro deconvolution package, with the Selcon3, Continll, and CDSSTR programs [63].

#### *4.12. Intrinsic Fluorescence Measurements*

The fluorescence emission measurements were obtained using an F-2500 fluorometer (Hitachi Ltd., Tokyo, Japan) at 25 ◦C in quartz cuvettes, with an optical path of 1 cm. Analyses were performed with the purified inhibitor (3 µM) in 10 mM PBA buffer (pH 7.0). The sample was excited at 280 or 295 nm, and the fluorescence emission was monitored in the range of 290–450 and 305–450 nm, respectively. The fluorescence emission spectra of buffers were subtracted from the spectra of the samples to minimize the effect of light scattering and to perform baseline corrections [64].

#### *4.13. Studies on the Influence of pH and Temperature on Structural Stability*

Purified inhibitor structural stability was analyzed under different conditions through the measurements of CD and intrinsic and extrinsic fluorescence emission. Samples of the inhibitor (4 µM) in 10 mM PBA buffer (pH 7.0) were incubated at 25, 40, 80, and 100 ◦C for 30 min and then cooled. Samples with the same concentration were incubated in solutions of different pH (pH 2, pH 4, pH 7, and pH 10) for 30 min. After each treatment, the measurements of CD and intrinsic fluorescence were performed under the same conditions described above. For the extrinsic fluorescence measurements, the treated samples were incubated with the 8-anilino-1-naphthalenesulfonate probe (ANS), 90 µM in 10 mM PBA buffer, pH 7.0 for 15 min, at 25 ◦C, and the fluorescence spectra were recorded at 400 to 650 nm with a 385 nm excitation, 30 min after adding the probe.

#### *4.14. N-Terminal Sequence Determination*

After reverse phase chromatography, the inhibitor (2 nM) was dissolved in 300 µL of buffer 0.25 M Tris-HCl (pH 8.5) containing 6 M guanidine, 1 mM EDTA, and 5 µL of β-mercaptoethanol and then incubated for 2 h at 37 ◦C, under a nitrogen-saturated atmosphere in the absence of light. Alkylation was performed with the addition of 5 mL of vinylpyridine and re-incubation for 90 min at 37 ◦C and subjected to HPLC/acetonitrile/isopropanol reverse phase chromatography at a constant flow rate (0.1 mL/min). The N-terminal sequence was obtained automatically by the Edman (1949) [65] degradation method in two independent laboratories. Similarity searches were performed using the FASTA program using the *UniProt Knowledgebase* database Larkin [66] and a *Blossom 80* (EBI) [67] matrix (www.ebi.ac.uk). Multiple alignments of similar sequences were performed using the ClustalW2 program (http://www.ebi.ac.uk/Tools/clustalw2/).

#### *4.15. E*ff*ect of pH and Temperature on Inhibitor Activity*

To verify the effect of pH, the inhibitor (1 µg) was pre-incubated for 30 min in 50 mM sodium citrate buffer (pH 3 and pH 6) or 50 mM Tris-HCl buffer (pH 7 and pH 9). After pre-incubation, the pH of the samples was adjusted to 8.0 with Tris-HCl, and the ability to inhibit papain was determined. The lability of the inhibitor at different temperatures was investigated by incubating the samples (1 µg) in 50 mM Tris-HCl pH 8.0 buffer, keeping them in a water bath at different temperatures (25, 40, 80, and 100 ◦C) for 30 min. The resistance of the inhibitor to boiling (100 ◦C) was also studied by incubation for different durations (30 min, 1, 2, 3, and 4 h). After different heat treatments, the samples were cooled in an ice bath for 5 min and tested for their inhibitory activity.

#### *4.16. Evaluation of the Inhibitor E*ff*ect on Insect Enzymes*

Twenty larvae of *C. maculatus* with 19–20 days post-hatching (4th instar) were removed from the infested seeds and immersed in 0.15 M NaCl solution for dissection of the intestines with the help of watchmaker tweezers and stereoscopic magnifying glass. Lysis was performed in the intestines by brief sonication wells at 150 µL solution NaCl 0.15 M. The obtained extract was centrifuged at 10,000× *g* at 4 ◦C for 10 min. The supernatant was collected and frozen at −20 ◦C. In a preliminary test, 50 µL of crude extract diluted ten times and containing 12 µg of protein was added in duplicates into two wells of a microplate with Na2PO<sup>4</sup> 0.1 M buffer (pH 6.3), 10 mM EDTA, and 0.4 M NaCl. The substrate (20 µL) Z-Phe-Arg-pNan (0.05 M) was added in a final volume of 250 µL. The plate was incubated at 37 ◦C and hydrolysis was followed photometrically for 40 min. Next, the purified inhibitor (10 µg) was added to one well, and substrate hydrolysis continued to be monitored for up to 4 h. To evaluate the effect of inhibitor concentrations, the extract diluted ten-fold in 0.1 M Na<sup>2</sup> PO<sup>4</sup> buffer (pH 6.3), 10 mM EDTA; 0.4 M NaCl; 8 mM DTT remained at 37 ◦C for 10 min for enzymatic activation. Next, 50 µL of the activated extract containing 11.5 µg of proteins were preincubated in the

absence and presence of different inhibitor concentrations (1 to 8 ug) for 10 min at 37 ◦C. After the addition of 20 µL of the Z-Phe-Arg-pNan substrate (5 mM) in a final volume of 250 µL, the hydrolysis was monitored for 60 min and then quenched with 50 µL of 40% acetic acid (*v*/*v*).

#### *4.17. Studies on the E*ff*ect of Inhibitors on Tumor Cells and Human Fibroblast Cells*

PC3, DU145, and Hs746T tumor cell lines were maintained in RPMI-1640 culture medium and the fibroblast cells were maintained in Dulbecco's modified Eagle's medium (DMEM), pH 7.4. Both culture media were enriched with 10% fetal bovine serum, penicillin (10 UI/mL), and 100 µg/mL streptomycin. Cells were sub-cultured weekly using the following protocol: The medium was removed from the confluent cell flasks (60 × 10 mm) and cells were washed with PBS solution (pH 7.4). For cell detachment, the cells were incubated with 1 mL trypsin solution (0.25%) for 1 min. Next, 1 <sup>×</sup> <sup>10</sup><sup>5</sup> cells were resuspended, transferred to a new plate in the appropriate media, cultured at 37 ◦C under 5% CO2, and the culture medium was changed every 3 days [46,47].

#### *4.18. Cell Viability Assay*

PC3, DU145, and Hs746T cells (5 <sup>×</sup> <sup>10</sup><sup>3</sup> cells/<sup>100</sup> <sup>µ</sup>L/well) and fibroblast cells (8 <sup>×</sup> <sup>10</sup><sup>3</sup> cells/100 µL/well) were incubated at 37 ◦C and 5% CO<sup>2</sup> for 24 h in RPMI-1640 medium containing 10% fetal bovine serum. A total of 100 µL of inhibitor (2.5–30 µM) diluted in RPMI-1640 medium, previously filtered through a Millipore filter (0.22 µm), was added to the adhered cells and incubated for 24, 48, and 72 h. At the end of each incubation period, 10 µL of MTT (tetrazolium salts) dissolved in PBS (5 mg/mL) was added to each well and the cells were again incubated for 2 h. Subsequently, the medium was removed and 100% DMSO was added to solubilize the formazan crystals and incubated for 20 min at 37 ◦C. The absorbance was measured at 540 nm using a spectrophotometer (SpectraCount model). Assays were performed in triplicate for each inhibitor concentration and experiments were performed twice as described by Gasperazzo Ferreira et al. [68].

#### *4.19. Statistical Analyses*

All assays were performed in triplicate and independently. The statistical analyses were expressed as the mean ± standard deviation (SD) and analyzed using GraphPad Prisma Software. Comparisons among the variables, measured in defined experimental groups, were conducted using one-way ANOVA, followed by Tukey's test. Statistical significance was defined as \* *p* < 0.05,\*\* *p* < 0.005, and \*\*\* *p* < 0.0001.

#### **5. Conclusions**

These findings provided relevant information about the insecticide and antitumor activity of the pine nuts, the potential biotechnology application in agriculture, and human health that can contribute to the preservation of the Araucária forest. Also, a protein named AaCI-2S, with a molecular mass of 18 kDa composed of two identical polypeptide chains linked by two disulfide bonds was characterized. The studies of the structure–activity relationship at different pH values and temperatures revealed its high functional and structural stability. The inhibitory activity demonstrated on cysteine proteases bromelain, ficin and cathepsin L, and the cysteine proteases of the larval midguts of *C. maculatus* suggests other endogenous roles for 2S albumin. AaCI-2S inhibited cell proliferation of gastric cancer and two lines of prostate cancer and did not affect the proliferation of non-tumorigenic cells. As the seeds of *Araucaria* are used as food, its antiproliferative effect is of nutritional significance for future studies and offers evidence regarding the benefits of including the pine nut as a functional food in our diet.

**Author Contributions:** All authors participated in this study. R.C.S., B.R.S., R.A.S.-L., F.L.A., T.H.N., and R.d.S.F. performed the assays and wrote the manuscript; M.L.V.O., M.U.S., and P.M.G.P. designed and interpretation of the data and made a critical review of the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) [2017/07972-9 and 2017/06630-7]; Coordenação de Aperfeiçoamento de Pessoal de Nível Superior-Brasil (CAPES) -Finance Code 001, and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) [401452/2016-6], and M.L.V.O. received a research fellowship from CNPq, Brazil.

**Acknowledgments:** Joana G Ferreira, Claudia de Paula, Lucimeire A de Santana and Reinhart Mentele merit our gratitude for the valuable and competent contribution to this work.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

## **Three** *Scrophularia* **Species (***Scrophularia buergeriana***,** *S. koraiensis***, and** *S. takesimensis***) Inhibit RANKL-Induced Osteoclast Di**ff**erentiation in Bone Marrow-Derived Macrophages**

### **Hyeon-Hwa Nam , A Yeong Lee , Yun-Soo Seo, Inkyu Park , Sungyu Yang, Jin Mi Chun, Byeong Cheol Moon , Jun-Ho Song \* and Joong-Sun Kim \***

Herbal Medicine Resources Research Center, Korea Institute of Oriental Medicine, 111, Geonjae-ro, Naju-si 58245, Korea; hhnam@kiom.re.kr (H.-H.N.); lay7709@kiom.re.kr (A.Y.L.); sys0109@kiom.re.kr (Y.-S.S.); pik6885@kiom.re.kr (I.P.); sgyang81@kiom.re.kr (S.Y.); jmchun@kiom.re.kr (J.M.C.); bcmoon@kiom.re.kr (B.C.M.) **\*** Correspondence: songjh@kiom.re.kr (J.-H.S.); centraline@kiom.re.kr (J.-S.K.)

Received: 20 October 2020; Accepted: 24 November 2020; Published: 26 November 2020 -

**Abstract:** Scrophulariae Radix, derived from the dried roots of *Scrophularia ningpoensis* Hemsl. or *S. buergeriana* Miq, is a traditional herbal medicine used in Asia to treat rheumatism, arthritis, and pharyngalgia. However, the effects of *Scrophularia buergeriana*, *S. koraeinsis*, and *S. takesimensis* on osteoclast formation and bone resorption remain unclear. In this study, we investigated the morphological characteristics and harpagoside content of *S. buergeriana*, *S. koraiensis*, and *S. takesimensis*, and compared the effects of ethanol extracts of these species using nuclear factor (NF)-κB ligand (RANKL)-mediated osteoclast differentiation. The harpagoside content of the three *Scrophularia* species was analyzed by high-performance liquid chromatography–mass spectrometry (HPLC/MS). Their therapeutic effects were evaluated by tartrate-resistant acid phosphatase (TRAP)-positive cell formation and bone resorption in bone marrow-derived macrophages (BMMs) harvested from ICR mice. We confirmed the presence of harpagoside in the *Scrophularia* species. The harpagoside content of *S. buergeriana*, *S. koraiensis*, and *S. takesimensis* was 1.94 ± 0.24 mg/g, 6.47 ± 0.02 mg/g, and 5.50 ± 0.02 mg/g, respectively. Treatment of BMMs with extracts of the three *Scrophularia* species inhibited TRAP-positive cell formation in a dose-dependent manner. The area of hydroxyapatite-absorbed osteoclasts was markedly decreased after treatment with the three *Scrophularia* species extracts. Our results indicated that the three species of the genus *Scrophularia* might exert preventive effects on bone disorders by inhibiting osteoclast differentiation and bone resorption, suggesting that these species may have medicinal and functional value.

**Keywords:** *Scrophularia buergeriana*; *S. koraiensis*; *S. takesimensis*; harpagoside; osteoclast differentiation; RANKL

#### **1. Introduction**

An imbalance between osteoclasts and osteoblasts affects bone formation, leading to weakened bone and the development of skeletal diseases such as osteoporosis, rheumatoid arthritis, lytic bone metastases, and chronic obstructive pulmonary disease. The function of osteoclasts, multinucleated giant cells, is bone resorption, but osteoporosis can occur if bone resorption exceeds formation due to an increase in the number of osteoclasts [1]. Most drugs used to treat osteoporosis inhibit osteoclast differentiation to control bone resorption. Receptor activator of nuclear factor (NF)-κB ligand (RANKL), a major osteoclastogenic molecule, is a member of the tumor necrosis factor (TNF) superfamily and is the initial stimulator of osteoclast differentiation, inducing the expression of osteoclast-associated

genes, such as tartrate-resistant acid phosphatase (TRAP) [2,3]. Therefore, bisphosphonates and anti-RANKL antibodies that inhibit osteoclast activity are currently used for the treatment of bone resorption diseases.

Scrophulariae Radix, an herbal medicine known as Korean Hyun-Sam, is derived from the dried roots of *Scrophularia ningpoensis* Hemsl. or *S. buergeriana* Miq., plants which are widely distributed throughout the temperate regions of the Northern Hemisphere, including Asia, Europe, and North America [4–6]. Scrophulariae Radix has been traditionally used as a therapeutic agent for blood cooling, yin nourishing, fire pursing, and toxin removal, and is widely used to treat rheumatism and arthritis in Southwest Asia [7–10]. It has also been reported to have neuroprotective, anti-inflammatory, anti-allergy, anti-amnesia, antioxidant, and hepatoprotective effects [11–15]. In Korea, *S. koraiensis* Nakai (Korean: To-Hyun-Sam) has been used as an antipyretic and anti-inflammatory agent in traditional medicine. *S. takesimensis* Nakai (Korean: Seom-Hyun-Sam) is restricted to Ulleung-do Island [16,17]. Although this species is a valuable endemic resource, its medicinal efficacy has not been assessed to date.

The therapeutic potential of *Scrophularia* species is associated with the functions of major secondary metabolites, such as phenylpropanoids and iridoid glycosides, which are present in the plant [18,19]. Harpagoside, an iridoid component present in the *Scrophularia* species, is a bioactive compound of *Harpagophytum procumbens* DC. (Devil's Claw) and has been used in Southern Africa to treat pain, arthritis, and ulcers. Pharmacological effects of harpagoside on RANKL-induced osteoclast differentiation have also been reported [20,21]. However, there are few reports concerning the pharmacological activity of *Scrophularia* species on RANKL-induced osteoclast differentiation, and studies on the biological activity of *S. koraiensis* and *S. takesimensis* have not been reported.

In the current study, we compared the morphological characteristics and harpagoside content of *S. buergeriana*, *S koraiensis*, and *S. takesimensis*, and compared the effects of *Scrophularia* species extracts on RANKL-mediated osteoclast differentiation.

#### **2. Results**

#### *2.1. Comparative Morphology of Scrophularia Species*

The three species can be distinguished on the basis of leaf shape, apex, margins, pubescence of stems, and calyx shape (Table 1). The leaf blade of *S. buergeriana* is ovate, with an acute apex (Figure 1A), and serrate with a spinose tooth marginal shape (Figure 1D). The stem is glabrous (Figure 1G), and the calyx is ovate with an obtuse apex (Figure 1J). *S. koraiensis* has lanceolate to rarely ovate-shaped leaf blades with an acuminate apex (Figure 1B) and is serrate with a spinose tooth (Figure 1E). The stem of *S. koraiensis* is sparsely pubescent with non-glandular trichomes (Figure 1H), and the calyx is lanceolate with an acute to attenuate apex (Figure 1K). However, *S. takesimensis* has ovate leaf blades with an acute apex (Figure 1C), and has a serrate, almost without spinose tooth, marginal leaf blade (Figure 1F). The stem surface of *S. takesimensis* is glabrous (Figure 1I), and its calyx is semicircular, with a rounded apex (Figure 1L).


**Table 1.** Major determinants of *Scrophularia* species.

**Figure 1.** Stereomicroscope micrographs showing the morphology of the three *Scrophularia* species studied. (**A**–**C**) Apex of leaf blade. (**D**–**F**) Margin of leaf blade. (**G**–**I**) Surface of stem. (**J**–**L**) Calyx of flower and/or fruit. (**A**,**D**,**G**,**J**) *S. buergeriana*. (**B**,**E**,**H**,**K**) *S. koraiensis*. (**C**,**F**,**I**,**L**) *S. takesimensis*. TR, Trichomes. All scale bars = 1 mm.

#### *2.2. Harpagoside Content of Scrophulariae Species*

High-performance liquid chromatography (HPLC) chromatograms of *S. buergeriana*, *S. koraiensis*, and *S. takesimensis* are shown in Figure 2A,B. Figure 2A shows the HPLC chromatograms of the harpagoside standard compound and the three species of *Scrophulariae* monitored at 280 nm, because the maximum wavelength of harpagoside is 279.5 nm. Harpagoside was detected at approximately 50.1 min. The harpagoside content of *S. buergeriana*, *S. koraiensis*, and *S. takesimensis* was 1.94 ± 0.24 mg/g, 6.47 ±0.02 mg/g, 5.50 ± 0.02 mg/g, respectively (Table 2). The total ion chromatography (TIC) of the mass spectrometry (MS) spectrum was confirmed from 190–850 *m*/*z*, and the extracted ion chromatogram (XIC) of harpagoside in the samples was analyzed at 517.11 *m*/*z*, because harpagoside was detected at 517.11 *m*/*z* [M-H + Na]<sup>+</sup> (Figure 2B).

(1) Values are expressed as means <sup>±</sup> standard deviation (SD) of three samples for each species. λ

### *2.3. Cytotoxic E*ff*ects on Primary Murine BMM Growth*

The XTT assay was conducted to assess cytotoxicity during osteoclast differentiation. *S. buergeriana*, *S. koraiensis*, and *S. takesimensis* did not reduce cell viability at most of the concentrations tested. Treatment with 200 µg/mL *S. koraiensis* reduced cell viability from that of the normal control group (Figure 3A–C), but the difference was not statistically significant. μ

**Figure 3.** Cell viability affected by ethanol extracts of the three *Scrophularia* species (**A**) *S. buergeriana* (**B**) *S. koraiensis* and (**C**) *S. takesimensis*. \* *p* < 0.05, \*\* *p* < 0.01, and \*\*\* *p* < 0.001 vs. control (dimethyl sulfoxide; DMSO).

### *2.4. E*ff*ects on Osteoclast Di*ff*erentiation*

To compare the effect of *S. buergeriana*, *S. koraiensis*, and *S. takesimensis* on osteoclast differentiation, mouse BMMs treated with macrophage colony stimulating factor (M-CSF) and RANKL were cultured

in the presence or absence of ethanol extracts of the three *Scrophularia* species. RANKL and M-CSF induced differentiation of BMMs after incubation for 4 days. TRAP-positive osteoclasts were present in higher numbers in the control group, whereas treatment with *Scrophularia* species extracts inhibited the formation of TRAP-positive cells in a dose-dependent manner (Figure 4B). The *S. koraiensis* treatment group showed suppressed formation of RANKL-induced TRAP activity at concentrations of 100 µg/mL and 200 <sup>µ</sup>g/mL. <sup>μ</sup> <sup>μ</sup>

μ μ μ κ **Figure 4.** Effects on osteoclast differentiation of ethanol extracts of *Scrophularia buergeriana*, *S. koraiensis*, and *S. takesimensis* at concentrations of 50 µg/mL, 100 µg/mL, and 200 µg/mL. (**A**) Tartrate-resistant acid phosphatase (TRAP)-positive cells photographed (100× magnification) after bone marrow macrophages were cultured with macrophage colony stimulating factor (M-CSF) and nuclear factor (NF)-κB ligand (RANKL) in the presence of ethanol extracts of the three *Scrophularia* species. (**B**) TRAP-positive cells were counted as osteoclasts. \* *p* < 0.05, \*\* *p* < 0.01, and \*\*\* *p* < 0.001 vs. control (DMSO).

#### *2.5. E*ff*ects on Bone Resorption*

To evaluate the effects of *Scrophularia* species on bone resorption, mature osteoclasts and extracts of *S. buergeriana*, *S. koraiensis*, and *S. takesimensis* were applied to plates coated with hydroxyapatite for symbiotic culture of the osteoblasts. Although the area of hydroxyapatite-adsorbed osteoclasts was increased in the control, the resorption area was markedly decreased by treatment with extracts of the three *Scrophularia* species. The resorption inhibition effect was highest for *S. buergeriana*, followed by *S. koraiensis*, and then *S. takesimensis.* (Figure 5A,B).

**Figure 5.** Effects of *S. buergeriana*, *S. koraiensis*, and *S. takesimensis* on bone resorption by mature osteoclasts. (**A**) Hydroxypatite-adherent cells were collected and imaged under a light microscope. (**B**) Resorption areas quantified on hydroxyapatite-coated plates. \*\* *p* < 0.01 and \*\*\* *p* < 0.001 vs. control (DMSO).

#### **3. Discussion**

Osteoporosis, one of the major diseases attracting attention worldwide, is associated with lowered bone mass density as a result of an imbalance between the osteoblasts and osteoclasts that influence bone homeostasis [22,23]. Therapeutic agents such as bisphosphonates are currently used to treat bone resorption diseases, but the long-term use of these drugs leads to the suppression of bone formation and osteonecrosis [24]. Therefore, the beneficial pharmacological effects of plant-derived natural compounds have been advocated [10].

Several studies have reported that iridoid glycosides in *Scrophularia* plant extracts demonstrate anti-inflammatory, neuroinflammation, antioxidant, and hepatoprotective effects [11–13,15,25] Harpagoside, an iridoid glycoside, is a main bioactive component of *Harpagophytum procumbens* (family Pedaliaceae) root used to treat chronic rheumatism, osteoarthritis, and arthritis, and is an active constituent of *Scrophularia* species [21,26–28]. We quantified the harpagoside content in *S. buergeriana*, *S. koraiensis*, and *S. takesimensis* using HPLC/MS (Table 2, Figure 2). Harpagoside was detected in three *Scrophularia* species, with the highest content found in *S. koraiensis* extract.

In this study, we investigated the inhibitory effects exerted by three *Scrophularia* species containing harpagoside against osteoclast differentiation and bone resorption without cytotoxic effects in RANKL-induced cells. The cytotoxicity of *S. buergeriana*, *S. koraiensis*, and *S. takesimensis* was evaluated in BMMs and measured by XTT assay during osteoclast differentiation. Treatment with ethanol extracts of the three *Scrophularea* species did not reveal cytotoxic effects on BMMs up to 200 µg/mL, with more than 90% cell viability being observed.

μ RANKL, a bone formation biomarker, is associated with stimulation of osteoblasts and osteoclast differentiation [22]. The balance of bone formation is influenced by osteoblasts and bone resorption activity by osteoclastic cells [29]. Previous studies have demonstrated that harpagoside improved bone properties by inhibiting the formation of osteoclasts from BMMs and the maturation of osteoblast cells [21,30].

TRAP expression is associated with osteoclast maturation and differentiation and is a standard approach to the detection of osteoclasts [24]. In our study, TRAP staining indicated that the numbers and areas of TRAP-positive cells increased, whereas the *S. buergeriana*, *S. koraiensis*, and *S. takesimensis* treatment groups exhibited considerably fewer TRAP-stained osteoclasts, without any cytotoxicity. The results suggested that the three *Scrophularia* species inhibit osteoclast differentiation and formation in BMMs. The inhibitory effect of harpagoside was increased in a dose-dependent manner by downregulating TRAP expression in BMCs [30]. Therefore, it is presumed that the difference in efficacy of the three Scrophularia species in our results would be affected by the content of harpagoside. To study the inhibitory effects of *Scrophularia* species on bone resorption by osteoclast formation induced by RANKL, we investigated bone resorption pit generation in mature osteoclasts. Although the area of hydroxyapatite-absorbed osteoclasts was increased in DMSO controls, the resorption area was markedly decreased by treatment with the three *Scrophularia* species. The highest resorption inhibition effect was exerted by *S. buergeriana*, followed by *S. takesimensis*, and then *S. koraiensis.*

#### **4. Materials and Methods**

#### *4.1. Chemicals*

Harpagoside was purchased from Shanghai Sunny Biotech (Shanghai, China). HPLC-grade solvents were obtained from Merck (Darmstadt, Germany). Sodium 3′ -[1-(phenyl-aminocarbonyl)-3,4 -tetrazolium]-bis(4-methoxy-6-nitro) benzene sulfonic acid hydrate (XTT reagent) was purchased from Sigma-Aldrich (St. Louis, MO, USA).

#### *4.2. Plant Materials*

Samples of the three *Scrophularia* species *(S. buergeriana*, *S. koraiensis*, and *S. takesimensis*) were obtained from the Department of Herbal Crop Research (Eumseong-gun, Chungcheongbuk-do, Korea). The same vouchers were used for both chemical and morphological analyses (*S. buergeriana*; 2-18-0144, *S. koraiensis*; 2-18-0145, *S. takesimensis*; 2-18-0146) from the Korean Herbarium of Standard Resources (KHSR), Korean Institute of Oriental Medicine.

#### *4.3. Comparative Morphology of Scrophularia Species*

*S. buergeriana* and *S. koraiensis* were originally collected from the experimental field of the National Institute of Horticultural and Herbal Science (NIHHS, Korea), and *S. takesimensis* was collected from the natural population on Ulleung-do Island. Vouchers of the studied medicinal materials (*S. buergeriana* (2-18-0144), *S. koraiensis* (2-18-0145), and *S. takesimensis* (2-18-016)) were deposited in the Korean Herbarium of Standard Herbal Resources (Index Herbariorum code: KIOM) at the Korea Institute of Oriental Medicine (Naju, Korea). We observed the voucher specimens of the three species (Figure S1) sing an Olympus SZX16 stereomicroscope (Olympus, Tokyo, Japan) for morphological comparison.

#### *4.4. HPLC*/*MS Analysis*

*S. buergeriana* (778.14 g), *S. koraiensis* (69.15 g), and *S. takesimensis* (92.5 g) were refluxed in 70% ethanol (*v*/*v*) for 2 h, and these extracts were filtered through filter paper and evaporated in vacuo. The powders of the 70% ethanol extract for *S. buergeriana*, *S. koraiensis*, and *S. takesimensis* were 363.83 g (46.76% of yield, *w*/*w*), 30.40 g (43.96% of yield, *w*/*w*), and 53.91 g (58.28% of yield, *w*/*w*), respectively, and these extracts were stored at 4 ◦C. Accurately weighed powders of three 70% ethanol extracts (10 mg) were dissolved in 10 mL 70% ethanol and filtered through a 0.2 µm syringe filter prior to HPLC analysis. HPLC was performed using a Waters e2695 Separation Module (Waters Corporation, Milford, MA, USA) and a 2998 photodiode array detector (Waters), combined with an Acquity QDa detector and a micro-splitter (IDEX Health & Science LLC, Oak Harbor, WA, USA) for MS. A Kinetex

Phenyl-hexyl 100A column (4.6 × 250 mm, 5 µm, Phenomenex Inc., Torrance, CA, USA) was utilized. The mobile phase was 0.05% aqueous formic acid (A), methanol (B), and acetonitrile (C) using the following gradient program: 100% A (2 min) → 96% A (3% B and 1% C, 7 min) → isocratic 85% A (11% B and 4% C, from 15 min to 20 min) → 70% A (23% B and 7% C, 35 min) → 50% A (35% B and 15% C, 45 min) → 30% A (50% B and 20% C, 55 min). The following conditions were applied: The flow rate was 0.8 mL/min, injected volume was 10 µL, and UV wavelength was monitored from 195 nm to 400 nm. The QDa mass detector employed the following conditions: Nitrogen carrier gas, 0.8 kV electrospray ionization (ESI) capillary, 600 ◦C probe temperature, 15 V con voltage, 120 ◦C source temperature, 20:1 split. TIC was monitored from 150 *m*/*z* to 850 *m*/*z*. Harpagoside was detected at 280 nm and 517.11 *m*/*z*.

#### *4.5. BMM (Bone Marrow Macrophage) Isolation and Culture*

Five-week-old ICR mice were sacrificed by cervical dislocation, the thigh and shin bones were aseptically excised, and the soft tissue was removed. After cutting both ends of the iliac crest, bone marrow cells were obtained by flushing both ends of the bone material using a 1 mL syringe. The isolated BMMs were incubated in a culture dish for 1 day in α-MEM medium containing 10% FMS and 1% penicillin/streptomycin, and unsaturated cells were collected. After 3 days, attached macrophages were used in this experiment. Macrophages were treated with M-CSF (30 ng/mL) and RANKL (100 ng/mL), and incubated with *S. takesimensis*, *S. koraiensis*, and *S. buergeriana* extracts at concentrations of 50 µg/mL, 100 µg/mL, and 200 µg/mL. The day following subculture, the cultured cells were stained with a TRAP solution. Cells with three or more nuclei among red-stained cells were considered differentiated osteoclasts, and the degree of differentiation was measured.

#### *4.6. Cell Cytotoxicity*

BMMs were seeded in 96-well plates at a density of 1 <sup>×</sup> <sup>10</sup><sup>4</sup> cells/well and were treated with M-CSF (30 ng/mL) and 50 µg/mL, 100 µg/mL, and 200 µg/mL ethanol extracts of *Scrophularia* species (*S. takesimensis*, *S. koraiensis*, and *S. buergeriana*) for 3 days. After 3 days, 50 µL XTT reagent (Sigma-Aldrich) was applied for 4 h. The optical density was measured at 450 nm using an ELISA reader (Biotek Instruments, Winooski, VT, USA).

#### *4.7. Bone Resorption Pit Assay*

To obtain mature osteoclasts, BMMs obtained from the shin and thigh bones of 5-week-old ICR mice and osteoclasts isolated from the skull of 1-day-old ICR mice were added to a 90 mm culture plate coated with collagen. To produce a symbiotic culture, 1α,25-Dihydroxyvitamin D3 (VitD3) and prostaglandin E2 were added. After incubation for 6 days, cells were removed by treatment with 0.1% collagenase and added to a 96-well plate coated with hydroxyapatite. At the same time, *S. takesimensis*, *S. koraiensis*, and *S. buergeriana* extracts were added to the hydroxyapatite-coated plates at a concentration of 200 ug/mL and cultured for 24 h. The cells were washed with distilled water and observed through an optical microscope (Nikon, Tokyo, Japan). The hydroxyapatite-absorbed portion was imaged using micrography, and the area was measured using ImageJ software version 1.51 (National Institutes of Health, Bethesda, MD, USA).

#### *4.8. Statistical Analysis*

Data were expressed as the mean ± standard deviation (SD). Statistical analysis was performed by analysis of variance (ANOVA), followed by a multiple comparison procedure using Dunnett's test. A value of *p* < 0.05 indicated significant difference.

### **5. Conclusions**

Our results suggest that the *Scrophularia* species may prevent bone loss by inhibiting osteoclast differentiation and born resorption without causing cytotoxicity. It was confirmed that similar metabolites were contained in the *Scrophularia* species, and their extracts exerted similar efficacy. *S. buergeriana* and *S. koraiensis* demonstrated inhibitory effects against osteoclast differentiation and bone resorption and had the highest harpagoside content. Therefore, the three *Scrophularia* species, including *S. koraiensis*, may be potentially therapeutic for treating bone disorder diseases, but the mechanism underlying their bioactivity requires further study.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2223-7747/9/12/1656/s1.

**Author Contributions:** Conceptualization, J.-S.K. and J.-H.S.; the morphology study, J.-H.S.; in-vitro analysis, H.-H.N., Y.-S.S. and J.M.C.; resources, I.P. and S.Y.; chemical analysis, A.Y.L.; genetic analysis, B.C.M.; funding acquisition, B.C.M., J.-H.S. and J.-S.K.; writing—review and editing, H.-H.N., J.-H.S. and J.-S.K.; All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by the Korea Institute of Oriental Medicine, Daejeon, South Korea (grant numbers K18402, KSN2012320) and the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (NRF-2020R1A2C1004272 and NRF-2020R1A2C1100147).

**Acknowledgments:** We thank the Korean Herbarium of Standard Herbal Resources (herbarium code KIOM) for the provision of materials. J.-H.S. sincerely thanks Jeong Hoon Lee (Department of Herbal Crop Research, NIHHS) for assistance in material sampling.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Abbreviations**

BMM, bone marrow-derived macrophage; RANKL, receptor activator of nuclear factor (NF)-κB ligand; TIC, total ion chromatography; TRAP, tartrate-resistant acid phosphatase; XIC, extracted ion chromatogram; XTT, sodium 3 ′ -[1-(phenyl-aminocarbonyl)-3,4-tetrazolium]-bis(4-methoxy-6-nitro) benzene sulfonic acid hydrate.

#### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

### **Functional Attributes and Anticancer Potentialities of Chico (***Pachycereus Weberi***) and Jiotilla (***Escontria Chiotilla***) Fruits Extract**

**Luisaldo Sandate-Flores <sup>1</sup> , Eduardo Romero-Esquivel <sup>1</sup> , José Rodríguez-Rodríguez <sup>1</sup> , Magdalena Rostro-Alanis <sup>1</sup> , Elda M. Melchor-Martínez <sup>1</sup> , Carlos Castillo-Zacarías <sup>1</sup> , Patricia Reyna Ontiveros <sup>2</sup> , Marcos Fredy Morales Celaya <sup>3</sup> , Wei-Ning Chen <sup>4</sup> , Hafiz M. N. Iqbal 1,\* and Roberto Parra-Saldívar 1,\***


Received: 30 September 2020; Accepted: 30 October 2020; Published: 22 November 2020 -

**Abstract:** Mexico has a great diversity of cacti, however, many of their fruits have not been studied in greater depth. Several bioactive compounds available in cacti juices extract have demonstrated nutraceutical properties. Two cactus species are interesting for their biologically active pigments, which are chico (*Pachycereus weberi* (*J. M. Coult.) Backeb*)) and jiotilla (*Escontria chiotilla* (*Weber) Rose*)). Hence, the goal of this work was to evaluate the bioactive compounds, i.e., betalains, total phenolic, vitamin C, antioxidant, and mineral content in the extract of the above-mentioned *P. weberi* and *E. chiotilla*. Then, clarified extracts were evaluated for their antioxidant activity and cytotoxicity (cancer cell lines) potentialities. Based on the obtained results, Chico fruit extract was found to be a good source of vitamin C (27.19 ± 1.95 mg L-Ascorbic acid/100 g fresh sample). Moreover, chico extract resulted in a high concentration of micronutrients, i.e., potassium (517.75 ± 16.78 mg/100 g) and zinc (2.46 ± 0.65 mg/100 g). On the other hand, Jiotilla has a high content of biologically active pigment, i.e., betaxanthins (4.17 ± 0.35 mg/g dry sample). The antioxidant activities of clarified extracts of chico and jiotilla were 80.01 ± 5.10 and 280.88 ± 7.62 mg/100 g fresh sample (DPPH method), respectively. From the cytotoxicity perspective against cancer cell lines, i.e., CaCo-2, MCF-7, HepG2, and PC-3, the clarified extracts of chico showed cytotoxicity (%cell viability) in CaCo-2 (49.7 ± 0.01%) and MCF-7 (45.56 ± 0.05%). A normal fibroblast cell line (NIH/3T3) was used, as a control, for comparison purposes. While jiotilla extract had cytotoxicity against HepG2 (47.31 ± 0.03%) and PC-3 (53.65 ± 0.04%). These results demonstrated that Chico and jiotilla are excellent resources of biologically active constituents with nutraceuticals potentialities.

**Keywords:** pachycereus weberi; escontria chiotilla; antioxidant activity; phenolic compounds; betalains; food composition; food analysis; cytotoxicity

#### **1. Introduction**

Mexico has a great diversity of cacti [1], but many of the fruits from these plants have not been studied. Cacti fruits have health benefits such as anticlastogenic capacity [2], hepatoprotective effect [3], antimicrobial activity [4], notable antioxidant capacity [5,6], and anticancer activity [7,8]. Currently, the scientific community is interested in two cacti fruits for their pigments, chico (*Pachycereus weberi (J. M. Coult.) Backeb*) and jiotilla (*Escontria chiotilla (Weber) Rose)*. *Pachycereus weberi* is typical in the south-central Mexico region, specifically the arid and semi-arid regions in Puebla, Guerrero, Morelos, and Oaxaca [1]. The fruit of *Pachycereus weberi* is better known as "chico fruit" and "tuna de cardón" [9] (Figure 1). *P Weberi* fruits are ellipsoidal, 6 to 7 cm in diameter [10]. In Santiago Quiotepec, Oaxaca, fruits of this species harvested and commercialized in local markets [11]. The fruit production is 509.3 ton in Tehuacán Valley México from *Pachycereus weberi* [12].

**Figure 1.** (**A**) *Pachycereus weberi* plant and (**B**) *P. weberi* fruit without prickles (top) with prickles (bottom).

Nowadays, the knowledge about antioxidant activity and phenolic compounds of the fruit from the Mexican plant *Pachycereus weberi* is limited. *Escontria chiotilla,* a columnar cactus is found in Guerrero, Michoacán, Puebla, Oaxaca, and Morelos. The fruit of *Escontria chiotilla* has a red shell with the presence of scales, spherical, and are 5 cm in diameter (Figure 2). The period of growth of the fruit ranged 140 to 175 days [13] and is usually named as "chonostle" and "jiotilla" [14]. The production of this fruit is approximately 99.5 tons in the valley of Tehuacan [12] and is commercialized in local markets, and people from the community of Coxcatlan prepare ice cream and jellies for local trade [15]. Different studies have shown that the pigments present in the jiotilla are vulgaxanthin I, vulgaxanthin II, indicaxanthin, and betanin [16]. These pigments have been extracted using acidified mucilage and microencapsulated for subsequent characterization [17].

Based on the literature, cacti fruits have cytotoxic properties in cancer cell lines [7,8]. For instance, prickle pear juices (*Opuntia spp.*) have shown to produce decreasing viability of cancer cell in vitro, especially colon (HT29/Caco-2) and prostate cancer (PC-3) [7,8]. Indeed, these are promising results as around 70% of cancer deaths occur in low and middle incomes countries [18] and in 2018, 1.8 million colorectal and 1.28 million prostate cases were reported [18]. Furthermore, several studies have shown that fruits growth in semi-arid and arid lands present high mineral content [19,20]. This aspect is relevant as minerals are essential in metabolism and homeostatic of the human body, deficiency of minerals can lead to symptoms of common disorders and diseases as osteoporosis and anemia [21]. One of the essential micronutrients is zinc, and the deficiency of this restricts childhood growth and decreases resistance to infections [22]. In the world, around 20% of children under five years of age are stunted [23]. The principal food with a high concentration of zinc are oysters, red meat, and poultry [24]. The food mentioned before may be more challenging to access for low-income populations [25]. Oaxaca, Guerrero, and Puebla are Mexican states with low-income populations [26]. It is crucial to find other sources of zinc that are accessible to all communities.

**Figure 2.** (**A**) *Escontria chiotilla* plant, (**B**) *E. chiotilla* flower, and (**C**) *E. chiotilla* fruit.

Studies on potentialities of cacti fruits have gradually increased in recent years, but there is still a lack of information regarding antioxidant capacity, metal content, or phytochemical content of *P. weberi* and *E. chiotilla* fruits pulp and juice extracts. Therefore, this study aimed to evaluate the physiochemical characteristic, antioxidant activities, and anticancer potentialities of *P. weberi* and *E. chiotilla* fruits pulp and/or juice extracts.

#### **2. Materials and Methods**

#### *2.1. Chemicals*

2,2-Diphenyl-L-picryl-hydrazyl (DPPH) (C18H12N5O6, Item D9132, Lot 115K1319) was obtained from Sigma-Aldrich (Steinheim, Germany). Methyl alcohol (CH3OH, Item MS1922, Lot 17020429) was obtained from Tedia High Purity Solvents (Fairfield, OH, USA). Folin&Ciocalteu's Phenol Reagent (C7H6O5, Item F9252, Lot SHBD7847V) was obtained from Sigma Aldrich (St. Louis, MO, USA). Sodium Carbonate (Na2CO3, Item S7795, Lot BCBP4310V) was obtained from Sigma Aldrich (Steinheim, Germany). Gallic acid monohydrate (C7H6O5, Item 398225, Lot MKBD1204) was obtained from Sigma-Aldrich (Shanghai, China). Sodium chloride (NaCl, DEQS290000100, Lot CS180307-03) was obtained from Desarrollo de Especialidades Químicas, S.A. de C.V. (Nuevo Leon, Mexico). Potassium persulfate (K2S2O8, Item 216224, Lot MKBR1923V) was obtained from Sigma-Aldrich (St. Louis, MO, USA). Disodium phosphate (Na2HPO4, 35902, Lot 536405) obtained from Productos Quimicos Monterrey S.A. de C.V. (Nuevo Leon, Mexico). Potassium chloride (KCl, S32204, Lot 536161) obtained from Productos Químicos Monterrey S.A. de C.V. (Nuevo Leon, Mexico). 2,2′ -Azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) ABTS (C18H24N6O6S4, Item A9941, Lot SLBT7934) obtained from Sigma-Aldrich (St. Louis, MO, USA). Potassium phosphate monobasic (KH2PO4, Item P0662, Lot SLBL9298V) was obtained from Sigma-Aldrich (Tokyo, Japan). Nitric acid (HNO3, Item UN2031, Lot SCA7227127) was obtained from SPC Science (Baie-d'Urfé, QC, Canada). Ammonium acetate (CH3COONH4, Item 0596, Lot L15C82) was obtained from J. T. Baker (Estado de Mexico, Mexico). Milli-Q water purification system is used to obtain the water that was used to perform the procedures (Q-POD, Darmstadt, Germany). Glacial acetic acid (Item AE4001, Lot 1004007) was bought from Tedia (Fairfield, OH, USA). 2,4,6-Tris(2-pyridyl)-s-triazine (TPTZ) (C18H12N6, Item T1253, Lot BCBT8262) was bought from Sigma Aldrich (Buchs, Switzerland). Iron (lll) chloride hexahydrate (FeCl3·6H2O, ItemF2877, Lot MKBR8795V) was bought from Sigma Aldrich (Shanghai, China). Hydrochloric acid (HCl, Item 00636, Lot 71973) was bought from CTR Scientific (Nuevo Leon, Mexico). Gallic acid monohydrate (C7H6O5·H2O, item 398225, Lot MKBD1204) was bought from Sigma Aldrich (Shanghai, China). Caffeic acid (C9H8O4, item C0625, Lot 089K1114) was obtained from Sigma-Aldrich (Buch, Switzerland). p-Coumaric acid (C9H8O3, item C9008, Lot BCBN0412V) was bought

from Sigma-Aldrich (United Kingdom). Violuric acid (C4H3O4·H2O, item 95120, Lot BCBN0412V) and fumaric acid (C4H4O4, item 47910, Lot BCBM2191V) were obtained from Sigma-Aldrich (Vienna, Austria).

#### *2.2. Collection and Preprocessing of Chico and Jiotilla*

A single batch of chico fruits (10 kg) was collected from a crop field in Ahuatlán (Puebla, México). The sampling area was situated between 18◦34′ latitude (North) and 98◦15′ longitude (West). The collected fruit samples were stored at 5 ◦C for 48 h and processed within 72 h of harvest from the plant. Whereas, a batch of jiotilla fruits (22 kg) was collected from a crop field in Santa María Zoquitlan, Tlacolula, Oaxaca, Mexico 16◦33′ latitude (North) and 96◦2315′ longitude (West). The collected fruit samples were stored at 5 ◦C for 12 h and processed within 24 h of harvest from the plant. As collected fruits of both plants, i.e., chico and jiotilla, were preprocessed by washing with tap water and Extran MA05 (Merck, Item 1400001403, Lot Mx1400005004, Estado de Mexico, Mexico). Then, the prickles were removed manually with care using a sterile blade. The electric extractor (Model TU05, Turmix ML, Estado de México) was used to prepare a seed-free pulp sample. This method was used since it is a quick way to eliminate the seeds. Subsequently, around 5 g of seed-free pulp samples were placed in 40 mL vials, 10 mL of deionized water was added, and then the vials were agitated for 15 min. The resultant mix was filtered (Whatman paper grade 4, 150 mm, Item 1009150, GE Healthcare Life Sciences, Little Chalfont, UK) under dark conditions to avoid compounds degradation. The extract obtained (40 mL) was placed in a 100 mL (chico fruit) or 50 mL (jiotilla) volumetric flask and used in the subsequent analysis. Different dilutions were made because a great content of antioxidant capacity was expected in chico fruit. Three extractions for each fruit were carried out following the same conditions.

#### *2.3. Clarified Juice Extract Preparation*

Seed-free pulp (30 g) from each fruit was placed in 50-mL polypropylene centrifuge tubes (Corning®, Tewksbury, MA, USA). The tube was subjected to centrifugation (4000<sup>×</sup> *<sup>g</sup>*, 4 ◦C, 10 min, Model SL 40R, Thermo Fisher Scientific, Langenselbold, Germany). The supernatant was filtered in the dark through 150 mm Whatman paper grade 4 (item 1009150, GE Healthcare Life Sciences, Little Chalfont, UK). The pellets were removed, and the clarified supernatant from the pulp of each fruit was used as a clarified juice extract.

#### *2.4. Physiochemical Characterization*

Weight of pulp and peel were determined with an analytical balance (OHAUS, model Scout Pro SP4001, China). Equatorial and polar diameters were measured with a caliper (Vernier from Truper, model H87). The pH was determined with Termo Fisher Scientific, ORION 3 STAR pH Benchtop (Singapore, Singapore). Total soluble solids (Brix) were determined with a refractometer HI96811 (HANNA, Smithfield, RI, USA), and moisture content (%) was obtained according to the gravimetric method [27]. A general proximate analysis was applied to the pulps, the method used was AOAC 1999 [28].

#### *2.5. Betacyanins and Betaxanthins Quantification*

A spectrophotometric method is used to determine the concentration of betacyanins and betaxanthins in the pulp of the fruit. The spectrophotometer used was Model DR 500 (Hach Lange GmbH, Düsseldorf, Germany), following the methods described by Sandate-Flores et al. [29]. For the dilution of the samples, water Milli-Q was preferred, the dilution was 1:10, and it was carried out in 5 mL volumetric flasks. The extinction coefficients used were the following: for betacyanin (E1% = 60,000 L mol−<sup>1</sup> cm−<sup>1</sup> , λ = 540 nm) and for betaxanthin (E1% = 48,000 L mol−<sup>1</sup> cm−<sup>1</sup> , λ = 480) [30]. The betalains concentration was showed in milligrams per gram (mg/g, dry weight). The procedure for determining moisture content is described in Section 2.4.

#### *2.6. Vitamin C Quantification*

The vitamin C quantification was performed based on the procedure described earlier by Santos et al. [31]. The sample was passed through a nylon filter of 0.20 mL (2 mL of the volumetric extract), placed in vials for HPLC. The HPLC-FLD system (Agilent 1200 HPLC, Santa Clara, CA, USA) was injected with 25 µL of the sample, for which a Zorbax Eclipse XDB C18 column was used (5 µm, 150 9 4.6 mm i.d.) (Agilent, Santa Clara, CA, USA). As eluent solvents, a solution of 10 mM ammonium acetate at a pH of 4.5 was used (phase A), and a methanol solution with 0.1% acetic acid was phase B. Using 80% of phase A, elution was carried out in isocratic mode for 7 min with a flow rate of 1 mL/min. For the measurement of ascorbic acid, the chromatograms were monitored at Exc = 280 and Em = 440 nm. A six-point calibration curve was made using an external standard of ascorbic acid (0.01 to 10 mg/L), the curve used for the quantification of vitamin C.

#### *2.7. Antioxidant Activity*

#### 2.7.1. Folin-Ciocalteu Method (Total Antioxidant Capacity)

The Folin-Ciocalteu colorimetric method was used to determine the total antioxidant capacity [32]. This colorimetric method was carried out following the procedure of Singleton et al. [33]. The analysis was performed directly in 96 well plates. Briefly, around 20 µL of the diluted sample was placed in Milli-Q water, then 100 µL of folin reagent at a concentration of 10% was added. After 5 min of the reaction period, around 80 µL of sodium carbonate at a concentration of 7.5% *w*/*v* was added. The above reaction mixture was then incubated at 37 ◦C for 90 min in the dark. Finally, the sample containing 96-well microplate was placed in a plate reader and scanned at 765 nm. The calibration curve was made using different concentrations (50 to 200 mg/L) of gallic acid. All measurements were made in triplicate, and Milli- Q water was used as blank.

#### 2.7.2. ABTS

For the ABTS method, the reported method of Re et al. [34] was used. ABTS is a single electron transfer (ET) reaction-based assay. The preparation of PBS was carried out to add 0.8 g of NaCl, 0.02 g of KH2PO4, 0.115 g of Na2HPO4, 0.02 g of KCl, and 0.02 g of NaN3. Then the solution was filled to 100 mL of Milli Q water. For the ABTS reagent, it was added 38.4 mg of ABTs 1 mM, 6.62 mg of potassium persulfate 2.45 mM, and 10 mL of the solution of PBS. After mixing both solutions, they stirred for 16 h, keep in the dark. To measure the absorbance, the spectrophotometer (Model DR 500, Hach Lange GmbH, Düsseldorf, Germany) was used at 734 nm, a dilution of the initial reagent was needed until it has an absorbance of 0.7 (40 mL of PBS with 3 mL of ABTS solution). The procedure was 20 µL diluted sample, and 2 mL of ABTS solution (absorbance 0.7) were incubated at 30 ◦C in a water bath for 6 min, then absorbances were read. Trolox was used as standard with concentrations of 5 ppm to 200 ppm. All measurements were made in triplicate. Milli- Q water was used as blank.

#### 2.7.3. Ferric Reducing Antioxidant Power (FRAP)

Based on the procedure of Benzie and Strain [35], the antioxidant capacity of the samples was obtained by the spectrometric method. To prepare the FRAP reagent, three solutions were made. The first one was the acetate buffer 300 mM pH 3.6, for this solution sodium acetate trihydrate, glacial acetic acid, and distilled water were mixed. For the second solution, iron tripyridyltriazine (TPTZ), concentrate HCl and distilled water were used. Then the last solution was FeCl3·6H2O and Milli-Q water. The three solutions mixed in relation to 10:1:1. FRAP reagent was prepared with 100 mL acetate buffer, 10 mL TPTZ, and 10 mL FeCl3·6H2O. These compounds were mixed at 30 ◦C for 30 min. After the incubation, 100 µL of the sample was added to 3 mL FRAP reagent. Trolox was used as a standard with concentrations of 10 to 200 ppm. Finally, absorbance was measured at 593 nm against a blank of the reagent. All measurements were made in triplicate.

#### 2.7.4. α-α-. diphenyl-β-picrylhydrazyl (DPPH)

Based on the procedure described by Brand-Williams [36]. Around 0.0148 g of DPPH was weighed and placed in a volumetric flask (25 mL). Then the volumetric flask was filled to the mark with methanol (mother solution). 1 mL of the solution before mention was placed in a volumetric flask (25 mL) and filled to the mark with methanol (diluted solution). The absorbance was measured in a spectrophotometer (Model DR 500, Hach Lange GmbH, Düsseldorf, Germany). The readings were taken in 2.5 mL cuvette, mixing 75 µL of sample and 3 mL of DPPH diluted solution. The reaction takes place 16 min after mixing reactants. Then absorbances were read at 515 nm. The calibration curve was made using Trolox with concentrations of 5 ppm to 200 ppm. All measurements were made in triplicate.

#### *2.8. Minerals Content Analysis*

Mineral content was evaluated with digestion in nitric acid (HNO3), ICP from Thermo Scientific, model iCAP 6000 Series (Cambridge, England) was used. The following wavelengths (nm) were used, i.e., Ca (396.8), K (766.4), Fe (259.9), Mg (279.5), Mn (257.6), Na (589.5), P (177.4), Cu (324.7), Se (196.0), and Zn (213.8). The digestion of 1 g of sample was carried out during 4 h at a range 85 to 90 ◦C. The samples were passed through a paper filter (Whatman paper grade 41,110 mm, Item1441110, GE Healthcare Life Sciences, Little Chalfont, UK). The heating was performed with a hot plate (Model Type 2200, Thermo Fisher Scientific, Dubuque, IA, USA). Three samples for each fruit extract were carried out following the same conditions.

#### *2.9. Cell Lines and In Vitro Cancer Cell Viability*

Normal fibroblast cell line (NIH/3T3), as a control cell line, and four different mammalian cancer cell lines mammary (MCF-7), prostate (PC3), colon (Caco-2) and hepatic (HepG2) were propagated in DMEM-F12 medium containing 10% FBS (Fetal Bovine Serum) (Gibco, Grand Island, NY, USA) and maintained in 5% of CO<sup>2</sup> atmosphere at 37 ◦C and 80 % of humidity. The cytotoxicity assay was performed in plates of 96-wells each well was prepared with 100 µL of a cell suspension containing <sup>5</sup> <sup>×</sup> <sup>10</sup><sup>5</sup> cells/mL of cancer cells (MCF-7, PC3,Caco-2, and HepG2) and NIH/3T3 after 12 h 100 µL a dilution at 4% of filtrated clarified juice extract in Milli Q water were added to each well in the cell giving a final concentration of 2% of juice in the cell growth media in triplicate (Appendix A). The culture medium without cells was used as a blank. After incubation at 37 ◦C under 5% CO<sup>2</sup> for 48 h, 20 µL Cell Titer 96®AQueous One Solution Cell Proliferation Assay (Promega, Madison, WI, USA) was used to determine % of cell viability. Absorbance was measured at 490 nm in a microplate reader (Synergy HT, Bio-Tek, Winooski, VT, USA). Cell viability was computed using the average absorbance units obtained from the wells and expressed as a percentage of the untreated cells well.

#### *2.10. Phenolic Analysis by HPLC*

The determination of phenolic compounds in clarified juices extracts were analyzed by a Perkin Elmer HPLC (Altus 10, Waltham, MA, USA) coupled with a UV-Vis detector. A column Eclipse XDBC18, 5 µm, 150 mm × 4.6 mm (Agilent Technologies, Santa Clara, CA, USA) was used for the analysis. A reversed-phase HPLC was performed, and gradient elution was performed by varying the proportion of solvent A (acidified water with acetic acid pH 2.5) to solvent B (methanol), with a flow rate of 0.8 mL/min. The mobile phase composition started at 100 % solvent A for 3 min, followed by an increase of solvent B to 30% 3 to 8 min, 50% B 8 to 15 min, 30% B 15 to 20 min, and then bring mobile phase composition back to the initial conditions. The reference standards were gallic acid, caffeic acid, coumaric acid, and fumaric acid. The calibration curve range was 10 to 80 mg/L. The aforementioned compounds were selected based on the literature of cacti fruits (*Opuntia ficus Indica and Opuntia littoralis*) [37,38].

#### **3. Results and Discussion**

#### *3.1. Physiochemical Characterization*

As is shown in Table 1, chico fruit averaged weight was 77.0 g, equatorial and polar diameters were 4.9 and 6.2 cm, respectively (Figure 1b). Regarding its epicarp, this was higher than its mesocarp, and the percentage of edible fruit was 36.6%. With respect to jiotilla, its average weight was 18.7 g, the edible fruit of jiotilla was 51.9%, and the equatorial and polar diameters were 2.8 and 3.4 cm, respectively. Table 2 shows the proximate analysis of chico fruit and jiotilla. Table 3 shows the total soluble solids, pH, and antioxidants activity in the fruits under study. The pH values were 4.5 (chico fruit) and 4.17 (jiotilla). Regarding the total soluble solids, they were 12.6 (chico fruit) and 7.4 (jiotilla) ◦Brix.

**Table 1.** Physical characterization of the collected plant samples.


<sup>1</sup> Results are shown as average <sup>±</sup> standard deviation.

**Table 2.** General proximate analysis of the collected plant samples.


<sup>1</sup> Results are shown as average <sup>±</sup> standard deviation.



<sup>1</sup> Results are shown as average <sup>±</sup> standard deviation. <sup>2</sup> Ascorbic acid. <sup>3</sup> Gallic acid. <sup>4</sup> Trolox were used in calibration curves. DW: dry weight; FS: fresh sample.

Comparing the edible percentage, the pitaya (*Stenocereus pruinosus*) fruit has a higher edible percentage (72.5%) [39] than chico fruit and jiotilla. Although the chico fruit (36.6%) and jiotilla (51.9%) have a pulp percentage higher than xoconostle (*Opuntia matudae*) (12% to 18%) [40]. The carbohydrates percentages are higher comparing them with the red pitaya (*Stenocereus pruinosus*) (10.2 ± 0.24%) and orange pitaya (*Stenocereus pruinosus*) (8.5 ± 0.16%) [39], but carbohydrates percentages are lower than presented in other fruits such as Banana (*Musa acuminata*) (22.84%) and Mango (*Mangifera indica*) (17%) [41]. The crude fiber in chico fruit and jiotilla was higher than the percentage in red pitaya (*Stenocereus pruinosus*) (0.67 ± 0.09 %), orange pitaya (0.53 ± 0.02 %) (*Stenocereus pruinosus*) [39]. Furthermore, the protein contents are lower than jackalberry tree fruit (*Diospyros mespiliformis*) (9.28 ± 1.14%) [42], but chico and jiotilla protein percentage are higher than apple (*Malus domestica*) (0.26%) [41].

#### *3.2. Antioxidant Capacity of Pulps*

Jiotilla (2.32 ± 0.23 mg/g dw) has a higher betacyanin concentration than chico fruit (1.55 ± 0.05 mg/g DW). The concentrations of betaxanthins in chico fruit and jiotilla were 1.46 ± 0.05 and 4.17 ± 0.35 mg/g DW, respectively. The concentration of vitamin C in the chico fruit was 27.19 ± 1.95 mg/100 g FS, while in jiotilla it was not detected. The total antioxidant capacities (Folin-Ciocalteu method) of chico fruit and jiotilla were 113.16 ± 5.82 and 83.40 ± 7.32 mg/100 g FS, respectively. The antioxidant activities determined as ferric reducing-antioxidant power (FRAP) for chico fruit and jiotilla were 0.27 ± 0.002 and 0.315 ± 0.003 mmol/100 g FS, respectively. In chico fruit and jiotilla, the values obtained in ABTS assay were 0.864 ± 0.023 and 0.708 ± 0.124 mmol/100 g FS, respectively. The antioxidant capacity of the fruits understudy was measured by the DPPH assay, which has been employed to measure the antioxidant capacity as well [43]. The free radical scavenging capacities of chico fruit and jiotilla were 34.52 ± 3.43 and 57.17 ± 5.28 mg/100 g FS, respectively. When compared with other cacti fruits, the betacyanins concentrations were lower than the values reported for red pitaya 2.86 ± 0.38 (*Stenocereus pruinosus*) mg/g DW [39] and prickly pear Rojo cenizo (*Opuntia ficus indica*) 5.95 ± 0.21 mg/g DW [44]. Betaxanthins concentration in jiotilla was higher than red pitaya (*Stenocereus Stellatus*) 1.51 ± 0.06 mg/g DW [45] and orange pitaya (*Stenocereus pruinosus*) 2.67 ± 0.27 mg/g DW [39]. Betalains concentration in chico fruit was lower compared with beetroot (*Beta vulgaris*) cultivar little ball in the flesh 3.6 ± 0.2 mg/g DW of betanin and 1.9 ± 0.1 mg/g DW in betaxanthins [46]. Vitamin C concentration in chico fruit, compared with other fruits is lower than guava (*Psidium guajava*) 131 ± 18.2 mg/100 g FS [47]. The fruits for (*Actinidia spp.*) cultivars and citrus species are an excellent source of vitamin C, comparing the (*Actinidia sp.)* species is reported that kiwi (variety Sanuki Gold) has 156 ± 31.2 mg/100 g FS [48] higher than chico fruit concentration. For citrus fruits like lemon (*Citrus limon*), the amount of vitamin C reported is 34 mg/100 g FS [49] and for orange fruit (*Citrus aurantium)* 36.1 mg/100 g FS [47]. It is important to highlight that chico fruits have a similar concentration of vitamin C like lemon juice and orange fruits. When compared total antioxidant capacity (Folin-Ciocalteu method) with some of the fruits recognized to have the highest values, such as raspberry and blackberry, the concentrations were lower. For instance, raspberry (*Rubus idaeus L.*) is reported to present a value of 1489 ± 4.5 mg GAE/100 g FS [50] and for cultivar Aksu Kırmızısı 1040.9 ± 15.9 milligrams of gallic acid per 100 g of FS [51]. Furthermore, a direct comparison of our results with other fruits such as apple red delicious (*Malus domestica*) (73.96 ± 3.52 mg GAE/100 g fw) [52], papaya (*Carcinia papaya Linn*) (54 ± 2.6 mg GAE/100 g FS) [53] and peach (*Prunus persica*) 27.58 ± 1.57 GAE/100 g FS [52] and it is reported for mango (*Mangifera indica L.*) in dry weight 1.64 ± 0.49 mg GAE/g DW [54] (in consideration of this the concentrations of chico fruit and jiotilla were changed to dry weight 9.19 ± 0.47 and 7.09 ± 0.54 mg GAE/g DW, respectively) Jiotilla and chico, represents higher antioxidant activity than fruits mentioned before.

The antioxidant activity determined as FRAP was lower than reported by red rose grape (*Vitis vinifera*) (0.49 ± 0.04 mmol/100 g FS) [55]. Nevertheless, antioxidant activity in chico fruit and jiotilla was higher than the amount in persimmon (*Spyros kaki)* (0.14 ± 0.03 mmol/100 g FS) and duck pear (*Pyrus bretschneideri*) 0.22 ± 0.03 mmol/100 g FS [55]. Dates (*Phoenix dactylifera*) reported amount between 406.61 ± 14.31 and 818.86 ± 21.91 µmol/100 g FS [56], compared with chico fruit and jiotilla. The antioxidant activities are higher in chico fruit (2201.56 ± 22.01 µmol/100 g DW) and jiotilla (2680.91 ± 26.80 µmol/100 g dry weight) than dates (*Phoenix dactylifera*). Comparing the antioxidant activity concentrations (ABTS method) with purple cactus pear (*Opuntia ficus-indica*) (0.61 ± 0.02 mmol/100 g FS) and orange pulp (0.37 ± 0.02 mmol/100 g FS) [6], the antioxidant activities in chico fruit and jiotilla resulted higher than the presented by *Opuntia ficus-indica* pulps. Antioxidant activity compared with fruits from apricot (*Prunus armeniaca*) for the variety Cöloglu, 0.45 ± 0.09 mmol/g

FS and the variety Zerdali, 0.37 ± 0.03 mmol/g FS [57], the chico fruit and jiotilla have higher activity than apricot (*Prunus armeniaca*). Concentration units of chico fruit and jiotilla (mmol/100 g FS) were changed to µmol/g fresh sample to compared with other fruits. Chico fruit (8.642 ± 0.204 µmol/g FS) and jiotilla (7.083 ± 1.247 µmol/g FS) antioxidant activity is higher than banana (*Musa acuminata*) 3.44 ± 0.29 µmol/g FS, apple red delicious 4.62 ± 0.03 µmol/g FS and pear (*Pyrussp.*) 4.30 ± 0.06 µmol/<sup>g</sup> FS [52]. Nevertheless, guava (*Psidium guajava*) 15.18 ± 0.81 µmol/g fs and sweetsop 23.60 ± 0.06 µmol/g fs [52] have a higher concentration than chico fruit and jiotilla. Antioxidant activity (DPPH method) reported in raspberry for Reveille *(Rubus idaeus L.)* (695.58 ± 11.56 mg/100 g FS) [58] is higher than chico fruit and jiotilla. Nevertheless, chico fruit (2.81 ± 0.26 mg/g DW) and jiotilla (4.86 ± 0.45 mg/g DW) antioxidant activities are lower than powder of wild plum tree (*Prunus domestica subsp. Insititia L*.) (26.47 ± 0.19 mg/g DW) [59]. Regarding raspberry extracts of (*Rubus Idaeus L.*), the antioxidant activity has been reported to be 29.0 ± 1.1 µmol/g FW [60] which represents a higher antioxidant activity than both fruits in the study, chico fruit (1.38 ± 0.13 µmol/g FW) and jiotilla (2.28 ± 0.21 µmol/g FS).

#### *3.3. Mineral Content Analysis*

Table 4 shows mineral content. Magnesium is essential for human metabolism; the enzymes use magnesium as a cofactor. The chico fruit and jiotilla have a magnesium content of 102.26 ± 4.24 mg/100 g and 33.12 ± 2.74, respectively. Regarding the potassium content, the chico fruit has 517.75 ± 16.78 mg/100 g. Comparing magnesium content with the almonds (*Prunus dulcis*) (270 mg/100 g), which represents a high magnesium content [61], chico fruit and jiotilla have lower contents in magnesium than almonds. Likewise, comparing with a cactus, *Opuntia spp*., it is reported for the pulp fruit content of 76 mg/100 g [62], the recommended intake per day is around 400–420 mg/per day [60]. Regarding the potassium content, the chico fruit has 517.75 ± 16.78 mg/100 g, which is a 159 mg/100 g superior to the value presented by the banana (*Musa acuminate*), a fruit with high potassium content, [61]. A comparison with (*Opuntia ficus indica)* fruits authors reported 559 mg of potassium/100 g [62]. It is important to highlight that the recommended intake per day is around 3400 mg potassium/per day [63]. Calcium is important to maintain strong bones, for the calcium content, the chico fruit and jiotilla have 69.99 ± 4.21 and 40.06 ± 1.43 mg/100 g respectively, these values are higher than reported data for banana (*Musa acuminate*) (26 mg/100 g), the pineapple (*Ananas comosus (L.) Merr*) (21 mg/100 g), papaya (*Carcinia papaya Linn*) (16 mg/100 g) and pitahaya (*Hylocereus undatus*) (31 mg/100 g) [64]. Zinc mineral also is important for children's growth, the chico fruit has 2.46 ± 0.65 mg/100 g, comparing this with reported values of different fruits, that the mango (*Mangifera indica*) has 0.14 mg/100 g, the papaya (*Carica papaya L.)* has 0.09 mg/100 g [64]. Zinc content is higher than fruits before mentioned. The zinc concentration in chico fruit is similar to borojo (*Borojoa sorbilis*) (2.47 mg/100 g) [64]. Nevertheless, the content of chico in zinc is lower than camajon fruit (*Sterculia apetala*) fruit 5.70 mg/100 g (our knowledge the fruit with the highest concentration of zinc) [64].


**Table 4.** Mineral content analysis profile of chico and jiotilla samples.

<sup>1</sup> Results are shown as average <sup>±</sup> standard deviation.

#### *3.4. Antioxidant Capacity of Clarified Juice Extracts*

Clarification is a procedure that removes the solids from a liquid, usually a beverage like wine. The concentration of betacyanins in clarified juice of chico fruit and jiotilla was 1.38 ± 0.01 and 2.93 ± 0.04 mg/g DW. Regarding the betaxanthin concentrations, in chico fruit and jiotilla clarified juice extracts were 1.16 ± 0.01 and 2.52 ± 0.35 mg/g DW. The total antioxidant capacity (Folin–Ciocalteu method) in chico fruit and jiotilla clarified juices were 129.47 ± 1.12 and 195.39 ± 7.82 mg/100 g FS. The antioxidant capacities of clarified juices are shown in Table 5 by methods FRAP, ABTS, and DPPH.


**Table 5.** Antioxidant activity clarified juices extracts.

<sup>1</sup> Results are shown as average <sup>±</sup> standard deviation. <sup>2</sup> Gallic acid. <sup>3</sup> Trolox were used in calibration curves. DW: dry weight; FS: fresh sample.

In clarification procedure some antioxidant capacity could be lost [65]. This phenomenon has been reported for green tea (*Camellia sinensis*) [66] and apple juice, where the clarification process reduces 2.5 times its antioxidant capacity [67]. However, in this work, the clarification increased antioxidant capacity. Similar results were observed in freeze-dried cherry laurels (*Prunus laurocerasus*) [68]. The increase in antioxidant capacity after centrifugation in liquids derived from vegetal material is probably due to the fact that antioxidants (polyphenols and betalains) are water-soluble and they are released from the pulp. The concentration of betacyanins in clarified juice of jiotilla (345.52 ± 5.04 µg/g FS) was higher than prickly pear juice (*Opuntia robusta*) betacyanins 300.5 ± 8.8 µg/g FS [7]. Contrary to the concentration found in clarified juice of chico fruit (151.6 ± 1.3 µg/g FS). Furthermore, this clarified juice is lower than the concentration reported for juice of *Opuntia rastrera* 152.6 ± 5.4 µg/g FS [7]. Regarding the betaxanthin concentration, in jiotilla clarified juice (296.8 ± 4.2 µg/g FS) is higher than prickly pear juice (*Opuntia robusta*) 189.9 ± 7.3 µg/g FS [7]. Also, chico fruit clarified juice has a higher betaxanthin concentration (127.5 ± 1.2 µg/g FS) than juice of *Opuntia rastrera* (86.2 ± 22.3 µg/g FS) [7]. Finally, the total antioxidant capacity (Folin-Ciocalteu method) in chico fruit (1294.76 ± 0.01 mg/100 g FS) and jiotilla (1953.39 ± 0.0701 mg/100 g FS) clarified juices are higher than prickly pear juice of Duraznillo rojo (*Opuntia leucotricha*) 226.3 ± 26.4 µg GA/g FS [7]. Comparing with other juices from commonly edible fruits, antioxidant activity in jiotilla juice clarified (3406.6±539.6µmol/100mL) is higher than pomegranate (*Punica granatum)* 3281.8 µmol/100 mL and aronia (*Aronia melanocarpa*) and 3277.9 µmol/100 mL juice with FRAP method [69].While in chico fruit juice clarified (665.9 ± 18.1 µmol/100 mL) antioxidant activity is higher than orange juice 203.4 µmol/100 mL[69]. Nevertheless, chico fruit (2983.9 ± 73.4 µmol/100 mL) and jiotilla (2462.6 ± 174.7 µmol/100 mL) juices have lower antioxidant activity than pomegranate 4537.3 µmol/100 mL and aronia 4261.8 µmol/100 mL juices in the ABTS method [69]. Chico fruit 319.7 ± 20.4 µmol/100 mL and Jiotilla 1122.2 ± 30.4 µmol/100 mLhave lower antioxidant activity than pomegranate (3138.3 µmol/100 mL) juice in DPPH [69]. Clarified juices extract of chico has a higher antioxidant activity than clarified juice extract of jiotilla, as evident by the ABTS method. This method reacts with any hydroxylated aromatics independently of their real antioxidant activity, including OH-groups, which do not contribute to antioxidant activity [69].

#### *3.5. Cytotoxicity of Clarified Juice Extracts*

The cytotoxicity of the juice extracts at 2% is shown in Figure 3. NIH/3T3 cell line was used as a control in order to screen the cytotoxicity of the clarified juices extract of jiotilla and chico on normal cells, as demonstrated in Figure 3, they were not cytotoxic in normal cell lines compared to cancer cell lines. The synergetic effect between betalains and phenolic compounds of cacti juice have been reported with benefits such as anticancer [7,8] and anticlastogenic effects [2]. Chico fruit and jiotilla showed a high concentration of betalains and phenolic compounds respect to different fruits. This contain could be responsible for the antioxidant capacity and cytotoxicity. Furthermore, phenolic acids were analyzed in order to identify the antiproliferative compounds. No cytotoxicity was observed in both juices in cell line NIH/3T3, used as a normal cell in order to evaluate the cytotoxicity effect of the juices relative to cancer cells, these values are congruent compared with other cacti fruits such as red pitahaya (*Hylocereus polyrhizus*) in normal human cell lines (HEK-293/human embryonic kidney and TPH-1/hummonocytes in the concentration of 0.39 to 0.78 mg/mL at 48 h) [4], and *Opuntia spp* Cardón (NIH/3T3 in concentration of 0.5% juice at 48 h) [7]. Chico fruit juice (49.7 ± 0.01%) inhibited the CaCo-2 growth and showed similar cell viability as gavia *Opuntia robusta* 52.50 ± 12.60% [7]. Cacti fruits have been tested in other colorectal cancer cell line HT29 and produced cytotoxicity an effective dose values (ED50 5.8 ± 1.0% *v*/*v* at 96 h) [8]. Cytotoxicity of chico fruit juice was observed (45.56 ± 0.05) in MCF-7, the significant effect was compared to *Opuntia rastrera* 75.40 ± 8.26% [7]. Other foods with a high concentration of betalains such as Beta vulgaris extract was cytotoxic in MCF-7 with IC50 value of 70 µg/mL, and cytotoxicity increase with the combination of Beta vulgaris extract and silver nanoparticles (IC50 47.6 µg/mL) [70]. The combination of chico fruit juice and silver nanoparticles could generate the same effect mentioned above. Jiotilla juice showed a higher value of cytotoxicity (47.31 ± 0.03%) in HepG2 than *Opuntia rastrera* 78.90 ± 9.00% [7]. Betanin from beetroot presented a 49% inhibition of HepG2 cell proliferation [71]. Other benefits that cacti juices have shown are hepatoprotective effect in-vitro and in-vivo [3,72]. Jiotilla juice (53.65 ± 0.04%) diminished the cell viability in PC-3 than moradillo *Opuntia violaceae* 61.20 ± 5.30% [7]. Extract of *Beta vulgaris L*. also has cytotoxicity in PC-3 (IC50 316.0 ± 2.1 µg/mL) [73].

**Figure 3.** Effect of clarified juices extracts of chico fruit (white) and jiotilla (black) at 2% on the cell viability. The error bars are standard deviations.

#### *3.6. Phenolic Analysis by HPLC*

Table 6 shows phenolic acid concentrations of chico fruit and jiotilla that compounds have cytotoxicity in cancer cell lines. *p*-Coumaric acid, a hydroxy derivative of cinnamic acid, is a compound that has a significant antiradical scavenging effect [74]. *p*-Coumaric acid is believed to reduce the risk of stomach cancer by reducing the formation of carcinogenic nitrosamines [75,76]. *p*-Coumaric acid has therapeutic benefits against cancer cell lines (Caco-2) [77]. The *p*-Coumaric acid-induced apoptosis in colon cancer cells (HTC-15) through the ROS-mitochondrial pathway [78]. Furthermore, *p*-Coumaric acid is shown to possess anti-inflammatory, anti-ulcer, anti-cancer, and anti-mutagenic properties [79,80]. Gallic acid, a polyhydroxy phenolic compound that can be found in green tea, grapes, strawberries, and bananas [81]. Gallic acid has demonstrated the potential anticancer activity in vivo and in vitro [82,83]. The anticancer activity of Gallic acid has been reported in various cancer

cells, including human ovarian cancer cells (HeLa), leukemia cell lines (C121) [84,85]. It was proven that the anticancer effect of Gallic acid is due to its ability to inhibit cell proliferation and to induce apoptosis [86,87]. Caffeic acid, the primary representative of hydroxycinnamic acids and phenolic acid in general, is widely distributed in plants [88]. Caffeic acid has an action against cervical (HeLa), mammary gland adenocarcinomas (MDA-MB-231), lymphoblastic leukemia (MOLT-3) [88], but it is not cytotoxic to healthy cells [89]. Further, caffeic acid treatment altered the mitochondrial membrane potential on HT-1080 human fibrosarcoma cell line. Other compounds that affect cancer cell lines are the betacyanins, and the main structure is betanin. Kapadia et al. found that betanin has cytotoxic properties in PC-3 cells [73] and Lee et al. observed that the same compound has cytotoxicity in HepG2 [71]. Clarified juice extract of jiotilla shows cytotoxicity in PC-3 and HepG2, this clarified juice has 2.12 times more betacyanins than clarified juice extracts of chico. Vitamin C promotes apoptosis in MCF-7 [90], although the concentration of vitamin C was not analyzed in the clarified juices, it was detected in the chico fruit (27.19 ± 1.95 mg/100 g FS). Probably, it is the main compound in clarified juice extract of chico that produces the cytotoxicity in MCF-7. In this work was identified the main compounds that have been reported in cacti fruits (betalains, p-coumaric acid, caffeic acid, ferulic acid, gallic acid) [37,38,91]. There is a synergy between betalains and phenolic compounds in antioxidant activity and cytotoxicity.


**Table 6.** Amount of the polyphenolic (mg/100 g fresh sample).

N.D.—Not detected.

#### **4. Conclusions**

The presence of a large variety of compounds of high-value in several fruits has been recognized in several studies. In this research, we investigated and demonstrated the nutritional characteristics of two desert fruits, i.e., chico fruit and jiotillain pulp and their clarified juice extracts. Chico fruit was seen to be an excellent resource of vitamin C, potassium, and zinc. Meanwhile, in jiotilla, betacyanins and betaxanthins were the main compounds that deserve to be highlighted, as their concentration was high as compared to earlier reported sources. All these results measured in the pulp of both fruits. On one hand, the determination of the cytotoxicity produced by clarified juice of both fruits in normal cell lines was absent. On the other hand, clarified juice extract of chico fruit showed cytotoxicity in CaCo and MCF-7. Regarding the clarified juice extract of jiotilla, the cytotoxicity activity was showing HepG2 and PC-3 cell lines. For both fruits, in the clarified juice and pulp, the nutritional profiles resulted are high and, in some instances, similar compared with some other and more common edible fruits. Overall, this work increased the knowledge of other different sources of nutrients that can be used to feed the population.

**Author Contributions:** Conceptualization, L.S.-F., J.R.-R., M.R.-A. and R.P.-S.; data curation, L.S.-F. and J.R.-R.; formal analysis, L.S.-F., E.R.-E., P.R.O. and M.F.M.C.; funding acquisition, R.P.-S.; investigation, L.S.-F., M.R.-A. and E.M.M.-M.; methodology, E.M.M.-M. and C.C.-Z.; project administration, R.P.-S.; resources, C.C.-Z.; supervision, R.P.-S.; validation, L.S.-F. and E.R.-E.; writing—original draft, L.S.-F., E.R.-E., J.R.-R., M.R.-A., E.M.M.-M., C.C.-Z., P.R.O. and M.F.M.C.; writing—review and editing, W.-N.C., H.M.N.I. and R.P.-S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research has been supported by Consejo Nacional de Ciencia y Tecnología (CONACYT) Doctoral Fellowship No. 492030 awarded to author L.S.-F. We want to express our profound gratitude to all the support from Opciones de vida para comunidades vulnerables, Water Center, Bioprocess Group, and Synthetic Biology Strategic Focus Group 0821C01004, and Research Chair Funds GIEE 0020209I13 at Tecnologico de Monterrey.

**Conflicts of Interest:** The authors declare that they have no competing interests.

### **Appendix A**

**Figure A1.** Dilution of clarified juice extract.

#### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*
