Influence of tallow replacement by oat β-glucan and canola oil on the fatty acid and volatile compound profiles of low-fat beef burgers

ABSTRACT In view of the growing expectations of meat products consumers (quality and sensory properties), the meat industry must develop functional products to meet the consumer requirements. In the experiment, the influence of oat β-glucan, canola oil and both on the proximate composition, fatty acid profile, volatile compounds profile and sensory properties of beef burgers was analyzed. The beef tallow replacement by β-glucan, canola oil or both decreased the SFA and increased PUFA content. The consumption of one portion of product with canola oil and β-glucan can provide 57.6% of the daily amount of linolenic acid (C18:3n3). Volatile compounds analysis of grilled samples showed the presence of 1-penten-3-one and 2-methylpropanal only in case of groups with canola oil and butane-2-one in β-glucan group. This strategy improves not only the nutritional value but also the sensory properties and consumer acceptance. The obtained product may be labelled with health claims regarding oat β-glucan.


Introduction
Beef burgers are one of the most common fast food products in many developed countries in Europe, America and East Asia (Market Research Engine, 2018;Nathwani, 2017). Unfortunately, despite the increasing number of studies proving the lack of influence of SFA cardiovascular disease, present meat fat phobia, (especially fear of saturated or "solid" animal fats), nutritional value of beef fat is questioned by some public health authorities (Frank, Oytam, & Hughes, 2017;Siri-Tarino, Sun, Hu, & Krauss, 2010). Previous generations of consumers, among others in East and Southeast Asia, were choosing fat pieces of meat because of their better sensory features. Today, also some consumers appreciate the unique sensory characteristics of marbled beef. Health authorities, especially of the English-speaking countries, persuaded people that consumption of animal fat, including SFA and solid fats, should be reduced to maintain a healthy nutrition. Years of negative information about animal fats led to a general avoidance of fatty meat (Frank, Joo, & Warner, 2016). Twenty-first-century consumers pay significant attention to the quality of food products, including their fatty acid profile, which has an impact on their health (McManus, Merga, & Newton, 2011). Simultaneously, they are looking for products with attractive sensory features (Hathwar, Rai, Modi, & Narayan, 2012). Therefore, the meat industry must develop functional products to meet the consumer requirements (de Oliveira Fagundes et al., 2017;Marcinkowska-Lesiak, Poławska, Półtorak, & Wierzbicka, 2017;Munekata et al., 2017;Półtorak et al., 2018). There are several methods to modify the fatty acid profile of meat products (Ospina-E, Sierra-C, Ochoa, Pérez- Álvarez, & Fernández-López, 2012;Poławska et al., 2013). The fatty acid profile, cholesterol content and precursors of vitamins A and E in beef are mainly determined by marbling (amount of intramuscular fat -IMF) which depends on among others the animal breed and the method of animal feeding (Daley, Abbott, Doyle, Nader, & Larson, 2010;Troy, Tiwari, & Joo, 2016). Changing the feeding method can lead to a reduction of SFA concentrations and an increase of PUFA concentrations. These are the two main components of the fatty acid profile that are most criticised (Wood & Enser, 2017) Another solution to improve the unfavourable fatty acid profile in burgers is the substitution of beef tallow with vegetable fats (e.g. canola oil, chia oil, walnut oil or olive oil) (Bolumar, Toepfl, & Heinz, 2015;de Oliveira Fagundes et al., 2017;Pintado, Herrero, Jiménez-Colmenero, & Ruiz-Capillas, 2016;Serdaroğlu, Nacak, & Karabıyıkoğlu, 2017). Among the many vegetable fats, canola oil has a very beneficial fatty acid profile [rich in polyunsaturated fatty acids (PUFA 28.14%)] and the lowest SFA level among all oils (7.37%). Additionally, similar to vegetable fat, canola oil does not contain cholesterol and has valuable unsaturated/saturated fatty acids (UFA/SFA ratio) (USDA, 2016). Fatty acid profile modification by replacing beef tallow with vegetable oils, however, has many technological drawbacks. Animal fat plays an important technological role in meat products affecting, among others, texture, water-holding capacity (WHC) and rheological properties (Keeton, 1994;Selani et al., 2016). Intramuscular fat present in beef is responsible for volatile compounds development and more intense flavour perception (Frank, Joo, et al., 2016;Frank, Kaczmarska, Paterson, Piyasiri, & Warner, 2017). Additionally, animal fat replacement has a negative influence on sensory quality parameters (mainly on aroma, taste, juiciness and texture) and, consequently, decreases consumer acceptance of the product (Al-Abdullah, Al-Ismail, Al-Mrazeeq, Angor, & Ajo, 2002;Jiang, Busboom, Nelson, & Mengarelli, 2011;Selani et al., 2016). The addition of vegetable oil to meat products may also affect the profile of volatile compounds of the final product. The fatty acid profile composition of meat and the presence of large amounts of unsaturated fats (MUFA and PUFA), which are more vulnerable to lipid oxidation will determine the profile of volatile compounds. This type of fatty acids determinate the presence of among other aldehydes which are forming during oxidation (Frank, Ball, et al., 2016;Resconi et al., 2010).
A solution to the technological and quality problems connected with the replacement of animal fat by vegetable oils can be the addition of β-glucan. Oat β-glucan is a polysaccharide built from β-D-glucose molecules contained in the aleurone layer. Due to, nutritional and low caloric value oat β-glucan has positive European Food Safety Authority (EFSA) recommendations (EFSA, 2010(EFSA, , 2011. Thanks to its positive effect on human health, βglucan obtained by hydrothermal processing (without chemical reagents), can be used in shaping the sensory characteristics of low-fat products (Gangopadhyay, Hossain, Rai, & Brunton, 2015). Oat β-glucan, among others, contributes to the reduction of blood glucose rise after a meal and has beneficial influence on blood cholesterol levels (Commision Regulation (EU) No 432/2012, n.d;Fulgoni, Chu, Shea, Slavin, & Dirienzo, 2015;Yangilar, 2013). Additionally, it has texturing properties and also improves the rheological properties of food products (Afshari et al., 2015;Álvarez & Barbut, 2013;Pintado et al., 2016). Addition of oat β-glucan concentrates can increase cohesiveness, adhesiveness, gumminess and hardness of food products, also some products are more grainy. Oat β-glucan has also influence on flavour intensity scores. Food with β-glucan concentrates is characterized by a higher intensity of cereal and starchy. β-glucan can also decrease oily flavour intensity (Lee, Inglett, Palmquist, & Warner, 2009).
As part of the research was to develop a recipe of acceptable to consumers low-fat meat product, with sensory properties similar to the control sample, which would be composed of β-glucan by at least 1%. The resulting product should be able to adhere to the following health claims: 'β-glucans contribute to the maintenance of normal blood cholesterol levels' and 'the consumption of β-glucans from oats or barley as part of a meal contributes to the reduction of blood glucose rise after a meal' (Commision Regulation (EU) No 432/2012, n.d).

Treatments and beef burgers preparation
Beef chucks (source: Meat Plant Wierzejki Ltd, Poland) peeled on the outside from subcutaneous fat and connective tissue were chopped and minced in a meat grinder using a Ø 8 mm plate (PI-22-TU-T, Edesa, Spain). Samples were prepared for four different formulations: Control; Osubstitution of 40% beef tallow with canola oil; Bbeef tallow replacement by 30% oat β-glucan concentrate (4% of product); O + Bsubstitution of 40% beef tallow with canola oil and total fat replacement by 30% oat β-glucan concentrate (4% of product). The composition of control and low-fat burgers with added β-glucan concentrates and canola oil is presented in Table 1. Thirty percent β-glucan concentrate extracted without chemicals, using the hydrothermal extraction method, from the oat grain Avena Sativa aleuronic layer (Microstructure Inc., Poland) containing: 20.36% proteins, 10.69% carbohydrates, 8.65% fat and 64.12% fibre (34.10% insoluble fractions and 30.03% soluble fractions, including 30 g of β-glucan per 100 g) was used. All ingredients were mixed and 150 g burgers (2,1 cm thickness) were formed using an aluminium forming press (10 cm in diameter).
Products were heat treated in standardised conditions by electrical grill (Silex S-Tronic Single Grill S161GR, Germany), and pre-heated using top (210°C) and bottom (190°C) heating plates. Samples were heat treated for 180 s to reach 75°C at their geometric centre (TrackSense® Pro, Ellab, Denmark).

Proximate composition
Raw samples' basic composition (moisture, protein, fat ash and connective tissue content) was determined using a near-infrared Table 1. Proportion of ingredients used in prepared beef burgers with canola oil and fat replacement levels by 30% oat β-glucan concentrate. Treatments: Control; Osubstitution of 40% beef tallow with canola oil; Bbeef tallow replacement by 30% oat β-glucan concentrate (4% of product); O + Bsubstitution of 40% beef tallow with canola oil and total fat replacement by 30% oat β-glucan concentrate (4% of product).
Tabla 1. Proporción de ingredientes usados en las hamburguesas de carne de res preparadas con aceite de canola y niveles de remplazo de grasa por 30% de concentrado de β-glucano de avena. Tratamientos: Control; Osustitución de 40% del sebo de res con aceite de canola; Bremplazo de 30% del sebo de res por el concentrado de β-glucano de avena (4% del producto); O + Bsustitución de 40% del sebo de res con aceite de canola y remplazo total de grasa por 30% de concentrado de β-glucano de avena (4% del producto). spectrometer NIRFlex N-500 Büchi contained in a NIRFlex Solids module (Büchi Labortechnik AG, Switzerland) in the spectral range of 12,500-4,000 cm −1 in reflectance mode, and the application of Büchi Art. N. N555-501. Evaluation of basic composition was carried out in a near-infrared laboratory accredited by the Polish Centre for Accreditation FT-NIR (Accreditation No. AB 1670), according to the method described by Wyrwisz, Półtorak, Zalewska, Zaremba, and Wierzbicka (2012). One hundred grams of homogenised meat product was placed on a Petri dish and measured six times at a 32 scanning rate for each sample.

Fatty acid profile evaluation
Fatty acids were extracted from the raw and grilled samples with chloroform-methanol, according to the procedure by Folch, Lees, and Sloane (1956). The fatty acid methyl esters (FAMES) were formed according to method presented by Ichihara and Fukubayashi (2010 using a KOH solution in methanol. FAMES were extracted using water and hexane. The hexane layer (containing FAME) was dehydrated through anhydrous Na 2 SO 4 . For analysis extracted and dehydrated hexane was transferred to a vials. FAMES were analysed using a Shimadzu GC-2010 gas chromatograph with a flame ionisation detector (FID) and equipped with an RT® 2560 silica column (100 m × 0.25 mm ID and 0.2 µm film thickness) (RESTEK, USA). Chromatographic conditions were as follows: initial oven temperature of 140°C (held for 5 min), ramp at 48°C/min to 240°C and then held for 30 min. Injector and detector temperatures were 240°C and 260°C, respectively. Split ratio was 80:1 and the volume of injection was 1 μl. The carrier gas was helium with a flow rate of 1.0 ml/ min. Each sample was measured in triplicate. FAMES were identified by comparison of their retention times with those of the reference standards (Supelco TM 37 Component FAME mix, Sigma, St Louis, MO, USA). The results were expressed in grams per 100 g of fat of detected FAMEs.

Volatile compound profile evaluation
The profile of volatile compound was analysed in raw and grilled samples using Electronic Nose Heracles II (Alpha M.O. S., Toulouse, France). Analysis was based on ultrafast gas chromatography with headspace (HS-GC) according the method described in work Wojtasik-Kalinowska et al. (2016). The device is included: an automatic sampling system and system of detection containing non-polar MXT-5 column and slightly polar MXT-1701 (length: 10 m, diameter: 180 mm) connected to a FID. The device was operated using AlphaSoft software (Alpha Software Corporation, Massachusetts, USA). The burger samples (3 g) were placed in 20 ml headspace vials. The samples were capped with a Teflon-faced silicon rubber cap and placed in the automatic sampler of the headspace system. Each sample was incubated at 55°C for 15 min under 8.33 Hz agitation. Hydrogen (the carried gas) was circulated at 1 mL min −1 . The accumulated gas was injected by autosampler from the headspace into GC (injection speed 125 ml s −1 ). The injected volume was 3500 mL and the injector temperature was 200°C. The analytes were accumulated in a trap at 15°C (Tenax). The temperature program included: 60°C for 2 s; 3°C s −1 ramp to 270°C and kept for 20 s. The FID was 280°C. The method was calibrated by an alkane solution [n-butane to n-hex-adecane (Restek)] in order to convert retention time into Kovats indices (Goodner, 2008). The profile of volatile compound was evaluated in six repetitions in three independent biological replicates. Volatile compounds were recognised using the AroChemBase library. The library contains 44 000 compounds and includes also a base of sensory descriptors for each single compound. In this study, the C11 peak was chosen as the reference peak to be monitored in time by the software.

Sensory evaluation
Sensory evaluation of the grilled products was performed by 120 untrained panellists recruited among students (19-24 years old) of the Warsaw University of Life Sciences (WULS) according to modified Meat Standards Australia (MSA, 2008) method. Sensory session of burgers were performed at 23 ± 1°C in isolated rooms with white light. In each session attended 20 consumers. Coded burgers (three-digit codes) were served on plastic plates in random order (four samples per consumer + one link sample). Link product was served as a first before four test samples. Position and carryover effects were completely balanced using a 4 × 4 Latin square design to allocate products to consumers. Information about consumer acceptance were recorded according to method described by Szpicer, Onopiuk, Półtorak, and Wierzbicka (2018). For each sample, consumers were asked to indicate their degree of liking on a 100 mm line scale. Consumers placed a vertical line in range samples from 0 mm ('extremely dislike') to 100 mm ('extremely like'). Consumers were asked to evaluate the following descriptors: external appearance, colour, aroma, taste, juiciness, texture, overall acceptability. Variance due to panellists and sessions was considered for the sensory data.

Statistical analysis
The statistical analysis of the obtained results was carried out with Statistica 13.3 (StatSoft Inc., Tulsa, USA). Production and measurements were taken for each of three independent biological replications, where different ingredients were used. The results are presented as the mean. The effects of the treatment group were analysed using the general linear model (GLM) procedure with Tukey's test at a level of significance of α = 0.05.

Proximate composition
The proximate composition (content of moisture, fat, protein, ash and connective tissue) of samples from different treatments is presented in Table 2. The oat β-glucan waterbinding capacity was the reason of increase moisture value. Consequently, the moisture level of all samples with β-glucan addition (B and O + B samples) caused a significant increase of the moisture level (P < 0.05) (Ahmad, Muhammad, Zahoor, Nawaz, & Ahmed, 2010). Also, protein and ash value increased after addition of β-glucan and β-glucan with canola oil (B and O + B groups) (P < 0.05). Significant decrease of fat level was observed in samples oat β-glucan concentrate (B and O + B) (P < 0.05). The recipe modifications did not cause significant changes in connective tissue level (P > 0.05).

Fatty acid profile evaluation
The beef tallow and canola oil fatty acid profiles were analysed. Results of fatty acid profile analysis of raw material used in experiment (Table 3) are in line with expectations and are confirmed with reports presented by Daley et al. (2010) and Orsavova, Misurcova, Ambrozova, and Vicha (2015). The most abundant fatty acids in both samples was C18:1n9c, respectively, 33.66 g/100 gbeef tallow and 65.02 g/100 gcanola oil). Unfortunately, beef tallow was also characterized by a high content of C16:0 31.73 g/100 g in contrast to canola oil, which contained a high content of PUFA like C18:2n6c and C18:3n3c (respectively, 18.57 g/100g and 7.57 g/100g) (Daley et al., 2010;Orsavova et al., 2015). Comparison of fatty acid profiles showed that the canola oil has over 9 times lower total SFA content than beef tallow. This is one of the most important factors confirming the high quality of canola oil and its nutritional value (Bree, De, Laitinen, & Flöter, 2009).
The fatty acid profile of raw (Table 4) and grilled (Table 5) burgers is presented in Tables 4 and 5 respectively. In quantitative terms, palmitic acid (C16:0) and stearic acid (C18:0) were the major SFAs detected in raw and grilled burgers in all analysed groups. The most abundant MUFA was oleic acid (C18:1n9c). The beef tallow replacement by β-glucan (B group), canola oil (O group) or both (O + B group) decreased significantly the SFA content and increased PUFA level (P < 0.05). These observations are in agreement with data presented by Haghshenas et al. (2015), who evaluated the influence of carboxymethyl cellulose and β-glucan in shrimp nuggets and Belichovska, Pejkovski, Belichovska, Uzunoska, and Silovska-Nikolova (2017), who analysed the influence of vegetable oil content in chicken frankfurters, and noticed a decrease in SFA levels in meat products with fat replaced by β-glucan and canola oil. Linoleic acid (C18:2n6c) was the main PUFA found in raw and grilled products. As canola oil has a high content of linolenic acid (C18:3n3) (Table 3), a significant increase of this fatty acid was obtained in O and O + B groups (P < 0.05). As a result, the consumption of one portion (150 g) of grilled product can provide 67.8% (O group) and 57.6% (O + B group) of the daily amount of linolenic acid (C18:3n3) recommended by Table 2. Proximate composition of beef burgers (mean). Treatments: Control; Osubstitution of 40% beef tallow with canola oil; Bbeef tallow replacement by 30% oat β-glucan concentrate (4% of product); O + Bsubstitution of 40% beef tallow with canola oil and total fat replacement by 30% oat βglucan concentrate (4% of product).
Tabla 3. Perfil de ácidos grasos de sebo de res y aceite de canola utilizado en la elaboración de hamburguesas bajas en grasa (expresado como g/100 g de ácidos grasos).  the European Food Safety Authority EFSA (2010). This high linolenic acid concentration (C18:3n3) caused higher PUFA levels (P < .05) and an increase of the PUFA/SFA ratio (P < .05) in the raw and grilled O and O + B groups, compared to the control samples. A very advantageous content of linoleic acid (C18:3n3) is the discovery of the beta-glucan structure-binding function. A similar trend was noted for group B but this increase was much lower than in the case of samples with Table 4. Effect of the partial replacement of beef tallow by canola oil; oat β-glucan concentrate or both of theme on fatty acids profile (expressed as g/100 g of fatty acids) of raw beef burger.   (2007); LA/LNA: relación de ácido linoleico/α-linolénico n6/n3: relación omega-6/omega-3 ND: no detectado. (a, b, c, d)significa con letras diferentes en mostrar un efecto significativo del grupo de tratamiento P ≤ 0.05 canola oil. Analysis showed higher n6/n3 ratio in raw control samples compared to the recommended value (between 1:1 and 2:1) (Simopoulos, 2011). Modification of the recipe had a significant effect on the n6/n3 ratio in raw and gilled samples. The substitution of beef tallow with canola oil (O and O + B samples) significantly lowered the n6/n3 ratio in relation to the control samples (P < 0.05). In contrast to the samples with canola oil, the raw burgers with β-glucan (B) were characterized by a significantly higher n6/n3 ratio (P < 0.05). The same tendency of n6/n3 ratio significant changes was observed in Table 5. Effect of the partial replacement of beef tallow by canola oil; oat β-glucan concentrate or both of theme on fatty acids profile (expressed as g/100 g of fatty acids) of grilled beef burger.
Tabla 5. Efecto del remplazo parcial de sebo de res por aceite de canola, por concentrado de β-glucano de avena o por ambos en el perfil de ácidos grasos (expresado como g/100 g de ácidos grasos) de hamburguesa de carne de res a la parrilla.   (2007); LA/LNA: relación de ácido linoleico/α-linolénico n6/n3: relación omega-6/omega-3 ND: no detectado. (a, b, c, d)significa con letras diferentes en mostrar un efecto significativo del grupo de tratamiento P ≤ 0.05 grilled samples (P < 0.05). On the other hand, Afshari, Hosseini, Khaneghah, and Khaksar (2017) showed the decrease of n6/n3 ratio in burgers formulated with total beef fat replacement by canola oil and olive oil with soy protein isolate containing inulin and β-glucan. Thermal treatment of the samples resulted in an increase of SFA content in all analyzed samples. The highest increase (about 10 g/100 g) was recorded for the O + B sample. A decrease in MUFA content caused by grilling was observed in all analysed samples; however, the highest changes were observed in samples O and O + B. Similar changes caused by grilling were observed for PUFA content. Despite the changes caused by thermal treatment, O and O + B burgers were characterized by the highest PUFA/SFA ratio among the remaining samples. Obtained products will ensure adequate supply of linoleic acid (C18:2n6) and alpha-linolenic acid (C18:3n3) in dietsufficient intake (AI) set at 4% and 0.5%, respectively, of the energy value of the diet (EFSA, 2011), and help in the implementation of the 'Twelve steps to healthy eating' recommended by the World Health Organization (replacing the majority of saturated fats by unsaturated fats) (WHO, 2000). Tables 6 and 7 show the 26 characteristic volatile compounds identified in all analysed groups of raw and grilled samples, respectively. In the case of raw burgers, 12 compounds were identified in the control group, 13 in O group, 14 in B group and 12 in O + B group. In the case of grilled meat, there were 18 compounds in the control group, O and O + B groups, and 17 in the B group. Raw meat is characterized by a very weak odour; however, it constitutes a matrix rich in non-volatile precursors of volatile compounds responsible for the development of meat products flavor (Kosowska, Majcher, & Fortuna, 2017). Free amino acids, particularly the sulfuric ones like cysteine and methionine, are the basic substrates in Maillard reactions and Strecker's degradation reaction (Whitfield, 2009). The Strecker degradation of amino acids is one of the main reactions leading to the final aroma compounds in the Maillard reaction (Villaverde, Ventanas, & Estévez, 2014). Only in the case of grilled beef the formation of Strecker's aldehydes (3-methylbutanal) has been observed. Among other reported  previously in literature compounds derived from Maillard and Strecker reactions are for example; methylpyrazine, 2-ethyl-3,5-dimethyl-pyrazine, trimethylpyrazine, 2-acetyl-1-pyrroline and 2-acetyl-2-thiazoline . Keton (butane-2,3-dione) was present in all analyzed groups, except grilled sample (B). The compound can be formed from the 2,3-enolization pathway which form part of the Maillard reaction. The presence of these compound in raw meat can be associated with pH value (Legako, Dinh, Miller, Adhikari, & Brooks, 2016). Aldehydes like propanal and hexanal are often used as indicators of the lipid oxidation in foods, because they can be measured in the sample headspace, and their lack of double bonds makes them more oxidatively stable than unsaturated aldehydes. Modification of the recipe also caused an increase in relative area of peaks of (E,E)-2,4-heptadienal in grilled samples. This compound is also characteristic for the meat of animals fed only in pastures (Resconi et al., 2012). Lipid oxidation is one of the main reasons of alcohols generation in meat (Martin, Antequera, Muriel, Perez-Palacios, & Ruiz, 2009). 1-propanol is produced by reducing corresponding aldehydes derived from lipid oxidation (Garcia et al., 1991). The current study showed that 1-propanol was detected in the control and B groups only in raw burgers, whereas the presence of 2-furanmethanol was noticed in the control group of grilled burgers. 2-furanmethanol is a product of Maillard reactions (Ames, Guy, & Kipping, 2001). The formation of alcohol compounds like 2-furanmethanol is triggered by thermal degradations of fatty acids (Elmore, Campo, Enser, & Mottram, 2002;Shahidi, 1994) and Maillard reactions (Ba, Ryu, & Inho, 2012). In the case of canola oil, the off-flavours formed during heating are characterised by a fishy smell. A compound, 1-penten-3-one, is suspected to be responsible for this (Sghaier et al., 2016). In this study, the presence of 1-penten-3-one was observed only in the case of groups where canola oil was added. Jiang et al. (2011) reported that the addition of PUFAs to the minced beef resulted in a decrease in beef flavour and an increase in off-flavour. Dimethyl sulfide and dimethyl disulfide were present in grilled burgers (control, B, O + B groups). Both dimethyl sulfide and dimethyl disulfide were not detected in groups where canola oil was added. It can be assumed that in groups with oil addition, the characteristic meaty flavour was not as intense, which was correlated with the consumers' evaluation. 2,3 dimethylpyrazine was identified in all grilled burgers. Compounds identified both in raw and grilled meat included: propane, ethanol, nonane, ethyl isobutyrate, 1R-(+)alpha-Pinene and beta-Pinene. Changes in the volatile compound profile in samples with β-glucan addition are probably caused by the addition of fibres, which by consequence decrease the meaty aroma (Sánchez-Zapata et al., 2010;Troutt et al., 1992). Additionally, the source of flavours in grilled meat products are mainly amino acids and lipids. Substitutes of meat fat also influence the volatile compound profile by altering flavour compounds and/or by reducing the primary flavour-generating source (meat fat) (Brewer, 2012). The most common fat substitutes, including oat fibre, can increase volatile compound generation in pyrazines and sulphur-containing compounds (Wood, 2011) and some, like maltodextrins and tapioca, can decrease their release (among others they inhibit the release of Maillard products) (Chevance et al., 2000).

Sensory evaluation
The results of sensory evaluations are shown in Table 8. Sensory parameters acceptance changes were connected with changes caused by recipe modifications. The consumers showed different acceptance between the samples. Sensory evaluation showed that the most appreciated samples with modified composition were those with canola oil and β-glucan addition (O + B burgers). Acceptance of all sensory parameters of O + B samples was very similar to control burgers' acceptance. The addition of β-glucan (B group) decreased the acceptance of most of the quality parameters (aroma, taste, texture, juiciness and overall acceptance) significantly (P < 0.05). A decrease of juiciness acceptance was also detected in samples with canola oil (O group). The results of work presented by Piñero et al. (2008) showed no influence of oat soluble fibres on the tenderness, appearance and colour acceptance of meat products. The study of Afshari et al. (2017), however, revealed that barley βglucan does not change the overall acceptability and colour of sausages.

Conclusions
The obtained results showed that the addition of canola oil and β-glucan as tallow replacements in beef burgers significantly improved the quality of fatty acid profile and, by consequence, changed the profile of volatile compounds. The consumption of one portion (150 g) of product with canola oil and β-glucan addition can provide 57.6% of the daily amount of linolenic acid (C18:3n3). This strategy improves not only the nutritional value but also the sensory properties and consumer acceptance of functional low-fat meat products. Thus, there is potential in the use of canola oil and β-glucan for the creation of functional low-fat beef products with acceptable volatile compound and beneficial fatty acid profiles. Table 8. Effect of the partial replacement of beef tallow by canola oil; oat β-glucan concentrate or both of theme on consumer acceptance of grilled beef burgers (mean).
Tabla 8. Efecto del remplazo parcial de sebo de res por aceite de canola, por concentrado de β-glucano de avena, o por ambos sobre la aceptación del consumidor de hamburguesas de carne de res a la parrilla (media).  (A, B, C) means with different letters in column show significant effect of treatment group; * -P ≤ 0.05 (A, B, C) -Las medias con letras diferentes en la columna muestran un efecto significativo del grupo de tratamiento; * -P ≤ 0.05