Comparison of Flavor Substances in Dried Shrimp Products Processed by Litopenaeus Vannamei from Two Aquaculture Patterns

HU Mengyue , XUE Yong ZHAO Ling, LIU Qi, and CAO Rong,

1)College of Food Science and Technology, Ocean University of China, Qingdao 266237, China

2)Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Qingdao 266071, China

3)Laboratory for Marine Drugs and Bioproducts of Qingdao National Laboratory for Marine Science and Technology, Qingdao 266237, China

Abstract Dried shrimp is a popular aquatic food worldwide due to its unique flavor. However, its flavor qualities on the market remain uneven. This study aimed to compare flavor substances in dried shrimp processed by Litopenaeus vannamei from two aquaculture patterns (earthen pond and greenhouse)using amino acid automatic analyzer, high-performance liquid chromatography(HPLC), gas chromatography (GC)-ion mobility spectrometry (IMS), and GC-mass spectrometry (GC-MS). The results revealed that dried shrimp from earthen pond had significantly higher free amino acids and 5’-nucleotides than those from greenhouse (P < 0.05),and their corresponding equivalent umami concentrations (EUCs)were 23.45 g MSG (100 g)-1 and 14.86 g MSG (100 g)-1, respectively. For volatile compounds, GC-IMS analysis indicated significant differences in volatile compounds between dried shrimp from different aquaculture patterns. According to the quantitative analysis of GC-MS, 11 volatile compounds were identified as aroma-active compounds (AACs), of which six AACs (1-octen-3-ol, 3-methylbutanal, nonanal, trimethylamine, 2,5-dimethylpyrazine, and 2,3,5-trimethylpyrazine)were found in all dried shrimp samples. Although dried shrimp from greenhouse possesses higher OAVs, OAVs of(E,E)-2,4-nonadienal (fishy smell), and 3,5-diethyl-2-methylpyrazine (burnt smell)reached 171.2 and 333 respectively, imparting negative effects on flavor. In general, dried shrimp from earthen pond had a better flavor profile than those from greenhouse.

Key words dried shrimp; earthen pond; greenhouse; free amino acids; flavor substances; 5’-nucleotides; GC-IMS; GC-MS

1 Introduction

Dried shrimp is one of the most widely consumed marine products with a desirable flavor. Dried shrimp production is inexpensive and convenient processing that comprises two steps: boiling and drying, which promote a series of quality changes, including nutrition, texture,color, and flavor. It has previously been observed that boiling resulted in a significant loss of carotenoid and that increasing brine concentration and boiling time could enhance astaxanthin retention in cooked shrimp, whereas most astaxanthin was subsequently degraded during solar drying (Hernández-Becerra et al., 2014). Besides solar drying, several drying methods have been applied in dried shrimp processing, including hot-air-drying (Prachayawarakorn et al., 2002), freeze-drying (Ying et al.,2021), and microwave-drying (Lin et al., 1999). As another essential step, drying profoundly influences the quality of dried shrimp and quality stability during storage. A recent study revealed that using ultrasound as a pretreatment step before drying reduced processing time and improved the quality of dried shrimp (Castañeda-López et al., 2021). Li et al. (2019a)revealed that compared to hot-air drying shrimp, freeze-drying shrimp had lower levels of oxidation parameters and better-quality stability during storage. Due to boosting demand for delicious, more researchers have focused on flavor profiles of dried shrimp. We previously clarified the flavor profiles of dried shrimp at different processing stages (Lin et al.,1999; Hu et al., 2021). Furthermore, Zhang et al. (2020)investigated the similarity of aroma attributes in different parts of dried shrimp. They concluded that different parts of raw shrimp contributed significantly different aroma profiles to dried shrimp and that shrimp epidermis was the main source site of shrimp aroma. However, the flavor quality of dried shrimp on the market remains inconsistent and further flavor-influencing factors should be investigated.

The flavor of aquatic products directly influences consumer tendencies and consumption, determining the final commercial value of farmed aquatic animals (Luo et al.,2021). Flavor, consisting of taste and aroma, results from complex interactions between non-volatile and volatile compounds. Previous studies have confirmed that several factors influence the flavor of products. For fresh products, Cao et al. (2021)indicated that during one harvest cycle of laver, the umami and richness tastes are more intense in the early crops and that significant differences also exists in the volatile compounds of different crops.Additionally, Song et al. (2019)and Luo et al. (2021)suggested that environmental factors had significant influences on non-volatile and volatile compounds of aquatic flesh. Concerning processed aquatic products, extensive research has investigated differences in flavor profiles of aquatic products submitted to different processing conditions, such as microbial communities (Wang et al., 2020b), cooking temperature (Xu et al., 2016), and processing time (Qin et al., 2020). However, the effects of raw materials on the flavor of their processed products should also be further investigated, since water-soluble compounds in raw materials, such as free amino acids,nucleotides, organic acids, peptides, etc., not only intensify the palatability of products but also serve as precursors in forming volatile compounds responsible for aroma(Khan et al., 2015; Alves et al., 2020).

Dried shrimp generally uses Litopenaeus vannamei as raw materials. At present, L. vannamei accounts for a high proportion of all aquatic products in daily food consumption, and to sustain the high production of white shrimp, farmed shrimp has recently developed rapidly worldwide, especially in China (Zhang et al., 2019). Earthen pond farming is a traditional aquaculture mode in China. It usually adopts a mixed-cultured mode with fish and shellfish in open-air seawater ponds (30 – 32 ppt salinity), which could provide a better environment and sufficient feed. The shrimp in earthen pond is generally harvest from July to October. Unlike earthen pond farming, plastic film is applied to keep warming in greenhouse aquaculture pattern which has a higher stocking density and lower salinity water (25 – 28 ppt salinity), and this aquaculture mode could be harvested in the early spring,avoiding the sale peak to maximize the economic benefits.

This study processed L. vannamei from different farmed patterns (earthen pond and greenhouse)into dried shrimp. The contents of free amino acids, 5’-nucleotides,and volatile compounds in dried shrimp samples were determined, and then the taste activity value (TAV)and odor activity value (OAV)were calculated to explore the flavor profile of dried shrimp. The results of this study will provide statistical information about the flavor-influencing factors of dried shrimp.

2 Materials and Methods

2.1 Materials

Fresh shrimp (L. vannamei)was obtained in August from two aquaculture patterns of one farmer namely,earthen pond (EP)and greenhouse (GH). Fresh samples with an average weight of 22.5 g were delivered alive to the laboratory and subsequently processed into dried shrimp in accordance with our previous study (Hu et al.,2021). The dried shrimp samples from the earthen pond(EPDS)and greenhouse (GHDS)were manually peeled and stored at -50℃ until analysis.

2.2 Analysis of Taste Compounds

2.2.1 Free amino acids analysis

1.0 g of samples were homogenized in 15 mL of 0.02 mol L-1HCl for 5 min and centrifugated (5000 r min-1, 10 min, 4℃). These steps should be repeated twice. And the collected supernatant was diluted to 50 mL by 0.02 mol L-1HCl. 2 mL of solution was mixed with 2 mL of 5-sulpho- salicylic acid and then centrifugated (10000 r min-1, 10 min, 4℃). To remove residues or impurities,this supernatant was filtered through a 0.22 μm membrane.Finally, 20 μL of this solution was assayed for free amino acids using a LA8080 amino acid automatic analyzer(Hitachi Technology Co., Ltd., Japan).

2.2.2 5’-nucleotides analysis

According to a method described previously by Cao et al.(2021), 5’-nucleotides were measured using HPLC system (Agilent Technology Co., Ltd., USA)equipped with a Waters-C18column (250 mm × 4.6 mm, 5 μm)and UV detector at 254 nm.

2.2.3 Taste active value (TAV)

Taste active value (TAV)was calculated according to the following formula:

2.2.4 Equivalent umami concentration (EUC)

Equivalent umami concentration (EUC)was calculated according to the following formula (Chen, and Zhang,2007):

where MSG represents monosodium glutamate; 1218 is a synergistic constant; airepresents the quantity of umami amino acid (g (100g)-1); birepresents umami coefficient of amino acid to MSG (Glu 1.0, Asp 0.077); ajrepresents the quantity of umami nucleotides (g (100g)-1); bjrepresents umami coefficient of nucleotides to IMP (IMP 1.0,AMP 0.18, GMP 2.3).

2.3 Analysis of Volatile Compounds

2.3.1 GC-IMS analysis

The volatile compounds of dried shrimp were identified using GC-IMS (FlavourSpecR, Dortmund, Germany)in accordance with a slightly modified method of Li et al.(2020b). For analysis, 2 g of sample was put into a 20 mL headspace injection bottle and incubated at 60℃ in an agitator (500 r min-1, 20 min). After that, the heated syringe was used to inject 500 μL of headspace into the heated injector (80℃). The pre-separation was completed using gas chromatography equipped with an FS-SE-54-CB capillary column (15 m × 0.53 mm)at 40℃. The flow rate of carrier gas nitrogen (99.99% purity)was as follows: 5 mL min-1for 2 min, 15 mL min-1for 8 min, 50 mL min-1for 5 min, and 100 mL min-1for 5 min. Following that, the analytes were ionized and further separated in IMS ionization chamber (45℃). When determining volatile compounds, the retention index (RI)of each volatile compound was calculated using n-ketones C4 – C8 (Sinopharm Chemical Reagent Beijing Co., Ltd., China)as external references, and then compared RI and the drift time (DT)to those of GC×IMS Library Search.

2.3.2 GC-MS analysis

7980A/5975C Gas chromatography-mass selective detector (GC-MS, Agilent Technologies Inc., USA)equiped with a DB-5ms capillary column (30 m × 250 μm × 0.25 μm, Agilent Inc., USA)was used to detect the volatile compounds. Briefly, 2 g of sample was added with 10 μL of 2,4,6-trimethylpridine (TMP, 1000 mg L-1), as an internal standard, in a 20 mL headspace vial in a 60℃ agitator at 500 r min-1for 5 min. After that, SPME was penetrated the headspace vial at 60℃ to absorb volatile compounds. After 50 min, SPME was immediately inserted into GC injector and desorbed at 250℃ for 5 min. The programmed column temperature was as follows: the initial was 40℃ and maintained for 5 min, then increased to 250℃ at the rate of 8℃ min-1, and maintained for 5 min. Helium (99.999%, purity)at a flow of 9 mL min-1was carrier gas. The temperature of the front inlet was 250℃ in the split mode with a ratio of 5:1. The mass selective detector was performed in scan mode (m/z 30 –500, EI (70 eV), ion source temperature: 230℃, and quadrupole temperature: 150℃).

The content of volatile compounds was calculated using the following equation:

N-Alkanes (C6–C30)were employed to calculate linear retention indices of volatile compounds. In addition,identification of volatile compounds was based on a comparison of GC retention indices (RI)and mass spectra(MS)with NIST 14 and Wiley11 libraries.

2.3.3 Odor activity value (OAV)

Odor activity value (OAV)was calculated as follows:

where Ciis the content of volatile compound detected in the sample; OTiis the odor threshold of this compound found in literature (Karahadian and Johnson, 1993; Grosshauser and Schieberle, 2013; Gu, et al., 2013; Mall and Schieberle, 2017).

2.4 Statistical Analysis

The experiments were performed in triplicate and the results were presented by ‘mean value ± standard deviation’. SPSS 25 software (IBM, Armonk, USA)was employed for the experimental data analysis. The significance of difference was analyzed using a t-test, and P <0.05 indicates a significant difference.

3 Results and Discussion

3.1 Free Amino Acids

Free amino acids provide key nutritional values and significantly improve the taste characteristics of many foodstuffs. Additionally, the variation of FAA confers unique flavor profiles for different aquatic products (Wang et al., 2019). In the present study, 15 FAAs were detected in EPDS and GHDS, and an independent-sample t-test revealed that their contents in dried shrimp were significantly affected by aquaculture modes, with dried shrimp from the earthen pond exhibiting the highest content (Table 1). It could be explained that FAAs in raw shrimp,osmotic regulators in crustaceans for adapting salinity,would be enhanced in high salinity earthen pond aquaculture (Luo et al., 2021).

The contribution of FAA to taste is determined by both contents and thresholds, which is calculated through taste active value (TAV). Six FAAs with TAV > 1, including arginine (Arg), glycine (Gly), alanine (Ala), glutamic acid(Glu), proline (Pro), and histidine (His)in descending order, could have directly contributed to the overall taste profile of dried shrimp. This result was also reported in the study of Xu et al. (2016), who analyzed the taste compounds in steamed shrimp. It has been reported that typical tastes of cooked crustacean meat centering around sweet and umami notes mainly are attributed to the FAAs,such as Gly, Pro, and Ala for sweet and Glu comprising umami-tasting molecules (Meyer et al., 2016). Furthermore, although Arg has a bitter taste, its widespread presence in kinds of seafood contributes to an overall pleasant preference (Shiau et al., 2018). These tasty FAA with higher TAV values observed in EPDS samples imply that dried shrimp from earthen pond might be more palatable. In addition, TAV of Val in GHDS samples was more than 1, imparting a negative effect on the taste of GHDS samples.

3.2 Flavor Nucleotides

After death, ATP degradation in shrimp produces a series of derivatives, such as ADP, AMP, IMP, and HxR(Seki et al., 2017). Among them, flavor nucleotides, including AMP, IMP, and GMP are crucial taste activecompounds to produce umami taste in aquatic products(Zhu et al., 2021). Table 2 summarizes the contents and TAVs of 5’-nucleotides in dried shrimp. AMP was a dominant nucleotide in dried shrimp, followed by GMP,implying thermal degradations of ATP mainly accumulated AMP during dried shrimp processing. Furthermore,earthen pond aquaculture resulted in significantly increased AMP and GMP in dried shrimp. Although IMP in dried shrimp was present at low concentrations, its content was slightly higher in EPDS than in GHDS samples.It is acknowledged that high levels of AMP would impart sweet and umami characteristics and enhance overall complexity in food (Fuke, and Ueda, 1996). IMP contributes to an extremely pleasant taste which is also responsible for taste complexity and sweetness even at low concentrations (Liu et al., 2021). Furthermore, a greater effect of GMP than AMP and IMP on umami taste in chicken nuggets was observed in the study of Sabikun et al. (2021). Additionally, there is a synergistic impact between flavor nucleotides and umami amino acids,which produce a stronger umami taste when combined together in certain ratios. This effect is calculated by EUC.In the present study, the EUC value of EPDS samples reached 23.45 MSG (100 g)-1, almost double that of GDHP (14.85 MSG (100 g)-1), indicating that EPDS samples had a more intense umami taste.

Table 1 Contents and TAVs of free amino acids in dried shrimp from different aquaculture patterns

Table 2 Contents and TAVs of 5’-nucleotides in dried shrimp from different aquaculture patterns

3.3 GC-IMS Analysis

In shrimp samples, 45 volatile compounds were identified using GC-IMS, including alcohols, aldehydes, ketones, esters, and O-/N-containing heterocyclic compounds, and based on signal intensities of these compounds, principal component analysis (PCA)was applied to highlight the difference in flavor profiles of the dried shrimp samples. In PCA plot (Fig.1), the extracted principal components, PC1 and PC2, represented 82% and 11% of the variation, respectively, indicating that the two main components cover the main information about volatile compounds in samples. The locations of clusters clearly demonstrated that the group of EPDS samples were far from that of GHDS samples, implying that flavor profiles of dried shrimp samples from different aquaculture modes could be effectively separated. This result revealed that aquaculture modes of raw materials greatly affected the volatile compounds in dried shrimp.

Fig.1 PCA scoring plot of the result of GC-IMS.

For comparing volatile compounds in different samples,the changes in each substance were revealed intuitively by fingerprint (Fig.2), in which each row represented a sample, each column represented a compound, and the signal peak color represented substance concentration.The darker the red color, the higher the density. In addition, because of their high concentrations, some compounds may produce multiple signals (monomers, dimers,or trimers).

As depicted in Fig.2, volatile compounds (2-heptanol,heptanal, 3-methylbutanal, 2,3-butanediol, etc.)were generally found in dried shrimp samples, and only some variations in signal intensity were observed. Alcohols,attributed to oxidation and hydrolysis of polyunsaturated fatty acids, are important volatile compounds in food(Marušić Radovčić et al., 2016). Alcohol compounds, including 2-heptanol, pentan-1-ol, 3-methyl-3-buten-1-ol,2-methylbutan-1-ol, 3-methyl-3-butenol, 4-methyl-2-pentanol had weaker signal intensities in EPDS than in GHDS samples, and the dimer of 1-propanaol was detected only in GHDS samples. Consistent with alcohols,more abundant aliphatic aldehydes were identified in GHDS samples. These aldehydes are also mainly associated with oxidation and degradation of lipids, such as hexanal and heptanal derived from the oxidation of linoleic and arachidonic acids, respectively (Guo et al., 2019),and pentanal from 13-linoleic pyrolysis (Zang et al.,2020). It is reported that ketones are mainly produced from the oxidation of polyunsaturated fatty acids, or through thermal degradation and oxidation of amino acids(Dong et al., 2018). In this study, 2,3-pentanedione concentration moderately increased in GHDS samples, whereas 3-hydroxybutan-2-one concentration was significantly higher in EPDS samples. These results indicated that lipid oxidation was one of the most important generation sources of odor compounds in dried shrimp and this reaction would be enhanced when raw shrimp are derived from greenhouse aquaculture mode. In addition,the stronger lipid oxidation produced 3-pentanone, propanone, n-butanol, and 2-hexanol, which were identified only in GHDS samples. These compounds might be potential markers for distinguishing dried shrimp samples from different aquaculture modes.

Fig.2 Gallery plot of volatile compounds in dried shrimp samples from two aquaculture patterns. EPDS, earthen pond dried shrimp; GHDS, greenhouse dried shrimp. Each row represents a sample, each column represents a volatile compound. M means monomer, D means dimer. 3 parallels in each group.

Other volatile compounds in dried shrimp, namely furfural, thiazole, 3-methylbutanal, ethyl pyrazine, 2-ethyl-5-methylpyrazine, 2-acetylfuran, and benzaldehyde, were also detected in shrimp samples, with higher concentrations in GHDS samples. These compounds mainly resulted from Maillard reaction, which involved amino acids and reducing sugars. For instance, 3-methylbutanal and benzaldehyde were derived from Strecker reaction of leucine and phenylalanine, respectively (Feng et al.,2017). Furthermore, 2-methylbutual, a product of Strecker reaction of isoleucine, was found only in GHDS samples. As a result, it could be concluded that stronger Maillard reaction occur when raw shrimp is derived from greenhouse aquaculture mode.

3.4 GC-MS Analysis

GC-MS isolated 28 volatile compounds in dried shrimp,categorized as alcohols, aldehydes, ketones, N-containing compounds and other compounds (Table 3), and Fig.3 showed contents and numbers of these compounds in two samples. As illustrated in Fig.3, aldehydes and N-containing compounds accounted for the majority of 28 volatile compounds in the dried shrimp. Additionally, it could be observed that the content of these volatile compounds in GHDS samples was much higher than in EPDS samples,consistent with GC-IMS analysis results. The differences of volatile compounds in dried shrimp could be due to variations in the microstructure of raw shrimp. Refaey et al. (2018)and Luo et al. (2021)have reported that higher stocking density and lower environmental salinity significantly decreased the intensity of muscular bundles,resulting in the susceptivity to thermal processing. Addi-tionally, previous research indicated that physical disruption of the microstructure during thermal processing was critical in determining the substrates available for chemical reactions, such as lipid oxidation, Maillard reaction,and Strecker reaction, which consequently greatly impact volatile compounds (Klevorn and Dean, 2018; Maggiolino et al., 2019; Xu et al., 2019). In this case, the microstructure of shrimp from greenhouse aquaculture mode is tended to be disrupted, releasing more volatile compounds during dried shrimp processing.

Table 3 Contents of volatile compounds in dried shrimp from different aquaculture patterns

Fig.3 Contents (a)and numbers (b)of volatile compounds in dried shrimp from two aquaculture patterns.

According to the aforementioned concentrations and thresholds of volatile compounds, their contributions to the overall aroma were determined by calculating OAV. If OAV＞1, that volatile compound is considered to be an aroma-active compound (AAC), which exerts a great effect on the food flavor. In the present study, 11 AACs were identified in two dried shrimp samples, comprising one alcohol, six aldehydes, and four N-containing compounds, as indicated in Table 4. Among them, 1-octen-3-ol, nonanal, 3-methyl-butanal, trimethylamine, 2,5-dimethylpyrazine, and 2,3,5-trimethylpyrazine were found in both dried shrimp. 1-octen-3-ol, derived from linoleic acid and arachidonic acid degradation (Wang et al., 2020),was found to be the only alcohol identified as an aromaactive compound in dried shrimp. This compound imparts a mushroom-like odor to aquatic products (Wu et al.,2014). Aldehydes are more crucial volatile compounds than alcohols because they have lower odor thresholds,and can exert stronger effects on the aroma of marine products. Nonanal, pentanal, and octanal are usually associated with a characteristic odor of oil and grass (Zhao et al., 2015), and it is confirmed that these compounds were produced from thermal oxidation and degradation of oleic acid and linoleic. In this study, OAVs of these saturated aldehydes were considerably higher in GHDS than EPDS samples, implying that these compounds have a greater impact on the flavor profile of dried shrimp from greenhouse farmed mode. Meanwhile,E,E-2,4-nonadienal, which is also generated from linolenic acid oxidation, was detected exclusively in GHDS with a relatively high OAV (171.2). Sghaier et al. (2015)indicated that nonadienal formation could lead to off-flavor characterized by a fishy odor. Additionally, An et al. (2020)suggested that nonadienal was one of the most potent aroma compounds in surimi, which imparted a fishy odor. Furthermore, Wang et al. (2020a)revealed that this compound was responsible for a strong fish smell in fish grill and skin. Therefore, (E, E)-2,4-nonadienal,with a characteristic fishy odor, might be a key volatile compound for distinguishing the aroma profile of dried shrimp from different aquaculture modes. In addition to lipid derivatives, other aroma-active compounds were mainly originated from Maillard reaction, such as 3-methylbutanal, 2,5-dimethylpyrazine, and 2,3,5-trimethylpyrazine. These Maillard reaction products impart a palatable roasted aroma in kinds of cooked seafoods, whereas a high content of pyrazines is associated with decreased aroma palatability, which smells burnt(Vazquez- Araujo et al., 2010; Lipan et al., 2020). In Table 4, it is worth noting that OAVs of pyrazines in GHDS samples, especially 3,5- diethyl-2-methylpyrazine, were vastly larger than those in EPDS, implying that GHDS samples exhibited irritant burnt odors due to high contents of pyrazines resulted from Maillard reaction.

Table 4 OAVs of major aroma-activity compounds in dried shrimp from different aquaculture patterns

4 Conclusions

According to experimental data in this study, there are significant differences found in taste and aroma compounds in dried shrimp products processed by L. vannamei from two aquaculture patterns. For taste compounds, dried shrimp from earthen pond possessed more free amino acids and flavor nucleotides contributing to sweet and umami taste. For volatile compounds, dried shrimp from greenhouse aquaculture exhibited higher TAVs of aroma active compounds, particularly E,E-2,4-nonadienal and pyrazine compounds, imparting undesirable fishy and burnt odor to dried shrimp. As a result,dried shrimp from earthen pond had a better flavor profile.

Acknowledgements

This work was supported by the National Key R&D Program of China (No. 2018YFD0901004); and the Central Public-Interest Scientific Institution Basal Research Fund, YSFRI, CAFS (No. 20603022020013).