International Journal

of Food Microbiology
International Journal of
ELSEVIER Food Microbiology 29 (1996) 49-58

Microbiology of mew, a traditional fermented bamboo

shoot product
Jyoti P. Tamang *, Prabir K. Sarkar *
Microbiology Laboratory, Department of Botany, University of North Bengal, Siliguri-734 430, India
Received 21 September 1994; accepted 10 April 1995

Abstract

The use of mesu as a pickle and as the base of curry is a tradition in the Darjeeling hills
and Sikkim of India. A total of 327 strains of lactic acid bacteria, representing Lactobacillus
plantarum, L. brevis and Pediococcus pentosaceus were isolated from 30 samples of mesu.
These species were present in all samples of raw bamboo shoots tested. Mesu was
dominated by L. plantamm followed by L. breuis; P. pentosaceus was isolated less fre-
quently and recovered from only 40-50% of the mesu samples. The fermentation was
initiated by P. pentosaceus, followed by L. breuis, and finally succeeded by L. plantarum
species. During fermentation, the titratable acidity increased from 0.04 to 0.95%, resulting
in the decline in pH from 6.4 to 3.8.

Keywords: Mesu; Traditional fermented product; Pickled vegetable; Microbial succession;

Young bamboo shoot

1. Introduction

Mesu, a non-salted fermented bamboo shoot product, has traditionally been

consumed by the people in the bamboo-growing regions of the Darjeeling hills of
West Bengal and Sikkim in India (Tamang et al., 1988). History of the art of
pickling this vegetable is lost in antiquity. Mesu is prepared only during the months
of June to September when bamboo shoots sprout. In the traditional method of its
preparation, young edible shoots of bamboo (Dendrocalamus hamiltonii Nees and
Arnott, Bambusa tulda Roxb. and Dendrocalamus sikkimensis Gamble, locally

* Corresponding author. Tel.: 91-353-450411. Fax: 91-353-450546.

1Present address: Department of Botany, Sikkim Government College, Gangtok-737 102, India

0168-1605/96/$15,00 0 1996 Elsevier Science B.V. All rights reserved

SSDI 0168-1605(95)00021-6
50 J.P. Tamang P.K. Sarkar /ht. J. Food Microbiology 29 (1996) 49-58

known as “choya bans”, “karati bans” and “bhalu bans”, respectively) are
defoliated, chopped, pressed tightly into a hollow bamboo stem, covered tightly
with leaves of bamboo or other wild plants and left at ambient temperature
(20-25°C) to ferment for 7-15 days. Completion of fermentation is indicated by
the typical mesu flavour and taste. Mesu is commonly used as a pickle by mixing it
with salt, mustard oil and green chillies. It is also used for preparing curry by frying
and mixing it with cooked meat. While fresh mesu has a shelf-life of only about a
week, its pickle can be stored for a year or more.
To our knowledge, mesu has so far not been scientifically investigated. The
present investigation examined the succession of microflora during fermentation of
young bamboo shoots to produce mesu.

2. Materials and methods

2.1. Samples

Samples were collected aseptically in sterile glass bottles from different weekly
markets of Kalimpong in the Darjeeling hills and Gangtok in Sikkim. The bottles
were kept in an ice-box and transported immediately (within 24 h) to the labora-
tory for analyses. Mesu prepared in laboratory following the traditional method
was also analyzed.
In order to study the microbial succession, mesu was prepared in the laboratory
by fermenting pieces (l-l.5 cm x 0.3-0.7 cm x 0.3-0.7 cm) of young shoots of
bamboo (Dendrocafamus hamiltonii) in a glass jar at 30°C for 10 days. Quintuplet
bottles were removed as indicated for analyses.

2.2. Determination of pH and acidity

A 10 g sample was blended for 15 s with 20 ml deionized water in a homoge-

nizer (Remi, Bombay, India), and pH of the slurry was determined using a pH
meter (Type 335, Systronics, Naroda, India) calibrated with standard buffer solu-
tions (Merck, Bombay, India). Titratable acidity was calculated by titrating the
filtrate of the homogenate with 0.1 N sodium hydroxide to an end point of
phenolphthalein (0.1% w/v in 95% ethanol (AOAC, 1990).

2.3. Isolation of micro-organisms

A 10 g sample was blended in 90 ml sterile diluent (0.85% w/v sodium chloride

in water) by use of a Stomacher lab-blender 400 (Seward Medical, London, UK)
for 2 min at “normal” speed, and 1 ml of the appropriate decimal dilutions were
mixed with molten (45°C) medium and poured into plates. The nutrient agar
(M087, HiMedia, Bombay, India), de Man, Rogosa and Sharpe (MRS) agar
(HiMedia M641), all purpose Tween (APT) agar (HiMedia M226) and glucose-yeast
 J. P. Tamang, P. K. Sarkar / ht. J. Food Microbiology 29 (I 996) 49-58 51

extract-peptone [(GYP)-CaCO,] agar (Okada et al., 1986) plates were incubated at

30°C in a candle jar for 48-72 h; and yeast extract-malt extract (YM) agar
(HiMedia M424) and potato dextrose agar (PDA) (HiMedia M096), supplemented
with benzylpenicillin (10 III/ml) and streptomycin sulphate (12 pg/ml), plates
were incubated at 28°C for 48 h. The colonies appeared were counted as colony
forming units (cfu) per g wet weight of sample. The colonies were randomly picked
from selected plates (having 50-300 colonies per plate) to obtain representative
strains at regular intervals of fermentation time, and the isolates were purified by
successive subculturing in the corresponding broth and streaked onto the agar
surface. After microscopic examination, purified cultures were grown on slants of
the same medium and stored at 5°C.

2.4. Characterization of isolates

Colonies of young cultures were Gram stained (Bartholomew, 1962). General

morphology and catalase activity were determined by the methods described by
Norris et al. (1981). Cell dimension was measured with a standardized ocular
micrometer. Motility of the isolates was checked using a phase-contrast micro-
scope. Growth of the isolates in sodium chloride was observed in tubes of
tryptone-glucose-yeast extract (TGYE) broth (HiMedia M952) (5 ml/tube), con-
taining sodium chloride (4, 6.5 or 18% w/v), for 7 days at 30°C. The production of
acid and gas from sugar was observed using 10 ml sugar basal broth (Garvie, 1984)
containing 2% w/v sugar and inverted Durham tubes for 10 days at 30°C (Okada
et al., 1986). Production of indole was tested by inoculating the isolates in 10 ml
modified Davis and Mingioli’s broth (Davis and Mingioli, 19501, prepared by
replacing ammonium sulphate with L-tryptophan (0.1% w/v) and supplementing
with yeast extract (0.2% w/v>. Ehrlich BShme reagent (l-2 ml> was layered on 3,.5
and 7 days-old broth cultures and observed for the formation of red ring at the
culture-reagent interface (Iswaran, 1980). Hydrolysis of gelatin was tested by
streaking MRS agar-gelatin (1% w/v) plates, incubating for 3 and 5 days at 30°C
and then flooding the plates with 10 ml 1 N sulphuric acid saturated with
ammonium sulphate (Sneath and Collins, 1974). Arginine hydrolysis test medium
(Thornley, 1960) was inoculated by stabbing, and the method of Lelliott et al.
(1966) was followed to determine its hydrolysis. Hydrolysis of fat was tested by
streaking the surface-dried plates of tributyrin agar (HiMedia Ml571 and incubat-
ing for 4 days at 30°C. Slants of esculin hydrolysis test medium (g/l distilled water:
esculin, 1.0; FeCl,, 0.5; peptone, 5.0; yeast extract, 1.0; agar, 20.0) were inoculated
and incubated at 30°C for 7 days and observed for any blackening of the medium.
Reduction of nitrate was tested in nitrate broth (HiMedia M439) after 3, 7 and 14
days of incubation at 30°C using nitrate reduction test reagent (Norris et al., 1981).
Growth at different pH values was studied by adjusting the pH of MRS broth to
different levels using 1 N hydrochloric acid or 10% w/v sodium hydroxide. The
tubes were inoculated and incubated at 30°C for 24 h (Hesseltine and Ray, 1988).
Growth at different temperatures was tested by inoculating and incubating the
52 J.P. Tamang P.K. Sarkar / ht. J. Food Microbiology 29 (1996) 49-58

isolates in TGYE agar slants immersed in water bath for a maximum period of 21
days.

2.5. Statistical analysis

The data obtained were analyzed statistically by determining analysis of vari-

ance (Snedecor and Cochran, 1989).

3. Results

Pure culture colonies developed well on MRS agar, APT agar and GYP-CaCO,
agar, and poorly on nutrient agar. But, there was no growth on YM agar and
antibiotics-supplemented PDA. A total of 327 representative and dominating
strains of bacteria, isolated from 30 samples of mesu were grouped on the basis of
Gram reaction, catalase activity, cell shape, gas from glucose, ammonia from
arginine, and acid from a range of sugars (Table 1). Twelve representative strains,
one from each group, were selected randomly for the purpose of identification.
All the isolates were non-motile, non-sporeforming, microaerophilic, catalase-
negative and Gram-positive. The isolates having rod shaped cells were assigned to
the genus Lactobacih, whereas the spherical cells in tetrads belonged to the
genus Pediococcus, according to the taxonomic keys of Kandler and Weiss (1986)
and Garvie (1986), respectively. The detailed characteristics of the representative
strains of Lactobacillus are shown in Table 2. The strains KM-Rl, KM-R32,
GM-RI, LM-RI and LM-R14 were referred to as Lactobacillus pfantarum Orla-
Jensen, and the strains KM-SRl, GM-SRl and LM-SRl were referred to as L.
brevis Orla-Jensen.
The four representative strains of Pediococcus differed only in the production
of acid from D-xylose and lactose (Table 1). In every other respect tested, all those
strains were identical. They were referred to as Pediococcus pentosaceus.
The average microbial loads studied in 10 samples of mesu collected from each
of Kalimpong and Gangtok markets and laboratory-made as well as young shoots
of bamboo are presented in Table 3. All the three species, including L. plantarum,
L. brevis and P. pentosaceus were present in both the bamboo shoots and the
fermented product. In the raw material P. pentosaceus was the most predominant
species, whereas in mesu L. plantarum was the predominant organism. P. pen-
tosaceus was only detected in 40-50% of the samples tested.
Fig. 1 shows the changes in microflora in young bamboo shoots during their
fermentation to produce mesu. Pediococcus pentosaceus predominated in the early
stages of fermentation. Its initial load increased significantly (p < 0.05) after 2
days, followed by a steady fall in the count. On each successive day, the decrease
of this species was significant (p < 0.05). The load of L. brevis was maximum on
the fourth day of fermentation, after which there was significant (p < 0.05) fall in
Table 1
Characteristics of representative strains of bacteria isolated from mesu samples a
Source Number of Cell Gas from NH, from Acid produced from ’ Grouped strains Represen-
strains b shape glucose arginine (No. of isolates) tative
Ara Xyl Tre Man1 Sorl Cel Lac
isolated strains

Kalimpong market 109 Rod + _ + + + + + 52 KM-R1

Rod _ _ _ _ + + + + + 23 KM-R32
Rod + + + - _ _ _ _ _ 19 KM-SRl
coccus _ + +-+- - + - 15 KM-T1

Gangtok market 103 Rod _ _ + + + + + + + 49 GM-R1

Rod + + + + - - _ _ + 36 GM-SRl
coccus - + + + + - _ + + 18 GM-T1

Laboratory-made 115 Rod _ _ + + + + + 38 LM-Rl

Rod - - + + + + + + + 20 LM-R14
Rod + + + _ _ _ _ _ + 30 LM-SRl
coccus _ + + _ + - - + + 13 LM-Tl
coccus _ + + + + - _ + - 14 LM-T2

a Number of samples was 10 from each source.

b All the isolates were non-motile, non-sporeforming, Gram-positive and catalase-negative.
’ Ara, L-arabinose; Xyl, o-xylose; Tre, trehalose; Manl, mannitol; Sorl, sorbitol; Cel, cellobiose; Lac, lactose.

Y
54 J.P. Tamang, P.K Sarkar / ht. J. Food Microbiology 29 (1996) 49-58

Table 2
Characteristics a of representative strains of rod-shaped bacteria isolated from mesu
Parameters KM-R1 KM-R32 GM-R1 LM-Rl LM-R14 KM-SRl GM-SRI LM-SRl
Cell width (Km) 0.7-0.8 0.7-0.8 0.7-0.9 0.7-0.8 0.7-0.9 0.6-0.7 0.6-0.7 0.6-0.7
Cell length (pm) 2.0-4.5 2.0-4.7 2.0-5.0 1.5-4.5 2.0-5.0 2.0-3.0 2.0-3.5 2.0-3.0
Hydrolysis of
Arginine _ _ _ - _ + + +
Esculin + + + + + + + _
Growth at 45°C _ + + + _ _ _ +
Acid produced from
r_-Arabinose + _ + - + + + +
D-Xylose _ _ + - + _ + _
o-Galactose + + + + + + + _
Lactose + + + + + + +
Cellobiose + + + + + _ _ _
Sucrose + + + + + + _ _
Trehalose + + + + + _ _ _
Raffinose + - _ - + _ + _
Mannitol + + + + + _ _ _
Sorbitol + + + + + _ - _
Gas from glucose - _ _ - _ + + +

a Characters common to all strains: non-motile, microaerophilic, Gram-positive and catalase-negative

rods in chains of 2-4 cells; fat and gelatin not hydrolysed; growth occurred at 15°C; optimum
temperature of growth, 30°C; nitrate not reduced; indole not produced; acid produced from o-ribose,
D-glucose, D-fructose, maltose and melibiose; acid not produced from o-mannose and starch.

the count. On the other hand, the population of L. plantarum increased continu-
ously from the start of fermentation till the formation of mesu on the tenth day;
again, every successive day, the increase was significant (p < 0.05) till the sixth day.
The changes in pH and acidity are shown in Fig. 2. During fermentation, the

Table 3
Microbial load ’ of bamboo shoots and mesu from different sources
Micro-organisms x lo6 cfu/g wet weight

Bamboo shoot Mesu

Kalimpong market Gangtok market Laboratory-made
Lactobacillus plantarum 0.1 283.0 290.0 300.0
(0.006-0.2) (227.0-331.0) (234.0-395.0) (240.0-390.0)

Lactobacillus brevis 0.08 0.004 0.005 0.004

(0.005-0.2) (0.0008-0.008) (0.0008-0.008) (0.0008-0.009)

Pediococcus pentosaceus 0.2 0.00002 0.00003 0.00002

(0.007-0.4) ( < DL b-0.0OO04) ( < DL-0.00006) ( < DL-0.00004)

a Data represent the means of 10 samples. Ranges are given in parentheses

b Less than detection limit (10 cfu/g).
 J.P. Tamang, P.K Sarkar / ht. J. Food Microbiology 29 (1996) 49-58 55

0 2 4 6 8 IO

Fermentotlon J/me tdoys)

Fig. 1. Changes in microflora (0, Pediococcus pentosacew; 0, Lactobacilh brevis; A, Lactobacillus
pfantarum) in bamboo shoots during the production of mesu. Values are the means of five batches of
fermentation with ranges. Where ranges overlap, bars for the lowest mean are shown to the left; and
where three ranges overlap, bars for the highest mean are shown to the right.

mean pH value decreased from 6.4 at 0 day to 3.8 at the end. The titratable acidity
increased significantly (p < 0.05) on each day till the eighth day, after which the
rise was not significant (p < 0.05).

4. Discussion

Microbial analysis of young bamboo shoots demonstrated the presence of L.

plantarum, L. brevis and I? pentosaceus. This finding is in agreement with the
report of Garvie (1986) about the occurrence of pediococci and lactobacilli in
plants. While I? pentosacew was the most predominant micro-organism in the
56 J.P. Tamang, P.K. Sarkar/Int. J. Food Microbiology 29 (1996) 49-58

s
Q
5

d
3 1 I I I I I

0 2 4 6 8 IO

Fermentation fime /doysl

Fig. 2. Changes in pH (0) and titratable acidity (0) in bamboo shoots during the production of mew.
Values are the means of five batches of fermentation and bars indicate ranges.

substrate, L. pluntarum predominated the final product. It seems that this fermen-
tation is initiated by P. pentosaceus, but lastly dominated by L. plantantm.
Mesu is a naturally fermented product; no starter culture is added during its
preparation. All the three bacterial species occurred in all of the samples of raw
bamboo shoots. In the first two days of fermentation, P. pentosaceus predomi-
nated, reaching a maximum of about 10’ cfu/g wet weight. The load of heterofer-
mentative L. brevis reached its peak on the fourth day. Rapid growth, production
of significant (p < 0.05) amount of acid by these lactics and the possible produc-
tion of gas created an oxygen-depleting condition, suitable for the last successor,
L. plantarum. After reaching their peaks, the counts of P. pentosaceus and L.
breuzk declined rapidly till the end of fermentation when there was predominance
by L. plantarum. The increase in load of L. plantarum and the increase in acidity
in the fermenting substrate were almost parallel. At the time of completion of
fermentation, the total acidity was as high as 0.95% and had a pH value as low as
3.8.
This observation is in agreement with the report of Dhavises (1972). In naw-
 JP. Tamang, P.K Sarkar /Int. J. Food Microbiology 29 (1996) 49-58 57

mai-dong, a similar Thai product, P. cerevisiae predominates during the initial

stages of fermentation. In contrast to the succession of microflora in mesu, in
naw-mai-dong the order is Pediococcus, L. plantarum and L. brevis. In sauerkraut
fermentation it was observed that at a temperature above 18°C L. brevis and L.
plantarum predominated, and at a temperature of 32°C or above the fermentation
became essentially dominated by L. plantarum (Pederson and Albury, 1969).
During the production of mesu at 3o”C, the temperature of the fermenting mass
rose above 30°C and as such L. plantarum was finally selected.

Acknowledgements

This work was supported by grants from the University Grants Commission,
New Delhi and the Sikkim Science Society, Gangtok.

References

AOAC (1990) Official Methods of Analysis of the Association of Official Analytical Chemists, 15th edn,
K. Heldrich (editor), Association of Official Analytical Chemists, Virginia, pp. 780, 842.
Bartholomew, J.W. (1962) Variables influencing results, and the precise definition of the steps in Gram
staining as a means of standardizing the results obtained. Stain Technol. 37, 139-155.
Davis, B.D. and Mingioli, E.S. (1950) Mutants of Escherichia coli requiring methionine or vitamin B,,.
J. Bacterial. 60, 17-28.
Dhavises, G. (1972) Microbial studies during the pickling of the shoot of bamboo, Bumbusa urundi-
nuceu, Willd., and of pak Sian, Gynandropsis pentaphylla, D.C. MS. Thesis, Kasetsart University,
Bangkok.
Garvie, E.I. (1984) The separation of species of the genus Leuconostoc and the differentiation of the
leuconostocs from other lactic acid bacteria. In: T. Bergan (editor), Methods in Microbiology, Vol.
16. Academic Press, New York, pp. 147-178.
Garvie, E.I. (1986) Genus Pediococcus Claussen 1903, 68. In: P.H.A. Sneath, N.S. Mair, M.E. Sharpe
and J.G. Holt (editors), Bergey’s Manua! of Systematic Bacteriology, Vol. 2. Williams and Wilkins,
Baltimore, MD, pp. 1075-1079.
Hesseltine C.W. and Ray, M.L. (1988) Lactic acid bacteria in murcha and ragi. J. Appl. Bacterial. 64,
395-401.
Iswaran, V. (1980) A Treatise on Media and Methods Used in Bacteriological Techniques. Today and
Tomorrow’s Printers and Publishers, New Delhi.
Kandler, 0. and Weiss, N. (1986) Genus tactobucillus Beijerinck 1901, 212. In: P.H.A. Sneath, N.S.
Mair, M.E. Sharpe and J.G. Holt (editors), Bergey’s Manual of Systematic Bacteriology, Vol. 2,
Williams and Wilkins, Baltimore, MD, pp. 1209-1234.
Lelliott, R.A., Billing, E. and Hayward, A.C. (1966) A determinative scheme for the fluorescent plant
pathogenic pseudomonads. J. Appl. Bacterial. 29, 470-489.
Norris, J.R., Berkeley, R.C.W., Logan, N.A. and O’Donnell, A.G. (1981) The Genus Bacillus and
Sporoluctobucillus. In: M.P. Starr, H. Stolp, H.G. Trfper, A. Balows and H.G. Schlegel (editors),
The Prokaryotes: A Handbook on Habitats, Isolation and Identification of Bacteria, Vol. 2.
Springer-Verlag, New York, pp. 1711-1742.
Okada, S., Daengsubha, W., Uchimura, T., Ohara, N. and Kozaki, M. (1986) Flora of lactic acid
bacteria in Miang produced in northern Thailand. J. Gen. Appl. Microbial. 32, 57-65.
Pederson, C.S. and Albury, M.N. (1969) The sauerkraut fermentation. New York State Agric. Exp. Sta.
Bull. 824.
58 J.P. Tamang, P.K. Sarkar / Int. J. Food Microbiology 29 (1996) 49-58

Sneath, P.H.A. and Collins, V.G. (1974) A study in test reproducibility between laboratories: Report of
a Pseudomonas working group. Antonie van Leeuwenhoek 40,481-527.
Snedecor, G.W. and Cochran, W.G. (1989) Statistical Methods, 8th edn. Iowa State University Press,
Ames, pp. 38-63.
Tamang, J.P., Sarkar, P.K. and Hesseltine, C.W. (1988) Traditional fermented foods and beverages of
Darjeeling and Sikkim - a review. J. Sci. Food Agric. 44, 375-385.
Thornley, M.J. (1960) The differentiation of Pseudomonas from other Gram-negative bacteria on the
basis of arginine metabolism. J. Appl. Bacterial. 23, 37752.