Aroma profile and consumer liking of salted and dried chicken meat: Effects of desalting and cooking methods

ABSTRACT

Twenty-four dried and salted broiler breasts, a chicken charqui–like muscle food, were submitted to desalting in cold water and different cooking methods (grilling, roasting, frying with olive oil, and sous-vide cooking). Samples were assessed by levels of thiobarbituric acid reactive substances, volatile compounds, and odor-liking analysis. Desalted and sous-vide cooked samples presented the highest levels of thiobarbituric acid reactive substances and volatile compounds derived from lipid oxidation. Grilling, roasting, and frying promoted an increase of Maillard-derived volatiles in dried salted broiler meat. This difference in aromatic profile led to a high odor acceptability of samples submitted to more intense thermal treatments such as grilling, roasting, and frying. In contrast, prolonged cooking (sous-vide technique) and desalting in cold water led to a marked decline in consumers’ liking due the rise in volatile compounds derived from lipid oxidation. In conclusion, based on the results, it seemed to be reasonable to hypothesize that dried and salted broiler meat submitted to high temperatures during cooking (≈180–200°C) tended to generate high Maillard-derived volatiles, while long cooking times led to a great thermal degradation/oxidation of lipids, which generates high content of lipid-derived volatile components.

KEYWORDS:

• Broiler meat
• Lipid oxidation
• Cooking technique
• Volatile compounds
• Odor-liking

Introduction

The application of heat treatment has become necessary for consumption of meat and meat products due to the improvement of their palatability and microbiological stability.^[1] The thermal treatment promotes flavor changes in meat mainly due to their impact on lipid oxidation reactions, being the major concern in cooked meat products.^[2] Lipid and water-soluble compounds present in raw meat such as unsaturated fatty acids, amino acids, peptides, reducing sugars, and other organic compounds interact during cooking to produce characteristic volatile compounds of cooked meat.^[3]

The oxidative deterioration of unsaturated lipids has been related to meat aroma depletion during its storage.^[4] In addition, cooking has a remarkable impact on the generation of reactive oxygen species (ROS), and may increase the rate of lipid and protein oxidation in meat systems.^[5] The rancid flavor in meat products results from lipid oxidation and is directly linked to the formation of specific volatile compounds such as acids, aldehydes, ketones, and alcohols. In fact, hexanal and pentanal has been recognized as a primary marker of meat flavor deterioration.^[6] However, it is worth noting that oxidized lipids have a marked role in aroma formation of cured and fermented meat products,^[7,8] where the oxidative breakdown may increase the amount of free fatty acids during ripening and promote the formation of desirable aged aroma.

Salted and subsequently dried meat products are very common worldwide. Charqui is a salted and dried meat product traditionally made from beef and very popular in South American countries. Charqui meat has to be desalted before consumption due to their high salt content. During the processing of charqui, a series of biochemical reactions responsible for flavor development occur. The high salt content plays an important role as a catalyst of lipid oxidation in charqui, which increases the generation of volatile compounds responsible for aroma development. In fact, most of volatile compounds of charqui meat are generated from lipid oxidative reactions.^[9]

There is no information about the influence of desalting and cooking methods on the formation of volatile compounds and aroma acceptability of chicken charqui–like meat. Hence, the aim of this study was to evaluate the effect of desalting and four different cooking procedures (roasting, grilling, frying with olive oil, and sous-vide cooking) on lipid oxidation, odor-liking, and aroma profile of salted and dried broiler meat (chicken charqui–like product).

Material and methods

Processing of salted and dried broiler meat and sample preparation

Dried salted broiler meat processing was performed as follows. Briefly, 24 broiler breasts (pectoralis major) were submitted to dry-salting (72 h) using coarse marine salt and 50 mg/kg of sodium nitrite, followed by a drying phase (45°C) in a drying chamber (Excalibur®, Food dehydrator 3000, Sacramento, USA) until the moisture and water activity (Aw) levels reached values below 50 g/100 g and 0.75, respectively. After processing, dried salted broiler samples were sliced into 30 mm thick steaks and divided into six groups. One group was used as a control (raw charqui) and the other groups were desalted in distillated water at 4°C for 24 h (5 L of distillated water/kg charqui). The water was changed at each 8 h. Four desalted samples were separated for further analysis (desalted charqui). Afterward, four types of heat treatment were performed in the remaining desalted samples using the following methods: (A) grilled at 180–200°C for 5 min at each surface using an electrical griddle (Repagas, 550 series, Madrid, Spain); (B) roasted at 200°C for 15 min using an electrical oven (Unox®, Model GN2.1, Cadonegue, Italy); (C) fried using 15 mL olive oil at 160–180°C for 5 min at each surface; and (D) sous-vide cooked in a thermostatized water bath at 65°C for 8 h. A heating treatment was considered completed when samples reached an internal temperature of 75°C, which was monitored during cooking using a digital probe thermometer (Testo thermocouple, Model 735-1, Lenzkirch, Germany). After cooking and cooling, samples were vacuum-packed and stored at –80°C until laboratory analysis was carried out.

TBARs measurement

Thiobarbituric acid reactive substance (TBAR) values of dried salted broiler meat were determined using the 2-thiobarbituric acid (TBA) method of Estévez and Cava^[10] and calculated from a standard curve prepared using a 1,1,3,3-tetraethoxypropane (TEP) solution (25 nmol/mL) in 3.86% perchloric acid. Briefly, 2.5 g of minced samples was dispensed in conical plastic tubes and homogenized with 7.5 mL of perchloric acid (3.86%) and 0.25 mL of BHT (4.2% in ethanol) using an ultraturrax homogenizer for 1 min. During homogenization, plastic tubes were immersed in an ice bath to minimize the oxidative reactions during extraction of TBARs. The homogenate was filtered and centrifuged at 3500 RPM for 3 min. Aliquots (2.0 mL) were mixed with 2 mL thiobarbituric acid (0.02 M) in glass tubes. Then, the tubes were placed in a hot water bath (90ºC) for 30 min together with a standard curve and the blank (2 mL perchloric acid + 2 mL TBA). After cooling, the absorbance was measured at 532 nm. Results were expressed as milligrams of MDA per kilogram of sample.

Odor-liking assessment

The odor-liking assessment of dried salted broiler samples was performed by 52 semi-trained panelists (staff and students from the Veterinary Faculty in Caceres, Spain, who frequently participate in sensory assessments of meat and meat products). A 9 cm linear nonstructured quantitative scale from extremely disliked to extremely liked was used. Five grams of each sample were finely minced, dispensed in falcon tubes, sealed, and wrapped with aluminum foil and offered to the panelists after being warmed up to 37°C for 10 min in an electric oven.

Volatile compound profile

Volatile compounds were analyzed in raw, desalted, and cooked samples by headspace-SPME and GC/MS (gas chromatograph Hewlett–Packard 5890 series II coupled to a mass selective detector Hewlett–Packard HHP-5791A) following the method described by Estévez et al.^[11] One gram of minced sample was weighed into a 4 mL vial, which was closed with a teflon/silicone septum (National Scientific) and preconditioned for 10 min at 37°C. An SPME fiber (50/30 μm divinylbenzene–carboxen–polydimethylsiloxane coating) was preconditioned prior analysis at 220°C during 45 min and then inserted through the septum and exposed to the headspace of the vial. After extraction, the SPME fiber was immediately transferred to the injector of the chromatograph (splitless mode). The separation of volatile compounds was performed on a 5% phenyl–methyl silicone (HHP-5)-bonded-phase fused-silica capillary column (Hewlett–Packard, 50 m × 0.32 mm i.d., film thickness 1.05 m). The carrier gas was helium at 18.5 psi (flow rate of 1.6 mL/min) at 40°C. The SPME fiber was desorbed and maintained in the injection port at 220°C during the whole chromatography run. The temperature program was isothermal for 10 min at 40°C and then increased at the rate of 7°C/min to 250°C and held for 5 min. The GC-MS transfer line temperature was 270°C. The mass spectrometer operated in the electron impact mode with electron energy of 70 eV and a multiplier voltage of 1650 V, and data were collected at a rate of 1 scan/s over a range of m/z 40–300. Compounds were tentatively identified by comparing their linear retention indexes/mass spectra (LRI/MS) with those from the libraries Wiley/NIST, or were positively identified by comparing their LRI/MS with those from standard compounds (Sigma-Aldrich, Steinheim, Germany).

Statistical analysis

A one-way analysis of variance (ANOVA) followed by Tukey’s test was used to analyze the influence of desalting and cooking methods on each measured parameter of dried salted broiler meat samples. A Kruskal–Wallis analysis was applied to sensory data. A principal component analysis (PCA) was also applied to chemical and sensory variables.

Results and discussion

Impact of desalting and cooking methods on TBARs

The evaluation of TBARs was performed in raw, desalted, and cooked samples (Fig. 1). Overall, TBARs content remained at low levels and ranged from 0.09 to 0.23 mg of MDA/kg of sample. The low TBARs numbers observed may be attributed to a possible antioxidant effect of NaCl at high concentrations. Several authors have reported the pro-oxidative activity of sodium chloride (NaCl) in meat systems.^[12,13] However, according to Rhee et al.,^[14] high NaCl concentrations (above 2.5%) in meat products formulation may prevent oxidative reactions. Although this antioxidant mechanism of NaCl is not fully elucidated, Srinivasan and Xiong ^[15] explained that the capability of NaCl to displace iron ions from binding sites might alter the speed of lipid oxidation. However, the desalting procedure promoted a remarkable increase in TBAR values compared with raw and cooked samples. In fact, a rise of 64% in TBARs content was observed after desalting compared with raw samples. This finding may be due to the pro-oxidative environment, such as increased water activity, created during prolonged water-based desalting procedures applied to salted and dried meat products.

Figure 1. TBARs (mean ± standard deviation) in chicken charqui: raw, desalted, and subjected to assorted cooking methods.

Different letters on top of bars denote significant differences between treatments in ANOVA (p < 0.05).

The cooking methods did not increase TBARs content compared with uncooked samples. In fact, a significant decrease of TBARs was noted in roasted treatments compared with raw ones. There were no significant differences in TBARs between raw, grilled, fried, and sous-vide cooked samples, which remained below 0.17 mg of MDA/kg of sample. The low levels of TBARs found in all samples may not be interpreted as if lipid oxidation did not occur. The thermal degradation of fatty acids has been postulated to form lipid oxidation derivatives with low odor thresholds such as aldehydes, which react with other unstable compounds formed by Maillard and Strecker reactions during the cooking of the meat, decreasing TBARs content.^[16] The lowest values of TBARs observed in roasted and fried samples may also be attributed to a higher formation of Maillard products (data not shown), which has been recognized to have antioxidant properties.^[17] Rocha Garcia et al.^[18] also reported low levels of TBARs content in salted and dried breast meat from laying hens and ascribed this behavior to the addition of sodium nitrite in product formulation. However, Özcan and Bozkurt^[19] observed an increased TBAR value after cooking processing of a ready-to-eat meat product. Different thermal treatment conditions (time and temperature) and also the extraction conditions of TBARs analysis may explain the low concentration of these compounds in our samples.

Impact of desalting and cooking method on volatiles’ profile

The influence of desalting and different cooking procedures on the volatiles’ profile of salted and dried broiler meat submitted to four different cooking methods is presented in Table 1. A total of 65 volatile compounds were identified and classified in 11 chemical families: acids (4), aldehydes (23), alcohols (4), esters (5), furans (3), hydrocarbons (7), ketones (7), nitrogen compounds (3), sulfur compounds (1), terpene (1), and other compounds (7).

Table 1. Volatile compounds (AU × 10⁶) detected in headspace of raw, desalted, and cooked chicken charqui.

				Cooked
	Reliability of detection^a	Raw	Desalted	Grilled	Roasted	Fried	Sous-Vide	SEM^b	p-Value^c
Acids
Hexanoic acid	A	0.53c	5.89a	0.89b	1.54b	0.98b	4.69a	0.31	*
Heptanoic acid	B	0.00b	1.56a	0.45a	0.98a	0.78a	1.56a	0.08	*
Octanoic acid	A	0.23b	0.89a	0.94a	1.00a	0.00c	1.78a	0.09	*
Nonanoic acid	B	0.00c	1.48a	0.00c	0.89a	0.00c	1.04a	0.10	*
Total acids		0.76c	9.82a	2.28bc	4.41b	1.76bc	9.07a	0.56	**
Aldehydes
2-(E)-propenal	A	0.00c	2.56b	0.00c	2.36b	0.00c	4.56a	0.24	**
2-Methyl-propanal	A	0.00c	0.63bc	5.69a	0.56bc	3.95ab	1.87b	0.31	**
3-Methyl-butanal	A	0.00c	0.98b	9.69a	1.58b	8.55a	0.98b	0.63	**
2-Methyl-butanal	A	0.00c	1.45b	12.58a	1.89b	10.25a	0.85b	0.85	***
2-(E)-butenal	A	0.00c	3.60a	0.89b	2.88ab	1.56ab	4.39a	0.19	*
Pentanal	A	1.81b	9.87a	1.89b	5.71ab	2.47b	3.76b	0.42	*
2-(E)-hexenal	A	0.00c	1.06b	0.00c	0.00c	0.00c	3.25a	0.18	*
Hexanal	A	2.95d	30.04a	8.86c	21.57b	11.65bc	18.83b	1.50	***
Heptanal	A	1.08c	14.87a	5.89b	9.99a	4.20b	10.02a	0.79	***
2-(E)-heptenal	A	0.00c	2.12ab	0.00c	1.08b	0.00c	3.62a	0.29	*
3-Methyltiopropanal	A	0.00c	0.00c	1.61ab	0.56b	2.30a	0.00c	0.15	*
Octanal	A	0.96c	19.28a	6.98b	4.44b	8.52b	14.63a	0.97	***
2,4-(E,Z)-heptadienal	B	0.00b	2.14a	0.00b	0.89a	0.00b	1.45a	0.15	*
Benzaldehyde	A	0.00c	1.45ab	2.88a	1.22ab	0.88b	2.69a	0.18	*
Benceneacetaldehyde	A	0.00c	0.00c	4.12a	0.66b	3.66a	0.00c	0.29	*
2-(E)-octenal	A	0.00d	0.25c	0.00d	0.98b	0.00d	2.74a	0.17	*
Nonanal	A	1.99c	21.36a	9.36b	16.63a	9.69b	12.08ab	1.09	***
2-(E)-nonenal	A	0.00b	0.98a	0.00b	0.87a	0.00b	1.40a	0.09	*
Decanal	A	0.77c	7.08a	3.00b	5.69a	3.24b	6.64a	0.40	***
2-(E)-decenal	A	0.00c	0.74a	0.00c	1.54a	0.00c	2.10a	0.14	*
2,4-(E,Z)-decadienal	B	0.00b	1.26a	0.00b	0.89a	0.00b	1.04a	0.08	*
2,4-(E,E)-decadienal	B	0.00c	1.89a	0.00c	0.24b	0.00c	0.96a	0.11	*
Dodecanal		0.00b	3.10a	1.14a	2.00a	1.78a	3.65a	0.19	*
Total aldehydes		9.56c	126.71a	74.58b	84.23b	72.70b	101.51a	6.09	***
Alcohols
1-Pentanol	A	1.01c	9.36a	2.15b	5.69ab	3.10b	9.54a	0.64	***
1-Octen-3-ol	A	1.23b	4.69a	1.27b	3.87ab	2.74b	6.75a	0.39	*
1-Hexanol	A	0.96c	7.40a	3.60b	3.65b	2.55b	7.66a	0.44	**
1-Octanol	A	0.00c	2.39a	1.65ab	3.33a	0.80b	2.36a	0.12	*
Total alcohols		3.20c	23.84a	8.67b	16.54ab	9.19b	26.31a	1.53	***
Esters
Acetic acid ethyl ester	A	1.23a	3.25a	0.00b	0.00b	0.00b	0.00b	0.19	**
Hexanoic acid ethyl ester	B	0.00b	4.01a	0.00b	0.00b	0.00b	0.00b	0.27	***
Heptanoic acid ethyl ester	B	0.00b	2.01a	0.00b	0.00b	0.00b	0.00b	0.15	*
Octanoic acid methyl ester	B	0.00b	1.99a	0.96a	0.54a	0.88a	0.36a	0.10	*
Acetic acid phenyl metil ester	C	0.00c	3.55a	1.02b	0.98b	1.23b	0.87b	0.18	*
Total esters		1.23b	14.81a	1.98b	1.52b	2.11b	1.23b	0.69	***
Furans
2,5-Dihidro-furan	A	0.00b	0.00b	1.25a	0.69a	1.58a	0.99a	0.04	*
2-(3H)-Dihidrofuranone	A	0.00c	0.00c	2.69b	0.98b	4.69a	1.02b	0.36	*
2-Pentylfuran	A	0.00c	5.69a	2.66b	1.98b	2.65b	5.60a	0.44	*
Total furans		0.00c	5.69a	6.60a	3.65b	8.92a	7.61a	0.84	***
Hydrocarbons
2-Methyl-pentane	A	0.00b	2.36a	0.00b	0.00b	0.00b	0.00b	0.18	**
3-Methyl-pentane	A	0.00b	4.52a	0.00b	0.00b	0.00b	0.00b	0.28	***
Heptane	A	0.98	1.55	0.60	0.41	0.95	0.47	0.08	ns
1-Octene	B	0.00c	6.17a	2.54b	2.01b	1.99b	4.10a	0.31	*
Decane	A	4.23	2.60	5.60	5.14	3.55	1.22	0.22	ns
Undecane	A	2.36	2.64	2.31	6.14	6.11	2.11	0.31	ns
Dodecane	A	0.00b	3.10a	2.55a	3.65a	1.11a	3.61a	0.20	*
Total hydrocarbons		7.57b	22.94a	13.63a	17.35a	13.71a	11.51a	1.24	***
Ketones
2-Heptanone	A	0.31c	7.69a	2.15b	6.54a	1.99b	8.99a	0.78	***
2,5-Hexanedione	B	0.00b	3.74a	0.00b	1.88a	0.00b	2.11a	0.18	**
1-Phenyl-ethanone	B	0.00b	1.26a	0.00b	0.00b	0.00b	0.00b	0.04	*
2,3-Octanedione	B	0.00c	11.69a	0.00c	4.11b	0.00c	6.77b	0.81	***
4-Octanone	A	0.00c	3.64a	0.88b	2.30a	0.00c	2.10a	0.19	**
2-Nonanone	A	0.23c	4.88a	1.21b	2.08b	1.32b	1.99b	0.21	*
3-Nonen-2-one	A	0.00c	2.31a	0.77b	1.47a	0.00c	1.14a	0.18	*
Total ketones		0.54d	35.21a	5.01c	18.38b	3.31c	23.10ab	1.99	***
Nitrogen compounds
2-Pyrrolidone	A	0.00c	0.00c	6.30a	0.87b	5.43a	0.00c	0.66	***
Pyridine	A	0.00c	0.00c	2.10a	0.96b	2.30a	0.00c	0.13	**
2-Methyl-pirazine	A	0.00c	0.00c	2.99ab	0.96b	4.12a	0.00c	0.20	**
Total nitrogen compounds		0.00c	0.00c	11.39a	2.79b	11.85a	0.00c	0.87	***
Sulphur compounds
Dimethylsulphide	A	1.02b	5.66a	6.20a	2.63b	6.32a	1.22b	0.39	**
Terpenes
Limonene	A	3.33ab	1.36b	0.84b	0.26	6.58a	1.23b	0.26	**
Others
Methyl-benzene	A	2.36b	5.66ab	8.22a	6.33ab	4.55ab	2.66b	0.31	*
Ethyl-benzene	A	3.48	2.71	2.99	2.74	2.99	2.10	0.07	ns
1,2-Dimethyl-benzene	B	0.00b	6.00a	4.33a	4.78a	6.44a	7.15a	0.41	**
1,3-Dimethyl-benzene	B	0.00c	5.31ab	2.55b	6.41ab	4.56ab	9.65a	0.46	**
1-Methylethyl-benzene	B	0.00c	2.11b	1.47b	2.00b	6.36ab	9.69a	0.59	**
1-Mmethyl-4-(1-methyl-ethyl)-benzene	C	0.00c	1.88b	6.50a	6.44a	5.41a	7.56a	0.46	**
Total others		5.84b	23.67a	26.06a	28.7a	30.31a	38.81a	1.49	***

SEM, standard error of the mean. Different letters in the same row denotes significant differences between treatments.

^aA, positive identification by comparing LRI/MS with that of standard compounds; B, tentative identification by comparing LRI/MS with those from the libraries Wiley/NIST; C, tentative identification by comparing MS with those from the libraries Wiley/NIST.

^bStandard error of the mean.^cSignificance level in ANOVA: *p < 0.05; **p < 0.01; ***p < 0.001.

Significant differences were observed for most volatiles extracted from experimental samples, indicating a high variability in the volatile profile of samples as affected by the processing stages. Desalted and sous-vide cooked samples had the highest volatile content (269.7 and 221.6 AU × 10⁶, respectively), followed by roasted (180.4 AU × 10⁶), fried (166.8 AU × 10⁶), grilled (157.2 AU × 10⁶), and raw (33.1 AU × 10⁶) treatments. Aldehydes were the main chemical family present in the headspace of all treatments.

Raw samples showed a high amount of aldehydes and hydrocarbons, which presented six and three different compounds, respectively. Among lipid-derived aldehydes, hexanal was the most abundant in all treatments. This indicates that most volatiles responsible for the characteristic aroma of raw and cooked chicken charqui are formed from the oxidation of polyunsaturated fatty acids. According to Jónsdóttir et al.,^[20] characteristic flavor of salted products is generated from lipid and protein degradation. The harsh conditions applied to charqui-type meat products such as the heavy salting and the subsequent drying phase may be responsible for this lipid degradation and flavor formation. Salvá et al.^[21] also found high content of aliphatic hydrocarbons and aldehydes in alpaca charqui, a type of dried salted meat product. Also, a high area percentage of hexanal was reported by Silva et al.^[7] in charqui prepared from broiler and laying hen meat. Furthermore, Gianelli et al.^[9] showed that an important percentage of volatiles in charqui came from the lipid oxidation.

A remarkable rise in acids, aldehydes, esters, hydrocarbons, and ketones were noted after desalting procedure. As previously observed, raw samples with high salt content had low lipid oxidation values (measured by TBARs content) compared with desalted ones. In fact, this behavior is now confirmed by the analysis of lipid-derived volatiles. The formation of aldehyde compounds after desalting was more intense than in those generated from the other treatments. The oxidation/degradation of polyunsaturated fatty acids leads to the formation of saturated and unsaturated aldehydes, which present low odor thresholds and are major contributors to the volatile profile of cooked meats. According to Shahidi and Pegg,^[6] the flavor note of hexanal is described as intense “grass-like,” while octanal and nonanal are associated to pleasant notes described as “floral” and “sweet.” In addition, the reaction of these aldehydes with products of Maillard reaction provides the generation other volatile compounds.^[3]

The cooking process had a significant impact on most of chemical families, except for esters and hydrocarbons, indicating that volatile profiles of salted and dried broiler meat are severely affected by the cooking method. Sous-vide cooked samples showed the highest volatile content, which suggests that a prolonged cooking time increases the generation and release of volatile compounds. High amounts of aldehyde compounds, representing 45.9% of the total volatile compounds in cooked samples, were found in grilled (74.6 AU × 10⁶), roasted (84.2 AU × 10⁶), fried (72.7 AU × 10⁶), and sous-vide cooked (101.5 AU × 10⁶) samples. Within aldehydes, hexanal, heptanal, octanal, nonanal, were the most abundant. Strecker aldehydes such as 2-methyl-butanal and 3-methyl-butanal were also identified and showed in high contents in grilled and fried samples. In particular, 3-methyl-butanal has been described to be generated from the Strecker degradation of leucine in the presence of dicarbonyls^[20]. During cooking, meat is submitted to high temperatures, which accelerates oxidative degradation of unsaturated fatty acids such as oleic, linoleic, and arachidonic acids, which leads to the formation of hexanal, heptanal, octanal, and nonanal^[22]. In agreement with the present results, Chen et al.^[23] reported that about 90% of volatile compounds in cooked meat arise from lipid oxidative reactions. Hexanal may be generated both from fatty acids oxidation and deterioration of other unsaturated aldehydes such as 2,4-decadienal, which may explain the predominance of hexanal over the other compounds in dried salted cooked samples^[24]. Grilled and fried samples had the lowest hexanal content compared with roasted and sous-vide cooked ones, indicating that lipid oxidation was less severe in grilled and fried samples compared with those submitted to the other cooking procedures. Consistently, Broncano et al.^[25] reported low hexanal levels in grilled pork meat compared with other cooking methods. In fact, grilled and fried samples presented higher percentage of Maillard-derived volatiles compared with other treatments.

In sous-vide cooked samples, aldehydes were the major chemical family (45.8%), followed by alcohols (11.9%), ketones (10.4%), hydrocarbons (5.2%), acids (4.1%), furans (3.4%), esters (0.6%), terpenes (0.6%), and sulfur compounds (0.6%), while nitrogen compounds were not detected in this treatment. This outcome may be attributed to the low content of Maillard product samples submitted to sous-vide cooking (data not shown), since nitrogen compounds have been postulated to be synthesized via Maillard reaction^[26]. Del Pulgar et al.^[27] reported that meat cooked at milder time/temperature combinations such as those used in sous-vide cooking leads to higher levels of volatile compounds generated via fatty acid degradation. Furthermore, the longer heating time applied in sous-vide cooking technique may explain the significant increases in aldehydes and ketones of dried salted broiler samples, as previously reported by Turner and Larick^[28] in sous-vide cooked chicken breast.

Nitrogen and sulfur compounds were most pronounced in grilled and fried samples. Sulfur compounds are derived from water-soluble precursors and may be synthesized via lipid hydroperoxides or through degradation of cysteine and methionine. These volatile compounds present low odor thresholds and have been reported to play an important role in cooked meat aroma^[3]. Fried samples showed relatively high terpene content (limonene) compared with other cooking treatments, which may be derived from the olive oil used in frying process. Alcohols (1-pentanol, 1-octen-3-ol, 1-hexanol, and 1-octanol) were significantly affected by cooking and most of them presented low content compared with the other volatile compounds. The presence of short-chain alcohols indicates a microbial spoilage index,^[29] suggesting that cooking methods applied in dried salted broiler samples were effective with respect to preservation of the product.

In our study, both temperature and cooking time promoted a significant alteration in development of volatile compounds. As expected, aroma of salted and dried broiler meat was primarily derived from both lipid degradation/oxidation and Maillard reaction derivatives, with aldehydes being the most abundant. These results are in agreement with those reported by Silva et al.^[7] and Gianelli et al.^[9] in chicken charqui (dried salted broiler and laying hen meat) and horse charqui, respectively.

Effect of desalting and cooking methods on odor-liking

Figure 2 shows the results for odor-liking assessment of salted and dried broiler meat submitted to desalting and different cooking methods. Overall, roasted and fried samples presented the highest odor-liking rating compared with other samples. No differences were observed between fried, grilled, and raw treatments. Desalted and sous-vide cooked samples showed the lowest odor-liking.

Figure 2. Odor-liking (mean ± standard deviation) in chicken charqui: raw, desalted, and subjected to assorted cooking methods.

Different letters on top of bars denote significant differences between treatments in Kruskal–Wallis analysis (p < 0.05).

Regarding to uncooked samples, there were a remarkable decrease (29.1%) in odor-liking scores after desalting procedure. This result confirms the previous hypothesis observed in TBAR values and volatile compounds that high salt content may inhibit lipid oxidation, while desalting procedure promotes rancid aroma. Furthermore, the lower scores attributed to desalted samples compared with raw ones may be associated to a more balanced volatile profile in raw samples than desalting ones. In fact, aldehyde content was more pronounced in desalted samples compared with raw treatment (Table 1).

The consumers’ liking and acceptability of meat products are strongly affected by aroma profile of cooked meat. Concerning to cooked samples, no significant differences were noted between roasted and fried treatments, which presented the highest scores of odor-liking (6.4 and 5.9, respectively). In fact, roasted cooking promoted a remarkable increase (63.2%) in odor-liking scores compared with desalted samples. Roasted-like odor has been described to be generated from Maillard and caramelization reactions^[30] that occur at high cooking temperatures (≈180–200°C). Interestingly, in our experiment, roasted and fried sampled showed the lowest results for TBARs and relatively low content of total aldehydes, which may explain the high odor acceptability by the sensory panel. Sous-vide technique promoted a marked decrease in odor-liking rating, indicating that, as previously observed in desalted samples, the high content in aldehyde compounds and the unbalanced volatile profile promoted an undesirable aroma in samples submitted to the sous-vide technique. In addition, most of volatile compounds in sous-vide cooked samples were derived from lipid degradation/oxidation reactions, since the low temperature applied in this cooking technique reduces the formation of Maillard products on the surface of the meat^[31]. It is worth noting that salted and dried broiler meat submitted to cooking methods that promote high Maillard products formation (i.e., grilling, roasting, and frying) presents the highest odor-liking by consumers. Compounds formed during the Maillard reaction provide savory, meaty, roasted, and boiled flavors. Furthermore, Maillard-derived compounds may interact with lipid oxidation derivatives such as aldehydes and other carbonyls, controlling the formation of sulfur compounds and other volatile compounds, which optimize cooked meat flavor characteristics^[3].

Principal component analysis

Data from chemical (volatile compounds and TBARs) and sensory analyses of salted and dried broiler meat submitted to desalting and different cooking techniques were used to perform a PCA in order to determine the relationship between chemical and sensory parameters and to discriminate between samples from different treatments. The similarity map of the measured parameters is shown in Fig. 3A, and the principal components (PC#1 and PC#2) accounted for the 81.8% of the total variability. Most lipid-derived volatiles were positioned on the positive axis of PC#1, far from the origin, whereas Maillard-derived volatiles were located in an opposite position. Interestingly, TBARs was also located far from the origin, but on the left upper quadrant, presenting a negative relationship with odor-liking by consumers. It is worth noting that odor-liking was not closely related to any particular group of volatiles indicating that consumer perception of a pleasant aroma may be driven by a balanced combination of both lipid-derived volatiles and Maillard products. In fact, according to Mottram,^[3] cooked meat flavor results from many different interactions between lipid- and water-soluble precursors. In addition, these interactions appear to improve the cooked meat flavor^[3].

Figure 3. Cooking methods applied to chicken charqui: projection of chemical and sensory variables (A) and the samples (B) onto the space defined by the principal components (PC#1/PC#2).

Analyzing the projection of the samples onto the space of the two principal components (Fig. 3B), it was possible observe a clear discrimination between roasted and sous-vide cooked samples. On the other hand, as expected, grilled and fried cooked samples were projected on the same space as the Maillard-derived volatiles formation, suggesting that thermal treatment applied in these cooking techniques (grilling and frying) promoted the formation of similar characteristics in terms of volatile profile of chicken charqui samples. Conversely, Maillard-derived volatiles have been considered as the major component in well-done cooked meat, since that more intense heating treatments favored the formation of Strecker degradation compounds as suggested by Roldán et al.^[32] in sous-vide cooked lamb loins. According to the results obtained in the PCA, it is possible to confirm that the formation of most of volatile compounds derived from lipid oxidation reactions was more intense in sous-vide cooked samples. These samples were projected on the opposite space of odor-liking, suggesting a negative relationship between prolonged cooking time at low temperature and aroma acceptation of salted and dried broiler meat. Roldán et al.^[32] reported higher proportion of volatile compounds from lipid oxidation in sous-vide cooked lamb loins submitted to mild heating conditions similar to those applied in our experiment. Furthermore, roasted cooked samples were projected relatively close to odor-liking indicating that this cooking method may comprise a reasonable good balance of volatiles leading to a pleasant odor profile.

Conclusion

Desalting and different cooking methods have an impact on lipid oxidation rate and volatile compounds formation in salted and dried broiler meat (chicken charqui). Moreover, this difference in aroma profile directly affects consumers’ liking. The salted and dried broiler meat submitted to sous-vide cooking shows a remarkable degradation of lipids and impairment of the aromatic profile, which is negatively related with aroma acceptation by consumers. Grilled, roasted, and fried cooking treatments present low lipid oxidation rate and led to a high content of Maillard-derived volatiles in samples. The high percentage of specific volatile compounds from lipid oxidation seems to be strongly related to odor rejection by consumers of this charqui-like product. On the other hand, Maillard-derived volatiles generated in thermal treatments that applies high temperature (i.e., grilling, roasting, and frying) appears to contribute positively to the high consumers’ liking of dried salted broiler meat. However, a balanced volatile profile and a moderate oxidation level play an important role in the aroma acceptability of salted and dried broiler meat.

Funding

The authors are thankful to the CNPq and CAPES for providing support through projects 474300/2011-0 (Universal), 401167/2014-3 (PVE), and 8851/14-0 (PDSE).

Additional information

Funding

The authors are thankful to the CNPq and CAPES for providing support through projects 474300/2011-0 (Universal), 401167/2014-3 (PVE), and 8851/14-0 (PDSE).