- Email: [email protected]
Sapal: a traditional fermented taro [Colocasia esculenta (L.) Schott] corm and coconut cream mixture from Papua New Guinea
International of hod
International Food Microbiology
ELSEVIER
Journal
Microbiology
Journal of 28 (1996) 361-367
Sapal: a traditional fermented taro [ Colocasia esculenta (L.) Schott] c orm and coconut cream mixture from Papua New Guinea R. Gubag h Food
a, D.A. Omoloso
a, J.D. Owens ‘9*
aDepartment ofApplied Science, University of Technology, Lae, Papua New Guinea Microbial Interactions Laboratory, Department of Food Science and Technology, University of Reading, Reading RG6 2AP, UK Received
26 September
1994; accepted
17 January
1995
Abstract Sapal is a traditional fermented food made by mixing cooked, grated taro [Colocasia esculenta CL.1 Schott] corm with coconut cream and allowing it to ferment at ambient temperature. The fermentation was primarily due to heterofermentative lactic acid bacteria, which reached 10” cfu/ml. Seven out of 10 isolated bacterial strains were identified as Leuconostoc mesenteroides or Leuc. parumesenteroides. The initial microbial flora was derived from the coconut cream. Yeasts grew on the surface of the sapal in the later stages of the fermentation. Overnight storage of the grated taro corm resulted in the glucose concentration increasing from 1.1 to about 5 g/l. During the fermentation the glucose concentration decreased to undetectable levels. The pH value fell from an initial value of 6.1 to 4.1 after 24 h.
Keywords: Sapal; acid bacteria
Taro;
Coconut;
Fermentation;
Colocasia
esculenta;
Leuconostoc;
Lactic
1. Introduction Sapal is a traditional fermented food made on the northern coast of Papua New Guinea and nearby offshore islands. It is made by mixing cooked, grated taro [Colocasiu esculenta (L.) Schott] corm with coconut cream and allowing it to ferment
at ambient
* Corresponding
temperature.
It is also
made
using
author.
016%1605/96/$15.00 0 1996 Elsevier SSDZO168-1605(95)00012-7
Science
B.V. All rights reserved
other
starchy
substrates,
R. Gubug et al. / Im. J. Food Microbiology 28 (1996) 361-367
362
such as tapioca or cassava, instead of taro. Traditionally, sapal is made seasonally in large quantities for social occasions and the exchange of gifts with neighbours. A product similar to sapal, called porong or polom, is made all around the Huon Gulf and the Siassi Islands (off the North coast of Papua New Guinea) but in these areas the mixture of cooked taro corm and coconut cream is consumed without being fermented. A fermented tare/coconut cream product similar to sapal, except for being in the form of small balls rather than a large mass, is made in the Solomon Islands (Mollison, 1993). The aim of this study was to describe the traditional process for making sapal and to investigate the microbiological and chemical changes occurring during the fermentation.
2. Materials
and methods
2.1. Preparation
of sapal by the traditional
method
Taro corms, which had been harvested 1 or 2 days previously, were peeled. The peeled corms were speared onto pointed wooden sticks and shredded by rubbing them against a grater consisting of a piece of galvanised iron sheet having jagged protuberances made by driving nails through it. The shredded taro (approximately 20 kg) was wrapped in a dry palm leaf and allowed to stand overnight (approximately 15 h) at ambient temperature (26-29”(Z). In the morning the shredded taro was wrapped in pieces of banana leaf and tied with strings (peeled from stems of banana) to make small packages (each approximately 0.5-l kg). The packages were added to an iron caudron of boiling water and boiled until cooked (1.5 h). Meanwhile the flesh was scraped from mature coconuts. The cream was extracted by placing a small amount (approximately 0.5 kg) of shredded coconut in a sausage-shaped pouch made of polypropylene sacking and having a stick through each end. The contents of the pouch were squeezed as hard as possible by rotating the sticks in opposite directions. Liquid from the first extraction was collected onto fresh shredded coconut. This cream-enriched shredded coconut was then subjected to a second extraction and the expressed cream collected in a large pot. Some of the coconut cream was put in a large boat-shaped wooden bowl (total capacity approximately 50 I>. The taro packages were removed from the caudron, unwrapped and immediately added to the bowl. The gelatinous cakes of cooked taro were broken up and mixed with the coconut cream, using large wooden paddles, to make a thick porridge. More coconut cream was added as required to obtain the desired texture. In total, 4.8 kg of coconut cream was used. The temperature of the final mixture was 56°C. The bowl was covered with banana leaves, leaving some airspace above the sapal to allow for expansion during fermentation, and left at ambient temperature (26-33°C). The ratio (wet weights) of cooked taro corm to coconut cream in the sapal was, approximately, 5:l. The sapal had the texture of a thick gruel but, according to the makers, it should normally have a firmer, almost yelly-like, consistency. After 2.5 h of incubation the temperature in the centre of the mass was still at 53°C and at 44°C after 6.5 h. After approximately 4-5 h of
R. Gubag et al. /ht.
incubation considerable bowl and some material
J. Food Microbiology
28 (1996) 361-367
gas evolution occurred, some material overflowed was skimmed off to prevent further over-flowing.
363
the
2.2. Sampling Samples were collected of freshly shredded taro, shredded taro after overnight storage, cooked taro, freshly shredded coconut, shredded coconut after storage, freshly prepared sapal and sapal after different times of incubation. Samples were examined within 1 h of collection for pH value, glucose concentration and microflora. 2.3. Determination of microbiological populations Approximately 5 g of material was added to 5 ml of sterile distilled water in a new plastic bag. The bag was pummelled by hand and further dilutions were made as required in 0.1% (w/v) peptone water (Oxoid CM9). Lactic acid bacteria were enumerated on duplicate surface inoculated plates of MRS agar (Oxoid CM361) incubated in an atmosphere with 5% oxygen and 10% carbon dioxide (Oxoid Gas generating kit BR56 for microaerophilic organisms) at ambient temperature for 2 days. Yeasts and moulds were enumerated on duplicate surface inoculated plates of Dichloran-rose bengal-chloramphenicol agar (Oxoid CM727 with SR78) incubated at ambient temperature for 2 days. Counts of aerobic mesophilic bacteria were made on duplicate spread plates of Standard plate count agar (Oxoid CM463) incubated at ambient temperature for 1 day. Colonies, 2 mm or more in diameter, were counted. These were presumed to represent the aerobic microflora and to exclude lactic acid bacteria which produce colonies less than 2 mm in diameter after one day’s incubation. Coliform bacteria were enumerated on duplicate spread plates of Violet red bile agar (Oxoid CM107) incubated at ambient temperature for 1 or 2 days. 2.4. Determination of pH rakes,
moisture contents and glucose concentrations
Sapal was mixed with an equal volume of distilled water and the pH value measured with a glass electrode (Checker pH Tester, Hanna Instruments). Moisture contents were determined on samples dried at 105°C. Glucose concentrations were determined with glucose test strips (BM-Test l-44, Cat. no. 1179446; Boehringer Mannheim UK Ltd.) according to the manufacture’s instructions. Shredded taro and shredded coconut were first mixed with an equal volume of distilled water while sapal was tested undiluted. 2.5. Isolation and characterisation of lactic acid bacteria Representative broth and purified colonies.
colonies on APT agar plates were subcultured to tubes of APT by streaking on APT agar plates and sub-culturing from isolated
R. Gubag et al. /ht. .I. Food Microbiology 28 (1996) 361-367
364
Cell morphology and Gram’s reaction of cultures were examined on cultures grown on APT agar (Difco 06.54) at 30°C for 48 h. Production of carbon dioxide from glucose was determined using a modified APT medium containing (g/l): yeast extract (Difco 0127), 7.5; tryptone (Difco 01231, 12.5; glucose, 50; NaCl, 5; K,HPO,, 5; MgS0,.7H,O, 0.8; MnS0,.4H,O, 0.14; FeS0,.7H,O, 0.04; Tween 80, 0.2; agar (Oxoid Ll l), 2; pH, 6.3. The medium was dispensed in 10 ml amounts in 19 X 150 mm test-tubes and sterilised by autoclaving at 121°C for 15 min. Prior to inoculation the medium was heated to expel dissolved oxygen, cooled, and each tube was inoculated with 0.5 ml of culture in APT broth (Difco 0655). The surface was sealed with a 2-3 cm plug of sterile 1.5% water agar. Cultures were incubated at 30°C for up to 5 days and examined for the production of gas. Production of ammonia from arginine was examined by growing cultures in MRS broth (Oxoid CM359) containing L-arginine, 3 g/l, and testing for ammonia with Nessler’s reagent (Harrigan and McCance, 1976). Production of dextran from sucrose was determined on sucrose agar (Harrigan and McCance 1976). The ability to grow at 10, 15, and 45°C was examined in APT broth medium (Difco 0655-01). Cultures were tested for fermentation of carbohydrates in API 50CH galleries (API bioM&ieux) using API CHL medium (API 5041) according to the manufacturers instructions. Cupules were sealed with sterile liquid paraffin and cultures were incubated at 30°C for 7 days. Cultures were identified with APILAB database.
3. Results
and discussion
3. I. Characteristics
of the fermentation
The fermentation was primarily due to lactic acid bacteria, which reached populations of 10” cfu/ml (Fig. 1). Growth of lactic acid bacteria was accompanied by a decline in pH value from an initial value of 6.0 to a final value of 4.1. Concentrations of yeasts changed little during the first 24 h of the fermentation and then increased. After 4 days, yeasts were present as a whitish surface film on the sapal and it seems likely that they were primarily aerobic contaminants rather than directly involved in the fermentation. Aerobic mesophilic bacteria also reached very high concentrations after 7 days (Table l), indicating that conditions in the fermenting sapal were not inhibitory to the growth of many bacteria. Twenty lactic acid bacterial strain were isolated from different stages of the fermentation and 10 were identified to species by their sugar fermentation patterns. All of the lactic acid bacteria were heterofermentative and 7 of the 10 identified cultures were Leuconostoc mesenteroides or Leuc. paramesenteroides strains (Table 2). In spite of the high temperature (2 44°C) of the sapal during the initial 6 h of the fermentation, very few of the lactic acid bacteria isolated were able to grow at 45°C (Table 2).
R. Gubag et al. /ht.
J. Food Microbiology 28 (1996) 361-367
. 0
24
TI:
72
365
3
88
(h) Fig. 1. Changes in microbial concentrations, glucose concentration and pH value during the fermentation of sapal. 0, lactic acid bacteria; A, yeasts; n , glucose concentration; 0, pH value.
3.2. Sources of microorganisms Shredded taro corm supported a substantial growth of lactic acid bacteria, aerobic mesophiles and yeasts during overnight storage (Table 1). Cooking of the shredded taro prior to mixing with coconut cream can be presumed to have eliminated most of these. Hence, the initial flora of the sapal derived mainly from the coconut cream. The cooked taro corm was added to the coconut cream while it
Table 1 Microbial flora, pH value coconut cream, unfermented Process
and glucose concentrations sapal and sapal fermented
stage
Moisture
Glucose (g/l)
a
of shredded for 7 days pH value
(%)
1. 2. 3. 4. 5. 6. 7.
Taro root, freshly shredded Taro, shredded, stored 15 hC Taro, shredded, cooked Coconut, freshly shredded Coconut cream Sapal’, unfermented Sapal, fermented 7 days
34 34 nd nd 42 37 nd
1.1 4.0-6.1 3.1 < 0.4 0.4-0.7 3.1 < 0.2
a g/l of wet material. b No data. ’ Temperature during storage was 24 to 29°C. d A mixture of cooked, shredded taro and coconut
6.05 5.8 nd 6.4 6.3 6.0 4.1
cream.
taro
corm,
shredded
coconut,
Microbial concentration (log,, cfu/g wet wt) Lactic acid bacteria
Aerobic mesophiles
Yeasts
ndb > 7.7 nd 5.5 6.5 3.4 9.9
nd > 7.7 nd 4.1 7.3 2.8 8.2
nd 5.3 nd 4.4 6.0 nd nd
366
R. Gubag et al. / Int. .I. Food Microbiology
Table 2 Properties Strain
2 8 12 18 28 30 36 40 44 46
no.
of lactic acid bacteria Sourceh
(h)
0 0 2 2 6 6 24 24 48 4x
from fermenting Dextran
production
28 (1996) 361-367
sapal” Growth
at:
10°C
45°C
+ + + _
+ + +
_ _
+ + -
+ + _
+ + +
+ _ +
+ _ _ _ _ _
” All strains were Gram-positive, catalase-negative, produced CO,.from arginine and grew at 15 and 30°C. ’ Isolated from sapal incubated for the indicated time. ’ Identified from pattern of fermentation in API CHL50 system.
Identificationc
Lactobacillus brer,is Lactobacilbts brer.i.7 LactobaciNus confi(sus Leuc. paramesenteroide~ Leuc. mesenteroides Leuc. rnesenteroides Leuc. paramesenteroides I.euc. mesenteroides I.euc. mesenteroides Leuc. mewnteroides glucose,
produced
NH,
from
was still hot and, consequently, the temperature of the fresh sapal was sufficiently high (56°C) to reduce the initial microflora to about 10’ cfu/g. 3.3. Source qf fermentable
carbohydrate
Overnight storage of the shredded taro corm resulted in a substantial increase in the glucose concentration (Table 1) and it is clear that the taro corm contributed most of the glucose present in the sapal at the start of the fermentation. The glucose was consumed rapidly in the fermentation and was undetectable (< 0.2 g/l) within 24 h (Fig. 1). Since analyses were not done for other sugars, it is possible that other fermentable sugars were contributed by the taro corm and/or by the coconut cream. 3.4. Microbiological
safety
The unfermented shredded tare/coconut cream mixture is an oily gelatinous pudding with a pleasant sweet taste and it appears that the majority of Papua New Guineans prefer to consume the mixture in this unfermented state. One former resident of the Siassi Islands mentioned that the tradition there was to heat left-over mixture in order to prevent it going sour. Consumption of the soured product appears to be restricted to a relatively few areas. The cooking of the shredded taro corm will ensure that any food poisoning microorganisms present on the taro are killed and the initial temperature of the cooked taro and coconut cream mixture was sufficiently high to greatly reduce the
R. Gubag et al. / Int. J. Food Microbiology 28 (1996) 361-367
367
numbers of bacteria added with the coconut cream. Thus, the freshly made, unfermented mixture will generally be safe from risks of food poisoning. The quantity of glucose present was sufficient to provide an estimated 0.3% lactic acid. During the fermentation, the pH value was reduced only to a final value of 4.1, suggesting that insufficient fermentation acids were produced to overcome buffering by the substrate. Hence, the growth of a large population of aerobic, catalase positive bacteria was supported in the fermenting sapal, with a final population of 1.6 x 10’ cfu/ml. It is possible that these bacteria could include food poisoning types and it must be concluded that sapal made by traditional methods is liable to carry some risk of transmitting food poisoning bacteria.
Acknowledgements We thank the families and neighbours of Mr John Gubag and Mr Mathew Gubag, Bat Village, Karkar Island, Madang Province for providing the taro and coconuts and for making the sapal for us, and The Research Committee of The University of Technology, Lae, Papua New Guinea and The British Council for financial support.
References Harrigan, W.F. and McCance, M.E. (1976) Laboratory Methods in Food and Dairy Microbiology, 2nd edn. Academic Press, London, pp. 75, 321. Mollison, B. (1993) Ferment and Nutrition. Tagari Publications, Tylagum, Australia, p. 87.
International Food Microbiology
ELSEVIER
Journal
Microbiology
Journal of 28 (1996) 361-367
Sapal: a traditional fermented taro [ Colocasia esculenta (L.) Schott] c orm and coconut cream mixture from Papua New Guinea R. Gubag h Food
a, D.A. Omoloso
a, J.D. Owens ‘9*
aDepartment ofApplied Science, University of Technology, Lae, Papua New Guinea Microbial Interactions Laboratory, Department of Food Science and Technology, University of Reading, Reading RG6 2AP, UK Received
26 September
1994; accepted
17 January
1995
Abstract Sapal is a traditional fermented food made by mixing cooked, grated taro [Colocasia esculenta CL.1 Schott] corm with coconut cream and allowing it to ferment at ambient temperature. The fermentation was primarily due to heterofermentative lactic acid bacteria, which reached 10” cfu/ml. Seven out of 10 isolated bacterial strains were identified as Leuconostoc mesenteroides or Leuc. parumesenteroides. The initial microbial flora was derived from the coconut cream. Yeasts grew on the surface of the sapal in the later stages of the fermentation. Overnight storage of the grated taro corm resulted in the glucose concentration increasing from 1.1 to about 5 g/l. During the fermentation the glucose concentration decreased to undetectable levels. The pH value fell from an initial value of 6.1 to 4.1 after 24 h.
Keywords: Sapal; acid bacteria
Taro;
Coconut;
Fermentation;
Colocasia
esculenta;
Leuconostoc;
Lactic
1. Introduction Sapal is a traditional fermented food made on the northern coast of Papua New Guinea and nearby offshore islands. It is made by mixing cooked, grated taro [Colocasiu esculenta (L.) Schott] corm with coconut cream and allowing it to ferment
at ambient
* Corresponding
temperature.
It is also
made
using
author.
016%1605/96/$15.00 0 1996 Elsevier SSDZO168-1605(95)00012-7
Science
B.V. All rights reserved
other
starchy
substrates,
R. Gubug et al. / Im. J. Food Microbiology 28 (1996) 361-367
362
such as tapioca or cassava, instead of taro. Traditionally, sapal is made seasonally in large quantities for social occasions and the exchange of gifts with neighbours. A product similar to sapal, called porong or polom, is made all around the Huon Gulf and the Siassi Islands (off the North coast of Papua New Guinea) but in these areas the mixture of cooked taro corm and coconut cream is consumed without being fermented. A fermented tare/coconut cream product similar to sapal, except for being in the form of small balls rather than a large mass, is made in the Solomon Islands (Mollison, 1993). The aim of this study was to describe the traditional process for making sapal and to investigate the microbiological and chemical changes occurring during the fermentation.
2. Materials
and methods
2.1. Preparation
of sapal by the traditional
method
Taro corms, which had been harvested 1 or 2 days previously, were peeled. The peeled corms were speared onto pointed wooden sticks and shredded by rubbing them against a grater consisting of a piece of galvanised iron sheet having jagged protuberances made by driving nails through it. The shredded taro (approximately 20 kg) was wrapped in a dry palm leaf and allowed to stand overnight (approximately 15 h) at ambient temperature (26-29”(Z). In the morning the shredded taro was wrapped in pieces of banana leaf and tied with strings (peeled from stems of banana) to make small packages (each approximately 0.5-l kg). The packages were added to an iron caudron of boiling water and boiled until cooked (1.5 h). Meanwhile the flesh was scraped from mature coconuts. The cream was extracted by placing a small amount (approximately 0.5 kg) of shredded coconut in a sausage-shaped pouch made of polypropylene sacking and having a stick through each end. The contents of the pouch were squeezed as hard as possible by rotating the sticks in opposite directions. Liquid from the first extraction was collected onto fresh shredded coconut. This cream-enriched shredded coconut was then subjected to a second extraction and the expressed cream collected in a large pot. Some of the coconut cream was put in a large boat-shaped wooden bowl (total capacity approximately 50 I>. The taro packages were removed from the caudron, unwrapped and immediately added to the bowl. The gelatinous cakes of cooked taro were broken up and mixed with the coconut cream, using large wooden paddles, to make a thick porridge. More coconut cream was added as required to obtain the desired texture. In total, 4.8 kg of coconut cream was used. The temperature of the final mixture was 56°C. The bowl was covered with banana leaves, leaving some airspace above the sapal to allow for expansion during fermentation, and left at ambient temperature (26-33°C). The ratio (wet weights) of cooked taro corm to coconut cream in the sapal was, approximately, 5:l. The sapal had the texture of a thick gruel but, according to the makers, it should normally have a firmer, almost yelly-like, consistency. After 2.5 h of incubation the temperature in the centre of the mass was still at 53°C and at 44°C after 6.5 h. After approximately 4-5 h of
R. Gubag et al. /ht.
incubation considerable bowl and some material
J. Food Microbiology
28 (1996) 361-367
gas evolution occurred, some material overflowed was skimmed off to prevent further over-flowing.
363
the
2.2. Sampling Samples were collected of freshly shredded taro, shredded taro after overnight storage, cooked taro, freshly shredded coconut, shredded coconut after storage, freshly prepared sapal and sapal after different times of incubation. Samples were examined within 1 h of collection for pH value, glucose concentration and microflora. 2.3. Determination of microbiological populations Approximately 5 g of material was added to 5 ml of sterile distilled water in a new plastic bag. The bag was pummelled by hand and further dilutions were made as required in 0.1% (w/v) peptone water (Oxoid CM9). Lactic acid bacteria were enumerated on duplicate surface inoculated plates of MRS agar (Oxoid CM361) incubated in an atmosphere with 5% oxygen and 10% carbon dioxide (Oxoid Gas generating kit BR56 for microaerophilic organisms) at ambient temperature for 2 days. Yeasts and moulds were enumerated on duplicate surface inoculated plates of Dichloran-rose bengal-chloramphenicol agar (Oxoid CM727 with SR78) incubated at ambient temperature for 2 days. Counts of aerobic mesophilic bacteria were made on duplicate spread plates of Standard plate count agar (Oxoid CM463) incubated at ambient temperature for 1 day. Colonies, 2 mm or more in diameter, were counted. These were presumed to represent the aerobic microflora and to exclude lactic acid bacteria which produce colonies less than 2 mm in diameter after one day’s incubation. Coliform bacteria were enumerated on duplicate spread plates of Violet red bile agar (Oxoid CM107) incubated at ambient temperature for 1 or 2 days. 2.4. Determination of pH rakes,
moisture contents and glucose concentrations
Sapal was mixed with an equal volume of distilled water and the pH value measured with a glass electrode (Checker pH Tester, Hanna Instruments). Moisture contents were determined on samples dried at 105°C. Glucose concentrations were determined with glucose test strips (BM-Test l-44, Cat. no. 1179446; Boehringer Mannheim UK Ltd.) according to the manufacture’s instructions. Shredded taro and shredded coconut were first mixed with an equal volume of distilled water while sapal was tested undiluted. 2.5. Isolation and characterisation of lactic acid bacteria Representative broth and purified colonies.
colonies on APT agar plates were subcultured to tubes of APT by streaking on APT agar plates and sub-culturing from isolated
R. Gubag et al. /ht. .I. Food Microbiology 28 (1996) 361-367
364
Cell morphology and Gram’s reaction of cultures were examined on cultures grown on APT agar (Difco 06.54) at 30°C for 48 h. Production of carbon dioxide from glucose was determined using a modified APT medium containing (g/l): yeast extract (Difco 0127), 7.5; tryptone (Difco 01231, 12.5; glucose, 50; NaCl, 5; K,HPO,, 5; MgS0,.7H,O, 0.8; MnS0,.4H,O, 0.14; FeS0,.7H,O, 0.04; Tween 80, 0.2; agar (Oxoid Ll l), 2; pH, 6.3. The medium was dispensed in 10 ml amounts in 19 X 150 mm test-tubes and sterilised by autoclaving at 121°C for 15 min. Prior to inoculation the medium was heated to expel dissolved oxygen, cooled, and each tube was inoculated with 0.5 ml of culture in APT broth (Difco 0655). The surface was sealed with a 2-3 cm plug of sterile 1.5% water agar. Cultures were incubated at 30°C for up to 5 days and examined for the production of gas. Production of ammonia from arginine was examined by growing cultures in MRS broth (Oxoid CM359) containing L-arginine, 3 g/l, and testing for ammonia with Nessler’s reagent (Harrigan and McCance, 1976). Production of dextran from sucrose was determined on sucrose agar (Harrigan and McCance 1976). The ability to grow at 10, 15, and 45°C was examined in APT broth medium (Difco 0655-01). Cultures were tested for fermentation of carbohydrates in API 50CH galleries (API bioM&ieux) using API CHL medium (API 5041) according to the manufacturers instructions. Cupules were sealed with sterile liquid paraffin and cultures were incubated at 30°C for 7 days. Cultures were identified with APILAB database.
3. Results
and discussion
3. I. Characteristics
of the fermentation
The fermentation was primarily due to lactic acid bacteria, which reached populations of 10” cfu/ml (Fig. 1). Growth of lactic acid bacteria was accompanied by a decline in pH value from an initial value of 6.0 to a final value of 4.1. Concentrations of yeasts changed little during the first 24 h of the fermentation and then increased. After 4 days, yeasts were present as a whitish surface film on the sapal and it seems likely that they were primarily aerobic contaminants rather than directly involved in the fermentation. Aerobic mesophilic bacteria also reached very high concentrations after 7 days (Table l), indicating that conditions in the fermenting sapal were not inhibitory to the growth of many bacteria. Twenty lactic acid bacterial strain were isolated from different stages of the fermentation and 10 were identified to species by their sugar fermentation patterns. All of the lactic acid bacteria were heterofermentative and 7 of the 10 identified cultures were Leuconostoc mesenteroides or Leuc. paramesenteroides strains (Table 2). In spite of the high temperature (2 44°C) of the sapal during the initial 6 h of the fermentation, very few of the lactic acid bacteria isolated were able to grow at 45°C (Table 2).
R. Gubag et al. /ht.
J. Food Microbiology 28 (1996) 361-367
. 0
24
TI:
72
365
3
88
(h) Fig. 1. Changes in microbial concentrations, glucose concentration and pH value during the fermentation of sapal. 0, lactic acid bacteria; A, yeasts; n , glucose concentration; 0, pH value.
3.2. Sources of microorganisms Shredded taro corm supported a substantial growth of lactic acid bacteria, aerobic mesophiles and yeasts during overnight storage (Table 1). Cooking of the shredded taro prior to mixing with coconut cream can be presumed to have eliminated most of these. Hence, the initial flora of the sapal derived mainly from the coconut cream. The cooked taro corm was added to the coconut cream while it
Table 1 Microbial flora, pH value coconut cream, unfermented Process
and glucose concentrations sapal and sapal fermented
stage
Moisture
Glucose (g/l)
a
of shredded for 7 days pH value
(%)
1. 2. 3. 4. 5. 6. 7.
Taro root, freshly shredded Taro, shredded, stored 15 hC Taro, shredded, cooked Coconut, freshly shredded Coconut cream Sapal’, unfermented Sapal, fermented 7 days
34 34 nd nd 42 37 nd
1.1 4.0-6.1 3.1 < 0.4 0.4-0.7 3.1 < 0.2
a g/l of wet material. b No data. ’ Temperature during storage was 24 to 29°C. d A mixture of cooked, shredded taro and coconut
6.05 5.8 nd 6.4 6.3 6.0 4.1
cream.
taro
corm,
shredded
coconut,
Microbial concentration (log,, cfu/g wet wt) Lactic acid bacteria
Aerobic mesophiles
Yeasts
ndb > 7.7 nd 5.5 6.5 3.4 9.9
nd > 7.7 nd 4.1 7.3 2.8 8.2
nd 5.3 nd 4.4 6.0 nd nd
366
R. Gubag et al. / Int. .I. Food Microbiology
Table 2 Properties Strain
2 8 12 18 28 30 36 40 44 46
no.
of lactic acid bacteria Sourceh
(h)
0 0 2 2 6 6 24 24 48 4x
from fermenting Dextran
production
28 (1996) 361-367
sapal” Growth
at:
10°C
45°C
+ + + _
+ + +
_ _
+ + -
+ + _
+ + +
+ _ +
+ _ _ _ _ _
” All strains were Gram-positive, catalase-negative, produced CO,.from arginine and grew at 15 and 30°C. ’ Isolated from sapal incubated for the indicated time. ’ Identified from pattern of fermentation in API CHL50 system.
Identificationc
Lactobacillus brer,is Lactobacilbts brer.i.7 LactobaciNus confi(sus Leuc. paramesenteroide~ Leuc. mesenteroides Leuc. rnesenteroides Leuc. paramesenteroides I.euc. mesenteroides I.euc. mesenteroides Leuc. mewnteroides glucose,
produced
NH,
from
was still hot and, consequently, the temperature of the fresh sapal was sufficiently high (56°C) to reduce the initial microflora to about 10’ cfu/g. 3.3. Source qf fermentable
carbohydrate
Overnight storage of the shredded taro corm resulted in a substantial increase in the glucose concentration (Table 1) and it is clear that the taro corm contributed most of the glucose present in the sapal at the start of the fermentation. The glucose was consumed rapidly in the fermentation and was undetectable (< 0.2 g/l) within 24 h (Fig. 1). Since analyses were not done for other sugars, it is possible that other fermentable sugars were contributed by the taro corm and/or by the coconut cream. 3.4. Microbiological
safety
The unfermented shredded tare/coconut cream mixture is an oily gelatinous pudding with a pleasant sweet taste and it appears that the majority of Papua New Guineans prefer to consume the mixture in this unfermented state. One former resident of the Siassi Islands mentioned that the tradition there was to heat left-over mixture in order to prevent it going sour. Consumption of the soured product appears to be restricted to a relatively few areas. The cooking of the shredded taro corm will ensure that any food poisoning microorganisms present on the taro are killed and the initial temperature of the cooked taro and coconut cream mixture was sufficiently high to greatly reduce the
R. Gubag et al. / Int. J. Food Microbiology 28 (1996) 361-367
367
numbers of bacteria added with the coconut cream. Thus, the freshly made, unfermented mixture will generally be safe from risks of food poisoning. The quantity of glucose present was sufficient to provide an estimated 0.3% lactic acid. During the fermentation, the pH value was reduced only to a final value of 4.1, suggesting that insufficient fermentation acids were produced to overcome buffering by the substrate. Hence, the growth of a large population of aerobic, catalase positive bacteria was supported in the fermenting sapal, with a final population of 1.6 x 10’ cfu/ml. It is possible that these bacteria could include food poisoning types and it must be concluded that sapal made by traditional methods is liable to carry some risk of transmitting food poisoning bacteria.
Acknowledgements We thank the families and neighbours of Mr John Gubag and Mr Mathew Gubag, Bat Village, Karkar Island, Madang Province for providing the taro and coconuts and for making the sapal for us, and The Research Committee of The University of Technology, Lae, Papua New Guinea and The British Council for financial support.
References Harrigan, W.F. and McCance, M.E. (1976) Laboratory Methods in Food and Dairy Microbiology, 2nd edn. Academic Press, London, pp. 75, 321. Mollison, B. (1993) Ferment and Nutrition. Tagari Publications, Tylagum, Australia, p. 87.