Chemosphere,Vol.35, No. 7, pp. 1613-1621.

1997
~ Pergamon © 1997ElsevierScienceLtd
All rightsreserved.Printedin GreatBritain
0045-6535/97$17.00+0.00
PII: S0045-6535(97)00219-1
ANALYSIS OF OIL COMPONENTS AND HYDROCARBON-UTILIZING
MICROORGANISMS DURING LABORATORY-SCALE BIOREMEDIATION
OF OIL-CONTAMINATED SOIL OF KUWAIT

Byung-Hoon Cho 1, Hiroyuki Chino 2, Hirokazu Tsuji 2, Takashi Kunito 1, Hideo Makishima 3,

Hiromi Uchidal, Satoshi Matsumoto 1, and Hiroshi Oyaizu 1

1)Department of Applied Biological Chemistry, Graduate School of Agriculture and

Agricultural Life Sciences, Bunkyo-ku, Tokyo 113, Japan
2)Bio-Environmental Engineering Department, Technical Research Institute,
Obayashi, Inc. Kiyose-shi, Tokyo 204, Japan
3)Analytical Research Center, Central Research Laboratory, Japan Energy, Inc.,
Toda-shi, Saitama 335, Japan
(Receivedin Japan 7 February1997;accepted3 April 1997)
Abstract
A huge amount of oil-contaminated soil remains unremediated in the Kuwait desert. The
contaminated oil has the potentiality to cause pollution of underground water and to effect the
health of people in the neighborhood. We have been studying bioremediation of Kuwait oil-
contaminated soil. Chemical analyses of biodegraded compounds and isolation of petroleum
hydrocarbon-decomposing microorganisms were carried out. From the chemical analyses, it was
revealed that the decomposed compounds were mainly saturated fractions from alumina column
chromatography and that the aromatic fractions were not decomposed well. Isolation of bacteria
was carried out for eight kinds of hydrocarbons which are components of crude petroleum (n-
hexadecane, 2,6,10,14-tetramethylpentadecane, 1,4-diisopropylbenzene, naphthalene, 1-
methylnaphthaiene, phenanthrene, anthracene, and perylene). Many of the n-hexadecane- and
2,6,10,14-tetramethylpentadecane- decomposing bacteria were isolated, but aromatic compound-
decomposing bacteria were not enriched. It was concluded that the slow decomposition of
aromatic compounds was due to the low population of aromatic compound-decomposing bacteria
in the Kuwait desert soil. © 1997 Elsevier Science Ltd

1.Introduction
The gulf war started on January 17, 1991 and ended on February 28, 1991 resulting in huge
scale oil contamination of the Kuwait desert soil (1, 2). The Kuwait desert is an important resource
for Kuwaiti that serves as a wild life habitat, and for farming, grazing, and recreation; thus,
1613
1614

remediation of the oil-contaminated desert soil is desired. We have been studying bioremediation
of the Kuwait oil-contaminated soil. In the previous paper (3) we reported that coconut charcoal
enhanced the biodegradation of contaminated hydrocarbons and that the amount of toxic
compounds produced during bioremediation was very small. This paper reports the result of
analyses of biodegraded oil components and isolation of petroleum hydrocarbon-utilizing
microorganisms.

2.Materials and Methods

2.1.Analyses of decomposed oil components
Used soils, mixed amendment materials, and incubation conditions are the same as those in the
previous report (3). Analyses of decomposed oil components were carried out on dichloromethane
extracts of 18-week incubated soil of columns 7 (3) and 10 (3) and of 43-week incubated soils of
column 10. Analyses were also carried out on the oil sludge of oil lake No. 12 and on Kuwait
crude petroleum. About 0.5 g of total petroleum hydrocarbons (TPHs) were extracted from each
sample according to the previous paper (3). The TPHs were separated according to the following
procedure. The n-heptane insoluble fraction of the TPHs was separated as asphaltene. The heptane-
soluble fraction was separated into further fractions by activated alumina column chromatography.
Samples were loaded on an alumina column, and the column was washed with excess water (50

°C). The samples were then eluted with 300 ml of n-heptane, and the eluents were collected as

saturated hydrocarbon fractions. Then, alumina columns were eluted with 300 ml of toluene, and
the eluents were collected as aromatic hydrocarbon fractions. The alumina columns were further
eluted sequentially with 80 ml methanol, 80 ml toluene, and 100 ml methanol, and the total eluents
were collected as resin fractions. The weight of each fraction was measured. Structural group
analyses by Field Desorption Mass-Spectrometry (FDMS, FDMS was done by JEOL AX-505H
Mass Spectrometry) were carried out on saturated and aromatic fractions. Elementary analyses
were carried out for each total TPH using a Perkin Elmer 240C CHN analyzer and a Mitsubishi
Kasei QR02 S analyzer. Gel permeation chromatography (GPC) was carried out for each total TPH
using a Waters ALC/GPC 201.
2.2.Isolation ofpetroleum hydrocarbon-utilizing bacteria
Hydrocarbon-utilizing bacteria were enriched in the liquid medium circulating flasks (Fig. 1)

for 2 months at 30 °(2. Liquid medium of 1000 ml contained 0.8 g KH2PO4, 1.2 g K2HPO4, 1.0 g

NH4NO3, 0.2 g MgSO4 " 7H2 O, 5 mg yeast extract (Difco), 5 mg Fe2(SO4)3, 5 mg Na2MoO4

2H20, and 5 mg MnSO4 • 4H2 O, and the pH was adjusted to 7.0 with NaOH. About 0.1 g of the
 1615

35~ Aluminum cap

I -x

p P w w ~
Oil contaminated soil "-------T~ ~

Glass filter

Silicon rubber

Air

Liquid medium

Fig. 1 Flask for enrichment of hydrocarbon-

decomposing bacteria.

enriched soils was further enriched for 7 days at 30 °C in 10 ml of liquid medium containing 1% n-

hexadecane, 2,6,10,14-tetramethylpentadecane, 1,4-diisopropylbenzene, naphthalene, 1-

methylnaphthalene, phenanthrene, anthracene, or perylene. The enrichments were repeated three
times; then the growing microorganisms were isolated on nutrient agar (Difco) plates. The
purified microorganisms were incubated in the liquid medium containing the respective
hydrocarbons, and the hydrocarbon-utilizing ability was affirmed when apparent growth was
observed.
2.3.Identification of petroleum hydrocarbon-utilizing bacteria
The shape and Gram reaction of cells were observed by a microscope. Instead of physiological
characterization, molecular identification was carried out by reading the nucleotides sequences of
16S ribosomal DNA (16S rDNA). DNAs were extracted from cells by the method of Zhu et al. (4).
The 16S rDNAs were amplified by the polymerase chain reaction (PCR) with two primers
(5'AG'ITrGATCCTGGCTC3' and 5'AAGGAGGTGATCCAGCC3') that hybridize the 3' and 5'
1616

ends of 16S rDNA (5). The amplified DNA was purified with MicroSpin S-400 HR columns
(Pharmacia Biotech, Inc.), and the sequences were determined with the 373A DNA sequencer from
Perkin Elmer, Inc., with dye primers (5'CAGGACTACCAGGGTATCTAAT3' and
5'AGGGqTGCGCTCGTTG3') that hybridize positions 803 to 787 and positions 1100 to 1115 of
the Escherichia coli numbering system of 16S rDNA. Dye primers were synthesized according to
the protocol supplied by Perkin Elmer, Inc. Identification of bacteria from 16S rDNA sequences
was carried out to compare the obtained sequence and DNA database using Genetyx software
(Software Science, Inc., Tokyo, Japan).

3. RESULTS AND DISCUSSION

3.1.Analyses of decomposed oil components
The compositions of the oil sludge of oil-lake No. 12 and of Kuwait crude petroleum are
shown in Table 1. The oil sludge contained more resin fractions and less aromatic fraction than the
crude oil. The increase in the resin fraction and the decrease in the aromatic fraction in the oil
sludge are due to the weathering reactions after the oil spewed out. In Table 2, the compositions of
TPHs extracted from the untreated soil mixture, 18-week incubated soils of columns 7 and 10, and
43-week incubated soil of column 10. The contents of the saturated fractions drastically decreased
from 39.0 to 7.8% in column 10 for 43 weeks, while the contents of resin and lost fractions
increased. The content of the aromatic fraction decreased slightly from 29.6 to 20.9%. It is
concluded that the saturated fractions were degraded rapidly and probably converted to the
compounds of the resin fraction and the aromatic fractions were more resistant to biodegradation
than the saturated fractions. The problem of bioremediation of Kuwait oil-contaminated soil may
be concentrated on the biodegradation of aromatic fractions.

Table 1. Compositions of oil sludge.

Sample Wt % of each fraction

Asphaltene Saturated Aromatic Resin Lost
Oil sludge collected 6.0 54.9 15.6 14.8 8.7
at oil lake No.12
Crude petroleum 2.9 58.7 33.9 4.5 0.2
(Burgan oil)
 1617

Table 2. Compositions of TPHs extracted from soils.

Column Wt % of each fraction

Asphaltene Saturated Aromatic Resin Lost
Control (untreated) 9.7 39.0 29.6 13.2 8.5
7 13.0 21.1 33.4 16.8 15.7
10 14.4 7.8 20.6 29.5 27.7

Soils of columns 7 and 10 were collected after 18 weeks' incubation.

FD-MS spectrometry was carried out for the untreated soil and 18-week incubated soils of
columns 7 and 10 (Table 3). When the soils were incubated, the contents of larger z numbers
decreased clearly in the saturated fractions and slightly in the aromatic fractions. This fact
indicated that the saturated fractions were biodegraded more rapidly than the aromatic fractions.

Table 3. Structual group analysis of biodegraded oil

% of oil component

Fraction Z number*
+2(-6) 0(-8) -2(-10) -4(-12) -6(-14) -8(-16) -10(-18)

Control
Saturated 26 21 13 9 13 10 8
Aromatic 19 14 12 16 14 13 13
Column 7
Saturated 11 15 15 12 19 16 11
Aromatic 16 14 12 16 15 14 14
Column 10
Saturated 15 17 14 11 17 17 10
Aromatic 17 14 12 15 15 13 13

Columns 7 and 10 were sampled after 18 weeks' incubation.

* Z refers to the Z number of the molecular formula for hydrocarbons,CnH2n+Z. A number
without parenthesis means the Z number of saturated hydrocarbons, and a number in parentheses
means the Z number of aromatic hydrocarbons.

Table 4. Elementary analysis of biodegraded oil

Column C H(H/C) N 0 S

Control(untreated) 83.3 11.4(1.64) 0.2 0.8 4.2

7 83.0 10.6(1.53) 0.2 1.3 4.9
10 81.6 10.3(1.51) 0.3 2.7 5.1

Soils of columns 7 and 10 were collected after 18 weeks' incubation.

1618

Therefore, the results of FD-MS correspond with the results of the composition analyses. The
results of elementary analyses are shown in Table 4. The content of sulfur was higher than the
average content of crude petroleum. When the soils were incubated, the content of sulfur
increased slightly. This fact indicated that the compounds containing sulfur were more stable
toward the bioremediation. The results of GPC analyses are shown in Table 5. The average
molecular weight decreased after incubation.

Table 5. Average molecular weight of biodegraded oil

determined by gel permeation chromatography

Molecular weight

Column Top peak Average

Con~ol(un~eated) 880 2070

7 810 1750
10 810 1660

Soils of columns 7 and 10 were collected after 18 weeks' incubation.

3.Zlsolation and identification of petroleum hydrocarbon-utilizing bacteria in the bioremedied

soils
Isolation was carried out using soil samples of 8-month incubated samples from columns 4
and 7. The soil samples were filled in the liquid medium circulating flasks and stored for two
months. On the other hand, some of the soil samples were subjected to the direct enrichment in the
liquid medium containing each petroleum hydrocarbon. Apparent bacterial growth was observed in
the enrichment liquid cultures containing aliphatic compounds (n-hexadecane and 2,6,10,14-
tetramethylpentadecane). However, in the liquid medium with seven aromatic compounds, bacterial
growth was very weak. Three n-hexadecane-utilizingbacteria (strains H-4-1, H-4-2, and H-7-1)
were isolated from the enrichment culture of n-hexadecane. These three strains also utilized
2,6,10,14-tetramethylpentadecane. Strain H-4-1 was isolated from the enrichment in the liquid
medium circulating flask of a soil sample from column 4, and strain H-4-2 was isolated from the
same soil sample by direct isolation. Strain H-7-1 was isolated from the enrichment in the liquid
medium circulating flask of a soil sample from column 7. Approximately 30 strains were isolated
from the enrichment culture of aromatic compounds (1,4-diisopropylbenzene, naphthalene, 1-
methylnaphthalene, phenanthrene, anthracene, and perylene), and they were subjected to a growth
test in the liquid medium containing the respective aromatic compounds. However, all of the
purified strains did not grow in the liquid medium.
 1619

The three n-hexadecane-utilizingbacteria were submitted for identification. The sequences of

16S rRNA positions from 535 to 1092 were determined for three isolated strains. Homology
searches were carried out using these sequences~ and the strains were identified based on the
sequence homology. The sequence of the strain H-4-1 showed higher homology values to the
sequences of Rhodococcus species (95.0% to Rhodococcus rhodochrous, 93.8% to Rhodococcus
equi, 93.7% to Rhodococcus glovelurus, and 93.5% to Rhodococcusfasciens). The sequence of
the strain H-4-2 also showed higher homology values to the sequences of Rhodoeoeeus species
(96.7% to Rhodococeus rhodochrous, 94.4% to Rhodococcus glovelurus, 94.0% to Rhodococcus
equi, and 92.9% to Rhodococcusfasciens). Judging from the sequence homology, the two strains
(H-4-1 and H-4-2) were identified as Rhodococcus species. However, because exactly the same
sequence was not found in the DNA databases (EMBL, GenBank, and DDBJ), these strains were
not accommodated into some special species. The sequence of the strain H-7-1 showed very high
homology value (99.3%) to the sequence ofNocarida asteroides and relatively lower homology
values to those of Rhodococeus species (92.9% to Rhodococcus glovelurus, 92.3% to Rhodococcus
fasciens, and 91.0% to Rhodococcus rhodochrous). The value of the DNA-DNA homology is a
conclusive factor in identifying bacteria; the strains which show values higher than 70% are
included in the same species (6). There are several reports on the correlation between the DNA-
DNA homology and 16S rDNA sequence homology (7, 8). According to the reports, when the
strains show sequence homology higher than 99.5%, the strains show DNA-DNA homology higher
than 70%. Judging from the value (99.3%) of 16S rDNA sequence homology between strain H-7-1
and N. asteroides, strain H-7-1 should not be identified as N. asteroides but identified as a
Nocardia species.
3.3.Problems relevant to the bioremediation of Kuwait oil-contaminated soil
In the previous study (3), commercial petroleum hydrocarbon-utilizingbacteria were applied
to the bioremediation of Kuwait oil-contaminated soil. However, the addition of bacteria did not
enhance the decomposition of contaminated petroleum hydrocarbons. Therefore, it was concluded
that the indigenous bacteria were associated with the biodegradation of the contaminated petroleum
hydrocarbons.
Kuwait is located in an extremely dry area of the world. The average annual precipitation is
recorded at about 120 mm. Therefore, the soil of Kuwait contains accumulated salt and is alkaline.
Salts were also supplied from contaminated oil and were accumulated in the surface layer (3). The
physicochemical characteristics of Kuwait soil are not suitable for the growth of microorganisms.
Many kinds of microorganisms iricluding Gram negative bacteria, Gram positive bacteria, and
fungi are associated with the decomposition of aliphatic compounds (9). On the other hand,
aromatic compounds are considered to be biodegraded mostly by Gram negative bacteria (9).
•620

While Gram positive bacteria and fungi can survive under dry conditions, Gram negative bacteria
dislike dry conditions. Judging from the physicochemical characteristics of Kuwait desert soil, the
numbers of Gram negative bacteria are considered to be very small. Therefore, the Kuwait desert
soil has the most unfavorable characteristics for the biodegradation of aromatic compounds. In fact
the analyses of chemical components of biodegraded oil revealed that the biodegradation of
aromatic compounds was very slow. The fact that the liquid culture enriched with aromatic
compounds did not show apparent growth of microorganisms also indicates the low population of
aromatic compound-utilizing bacteria.

4.Conclusions
In the present study, we carried out laboratory scale bioremediation experiments for Kuwait
oil-contaminated soil and reached the following conclusions from the chemical analysis of
bioremediated oil and isolation of petroleum hydrocarbon-utilizing bacteria.
1)The saturated fraction was biodegraded more rapidly than the other fractions.
2)The aromatic fraction was biodegraded but its biodegradation speed was very slow.
3)Compounds containing sulfur were relatively resistant to the biodegradation.
4)The population of aromatic compound-utilizing bacteria is probably low in the Kuwait oil-
contaminated soil, and this fact may be related to the slow biodegradation of aromatic compounds.

5_..Acknowledgements
This research was carried out as a joint program between the Petroleum Energy Center (PEC),
Tokyo, Japan, and the Kuwait Institute for Scientific Research (KISR), Safat, Kuwait. The authors
are grateful for the kind suppor{ of Drs. N. A1-Awadhi, R. AI-Daher, A. ElNawawy, and M. T.
Balba of KISR. The authors are also grateful for the kind assistance of A. Hirano of the Japan
Bioindustry Association, and of H. Nakamura and K. Hirano of PEC.

References
1. N. AI-Awadhi, M. S. Abdal, E. J. Briskey, and K. Williamson, Assessment of technologies for
the remediation of oil-contaminated soil resulting from exploded oil wells and burning oil
fires in Kuwait, Proceeding of Air and Waste ManagementAssociation Meeting, pp.21-26.
Kansas City, Missouri, USA (1992).
i

2. N. A1-Awadhi, K. J. Williamson, and J. D. Isok, Remediation of Kuwait's oil-contaminated

soils. In Hydrocarbon Contaminatedsoil and groundwater, (Edited by P.T. Kostecki and
E. J. Calabrese), pp.9-21. Lewis Publishers, London (1993).
3. B.-H. Cho, H. Chino, H. Tsuji, T. Kunito, K. Nagaoka, S. Otsuka, K. Yamashita, S.
 1621

Matsumoto, and H. Oyaizu, Chemosphere, submitted, (1997)

4. H. Zhu, F. Qu, and L.-H. Zhu, Nucleic Acids Res., 21, 5279-5280 (1994)
5. H. Sawada, H. Ieki, H. Oyaizu, and S. Matsumoto, Int. J. Syst. Bacterioi., 43, 694-702
(1993)
6. L.G. Wayne, D. J. Brenner, R. R. Colwell, P. A. D. Grimont, O. Kandler, M. I. Krichevsky,
L. H. Moore, W. E. C. Moore, R. G. E. Murray, E. Stackebrandt, M. P. Starr, and H. G.
Truper, Int. J. Syst. BacterioL, 37, 463-464 (1986)
7. A. Hiraishi and Y. Ueda, Int. J. Syst. Bacteriol., 44, 15-23 (1994)
8. R.I. Amann, C. Lin, R. Key, L. Montgomery, and D. A. Stahl, Syst. AppL Microbiol., 15,
23-31 (1992)
9. J.G. Leahy and R. R. Colwell, Microbiol. Rev., 54, 305-315 (1990)