Microbes Environ. Vol. 23, No.

4, 350352, 2008

http://wwwsoc.nii.ac.jp/jsme2/ doi:10.1264/jsme2.ME08105

Short Communication
Isolation and Characterization of Aerobic Thermophilic Bacteria from the
Savusavu Hot Springs in Fiji
VINAY V. NARAYAN1, MOHAMED A. HATHA1,2, HUGH W. MORGAN3, and DHANA RAO1*
1Division

of Biological Sciences, Faculty of Science and Technology, University of the South Pacific, Fiji;
of Marine Biology, Microbiology & Biochemistry, School of Marine Sciences, Cochin University
of Science & Technology, Cochin682 016 Kerala, India; and 3Division of Biological Sciences, Thermophile
Research Unit, University of Waikato, Hamilton, New Zealand
2Department

(Received March 25, 2008Accepted July 28, 2008Published online October 1, 2008)

The relative isolation and unique physical properties of the Savusavu Hot Springs in Fiji may yield unique thermophiles. This study was conducted to determine the presence of aerobic thermophilic bacteria in these hot springs. A
total of 104 thermophilic bacterial isolates were characterized and using Thermus and Bacillus strains as controls,
58% of the isolates were identified as Anoxybacillus flavithermus, 19% as Geobacillus stearothermophilus/Bacillus
licheniformis, 10% as Thermus sp. TG153 and 10% as Thermus sp. TG206. Four isolates were unique in their molecular patterns suggesting there may be novel bacteria in the Savusavu hot springs.
Key words: Hot Springs, thermophilic bacteria, novel bacteria, hydrolytic enzymes, 16S rDNA

The Savusavu Hot Springs are the most extensive and

active hot springs in Fiji. There are inter-tidal hot springs
scattered for about a kilometer along the north coast of the
Savusavu peninsula towards the eastern bay. There are also
major hot springs located on land, approximately 165 meters
from the shore in a hollow depression1,4). A significant feature of these hot springs is that they show both daily and seasonal variations in activity. Thus, the microbial composition
of the springs may also show significant variation.
Not much is known of the microorganisms present in these
hot springs. The only recorded work is that conducted by
Hudson6) who isolated a Thermus strain from the Savusavu
beach (Fiji3) and compared it with various isolates from New
Zealand and Icelandic hot pools7). Hudson found that the
isolate from Fiji did not fall into any of the three major clusters in their results and so concluded that different thermal
regions have different insular phenotypes.
Fiji is an isolated island nation, the formation of the
islands and its hot springs resulting from volcanic activity,
without interference from any nearby continent or major
island. Therefore, the thermophiles that are present in these
hot springs could have unique evolutionary lineages that are
not found in other parts of the world. Such hot springs have
often been regarded as the equivalent of island ecosystems in
that their relative isolation and unique physical properties
can support novelty and biodiversity in the colonizing flora.
In the case of the Savusavu springs, this feature is further
enhanced by the closeness of the tidal flow, the daily variation in activity, and in some instances, the diurnal flooding of
the springs with seawater. This introduces unique constraints
on bacterial development and makes these springs particularly interesting environments for study, as they may yield
some novel thermophiles. This is the first detailed micro* Corresponding author. E-mail: rao_dl@usp.ac.fj; Tel: +679323
3550; Fax: +6793231512.

biological study of the Savusavu Hot Springs, which aims to

characterize thermophilic biodiversity using culture-dependent techniques.
Combined sediment and water samples were collected
from the Savusavu Hot Springs in sterile wide-mouthed glass
bottles and transported to the laboratory, where they were
maintained in water baths at a temperature of 65C until further analysis later on the same day. In situ measurements of
temperature and pH were taken to create a profile of the hot
spring. The hot springs were located in Nakama, Savusavu,
Vanua Levu, Fiji (1644'48''S, 17921'30''E). Microorganisms were isolated by direct streaking on Nutrient agar
plates12) at pH 7 and incubating at 65C for 24 hours. Individual colonies were picked and pure cultures obtained for biochemical characterization.
The bacteriological and biochemical characterization were
conducted according to the methods of Nair9). These were
total plate counts, temperature tolerance test, Gram staining,
endospore staining, motility tests, Kovacs oxidase, oxidation/fermentation and catalase tests. Extracellular and hydrolytic enzyme (amylase, lipase and gelatinase) production was
screened by streaking pure thermophilic bacterial cultures
onto nutrient starch agar (NSA)3), nutrient gelatine agar
(NGA)13) and nutrient lipid agar (NLA)14) plates, and observing zones of clearing.
DNA extraction was performed using the CTAB chloroform/isoamyl alcohol method2), and the concentration was
determined using a Nanodrop ND 1000 Spectrophotometer
(Nanodrop Technologies, USA). DNA was used as the template in PCRs with primers EubA and EubB11) designed to
amplify the eubacterial 16S rRNA gene. PCR products were
digested with EcoRI and HaeIII and analyzed. Bacterial cultures used as positive controls were of thermophiles from
New Zealand hot pools kept at the Thermophile Research
Unit (University of Waikato, New Zealand) and included
the following: Anoxybacillus flavithermus (DSM 2641),

Thermophilic Bacteria in Fiji Hot Springs

351

Geobacillus stereothermophilus strain (ATCC 12980),

Bacillus licheniformis (DSM 8785), Thermus sp. strain
Rt41.A (NCBI taxon identifier 32063), Thermus sp. strain
TG 275 (TRU HS5.A1), Thermus sp. strain TG 206 (TRU
ZV1D.A2), Thermus sp. strain TG 153 (TRU HWB.A1) and
Thermus sp. strain TG351 (ATCC 31674).
The pH of the Savusavu Hot Springs was within a range of
6.57.5. The temperature varied according to the activity of
the springs, ranging from 66 to 102C.
Table 1 shows that thermophile numbers (determined by
total plate counts: aerobic incubation at 75C using nutrient
agar medium12) for 48 hours) in the Savusavu Hot Springs
vary significantly with the different seasons. During September to November, the springs have a minimum population
level. Also, during this time, there is a lot of rainfall, which
could result in the dilution and washing off of thermophiles.
This is reflected in the low number of thermophiles as compared to January and March, when the springs are at peak
activity/water flow and there is little rainfall.
The lowest growth temperature for the isolates obtained
from the Hot Springs was 45C and growth was evident up to
85C, with very weak growth seen at 90C. Growth was
determined by the degree of turbidity of thermophilic microTable 1.

Thermophilic bacterial load and physical conditions of the

Savusavu Hot Springs at different times of the year

Sampling
Date

Avg. load of
Thermophilic
SD of
Bacteria
cfu values
1
(cfu mL ) n=3

September
2005
November
2005
January
2006
March
2006

Table 2.

Temp. of
pool water
(C)

pH of
pool water

960

1.6

60.5

7.0

1260

3.2

63.0

7.5

12200

6.6

70.0

6.5

21600

5.3

73.0

6.5

bial cultures in nutrient broth. This indicates that there are

obligate and facultative aerobic thermophiles present in the
hot springs.
The isolates could tolerate a salinity of 0.1% up to 3.0%,
and a pH of 58, meaning that they are near neutral pH
thermophiles. Although the Savusavu Hot Springs are
located very close to the sea, the above results indicate that
there is no mixing of seawater with the underground hot
spring waters. This was also confirmed by the chemical analysis of the hot spring waters4).
As shown in Table 2, most of the isolates obtained
were Gram positive (76.9%), had the ability to produce
endospores (51%), and were motile (71.2%). Catalase and
oxidase activity was also detected, with approximately 4% of
isolates also exhibiting fermentative activity. Levels of extracellular amylase, lipase and gelatinase activity varied.
Of the isolates analyzed, 60% gave restriction patterns that
were similar to Anoxybacillus flavithermus (DSM 2641),
20% were similar to Geobacillus stearothermophilus (ATCC
12980), 10% were similar to Thermus sp. strain TG 206
(TRU ZV1D.A2) and 10% were similar to Thermus sp. strain
TG 153 (TRU HWB.A1). Four of the thermophilic bacterial
isolates that were obtained from the Savusavu Hot Springs
did not match DNA patterns of any of the comparative cultures used. These strains have very different restriction patterns from the type strains they were compared with. These
were the isolates labeled as strain 70C8, strain 70C35, strain
70C37 and strain 75C19. Their unique DNA band patterns
are shown in Figs. 13. Fig. 1 shows the DNA patterns of
strain 70C8, strain 70C35 and strain 70C37 after digestion
with EcoRI, while Fig. 2 shows the patterns of the same
isolates after digestion with HaeIII. Fig. 3 shows the unique
pattern formed by the fourth isolate; strain 75C19; after
digestion with EcoRI.
The molecular analysis showed that all of the identified
thermophiles have been found previously in other hot springs
around the world. This was also the case for the isolates
obtained at 85C, as their DNA bands matched isolates

Morphological and physiological characteristics of the

thermophilic bacterial isolates from Savusavu Hot Springs,
Fiji

Characteristics
Morphological:
Gram positive
Gram negative
Gram variable
Endospore producers
Endospore absent
Motile
Immotile
Biochemical:
Catalase positive
Catalase negative
Oxidative
Fermentative
Physiological:
Amylase producers
Amylase negative
Lipase producers
Gelatinase producers

Percentage
76.9
19.2
3.8
51
49
71.2
28.8
86.5
13.5
59.6
4.04
81.7
16.3
11.5
23.1

Fig. 1. EcoRI/React 3 showing unique bands for the thermophilic

strains 70C8, 70C35 and 70C37 in lanes 13, 14 and 16, respectively.
Lanes: 1, DNA ladder; 2, CN; 3, Thermus sp. strain RT 41 A; 4, Thermus
sp. strain TG 275; 5, Thermus sp. strain TG 206; 6, Thermus sp. strain
TG 153; 7, Thermus sp. strain TG351; 8, strain 70C57; 9, strain 70C61;
10, strain 70C64; 11, strain 70C65; 12, strain 70C66; 13, strain 70C8;
14, strain 70C35; 15, strain 70C60; 16, strain 70C37; 17, DNA ladder.

NARAYAN et al.

352

casein and tributyrin hydrolytic properties. Anoxybacillus

flavithermus has been the subject of study for its gelatinases,
which are used in gelatin processing10). In addition, the four
strains with variant DNA patterns might be novel bacteria
with possibly useful properties.
This study shows that there are industrially important thermophilic bacteria present in the Savusavu Hot Springs. Further work is underway to conduct a complete profile of the
microbial community in these hot springs using culture-independent methods.
Acknowledgements

Fig. 2. HaeIII/React 2: 16S rDNA PCR showing unique bands in

thermophilic strains 70C8, 70C35 and 70C37 in lanes 1518. Lanes: 1,
DNA ladder; 2, CN; 3, Thermus sp. strain RT 41 A; 4, Thermus sp.
strain TG 275; 5, Thermus sp. strain TG 206; 6, Thermus sp. strain TG
153; 7, Thermus sp. strain TG 351; 8, strain 70C57; 9, strain 70C61; 10,
strain 70C64; 11, strain 70C65; 12, strain 70C66; 13, DNA ladder; 14,
DNA ladder; 15, strain 70C8; 16, strain 70C35; 17, strain 70C37; 18,
strain 70C60.

Fig. 3. EcoR1/React 3 showing unique band in thermophilic strain

75C19 in lane 20. Lanes: 1, Thermus sp. strain RT 41 A; 2, Thermus sp.
strain TG 206; 3, Thermus sp. train TG 351; 4, strain DNA ladder; 5,
strain CN; 6, strain ATCC 12980; 7, strain DSM 2641; 8, strain 75C4;
9, strain 75C9; 10, strain 75C10; 11, strain 75C11; 12, strain 75C12; 13,
strain 75C13; 14, strain 75C14; 15, DNA ladder; 16, strain 75C15; 17,
strain 75C16; 18, strain 75C17; 19, strain 75C18; 20, strain 75C19.

obtained at 70C and the comparative cultures. However,

these known strains are of significance in industry and
applied science. Geobacillus stearothermophilus spores are
commonly included among the materials being autoclaved in
industries. Death of the spores indicates the autoclave is
functioning properly and sterilization was successful8).
Recently, it has also been shown that Geobacillus stearothermophilus has very high cadmium ion adsorption potential and can be used for metal mobilization in the environment (such as the contamination of drinking water). It can also
be used to improve waste treatment of polluted water and
soil5). Thermus spp. TG153 (Thermus sp. strain HWB.A1)
and TG206 (Thermus sp. strain ZV1D.A2) have significant

We thank the staff at the Thermophile Research Unit (TRU),

University of Waikato (UOW), New Zealand for providing support
and facilities. We also thank the University of the South Pacific
Research Committee for awarding a scholarship (VVN) and financial support. Mr. Abhineshwar Vinay Prasads technical support is
acknowledged and Ms. Sujlesh Sharma is to be thanked for inspirational support and motivation.
References
1) Autar, R.K. 1996. Fiji geothermal resource assessment and development programme. Proceedings: 21st Workshop on Geothermal
Reservoir Engineering. Stanford University. Stanford, California.
Jan. 2224. SGP-TP 151.
2) Dempster, E.L., K.V. Pryor, D. Francis, J.E. Young, and H.J. Rogers.
1999. Rapid DNA extraction from ferns for PCR-based analyses. Biotechnol. 27:6668.
3) Dhanasekaran, D., P. Sivamani, G. Rajakumar, A. Panneerselvam,
and N. Thajuddin. 2006. Studies on free and immobilised cells of
Bacillus species on the production of alpha-amylase. Int. J. Microbiol.
2(2).
4) Healy, J. 1960. The hot springs and geothermal resources of Fiji. New
Zealand. Department of Scientific and Industrial Research. Bulletin;
136. Wellington, New Zealand.
5) Hetzer, A., C.J. Daughney, and H.W. Morgan. 2006. Cadmium ion
biosorption by the thermophilic bacteria Geobacillus sterothermophilus and G. thermocatenulatus. Appl. Environ. Microbiol. 72:4020
4027.
6) Hudson, J.A., H.W. Morgan, and R.M. Daniel. 1986. Numerical
classification of Thermus isolates from Icelandic Hot Springs. Syst.
Appl. Microbiol. 9:218223.
7) Hudson, J.A., H.W. Morgan, and R.M. Daniel. 1988. Numerical classification of Thermus isolates from globally distributed hot springs.
Syst. Appl. Microbiol. 11:250256.
8) Lemieux, P., R. Sieber, A. Osborne, and A. Woodward. 2006.
Destruction of spores on building decontamination residue in a commercial autoclave. Appl. Environ. Microbiol. 72:76877693.
9) Nair, P.S., and P.K. Surendran. 2004. Biochemical characterisation of
lactic acid bacteria isolated from fish and prawn. J. Cult. Col. 4:48
52.
10) Rivers, S., and R.E. Amelunxen. 1973. Proteins from thermophilic
microorganisms. Bacteriol. Rev. 37:320342.
11) Ronimus, R.S., A. Rckert, and H.W. Morgan. 2004. A RAPD-based
survey of thermophilic bacilli in milk powders from different countries. Int. J. Food Microbiol. 15:263272.
12) Tandon, P., S. Chhibber, and R.H. Reed. 2006. Survival and detection
of the fecal indicator bacterium Enterococcus faecalis in water stored
in traditional vessels. Indian J. Med. Res. 125:557566.
13) Whaley, D.N., V.R. Dowell, L.M Wanderlinder, and G.L Lombard.
1982. Gelatin agar medium for detecting gelatinase production by
anaerobic bacteria. J. Clin. Microbiol. 16:224229.
14) Willis, A.T. 1960. The lipolytic activity of some clostridia. J. Path.
Bacteriol. 80:379390.