Letters in Applied Microbiology 2000, 31, 59

Indigo production by naphthalene-degrading bacteria

B. Bhushan, S.K. Samanta and R.K. Jain
Institute of Microbial Technology, Chandigarh, India

69/99: received 13 December 1999, revised 1 March 2000 and accepted 3 March 2000

A wild-type naphthalene-degrading strain

B . B H U S H A N , S . K . S A M A N T A A N D R . K . J A I N . 2000.
Pseudomonas putida RKJ1 and two recombinant strains each of Ps. putida and Escherichia coli
carrying the genes for naphthalene degradation on a recombinant plasmid pRKJ3, produced
indigo and indirubin pigments from indole. Naphthalene, salicylate and IPTG induced cells
of naphthalene-degrading recombinant bacteria produced up to two times higher indigo
compared with the uninduced cells. The maximum rates of indigo formation by Ps. putida
RKJ1, Ps. putida RKJ5/pRKJ3, Ps. putida KT2442/pRKJ3, E. coli TB1/pRKJ3 and E. coli
AB1157/pRKJ3 were 0
60, 0
80, 0
60, 1
20 and 1
50 nmol min1 mg dry biomass1,
respectively, using indole as the substrate. The apparent Km values of indigo formation by
these same bacteria were 0
22, 0
15, 0
10, 0
21 and 0
20 mmol l1, respectively, again using
indole as the substrate. The present study revealed that E. coli AB1157 was the most
efcient of the hosts tested for the expression of the plasmid encoded genes (pRKJ3) from
the wild-type strain Ps. putida RKJ1. In addition, both recombinant E. coli strains were
capable of producing indigo directly from nutrient medium.

INTRODUCTION Recently, a method has been patented by O'riel and Kim

Indigo is one of the oldest textile dyes and was traditionally
produced from an extract of the plants of the genus
Indigofera. Presently, it is being produced by chemical
synthesis and is marketed by many leading companies.
Very few reports are available regarding the microbial bio-
synthesis of indigo. Among the microbial indigo producers,
the majority of micro-organisms are aromatic hydrocarbon-
degrading bacteria such as naphthalene-degrading
Pseudomonas putida NDO (Murdock et al. 1993), Ps. putida
PpG7 (O'Connor and Hartmans 1998), p-cumate and m-
and p-toluate-degrading Ps. putida F1 and Ps. putida mt-2,
respectively (Eaton and Chapman 1995), styrene-degrading
Ps. putida S12 and CA-3 (O'Connor et al. 1997) and
toluene-degrading Ps. mendocina KR1 (Yen et al. 1991).
The enzyme system responsible for indigo formation gen-
erally consists of one or more enzymes, typically monooxy-
genases, dioxygenases or hydroxylases (O'Connor et al.
1997; Doukyu et al. 1998). The genes encoding these
enzymes have been cloned from Pseudomonas and
Rhodococcus spp. into Escherichia coli strains (Yen et al.
1991; Hart et al. 1992; Eaton and Chapman 1995) so as to
produce indigo directly from either nutrient medium
(Ensley et al. 1983) or from glucose (Murdock et al. 1993).
Correspondence to: R.K. Jain, Environmental Biotechnology Laboratory,
Institute of Microbial Technology, Sector-39 A, Chandigarh-160 036, India
(e-mail: rkj@imtech.ernet.in).
= 2000 The Society for Applied Microbiology
(1998) for producing indigo and indirubin dyes using a
recombinant Escherichia coli containing a gene encoding a
phenol hydroxylase from Bacillus stearothermophilus. In the
present study, we report on the formation and identica-
tion of indigo and indirubin pigments by one wild-type
and four recombinant bacterial strains.
MATERIALS AND METHODS
Bacteria, media and culture conditions
Of the ve bacteria used in the present study, Ps. putida
RKJ1 (Samanta et al. 1998) was a wild-type strain and was
capable of degrading naphthalene through salicylate. Four
recombinant bacteria namely Ps. putida RKJ5/pRKJ3, Ps.
putida KT2442/pRKJ3, E. coli TB1/pRKJ3 and E. coli
AB1157/pRKJ3 contained naphthalene-degrading genes on
a recombinant plasmid pRKJ3 from the wild-type Ps.
putida RKJ1 (Samanta et al. 1998). The naphthalene-
degrading genes were found to be located on a 25-kbp
EcoR1 fragment of a 83-kbp plasmid of wild-type host Ps.
putida RKJ1. The 25-kbp DNA fragment was cloned into a
broad host-range cosmid vector pLAFR3 (Straskawicz et al.
1987) which formed the recombinant plasmid pRKJ3
(Samanta et al. 1998). This plasmid was then introduced
into two genetically compatible hosts Ps. putida RKJ5 (plas-
mid-cured strain of Ps. putida RKJ1) and Ps. putida
KT2442 (Franklin et al. 1981), and two indole-producing
6 B. BHUSHAN ET AL.
strains of E. coli, i.e. TB1 (Li et al. 1979) and AB1157
(Bachmann 1972) by conjugation using E. coli pRK2013 as
the helper strain (Jain et al. 1984).
The mineral salts medium (MSM) used in the present
study was same as described earlier (Rani et al. 1996).
Naphthalene, salicylate and isopropyl-(-D-thiogalactopyra-
noside (IPTG) were added to the basal medium either
alone or in combination, so as to induce the plasmid-
encoded genes. In case of the media supplemented with
naphthalene, the MSM was prepared in naphthalene-satu-
rated distilled water (Strandberg et al. 1986). Salicylate and
IPTG were added to the media at a nal concentration of
2 mmol l1 and 1 mmol l1, respectively, for induction pur-
poses. In order to detect indigo production, the seed cul-
tures were inoculated into 100 ml of respective medium
and incubated at 30  C under shaking conditions (200 rev
min1). At mid-log phase of growth, indole was added in
the fermented media at a nal concentration of 2 mmol l1.
The cultures were further incubated for up to 12 h under
the same conditions until blue pigments appeared in the
culture broth.
Purification and identification of indigo and indirubin
The pigments from the whole culture broth were extracted
by the method of Hill et al. (1989) with some modica-
tions. The whole culture broth was mixed and shaken with
equal volume of chloroform in a separating funnel. The
mixture was centrifuged at 16 200 g for 20 min and the
chloroform layer was separated out and evaporated to dry-
ness in a round-bottomed ask using a rotary evaporator.
The dried pigments were then suspended in chloroform.
The pigment mixture thus obtained was subjected to analy-
tical as well as preparative layer chromatography (silica gel
60 F254 plates 20 (20 cm, Merck, Darmstadt, Germany)
along with standard indigo as described by Hart et al.
(1992) with some modications. The running solvent used
was chloroform-methanol (15 : 1). The separated pigments
were recovered from the silica plates in the same manner as
described by Hart et al. (1992) and further analysed by gas
chromatography-mass spectrometry (GC-MS). The GC-
MS analyses were carried out using a Shimadzu QP5000
GC-MS (Tokyo, Japan) equipped with a quadrupole mass
lter and DB-1 (100% dimethyl polysiloxane) capillary col-
umn (30 m  0
25 mm) with a lm thickness of 0
25 mm
(J & W, Tokyo, Japan). Other conditions were elec-
tron-impact ionization of 70 eV, scan intervals 1
5 s and
mass range 40500. The oven temperature was 150  C for
5 min with 10  C increase per minute to a nal temperature
of 300  C for 5 min, the injector temperature was kept at
250  C. A splitless injection was performed.
= 2000 The Society for Applied Microbiology, Letters in Applied Microbiology, 31, 59
Kinetics of indigo formation
The cell biomass used in the present study was obtained by
growing the bacterial culture for overnight in MSM
(described earlier), harvesting at 4  C by centrifugation,
washing with MSM and suspending in the same medium
to a concentration of 1 mg dry biomass ml1. The specic
rates of indigo formation by various bacteria were deter-
mined by the method of O'Connor and Hartmans (1998)
except that the nal indole concentration used in the assay
mixture was 0
50 mmol l1. The samples were taken every
10 min for up to 60 min. The specic rate of indigo forma-
tion was calculated from the increase in O.D. at 610 nm
over time. In order to determine the apparent Km and
maximum rates of indigo formation by different bacteria,
the specic rates of indigo formation were determined at
different concentrations of indole from 0
1 mmol l1 to 1
0
mmol l1. The assay time was kept constant at 10 min. The
apparent Km values were calculated from the Michaelis
Menten curves drawn between indole concentration and
indigo formation rates. The maximum rates of indigo for-
mation were calculated from the LineweaverBurk plot
drawn separately for each bacteria. All the experiments
except the purication of indigo pigments were performed
independently in triplicates and the results given here are
the mean of the three. The standard deviations were within
15%.
Chemicals
Indole, indigo, naphthalene, salicylate and IPTG were pur-
chased from Sigma Chemical Co., St Louis, MO, USA.
All other chemicals were of highest quality commercially
available.
RESULTS AND DISCUSSION
The ve bacteria used in the present study produced two
types of pigments, i.e. blue (pigment 1) and purple (pig-
ment 2). The Rf values of pigments 1 and 2 from all ve
bacteria in thin-layer chromatography (TLC) were 0
79
and 0
52, respectively, using chloroform-methanol (15 : 1)
as the solvent mixture. The Rf value of pigment 1 exactly
matched that of authentic indigo. Further identication of
the pigments were carried out using GC-MS technique.
The retention times of pigments 1 and 2 were 22
12 and
21
57 min, respectively, in GC-MS analyses. The m/z-
values (relative intensity, [proposed composition of ions])
of major ions found in the mass spectrum of pigment 1
were 262 (100, [M]), 234 (24, [M-CO]), 205 (37, [M-
CO-CHO]), 179 (9, [M-CO-CO-C2H2]), 151 (7), 131
(14, [M/2]), 104 (48, [M/2-HCN]) and 76 (53, [M/2-
CO-HCN]) which matched exactly the mass spectrum of
 INDIGO PRODUCTION BY BACTERIA 7

authentic indigo (data not shown). On the other hand, the
mass spectrum of pigment 2 was 262 (100, [M]), 234 (33,
[M-CO]), 205 (31, [M-CO-CHO]), 179 (43, [M-CO-
CO-C2H2]), 151 (8), 131 (18, [M/2]), 104 (53, [M/2-
HCN]) and 76 (49, [M/2-CO-HCN]) which exactly
matched the mass spectrum of indirubin published earlier
by Eaton and Chapman (1995). Although indigo and indir-
ubin were produced simultaneously to a variable extent by
all the bacterial strains used here, we have only described
the kinetics of indigo production.
The results obtained from indigo formation studies, fol-
lowing induction with naphthalene, salicylate, IPTG,
naphthalene IPTG and salicylate IPTG, revealed that
in case of wild-type Ps. putida RKJ1, all inducing condi-
tions except IPTG, stimulated the indigo production sig-
nicantly (Fig. 1a). On the other hand, the cells of
recombinant Ps. putida RKJ5/pRKJ3 (Fig. 1b) and Ps.
putida KT2442/pRKJ3 (Fig. 1c), under all induction con-
ditions, produced indigo to a greater extent compared with
their respective uninduced cells. Although there was not a
signicant difference between the indigo production by
induced and uninduced cells of E. coli TB1/pRKJ3 (Fig.
1d) and E. coli AB1157/pRKJ3 (Fig. 1e), it was observed
that both these strains produced indigo up to 2
02
5 times
more than the wild-type and recombinant Ps. putida strains
(Fig. 1).
The specic rates of indigo formation by different strains
were also determined. It was found that the maximum rate
of indigo formation by the wild-type Ps. putida RKJ1
induced with naphthalene was 0
60 nmol min1 mg dry
biomass1, whereas, the maximum rate of indigo formation
obtained from recombinant strains Ps. putida RKJ5/
pRKJ3, E. coli TB1/pRKJ3 and E. coli AB1157/pRKJ3
were 1
3, 2
0 and 2
5 times higher, respectively, compared
with Ps. putida RKJ1 (Table 1). Since the parents of E. coli
recombinant strains were capable of producing indole from
tryptophan, the recombinant E. coli TB1/pRKJ3 and E.
coli AB1157/pRKJ3 strains produced indigo directly from

Fig. 1 Progress curve of indigo formation by Pseudomonas putida RKJ1 (a), Ps. putida RKJ5/pRKJ3 (b), Ps. putida KT2442/pRKJ3 (c),
Escherichia coli TB1/pRKJ3 (d) and E. coli AB1157/pRKJ3 (e). Symbols for various conditions used: uninduced cells (W), naphthalene-
induced cells (*), salicylate-induced cells (!), IPTG-induced cells (!), naphthalene IPTG induced cells ( & ), salicylate IPTG-
induced cells (&). All the experiments were performed independently in triplicates and the results presented here are the mean of the
three. The standard deviations were within 15%

= 2000 The Society for Applied Microbiology, Letters in Applied Microbiology, 31, 59
8 B. BHUSHAN ET AL.

Table 1 Rate of indigo formation by various bacteria

Inducing Indigo formation rate (nmol min1 mg dry biomass1)

compound
Ps. putida Ps. putida Ps. putida E. coli E. coli
RKJ1 RKJ5/pRKJ3 KT2442/pRKJ3 TB1/pRKJ3 AB1157/pRKJ3

Control 0
30 0
03* 0
40 0
01 0
30 0
01 0
80 0
05 0
80 .0
03
(uninduced)
Naphthalene 0
60 0
03 0
70 0
05 0
60 0
04 1
10 0
06 1
30 0
05
Salicylate 0
60 0
04 0
80 0
06 0
60 0
07 1
20 0
05 1
40 0
05
IPTG ND{ 0
70 0
04 0
60 0
02 1
20 0
06 1
40 0
06
Naphthalene IPTG ND 0
80 0
07 0
60 0
02 1
20 0
07 1
40 0
06
Salicylate IPTG ND 0
80 0
06 0
60 0
02 1
20 0
05 1
50 0
05

*Mean of three values standard deviation.

{ND not determined.

nutrient medium. As evident from Table 1 the recombinant
Ps. putida RKJ5/pRKJ3 enhanced the rate of indigo forma-
tion from indole by up to 30% compared with its wild-
type strain Ps. putida RKJ1. Furthermore, the rates of
indigo formation from indole by the recombinant E. coli
strains were almost 2
5 times higher compared with that of
cloned Ps. putida strains (Table 1) which could be due to
the better expression of the genes responsible for indigo
production in E. coli strains compared with the Ps. putida
strains which are otherwise genetically more compatible
hosts. This may be due to the fact that the genes encoding
indole production in E. coli strains might be acting syner-
gistically with the genes encoding the process of biotrans-
formation of indole to indigo. The overall study revealed
that E. coli AB1157/pRKJ3 was the most efcient of the
hosts tested for the expression of plasmid-encoded genes
for the production of indigo and indirubin originally
obtained from Ps. putida RKJ1. On the contrary, O'Connor
and Hartmans (1998) have reported the downregulation of
Pseudomonas genes encoding naphthalene dioxygenase in E.
coli constructs which in turn resulted in lower rates of
indigo formation by the latter.
The apparent Km values of indigo formation from Ps.
putida RKJ1, Ps. putida RKJ5/pRKJ3, Ps. putida KT2442/
pRKJ3, E. coli TB1/pRKJ3 and E. coli AB1157/pRKJ3
were determined as 0
22, 0
15, 0
10, 0
21 and 0
20 mmol
l1, respectively. The apparent Km values of indigo forma-
tion as determined from MichaelisMenten curves indi-
cated that higher indole concentration (> 0
5 mmol l1) is
toxic to the bacteria thus hindering the biotransformation
of indole to indigo. The maximum rates of indigo forma-
tion as calculated from the LineweaverBurk plot were
0
60, 0
90, 0
60, 1
00 and 1
40 nmol min1 mg dry
biomass1, respectively. The maximum rates of indigo for-
mation were exhibited by E. coli AB1157/pRKJ3 which
was almost two times greater than that given by wild-type
Ps. putida strain, and 1
52 times greater than that given by
the recombinant Ps. putida strains. No similar data is avail-
able in the literature for the comparison of apparent Km
values and indigo formation rates from naphthalene-
degrading strains. Indigo formation under different indu-
cing conditions also revealed that both E. coli strains pro-
duced indigo more efciently than the Ps. putida strains
under all the conditions, especially when cells were induced
with salicylate IPTG.
ACKNOWLEDGEMENTS
The authors thank Mrs Gitika Maingi for technical assis-
tance and Mr Ashvini Chauhan for fruitful discussions.
This work was supported by CSIR and DBT, Government
of India. This is IMTECH communication no. 045/99.
REFERENCES
Bachmann, B.J. (1972) Pedigrees of some mutant strains of
Escherichia coli K-12. Bacteriological Reviews 36, 525557.
Doukyu, N., Arai, T. and Aono, R. (1998) Effects of organic sol-
vents on indigo formation by Pseudomonas sp. strain ST-200
grown with high levels of indole. Bioscience Biotechnology and
Biochemistry 62, 10751080.
Eaton, R.W. and Chapman, P.J. (1995) Formation of indigo and
related compounds from indolecarboxylic acids by aromatic
acid-degrading bacteria: chromogenic reactions for cloning
genes encoding dioxygenases that act on aromatic acids. Journal
of Bacteriology 177, 69836988.

= 2000 The Society for Applied Microbiology, Letters in Applied Microbiology, 31, 59

Ensley, B.D., Ratzkin, B.J., Osslund, T.D., Simon, M.J.,
Wackett, L.P. and Gibson, D.T. (1983) Expression of naphtha-
lene oxidation genes in Escherichia coli results in the biosynth-
esis of indigo. Science 222, 167169.
Franklin, F.C.H., Bagdasarian, M., Bagdasarian, M.M. and
Timmis, K.N. (1981) Molecular and functional analysis of the
TOL plasmid pWWO from Pseudomonas putida and cloning of
genes for the entire regulated aromatic ring meta-cleavage path-
way. Proceedings of the National Academy of Sciences of the USA
78, 74587462.
Hart, S., Koch, K.R. and Woods, D.R. (1992) Identication of
indigo-related pigments produced by Escherichia coli containing
a cloned Rhodococcus gene. Journal of General Microbiology 138,
211216.
Hill, R., Hart, S., Illing. N., Kirby. R. and Woods, D.R. (1989)
Cloning and expression of Rhodococcus genes encoding pigment
production in Escherichia coli. Journal of General Microbiology
135, 15071513.
Jain, R.K., Bayly, R.C. and Skurray, R.A. (1984) Characterization
and physical analysis of a 3,5-xylenol degradative plasmid in
Pseudomonas putida. Journal of General Microbiology 30, 3019
3028.
Li, X., Robbins, J.W. Jr and Taylor, K.B. (1979) Effect of the
levels of dissolved oxygen on the expression of recombinant
proteins in four recombinant Escherichia coli strains. Journal of
Industrial Microbiology 9, 110.
Murdock, D., Ensley, B.D., Serdar, C. and Thalen, M. (1993)
Construction of metabolic operons catalyzing the de novo bio-
synthesis of indigo in Escherichia coli. Bio/Technology 11, 381
386.
= 2000 The Society for Applied Microbiology, Letters in Applied Microbiology, 31, 59
INDIGO PRODUCTION BY BACTERIA 9
O'Connor, K.E., Dobson, A.D. and Hartmans, S. (1997) Indigo
formation by microorganisms expressing styrene monooxygen-
ase activity. Applied and Environmental Microbiology 63, 4287
4291.
O'Connor, K.E. and Hartmans, S. (1998) Indigo formation by
aromatic hydrocarbon-degrading bacteria. Biotechnology Letters
20, 219223.
O'riel, P.J. and Kim, I.C. (1998) Method for production of indigo
and indirubin dyes. US Patent US 5834297.
Rani, M., Prakash, D., Sobti, R.C. and Jain, R.K. (1996) Plasmid
mediated degradation of a o-phthalate and salicylate by a
Moraxella sp. Biochemical and Biophysical Research
Communications 220, 377381.
Samanta, S.K., Rani, M. and Jain, R.K. (1998) Segregational and
structural instability of recombinant plasmid carrying genes for
naphthalene degrading pathway. Letters in Applied Microbiology
26, 265269.
Strandberg, G.W., Abraham, T.J. Jr and Fraizer, G.C. (1986)
Phenanthrene degradation by Beijerinckia sp. B8/36.
Biotechnology and Bioengineering 28, 142145.
Straskawicz, B., Dahlbeck, D., Keen, N. and Napoli, C. (1987)
Molecular characterization of cloned avirulence genes from race
O and I of P. syringae pv. glycinae. Journal of Bacteriology 169,
57895794.
Yen, K.M., Karl, M.R., Blatt, L.M., Simon, M.J., Winter, R.B.,
Fausset, P.R., Lu, H.S., Harcourt, A.A. and Chen, K.K. (1991)
Cloning and characterization of a Pseudomonas mendocina KR1
gene cluster encoding toluene-4-monooxygenase. Journal of
Bacteriology 173, 53155327.