Taste compounds, affecting factors, and methods used to evaluate chicken soup: A review

Abstract The taste of chicken soup is dependent upon various taste substances and human senses. More than 300 nonvolatile compounds reportedly exist in chicken/chicken soup. The primary purpose of this review was to elaborate on the prominent taste substances, the taste evaluation methods, and the factors affecting the taste of chicken soup. Most taste‐active compounds with taste descriptions and thresholds in chicken soup were summarized. The application of sensory evaluation, liquid chromatography, electronic tongue, and other evaluation methods in chicken soup taste analysis were elaborated. The effects of genetic constitution, preslaughter, processing, and storage on chicken soup taste had been discussed. Nucleotides (especially inosine 5′‐monophosphate), amino acids and their derivatives, organic acids, sugars, and peptides play a vital role in the taste attributes of chicken soup. Combining of liquid chromatography and mass spectrometry enables qualitative and quantitative analysis of taste‐active compounds in chicken soup, aiding the exploration of key taste‐active compounds. The electronic tongue application helps the overall taste perception of the soluble taste‐active compounds present in chicken soup samples. Postmortem aging and stewing for a prolonged duration are effective techniques for improving the taste quality of chicken soup. The washing of preprocessing, the cooking temperature of processing, and the storage conditions also exert a significant impact on the taste of chicken soup.

characteristics of low price, convenient cooking, and timely supply of processed products (Fu et al., 2006). Available data indicate that the proportion of chicken consumption in China has shown a steady growth trend in recent years (https://downl oads.usda.libra ry.corne ll.edu/usda-esmis/ files/ 73666 448x/t435h 4995/vt151 9922/lives tock_poult ry.pdf [the United States Department of Agriculture -USDA]). The reason is not only the substitution of poultry for pork under the influence of African swine fever, but also that consumers prefer to choose poultry for their health (Woźniak et al., 2016).
Stew is one of the popular ways to cook chicken in China. As a result, chicken soup has become an indispensable part of several meals (Kurobayashi et al., 2008). As the taste of chicken soup remarkably affects its palatability and consumer acceptance, it is one of the main characteristic factors of chicken soup (Drewnowski & Darmon, 2005). Considering consumers' pursuit of delicacy, any subtle changes in chicken soup taste can considerably affect consumer choice and purchase. Therefore, evaluation and research on the taste of chicken soup are of significance for promoting the industrialization of high-quality chicken soup. Over the past half-century, several researchers have engaged efforts for the improvement of characteristic taste-active compounds of chicken soup to facilitate its industrial production. However, the complex interactions between various flavor compounds, coupled with multiple influencing factors during breeding and processing, pose challenges to maintain of consistency in the sensory quality of chicken soup. In this article, the taste-active compounds contributing to the taste of chicken soup, the factors affecting chicken soup taste, and the methods used to evaluate the taste of chicken soup were discussed.

| TA S TE-AC TIVE COMP OUNDS IN CHI CKEN SOUP
Identification of characteristic taste-active compounds in chicken soup is essential for the analysis of chicken soup taste. Thus far, more than 300 small molecular compounds have been reported in chicken/chicken soup (Table 1), of which 91 are taste-active compounds (Jayasena et al., 2013a;Pippen et al., 1960;Wang et al., 2020;Yang et al., 2018;Zhang, Zhao, et al., 2019). Identifying the key flavor components in chicken soup remains difficult when studying the flavor characteristics of chicken soup due to the presence of a large number of taste-active components in the soup. The characteristic taste of chicken soup could be attributed to the taste-active compounds in chicken as well as those generated during stewing. As a complex attribute, the taste of chicken soup is determined by the comprehensive analysis of several different taste-active compounds rather than one or two taste-active compounds.
The taste of chicken soup is attributed to the following: (a) chemical components transferred from the chicken carcass to the soup, including proteins, fats, carbohydrates, and minerals; and (b) the taste compounds produced by the reaction of water-soluble taste precursors during heating. These reactions include the formation of flavor nucleotides, the thermal reaction of carbohydrates, the reaction of TA B L E 1 Small molecular compounds have been reported in chicken/chicken soup Phosphoserine Aliani and Farmer (2005) 38 Oxalic acid Zhang et al. (2020)  flavor peptides to amino acids, the reaction of reducing sugars to amino acids, and the oxidative degradation of lipids. In general, the main contributing components to the taste of chicken soup include sugars, amino acids, nucleotides, organic acids, salts, and peptides (Fukuuchi et al., 2018;Lin et al., 2016;Rikimaru & Takahashi, 2010).

| Sweet-tasting compounds
Sweet taste is one of the basic taste sensations and the most predominant preference among individuals. Natural sweet substances mainly include aliphatic hydroxyl compounds such as alcohols and sugars. However, amino acids, amides, lipids, and other compounds also have a sweet taste. The main sweet-tasting compounds reported in chicken/chicken soup include sugars, sugar alcohols, and sweet amino acids. In the study reported by Dunkel and Hofmann (2009), 4 sugars (glucose, fructose, sucrose, and lactose), 4 sugar alcohols (inositol, xylitol, ethylene glycol, and ribitol), and 7 sweettasting amino acids (alanine, serine, glycine, proline, threonine, cysteine, and methionine) were quantified in double-boiled chicken broth. The taste threshold concentrations (TCs) of the mentioned above 15 sweet-tasting compounds were also determined in bottled water by means of a triangle test. Among them, cysteine had the lowest taste threshold (TC = 2,000 μM), followed by methionine (TC = 5,000 μM) and alanine (TC = 8,000 μM). Since the side chain of methionine contains a hydrophobic methylthio group (CH 3 S-), most researchers preferred to classify it as a bitter amino acid. In addition to inositol, glucose, and fructose, 4 free sugars (sedoheptulose, mannose (TC = 199,822 μM), ribose, and ribulose) were also identified in Single-comb White Leghorn and New Hampshire chickens (Lilyblade & Peterson, 1962). The content of maltose (TC = 31,259 μM) was detected in the breast and thigh of Sanhuang chicken and Blackbone silky fowl; however, the maltose content was found to be <0.1% in four samples . In the study reported by Aliani and Farmer, and glucose-6-phosphate [G-6-P]) sugars were quantified, which showed that their content in breast meat was higher than that in leg meat (Aliani & Farmer, 2002. Recently, two sweet-tasting amino acid derivatives (sarcosine and β-alanine) were studied in chicken (Hy-line brown) soup ; however, the contribution of these two compounds to the sweetness of chicken soup is unclear due to the lack of data on their taste threshold. Alkaloid is a type of basic organic compound that contains nitrogen. Almost all alkaloids have a bitter taste, except betaine that has a sweet taste. It was found that the content of betaine (CT = 50,021 μM) in freezedried chicken soup prepared with Korean native chicken (KNC) was significantly higher than that prepared with commercial broiler (CB) (Jayasena et al., 2014).

| Salty compounds
Saltiness is the characteristic taste of neutral salt, which plays an essential role in cooking. As cations are easily adsorbed by the carboxyl group or phosphoric acid group of the taste receptor proteins, the salty taste perceived by the tongue is mainly attributed to metal cations, such as sodium ions and potassium ions. The metal cations in chicken/chicken soup include Na + , K + , NH + 4 , Ca 2+ , and Mg 2+ (Choi, 2011). As the taste of divalent salts is complex and mainly characterized by bitterness and saltiness, the taste TC of CaCl 2 and MgCl 2 was found to be bitter in the study reported by Dunkel & Hofmann (2009;Lawless et al., 2003). The contribution of anions to saltiness has long been proven (Elliott & Simon, 1990;Roebber et al., 2019).
The study conducted by Lawless et al. (2003) have revealed that salts with larger anions are not effective stimulants because of their limited expansion at the tight junction and at the basolateral area F I G U R E 1 Taste-active compounds and their taste threshold in chicken/chicken soup of taste receptor channels. The anions reported in chicken/chicken soup mainly included Cl − , OH − , HPO 2 − 4 , and PO 3 − 4 (Nishimura et al., 1988(Nishimura et al., , 2016Qi et al., 2017).

| Bitter-tasting compounds
The bitter taste is easily perceived and lasts for a longer duration in the mouth. The taste threshold of caffeine was only 750 μM (Stark et al., 2006). Bitter taste exerts a double effect on the quality of chicken soup. An ideal extent of bitter taste in chicken soup can help enrich and improve the taste of chicken soup. However, a considerable extent of bitter taste will affect the taste of chicken soup and result in consumer dissatisfaction. The bitter-tasting compounds in chicken/chicken soup mainly include bitter amino acids, nucleotides, and bitter peptides. Strong hydrophobic sites reportedly exist in the molecular structure of several bitter organic compounds. The intensity of the hydrophobic interaction between hydrophobic sites and taste cell membrane may help determine the extent of bitterness (Katsuragi et al., 1996). In Dunkel and Hofmann′s study,16 bitter-tasting amino acids and nucleosides (arginine, histidine, isoleucine, leucine, lysine, phenylalanine, tyrosine, valine, tryptophan, taurine, xanthine, hypoxanthine, inosine, guanosine, adenosine, and 3′:5′-cAMP) were quantified in double-boiled chicken broth.

| Sensory evaluation
Sensory evaluation using the human tongue as a detector can provide direct, effective, unique, and intuitive taste information for chicken soup. Moreover, it provides a human feeling and is considered to be the most direct and intuitive tool for such evaluation.
The detection threshold of the human tongue (45 years of age and below) for sodium chloride, quinine sulfate, sucrose, and citric acid was reported to be 2.49 × 10 -3 M, 1.24 × 10 -6 M, 5.92 × 10 -3 M, and 1.04 × 10 -4 M, respectively (Weiffenbach et al., 1982). Thereby sensory evaluation becomes a valuable and sensitive method in analyzing chicken soup taste and taste-active compounds. The sensory evaluation of chicken soup is usually conducted by directly tasting the warm chicken soup (45-60°C). A sensory panel often includes 6-19 well-trained panelists. Linear scale , category scale (Kurobayashi et al., 2008), and quantity scale (Kong et al., 2017) are widely used to evaluate the taste attribute strength of chicken soup.
Sensory evaluation is mainly performed to distinguish between the taste characteristics of different chicken soup or to score the taste attributes of chicken soup. The five sensory attributes (chicken meat-like, fatty, off-flavor, kokumi, and umami) of chicken soup prepared using traditional clay stew-pot and commercial ceramic electrical stew-pot could be distinguished based on the taste sensed . The overall acceptance of chicken soup subjected to refrigerated (4 ± 1°C) and superchilled (−2.5 ± 1°C) storage conditions could be identified by the human tongue .
Umami and saltiness of chicken meat after stewing for 1 and 2 hr could be distinguished based on the taste sensed (Qi et al., 2018). The umami, saltiness, and sweetness of chicken broth containing a nonvolatile fraction of celery (7%) and chicken broth (control) could also be identified by the human tongue (Kurobayashi et al., 2008). The sensory evaluation of chicken soup taste analysis is fast; however, it is inevitably affected by the physiological, physical, psychophysical, and other factors of panelists. The sensory abilities of human tongue markedly vary among different individuals. Additionally, sensory evaluation is nonquantitative and markedly affected by individual subjectivity. Sensory fatigue, a source of error, also restricts the efficiency of intensive evaluation. Therefore, effective technical training of panelists, uniformity of the description, and consideration of reference standards for chicken soup taste are of significance in sensory evaluation.
The quantitative descriptive analysis (QDA) method was performed to analyze the taste characteristics of chicken soup. In the QDA method, panelists were requested to describe the perceived sensory attributes and the intensities of the attributes. Many descriptive words have been reported, including taste, flavor, and mouthfeel (You et al., 2019;Zhang et al., 2018) (Table 2), rendering the descriptive analysis more accurate and consistent. The descriptive words covered both positive taste attributes (sweet, salty, umami, chicken meat-like, fatty, meaty, and kokumi) and negative ones (bitter, sour, greasy, rubbery, and warm-over flavor) (Hooge & Chambers, 2010;Miyaki et al., 2015;Zhan et al., 2020;Zhang et al., 2018).

| Electronic tongue
Electronic tongue, a tool used to evaluate food taste by simulating human taste, is based on sensor array, signal acquisition, and pattern recognition systems (Makkliang et al., 2015). The electronic tongue uses a material similar to that observed in biological systems as a sensitive membrane of the sensor. When one side of the lipid film establishes contact with the taste substance, the membrane potential changes, resulting in a response and the detection of the relationship between various substances (Makkliang et al., 2015;Xiao, Tahara, et al., 2019). Electronic tongue, a simple and rapid tool, provides global taste perception for soluble taste compounds in chicken soup samples. It is mainly used to analyze the difference among chicken soup samples because it can rapidly distinguish between and quantify the different taste senses in the sample. Furthermore, it can be used to detect subtle changes in chicken soup samples. Electronic tongue combined with PCA data processing was used to investigate the effects of stewing time (1, 2, and 3 hr) on traditional Chinese chicken soup . The electronic tongue could successfully distinguish between three types of chicken soup samples with high accuracy . The extract of Dezhou braised chicken in different processing stages was analyzed, and the radar fingerprint was formed according to the output data of 8 taste sensors. The electronic tongue could thus be applied to explore the evolution of different taste components in Dezhou stewed chicken . The electronic tongue could also be used to estimate the degree of adulteration (Tian et al., 2019). When the chicken was mixed with mutton in proportions of 0%, 20%, 40%, 60%, 80%, and 100%, the meat extract was detected using the electronic tongue (Tian et al., 2019). Results showed that the electronic tongue with least squares support vector machines (LS-SVM) was the most effective method for predicting chicken content (Tian et al., 2019). Of note, all data obtained by using the electronic tongue are analyzed as a whole, and detailed taste-active compound information is not required. Therefore, the use of the electronic tongue has advantages in identification and classification but exhibits limitations in quantitative analysis. Furthermore, the selectivity and limitations of the sensors prevent their widespread application in food evaluation.

| Liquid chromatography method
The liquid chromatography (LC) technology is the most common and effective method for conducting chicken soup taste analysis. In fact, it is widely used for performing qualitative and quantitative analysis of taste-active compounds in chicken soup. The complex composition of chicken soup, with insufficient concentration of certain taste compounds, requires higher sample pretreatment requirements.
High-speed centrifugation, filtration, ultrafiltration, selective precipitation, extraction, derivatization, concentration (rotoevaporation, freeze-drying, and so on), and other technologies are widely used in the preprocessing stage of samples (Kong et al., 2017;Qi et al., 2018;Wang et al., 2018;Zhang et al., 2020). In the determination of water-soluble taste substances in chicken soup, degreasing is an inevitable operation step used for sample pretreatment.
Additionally, the removal of macromolecular proteins is necessary to protect the chromatographic column. Perchloric acid and methanol are commonly used for protein precipitation in chicken soup samples . For the identification of unknown taste compounds (such as peptides), most are subjected to ultrafiltration separation, gel filtration separation, and high-performance liquid TA B L E 2 Descriptive sensory analysis attributes and references used to evaluate chicken soup

Sensory attributes Definitions References intensity Source
Taste Chicken meat-like -100 g lean minced chicken meat was cooked in water at 100°C for 2 hr One of the basic taste, common to MSG. The taste and mouth-filling sensation of compounds such as glutamates that is savory, brothy, meaty, rich, full, and complex, common to many foods such as soy sauce, stocks, ripened cheese (especially parmesan), shellfish (crab, lobster, scallops, clams), mushrooms (especially porcini), ripe tomatoes, cashews, and asparagus Saltiness A basic taste of which the taste of sodium chloride in water is typical 2.25 g NaCl in 500 ml filtered water (Score: 7.5) 2.75 g NaCl in 500 ml filtered water (Score: 10.0) 3.10 g NaCl in 500 ml filtered water (Score: 12.5)

Hooge and
Chamber (2010) One of the basic taste, common to sodium chloride 0.2% NaCl in water (Score: 2) 0.5% NaCl in water (Score: 5) Taihe and crossbred black-boned silky fowls were analyzed using ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) at a mass scan mode, and the sample "fingerprint" formed by quantified lipids was used for conducting discrimination by statistical analysis (Mi et al., 2018). The orthogonal partial least squares discriminant analysis results showed that Taihe and crossbred black-boned silky fowls could be effectively distinguished from each other (Mi et al., 2018). Based on a study conducted on the effects of quercetin and cinnamaldehyde on the chemical composition of beef soup, 636 molecular features were identified by performing UHPLC-MS/MS (Li et al., 2020).
Compared to databases based on data obtained by gas chromatography, most databases based on liquid chromatography data are nonuniversal. Furthermore, a database for taste compounds has not been established. Thus, the engagement of remarkable efforts is necessary for data processing.

| Nuclear magnetic resonance
Nuclear magnetic resonance (NMR) is a nondestructive and rapid method used for the identification and characterization of small molecules. For the past few years, NMR techniques have been rapidly developed and it has been successfully used for the analysis of chicken quality. The focus of NMR studies includes the estimation of the meat quality variation caused by raw materials (variety Wang et al., 2016), age , and position ) and external conditions (storage (Graham et al., 2010;Li et al., 2014), and processing (Li et al., 2000;Shaarani et al., 2006)). Visual analysis of the correlation between taste precursors and chicken samples was achieved using NMR and showed that there were eight taste compounds in 230-day-old chickens, which were significantly different from the other four age groups. Furthermore, organic acids and small peptides were

Mouthfulness
The perception that the sample fills the whole mouth is blooming, or growing, a full-bodied sensation when the sample is held in the mouth Kitchen Basics chicken broth (Score: 1.5) 0.5% MSG in Kitchen Basics chicken broth (Score: 3) Miyaki et al, (2015) Mouth coating The degree to which there is a leftover residue, a slick, powdery, or fatty coating or film in the mouth that is difficult to clear.
0.5% MSG in water (Score: 4) Half and half (Score: 5) Miyaki et al. (2015) Tongue coating The degree to which there is a leftover residue, a slick, powdery, or fatty coating or film on the tongue that is difficult to clear 0.5% MSG in water (Score: 3) Miyaki et al. (2015) Total trigeminal The intensity of the total sensation, including numbing, burning, tingling, or irritation, impaired on the soft tissues of the oral cavity, particularly the tongue Wintergreen breathsaver (not scored) 0.5% MSG in water (Score: 5) Miyaki et al. (2015) Salivating The degree to which the sample caused a perceived increase in salivation No reference Miyaki et al. (2015) Swelling of cheeks and lips The feeling of swelling of the soft tissue in the oral cavity, specifically the cheeks and lips, reminiscent of the perception of swelling produced by antithetic treatments at a dental office, but without a distinct numbing effect 0.5% MSG in water (Score: 4) Miyaki et al. (2015) TA B L E 2 (Continued) identified as the main taste precursors of chicken breast and thigh meat . The 1 H-NMR results showed that boiling treatment had a significant effect on the distribution of watersoluble low-molecular-weight compounds in Wuding chicken (Xiao, Luo, et al., 2019). NMR, mainly used to analyze polar metabolites, permits the detection of a broad scope of analytes within a single run. However, NMR typically exhibits disadvantages such as lower sensitivity compared to GC-MS and LC-MS (Lei et al., 2011;Ruiz-Aracama et al., 2012).

| Gas chromatography-mass spectrometry
Gas chromatography-mass spectrometry (GC-MS) is typically used to analyze (semi-)volatile compounds and active aroma ingredients in chicken soup (Feng et al., 2018), due to its robustness, reproducibility, and selectivity, and owing to the availability of a considerable number of commercial and "in-house" databases. In certain studies, the derivatization of polar compounds has expanded the scope of this approach, thus highlighting it as one of the most effective, repeatable, and widely used analytical platforms in taste-active compound research (Beale et al., 2018).
In the food industry, the GC-MS combined with derivatization has been used for the quantitative analysis of taste-active compounds (amino acids and their derivatives, organic acids, sugars and sugar alcohols, etc.) in the fermentation of soybean paste (Sun et al., 2019), rice koji (Lee et al., 2016), and so forth. However, this method is rarely performed to assess the taste-active compounds in chicken soup. This may be due to the complexity of derivatization. During the addition of appropriate derivatization reagents, it is important to ensure the transformation of all target compounds to their derivatives to confirm that the corresponding peaks do not dominate the total ion chromatography peaks or mask other peaks (Beale et al., 2018).

| Others
In addition to the above-mentioned techniques, the ambient desorption ionization techniques have been used to assess chicken quality. This technique is a modern approach for metabonomics fingerprint and profile analysis, which can be used for di-

| Genetic factors
The taste of chicken is a markedly inherited trait. The genes controlling the taste trait have been studied. Based on findings reported by previous studies, the taste trait deemed is relatively complex and genetically controlled. There are different genes involved in the synthesis of taste compounds in chicken. 5′-IMP, as an essential index for meat flavor determination (Fujimura, 1998;Kawai et al., 2002), has been systematically studied in the literature. The GPAT, AIRC, PurH, GARS-AIRS-GART, ADSL, and AMPD1 genes were identified as the most common candidate genes for 5′-IMP content in chicken.
GPAT and AIRC encode two enzymes that catalyze step 1 and steps 6 plus 7, respectively, of the de novo purine biosynthetic pathway (Zhang et al., 2009). The chicken GPAT/AIRC genes are located on chromosome 6. The PurH gene is responsible for encoding the attic enzyme, a 64-kDa bifunctional enzyme that catalyzes the final two reactions in de novo purine biosynthesis and possesses two enzymes, namely AICAR transformylase and 5′-IMP cyclohydrolase (Asby et al., 2015). The crystal structure of chicken ATIC showed the absence of an intermediate connecting channel between the Nextremity and C-extremity activity centers, which plays an important regulatory role in the synthesis of 5′-IMP (Greasley et al., 2001). Shu et al. (2010) showed that the GPAT/AIRC and purH genes affected muscle 5′-IMP content. As a result, they might be candidate loci or linked to major genes that affect muscle 5′-IMP content; the epistatic effects were found to be higher than the single genotype effects in Chinese Baier chicken. Three SNPs in GPAT gene (exon 2), AIRC gene (exon 3 and 8), purH gene (exon 16), and GPAT/AIRC (promoter region) were associated with muscle 5′-IMP content in chickens (Shu et al., 2008(Shu et al., , 2010. The GARS-AIRS-GART genes are located on chromosome 1, which catalyze steps 2, 3, and 5, respectively, of the de novo purine biosynthetic pathway. Chicken GARS-AIRS-GART gene has been studied as a candidate gene for determination of the effects on the 5′-IMP content in the muscle (Shu et al., 2007;Ye et al., 2010). Furthermore, a favorable association between genotypes and higher 5′-IMP content has been demonstrated in several Chinese native chicken breeds.
The ADSL gene plays an important role in the biological pathways of purine nucleotide de novo synthesis, and its mutation can reduce the ability of 5′-IMP synthesis in vivo (Lundy et al., 2010).
Studies have revealed no significant correlation between the content of exogenous 5′-IMP and the expression of ADSL. This may be attributable to the purine biosynthetic pathway that is ultimately responsible for the generation of 5′-IMP from α-D-ribose-5-phosphate .
The AMPD1 gene is primarily expressed in muscle tissues and is involved in the metabolism of 5′-IMP. Hu et al. revealed that animals with the homozygous genotype AA at positions 4,064 and 6,805 presented with significantly higher 5′-IMP contents than those with the GG genotype (p < .05). Further, they found that the homozygous genotype AA at position 6,805 resulted in a significantly higher 5′-IMP content than the genotype GG for both cock and hen (Hu et al., 2015). Chen et al. (2008) estimated the genetic parameters of 1,069 purebred Beijing-You full-sib male chickens and found that the heritability of muscle 5′-IMP content was moderate (0.23) . Therefore, 5′-IMP content in chicken meat could be increased through genetic selection. Owing to the adequately high heritability of breast and fat yield characteristics (Le Bihan-Duval et al., 1998), poultry body composition has been substantially improved by selection.

| Preslaughter factors
Many studies have shown that under the same cooking conditions, factors such as chicken breed, age, sex, and diet affect the content of taste substances in chicken and thus affect the quality of chicken soup (Jayasena et al., 2013b;De Zwart et al., 2003). In Korea, KNC is more popular among consumers because of its characteristic flavor and texture compared to CB (Jayasena et al., 2013b). As the production of KNC is insufficient to meet consumer demand (Jeon et al., 2010), traditional dishes such as samgyetang and baeksuk are made with CB instead of KNC. Studies are underway to clarify the effect of chicken breed on the quality of soup, which is important for maintaining the characteristics of chicken soup. The analysis of the taste components in defatted freeze-dried chicken soup (DFDS) prepared with KNC and CB showed that the DFDS prepared with KNC had higher contents of 5′-IMP, betaine, inosine, cysteine, and carnitine (Jayasena et al., 2015). Additionally, the lipid layer isolated from KNC soup showed significantly higher levels of linoleic (C18:2), α-linolenic (C18:3), arachidonic (C20:4), and docosahexaenoic acid (C22:6, DHA) and low saturated fatty acid Jeon et al., 2010). In general, the DFDS prepared using KNC was superior to the DFDS prepared using CB in terms of nutrition and sensory quality (Jung et al., 2011).
The sex of KNC also significantly influences the taste characteristics of freeze-dried broth (FDB) (Jayasena et al., 2014). A study conducted on the taste compounds of FDB samples prepared using 100-day-old male and female KNC showed that the 5′-IMP and arachidonic acid (C20:4) contents in samples prepared using female FDB were significantly higher than those prepared using male FDB.
The content of taste-active compounds in DFDS was significantly affected by the age of KNC (Jayasena et al., 2015). The study of taste-active compounds in DFDS prepared with KNC of 5 ages (10,11,12,13, and 14 weeks of age) revealed that inosine and linoleic acid contents in DFDS increased with an increase in age. In contrast, the contents of 5′-AMP, oleic acid, and hypoxanthine decreased. The 5′-IMP content of DFDS fluctuated significantly with the increase in KNC age. Studies conducted by Jayasena et al. revealed that age exerted a significant effect on the content of oleic acid, arachidonic acid, and DHA in KNC meat. The content of free amino acids responsible for the umami taste (glutamic acid and aspartic acid), sweet taste (alanine, serine, and glycine), and bitter taste (valine, isoleucine, leucine, phenylalanine, methionine, arginine, and histidine) decreased significantly as the age of KNC increased. However, the age of KNC exerted a positive effect (p < .05) on lysine content that is also responsible for the sweet taste. Therefore, the effect of age on the content of 5′-IMP and glutamate in meat depends on the state of slaughter and cooking status (Jayasena et al., 2015). In cooked KNC leg meat, a positive correlation between bird age and reducing sugar content (p < .05) was observed. However, in KNC breast, there was a negative correlation between age and DHA content, especially after 13 weeks.
Studies have shown that dietary nutrients play a significant part in determining taste compounds of chicken meat (Fujimura and Kadowaki, 2006). Fujimura and Kadowaki (2006) reported that free glutamic acid and sensory score in chicken meat were increased in high crude protein diet. The free glutamic acid content in chicken muscle was negatively correlated with the level of leucine in the diet (Fujimura and Kadowaki, 2006). Laksesvela (1960) reported that the taste of chicken meat was significantly improved when the chicken were provided with 36.7 mg/kg of dα-tocopheryl acetate. In addition, reducing dietary lysine content was proved to increase the contents of free glutamic acid, glycine, valine, isoleucine, leucine, histidine, and threonine content in chicken meat and significantly improved the umami and kokumi tastes of chicken meat (Watanabe et al., 2016).
Other preslaughter factors such as heat stress and preslaughter shackling also contribute to the taste quality of chicken. Ali et al. (2008) reported that chickens exposed to heat stress before slaughter showed the lowest limit pH, which had a negative impact on meat quality. The lactate concentration in chicken breast muscle was found to increase with struggling activity (Papinaho et al., 1995).

| Preprocessing
After subjection to slaughter, the circulatory system and oxygen ssupply of chicken tissues are terminated. As a result, metabolism conditions in the muscle tissues change from aerobic metabolism to anaerobic fermentation, during which lactic acid is produced and accumulated. The lactic acid content in chicken was found to reach the maximum value (6.75 mg/g muscle) within 1 day after slaughter and reduce gradually to 5.47 mg/g muscle at 3 days after slaughter (Nishimura et al., 1988). The lactic acid content in the heated chicken soup was higher under the condition of additional storage than that without additional storage. Considering the sourness and water solubility of lactic acid, it might contribute to the taste quality of chicken soup. Gault (1985) reported that lower pH value has a negative influence on the water-holding capacity and tenderness of meat.
In addition, the increase of acid condition in chicken muscle leads to denaturation of muscle protein. With the occurrence of meat stiffness, denatured proteins (such as sarcoplasma, myofibril, and myoglobin) are easily affected by proteases, and the N-terminal of protein is separated one by one, forming a variety of low molecular peptide and amino acid compounds. Studies have shown that the levels of oligopeptides and all free amino acid increased with the degradation of protein during storage, which were positively correlated with the taste of chicken soup (Nishimura et al., 1988). Meanwhile, the ATP in the chicken muscle after slaughter could be degraded by the pathway ATP→ADP→AMP→IMP→HxR→Hx by the corresponding intracellular enzymes (Nakatani et al., 1986), and its metabolite 5′-IMP is a kind of compound which has important contribution to the umami taste of chicken. Studies have shown that the 5′-IMP content of chicken breast muscle reached a maximum at 8 hr after slaughter and then decreased gradually (Nishimura et al., 1988). Although the level of 5′-IMP in the heated soup of meat after additional storage was less, the importance of 5′-IMP in meat taste should be studied in detail, considering its synergistic effect with glutamic acid.
Immersion in hot water is one of the strategies for reducing the abundance of pathogenic bacteria on the surface of poultry meat.
However, subjection to washing and heat treatment may result in a loss of fat content (Pereira et al., 1976). Piette et al. indicated that high temperature (80°C) could effectively be used to extract fat from chicken skin (Piette et al., 2001). According to the methods adopted by the manufacturers of ready-to-eat ginseng chicken soup, the raw chicken is cooked before retorting to reduce the fat content and to improve the quality characteristics of their products (Triyannanto & Lee, 2015). In the study conducted by Triyannanto and Lee (2015), the fat content of ginseng chicken soup obtained by adding a washing step to the standard methodology was the lowest in three different treatments (addition of emulsifier; addition of emulsifier/precooking; or addition of emulsifier/precooking/ washing). Fat in conventional food products is a source and carrier of flavor as well as a flavor enhancer. Several unsaturated fatty acids in fat have been proven to intensify the flavor of chicken (Jayasena et al., 2015). Among them, hexanal and 2,4-decadienal, the most abundant aldehydes in chicken, are derived from the oxidative decomposition of linoleic acid (Shi & Ho, 1994). Arachidonic acid contributes to the umami taste of chicken Kiyohara et al., 2011). Docosahexaenoic acid increases the sweetness and umami characteristics while inhibiting the sour and bitter tastes (Koriyama et al., 2002). Oleic acid reportedly increases taste panel scores for meat flavor (Wood & Enser, 1997). Altogether, the loss of fat directly leads to the reduction of flavor-related fatty acids, which affects the flavor characteristics of chicken/chicken soup.
Postmortem aging is a necessary process for muscle to meat transformation, which is conducive to improving meat quality (Nishimura et al., 1988). Postmortem aging of chicken is another crucial factor affecting the taste of chicken soup. Nishimura et al. (1988) reported that chicken flavor was more pleasant at 8 hr than immediately after slaughter. In the process of postmortem aging of chicken, the glutamic acid level increases rapidly owing to the high activity of aminopeptidase and its hydrolytic activity toward glutamic acidβ-naphthylamide (Glu-NA) (Nishimura et al., 1988).
Considerable increases in alanine, serine, glutamic acid, and leucine could be obtained by postmortem aging of chicken breast muscle for 6 days at 4°C. Studies also revealed a marked increase in 5′-IMP and 5′-GMP content during postmortem aging of chiller, which results in a significant increase in the intensity of the savory, brothy taste of chicken (Nishimura et al., 1988;Tikk et al., 2006).

| Cooking
Cooking is a key factor affecting the flavor of chicken soup (Al- Khalifa & Dawood, 1993). During stewing, water-soluble taste substances produced in chicken and during heat treatment migrate to water, thereby resulting in a delicious taste. The ratio of meat to water, heating temperature and time, cooking method, and seasoning are important parameters affecting the quality of chicken soup.
Studies have shown that higher meat to water ratio aided the extraction of the taste compounds from chicken. Furthermore, the contents of 5′-IMP, 5′-GMP, and lactic acid were found to increase . Prolongation of the stewing time also aided the extraction of the taste substances from chicken. Chen et al. (2007)  Seasonings are often used to increase the acceptability of chicken soup. As early as 1992, monosodium glutamate (MSG) and sodium chloride (NaCl) concentrations had been proved to have a significant effect on the hedonic score of chicken soup (p < .05) (Chi & Chen, 1992). Chi and Chen (1992) reported that spiced chicken soup (adding 0.125% granulated onion, 0.0125% garlic powder, 0.0125% ground white pepper, and 0.0125% whole celery seed) had a higher maximum hedonic score (7.81, near to like very much) than nonspiced chicken soup (7.28, near to like moderately), which indicated that the spice contributes to the hedonic score of chicken soup (Chi & Chen, 1992). In addition, the addition of scallop and celery components were also contributed to the sweetness and umami taste of chicken soup (Kurobayashi et al., 2008;Yoneda et al., 2005).

| Storage
With the degradation of ATP and protein, several taste compounds are produced during the postmortem storage of chicken, which significantly improves the taste of chicken soup. Furthermore, most free and combined amino acids, ammonia, inosine, and hypoxanthine increase during storage. Studies have shown that soup prepared from the chicken muscle stored for 7 or 8 days at 0°C exhibited the highest taste intensity based on sensory evaluation (Sen & Endo, 1990).
The chicken soup as a system is complicated. In fact, the chemical and biochemical changes caused by microorganisms during the storage process lead to the deterioration of the sensory properties of chicken soup. Studies have shown that chicken whey soup has a shelf life of 6 days when refrigerated (Chidanandaiah et al., 2002). In the study reported by Gadekar et al. (2009), there was a significant

| CON CLUS ION
The taste of chicken soup is a result of the interaction of several taste compounds. Identification of the taste substances in chicken soup has remarkable significance. To date, more than 91 taste-active compounds have been reported in chicken/chicken soup. Among these compounds, 5′-IMP is the component responsible for the umami taste of chicken meat. Apart from 5′-IMP, amino acids and their derivatives, organic acids, and peptides also strongly affect the taste quality of chicken soup. Liquid chromatography-mass spectrometry is the most commonly used method for conducting qualitative and quantitative analysis of the taste components in chicken soup. The combination of human tongue and the innovative electronic tongue system can not only enable the identification of the taste attributes and strength of chicken soup, but also facilitate the quantification of less remarkable differences among samples. Chicken soups prepared with the same breed of chicken have different tastes and qualities due to differences in preslaughter factors, processing, and storage. Washing can significantly reduce the fat content in chicken soup and affect its flavor characteristics. Aging and prolonged duration of stewing are effective methods for improving the taste of chicken soup. The production of off-flavor substances and the loss of characteristic taste substances are the main reasons for the deterioration of the taste quality of chicken soup. Storage time and storage conditions were found to significantly affect the biochemical reaction of chicken soup, thereby affecting its taste during storage.
More than 300 compounds have been reported in chicken/chicken soup. Furthermore, its taste characteristics are attributed to several taste compounds. Different proportions of these characteristic taste substances may lead to different perception (taste intensity, attributes, and so on) results. In general, the relationship between tasteactive compounds and the taste of chicken soup has not been well established. Therefore, characteristic taste-active compounds and their relationship with the sensory attribute must be investigated in future studies.

ACK N OWLED G M ENT
This work was supported by the National Key R&D Program of China (No. 2016YFD0400705).

CO N FLI C T S O F I NTE R E S T
The authors declare that there are no conflicts of interest.