The impact of quinoa flour on some properties of ayran

Abstract In this study, some physical, chemical, microbiological, and sensory properties of ayran produced from quinoa flour addition at different ratios (0.1, 0.2, 0.3, and 0.4%, w/v) were investigated. The effect of quinoa addition and storage time on pH, titration acidity, serum separation, L values and microorganism counts were statistically significant (p < .05). The counts of Streptococcus salivarus subsp. thermophilus and Lactobacillus delbruecki subps. bulgaricus had a wide range between 7.13 and 7.52 log CFU/mL and 3.62 and 3.98 log CFU/mlL At the end of the storage, the general appreciation score of the sample containing 0.2% quinoa flour was found to be higher than the other samples.

The effect of quinoa flour on physical, chemical, sensory, and microbiological properties of ayran was also investigated during storage period.

| Materials
Ayran was produced in the Department of Food Engineering at Pamukkale University, Faculty of Engineering. UHT milk used as material was obtained from the local market of Denizli. Y 811 (10 U)-coded DVS lyophilized yogurt culture (Streptococcus salivarus subsp. thermophilus and Lactobacillus delbrueckii subsp. bulgaricus) was provided by the Maysa Food (İstanbul, Turkey). White quinoa used in the study belonged to A La Çiftçi brand. White quinoa was first shredded in a blender (Waring 8011 ES HGB2WTS3, USA), and then, quinoa flour was passed twice through a laboratory sieve (Retsch, 355 µm pore diameter, Germany). Salt was purchased from local market.

| Use of starter culture
Ten units (500 L milk) of Y 811-coded DVS culture was calculated according to the amount of milk used in production and weighed under aseptic conditions. After heat treatment, starter culture was added to chilled milk at 43°C.

| Ayran production
Cow milk was standardized up to 6% dry matter by using water.
The heat treatment (90°C-10 min) was applied and milk was cooled to 43 ± 1°C. Starter culture was inoculated at 42°C. Then, divided into 5 groups, the first ayran group (C) did not contain quinoa flour.

| Chemical analysis
The fat content and dry matter content were determined by using Gerber method (Bradley et al., 1992) and gravimetric method (Metin & Öztürk, 2002), respectively. The protein content was determined by using Kjeldahl method (AOAC, 1990). The titration acidity was expressed as %lactic acid (Bradley et al., 1992). The pH was measured by using a pH meter (Crison pH-Meter BASIC 20+, Barcelona, Spain).

| Phenolic content and antioxidant activity analysis
The total phenolic content was evaluated by Ertan et al. (2017) with modified procedure. Sodium carbonate (75 g/L) and Folin-Ciocalteu phenolic reagent (1:10, Folin-Ciocalteu phenolic reagent: water) were used during this analysis. 1 ml of ayran samples was placed in the test tube, and 5 mL of FCR and 4 ml of Na 2 CO 3 were added and stored in the dark for 2 hr. Samples were centrifuged (Nuve 1200 NF, Ankara, Turkey) at 3600 x g for 10 min at 4°C. At the end of the centrifugation, the absorbance of the samples was read at 760 nm on a spectrophotometer (PG Instruments T80 UV/Vis Spectrophotometer, UK).
Total phenolic content of the samples was given in mg GAE/L. For antioxidant activity analysis, Thaipong, Boonprakob, Crosby, Cisneros Zevallos, and Byrne (2006) proposed method has been partially modified. Stock solution of DPPH (2,2-diphenyl-1-picrylhydrazyl) was prepared as 24 mg/100 mL methanol and stored at −18°C. Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid) solution was used for the calibration curve. For samples or standards, 600 µL was added to 2,400 µL DPPH working solution and allowed to stand in the dark for 1 hr at room temperature. The samples were centrifuged (Nüve NF 1200R, Ankara, Turkey) at 4 ○ C at 3600 x g for 10 min. The end of this period, the absorbances of each mixture were read on a spectrophotometer (PG Instruments T80 UV/Vis Spectrophotometer, UK) at 515 nm. Antioxidant activity results were expressed as µmol Trolox equivalent (TE)/L.

| Physical analysis
For serum separation, samples were placed in 100-mL graduated cylinder and stored at 4°C. On the 1st, 7th, and 14th days of the samples, serum separations were measured by looking at the measurements. The results were given as percentages (Tamuçay Özünlü, Koçak, & Aydemir, 2007). Color was measured on yogurts using a HunterLab colorimeter (Hunter MiniScan Xe, Hunter Associates Laboratory, USA) according to the HunterLab scale that is L (lightness), a (red/greenness), and b (yellow/blueness) (Arslan & Bayrakçı, 2016).

| Fat analysis
The extraction of the samples was made according to the modified Folch method (Folch, Lees, & Stanley, 1957). Fatty acid methyl esters (FAMEs) were prepared according to the IDF standard (Anonymous, 1999). Approximately 0.1 g of the sample extract was transferred into a centrifuge tube and dissolved in 2 mL of hexane. Subsequently, 0.2 mL of a 2 N KOH solution prepared in methanol was added and centrifuged at 2770 x g for 5 min. The were used. The injection volume was 1 μL. According to the temperature program, it increased from 50°C in 2 min and from 50°C to 174°C in 14 min. It was then increased by 2°C per minute to 215°C. It was maintained at this temperature for 25 min. Hydrogen was used as the carrier gas. The injector and detector temperatures were 280°C.

| Sensory properties
Ayran sample was evaluated in terms of appearance, color, odor, consistency, taste, and general appreciation by 40 panelists group from Pamukkale University Food Engineering Department. Panelists scored on the sensory form according to the hedonic scale of 1-7 (Altuğ & Elmacı, 2005;Er Gürmeriç, 2008).

| Statistical analysis
The results were evaluated statistically by using SPSS program (SPSS package program, Version 20). Analysis of variance (ANOVA) was used for comparison. Duncan test was used in cases where the difference between samples was significant. Statistically differences were determined at p < .05 level.

| Chemical composition
The contents of dry matter, protein, fat, and pH value of milk used in yogurt production were 11.08%, 2.90%, 3.00%, and 6.65, respectively. The contents of protein and fat of quinoa flour were 13.13% and 3.76%, respectively. Table 1 shows some chemical analysis results for the first day of storage of ayran samples. The differences between the samples of ayran on the protein and dry matter contents were statistically significant (p < .05). Ayran, according to Turkish Food Codex, was included in the semiskimmed ayran group (0.8%-1.2% milk fat) (Anonymous, 2009). The protein content of ayran samples changed by 2.06%-2.39%. The addition of quinoa flour affected the protein content. Because quinoa is one of the foods rich in protein, the protein content of quinoa flour used in the study was determined as 13.13%.
The pH value and the titration acidity values of samples are shown in Table 2. The differences of samples and storage time on pH and titration acidity values of the samples were statistically significant (p < .05). It was found that the pH value decreased gradually in the samples with addition of quinoa flour and the Q4-coded ayran sample had the lowest pH value compared to the others. This result shows that the use of quinoa flour reduced the pH value in ayran. Gursoy et al. (2016) reported that pH values of ayran samples ranged from 4.53 to 4.10 during storage period. Table 2 shows that titratable acidity in ayran samples varied between 0.48% and 0.62%. The difference between the titratable acidity values between the samples and storage period was found to be statistically significant (p < .05).
samples decreased throughout the storage period. Codină, Franciuc, and Mironeasa (2016) investigated the effects of quinoa flour (0%, 0.2%, 0.6%, 1%, 1.4%, and 2%) in yogurt production. As a result of the research, they found that the addition of quinoa flour caused a decrease in pH values. Curti, Vidal, Curti, and Ramón (2017) found that when adding different amounts of quinoa (1%, 3%, and 5%) to yogurts, the pH of yogurts decreased due to storage, and that the yogurt affected the gel structure, and that it was important to affect consumers.
It was found that these studies supported the pH decrease in our study and the addition of quinoa had a significant effect on pH in fermented products. Acidity influences to the serum separation and rheological properties (Gursoy et al., 2016).

| Phenolic content and antioxidant activity
The total phenolic content decreased during storage (Table 3).
The total phenolic contents of samples at 1 and 14 day of storage were observed between 197.59 and 225.38 mg GAE/L and 100.01 and 105.32 mg GAE/L, respectively. It was determined that the addition of quinoa did not affect the total phenolic content and antioxidant activity. The total antioxidant activity increased at the 7th day and continued to increase at the 14 th day for all samples. This increase was found statistically significant (except Q4-coded sample).
In a study on the antioxidant capacity of various fermented milk, the antioxidant capacity of ayran was found to be 0.092 mM TE/kg (Najgebauer -Lejko & Sady, 2015).
Çelik (2016) reported that the total phenolic content of kefir produced from different proportions of propolis during the storage period. The study showed that the phenolic content of samples ranged from 0.05 to 1.15 mg GAE/g at the beginning of the storage.
Also in this study was observed a reduction in total phenolic content during storage. This result supported the decrease in total phenolic content during storage in our study. Lorusso, Coda, Montemurro, and Rizzello (2018) in their study found that the total phenolic content in yogurt-like drinks contained quinoa flour between 4.00 and 9.60 mmol/kg.

| Rheological properties
The values of the consistency coefficient (k) of ayran samples and the flow behavior index (n) were determined as a result of rheological measurements performed at 4°C on the first day of storage and shown in Table 4. Flow behavior index of ayran samples varied from 0.72 to 0.93. In this study, the flow behavior index of ayran samples was less than 1. It showed non-Newtonian pseudoplastic flow behavior. Similar results were also reported by Gursoy et al. (2016).
Apparent viscosity of samples decreased with the severity of shear rate (Figure 1). In the present study, the highest apparent TA B L E 3 Changes of total phenolic content (mg GAE/L) and total antioxidant activity (µmol TE/L) of ayran samples during storage  In our study, it was found that flow behavior index decreased and consistency coefficient increased with an increase in quinoa flour. In a study examining the rheological properties of ayran using different levels of water and salt, the researchers explained that ayran showed non-Newtonian behavior based on the power law model (Köksoy & Kılıç, 2003).

| Physical properties
The storage period and quinoa addition were a significant factor for serum separation values of ayran samples (p < .05). The serum separation value of the control group (C) ranged from 1.50% to 15.75%, while samples containing %0.4 quinoa flour ranged from 5.50% to 26.75% during storage. Sample with 0.4% added quinoa exhibited the highest serum separation values compared with other samples (Table 5).
The differences between the formulations of the samples and storage on the L values were found to be statistically significant (p < .05). On the 1st and 14th days of storage period, whiteness index (L) value was observed to increase gradually. It was observed  (A,B,C) in the same ayran samples were significant (p < .05). Differences between ayran samples (a,b,c) indicated in different lowercase letters at the same storage time (p < .05).
that the whiteness index of the samples decreased as the concentration of quinoa used increased.
Ayran samples had negative a value and positive b value during storage.

While the differences between the a and b values during storage
were not statistically significant (p > .05), the differences between the samples were found to be statistically significant (p < .05). The lowest a (greenness) values among the samples during storage were determined as (Q4) ayran sample containing the most quinoa flour.
While the b value of the control sample was determined as the lowest value at the beginning of storage, the difference between the b values of the samples on the 14th day was not found to be statistically significant (p < .05).

| Fatty acid composition
The saturated fatty acid content and unsaturated fatty acid content of quinoa flour were 13.00% and 86.97%, respectively. Linoleic acid was the most abundant fatty acid in quinoa flour. Elaidic + oleic acid and palmitic acid were the second and third highest fatty acids.
Palmitic acid was the principal saturated fatty acids.

| Microbiological properties
The differences between the treatments and storage period on In a study investigating the effect of quinoa added on fermented milk in different proportions (0, 1, 2, and 3 g/100 g), Casarotti, Carneiro, and Penna (2014) found that quinoa flour did not affect the fermentation time during production but increased acidity during storage. Codină et al. (2016) stated that the addition of quinoa flour had a positive effect on the development of starter yogurt bacteria due to the decrease in pH and increase the total acidity.
The results of these studies were compared with the results of our study, and the effect of quinoa flour on microorganism was similar to other studies. The addition of quinoa flour was observed to play an encouraging role in L. delbrueckii subsp. bulgaricus development.

| Sensory properties
The sensory analysis results for the appearance, color, odor, consistency, taste, and general appreciation of ayran samples are shown in Note: The differences between storage times shown in different capital letters (A,B) in the same ayran samples were significant (p < .05). Differences between ayran samples (a,b) indicated in different lowercase letters at the same storage time (p < .05). appreciate score was influenced by the type of sample and storage time.
The highest appearance and odor scores were determined in C-coded ayran sample at the beginning of storage. The Q1-coded sample was higher of both appearance and color scores than those of other sample at 14 days of storage. Color scores of all samples (except Q1) gradually increased after the first day of storage until the 7th day. Odor and taste scores varied from 4.77 to 5.40 and from 4.35-5.27, respectively. The highest consistency scores were obtained the Q1-coded sample on the first day of storage and the Q2coded sample at the end of storage. It was observed that consistency score decreased below 5 points at the end of storage. A decrease in consistency scores content of ayran may be released of quinoa flour in the structure depending on the time.
The Q2 sample had the highest general appreciate scores (p < .05) followed by Q1, C, Q3, and Q4 samples at the end of storage. The general appreciation score was over 5, and general appreciation of all samples was similar at 7 days of storage. The general appreciation of samples decreased at the end of storage.

| CON CLUS ION
It was found that the pH values of the samples decreased during the storage period and the pH values of samples ranged between 3.99 and 4.25. The addition of quinoa flour reduced the pH of ayran, and the lowest pH value was determined in the Q4-coded ayran sample containing the highest quinoa flour. The content of protein content and dry matter increased with an increase in the quinoa flour.
Different treatment did not cause a significant change in fat content, total phenolic content, and antioxidant activity. Erucic acid was found in samples containing quinoa.