Introduction

Chicken has been perceived as body-warming food in traditional medicine, with carnosine and anserine being the key components for its bioactivity (Li et al., 2012). Carnosine (β-Ala-His) and anserine (β-Ala-His(3-Me)) are imidazole-containing dipeptides of β-alanine and L-histidine, found in the skeletal muscle, heart, and central nervous system (Mora et al., 2008; Tian et al., 2007) of vertebrates (chicken, pork, cattle, etc). Carnosine and anse-rine play important roles in physiological functions such as potent intracellular pH buffering, anti-oxidation, and anti-glycation (Decker et al., 2000; Dukic-Stefanovic et al., 2001; Intarapichet and Maikhunthod, 2005; Lee et al., 1998). These peptides have gained considerable attention as anti-oxidant food supplements for eye disorders (Bold-yrev et al., 2004; Chan and Decker, 1994; Yanai et al., 2007). Since the formation of advanced glycated end-products (AGEs) and lipid peroxidation are closely related to dege-nerative diseases such as diabetes, aging, Alzheimer’s di-sease, and cancer (Ahmed, 2005; Gunter et al., 2007, John-son and Lund, 2007), carnosine and anserine are receiv-ing much attention as natural antioxidants (Hipkiss and Brownson, 2000; Kohen et al., 1988; Nabetani et al., 2012).

White muscle is known to contain a higher amount of carnosine than does red muscle tissue (Boldyrev et al., 2004; Brown, 1981; Hipkiss, 2005; Plowman and Close, 1988). Chicken is a representative animal with higher car-nosine content among the white muscle tissue animals (Boldyrev and Severin, 1994; Kim et al., 2012). A few efforts have been made by other researchers to identify the optimum extraction conditions of the dipeptides. For example, Maikhunthod and Intarapichet (2005) attempted heat and ultrafiltration processes with acidic extraction, and Nabetani et al. (2012) applied membrane separation to prepare a functional peptide extract containing carnos-ine and anserine from chicken. Most of the reported car-nosine extract preparation methods involve acid extrac-tion, followed by de-proteinization using ethanol treat-ment, which could alter the actual carnosine concentra-tion. In addition, no systemic investigation has been at-tempted for optimizing carnosine and anserine extraction while preserving the maximum physiological activities. Herein, we have optimized the extraction process for pep-tide mixtures with the highest anti-oxidation and anti-gly-cation activity through response surface methodology (RSM) using multiple objective optimizations.

Materials and Methods

Chemicals and reagents

All the reagents used in this study were of the highest purity available. Acetonitrile, methanol, ethanol, bovine serum albumin (BSA), methylglyoxal, L-ascorbic acid, α,α-diphenyl-β-picrylhydrazyl (DPPH), hydrochloric acid (HCl) 35%, sodium chloride (NaCl), ferrous chloride, boric acid, o-phthaldialdehyde (OPA), L-carnosine, and L-anserine were purchased from Sigma Aldrich Co. (USA).

Experimental design and optimization of extract-ing conditions by statistical analysis

RSM, an empirical modeling technique, was used to estimate the relationships between a set of experimentally observed results. Central composite design (CCD) was used for predicting the optimum extraction conditions for carnosine and anserine from chicken breast using Design-Expert (version 8.0, State-Ease Inc., USA). The details of CCD for a given range of parameters based on coded and actual levels are shown in Table 1, with two independent variables, extraction time and ethanol concentration. Low, middle, and high levels of each variable were designated as -1, 0, and +1, respectively. Overall, thirteen combina-tions were chosen in random order according to the CCD configuration for four dependent variables: (i) concentra-tions of carnosine and anserine, (ii) anti-glycation activity, (iii) DPPH radical scavenging, and (iv) inhibition of lipid peroxidation. To obtain a reliable estimate of the opti-mum extracting conditions, all data were fitted to a mod-ified Gompertz function using a non-linear regression model. The results of the thirteen combinations were fit-ted to a quadratic polynomial model. Statistical testing of the model was performed at p<0.05 by analysis of vari-ance (ANOVA) to test the significance and adequacy of the model. Finally, simultaneous optimization for the ex-traction was performed using Design-Expert (version 8.0, State-Ease Inc., USA) to identify the conditions for the highest physiological activities such as anti-glycation activity, DPPH radical scavenging, and inhibition of lipid peroxidation.

Table 1.Factors for the central composite design for response surface methodology (RSM)

Extraction of carnosine and anserine from chicken muscle

Chicken samples (20 and 90 wk laying hen) were recei-ved from Department of Animal Science and Technology, Chung-Ang University, Korea. At the end of each week, the chickens were sacrificed by cervical dislocation, a method approved by Animal Care Committee of Chung-Ang University. Carnosine and anserine were extracted according to the method proposed by Auh et al. (2010), with a slight modification. Muscle samples from chicken breast were isolated, frozen at -80°C, and freeze-dried for 72 h. The dried muscle samples were finely ground, and 0.5 g of the ground muscle was suspended in 12 mL of solvent (0-70% of ethanol) and incubated at 25°C with moderate shaking for 1 to 3 h. Subsequently, the mixture was centrifuged at 14,000 g for 30 min, and then, 300 mL of the supernatant was mixed with 900 mL of cold etha-nol and kept for 30 min at 4°C. The supernatant obtained after centrifugation was filtered through a membrane syringe filter (MCE type, 0.45 mm, Adavantec, MFS Inc., Japan) and freeze dried. Lyophilized samples were stored at -40°C until further analysis.

HPLC analysis

Carnosine and anserine concentrations were determined in accordance with the methods reported by Aristoy et al. (2004) and Auh et al. (2010). An HPLC system (Gilson Medical Electronics, USA) equipped with a fluorescence detector (Agilent, USA) and a Spherisorb SCX column (4.6×250 mm, Waters Co., USA) was used for the analy-sis. 10 mL of the sample solution (10%, w/v) was mixed with 100 mL of OPA solution and placed for 2 min at room temperature in the dark. Subsequently, 20 mL of the derivatized sample was injected into the HPLC system. The mobile phase was composed of 80% of phase A (20% acetonitrile in 6.6 mM HCl) and 20% of phase B (phase A containing 0.8 M NaCl), and the elution was effected at 0.6 mL/min for 30 min at 40°C. Fluorescence detection was performed on Spectrofluorimeter RF-1500 (Shimadzu, Japan) using 338 and 445 nm as the excita-tion and emission wavelengths, respectively (Plowman and Close, 1988).

Anti-glycation activity (AGE inhibition)

The anti-glycation activity of carnosine-containing ext-ract from chicken breast (20- and 90-wk laying hen) was measured by % inhibition of AGE formation, according to the method reported by Li et al. (2008) but with a slight modification. The extract was mixed with BSA (50 mg/mL) and methylglyoxal (3 mM) in 50 mM sodium phos-phate buffer (pH 7.4) and incubated at 25°C for 24 h under dark conditions. The formation of AGE was quantified by measuring the fluorescence intensity (excitation at 350 nm, emission at 450 nm). All the measurements were conduc-ted in triplicate.

Anti-oxidation activity

The anti-oxidation activity of carnosine-containing ext-ract from chicken breast (20- and 90-wk laying hen) on DPPH radicals was measured by Wu et al. (2002)’s method, with a slight modification. The samples were mixed with 0.2 mM DPPH (in 95% methanol) and kept for 30 min at room temperature; then, the absorbance at 517 nm was measured and the anti-oxidation activity was calculated. All the measurements were performed in triplicate.

Inhibition of lipid peroxidation (%)

The inhibition of lipid peroxidation of carnosine-con-taining extract from chicken breast was determined by a slight modification of the ferric thiocyanate method (Shimada et al., 1992). To accelerate the oxidation, 190 mL of the extracted sample, 400 mL of 50 mM potassium phos-phate buffer (pH 7.0), 200 mL of 2.5% linoleic acid, 200 mL of ethanol, and 10 mL of 10% Tween 20 were mixed and incubated for 24 h at 40°C. Various concentrations of ascorbic acid (0.001 to 10 mM) were used as the positive control. To 15 mL of the reaction mixture were added 1455 mL of 75% ethanol, 15 mL of 30% ammonium thiocyan-ate, and 15 mL of 20 mM ferrous chloride solution in 3.5% HCl. After the mixture was stirred for 3 min, the peroxide values were determined by monitoring the absorbance at 500 nm, and subsequently, the % inhibition of lipid per-oxidation was calculated.

Results and Discussion

Optimizing extracting conditions and physiological activities

The extraction conditions for carnosine-containing mix-tures with physiological activities such as anti-glycation activity, DPPH radical scavenging, and inhibition of lipid peroxidation was optimized to achieve the highest activi-ties. Thirteen combinations of CCD were used with vary-ing ethanol concentrations and lengths of time, and the resulting responses of four different factors were deter-mined (Tables 2 and 3). Subsequently, multiple objective optimization was applied to identify the optimum condi-tions with the highest amounts of carnosine and anserine, along with maximized levels of physiological activities. Carnosine and anserine could be easily extracted within 2 h at low ethanol concentrations of less than 35%.

Table 2.Experimental design used in response surface analysis for optimization of extraction conditions from 20-wk chicken breast

Table 3.Experimental design used in response surface analysis for optimization of extraction conditions from 90-wk chicken breast

The AGE inhibition activity of the extract was sensitive to ethanol concentration, and higher activity was observed when the extraction was performed using water within 3 h. However, the DPPH radical scavenging activity was not significantly affected by the extraction time, and higher activity was observed with lower ethanol concen-trations. Inhibition of lipid peroxidation was the highest when lower concentrations of ethanol were used with 3 h of extraction time. Further, the response of each factor followed the quadratic polynomial models as expected, thus, multiple objective optimization was clearly achieved, as envisioned. Fig. 1 clearly shows the successfully iden-tified optimum conditions for carnosine and anserine extraction from 20- and 90-wk old chicken breast. The optimum conditions were identified as water extraction for 1.60 h and 2.12 h, in 20-wk and 90-wk laying hen, respectively. Although carnosine and anserine are highly water soluble dipeptides, they can be easily isolated with water extraction within 2 h from chicken muscle. Table 4 shows the comparison of the estimated values under the optimum conditions with the experimental values to con-firm the reliability of the method. In general, most of the responses under the optimum conditions were higher in the 20-wk laying hen than in the 90-wk samples, in accor-dance with the previous reports (Intarapichet and Maik-hunthod, 2005; Kim et al., 2012), which stated that the carnosine and anserine concentrations would gradually decrease with aging in chicken muscle. The carnosine and anserine contents were 3944.53 mg/kg in the 20-wk sample and 3220.98 mg/kg in the 90-wk muscle sample. AGE inhibition activity was significantly higher in the 20-wk laying hen extract, which reflected the higher amounts of carnosine and anserine in younger muscles; the anti-oxidation activity, too, was also higher in the 20-wk muscle extract than in the 90-wk samples. Signifi-cantly, in the extract obtained from the 20-wk laying hen muscle, the experimental values of carnosine and anse-rine content were within the range of estimated values, while the AGE inhibition and DPPH radical scavenging activity were slightly higher than the expected levels. For the 90-wk laying hen muscle, although the estimated car-nosine and anserine contents were comparable with the experimental values, inhibition of AGE formation and lipid peroxidation were slightly higher in the experimen-tal measurements.

Fig. 1.Contour plots for simultaneous optimization of chicken breast extract with the highest physiological activities. A, 20-wk laying hen (25°C); B, 90-wk laying hen (25°C).

Table 4.Comparison of estimated and experimental values under the optimized extraction conditions

Effect of carnosine content on physiological activities

We successfully established the optimum conditions for the isolation of carnosine- and anserine-containing extracts and simultaneous maximization of three different physio-logical activities; further, we compared the relative contri-bution of carnosine and anserine to the activities (Fig. 2). Physiological activities of the extract were primarily affec-ted by the amount of carnosine and anserine, and the ratio of carnosine seemed to be more critical on their activities. Carnosine content in 20 wk muscle sample (700.80 mg/kg) was significantly reduced in 90 wk muscle (537.60 mg/kg), thus AGE inhibition and anti-oxidation activities decreased in proportion to the carnosine ratio in the ext-ract. Relative amount of carnosine was significantly dec-reased by aging with a loss in AGE inhibition and anti-oxidation activity. In anti-oxidation effect, DPPH radical scavenging activity was more influenced by a reduction of carnosine. Although carnosine and anserine are both effective antioxidants as extensively demonstrated by Wu et al. (2003), carnosine content would be more critical determinant for anti-oxidation Thus, anti-glycation and anti-oxidation activity of carnosine and anserine extract from chicken muscle increased proportionally to the car-nosine content as well as the relative amount of two pep-tides in the muscle.

Fig. 2.Comparison of physiological activities of chicken ext-racts obtained under optimized conditions. A, AGE in-hibition; B, DPPH radical scavenging effect; C, inhibition of lipid peroxidation.