Original papers
Characterisation of Milk Clotting Properties of Latex from Japanese Milkweed (Metaplexis japonica)
2020 年 26 巻 2 号 p. 257-263
詳細
2020 年 26 巻 2 号 p. 257-263
The crude enzyme of latex from Japanese milkweed (Metaplexis japonica) clotted 10% reconstituted skim milk, and the curd was stable with no visible collapse. Additionally, milk-clotting properties of the crude enzyme were investigated. Multiple proteases were suggested to be involved in milk clotting, and cysteine proteases played a major role. The crude enzyme showed a higher milk clotting activity under acidic conditions and at 60 °C when a reconstituted low-heat skim milk ranging from pH 5.5 to 8.0 and from 25 to 60 °C was used as a substrate, respectively. Finally, the casein hydrolysis pattern of the crude enzyme showed a similar pattern to microbial rennet (degradation primarily of κ-casein).
Milk-clotting enzymes such as chymosin are essential for cheese making. Chymosin is a main milk-clotting enzyme in calf rennet, which is obtained from the fourth stomach of calves. Chymosin cleaves κ-casein at Phe105-Met106, and a glycomacropeptide is released from the casein micelle. Subsequently, the micelles are polymerised and precipitated, allowing for milk coagulation. A decrease in the supply of calf rennet was accompanied by an increase in worldwide cheese production (Cavalcanti et al., 2002). Recombinant DNA technology was used to develop microorganisms capable of producing chymosin in the 1980s. In recent years, recombinant chymosin is produced in large-scale and is widely used for cheese production. The chymosin is called fermentation-produced chymosin (FPC), and is produced by a fermentation process involving Aspergillus niger or Kluyveromyces lactis (Andren, 2010).
Some proteases are able to clot milk, and they named calf rennet, microbial rennet and plant rennet after their producers. At present, calf and microbial rennet are practically used for cheese making. Although microbial rennet produces some bitter peptides during cheese ripening, it has been widely used in cheese manufacturing because it is a demonstrated substitute for calf rennet (Jacob et al., 2010). Bitterness is a common problem in cheddar, gouda, and other ripened cheeses (McSweeney, 2011). It has been recognized for a long time that bitterness, which can limit cheese acceptability if too intense, is due to an excessive concentration of low-molecular weight, mainly hydrophobic peptides, which accumulate during ripening as a result of proteolysis (Le Quere, 2011). During cheese maturation, a bitter flavour defect can occur, associated mainly with the production (usually by the coagulant but with some contribution from the starter) of short bitter peptides containing predominantly hydrophobic amino acid residues (Broome et al., 2011). The use of plant rennet is limited because most plant enzymes have higher proteolytic activity and cause cheeses to have undesired textures and bitter flavours (Lo Piero et al., 2002; Anusha et al., 2014). Some plant rennets have been used for traditional cheeses. Proteases from cardoon flowers cleave κ-casein, similar to chymosin, and are used for several traditional Portuguese and Spanish cheeses (Chazarra et al., 2007). Further, papaya, pineapple, and fig are used for traditional cheese-like products in Indonesia (Hosono et al., 1983).
Japanese milkweed (Metaplexis japonica), “gagaimo” in Japanese, is a creeper plant widely distributed in East Asia, that produces latex. The plant appears in the oldest Japanese book about the myth ko-ji-ki, completed in 712. The seed is used as a tonic medicine in Chinese medicine, and people say that the latex is an antidote against snake and insect bites. Japanese milkweed was reported to be eaten as vegetable in Ohshika, Nagano in Japan (Matsushima et al., 2013). Although eating this plant as a vegetable is rare in other areas, Japanese milkweed are thought to be edible. In this paper, we prepared curd from milk using the latex as rennet. Next, we studied the milk-clotting properties of the latex crude enzyme to understand its optimal pH and temperature, protease type, and casein hydrolysis pattern.
Preparation of crude enzyme Latex of M. japonica was collected in Hiroshima and Aomori, Japan. The latex was diluted five-fold with 0.1 M citrate buffer (pH 5.5), and then the dilution was centrifuged at 4 350 ×g for 5 min. The supernatant was used as a crude enzyme. The crude enzyme was stored at −20 °C.
Trial cheese making We tried making cottage cheese to obtain basic information about cheesemaking using the crude enzyme. The cottage cheese was made according to Farkye (2004). The crude enzyme was diluted with 0.1 M citrate buffer (pH 5.5) to approximately 600 Soxhlet unit/mL (1:600). Soxhlet unit was defined as 1 mL of milk clotted by one-unit volume of enzyme for 40 min at 35 °C (Harboe et al., 2010). Strengths of the crude enzyme were measured as follows. Low-heat skim milk powder (Yotsuba Milk Products Co., Hokkaido, Japan) was reconstituted with 10-fold water. Ten mL of the reconstituted skim milk (RSM) and 0.1 mL of 5% calcium chloride solution were mixed. The mixture was warmed to 35 °C in a water bath, and 100 µL of the crude enzyme was added.
The 1.0 mL of the prepared dilution was added into 1 L pasteurised milk (heated at 65 °C for 30 min) (Kisuki Nyugyo Co., Shimane, Japan). Streptococcus thermophilus NBRC 13957 was used as a starter culture. One mL of subcultured strain 13957 was inoculated into 100 mL of 10% RSM, followed by incubation at 37 °C for 12 h. This starter culture was added to the 1 L pasteurised milk. This milk was held in a water bath at 40 °C for 6 h to reduce the pH below 5.2. Subsequently, the curd was cut into 20-mm sections. The temperature was increased gradually to 50 °C over 30 min to firm the curd. Whey was drained, and the curd was washed twice with water. The curd was milled with hand. Microbial rennet from Mucor miehei (Sigma-Aldrich, St. Louis, MO, USA) was used as a control. The rennet solution was prepared to have approximately 600 Soxhlet unit/mL, and it was substituted for the latex dilution in cheese making.
Determination of milk-clotting activity Milk-clotting time was measured in accordance with the IDF method (2007), and 10% RSM was used as a substrate. Milk-clotting time was determined by visual inspection of a clot on the tube wall by manually rotating the test tube. In this study, milk-clotting time was used as an indicator of milk-clotting activity (MCA) (i.e., the shorter the milk-clotting time, the higher the MCA).
Analysis of the types of protease present in the crude extract The effect of protease inhibitors on MCA of the crude enzyme was tested using pepstatin A, ethylenediaminetetraacetic acid (EDTA) and iodoacetamide (IAA), which are known inhibitors of aspartic proteases, metalloproteases and cysteine proteases, respectively. Briefly, 150 µM pepstatin A, 10 mM EDTA and 10 mM IAA solutions were prepared with 0.1 M citrate buffer (pH5.5). The pepstatin A, EDTA and IAA solutions were mixed with an equivalent volume of the crude enzyme, respectively. Final concentrations of inhibitors were 75 µM for pepstatin A and 5 mM for EDTA and IAA. The mixture was maintained for 60 min at room temperature, and then 100 µL of the mixture was added to 10 mL of 10% RSM at 35 °C. The control solution was prepared same as above without inhibitor. The Soxhelt unit of each mixture was calculated, and a residual activity was obtained as follows:
The effect of phenylmethylsulphonyl fluoride (PMSF) on MCA of the crude enzyme was tested. PMSF is an inhibitor of both serine proteases and cysteine proteases. This cysteine proteases inhibition is reversible with dithiothreitol (DTT) and the serine proteases inhibition is irreversible with DTT. To inhibit only serine proteases, 100 mM PMSF was prepared with methanol, and the PMSF solution was mixed with an equivalent volume of 100 mM DTT solution, 10-fold crude enzyme and eight-fold 0.1 M citrate buffer (pH5.5). The final concentration of the inhibitor was 5 mM. The mixture was maintained for 60 min at room temperature, and then 100 µL of the mixture was added to 10 mL of 10% RSM at 35 °C. The control solution was prepared using the same method as above without PMSF. The MCA of each mixture was compared with the control to calculate residual activity.
Effect of temperature, pH, and mineral on milk-clotting activity Ten mL of 10% RSM and 0.1 mL of 5% calcium chloride solution were mixed. This RSM was warmed to different temperatures (20–60 °C), and 200 µL of the crude enzyme was added to evaluate the effect of temperature on MCA.
The low-heat skim milk powder was reconstituted with 10-fold 0.01 M citric acid-phosphate buffer (pH 5.5–8.0) (Luo et al., 2018). Then 0.1 mL of 5% calcium chloride was added to 10 mL of the RSM. This RSM was held at 35 °C, and 200 µL of the crude enzyme was added to evaluate the effect of pH on MCA.
The low-heat skim milk powder was reconstituted with distilled water containing different concentrations of calcium chloride, sodium chloride, and magnesium chloride, and 200 µL of the crude enzyme was added to 10 mL of the RSM to evaluate the effect of minerals on MCA.
Casein hydrolysate analysis A casein hydrolysate analysis was performed by SDS-PAGE. A casein (Hayashi Pure Chemical Ind. Ltd., Osaka, Japan) suspension was prepared by dissolving 1.0% casein in 0.1 M acetic acid buffer (pH 5.5). Supernatant (5 µL) of various diluted latex solutions (1, 5, 10, 50, and 100-fold dilution) was added to 10 mL of the casein suspension. The reaction mixture was maintained at 50 °C for 30 min. Then, sample buffer was added, and the mixture was heated at 100 °C for 5 min. This sample solution was loaded to a Tricine gel (P-T16.5S) (ATTO Corp., Tokyo, Japan). Electrophoresis was performed at 175 V for 60 min. The gel was placed into a fixative solution (50% methanol and 20% acetic acid) and warmed in a microwave for 1 min. Subsequently, the gel was stained for 1 h in a staining solution (50% methanol, 10% acetic acid, and 0.25% Coomassie Brilliant Blue R-250), followed by destaining. Microbial rennet produced by M. miehei was used to compare its casein hydrolysis pattern against the crude enzyme. The strength of microbial rennet was adjusted to 7 500, 18 000, and 30 000 Soxhlet unit/mL. Casein without enzyme treatments was used as a control (Anusha et al., 2014).
Statistical analysis All assays were performed at least in triplicate. To identify the influence of protease inhibitors on the milk-clotting activity, results were subjected to the Student's t-test using Statview 5.0 software (SAS Institute, Cary, NC, USA). The significant difference (p < 0.01) was determined by comparing with the values of non-treated samples. To identify the differences in the milk-clotting time, one-way analysis of variance (ANOVA) was applied to the means, and significant differences between the means were determined by the Student-Newman-Keuls test (p < 0.01) applied using Statview 5.0 software.
Trial cheese making The latex of M. japonica had approximately 5400 Soxhlet unit/mL. RSM was coagulated by the crude enzyme (Fig. 1). The curd was stable and there was no visible collapse. The pH of the curd by the crude enzyme was 5.0 after a 6-h fermentation, and the value was the same as rennet. Further, the yield was 85±6 g and 112±33 g curd from 1 L milk when the crude enzyme and microbial rennet were used, respectively.
Illustration of curd prepared using M. japonica diluted latex (3000 Soxhlet unit/mL).
The curd yield was significantly low in case of the crude enzyme. Most vegetable coagulants have an excessive proteolytic nature, and this causes lower cheese yield (Shah et al., 2014). We think that the proteolytic activity of the latex was higher than that of the microbial rennet.
The pH of the curd dropped, similar to that of microbial rennet. This suggested that crude enzyme did not suppress the growth of S. thermophilus. Next, we investigated the MCA properties of the crude enzyme.
Types of protease present in the crude extract Four protease inhibitors were used to identify the types of protease in the crude enzyme (Table 1). There were significant decreases (p < 0.01) in milk-clotting activity when the latex crude enzyme was incubated with IAA and PMSF, respectively. IAA, which is a cysteine protease inhibitor, greatly decreased MCA to below 17% compared to the MCA of the untreated sample. PMSF, which is an inhibitor against serine proteases and cysteine proteases, decreased MCA to 58%. However, PMSF with DTT did not decrease MCA. The inhibitory activity against cysteine proteases is reversible with DTT, and the inhibitory activity against serine proteases is irreversible with DTT. Hence these results suggest that cysteine proteases are mainly involved in the MCA of the crude enzyme.
| Protease inhibitor | Residual activity (%)† |
|---|---|
| Non-treated | 100 ± 11 |
| Pepstatin A | 85 ± 15 |
| IAA | < 17* |
| EDTA | 107 ± 23 |
| PMSF | 58 ± 11* |
| PMSF with DTT | 114 ± 14 |
Most plant-derived proteases have been classified into cysteine and aspartic proteases (Sun et al., 2016). Cynarase, which is a milk-clotting protease in cardoon, is an aspartic protease (Heimgartner et al., 1990). Cynanchum otophyllum proteases have been suggested as cysteine proteases (Luo et al., 2018). We presume that at least one cysteine protease was involved in milk-clotting activity of M. japonica.
Influence of temperature and pH on milk-clotting activity The influence of temperature on MCA of the crude enzyme of latex (water-soluble fraction) was investigated (Fig. 2). We observed no milk-clotting activity with gummy precipitation. Prior to these experiments, we examined MCA stability of the crude enzyme with water, methanol, and 0.1 M citrate buffer (pH 5.5), and the activity was the most stable using the buffer (data not shown). Therefore, we used 0.1 M citrate buffer (pH 5.5) for the dilution. The crude enzyme exhibited a broad optimal temperature range between 40 °C and 60 °C for MCA.
Influence of temperature on milk-clotting time of crude enzyme of M. japonica latex. Data are presented as means ± standard deviation. Different letters indicate statistically significant differences (p < 0.01).
This result indicates that the proteases in the crude enzyme are highly resistant to heat, and this property agrees with previous reports. Sidrach et al. (2005) and Chazarra et al. (2007) reported that the protease of cardoon (Cynara scolymus L.) exhibited the highest activities at 60 °C and 70 °C, respectively. Luo et al. (2018) reported that proteases of C. otophyllum showed the highest MCA at 65 °C and 70 °C, and those maintained approximately 60% MCA at 80 °C. These plants are actually used for a cheese making as rennet in Spain and China, respectively. Heat-stable rennet is generally not suitable for long matured cheeses because protein degradation by residual rennet leads to a bitter taste.
The crude enzyme exhibited high MCA under acidic conditions at pH 5.5 and 6.0 (Fig. 3). This result agreed with previous reports, as cardoon and C. otophyllum also showed a higher MCA at pH 5.5 (Chazarra et al., 2007; Luo et al., 2018).
Influence of pH on milk-clotting time of crude enzyme of M. japonica latex. Data are presented as means ± standard deviation. Different letters indicate statistically significant differences (p < 0.01).
Influence of mineral on milk-clotting activity The influence of minerals on MCA was investigated with RSM supplemented with calcium chloride, sodium chloride, and magnesium chloride (Fig. 4). MCA increased significantly in a dose-dependent manner when calcium chloride or magnesium chloride was added, reaching a maximum at a concentration of 15 and 30 mM. MCA increased moderately when sodium chloride was added, and MCA at a concentration of 100 mM sodium chloride was less than at a concentration of 8 mM calcium chloride or magnesium chloride.
Influence of calcium ion, sodium ion, and magnesium ion on milk-clotting time of crude enzyme of M. japonica latex. Data are presented as means ± standard deviation. A: calcium ion, B: sodium ion, C: magnesium ion. Different letters indicate statistically significant differences (p < 0.01).
This result agrees with the report by Chazarra et al. (2007). They also found that 50 mM calcium ion was sufficient as a cardoon extract to coagulate milk. This agreement might be owed to calcium ion having no influence on the action of the enzyme. Calcium ion is thought to have an influence on milk coagulation (i.e., combination of paracasein), but not on rennet activity (i.e., hydrolysis of casein protein) (Sandra et al., 2012; Sandra and Corredig, 2013). In the same way, calcium ions may have no influence on milk-clotting enzymes of M. japonica and cardoon, and a concentration of 50 mM calcium chloride is probably sufficient for a combination of paracasein. Magnesium chloride supplementation showed similar results to calcium chloride supplementation. There remains a possibility that the bivalent cations are prosthetic groups of the milk-clotting enzyme.
Casein hydrolysis Casein hydrolysis pattern by the crude enzyme was assessed in Tricine-SDS-PAGE (Fig. 5). This analysis was done at 50 °C. This temperature is thought to be suitable for the crude enzyme to hydrolyse casein, and not for rennet from M. miehei because the microbial rennet is not thermostable (Preetha and Boopathy, 1997). Therefore, we used high Soxhlet unit solutions of microbial rennet in this analysis. Both casein hydrolysate patterns of microbial rennet and the supernatants of various diluted latex showed an absence of κ-casein and presence of αs1-, αs2- and β-casein. The casein hydrolysate band pattern was slightly different. Microbial rennet showed three casein hydrolysate bands at approximately 18, 16, and 14 kDa. The supernatants of above 50-fold diluted latex showed three casein hydrolysate bands at approximately 19, 15, and 14 kDa, and α-, β-, and κ-caseins were completely degraded by supernatants below 10-fold dilutions.
Casein hydrolysis pattern by microbial rennet and latex supernatant.
Lane 1: Low molecular marker (6.5, 14.3, 17, and 29 kDa)
Lane 2 – 4: microbial rennet 7500, 18000, and 30000 Soxhlet unit/mL
Lane 5 – 11: supernatant of diluted latex; 1, 5, 10, 50, 100, 75, and 50-fold, (approximately 7200, 1440, 720, 144, 72, 96, and 144 Soxhlet unit/mL, respectively)
Lane 12: casein without enzyme treatment
Previous studies about plant milk-clotting enzymes have shown that purified actinidin and purified proteases from Dregea sinensis Hemsl. and C. otophyllum Schneid. could completely degrade β- and κ-casein and partially degrade α-casein (Lo Piero et al., 2011; Zhang et al., 2015; Luo et al., 2018). Further, crude enzyme extracts of cardoon could completely degrade κ-casein and partially degrade α- and β-caseins ater a 60-min incubation (Chazarra et al., 2007). In that study, the authors extracted the crude enzyme from 15 g of dried cardoon stigmas by homogenisation in 50 mL buffer. In this study, we used latex itself without any processing, except for dilution in the buffer. Therefore, the supernatant might maintain the higher protease activities, such that α-, β-, and κ-caseins were degraded completely by the latex supernatant after a 30-min incubation. The casein hydrolysis profile of the diluted latex solution was similar to microbial rennet when the latex was diluted more than 50-fold.
The crude enzyme of latex from Japanese milkweed (Metaplexis japonica) clotted 10% RSM, and the curd was stable with no visible collapse. Further, the yield was approximately 85 g and 112 g from 1 L milk when using the crude enzyme and microbial rennet, respectively. Next, milk-clotting properties of the crude enzyme were investigated. Four protease inhibitors were used to investigate proteases involved in milk-clotting processes. Cysteine proteases were mainly involved in milk clotting of the crude enzyme. The crude enzyme showed a higher milk clotting activity under acidic conditions and at 60 °C when a reconstituted low-heat skim milk ranging from pH 5.5 to 8.0 and from 25 to 60 °C was used as a substrate, respectively. Finally, the casein hydrolysis pattern of the crude enzyme was assessed by Tricine-SDS-PAGE after 30 min of incubation, which showed a similar pattern to microbial rennet (degradation primarily of κ-casein). To our knowledge, this is the first report of curd preparation and milk-clotting properties of the crude enzyme from M. japonica latex. Japanese milkweed has been reported to be eaten parts of Japan, and the crude enzyme could be used for food processing. The tendency of the curd to fuse and the production of bitter taste in cheese after ripening are worthy of future study to evaluate the use of crude enzyme as a milk-clotting agent in rennet-curd cheese making.
Acknowledgements We would like to thank Prof. Kohei Irifune and Prof. Hiroyuki Koumura at Prefectural University of Hiroshima for their useful advice regarding plant collection.
