Molecular diversity of microbial community in acid mine drainages of Yunfu sulfide mine.

Abstract

Two acid mine drainage (AMD) samples were studied by a PCR-based cloning approach, which were from Yunfu sulfide mine in Guangdong province, China. A total of 15 operational taxonomic units (OTUs) were obtained from the two AMD samples. The percentage of overlapped OTUs in two AMD samples was 42.1%. Phylogenetic analysis revealed that the bacterium in the two samples fell into four putative divisions, which were Nitrospira, alpha-Proteobacteria, beta-Proteobacteria, and gamma-Proteobacteria four families. Organisms of genuses Acidithiobacillus and Gallionella, which were in gamma-Proteobacteria family and beta-Proteobacteria family, respectively, were dominant in two samples. The proportions of clones affiliated with Gallionella in each sample were 47.2% (G2) and 16.9% (G1). The result suggested that organisms of Gallionella were a very important composition in microbial communities of the two AMD samples we studied. In addition, the PCR amplification of archaeal 16S rDNA genes form these two AMD samples have been performed with two sets of archaea-specific primers, but no PCR product found.

Full text

Page 1

ORIGINAL PAPER
Molecular diversity of microbial community in acid mine
drainages of Yunfu sulfide mine
Zhiguo He Æ Æ Shengmu Xiao Æ Æ Xuehui Xie Æ Æ Hui Zhong Æ Æ Yuehua Hu Æ Æ
Qinghua Li Æ Æ Fenglin Gao Æ Æ Guiyuan Li Æ Æ Jianshe Liu Æ Æ Guanzhou Qiu
Received: 9 July 2006/Accepted: 16 October 2006/Published online: 19 December 2006
? Springer 2006
Abstract
were studied by a PCR-based cloning approach, which
were from Yunfu sulfide mine in Guangdong province,
China. A total of 15 operational taxonomic units
(OTUs) were obtained from the two AMD samples.
The percentage of overlapped OTUs in two AMD
samples was 42.1%. Phylogenetic analysis revealed
that the bacterium in the two samples fell into four
putative divisions, which were Nitrospira, a-Proteo-
bacteria, b-Proteobacteria, and c-Proteobacteria four
families. Organisms of genuses Acidithiobacillus and
Gallionella, which were in c-Proteobacteria family and
b-Proteobacteria family, respectively, were dominant in
two samples. The proportions of clones affiliated with
Gallionella in each sample were 47.2% (G2) and 16.9%
(G1). The result suggested that organisms of Gallio-
nella were a very important composition in microbial
communities of the two AMD samples we studied. In
addition, the PCR amplification of archaeal 16S rDNA
Two acid mine drainage (AMD) samples
genes form these two AMD samples have been per-
formed with two sets of archaea-specific primers, but
no PCR product found.
Keywords
AMD
Microbial diversity ? Ecology ? RFLP ?
Introduction
Dissolution of sulfide ores exposed to oxygen, water,
and microorganisms results in acid production and
environmentally detrimental
(AMD) (Nordstrom and Alpers 1999). Microorgan-
isms that are able to develop under extreme conditions,
especiallyiron-oxidizing
responsible for the solubilization of metals from sulfide
minerals in acidic environments via direct action of
their enzymes or indirectly through chemical action of
their metabolic products (Southam and Beveridge
1992; Wulf-Durand et al. 1997; Wielinga et al. 1999).
Because of the limited types of substrates available
in mining environments, the biotopes were initially
expected to be extremely poor with respect to the
diversity of the microbial flora. However, cultivation-
based studies revealed a great diversity of the micro-
bial community in AMD (Johnson 1998; Hallberg and
Johonson 2001). The presence of various bacterial
species including common prokaryotic chemolitho-
trophs other than Acidithiobacillus ferrooxidans, such
as A. thiooxidans, A. caldus, and Leptospirillum fer-
rooxidans and so on, has been reported in mining
environments. The presence of archaea including a
group of sulfur and/or iron-oxidizers, such as Sulfolo-
bus, Acidianus, Metallosphaera, Sulfurisphaera, and
acidminedrainage
chemolithotrophics are
Communicated by K. Horikoshi.
Zhiguo He, Shengmu Xiao, Xuehui Xie, and Hui Zhong equally
contributed to this work.
Z. He ? S. Xiao ? X. Xie ? Y. Hu (&) ?
Q. Li ? F. Gao ? J. Liu ? G. Qiu
School of Resources Processing and Bioengineering,
Central South University, Changsha 410083, China
e-mail: hzg13@163.com
H. Zhong ? G. Li (&)
Cancer Research Institute, Central South University,
Changsha 410078, China
e-mail: hzg13@163.com
123
Extremophiles (2007) 11:305–314
DOI 10.1007/s00792-006-0044-z

Page 2

Ferroplasma has also been reported in acidic environ-
ments (Fuchs et al. 1995, 1996; Kurosawa et al. 1998;
Edwards et al. 2000; Golyshina et al. 2000).
Cultivation-based analysis is not considered a suit-
able method for characterizing microbial communities
(Ward et al. 1990; Delong 1992), and researchers have
used methods based on the analysis of 16S rDNA se-
quences to study microbial communities of acidic
environments (Edwards et al. 1999; Bond et al. 2000;
Burton and Norris 2002; Simmons and Norris 2002;
Baker and Banfield 2003).
Tobetterunderstandthecompositionsandstructures
of microorganisms in different AMD environments, a
PCR-basedcloningapproachwasusedinthiswork.Two
AMD samples from Yunfu sulfide mine were studied.
The Yunfu sulfide mine was in Guangdong province,
China, which began to mine pyrite for sulfuric acid
manufacture since 1988, for its high concentration of
sulfur and low concentrations of other elements. The
microbial communities in AMD samples from Yunfu
sulfide mine have not been investigated to date yet.
Materials and methods
Sites description and samples collection
Samples were colleted from Yunfu sulfide mine, in
Guangdong province, China. The mine had mainly
produced pyrite for sulfuric acid manufacture since
1988, for its high concentration of sulfur and low con-
centrations of other elements.
There were three aqueous AMD samples collected
from two separated sites in Yunfu sulfide mine. pH and
temperature were same in two samples, which were 2.5
and 25.0?C, respectively. One of two samples was
named as G1, from which the clones with 16S rDNA
inserts were given the prefix GY. Similarly, the other
sample was named as G2 and from which clones with
16S rDNA inserts were given the prefix G.
Ten-liter water sample was colleted from each posi-
tion. Samples were processed within 24 h after collec-
tion. Two water samples were filtered through 0.22 lm
hyper filtration membrane with vacuum pump, respec-
tively. The sediments on the membrane were washed by
sterile deionized water twice. Then the sediments were
stored at –70?C for reservation. The filtered water
samples were prepared for chemical analysis.
Chemical analysis of water samples
The element analysis of filtered water samples was
carried out by Inductively Coupled Plasma-Atomic
Emission Spectrometry (ICP-AES).
Twenty-nine elements were tested in each water
sample. They were as following: Hg, As, P, Ni, Co, Cr,
Be, Ti, W, Zn, In, Mg, Mn, Ca, S, Mo, Bi, Au, Fe, Si,
Cu, Sn, Sb, Cd, Ga, Pt, Al, Ag.
DNA extraction and purification
Extraction of nucleic acids was according to procedure
described by Zhou et al. (1996). Five gram of sediment
was mixed with 13.5 ml extraction buffer (0.1 M
phosphate [pH 8.0], 0.1 M EDTA, 0.1 M EDTA,
1.5 M NaCl, 1% CTAB) and 50 ll protenase K
(10 mg/ml) in 50 ml centrifuge tube, then incubated at
37?C for 30 min. 1.5 ml of 20% SDS was added and
mixed gently, then incubated at 65?C for 2 h. The
mixture was centrifuged and the supernatant was
transferred into a new 50 ml of centrifuge tube. The
soil pellet was resuspended with extraction buffer, and
0.5 ml 20% SDS was added. The mixture was incu-
bated at 65?C for 15 min, then centrifuged and the
supernatant was collected and combined with the
previous supernatant. The combined supernatant was
extracted with chloroform. 2-Isopropanol was added to
the supernatant collected and then mixed gently. The
mixture was kept at the room temperature for an hour
or overnight, then centrifuged. The pellet was washing
with 70% ethanol and dissolved with 200–500 ll sterile
water. By using combined methods that included
grinding, freezing and thawing, and treatment with
sodium dodecyl sulfate, various types of bacterial could
been effectively lysed. The crude DNA was purified by
using Wizard plus sv Minipreps DNA purification
system (Promega Corporation, USA) and quantified
by ethidium bromide-UV detection on an agarose gel.
PCR and fractionation of 16S rDNA genes
Bacterial 16S rDNA genes were amplified with the
primer set was 1492R (5¢-CGGCTACCTTGTTACG-
ACTT-3¢), and 27F (5¢-AGAGTTTGATCCTGGCTC
AG-3¢) (Lane 1991). A gene amp (Biometra, T-Gran-
dient, Genman) was used to incubate reactions through
an initial denaturation at 94?C for 2 min, followed by
35 cycles of 94?C for 40 s, 55?C for 30 s, and 72?C for
1 min, and completed with an extension period of
10 min at 72?C. Products from the amplification reac-
tions of expected size (about 1,500 bp) were pooled
and purified before ligation later.
PCR amplification of archaeal 16S rDNA genes was
carried out following the PCR reactions described as
above with two different sets of archaea-specific
primers, which were as follows: S-D-Arch-0025-a-S-17
306Extremophiles (2007) 11:305–314
123

Page 3

(5¢-CTGGT TGATCCTGCCAG-3¢) (Robb et al. 1995)
or S-D-Arch-0344-a-S-20 (5¢-ACGGGGCGCAGCAG
GCGCGA-3¢) (Weisburg et al. 1991) with S-*-Univ-
1517-a-A-21 (5¢-ACGGCTACC TTGTTACGACTT-3¢)
(Raskin et al. 1994) to yield 1,500 or 1,120 bp PCR
products, respectively.
Cloning, RFLP, and sequencing
The purified PCR products were ligated to the vector
PGEM-T (Promega Corporation), and used to trans-
form DH5a competent host cells. About 120 white
colonies were randomly selected from each library.
The transformation efficiency in our study was around
5 · 108 cfu/lg DNA. And we used an external control
to track efficiency, offered by the PGEM-T vector
system (Promega company). For restriction fragment
length polymorphism (RFLP) and sequencing, the
inserted fragments were amplified with the vector-
specific T7 and SP6 primers. These unpurified PCR
products were digested with two restriction endonuc-
leases AfaI and MspI (TaKaRa Biotechnology Co,
Ltd.), incubated at 37?C for 3 h. The restricted frag-
ments were separated by gel electrophoresis in 3.0%
agarose with ethidium bromide staining and observed
on UV illumination. RFLP patterns were identified
and grouped, and representative clones were selected
for nucleotide sequencing.
Phylogenetic analysis
Phylogenetic affiliations of the partial sequences were
initially estimated using the program BLAST (Basic
alignment search tool) (Bond et al. 2000). Similarity of
partial sequences was determined using ARB (a soft-
ware environment for sequence data) (Strunk and
Ludwig 1995). The initial phylogenetic trees were
based on all available sequences and were constructed
by using the DNA distance program Neighbor-Joining
with Felsenstein Correction in ARB (Smith et al.
1994). Based on the initial phylogenetic results,
appropriate subsets of 16S rDNA sequences were se-
lected and subjected to a final phylogenetic analysis
with CLUSTAL X.
Statistical methods
The rarefaction analysis was performed with SigmaPlot
software. An exponential model, y = a · [1 – exp
(–b · x)], was used with SigmaPlot 8.0 nonlinear
regression software to fit the clone distribution data.
Nucleotide sequence accession numbers
Sequences have been submitted to GenBank with
accession numbers are as follow: DQ480487 (G77),
DQ480488 (GY28), DQ480486 (G74), DQ480485
(G51-19),DQ480484 (G31),
DQ480481 (G5), DQ480482 (G71), DQ480479 (G28),
DQ480480 (G44), DQ480478 (G51-12), DQ480476
(G27), DQ480477 (G66), DQ480475 (G52), DQ480474
(G8).
DQ480483(G24),
Results
Biogeochemical properties of two AMD samples
Although pH and temperature in two AMD samples
both were 2.5 and 25.0?C, respectively, elements’
concentration of them was very different by analysis of
ICP-AES. Sample G2 had higher elements’ concen-
tration than those in sample G1, except element cop-
per. The data of 29 elements’ concentration are shown
in Table 1.
RFLP analysis of 16S rDNA clone libraries
Two sets of archaea-specific primers used to amplify
archaeal 16S rDNA from two AMD samples both
failed, while the bacterial-specific 16S rDNA primer
set 27F and 1492R worked.
Consequently, the primer set 27F and 1492R was
used to amplify bacterial 16S rDNA gene from two
AMD samples in Yunfu sulfide mine. The PCR prod-
ucts formed a single band approximately 1,500 bp in
length. After T-A cloning, 120 clones containing 16S
rDNA inserts were obtained from each sample. The
profiles of RFLP in two samples are shown in Figs. 1, 2.
The rarefaction analysis was used in RFLP analysis.
The results are shown in Fig. 3. Nonlinear regression
suggested that saturations were at 70 clones and 40
clones for samples G2 and G1, respectively. The result
also suggested that the clones tested in the experiment
were sufficient to detect the level of microbial com-
munities’ diversity and infer the level of distribution
within communities of two samples.
The RFLP analysis revealed extensive diversity of
16S rDNA for two AMD samples. A total of 15 OTUs
(unique RFLP patterns) were obtained. There were
14 OTUs in sample G2; 5 OTUs were detected in
sample G1. The distributions of OTUs, which were
ranked in the order of abundance in each sample, were
shown in Fig. 4.
Extremophiles (2007) 11:305–314307
123

Page 4

The RFLP patterns of clones GY10, GY8, GY81,
GY21, and GY28 represented 48.2, 33.7, 8.4, 8.4, and
1.2% of the total clone populations in AMD sample
G1, respectively. In the other AMD sample G2, the
RFLP patterns of clones G8, G52, G27, G66, G51-12,
G28, G44, and G5 represented 31.5, 17.6, 10.2, 10.2,
10.2, 5.6, 3.7, and 3.7% of the total clone populations,
respectively. Some RFLP patterns in the two AMD
samples were overlapped with each other, such as
OTUs GY10 and G66, OTUs GY8 and G28, and so on.
The percentage of overlapped OTUs between samples
G1 and G2 was 42.1%. There were only five OTUs
detected in sample G1, four of which could be also
detected in sample G2.
Phylogenetic analysis
To determine the phylogenetic diversity, representa-
tive OTUs that occurred more than once in libraries, as
well as representatives of the unique OTUs, were fully
sequenced. The results of similarity among the se-
quences are shown in Table 2. Sequences’ comparison
showed that the clones in two AMD samples had high
similarity, which was 72–98% similar.
The phylogenetic analysis in two samples was
established with a bootstrap neighbor-joining method
with the sequences. The phylogenetic tree is shown in
Fig. 5. Of note is that there were 15 OTUs with RFLP
analysis, corresponding to 12 kinds of particular
organisms. The accuracy of RFLP screening was only
about 80% evaluated by sequencing representative
clones, which may be due to the comparatively
simplicity of microbial community in AMD, and this
method has been commonly used for it is compara-
tively accurate and fast.
These 16S rDNA sequences fell into four putative
phylogenetic divisions. They were Nitrospira, a-Prote-
obacteria, b-Proteobacteria, and c-Proteobacteria four
families. In sample G2, a-Proteobacteria, b-Proteobac-
teria, and c-Proteobacteria, Nitrospira four families
were all detected, proportions of which were 0.9, 48.1,
37.1, and 13.9%, respectively. In sample G1, there were
only two families detected, which were b-Proteobacte-
ria (16.9%) and c-Proteobacteria (83.1%). The distri-
butions of the four families divided by phylogenetic
analysis in two AMD samples are shown in Fig. 6.
c-Proteobacteria family was a predominant one in
two AMD samples. There were two groups in the
family. One group was clustered with the genus Acid-
ithiobacillus, including OTUs G52, G28, and G66. The
other group was divided into two sub-groups. One with
OTUs GY28 and G24 was clustered with Acinetobacter
sp. The other with OTUs G51-19 and G77 was clus-
tered with uncultured bacterium.
b-Proteobacteria family was the other predominant
family in two AMD samples. It also fell into two
groups. One group including OTUs G27, G71, G5, and
G8 was clustered with the genus Gallionella. The other
group was clustered with Pseudomonas testosterone,
just including one OUT, G31.
Nitrospira family was only detected in sample G2, in
which there were two OTUs G44 and G51-12. The two
OTUs were both affiliated with the genus Leptospiril-
lum. G44 was affiliated with L. ferrooxidans, while
G51-12 was affiliated with the bacterium of the group
III of Leptospirillum.
Table 1 Twenty-nine elements’ concentration in two AMD samples G1 and G2
SampleAs (mg/l)P (mg/l) Ni (mg/l) Zn (mg/l)Fe (g/l) Cu (mg/l)
G1
G2
2.5
5.9
5.4
13.1
0.8
2.2
24.1
404
1.07
4.025
0.9
0.8
Mn (mg/l)
76.7
547.6
Mg (mg/l)
75.6
352.5
Ca (mg/l)
320.1
514.2
S (g/l)
1.36
4.3
Al (mg/l)
169.4
617.8
Mo (mg/l)
0.5
1.2
G1
G2
Co (mg/l)
0.5
1.2
Cr (mg/l)
0.4
0.8
Be (mg/l)
0
0.2
Ti (mg/l)
0.3
0.3
W (mg/l)
2.1
10
Si (mg/l)
31.7
71.7
G1
G2
Pb (mg/l)
2.7
4.3
In (mg/l)
4.7
6.9
Bi (mg/l)
4.3
6.1
Au (mg/l)
2.2
8.4
Ag (mg/l)
0.6
0.5
Hg (mg/l)
1.1
2.5
G1
G2
Sn (mg/l)
0.9
2.1
Sb (mg/l)
2.7
5.5
Cd (mg/l)
0.3
0.5
Ga (mg/l)
4.4
8.7
Pt (mg/l)
5.1
7.6
G1
G2
308Extremophiles (2007) 11:305–314
123

Page 5

The last family was a-Proteobacteria family, which
was also just detected in AMD sample G2. Only one
OTU G74 was detected in this family. It had 99%
similarity to Hyphomicrobium vulgare strain IFAM
MC-750.
Discussion
Currently, AMD environments have been set as model
systems for analysis of biogeochemical interactions and
microbial community structure and function (Hallberg
Fig. 1 Restriction fragment length profiles of 16S rDNA
fragments amplified from sample G1 in Yunfu sulfide mine.
The 16S rDNA fragments were amplified using the primer set
27F and 1492R, digested with the restriction endonucleases AfaI
and MspI, and then analyzed by 3.0% ararose gel electropho-
resis. The clones tested were shown in (a–e). M: 100 bp DNA
ladder plus
Extremophiles (2007) 11:305–314 309
123