JOURNAL OF BIOSCIENCE AND BIOENGINEERING

Vol. 88, No. 6, 617-621. 1999

Production of Canthaxanthin by Extremely Halophilic Bacteria

DALAL ASKER AND YOSHIYUKI OHTA*
Faculty of Applied Biological Science, Hiroshima University, l-4-4 Kagamiyama, Higashi-Hiroshima 7394.528, Japan
Received 25 June 1999IAccepted 14 September 1999

Soil samples from a salt farm were used as a source for the isolation of carotenoid-producing bacteria. The
conditions for optimum growth and carotenoid production were established for the isolated bacteria. Carote-
noids were analysed by spectrophotometry and High Performance Liquid Chromatography (I-IPLC). Thirty-one
red extremely halophilic bacteria were isolated from saline soil samples collected from a salt farm in Alexandria,
Egypt. Among the isolated strains, strain TM exhibited the highest carotenoid-producing ability. Maximum
growth of strain TM occurred in the presence of high concentrations of sodium chloride and magnesium sulfate.
Growth did not occur when NaCl concentration was lower than 10% and the cells lysed at this concentration.
Optimum growth of and carotenoid production by strain TM were realized at 37’C in the presence of 1% yeast
extract, 0.75% casamino acids, 25% NaCl, 4% MgS04, 0.2% KC1 and at pH 7.2 with shaking for 6 d. Strain
TM produced 2.06 mg total carotenoids g-l dry cells, including 0.06 mg of ,&carotene and 0.70 mg of can-
thaxanthin. This is the first report of an extremely halophilic bacterium that produces canthaxanthin.
[Key words: extremely halophilic bacteria, Halobacterium sp., carotenoid, canthaxanthin, b-carotene, bac-
terioruberins, HPLC, spectrophotometer]

Carotenoids are yellow to orange-red pigments that in 90ml of complex medium (CM) containing (per liter):
are present in a wide variety of bacteria, algae, fungi log yeast extract, 7.5 g casamino acids, 250g NaCl, 40g
and plants (1). The functions of carotenoids include the MgS04.7Hz0, 2 g KCl, and 3 g trisodium citrate. The
following: food colorants, absorbers of light energy, pH was adjusted to 7.2 (18). The flasks were incubated
oxygen transporters, provitamin A, scavengers of active on a shaker at 37°C until red turbidity appeared. The
oxygen, antitumor and enhancers of in vitro antibody turbid cultures were streaked on CM broth containing
production (2-6). The use of colorants in food has been 1.2% (w/v) agar and incubated at 37°C. After 7 d of
acknowledged to play a part in consumer acceptability incubation, red and yellow colonies were selected and
of processed foods. Canthaxanthin as a ketocarotenoid purified.
pigment is a useful food colorant in the red wavelength Screening of highest carotenoid-producing strain
range; furthermore, it is used as an agent for pigmenting Each of the isolated strains was grown in 100ml CM
cultured salmonids and crustaceans (7). The red extreme- broth in a 500ml Erlenmeyer flask and incubated on a
ly halophilic bacteria (non-photosynthetic bacteria) pro- rotary shaker at 240 rpm (200; Takasaki Co., Saitama)
duce phytoene, p-carotene, lycopene and derivatives of for 7 d at 37°C. One milliliter of the culture was diluted
acyclic Cso bacterioruberin (8). Recently, the production to 10 ml using 25% NaCl solution, to measure growth in
of ketocarotenoids such as 3-hydroxy echinenone or trans- terms of optical density at 660 nm with a spectrophotom-
astaxanthin from Halobacterium salinarium was reported eter (U 2001; Hitachi, Tokyo). The results were express-
by Calo et al. (9). These bacteria grow in highly saline ed as (OD values x 10).
environments, giving a pink to red color to these habitats. Carotenoid extraction and analysis Ten ml aliquots
In general, they require at least 1.5 M (9%, w/v) NaCl of cultures were centrifuged at 6000 xg for 10 min. The
for growth, and 2-4M (12-23x, w/v) NaCl for optimal harvested cells were resuspended in distilled water. Spon-
growth (10-14). taneously, cellular lysis occurred, and then the pigments
With the rising global concern to avoid the undesira- were extracted by acetone and transferred to hexane (9).
ble effects of synthetic food colorants such as allergy, The carotenoid extracts were analysed by scanning the
hypersensitivity, intolerance and childhood hyperactivity absorbance in the wavelength region of 400-600 nm
(15-17), an attempt was made to find a new source of using the spectrophotometer. The total carotenoid con-
canthaxanthin by isolating red halobacteria strains from tent in the hexane extract was estimated by measuring the
a salt farm in Egypt. To maximize the yield of carote- absorbance at R,, (490nm) (19). The results are given
noid production by the isolated strains, the optimum as OD per 100ml culture. The highest carotenoid-pro-
conditions for the production of carotenoid were also ducing strain was selected on the basis of high growth,
investigated. carotenoid and degree of cell pigmentation (20). The
degree of cell pigmentation was calculated as the ratio
MATERIALS AND METHODS of OD of the carotenoid to OD of the growth (OD,dODm).
The selected strain was used in subsequent experiments.
Isolation of halobacteria Soil samples were ob- HPLC analysis of carotenoid Red pigments ob-
tained from the seawater evaporation pond (El-Malahat) tained were separated and quantified using HPLC (L-
near Alexandria City in Egypt in April 1996. This farm 6200; Hitachi) equipped with a UV-VIS detector (L-4200;
is used in the production of common salt. Approximate- Hitachi), according to the method of Sedmak et al. (21)
ly 10 g (fresh weight) each of soil samples was suspended and Calo et al. (9) with slight modification. Carotenoid
extracts were subsequently filtered through a 0.5 pm hydro-
* Corresponding author. phobic Teflon membrane (Advantec, Tokyo). Chromato-
617
618 ASKER AND OHTA J. BIOSCI. BIOENG..

graphic separation was performed on a normal-phase of the soil around the salt farm is characterized by an
column (Silica, 250 x 4.6 mm, Inertsil SIL, GL Sciences orange-red color, corresponding to the growth of pig-
Inc., Tokyo). The mobile phase was hexane: ethyl acetate mented microorganisms. Soil samples were collected
(50 : 50, v/v) at a flow rate of 1 ml/min. This was later from this environment as a source for the isolation of
modified to 30 : 70 (v/v) of hexane: ethyl acetate. The carotenoid-producing halobacteria. Thirty-one red and
eluant was monitored at 490nm. The results are given as six yellow pure halobacteria were isolated. All the red
the percentage of total carotenoids. halobacteria strains isolated gave identical absorption
Optimization of culture conditions Determination spectra of carotenoid (Fig. lc). The spectra are character-
of optimum conditions for growth and carotenoid ized by maximum peaks at 493 and 527 nm with a broad
production was carried out by inoculating the selected shoulder at 467 nm, indicating that the main carotenoid
strain in 100ml CM broth in a 5OOml Erlenmeyer flask is w-bacterioruberin (20, 23). However, the yellow-pig-
and incubating on a rotary shaker at 240 rpm for 7 d at ment-producing isolates did not produce any carotenoid.
37°C. Growth and carotenoid production were deter- Table 1 summarizes the growth and carotenoid produc-
mined as described above. The tested parameters were: tion by red halophilic bacterial isolates. It was observed
NaCl concentration between 0 and 35% in 5% incre- that strain TM had the highest value of the degree of
ments; MgS04. 7HzO concentration at 0, 1, 2, 3, 4, 5 pigmentation, indicating that it exhibited the highest
and 6%; using MgCIZ or sodium sulfate instead of mag- carotenoid-producing ability. Figure 2 shows the HPLC
nesium sulfate; KC1 concentration at 0, 0.1, 0.2, 0.3, chromatogram of carotenoids produced by strain TM
0.4, 0.5 and 1%; trace elements (FeC12, 0.23 mg; CaC12. in comparison with standard carotenoids. It was found
7H20, 0.7 mg; MnS02.Hz0, 0.03 mg; ZnS04, 0.044 mg; that the separation of total carotenoids from strain TM
CuS04. 5Hz0, 5 jig per 100 ml) (22); incubation time, 1 with ethyl acetate: hexane (50 : 50, v/v) as a mobile phase
to 8 d; temperature, 15 to 55°C in 5” increments; different required a longer time than that from Halobacterium
levels of pH, 4.5 to 9; shaking; medium volumes, 100, salinarium (9). The separation time reached 40min at
200, 300 and 400ml of medium per 500ml Erlenmeyer least until all pigments were eluted (Fig. 2~). Therefore,
flask; and light illumination. The optimum parameter an increase in solvent polarity seemed to be essential for
from each experiment was utilized in the next experiment eluting the polar red pigments from the column (Fig.
and so on until all the parameters were optimized. 2b). Retention times (R,) of peaks 1 and 3 were iden-
tified to be those of j-carotene and canthaxanthin, since
their R, values were identical to those of authentic stan-
RESULTS
dards (Fig. 2a). This bacterium also produced 3-hydroxy
Isolation and screening of carotenoid-producing halo- echinenone (peak 2) and bacterioruberins (peaks 4, 5,
bacteria In summer, the temperature of the salt farm 6 and 7), as shown in Fig. 2b (9). Consequently, strain
in Alexandria City ranges from 37 to 40°C. As a result,
the total dissolved salt concentration increases to TABLE 1. Growth of and carotenoid production by red halophihc
bacterial isolates
saturate at pH 7.2. Under these conditions, the surface
Isolate Growth Pigment Degree of pigmentation
IlO. (O&d U-bd KQxmd

L
a R 3.35 7.25 2.16
H 3.35 5.08 1.52
T 3.53 6.94 1.97
DT 3.43 6.49 1.89
DM 3.40 4.86 1.43
X 3.05 6.73 2.21
I/“ MT 2.90 6.90 2.38
I I I I Z 2.73 5.26 i .93
TM 2.95 10.31 3.56
b N 2.90 6.16 2.12
U 2.55 5.30 2.08
0 2.40 4.45 1.89
S 2.30 4.62 2.01
B 2.55 4.09 1.60
D 2.40 5.51 2.30
I 2.50 4.66 1.86
C 2.55 3.41 1.34
G 3.00 6.98 2.19
A 2.95 4.08 1.38
w 2.85 3.15 1.11
F 2.95 3.96 1.34
E 3.10 5.16 2.10
P 2.55 4.41 1.75
Y 2.85 3.27 1.15

t 2.90
2.80 4.18
3.85 1.38
1.44
V 2.85 4.70 1.65
M 2.90 4.93 1.70
Wavelength (nm) TD 3.15 3.37 1.07
K 2.95 7.48 2.54
FIG. 1. Absorption spectra of (a) p-carotene standard, (b) can- MD 2.75 4.95 1.80
thaxanthin standard and (c) carotenoid produced by strain TM.
VOL. 88, 1999 CANTHAXANTHIN-PRODUCING HALOBACTERIA 619

3.5 12.0
a b
1
3.0
10.0
3
2.5
8.0
^o 3
g 2.0
%
e 6.0 +j
S
5 $
e 1.5
e
u
J
4.0
1.0

2.0
0.5

0.0 L 0.0
0 s 10 15 20 25 30 35
NaCl cone. (%)
30 35 40 45
FIG. 4. Effect of NaCl concentration on growth of and carote-
Time (mm) noid production by strain TM. Symbols: bar, growth (OD&; line,
FIG. 2. HPLC profiles of (a) p-carotene and canthaxanthin stan- carotenoid (OD,,).
dards, (b) and (c) carotenoids produced by strain TM were separated
by hexane: ethyl acetate 50 : 50 and 30 : 70 (v/v), respectively. Peaks: both low growth and carotenoid production. Replacing
1, p-carotene; 2, 3-hydroxy echinenone; 3, canthaxanthin; 4, 5, 6 and magnesium sulfate with magnesium chloride caused very
7, bacterioruberins. slow growth even at a high concentration of MgC12
(3X), and also, led to low growth with a very low con-
TM was selected for further studies. tent of carotenoid. The experiment was repeated with
Optimization of culture conditions As shown in Na2S04 to determine whether the cells require sulfate ion
Fig. 3, the growth curve of strain TM was characterized or magnesium cation; however no growth was observed
by a long lag phase (about 2d). Pigment production by at all. This indicated that strain TM required high con-
the strain was observed after 4d of incubation, as the centrations of magnesium and sulfate ions.
culture became slightly orange in color. Growth and The maximum growth of and carotenoid production
carotenoid production increased to a maximum after 6 d by the strain were found at 0.2% KCl. However,
of incubation, followed by a decrease in both growth without the addition of KCl, strain TM could grow but
and carotenoid production. with little pigment production (24). By increasing the
Figure 4 shows that the growth of and carotenoid concentration of KCl, both growth and pigment produc-
production by strain TM did not occur at less than 10% tion increased. However, increasing the concentration of
NaCI. The maximum growth and production of carote- KC1 above 0.2% caused a slight decrease in the growth
noid were observed at 25% NaCl. and carotenoid production. The addition of trace ele-
Growth of and carotenoid production by strain TM ments increased both growth of and carotenoid produc-
did not occur at less than 1% MgS04, as shown in Fig. tion by strain TM.
5. Increasing the concentration of MgS04 enhanced the
growth and production of carotenoid by the strain. It
was found that this strain required 4% (w/v) of MgS04
for maximum growth and carotenoid production. In-
creasing MgS04 concentration above 4% resulted in

12.0 h 2.0 8.0 2

3.5I 3
%
3.0 8
10.0
2 1.5 6.0 .%
x
8.0 3 k 2
” e
8 1.0 4.0 9
6.0 g
P
4.0 2
f
0.5 2.0 u
0.0 0.0
0.0 0.0 0 1 2 3 4 5 6
0 2 4 6 8 MgSO, cone. (%)
Incubation time (d)
FIG. 5. Effect of MgS04 concentration on growth of and carote-
FIG. 3. Growth and carotenoid production by strain TM. Sym- noid production by strain TM. Symbols: 0, growth (OD,&; 0,
bols: 0, growth (OD.&; 0, carotenoid (OD,,). carotenord (OD&.
620 ASKER AND OHTA I. BIOXI. BIOPNC;.1

used in both food and feed industries, but also a source

which has features to facilitate its industrial application.
Extremely halophilic bacteria producing red pigments
were isolated. Among the thirty-one strains isolated,
strain TM was selected as the highest carotenoid-produc-
ing strain. It was interesting to find that some ketocarote-
noids produced by the strain were similar to those
produced by the marine bacterium, Agrobacterium
auranticum (25, 26), as well as the yeast, P. rhodozyma
(27, 28). Strain TM produced 2.06mg total carotenoids
g I. dry cells. In comparison, Phafia rhodozyma
produces 0.41 mg.g ’ dry yeast and Agrobacterium
auranticum produces 0.46 mg total carotenoids.1 I cul-
ture. The microbial sources of canthaxanthin are limited
to only a few microorganisms (29). The presence of can-
thaxanthin (33.88% of the total carotenoids) and 3-
15 20 25 30 37 40 45 50 55 hydroxy echinenone in strain TM, especially astaxanthin
Temperature (e) (data not shown) indicates its potential industrial ap-
FIG. 6. Effect of temperature on growth of and carotenoid plicability.
production by strain TM. Symbols: 0, growth (ODsso); 0, carote- The unique features of strain TM are as follows: (i)
noid (OD,,). the extremely high NaCl concentration used in the
growth medium, which is useful to prevent contamina-
Figure 6 demonstrates that the growth of and carote- tion by other organisms. Because of this, no sterilization
noid production by strain TM did not occur below 20°C procedures are required (data not shown); (ii) NaCl con-
and above 50°C. The maximum growth and production centration below 10% induces cell lysis (unpublished
of carotenoid occurred at 37°C. At temperatures above data), so no cell disrupting devices are required as cells
37”C, low growth and pigment content was observed. lyse spontaneously in freshwater; and (iii) furthermore, the
The growth of and carotenoid production by strain procedures for extraction and purification of carotenoids
TM did not occur below pH 5 and above 7.5, as shown from strain TM seem to be simpler than those from
in Fig. 7. Maximum growth and production occurred at other sources, as bacterial carotenoids are usually easier
pH 7.2. to isolate than carotenoids from other sources (25). The
The strain when cultivated under shaking conditions strain requires extremely high concentrations of NaCl
exhibited high growth and carotenoid production. and MgS04 for optimum growth. The concentration of
Without shaking, little growth with no carotenoid pro- the latter was about twice that reported for Halobacte-
duction was observed. Maximum growth and carotenoid rium sp. which required 2% MgS04. Magnesium has been
production by strain TM occurred in 100 ml medium per reported previously to be an essential element for red
500 ml Erlenmeyer flask. No effect of light illumination halophiles; it is possible that Mg” is required for cell
was found on growth of and carotenoid production by division (12). The addition of trace elements solution
the strain. increased both growth of and carotenoid production by
Strain TM, grown in CM broth under optimum condi- the strain, because it fulfills the requirements of iron,
tions, produced 2.06mg total carotenoids gg’ dry cells. manganese, and zinc of the strain (18). Shaking the cul-
The relative percentages of p-carotene, canthaxanthin ture significantly increased the growth of and carotenoid
and bacterioruberins to the total carotenoids were production by the strain, because it is a strict aerobe.
2.94%, 33.88%, and 63.17%, respectively. Increasing the volume of the medium above 100ml in
a 500ml Erlenmeyer flask decreased the amount of dis-
solved oxygen; therefore, growth of and carotenoid
DISCUSSION
production by the cells decreased under these conditions.
It is important to isolate not only a new biological Yokoyama and Miki (26) reported that varying the me-
source which produces a natural colorant which can be dium volume controlled the growth of and carotenoid
production by Agrobacterium auranticum.
3.5 12.0 In conclusion, strain TM is a promising source of can-
thaxanthin for fish and poultry industries, and could
3.0 10.0 be used as a colorant in the food industry. Improved
- 2.5 -3 production can be expected by obtaining a mutant with
0 8.0 a"
o_
higher productivity or by isolating a cloned gene related
g 2.0
6.0 = to a specific carotenoid biosynthesis from the strain.
35 1.5
E
4.0 E
6 1.0 ACKNOWLEDGMENTS
B
2.0
0.5 We would like to thank Professor Ahmed R. El-Mahdy of Alex-
andria University for useful suggestions and Mr. M. Adjei for criti-
0.0 0.0
cal reading of the manuscript.
4.5 5 5.5 6 6.8 7.2 7.5 7.8
PH
REFERENCES
FIG. 7. Effect of pH on growth of and carotenoid production by
strain TM. Symbols: 0, growth (OD,&; 0, carotenoid (ODdw). 1. Goodwin, T. W. and B&ton, G.: Distribution and analysis of
VOL. 88, 1999 CANTHAXANTHIN-PRODUCING HALOBACTERIA 621

carotenoids, p. 61-132. In Goodwin, T. W. (ed.), Plant pig- 16. Rosette, L.: Food colors: a scientific status summary by the
ments. Academic press, London, United Kingdom (1980). Institute of Food Technologists Expert panel on food safety &
2. Krinsky, N.I.: Carotenoid protection against oxidation. Pure nutrition. Food Tech., 40, 49-56 (1986).
Appl. Chem., 51, 649-660 (1979). 17. Salem, E. A. G., El-Ahwal, A.M. A., El-Dera, H. B., Omara,
3. Mathews-Roth, M. M.: Carotenoids in medical applications, p. A.A., and El-Henawy, N.A.: A survey on the added food
755-785. In Bauernfeind, J. C. (ed.), Carotenoids as colorants colors in food products. The Bulletin of the High Institute of
and vitamins A precursors. Academic Press, New York (1981). Public Health, Egypt, 20, 127-139 (1990).
4. Palozza, P., Luberto, C., Ricci, P., and Sgarlata, E.: Effect 18. Sehgal, S. N. and Gibbons, N. E.: Effect of some metal ions on
of p-carotene and canthaxanthin on tert-butyl hydroperoxide- the growth of Halobacterium cutirubrum. Can. J. Microbial., 6,
induced lipid peroxidation in murine normal and tumor thymo- 165-169 (1960).
cytes. Arch. Biochem. Biophys., 297, 291-295 (1996). 19. Daves, B. H.: Analysis of carotenoid pigments, p. 489-532. In
5. Tomita, Y.: Immunological role of vitamin A and its related Goodwin, T. W. (ed.), Chemistry and biochemistry of plant
substances in prevention of cancer. Nutr. Cancer, 5, 187-194 pigments, vol. 2. Academic Press Inc., New York (1965).
(1983). 20. Baxter, R. M.: Carotenoids pigments of halophilic bacteria.
6. Tee, E. S.: Carotenoids and retinoid in human nutrition. Crit. Can. J. Microbial., 6, 417-424 (1960).
Rev. in Food Sci. & Nutr., 31, 103-163 (1992). 21. Sedmak, J. J., Weerasinghe, D. K., and Jolly, S. 0.: Extrac-
7. Torrissen, 0. J.: Pigmentatiqn of sahnonids-A comparison of tion and quantitation of astaxanthin from Phafia rhodozyma.
astaxanthin and canthaxanthin as pigment sources for rainbow Biotechnol. Tech., 4, 107-112 (1990).
trout. Aquaculture, 53, 271-278 (1986). 22. Onishi, H., McCance, M. E., and Gibbons, N. E.: A synthetic
8. Goodwin, T. W.: Non-photosynthetic bacteria, p. 290-317. In medium for extremely halophilic bacteria. Canad. J. Micro-
Goodwin, T. W. (ed.), The biochemistry of the carotenoids, biol., 11, 365-373 (1965).
vol. 1. Chapman & Hall Ltd., London (1980). 23. Petter, H. and Lochhead, A.G.: On bacteria of salted fish.
9. Calo, P., de Miguel, T., Sieiro, C, Velazquez, J. B., and Villa, Kon. Akad. Wetenschap. Amsterdam Proc., 34, 1417-1423
T. G.: Ketocarotenoids in halobacteria: 3-hydroxy echinenone (1931).
and trans-astaxanthin. J. Appl. Bacterial., 79, 282-285 (1995). 24. Brown, H. J. and Gibbons, N. E.: The effect of magnesium,
10. Gibbons, N. E.: Turbidity of suspensions and morphology of potassium, and iron on the growth and morphology of red
red halophilic bacteria as influenced by sodium chloride concen- halophilic bacteria. Can. J. Microbial., 1, 486-494 (1955).
tration. Can. J. Microbial., 6, 535-543 (1960). 25. Yokoyama, A., Izumide, H., and Miki, W.: Production of
11. Gibbons, N. E. and Payne, J. I.: Relation of temperature and astaxanthin and 4-ketozeaxanthin by the marine bacterium,
sodium chloride concentration to growth and morphology of Agrobacterium aurantiacum. Biosci. Biotech. Biochem., 58,
some halophilic bacteria. Can. J. Microbial., 7, 483-489 1842-1844 (1994).
(1961). 26. Yokoyama, A. and M&I, W.: Composition and presumed
12. Javor, B. J.: Growth potential of halophilic bacteria isolated biosynthetic pathway of carotenoids in the astaxanthin-produc-
from solar salt environments: carbon sources and salt require- ing bacterium Agrobacterium aurantiacum. FEMS Microbial.
ments. Appl. Environ. Microbial., 48, 352-360 (1984). Lett., 128, 139-144 (1995).
13. Larsen, H.: Halophilism, p. 297-342. In Gun&us, I. C. and 27. Johnson, E. A., Conklin, D. E., and Lewis, M. J.: The yeast
Stanier, R. Y. (ed.), The bacteria. Academic Press, New York Ph@u rhodozyma as a dietary pigment source for salmonds
(1962). and crustaceans. J. Fish Res. Bord Can., 34, 2417-2421 (1977).
14. Larsen, H.: Family Halobacteriaceae, p. 261-267. In Krieg, 28. Johnson, E. A., Villa, T. G., and Lewis, M. J.: Phujgia
N. R. (ed.), Bergey’s manual of systematic bacteriology. Wil- rhodozyma as an astaxanthin source in salmond diets. Aquacul-
liams & Wilkins, Baltimore (1984). ture, 20, 123-134 (1980).
15. Cooners, C. K., Goyelle, C. II., Deborah, A., and Andrrnlonis, 29. Nelis, H. J. and Leenheer, A. P.: A review: microbial sources
P. P.: Food additives and hyperkinesis, a controlled double of carotenoid pigments used in foods and feeds. J. Appl. Bac-
blind experiment. Pediatrics, 58, 154-166 (1976). teriol., 70, 181-191 (1991).