*Article* **Cornelian Cherry (***Cornus mas***) Powder as a Functional Ingredient for the Formulation of Bread Loaves: Physical Properties, Nutritional Value, Phytochemical Composition, and Sensory Attributes**

**Veronika Šimora 1,2, Hana Dˇ úranová 1, Ján Brindza 3, Marvin Moncada 4, Eva Ivanišová 5, Patrícia Joanidis 1, Dušan Straka 1, Lucia Gabríny <sup>1</sup> and Miroslava Kaˇcániová 2,6,\***


**Abstract:** In the current study, Cornelian cherry powder (CCP, *Cornus mas*) was investigated as a functional ingredient for bread production. Experimental bread loaves were prepared using five levels of CCP (0, 1, 2, 5, and 10% *w*/*w*) to replace wheat flour in bread formulation. The final products were analyzed regarding their proximate composition, content of selected biologically active substances, antioxidant activity (AA), volume, and sensory attributes. Increasing the incorporation of CCP led to significantly (*p* < 0.05) higher concentrations of carbohydrate, ash, energetic value, total polyphenols, phenolic acids and AA, and reduced fat and protein contents (*p* < 0.05). Moreover, up to 5% addition of CCP positively affected the volume (642.63 ± 7.24 mL) and specific volume (2.83 <sup>±</sup> 0.02 cm3/g) of bread loaves, which were significantly (*<sup>p</sup>* < 0.05) higher compared to the control (no addition of CCP; 576.99 <sup>±</sup> 2.97 mL; 2.55 <sup>±</sup> 0.002 cm3/g). The sensory attributes chewiness, crumb springiness, bitterness, and sourness had lower scores (*p* < 0.05) in bread formulated with 10% CCP compared to the control. Overall, results show that the bread loaves produced with up to 5% CCP addition were considered the preferred formulation among the experimental samples tested, taking into consideration their composition, bioactive content, sensory, and physical properties.

**Keywords:** bakery; biologically active substances; DPPH assay; nutritional value; sensory properties

#### **1. Introduction**

At a global level, bakery goods constitute an essential part of human nutrition [1]. The development of innovative products using value-added ingredients has become an important trend in the bread manufacturing industry in an effort to meet the demand of a new generation of consumers seeking healthier lifestyles [2]. Fruits are a remarkable source of natural bioactive compounds with great potential of incorporation in the formulation of multiple products [3–5]. However, fresh fruits are highly perishable due to their extremely high moisture content (MC; 75–95%) [6], which leads to estimated post-harvest losses of around 30% [7]. Therefore, efficient preservation techniques are necessary to extend their shelf-life and marketability [8]. In this regard, drying technology has proven to be a feasible, convenient, and well-accepted processing strategy for improving food storage

**Citation:** Šimora, V.; Dúranová, H.; ˇ Brindza, J.; Moncada, M.; Ivanišová, E.; Joanidis, P.; Straka, D.; Gabríny, L.; Kaˇcániová, M. Cornelian Cherry (*Cornus mas*) Powder as a Functional Ingredient for the Formulation of Bread Loaves: Physical Properties, Nutritional Value, Phytochemical Composition, and Sensory Attributes. *Foods* **2023**, *12*, 593. https://doi.org/ 10.3390/foods12030593

Academic Editors: Donatella Bianca Maria Ficco and Grazia Maria Borrelli

Received: 17 December 2022 Revised: 25 January 2023 Accepted: 28 January 2023 Published: 31 January 2023

**Copyright:** © 2023 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 (https:// creativecommons.org/licenses/by/ 4.0/).

stability [9,10]. Several types of dehydration technologies are available, e.g., solar drying, hot air drying, vacuum drying, spray drying, osmotic drying, microwave drying, and freeze drying [11]. Among them, hot air drying is currently the most widely used method in the agri-food sphere. Indeed, a more homogeneous, sanitary, and coloured dehydrated product can be quickly obtained using this process [12]. Ultimately, powdered dried fruits provide the producers with enhanced nutritional and health properties for the many types of cereal products [13]. In effect, the addition of powdered fruits to the recipe of the bakeware offers an important protective capacity against a number of civilization disorders due to their antioxidant, anti-inflammatory, anti-mutagenic, and anti-carcinogenic characteristics [14]. For example, enriched foods showed enhanced antioxidant power, which can help prevent some chronic diseases [15].

In this regard, many fruit-derived materials have been examined, including mango peel [4], banana [3], pomegranate peel [16], orange peel [17], and apple pomace [18]. From a technologically advanced point of view, fruit-derived ingredients are naturally glutenfree, and this might modify important attributes of the flour blend and consequently the properties of the final bread [19]. For instance, decreased volume and cohesiveness of bread loaves formulated with mango peel powder (more than 5% *w/w*) addition has been noted by Chen et al. [4]. Moreover, decreased volume, specific volume, moisture, elasticity, and cohesiveness of bread loaves with increasing concentrations of pomegranate peel powder (0%, 1%, 3%, 5%, and 7%) have been found by Zhang et al. [16]. On the other hand, sensory ratings shown by Baba et al. [3] revealed no significant differences in taste, aroma, and appearance of wheat bread containing banana powder (up to 30%) as compared to the control sample, but the bread with the 30% addition had lower overall acceptability. Hence, the careful selection of ingredients and the determination of their adequate concentration [20] are of great importance for a successful food practice.

Cornelian cherries (CC; *Cornus mas* L.) are currently gaining increasing attention from the scientific research community [21]. These oval or pear-shaped edible fruits with color ranging from red to purple constitute an important source of vitamin C and polyphenols, mainly flavonoids, anthocyanins, and iridoids [21–23]. Such substances are linked to a wide range of biological effects and pharmacological properties, including antimicrobial, anti-inflammatory, anti-cancer, anti-diabetic, and anti-atherosclerotic activities [24–26]. Most often, the cherries are consumed fresh or as a dried delicacy. Due to their beneficial properties, they can serve as purposeful ingredients for the commercial food sector [27]. In recent years, CC has also been applied as a flavoring ingredient in ice creams, desserts, and cakes [28]. Furthermore, CC and their juices or extracts have inspired the production of novel foods, such as beer [29], soup [30], vinegar [31], or burgers [32]. Regarding baked goods, ˙ Ilyaso ˘glu et al. [33] replaced wheat flour (WF) with CC (10 g 100 g−<sup>1</sup> composite flour) in cookie formulations and reported enhanced contents of omega-3 fatty acids, total phenolic content, and antioxidant capacity in the final product.

To the best of our knowledge, the addition of Cornelian cherry powder (CCP) to bread formulation has not yet been investigated. Therefore, this study is the first report showing the production and quality assessment of bread loaves produced with CCP. For this, the incorporation of different CCP concentrations (1, 2, 5, and 10% *w*/*w*) for the partial substitution of WF in the formulation of bread loaves was performed, and the quality characteristics of the final products were determined. To achieve our goal, we assessed the compositional profile, antioxidant activity (AA), phenolic compounds, key physical attributes, and sensory properties of bread loaves. As a result, recommendations about the most appropriate level of CCP for bread production are discussed. This report unveils the potentiality of Cornelian cherry powder as a functional ingredient for bakery products.

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

#### *2.1. Materials*

Wheat flour (WF; T-650 type) was purchased from a grinding mill (Pohronský Ruskov JSC, Pohronský Ruskov, Slovakia). Other raw materials for bread making, such as salt (Solivary Trade Ltd., Trenˇcín, Slovakia), saccharose (Slovenské Cukrovary Ltd., Sered', Slovakia), and compressed yeast (Thymos Ltd., Vel'ká Lomnica, Slovakia), were obtained from a local grocery store. The chemical reagents were of analytical quality, and all of them were purchased from Reachem (Bratislava, Slovakia) or Sigma Aldrich (Saint Louis, MO, USA).

#### *2.2. Preparation of Cornelian Cherry Powder (CCP)*

Fresh, fully ripe Cornelian cherry (CC; *Cornus mas*) fruits were collected from the SUA Botanical Garden (Slovak University of Agriculture, SUA, Nitra, Slovakia) and subsequently selected, cleaned, and pitted. For the preparation of Cornelian cherry powder (CCP), the fruits were dried at 45 ◦C until complete dehydration using a cabinet dryer (Universal oven UF 160, Memmert GmbH + Co.KG, Büchenbach, Germany). After this, the dried CC was homogenized (ETA Gratus 0028 90030, ETA-Slovakia Ltd., Bratislava, Slovakia) and sieved to obtain powder particles with a diamter of 0.5 mm. The produced CCP was packed in polyethylene (PE) bags and stored at room temperature in a dark place until analysis and/or use.

#### *2.3. Composite Flour Preparation*

Five composite flours were prepared by partially replacing WF with CCP according to the following ratios: WF:CCP, 100:0 (control sample), 99:1 (1% addition of CCP), 98:2 (2% addition of CCP), 95:5 (5% addition of CCP), and 90:10 (10% addition of CCP), respectively. The WF/CCP blends were individually packaged in PE bags and kept at room temperature until the bread making process.

#### *2.4. Bread Loaf Preparation*

Bread loaf experimental treatments (Figure 1) were prepared following the methodology described by Valková et al. [18]. The bread formulation consisted of 500 g of WF or WF/CCP blends, saccharose (1% of flour), salt (2% of flour), water (60% of flour), and yeast (2% of flour). Initially, compressed yeasts were reactivated in a saccharose solution at 32 ◦C for 5 min. All ingredients were blended for 10 min in two steps (first step: 4 min at 1500 rpm; second step: 6 min at 3000 rpm) in a mixer (DIOSNA SP 12; DIOSNA Dierks and Söhne GmbH, Osnabrück, Germany) using a dough hook accessory to ensure proper hydration of flour. Then, the dough was carefully portioned into 250 ± 5 g pieces and placed into oiled and floured tins. The tins were transferred to a fermentation cabinet (MIWE cube, Pekass Ltd., Pilsen, Czech Republic) set at 32 ◦C and 85% relative humidity, and allowed to proof for 40 min. The bread loaves were baked in two phases (Phase I: 180 ◦C with the addition of 160 mL steam at the same temperature; Phase II: 210 ◦C for 7 min, no steam) in a steamer oven (Laboratory oven MIWE cube, Pekass Ltd., Pilsen, Czech Republic). The baked loaves were removed from the tins and left to cool at room temperature for 2 h until cutting. In total, three batches of each type of bread were produced (15 bread loaves in total).

**Figure 1.** Experimental bread loaves. (**A**) control sample, (**B**) 1% Cornelian cherry powder addition, (**C**) 2% Cornelian cherry powder addition, (**D**) 5% Cornelian cherry powder addition, and (E) 10% Cornelian cherry powder addition.

#### *2.5. Determination of Compositional Profile and Energetic Value*

The compositional value of CCP and experimental bread loaves was assessed. For this purpose, moisture, ash, crude protein, fat, and total carbohydrate contents, as well as the energetic values, were determined.

The moisture content (MC) was measured with an automatic moisture analyzer, DBS 60-3 (Kern and Sohn GmbH, Altstadt, Germany), according to the manufacturer's instructions and the ASTM D 6980 method. Briefly, 1 g of sample was placed on the sample plate and tested at 120 ◦C for the required time (10–15 min). The total ash and crude protein contents were determined in accordance with AACC standard 08-01 using a muffle furnace (Neberterm, Germany) and the semi-micro Kjeldahl method (factor of converted nitrogen to protein was 6.25). The total fat content was analyzed using the Ankom XT15 Fat Extractor (Ankom Technology, Fairport, NY, USA) in accordance with the manufacturer's instructions. The total carbohydrate content (TCC) and energy were calculated by Equations (1) and (2), according to Valková et al. [18] and Arraibi et al. [34]:

TCC (%) = 100 − moisture (%) − protein (%) − lipids (%) − ash (%) (1)

Energy (kcal/100 g) = 4 × (% proteins + % carbohydrates) + 9 × (% fat) (2)

#### *2.6. Determination of Radical Scavenging Activity and Polyphenolic Compounds*

Firstly, samples of ethanolic extracts (CCP, and bread treatments) were prepared. For each extraction, 0.2 g (CCP) and 0.5 g (bread treatments) of sample received 20 mL or 40 mL of 80% ethanol, respectively, and were extracted for 2 h, followed by centrifugation at 4000× *g* for 10 min in a Rotofix 32A (Hettich, Spenge, Germany). The supernatant was used for the determination of antioxidant activity (AA), total polyphenols content (TPC), total phenolic acids content (TPAC), and flavonoid content (FC).

The AA, TPC, TPAC, and FC of the samples were analyzed using the 2,2-diphenyl-1 picrylhydrazyl (DPPH) assay [18], the colorimetric assay utilizing the Folin-Ciocalteu (F-C) reagent [18], and according to the procedures of Valková et al. [18] and Ivanišová et al. [35], respectively. The AA was expressed as Trolox equivalent antioxidant capacity (TEAC) in milligrams per gram of dry weight (dw). Gallic acid (for TPC analysis), caffeic acid (for TPAC analysis), and quercetin (for FC analysis) standards were used, and the results were expressed as gallic acid equivalents (GAE), caffeic acid equivalents (CAE), and quercetin equivalents (QE) in milligrams per gram of dw, respectively.

#### *2.7. Volume Analysis*

The volume (mL) and specific volume (cm3/g) of bread samples were assessed using an automatically laser-based scanning device, the VolScan Profiler VSP 300 (Stable Micro Systems, Godalming, UK), according to the manufacturer's recommendation (AACC approved method 10.16.01).

#### *2.8. Sensory Assessment*

Sensory analysis was performed by 10 panelists (three men and seven women, aged 26–47), trained according to the standard STN EN ISO 8586. The evaluation of bread samples took place at the Sensory Laboratory of the Research Centre AgroBioTech (Slovak University of Agriculture, SUA, Nitra, Slovakia) during the late morning. The experimental samples (control and four treatments, 1, 2, 5, and 10% (*w*/*w*)) were coded with 3-digit numbers and presented to the panelists at the same time. Between evaluations of individual samples, the panelists were instructed to drink water.

Eleven sensory descriptors were evaluated (on a 15-point unstructured scale) and divided into the following categories: color (crust color, crumb color), texture (pore uniformity, crumb springiness, chewiness), aroma, taste (sweet taste, bitter taste, sour taste, aftertaste), and overall impression. The descriptors for sensory rating were selected according to García-Gómez et al. [36], and based on preliminary training sessions with selected panelists.

#### *2.9. Statistical Analysis*

All the analyses were conducted at least in triplicate, and the data was reported as mean value ± standard deviation. One-way analysis of variance (ANOVA) and the Tukey test (Prism 8.0.1 program, GraphPad Software, San Diego, CA, USA) were applied to establish statistically significant differences between the samples at the level of *p* < 0.05.

#### **3. Results**

#### *3.1. Characterization of Cornelian Cherry Powder*

The compositional profile, energetic value, polyphenolic compounds, and AA of CCP are shown in Table 1. CCP had a low energetic value, and low fat and protein contents, but a high total carbohydrate and ash contents. Regarding DPPH free radical scavenging activity and polyphenolic compounds, the CCP exhibited high AA, and high total polyphenols, total phenolic acids, and flavonoid contents.

**Table 1.** Composition, energetic value, antioxidant activity, and polyphenolic compounds of Cornelian cherry powder.


Values are expressed as the mean ± standard deviation (*n* = 3). CCP—Cornelian cherry powder; AA—antioxidant activity expressed as mg of Trolox equivalents per gram dry weight; TPC—total polyphenols content expressed as mg of gallic acid equivalents per gram dry weight; TPAC—total phenolic acids content expressed as mg of caffeic acid equivalents per gram dry weight; FC—flavonoid content expressed as mg of quercetin equivalents per gram dry weight.

#### *3.2. Compositional Profile and Energetic Value of Experimental Bread Loaves*

Increasing the percentage of CCP in the bread formulations led to a progressive and significant (*p* < 0.05) increase in total carbohydrate and ash contents, and energetic value, as shown in Table 2. Interestingly, no differences in total carbohydrates and energetic value were noted between bread loaves with 5% and 10% of CCP. In addition, a significantly (*p* < 0.05) linear reduction in fat and protein contents was observed when more CCP was added to the formulation of bread loaves (Table 2), which agrees with the low fat and protein contents found for CCP (Table 1). No significant differences were observed for the MC of bread treatments (*p* > 0.05).

**Table 2.** Compositional profiles of experimental bread loaves.



**Table 2.** *Cont*.

<sup>1</sup> Incorporation ratio of partial substitution of wheat flour; please see item 2.3 for further details. Values are expressed as the mean ± standard deviation (*n* = 3). Data in the same line with different superscript letters are significantly different (Tukey's test, *p* < 0.05). CCP—Cornelian cherry powder.

#### *3.3. Antioxidant Activity and Polyphenolic Compounds of Experimental Bread Loaves*

The AA and content of selected polyphenolic compounds in experimental bread loaves enriched with CCP are summarized in Table 3. For all analyzed parameters, a significant and linear increase in bioactivity was demonstrated as higher ratios of CCP were used for the production of bread loaves (*p* < 0.05), with the exception of the 1% CCP treatment, which had similar results compared to the control sample (*p* > 0.05). Moreover, flavonoid compounds were not detected in any of the bread treatments (Table 3).

**Table 3.** Antioxidant activity and selected polyphenolic compounds of experimental bread loaves.


<sup>1</sup> Incorporation ratio of partial substitution of wheat flour; please see item 2.3 for further details. Values are expressed as the mean ± standard deviation (*n* = 3). Data in the same line with different superscript letters are significantly different (Tukey's test, *p* < 0.05). CCP—Cornelian cherry powder; AA—antioxidant activity expressed as mg of Trolox equivalents per gram dry weight; TPC—total polyphenols content expressed as mg of gallic acid equivalents per gram dry weight; TPAC—total phenolic acids content expressed as mg of caffeic acid equivalents per gram dry weight; FC—flavonoid content expressed as mg of quercetin equivalents per gram dry weight. ND—not detected.

#### *3.4. Volume of Experimental Bread Loaves*

Breads produced with WF/CCP blends of 1%, 2%, and 5% CCP showed significantly different (*p* < 0.05) volume and specific volume (Table 4). The addition of CCP led to a significant increase in both parameters, volume and specific volume, which is a desirable attribute for bread loaves. However, bread loaves prepared with 10% CCP had significantly (*p* < 0.05) lower volume and specific volume compared to all treatments and the control sample. The highest results were observed for 2% CCP, which resulted in an increase of 15.8% and 16.4% for volume and specific volume, respectively.

**Table 4.** Volume of the experimental bread loaves.


<sup>1</sup> Incorporation ratio of partial substitution of wheat flour; please see item 2.3 for further details. Values are expressed as the mean ± standard deviation (*n* = 3). Data in the same column with different superscript letters are significantly different (Tukey's test, *p* < 0.05). CCP—Cornelian cherry powder.

#### *3.5. Sensory Properties of Experimental Bread Loaves*

The crust and crumb colors of bread loaves formulated with ≥2% and ≥5% CCP, respectively, were reported as significantly (*p* < 0.05) darker compared to the control (Table 5). Significant differences in the bread aroma were identified between the control sample and the bread enriched with 2% and 10% CCP, respectively. Further, the evaluators perceived an aftertaste (*p* < 0.05) in samples produced with ≥1% CCP, while the highest sour and bitter taste scores (*p* < 0.05) were reported for bread produced with the highest (10%) CCP ratio. Additionally, this sample was reported as the least chewable and having poor crumb springiness. Regarding the pore uniformity and sweet taste parameters, there were no significant differences between experimental samples. When evaluating the overall impression, the sample with the 2% addition was perceived as the tastiest and had the highest score, which is significantly superior to the control (*p* < 0.05).

**Table 5.** Sensory analysis of the experimental bread loaves.


<sup>1</sup> Incorporation ratio of partial substitution of wheat flour; please see item 2.3 for further details. Values are expressed as the mean ± standard deviation (*n* = 3). Data in the same line with different superscript letters are significantly different (Tukey's test, *p* < 0.05). CCP—Cornelian cherry powder.

#### **4. Discussion**

In general, the functional characteristics of a raw material affect its interaction with other food components and strongly determine its final application [37]. Wheat flour is used as a major staple raw material in bread production [38], and it is the most abundant source of calories and protein in the human diet [39]. Although it is also a great source of nutrients, its content of bioactive compounds and AA is poor as a consequence of the refining during processing [40]. To improve the nutritional profile and the biological activity of bakery products, the partial replacement of WF with phytochemical-rich, functional plant-based flours or powders is an interesting strategy. In this sense, horticultural crops are known to provide a rich source of diverse nutritional molecules, many of them possessing antioxidant activity, which has been reported as capable of protecting the human body against oxidative cellular reactions [41]. Cornelian cherry-derived ingredients have been tested as part of several formulations to produce enhanced food products [28,32,33,42–52], but not in bread formulations yet. Therefore, this is the first report of such research activities with solid potential for practical applications to bread manufacturing.

Regarding the characterization of CCP, Tontul et al. [53] recorded similar values for MC (8.03 ± 0.14%) in CCP dried at 50 ◦C. Since moisture in sugar-rich powders acts as a plasticizer [54], the lower MC of CCP used in our study may have a positive effect on its cohesive properties. Importantly, foods with reduced MC are considered safe due to the growth mitigation of undesirable microorganisms (especially molds), thus improving the shelf-life of the product [55]. We found CCP to have a higher ash content compared to dried CC fruits (2.83 ± 0.35%), as demonstrated by Petkova and Ognyanov [56]. Further, total carbohydrates represent the most essential source of energetic value in CC fruits [57], which is also in line with our findings. Likewise, relatively low crude protein levels (ranging from 1.43 to 2.71%) have also been demonstrated in CC fruits by Serbia by Bijeli´c et al. [58]. In fact, raw CC fruits typically have low fat content (1.49 ± 0.02%) [59], which was also confirmed by our results. Differences between our results and those of cited research studies can be explained by different genotypes used, as well as the influence of environmental growth conditions [60].

It is important to characterize not only the overall AA but also the individual antioxidant components responsible for such activity, which are present in diverse fruits [61–64]. Our findings demonstrate that oven-dried CCP produced in this study has high contents of total polyphenols, phenolic acids, and flavonoids along with a strong AA and reiterate results found in fresh CC fruits determined by Dupak et al. [65] and Szczepaniak et al. [66]. In contrast, Popovi´c et al. [67] have identified a lower content of total polyphenols in dried samples of 10 CC genotypes. In addition, AA and the contents of total phenolic acids and flavonoids were lower in CC pulp analyzed by Dupak et al. [65] in comparison with our CCP. In fact, these discrepancies are expected since different processing protocols and parameters, fruit genotypes and varieties, and maturity stages affect the aforementioned results [25,68]. Considering our results, it can be hypothetically assumed that eating both our CCP and/or products enriched with the CCP could be beneficial for human health in the sense of their ability to eliminate harmful oxidative stress in the organism, thus reducing the risk of chronic disease incidence.

The effect of CCP addition on the nutritional composition and key quality attributes of bread loaves enriched with four concentrations (1%, 2%, 5%, and 10%) of CCP was further evaluated. Carbohydrates are the prime macronutrients in bread. The content of carbohydrates in the bread formulations increased linearly with the incorporation of CCP, because CCP is also a major source of this macronutrient (>85%, Table 1). In addition, increasing additions of CCP led to progressively lower protein and fat contents and higher ash contents in the enriched bread loaves, also as a reflex of the original CCP composition (Table 1). The same trend was noted by Topda¸s et al. [28], who analyzed the impact of different CC fruit paste (5%, 10%, and 15%) additions to the composition of ice cream. Similarly, a lower fat content in CCP-enriched biscuits compared to control samples was reported by ˙ Ilyaso ˘glu et al. [33]. Given these findings, CCP may be a promising ingredient in the preparation of low-fat goods for the food sector.

Moreover, our results pointed to markedly higher concentrations of phenolic compounds (TPC and TPAC) and AA in the CCP-containing bread loaves in comparison to the control. This finding agrees with the research conducted by ˙ Ilyaso ˘glu et al. [33], Topda¸s et al. [28], and Haghani et al. [51], which revealed increased levels of TPC and higher AA, as well as higher concentrations of CC in products used to prepare biscuits, conventional ice cream, and probiotic ice cream, respectively. Enhanced TPC and antioxidant capacity of white chocolate and dairy desserts after addition of CCP and CC juice, respectively, were also observed in the studies performed by Cerit et al. [45] and Ivanova et al. [47]. On the other hand, the total absence of FC identified in all our bread samples may be related to the thermolability of these biologically active substances [69] and possible complete degradation during the baking process. Indeed, flavonoids are major phenolic compounds with natural antioxidant capacity (mediated via their functional hydroxyl groups in their structure) [70], reported in CC fruits [63] and also here in our developed CCP (Table 1). The microencapsulation technique using appropriate wall materials has been reported in the literature [71] as an efficient stabilization approach for the preservation of polyphenolic extracts from Cornelian cherries [47]. Therefore, the spray-drying and microencapsulation of plant extracts to produce powdered ingredients with preserved biologically active compounds destined for bakery products will be considered in future studies by our research team.

The addition of non-traditional ingredients to the bakery goods not only affects their nutritional composition and bioactivity properties but may also influence important physical attributes. One of the most crucial physical properties of bread is its volume, which

strongly determines the consumers' preferences and predicts its quality, as well [72]. In our study, bread samples enriched with up to 5% CCP displayed significantly higher values for volume and specific volume compared to control wheat bread. We hypothesize that this increase may be related to the presence of pectin and other hydrocolloids (estimated to be about 5.7% in dried CC fruits) in the CCP composition [73]. Confirming our hypothesis, Rosell et al. [74] reported that mixing WF with hydrocolloids increases dough stability and loaf volume due to their ability to absorb water and gelling properties, as well. In view of this, during heating a gel network may be formed, which can consequently strengthen the expanding dough cells, thereby improving gas retention and bread volume [75]. In addition, Das et al. [76] found that hydrocolloids prevent the small cells found in bread dough from clumping together to generate larger cells. Having a larger number of small cells can form a more uniform matrix that acts as a CO2-trapping network. Hydrocolloids improving the volume of bread loaves were also recognized in the research conducted by Kang et al. [77] and Zhao et al. [78]. Conversely, the reduction in volume and specific volume observed in our bread samples supplemented with 10% CCP may be attributed to an undesirably higher content of hydrophilic compounds (including carbohydrates) in these bread loaves, which could theoretically cause excessively higher viscosity. Furthermore, the reduced volume and increased stiffness of bread loaves enriched with higher CCP addition may also be due to a relevant reduction in the amount of gluten in the doughs [79], caused by the significant depletion of WF content in that formulation. Overall, our findings suggest that the addition of CCP up to 5% is a promising strategy to produce bread loaves and other related products with enhanced bioactivity and preserved bread quality parameters.

The sensory properties of a food product play a major role in its consumer acceptance and marketability [80]. When dealing with the incorporation of alternative ingredients into food formulations, the goal is to enhance the nutritional, bioactive, and physical attributes of the product without compromising its sensory acceptability [81]. In effect, higher levels of non-traditional bakery ingredients in bread formulations can greatly affect their taste and aroma [15], which are considered important sensory characteristics along with the texture [82]. Fresh, mature CC fruits have an intrinsic cherry-like, tart-sweet, and sour flavor with characteristic aroma [27,63] which might interfere with the sensory perception of a final product. Indeed, a gradual increase of aroma and taste (sweet, bitter, sour taste, aftertaste) scores of bread samples with the increment of CCP incorporation ratio was observed. Lowest scores for bitter, sour, and aftertaste were observed for bread formulated with 1% and 2% CCP addition. We hypothesize that bioactive compounds such as phenolics and iridoids found in CCP, which display many biological activities, such as anti-inflammatory, antioxidative, anti-cancer, anti-atherogenic, antidiabetic, and neuroprotective attributes [83], may play an important role in the observed sensory findings [84,85]. Further, as the CCP ratio increased, the crumb and crust color of CCP-enriched samples became darker (Figure 1). We suppose that this color modification could affect consumers' acceptability in a positive manner. In effect, previously it was shown that the red color of bread (caused by red beetroot addition) was preferred by consumers as compared to that supplemented only with white beetroot [86]. In addition to these observations, the texture of our bread loaves was also modified by CCP addition. The increased chewiness of bread was positively correlated with the amount of CCP (10% addition), which is consistent with a previous study [87] documenting the superior chewiness of bread samples enriched with artichoke fiber. Similarly, the significant difference in springiness identified only between the bread with 10% CCP addition and the control reflects a dose-dependent effect of CCP addition on this specific parameter. Finally, the best overall impression score was reported for samples with 2% CCP addition, and it was significantly higher compared to the control, whereas the other treatments showed similar results (*p* > 0.05). Altogether, bread loaves prepared with 2% or 5% replacement ratios of WF with CCP proceed to be promising bakery products with desirable overall characteristics. Our results present a novel, functional approach for the development of wheat bread with enhanced attributes for the current health-oriented bakery market.

#### **5. Conclusions**

Our study investigated the production and application of Cornelian cherry powder as a potential functional food ingredient for the partial replacement of WF in bread formulations. Our results show that the incorporation of CCP (replacement ratios between 1–10% *w*/*w*) in wheat bread formulations produces final products with significantly different composition, bioactivity, volume, and sensory attributes compared to the control. Indeed, increasing ratios of CCP lead to bread loaves with significantly higher (*p* < 0.05) carbohydrate and ash contents, energetic value, TPC, TPAC, and AA, but lower fat and protein contents. Further, our findings showed that CCP added at 1% to 5% ratios significantly (*p* < 0.05) improved the volume and specific volume of experimental bread loaves, and the highest overall impression score, significantly higher compared to the control, was reported for samples with 2% CCP addition. Overall, here we demonstrate that CCP can partially replace WF when used up to 5% (*w*/*w*) for bread formulations without negatively impacting key physical properties and sensory attributes, while enhancing the concentration of phenolic antioxidants. Moreover, we believe that the incorporation strategy shown here can be successfully applied to multiple bakery products for the production of healthier and more functional food products for the emerging health-oriented market.

**Author Contributions:** Conceptualization, V.Š. and J.B.; methodology, V.Š., E.I., P.J., and D.S.; software, V.Š., H.D., and M.M.; validation, V.Š., L.G., and M.K.; formal analysis, V.Š., H. ˇ D., J.B., E.I., ˇ and M.M.; investigation, V.Š., P.J., and D.S.; resources, L.G. and M.K.; data curation, L.G. and M.K.; writing—original draft preparation, V.Š. and H.D.; writing—review and editing, V.Š. and H. ˇ D.; visu- ˇ alization, V.Š., H.D., L.G., and M.K.; supervision, L.G. and M.K.; project administration, L.G.; funding ˇ acquisition, M.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This publication was supported by the Operational Program Integrated Infrastructure within the project: Demand-driven research for the sustainableand innovative food, Drive4SIFood 313011V336, co-financed by the European Regional Development Fund.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** This work has been supported by grants from the KEGA, grant number 010SPU-4/2021. The authors are thankful to Roberta Hoskins for language corrections.

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

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

**Sonia Bonacci 1, Vita Di Stefano 2,\*, Fabiola Sciacca 3, Carla Buzzanca 2, Nino Virzì 3, Sergio Argento <sup>4</sup> and Maria Grazia Melilli <sup>4</sup>**


**Abstract:** The rheological and chemical quality of pasta samples, which were obtained using the durum wheat semolina fortified with the hemp seed solid residue, after oil extraction, sieved at 530 μm (Hemp 1) or 236 μm (Hemp 2) at different percentages of substitution (5%, 7.5%, and 10%, were evaluated. The total polyphenolic content in hemp flour was quantified in the range of 6.38–6.35 mg GAE/g, and free radical scavenging was included in the range from 3.94–3.75 mmol TEAC/100 g in Hemp 1 and Hemp 2, respectively. The phenolic profiles determined by UHPLC-ESI/QTOF-MS showed that cannabisin C, hydroxycinnamic and protocatechuic acids were the most abundant phenolic compounds in both hemp flours. Among the amino acids, isoleucine, glutamine, tyrosine, proline, and lysine were the most abundant in raw materials and pasta samples. Although the hemp seeds were previously subjected to oil extraction, hemp flours retain about 8% of oil, and the fatty acids present in the largest amount were linoleic acid and α-linolenic acid. Characterization of the minerals showed that the concentration of macro and trace elements increased according to fortification percentage. Sensory evaluation and cooking quality indicated that the best performance in terms of process production and consumer acceptance was obtained using Hemp 2 at 7.5%. Hemp supplementation could be a potential option for producing high-quality, nutritionally rich, low-cost pasta with good color and functionality.

**Keywords:** pasta fortification; hemp flour; durum wheat cultivar; amino acids; fatty acids; mineral fortification

#### **1. Introduction**

In recent years, consumers' eating habits have changed significantly. Food that, in the past, was intended to satisfy hunger and provide the necessary nutrients, today, combined with an active lifestyle, can help with harmonious physical and mental well-being.

Wheat is mainly used for the production of pasta, bread, and sweet and savory baked products. Pasta is one of the basic foods in the Mediterranean diet. Pasta traditionally made with durum wheat semolina can be prepared with "non-wheat flour" or by incorporating by-products from the agro-food industry, in variable percentages, which can increase its nutritional value [1,2]. Food by-products (grape, cereal bran, sunflower, artichoke, etc.) could represent interesting sources of bioactive compounds [3–5]. Fortification is the process by which nutrients with beneficial health effects are added to a food product in order to improve its nutritional quality and to increase its intake levels in the population. Food fortification (or enrichment) often negatively affects the quality of products, in terms of texture, color, cooking quality and sensory properties. Therefore, one of the main

**Citation:** Bonacci, S.; Di Stefano, V.; Sciacca, F.; Buzzanca, C.; Virzì, N.; Argento, S.; Melilli, M.G. Hemp Flour Particle Size Affects the Quality and Nutritional Profile of the Enriched Functional Pasta. *Foods* **2023**, *12*, 774. https://doi.org/10.3390/ foods12040774

Academic Editors: Donatella Bianca Maria Ficco and Grazia Maria Borrelli

Received: 5 January 2023 Revised: 27 January 2023 Accepted: 6 February 2023 Published: 10 February 2023

**Copyright:** © 2023 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 (https:// creativecommons.org/licenses/by/ 4.0/).

challenges in the food industry is to increase the healthiness of food without sacrificing sensory attributes [6].

Hemp (*Cannabis sativa* L.) is a plant used as textile fibers (from vegetative organs) native to the regions north and south of the Himalayas. Its use dates back to the Neolithic, and China is the country where it has been cultivated for the longest time. Its introduction in Europe probably dates back to the second millennium BC. Worldwide, it is cultivated mainly in Asia (China and India), Eastern Europe, and Russia. Today it is mainly used for textile, biocomposite, papermaking, construction, biofuel, and cosmetic purpose.

*Cannabis sativa* seeds can be used by the agri-food industry to produce flour, pastry, and oil, while the stem through to the canapulo (woody part of the stem) is in the green building sector. Its fiber (the external part of stem) will find new applications in the textile industry. Hemp inflorescences and roots, thanks to the extraction of bioactive molecules, will play an important role in the pharmaceutical and para-pharmaceutical industry [7].

Hemp seeds are mainly used as animal feed, but there is growing interest in their usage for human nutrition as a source of nutrients. They contain 25–35% of oil, 20–25% protein, 20–30% carbohydrates, and 10–15% insoluble fiber, vitamins, and minerals [8]. In particular, hemp seed oil is high in polyunsaturated fatty acids, with an ideal ratio (3:1) of linoleic acid (ω-6) and α-linolenic acid (ω-3) for human nutrition [9,10]. Merlino et al. [4] incorporated hemp seed flour (HSF) as a fortifying ingredient in the production of gnocchi, a typical Italian potato-based fresh pasta, from 5 to 20% HSF in substitution of soft wheat flour. Addition of HSF allowed for the enhancement of the nutritional value of gnocchi as a "source of fiber" in formulations with ≥ 10% of HSF. The fortified gnocchi had a high technological quality for cooking loss, cooking resistance, and textural properties, and average sensory quality; however, improving the HSF sensory quality for consumers' satisfaction was necessary in terms of odor and bitter taste. Hemp seeds were used to enrich pasta (15%), and its effects on osteoarticular pain and bone formation markers in patients with osteoarthritis in post-arthroplasty rehabilitation were evaluated. The first results showed that hemp seed can improve pain symptoms in patients with osteoarthritis undergoing arthroplasty surgery and improves bone metabolism both in humans and in vitro [11]. Pasta samples fortified with 5–40% commercially available hemp flour or 2.5–10% of hemp cake obtained from hemp seed oil pressing were studied [12]. The addition of hemp seed raw materials led to an increase in the protein, total dietary fiber (TDF), ash, and fat contents in the pasta samples. Due to its lower granulation and higher nutritional value, hemp flour was found to be a better raw material for the fortification of pasta than hemp cake.

In this study, the rheological and chemical qualities of pasta obtained by using the durum wheat cultivar "Ciclope", fortified with different percentages of hemp flour (cv. Futura 75), were investigated. The influence of semolina replacement at different percentages (5%, 7.5%, and 10%) and using two hemp flours (with different particle sizes) was evaluated, highlighting the effect of the particle size of hemp flours on the sensory properties, cooking quality, mineral composition, and nutritional characteristics of cooked functional pasta.

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

#### *2.1. Raw Material*

The durum wheat cv Ciclope was chosen among the durum wheat varieties made up by CREA Research Centre for Cereal and Industrial Crops of Acireale (Catania–Sicily, Italy). Durum wheat grain sample was milled using an experimental laboratory mill (Bona Labormill 4RB, Monza, Italy) and fine particles were separated (250 μm), with an extraction rate of about 55–60%.

Hemp flours were obtained from cv Futura seeds provided by Mulino Crisafulli (Caltagirone, Catania, Italy). The hemp seeds were subjected to mechanical extraction of the oil; the residue was milled into powder using a blender model M20 (IKA, Staufen, Germany) and was kept in hermetic bottles at room temperature (20 ± 2 ◦C) and sieved at 530 μm (namely Hemp 1) and 236 μm (namely Hemp 2).

#### *2.2. Pasta Making, Cooking Quality and Sensorial Analysis*

A Ciclope semolina pasta sample (CTRL) and pasta fortified with variable percentages of hemp flour (Hemp 1 and Hemp 2 at 5, 7.5, and 10% of substitution) were prepared (Table 1).

**Table 1.** Ciclope and fortified pasta samples.


Hemp 1 is hemp flour sieved at 0.530 mm; Hemp 2 is flour sieved at 0.236 mm. The fortified pastas take into account the different percentages of substitution of semolina with the two different hemp flours.

Pasta samples were prepared using a Pastamatic ARIETE 1591 equipped with a mixer and an extruder (De Longhi Appliances s.r.l., Florence, Italy), mixing 500 g flour (durum wheat semolina + x% *w*/*w* of hemp flours) with distilled water for 10.5 min. in order to obtain a dough with 40% moisture. The dough was extruded into a mancherons shape (5 cm long) following the procedures described by Cardullo et al. [13]. The optimal cooking time (OCT), the cooking loss, and the amount of solid substance in the cooking water were evaluated according to the AACC-approved method 66–50 (2000). The swelling index of cooked pasta was determined according to the procedure described in previous papers [2,3,8–10].

In order to evaluate the sensory attributes, a panel of 8 trained tasters (4 men and 4 women, aged between 30 and 64 years) analyzed the cooked pasta in OCT. Panelists' lists were developed on the basis of their sensory skills (ability to accurately determine and communicate sensory attributes such as the appearance, odor, flavor, and texture of a product). They judged bulkiness, firmness, adhesiveness, fibrous, color, odor, and taste. Based on the above-mentioned attributes, panelists were also asked to score the overall quality (OQS) of the product.

A 9-point scale was used: 1—very clear, 9—very dark in terms of color; 1—extremely unpleasant, −9—extremely pleasant in terms of bulkiness, firmness, adhesiveness, and fibrousness; 1—extremely unpleasant, 9—extremely pleasant in terms of odor, taste, and OQS [14–16].

#### *2.3. Chemical Characterization*

#### 2.3.1. Polyphenols Extraction

Phenolic compounds (PCs) can be classified as free, conjugated (to sugars and low molecular-weight compounds), and insoluble bound phenolics (BPs); these latter are covalently bound to the structural components of the cell wall [17,18]. BPs are not extractable in aqueous and/or organic solvents; therefore, preventive hydrolysis based on alkaline or acidic treatments is one of the most valuable strategies for targeting these compounds. The alkaline treatment can cleave the ester bonds linking the compounds to the cell wall, thus allowing for the release of PCs (mainly phenolic acids) from the insoluble residues. Free and bound phenolics were extracted using modified methods [19–21].

Eight grams of sample (Ciclope, Hemp 1 and Hemp 2 flours and ground, cooked pasta samples) were homogenized for 45 min in 40 mL 80% methanol solution using an ultrasonic bath. The samples were centrifuged at 5000× g for 15 min, and the supernatant was recovered. The pellet was re-extracted four times (repeating the protocol described above) and the supernatant was collected and evaporated using a rotary evaporator under vacuum at 45 ◦C. The residue was redissolved in 2 mL of methanol. This solution, containing free phenolic compounds (PCs), was filtered through a 0.22 μm nylon syringe filter into glass vials prior to HPLC-ESI/QTOF-MS analysis. In order to obtain the bound phenolics (BPs) extract, the residues separated after centrifugation were digested in 40 mL of NaOH 4 M for 1 h at room temperature and acidified using hydrochloric acid to pH 2. Subsequently, the acid solution was extracted with ethyl acetate (50 mL) four times and the organic fraction was evaporated in a rotary evaporator at a temperature of 45 ◦C. The residue was redissolved in 2 mL of methanol and solution filtered through a 0.22 μm nylon syringe filter prior HPLC-ESI/QTOF-MS analysis of bound phenolic (BPs) fraction. Both extractions for free (PCs) and bound phenolics (BPs) in samples were performed in triplicate.

#### 2.3.2. HPLC-ESI/QTOF-MS Analysis of Phenolic Compounds

The phenolic profile of hemp flour and pasta was investigated through an untargeted metabolomics-based approach using a HPLC-ESI/QTOF-MS method previously optimized [21]. The equipment consisted of an Alliance 2695 (Waters) HPLC system equipped with an autosampler, degasser, and column heater coupled with a Quadrupole Time-of-Flight (Waters Q-ToF Premier) mass spectrometer. The compounds were separated using a Phenomenex Luna C18 column (100 cm, 2 mm, 3 μm particle size). The phenolic compounds identified in different flours and pasta samples, were next quantified according to their class and sub-class, using calibration curves in a range of 2.5 μg mL<sup>−</sup>1–25 μg mL<sup>−</sup>1, built from pure reference standards (chlorogenic acid, catechin hydrate, rutin, caffeic acid, kaempferol, sinapic acid, and benzoic acid; Supplementary Table S1). When reference compounds were not available, the quantitation was based on structurally related substances. Specifically, rutine in negative mode was the reference compound for the determination of cannabisin B and cannaflavin C. Sinapic acid was the reference compound in negative mode for ferulic acid, chlorogenic acid was the reference compound for N-trans-caffeoyltyramine, and benzoic acid was the reference compound used for protocatechuic acid and vanillic acid semi-quantification.

#### 2.3.3. Total Phenolic Content (TPC)

The content of total phenolics (TPC) was determined using the Folin–Ciocalteau method [22]. A calibration curve was set with gallic acid ranging from 0.001 to 0.25 mg mL−<sup>1</sup> methanol/water (80:20 *v*/*v*) (*y* = 10.955x + 0.1405, *R*<sup>2</sup> = 0.992). The results were expressed as mg gallic acid equivalents per g (mg GAE g−1) of sample. In this method, 5 mL methanol/water (80:20 *v*/*v*) was added to 0.5 g of the flour samples (Hemp 1, Hemp 2, Ciclope flours, pasta samples obtained from semolina replacement with the two hemp flours at 5, 7.5, and 10%), then, the obtained mixture was filtered through a 0.45 μm PTFE syringe filter. Next, 125 μL of the solution was mixed with 625 μL of diluted (1:5) Folin– Ciocalteau reagent in water and 120 μL of 7% Na2CO3. The samples were left in the dark for 1 h at room temperature. The TPC was measured four times for each sample.

#### 2.3.4. Fatty Acid Composition

After the basic hydrolysis of triglycerides, it was necessary to convert fatty acids into their methyl esters (FAMEs). Quali-quantitative determinations of FAMEs were conducted according to Melilli et al. [23] using a gas chromatography–mass spectrometry (GC/MS) ISQ™ 9000 Quadrupole GC-MS System (Thermo Fisher Scientific, Waltham, MA, USA). The identification of FAMEs was performed by comparing their retention times with those of reference standards (mixture FAME Mix, SUPELCO, which included 37 FAMEs). The results of the FAMEs were expressed as relative percentages (%).

#### 2.3.5. Amino Acids (AAs) Quantification by HPLC-FLD Method

Some procedures are needed for amino acid analysis, such as proteins hydrolysis. The modified procedure employing the acid hydrolysis of protein and the derivatization of the free amino acids using FMOC-Cl (9-fluorenylmethylchloroformate) was required prior to analysis with HPLC-FLD. Five hundred milligrams of the sample were added with 1 mL HCl 6 M in order to support the subsequent hydrolysis of proteins and were incubated in an oven at 110 ◦C for 24 h. After cooling to room temperature, the sample was diluted with 2 mL of deionized water and filtered with 0.45 μm PTFE syringe filters. The solution was subjected to pre-column derivatization by reaction of the sample with FMOC-Cl: 200 μL of 3 mM FMOC-Cl acetonic solution and 200 μL of borate buffer were added to 50 μL of the solution containing amino acids. The solution was heated at 70 ◦C for 10 min. Subsequently, 50 μL of a heptylamine solution (3 mL heptylamine, 15 mL ACN, and 175 mL HCl 0.1 M) was added to the solution and mixed for 3 min. Eighty microliters of the latter solution were taken, and 320 μL ACN and 600 μL hexane were added. A volume of 20 μL of this solution was injected into the HPLC-FLD instrument. Derivatized amino acids analyses were carried out using an HPLC Agilent 1100 Series chromatographic system equipped with a G1312A binary gradient pump and a fluorescence FLD detector and controlled by Chemstation software. For the chromatographic separation of derivatized amino, a Discovery HS C18 column was used (4.6 mm × 150 mm. 3.5 μm) (Supelco, Bellefonte, PA, USA), which fitted with guard column. The column operated at 40 ◦C, the flow rate was maintained at 1mL min−1. Mobile phases were 0.1% formic acid as eluent A and ACN as eluent B. The program of gradient elution was as follows: 0–10 min, 3% B; 3–17 min, linear increase to 10% B; 17–47 min, linear increase to 50% B; 47–57 min, linear increase to 100% B; 57–60 min, hold 100% B; 60–63 min, equilibration and return to the initial conditions. Each derivative eluted from the column was monitored by a fluorometric detector (FLD) set to an excitation wavelength of 254 nm and an emission wavelength of 630 nm. A comparison of the retention times of the standards for peak identification was carried out, and a fortification technique (spiking) was applied. Quantitative determination of the derivatized amino acids was performed using calibration curves. Standard solutions of the derivatized amino acids were prepared at five concentration levels in a range from 0.025 mM to 0.4 mM (Supplementary Table S2). The results were expressed in terms of grams of amino acids in 100 g of sample.

#### 2.3.6. Antiradical Properties of Raw Materials and Functional Pasta

The antiradical activity of samples (flours and fortified pasta) was measured using the DPPH assay.

One gram of each sample was extracted with 4 mL of methanol for 40 min in an ultrasonic bath. The supernatant was filtered using a 0.45 μm PTFE syringe filter. One hundred microliters of the filtrate were mixed with 3 mL DPPH (60 μM in methanol) and placed in the dark for 30 min. Absorbance at 515 nm was measured with a spectrophotometer (Varian Cary® 50 UV-Vis spectrophotometer) using methanol as a blank. Antiradical scavenging activity was expressed as the percentage inhibition of the DPPH radical and was calculated using the following Equation:

$$\text{Scavenging}\% = (\text{A0} - \text{Ai/A0}) \times 100$$

where A0 is the absorbance of DPPH without the sample, and Ai is the absorbance of the sample and DPPH. The results were also reported as TEAC (Trolox equivalent antioxidant activity) and expressed in terms of mmol Trolox equivalents (TE)/100 g of sample. Trolox was utilized as the standard, and the calibration curve in a range between 5 and 400 μM was prepared using methanol as solvent (y = 0.0037x + 0.1655 and R2 = 0.987). All of the experiments were carried out in triplicate.

#### 2.3.7. Mineral Profile of Pasta

In order to assess the influence of cooking on the exchange of mineral contents, the elemental composition of the raw and cooked hemp pasta samples and of the different cooking waters was established. Elemental analysis of microelements (As, Be, Cd, Co, Hg, Li, Ni, Sb, Se, Sn, Sr, and V) was performed using an inductively coupled mass spectrometer ICP-MS iCAP RQ, (Thermo Fisher Scientific Inc., Bremen, Germany) operating with argon gas of spectral purity (99.999 sample solutions were pumped by a peristaltic pump from tubes arranged on a CETAC ASX-520 auto-sampler (Thermo Scientific, Omaha, NE, USA). Instrument sensitivity, resolution, and mass calibration were optimized daily with the tuning solution (iCAP Q/RQ Tune aqueous multielement standard solution (Thermo

Scientific, Bremen, Germany) in order to maximize ion signals and minimize interference due to high oxide levels, optimizing torch position, ion lenses, gas output, resolution axis, and background. The optimal parameters are shown in Table 2.

**Table 2.** Operating conditions and acquisition parameters for ICP-MS.


The Al, B, Ba, Ca, Cu, Fe, Mg, Mn, Mo, Na, P, and Zn contents were determined using an inductively coupled plasma optical emission spectrometer (ICP-OES Analyzer, iCAP 7400, Thermo Fisher Scientific Inc., Waltham, MA, USA) equipped with a concentric nebulizer and a cyclonic spray chamber. The operating conditions are shown in Table 3.

**Table 3.** Operating conditions and acquisition parameters for ICP-OES.


Sample preparation was carried out using an Anton Paar Multiwave 5000 digestion system equipped with an XF100 rotor. In order to decontaminate PTFE vessels, a cleaning procedure was carried out by adding 4 mL of HNO3 and 4 mL of H2O to each vessel under the following conditions: 1100W for 15 min. After cleaning, vessels were rinsed with ultrapure water and dried [24]. Aliquots of 0.5 g of each pooled sample were weighted directly into the PTFE vessel of the microwave system. Digestion was performed by adding 8 mL of HNO3. The operating conditions used for the microwave digestion were 800 W over 15 min and then hold at this power for 30 min. After digestion, samples were quantitatively transferred to a graduated polypropylene test tube and diluted with ultrapure water to 50 mL and stored at 4 ◦C until analysis. Each sample's digestion was performed in triplicate. The analytical batch consisted of a set of calibration standard samples and a minimum of three procedural blanks. Each solution was measured in triplicate, and analyses were carried out by a classical external calibration approach. For each element at least six calibration points were considered for calculation. The concentration range was selected based on the expected elemental values and sample dilution. The calibration ranges were: 0.005–100 μg L−<sup>1</sup> for microelements; 0.002–1 mg L−<sup>1</sup> for Ba, Cu, Mn, Mo, and Zn; 0.1–100 mg L−<sup>1</sup> for Ca, Mg, Na, and K; and 0.01–10 mg L−<sup>1</sup> for Al, B, Fe, and P. Stock solutions of calibration standards were properly diluted with 5% HNO3.

#### *2.4. Data Analysis*

Data were submitted to the Bartlett's test for homogeneity of variance and then analyzed using two-way analysis of variance (ANOVA), based on a factorial combination of particle size (PS) × percentage of substitution (S) for the sensory characteristics of pasta. A separate ANOVA was conducted for each measurement and each main factor (particle size or percentage of substitution); in this case, means were statistically separated based on the Student–Newman–Keuls test. The CTRL was excluded when comparing pasta fortified with the two types of hemp flours. All other data, following Bartlett's test for the homogeneity of variance, were analyzed using one-way ANOVA, and means were compared by *LSD* test when the *F*-test was significant, at least at the 0.05 probability (CoHort Software, CoStat version 6.451).

#### **3. Results**

#### *3.1. Chemical Characterization of Raw Material*

The chemical characterization of the raw material is shown in Table 4. The total phenolic contents (TPC) of Ciclope durum flour and Hemp 1 and Hemp 2 were examined. For Ciclope flour, total polyphenol content, determined using the Folin–Ciocalteau method, was 2.45 mg GAE/g. In the case of hemp flours, similar total polyphenol values were obtained (6.38 and 6.35 mg GAE/g, respectively, for Hemp 1 and Hemp 2). The DPPH method has been widely used in antiradical activity studies of plant extracts [25]. DPPH radical scavenging activity was also expressed as the % scavenging value. The results of the radical scavenging activity of sample flours showed that all extracts had the ability to scavenge DPPH radical with values of 29.7, 53.2, and 51.5 for Ciclope, Hemp 1, and Hemp 2, respectively. The results confirmed that Hemp 1 and Hemp 2 flours showed similar antiradical activity, which was higher than Ciclope durum wheat flour.

**Table 4.** Chemical characterization of raw materials.


Data are expressed as the means ± SD of triplicate experiments. Hemp 1 and Hemp 2 were diluted 10-fold for anti-scavenger determination. Different letters indicated differences at *p <* 0.05 between Hemp 1 and Hemp 2 samples. Values without letters are not significantly different.

Although the hemp flours come from the shredding and sieving of the defatted seeds, a small percentage (about 8%) of oil remained in the solid matrix. The fatty acid profile was evaluated by GC-MS analysis and reported in Table 4. The main fatty acids identified in the lipid fractions of Ciclope flour were palmitic, oleic, and linoleic acids, with lower percentages of α linolenic acid. The lipidic profiles of hemp flours were shown in linoleic and α linolenic acid as major fatty acids, and oleic and palmitic acids as minor. Regarding the quantity of fatty acids, there was no differences in the lipid profile of Hemp 1 and Hemp 2 flours. The fatty acid present in largest amount was linoleic acid in 53% and α linolenic acid with 15.5%. Results were consistent with the analyses of other authors; Siano et al. [26] in agreement with our results, identified a similar fatty acid composition with linoleic acid as prevalent (56.42%), followed by linolenic (14.55%), oleic (12.79%), γ-linolenic (3.03%) and as saturated, palmitic (7.35%) and stearic acids (2.26%). Pojic work [27] highlighted high content of linoleic (54.09%–55.43%), linolenic (17.31–18.42%) and oleic (12.96–13.93%) acids, followed by palmitic (6.48–7.90%), stearic (3.18–3.86%) and γ-linolenic (2.61–2.76%).

There are no studies on the amino acid composition of Ciclope flour and hemp flour. Table 4 moreover summarizes the mean individual and total free amino acid (AAs) contents observed in the studied flours samples. According to the AAs profiles isoleucine, leucine, tyrosine and serine were among the amino acids with the highest content in Ciclope. The total AA content was 19.04 g/100 g in Ciclope. In Hemp 1 and Hemp 2 isoleucine, glutamine, tyrosine, proline and lysine were the most abundant. The particle size affected the amounts of the total AA content with 32.3 g/100 g (Hemp 1) and 34.4 g/100 g (Hemp 2). Particularly interesting was the amount of essential amino acids determined in the hemp flours which resulted in 15.9 g/100 g and 17.5 g/100 g in Hemp 1 and Hemp 2, respectively.

Hydroxycinnamic and protocatechuic acids (Table 5) represented the most abundant phenolic compounds quantified in hemp flour samples. The samples were subjected to alkaline treatment in order to determine the bound phenolic fraction (BPs). Results show a higher amount of bound hydroxycinnamic acid in Hemp 1 than Hemp 2 (1687.4 μg 100 g−<sup>1</sup> and 1589.0 μg 100 g<sup>−</sup>1, respectively). The presence of cannaflavin C is highly relevant both in PCs and BPs form. Values ranging from 1384.1 to 3367.0 μg 100 g−<sup>1</sup> for Hemp 1 and from 1139.2 to 2207.0 μg 100 g−<sup>1</sup> for Hemp 2.


**Table 5.** Phenolic compounds (μg 100 g−1): Free and bound quantification (PCs and BPs) of raw materials.

Data are expressed as means ± SD of triplicate experiments. n.d. not detected.

As far as phenol amides were concerned, N-trans-caffeoyltyramine was quantified uniquely in Hemp 2 as bound phenol (1817.1 μg 100 g−1). In Ciclope flour modest amounts of bound phenols have been determined, among these protocatechuic acid, *p*hydroxycinnamic acid and vanillic acid (189.5, 329.6 and 75.2 μg 100 g<sup>−</sup>1, respectively). No free-form phenolic compounds were found.

#### *3.2. Chemical Characterization of the Functional Pasta*

Total phenolic contents and antiradical activity were determined on cooked Ciclope pasta and the fortified pasta samples as shown in Table 6. The data reveal TPC values in Hemp 1\_10% and in Hemp 2\_10% pasta samples (4.92 ± 0.31–4.21 ± 0.35 mg GAE/g, respectively), in agreement with the activity of radical scavenging (3.86 ± 0.07–3.14 ± 0.06 mmol TE/100 g, respectively), higher than Ciclope semolina pasta.


**Table 6.** Chemical characterization of cooked pasta.

Data are expressed as means ± SD of triplicate experiments. Means were separated by LSD test at *p* < 0.05.

Methyl ester fatty acids profile, as shown in Table 6, was also studied on cooked pasta samples. The main differences concern γ linolenic acid and α linolenic acid. In Hemp 1\_10% and in Hemp 2\_10% pasta samples reported, respectively, 5.26% and 5.71% of γ linolenic acid, and 15.89 of α linolenic acid.

Table 6 also highlights the amino acid content in the cooked pasta sample. Amino acids such as tyrosine, glutamine, proline and isoleucine are among the main in Hemp 1\_10% and in Hemp 2\_10% pasta. The concentration of essential amino acids in the two different pasta samples was interesting. In particular, there was a concentration of 4.30 g/100 g in Hemp 1\_10% and 4.62 g/100 g in Hemp 2\_10%. The concentration of total amino acids in the 10% fortified pasta was more than double that of the Ciclope durum wheat pasta.

In hemp-fortified pasta, phenols were detected and quantified (Table 7).

As expected, the alkaline treatment of the samples allowed for the cleavage of the ester bonds that bind the compounds to the cell wall, thus allowing for the release of PCs (mainly phenolic acids) from the insoluble residues.

In the pasta samples fortified, after the alkaline treatment was observed a greater presence of bound phenolics, particularly for the samples obtained with 10% substitution. Cannaflavin C, *p*-hydroxy benzoic acid, protocatechuic acid, hydroxycinnamic acid, and caffeic acid were predominantly found in Hemp 1 pasta in bound form. Conversely, *p*-hydroxycinnamic acid, caffeic acid, *p*-hydroxy benzoic acid, protocatechuic acid, and trans-N-caffeoyl-tyramine resulted in higher quantities in Hemp 2 pasta samples. As expected, the amount of free and bound phenols decrease the percentages of fortification.


**Table 7.** Phenolic compounds (μg 100 g<sup>−</sup>1): free and bound quantification in cooked pasta (PCs and BPs).

Data are expressed as the means ± SD of triplicate experiments. n.d., not detected.

The fortification at 7.5% especially with Hemp 1, showed good values of bound phenols. A similar phenolic profile was also reported by Pannico et al. [28] and Izzo et al. [29].

The quantification of minerals reported in Table 8 was carried out using external calibration curves. The data allow for the assessment of the contribution of hemp enrichment to the macro-element's composition of pasta.

Most of the minerals' concentrations increased according to the fortification percentage. In particular, the addition of hemp to pasta increased the content of iron, potassium, magnesium, and phosphorus. The iron concentration increased from 0.035 mg g−<sup>1</sup> to 0.051 mg g−1, the potassium concentration increased from 1.953 mg g−<sup>1</sup> to 2.020 mg g−1, the magnesium concentration increased from 0.873 mg g−<sup>1</sup> to 1.191 mg g−1, and the phosphorus concentration increased from 2.502 mg g−<sup>1</sup> to 3.086 mg g−<sup>1</sup> in Hemp 1\_5% pasta and Hemp 1\_10% pasta, respectively. As for trace elements, the copper concentration increased from 7.903 ug g−<sup>1</sup> to 9.669 ug g−<sup>1</sup> in Hemp 1\_5% pasta and Hemp 1\_10% pasta, respectively. The same results were recorded in pasta fortified with Hemp 2. The mineral element content in hemp seeds was nutritionally interesting, as reported by Alonso et al., 2022 [30]. Phosphorus, potassium, magnesium, calcium, iron, zinc, manganese, and copper are essential dietary elements for mammals and are involved in many physiological processes [31]. In Figure 1 the percentage of variation is reported for the most important minerals in the hemp-enriched pasta vs the CTRL. Pasta enriched with Hemp\_2 yielded the best results for all of the considered minerals except Fe.

**Table 8.** Mineral content in cooked, fortified pasta.



**Table 8.** *Cont.*

Data are expressed as the means ± SD of triplicate experiments.

**Figure 1.** Percentage variation (%) of the hemp-enriched pasta samples vs the CTRL (durum wheat semolina cv. Ciclope).

#### *3.3. Pasta Quality*

Semolina particle size is a key quality factor in pasta making. Semolina used for pasta processing typically ranges in particle size from 550 to 150 μm [31]. The semolina used in this study had fine particles (<250 μm), similar to Hemp 2 flours. The addition of hemp flour significantly affected the sensory attributes of cooked pasta (Table 9). Substantial differences were recorded between the two hemp flours for all of the sensory attributes except bulkiness. In general, the use of Hemp 2, with a similar particle size to semolina, to enrich pasta yielded better results than Hemp 1. The absence of proteins such as glutenins and gliadins, responsible for the formation of gluten, has inevitably influenced the characteristics of the product. CTRL recorded the greatest OQS, mainly in terms of firmness and adhesiveness (Table 9); the enrichment of durum wheat pasta with non-gluten flours may affect the parameters, resulting in an increase in adhesiveness. The odor and taste of pasta fortified at different substitution levels resulted similarly to CTRL pasta, suggesting that the particle size of hemp did not affect these traits. The best results in terms of OQS were obtained using Hemp 2 at 7.5% substitution.

**Sample Bulkiness <sup>I</sup> Firmness <sup>I</sup> Adhesiveness <sup>I</sup> Fibrous <sup>I</sup> Color II Odor III Taste III OQS III** CTRL 3.7 ± 0.012 5.7 ± 0.018 4.3 ± 0.014 5.0 ± 0.016 5.0 ± 0.016 4.7 ± 0.015 6.3 ± 0.021 6.0 ± 0.020 Hemp 1\_5 3.0 <sup>±</sup> 0.010 4.7 <sup>±</sup> 0.015 5.0 <sup>±</sup> 0.016 4.7 <sup>±</sup> 0.015 4.7 <sup>±</sup> 0.015 4.7 <sup>±</sup> 0.015 5.3 <sup>±</sup> 0.017 5.3 <sup>±</sup> 0.017 Hemp 1\_7.5 3.7 <sup>±</sup> 0.012 4.0 <sup>±</sup> 0.016 4.7 <sup>±</sup> 0.015 5.7 <sup>±</sup> 0.018 5.0 <sup>±</sup> 0.016 5.7 <sup>±</sup> 0.018 5.7 <sup>±</sup> 0.018 5.3 <sup>±</sup> 0.017 Hemp 1\_10 4.5 <sup>±</sup> 0.015 5.0 <sup>±</sup> 0.013 3.5 <sup>±</sup> 0.011 4.5 <sup>±</sup> 0.015 5.0 <sup>±</sup> 0.016 4.5 <sup>±</sup> 0.015 6.0 <sup>±</sup> 0.020 4.0 <sup>±</sup> 0.013 Hemp 2\_5 3.7 <sup>±</sup> 0.012 4.0 <sup>±</sup> 0.013 4.3 <sup>±</sup> 0.014 5.0 <sup>±</sup> 0.016 5.3 <sup>±</sup> 0.017 5.3 <sup>±</sup> 0.017 5.3 <sup>±</sup> 0.017 5.3 <sup>±</sup> 0.017 Hemp 2\_7.5 5.3 ± 0.017 5.7 ± 0.018 6.0 ± 0.020 5.7 ± 0.018 6.3 ± 0.021 5.7 ± 0.018 6.7 ± 0.022 6.3 ± 0.021 Hemp 2\_10 4.3 <sup>±</sup> 0.014 5.0 <sup>±</sup> 0.016 4.7 <sup>±</sup> 0.015 5.7 <sup>±</sup> 0.018 5.0 <sup>±</sup> 0.016 4.7 <sup>±</sup> 0.015 5.3 <sup>±</sup> 0.017 5.0 <sup>±</sup> 0.016 LSD (*<sup>p</sup>* < 0.05) 0.034 0.039 0.036 0.040 0.041 0.039 0.045 0.042 Percentage of substitution (S) CTRL 3.7 ± 0.012 b 5.7 ± 0.018 a 4.3 ± 0.014 c 5.0 ± 0.016 b 5.0 ± 0.016 b 4.7 ± 0.015 c 6.3 ± 0.021 b 6.0 ± 0.020 a 5% 3.3 ± 0.013 b 4.3 ± 0.014 b 4.6 ± 0.015 b 4.8 ± 0.015 c 5.0 ± 0.016 b 5.0 ± 0.016 b 5.3 ± 0.017 d 5.3 ± 0.017 b 7.5% 4.5 ± 0.015 a 4.8 ± 0.017 a 5.3 ± 0.018 a 5.6 ± 0.018 a 5.6 ± 0.019 a 5.7 ± 0.018 a 6.7 ± 0.020 a 5.8 ± 0.024 a 10% 4.4 ± 0.015 a 5.0 ± 0.015 a 4.0 ± 0.013 c 5.1 ± 0.017 b 5.0 ± 0.016 b 4.6 ± 0.015 c 5.7 ± 0.019 c 4.5 ± 0.015 c Particle size (PS) Hemp 1 3.7 ± 0.012 a 4.5 ± 0.015 a 4.4 ± 0.014 b 4.9 ± 0.016 b 4.9 ± 0.016 b 4.9 ± 0.016 a 5.7 ± 0.018 a 4.9 ± 0.016 b Hemp 2 4.4 ± 0.014 a 4.9 ± 0.016 b 4.8 ± 0.016 a 5.4 ± 0.018 a 5.5 ± 0.018 a 5.2 ± 0.017 a 5.8 ± 0.019 a 5.5 ± 0.018 a ANOVA Main effects S \*\*\* \*\*\* \*\*\* \*\*\* \*\*\* \*\* \*\*\* \*\*\* PS \*\*\* \*\*\* \*\*\* ns \*\*\* ns ns \*\*\* Interaction S × PS \*\*\* \*\*\* \*\*\* \*\*\* \*\*\* \* \*\*\* \*\*\*

**Table 9.** Sensory characteristics of cooked pasta.

Data are expressed as the means ± SD of triplicate experiments. \*\*\*; \*\*, and \* indicate significance at *p* < 0.001, *p* < 0.01, and *p* < 0.05, respectively, while "ns" is not significant. The differences among all samples were detected by LSD test at *p* < 0.05. Different small letters in a column for "PS" and "S" factors indicate statistical differences among samples (*p <* 0.05) (Student–Newman–Keuls test). OQS means "over quality score". I: 1—low sensation, -9—high sensation; II: 1—very clear, -9—very dark; III: 1—extremely unpleasant, -9—extremely pleasant; n = 24.

As regards to cooking quality, the replacement of durum wheat semolina with both types of hemp flours in the pasta statistically influenced (*p* < 0.05) the water absorption, most probably due to the high dietary fiber content and resultant strong water absorption capacity [13,15], while the optimal cooking time compared to the CTRL sample increased, particularly when using the Hemp 1 flours (Table 10).

The amount of solid substance lost in the cooking water (cooking loss) did not result as being influenced by the hemp particle size or the percentage of substitution, meaning that the hemp flours were well retained in the pasta. According to Sicignano et al., [31] the hydration of semolina with a wide range of particle sizes affects dough development and pasta quality; the different particle sizes between the semolina and Hemp 1 flours probably led to an over-hydration of the finer fraction and under-hydration of the coarser fraction, affecting the WA and OCT (Table 10).


**Table 10.** The cooking quality, optimum cooking time (OCT), water absorption capacity (WA), and cooking loss of fortified pasta samples.

Different small letters in a column. indicate statistical differences between the hemp flours used for the same concentration at *p <* 0.05; n = 3. Data are expressed as the means ± SD of triplicate experiments.

#### **4. Discussion**

The main aim of this study was to develop pasta fortified with variable percentages of hemp flour with different particle sizes. Our results revealed that the incorporation of hemp flour into the pasta formulas led to significant increases in the TPC and DPPH values, AAs, FA composition, and sensory qualities, in addition to obtain satisfactory properties

and good cooking qualities, related to the percentage of hemp substitution used in the production recipe. The fibrous sensation recorded was probably due to the different particle sizes between the semolina and Hemp 1 flour and to the different percentages of water required for the dough development detected by farinograph analysis. Blends of semolina and Hemp 2 required less water (on average 58%) vs Hemp 1 (on average 61%) (data not shown).

The Folin–Ciocalteau method was used for raw material; similar polyphenol content values were obtained for Hemp 1 and Hemp 2 (6.38 and 6.35 mg GAE/g, respectively), while the TPC value was lower for Ciclope flour (2.45 mg GAE/g). TPC values increased with the addition of hemp flour in pasta. The highest increase in TPC was observed in pasta samples containing 10% hemp flour (4.92 ± 0.31 mg/GAE and 4.21 ± 0.35 mg/GAE for Hemp 1 and Hemp 2, respectively) while the lowest was recorded for the CTRL pasta sample (1.11 ± 0.18 mg/GAE).

However, despite the loss in amino acids and phenolics during cooking (about 40%), the enriched pasta still showed good antioxidant activity. The improvements to DPPH values were found to be higher in pasta formulas with the addition of hemp flour than in CTRL samples prepared with 100% Ciclope wheat flour. In fact, the supplementation of 10% hemp flour also enhanced the antioxidant activity (3.86 ± 0.07 mmol TE/100 g and 3.14 ± 0.06 mmol TE/100 g for Hemp 1 and Hemp 2 pasta samples, respectively) compared to the CTRL samples (1.14 ± 0.05 mmol TE/100 g).

This study also focused on the AAs composition of fortified pasta. The contents of some amino acids considered essential in the human diet can be low in wheat proteins, especially lysine and threonine. The preparation of a functional pasta enriched with variable percentages of hemp flour could affect the content of these two amino acids.

From observed data, the lysine content was found to be 0.16, 0.85–0.36 g/100 g in the CTRL, Hemp 1\_10, and Hemp 2\_10 flours respectively, while threonine was found to be 0.38, 0.72–1.17 g/100 g in the CTRL, Hemp 1\_10, and Hemp 2\_10 flours respectively). The total essential amino acids in the fortified pasta samples were 4.30 and 4.62 g/100 g (in Hemp 1\_10 and in Hemp 2\_10, respectively), while in the durum wheat pasta they were 2.05 g/100 g.

An increase in mono- and polyunsaturated fatty acids was also observed in fortified pasta. The total ω 3 contents varied between 15.9 and 15.3% in Hemp 2 pasta samples and between 15.9–14.7% in Hemp 1 pasta samples. Linoleic was the fatty acid present in the largest amount in Hemp 2 pasta samples (47.94%). The control sample had a lower amount of total ω 3 (4.02%).

Phenols represent the most relevant compounds found in hemp, including some phenylamides, phenolic acids, lignanamides, and flavonoids, such as flavonols, flavones, and flavanols. Current literature suggests that the long-term consumption of diets rich in phenolic compounds protects against certain cancers, cardiovascular diseases, type 2 diabetes, osteoporosis, lung damage, and neurodegenerative diseases [32,33].

An untargeted metabolomics-based approach was used to comprehensively screen and profile phenols in different hemp flours and pasta samples through UHPLC-ESI/QTOF-MS analysis. A total of 12 phenols were identified (Table 8) by comparison with the retention time, MS spectra, and accurate mass measurement obtained from the literature data [17–21] and by phenolic reference standards (Supplementary Table S3).

As is known, phenolic compounds are contained in plant materials in the free (PCs) but also in the insoluble bound form (BPs); these latter are covalently linked to the structural components of the cell wall. Particularly interesting are the implications of BPs in foods, in terms of bioaccessibility, transformation during digestion, and modulation of the gut microbiota [34]. For this reason, the study of the bound polyphenolic fraction (BPs) in pasta samples was of interest.

In raw samples, the contents of free phenolic compounds (PCs) were rather low in Ciclope flour; hydroxybenzoic and protocatechuic acids represented the most abundant phenolic compounds quantified in hemp flour samples. The results of the phenolic component after alkaline treatment of the matrices showed a higher amount of these in flour. Regarding fortified pastas, the phenol content is good, especially in the bound form. As could easily be predicted, the content of these increases with the percentage of replacement. Cannabisin B was only found in Hemp 1 as a bound phenol and at lower levels with respect to cannaflavin C. Cannaflavin A, cannaflavin B, and cannabisins A, B, and C are nonpsychoactive molecules exclusively present in hemp plants that suppress PGE2 production in synovial membrane cells, exhibiting anti-inflammatory power 30-times stronger than acetylsalicylic acid [35]. Among hemp's exclusive lignanamides, these exhibit remarkable beneficial effects on human health [36].

Caffeoyltyramine and its phenolic amides, including cis-N-caffeoyltyramine and trans-N-caffeoyltyramin, are known to have anti-fungal, antioxidant, anti-inflammatory, and antihyperlipidemic activities [28]. As observed from Table 8, the phenols cannaflavin C, *p*-hydroxybenzoic acid, protocatechuic acid, hydroxycinnamic acid, and caffeic acid were predominantly found in Hemp 1 pasta in bound form.

In the pasta samples obtained by replacing the semolina with Hemp 2 flour, *p*hydroxycinnamic acid, caffeic acid, *p*-hydroxybenzoic acid, protocatechuic acid, and trans-N-caffeoyltyramine always resulted to a greater extent in the bound form. As expected, the amount of free and bound phenols is strongly related to the fortification rate.

The incorporation of hemp flours to produce pasta has been studied by other research teams, but the contribution of mineral content was not investigated. Our results represent the first finding of the addition of hemp flours producing fortified pasta rich in minerals. However, in this study the presence of phytates in hemp flours was not detected, and even if the pasta samples could be an excellent source of mineral elements, their nutritional quality could be reduced.

Each increase in the addition of hemp flour resulted in an increase in pasta quality (water absorption, due to gluten dilution) and in the satisfactory organoleptic properties (until 7.5%); the fibrous sensation recorded at higher percentages of substitution was probably due to the different particle sizes between semolina and Hemp 1 flour and to the different percentages of water required for the dough development detected by farinograph analysis. Blends of semolina and Hemp 2 required less water (on average, doses of 58%) compared to Hemp 1 (on average, doses of 61%). Additionally, the contribution of hemp enrichment improved the micro-elemental composition of pasta (iron, potassium, magnesium, and phosphorus) compared to the control sample. The number of health claims relating to mineral elements which could be used for hemp flours could be high, but further studies are needed on the bioaccessibility and bioavailability in order to clarify the role of hemp flour as a dietary source of mineral elements due to the presence of phytates in the raw material.

#### **5. Conclusions**

The present study concluded that the incorporation of hemp flours with different particle sizes represents the best compromise between pasta properties and nutrient content; the addition of hemp flour to durum semolina cv Ciclope flour effectively increased the anti-radical potential. The best results in terms of overall quality score (OQS) were obtained using flour with a minor particle size (Hemp 2) for the preparation of the pasta. The maximum substitution level, which showed the best performance during the production process, was 7.5% for both types of hemp flour used. However, some differences can be highlighted: specifically, the pasta obtained by the replacement of 7.5% of the Ciclope semolina with Hemp 2 flour showed a better profile in terms of mineral salts and amino acids and a greater quantity of polyunsaturated fatty acids, while the pasta obtained from the same percentage of substitution but with Hemp 1 flour showed a better phenolic profile and TCP and better anti-radical activity. Hemp supplementation could be a potential option for the production of high-quality, nutritionally rich, low-cost pasta with good organoleptic properties.

In summary, the enrichment of wheat pasta with hemp flours is a very interesting future trend that enables more attractive pasta products to be obtained for consumers in terms of increased nutritional and pro-health value. On the basis of these results, hempenriched pasta may have great potential in the industry for the development of functional products. Further studies on the bioavailability of nutrients, the glycemic index, and the effects on the intestinal microbiota of Hemp 1- and Hemp 2-fortified pasta will be conducted in the near future.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/foods12040774/s1. Table S1: Retention time (min.), coefficient of determination (R2) and linear regression model of external standards used for phenolic compounds calibration; Table S2: Retention time (min.), coefficient of determination (R2) and linear regression model of external standards used for amino acids calibration; Table S3: Retention time (min.), formula, MS spectra, and accurate mass measurement of phenolic compound identified.

**Author Contributions:** Conceptualization, M.G.M. and V.D.S.; methodology, V.D.S. and M.G.M.; formal analysis, S.B., V.D.S., C.B. and S.A.; investigation, S.B. and F.S.; resources, N.V.; data curation, V.D.S., C.B., and M.G.M.; writing—original draft preparation, M.G.M., F.S., C.B. and S.B.; writing review and editing, V.D.S., C.B. and M.G.M.; supervision, V.D.S. and M.G.M.; project administration, M.G.M.; funding acquisition, M.G.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** CNR-DISBA project NutrAge (project number 7022).

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Acknowledgments:** We would like to thank Giuseppe and Francesco Sammartino, project leaders of "The hemp supply chain-from hemp seed to table" SEMINCANTA", for supplying hemp flours.

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

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

## *Article* **Functional Bread Produced in a Circular Economy Perspective: The Use of Brewers' Spent Grain**

**Antonietta Baiano \*, Barbara la Gatta, Mariacinzia Rutigliano and Anna Fiore**

Dipartimento di Scienze Agrarie, Alimenti, Risorse Naturali e Ingegneria (DAFNE), University of Foggia, 71122 Foggia, Italy

**\*** Correspondence: antonietta.baiano@unifg.it; Tel.: +39-(0)881589249

**Abstract:** Brewers' spent grain (BSG) is the main by-product of the brewing industry, corresponding to ~85% of its solid residues. The attention of food technologists towards BSG is due to its content in nutraceutical compounds and its suitability to be dried, ground, and used for bakery products. This work was aimed to investigate the use of BSG as a functional ingredient in bread-making. BSGs were characterised for formulation (three mixtures of malted barley and unmalted durum (Da), soft (Ri), or emmer (Em) wheats) and origin (two cereal cultivation places). The breads enriched with two different percentages of each BSG flour and gluten were analysed to evaluate the effects of replacements on their overall quality and functional characteristics. Principal Component Analysis homogeneously grouped BSGs by type and origin and breads into three sets: the control bread, with high values of crumb development, a specific volume, a minimum and maximum height, and cohesiveness; Em breads, with high values of IDF, TPC, crispiness, porosity, fibrousness, and wheat smell; and the group of Ri and Da breads, which have high values of overall smell intensity, toasty smell, pore size, crust thickness, overall quality, a darker crumb colour, and intermediate TPC. Based on these results, Em breads had the highest concentrations of nutraceuticals but the lowest overall quality. Ri and Da breads were the best choice (intermediate phenolic and fibre contents and overall quality comparable to that of control bread). Practical applications: the transformation of breweries into biorefineries capable of turning BSG into high-value, low-perishable ingredients; the extensive use of BSGs to increase the production of food commodities; and the study of food formulations marketable with health claims.

**Keywords:** brewers' spent grain; bread-making; circular economy; common wheat; dietary fibre; durum wheat; emmer; phenolics; proteins; sustainable food production

#### **1. Introduction**

Topics related to a circular and sustainable economy are gaining more and more attention as means to combat climate change, reduce the use of fossil fuels and other natural resources, recover all products from resources without generating waste, and thus creating new opportunities for long-term economic growth [1]. In this perspective, the sustainable management of by-products and waste is an integral aspect of the circular economy.

Brewers' spent grain (BSG) is a lignocellulosic material representing the main byproduct generated from the brewing industry, since it corresponds to about 85% of the total solid residues [2]. It has been estimated that between 14 and 20 kg of wet BSG are obtained per 100 L of finished beer [3]. The average annual global production is estimated to be around 39 million tons, with ~3.4 million tons (which could become 8.5 be the end of 2030) produced in the European Union [4]. In Italy, a production of 188 thousand tons/year of BSG is estimated. Thirty percent of them are reused, mainly in the livestock sector. Furthermore, the BSG market value as cattle feed is very low, being around ~€35/ton [5]. The remaining 70% must be disposed of by breweries, with costs between 75 and 100 €/t (in the EU) and environmental impacts deriving from the release

**Citation:** Baiano, A.; la Gatta, B.; Rutigliano, M.; Fiore, A. Functional Bread Produced in a Circular Economy Perspective: The Use of Brewers' Spent Grain. *Foods* **2023**, *12*, 834. https://doi.org/10.3390/ foods12040834

Academic Editors: Donatella Bianca Maria Ficco and Grazia Maria Borrelli

Received: 30 January 2023 Revised: 11 February 2023 Accepted: 13 February 2023 Published: 15 February 2023

**Copyright:** © 2023 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 (https:// creativecommons.org/licenses/by/ 4.0/).

of 513 kg of CO2 equivalent of greenhouse gases per ton of landfilled BSG [6]. The research of Bolwig et al. [7] highlighted that an increasing number of farmers are refusing to use BSG as animal feed. This situation, together with the absence of on-site storage facilities, slows down the BSG disposal, sometimes forcing breweries to halt production.

The composition of BSG paves the way for the possibility of recovering and recycling them as secondary starting materials to produce a variety of value-added goods, included functional foods. Although BSG composition depends on several factors including the type of grains (mainly malted barley and wheat, but also unmalted grains and other starchy raw materials) and the malting and mashing methods [8], it includes on average: hemicellulose (~20%), cellulose (15.2–28.7%), lignin (3.35–21%), proteins (18.5–24.7%), lipids (8.4%), starch (5.3%), and ash (3.7%) [9]. From a nutritional standpoint, the interest towards BSG is mainly due to its content in phenolic compounds and soluble and insoluble fibre components such as β-glucans, and the less known arabinoxylans. Since most of the phenolic compounds of cereals are contained in the husk, BSG is rich in valuable antioxidant compounds such as phenolic acids (both benzoic and cinnamic acid derivatives), flavonoids, tannins, proanthocyanidins, and amino-phenolics and, for this reason, can be considered a suitable target for exploitation as a health promoting food supplement [10,11]. β-glucans and arabinoxylans are known to exert physiological functions in the gastro-intestinal tract. In particular, β-glucans are implicated in lowering plasma cholesterol and reducing glycaemic index and the risk of colon cancer, while arabinoxylans exert positive effects such as the production of short-chain fatty acids, the reduction of serum cholesterol, the enhancement of calcium and magnesium adsorption and, because of their esterification with hydroxycinnamic acids, they may also have some antioxidant properties [12]. Regardless of their health effects, the incorporation of arabinoxylans or β-glucans into wheat flour can also negatively influence bread-making. Loaf volume is depressed as a consequence of the effects of gluten dilution; the disruption of the gluten network related to the presence of fibres; the ability of β-glucans to bind water thus reducing its availability for the gluten network development; and, consequently, the gas holding capacity [13,14]. Although the use of BSG in bread-making has been extensively investigated, most research has been focused on the effects of the level of addition of spent grains on the characteristics of doughs and breads [15–18]. Topics related to the effects of nature (composition and brewing style) and geographical origin of BSG on concentrations of functional compounds in both spent grains and final products have been thoroughly investigated in very few studies and generally not together [19,20].

Our work was aimed at investigating the feasibility of using BSGs from the brewing of Belgian style white beers as functional ingredients in bread-making from a perspective of the sustainable use of natural resources, the reduction of waste disposal, and the reuse of by-products. BSGs were firstly characterised in order to evaluate their content in nutraceutical compounds as a function of both formulation (three different mixtures of malted barley and unmalted cereals) and geographical origin (two different places of cereal cultivation) of the starting brewing materials. To our knowledge, this is the first time that such an approach has been addressed. Successively, the bread samples obtained by replacing the Manitoba wheat flour with two different percentages of each BSG flour were submitted to physical, chemical, image, and sensory analyses with the aim of evaluating the effects of such replacements on their overall quality and functional characteristics.

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

#### *2.1. Production of Cereals Used in Brewing*

Barley malt cv. Fortuna was supplied by Agroalimentare Sud (Melfi, Potenza, Italy). The unmalted cereals, i.e., durum wheat (*Triticum durum*) cv. Dauno III, soft wheat (*Triticum aestivum*) cv. Risciola, and emmer (*Triticum dicoccum*) were produced from seeds supplied by CREA-CI Research Centre for Cereal and Industrial Crops (Foggia, Italy) cultivated under the same agricultural practices in two areas of Puglia region, namely Daunia (Soc. Cooperativa Agricola Valleverde, Bovino, Foggia, Italy) and Salento (Birra Salento

Societa' Semplice Agricola Di Leo Consolata and Co, Leverano, Lecce, Italy). The two areas strongly differ for geographical position and climate as follows:

Bovino: Height above sea level: 620 m; Latitude 41.250944 N; Longitude 15.342204 E; Climate zone: E; Degree days: 2243;

Leverano: Height above sea level: 37 m; Latitude 40.289737 N; Longitude 18.001176 E; Climate zone: C; Degree days: 1197.

#### *2.2. Brewing Procedures and Obtainment of BSG Flour*

BSGs were collected after mashing during brewing of craft Belgian style white beers performed in a 30 L Braumeister system (Speidel Tank-und Behälterbau GmbH, Ofterdingen, Germany) as follows: cereal mixtures made of 65% barley malt and 35% unmalted cereals were crushed with a 2-roller mill (Albrigi Luigi, Stallavena, Verona, Italy) under mill gaps of 0.5 ± 0.1 mm; water-to-cereal ratio was set to 20:5 (*v*/*w*); mash-in temperature was 52 ◦C, followed by a 13 min stand at 55 ◦C, a 35 min stand at 64 ◦C, a 20 min stand at 72 ◦C, and a final mash-off at 78 ◦C for 15 min; 7 L of sparge water at 80 ◦C was passed through the grain bed. Table 1 lists the resulting six types of BSGs differing for type and geographical origin of the unmalted cereals in the mixtures.

**Table 1.** BSG composition (%) and geographical origin.


Since freezing techniques are not appropriate to preserve BSG quality as they affect its arabinose content [21], after separation of wort, BSG samples were dried at 20 ◦C for 24 h through a Forced-air-drying system. The dried BSGs were ground (particle size distribution: <15% higher than 500 μm; 35–45% between 500 and 250 μm; 30–40% between 250 and 125 μm; <15% between 125 and 63 μm; <1% lower than 63 μm) with a blade mill, taking care to keep the temperature below 35 ◦C. The resulting BSG flours were sealed in polyethylene bags and stored at –20 ◦C until further use and analysis.

The following ingredients were used in bread formulations: BSGs, Manitoba soft wheat flour type 0 (COOP, Casalecchio di Reno, Italy), water, extra-virgin olive oil (Pazienza, Foggia, Italy), gluten (Elgranero, Madrid, Spain), sodium chloride (Compagni Europea Sali, Margherita di Savoia, Italy), and dehydrated *Saccharomyces cerevisiae* yeast (Cameo, Desenzano del Garda, Italy).

#### *2.3. Formulations and Bread-Making*

Thirteen types of bread were produced according to the formulations reported in Table 2: a control, made of 100% Manitoba soft wheat flour type 0 (Figure 1a); and twelve samples, obtained by replacing Manitoba flour with two different amounts (20 and 25%) of each of the six BSG flours (Figure 1b). When 20% of Manitoba flour was replaced, gluten was also added in an amount corresponding to 3% of the total weight of flours (BSG + Manitoba), while at 25% of flour substitution, a corresponding 4% of gluten was added, always referred to in the total amount of flours. Relying on the results of preliminary tests, we decided to use the same water amount for all formulations, since it was sufficient to ensure the right level of hydration and the correct development of the dough without changing the production times. For each type of breads, three technological replicates were performed.


**Table 2.** Ingredients (g) used in bread formulations.

<sup>1</sup> The percentage of replacement of Manitoba flour with BSG is reported in brackets. <sup>2</sup> The percentage of gluten reported in brackets is referred to the total amount of flours (Manitoba + BSG).

**Figure 1.** Slices of (**a**) control bread and (**b**) the 12 functional breads produced.

Loaves of regular shape were produced using a mould of 20 cm length and 11 cm width. In order to standardise the production, the entire procedure was performed in a bread-making machine (Zero-Glu Pro, Imetec, Azzano S. Paolo, Italy) using the program number 14 (total duration 3 h and 12 min) that includes the following steps: kneading, 22 min; 1st leavening, 40 min; 1st stirring, 5 s; 2nd leavening, 73 min; 2nd stirring, 5 s; 3rd leavening, 50 min; cooking, 47 min.

#### *2.4. Analysis of BSGs*

Moisture and ash contents were determined according to the AACC Methods 44-15.02 and 08-01.01, respectively, and expressed as % [22].

BSG proteins were extracted and analysed through electrophoretic and chromatographic techniques. Total proteins from BSG samples (~1 g) were extracted for 3 h with 10 mL of a buffer containing Tris-HCl 0.0625 M (pH 6.8), SDS 2%, and glycerol 10% (*v*/*v*) and then the extracts were centrifuged at 13,000× *g* for 10 min at 25 ◦C. Supernatants were carefully removed and stored at −20 ◦C until the analysis. Protein content was determined using Quick Start™ Bradford Protein Assay (Bio-Rad, Hercules, CA, USA), according to the supplier instructions. Electrophoretic separation (SDS-PAGE) was performed through a Mini-PROTEAN Tetra system electrophoresis cell (Bio-Rad, Hercules, CA, USA), according to la Gatta et al. [23] as follows: loading of 10 μg of sample onto the gel; running buffer consisting of Tris 0.025 M, Glycine 20 mM, 1% SDS; application of a potential of 200 V for one hour; use of a Prestained SDS-PAGE standard (Bio-Rad, Richmond, CA, USA). The gels were stained with a 0.25% (*w*/*v*) solution of Coomassie Brilliant Blue G-250 (CBB), fixed with a 7% (*v*/*v*) solution of acetic acid and 40% (*v*/*v*) of methanol, and destained with water. The polymeric components of BSG proteins were analysed using a two-step extraction procedure. An aliquot of each sample (20 μL) was analysed according to la Gatta et al. [23] using a Biosep SEC-S4000 column (300 × 7.8 mm, Phenomenex, Torrence, CA, USA). The fractions of total unextractable polymeric proteins (tUPP%) and of large unextractable polymeric proteins (lUPP%) were calculated according to Kuktaite et al. [24]. In addition, the incidence of unextractable (both monomeric and polymeric forms) proteins on the total protein content was also calculated (UP%).

The extraction of total phenolics was performed according to Gandolpho et al. [25] with some modifications. Briefly, 1 g of each BSG was suspended in a hydroalcoholic solution (ethanol 58%, solid-to-liquid ratio 1 g to 30 mL) and treated in an ultrasonic bath according to the following conditions: 37 kHz, 30 ◦C, 30 min. After extraction, the solutions were centrifuged (2000× *g*, 25 min, 20 ◦C) and the supernatants were removed and filtered through nylon filter (0.45 μm). The filtered extracts were analysed for their phenolic content and antioxidant activity. TPC was determined using the Folin–Ciocalteu reagent [26] and expressed as mg of gallic acid/g of dry matter. The phenolic profile was analysed by a HPLC system equipped with a diode array detector (Agilent 1100 Liquid Chromatograph, Santa Clara, CA, USA) according to Aliakbarian et al. [27] using a 100 × 4.6 mm × 3 μm RP-C18 Gemini column (Phenomenex, Aschaffenburg, Germany). The wavelengths used were 250, 280, and 320 nm. Identification of phenolics was made taking into account their retention times and comparing their spectra with those of standard materials while their quantification (mg/g dm) was obtained by comparing their peak areas with those of standard curves. The antioxidant activity was measured using 2,2-diphenyl-1-picrylhydrazyl (DPPH) free radical scavenging [28] and results were expressed as mg of Trolox (6-hydroxy-2,5,7,8-tetramethylchroman2-carboxylic acid) per g of dry matter.

Soluble and insoluble dietary fibres were determined according to AACC Methods 32-05.01 and 32-21.01 [22] using the K-TDFR-200A Megazyme kits (Neogen Europe Ltd., Ayr, Scotland) and expressed as %.

#### *2.5. Analysis of Functional Bread*

Previously ground bread slices were submitted to the determination of moisture, ash, TPC, antioxidant activity, phenolic profile, and dietary fibre, and to the protein characterization as already described for BSG in Section 2.3.

In order to evaluate the chromatic and structural characteristics of the bread samples, the images of the slices were acquired with an Epson scanner (mod. XP-3100, Cinisello Balsamo, Italy) at a resolution of 1200 dpi and saved in the tiff format. A ruler was placed next to the samples in order to convert pixels to centimetres through the measurement tools of the image analysis software. The image processing was performed with ImageJ ver. 1.52a, a free public domain software developed at the National Institutes of Health (USA). The crust and crumb colours were expressed according to the coordinates of the colour space defined by the International Commission on Illumination: L\*, lightness/brightness from black to white on a scale of zero to 100 and a\* and b\*, which represent chromaticity with no specific numeric limits. Negative and positive a\* values correspond with green and red, respectively, while negative and positive b\* values correspond with blue and yellow, respectively. The following structural characteristics were determined: slice height; crumb specific volume (cm3/g); number of pore/cm2; average pore size (mm2); porosity %, i.e., the surface of the slice occupied by pores; pore circularity, calculated as 4π\*area/perimeter2 (it ranges between 1.0 and 0.0, with a value of 1.0 indicating a perfect circle and value near to 0.0 indicating increasingly elongated shapes).

A panel of 10 trained judges (5 females and 5 males) between 20 and 65 years of age, experienced in the sensory evaluation of baked foods, carried out a Quantitative Descriptive Analysis (QDA) in a sensory laboratory equipped with booths at 23 ± 1 ◦C, under a white light. All samples were assessed 2 h after baking. A slice of each bread sample was randomly labelled with a three-digit numeric code and provided to assessors in a double-blind presentation to avoid any expectation error. The attributes were selected among those found in literature and generated by the same panel so as to give an overall description of the products without overlapping. As a result of this selection, the profile sheet used by panellists for the QDA of bread samples included 5 visual (for crust: colour and thickness; for crumb: colour, pore size, and development), 6 olfactory (on crust and crumb together: overall, freshly baked bread, wheat, and malty smell; toasty smell on crust; yeast smell on crumb), 4 gustatory (sweetness, saltiness, and acidity/sourness of crumb; crust bitterness), and 6 tactile (for crust: hardness and crispiness; for crumb: resistance to chewing, cohesiveness, graininess/fibrousness, stickiness) attributes to evaluate on numeric category scales. Unipolar category scales were used for all the attributes with the exception of tactile attributes, which were evaluated on bipolar category scales. The attribute definitions were retrieved from Callejo [29] and supplied to the panellists that were also asked to evaluate the overall quality of each bread, i.e., the comprehensive likeness/dislikeness of the sample expressed considering all the sensory attributes previously evaluated [30]. The panellists rated the intensity of each parameter from 0 (minimum scale) to 9 (maximum scale). Judges were instructed to rinse their mouths between samples with natural water. To prevent sensory fatigue, the samples were divided into two subsets, one completely evaluated in the morning tasting session and another one completely evaluated in the session of the afternoon of the same day.

#### *2.6. Statistical Analysis*

Each analysis was replicated at least three times, with the exception of the chromatographic analyses of proteins, which were performed in duplicates, and the image analysis, with five acquisitions for each sample. The averages and the standard deviations were calculated. Analysis of Variance (ANOVA) and LSD test were applied (*p*-value < 0.01) to study the single and interactive effects of type of BSG and geographical origin (of the unmalted cereal in the mixtures used in brewing) on the characteristics of BSGs. The same statistical analyses were applied to evaluate the single and interactive effects of type of BSG, geographical origin (of the unmalted cereal in the mixtures used in brewing), and amounts of BSG-gluten used in the formulation of the characteristics of the functional breads. Principal Component Analysis (PCA) was applied to evaluate the possibility of homogeneously grouping both BSGs and breads according to the experimental data. Pearson correlation coefficients at *p*-value < 0.01 was applied to individuate significant correlations among pairs of characteristics of bread samples. The statistical analyses were carried out using Excel software V. 14.0.0 for Mac and Statistica for Windows V. 7.0. (Statsoft, Tulsa, OK, USA).

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

*3.1. BSGs*

3.1.1. Physicochemical Characteristics of BSGs

Moisture was determined to check that drying was successful and to verify that the amount of water indicated in the bread formulations (Table 2) ensured the right dough hydration. According to the results of Table 3, neither type of BSGs nor their origin significantly affected the residual moisture. The mean moistures ranged from 2.9 to 3.5%, thus ensuring the inhibition of microbial growth [31]. In agreement with the findings of previous studies, BSG is a good source of ash with a great variability among samples depending on both the type and geographical origin of BSG [32,33]. The highest ash content was observed in Em spent grains coming from Daunia area while the lowest concentrations were detected in **Ri** samples (Table 3). The differences among BSG types depended on the use of not de-hulled emmer while the influence exerted by the geographical origin was due to the substantial differences in soil and climate conditions between Daunian sub-Apennines (D) and Salento peninsula (S), being the same agricultural practices adopted [34,35].

**Table 3.** Single and interactive effects of type of BSG and geographical origin of the cereal mixtures used in brewing on physical and chemical characteristics of BSGs.


In column, different letters indicate significant differences at *p* < 0.01; \* *p* < 0.01; n.s.: not significant.

BSG showed the common chromatographic profiles (Figure 2) of SDS-extractable and SDS-unextractable cereal proteins with the elution of four main peaks between 7 and 14 min: peak one, related to the elution of Large Polymeric Proteins (LPP), peak two related to the elution of Small Polymeric Proteins (SPP), peak three related to the elution of Large Monomeric proteins (LMP), and peak four related to the elution of Small Monomeric Proteins (SMP) [23,24]. The importance of studying the size distribution of polymeric proteins is due to its influence on the technological properties of flour [36]. Since UPP% of wheat cultivars is affected by both genetic and environment, this index has been suggested as an effective evaluation parameter in breeding programs of locally adapted wheats intended for bread-making, even considering the effects of environmental factors such as temperature, and nitrogen application and timing [37–39]. In agreement with the finding of a previous work [36], the proportion of the total unextractable polymeric proteins (tUPP%) in BSG samples was significantly affected by genotype and environment, whether considered individually or not. In addition, our results highlighted that the interactions of the two factors was also statistically significant, and that some considerations can be extended to the fate of the large unextractable polymeric fraction (lUPP%). The highest tUPP% and lUPP% were quantified in Em spent grains from Salento while the lowest values were found in Da spent grains from Daunia. Regarding the effects of BSG type, a previous research already demonstrated a significantly higher aggregation of emmer gluten proteins with respect to that of durum wheat, which was due to differences in the amino acid sequences between the glutenin subunits of the two cereals [23]. As can be inferred from Table 3, Da-D, Da-S, and Fa-S showed a percentage of total unextractable polymeric proteins higher than that of the large unextractable polymeric proteins, while Ri-D, Ri-S, and Fa-D showed the opposite behaviour. These results would suggest a different polymeric protein distribution between the two groups of samples, with the second group characterised by the presence of high molecular weight aggregates. These findings are in agreement with the results obtained by Johansoon et al. [40], who observed that the protein size distribution is affected by genotype and, for the same cultivar, by weather conditions. The percentage of unextractable (both monomeric and polymeric forms) proteins was also calculated. The highest and the lowest UP% were quantified in Em and Da spent grains, respectively, while any differences could be attributed to geographical origin. This independence of environment would indicate UP% as a more useful tool than total and large UPP% in screening genotypes suitable for bread-making.

**Figure 2.** Chromatographic profiles of SDS-Extractable and SDS-Unextractable proteins of Em-D spent grains: peak 1, Large Polymeric Proteins (LPP); peak 2, Small Polymeric Proteins (SPP); peak 3 and 4, Large Monomeric and Small Monomeric Proteins (LMP + SMP).

During brewing most of the water-extractable and soluble compounds are released into the wort, so BSG are mainly constituted of water-unextractable or insoluble compounds (cellulose and non-cellulosic polysaccharides) and the few water-extractable and soluble compounds (feruloylated arabinoxylans with a wide range of molecular mass) are entrapped in the complex matrix made of cellulose, protein, and lignin [41]. In our work, the percentage of insoluble dietary fibre ranged between 21.58% and 34.44% and both interactive and single effects of BSG type and geographical origin were significant. The highest IDF% were detected in Em, followed by Ri, and Da. BSGs from Salento generally had higher insoluble fibres than the corresponding samples from Daunia. The interactive and single effects of the considered factors were statistically significant for soluble dietary fibres as well. The percentages of SDF were very low in all brewers' spent grains, ranging from 0.42% to 2.06%. The lowest percentages of soluble fibres were detected in Da samples too, while Ri spent grains had the highest SDF%. The percentages of soluble fibre on the total fibre ranged from 1.89% (Da-S) to 7.04% (Ri-D). Concerning the influence of geographical origin, the soluble fibres showed a behaviour opposite to that of the insoluble fraction, with the highest percentages of SDF detected in samples from Daunia. The amounts of IDF and SDF were consistent with those found by Shih et al. [42] in BSGs from India Pale Ale brewers (27.94–48.93% for IDF; 1.44–2.10% for SDF). The total dietary fibre was between 21.99% of Da-S and 35.34% of EmS, resulting within the ranges found in the literature

(28.22–42.6%) [43,44]. The highest IDF% and TDF% of Em spent grains strongly depended on the use of the non-dehulled emmer while the differences between Ri and Da were determined by genotype [45].

#### 3.1.2. Phenolic Concentration and Antioxidant Activity of BSGs

The interest towards phenolic content and the profile of BSG is due to the bioactive and antioxidant effects of these compounds and to their contribution, together with fibres, to the functional effects of bread and other cereal derivatives obtained with a partial replacement of wheat flour with nutraceutical ingredients. Nevertheless, it must be pointed out there is still no scientific evidence to support health claims concerning the phenolic compounds of cereals, as a consequence of the co-occurrence of other bioactive compounds (just like fibre in wholegrain products) that hinders verification of any health benefits exerted by these compounds [46]. During the mashing and sparging steps, phenolic compounds are released from the starting cereal mixtures (where they accumulate mainly in bran fractions) to wort but significant amounts remain in the resulting BSG, mainly bound to cell walls as polysaccharides or proteins but also, in smaller amounts, as free compounds. The mean total phenolic content of BSG, comprised between 3.114 and 4.868 mg/g d.m. was significantly influenced by the type of BSG with the highest and the lowest phenolic contents detected in Em and Ri spent grains, respectively. The effects of the place of origin were not statistically significant (Table 4). Since the six starting cereal mixtures were formulated by adding the same barley malt to a constant amount (35%) of each unmalted cereal, the influence exerted by the type of BSG on its phenolic content must be unquestionably attributed to the species and variety of the unmalted grains [47] or, in other words, to their specific phenolic profile and the distribution of free, conjugated, and bond forms. This idea is reinforced by the finding that different phenolic compounds are released in different amounts during mashing. In fact, according to Langos and Granvogl [48], ferulic and *p*-coumaric acids are released in higher extension (up to 9-fold) compared with cinnamic acid and, as inferred by Schwarz et al. [49], the optimal pH ranges for the extraction of ferulic acid, *p*-coumaric acid, and cinnamic acid are different, being 5.4–6.6, 5.8–6.0, and 6.8, respectively.

The comparison of our results with the findings of other researchers is not simple since the latter generally concerns BSGs deriving from a mixture of various malts (instead of mixtures of malted and unmalted grains as in our work) or spent grains of unknown origin and nature [15,50]. Nevertheless, TPC of our BSG samples were within the range of concentrations found by Birsan et al. [51] in light and darker BSGs (from 3.01 ± 0.19 to 4.71 ± 0.28 mg/g d.m.) but much greater than those detected by Farcas et al. [20] in 100% malt BSGs (from 0.36 to 2.79 mg/g d.m.).

The mean antioxidant activity, ranging from 0.64 mgTrolox/g d.m. of Em-D to 2.89 mgTrolox/g d.m. of Ri-S, was significantly influenced by the type of BSG with the highest value measured on **Ri**. The effects of the place of origin were not statistically significant (Table 4). Our data confirmed that TPC and antioxidant activity were poorly correlated (R2 0.4997, *p* < 0.01); regardless, phenolic compounds are the most important group of antioxidants [52].

Regarding the phenolic profiles (Table 4), eleven compounds were detected in all the BSGs: six phenolic acids (gallic, 4-hydroxybenzoic, vanillic, caffeic, ferulic, *p*-coumaric); two flavanols (catechin and epicatechin); two flavonols (quercetin and kaempferol); and one hydroxystilbene (resveratrol). Kaempferol (0.795–1.018 mg/g d.m.), epicatechin (0.160–0.168 mg/g d.m.), and caffeic acid (0.014–0.178 mg/g d.m.) were the main phenolics. The other compounds had concentrations always lower than 0.050 mg/g d.m. Regarding the single effects exerted by the type of BSGs on their phenolic profiles, Em showed the highest amounts of all phenolic compounds except gallic acid, detected in higher concentrations in Ri, and vanillic acid, whose concentrations was the same in all BSGs. The single effects of geographical origin was not statistically significant for vanillic, caffeic, and *p*-coumaric acids. The highest amounts of other compounds were retrieved in BSGs from Daunia.


 activity of

**Table 4.** Effects of type of BSG and

geographical

 origin of the cereal mixtures used in brewing on phenolic contents and antioxidant

#### *3.2. Functional Breads*

#### 3.2.1. Physicochemical Characteristics of the Functional Breads

Colour, moisture, ash, and the protein distribution of breads enriched with BSGs are reported in Table 5 and compared with the characteristics of the control bread.

Regarding colour indices, the crust of the control bread showed the highest L\* (together with Em-D breads) and b\* (with Da-S-20G3) values and the lowest a\* indices (with Da-D-25G4, Da-S-20G3, and Em-D-20G3). Em-S-25G4 had the lowest L\* and b\* values, while the highest a\* was measured on Da-S-25G4. The crumb of the control bread also showed the highest L\* and b\* (with Ri-D-20G3) values and the lowest a\* indices.

The type of BSG exerted significant effects on all colour indices. Concerning crust, the highest brightness and yellow index were measured on Da samples and the lowest red index was detected in Ri breads. The crumb of Ri breads was the darkest, the reddest (with Da samples), and showed the lowest yellow index (together with Em). The different behaviour observed for the various types of BSG was correlated to the colour of the starting spent grains.

The level of BSG-gluten addition significantly affected the colorimetric indices of bread with the exception of the crust a\* values, but the effects were opposite between crust and crumb. The increase in BSG-gluten content generally decreased the L\* and b\* values of the crust, i.e., determined the formation of a darker colour. This behaviour is explained by the higher protein content in the formulation which caused intensive Maillard reactions [53]. Contrary to what happened to the crust, the increase in BSG-gluten content corresponded to a slightly lighter crumb (due to the clear colour of the fibre added) but also to a more intense golden colour (increased a\* and b\* values). In most previous works, a darker crumb was observed as a consequence of BSG addition [15] but Gómez et al. [54] pointed out that the effects depend on the colour of fibres contained in the added ingredient.

Remarkable water content was quantified in the control bread. However, other bread samples showed higher crust moisture (Em-D-20G3) and similar amounts of water in crumb (Em-D-25G4). The type of BSG did not affect crust moisture while the crumb of breads produced with Da spent grains contained generally higher water in their composition. The crumb moisture was significantly and positively affected by the BSG-gluten level, and this behaviour may be attributed to the higher fibre and protein contribution [17].

Regarding ash, the lowest and the highest amounts were quantified in control (3.1%) and Em-S-25G4 (3.6%) breads, respectively. The addition of all the BSG types significantly increased bread ash content but, consistently with the BSG composition reported in Table 3, the greatest ash amount was contributed by the Em type. Since BSG is mainly composed of the husk of grain and minerals are present in a greater amount in their outer layer, the amount of BSG-gluten added also had a significant and positive effect on the bread ash content [17,55].

Concerning protein size distribution in bread samples, the chromatographic profiles highlighted the presence of only two main peaks, eluted between 10 and 12.5 min (i.e., the range of large and small monomeric proteins) and the total lack of peaks in the elution range of polymeric protein aggregates. This result could be due to a variety of factors including technological process; BSG diluting effect on the protein network; and possible accumulation of low-molecular weight metabolites (mainly glutathione) deriving from the lysed yeast cell. Glutathione (GSH) was found to be responsible for the chemical modification of the gluten protein structure, leading to its depolymerization [56] and the formation of lower molecular weight gluten proteins [57,58]. Consequently to these changes, unextractable proteins included only monomeric forms (Table 5). From the comparison of data in Tables 3 and 5, a remarkable increase in UP% from BSGs to the corresponding breads can be inferred. It could be due to baking, whose high temperatures are known to induce interactions among different class of proteins, thus leading to the formation of larger aggregates or a supramolecular structure. The average unextractable monomeric proteins ranged from a minimum of 35.91% in Em-S-20G3 to a maximum of 42.07% in Ri-S-20G3. Percentages higher than 40% were also detected in Ri-D-25G4, Ri-S-25G4, and the control bread. Significant single effects of BSG type were observed, with UP% higher in Ri breads and lower in Em breads.


 column, different letters indicate significant differences at *p* < 0.01; \* *p* < 0.01; n.s.: not significant.

D

S

Significance

20 G3 25 G4 Significance

 55.7 b

 53.4 a

 \*

 n.s. In

18.1

18.2

 41.1 b

 39.5 a

 \*

 \*

 53.8

 55.3

 n.s.

 18.4

 18.0

 n.s.

 40.4

 41.1

 n.s.

Single Effect (Geographical Origin of Cereal Mixtures)

 55.6

 54.8

 n.s. Single Effect (Amount BSG-gluten)

> 54.0 a

56.4 b

7.7 a

9.0 b

\*

32.6 a

34.0 b

 \*

 n.s.

9.6

9.6

 20.1 a

 22.4 b

 \*

3.3 a

3.4 b

 \*

 n.s.

38.92

38.75

 8.4

 8.3

 n.s.

 33.5

 33.2

 n.s.

 10.3

 9.3

 n.s.

 21.4

 21.0

 n.s.

 3.4

 3.4

 n.s.

 38.32

 39.35

 n.s.

The geographical origin of the starting cereal mixtures did not significantly influence any of the parameters of Table 5.

#### 3.2.2. Phenolic Concentration, Antioxidant Activity, and Fibre Content of the Functional Breads

The amounts of antioxidants and fibres are among the most important factors to consider in assessing the nutraceutical quality of breads.

The total phenolic contents of functional breads were always higher than those of control bread (1.555 ± 0.158 mg/g d.m.) and comprised between 1.793 ± 0.183 mg/g d.m. (Ri-D-20G3) and 2.833 ± 0.772 mg/g d.m. (Da-S-25G4) (Table 6). These were the results of the interactions among the type of BSG, the geographical origin of cereal mixtures, and the amount of BSG-gluten, although the single effects of these variables were also statistically significant (with the exception of the geographical origin). In particular, the highest TPCs were detected in Em breads, consistently with the high phenolic concentrations of the starting spent grains. Moreover, the bread antioxidant content increased with the increase in BSG-gluten amount in the bread formulation. Finally, it is appropriate to consider that the phenolic content of the final breads was also affected by bread-making and, although the same process was applied for all samples, the magnitude of these changes could not be the same. According to a recent research of Tian et al. [59], bread-making generally improved the potential health benefits of whole wheat products. The authors pointed out that fermentation and baking generally increased soluble phenolic content and its antioxidant activity due to the contribution of Maillard reaction products, and that those steps only slightly increased the insoluble phenolic fraction and its antioxidant activity. The TPCs of our experimental functional breads were considerable higher than those (0.47 ± 0.06 mg/g d.m.) retrieved in the recent literature [60].

The antioxidant activity values of the functional breads showed trends similar to those of TPCs. They were always higher than those of the control bread (0.36 ± 0.02 mg Trolox/g d.m.) and comprised between 0.67 ± 0.16 mg/g d.m. (Ri-D-20G3) and 3.45 ± 0.47 mg/g d.m. (Em-D-25G4) (Table 6). The single effects of the type of BSG and the amount of BSG-gluten were statistically significant. In particular, the highest and the lowest antioxidant activity values were detected in breads produced with Em and Ri spent grains, respectively. Furthermore, the bread antioxidant activity increased with the increase in BSG-gluten amount in the bread formulation. TPC and antioxidant activity showed a strong correlation (R = 0.885, *p*-value < 0.01).

The bread phenolic profiles were simpler than those of the starting BSGs. Five phenolic compounds were identified in all the functional breads, including four phenolic acids (gallic, vanillic, caffeic, and sinapic) and a flavanol (epicagallotechin gallate). Nevertheless, their concentrations depended on BSG type, the geographical origin of the stating cereal mixtures, and BSG-gluten amount. *p*-Coumaric acid was detected only in the control bread that, instead, did not contain phenolics such as epigallocatechin gallate, and vanillic, caffeic, and sinapic acids. The interactive effects of the three factors were also statistically significant. The type of BSG showed significant single effects, with the highest concentrations of almost all compounds detected in Em breads. The higher level of BSG-gluten supplementation allowed to obtain breads with a higher content of almost all compounds. The single effects of the geographical origin of the starting mixtures were less significant.



The interest towards the fibre content of BSG-enriched bread is due to the possibility, established by Reg. (EU) N◦. 432/2012 [61], to use the following two health claims: "Barley grain fibre contributes to an increase in faecal bulk" for foods which are high in that fibre, i.e., for those foods that contain at least 6 g of fibre per 100 g or at least 3 g of fibre per 100 kcal; and "Beta-glucans contribute to the maintenance of normal blood cholesterol levels" for food which contains at least 1 g of beta-glucans from barley/barley bran per quantified portion. [62]. The increase of faecal bulk is related to the ingestion of insoluble dietary fibre. Other long-known benefits of dietary fibre intake include the modulation of glycaemic index and potential prebiotic capacity, which are known to be linked to arabinoxylan and arabinoxylan-oligosaccharides, the latter primarily deriving from wheat [63,64]. USDA recommends daily intakes of fibres equal to 25 g for women and 38 g for men up to 50 years old and to 21 and 30 g for elder women and men, respectively [65]. Since the actual intake is generally lower, especially in Western countries, the regular consumption of BSG-enriched breads could help consumers to meet such recommendations. As can be inferred from Table 6, the amounts of IDF and SDF in BSG- enriched breads were significantly lower than in the corresponding spent grains, due to the dilution effect of Manitoba flour. IDF% and SDF% felt both interactive and single effects of the factors. More in depth, IDF% ranged from 2.05% in the control bread to 7.15% of Em-S-25G4 while SDF% was in the 0.004% (Ri-S-25G4)–1.56% (Da-S-25G4) range and the percentages of soluble fibre on the total fibre were appreciably higher than in the BSG samples, ranging from 0.06% (RiS25-G4) to 22.51% (DaS25-G4), thus contributing beneficial effects that go beyond the simple increase in faecal bulk. A first reason for this behaviour is the high percentages of soluble fibre in the Manitoba flour. In fact, the control bread had a percentage of soluble fibre on the total fibres equal to 23.49%. Another reason is that, during bread-making, a decomposition of dietary fibre (first hemicellulose and afterwards cellulose) occurred, reducing the fibre molecular weight [66]. The reason for this degrading action could be both the yeast, since some *Saccharomyces cerevisiae* strains are able to produce cellulase and xylanase [66], and the first step of the baking process when both temperatures and moisture are elevated, thus simulating the conditions of an autoclave treatment [67]. In the control bread, the incidence of soluble fibre on the total amount of fibre was equal to 23.49%. However, while data indicate that the single effects exerted by BSG type and the geographical origin on bread IDF% were similar to those observed for spent grains, they also describe the opposite effects of BSG type and geographical origin on the soluble fibre fraction between breads and spent grains. Significant increases in both IDF% and SDF% were evaluated by increasing the amount of BSG-gluten added. The total dietary fibres ranged between 4.50% of RiD20-G3 and 7.85% of EmS25-G4, resulting as significantly higher than those quantified in the control bread (2.68%) and slightly higher than the ranges found in recent literature (3.32–6.37%) [44]. According to the data concerning the dietary fibre contents of the enriched breads, the inclusion of Em spent grains in percentages equal or higher than those used in these experiments could allow the use of one or both of the health claims mentioned above [61].

#### 3.2.3. Structural Characteristics of the Functional Breads

Consumers are more likely to purchase well-leavened and regularly shaped breads. For this reason, height and specific volume are often considered as the key quality parameters.

In our work, the minimum and maximum slice heights were used as indices of bread shape regularity. The highest values of the minimum (9.36 ± 0.24 cm) and maximum (10.14 ± 0.17 cm) height were measured on the control bread, while the lowest values were found for Da-S-20G3 (7.72 ± 0.84 cm) and Ri-D-25G4 (8.39 ± 0.20 cm), respectively (Table 7). As can be inferred from these data, slice height values were significantly affected by interactions among BSG type, origin, and the amount of BSG-gluten, but the BSG type also exerted significant single effects, with the highest values observed in Em breads. The crumb specific volume is another key parameter since superior bread quality is often characterised by high specific volumes [44]. Crumb specific volume (Table 7) was only slightly reduced

by the BSG addition, ranging from 2.11 ± 0.9 cm3/g of Da-D-20G3 to 2.64 cm3/g of Em-D-25G4 and control bread. These results were significantly better than those observed by Amoriello et al. [15] in breads produced with medium or strong wheat flour supplemented with 5 or 10% of BSG. In that work, the authors attributed the limited dough development of BSG-enriched breads to a reduction of extensibility and the gas-retention ability of gluten, in turn caused by dilution with non-gluten proteins and disruption due to the interference of fibres. The better specific volume of supplemented breads obtained in our work was determined by the concurrent addition of gluten. As always in our work, only the interactive effects of the three factors were significant, but it is interesting to point out that, although not significant, the highest specific volumes were quantified in Em breads, especially at increasing amounts of BSG-gluten added, i.e., in the samples with the highest fibre contents and deriving from the spent grains that had the highest amounts of total and large unextractable polymeric proteins.

Pores are created within the dough structure as a consequence of CO2 production during leavening. Their characteristics are described in Table 7. The control bread had a high pore density (0.73 ± 0.02 pores/cm2) but not the highest, since that index ranged from ~0.3 pores/cm<sup>2</sup> (Ri-S breads, Em-D-25G4, and Em-S-25G4) to ~0.82 pores/cm<sup>2</sup> (Em-D-20G3). This result was not in agreement with the findings of Neylon et al. [44], who observed a significant decrease in the number of cells in BSG-fortified breads. BSG type, origin, and the amount of BSG-gluten exerted statistically significant single and interactive effects, with the higher density measured on breads produced with Da-D spent grains at the lowest level of addition. The average pore size was inversely correlated (R = 0.938, *p*-value < 0.01) with pore density, ranging from 0.10 mm<sup>2</sup> (control bread and Da-S-25G4) to 0.30 mm2 (Em-D-25G4 and Ri-S-20G3). As for pore density, the results were not in agreement with the findings of Neylon et al. [44], who observed a decrease in the cell diameter in BSG-fortified breads. Only the BSG type showed a significant effect on pore size, with the lowest values observed in Da breads. Porosity % ranged from ~35% (control bread, Da-D-20G3, and Ri-D-25G4) to ~47% (Em-S-25G4) and, as for the average pore size, only the BSG type showed a significant effect, with the highest values observed in Em breads. According to these results, the crumb of Em breads had a less compact structure than those of all other breads. Porosity % was well correlated with the total pore surface (pore density × verage pore size), showing an R value of 0.694 (*p*-value < 0.01). Circularity ranged from 0.774 ± 0.029 (Ri-S-20G3) to 0.829 ± 0.029 (Em-S-25G4), thus indicating a predominantly circular shape. Nevertheless, data showed a remarkable variability already within the samples of each type of bread and only the interactive effects of the three factors were statistically significant, making it difficult to understand the weight of each independent variable.

#### 3.2.4. Sensorial Characteristics of the Functional Breads

Control bread and breads enriched with the BSGs were evaluated by a trained panel through a Quantitative Descriptive Analysis and the results are reported in Table 8. Two of the characteristics that judges were requested to evaluate, namely crust bitterness and crumb stickiness, were not detected for any of the experimental breads. The interactive effects of BSG type, the origin of cereal mixtures, and the amount of BSG-gluten added were significant for all the other sensorial parameters.


**Table 7.** Effects of type of BSG,

geographical

 origin, and amounts of BSG-gluten

 in the formulation

 on structural

characteristics

 of the functional

In column, different letters indicate significant differences at *p* < 0.01; \* *p* < 0.01; n.s.: not significant.



20 G3

25 G4 Significance

 n.s.

 n.s.

 n.s.

 n.s. In column, different letters indicate significant differences at *p* < 0.01; \* *p* < 0.01; n.s.: not significant.

 n.s.

 n.s.

 n.s.

 \*

 n.s.

 n.s.

 \*

 n.s.

 n.s.

 n.s.

 \*

 \*

 \*

 n.s.

 n.s.

 \*

 6.8

 7.2

 4.2

 6.8

 2.7

 6.7

 6.3

 3.0

 4.7

 6.2

 3.3

 6.3

 6.3

 3.2

 2.3 a

3.8 b

1.7

1.3

 1.5

 1.3

 1.5 a

2.7 b

5.8

6.2

 3.3

 3.2

 0.9

 1.0

 4.3 a

4.7 b

3.0 a

2.5 b

6.7

4.2 b

1.8 a

6.5

 2.7

 2.2

 6.3 a

6.7 b

Bread colour is the first variable evaluated by consumers and strongly affects their willingness to purchase and the product acceptability. In agreement with the findings of Ginindza et al. [68], the addition of BSG always resulted in a significant colour change compared with the control bread and the reason is that BSG contributed a higher amount of aminoacids, thus favouring the non-enzymatic browning reactions [69]. The Em breads were evaluated as the darkest. The single effects of geographical origin and the amount of BSG-gluten added were not significant either for crust or for crumb colour. Regarding the other visual characteristics, the type of BSG had significant single effects on crust thickness (higher in Da breads while control bread had intermediate scores), crumb pore size (higher in Ri breads while control bread had intermediate scores), and crumb development (higher in Em samples among the BSG-enriched breads but lower than that of the control bread).

The fortified breads generally showed the highest intensity of wheat, malt, and yeast flavours. According to Ktenioudaki et al. [16], these flavours were due to volatile compounds already present in BSG (2-heptane, butanal, 2-methylbutanale, benzene, and 2,3 butanedione) and arising from fermentation (ethanol, butanol, and other acholic compounds) and Maillard reactions (furfural, pyrazine). The type of BSG exerted significant effects on overall intensity, malty, and toasty flavours (lower in Em breads), and on wheat and yeast smells (higher in Da breads). The amount of BSG-gluten added exerted significant single effects only on wheat and yeast smells, which increased with the level of replacement.

Concerning crumb taste, the single effects of the three factors were not statistically significant. The saltiness and acidity/sourness of the functional breads were evaluated as lower or equal to those of the control bread. The latter obtained intermediate scores for sweetness, perhaps as a consequence of the maltose and glucose contained in BSG. Moreover, the sweet taste was also enhanced by the volatile compounds responsible for sweet and malty flavours.

Regarding bread texture, the addition of BSG-gluten always increased resistance to chewing and the fibrousness of the functional breads with respect to the control. The control bread obtained intermediate scores for crispness. The highest crust hardness and resistance to chewing and the lowest cohesiveness and fibrousness were attributed to breads fortified with Da spent grains. The amount of BSG-gluten added significantly affected the bread texture, except for crumb cohesiveness. The hardness of the crust and the resistance to chewing of crumb increased with BSG-gluten content. This behaviour could be explained by both the higher fibre and protein contents of functional breads, which caused a greater water absorption [17,70], and also the presence of pentopans, a fibre BSG component, responsible for the gluten protein cross linking [71]. Crust crispness decreased with the increase in BSG-gluten content due to the increase in water absorbed, while crumb cohesiveness and fibrousness did not show significant changes. These results were only partially in agreement with those of Yitayew et al. [17], who found that the hardness and breakage of crumb increased with BSG level.

Finally, the best and the worse overall ratings were attributed to Ri (together with control) and Em breads, respectively, and the panellists accorded their preferences to breads produced with the lowest amount of BSG-gluten added only for DaS and Em breads. In the other case, the % of BSG added did not affect the overall sensory quality. This is an interesting finding since the percentages of BSG in our products (20 or 25%) was much higher than those (≤10%) indicated as optimal by Yitayew et al. [17] and Ginindza et al. [68] who found the maximum and minimum acceptability score for the control sample and the bread supplemented with the highest amount of BSG, respectively. The interest of our finding is also due to the fact that, in other research, the BSG-fortified foods exerted a higher appeal [42] than the not-enriched counterparts as a consequence of consumer interest for health issues, characteristics that overshadows the product sensory properties [72].

#### 3.2.5. Principal Component Analysis and Pearson Coefficients

Principal Component Analysis was carried out to verify the possibility of grouping BSGs by type and/or by place of origin on the basis of their physical characteristics and the content of antioxidants, fibres, and non-extractable proteins. The ability of PCA to group BSG samples belonging to the same base malts group with a high percentage of explained variance (75%) was already highlighted by di Matteo et al. [73]. Figure 3 shows a biplot of Factors 1 and 2 that accounted for 67.63% of the variance in the whole data set. Regarding the projection of the BSG samples on the factorial plan (Figure 3a), Em samples differed from the others for their negative loading of Factor 1, accounting for 43.19% of the explained variance. Em samples were furtherly divided into two geographically homogeneous groups, one including spent grains from Daunia and another including spent grains from Salento, characterised by the negative and positive loading of Factor 2, respectively. The negative loadings of this factor were associated with high amounts of ash and most of the phenolic compounds while the positive loadings of these factors were associated with high amounts of insoluble fibres, unextractable proteins, *p*-coumaric acid, and epigallocatechingallate (Figure 3b). Ri-D and Ri-S spent grains were homogeneously grouped but close to each other in the upper right quadrant, the first with loadings near to 0, the latter with more positive loadings of Factor 1 (Figure 3a), characterised by high antioxidant activity values and low phenolic concentrations (Figure 2b). Da-D and Da-S spent grains were homogeneously grouped but close to each other in the lower right quadrant (Figure 3a). This quadrant was associated with low-to-intermediate values of all variables (Figure 3b).

Principal Component Analysis was also performed to highlight the relationship between bread samples and their chemical, physical, structural, and sensory characteristics. Figure 4a,b shows the projection on the factorial plan of breads and analytical information, respectively. Factors 1 and 2 only explained 44.21% of the variance in the whole data set and made it possible to group the thirteen types of samples into just three sets that stood out for their position in the factorial plan: one including the control bread, placed in the lower left quadrant and associated with high values of crumb development, specific volume, minimum and maximum height, cohesiveness, and amount of *p*-coumaric acid; another group, comprising all the Em breads distributed within the lower right quadrant and characterised by high concentrations of IDF, TPC, gallic acid, vanillic acid, sinapic acid, epigallocatechingallate, and high values of crispiness, porosity, fibrousness, and wheat smell; and the last group, including Ri and Da breads, which were partially overlapped and concentrated straddling the two upper quadrants, associated with high values of overall smell intensity, a toasty smell, pore size, crust thickness, overall quality, a high concentration of caffeic acid, a darker crumb colour, and intermediate TFC. As can be inferred from Figure 4a, PCA of the overall data set was not able to discriminate bread samples on the base of the two different percentages of the BSG-gluten used in bread-making. In the paper of Ktenioudaki et al. [16], PCA analysis of volatile compounds was able to clear separate snack samples in homogeneous clusters for the percentage of BSG added but it is appropriate to point out that they worked with a single type of BSG instead of three types as we did.

**Figure 3.** Projections of (**a**) BSG samples and (**b**) the related experimental data on the factorial plane. Phenolics are in blue font; Insoluble and Soluble Dietary Fibres are in red; tUPP%, lUPP%, and UP% are in purple; Moisture % and Ash % are in brown.

**Figure 4.** Projections of (**a**) control and functional breads and (**b**) the related experimental data on the factorial plane. Phenolics are in green font; Insoluble and Soluble Dietary Fibres are in purple; UP% is in dark green; Moisture %, Ash %, and colorimetric indices are in red; structural parameters are in black.

Pearson correlation coefficients were calculated to individuate correlations among the quality characteristics of breads and measured variables. For the sake of synthesis, only the main statistically significant correlations (*p* < 0.01) are discussed.

The percentage of unextractable proteins was positively correlated with the crumb pore size (+0.61) and negatively correlated with porosity (−0.56), sensory crust hardness (−0.53), and antioxidant activity (−0.60). A possible explanation of this behaviour is that the formation of larger protein aggregates made the crumb structure more compact and entrapped phenolic compounds through protein-polyphenol complexation while increasing the water holding capacity of the crust structure.

The insoluble dietary fibre content was positively correlated with ash content (+0.83), porosity (+0.47), and crumb fibrousness (+0.51) while the soluble dietary fibre content was positively correlated with crust thickness (+0.51) and crumb sweetness (+0.70) as a consequence of the depolymerization of hemicellulose and cellulose and the production of simple sugars. To confirm the significant effects of the addition of BSGs in terms of colour changes, IDF% resulted in a significant correlation with crust b\* (−0.54), crumb a\* (+0.58), and the colour of the crust (+0.79) and crumb (+0.62). However, contrary to what Ginindza et al. [68] highlighted, the darkening of the colour did not negatively affect the overall quality of the bread. IDF% as an index of the amount of BSG added also affected bread taste by reducing saltiness (−0.82) and olfactory characteristics, and enhancing the intensity of the wheat (+0.80) and yeast (+0.58) smell. Contrary to what Ginindza et al. [68] highlighted, the flavour changes caused by the addition of BSG were not correlated with the overall sensory quality of bread.

TPC was positively correlated with ash content (+0.66), IDF% (+0.74), antioxidant activity (+0.70), and with the individual concentration of most of the phenolic compounds, thus demonstrating that wholegrain derivatives can be considered good sources of phenolic antioxidants.

Further significant correlations were found between the physical, structural, and sensory characteristics of breads, as in the following cases: crumb specific volume and crumb development (+0.43); porosity and crumb pore size (−0.48); crust b\* and crust colour (−0.51); and crumb L\*, a\*, and b\* and crumb colour (−0.38, +0.68, +0.43).

Finally, the overall quality of breads was positively correlated with a malty smell (0.68) and crumb saltiness (0.70), and negatively correlated with ash and IDF% (−0.72 and −0.52), and with crumb fibrousness and porosity (−0.56 and −0.58). Our results were only partially in agreement with the findings of Ktenioudaki et al. [16], who found that taste and texture were the attributes that mostly affected the overall acceptability, and with the results of Combest and Warren [74], who found significant correlations only between taste and overall liking. In fact, in our study, some smell, taste, and texture attributes are correlated with the overall sensory quality of breads.

#### **4. Conclusions**

The partial replacement of wheat flour with BSG resulted in significant increases in phenolic content, and insoluble and soluble dietary fibres of the enriched breads with respect to the control thus confirming the nutraceutical and functional nature of BSGs and BSG-enriched breads, respectively. Concerning BSG samples, single and interactive effects of the type and the geographical origin of the starting cereal mixtures were highlighted, to point out that their composition is mainly determined by genetics but can be significantly influenced by environmental conditions.

The highest amounts of phenolic compounds were detected in Em spent grains, followed by Da, and Ri. Em spent grains also showed the highest concentrations of both insoluble and soluble dietary fibres, followed by Ri, and Da. The supplementation with the highest percentage of BSGs exerted a positive influence on the contents of phenolics and dietary fibres without worsening the structural and sensory attributes. These results were probably due to the simultaneous addition of gluten in amounts able to counterbalance the negative effects exerted by the fibres added on the gluten network. Moreover, the increase

in soluble fibres from BSGs to the corresponding breads may be partially responsible for the good structural and sensorial performance of the enriched breads compared with the control bread.

The breads enriched with Em spent grains showed the highest percentage of phenolics and insoluble fibres and are therefore the best breads in terms of the content of nutraceuticals. However, since these desirable characteristics were accompanied by a worsening of the overall sensory quality evaluated by a trained panel, the production of bread supplemented with Ri or Da spent grains represented the optimal choice, since their intermediate phenolic and fibre contents and their overall quality scores were comparable to that of the control bread.

As a practical application, the large amount of BSG produced annually in the world makes the transformation of breweries into biorefineries capable of turning this by-product into high-value, low-perishable ingredients for food and feed industries economically convenient. Since BSG composition depends on their type and origin, they should be offered for sale with labels showing the precise information on their composition, nutritional value, and the suggested percentages of use in the formulation of the finished products. On the other hand, food companies should explore the huge variety of uses of BSG, in particular to a) increase the production of bread and other food commodities to meet the needs of the growing world population, and b) study food formulations that can be labelled with health claims.

**Author Contributions:** Conceptualization, A.B.; methodology, A.B., A.F., B.l.G. and M.R.; software, A.B.; validation, A.B.; formal analysis, A.B.; investigation, A.B., A.F., B.l.G. and M.R.; resources, A.B.; data curation, A.B., A.F.; writing—original draft preparation, A.B.; writing—review and editing, A.B.; visualization, A.B.; supervision, A.B.; project administration, A.B.; funding acquisition, A.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** PSR PUGLIA 2014/2020—Misura 16—Cooperazione—Sottomisura 16.2—"Sostegno a progetti pilota e allo sviluppo di nuovi prodotti, pratiche, processi e tecnologie"—Project entitled "Dal Campo al Boccale" (BEˆ2R). Grant Number: DDS 94250036020.

**Data Availability Statement:** No new data were created or analyzed in this study. Data sharing is not applicable to this article.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

#### **Abbreviations**


#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

## *Article* **A Functional End-Use of Avocado (cv. Hass) Waste through Traditional Semolina Sourdough Bread Production**

**Enrico Viola 1, Carla Buzzanca 2, Ilenia Tinebra 1,\*, Luca Settanni 1, Vittorio Farina 1,3, Raimondo Gaglio <sup>1</sup> and Vita Di Stefano <sup>2</sup>**


**Abstract:** In recent years, a main goal of research has been to exploit waste from agribusiness industries as new sources of bioactive components, with a view to establishing a circular economy. Non-compliant avocado fruits, as well as avocado seeds and peels, are examples of promising raw materials due to their high nutritional yield and antioxidant profiles. This study aimed to recycle avocado food waste and by-products through dehydration to produce functional bread. For this purpose, dehydrated avocado was reduced to powder form, and bread was prepared with different percentages of the powder (5% and 10%) and compared with a control bread prepared with only semolina. The avocado pulp and by-products did not alter organoleptically after dehydration, and the milling did not affect the products' color and retained the avocado aroma. The firmness of the breads enriched with avocado powder increased due to the additional fat from the avocado, and alveolation decreased. The total phenolic content of the fortified breads was in the range of 2.408–2.656 mg GAE/g, and the antiradical activity was in the range of 35.75–38.235 mmol TEAC/100 g (*p* < 0.0001), depending on the percentage of fortification.

**Keywords:** avocado wastes/by-products; functional bread; lactic acid bacteria; sourdough; peels; pulp; seeds; polyphenols; antioxidant properties

#### **1. Introduction**

Avocado (*Persea americana* Mill.) is a subtropical/tropical fruit native to Mexico and Central America and is widely produced and consumed worldwide [1]. In recent years, avocado production has steadily increased globally due to the growing popularity and demand for the fruit [2]. The main avocado producers are Mexico (33%), the Dominican Republic (10.5%), Peru (7.8%), Indonesia (5.7%), and Colombia (5.1%) [3]. Spain and Italy are the only European countries with significant commercial production of avocados, which are cultivated, respectively, on the Andalusian Mediterranean coast, mainly in the provinces of Malaga and Granada, and in Sicily, along the Tyrrhenian coastal areas and close to Catania [4]. In Europe, avocado consumption per capita increased by an average of 180% between 2012/13 and 2018/20, with industry expectations for further increases [5]. A significant portion of this demand is driven by the young millennial generation in Europe and increased consumer interest in so-called "superfoods". Avocado fruit has great potential to meet consumers' desired requirements due to its high nutritional value, particularly its antioxidants, fiber, and low sugar content [6–8]. For these reasons, eating avocados is generally recommended for people with diabetes because it is a high-energy food [9] and can be used in a wide range of food products [10]. However, consumers' avocado

**Citation:** Viola, E.; Buzzanca, C.; Tinebra, I.; Settanni, L.; Farina, V.; Gaglio, R.; Di Stefano, V. A Functional End-Use of Avocado (cv. Hass) Waste through Traditional Semolina Sourdough Bread Production. *Foods* **2023**, *12*, 3743. https://doi.org/ 10.3390/foods12203743

Academic Editors: Grazia Maria Borrelli and Donatella Bianca Maria Ficco

Received: 14 September 2023 Revised: 9 October 2023 Accepted: 10 October 2023 Published: 11 October 2023

**Copyright:** © 2023 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 (https:// creativecommons.org/licenses/by/ 4.0/).

preference may depend on several quality attributes [11,12]. These quality attributes refer to physical product characteristics such as freshness, color, and size and experience attributes such as taste, aroma, and stage of maturation [13,14]. External factors that most influence consumer choice include fruit weight (commercial size), peel color (green or black), absence of defects (crick side, blanch, terminal spot), and ripening stage, which is closely related to fruit firmness, given the high perishability of avocados [14,15]. For these reasons, consumers prefer to buy unripe and/or not fully ripe avocado fruit [14], avoiding fruit that is already ripe or overripe.

Avocados are mostly consumed fresh, but they are also processed to extract oil and other products, such as guacamole [16]. Therefore, several components of the fruit, including the peel and seeds, are not used and are wasted, becoming a source of environmental contamination. However, these components are rich in protein, fiber, and numerous bioactive compounds [17,18]. For instance, the seed and peel of the "Hass" avocado account for about 15% and 14% of the weight of the fruit, respectively [18–20]. This is equivalent to at least 1.6 million tons of avocado seeds and peels annually discarded worldwide [1], which adds to the global share of food waste. Among the various processing techniques that can be used to recycle both the discarded (due to overripeness) avocado fruit and its by-products is dehydration.

Drying is probably one of the oldest methods of food preservation [21] and consists of the removal of water to a final concentration, which assures microbial stability and ensures the expected shelf-life of the product [22]. In addition, this technique is the most widely used for creating powders from fresh fruits [23].

Fruit and vegetable powders can be used as intermediates in the beverage industry, functional food additives that improve the nutritional value of foods, flavoring agents, or natural coloring agents [24]. Fruit and vegetable powders also serve as ingredients for pasta, breads, dry soups, and other food recipes [25–29].

Powder quality depends largely on the drying and milling conditions as well as the composition and quality of the raw material [30,31].

Fruit and vegetable powder ingredients for dough and/or bread preparation must be strategically selected to achieve the optimal composition and physical properties and avoid adverse effects [32]. In fact, vegetable powder can potentially decrease the stability of dough because the fiber in it slows down the rate of hydration and gluten development. This depends on the amount of vegetable powder incorporated [33]. Similarly, vegetable powder may affect texture differently depending on the type of by-product. On the other hand, fruit and vegetable powders can impart coloring and stabilizing properties to the final product due to the presence of carotenoids and polyphenols [33]. This work aimed to recycle avocado waste to produce new value-added ingredients. To this end, avocado waste was dehydrated and milled. The resulting powder was then used as ingredients for processing sourdough semolina bread to functionalize this food, widely consumed daily in Southern Italy.

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

#### *2.1. Chemicals and Reagents*

Methanol, sodium carbonate, gallic acid, Folin-Ciocalteu s phenol reagent, DPPH (2,2-diphenyl-1-picrylhydrazyl), ABTS (2,2 azino-bis (3-ethylbenzothiazoline-6-sulfonicacid), potassium persulphate, sodium hydroxide (NaOH) and Trolox (6-hydroxy-2,5,7,8 tetramethylchroman-2-carboxylic acid) were obtained from Fluka (Buchs, Switzerland). HPLC-grade water was obtained by purifying double distilled water in a Milli-Q Gradient A10 system (Millipore, Bedford, MA, USA), 0.45 μm PTFE syringe filter (Whatman, Milan, Italy).

#### *2.2. Production of Avocado Waste Powder (AWP) and Commercial Semolina*

Avocado fruit (*Persea americana* Mill.) cv. Hass was harvested at the experimental field of the Department of Agricultural, Food, and Forestry Sciences, University of Palermo. After being harvested, the fruits were left to ripen at a temperature of 20 ± 5 ◦C, and the progress of ripening was assessed by the change in epicarp color; therefore, the shade angle indicator was used, as described by Sánchez-Quezada et al. [34].

To obtain a uniform sample representative of the ripening stage, the researchers chose to pick avocado fruit at the overripe stage, i.e., fruits with hue angle values of ≥45 ± 7 h◦, as determined by Sánchez-Quezada et al. [34]. After being sanitized in chlorinated water (2% *v*/*w*) for 10 min, the avocado fruits were peeled and their seeds were removed.

The different components of the avocado fruit (pulp, seeds, and peels) were separated as they required different dehydration times and temperatures. After several preliminary tests the avocado pulp and its by-products were dried as follows:


Before dehydration, the seeds were washed with water, and the outer covering of the seeds was removed manually during washing. A tray dryer (Ausla, 1000 Watt, Milan, Italy) was used to dry the pulp and by-products.

The time/temperature binomials chosen generated a moisture content of less than 12–13%, which is the range selected to avoid microbial proliferation and to achieve rapid drying that would not lead to degradation of the bioactive components [35]. The dehydrated products obtained (pulp, seed, and peel) were separately processed into "powder" by an ultra-centrifugal mill (Fritsch, Pulverisette 14, Lainate, Italy). To obtain a powder with particle sizes between 1.5 and 2 mm from each fruit part (pulp, seeds, and peels), they were processed at 700 rpm for 10 s.

For breadmaking, a "powder mixture" (Avocado Waste Powder: AWP) comprised of pulp, seeds, and peels was used. The AWP consisted of the following percentages of dried fruit: 50% pulp, 25% seeds, and 25% peel.

A commercial semolina (Cuore Mediterraneo, Santa Giusta, Italy) was used to process the bread for this study. Its nutritional values (per 100 g) were: 12.5 g of protein; 1.5 g of fat; 0.3 g of saturated fats; 69 g of carbohydrates, and 26 g of fiber.

#### *2.3. Determination of Color Characteristics of AWP*

The color of both semolina and AWP samples was measured using a Minolta colorimeter (Chroma Meter CR-400, Konica Minolta Sensing Inc., Tokyo, Japan), and the L\* (brightness), Chroma (C\*), and hue angle (h◦) parameters were evaluated [21]. The instrument was calibrated using a standard white plate. Chroma (C\*) values and hue angles (h◦) were calculated using Equations (1) and (2), respectively:

$$\mathbf{C}^\* = \left(\mathbf{a}^2 + \mathbf{b}^2\right)^{\frac{1}{2}} \tag{1}$$

$$\mathbf{h}^{\odot} = \arctan\left(\frac{\mathbf{b}}{\mathbf{a}}\right) \tag{2}$$

Using the obtained values of L\*, a\*, and b\*, a color table was created by converting the CIEL\*a\*b\* color space to the red/green/blue (RGB) scale through the e-paint.co website (accessed on 15 June 2023).

#### *2.4. Hygienic Characteristics of AWP*

The AWP was microbiologically analyzed for some microbial groups that are unwanted during food fermentation, as reported by Messina et al. [36]. Briefly, 10 g of AWP was first homogenized by a BagMixer® 400 stomacher (Interscience, Saint Nom, France) and then serially diluted. The diluted samples were analyzed for the following microbial groups: total mesophilic microorganisms (TMM), members of the Enterobacteriaceae family, total coliforms, and spore-forming aerobic bacteria. The analyses were performed in duplicate.

#### *2.5. Bacterial Strains*

Lactic acid bacteria (LAB) isolated from Sicilian sourdoughs and previously tested to produce semolina breads with the addition of by-product ingredients [37] were used to prepare a multiple-strain sourdough starter. The strains *Lentilactobacillus diolivorans* SD4, *Fructilactobacillus sanfranciscensis* SD22, *Levilactobacillus brevis* SD46, *Lactiplantibacillus plantarum* SD96, *Weissella cibaria* SD123, *Lactiplantibacillus pentosus* SD130, *Leuconostoc citreum* SD142, and *Leuconostoc holzapfelii* SD148, all belonging to the Culture Collection of the Agricultural Laboratory of the University of Palermo, Italy, were defrosted from −80 ◦C and cultivated in de Man-Rogosa-Sharpe medium modified as described by Lhomme et al. [38] at 30 ◦C for 24 h.

#### *2.6. Sourdough Propagation*

After reactivation in a synthetic medium, all the LAB strains were propagated in sterile semolina extract (SSE) broth [39]. Commercial semolina was used to both prepare liquid SSE broth and propagate solid sourdough. The individual cultivation of LAB and the mixed cell culture representing sourdough inoculum were performed as reported by Gaglio et al. [40].

The LAB mixed culture was diluted in sterile tap water to reach a final volume of 187.5 mL. This cell suspension was added to 312.5 g of semolina to obtain a 500 g dough with a dough yield (DY = weight of the dough/weight of semolina × 100) of 160 and a cell density of about 106–107 CFU/g. The dough was then left to ferment at 28 ◦C for 16 h and subjected to seven consecutive daily refreshments to generate a mature sourdough inoculum [41].

#### *2.7. Bread Doughs and Baking Process*

Bread production was carried out solely with the sourdough developed from the selected LAB strains. No baker's yeast and kitchen salt were added to evaluate the effect of AWP on the performance of LAB. The control (CTR) doughs (800 g) to be leavened before baking were processed by adding 228.6 mL of sterile tap water and 457.2 g of semolina to 114.2 g of mature sourdough (DY = 175). The experimental AWP doughs were produced with the same amount of water and sourdough, but the amount of semolina was reduced to 434.3 g and 411.4 g for the 5-AWP [containing 5% (*w*/*w*) AWP] and 10-AWP [containing 10% (*w*/*w*) AWP] trials, respectively. A planetary mixer (model XBM10S; Electrolux Professional SpA, Pordenone, Italy) was used to mix all the ingredients for 15 min with a paddle turning on Speed 4. Aliquots of 100 g per dough were transferred into trapezoidal stainless steel baking pans [42], kept at 28 ◦C for 8 h, and then baked as reported by Alfonzo et al. [43]. Two technical repeats were obtained from each bread trial (performed in duplicate), and all bread baking was repeated after two weeks to obtain two independent replicates.

#### *2.8. Acidification Process*

Sourdough fermentation was monitored by pH measurement, total titratable acidity (TTA) determination, and the evolution of LAB numbers following the approach of Francesca et al. [44]. To perform LAB viable counts and that of other microbial groups relevant during dough fermentation, the sourdough and bread doughs were microbiologically evaluated to enumerate TMM, sourdough LAB, yeasts, members of the Enterobacteriaceae family, and total coliforms, as reported by Gaglio et al. [37]. All analyses were performed in duplicates.

#### *2.9. Quality Characteristics of Breads*

The breads were cooled at room temperature for approximately 30 min after baking and investigated for several quality parameters, as reported by Cirlincione et al. [45]. In particular, the following parameters were considered: weight loss (WL, %), specific volume (cm3/g bread), firmness (N/mm2), crust and crumb color [Lightness (L\*), redness

(a\*) and yellowness (b\*)], void fraction (%), cell density (number of cells/cm2), and mean cell area (mm2). The analyses were performed in duplicate.

#### *2.10. Chemical Characterization*

#### 2.10.1. Total Phenolic Content Analysis

Total phenolic content (TPC) was determined using the optimized Folin–Ciocalteu method previously published [46]. One gram of each bread sample (CTR-Bread, bread produced with control dough; 5-AWP Bread, experimental bread enriched with 5% (*w*/*w*) of avocado waste powder (AWP); 10-AWP Bread, experimental bread enriched with 10% (*w*/*w*) of AWP) was added to 5 mL of methanol/water (80:20 *v*/*v*) and sonicated and filtered through Whatman 0.45 μm PTFE filters. This was followed by a reaction with the Folin-Ciocalteu reagent in the presence of sodium carbonate to form a blue-colored complex. The intensity of the color was proportional to the phenolic compounds in the sample. The resulting colorimetric reaction was measured at 765 nm using a UV-VIS spectrophotometer (Varian Cary 50, Agilent, Santa Clara, United States). The amount of TPC was calculated by interpolation from a calibration curve of gallic acid [0.001 to 0.25 mg/mL] (y = 10.945x + 0.1305, R2 = 0.993). The results were expressed as mg gallic acid equivalents per g (mg GAE g<sup>−</sup>1) of the sample.

#### 2.10.2. Radical Scavenging Properties Evaluation, DPPH and ABTS Assay

The measurement of the powder and fortified bread samples' antiradical activity (DPPH (2,2-diphenyl-1-picrylhydrazyl) and ABTS (2,2 azino-bis (3-ethylbenzothiazoline-6 sulfonic-acid) assays) followed a procedure previously described by Di Stefano et al. [47]. The DPPH assay was used for the in vitro evaluation of the scavenger activity toward free radicals. One g of each bread sample (AWP and semolina) was extracted with 4 mL of methanol, mixed with 3 mL of DPPH (60 μM) and placed in the dark for 30 min. Scavenging activity was monitored by spectrophotometric analysis of the absorbance at a wavelength of 517 nm with a UV-VIS spectrophotometer (Varian Cary® 50, Agilent, Santa Clara, United States) and using methanol as the blank. The results were reported as Trolox equivalent antioxidant activity and expressed as mmol Trolox equivalent (TE)/100 g of the sample. The absorbance signal was translated into antioxidant activity using Trolox as the standard and the calibration curve in the range of 5–400 <sup>μ</sup>M (y = −0.0008x + 0.4036, R2 = 0.998). All experiments were performed in triplicate. According to Re et al. [43], for the ABTS assay, the ABTS<sup>+</sup> radical cation was produced by reacting ABTS stock solution with 2.45 mM potassium persulfate; the solution turned a dark blue-green at the end of the reaction time. One gram of each sample was added to 4 mL of methanol, sonicated, and filtered through Whatman 0.45 μm PTFE filters. A calibration curve using Trolox at increasing concentrations [2.5–30 μM] was constructed. The assays were performed in triplicate.

#### *2.11. Sensory Analysis*

A panel of 17 judges, including 10 women and seven men whose ages ranged from 26 to 57, was recruited to perform a descriptive sensory analysis of the breads produced with different percentages of AWP. First, the judges were trained to acquire familiarity with bread attributes by tasting commercial semolina bread. The judges were asked to score appearance, texture, odor, and taste descriptors from among those reported by Ruisi et al. (2021) [48]. The evaluation for each attribute was expressed on a 9-point scale (1 = extremely bad; 9 = extremely good). The evaluation was performed by the judges in individual chambers following the ISO 13299 guidelines [49].

#### *2.12. Statistical Analysis*

Differences between the microbiological and physicochemical data were identified by way of one-way variance analysis, while Tukey's test was used for multiple mean comparisons (statistical significance *p* < 0.05). In addition, a hierarchical cluster analysis (HCA) was performed to group the produced breads according to their dissimilarity, as reported by Martorana et al. [50]. The data were statistically processed using XLStat software version 7.5.2 for Excel (Addinsoft, New York, NY, USA).

#### **3. Results**

#### *3.1. Color Characteristics*

From the analysis of colorimetric data (presented in Table 1), the semolina sample showed an elevated brightness (86.90 ± 1.63) compared to the AWP sample but a lower C\* index. This was due to the color of the semolina, which, as can be seen from the RGB data, tended toward white. The AWP, on the other hand, showed lower values of L\* but higher values of Chroma. From the values obtained and the evaluation of the color table, the powders maintained a color tending toward green. Therefore, processing the dehydrated avocados into powder did not alter their color.

**Table 1.** Brightness (L\*), chroma (C\*), and hue angle (h◦\*) of the CTR and AWP samples. CIELab\* values were then converted to RGB. Abbreviation: AWP (avocado waste powder). Results indicate mean values ± S.D. (standard deviation). n.a. = not analyzed. Data within a column followed by different letters are significantly different according to Tukey's test.


#### *3.2. Monitoring of the Fermentation Process*

The growth of the selected LAB strains in SSE showed a consistent decrease in pH until the average value of 4.03 ± 0.10. The lowest pH (3.82 ± 0.18) was reached by the strains *L. pentosus* SD130 and *L. plantarum* SD96. The mixture with the eight LABs developed individually in SSE that was obtained after three propagation days was used as a liquid inoculum to produce sourdough for bread production. Soon after inoculation of semolina and tap water with the eight strains, the obtained dough showed a pH of 5.44 ± 0.21; this value decreased to 3.92 ± 0.04 in the sourdough obtained after seven days with daily refreshments. The TTA of the mature sourdough was 12.10 ± 0.07 NaOH 0.1 N/10 g. Table 2 presents the pH and TTA data for the unbaked doughs up to 8 h of leavening. A pH value of 5.5 was measured for the CTR dough at the start of fermentation.

The initial pH of the 5-AWP dough was slightly lower (5.42) than that of the 10- AWP dough (5.56) and the CTR dough. During fermentation, the pH values decreased progressively until they reached almost similar values of 4.19, 4.33, and 4.37 for the CTR, 5-AWP, and 10-AWP doughs, respectively, at the end of the monitoring period. An inverse trend was noted for the TTA, which increased linearly over time. After 8 h of fermentation, the TTA of the CTR dough (8.93 mL NaOH 0.1 N/10 g) was slightly higher than the value expressed by the AWP doughs (8.65 and 8.60 for 5-AWP and 10-AWP, respectively). The results of the plate count of the doughs are presented in Table 3. Sourdough developed from LAB selected at the seventh refreshment was characterized by 8.64 and 7.74 Log CFU/g of LAB and yeasts, respectively. Regarding the doughs leavened for bread production, LAB accounted for 7.60–7.71 Log CFU/g soon after ingredient mixing. These data were a little higher than those shown by the TMM (6.60–6.66 Log CFU/g) and confirmed that LAB from the sourdough inoculum was significantly transferred to the bread doughs. The levels of yeast immediately after production (0 h) were one order of magnitude lower than the LAB. After 8 h of fermentation, the cell densities of LAB increased in all the trials. The LAB levels of the CTR and 5-AWP doughs were almost comparable (8.94 and 8.98 Log CFU/g, respectively), while a slightly lower density (8.51 Log CFU/g) characterized the 10-AWP dough. A very limited increase in cell density was registered for the yeasts, barely overcoming 7.0 Log CFU/g for the CTR and 5-AWP doughs at the end of fermentation. Regarding hygiene indicators, although the levels of Enterobacteriaceae and total coliforms

in the AWP were below the detection limit (for this reason, these results are not included in Table 3), after 8 h of fermentation, their presence was revealed in both AWP bread doughs at levels around 10<sup>3</sup> CFU/g. No spore-forming bacteria were detected in the AWP and corresponding bread doughs.

**Table 2.** Chemical parameters of doughs. Results indicate mean values ± S.D. (standard deviation) of four determinations (carried out in two technical repeats for two independent experiments). Data within a line followed by different letters are significantly different according to Tukey's test. Abbreviations: TTA, total titratable acidity; CTR, control dough; 5-AWP, experimental dough enriched with 5% (*w*/*w*) of avocado waste powder (AWP); 10-AWP, experimental dough enriched with 10% (*w*/*w*) of AWP; n.a. = not analyzed.


**Table 3.** Microbial loads of doughs. Results indicate mean values ± S.D. (standard deviation) of four plate counts (carried out in two technical repeats for two independent experiments), expressed as Log CFU/g. Data within a line followed by different letters are significantly different according to Tukey's test. Abbreviations: TMM, total mesophilic microorganisms; LAB, lactic acid bacteria; CTR, control dough; 5-AWP, experimental dough enriched with 5% (*w*/*w*) of avocado waste powder (AWP); 10-AWP, experimental dough enriched with 10% (*w*/*w*) of AWP; n.a. = not analyzed; n.d. = not detected.


#### *3.3. Bread Quality Attributes*

The characteristics of the final breads produced are summarized in Table 4.

Weight loss after baking was 11.30% in the CTR bread, while lower values were displayed by the AWP breads. The specific volume of the breads decreased with the different percentages of AWP; a value of 3.18 cm3/g was registered for the CTR bread compared to a value of 2.74 cm3/g for the 10-AWP bread. The addition of AWP determined a linear increase in firmness, with the highest value (0.113 N/mm2) recorded for the 10-AWP bread. Furthermore, the addition of AWP determined a change in the color parameters of both the crust and crumb of the breads, especially for L\* and a\*. Both parameters decreased progressively with the AWP percentages. Negative values were registered for the crumbs of all the trials. Image analysis of the breads indicated an increase in the void fraction and cell density of the crumbs with increasing percentages of AWP and a decrease in alveolation.

**Table 4.** Quality attributes of bread samples. Results indicate mean values ± S.D. (standard deviation) of four determinations (carried out in two technical repeats for two independent experiments). Data within a line followed by different letters are significantly different according to Tukey's test. Abbreviations: CTR-Bread, bread produced with control dough; 5-AWP Bread, experimental bread enriched with 5% (*w*/*w*) of avocado waste powder (AWP); 10-AWP Bread, experimental bread enriched with 10% (*w*/*w*) of AWP.


#### *3.4. Chemical Characterization of Raw Materials and Bread Samples*

The antioxidant activity and antiradical scavenging activity of the raw materials were measured; in particular, as shown in Table 5, high TPC was mostly highlighted in the AWP (197.775 mgGAE/g) compared to the semolina (3.676 mgGAE/100 g). The highest increase in antiradical activity was observed in the AWP, with values of 38.235 mmol TE/100 g and 35.175 mmol TE/100 g for the DPPH and ABTS assays, respectively, while the lowest was recorded for semolina (2.656 and 2.408 mmol TE/100 g for the DPPH and ABTS assays, respectively).

**Table 5.** Antioxidant and antiradical activity of semolina and avocado waste powder (AWP). Results indicate mean values ± S.D. Data within a column followed by different letters are significantly different according to Tukey's test. Abbreviations: AWP (avocado waste powder); TPC (total phenolic content).


The same analyses were carried out on the bread samples fortified with different percentages of AWP. The addition of AWP enhanced the samples' antiradical and antioxidant activity. As Table 6 shows, 10-AWP bread had higher values of antioxidant (23.882 mg-GAE/100 g) and antiradical activity (6.656 and 9.234 mmol TE/100 g for DPPH and ABTS assays, respectively) compared to the CTR bread, which was made with only semolina.

**Table 6.** Antioxidant and antiradical activity of fortified and control bread samples (5-AWP Bread, 10-AWP Bread and CTR-Bread). Results indicate mean values ± S.D. Abbreviations: CTR-Bread, bread produced with control dough; 5-AWP Bread, experimental bread enriched with 5% (*w*/*w*) of avocado waste powder (AWP); 10-AWP Bread, experimental bread enriched with 10% (*w*/*w*) of AWP; TPC (total phenolic content).


#### *3.5. Bread Sensory Attributes*

Figure 1 presents the spider plot resulting from the sensory evaluations of the CTR and AWP breads.

**Figure 1.** Spider diagrams of descriptive sensory analysis of breads. Abbreviations: CTR Bread; 5-AWP Bread, experimental bread enriched with 5% (*w*/*w*) of avocado waste powder (AWP); 10-AWP Bread, experimental bread enriched with 10% (*w*/*w*) of AWP; n.s., not significant (*p* > 0.05).

The addition of AWP, at both percentages, greatly affected the sensory characteristics of the semolina breads. The sensory traits significantly different from those of the control breads were crust and crumb color; crispy crust; bread and strange odor; astringent, bitter taste persistency; bread and strange aroma; and, especially, aroma intensity. Except for bread odor and bread aroma, which were scored at a lower level than the CTR bread, all the other traits mentioned had higher scores for the AWP breads. Regarding bread structure, although alveolation and adhesiveness are lower in AWP breads, the differences are not significant. Considering the overall assessment based on all these traits, the CTR bread received the highest scores, and between the 5-AWP bread and 10-AWP bread trials, the breads processed with 5% AWP were more appreciated.

#### *3.6. Multivariate Analysis*

The HCA clustered the breads based on their dissimilarity and relationship using a total of 37 variables, including quality attributes, antioxidant and antiradical properties, and sensory traits. The resulting cluster presented in Figure 2 shows low levels of dissimilarity (0.099%) among the breads. However, the breads enriched with AWP formed a single cluster and were clearly separated from control production.

**CTR-Bread 5-AWP Bread 10-AWP Bread**

**Figure 2.** Dendrograms obtained from hierarchical cluster analysis based on values of quality attributes, antioxidant and antiradical properties and sensory traits of breads. Abbreviations: CTR Bread; 5-AWP Bread, experimental bread enriched with 5% (*w*/*w*) of avocado waste powder (AWP); 10-AWP Bread, experimental bread enriched with 10% (*w*/*w*) of AWP.

#### **4. Discussion**

In the production of cereal-based foods, fiber has grown in popularity as an added functional ingredient. Fiber, such as inulin, improves the rheological and technological characteristics of food, as well as its consistency, acceptability, and healthy properties. It also targets the prevention of metabolic syndromes. In addition to these improvements, when a food product is enriched with fiber, its shelf life is extended [51]. Other fiber-rich matrices, such as wheat or oat bran, have been used to replace wheat flour in baking [52]. One of the main reasons to supplement foods with dietary fiber is that it produces a wide variety of flavors that make products more palatable [53]. In a recent study, it was also shown that fortifying semolina bread with hemp seed flour improves its nutritional and antioxidant properties without significant changes in rheological properties [54]. In the work of Gómez and Martinez [32], the incorporation of fruit and vegetable by-products in baked products was evaluated to create foods with a higher fiber content. The authors highlighted a slowdown in the digestion of starch and other carbohydrates present in cereals and an improvement in rheological properties and interactions with digestive enzymes

in the stomach. The improvement of antioxidant activity in fortified foods is due to an increase in the content of bioactive compounds, such as polyphenols and carotenoids [32]. Additionally, in the study of Gaglio et al. [40], the reuse of by-products for the production of fortified bread was evaluated. In particular, the authors used powdered almond skin at different percentages (5% and 10%) to produce functional products by modifying a traditional sourdough bread recipe. The final characteristics of the bread were influenced by the fortification and its percentage. The powdered almond skin positively influenced the sensory characteristics of the bread, with an increase in the intensity of the odor and the color of the crust and crumb. Thanks to the phytochemicals released by the fortified bread, an increase in the antioxidant capacity that can provide antioxidant protection at the level of human intestinal cells was also highlighted. Moreover, the microbiological parameters during fermentation were influenced by the development of coliforms due to the presence of spores after baking [37]. A recent study investigated Cava lees, another type of by-product that represents 25% of wine industry waste and is rich in antioxidant compounds and dietary fiber. This study aimed to evaluate the effect of Cava lees on microbial populations during natural leavening and bread fermentation. The results showed that the bread formulation with 5% Cava lees promoted the growth of both LAB and yeast and increased the concentration of volatile substances typically present [55].

Considering managing agro-wastes and food by-products while avoiding environmental concerns, the present work focused on valorizing the avocado production chain through the reuse of waste arising from non-compliant avocado fruits as well as avocado seeds and peels, to produce functional bread. From the results obtained in this study, it can be stated that the right time-temperature combination for the dehydration of avocado pulp and by-products, which maintained colorimetric and antioxidant characteristics after the dehydration process, was found. In particular, the grinding process did not affect or alter the organoleptic characteristics of the powder. It is important to emphasize that color is an extremely important characteristic because it makes the product attractive and acceptable, inducing consumers to purchase it; in other words, it is the first quality that guides consumers' purchase choices [56]. The dehydration process tends to alter the surface characteristics of the food and, consequently, alters both reflectivity and color properties; particularly in fruit, alterations occur at the expense of carotenoids and chlorophyll. Such alterations were not observed in this study, and as reported in the results, the powders retained a greenish color reminiscent of fresh avocado. In addition, due to the dehydration process, loss of the aromatic substances could occur [57], depending on the amount of heat energy absorbed by the product, in the form of sensible and latent heat, for the vaporization of water. The amount of this loss depends on temperature, moisture content of the food, vapor pressure, and the solubility of the volatile compounds in water vapor. However, the extent of this loss is also related to lipid oxidation reactions [58]. The oxidation of fatty acids gives rise to aldehydes, ketones, and acids that cause rancidity and off-flavors [59]. In this case, the products obtained by the dehydration protocol used did not cause such alterations, and this is deducible from the sensory evaluation of the functional bread. For bread making, the sourdough inoculum was developed from selected starter strains. They all acidified the SSE used for pH values in the range of 3.82–4.34, which are generally registered for sourdough *Lactobacillus*, *Leuconostoc*, and *Weissella* grown in this semolina-derived medium [39,60]. Looking at the TTA data for the starter strains and bread doughs, this parameter was confirmed to evolve (increase) inversely with pH [61]. The acidification process was also followed through the LAB development, and at the end of fermentation (8 h), they increased about two Log cycles in all the bread doughs, as commonly observed in sourdough bread production [62]. The LAB levels were slightly higher than the TMM levels, indicating the absolute dominance of the added strains during fermentation. Furthermore, the low TMM levels registered are a consequence of the high nutritional requirements of LAB that are not fully satisfied by principal component analysis [63,64]. Of course, during sourdough fermentation, the development of yeasts is also particularly important [65]. Yeast cell densities estimated at the end of the leavening duration were between 6.78 and

7.06 Log CFU/g, although they were not deliberately inoculated. However, yeasts develop spontaneously in sourdough [66], and the results of this study are generally found in semolina sourdough fermentation [41], even in the presence of waste/by-product addition [40,64]. Furthermore, in this research, the ratio between yeasts and LAB was optimal at 1:100, which is considered optimal for sourdough preparations [67]. The AWP was also investigated for several undesired groups such as *Salmonella* spp., *Listeria monocytogenes*, coagulase-positive staphylococci, members of Enterobacteriaceae, coliforms, *Escherichia coli*, spore-forming bacteria, and *Pseudomonas*; none of these bacterial groups exceeded the detection limit. However, after 8 h of fermentation, members of Enterobacteriaceae and total coliforms, but no *E. coli*, were detected in both AWP bread doughs. Although undesirable in dough because they compete with LAB and yeast development, the levels estimated were particularly low, and this is imputable to the fact that Enterobacteriaceae are limited in their growth by the low pH encountered in sourdough during fermentation [68]. The quality attributes of the breads were impacted by the addition of AWP. In general, a diminution in WL and the specific volume of the breads is generally reported when food waste is added [37]. The data presented in this work confirmed this trend, but only the 10-AWP breads were characterized by a WL and specific volume significantly different from those displayed by the CTR bread. The browning of the crust and crumb was due to the darker color of AWP compared to semolina. This browning is a positive sign since, as reported by Sandvik et al. [69], the dark color of bread is linked to health among consumers.

The firmness of the breads increased as a consequence of the increase in dietary fiber, as observed by Ruisi et al. [48]. Regarding the image analysis of the central slices of the breads, alveolation diminished with the AWP-enriched breads. This phenomenon was also observed with the addition of pumpkin pomace and dry tomato waste [70,71] and is due to the low percentage of gluten [72]. The antiradical and antioxidant activity was mostly highlighted in the AWP breads compared to the semolina breads. The highest value of TPC was also observed in the AWP sample, while the lowest was recorded for the semolina sample. The supplementation of AWP in bread enhanced the food's antiradical and antioxidant activity. The 10-AWP bread had higher values of antioxidant and antiradical scavenging activity compared to the CTR bread, which was made with only semolina. According to TPC values, the chemical analyses showed that the fortification of bread with AWP in different percentages increased its antiradical and antioxidant activity and organoleptic and baking qualities proportionally to the percentage of fortification used.

#### **5. Conclusions**

The overall data collected showed the excellent suitability of AWP for functional bread making. The present research found that avocado waste products that have been dehydrated and processed into powder can be successfully incorporated into leavened baked products to improve their characteristics, particularly in terms of antioxidant content. A positive relationship between the proportion of added powder and antioxidant content appeared, as well as the organoleptic and baking qualities of bread. The addition of 10% AWP produced dough with a higher antioxidant profile than the control bread. In addition, the bread produced from this dough was highly appreciated on a sensory level in terms of aroma and color. Fortified bread, therefore, has shown great potential to serve as a functional food among consumers.

**Author Contributions:** Conceptualization, V.F. and I.T.; methodology, V.F., I.T., V.D.S., R.G., L.S. and C.B.; formal analysis, I.T., E.V. and C.B.; data curation, C.B., I.T., R.G. and E.V.; writing—original draft preparation, I.T., C.B., R.G., E.V. and L.S.; writing—review and editing, V.F., L.S., V.D.S. and R.G.; supervision, V.F., V.D.S., R.G. and L.S.; funding acquisition, V.F. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the TINFRUT project, funded by PSR Sicilia 2014–2020 Programma di Sviluppo Rurale SOTTOMISURA 16.1 "Sostegno per la costituzione e la gestione dei gruppi operativi del PEI in materia di produttività e sostenibilità dell'agricoltura".

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

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

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

## *Review* **The Wheat Aleurone Layer: Optimisation of Its Benefits and Application to Bakery Products**

**Lucie Lebert 1, François Buche 2, Arnaud Sorin <sup>1</sup> and Thierry Aussenac 2,\***

<sup>2</sup> Institut Polytechnique UniLaSalle, Université d'Artois, ULR 7519, 60026 Beauvais, France

**\*** Correspondence: thierry.aussenac@unilasalle.fr

**Abstract:** The wheat aleurone layer is, according to millers, the main bran fraction. It is a source of nutritionally valuable compounds, such as dietary fibres, proteins, minerals and vitamins, that may exhibit health benefits. Despite these advantages, the aleurone layer is scarce on the market, probably due to issues related to its extraction. Many processes exist with some patents, but a choice must be made between the quality and quantity of the resulting product. Nonetheless, its potential has been studied mainly in bread and pasta. While the nutritional benefits of aleurone-rich flour addition to bread agree, opposite results have been obtained concerning its effects on end-product characteristics (namely loaf volume and sensory characteristics), thus ensuing different acceptability responses from consumers. However, the observed negative effects of aleurone-rich flour on bread dough could be reduced by subjecting it to pre- or post-extracting treatments meant to either reduce the particle size of the aleurone's fibres or to change the conformation of its components.

**Keywords:** wheat aleurone; dietary fibre; extraction process; antioxidant; bread; arabinoxylans

#### **1. Introduction**

Wheat is indispensable in producing many staple foods around the world, including bread, biscuits, cakes and noodles. As epidemiological studies have demonstrated that an increase in the intake of whole grain products is related to a lower incidence of cardiovascular disease, obesity, diabetes and cancer, the composition of wheat makes it a valuable asset in the diet for the prevention of chronic diseases. Consequently, in relation to the growing number of metabolic diseases, nutritional guidelines worldwide advise an increase in the consumption of whole grain products, partly as they contain fibres that many consumers lack in their diet [1,2].

These health benefits are mostly due to the presence of micronutrients, dietary fibres (DF) and bioactive components, which are mainly located in the outer layers of the grain: the bran and the aleurone layer [1,3,4]. Many researchers and industries have aimed to extract, isolate and introduce these grain fractions in food products as ingredients for added nutritional value. However, it seems that the addition of bran or fibre to wheat-flour-based products changes not only the technological properties of the end-product but also its sensory acceptance by consumers [3,5]. Moreover, the main challenge of the extraction process is improving the nutritional properties of the ingredient without impairing its technological properties during breadmaking. For instance, soluble fibres, such as waterextractible arabinoxylans (WEAX), answer this problem. The use of fibres from wheat bran has also been deeply investigated [3,6–8].

Recently, researchers have focused on the wheat aleurone layer, considered by millers to be the main bran layer. As it contains the majority of the grain's minerals and is also rich in protein, DF and bioactive components (mostly ferulic acid), the aleurone layer may be the source of many bran's reported health benefits [4,9–11]. However, despite its known nutritional and health-prevention properties, the aleurone layer is scarce on the market,

**Citation:** Lebert, L.; Buche, F.; Sorin, A.; Aussenac, T. The Wheat Aleurone Layer: Optimisation of Its Benefits and Application to Bakery Products. *Foods* **2022**, *11*, 3552. https:// doi.org/10.3390/foods11223552

Academic Editors: Donatella Bianca Maria Ficco and Grazia Maria Borrelli

Received: 28 September 2022 Accepted: 3 November 2022 Published: 8 November 2022

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

**Copyright:** © 2022 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 (https:// creativecommons.org/licenses/by/ 4.0/).

whether included in cereal food products or as an ingredient. This could be related to the challenges posed by its extraction. Although multiple processes have been patented, the end product is often obtained with either low purity or yield [10,12–14]. Moreover, it seems that its incorporation into bakery products yields contrasting results, as some negative technological effects could be observed, such as a reduction in loaf volume and increased crumb hardness [15–17]. These limiting technological aspects on final food products could be improved by making specific modifications to the aleurone's components without losing any of the health benefits, prior to its incorporation into a food process, as is already carried out for wheat [18].

The objective of this review, therefore, is to provide readers with the elements of understanding to optimise the potential of the wheat aleurone layer, both from a nutritional and technological point of view. To do so, the wheat aleurone layer will first be described in terms of function and composition. Its potential nutritional and health benefits will then be presented based on clinical studies and in relation to the individual effects of its compounds. Aleurone's potential as an ingredient will also be reviewed, starting with the processes to extract it and related issues. Then, the applications of cereal-based food products performed in the literature will be investigated, with a tentative explanation of the underlying mechanisms of the observed effects. Finally, the last section will focus on the existing processes meant to optimise the aleurone layer's potential, both for its nutritional, health-related and technological benefits.

#### **2. Aleurone Layer**

#### *2.1. Description: Histology and Functions*

The aleurone layer is a tissue of wheat grain made of unicellular block-shaped cells (37–65 μm vs. 25–75 μm) [19]. Among the seven layers comprising the mature bran, the aleurone is the only one with the remaining living cells [20]. It is located between the endosperm and the nucellar epidermis (or hyaline layer), as shown in Figure 1 [10,21]. Although botanically part of the endosperm, it is considered by millers as a bran layer since it remains attached to the hyaline layer during milling [10,22]. It represents around 50% of wheat bran (or 75% *w*/*w* of its dry weight), making it the major bran layer [10,12]. Indeed, the aleurone layer is thick and can reach up to 65 μm [23]. It also corresponds to 5–8% (*w*/*w*) of the whole kernel [21].

**Figure 1.** Wheat grain histology [10]. Adapted from Surget and Barron (2005) [21]. (Copyright for reprinting was requested and obtained through Taylor & Francis and Copyright Clearance Center. License number: 5386341293998. License date: 12 September 2022).

Although the aleurone layer is singular in wheat, it can be found multi-layered in barley, rice and oats [10].

Multiple functions in the wheat grain are allotted to the aleurone layer, namely: accumulation and transport of nutrients for seed germination, decomposition of storage materials of the endosperm for embryo growth, and protection and maintenance of caryopsis activity [24]. More specifically, its major role is during germination, where it is involved in the synthesis and release of hydrolases in the endosperm, as induced by gibberellin [11,25]. These enzymes then break down starch polymers and proteins in starchy endosperm cells, which undergo programmed cell death [22,26]. To facilitate the transfer, the aleurone's outer cell wall is degraded by endogenous hemicellulases, and only the inner resistant layer remains [25,27].

The aleurone layer is also involved in seed dormancy, induced by abscisic acid. At the same time, this hormone induces programmed cell death in endosperm cells [11]. However, the aleurone layer serves as grain storage for metabolites, minerals and amino acids [9,26,28]. It is equally involved in the regulation of water diffusion and distribution through its cell-walls [11,29].

Finally, in the crease region of the grain, some modified aleurone cells, called transfer cells, also participate in grain filling via solute uptake [30]. Due to their arabinoxylan's higher degree of arabinose substitution and lower degree of feruloylation, the transfer cells show specific cell wall hydration and porosity properties that are compatible with water diffusivity and uptake for grain filling [31].

#### *2.2. Composition*

#### 2.2.1. Cell Wall

The aleurone cell-wall represents 35% (*v*/*v*) of the total cellular volume. It is bilayered, with a thicker outer layer (of 2 μm vs. 0.5 μm for the inner layer) and mainly composed of arabinoxylans (65%), β-glucans (29%) and phenolic acids [9], as presented in Tables 1 and 2. This high proportion of dietary fibres (DF), especially pentosans (44% *w*/*w* of total grain [32,33]), makes the aleurone's cell walls valuable from a technological and nutritional point of view, thanks to their gelling ability [6], and cancer-prevention properties [34,35].

**Table 1.** Carbohydrates repartition in the wheat aleurone layer, bran and whole grain.


\* The aleurone layer and the bran amount respectively to 5–8% (*w*/*w*) and 5.5–9.5% (*w*/*w*) of the wheat grain [21]. All data were placed in the same unit to facilitate comparison.

Its major component, arabinoxylan (AX), is a polysaccharide made of a linear xylose chain with β-(1,4) linkages between its xylopyranosyl residues (Figure 2 [38]). These residues can either remain unsubstituted, mono-substituted at the third carbon position (C3), or with a di-substitution at both the C2 and C3 positions. Indeed, α-L-arabinofuranose residues can be esterified to the xylan backbone via α-(1,2) and α-(1,3) linkages but not in an even pattern. These arabinose residues can, in turn, be esterified by phenolic acids (mainly

ferulic acid) on their primary alcohol function (O5) [9,10,39]. The feruloylation of AXs is progressive throughout grain filling [31] and takes place on average every 15 arabinofuranose residues [22]. In addition to arabinose substitution, some acetyl groups can be esterified to xylan instead [30]. The presence of these side groups confers technological value to aleurone's AXs as they can interact with AX chains (dimerization) or other molecules, such as proteins or fibres, thus strengthening the gluten network [6].

**Figure 2.** Arabinoxylan structure.

With a low arabinose to xylose ratio (A:X = 0.41–0.47), aleurone cell wall AXs are considered poorly cross-linked [23]. AX solubility is thus influenced since the presence of arabinose residues causes the formation of long asymmetrical polysaccharides; hence, an unsubstituted region has a tendency to aggregate (stabilisation by hydrogen bonds) and become water insoluble [6,25,40]. This insolubility increases with the presence of FA, which is shown by a high FA monomer-to-arabinose ratio (7.2–7.4) [10,23]. Only a few waterextractable AXs (WEAX) can be found at the cell-wall surface (1.5 to 2.5%), probably related to incomplete cross-linking with other components [39]. In addition, compared with other bran layers, there is less cross-linking between polysaccharides and phenolic compounds in aleurone. This might be due to the role of aleurone in the enzymatic degradation of the stored compounds, which requires facilitated movement through the cell walls [41]. Although, most health benefits related to the presence of AXs in the diet stem from the addition of WEAX, the predominance of water-unextractable AXs (WUAX) in the aleurone cell-walls can also be valued. However, it is often associated with negative rheological traits in breadmaking [39].

The second main component in wheat aleurone cell walls, β-glucan (BG), consists of a linear chain of glucose residues joined by glycosidic linkages [9]. The majority of the chain (90%) is made of cellotriosyl (DP3) and cellotetraosyl (DP4) units bonded by β-(1,3) linkages. The remaining 10% refer to β-(1,4)-linked side chains [25]. According to Jamme et al. [42], BGs are not present in the junction zone, whereas they are abundant in the inner periclinal face. As presented in Table 1, there are more BGs in aleurone cell walls compared to bran, which can be related to the layer's reported health benefits, such as the modulation of digestion [43].

Finally, phenolic acids are present in the aleurone cell walls (Table 2). They are mostly (95%) represented by ferulic acid (FA) in its trans form (90%) [44]. Their presence increases cell wall resistance by forming covalent cross-links (esterification) between AX chains through oxidative dimerization [25]. The most common dimerised forms are 5-5 -diferylate, 8-O-4 diferulate, 8-5 diferulate and 8-5 benzo diferulate [9,44]. Moreover, 92% of FA is found in bound form in the aleurone layer and accounts for 55 to 60% (*w*/*w*) of the whole

FA grain concentration according to Barron et al. [23], but can reach up to 70% (*w*/*w*) as per Brouns et al. [10]. This is highly beneficial, as FA is renowned for its antioxidant properties, which increase the antioxidant capacity of the aleurone layer [45].

Other phenolic acids include para-coumaric acid (PCA), which constitutes 67% (*w*/*w*) of the whole grain concentration and about 10% (*w*/*w*) of the total phenolics in aleurone cells [23]. It is also 3 to 5 times more concentrated in this bran layer than in the others but is not found in the endosperm [46]. Mainly in bound form (63% *w*/*w*), PCA is substituted to AX on average once every 90 arabinofuranose residues [10,22]. Sinapic and vanillic acids can also be found, mostly in conjugated form, amounting to 69 and 67%, respectively [10,23]. Traces of free syringic acid and bound flavonoids (apigenin and lutolein) have also been reported in the cell walls of the aleurone layer [10,47,48]. These phenolic compounds contribute to the total antioxidant capacity of the aleurone layer.


**Table 2.** Phenolic acids repartition in the wheat aleurone layer, bran and whole grain.

\* The aleurone layer and the bran amount respectively to 5–8% (*w*/*w*) and 5.5–9.5% (*w*/*w*) of the wheat grain [21]. All data were placed in the same unit to facilitate comparison.

Aleurone cell wall polysaccharides (AX and BG) are in different proportions depending on their position in the bilayered cell wall. Although the thinner inner layer seems to comprise more BGs, the thicker outer layer contains more AXs. FA residues in both layers appear unchanged, although there seem to be more phenolic acids in the anticlinal than periclinal walls [25,51]. The cell walls of the transfer cells (modified aleurone cells in the crease region) contain less FA, and PCA is present in negligeable amounts. Moreover, their AXs are more substituted (A:X = 0.6), and they contain higher amounts of BGs once they mature [31]. Thus, the origin of the aleurone cell wall is related to its function and impacts on the layer's technological and health properties.

However, some minor components, such as proteins, can be found in the cell walls of the aleurone layer, cross-linked to AXs, BGs or sometimes to hydroxycinnamic acids [9,52]. They amount to about 1% of the total aleurone cell walls and show a similar amino acid composition as the proteins found in the endosperm (Table 3). These proteins can be classified as either glycine (37–86%), proline (11–39%), or serine rich (up to 23%) [52].

Compared with other bran layers, the aleurone cell-wall comprises low, if any, amounts of cellulose, glucomannan and lignin (Table 1) [10,25,44,53].


**Table 3.** Proteins and amino acids repartition in the wheat aleurone layer, bran and whole grain.

\* The aleurone layer and the bran amount respectively to 5–8% (*w*/*w*) and 5.5–9.5% (*w*/*w*) of the wheat grain [21]. All data were placed in the same unit to facilitate comparison.

#### 2.2.2. Intracellular Medium

The intracellular medium represents 70% of aleurone cell dry mass [19]. It is composed of a large nucleus and aleurone grains (or granules). The latter are inclusion bodies, or vacuolar units of two types with a spherical structure (2–4 μm in diameter), surrounded by non-polar lipid droplets. Type I inclusions comprise phytic acid crystals. In those, dihydrophosphate traps minerals, including calcium, magnesium and zinc, by chelation, forming phytate complexes. This chelation of minerals results in a decrease in their bioaccessibility, which subsequently reduces the nutritional value of the aleurone intracellular medium unless they are released, for instance by the actions of endogenous enzymes (phytase) [9,57].

The other type of aleurone grain (type II) contains niacin (B3 vitamins) and proteins [9,10,58–60]. The proteins contained in type II inclusions represent 15 to 20% of total wheat grain proteins (dry mass) [5,61]. They are mostly storage proteins (mainly 7S globulins), despite the presence of numerous metabolic and defence enzymes [62]. The presence of lysine in these proteins is beneficial as it is a limiting amino acid in cereal products [9].

Overall, the aleurone grains provide numerous minerals accounting for 40% of total grain minerals, such as phosphate (80% in phytate form), magnesium, manganese, iron, potassium and sodium [10,11], as presented in Table 4. They are also a source of vitamins (Table 5). Indeed, thiamin and riboflavin are mostly present in this bran layer. Moreover, it is a source of lignans, especially syringaresinol, which are renowned for their antioxidant properties [10]. Significant levels of carotenoids (lutein and zeaxanthin) can also be found [63], as well as betaine and choline, which are twice as concentrated as they are in bran [64], as presented in Table 6. These bioactive components contribute to the nutritional value of the aleurone layer, and demonstrate the potential benefits of its addition to cereal products for the consumer.


**Table 4.** Minerals repartition in the wheat aleurone layer, bran and whole grain.

\* The aleurone layer and the bran amount respectively to 5–8% (*w*/*w*) and 5.5–9.5% (*w*/*w*) of the wheat grain [21]. All data were placed in the same unit to facilitate comparison.

**Table 5.** Vitamins repartition in the wheat aleurone layer, bran and whole grain.


\* The aleurone layer and the bran amount respectively to 5–8% (*w*/*w*) and 5.5–9.5% (*w*/*w*) of the wheat grain [21]. All data were placed in the same unit to facilitate comparison.

**Table 6.** Repartition of bioactive components in the wheat aleurone layer, bran and whole grain.



**Table 6.** *Cont.*

\* The aleurone layer and the bran amount respectively to 5–8% (*w*/*w*) and 5.5–9.5% (*w*/*w*) of the wheat grain [21]. All data were placed in the same unit to facilitate comparison.

In addition, the aleurone intracellular medium is rich in endoplasmic reticulum, mitochondria, membrane-bound vesicles and lytic vacuoles [26]. No starch granules are present in the aleurone cells despite their proximity to the starchy endosperm [51].

In both cell walls and intracellular medium, most compounds are susceptible to great variations in concentration. Indeed, the effects of genetics, cultivar, and culture and storage conditions must be considered, as they can influence the amount of aleurone's compounds, thus having repercussions on its technological, health and nutritional properties [11,69]. For instance, coloured wheat varieties will differ greatly in their phytochemical composition [70–72].

#### **3. Health and Nutritional Benefits**

As presented, the aleurone layer is a source of many bioactive compounds that can potentially exhibit nutritional and/or health benefits for consumers. The benefits related to the layer's composition are impacted by the natural variability of the components. Moreover, as the following studies have been performed on materials of different purity and in various forms due to non-identical transformations and processes of obtention, the concluding remarks must then be approached as potential beneficial effects, as they are related to the specific conditions of materials and processes.

#### *3.1. Digestibility and Colonic Fermentation*

When ingested, the aleurone seems to have better digestibility than the other bran layers, with an overall digestibility value of 30%. This might be related to the absence of lignin in the aleurone, which is known to impair digestion [10,20]. However, this layer shows very poor digestibility in the upper digestive tract, especially when it comes to mineral absorption [57]. Furthermore, the release of minerals from the food matrix may be hindered by the presence of DFs, which cannot be digested in this part of the human gastrointestinal system [9]. Their bioaccessibility is even more decreased due to the presence of phytates, by which they are complexed [9].

In contrast, the in vitro colonic fermentation potential of aleurone seems to be better than that of wheat bran [73]. In multiple studies, the consumption of aleurone led to the stimulation of microbial activity (*Bifidobacteria dorea* and butyrate-producing *Roseburia* spp.) in the caecum and colon, thus leading to a higher yield of propionate and butyrate. These important short-chain fatty acids (SCFA) are renowned for their health benefits, namely in cancer prevention [57,74,75]. A decrease in health-detrimental bacteria, such as *Bilophila*, *Escherichia* and *Parabacteroides*, was also reported [75].

These observed effects probably arise from the presence of DFs in the aleurone layer. For instance, AXs in the layer's cell walls can exert beneficial health effects by inducing the proliferation of healthy gut microbiota and providing a substrate for increased SCFA production. Additional effects could include increased faecal bulk and viscosity, accelerated transit time and potential binding to cancer-inducing molecules [34,35]. These prebiotic effects could be accompanied by the cholesterol-lowering activity of AXs [76]. Slightly

branched AXs are more easily degraded than AXs with a higher substitution degree, as they ferment more slowly or remain unfermented [57]. As the aleurone layer's AXs are poorly cross-linked, this might be beneficial when compared with other bran layers.

Nevertheless, the presence of DFs can also hinder the bioavailability of other bioactive compounds present in the aleurone layer by decreasing their bioaccessibility. Since most fibres in this layer are either only partially or not digested by the intestinal flora, most of these important compounds cannot be released for absorption [9]. This includes the absorption of minerals complexed with phytate. However, DF fermentation in the colon can increase minerals' colonic absorption by increasing the production of SCFA, in turn reducing the pH in the intestine, thus solubilising phytic complexes and releasing minerals [77].

#### *3.2. Health Benefits of Aleurone Consumption*

In various clinical studies, a diet rich in wheat aleurone has been shown to reduce the risk of cancer, cardiovascular disease, obesity and diabetes [12,78]. Symptoms such as hypertension and hyperglycaemia reduction have been reported after consumption of aleurone in the long term, which contributes to the prevention of obesity and hyperlipidaemia [79]. This can also be seen with a decreased low density lipoprotein (LDL) cholesterol level post-ingestion [80]. Moreover, aleurone ingestion seems to reduce the risk of colon cancer after fermentation in the colon by decreasing secondary bile acid production, inducing apoptosis and cell differentiation, as well as detoxification [81–84].

In human studies, a moderate intake of aleurone raised red blood cell folate levels, decreasing plasma homocysteine levels. This effect is beneficial since high plasma homocysteine is a risk factor for both cardiovascular disease and cancer via DNA damage [85–87]. Moreover, the authors concluded that the folate contained in the aleurone was then highly bioavailable and bioefficient [86].

However, Keaveney et al. [88] reported contrasting results, as the consumption of 50 g of a wheat aleurone fraction from Bühler A.G. increased the betaine plasma concentration by 2.5 times, but neither plasma tocopherols, folate nor choline levels were increased. Another study confirmed this effect of aleurone on metabolic risk factors; higher plasma betaine and related lower plasma homocysteine were observed [80]. Nonetheless, the most recent study reported that upon consumption of 27 g of aleurone daily by overweight and obese subjects, no significant changes were observed in health-related biomarkers, including plasma homocysteine and SCFA levels [89].

Studies on the effect of wheat aleurone consumption on inflammation markers have reported a decrease in pro-inflammatory tumour necrosis factor (TNF)-α (one of the most important cytokines in the immune system [90]) in LPS-stimulated U937 macrophages [91]. Moreover, a decrease in C-reactive protein (CRP) was also noticed (a biomarker of inflammation used in the prediction of coronary heart disease [92]), probably via a reduction in LDL-cholesterol levels. This could imply that aleurone consumption may change hepatic metabolism either by the action of its independent bioactive components or by their interaction [93].

Although the health benefits observed upon aleurone consumption in clinical studies are numerous, they are still hypothetical. As the studies did not involve the same test conditions and used different materials, they are thus not comparable. Moreover, it does not consider the natural variability in the composition of the aleurone layer, nor does it wholly consider the matrix effects of its consumption in a normal diet with other food products. In conclusion, the observed health benefits may arise upon aleurone consumption, but they should be referred to as potential effects.

#### *3.3. Antioxidant Capacity*

In addition to the aforementioned beneficial health effects, aleurone is also a source of many bioactive compounds with antioxidant activity. It is the bran layer with the highest total antioxidant capacity and can provoke a prolonged anti-inflammatory effect after consumption [91]. This is partly due to its high phenolic acid content and mainly allotted to FA, which accounts for 60% of the aleurone antioxidant capacity [45].

The consumption of 50 g of this milling fraction from Bühler A.G. has also been reported to increase the FA plasma concentration [94]. Furthermore, the latter phenomenon seems to be related to enhanced FA bioavailability [95], and more generally an increased phenolic acid bioavailability, which is apparently conserved even when the aleurone fractions are incorporated into bread [96].

Due to its resonance-stabilised phenoxy radical structure, FA can inhibit lipid peroxidation via superoxide scavenging [97,98]. Moreover, the presence of a methyl group in C3 enhances resonance stabilisation, thus making it a very stable antioxidant that does not initiate an oxidative chain reaction [99,100]. Although this activity is increased when FA is in its dimerised form (whereas it is mostly in monomer form in the aleurone layer, Table 2), it still exhibits better LDL inhibition than ascorbic acid, a powerful antioxidant [9,101].

However, the bioavailability of FA is decreased when in this dimerised form or if esterlinked to AX, the latter amounting to the majority of FA in the aleurone layer [45,73,102]. The low degradation of DFs by the microbiota could present better health effects than a high peak after ingestion since phenolic compounds are continuously released in the plasma [103]. Moreover, the bioavailability of such compounds could be improved by the addition of cell wall degrading enzymes that release them [104].

In addition to FA, other phenolic acids (mainly PCA and sinapic acid) also contribute to the total antioxidant capacity of the aleurone layer [12]. Bioactive components, such as phytate, phytoestrogens (including lignans), and anthocyanins, are also involved [77,105–107]. Vitamin E participates equally by quenching singlet oxygen and nitrogen-oxide radicals. More precisely, α-tocopherol works as part of an antioxidant network in breaking lipidsoluble chains [108]. Vitamin B6 also possesses antioxidant activity and thus participates in protection from oxidative stress [109]. The combined effects of these compounds present in the aleurone layers make them a valuable source of antioxidants, provided they retain this property after transformation (i.e., breadmaking).

#### *3.4. Bioactive Components of Aleurone and Related Potential Nutritional and Health Benefits*

The aforementioned effects have been reported for the consumption of aleurone as a whole, but the health benefits of the bioactive components it contains have also been investigated separately.

The presence of vitamins in the aleurone layer (Table 5) can be beneficial from both nutritional and health points of view. For instance, tocopherols are involved in the regulation of cell signalling and gene expression and are also known to delay the progress of degenerative diseases. The other type of E vitamins, tocotrienols, are involved in the prevention of neurodegeneration. They can also induce immune responses as well as lower cholesterol and prevent cancer [108]. Unlike E vitamins, folate (vitamin B9) is renowned for its biological activity in normal foetus neural tube development [11]. In addition, niacin is involved in carbohydrates and fats metabolism [110].

However, anthocyanins may exhibit anticancer properties [106,107], and lignans may lower cholesterol and show potential estrogenic activity [12,105]. Moreover, the benefits of betaine include its role as an osmolyte, remethylating total homocysteine, therapeutic agent for non-alcoholic fatty liver disease, and lipotrop [80].

Lastly, the aleurone layer contains an essential amino acid: lysine (Table 3). Its presence is beneficial for the end-product nutritional value since lysine is a limiting amino acid in cereal grains [9]. In addition, the presence of arginine in aleurone proteins can also be valorised due to its role in vascular dynamics and endothelial function, which can improve [12].

As these compounds are present in the aleurone layer, this adds to its value, provided that they remain present in the end-product and conserve their potential health-benefit properties through the process.

#### **4. Potential of the Aleurone Layer as an Ingredient in Bread- and Cereal-Based Products**

*4.1. Extraction of the Aleurone Layer and Its Challenges*

The aleurone layer's potential can only be revealed if it is first extracted. However, no universal process yet exists. In addition, as the grain's milling properties (friability) are dependent on its constituents, themselves related to the culture conditions and genetic background, the process must constantly adapt to the raw material. Therefore, this represents a major hurdle in the utilisation of the aleurone layer and in the exploration of its properties.

Another challenge arises from its composition. The aleurone layer, although botanically part of the endosperm, is considered by millers as belonging to the bran fraction. Since it is tightly adhered to the pericarp, it is usually removed from the endosperm during conventional milling. This tight adherence to seed coats also makes it difficult to separate the aleurone from the rest of the bran [10,12]. Consequently, most existing processes for the retrieval of the aleurone layer start with bran material. Multiple procedures have been patented [14,111–114] and two companies have mainly been known to produce and commercialise aleurone-rich flour: Bühler A.G. and Cargill Limited (through Horizon Milling with the GrainWise brand). However, it seems that most existing processes result in the obtention of aleurone-rich flour that is either not highly concentrated [13] or low yielding [14]. A summary of the composition of aleurone-rich fractions issued from the existing processes mentioned below, available in the literature, is shown in Table 7. As shown in the table, the purity of the obtained aleurone-enriched fraction depends greatly on the extraction process performed, which explains the discrepancies observed in their composition when compared to that of the pure hand-isolated aleurone layer. These differences in composition may also arise from the use of different analytical methods among publications.

Most of these processes extract the aleurone layer from bran components by dryfractionation, a succession of mechanical or physical unitary steps [10,115,116]. Many researchers first aim to dissociate the different bran tissues, which can be conducted by grinding. They include a separation step that then enables sorting out the particles according to their size, mass, density or electrostatic properties [116]. The obtained milling fraction can thus be added to basic wheat flour to enrich it with the aleurone layer [115].

The aleurone layer is extensible, similar to the intermediate strips of wheat bran, and it has an elastoplastic rheological behaviour. Its mechanical characteristics are impacted by the degree of feruloylation of its AX, particularly by the presence of FA dehydrodimers [44]. Hence, the mechanical stress generated by dry fractionation processes first affects the aleurone cell walls, which crack, allowing the cell contents to be released [117]. According to Rosa et al. [118], the velocity of phytic acid release could thus be used as a marker to estimate aleurone cell opening.

Electrostatic separation of aleurone from other bran tissues is an interesting process since the aleurone layer presents unique electrostatic properties compared to the other strips. However, this process can be influenced by multiple parameters, such as particle size, composition, microstructure and moisture content [68].

The main advantage of physical or mechanical extraction methods is that they do not require the use of chemical products that can interact with the matrix and decrease the product's purity and phytochemicals' functionality [116]. Compared with wet processes (chemical and enzymatic treatments), they also enable higher energy efficiency [10,116]. However, the succession of unit operations may impact the antioxidant and secondary metabolites of the aleurone layer [3]. Moreover, grinding generates various particles from bran tissues of different sizes and densities, which are hard to differentiate, hence the reported end-product's low purity [116].

This type of process was used in the patented method by Stone and Minifie [14], who first used hammer-milling in wheat bran containing 34% of aleurone cells, followed by sieving, electrostatic fractionation, and a final separation through an electric field. A 95% purity of aleurone cells was obtained with a 10% yield [14,68]. Nonetheless, alternative methods

exist: humidification then micro-grinding of wheat bran with a friction roller mill [114]; sequential pearling cycles in a vertical abrasive polishing machine [74]; centrifugal impact milling [68]; ultrafine grinding and electrostatic separation [119]. Different outcomes have been reported with these processes, with varying aleurone purity and yield.

However, there are limited studies on the extraction of the aleurone layer by wet processes (chemical and enzymatic treatments). For instance, the maceration of wheat bran in chemical reagents, such as organic solvents, has been tested but not in an industrial scale [10].

Nonetheless, dry and wet processes can be coupled. A patented method isolated aleurone by successive steps of cleaning, steaming, stabilising, roller-milling, sieving, fine grinding and air-classifying. However, the end-product still contained 36.5% starch, demonstrating low purity [87,120]. In addition, the patent deposited by Kvist et al. [121] subjected wheat bran to several enzymatic treatments, wet milling steps, sequential centrifugation, and ultrafiltration. Other researchers coupled successive steps of milling, sieving, air classification and centrifugation with benzene-carbon tetrachloride mixtures at laboratory scale [10].

Although many experiments have been conducted and sometimes patented to extract and isolate the aleurone layer, the challenge of measuring end-product purity has arisen. Thus, researchers have defined biochemical markers to differentiate grain parts. These biochemical markers can be used to determine the extent to which the aleurone layer is extracted from other grain components. Starch, phytate, p-coumaric acid, alkylresorcinols and FA trimer are used to estimate the proportion of the endosperm, aleurone cell content and cell-walls, intermediate layer and outer pericarp, respectively [46,65,122]. However, the relative amount of grain tissue can only be calculated when compared to the reference values. The latter were values of the same parameters from pure isolated tissues of identical wheat cultivars. Thus, it limits their utilisation for characterisation since pure fractions are obtained from hand-isolated tissues, a long and labourous process. Moreover, these markers are susceptible to natural variability among wheat cultivars due to the culture conditions and genetic background [50,123]. As an alternative to these biochemical markers, microscopy analyses can be performed to estimate the purity of the extracted fractions [115].


**Table 7.** Composition of aleurone-enriched fractions issued from different extracting processes in literature.


**Table 7.** *Cont.*

\* Origin of products, as listed below. **1a**: Bühler A.G. (55–70% aleurone purity); **1b**: Bühler A.G. (75–90% aleurone purity)—patented method [111]; **2**: Goodman Fielder Milling and Baking Pty. Ltd. (90% aleurone-rich flour with 10% of waxy maize starch); **3**: Cargill Limited and Horizon Milling (Grainwise); **4**: Jiaxing Zhishifang Food Science and Technology Co. (Shandong, China), 14% wb. All data were placed in the same unit to facilitate comparison.

#### *4.2. Application to Breadmaking*

#### 4.2.1. Aleurone Bread Nutritional Profile

According to past reviews and experiments, there are many benefits to incorporating an aleurone-rich flour into bread and bakery products, starting with an improved nutritional profile of the end-products. This amelioration is related to increased DF and protein (mainly albumin and globulin) content at the expense of readily digestible carbohydrates [13,15,119,131,136]. The enhancement of minerals, including phosphate, magnesium, manganese and iron, and bioactive compounds such as phenolic acids, antioxidants, phytoestrogens and sterols, also increase the value of the obtained end-products [16,136]. This improved composition confers the end product a nutritional profile similar to that of whole wheat products [128], while equally making it a good source of fibre [12]. However, the nutritional benefits of aleurone-rich products are accompanied by increased phytate content, which is known for its antinutritional effect [57].

#### 4.2.2. Aleurone Bread Dough Characteristics

Despite these beneficial nutritional properties, the incorporation of the aleurone layer for breadmaking leads to changes during dough formation, which affects the sensory attributes of the end-product. With its high DF content (Tables 1 and 7), the aleurone layer impairs dough hydration properties. The AXs and BGs contained by the aleurone layer compete for water with the proteins forming the gluten network, thus increasing the water absorption of the dough and retarding the dough development time [136,137]. The water retention capacity is also affected by fibres that take up a large amount of water (3.5 to 6.3 times their weight for WEAX and 6.7 to 9.9 for WUAX) by binding through hydroxyl groups, resulting in a longer mixing stability due to the alteration of the gluten structure [6,131,136].

In addition, the presence of these fibres has a diluting effect on the starch granules. Damaged starch content is then decreased as well as the falling number. The latter effect is further reduced by the increase in α-amylase activity in the presence of calcium. Indeed, this metalloenzyme requires calcium for its performance, which is provided by the aleurone layer (Tables 4 and 7). In addition, these observed properties seem to increase in relation to the aleurone-rich flour dosage [136].

The effect of aleurone incorporation on starch also influences the pasting properties of dough. Multiple studies have shown a decrease in peak viscosity, as well as in the retrogradation of dough [131,132]. This might not only stem from the presence of fibres that interfere with starch granule swelling and increased amylase activity but also from the combined effect of other aleurone constituents. For instance, the presence of fat and FA can also impact pasting properties, in addition to an already low starch content [132,138].

Nevertheless, the aleurone dough exhibits higher Rapid Visco Analyzer (RVA) parameters, revealing a strong gel ability greater than that of whole wheat flour [131]. According to Bucsella et al. [136], this could be due to the swelling of fibres, which form a strong gel despite the lower starch content. This gel is described as being more resistant to heat and mechanical stress.

The gluten network can also be strengthened following the addition of aleurone-rich flour to bread dough (up to 40%). However, according to Mixolab (Chopin) measurements, the dough development time is increased due to the presence of fibres that compete for water and hinder gluten network formation by intercalating between the proteins, resulting in a more heat-stable and stress-resistant dough [131].

This increase in dough stability is also depicted by firm elastic-like behaviour due to the stronger gluten complex [131,136]. The increased protein content (albumin) and the strengthening effect of AX binding to gluten via the oxidative dimerization of FA also contribute to these observed effects [139]. Instrumentally, this translates into an increase in dough stability and break time, as well as delayed weakening [136].

Despite the aforementioned beneficial traits observed due to the aleurone components, most of them are dose dependent. Excessive addition of aleurone-rich flour to the dough can lead to deleterious effects on dough rheology.

#### 4.2.3. End-Product: Aleurone Bread Characteristics

The addition of aleurone-rich flour to bread-making has an impact on end-product quality, although the results of the researcher's findings are contradictory. This may be related to the aleurone-enrichment level, the purity of this material as well as the bread formulation process, compiled in Table 8.


**Table 8.** Overview of existing aleurone-enriched bread formulation processes in literature.

\* Basic flour type as described in the corresponding articles, defined as commercial flour.

Some report a decrease in loaf volume, accompanied by reduced height and increased weight [13,15–17]. For instance, Bagdi et al. [15] observed a diminution of 27% of the specific volume for a bread prepared with 100% of aleurone-rich flour compared to a control white bread, as well as a reduction in loaf height of 13%. Using the same breadmaking process (ICC Standard Method 131), Bartalné-Berceli et al. [16] obtained a height decrease of 15 and a 7.2% weight increase with a bread containing 25% of aleurone-rich flour compared with a control white bread. These tendencies are further incremented with a higher aleurone flour input (50%), where almost half of the height was decreased and 3.6% of the weight was increased compared with the control.

Other studies have observed a higher loaf volume than white bread upon aleurone incorporation or have found insignificant changes. The texture in these experiments was also reported to be softer than white bread, which means that the crumb was less dense [12,136,140]. Indeed, Tian et al. [140] described an increase of 40.91% in bread specific volume using aleurone-rich flour (modified GB/T 35869-2018 procedure with 54.11% of aleurone layer content). However, this beneficial effect could be attributed to the presence of hemicellulases (at 40 mg/kg) that enable the formation of WEAX from WUAX. Similar results were obtained from breads made with a sourdough preparation (MSZ-6369-8:1988) incorporating an aleurone fraction, even though the observed volume increase was not significantly different from that of the control bread [136]. Breads made from straight dough and sponge dough processes with additives containing aleurone-rich flour (20%) also show this beneficial trait [12].

Overall, it seems that these beneficial effects could be related to the presence of hemicellulose-degrading enzymes—either endogenous (sourdough) or exogenous (as an additive)—or additives or a special breadmaking process, each enabling the revelation of the aleurone layer's full technological potential.

Unlike for the volume and texture of the breads, the appearance of the end-product is equivocal: the crumb colour (measured by colorimetry) is darker than white bread, even brownish, which can be a limiting factor for some consumers. However, it is still lighter than whole wheat products [13,15,57].

As for the taste of the bread, diverging results also occur. Whereas some report a flavour similar to that of white bread, especially when a long fermentation process takes place [12,128]; others describe a bread that is more bitter and sour, even rye-like [13,15]. In addition to the last finding, Amrein et al. [57] outlined a gritty mouth-feel, which is a limiting factor for the consumers of the study. However, the smell of the products is reported to be more intense and sour [13,15].

Overall, bread made of aleurone-rich flour in different proportions showed similar properties to that of white breads but with the nutritional profile of whole wheat breads. More thorough experiments on the relevance of aleurone addition in cereal products compared to other wheat kernel layers should be conducted, as the diversity of breadmaking methods and starting materials is great in the existing studies. Nevertheless, the results are still contradictory and lead to either a decrease or increase in end-product consumer acceptability. The use of special breadmaking technologies or additives, such as cell wall degrading enzymes, could thus reveal the aleurone layer's full technological potential.

#### 4.2.4. Underlying Mechanisms

Many of the adverse or positive technological effects due to the addition of aleuronerich flour to bakery products stem from its unique composition and, more specifically, its high protein and DF content. Indeed, studies investigating the effect of DF, AX and bran incorporation into bakery products showed similar properties to those described in aleurone-enriched products.

Most experiments on this subject describe that the addition of fibres to bread dough increases dough development time, water absorption and strength. However, it also seems to weaken the dough's tolerance to mixing and fermentation [141]. This results for most studies in a reduction in loaf volume [5,141–144], an increase in crumb firmness [142–144], and a darkened crumb appearance [141–144].

Hypotheses exist to explain the mechanisms underlying these results, which corroborate those of aleurone-rich products. First, fibres with their high water binding capacity might compete for water with starch and gluten, thus keeping wheat proteins from sufficient hydration for the formation of the gluten network [5,141,142,145–149]. Another explanation is that fibres dilute gluten, thus affecting its gas-holding capacity [5,141,142,145–149]. Nonetheless, this impairment in gas retention that causes a loss of loaf volume could also be due to the shortened and lowered resistance to dough extension upon DF addition, which increases the concentration of soluble cell wall materials and disrupts the gluten network [142,150].

In addition to the previous mechanisms of action that hint at a physical mode of action, a chemical hypothesis also exists that states that FA linked to DFs could mediate AX– AX and AX–protein cross-linking (through FA–tyrosine linkages), thus impacting gluten properties [146–149]. This would be possible in the presence of oxidants or enzymes (such as laccase and peroxidases) that provoke the dimerization of FA, thus creating covalent linkages between AX chains. Moreover, this dimerization increases the water retention capacity of AXs, which directly affects the gluten network [6].

More specifically, studies conducted on the addition of AXs to bakery products could be helpful in understanding the mechanisms underlying the aleurone-enriched bread properties, since they represent a large part of this layer. Similar to the addition of general DFs, an increase in the water absorption of the dough is reported due to the high water retention capacity of AXs, which increases dough consistency [151]. According to Berger and Ducroo [6], to reach the same dough consistency as the control dough on the Brabender farinograph, 0.5 to 2% of additional water should be incremented per percent of AX supplemented.

As for the negative effects on gluten network formation due to AX addition, they could stem from the steric hindrance of the increased batter viscosity that limits components mobility, thus decreasing the formation of gluten aggregates and starch entrapment in its matrix [151]. Nonetheless, the observed effects are not as important as the extent of their addition and their molecular size, but most importantly, depend on the breadmaking quality of the initial flour used for the experiments [152].

Furthermore, the water extractability of AXs is also a determining factor in the adverse effects it causes on bread and bakery products. For instance, water-unextractable AXs (WUAX) seem to generate more deleterious side effects upon their addition than water-extractable AXs (WEAX). This might explain the contradictory results with aleurone applications since it mainly contains WUAXs, which can be transformed into WEAXs during breadmaking.

The use of WUAXs in bakery products is often reported with breads of lower volume, coarser crumb and higher firmness [39]. To explain this phenomenon, there are three hypotheses: (i) WUAXs form physical barriers for wheat proteins during dough development [39]; (ii) these AXs form intrusions in the gas cells during fermentation [39,153]; (iii) the WUAXs compete for water with the gluten network, thus impairing its formation and leading to a fracture effect that increases dough resistance to extension [6,7,39]. The latter hypothesis is believed to be more accurate since a correction of dough hydration (2% per percentage of AXs added) improves its extensibility [6].

In contrast, the use of WEAXs in bread dough yields contradictory results, even though they are mostly beneficial. Globally, a finer and more homogenous breadcrumb is depicted, with a bread that is softer [39]. The loaf volume is also impacted, but contradictory results are obtained. The observed higher volume is usually obtained with the use of high molecular weight WEAXs [39,153,154]. The underlying mechanisms explaining these effects include an increase in liquid film stability and thus in dough aqueous phase viscosity [7,39]. Moreover, WEAXs of higher molecular weight can form a secondary network, weaker than gluten, which enforces the latter and stabilises it through the dimerization of FA and by physical entanglement, either of gluten or between WEAXs [39]. By increasing the dough's gas retention capacity, the resulting breads become higher [6]. According to [6], the higher the WEAXs molecular weight, the highest beneficial effect is observed.

#### *4.3. Application to Other Food Products*

Although the incorporation of aleurone-rich flour into bakery products have mostly been studied, its application to other product categories exists. For instance, Cargill Limited [128] developed cereal flakes and extruded snacks with 35% of aleurone-rich flour, as well as high-protein bars containing 20%. Ready-to-eat cereals enriched in aleurone were also studied by Byrne [155].

Other applications entail aleurone-enriched pasta and noodles. Whereas the former was described as healthier than wholemeal spaghetti due to higher protein, fat, and DF content, its consumer acceptance was decreased. Although it showed improved quality characteristics (lower water uptake, higher cooked pasta firmness, higher tensile strength and lower stickiness), the darker, more intense, bitter and sour taste of the pasta influenced consumer appreciation of the product [138]. The second used a combination of aleuronerich flour and transglutaminase, which resulted in noodles with less cooking loss and the best sensory evaluation when compared to traditional noodles [130].

Finally, Yang et al. [134] used aleurone-enriched flour and cell wall degrading enzymes for the production of Chinese buns. They found that the action of enzymes promoted the WEAX content while also increasing the water availability to the gluten-forming proteins, resulting in softer dough, especially when enzyme activities were combined.

#### **5. Optimization of the Aleurone Layer's Potential**

Despite the aleurone layer's nutritional and potential health-benefit properties, its use on the market is scarce. This might be related to the hurdles in its extraction, as well as the rheological issues it has with the end-product. To overcome the latter and obtain a product with improved nutritional traits without alteration of its technological properties, aleurone-rich flour can be subjected to different processes, whether physical, chemical or biological. By modifying the aleurone constituents and, most specifically, its DFs, which are mainly responsible for the observed adverse effects, the technological potential of the aleurone layer could thus be optimised. Moreover, these processes could also enhance the health benefits associated with the intake of this bioactive milling fraction.

Many physical treatments, whether thermal, non-thermal, dry or wet, focus on the particle size reduction in the material, as it can lead to many beneficial effects. It includes an improved antioxidant and bile acid-binding capacity, a greater bioavailability of phenolic compounds and vitamin E, a higher production of colonic SCFA, and a faster digestion rate, which can increase transit time and decrease faecal bulking [13,19]. For this purpose, different techniques can be performed. Whereas milling refers to the process of separating the endosperm (known as white flour) from the bran (outer layers and germ) [156], grinding uses shear stress and compression to reduce the particle size [157]. Both can be used to obtain a particle size below that of aleurone cells (around 50 μm) to release their content [117, 158].

The results of those studies concluded that as the particle size of the material decreased, phytate extractability was enhanced, and phenolic acids were released [117], thus improving the mineral bioaccessibility of aleurone [158]. The hydration properties (namely water holding and binding capacity) of the milling fraction are also reduced, which subsequently negatively affects its fermentability [127]. In addition, an increase in conjugated and free FA post-grinding has been observed, as well as the release of aleurone intracellular compounds (soluble proteins, vitamin E and phytic acid). The combined effects of improved bioaccessibility of antioxidant compounds, as well as the greater exposure of phenolic moieties, result in an enhanced antioxidant capacity of the modified aleurone ingredient [9, 118,126,127,159].

In addition to physical treatments, the use of chemical processes to modify, inter alia, the solubility of fibres is of equal interest for the improvement of the aleurone layer's technological potential. Their benefits lie in the fact that they are the only ones that can provoke a polymerisation of the materials. However, the final product can be of low purity, with a high degree of hydrolysis and modified functional groups [160,161].

Experiments conducted by Bagdi et al. [162] revealed that hydroxyl radical treatment, •OH oxidation and cross-linking of AXs extracted from aleurone-rich flour modified its bile acid-binding capacity and could even enhance the cholesterol-lowering effect of AXs. Besides these improved health effects, the use of an alkaline treatment on the aleurone's WUAXs could be beneficial, as it was reported to maintain the fibre's molecular weight while increasing its water solubility [154]. Moreover, as carboxymethylation of wheat bran enhanced its health properties (increased total antioxidant capacity, total reducing power, Fe2+ chelating capacity and DPPH radical scavenging capacity), the use of this chemical mean of treatment on the aleurone layer could equally raise its value [163].

Nonetheless, most of the treatments performed on aleurone-rich flour in the literature have been conducted by biochemical means. Rhodes and Stone [124] studied the effect of combined methods, namely ultra-fine grinding and enzymatic treatment with xylanase and feruloyl esterase. Upon those treatments, changes in the aleurone layer structure were observed, as well as an increase in free FA. Moreover, associated beneficial health effects, such as reduced mouse body weight and improved glucose metabolism, led the authors to conclude that a partial depolymerisation of the wheat aleurone cell wall could be favourable for their metabolism.

These conclusions are consistent with those from Rosa et al. [127], which upon xylanase and feruloyl esterase treatment of aleurone, described a release in FA, both in free and conjugated forms. Furthermore, although SCFA production was not improved, the fast metabolization of FA by the colonic microbiota promoted the production of FA colonic metabolites.

Moreover, these findings are complementary to those from Rosa, et al. [126], whom upon enzymatic treatment of an aleurone-rich fraction, also found an increase in the release of cell-wall bound phenolic acids in conjugated and free forms. However, this was also associated with an increase in the antioxidant capacity of the fraction due to both the released components and the increase in their bioaccessibility.

However, experiments conducted by Vangsøe et al. [133] demonstrated that the enzyme susceptibility to the aleurone cell wall AXs was correlated to its arabinose-to-xylose ratio. More specifically, it seems that xylanase activity was enhanced in the presence of AXs with lower substitution degrees, which are mainly found in the cell walls of the aleurone layer.

Overall, xylanase has mainly been used for the transformation of the aleurone layer in previous experiments. As those enzymes are capable of hydrolysing WUAXs, known for their negative technological effects, this explains their success. Nevertheless, complete hydrolysis should not be performed, as the dough's stickiness increases, and the crust darkens. Instead, the release of WUAX-hydrolysed fragments of high molecular weight is preferred. By increasing the dough's viscosity, decreasing the fermentation gas migration rate and thus improving its retention, the dough shows an improved tolerance to fermentation and oven baking. The final products are reported to increase in volume from 15 up to 30% [6].

In conclusion, only a few modification processes were performed on the aleurone layer despite the large range of known physical, chemical and biological processes [18,157]. As the transformation of wheat bran has been extensively studied, it would be beneficial to rely on the obtained results to undertake new experiments focusing on improving the milling layer's technological and health benefits in bakery products.

However, the experiments were performed on starting materials of different purity, which explains the disparity in the results. It would then be interesting to work on the aleurone-extracting process to make it efficient and reproduceable. Little else can be accomplished on this part besides aiming to erase the traits that are not beneficial to the rheology of the end-products.

#### **6. Conclusions**

The aleurone layer is a major bran component that exhibits numerous nutritional and potential health benefits. Multiple processes exist and have been patented to extract this layer, but it seems that a choice must be made between the extraction yield and the purity of the fraction. Although its addition to bakery products has been studied and claimed to be beneficial from a nutritional point of view, some technological negative effects still arise, such as a decrease in loaf volume. These are partly due to the presence of a large number of DFs, mainly water-unextractable. However, those side effects could be reduced, and the aleurone-rich flour ingredient's functionality improved following some transformation processes preceding its addition to a dough matrix.

Despite all the known benefits of the aleurone layer, as proved in the experiments conducted by researchers, only a few applications exist nowadays with even fewer producers and furnishers. This may be related to the challenges of its extraction—more specifically, to the fact that no universal process exists. Even though it may be created, the natural variability of the grain's constituents and the impact of the environment would still require this process to be adapted to each raw material, as the grain's technological properties would change in accordance with its composition. In addition, the negative rheological traits observed upon aleurone addition into bakery products may also explain the absence of its use on the market, unless its presence is not communicated as such.

The appeal of the aleurone layer may also be dwindled by its more expensive extraction process when compared with wholegrain flours and other by-milling products. The latter can be nutritionally relevant for the consumer and would be easier to obtain. However, as demonstrated in this review, the health benefits obtained upon aleurone consumption may be higher (i.e., antioxidant capacity, reduced risk of cancer, etc.). In addition, the presence of the outer parts of the grain could increase the concentration of pesticides residues and impair the technological and organoleptic properties of the cereal end-products, thus favouring the use of the innermost part of the bran which shows those undesirable effects to a lesser extent [18].

Perhaps investigating other types of applications, as well as transformations pre- and post-extraction of the milling fraction, could instigate increased interest in this nutritionally beneficial aleurone layer. The lack of an actual market would also make it an opportunity for a company to renew its interest in the aleurone as an ingredient, not unlike bran fibres, provided its rheological properties are improved. Investigating the limits of aleurone layer incorporation to the formulation of other bakery products than bread in a design of experiments format could be interesting.

**Author Contributions:** Conceptualization, F.B., A.S. and T.A.; writing—original draft preparation, L.L.; writing—review and editing, L.L., F.B., A.S. and T.A.; supervision, A.S. and T.A.; project administration, F.B. and A.S.; funding acquisition, A.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Foricher les Moulins, grant number CIFRE 2021/1733.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data is contained within the article.

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

#### **References**


MDPI St. Alban-Anlage 66 4052 Basel Switzerland www.mdpi.com

*Foods* Editorial Office E-mail: foods@mdpi.com www.mdpi.com/journal/foods

Disclaimer/Publisher's Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Academic Open Access Publishing

mdpi.com ISBN 978-3-0365-9509-2