FEMS Microbiology Ecology 43 (2003) 191^194

www.fems-microbiology.org

Biodegradation of the herbicide tri£uralin by bacteria

isolated from soil
Maria De Lourdes Bellinaso a;b , Charles William Greer c , Maria do Carmo Peralba d ,
Joa‹o Anto“nio Pe“gas Henriques e , Christine Claire Gaylarde e;

Downloaded from https://academic.oup.com/femsec/article/43/2/191/533798 by guest on 26 April 2025

a
Department of Biology and Chemistry, UNIJUI, IJUIŁ RS, Brazil
b
Dept. Biochemistry, UFRGS, Porto Alegre, Brazil
c
Biotechnology Research Institute, National Research Council of Canada, Montreal, QC, Canada
d
Institute of Chemistry, UFRGS, Porto Alegre RS, Brazil
e
Department of Biophysics/Biotechnology Center, UFRGS, Av. Bento Goncalves 9500, 91501-970 Porto Alegre RS, Brazil

Received 28 June 2002 ; received in revised form 23 August 2002 ; accepted 4 September 2002

First published online 9 October 2002

Abstract

Trifluralin (K,K,K-trifluoro-2,6-dinitro-N,N-dipropyl-p-toluidine ; TFL) is a pre-emergence, soil-incorporated herbicide that has been in

agricultural use since the early 1960s and is moderately persistent in soil. The purpose of this study was to isolate and characterise TFL-
resistant bacteria from a soil in which this pesticide has been used for the last four decades and to determine their ability to degrade TFL
using HPLC. Eight bacteria were isolated by repeated subculture in liquid medium with TFL as carbon source and a ninth (isolate 9) from
growth around TFL crystals on solid medium. The bacteria from enriched liquid culture were identified by biochemical tests and 16S
rDNA sequencing. In a mineral salts medium with 0.1% succinate, 0.1% yeast extract and 50 mg l31 TFL, reductions in the level of
pesticide of 24.6% for Klebsiella sp., 16.4% for Herbaspirillum sp., 25.0% and 16.0% for two strains of Bacillus sp. and 21.0% for
unidentified isolate number 9 were obtained after 30 days. These were similar to the level obtained using a known TFL-degrading
bacterium, Brevundimonas diminuta (NCIMB 10329). Three Pseudomonas sp. and one Bacillus sp. reduced levels by less than 5%. The five
positive isolates can be used to study the biochemical and molecular biology of TFL biodegradation with the aim of optimising the
degradative ability of one or more of the isolates for future use in bioremediation processes.
? 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.

Keywords : Biodegradation; Bioremediation ; Pesticide; Tri£uralin; Soil bacterium

1. Introduction extractable transformation products and 43% as non-ex-

tractable soil-bound residues. They also identi¢ed 28 deg-
Tri£uralin (TFL) is a pre-emergence, soil-incorporated radation products and suggested a degradation pathway
herbicide that has been in agricultural use since the early for the molecule. Camper et al. [4] studied TFL degrada-
1960s and is widely used for soybean production in Brazil. tion in various types of agricultural soil under aerobic and
It shows moderate persistence in soil. TFL is degraded by anaerobic conditions and found that breakdown was
photolysis and biological activity. The majority of work highly dependent on soil type. In a soil used for the dis-
on its biodegradation has studied soil systems. Golab et al. posal of various pesticides (diuron, TFL, carbofuran),
[6], in an extensive ¢eld study using radiolabelled TFL, only TFL was still detected after many years [10].
found that after one year, 69% of the applied 14 C was Although there has been considerable research on the
present in the top 0^15 cm of soil: 14% as TFL, 12% as degradation of TFL in soil [8], few studies have reported
degradation by isolated microorganisms, a necessary pre-
requisite if the microbial enzymes and pathways involved
are to be discovered. Hamdi and Tew¢k [9] and Carter
* Corresponding author, Tel. : +55 (51) 3316 6026 ;
and Camper [5] isolated a species of Pseudomonas that
Fax : +55 (51) 3316 6050. decomposed TFL in liquid medium containing a supple-
E-mail address : cgaylarde@yahoo.com (C.C. Gaylarde). mentary carbon source. Zeyer and Kearney [21] also used
0168-6496 / 02 / $22.00 ? 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.
PII : S 0 1 6 8 - 6 4 9 6 ( 0 2 ) 0 0 3 8 8 - 4

FEMSEC 1438 5-2-03

192 M.D.L. Bellinaso et al. / FEMS Microbiology Ecology 43 (2003) 191^194

complex media and demonstrated degradation of TFL by TFL that had been allowed to form on the surface of
a species of Candida isolated from soil. Sato [17,18] iso- Thornton’s agar prior to spreading with soil suspension.
lated two bacteria from agricultural soil contaminated
with TFL. The isolates degraded the molecule and three 2.4. Characterisation of isolates
of the degradation products cited by Golab et al. [6] were
detected. We aimed to isolate TFL-resistant bacteria from The isolates were identi¢ed by biochemical tests (Gram
a Brazilian soil in which this pesticide has been used for staining, the GNI card (Vitek Systems, Hazelwood, MO,
four decades, characterise the resistant isolates and deter- USA), and the API 20NE identi¢cation system (bioMe¤r-
mine their ability to degrade TFL. iuex, France)) and by 16S rDNA sequencing of PCR
products using extracted genomic DNA and the primers
F1 (5P-GAG TTT GAT CCT GGC TAC G-3P) and R13
2. Materials and methods (5P-AGA AAG GAG GTG ATC CAG CC-3P). The PCR
was performed according to Larame¤e et al. [11] in a total
2.1. Soil samples volume of 50 Wl. Samples were subjected to 30 cycles of

Downloaded from https://academic.oup.com/femsec/article/43/2/191/533798 by guest on 26 April 2025

ampli¢cation consisting of 1 min denaturation at 94‡C,
The soil was a silty clay agricultural soil, typical of 1 min annealing at 65‡C and 1 min extension at 72‡C.
North-eastern Rio Grande do Sul. It had been used for PCR products were sequenced with the Abi prism dye
the cultivation of soya and wheat for approximately four terminator cycle sequencing ready reaction kit (Perkin-El-
decades and TFL had been employed on the land through- mer), using primers F1 or R6 (5P-CCG TCA ATT CAT
out this time. Three soil samples from a depth of 10^15 cm TTG AGT TT-3P). The reaction was carried out in 10 Wl,
were taken just before herbicide treatment, prior to the with 25 cycles of 10 s at 96‡C, 5 s at 50‡C and 4 min at
planting of soya. The soil was dried at room temperature, 60‡C. Nucleotide sequences were obtained in an ABI
mixed thoroughly, sieved in a 2 mm sieve to remove stones Prism 373 automatic sequencer. The partial 16S rDNA
and plant material and quartered. One part was used for sequences (400^500 bp) were submitted for comparison
the analyses. to the GenBank database using the BLAST algorithm
[1], the EMBL database using the FASTA algorithm
2.2. Chemicals [16], and to the Ribosomal Database Project (RDP) [12].

Tri£uralin (K,K,K-tri£uoro-2,6-dinitro-N,N-dipropyl-p- 2.5. Microcosms for degradation of Tri£uralin

toluidine), pure grade (98%), and tri£uralin emulsion
(445 g l31 ) were donated by Milenia Agrocie“ncias (Ta- Five di¡erent media, containing 100 mg l31 or 50 mg l31
quari, Rio Grande do Sul, Brazil). The composition of of pure TFL or TFL emulsion and inoculated with 100 Wl
the emulsion was not supplied. of bacterial suspension (A600 = 0.1), were used. The bottles
were hermetically sealed to prevent leakage of liquids and
2.3. Isolation of microorganisms resistant to Tri£uralin of TFL, which is moderately volatile, placed in the dark
and incubated at 28‡C, 120 rpm, for 30 days. The ¢ve
The microorganisms were isolated in mineral medium media tested were based on Greer et al. [7]. The basic
containing 50 and 100 mg l31 TFL (pure grade). The basal medium contained 13 mM K2 HPO4 , 6.4 mM NaH2 PO4 ,
medium was Bushnell^Haas mineral salts medium [3]; 8.33 mM (NH4 )2 SO4 , 0.395 mM MgSO4 , 1 M AlK(SO4 )2 ,
TFL was used as sole carbon source in this medium and 10 mM FeSO4 , 10 mM ZnSO4 , 10 mM MnSO4 , 1 mM
as carbon and nitrogen source in medium without the CuSO4 , 1 mM Co(NO3 )2 , 10 mM Ca(NO3 )2 , 2 mM
0.1% ammonium nitrate component. TFL was added to NaMoO4 . Alterations were made to this as follows : me-
the £asks as a solution in methanol and the solvent was dium 1, without inorganic nitrogen source, with TFL (100
allowed to evaporate for 30 min before adding the sterile mg l31 ); medium 2, with TFL (100 mg l31 ); medium 3,
mineral medium. without inorganic nitrogen source, with 0.1% succinate
10 g soil were added to 20 ml medium in 125 ml £asks. and 100 mg l31 TFL; medium 4, with 0.1% succinate
These were incubated at 28‡C in a rotary shaker (120 and 100 mg l31 TFL; medium 5, with 0.1% succinate,
rpm), with £asks protected from the light to avoid photo- 0.1% yeast extract and 50 mg l31 TFL.
degradation of TFL. An aliquot of 1 ml was subcultured Two controls were used: a chemical control, without
to fresh medium every six days for 36 days and growth microorganisms (used for all media), and a metabolic con-
monitored by plating on the same media solidi¢ed with trol without TFL for medium 5 only, to check for the
agar, and on Thornton’s agar medium [15] containing 50 presence of any metabolites produced in this richer me-
mg l31 TFL. Colonies that grew best on the latter medium dium in the absence of TFL. Isolates 1 to 8 were incubated
were isolated and stored at 0‡C on the same medium. One in all the media. Isolate 9 (from TFL crystals) and the
isolate was obtained from the growth around crystals of positive control were incubated only in medium 5. The

FEMSEC 1438 5-2-03

 M.D.L. Bellinaso et al. / FEMS Microbiology Ecology 43 (2003) 191^194 193

Table 1 ing agent is a strategy that has been used extensively to

Partial identi¢cation of bacterial isolates based on biochemical tests and
increase the aqueous concentrations of poorly soluble
16S rDNA gene sequence analysis
xenobiotics [14,19].
Isolate Results of 16S rDNA sequence(primer/% homology to There was no signi¢cant di¡erence in the amount of
number biochemical closest database match; accession number)
test kits
TFL in the chemical controls after 30 days incubation,
indicating that volatilisation and chemical degradation
1 Klebsiella oxytoca Klebsiella oxytoca (F1/93% ; AF440521)
2 No id Herbaspirillum seropedicae (F1/100% ;
were not important mechanisms of loss in these experi-
AB027694) ments. Apart from the controls described, a chemical con-
3 Pseudomonas Pseudomonas montellii (R6/98%; trol with heat-killed bacteria (autoclaved at 121‡C for 20
aeruginosa AB021409, AF064458) min before inoculation into the £asks) also gave negative
4 Bacillus sp. Bacillus megaterium (F1/98% ; AF142677) results. After 30 days, TFL concentrations, measured by
5 Bacillus sp. Bacillus megaterium (F1/99% ; AF142677)
6 Bacillus sp. Bacillus megaterium (F1/99% ; AY030338.1)
gas chromatography, were not signi¢cantly di¡erent from
7 Pseudomonas Pseudomonas sp. (R6/98%; AJ011507, the chemical control. The increased disappearance of TFL
aeruginosa AB013254) over the chemical control in inoculated £asks was: 24.6%

Downloaded from https://academic.oup.com/femsec/article/43/2/191/533798 by guest on 26 April 2025

8 Pseudomonas Pseudomonas montellii (R6/98%; for Klebsiella sp.; 16.4% for Herbaspirillum sp.; 25.0% for
aeruginosa AB021409, AF064458) Bacillus sp. 2; 16.0% for Bacillus sp. 3; 21.0% for uniden-
ti¢ed isolate 9. The percentage disappearance for the pos-
itive metabolic control, Brevundimonas diminuta, was sim-
positive control was Brevundimonas diminuta (NCIMB ilar at 26.0%. The degradation rate of TFL by this
10329), a known TFL-degrading bacterium [9]. organism has not previously been recorded in the litera-
ture.
2.6. Tri£uralin quanti¢cation The other four isolates, identi¢ed by these tests as three
strains of Pseudomonas and one Bacillus sp., increased the
TFL was quanti¢ed by HPLC (Varian 5000, Cds-111L disappearance of TFL by less than 5%. The one-way AN-
microprocessor and UV-50 detector). 8 ml of methanol OVA showed that there was a signi¢cant di¡erence be-
was added to the microcosms after 30 days incubation. tween the values of the means (F = 146.35, P 6 0.01).
The bottles were hermetically sealed, sonicated for 5 min There was excellent correlation between duplicate £asks
and centrifuged at 1000Ug for 5 min. The supernatant (R = 0.998). Species of the genus Pseudomonas have been
was ¢ltered with a Millipore ¢lter (0.22 Wm) and 50 Wl reported in the literature as degraders of TFL, dinitroani-
injected directly into the HPLC. The mobile phase con- line and other nitroaromatic compounds [17,18]. However,
sisted of an isocratic ratio of 75% acetonitrile/25% water in our experiments the isolates identi¢ed as Pseudomonas
at a £ow rate of 1 ml min31 . The HPLC system was did not appear to degrade TFL.
equipped with a C18 column (25 cm, 4.6 mm, 5 Wm) Chromatograms for degrading isolates showed reduced
and the UV detector was operated at a wavelength of TFL peaks and extra peaks at shorter retention times ; the
275 nm. Peak areas and retention times were compared latter are putative degradation products of TFL. A more
to a standard curve and the percentage reduction in detailed analysis of these peaks by mass spectroscopy and
TFL calculated attributing 100% to the value of the chem- comparison to appropriate reference compounds would be
ical control peak. Duplicate samples were each analysed necessary to identify these products as TFL metabolites.
twice and the results analysed statistically using the one- All the genera isolated, Pseudomonas, Bacillus, Klebsiel-
way ANOVA. la and Herbaspirillum, are representative of the degrada-
tive micro£ora of the soil [20] and the ¢rst three have been
cited as biodegraders of xenobiotics [13,17,18]. We found
3. Results and discussion that positive degradation occurred only in microcosms
containing medium 5. This could have been due to :
Nine bacterial isolates were selected for the microcosm (1) increased bioavailability due to the use of the emulsi-
tests. The tentative identi¢cation of eight of these (isolated ¢ed form of TFL, which has been shown to be more read-
by repeated subculture in liquid medium containing TFL) ily degraded in some soils [14] ; (2) increased nutrients
is shown in Table 1. Based on the test results, it is possible provided by the yeast extract ; (3) reduced toxicity of
that isolates 3 and 8 may be the same organism, as may be TFL or its products because of the lower concentration
isolates 5 and 6. Only four of the eight and the strain used [2]. Zeyer and Kearney [21] showed no e¡ect of
isolated on TFL crystals (isolate 9) caused degradation changing TFL concentration on degradation levels in
and only in medium 5. This medium contained more com- complex media after 21 days. However, Carter and Camp-
plex nutrients, including vitamins, not present in the other er [5] demonstrated that two soil isolates capable of de-
media, and a lower concentration of TFL (50 mg l31 ), grading TFL grew optimally at 50 mg l31 TFL; they sug-
which was added to the medium in its emulsi¢ed form gested that 100 mg l31 could be inhibitory both to growth
to increase its bioavailability. The addition of an emulsify- and to biodegradation.

FEMSEC 1438 5-2-03

194 M.D.L. Bellinaso et al. / FEMS Microbiology Ecology 43 (2003) 191^194

Our results show that bacteria isolated from Brazilian anaerobic degradation of pro£uralin and tri£uralin. J. Environ. Sci.
Health Part B 15, 457^473.
soils in which TFL has been used were able to degrade
[5] Carter, G.E. and Camper, N.D. (1975) Soil enrichment studies with
TFL to an extent comparable to the only TFL degrader tri£uralin. Weed Sci. 23, 71^74.
found in a commercial culture collection. Three of the [6] Golab, T., Althaus, W.A. and Wooten, H.L. (1979) Fate of (14-
isolates increased degradation by approximately 20%, sim- C)Tri£uralin in soil. J. Agric. Food Chem. 27, 163^179.
ilar to the known TFL degrader. This activity was seen in [7] Greer, C.W., Hawari, J. and Samson, R. (1990) In£uence of environ-
mental factors on 2,4-dichlorophenoxyacetic acid degradation by
only one of the ¢ve media used, con¢rming that the me-
Pseudomonas cepacia isolated from peat. Arch. Microbiol. 154,
dium can have a strong in£uence on biodegradation and 317^322.
that it is important to use various media when testing for [8] Grover, R., Wolt, J.D., Cessna, A.J. and Schiefer, H.B. (1997) Envi-
xenobiotic biodegradation. The isolates can be used to ronmental fate of tri£uralin. Rev. Environ. Contam. Toxicol. 153, 1^
study the biochemical and molecular biology of the deg- 64.
[9] Hamdi, Y.A. and Tew¢k, M.S. (1969) Decomposition of the herbi-
radation of TFL, leading to the possibility of optimising
cide tri£uralin by a Pseudomonas. Acta Microbiol. Polonica Ser. B 1
the degradative ability of one or more of the isolates for (18), 83^84.
future use in bioremediation processes. [10] Johnston, W.H. and Camper, N.D. (1991) Microbial degradative

Downloaded from https://academic.oup.com/femsec/article/43/2/191/533798 by guest on 26 April 2025

activity in pesticide pretreated soil. J. Environ. Sci. Health Part B
26 (1), 1^14.
[11] Larame¤e, L., Lawrence, J.R. and Greer, C.W. (2000) Molecular anal-
Acknowledgements
ysis and development of 16S rDNA oligonucleotide probes to char-
acterize a diclofop-methyl-degrading bio¢lm consortium. Can. J. Mi-
The gas chromatography was performed by the Pesti- crobiol. 46, 133^142.
cide Analysis Laboratory of UNIJUIŁ (Universidade da [12] Maidak, B.L., Cole, J.R., Lilburn, T.G., Parker Jr., C.T., Saxman,
Regia‹o Noroeste do Estado do Rio Grande do Sul). We P.R., Farris, R.J., Garrity, G.M., Olsen, G.J., Schmidt, T.M. and
Tiedje, J.M. (2001) The RDP-II (Ribosomal Database Project). Nu-
also wish to thank the Hospital de Clinicas de Porto Ale-
cleic Acids Res. 29, 173^174.
gre for performing the biochemical identi¢cation tests and [13] Melo, I.S. and Azevedo, J.L. (1997) Microbiologia Ambiental. EM-
Nathalie Fortin and Louise Paquet (BRI-NRC) for per- BRAPA-CNPA, Brazil, pp. 217^218.
forming 16S rDNA sequence analysis and helping with [14] Nikolova, P. and Ward, O.P. (1993) Whole cell biocatalysis in non-
HPLC quanti¢cation of TFL, respectively. MLB was sup- conventional media. J. Ind. Microbiol. 12, 76^86.
[15] Parkinson, D., Gray, T.R.G and Williams, S.T. (1971) Methods for
ported by a CAPES doctoral grant and by research grants
Studying the Ecology of Soil Microorganisms. Adlard, Oxford.
from FAPERGS and the Genotoxicity Laboratory, Bio- [16] Pearson, W.R. and Lipman, D.J. (1988) Improved tools for biolog-
technology Center/UFRGS. ical sequence analysis. Proc. Natl. Acad. Sci. USA 85, 2444^2448.
[17] Sato, Y. (1992) Degradation of tri£uralin by bacteria isolated from
soil. Weed Res. Japan 37, 213^219.
References [18] Spanggord, R.J., Spain, S.F., Nishino, S.F. and Mortelmans, K.E.
(1991) Biodegradation of 2,4-dinitrotoluene by a Pseudomonas sp.
[1] Altschul, S.F., Madden, T.L., Scha«¡er, A.A., Zhang, J., Zhang, Z., Appl. Environ. Microbiol. 57, 3200^3205.
Miller, W. and Lipman, D.J. (1997) Gapped BLAST and PSI- [19] Tiehm, A. (1994) Degradation of polycyclic aromatic hydrocarbons
BLAST: a new generation of protein database search programs. Nu- in the presence of synthetic surfactants. Appl. Environ. Microbiol.
cleic Acids Res. 25, 3389^3402. 60, 258^263.
[2] Boyette, K.D., Moorman, T.B. and Koskinen, W.C. (1988) E¡ects of [20] Van Elsas, J.D., Trevors, J.T. and Wellington, E.M.H. (1997) Mod-
tri£uralin and metabolites on decomposition of selected substrates by ern soil microbiology. In: Topp, E., Vallaeys, T., Soulas, G. (Eds.),
soil microorganisms. Biol. Fertil. Soils. 6, 100^105. Pesticides Microbial Degradation and E¡ects on Microorganisms.
[3] Bushnell, C.D. and Haas, H.F. (1941) The utilization of certain hy- Marcel Dekker, New York, pp. 547^575.
drocarbons by microorganisms. J. Bacteriol. 41, 653^673. [21] Zeyer, J. and Kearney, P.C. (1983) Microbial dealkylation of tri£ur-
[4] Camper, N.D., Stralka, K. and Skipper, H.D. (1980) Aerobic and alin in pure culture. Pestic. Biochem. Physiol. 20, 10^18.

FEMSEC 1438 5-2-03