Screening and characterisation of β-glucosidase production strains from Rosa roxburghii Tratt

2.1 Yeast strains

The commercial Saccharomyces cerevisiae (S. cerevisiae) X16 obtained from Laffort Company (France) was used as a reference strain. A total of 65 strains of yeasts isolated from R. roxburghii fruit in Longli County, Guizhou Province were analysed in this study to investigate the β-glucosidase activities. All yeasts cells were cultured on YPD solid medium (1% yeast extract, 2% peptone, 2% glucose, and 2% agar) containing 100 mg/L of chloromycetin at 28 °C for 72 h, and then kept at 4 °C for later use.

2.2 β-glucosidase activity assay

Esculin semi-quantitative colouration analysis was applied to screen β-glucosidase production yeast strains. Yeasts were incubated in 200 μL esculin medium with 10⁵ live cells/mL in 96-well plates, and then incubated at 28 °C for 72 h. β-glucosidase activity was determined according to the colour depth.

To confirm the preliminary screening results, β-glucosidase activity was determined using p-nitrophenyl-β-d-glycoside (pNPG) as the substrate and the amount of p-nitrophenol (pNP) released from pNPG was checked. Firstly, yeasts were incubated in 50 mL YPD liquid medium with 5% (v/v) seed culture, cultured at 28 °C while being shaken at 160 rpm for 72 h, and collection via centrifugation at 3000 g for 10 min. The supernatant was used as an enzyme solution. Enzyme solution (0.1 mL) was mixed with pNPG solution (0.2 mL, 0.002 M) in citrate phosphate buffer (0.1 M, pH 5.0). The reaction was maintained at 40 °C for 30 min, and stopped by adding Na₂CO₃ (2.0 mL, 0.25 M). The released pNP was measured spectrophotometrically at 405 nm. One unit (U) of enzyme activity was defined as the amount of enzyme that released 1 nmol of pNP per min under the aforementioned experimental conditions.

2.3 Identification of yeast strain

The selected isolate strain was streaked on the WL nutrient agar and incubated at 28 °C for 5 days. The micromorphology of the selected yeast was captured by using a microscope (Olympus, Tokyo, Japan) and classified according to its morphotype.

The selected isolate strain was further identified by amplifying the D1/D2 domain of 26S rDNA via polymerase chain reaction (PCR) with forward primer NL1 (5′GCATATCAATAAGCG GAGGAAAAG-3′), and reverse primer NL4 (5′-GGTCCGTGTTTCAAGACGG-3′). The sequences of the D1/D2 domain of 26S rDNA were compared by BLASTN Homology Search.

2.4 Effect of brewing conditions on β-glucosidase activity

In order to ascertain the effect of temperature or pH on β-glucosidase activity, the enzyme reactions were incubated at different temperatures over the range between 20 and 60 °C (20, 30, 40, 50 and 60 °C) at pH 5, or with citrate-phosphate buffer across the pH range from 3.0 to 7.0 (3.0, 4.0, 5.0, 6.0 and 7.0). The enzyme assay was conducted with pNPG as the substrate as previously described.

Effects of glucose or ethanol on enzyme activities were identified by using glucose concentrations over a range of 0–20% w/v or ethanol concentrations over a range of 0–20% v/v, respectively. Stock solutions of glucose or ethanol were prepared in citrate-phosphate buffer (pH 5.0) and were added to the reaction mixture resulting in final concentrations from 0 to 20% w/v or 0–20% v/v, respectively. The enzyme assay was conducted as previously described.

To verify the effect of metal ion on the β-glycosidases activity, K⁺, Na⁺, Mg²⁺, Cu²⁺ and Al³⁺ were added to citrate-phosphate buffer (pH 5.0) with the final concentration of 10 mmol/L. The reaction mixtures were incubated at 40 °C for 30 min, and then enzyme activities were measured.

2.5 Kinetic of β-glucosidase enzyme

Kinetic parameters, K_m and V_max, were detected from Lineweaver-Burk plots of β-glucosidase activity at pNPG concentrations ranging from 0.2 to 2.0 mM (at pH 5.0 and 40 °C). Each sample was detected in triplicate.

2.6 Laboratory-scale fermentation of R. roxburghii wine

The R. roxburghii wine fermentation was conducted as described by Huang [10] with some modifications. Fresh R. roxburghii was washed with sterile water and crushed to acquire its R. roxburghii must. The must was filtrated with microporous film, poured into a 2 L bottle and divided into three groups: an S. cerevisiae group, a Wickerhamomyces anomalus group and a mixed group. The initial R. roxburghii must contained 50 mg/L SO₂ and 20 mg/L pectinase at pH 3.5. Logarithmic growths of S. cerevisiae X16 (10⁷ CFU/mL) and W. anomalus (10⁸ CFU/mL) were inoculated into the S. cerevisiae and W. anomalus groups, respectively. The mixed group was simultaneously inoculated with 10⁷ CFU/mL of S. cerevisiae X16 and 10⁸ CFU/mL of W. anomalus. Fermentation was performed at 25 °C without agitation until finished. Three replicates for each group were tested.

2.7 Chemical compositions of R. roxburghii wine

pH measurements were conducted using a digital pH meter (Model PT-10, Sartorius AG, Germany). The concentrations of ethanol, residual sugars, total acidity and volatile acids were determined using methods recommended by the International Organization of the Vine and Wine (OIV 2005). Each sample was detected in triplicate.

2.8 Volatile compound analysis of R. roxburghii wine

Fermented R. roxburghii wine was centrifuged at 3000 g for 15 min and supernatant was prepared for volatile compound analysis. Cyclohexanone was added to the R. roxburghii wine as an internal control. A headspace solid-phase micro extraction (HS-SPME) method was performed at 40 °C for 30 min to extract volatile compounds. An Agilent TQ8040 gas chromatograph mass spectrometer was used to analyse each specimen. The flow rate of the carrier gas (helium) was maintained at 1.2 mL/min. The column temperature was programmed as follows: 40 °C for 2 min, increased to 230 °C at a rate of 5 °C/min and then held at 230 °C for 15 min. The temperatures for the injector and detector were set to 250 and 280 °C, respectively. The mass spectrometer was operated in electron-impact mode at 70 eV. Volatile compounds of orange wine were identified by comparing their linear retention indices with those of pure standards, published data or MS fragmentation patterns obtained from databases. The odour active value (OAV), as the ratio of the concentration of a flavour compound to its odour threshold, is a parameter widely used to obtain odour patterns starting from quantitative compositions. In this research, compounds with OAV > 1 are considered to be responsible for aroma, and the higher their OAV, the more they contribute to the aroma profile.

2.9 Statistical analysis

The results were shown as mean ± SD. One-way analysis of variance, Student’s t-test, and principal component analysis (PCA) were conducted using SPSS 25.0 statistical software (SPSS). Differences were considered significant when P < 0.05.

3.1 Screening of high β-glucosidase production of yeasts

β-glucosidase production yeasts were screened from “Guinong 5” R. roxburghii fruit (Figure 1), obtained from Longli County, Guizhou Province, China. A total of 65 yeast isolates were chosen from YPD plates and then evaluated for β-glucosidase production using the esculin reaction method. As shown in Figure 2, eight yeast isolates (B3, B9, B11, C4, C5, C9, D8, and E9) were found to exhibit high β-glucosidase activity on account of their dark brown colour in esculin medium.

Figure 1:

“Guinong 5” R. roxburghii fruit used in this study.

Figure 2:

Detection of β-glucosidase production using the esculin reaction method.

To verify the β-glucosidase activities of these eight candidate yeast isolates, quantitative analysis was undertaken using p-NPG colorimetry. The results demonstrated that the highest β-glucosidase activity was produced by Strain C4 (Table 1). Therefore, the C4 strain was subjected to selected for further studies.

3.2 Identification of selected yeast strain

Firstly, morphological characteristics were checked to identify the C4 strain. The morphology of the colony on WL (Wallerstein laboratory nutrient agar) identification medium was white on the surface and dark green within, villiform and flat. The cellular microscopic morphology was assessed after staining by crystal violet (Figure 3). These observations indicated that the yeast C4 isolate might belong to the species W. anomalus.

Figure 3:

Morphology of cells and colonies of C4 strain. A, Microscopic morphology of C4 strain via crystal violet staining. B-D, Colony morphology of C4 strain on YPD medium (B), WL medium (C), and lysine medium (D).

Then, a molecular biological method was applied to determine the C4 strain. The sequence of D1/D2 domain of 26S rDNA was amplified and compared by BLASTN Homology Search. The results revealed that the sequence of yeast isolate, C4, was 99.66% identical to W. anomalus TC2-5. Taken together, it was confirmed that the isolate C4 (GenBank accession: MT705624) was identical to Wickerhamomyces anomalus.

3.3 Effects of brewing conditions on β-glucosidase activity

To identify the effect of temperature on the β-glycosidases activity from the selected yeast strain C4, enzymatic activity was detected in a temperature range from 20 to 60 °C. Data indicated that the β-glycosidases of C4 strain had an optimum temperature of 40 °C (Figure 4A).

Figure 4:

Effects of brewing conditions on β-glucosidase activity. A, Effect of temperature on β-glucosidase activity. B, Effect of pH on β-glucosidase activity. C, Effect of glucose on β-glucosidase activity. D, Effect of ethanol on β-glucosidase activity.

Figure 4B shows the effect of pH on enzymatic activity. The β-glucosidase extracted from W. anomalus C4 was strongly inhibited at pH 3.0. Maximum activity was recorded at a pH of around 5.0, therefore, the optimum pH for W. anomalus C4 was 5.0.

The effects of glucose concentrations in the range from 0–20% w/v on β-glucosidases were determined. The results of glucose inhibition are shown in Figure 4C. The β-glucosidase activity from W. anomalus C4 was strongly inhibited by 5% w/v glucose and the relative activity remained at around 50%. In the presence of 20% w/v glucose, the relative activity of this β-glucosidase was around 30%.

The effect of ethanol concentration in the range of 0–15% v/v on β-glucosidase activity was also investigated (Figure 4D). The activity of β-glucosidase from W. anomalus C4 was enhanced by ethanol concentrations in the range of 0–10% v/v with the maximum activity at a concentration of 10% v/v, however, when the concentration of ethanol exceeded 10% v/v, the β-glucosidase activity of W. anomalus C4 was inhibited.

In addition, the effect of metal ion on the β-glycosidases activity was assessed. As seen in Figure 4, K⁺, Na⁺, Mg²⁺ and Al³⁺ enhanced β-glycosidases activity, in contrast, Cu²⁺ inhibited β-glycosidases activity (Figure 5).

Figure 5:

Effect of metal ion on β-glucosidase activity.

3.4 Kinetic parameters of β-glucosidase

The kinetic parameters, K_m and V_max of β-glucosidase of W. anomalus C4 were calculated by using p-nitrophenol-β-d-glucopyranoside as a substrate. The results indicated that K_m and V_max were 0.53 ± 0.03 mM and 1.08 U/g, respectively Table 1.

3.5 Physico-chemical characteristics of R. roxburghii wine fermented with different yeast strains

The physico-chemical characteristics of R. roxburghii wine fermented with different yeast strains, including pH, ethanol content, residual sugar total acidity, and volatile acidity were further evaluated. As shown in Table 2, a lower ethanol concentration and higher residual sugar were detected in R. roxburghii wines produced using S. cerevisiae X16 compared with pure W. anomalus C4, or mixed fermentation with W. anomalus C4 and S. cerevisiae X16. In addition, fermentation with W. anomalus C4 alone and co-fermentations with S. cerevisiae X16 and W. anomalus C4 could decrease the volatile acids content compared to S. cerevisiae X16 control.

An electronic tongue assay system (INSENT SS402B, Japan) was also used to compare the sensory characteristics of R. roxburghii wines. The results showed that there were no significant sensory differences between three R. roxburghii wines in bitterness, astringency, aftertaste-A, aftertaste-B, umami, richness, and saltiness except for sourness. The sourness values measured in R. roxburghii wines produced by W. anomalus C4 pure inoculation or mixed inoculation were smaller than that by S. cerevisiae X16 alone (Figure 6).

Figure 6:

The R. Roxburghii wine taste attribute radar chart.

3.6 Volatile chemical profiles of finished R. roxburghii wines

A total of 64 volatile chemicals were founded in three different fermented R. roxburghii wines, including five volatile acids, 10 volatile alcohols, 25 volatile esters and 24 other volatile compounds (Table 3). There were 26, 45 and 58 volatile compounds detected in the three types of R. roxburghii wines produced by pure S. cerevisiae X16, W. anomalus C4 or co-inoculation with and S. cerevisiae X16 and W. anomalus C4, respectively. Moreover, wines inoculated with pure W. anomalus C4 or mixed with S. cerevisiae X16 contained lower amounts of volatile acids, increased amounts of volatile alcohols, volatile esters and other volatile compounds, compared to S. cerevisiae control. Taken together, inoculation with W. anomalus C4 could enhance the volatile aroma richness and complexity of R. roxburghii wines, regardless of type or amount.

The flavour active values (odour activity value, OAV) can evaluate the contribution of volatile compounds for wine aroma. When OAV > 1, the compound makes a significant contribution to the main aroma: OAV < 1 indicates that the compound does not contribute significantly thereto. The OAV values of the main volatile compounds in R. roxburghii wines are listed in Table 4. There were 21 and 23 compounds with OAV > 1 in pure W. anomalus C4 or in those mixed with S. cerevisiae X16 fermented R. roxburghii wine. In contrast, only 13 compounds had OAV > 1 in S. cerevisiae X16 solar-fermented R. roxburghii wine.

PCA was applied to analyse the chemical data pertaining to the wines obtained with each fermentation. PCA1 accounted for the 55.04% and PCA2 for an additional 44.96% of the variability (Figure 7): as seen in Figure 7A, complex odiferous compounds were clustered in the positive half-axis of PCA1, such as ethyl hexanoate, ethyl octanoate, ethyl cinnamate, isobutanol, linalool and 1-hexanol. 1-decanol and ethyl laurate, with fatty notes, were clustered in the negative half-axis of PCA1. The rose/floral odiferous compounds (phenylethyl alcohol, 1-octanol, and ethyl 9-decenoate) were the main components of the positive half-axis of PCA2. n-Decanoic acid and ethyl butyrate were distributed along the negative half-axis of PCA2.

Figure 7:

Principal component analysis of R. roxburghii wines fermentation with different yeasts. A, Principal component load plot of volatile aroma compounds. B, Principal component score of volatile aroma compounds.

This also indicated that the fruity odour (ethyl hexanoate, and ethyl octanoate) and the rose/floral odour (phenylethyl alcohol and 1-octanol) may be the main volatile characteristics of mixture fermented and pure fermented R. roxburghii wine made with W. anomalus C4, respectively (Figure 7B); however, it was difficult to identify the main volatile characteristics of S. cerevisiae X16 fermented R. roxburghii wine.