Quality of fat-reduced frankfurter formulated with unripe banana by-products and pre-emulsified sunflower oil

ABSTRACT

This research was undertaken to investigate the cooking properties, lipid oxidation, fatty acid profile, texture, and rheological behavior of fat-reduced frankfurter containing whole banana flour (WBF), pulp banana flour (BPF), and peel banana flour (BSF) obtained from unripe fruit and pre-emulsified sunflower oil. The addition of WBF, BPF and BSF in association with pre-emulsified oil lowered the cooking loss and increased the emulsion stability of the final product. These ingredients also enhanced the oxidative stability of the samples during storage (P < .05). Higher PUFA values were associated with the incorporation of pre-emulsified sunflower oil. BSF and BPF provided a more enhanced gel network structure, and subsequently retained more water molecules. WBF and BPF samples improved the sensory aspects of reduced-fat frankfurters compared to BSF (P < .05). Banana by-products, as ingredients, enable the use of vegetable oils in meat emulsion and subsequently improve the health benefits of final products by replacing up to 50% animal fat.

KEYWORDS:

• Unripe banana by-products
• fat-reduced frankfurter
• sunflower oil
• lipid oxidation
• ingredients
• fatty acids
• rheology
• sensory

Introduction

Processed meat products have an essential role for consumers as they contain essential protein sources with high nutritional value. However, these products contain moderate amounts of animal fat to achieve the desired functional and sensorial properties. Fat from animal sources is a crucial ingredient applied to improve the quality of meat emulsions by decreasing the cooking loss and increasing the binding properties while improving emulsion stability, mouth-feel, rheological and structural properties of the final product.^[1,2] However, excessive consumption of animal fat has been associated with many human health-related diseases, particularly cardiovascular disease, cancer, and type-2 diabetes, because some animal fat contain high amounts of particular saturated fatty acids.^[3] Therefore, an alternative approach to fully or partially reduce animal fat and saturated fatty acids associated with health problems is to produce a comminuted meat product using high molecular carbohydrates from different sources^[3,4] and pre-emulsified vegetable oil^[5–8] as a fat substitute. The incorporation of plant-based indigestible carbohydrates mixed with vegetable oil generated meat product with greater functional properties.^[6,7] These ingredients would enhance the technological quality and organoleptic characteristics of comminuted meat product manufactured with fat-reduced content and consequently provide an additional nutritional and health benefits due to the higher intake of dietary fiber together with unsaturated fatty acids.^[9–13] Sunflower oil is an important oilseed crop that contains high concentrations of unsaturated fatty acids, such as omega-3 and omega-6, which could be useful in reducing the risk of certain health problems, including cardiovascular diseases.^[14] Yılmaz et al.^[15] indicated that sausages manufactured with sunflower oil as ingredients would have health benefits resulting from higher contents of bioactive compounds such as unsaturated and essential fatty acids.

Unripe banana, milled into flour, provides gluten-free ingredients rich in indigestible and nutritional compounds such as fiber (6.0–15.5%), resistant starch (40.9–58.5%), proteins and bioactive compounds with antioxidant ability.^[16–18] These indigestible sources of high molecular carbohydrates have been used as ingredients in many food products, including cakes, bread and chicken nuggets due to their natural antioxidant properties capable of modulating lipid metabolism and improving the nutritional quality of the final product.^[19,20] Regarding meat products, Alves et al.^[21] indicated that flour from unripe banana in combination with pork skin could effectively replace animal fat in bologna sausages without negatively affecting their functional quality and organoleptic properties. Bastos et al.^[22] also reported that green banana flour increased the yield and water-binding capacity of meat burgers, but reduced its hardness and shrinkage. Even though banana by-products from unripe fruits, including peel and pulp flour, have been reported as food ingredients,^[18,23,24] we believe there remains a lack of studies regarding its effect and behind mechanism in the meat products. Thus, the current study was proposed to investigate the cooking properties, lipid oxidation, fatty acid profile, texture and rheological behavior of fat-reduced frankfurter type sausages containing three types of banana by-products including whole banana flour (WBF), pulp banana flour (BPF) and skin banana flour (BSF) obtained from unripe fruit in association with pre-emulsified sunflower oil.

Material and methods

Processing of unripe banana by-products

Banana by-products (WBF, BPF, BSF) were prepared from unripe fruits according to methods stated by Singh et al.^[18] with minor changes. Commercial unripe banana fruits (Musa acuminata Colla cv. Cavendish) purchased from Xianghui orchard in Zengcheng district (Guangdong province, China) were washed and immersed in water containing sodium hypochlorite (0.2 g/kg) for 20 min to sanitize. Following drying from the washing step, the entire banana, pulps and peel only, previously separated, were cut (5 mm thickness) into transverse slices and dipped in 1.5 g/L of the citric acid solution for about 1 h to avoid undesirable enzymatic color changes. The banana slices were rinsed with tap water, drained and dried in an oven at 55°C for 20 h. After dehydration, banana slices were coarse-ground using a grinder (HK-10B, Xulang Machinery, China) and separated using a sieve shaker (150 mesh). The coarse-ground banana by-products were stored in polyethylene bags at 4°C until further use. Duplicate measurements of the compositional properties of unripe banana by-products determined according to AOAC^[25] methods were presented as follows: WBF was composed by 7.73 ± 0.11 g/100 g of moisture, 5.03 ± 0.06 g/100 g of protein, 1.26 ± 0.05 g/100 g of fat, 2.30 ± 0.07 g/100 g of ash and 83.69 ± 0.22 g/100 g of carbohydrates. BPF had 8.78 ± 0.06 g/100 g of moisture, 4.18 ± 0.07 g/100 g of protein, 0.91 ± 0.04 g/100 g of fat, 1.42 ± 0.03 g/100 g of ash and 84.71 ± 0.04 g/100 g of carbohydrate. BSF presented 6.82 ± 0.0.12 g/100 g of moisture, 6.71 ± 0.16 g/100 g of protein, 3.53 ± 0.11 g/100 g of fat and 78.25 ± 0.17 g/100 g of total carbohydrate. Total dietary fibers content measured using K-TDFR kit (Megazyme Int. Ireland Ltd., Wicklow, Ireland) for WBF and BPS was 10.98 ± 0.85 and 7.30 ± 1.01 g/100 g, while BSF was 21.89 ± 1.17 g/100 g. Total starch content evaluated using K-RSTAR megazyme kit was 71.33 ± 1.44, 75.18 ± 1.01 and 65.32 ± 0.72 g/100 g for WBF, BPF and BSF, respectively. The content of total phenolic analyzed according to the method mentioned by Fatemeh et al.^[26] was 34.56 ± 0.12, 18.90 ± 0.14 and 38.40 ± 0.09 mg GAE/100 g for WBF, BPF and BSF, respectively.

Pre-emulsified sunflower oil preparation

Sunflower oil (Arawana brand) composed by 20 g/100 g of monounsaturated fatty acid, 67 g/100 g of polyunsaturated fatty acid and 55 mg/100 g of vitamin E was bought from a Suguo supermarket (Nanjing, China). The pre-emulsified sunflower oil was mechanically prepared by mixing sodium caseinate, hot water and sunflower oil at a ratio of 1:8:10. The pre-emulsified sunflower oil was kept at 4°C overnight before frankfurter manufacture.

Manufacture of frankfurter-type sausages

Fresh pork meat (70.46 g/kg moisture, 20.88 g/kg protein, 4.72 g/kg fat) and pork back fat (12.28 g/kg moisture, 86.07 g/kg fat) were obtained from a branch of Yurun Group Co., Ltd (Nanjing, China). Additives/ingredients were provided by the pilot plant center of meat quality and safety control of Nanjing Agricultural University. Separately, lean meat and pork back fat were coarse-ground using a chopper (TC, 12E, SIRMAN, Venezia, Italy) and stored at 4°C overnight. The fat-reduced frankfurters formulated with banana by-products and sunflower oil were manufactured, as mentioned by Pereira et al.^[27] with some changes. Accurately, 50 g/100 g of pre-emulsified sunflower oil was added to replace 50 g/100 g of pork-back fat in all treatments, except for control that was prepared with only pork back fat. Additionally, 3 g/100 g of pork back fat was replaced by the same proportion of banana by-products (WBF, BPF, BSF). Firstly, ground pork meat, half ice and salt were chopped for 2 min in a food mixer (BZBJ-15, Expro Stainless Steel Mechanical & Engineering Company Hangzhou, China) to facilitate the extraction of myofibrillar proteins. Then, the remaining ice, ingredients (Sodium tripolyphosphate, glucose, white pepper, garlic), banana by-products and pre-emulsified sunflower oil were added according to the formulation (Table 1) with continued mixing at high speed for a further 4 min to obtain the final emulsion. At all stages of the process, the temperature of the raw emulsion did not surpass 10°C. The prepared raw emulsion was transferred into a hand stuffer and filled into 30–32 mm diameter polyamide casings (Shijiazhuang Jipeng Company, China). Three batches of each formulation were performed on different days within a week and each batch was considered as one replication.

Table 1. Formulation of frankfurter-type sausages containing unripe banana by-products and pre-emulsified sunflower oil

Ingredients (gram)	Control (g)	T1 (g)	WBF (g)	BPF (g)	BSF (g)
Lean meat	700	700	700	700	700
Fat	300	150	120	120	120
Pre-emulsified oil	–	150	150	150	150
Ice/water	200	200	200	200	200
Additives	–	–	30	30	30
Salt	15	15	15	15	15
Glucose	5	5	5	5	5
Sodium Tripolyphosphate	3	3	3	3	3
White pepper	2.4	2.4	2.4	2.4	2.4
Garlic	0.4	0.4	0.4	0.4	0.4

Control = total pork-back fat; T1 = 50% of pork-back fat and 50% of sunflower oil; WBF = 3% of whole green banana flour; BPF = 3% of pulp banana flour; BSF = 3% of peel banana flour.

Cooking loss

Cooking loss of frankfurter samples was performed 24 h after manufacturing by cooking the samples for 30 min at 80°C.^[28] The cooking loss was determined by subtracting the weight difference of the samples before cooking (w _{raw sample}) from the sample after cooking (w _{cooked sample}) and expressed as a percentage:

$\mathrm{C}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{l}\mathrm{o}\mathrm{s}\mathrm{s}\mathrm{\%}={\mathrm{W}}_{\mathrm{r}\mathrm{a}\mathrm{w}\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}\mathrm{s}}-{\mathrm{W}}_{\mathrm{c}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{d}\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}\mathrm{s}}/{\mathrm{W}}_{\mathrm{r}\mathrm{a}\mathrm{w}\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}}\ast 100$

Proximate composition

Protein, fat and ash content was measured according to protocol cited in AOAC.^[25] Moisture content was evaluated by the mean difference after drying a sample to a constant weight for 12 h at 105°C in a drying oven. The pH of fat-reduced frankfurter was tested using a pH meter (Hanna, Italy). The water activity (a_w) was determined using an electric hygrometer apparatus (Novasina AG, CH-8863, Switzerland) as recommended by the manufacturer. All parameters were evaluated in triplicate per batch.

Emulsion stability

The emulsion stability of fat-reduced frankfurters was determined according to the protocol mentioned by Pereira et al.^[27] with slight changes. Accurately, 40 g of raw batter weighed into 80 mL test tubes was centrifuged (Allegra™ 64 R, Fisher Scientific, Pittsburgh, PA, USA) at 1,000 × g for 4 min at 4°C to eliminate possible air bubbles. Each test tube was heated (80°C) for 30 min in a water bath. After the samples were cooled, all expressible liquids were poured into pre-weighed crucibles, weighed and dried at 105°C for 16 h. Each formulation per batch was analyzed 6 times. The % of total fluid released (TFR), water released (WR) and fat loss (FR) was calculated as follows:

$\begin{array}{rl}& \mathrm{\%}\mathrm{T}\mathrm{F}\mathrm{R}=\mathrm{T}\mathrm{F}\mathrm{R}\mathrm{d}\mathrm{i}\mathrm{v}\mathrm{i}\mathrm{d}\mathrm{e}\mathrm{d}\mathrm{b}\mathrm{y}\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\times 100;\mathrm{W}\mathrm{R}\mathrm{\%}=\left(\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{c}\mathrm{r}\mathrm{u}\mathrm{c}\mathrm{i}\mathrm{b}\mathrm{l}\mathrm{e}+\mathrm{T}\mathrm{F}\mathrm{R}\right)-\\ & \left(\mathrm{d}\mathrm{r}\mathrm{i}\mathrm{e}\mathrm{d}\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{c}\mathrm{r}\mathrm{u}\mathrm{c}\mathrm{i}\mathrm{b}\mathrm{l}\mathrm{e}\right)\ast 100;\mathrm{\%}\mathrm{F}\mathrm{R}=\mathrm{\%}\mathrm{T}\mathrm{F}\mathrm{R}-\mathrm{\%}\mathrm{W}\mathrm{R}\end{array}$

Lipid oxidation and fatty acid profile

Oxidative stability of fat-reduced frankfurter was evaluated by the quantification of malonaldehyde (MDA) values using thiobarbituric acid reactive substances (TBARS), as mentioned in the protocol cited by Zhang et al.^[29] The level of lipid oxidation was evaluated in triplicate at 1, 4 and 7 d of storage. Absorbance values were observed at 532 nm, and the TBARS were expressed as mg MDA/Kg.

The fatty acid profile of fat-reduced frankfurters containing unripe banana by-products and pre-emulsified sunflower oil was determined by exposing the samples to gas chromatography, as indicated by Hu et al.^[7] with minor changes. The fatty acid compound was identified by matching the retention times of the frankfurter samples with a mixed fatty acids standard (Supelco™ 37 FAME Mix 47885-U, USA). Triplicate measurements of fatty acid of each sample per batch were conducted.

Rheological measurements

The gelation of raw emulsions samples was measured in situ by oscillatory rheology using a rheometer apparatus (MCR 301, Anton Paar, Graz, Austria). The device was set with 50 mm parallel steel plate geometry and adjusted to a gap of 1 mm. About 10 g of raw batters were placed onto the center of flat stainless plates, followed by moving upper plate into the position. Silicone oil was dropped around the plate edges to avoid water evaporation and the samples were continuously heated in a gradient from 20°C to 80°C with a scanning frequency fixed at 0.1 Hz. The oscillatory sweep from 0.1 to 100 rad/s was performed in the linear viscoelastic region with a strain of 1%. Storage modulus (G’) was measured in triplicate for each sample.

Texture profile analysis

The texture profile of frankfurter samples was determined using a texture analyzer XT Plus with a P50 probe (Stable Micro Systems Ltd., Godalming Surrey, UK). The operational settings were: pretest speed 2.0 mm/sec., posttest speed 10 mm/sec., test speed of 2 mm/sec and a half compression (50%). Hardness (N), springiness, cohesiveness, gumminess (N) and chewiness (N) properties were analyzed. Six replications of each batch were evaluated.

Sensory evaluation

Sensory characteristics of fat-reduced frankfurter formulated with WFB, BPF and BSF associated with pre-emulsified sunflower oil were assessed in duplicate by 18 panelists comprising 10 female and 8 male recruited from National Center of Meat Quality and Safety Control. All the sensorial participants were instructed to evaluate the sensory properties of fat-reduced frankfurter according to the Chinese standard criterion GB/T 22210–2008 used to analyze meat and meat products. Samples were cut in a slice with a 2 cm of diameter, heated in the microwave for 10 s and served to the panelists for sensory judgment through a nine-point hedonic scale.^[7] The panelists were requested to evaluate the descriptive attributes, including aroma/smell and taste/flavor (1-extremely bland, 9-extremely intense), texture/consistency (1-extremely softer, 9-extremely harderr), appearance (1-extremile unattractive, 9-most attractive) and overall acceptance (1-extremely dislike, 9-extremely like).

Statistical analysis

The collected data were submitted to statistical analysis using SPSS v. 20 (IBM Corp., Armonk, NY, USA) designed for Windows. The analysis of variance for each parameter was performed at least in triplicate using one way ANOVA and Duncan’s multiple range tests with 95% levels of significance (P < .05).

Results and discussion

Cooking properties and emulsion stability

Plant-based products containing indigestible carbohydrates such as resistant starch and fibers have been shown to enhance the functionality and health quality of comminuted meat products by positively affecting its cooking properties and thermal emulsification.^[3,28] Table 2 demonstrates the changes occurring for cooking loss of fat-reduced frankfurters formulated with and without WBF, BSF and BPF in association with pre-emulsified sunflower oil. The replacement of 50% of animal fat with pre-emulsified sunflower oil significantly augmented the cooking loss of frankfurter samples due to an increased exudation compared with control (P < .05). However, the incorporation of banana by-products (BPF and BSF), in combination with pre-emulsified sunflower oil to substitute animal back fat, significantly lowered the cooking loss of frankfurter compared to control and T1 (P < .05). Similarly, Choi et al.^[5,6] found lower cooking loss in frankfurters products after replacing levels of pork back fat with rice bran fibers and vegetable oil. Alves et al.^[21] also found that bologna type sausages prepared with green banana and pork skin gel significantly decreased its cooking loss and improved its technological quality. Among the fat-reduced frankfurters, those formulated with BPF and BSF (3%) decreased cooking losses compared to WBF and T1 samples (P < .05).

Table 2. Nutritional composition, cooking properties and texture of fat-reduced frankfurters formulated with unripe banana by-products and pre-emulsified sunflower oil

	Control	T1	WBF	BPF	BSF
Nutritional composition
pH	6.65 ± 0.02 ^c	6.64 ± 0.01 ^c	6.46 ± 0.01^a	6.54 ± 0.01^b	6.46 ± 0.02^a
Aw	0.950 ± 0.001^a	0.950 ± 0.002^a	0.955 ± 0.001^b	0.955 ± 0.001^b	0.954 ± 0.001^b
Protein (%)	13.42 ± 0.52^a	14.02 ± 0.22^a	15.32 ± 0.18^b	15.12 ± 0.16^b	15.20 ± 0.27^b
Moisture (%)	58.75 ± 1.80^a	63.83 ± 0.95^b	63.49 ± 2.87^b	64.71 ± 0.80^b	65.37 ± 0.65^b
Fat (%)	28.34 ± 0.18 ^c	17.46 ± 0.25^b	11.07 ± 0.09^a	11.42 ± 0.15^a	11.02 ± 0.17^a
Ash (%)	2.24 ± 0.14^a	2.43 ± 0.05^ab	2.53 ± 0.05^b	2.29 ± 0.07^ab	3.21 ± 0.09 ^c
Cooking quality
Cooking loss (%)	3.14 ± 0.14^b	4.39 ± 0.07 ^c	2.87 ± 0.23^b	2.50 ± 0.30^a	2.20 ± 0.40^a
Emulsion Stability (%)
TFR	2.76 ± 0.36^b	4.04 ± 0.41 ^c	2.37 ± 0.34^ab	2.66 ± 0.30^b	2.14 ± 0.37^a
WR	2.53 ± 0.32^b	3.80 ± 0.41 ^c	2.20 ± 0.31^ab	2.48 ± 0.28^b	2.00 ± 0.33^a
FR	0.23 ± 0.05^b	0.25 ± 0.11^b	0.17 ± 0.05ª	0.18 ± 0.02ª	0.14 ± 0.04ª
Textural properties
Hardness (N)	78.45 ± 4.95 ^c	59.25 ± 4.22^a	71.50 ± 4.57^b	70.90 ± 4.37^b	60.10 ± 3.32^a
Springiness (mm)	0.85 ± 0.03^ab	0.89 ± 0.01 ^c	0.89 ± 0.02 ^c	0.87 ± 0.03^bc	0.84 ± 0.01^a
Cohesiveness	0.63 ± 0.01^b	0.68 ± 0.01^d	0.66 ± 0.01 ^c	0.66 ± 0.02 ^c	0.58 ± 0.21^a
Chewiness (N)	43.83 ± 1.92^d	34.08 ± 2.01^b	41.35 ± 1.25 ^cd	40.90 ± 3.32 ^c	29.38 ± 1.14^a

Control = total pork-back fat; T1 = 50% of pork-back fat and 50% of sunflower oil; WBF = 3% of whole green banana flour; BPF = 3% of pulp banana flour; BSF = 3% of peel banana flour. TFR = total fluid released; WR = water released; FR = fat released.

^a–c: Means ± SD with different letter in the same row shows significant difference (p < 0.05)

Table 2 also indicates that adding WBF, BPF and BSF in the processing of frankfurters significantly increased the emulsion stability of the final product by decreasing the TFR and WR compared to T1 and Control (P < .05). The observation above indicates that the resistant starch and fibers present in unripe banana by-products allow the retention of more water due to its absorption capacity. Similar trends were observed in bologna type sausages formulated with levels of green banana and pork skin gel (PSGBF).^[21] Zhuang et al.^[30] also indicated that incorporating various levels of sugarcane dietary fibers (1, 2 and 3%) and pre-emulsified sesame oil decreased the TFR and WR of meat batters resulting in better emulsion stability compared to the products manufactured with the traditional method. The lowest TFR and WR values were observed in frankfurter prepared with BSF and pre-emulsified sunflower oil. FR was significantly lower in frankfurters containing WBF, BPF and BSF compared with the control and T1 (P < .05). It may be due to less fat in the emulsion and better oil binding induced by the resistant starch and fibers content present in these ingredients. No significant differences in FR among WBF, BPF and BSF samples were noted (P > .05). In general, the incorporation of WBF, BPF and BSF with high carbohydrate content to the formulation associated with vegetable oils increased water retention, thus decreased the cooking loss and improved the emulsification stability.

Proximate composition

The physicochemical composition of frankfurters formulated with and without WBF, BPF and BSF associated with pre-emulsified sunflower oil is also presented in Table 2. Significant variation in pH values from 6.46 to 6.65 was observed among samples formulated with banana by-products compared to control and T1 (P < .05). It was found that the addition of banana by-products associated with vegetable oil decreased the pH values of fat-reduced frankfurters. Previous studies also have reported changes in pH values of meat products containing vegetable and fruit by-products.^[23] In contrast, Alves et al.^[21] demonstrated that the pH values of bologna type sausages containing various concentrations of gel prepared with green banana flour and pork skin did not differ from each other. The range of Aw values from 0.954 to 0.955 showed no significant changes among fat-reduced frankfurters prepared with WBF, BSF and BPF in combination with pre-emulsified sunflower oil (P > .05). However, these treated samples significantly differed from the control and T1 samples on Aw (P < .05). Thus, the findings presented here are considered suitable for an emulsified meat product according to studies reported by Pietrasik and Janz.^[31]

Analysis of variance for moisture content indicated a significant difference between control and samples containing banana by-products and pre-emulsified sunflower oil (P < .05). Approximately 58.75% of moisture content found in control samples was different from 63.49, 64.71, and 65.37% presented in WBF, BPF and BSF samples, respectively. These differences among the samples could be explained by the fact that the components of unripe banana, including insoluble fibers (lignin, cellulose and hemicellulose) retain and absorb more water than fat tissue. No significant differences were observed among frankfurters containing green banana by-products in moisture contents (P > .05). Values ranging from 15.12% to 15.32% showed that significantly higher protein contents were found in the samples containing WBF, BSF and BPF compared to 13.42% of control (P < .05), while no differences were noted among samples prepared with unripe banana by-products. Our results agree with those of Kang et al.^[32] who found higher protein contents in meat batters manufactures with pre-emulsified sesame oil as the fat substitute. On the other hand, Alves et al.^[21] reported that the incorporation of unripe banana flour and pork skin gel did not differ in protein contents, probably because of the similar amounts of protein in the PSGBF gel (10.34%) and animal fat (8.52%).

Frankfurter samples formulated with banana by-products and pre-emulsified sunflower oil had significantly lower fat content compared to T1 samples and control (P < .05). Our results are similar to studies reported by Choi et al.^[6] who successfully decreased the fat content and improved the quality of meat emulsion by adding grape seed oil and rice bran fibers as a fat substitute. Ash contents were significantly higher in fat-reduced frankfurters formulated with BSF compared with other treatments (P < .05). Higher values of ash were likely to be instigated by the concentration of total dietary fibers, resistant starch and minerals present in the added ingredients. Similarly, Alves et al.^[21] showed that the presence of higher ash content in fat-reduced bologna was associated with the presence of resistant starch and minerals provided by the incorporation of green banana flour mixed with pork skin.

Lipid oxidation

Table 3 shows that TBARS values of fat-reduced frankfurters after storage were significantly affected by the incorporation of unripe banana by-products (P < .05). Control samples (30% pork back fat) with values increased from 0.390 to 0.706 mg MDA/Kg^−[1] during storage had higher levels of TBARs followed by samples formulated with 15% fat and 15% of the pre-emulsified sunflower oil. Our findings disagree with those of Choi et al.^[5] who indicated higher TBARS values in fat-reduced frankfurters containing canola oil compared to control without vegetable oil. The oxidative stability of fat-reduced frankfurters containing WBF, BSF and BPF associated with pre-emulsified sunflower oil was significantly improved compared to T1 and control (P < .05), with the lowest level attributed to BSF samples that was gradually increased from 0.131 to 0.196 mg MDA/Kg^−[1] at 7 d of storage. This phenomenon occurred due to a lessening of pork back fat content in the formulations, as well as the presence of antioxidant compounds from unripe banana by-product mainly peels that can limit the extent of lipid oxidation. Gao et al.^[33] showed that the incorporation of the pre-emulsified sunflower oil delayed the oxidation of frankfurters due to the presence of antioxidants compounds such as α-tocopherol in their composition, which improved the nutritional value and reduced the upcoming of unpleasant aromas. No significant differences in TBARS values were promoted among frankfurters containing unripe banana by-products (WBF, BPF, BSF) associated with vegetable oil during 1 d and 4 d of storage (P > .05). However, the TBARS values of samples containing BSF (0.196 mg MDA/kg^−[1]) significantly differed from WBF (0.331 mg MDA/kg^−[1]) and BPF (0.339 mg MDA/kg^−[1]) at 7 d. Results of the current study and previous reports showed that the flour of green banana skins contained more phenolic contents compared with flour of banana pulp^[26] which may lower TBARS. This fact could be used to explain the decreased TBARS values found in frankfurter samples containing BSF compared with other treatments at 7 d of storage. Our fat-reduced frankfurter presented a TBARS value within the range of 0.1 to 0.9 mg MDA/kg values reported in other frankfurters manufactured with oils from fish or vegetables as additives to enhance the lipid portion.^[34,35] The major total phenolic concentrations found in green banana peel flour is probably responsible for the higher antioxidant activity of BSF samples. It was noticeable that the storage time from 0 to 7 d significantly increased the TBARs values of all samples (P < .05), whereas it was retarded in samples with fat-reduced content. Previous studies of Selani et al.^[12] indicated that the oxidative rancidity was retarded with the incorporation of pineapple by-products but was not completely prevented, which corroborates with our findings.

Table 3. TBARs values of fat-reduced frankfurter prepared with unripe banana by-products and pre-emulsified sunflower oil

		Storage time
		1 Day	4 Days	7 Days
Tbars values (mg MDA Kg^-1)	Control	0.390 ± 0.007^cx	0.578 ± 0.018^cy	0.706 ± 0.006^dz
	T1	0.161 ± 0.003^bx	0.334 ± 0.005^by	0.418 ± 0.013^cz
	WBF	0.129 ± 0.002^ax	0.157 ± 0.006^ay	0.331 ± 0.008^bz
	BPF	0.127 ± 0.003^ax	0.157 ± 0.003^ay	0.339 ± 0.005^bz
	BSF	0.131 ± 0.003^ax	0.144 ± 0.010^ay	0.196 ± 0.004^az

Control = pork-back fat; T1 = 50% of pork-back fat and 50% of sunflower oil; WBF = 3% of whole green banana flour; BPF = 3% of pulp banana flour; BSF = 3% of peel banana flour.

^a–dMeans ± SD with different letter in the same column shows significant difference (p < 0.05)

^x–zMeans ± SD without a similar superscript within the row differ significantly (p < 0.05).

Fatty acid profile

The variations on the fatty acid profile caused by partial replacement of pork back fat to improve the nutritional quality of frankfurters are shown in Table 4. Control samples presented approximately 37.5 g/100 g of SFAs and 42 g/100 g MUFAs of the total of fatty acids, while T1, WBF, BPF, and BSF samples had SFA ranging from 28.76 to 29.23 g/100 g whereas MUFA corresponded to 38.51 g/100 g, 38.71 g/100 g, 37.88 g/100 g and 38.72 g/100 g, respectively. These significant differences in fatty acids compositions between control and treated samples were promoted by the incorporation of 50% of pre-emulsified sunflower oil in combination with unripe banana by-products to replace pork-back fat partially. WBF samples had higher total PUFA and MUFA values compared with T1, BPF and BSF (P < .05). Reports from other studies indicated that the use of vegetable oil as a fat animal substitute in meat products provided less proportion of SFA and generated a higher amount of PUFA, which improved the nutritional and healthy values of the final product.^[7,36,37] Similarly, our findings showed a significantly higher quantity of PUFA in samples formulated with partial replacement of fat compared to control. The incorporation of rice bran fibers associated with canola oil to partially replace animal fat decreased the amount of total SFA in frankfurter compared with treatments without vegetable oil.^[5] The fat-reduced sausage was mostly composed by oleic acid (35.96 to 39.49 g/100 g) followed by linoleic acid (18.29 to 31.98 g/100 g), palmitic acid (17.60 to 22.59 g/100 g) and stearic acid (9.44 to 12.53 g/100 g) whereas the other fatty acid compounds were lower than 4%. Previous studies have shown that there was a higher proportion of linoleic and oleic acid presented in the sunflower oil as well as high levels of vitamin E.^[15] The presence of these compounds in the pre-emulsified oil could be favorable to decrease the content of SFA in comminuted meat products. The increments of PUFA and the decrease of SFA could reduce the amount of plasma cholesterol and prevent the occurrence of coronary diseases.^[7] The decreased amount of SFA observed after the incorporation of WBF, BPF and BSF associated with pre-emulsified sunflower oil significantly augmented the ratio of PUFA/SFA from 0.57 to 1.17 (P < 0.05). These results are within the range established to assess the nutritional value recommended for the amounts of lipids (higher than 0.4) in food products.^[38] In general, fat-reduced frankfurter intake could provide beneficial effects to humans due to the enhanced fatty acid compounds present in the pre-emulsified sunflower oil.

Table 4. Fatty acid composition of fat-reduced frankfurter formulated with unripe banana by-products and pre-emulsified sunflower oil (g/100 g fatty acids)

Fatty acids	Control	T1	WBF	BPF	BSF
Miristic C14:0	1.13 ± 0.01^d	0.82 ± 0.01^b	0.81 ± 0.04^a	0.85 ± 0.01 ^c	0.85 ± 0.00 ^c
Palmitic C16:0	22.59 ± 0.31 ^c	17.60 ± 0.22^ab	17.28 ± 0.21^a	17.68 ± 0.18^b	17.77 ± 0.14^b
Heptadecanoic C17:0	0.27 ± 0.01^b	0.20 ± 0.01^a	0.20 ± 0.00^a	0.19 ± 0.01^a	0.20 ± 0.006^a
Stearic C18:0	12.53 ± 0.33 ^c	9.64 ± 0.08^ab	9.44 ± 0.06^a	9.69 ± 0.23^b	9.57 ± 0.12^ab
Arachidic C20:0	0.25 ± 0.02^a	0.28 ± 0.02^b	0.27 ± 0.00^ab	0.23 ± 0.00^a	0.24 ± 0.02^a
Others SFAs	0.65 ± 0.001^e	0.59 ± 0.004 ^c	0.61 ± 0.004^d	0.58 ± 0.001^b	0.57 ± 0.003^a
Palmitoleic C16:1	1.38 ± 0.01^d	1.02 ± 0.00 ^a	1.03 ± 0.00 ^b	1.05 ± 0.00 ^c	1.05 ± 0.01 ^c
Oleic acid C18:1n9 c	39.47 ± 0.58 ^c	36.42 ± 0.45^ab	35.96 ± 0.08^a	36.66 ± 0.11^b	36.75 ± 0.20^b
Eicosanoic C20:1	0.82 ± 0.02 ^c	0.66 ± 0.01^b	0.60 ± 0.01^a	0.65 ± 0.00^b	0.64 ± 0.00^b
Heptadecenoic C17:1	0.20 ± 0.00^a	0.17 ± 0.00^a	0.18 ± 0.01^a	0.19 ± 0.04^a	0.16 ± 0.01^a
Elaidic acid C18:1n9 t	0.10 ± 0.05^a	0.24 ± 0.05 ^c	0.11 ± 0.02^b	0.16 ± 0.08^b	0.12 ± 0.03^b
Linoleic c18:2n6 c	18.29 ± 0.15^a	30.60 ± 0.64^b	31.98 ± 0.39 ^c	30.45 ± 0.31^b	30.54 ± 0.22^b
Linolelaidic C18:2n6 t	0.15 ± 0.02^b	0.08 ± 0.01^a	0.08 ± 0.01^a	0.07 ± 0.01^a	N/D
γ Linolenic c18:3n6	0.04 ± 0.00^b	0.03 ± 0.00^a	0.03 ± 0.00^a	0.03 ± 0.00^a	0.03 ± 0.00^a
α linolenic C18:3n3	0.78 ± 0.003^d	0.54 ± 0.01^b	0.50 ± 0.001^a	0.55 ± 0.01 ^c	0.56 ± 0.01 ^c
Eicosadienoic C20:2	0.76 ± 0.01^d	0.52 ± 0.01^b	0.45 ± 0.001^a	0.54 ± 0.01 ^c	0.53 ± 0.01^bc
Eicosatrienoic C20:3n6	0.09 ± 0.00^b	0.08 ± 0.00^a	0.08 ± 0.01^a	0.08 ± 0.00^a	0.07 ± 0.01^a
Arachidonic C20:4n6	0.30 ± 0.00^a	0.28 ± 0.00^a	0.25 ± 0.05^a	0.24 ± 0.01^a	0.23 ± 0.06^a
Eicosatrienoic C20:3n3	0.14 ± 0.01^d	0.12 ± 0.00 ^c	0.08 ± 0.00^a	0.09 ± 0.00^b	0.09 ± 0.00^b
ΣSFA	37.46 ± 0.64 ^c	29.15 ± 0.42^b	28.67 ± 0.44^a	29.25 ± 0.33^b	29.23 ± 0.11^b
ΣMUFA	41.97 ± 0.66 ^c	38.51 ± 0.49^ab	37.88 ± 0.36^a	38.71 ± 0.23^b	38.72 ± 0.19^b
ΣPUFA	20.57 ± 0.28^a	32.25 ± 0.61^b	33.45 ± 0.46 ^c	32.04 ± 0.03^b	32.05 ± 0.20^b
PUFA/SFA	0.55 ± 0.07^a	1.11 ± 0.04^b	1.17 ± 0.09 ^c	1.10 ± 0.05^b	1.10 ± 0.05^b

Control = total pork-back fat; T1 = 50% of pork-back fat and 50% of sunflower oil; WBF = 3% of whole green banana flour; BPF = 3% of pulp banana flour; BSF = 3% of peel banana flour.

^a–dMeans ± SD within the row with different superscript letters (a-e) are significantly different (p < 0.05).

Saturated fatty acids (SFA); Monounsaturated fatty acids (MUFA); Polyunsaturated fatty acids (PUFA).

Rheological measurements

Figure 1 shows the dynamic viscoelastic changes of fat-reduced frankfurter containing unripe banana by-products and pre-emulsified sunflower oil during the heating process. The viscoelastic curves (G’) of frankfurter samples, with or without banana by-products, showed similar trends along with different phases that were similar to other studies reported using high molecular indigestible carbohydrates in meat emulsions.^[30,39] Firstly, G’ remained relatively constant during the initial heating process until 40°C. Then, it increased to a peak near 47°C, resulting from denaturation and temporary gelation of myofibrillar proteins. Following that, changes in oscillatory curves showed a major transition in energy loss (sudden fall) that occurred between 47 to 55°C. This fall might be caused by the denaturation of myofibrillar protein, mainly myosin tail and the disruption of protein matrix previously developed at the lower temperatures.^[27,32] Continuously, this phase was followed by a gradual increase in G’ at temperatures from 55°C to 80°C that were induced by the denaturation and aggregation of muscle protein during the heating process,^[7,27] showing changes from the viscous emulsion (raw emulsion) to the gelled or cooked emulsion. This suggests that the gel matrix structure was regenerated with increasing temperatures during the heating process. There was little difference in G’ transition curves observed in fat-reduced frankfurters samples suggesting that banana flour did not have any negative effects on the denaturation and interaction of myofibrillar protein during the heat-induced process.

Figure 1. Dynamic storage modulus (G’) of reduced-fat frankfurters containing banana by-products and pre-emulsified sunflower oil. Control (♦) = total pork-back fat; T1 (■) = 50% of pork-back fat and 50% of sunflower oil; WBF (─) = 3% of whole green banana flour; BPF (●) = 3% of pulp banana flour; BSF (+) = 3% of peel banana flour

The heating process causes the unfolding of the proteins to form a 3-dimensional gel matrix, which enables the structure to interact more readily with fat and water, thereby enhancing the product yield and rheological and textural properties of the final product.^[40] The incorporation of pre-emulsified sunflower oil without banana by-products (T1 sample) did not provide significant changes in the G’ values compared to control samples. Previous studies have indicated that unsaturated vegetable oils used to replace fat negatively affected the rheological properties of the meat emulsion by reducing its viscoelasticity. However, Álvarez et al.^[9] reported that the incorporation of canola-olive oils favored the formation of a gel network, which was enhanced by associating vegetable oil with the walnut extract. Therefore, this suggests that the addition of ingredients containing starch (mainly resistant starch) and fiber components could enable the use of vegetable oils, which may result in the enhancement of the gelling matrix. The formation of a higher gel structure (G’) observed in the fat-reduced frankfurter was more evident in samples containing BSF and BPF compared to control samples. Zhuang et al.^[30] found approached results after adding sugarcane dietary fiber as the fat substitute to produce comminuted meat products with fat-reduced content. It may result from the high concentration of indigestible carbohydrates such as resistant starch and fiber that combines with muscle proteins and vegetable oils to form an elastic and structured gel network after heating, which allows better binding of water and lipids molecules in the food matrix.

Instrumental texture profile analysis

Changes in the textural properties of fat-reduced frankfurters containing unripe banana by-products and pre-emulsified sunflower oil replacing pork back fat are shown in Table 2. Incorporation of 50% of pre-emulsified sunflower oil to replace fat lowered the hardness of frankfurters compared to control (P < .05), which gave a softer texture to the final product. Higher hardness was observed in control samples compared to other treatments (P < .05). These results agree with the previous study reported by Alves et al.^[21] The addition of WBF and BPF augmented the hardness values of fat-reduced frankfurter compared to T1 (P < .05), but it remained lower than 78.45 N values of control samples. Likewise, Choi et al.^[41] indicated lower hardness and gumminess values in fat-reduced frankfurters prepared with sunflower seed oil and makgeolli lees fiber when compared with high-fat samples. Bastos et al.^[22] also found lower hardness values in beef burgers formulated with flours from banana and oatmeal as fat substitutes. Regarding the use of vegetable oils, Álvarez et al.^[9] affirmed that the incorporation of vegetable oils lowered the hardness values of frankfurter type sausage, concluding that fat composition and quantity had a great influence on the textural characteristics of meat emulsions. Similarly, Salcedo-Sandoval et al.^[10] reported that the incorporation of healthier oil in the emulsified meat product as a fat substitute would generate a product with a softer texture. However, many studies have reported higher hardness values after the addition of pre-emulsified plant oil and dietary fiber from different sources into the meat emulsions,^[30,32] which was not observed in our study. Control samples presented lower springiness and cohesiveness values compared to T1, WBF and BPF (P < .05). Pre-emulsified sunflower oil added without unripe banana by-products significantly augmented the cohesiveness values of the samples (P < .05). The textural properties among types of unripe banana by-products showed that BSF significantly lowered the hardness values compared to WBF and BPF (P < .05). This significant trend on the textural parameter of frankfurters prepared with BSF was also observed in springiness, cohesiveness and chewiness values. Although BSF decreased the cooking loss, retarded the lipid oxidation and improved the thermal gelation of fat-reduced frankfurter, it negatively affected the textural parameters of the final cooked frankfurter. Similarly, Türker et al.^[20] reported that green banana peel had a negative effect on the hardness of free-gluten cakes.

Sensory evaluation

Sensory properties of fat-reduced frankfurters with WBF, BPF and BSF in combination with pre-emulsified sunflower oil are shown in Figure 2. These changes were mostly associated with the incorporation of WBF, BPF and BSF since T1 and control samples indicated a similar trend on the appearance, aroma and texture (P > .05). T1 and control samples had a significantly better appearance compared to other treatments (P < .05). This fact could be associated with processing method and enzymatic reaction that altered the appearance of BSF and WBF, which subsequently lowered the color of fat-reduced frankfurter. No difference in aroma/smell was detected among treated samples and control (P > .05). Similar trends in the sensory aspects, including aroma, texture and overall acceptability of bologna type sausage containing up to 60% of PSGBF gel as a fat substitute was previously reported.^[21] The flavor was more intense for T1 and BPF samples (P < .05). BSF samples had a less attractive appearance among all samples (P < .05). In addition, BSF also had a significantly lower impact on the sensorial parameters according to the panelists’ judgment (P < .05). The lower effects in BSF samples suggest that the flour extracted from unripe banana skin needs previous treatment to enhance particular properties such as appearance before its application as an ingredient in the meat emulsion. Previously, Pietrasik and Janz^[31] showed that replacing 50% of pork fat with pea flour rich in fiber and starch fraction significantly reduced the sensory quality of mortadella corroborating in part with the results in this study regarding the application of BSF. Overall, WBF and BPF associated with pre-emulsified sunflower oil showed the enhanced sensory qualities.

Figure 2. Sensory properties of fat-reduced frankfurter formulated with unripe banana by-products. Control (─) = total pork-back fat; T1 (─) = 50% of pork-back fat and 50% of sunflower oil; BPF (─) = 3% of pulp banana flour; WBF (─) = 3% of whole green banana flour; BSF (─) = 3% of peel banana flour

Conclusion

This study demonstrates that WBF, BPF and BSF when used as a source of dietary fiber in frankfurters, enabled the incorporation of a vegetable oil rich in PUFA n-3. Their combination with pre-emulsified sunflower oil to replace fat up to 50% lowered the cooking loss and enhanced the emulsion stability and rheological characteristics of fat-reduced frankfurters. Among the treated samples, BSF decreased cooking loss and improved the emulsion as well as oxidative stability compared to WBF and BPF, but it reduced the textural and sensorial aspects of the final product. The addition of pre-emulsified sunflower oil provided higher MUFA and PUFA values, which could enhance the healthier quality of fat-reduced frankfurter. Technologically, flour ingredients obtained from unripe banana associated with pre-emulsified sunflower oil can be applied as a valuable source of bioactive phenolic compounds, binder and filler agents to provide quality and nutrition values to emulsified meat products.

Additional information

Funding

This study was supported by The State Key Research and Development Plan “Modern Food Processing and Food Storage and Transportation Technology and Equipment” [2018YFD0400101], Jiangsu Province Agricultural Science and Technology Independent Innovation Funding [CX(19)2018] and the Fundamental Research Funds for the Central Universities [KYDZ201902].