Comparison between traditional and industrial soy sauce (kecap) fermentation in Indonesia

JOURNALOF FERMENTATION ANDBIOENGINEERING Vol. 81, No. 3, 275-278. 1996

Comparison between Traditional and Industrial Soy Sauce (Kecap) Fermentation in Indonesia WILFRED

F. M. RGLING,’

ANTON

APRIYANTON0,2

AND HENK W. VAN VERSEVELD’*

Department of Microbiology, Biological Laboratory, Vrije Universiteit, de Boelelaan 1087, I081 HV Amsterdam, The Netherlands’ and Department of Food Technology and Human Nutrition, Bogor Agricultural University, Kampus IPB Darmaga, P. 0. Box 220, Bogor 16002, Indonesia= Received 10 July 1995IAccepted 21 November 1995

Growth of lactic acid bacteria and amino acid production at an Indonesian soy sauce manufacturer, employing modem Japanese process technology, indicated that brine fermentation for one month is sufficient for industrial kecap production. Compared to traditional Indonesian soy sauce fermentation, appjlication of modem Japanese process technology resulted in an obvious but not essential yeast fermentation. [Key words:

soy sauce,

Tetragenococcus halophila, kecap] isms were enumerated and identified as previously described (1). For specific enumeration of salt-tolerant lactic acid bacteria, MRS agar (special agar for enumeration of lactic acid bacteria, supplied by Oxoid) with 1,000 ppm cycloheximide and 15% NaCl was used. Samples for enzyme and product analysis were prepared by centrifugation, and the supernatant was stored at -40°C. Protease activity was measured with azo-albumin as substrate (6) and expressed as GDm ml-r h-r. Leucine aminopeptidase and glutaminase activity were determined as described by Tomita et al. (7) and expressed as pmol ml-l h-l. Acetate was measured by the colorimetric determination of acetyl phosphate (8), which was formed enzymatically from acetate (9). Lactate, ethanol and glutamic acid were measured enzymatically as described by Bergmeyer (10). Glycerol, glucose, fructose and galactose were determined enzymatically according to the manual of Boehringer Mannheim (11). Form01 nitrogen (a measurement of free amino acid content) was determined by form01 titration and expressed as g/l nitrogen. Three overlapping phases could be distinguished in brine fermentation: first, amino acid production (judged from form01 nitrogen production), followed by fermentation by lactic acid bacteria (judged from acetate and lactate production) and yeast fermentation (judged from ethanol and glycerol production). Figure 1 shows a schematic representation of the occurrence of these phases in each batch. Amino acid production Amino acid production started directly after the preparation of brine. In general, amino acids were produced during the first three weeks (Fig. 1). The amino acid mainly responsible for flavor, glutamic acid, is produced during a longer period. In batches B and C, glutamic acid content continued to increase slowly after 4 months, although less than 15% of the final glutamic acid cantent is formed during the last 3 months (Fig. 2A). The activity of glutaminase, the enzyme responsible for conversion of glutamine to glutamic acid, rapidly decreased but did not completely disappear (Fig. 2B). Even after form01 nitrogen production had stopped, protease and leucine aminopeptidase activity were present (Fig. 2). Therefore, the exhaustion of digestible proteins was more likely the cause of the termination of amino acid production.

In Indonesia, soy sauce (kecap) is usually made by small-scale producers in a traditional manner, with little or no innovation in the process since ancient times. Black soybeans are used as the only plant material. Boiled soybeans undergo spontaneous solid-state fermentation (SSF) before being subjected to brine fermentation. After the brine is filtered, the filtrate is boiled together with caramel and spices, yielding the final product: kecap (1). In the last two decades several industrial soy sauce manufacturers have been established in Indonesia. These manufacturers simply apply modern Japanese process technology (for a review, see ref. 2) for kecap production, without taking into account possible differences with traditional kecap production regarding microbial and biochemical changes during brine fermentation. Industrial manufacturers use defatted yellow soybean flakes and wheat instead of black soybeans only. SSF is well controlled and inoculated but brine fermentation is spontaneous and subjected to tropical weather conditions for 4 months. During brine fermentation in traditional kecap production, amino acid production and growth of the lactic acid bacterium Tetragenococcus halophila [until recently known as Pediococcus halophilus (3)] take place (1). No obvious yeast fermentation is observed (1, 4, 5). In Japan usually an obvious yeast fermentation is observed when wheat and soybeans are used as plant materials (2). Since yeast fermentation is not essential for Indonesian kecap production (1, 5) and microbial and biochemical changes during brine fermentation at industrial Indonesian manufacturers are unknown, we investigated the changes that take place at an industrial manufacturer. Results were compared with traditional kecap fermentation (1, 4) in order to optimize kecap production with modern technology. At Brine fermentation at industrial manufacturer the manufacturer, SSF takes 43 h and brine fermentation 4 months. About 8,000 kg SSF material is mixed with 16,0001 brine. Final salt concentration of the brine is 15%. In total 8 brines were investigated: batches A, B and C were monitored for 4 months, batches D, E and F for one month and batches G and H for 6 weeks. Brine samples were taken at 30cm depth. Microorgan* Corresponding author. 275

276

J. FERMENT. BIOBNG..

RGLING ET AL. Batch : A

0 C D

5 Z .9

8

- 0.8

s

: 5 .z

6 4

- 0.6 - 0.4

E z Lc

2

- 0.2

z0 -2 m 2

E F G (A)

Ii

I

I 0

20

40

60 Time

80 (d)

I 100

I 120

140

0

Batch :

!

I

1

0

20

40

0

20

4.0

1

,

Tin?(d)

80

I

I

100

120

100

120

0

A B C D E F G H 0

20

40

60 Time

:d:

100

120

140

Batch : Tit&’

(d)

2

-I

FIG. 2. Changes during brine fermentation (batch C). (A) Nitrogen containing products: 0, form01 nitrogen (g N/Z) and + , glutamic acid (g N/I); (B) enzyme activities: 0, glutaminase (,nmol ml-r h-r); q , protease (ODW ml-r h-r); +, leucine amino-peptidase (,umol ml-r h-r).

I

G H 0

I 20

I 40

0

I 60 Time

I 80 (d)

I 100

I 120

1 140

FIG. 1. Schematic representation of occurrence (shown in black). (A) Amino acid formation, (B) fermentation by lactic acid bacteria and (C) yeast fermentation in eight industrial brine fermentations. Arrow indicates that yeast fermentation was still in progress when sampling was terminated.

Fermentation by lactic acid bacteria

Numbers of bacteria such as staphylococci and Enterobacteriaceae, derived from SSF, decreased rapidly after addition of salt. Only Bacilli could be observed on agarplates without salt after one week (data not shown). Numbers of salt-tolerant bacteria were high at the start of fermentation (1O7-8 cfu/ml), but dropped rapidly within 2 d before increasing again. The high initial numbers of salt-tolerant bacteria were caused by staphylococci derived from SSF. After one week of brine fermentation, only the lactic acid bacterium T. halophila was isolated from samples, its maximum number reaching around 108 cfu/ml. Obvious fermentation started after 5 to 10 d and during growth of T. halophila, pH of the brine dropped from 5.0-5.1 to 4.4-4.6 and concentrations of lactate and acetate increased up to 164 mM and 69 mM, respectively (Table 1). Fructose completely disappeared. Lactic acid fermentation took about 2.5-3 weeks, then no further changes in lactate and acetate concentrations occurred and the number of T. halophila declined. salt-sensitive

Yeast fermentation is the most Yeast fermentation variable phase in brine fermentation (Fig. 1): in batch F yeast fermentation started already after 5 d while in batches G and H even after 42d no yeast fermentation could be observed. Immediately after the preparation of brine, high number of yeast was observed (lWs cfu/ml). During yeast fermentation a slight increase in yeast number was observed as well as increases in ethanol and glycerol concentrations. Zygosaccharomyces rouxii was the dominant yeast species. Glucose concentration dropped during the yeast fermentation, but galactose concentration remained unchanged. During brine fermentation heatColor formation dependent browning reactions (Maillard reactions) took place (2). As can be judged from the changes in color (OD&, Maillard reactions occurred during the entire period of fermentation (Fig. 3). Comparison between industrial and traditional kecap Despite the fact that the brine is not production inoculated, microbial and biochemical changes at the Indonesian industrial manufacturer resemble soy mash fermentation in Japan: first, amino acid production takes place, then fermentation by the lactic acid bacterium T. halophila followed by fermentation by the yeast 2. rouxii. Due to the tropical weather, both form01 nitrogen production and microbial growth are faster than those during soy sauce production in Japan, which are executed in a moderate climate (2, 4). The rates of amino acid production and fermentation by T. halophila are comparable with those of traditional kecap fermentation; in general, these phases are finished within 4 weeks in both

NOTES

VOL. 81, 1996 TABLE 1. Batch

Microbial and biochemical composition of brine after 4 weeks of fermentation A

B

Initial pH 5.4 4.9 4.65 4.4 End pH Max. log counts (cfu/ml): 6.4 7.8 Total, 15% NaCP MRS, 15% NaCP 6.2 7.8 6.4 6.8 Yeast, 15% NaCP

C

D

E

F

G

H

5.0 5.05 nm 4.45 4.55 4.6

nm 5.1 4.15 4.5

nm 4.5

7.8 7.8

7.2 7.1

8.1 8.1

7.9 7.9

7.7 7.7

6.7 6.2 6.4 Form01 nitrogen (g N//)7.3 6.9 7.0 7.1 6.4 Glutamic acid (g N/[) 1.16 0.67 0.70 0.71 0.62 Glucose (mM)b 284 213 215 235 202 Fructose (mM)b 14.5 15.4 12.9 12.2 10.7 82 82 77 77 13 Galactose (mM) Lactic acid (mM) I_0 143 138 151 137 Acetic acid (mM) 32 62 53 54 52 Ethanol (mM) 232 145 18i 120 130 154 127 139 133 127 Glycerol (mM)

8.0 8.0

6.4 5.9 5.5 6.2 6.4 7.3 0.64 0.71 0.90 174 283 271 10.0 13.8 9.2 83 81 79 42 164 163 29 68 66 242 2 46 172 &4 102

B Salt percentage used for plating. b Concentration before the occurrence of fermentation. Maximum values are indicated in bold, minimum values are in italics and underlined. traditional (1) and industrial brine fermentations (Fig. 1). In contrast to traditional production, in many cases an obvious yeast fermentation was observed, which lasted up to 3 months after the start of brine fermentation (Fig. 1). Heat-dependent browning reactions still occurred after 4 months (Fig. 3). Both yeast fermentation and color formation during brine fermentation are very important in Japanese soy sauce production (3). However, in Indonesian kecap production the filtrate brine undergoes several post-fermentation treatments, such as the addition of caramelized sugar and subsequent boiling for several hours, resulting in a strong brown color and evaporation of volatile compounds such as ethanol. Thus, good amino acid production and lactic acid fermentation seem to be sufficient for kecap production. The organic acids formed during the growth of T. halophila have a preserving effect on the final product. Amino acids contribute to the flavor of soy sauce, directly (glutamic acid) or indirectly (via Maillard reactions during boiling of the mixture of brine extract and caramel) (2). Since amino acid content and lactic acid concentration do not change much after 4 weeks, brine fermentation for one month seems to be sufficient for industrial kecap production. Variations in brine composition after one month of For good-quality production a consisfermentation tent composition of the brine extract is required. In traditional kecap production, large variations in final composition of batches from the same manufacturer were observed (1). Surprisingly, a similar observation (see Table 1, Fig. 1) was made for the industrial manufacturer, despite the use of modern process technology and standardized conditions. In particular large differences were found for products related to lactic acid and yeast fermentation. In some batches no obvious lactic acid fermentation (batches A and F) or yeast fermentation (batches G and H) occurred (Fig. 1). These differences might be related to the low initial pH (generally 4.9 to 5.1) of the freshly prepared mash. Initial numbers of T. halophila and Z. rouxii have a large influence on soy mash fermentation, especially at low pH (12, 13). Since the mash is not inoculated, numbers of yeast and T. halophiia still present from previous

277

2016-

01 0

20

40

00

100

120

TimG,O(cl) FIG. 3.

Changes in color during brine fermentation

(batch C).

batches or propagating during SSF will determine the growth of yeast and lactic acid bacteria in brine. SSF samples showed large variations in final numbers of yeasts and bacteria (data not shown). This may explain why in some cases yeast or lactic acid fermentation is lacking. Biochemical composition of soy mash is strongly influenced by microbial interactions such as ethanol production by yeast, which inhibits T. halophila growth (2), and acidification through acetate and lactate production by T. halophila, which negatively affects yeast growth (14), glutamic acid formation (15) and enzyme activities (2, 4). The low initial pH seemed to be related to the acidity of the salt and water added, since 43-h SSF samples had a pH of 6.3-6.4. A higher initial pH and inoculation with T. halophila of brine will contribute to a more consistent quality. In conclusion, the results show that blindly applying process technology, designed for Japanese soy sauce production, to Indonesian kecap production, without knowledge about the microbial and biochemical changes that take place, results in inefficiently made kecap of irregular composition. The authors would like to thank the Inter University Centre for Food and Nutrition, Bogor Agricultural University, Bogor, Indonesia, for providing space in their laboratory to do the research described here. REFERENCES 1. Riiling, W.F.M., Timotius, K.H., Prasetyo, A.B., Stouthamer, A. H., and van Verseveld, H. W.: Changes in microflora and biochemical composition during the baceman stage of traditional Indonesian kecup (soy sauce) production. J. Ferment. Bioeng., 77, 62-70 (1994). 2. Yokotsuka, T.: Soy sauce biochemistry. Adv. Food Res., 30, 195-328 (1986). 3. Anonymous: Validation of the publication of new names and new combinations previously effectively published outside the IJSB. List No. 49. Int. J. Syst. Bact., 44, 370-371 (1994). 4. Riiling, W. F. M., Timotius, K. H., Stouthamer, A. H., and van Verseveld, H. W.: Physical factors influencing microbial interactions and biochemical changes during the baceman stage of Indonesian kecap (soy sauce) production. J. Ferment. Bioeng., 77, 293-300 (1994). 5. Riiling, W. F. M., Timotius, K. H., Schuurmans, F. P., Stouthamer, A. H., and van Verseveld, H. W.: Influence of prebrining treatments on microbial and biochemical changes during the baceman stage in Indonesian kecup (say sauce) production. J. Ferment. Bioeng., 77, 400-406 (1994). 6. Frankena, J., van Verseveld, H. W., and Stouthamer, A. H.: A

278

7.

8.

9.

10. 11.

RGLING ET AL. continuous culture study of the bioenergetic aspects of growth and production of exocellular protease in Bacillus licheniformis. ADDS.Microbial. Biotechnol.. 22. 169-176 (1985). Tomit&-K., Yano, T., Kumagai, H.,.and To&k&, T.: Formation of r-glutamylglycylglycine by extracellular glutaminase of Aspergillus oryzae. J. Ferment. Technol., 66, 290-304 (1988). Lipmano, F. and Tuttle, L. C.: A specific micromethod for the determination of acyl-phosphates. .I. Biol. Chem., 159, 21-28 (1945). Rose, I. A., Gruaberg-Manago, M., Koney, S. C., and Ochoa, S.: Enzymatic phosphorylation of acetate. J. Biol. Chem., 211, 737-756 (1954). Bergmeyer, I-I. U.: Methoden der enzymatischer analyse, 2nd ed. Verlag Chemie GmbH, Weinhem (1980). Boehringer Mannheim: Methoden der enzymatischen Lebensmittel-analytik mit Einzelreagentien. Boehringer Mannheim GmbH Biochemica (1983).

J. FERMENT.BIO~NG., 12. Inamori, K., Miyauchi, K., Ucbida, K., and Yoshino, H.: Interaction between Pediococcus halophilus and Saccharomyces rouxii (Microorganisms involved in shoyu moromi fermentation. Part I). Nippon Nogeikagaku Kaishi, 58, 771-777 (1984). (in Japanese) 13. Onishi, H.: Studies on osmophilic yeasts. II. Factors affecting growth of soy yeasts and others in the environment of a high concentration of sodium chloride. Bull. Agr. Chem. Sot. Japan, 21, 143-150 (1957). 14. Noda, F., Hayasbi, K., and Mizunuma, T.: Antagonism between osmophilic lactic acid bacteria and yeasts in brine fermentation of soy sauce. Appl. Environ. Microbial., 40,452457 (1980). 15. Kuroshima, E., Oyama, Y., Matsuo, T., and Shugimori, T.: Biosynthesis and degradation of glutamic acid in microorganisms relating to the soy sauce brewing. III. Some factors affecting the glutamic acid and its related substances formation in soy sauce brewing. J. Ferment. Technol., 47, 693-700 (1969).