*Article* **Breadmaking Performance of Elite Einkorn (***Triticum monococcum* **L. subsp.** *monococcum***) Lines: Evaluation of Flour, Dough and Bread Characteristics**

**Andrea Brandolini 1,\*, Mara Lucisano 2, Manuela Mariotti 2, Lorenzo Estivi <sup>2</sup> and Alyssa Hidalgo <sup>2</sup>**


**Abstract:** Einkorn flour, rich in proteins, carotenoids, and other antioxidants, generally has poor breadmaking value. In this research, the composition and technological characteristics of the flours and breads of two elite einkorns (Monlis and ID331) and a bread wheat (Blasco), cropped in four different environments, were evaluated. The einkorns confirmed better flour composition than bread wheat for proteins (on average, 16.5 vs. 10.5 g/100 g), soluble pentosans (1.03 vs. 0.85 g/100 g), and yellow pigment (10.0 vs. 1.0 mg/kg). Technologically, they had better SDS sedimentation values (89 vs. 66 mL), lower farinographic water absorption (52.6 vs. 58.8%), and a similar development time, stability, and degree of softening. Viscoelasticity tests showed lower storage and loss moduli and more prevalent elastic behaviour for Blasco, while rheofermentographic tests showed an anticipated development time (120.8 vs. 175.0 min), higher maximum height (73.0 vs. 63.0 mm), and superior retention coefficient (99.1 vs. 88.7%), but a lower CO2 total (1152 vs. 1713 mL) for einkorn doughs. Einkorn breads were bigger than the control (736 vs. 671 cm3); crumb pores percentage was similar, but medium-size pores were scarcer. Finally, a 52-h shelf-life trial demonstrated that einkorn bread had a softer texture, maintained for a longer time, and a slower retrogradation than the control. Therefore, choice of appropriate varieties and process optimisation allows the production of excellent einkorn breads with a superior nutritional value and longer shelf life.

**Keywords:** colour; farinograph; rheofermentograph; viscoelastic behaviour; bread shelf life; crumb porosity

#### **1. Introduction**

Einkorn (*Triticum monococcum* L. subsp. *monococcum*), a diploid hulled wheat (2 *n* = 2 x = 14) with high protein [1–3], lutein, and antioxidants content [4,5], is considered a cereal with a poor baking attitude and a dough characterized by excessive stickiness. The farinographic tests often show evidence of scarce stability and a high degree of softening, while breadmaking yields small volume loaves, due to reduced leavening and easy collapse of the dough [6,7]. However, a great variation for breadmaking quality exists within the einkorn gene pool, and selected ecotypes with high SDS sedimentation values (>70 mL), good farinographic stability (360–720 s), and a limited degree of softening (20–50 UB) yielding breads with volumes similar or even higher than wheat breads have been identified [6–8]. Additionally, because their doughs present poor tolerance to mechanical processing and prolonged fermentation, gentle processing at a low speed and for a short duration (3–4 min) have been proposed to improve loaves' volume [9]. The use of sourdough fermentation has also been suggested to improve texture, volume, and shelf life of einkorn bread [10]. Crust shape and colour are similar to those of wheat loaves, but

**Citation:** Brandolini, A.; Lucisano, M.; Mariotti, M.; Estivi, L.; Hidalgo, A. Breadmaking Performance of Elite Einkorn (*Triticum monococcum* L. subsp. *monococcum*) Lines: Evaluation of Flour, Dough and Bread Characteristics. *Foods* **2023**, *12*, 1610. https://doi.org/10.3390/ foods12081610

Academic Editor: Saroat Rawdkuen

Received: 8 March 2023 Revised: 28 March 2023 Accepted: 3 April 2023 Published: 10 April 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/).

the crumb shows an enticing yellow tinge [11] for the high carotenoids content [5]. Furthermore, the lower endogenous enzymatic activity [12,13] allows for better preservation of antioxidants during storage [14,15] and processing [16,17], and limits heat damage [18], thus safeguarding the favourable nutritional characteristics of einkorn-derived products.

The potential health benefits of einkorn foods have been the subject of recent investigations. Although einkorn, like other wheats, such as barley and rye, is not suitable for consumption by people with celiac disease [19–21], einkorn may be better suited than other *Triticum* species for patients with chronic dysmetabolic diseases. For example, einkorn bread consumption leads to more favorable metabolic responses and greater satiety compared with standard wheat breads [22]. Investigating the effects of einkorn bread on the intestinal physiology and metabolism of pigs, Barone et al. [23] observed a lower postprandial insulin rise after einkorn consumption compared with bread wheat consumption; furthermore, the intestinal ecosystem was enriched in health-promoting bacteria. Einkorn's anti-inflammatory effects were also recorded in cultured cells [24].

Although einkorn breads are often characterized by an inferior technological quality compared with wheat breads, their enticing taste, aroma, attractive color, and healthpromoting properties suggest that they may be a worthy addition to the increasing assortment of available products. Therefore, the aim of this research was to evaluate several breadmaking quality facets of two elite einkorn wheats (ID331 and Monlis) and, as control, of one bread wheat cultivar (Blasco); the two einkorn accessions were selected because in previous unpublished tests repeatedly showed the best breadmaking attitude. Hence, all-around information about the breadmaking properties of these two elite einkorns will allow the baking industry to develop new leavened products through adopting the most appropriate processing approaches. To this end, the composition and technological and rheological properties of their flours, doughs, and breads were assessed throughout the breadmaking process; furthermore, the shelf life of the bread loaves was investigated.

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

#### *2.1. Materials*

Two breadmaking-quality einkorns (ID331 and Monlis) and a breadmaking-quality bread wheat (cv. Blasco) were cropped in four different environments, two under conventional management (Sant'Angelo Lodigiano, LO, and Montelibretti, Rome, labelled as SAL and ROMA) and two under organic management (Sant'Angelo Lodigiano, LO, and Leno, BS, i.e., SALbio and LENObio). Some relevant agronomic parameters of the four environments are summarized in Table S1. The accessions were planted in 10 m2 plots according to a Randomized Complete Block Design with three replications. For weed control, the herbicide Ariane II (Clopiralid 1.8% + Fluroxypyr 3.6% + MCPA 18.2%; Dow AgroSciences, Milan, Italy) was applied just before heading to the conventional management plots, while the organic trials were manually weeded. After machine harvesting, by mid-July (Table S1), the Monlis and ID331 kernels were de-hulled with an Otake FC4S thresher (Satake, Japan), which is a step that was not necessary for the free-threshing bread wheat Blasco. All seeds were stored at 5 ◦C until further processing.

#### *2.2. Methods*

#### 2.2.1. Kernels

Kernel moisture was determined by a GAC2000 moisture analyzer (Dickey–John, Auburn, IL, USA) and corrected to 15% for the einkorns and 16% for Blasco (harder texture) by overnight tempering. Afterwards, the hectoliter weight (kg/hL) was determined with a GAC2000 instrument (Dickey–John, Auburn, IL, USA). The seeds were milled with a Bona–GBR laboratory mill (Bona, Monza, Italy), which separates flour from bran and germ, and the milling yield (% flour/kernels *w*/*w*) was computed.

#### 2.2.2. Flour and Dough

Flour particles' size was determined by sifting 100 g of flour for 5 min through a sieve with a 125 μm mesh; the ≥125 vs. <125 ratio was computed. Flour characteristics were then assessed according to the following methods: moisture (AACC 44-15.02 [25]), ash content (AACC 08-03.01 [25]), protein content (N × 5.7; AACC 46-10.01 [25]), yellow pigment (AACC 14-50.01 [25]), dry gluten content (AACC 38-12.02 [25]) with a Glutomatic (Perten, Hägersten, Sweden), Falling number (AACC 56-81.03 [25]) with a 1550 Falling Number (Perten, Hägersten, Sweden), SDS sedimentation test (a breadmaking attitude predictor [26]), starch and amylose contents and α-amylase activity with Megazyme Assay Kits (Megazyme International Ireland Inc., Bray, Ireland), and pentosan and soluble pentosan contents (colorimetric method) [27]. Dough mixing properties were evaluated with a Brabender farinograph (Brabender OHG, Duisburg, Germany) using a 50 g mixer according to the ICC method 115-D [28]. Briefly, after adjusting the maximum consistency of the dough to a fixed value (500 Brabender Units, BU) by altering the quantity of water added, the test was run for 12 min. The parameters recorded were water absorption (amount of water added to set the curve at 500 BU, expressed as percentage of flour at 14% moisture), development time (time in minutes between the origin of the curve and its maximum value i.e., 500 BU), stability time (difference in minutes between the time to the maximum and time when the top of the curve falls below 500 line), and degree of softening (difference in BU between the maximum value and the value at the end of the test).

Dough leavening properties were assessed by rheofermentographic tests and were performed with a Chopin F3 Rheofermentometer (Chopin SA, Villeneuve-La-Garenne, France), according to [29]. The indices recorded from the curves were Hm (mm; maximum development of the dough), h (mm; height of the dough at the end of the test), CO2 total (mL; total gas production during the test), CO2 loss (mL; total gas loss during the test), CO2 retained (mL; gas retained by the dough during the test), and retention coefficient (% CO2 retained/CO2 total).

To better understand the fundamental properties of the doughs, rheological analyses were carried out according to [30] with a Physica MCR300 Rheometer, supported by US200 v. 2.5 software (PHYSICA Messtechnic GmbH, Ostfildern, Germany). Measurements were carried out at 25 ◦C using a corrugated parallel plate system (diametre: 2.5 cm) at a gap of 2 mm and a special humidity cover (H-PTD 150) to prevent moisture losses. Frequency sweep tests were performed over the range 0.1–50 Hz at 0.1% strain on doughs prepared at the same consistency (500 UB) after a resting period of 30 min to equilibrate stresses. The selected 0.1% strain value was obtained from preliminary amplitude sweep tests performed in the range of 0.01–100% strain, at a constant frequency of 1 Hz, to determine the linear viscoelastic region (LVR) of the sample. Each test was performed at least three times. Data were analyzed with US200/32 v. 2.50 rheometer software (PHYSICA Messtechnic GmbH, Ostfildern, Germany) and the value of storage modulus (G , Pa) and loss modulus (G, Pa) at 1 Hz and tanδ (ratio between G and G ) were computed.

#### 2.2.3. Bread

Two breads per accession were produced according to method 10-10 B [25] with minor modifications [6]. Bread weight (g) was determined with a LP5200P balance (Sartorius AG, Göttingen, Germany), bread volume (cm3) by rapeseed displacement (method 10-05.01 [25]), and bread height (mm) with a caliper; specific volume (cm3/g) was computed as volume-to-weight ratio.

Bread quality changes during storage were monitored on eight breads per accession, prepared only from SAL trial flours. Two breads were immediately analyzed, while the others were packaged in food paper bags and stored at 25 ◦C and 60% relative humidity in an HC 0020 air-conditioned cell (Heraeus, Hanau, Germany). Four times (t0: 0 h, t1: 24 h, t2: 30 h, t3: 52 h), two bread loaves of each accession were weighed to evaluate weight decrease during storage, then they were transversely cut to obtain three uniform 25-mmthick slices. Crumb moisture was assessed on two slices while water activity (*aw*) analysis

was performed on the third central slice with an AQUALAB apparatus (Decagon Devices Inc., Pullman, DC, USA).

Crust color and crumb color of the three slices of each sample at t0 were assessed in the CIELAB space using a Minolta Chroma Meter CR 210 (Minolta Camera Co., Osaka, Japan) with a standard illuminator C. Additionally, images were captured using a flatbed scanner (Scanjet 6300c; Hewlett Packard, Palo Alto, CA, USA) in 256 grey level at 300 dots per inch and were processed using a dedicated software (Image Pro-Plus 4.5.1.29, Media Cybernetics Inc., Rockville, MD, USA). The parameters evaluated were density red (R), density green (G), density blue (B), and density mean. At the same time, crumb porosity was determined by assessing the number, area, diameter, and shape of the pores, their size distribution (three categories: C1: 0.1–1 mm2, C2: 1–5 mm2, C3: >5 mm2), and % of pore area.

Crumb texture characteristics were assessed by Texture Profile Analysis on the three central slices from each sample at each storage time with an HD.plus Texture Analyzer TA double-column dynamometer (Stable Micro System, Godalming, UK) connected to a Texture Exponent 32 recording system version 4.0.8.0 (Stable Micro System, Godalming, UK). The operating conditions adopted were: load cell 500 N, crossbar speed 2 mm/s, compression plate diameter 36 mm, sample compression up to 40% thickness, and 25 s waiting time between 1st and 2nd compression cycles. The parameters derived from the Force/Time curve were: young modulus (N/mm2), hardness (N), springiness or elasticity (adimensional), cohesiveness (adimensional), and chewiness (i.e., hardness × cohesiveness × springiness, N).

#### 2.2.4. Statistical Analysis

A two-way analysis of variance (ANOVA) was carried out for most parameters, using environment and genotype, or genotype and storage time, as factors. A one-way ANOVA was performed on the results of the color and porosity parameters, assessed on the breads prepared from the SAL flours. When significant differences (*p* ≤ 0.05) were found, Fisher's lowest significant difference (LSD) was computed. Both ANOVA and LSD testing were performed with the software STATGRAPHICS plus v.4 (STATPOINT Technologies Inc., The Plains, VA, USA). Means and standard errors were determined using the software Excel® (Microsoft Corporation, Redmond, DC, USA).

#### **3. Results**

#### *3.1. Kernels*

Figure 1 reports the values of hectoliter weight, flour yield, and proportion of flour <125 μm, while their behavior in the four locations is depicted in Figure S1. The results in the different locations (Figure S1) suggest that the hectoliter weight of the bread wheat was generally higher than that of the two einkorns, while the flour yield and the percentage of flour particles <125 μm were higher in the einkorns than in the bread wheat Blasco. It is interesting to note that an elevated hectoliter weight, besides indicating healthy caryopses with a compact endosperm typical of "hard" type wheats, such as Blasco, usually forecasts high flour yields [31]. However, despite the smaller seeds and consequent superior teguments proportion [2], as mentioned above, the flour yield of both einkorns seems superior to that of the bread wheat Blasco. This peculiar result was also spotted by Borghi et al. [6] in their survey of 25 einkorns and by Corbellini et al. [7] in their study of 24 einkorns, whose flour yields ranged between 53.0 and 64.4% and were similar to the soft wheat Veronese (59.0%), but were superior to the hard wheat Pandas (50.0–52.6%). The higher extraction rate may be linked to the extra-soft texture of einkorn kernels [1,7], which makes them more easily grindable, and originates a very fine flour [6,32,33]. In this study, indeed, the percentage of einkorn flour particles smaller than 125 μm was, on average, 86.0 g/100 g (in the range of the 84.9–91.7 g/100 g reported by Borghi et al. [6]), while it was 57.8 g/100 g for Blasco. Probably during milling the harder bread wheat kernels broke into larger fragments, thus originating, in comparison to einkorn, a coarser flour. No major differences between ID331 and Monlis were observed.

**Figure 1.** Bread slices prepared from flours of bread wheat Blasco, einkorn ID331, and einkorn Monlis.

#### *3.2. Flour and Dough*

A two-way ANOVA (Table S3) highlighted the existence of significant differences among accessions and among environments, as well as their interactions for moisture, ash, protein, total starch, amylose, soluble pentosans, total pentosans, and yellow pigment content. In all instances, the genotype was the most relevant factor. The results of the four environments, not discussed in this article, are summarized in Table S3.

The ash content of Blasco (Table 1) was significantly lower (0.58 g/100 g DM) than that of the two einkorns (on average, 0.63 g/100 g DM), as evidenced also by other authors [1,32,34]. Additionally, Blasco was less rich in protein and richer in starch than the two T. monococcum accessions.


**Table 1.** Kernel and flour parameters (mean values ± standard error) of two einkorns (ID331 and Monlis) and one bread wheat (Blasco). *n* = 8: four environments × two repetitions.

Different letters in the same row mean significant differences (*p* ≤ 0.05) among samples following LSD test. \* The kernel data were not subjected to statistical analysis.

Einkorn is well known for its high protein levels [1,6,7,32,33,35], which in part is due to genetic factors and in part to its smaller kernels having an inferior endosperm-toexternal layers ratio and hence, a superior incidence of the protein-rich aleuronic layer [2]. Nevertheless, despite the considerable difference in protein concentration, the amino acid composition of all wheats is similar, with lysine representing the main limiting factor [36–38].

Conversely, the plump Blasco kernels were richer in starch than the smaller, proteinrich einkorn seeds (79.1 vs. 69.0 g/100 g). However, the amylose proportion of total starch was similar between Monlis and Blasco, and was only slightly inferior in ID331 (Table 1). On

the other hand, Blasco had similar pentosan and inferior soluble pentosan contents than the einkorns (2.52 vs. 2.57 g/100 g and 0.82 vs. 1.03 g/100 g, respectively). The total pentosan concentration of Blasco was within the variation (2.0–3.0 g/100 g) described for bread wheat [39,40], while the soluble pentosan content fell into the range (0.49–1.23 g/100 g) observed in *T. aestivum* [27,41,42]. To the best of our knowledge, no information about pentosans content in *Triticum monococcum* is available in the literature. Interestingly, when pentosans content in flour increases, the retrogradation of starch decreases, due to their steric interference with the intermolecular associations of starch [43].

A peculiar characteristic of einkorn flour is its yellowness, due to the abundant presence of carotenoid [4,5]. This is confirmed by the yellow pigment content of both ID331 (8.72 mg/kg DM) and Monlis (11.30 mg/kg DM), which are about ten times larger than the value of Blasco (0.98 mg/kg DM) and in the range reported for whole meal flours of a collection of einkorns [1]. Remarkably, the two einkorns tested had concentrations that were twice higher than those commonly reported for yellow durum wheats [44].

The two-way ANOVAs (Table S2) verified the existence of highly significant differences among genotypes, environments, and their interactions for dry gluten content, Falling number, α-amylase activity, and SDS sedimentation value; the genotypic effect was largely the most relevant factor. The results of the four environments, not discussed in this article, are summarized in Table S3 for completeness of information.

As already hinted by the protein content, the dry gluten content was higher in ID331 and Monlis than in Blasco (Table 1). Falling number and α-amylase activity are inversely correlated traits [45]: the higher the α-amylase activity, the shorter the falling time. In this study, Falling Number results were all well above those values suggesting modest pre-germination phenomena (200–300 s) [46]. The significantly higher levels of both FN and α-amylase activities obtained for einkorn flours can be explained with their superior ash content, being that the amylolytic enzymes are mainly localized in the germ and in the external regions of the kernels.

The SDS sedimentation test underlined the good breadmaking propension of the two einkorns tested and of the bread wheat control. Einkorn generally has a poor breadmaking capacity [32,33], but some good accessions are reported [1,6,7,47]. Indeed, ID331 and Monlis are among the most suitable *T. monococcum* to prepare leavened products, and their SDS quality was superior even to the good breadmaking wheat variety Blasco.

The farinographic results (Figure S1) suggested differences between the bread wheat flour and the two einkorn flours only for water absorption (Table 2), as Blasco needed more water (around 6%) than ID331 and Monlis to reach optimum dough consistency. Development time, stability, and degree of softening did not look different. Borghi et al. [6] found that most of the einkorn accessions that they investigated had very low stability (<1 min), but some genotypes reached 2.0–4.5 min. Low development time, poor stability, and a strong degree of softening are indices of a weak flour, with scarce resistance to the mechanical action of kneading and are therefore unsuitable for preparing leavened products. Nevertheless, ID331 and Monlis showed a breadmaking attitude similar to Blasco, a cultivar fit for the manufacturing of high-quality leavened bakery products.

The unreplicated results of the rheofermentographic parameters are summarized in Table 2 and are presented across the four locations in Figure S2. This analysis is helpful to evaluate changes in dough during the leavening phase that can be linked to breadmaking quality [29]. The results in Figure S2 suggest that Monlis and ID331 reached a significantly higher dough development during the test in a shorter time in comparison to Blasco, maybe due to the smaller size of flour (see above) and starch granules [48], whose superior surfaceto-volume ratio may favor enzymatic reactions. They also exhibited a limited CO2 loss, which determined a significantly higher CO2 retention coefficient, probably attributable to a more compact gluten network due to the higher protein content of the two einkorn flours.

The two-way ANOVA (Table S2) of the viscoelasticity parameters hinted to significant differences between Blasco and the two einkorns, both for the limits of the linear viscoelastic region obtained from the strain sweep test, and the storage and loss moduli

obtained from the frequency sweep test. For completeness of information, the results of the four environments, not discussed in this article, are summarized in Table S3.

**Table 2.** Breadmaking parameters (mean values ± standard error) of two einkorns flours (ID331 and Monlis) and one bread wheat flour (Blasco). *n* = 8: four environments × two repetitions (Brabender, *n* = 4: four environments).


Different letters in the same row mean significant differences (*p* ≤ 0.05) among samples following LSD test. \* The farinograph and rheofermentograph data were not subjected to statistical analysis.

The evaluation of the region of linear viscoelasticity of a sample is an important step: when materials are tested in the linear range, their viscoelastic behaviors do not depend on the magnitude of the stress, the magnitude of the deforming strain, or the rate of application of the strain [49], but on their intrinsic features. The length of the linear viscoelastic region (LVR) can therefore be used as a measurement of dough stability: einkorn doughs remained in the linear viscoelastic region over greater strains than Blasco dough, indicating the presence of a stronger network. Indeed, for Blasco a drop in G LVR started to occur above the 0.33% strain and became larger at a higher strain, indicating a progressive disorganization of the dough structure beyond this deformation level. A 0.1% strain for the subsequent frequency tests was therefore adopted, as within the LVR of all the samples. Although all the dough samples that were analyzed had the same farinographic consistency (500 BU), significant differences were observed among the samples. Generally, the mechanical spectra of all the samples exhibited a solid-like behavior, with G always being higher than G. Einkorn doughs presented significantly higher values of both moduli, with a prevalence of the viscous behavior, as indicated by the higher values of the damping factor (G/G ).

#### *3.3. Bread*

#### 3.3.1. Characteristics

The two-way ANOVA for bread volume, height, and specific volume (Table S2) scored significant genetic effects, including especially highly significant environmental effects and their interactions. Interestingly, ID331 and Monlis outcompeted Blasco for loaf volume, height, and specific volume (Table 2), indicating that it is possible to have an einkorn flour with the same technological properties of good quality bread wheat. Cross-sections of the loaves prepared with the flours from the LENObio trial are depicted in Figure 1.

Due to the higher amount of water added following indication of the farinographic test (56.0% vs. 53.4% to reach 500BU), the bread produced from Blasco flour had a significantly higher loaf weight (Table 2), crumb humidity, and aw (Figure 2, t0) than the einkorn samples. Monlis exhibited the highest loaf volume (on average, 740 cm3), followed by ID331 (732 cm3) and Blasco (671 cm3). Similar values for loaves obtained from selected einkorn flours are reported [6,7]. These results were reflected in the breads' specific volumes: Monlis originated the breads with the highest value, that is the "lightest" breads (Table 2).

**Figure 2.** Moisture, water activity (aw), weight decrease, and texture parameters of breads from two einkorns (ID331 and Monlis) and one bread wheat (Blasco) during storage (h). Error bars represent standard error.

The one-way ANOVA (Table S4) indicated the presence of significant differences among genotypes for color indices in the cases of *a*\* (crust and crumb) and *b*\* (crumb). Table 3 reports the average values of *L*\*, *a*\*, and *b*\* evaluated on the crust and crumb of each bread sample. The crust color showed a low variation, with very similar *L*\*, *a*\*, and *b*\* values among samples, as observed also by D'Egidio et al. [31]. A difference was instead evident for crumb color: Blasco had significantly higher *a*\* values, but a much inferior *b*\* (yellow index) than both the einkorn genotypes. This reflects the differences in the yellow pigment content of the raw materials (Table 1), which was particularly abundant in the carotenoid-rich einkorns and gives their breads a characteristic and enticing golden yellow color. Table 3 also shows the average values of the color parameters evaluated by Image Analysis; the ANOVA (Table S4) highlighted significant differences among the samples for all parameters: Blasco had significantly lower R (red intensity) and G (green intensity), and had significantly higher B (blue intensity) than the einkorn samples.

**Table 3.** Crust and crumb colorimetric indices (mean values ± standard error) of breads prepared from flours of two einkorns (ID331 and Monlis) and one bread wheat (Blasco) cropped at Sant'Angelo Lodigiano under conventional management. *n* = 12: three sections from two slices of two loaves.


Different letters in the same row mean significant differences (*p* ≤ 0.05) among samples following LSD test.

Bread can be considered a solid food foam because the gas developed during fermentation is trapped in its solid matrix. Information about shape, size, and distribution of the pores is very important in order to identify the influence of processing conditions on the quality of the final products and to devise methods to manufacture products with peculiar textural characteristics [50]. Image Analysis was used in this study to assess the porosity of the different breads by determining the number of pores and their geometrical features (area, diameter, shape). All the identified pores were classified according to their size and the results are shown in Table 4, while the ANOVA is presented in Table S5. The Blasco breads presented as many small pores (0.1–1 mm2) as the einkorn breads, while the number of medium-sized pores (1–5 mm2) was significantly higher (*<sup>p</sup>* ≤ 0.05). Among einkorns, Monlis bread porosity was characterized by a greater number of large pores (>5 mm2), with an average area of 9.21 mm2, significantly higher than the Blasco and ID331 pores of the same class.

Information about the shape and regularity of the pores is obtained by considering that the larger the value, the greater the irregularity. Indeed, the smallest pores had the most regular and uniform shape (range: 1.78–1.89), while the large pores ranged from 2.34 to 2.60, indicating that as their size increased, their form became more irregular and less homogeneous. In our case, ID331 bread had the most regular pores for all the three size classes: its bread presented a porosity similar to Blasco, but was characterized by a more regular shape, thus making the porosity of the crumb more homogeneous.

**Table 4.** Porosity (mean values ± standard error) of the breads prepared from two einkorn (ID331 and Monlis) and one bread wheat (Blasco) refined flours cropped at Sant'Angelo Lodigiano under conventional management. *n* = 12: three sections from two slices of two loaves. Porosity classes: C1 = 0.1–1 mm2, C2 = 1–5 mm2, C3 = >5 mm2. Different letters within pores class in the same column indicate significant differences (*p* ≤ 0.05) among samples following LSD test.


The number of pores and their average area is reflected in the percentage of relative area occupied in the slice and, ultimately, in the total porosity area of the bread. For Blasco, the surface occupied by the small pores was similar to that found in the einkorn samples, while the surface associated with medium-sized pores was significantly higher. Overall, pores area was highest in Blasco bread (810.57 mm2), mainly because of the greater number of medium-sized pores. In Monlis bread, the large pores contributed considerably to the total porosity, as also confirmed by the less homogeneous internal loaf structure. Nevertheless, the loaves obtained from Monlis were the ones with the largest specific volumes, probably due to the high presence of these large pores. Despite the distribution of pores in the three classes being different among the three samples, the percentage of pores, computed as ratio between total alveolate area and total bread slice area, was similar among the three accessions.

The mechanical and geometrical characteristics of bread influence its behavior during oral processing [50]. To assess these features in a reproducible way, Texture Profile Analysis is used to create controlled stresses and to measure the mechanical characteristics of foods. Figure 2 shows the average values of the texture parameters of the three different breads. The two-way ANOVA (Table S6) stressed the existence of significant differences (*p* ≤ 0.05) between accessions only for young modulus (consistency index), hardness, gumminess, and chewiness. Storage times were always significant and so was the genotype x time interaction (except for elasticity). At t0, Blasco showed the highest consistency, hardness, gumminess, and chewiness, and Monlis exhibited the lowest values, while in general ID331 had results in line with Monlis. These data indicate how the bread loaves obtained from Monlis were the softest ones and were characterized by the best chewability, despite the finer and more regular porosity and the higher moisture content of Blasco breads, which are factors that are conventionally related to a softer structure. Possibly, the superior softness of the einkorn breads may be related to the lower starch content of their flours.

#### 3.3.2. Changes during Storage

Figure 2 also shows the evolution of aw, weight, young modulus, and texture parameters of the breads' slices during storage. Blasco bread lost most of its moisture in the first 24 h after the production, while the decrease was more progressive for the einkorns. Interestingly, Blasco bread did not present significant water activity changes during storage, while the einkorn breads showed a significant increase after the first 24 h, followed by a plateau. This increase could be due to a reorganization of starch molecules to a different composition of starch with particular reference to the amylose content whose molecules, involved in the phenomenon of retrogradation, have a greater tendency to associate with each other and form hydrogen bonds, releasing water, or even to a reorganization of the

protein matrix [51], more abundant in einkorn flours. As expected, the weight of all the breads significantly decreased during storage, but was steeper for the two einkorns after the initial 24 h.

The Young modulus of the Blasco bread at t0 was higher than that of the einkorn breads. During storage, this index increased for all samples, but was faster in the bread wheat. This behavior is also reflected in the hardness: in fact, the loaves from einkorn flours showed, already at t0, lower hardness values (ID331: 3.26 N; Monlis: 2.44 N) compared with Blasco (4.35 N), and this difference amplified over time, as the kinetics of the bread from bread wheat were more accelerated than those of einkorn breads. The speed of the process, as measured by the value of the angular coefficient of the regression curves, was 0.24 N/h for Blasco and 0.12–0.14 N/h for the two einkorn samples. Hence, the einkorn breads maintained their softness for a longer time and their retrogradation was slower than that of the bread wheat.

Elasticity and cohesiveness decreased with similar trends in einkorn and wheat breads over the storage time. For that reason, bread chewiness (i.e., the energy required to chew a solid food) was mainly related to the hardness values, which was much higher for wheat bread, particularly during storage. Since the superior softness of the einkorn breads cannot be attributed to a higher moisture content (indeed, Blasco breads were characterized by the highest moisture), such behavior may be related to the lower starch content of einkorn flours and/or to the lower presence of amylose in the samples. Additionally, the slightly higher amylase activity of Monlis and ID331 may have contributed because amylases decrease starch retrogradation, diminish rigidity of the starch gel network, and limit starch– protein interactions [52,53]. Their anti-staling effect is also related to the hydrolyzation of amylopectin and the production of soluble low-molecular-weight branched-chain polymers, which are less prone to retrogradation and influence water movement and accessibility [54].

#### **4. Conclusions**

This study demonstrates the possibility of obtaining einkorn breads with technological properties comparable to those of breadmaking-quality bread wheat. However, to achieve this result, high-quality genotypes are necessary. Furthermore, as the quality characteristics of the breads are influenced by processing conditions, einkorn bread manufacturing should utilize short mixing and leavening times to avoid overstressing the doughs. The choice of appropriate einkorn varieties and optimization of the breadmaking process will allow the production of breads with appropriate technological characteristics, endowing the market with an innovative cereal-based food possessing a superior nutritional value and longer shelf life.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/foods12081610/s1, Figure S1: Hectoliter weight, flour yield, and proportion of flour <125 μm of flours of ID331, Monlis, and Blasco cropped in four locations (SAL, SALbio, LENObio, and ROMA); Figure S2: Reofermentographic results of flours of ID331, Monlis, and Blasco cropped in four locations (SAL, SALbio, LENObio, and ROMA). Table S1: Cropping environments and field management information; Table S2: Two-way ANOVA (\* *p* ≤ 0.05; \*\* *p* ≤ 0.01; \*\*\* *p* ≤ 0.001) of composition, rheological analysis, and breadmaking test parameters; Table S3: Environmental values (mean ± s.e.) of flour and bread parameters of two einkorns (ID331 and Monlis) and one bread wheat (Blasco); Table S4: One-way ANOVA (\* *p* ≤ 0.05; \*\* *p* ≤ 0.01; \*\*\* *p* ≤ 0.001) of crust and crumb color parameters; Table S5. One-way ANOVA (\* *p* ≤ 0.05; \*\* *p* ≤ 0.01; \*\*\* *p* ≤ 0.001) of porosity; Table S6: Two-way ANOVA (\* *p* ≤ 0.05; \*\* *p* ≤ 0.01; \*\*\* *p* ≤ 0.001) of texture profile analysis parameters.

**Author Contributions:** Conceptualization, A.B. and A.H.; methodology, A.B., M.L., M.M. and A.H.; software, M.M. and A.H.; validation, A.B., M.L., M.M., L.E. and A.H.; formal analysis, A.B., M.L., M.M. and A.H.; investigation, A.B. and L.E.; resources, A.B.; data curation, A.B. and A.H.; writing—original draft preparation, A.B., M.M. and A.H.; writing—review and editing, A.B., M.L., M.M., L.E. and A.H.; supervision, A.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Data Availability Statement:** The data presented in this study are available in the article and the Supplementary Materials.

**Acknowledgments:** We thank Alessandro Cozzi and Erica Landoni for their technical assistance.

**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* **Processing and Bread-Making Quality Profile of Spanish Spelt Wheat**

**Ana Belén Huertas-García 1, Carlos Guzmán 1,\*, Maria Itria Ibba 2, Marianna Rakszegi 3, Josefina C. Sillero <sup>4</sup> and Juan B. Alvarez <sup>1</sup>**


**Abstract:** Spelt wheat (*Triticum aestivum* L. ssp. *spelta* Thell.) is an ancient wheat that has been widely cultivated for hundreds of years. Recently, this species has been neglected in most of Europe; however, the desire for more natural and traditional foods has driven a revival of the crop. In the current study, eighty-eight traditional spelt genotypes from Spain, together with nine common wheat cultivars and one modern spelt (cv. Anna Maria) were grown during a period of two years in Andalucia (southern Spain). In each, several traits were measured in to evaluate their milling, processing, and end-use quality (bread-making). The comparison between species suggested that, in general, spelt and common wheat showed differences for most of the measured traits; on average, spelt genotypes had softer grains, higher protein content (14.3 vs. 11.9%) and gluten extensibility (alveograph P/L 0.5 vs. 1.8), and lower gluten strength (alveograph W 187 vs. 438 <sup>×</sup> <sup>10</sup>−<sup>4</sup> J). In the baking test, both species showed similar values. Nevertheless, the analysis of this set of spelt genotypes showed a wide range for all measured traits, with higher values than common wheat in some spelt genotypes for some traits. This opens up the possibility of using these materials in future breeding programs, to develop either new spelt or common wheat cultivars.

**Keywords:** wheat quality; genetic resources; ancient wheat; bread-making

#### **1. Introduction**

Since the 1960s, the importance of plant genetic resources has gradually increased, as shown by the development of the "International Treaty on Plant Genetic Resources for Food and Agriculture" [1]. The lack of genetic diversity in crops, globalization, and climate change have shown how easy is for any pathogen or plague to quickly spread around the world [2]. This could be a threat for food security, as modern agriculture is increasingly focused on fewer crops and fewer varieties within each crop [3]. At the same time, a greater awareness of the need to use more sustainable and environmentally friendly agronomic techniques, together with the problems associated with global change, have boosted the search for alternative gene sources, which is one of the strategies used to develop more resilient cultivars under the conditions of global warming. In this context, ancient wheats and wild-wheat relatives, which have adapted to be grown in marginal zones under extreme conditions [4], are considered to host interesting and unexploited genetic variability that could be used in modern wheat-breeding programs to develop more resilient cultivars.

**Citation:** Huertas-García, A.B.; Guzmán, C.; Ibba, M.I.; Rakszegi, M.; Sillero, J.C.; Alvarez, J.B. Processing and Bread-Making Quality Profile of Spanish Spelt Wheat. *Foods* **2023**, *12*, 2996. https://doi.org/10.3390/ foods12162996

Academic Editors: Grazia Maria Borrelli and Donatella Bianca Maria Ficco

Received: 7 July 2023 Revised: 4 August 2023 Accepted: 6 August 2023 Published: 9 August 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/).

Among these ancient wheats, spelt (*Triticum aestivum* L. ssp. *spelta* Thell., 2n = 6× = 42, AuAuBBDD), originally obtained from the natural hybridization between emmer wheat (*T. turgidum* spp. *dicoccum* Schrank em. Thell., = 4× = 28, AuAuBB) and *Aegilops tauschii* ssp. *strangulata* Coss. (2n = 2× = 14, DD) in the Fertile Crescent (Near East), is, today, the most cultivated species, and several spelt cultivars have been bred in order to improve their productivity [5,6]. For this reason, some ancestral traits like the hulled grain or the semi-branching rachis, have been modified through crosses with common wheat [7–10]. Consequently, two different types of spelt are now present in the farmers' fields: the traditional or pure spelt, and the modern spelt derived from hybridization with common wheat [5]. The variability of these two types of spelt is notably different, with the traditional spelt holding a greater genetic variability than modern spelt. Nevertheless, more studies comparing the variability of modern and traditional spelt are needed.

On the other hand, several studies have suggested the exceptionality of the Iberian spelt (pol. *ibericum* Flaskb.) compared to the rest of the European spelt (Bavarian group, pol. *bavaricum* Vav.), including the old studies of N.I. Vavilov [11–14]. Therefore, while European spelt could derive from a secondary hybridization event between emmer wheat and hexaploid wheat, the Iberian spelt would have originated from the first hybridization event between emmer wheat and *Ae. tauschii* ssp. *strangulata* in Asia [11–14]. Furthermore, the spelt crop in Spain has been scarce until recent times and mainly based in traditional materials. The appearance of modern spelt in Spain is recent, and only two cultivars have been developed since 2018: cvs. Anna Maria and Viso. However, the traditional Spanish spelt stored in germplasm banks is abundant [14]. The current trend with this crop opens up the opportunity to add value to these old materials for their use in the current agricultural context, both as pure spelt and as a source of novel genetic diversity to develop modern spelt [14,15] or common wheat cultivars.

Before this new trend of growing old crops, spelt had already been used in breeding programs as a source of resistance genes for some wheat diseases [16–26]. Now, in the context of the renewed interest in artisan and "more natural" food, spelt is used as raw material for the making of food products (bread, biscuits, pasta, pancakes, etc.) which are present in many bakeries but also in large retailers. For this reason, studies on the processing quality of this crop have increased in importance [27–34].

Wheat processing quality is complex and varies depending on the wheat processor (millers or bakers) and on the final products. Wheat millers for example, value grain size, test weight, and texture, which are associated with the flour yield, and grain protein content, which is a partial indicator of wheat functionality [35,36]. Bakers, on the other hand, value the quantity and quality of the protein in flour, and the rheological properties of the dough made with it. For these reasons, the evaluation of new wheat materials must consider the quality requirements of all wheat processors including both millers and bakers.

Most of the studies conducted on the processing quality of ancient wheat only included a limited number of accessions [37] which reduced their ability to gain a clear understanding of the potential in terms of wheat processing, of such species. In general, the quality characteristics of these ancient wheats have been compared with modern wheat (common wheat–*T. aestivum* L. ssp. *aestivum*). Although this could be right, these data should be evaluated with caution. Many of these ancient wheats have been revived as a modern wheat substitute, and consequently, to establish the quality characteristics of modern wheat as the reference could be inadequate and lead to underestimating all ancient wheats. Obviously, spelt is not common wheat. The development of new cultivars of these ancient wheats should be complementary to modern wheat in the context of the new agri-food industry.

The main objective of this study was the evaluation of a wide collection of Spanish traditional spelt accessions for some grain and technological quality traits, together with their comparison with one modern spelt (cv. Anna Maria) and several common wheat cultivars widely cultivated in Andalucia (southern Spain).

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

#### *2.1. Plant Material and Field Trials*

Eighty-eight accessions of Spanish traditional spelt, together with ten modern wheat cultivars (nine common wheats and one modern spelt) were used (Table S1). These materials were planted in a randomized complete block design with two replicates during 2019–2020 and 2020–2021 crop seasons in Cordoba (Andalusia, Spain). Due to the high number of materials, the plot size was small (0.13 m2) and, consequently, the grain yield was limited for some accessions that could be only evaluated from small-scale tests.

The 88 traditional spelt accessions were selected according to their grain protein composition and origin from two wide collections originally provided by the National Small Grains Collection (USDA, Washington, DC, USA) and Centro de Recursos Fitogeneticos (INIA, Madrid, Spain) [38,39]. Of the 10 modern cultivars used as control, 9 were commercial Spanish common wheat cultivars commonly grown in the south of Spain (cvs. Antequera, Arthur Nick, Conil, Galera, Montemayor, Rota, Santaella, Setenil, and Tejada), and fall into different categories within the Spanish quality groups, depending on their performance in each environment. Cvs. Antequera, Conil, Galera, Rota, and Tejada often fall within the Spanish quality group 1 (strong gluten wheat for mechanized bread-making), while cvs. Arthur Nick, Montemayor, Santaella, and Setenil produce grains that are usually classified as quality groups 2–3 (strong–medium gluten wheat for semi-mechanized bread-making).The modern spelt (cv. Anna Maria) is a modern spelt cultivar obtained from hybridization between pure spelt and common wheat.

#### *2.2. Grain and Flour Quality Traits*

Thousand kernel weight (TKW, g) and test weight (TW, kg/hL) were obtained using the SeedCount SC5000 digital imaging system (Next Instruments, Australia). The grain (GPC, %) and flour (FPC, %) protein content were determined by near infrared spectroscopy (NIR Systems 6500, Foss, Hillerød, Denmark) based on AACC official methods 39-10.01 and 39-11.01, respectively, which were calibrated based on method 46-11.02 [40]. Grain hardness was measured on samples of 100 kernels with the single-kernel characterization system (SKCS) (Perten Instruments, Springfield, IL, USA) [40]. The polyphenol oxidase (PPO) activity was measured by absorbance at 475 nm according to Anderson and Morris [41], and expressed in Ug<sup>−</sup>1min−1.

For the milling, the two field replicates of each genotype were mixed in order to obtain enough flour. The grain samples were processed applying AACC method 26-95 [40]. All samples were milled into flour using a Brabender Quadrumat Senior mill (CW Brabender, Duisburg, Germany) and flour yield (%) was calculated.

Measurement of sodium dodecyl sulfate (SDS) sedimentation volume (ml) was carried out according to Dick and Quick methodology [42] with the modifications introduced by Peña et al. [43].

#### *2.3. Alveographic and Baking Traits*

The dough tenacity (P), extensibility (L), tenacity/extensibility ratio (P/L), tenacity/swelling index ratio (P/G), elasticity index (Ie), and strength (W) were determined by AACC method 54-30.02 using a Chopin alveograph [40]. Due to the limited flour available, dough rheological properties were measured only on 7 common wheat cultivars and 80 spelt accessions.

The bread-making process was conducted on 4 common wheat cultivars and 50 spelt accessions (only of those genotypes for which there was enough flour available to perform the test), using the direct dough method (AACC method 10-10.03), and loaf volume (cc) was determined by rapeseed displacement using a volume meter [40].

#### *2.4. Statistical Methods*

The comparison between both species sets was carried out for each trait analyzed by Student's *t*-test. A Pearson correlation analysis was carried out among the grain, flour, and rheological traits within the Spanish spelt set.

For the spelt set, data were analyzed by an analysis of variance (ANOVA) using genotype, year, and genotype × year as variation sources. Because cv. Anna Maria represented the current trend in spelt, the cv. Anna Maria data for each measured traits were used as reference to evaluate and compare the values of each Spanish traditional spelt genotype.

All statistical analyses were performed using Statistix software (version 9).

#### **3. Results**

#### *3.1. Comparison among Species*

The data obtained for all materials evaluated (Tables S2 and S3) were grouped according to the species (common wheat vs. spelt) in order to compare the two groups. The mean values of each set were analyzed by Student's *t*-test (Table 1).

**Table 1.** Average values of the common wheat and spelt groups (averaging genotypes and years) and result of the *t*-test done between both values.


TW, test weight; TKW, thousand kernel weight; GPC, grain protein content; FPC, flour protein contents; SDS-sed, sodium dodecyl sulfate sedimentation test; PPO activity, polyphenol oxidase activity; P, dough tenacity; L, dough extensibility; G: swelling index; W, dough strength; and Ie, elasticity index. \*\*\*, \*\*, \*: significant at 99.9, 99, and 95%; ns: not significant.

For grain or flour components, the differences between both species were, in general, small, but significant. The thousand kernel weight (TKW) of spelt was slightly higher than common wheat; however, spelt grains showed lower test weight (TW) values, probably due to the morphology of their grains that have, on average, an elongated shape. This had no influence on the flour yield, although the grain hardness, clearly lower in spelt, could also have played a role on the flour yield.

The protein content was higher in spelt, both in grain and in flour. But this scarcely influenced the gluten strength measured by the SDS-sedimentation test, because the Student's *t*-test analysis indicated that the differences between both species were not significant (Table 1). On the contrary, there were highly significant differences for polyphenol oxidase

(PPO) activity, for which the spelt group exhibited the double mean activity of the common wheat group.

The alveographic parameters showed that while common wheat presented dough with high tenacity (P) and low/moderate extensibility (L), the spelt genotypes showed, in general, low to moderate tenacity (P) and high extensibility (L). In any case, the dough strength (W) was larger in common wheat than in spelt (Table 1). Nevertheless, within both sub-sets there were no significant differences in the bread-making quality of the two groups (loaf volume), although the mean value of spelt was 30 cc higher than that of common wheat.

#### *3.2. Variability for Grain and Flour Quality Traits in Spelt*

When the comparison was carried out among the spelt genotypes (Figure 1), both traditional and modern spelt (cv. Anna Maria), the data showed high variation among these genotypes for all measured traits in grain and flour (Table S2). The ANOVA analysis suggested a high influence of the genotype in this variation (Table S4), although the differences between both years were also significant.

**Figure 1.** Frequency distribution of the spelt genotypes and average values of the common wheat cultivars for different grain and flour quality traits.

Most of the traditional spelt genotypes showed lower TW values than the cv. Anna Maria; however, the thousand kernel weight (TKW) of the latter was significantly lower than the values of the traditional spelt accessions (Figure 1).

The protein content, both in grain and flour, was highly variable, with cv. Anna Maria being in the low part of the distribution in both cases (Table S2). This high protein content had little effect on the grain hardness, because, in general, the spelt genotypes showed soft or very soft grain, although some accessions showed values of semi-hard grain (Figure 1). The general lower grain hardness associated with the spelt accessions was positively associated with flour yield.

The gluten strength measured as the SDS-sedimentation volume showed values of medium and high, with some exceptions (Figure 1).

For the PPO activity, the range was very wide (3–14 Ug−1min−1), and two groups of materials could be distinguished among the spelt genotypes: one with a mean value of 6.5 Ug<sup>−</sup>1min−1, and another with the mean values around 10.5 Ug−1min−1.

#### *3.3. Alveogram and Baking Traits in the Spelt Collection*

In the previous comparison with common wheat (Table 1), the data showed that spelt doughs had low tenacity, high extensibility, and low to medium gluten strength, as indicated by their W values. The analysis of the 80 genotypes that could be evaluated with the alveograph, showed a wide variation for the traits measured with this equipment (Tables 2 and S3), with different genotypes exhibiting values higher or lower than the average. In this respect, some spelt genotypes could be classified as medium to high gluten strength with W up to 388 × <sup>10</sup>−<sup>4</sup> J. Spelt genotype BGE 020900 (W = 320 × <sup>10</sup>−<sup>4</sup> J on average across the two years) could be considered a good donor of this trait for breeding programs. For gluten extensibility, several spelt accessions (such as BGE 001990, PI 348727, and PI 348747) showed very low P/L values (0.3), and could be considered interesting sources of this trait. Accession PI 348465 showed a very interesting combination of both gluten strength and extensibility (W = 283 × <sup>10</sup>−<sup>4</sup> J and P/L = 0.4) and could be considered the best material found in terms of gluten quality. The modern spelt (cv. Anna Maria) presented values around the average values of the spelt set (Table 2). In this case, the ANOVA analysis also showed the high influence of the genotype in the variation detected (Table S4).

**Table 2.** Comparison of the alveograph parameters obtained in the traditional spelt accessions and the modern spelt cultivar Anna Maria.


P, dough tenacity; L, dough extensibility; G: swelling index; W, dough strength; and Ie, elasticity index.

As already mentioned, the mean values of spelt and common wheat for loaf volume did not show significant differences (Table 1). However, when the 50 genotypes of the spelt set were independently analysed, these materials exhibited a high variability for this trait, with minimum and maximum values of 600 and 975 cc, respectively (Figure 2). Apart from that, almost 82% of these genotypes had loaf volume between 750 and 875 cc. Genotypes PI 469058 and PI 469051 (885 mL and 848 mL of loaf volume, respectively, on average across the two years) were the best performers for this trait and could be used by breeding programs aimed at the improvement of bread-making quality.

**Figure 2.** Frequency distribution of the spelt genotypes and average values of the common wheat cultivars for loaf volume.

Finally, a correlation analysis was carried out with the analyzed traits (except loaf volume, due to the lack of this data in many genotypes) within the spelt wheat set (Table S5). A negative correlation was found between test weight and protein content, and grain size (TKW) and alveograph W. Positive correlations were identified between protein content and SDS-sed and gluten extensibility (L), and among the different alveograph parameters.

#### **4. Discussion**

Recent changes in the agri-food industry have generated a growing interest in foods and old crops that have been practically lost during the last century [44]. In some cases, this renaissance has been associated with the supposed miraculous properties of these old crops. In general, such statements are not supported by any scientific basis and both the nutritional and nutraceutical properties of ancient wheats have been shown to be very similar to those of modern wheat [45–51]. However, other real benefits, such as the expansion of diversity in food, are little appreciated.

Within the wheat world, both the old varieties that had been replaced by more productive ones, and some of the species that were cultivated in the past have been recovered during recent decades [6,45]. Some of the latter are called "ancient wheats" and consists mainly of three species: einkorn (*T. monococcum* L. ssp. *monococcum*, 2n = 2× = 14, AmAm), emmer, and spelt. Some of the agronomic characteristics of these ancient wheats were those that led to their disappearance and abandonment when most of the agricultural processes were mechanized. In addition, due to being hulled grain cereals, the need for special dehulling treatment prior to grinding increased costs and affected their profitability. For this reason, their revival is linked to the boom in traditional and gourmet bakeries, where the higher prices of these products can offset production costs. Nowadays larger retailers also offer flour and products made from these types of wheat.

At the same time, this renewed interest has resulted in the development of numerous studies on these species, comparing their characteristics with the ones of modern wheat [50,51]. However, many of these studies have been carried out with a limited number of genotypes [37], which could create bias in the results and undervalue the true role of these old materials. For this reason, the evaluation of large collections of these ancient wheats, with the limitation of the storage materials in germplasm banks, could shed light on these questions, identifying new genotypes with potential utility for breeding programs. In the current study, one collection of 88 Spanish traditional spelt accessions, together with 9 common wheat cultivars and 1 modern spelt cultivar (cv. Anna Maria), were analyzed and compared for several traits related to milling, processing, and end-use quality.

When the analyzed materials are ancient or old wheats, the technological quality must be evaluated with caution. Changes in baking techniques throughout the last century have generated materials adapted to these techniques, which are different from those required for traditional baking, and consequently, the evaluation of ancient wheats according to modern parameters could not be the best strategy. In this regard, our previous studies on the grain composition of the spelt accessions evaluated in the current study showed the presence of rare HMWGs variants in spelt wheat (1, 13 + 16, and 2 + 12) [38,39]. However, it is possible that the high frequency of these variants is an empirical consequence of the way these wheats were and are used in traditional agri-food applications. These characteristics are mainly demanded by bakers, since all the studies carried out suggest the clear influence of glutenins on the viscoelastic properties of wheat dough [52]. However, millers are interested more in other traits more related to the physical and chemical characteristics of the grain such as TW, TKW, protein content, and grain hardness, mainly due to their influence on the flour yield.

Previous studies have shown that the grain size in spelt is larger for the modern material (with common wheat introgression) than for the traditional spelt [9,31,33]. In this study, the traditional spelt genotypes presented a TW and TKW very similar to the common wheat cultivars used as the control, and, compared with the modern spelt (cv. Anna Maria), the latter has a better TW but its TKW was lower. This reinforces the idea that variation in the traditional spelt is high and could be interesting for breeding programs aiming to develop new cultivars with very high grain size [53,54]. In parallel, the spelt genotypes with high grain size showed high protein content (indeed, a significant correlation between TKW and protein content was found) and a soft texture, which positively affected flour yield.

The PPO activity appeared comparatively higher in spelt than in common wheat. This trait is regulated by several enzymes, synthesized by the *Ppo-1* and *Ppo-2* loci at the homoeologous group 2 chromosomes [55–59], and has been associated with the discoloration and darkening of wheat products [60–63], which generates a certain amount of rejection among consumers [41,56,61–64]. Paradoxically, it may happen that today's consumers associate this dark color with the true presence of flour spelt in a food product, while the cream color suggests that the product is made with flour of modern common wheat and not from spelt (more "natural" vs. more "industrial"). Therefore, high PPO activity may not be an undesirable trait for spelt cultivars, although it does deserve attention if spelt is to be used in the breeding of modern common wheat as source of other traits of interest. In this regard, some traditional spelt genotypes showed low PPO activity values (≤5 Ug<sup>−</sup>1min−1) (Table S2), although this was not the general trend.

Previous studies conducted on spelt revealed that spelt mostly exhibits low to medium gluten strength [28–34]; however, our study revealed that it is also possible to identify genotypes with stronger gluten. In any case, the viscoelastic properties of spelt could be different from those of common wheat. The spelt genotypes showed, in general, more extensible doughs, which was favored by higher protein contents, and, in a few cases, the W values were reasonably high (up to 300 × <sup>10</sup>−<sup>4</sup> J). This was also due to the strong correlation found between alveograph W and P/L, which is also normal in common wheat sets [65]. However, it was not possible to identify an unambiguous relationship between the W values and loaf volume. As with other measured traits, the variation of these two parameters was high among the spelt genotypes. When the loaf volume was related to the flour protein content, some genotypes with low protein content showed high loaf volume, which suggests the high quality of these gluten proteins.

The current trend in the cereal's world has extended the search for other desirable traits within the grain components, mainly related with nutritional and nutraceutical properties, which would complement the technological properties of the doughs [37,66–68]. Nowadays, the presence of micronutrients such as Fe or Zn in the flour, or dietary fiber in form of soluble arabinoxylans, is highly recommended and this has increased the interest in the ancient wheats, with some studies suggesting that these old materials could be a good source for these traits [69–71]. In this respect, a previous study on these nutritional aspects has revealed that the current spelt genotypes show a notable variation in these traits [72]. These data, together with the data obtained in the current study, highlight the

need to increase the evaluation of wide collections of these ancient wheats, in order to detect the true variability in these old materials for different traits, including those associated with processing and nutritional quality. Such analyses will allow the identification of unique germplasms that could be used for the selection and purification of intra-accession variability for the development of traditional and homogeneous spelt varieties, both to be crossed with modern wheat to transfer the trait of interest, to improve modern wheat genetic diversity, and to develop better-adapted spelt cultivars.

#### **5. Conclusions**

Ancient wheats can be good sources of interesting agronomic features, mainly rustresistant genes and quality traits for wheat breeding. The evaluation of the large collections of these old materials would allow for evaluation of the true variability present in these species. In the current study, large variation was found in a set of Spanish spelt landraces, which, in general, showed soft grain, medium–high protein content, low gluten strength, high gluten extensibility, and medium bread-making quality; spelt genotypes showing outstanding values for some of these traits that could be useful for breeding purposes were identified. Additionally, this and similar studies could provide the opportunity to develop new cultivars of spelt with good characteristics for the food industry.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/foods12162996/s1, Table S1: Plant material used in the study; Table S2: Mean values of the grain and flour measured traits for each season in the materials evaluated (spelt and common wheat); Table S3: Mean values of the alveographic and baking traits for each season in the materials evaluated (spelt and common wheat); Table S4: Effects of genotype, season, and genotype x season (GxS) on quality traits in spelt accessions. Sum of squares, % of the total sum of squares from ANOVA analysis, and coefficient of variation (CV) are indicated; and Table S5: Correlation analysis of quality traits in spelt.

**Author Contributions:** Conceptualization, C.G. and J.B.A.; formal analysis, A.B.H.-G., M.I.I., M.R. and J.C.S.; data curation, M.I.I., C.G. and M.R.; writing—original draft preparation, A.B.H.-G. and J.B.A.; writing—review and editing, C.G., M.I.I. and M.R.; supervision, C.G. and J.B.A.; project administration, J.B.A.; funding acquisition, C.G. and J.B.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by grant PID2021-122530OB-I00 from the Spanish State Research Agency (Spanish Ministry of Science and Innovation)—MCIN/AEI/10.13039/501100011033 co-financed with the European Regional Development Fund (FEDER) from the European Union. Carlos Guzman gratefully acknowledges the European Social Fund and the Spanish State Research Agency (Spanish Ministry of Science and Innovation)—MCIN/AEI/10.13039/501100011033—for financial funding through the Ramon y Cajal Program (RYC-2017-21891).

**Data Availability Statement:** Data are included in the Supplementary Materials.

**Acknowledgments:** We thank to the National Small Grain Collection (Aberdeen, USA) and Centro de Recursos Fitogeneticos (INIA, Spain) for supplying the analysed materials.

**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.
