*2.7. Statistical Analysis*

The frankfurter manufacture was repeated twice on 2 di fferent days. One-way analysis of variance (ANOVA) was performed to evaluate di fferences between formulations using the SPSS program (v.22, IBM SPSS Inc., Chicago, IL, USA). Least squares di fferences were used for comparison of mean values between formulations and Tukey's HSD test to identify significant di fferences (*p* < 0.05) between formulations.

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

#### *3.1. Composition and Energy Value*

Table 2 shows the composition (proximate analysis, mineral content and total extractable polyphenols) and energy value of frankfurters.


**Table 2.** Proximate composition (%), mineral content (mg/100 g product), total extractable polyphenols (mg/100 g mf) and energy value (kcal/100 g product) of frankfurters.

\* For samples denominations, see Table 1. \*\* Dietary fibre content was estimated according to the content analysed in SUN, so no statistical analysis was performed. Data are expressed as means ± SD (*n* = 6). Different letters in the same row indicate significant differences (*p* < 0.05).

As expected, samples formulated with all pork backfat (F/AF) had the lowest (*p* < 0.05) moisture content, because they were formulated with the lowest water content (Table 1). It was observed that with the SUN addition, protein content significantly increased (Table 2) with increasing SUN addition, due to the protein content of this ingredient [8]. As a consequence of the use of SUN extract as an animal fat replacer, two fat levels were observed, ~19% in samples elaborated with all animal fat (F/AF) and nearer to 12% in frankfurters reformulated with SUN (F/2SUN and F/4SUN) (Table 2). This fat reduction meant a decrease in saturated fatty acids, as can be seen in Figure 1. The unsaturated fatty acids content also was reduced as a consequence of the use of SUN extract as an animal fat replacer in F/2SUN and F/4SUN samples (Figure 1). Overall the SUN addition led to a quantitative fatty acid modification, rather than to a qualitative change in fatty acids.

**Figure 1.** SFA, MUFA and PUFA of frankfurters (g/100 g of frankfurter). For sample denomination, see Table 1. Different letters in the same fatty acid group indicate significant differences (*p* < 0.05).

Regarding dietary fibre, SUN showed a relevant content of 28% in the freeze-dried sample, of which 87% corresponded to insoluble dietary fibre. Since dietary fibre intakes are still far from official recommendations [30], this shows the potential of SUN as a functional ingredient for different food applications. Due to the amount of SUN incorporated into the frankfurters, the final dietary fibre content in F/2SUN was 0.56%, while in F2/4SUN was 1.12%, representing low amounts, although higher than in F/AF, where dietary fibre was absent.

Although absolute values were different for ash content among samples, there was no statistical difference to report (Table 2). Ash content is a good indicator of total mineral content and according to the literature [31], sunflower oil cake on a dry basis contains 0.48% calcium, 0.84% phosphorus, 0.44% magnesium and 3.49% potassium. Not all minerals were analysed in the samples, but the content of some minerals was improved as a consequence of SUN addition in F/2SUN and F/4SUN samples (Table 2). Both F/2SUN and F/4SUN showed higher (*p* < 0.05) magnesium, potassium, copper and manganese concentration than reference samples (F/AF). Regarding zinc, F/2SUN and F/AF showed similar (*p* > 0.05) levels, while F/4SUN had significantly higher amounts. Iron levels increased as a consequence of SUN addition, but the increase was significant only in F/4SUN compared to the control; this was probably due to the fact that most of the iron came from the meat materials and, therefore, only the higher SUN inclusion made a significant difference to the iron content.

Energy value in the control samples (F/AF) was ~232 kcal/100 g, while reformulated samples with SUN, F/2SUN and F/4SUN, showed an energy value of proximately 173 and 180 kcal/100 g, respectively (Table 2). In that sense, the reformulation strategy carried out allowed to obtain products with ~25% less of energy value than control one. This di fference in energy value was mainly due to the lower fat content (but also, the higher fibre, water and protein contents) of samples with SUN compared to F/AF.

#### *3.2. Polyphenol Content and Profile*

Total extractable polyphenol content was determined in the analysed samples (Table 2). A value was obtained for the F/AF samples, despite the absence of SUN, since it is known that in matrixes of animal origin several interfering substances may provide a response to the Folin assay, as observed, for instance, in minced fish [32]. Nevertheless, the most relevant fact is that there was a significant increase in polyphenol content when adding SUN to the frankfurters, even at relatively low concentrations such as 2% and 4%. Whether this presence of phenolic compounds would delay the oxidation of frankfurters remains to be elucidated.

Besides, a total of 24 phenolic compounds were identified in SUN (Table 3): 18 phenolic acids, 5 flavonoids and 1 hydroxycoumarin. Phenolic acids, and particularly quinic derivatives of ca ffeic or ferulic acids, have been previously reported as the main phenolic compounds in sunflower kernel and shell [33] and, specifically, in sunflower meal [34,35]. These acids were detected in this study, together with some additional phenolic acids (hydroxybenzoic acid, vanillic acid, sinapic acid, a quinolactone from ca ffeic acid, 1,2-disinapoylgentiobiose and the hydroxyphenylpropionic acid 3,4-dihydroxyphenyl-2-oxypropanoic acid), which, as far as we know, had not been reported in sunflower products before. Regarding the other phenolic compounds, some of them are quite ubiquitous in plant foods, such as (+)-gallocatechin while, for instance, 5-hydroxy-4,4,6-trimethoxyaureone has not been reported in sunflower seeds, but it has in sunflower flowers as well as in di fferent oils; for these reasons, their presence in SUN was plausible.


**Table 3.** Phenolic compounds identified in upcycled defatted sunflower seed flour by HPLC-ESI-QTOF MS analysis.

#### *3.3. Overall Nutritional Value: Nutrition and Health Claims*

All samples were a "high in protein" and could be labelled with the corresponding nutritional and health claims according to the Regulation (EC) n<sup>º</sup> 1924/2006 and Regulation (EU) n<sup>º</sup> 432/2012 [36,37]. Moreover, as a consequence of the use of SUN as an animal fat replacer, other nutrition and health claims can be made. Taking into account the fat content, the claim "reduced fat content" in F/2SUN and F/4SUN can be made [36]. The increased mineral content due to the presence of SUN, makes it possible for the "source of magnesium" claim to be made on F/2SUN and F/4SUN. In addition, several health claims could be attributed to F/2SUN and F/4SUN according to Regulation (EU) n<sup>º</sup> 432/2012, due to the magnesium presence. Moreover, F/4SUN could be labelled as "source of copper" and the health claims corresponding to its presence. Regarding zinc and potassium, although their presence was generally higher in F/2SUN and F/4SUN (Table 2), all samples could be labelled as "source of zinc" and "source of potassium" and related health claims [36,37]. These results were similar to other studies where pork back fat in frankfurters was replaced with other high protein ingredients such as hydrolysed collagen [38], and chia seeds [14].

#### *3.4. Processing Loss, pH and Colour*

The processing loss and pH results are shown in Table 4.


**Table 4.** Processing loss, pH values, colour and texture profile analysis (TPA) of frankfurters.

\* For samples denominations, see Table 1. Data are expressed as means ± SD (*n* = 10 for processing loss, *n* = 3 for pH values, *n* = 20 for colour parameters and *n* = 8 for texture parameters). Different letters in the same row indicate significant differences (*p* < 0.05).

There was no significant di fference in the processing loss across the three recipes. These results were low and similar to other frankfurters' studies, which reported ranges of processing loss between 10% and 20% [39–41]. The processing loss obtained (Table 4) indicated good stability in terms of fat and water-binding properties of the meat matrix. The pH decreased with increasing SUN content. The SUN addition influenced the colour, with L\*, a\* and b\* values decreasing significantly with increased SUN content, indicating that the frankfurters became darker, less red and less yellow. It is important that future SUN applications are aimed at minimising colour di fferences in the food matrix to avoid possible consumer's rejection, as liking of foods with adequate appearance could favour healthier product consumption by consumers [42]. During the process of new product development, the colour of foods with SUN could be altered to obtain no noticeable colour di fference for consumers. Colour alterations have been reported in other studies where pork back fat in frankfurters was replaced by chia seeds [14] and pineapple fibre [15] as well as in other studies using SUN in baked goods [8,9].

#### *3.5. Texture Profile Analysis*

The results from the texture profile analysis are shown in Table 4. Results showed that hardness was similar (*p* > 0.05) in frankfurters containing 2% of SUN (F/2SUN) but significantly higher in those with 4% (F/4SUN) as compared with the control (F/AF) (Table 4). This is probably due to the water holding capacity of the SUN [8]. The SUN might have retained water, swollen and consequently increased firmness, similarly to the behaviour reported on frankfurters with collagen [13,38,43]. Cohesiveness, springiness and chewiness were higher (*p* < 0.05) in samples with SUN added (F/2SUN and F/4SUN), regardless of the amount of this ingredient used as animal fat replacer (Table 4). This textural behaviour could be related with the di fferences in composition (mainly protein and fibre content) in each type of frankfurter due to of addition of SUN as an animal fat replacer, although the muscle protein level was kept constant in the formulations.

#### *3.6. Attenuated Total Reflectance (ATR)-FTIR Spectroscopy Analysis*

The acyl chain region comprised between 2950–2830 cm<sup>−</sup><sup>1</sup> of the ATR–FTIR spectrum of the di fferent frankfurter are shown in Figure 2.

**Figure 2.** ATR–FTIR spectra in the 2950–2830 cm<sup>−</sup><sup>1</sup> of frankfurters. For sample denomination, see Table 1.

This spectral region was dominated by two strong bands resulting, respectively, from the asymmetric (asCH2) and the symmetric (sCH2) stretching vibrations of the acyl CH2 groups [44]. Partial replacement of animal fat by SUN produces a frequency upshift in asCH2 and sCH2 from 2919 to 2921 cm<sup>−</sup><sup>1</sup> and from 2851 to 2852 cm<sup>−</sup>1, respectively, in going to from F/AF to F/4SUN (Figure 2). This frequency upshift was generally attributed to the diminution of the conformational order of the lipid acyl chains and to the increase of their dynamics, which implied greater inter- and intramolecular lipid disorder [45,46]. Therefore, it is possible to assume that frankfurters elaborated with SUN showed greater inter- and intramolecular lipid disorder than control elaborated with all animal fat, being these phenomena more relevant in F/4SUN samples. The increase of lipid disorder in reformulated frankfurters as a function of SUN content could be attributed to more lipid interactions in the meat matrix in these samples (F/2SUN and F/4SUN), mainly protein-lipid interactions [47]. The lipid chain disorder or increased lipid interactions observed in F/2SUN and F/4SUN could be related with their specific textural behaviour (Table 3), greater hardness, springiness and chewiness than control (F/AF).

Accordingly, previous studies showed that the direct addition of vegetable proteins, such as soy protein isolate, in the reformulation process of heated meat batters produced protein secondary structural changes accompanied by textural properties modifications such as an increase in hardness and chewiness [48]. In addition, it has been described that the direct addition of mushroom powder, with similar composition as SUN in terms of protein, dietary fibre and bioactive compound content, in the reformulation of processed meats impacts on their rheological and structural characteristics. Similar to results found in the present work, it has been indicated that this reformulation process increased the conformational disorder of the lipid acyl chains due to modifications of lipid-protein interactions [49].

#### *3.7. Preliminary Sensory Analysis*

The results of the pilot sensory study are shown in Table 5.


**Table 5.** Pilot sensory analysis of frankfurters.

\* For samples denominations, see Table 1. Data are expressed as means ± SD (*n* = 27). Different letters in the same row indicate significant differences (*p* < 0.05).

The panel detected a significant difference only in the textural sensory parameter's firmness and juiciness. F/4SUN samples were considered firmer than F/AF, which agrees with findings from the texture profile analysis. F/4SUN samples were also considered significantly less juicy than F/AF. This might be related to another attribute, powdery, which tended to increase in samples with SUN, although not significantly. Samples with SUN tended to have a higher aroma and flavour intensity and lower overall acceptability than control, but these differences were not significant. The reduction of pork back fat in frankfurters has been associated with lower acceptability scores in several sensory attributes [14,16]. Since a limited number of consumers was used to in the present study, sensory tests run on a bigger sample size would need to be repeated in the future in order to ensure enough statistical power.
