African Journal of Microbiology Research Vol. 6(12), pp. 2929-2940, 30 March, 2012 Available online at http://www.academicjournals.org/AJMR DOI: 10.5897/AJMR11.1416 ISSN 1996-0808 ©2012 Academic Journals

Full Length Research Paper

Bacterial community compositions in response to sediment properties in urban lakes of

Dayong Zhao 1,2 *, Meng Wang 1, Jin Zeng 2,3 , Wenming Yan 2, Jianqun Wang 1, Ting Ma 1 and Rui Huang 1

1College of Hydrology and Water Resources, , Nanjing, 210098, China. 2State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University, Nanjing, 210098, China. 3State Key Laboratory of Lake Science and Environment, Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences, Nanjing 210008, China.

Accepted 14 February, 2012

Compared to other lakes, urban lakes are often shallow, highly artificial and hypertrophic due to the higher level of public interface. Bacteria in lake sediment are important participators in the nutrient cyclings in lake ecosystems. In this study, bacterial community compositions in surface sediment of three urban lakes (Lake Xuanwu, Lake Yueya and Lake Pipa) of Nanjing were investigated by using the terminal restriction fragment length polymorphism (T-RFLP) of 16S ribosomal RNA genes followed by cloning and sequencing. At the same time, the response of bacterial community composition to sediment properties was assessed by multivariate analysis. The results indicated that most of the sampling stations in Lake Xuanwu showed similar T-RFLP pattern, suggesting the similar bacterial community compositions in these stations. However, the bacterial T-RFLP patterns varied among different sampling stations in sediments of Lake Yueya and Lake Pipa. Chloroflexi were the most dominant bacterial group in the clone library constructed from Lake Yueya (26.0% of the total clones). Whereas Betaproteobacteria were the most abundant group in the clone library from Lake Pipa (18.6% of the total clones). The higher abundance of Chloroflexi in sediment of Lake Yueya could be attributed to the higher concentrations of organic matters (OM) and total nitrogen (TN) in the sediment. Canonical correspondence analysis (CCA) showed that the bacterial community compositions in lake sediment were significantly related to the concentrations of OM in the sediment, which was associated with the macrophytes and phytoplankton in the lake ecosystems.

Key words: bacterial community compositions, sediment, urban lakes, multivariate analysis.

INTRODUCTION

Bacteria in lake sediment play important ecological and Therefore, elucidating the composition of the bacterial biogeochemical roles in the freshwater ecosystem, which community is a key step for better understanding the include regulating the decomposition and transformation metabolic processes in the freshwater ecosystem (Nixdorf of organic matters (OM) and biogenic elements such as C, and Jander, 2003; Koizumi et al., 2004). In the past N, P, Fe, O, and S (Nealson, 1997; Zeng et al., 2008). decade, the culture-independent methods based on bacterial 16S rRNA sequences, such as terminal-restriction fragment length polymorphism (T-RFLP) and denaturing gradient gel electrophoresis *Corresponding author. E-mail: [email protected]. Fax: +86 (DGGE) have been widely employed to investigate 25 83787891. Tel: +86 25 83787891. bacterial diversity in various environments (Muyzer et al.,

2930 Afr. J. Microbiol. Res.

1993; da C Jesus et al., 2009; Yang et al., 2011; Zhao et spatial variability in nutrient and physicochemical al., 2011). parameters in sediment of these small urban lakes (Xue The composition and diversity of bacterial community et al., 2004). However, the distribution and composition of are closely related to the sediment properties. While both the bacterial community in lake sediment, and whether the sediment properties and the bacterial community the changes in microbial assemblages were associated structure in the freshwater lake ecosystem have been well with diverse environmental factors are also unknown. The documented, the two related aspects were often aims of this study were to examine the diversity and considered separately (Koizumi et al., 2004; Carini et al., community composition of bacteria in the three different 2005). Investigating the differences of bacterial diversities urban lakes of Nanjing. For that, T-RFLP analysis of PCR in lake sediments in respond to the variations of sediment amplified fragments and constructing clone libraries were characteristics, would provide useful data for better employed. At the same time, the contributions of nitrogen, understanding the complex bacterial ecology in phosphorus, OM and pH (which have been proven to be freshwater ecosystems. At present, multivariate analysis key factors for bacterial community structures in marine methods such as principal component analysis (PCA), and river ecosystem) on microbial community non-metric multidimensional scaling (NMDS) and compositions were assessed using multivariate analysis canonical correspondence analysis (CCA) have proven to techniques. be robust methods for interpreting the relationship between the bacterial community composition and environmental factors (Iwamoto et al., 2000; Salles et al., MATERIALS AND METHODS 2006). Based on these techniques, environmental variables such as salinity (Ikenaga et al., 2010), nutrient Sediment sample collection and physicochemical parameter analysis concentrations (Rubin and Leff, 2007; Wu et al., 2008), organic matters (OM) (Bissett et al., 2007), pH Sediment samples were taken from several stations of three urban (Stepanauskas et al., 2003) and plant cover type (Jensen lakes of Nanjing, including Lake Xuanwu (X1-X5), Lake Yueya et al., 2007) have been proven as important factors (Y1-Y3) and Lake Pipa (P1-P2). The location of each sampling site influencing the bacterial community composition in the was recorded with GPS (Table 1). Sediment samples were collected with a corer sampler (437405, HYDRO-BIOS, Germany). marine and river ecosystem. However, whether these Undisturbed sediment cores were collected from each sampling important environmental factors also have significant station in three replicates. Sediment samples were stored on ice and effects on the bacterial community in lake sediment has in dark during transport to the laboratory. The surface 0-1 cm received little attention. sediment was sectioned using the sterile spatula and each replicate Urban lakes are very different from other rural and was well mixed and stored in sterile 50 mL collection tubes. The natural lakes: they are shallow, highly artificial and often samples were stored at -80°C prior to further analysi s. Sediment pH was measured with specific electrodes (PHB-5, hypertrophic. Meanwhile, urban lakes always receive REX, China). TN, TP and OM concentrations were measured higher level of public interface, especially in the densely according to (Jin and Tu, 1990) after the sediment samples were populated cities such as Nanjing and Wuhan in China dried with a Freeze Dryer (ALPHA 1-2, CHRIST, Germany). + - - (Birch and McCaskie, 1999). Lake Xuanwu (surface area: Ammonia nitrogen (NH 4 ), nitrate (NO 3 ) and nitrite (NO 2 ) were 3.7 km 2; average depth: 1.43 m) in Nanjing is a typical extracted with 2 M KCl, and their concentrations were determined urban, shallow lake, as well as a famous resort lake of using a continuous flow analyzer (San++, SKALAR, Netherlands) (Wu et al., 2010). China. The increasing amount of domestic wastewater discharged into the lake has caused severe eutrophication and the recreational value of the lake is DNA extraction also affected. Lake Yueya in Nanjing (surface area: 0.17 km 2; average depth: 2.0 m) was originally part moat of the Sediment sample from each sampling station was used for DNA extraction. After frozen dried, 0.5 g sediment sample (dry weight) city. Domestic wastewater and tourism were the main was used for DNA extraction based on the methods described by pollution source, which caused the severe eutrophic Zhou et al. (1996). The amounts of DNA extracted from samples status in this lake [the average concentrations of total were quantified using a BioPhotometer (Eppendorf, Hamburg, nitrogen (TN) and total phosphorus (TP) in 2006 reached Germany). to 2.78 and 0.26 mg/L, respectively] (Yao et al., 2009). Lake Pipa is an urban mini-lake, located in the PCR amplification and T-RFLP analysis exterior margin of Zhongshan scenic spots (Nanjing 4 2 City), which has a surface area of 2.66×10 m . Water Variable region of the bacterial 16S rRNA gene was amplified by the quality of Lake Pipa was fine before the discharge of primer set 8f (5 ′-AGAGTTTGATCCTGGCTCAG-3′) and 926r domestic wastewater from a hotel named Pipa Villa. (5 ′-CCGTCAATTCCTTTGAGTTT-3′) (Liu et al., 1997). The 5 ′ end of The hotel was demolished in October of 2005, and the the forward primer was labeled with Cy5. The 50 l PCR mixture water quality is being recovered slowly. contained 1 l of each primer (10 pmol), 0.25 l (1.25 U) of Ex Taq DNA polymerase (Takara, Otsu, Japan), 5 l of 10 × Ex Taq buffer, There have been few studies on documenting the 5 l of dNTPs (2.5 mM each), 1 l of DNA template (approximately

Zhao et al. 2931

10 ng), and the sterilized ultrapure water up to 50 l. PCR rather than linear responses to the environmental variables (Lepš amplification was performed by using a Thermal Cycler (S1000, and Šmilauer, 2003); therefore, CCA was performed by CANOCO Bio-Rad, MA, USA) with the amplification program as follows: 94°C 4.5 software package (Biometris, Netherlands) to explain the for 3 min, followed by 30 cycles of 94°C for 1 min, 5 5°C for 1 min, relationship between the bacterial community structure and and 72°C for 1 min with the final extension step at 72°C for 8 min. environmental factors. Both the environmental data and species Negative controls (without DNA template) were run in all data were not transformed for analysis. Forward selection was used amplifications. to identify environmental factors that significantly affecting the Three replicate of PCR products were combined and processed bacterial community structure (Ter Braak, 1987; Lepš and Šmilauer, by Mung Bean Nuclease (Takara, Otsu, Japan) according to the 2003). manufacturer’s instructions, after which the products were purified Meanwhile, statistically significant differences in the bacterial by Axygen PCR cleanup purification kit (Axygen Biotechnology Ltd. community composition reflected in the clone libraries from Lake Hangzhou, China) and then digested by the restriction enzyme HhaI Xuanwu, Lake Yueya and Lake Pipa were identified using the (Takara, Otsu, Japan), purified again by Axygen PCR cleanup ∫-Libshuff software package with 10000 randomizations (Schloss et purification kit, and analyzed by capillary electrophoresis using a al., 2004). This program calculates the integral form of the CEQ 8000 Genetic Analyzer (Beckman Coulter, Fullerton, CA, USA). Cramér-von Mise statistic based on Monte Carlo methods. The Accounting for the small differences in running time among samples, calculated P values were used to determine whether there were fragments from different profiles with less than 1 bp difference were significant differences between the two clone libraries (Schloss et al., considered to be the same. Peaks of less than 60 bp or longer than 2004). 600 bp were discarded. Meanwhile, only the terminal restriction fragments (T-RFs) with a relative area percent over 1% were included for further processing. RESULTS

Cloning, sequencing and phylogenetic analysis Properties of the sediment samples collected from the three lakes Clone library was constructed with DNA samples extracted from sediments of each lake, respectively. To amplify equal amounts of The pH, nutrient and OM concentrations of the sediment the samples, DNA extracted from each lake were mixed, respectively. PCR amplification was performed using the same samples collected from three different lakes are shown in protocol as that mentioned above except that the forward primer Table 1. There was no large scale variation of pH was not labeled with Cy5. PCR products were purified and ligated (6.65-7.14) in sediment of three different lakes. TN into the pGEM-T vector (Promega, Madison, WI, USA) following the concentrations in sediment of Lake Yueya (3.56-4.22 g/kg) manufacturer’s instructions. Plasmids were transformed into were higher than those of the other two lakes. The competent Escherichia coli cells (DH5 α, Takara, Japan) and the average concentration of TP in the five sampling station of presence of the 16S rRNA gene in randomly selected positive Lake Xuanwu was 1.14 g/kg, which was lower than those clones was checked by PCR amplification using vector primers (T7 and SP6). Positive clones were sent to the Shanghai Majorbio of Lake Yueya (1.67 g/kg) and Lake Pipa (1.63 g/kg). Bio-technology Co., Ltd., China for sequencing. Sampling station P2 maintained the highest Chimeric sequences were identified using the Mallard software concentrations of ammonia (37.41 mg/kg) and nitrate package, and all suspicious sequences were excluded from further (13.45 mg/kg). The average concentration of OM in analysis (Ashelford et al., 2006). The remaining sequences were sediment of Lake Yueya was 6.52%, which was higher compared with GenBank entries using BLAST (http://www.ncbi.nlm.nih.gov/BLAST/). The Ribosomal Database than those of the other two lakes. Project classifier was applied to assign the acquired sequences to the taxonomic hierarchy (Wang et al., 2007). The 16S rRNA gene sequences were aligned using ClustalX T-RFLP analysis of the bacterial community (Thompson et al., 1997). A phylogenetic tree was constructed by the composition in sediment of three lakes neighbor-joining method based upon distances determined by Jukes and Cantor (1969) with 1,000 bootstraps using the MEGA 4.0 program (Tamura et al., 2007). In order to test the phylogenetic The T-RFLP profiles of bacterial community composition assignments based on in silico T-RF analysis, randomly selected in sediment of three lakes were shown in Figure 1. clones were analyzed by in-vitro T-RF by finding the first HhaI Distinct differences of the T-RFLP patterns were observed enzymatic digestion site downstream from 8f (Li et al., 2011). among the three different lakes. Sampling stations of X1-X4 in Lake Xuanwu showed similar pattern. In sediment of Lake Yueya and Lake Pipa, by contrast, the Nucleotide sequence accession number bacterial T-RFLP patterns varied among different

Sequences obtained in this study were uploaded and are available sampling stations. In the five samples of Lake Xuanwu, a at the GenBank database under accession numbers total of 58 distinct bacterial T-RFs were identified, JN849235-JN849369. including the T-RFs of 86, 87, 88, 95, 96, 131, 132, 176 and 218 bp with the relative abundance > 4%. Meanwhile, T-RFs of 110, 129, 134, 155, 156, 199, 202, 205, 211, 221, Statistical analysis 223 and 230 bp were only found in Lake Xuanwu. Other The initial detrended correspondence analysis (DCA) results specific T-RFs of 60, 92, 513, 93, 201, 203, 365 and 565 demonstrated that the data obtained in this study exhibited unimodal bp were only detected in the samples of Lake Yueya, and

2932 Afr. J. Microbiol. Res.

Table 1. The chemical characteristics of the sediment samples collected from three lakes.

+ - - Sampling stations Locations recorded by GPS pH TN (g/kg) TP (g/kg) NH 4 (mg/kg) NO 3 (mg/kg) NO 2 (mg/kg) OM (%) X1 32.06928N, 118.78914S 6.65 3.54±0.12 0.62±0.04 16.78±0.45 1.46±0.09 0.26±0.02 6.79±0.06 X2 32.07277N, 118.79202S 6.86 2.31±0.09 1.09±0.05 12.57±0.52 3.32±0.12 0.17±0.02 5.23±0.07 X3 32.07289N, 118.79613S 6.84 2.19±0.10 1.64±0.04 13.32±0.69 5.24±0.24 0.35±0.03 5.25±0.15 X4 32.08010N, 118.78980S 7.14 2.26±0.11 1.22±0.06 13.90±0.48 11.28±0.68 0.88±0.03 4.79±0.15 X5 32.08180N, 118.79436S 6.86 2.41±0.10 1.15±0.07 11.80±0.61 4.44±0.21 0.42±0.02 5.06±0.17 Y1 32.03120N, 118.82149S 6.65 3.56±0.11 1.53±0.06 12.93±0.44 4.29±0.18 0.38±0.05 6.29±0.21 Y2 32.03065N, 118.82213S 6.78 3.91±0.13 1.65±0.13 12.46±0.48 10.06±0.48 1.05±0.08 6.72±0.03 Y3 32.02956N, 118.82219S 6.76 4.22±0.10 1.82±0.11 13.30±0.38 5.87±0.36 0.39±0.02 6.54±0.22 P1 32.05627N, 118.81670S 6.70 1.81±0.06 1.68±0.08 17.63±0.57 5.15±0.24 0.81±0.05 5.83±0.16 P2 32.05516N, 118.81576S 6.99 2.37±0.07 1.58±0.07 37.41±0.68 13.45±0.39 0.42±0.02 3.73±0.11

+ - - GPS: global positioning system, TN: total nitrogen, TP: total phosphorus, NH 4 : ammonia nitrogen, NO 3 : nitrate, NO 2 : nitrite, OM: organic matters.

T-RFs of 62, 86, 87, 88, 96, 97 and 208 bp were were the most dominant bacterial group in the heterologous coverage curves based on the the dominant fragments with their relative clone library from Lake Yueya (26.0% of the total LIBSHUFF program are shown in Table 3. abundance > 4%. T-RFs of 86 bp was the most clones) whereas Betaproteobacteria were the Comparisons between the clone library of Lake dominated fragment in sediment of Lake Pipa with most abundant group in the clone library from Xuanwu and the libraries from the other two lakes the relative abundance of 16.7%, followed by 87 Lake Pipa (18.6% of the total clones) (Table 2). did not reveal significant differences ( P > 0.0083). bp (15.2%) and 475 bp (14.8%). Only three T-RFs Both Betaproteobacteria and Chloroflexi were the At the same time, result of the calculation between of 94, 99 and 564 bp were specifically found in most dominant groups (16.7%) in the clone library Lake Pipa and Lake Yueya libraries yielded a P Lake Pipa. of Lake Xuanwu, followed by the value of 0.8354 (the Lake Yueya library is Gammaproteobacteria group which covers the homologous), suggesting some overlap between 14.3% of the total clones (Table 2). Lower the two libraries. However, a P value of 0.0036 Clone library analysis percentages of clones affiliated with (the Lake Yueya library is heterologous) was also Epsilonproteobacteria and Actinobacteria phyla observed, suggesting that the Lake Pipa library Three clone libraries were constructed to identify were observed in the library of Lake Pipa, whereas contained more taxa that were not found in the the bacterial species generated from the sediment these groups were not observed in the libraries of Lake Yueya library. of Lake Xuanwu (n = 42), Lake Yueya (n = 50) and Lake Xuanwu and Lake Yueya. Dominated clones Lake Pipa (n = 43) (Table 2). At the same time, affiliated to Betaproteobacteria grouped together Phylogenetic tree of the clone sequences was with Burkholderiales and Rhodocyclales. The Assignment of T-RFs constructed using the neighbor-joining method majority of the Chloroflexi obtained in this study based on the MEGA 4.0 software package (Figure were found affiliated with Anaerolineae and Additionally, the randomly selected clones were 2). Caldilineae (Figure 2). analyzed by in-vitro T-RF by finding the first HhaI Clone libraries constructed from the sediment of enzymatic digestion site downstream from 8f to Lake Yueya and Lake Pipa showed remarkable investigate the phylogenetic assignments of T-RFs. differences. The most important difference was the LIBSHUFF analysis As shown in Table 4, many detected T-RFs could abundance of clones affiliating with both be assigned to defined taxonomic groups. At the Betaproteobacteria and Chloroflexi . Chloroflexi Statistical comparison results of homologous and same time, several T-RFs including 83 and 99 bp

Zhao et al. 2933

Figure 1. Relative abundance of the bacterial 16S rRNA amplicons recovered from the sediment samples. Those T-RFs of less than 1% of the total abundance were combined together as a single group termed as <1%.

Table 2. Phylogenetic analysis of the clone libraries constructed from the sediment of three lakes.

Number of clones Phylogenetic group Lake Pipa Acidobacteria 1 2 2 Actinobacteria N.D. N.D. 1 Bacteroidetes 5 5 4 Chloroflexi 7 13 1 Cyanobacteria N.D. 1 1 Firmicutes 1 N.D. 2 Gemmatimonadetes 2 N.D. N.D. OD1 N.D. 2 N.D. Planctomycete 3 3 1 Proteobacteria (Alpha-) 1 N.D. 3 Proteobacteria (Beta-) 7 7 8 Proteobacteria (Delta-) 2 4 5 Proteobacteria (Epsilon-) N.D. N.D. 1 Proteobacteria (Gamma-) 6 6 6 Verrucomicrobia 2 N.D. 3 Unclassified Bacteria 5 7 5 Total 42 50 43

N.D.: not detected.

2934 Afr. J. Microbiol. Res.

100 LPP151 100 Paracoccus sp. PA216 (AM900779) LPP126 100 Rhodobacter sp. THWCSN29 (AM888193) 100 alpha - 68 LXW22 100 Ochrobactrum anthropi strain W -7 (EU187487) LPP152 100 Uncultured Alphaproteobacteria bacterium clone QEDV3CG10 (CU919640) 100 LPP135 Uncultured Epsilon proteobacterium clone BB38 (AM167942) epsilon - 96 LYY267 100 Uncultured bacterium clone B0610R002_B03 (AB657284) 100 LXW25 Uncultured bacterium clone sb21.42 (HQ904205) 86 LXW12 100 Uncultured bacterium clone B0618R001_D14 (AB658268) 91 100 LXW39 Uncultured Delta proteobacterium clone MA -R97 (JN038658) 100 LYY242 Uncultured Syntrophaceae bacterium clone C4 (HQ003562) LYY239 72 100 Uncultured prokaryote clone Td1 -2 (GU208365) 100 LYY259 delta - 96 Uncultured cf. Anaeromyxobacter sp. clone 90D -55 (AJ504436) 64 LPP156 100 Uncultured bacterium clone B1001R001_B01 (AB660040) 100 LPP163 Syntrophus gentianae strain HQgoe1 (NR 029295) 100 LYY251 72 Uncultured Delta proteobacterium clone 2B5 (HQ003572) LXW32 100 Uncultured prokaryote clone 6T1 -22 (GU208253) 50 100 LPP133 Uncultured Geobacter sp. clone HrhB74 (AM159289) 99 LPP119 100 Uncultured Delta proteobacterium clone B -LO -T0_OTU7 (FM204949) 99 LYY246 51 Uncultured Beta proteobacterium clone R15 -77 (JF808905) LXW31 94 100 Uncultured prokaryote clone 6S1 -11 (GU208247) LXW3 58 55 98 Uncultured Beta proteobacterium clone Z17M24B (FJ484428) 100 LPP141 Proteobacteria Uncultured Beta proteobacterium clone CYC_17 (EF562577) 57 69 LPP160 100 Uncultured Beta proteobacterium clone MP -R41 (JN038714) 100 LPP143 Uncultured Rhodocyclaceae bacterium clone 2A7 (HQ003474) 100 LYY257 51 Uncultured Beta proteobacterium clone GuBH2 -AD -29 (AJ519655) 98 LPP146 74 100 Uncultured Beta proteobacterium clone MWR -H2 (FJ946626) 99 LPP153 71 LPP150 Uncultured Beta proteobacterium clone 2C12 (HQ003488) beta - 100 84 Uncultured Beta proteobacterium clone F1 (HQ003489) LXW42 78 Uncultured Beta proteobacterium clone MVS -96 (DQ676430) 100 LXW30 59 99 Uncultured prokaryote clone Fh4 -20 (GU208432) 99 LXW2 99 Beta proteobacterium HS5/S24542 (AY337603) 98 100 LXW36 Uncultured Burkholderiales clone GC12m -2-26 (EU641675) 100 LPP159 54 Delftia acidovorans strain FGQ5 (HQ327476) 100 LXW6 70 100 Acidovorax temperans (DQ111771) 100 LPP134 Uncultured Beta proteobacterium clone MVS -39 (DQ676287) 69 LYY221 100 Uncultured Rhodocyclaceae bacterium clone Amb_16S_962 (EF018674) 100 LPP125 56 Stenotrophomonas maltophilia (DQ813325) 100 LXW8 100 Uncultured prokaryote clone Td1 -20 (GU208372) 100 LYY252 Acinetobacter sp. NB4 (AM269472) 75 LXW28 100 Uncultured Pseudomonadaceae bacterium clone AMEH3 (AM935347) LYY222 100 Uncultured bacterium clone c5LKS1 (AM086102) 80 LPP139 100 Uncultured prokaryote clone Ftd3 -4 (GU208300) Uncultured Gamma proteobacterium clone AMEH9 (AM935352) 51 56 LXW38 LXW10 100 Uncultured Gamma proteobacterium clone GASP -MA4S2_F09 (EF664072) 53 100 LYY234 Uncultured Gamma proteobacterium clone 426T3 (DQ110125) gamma - 100 LPP140 55 Uncultured prokaryote clone Td1 -10 (GU208368) 100 LXW44 Uncultured bacterium clone d0 -14 (AM409812) 99 LPP162 100 Methylosarcina fibrata strain AML -C10 (NR 025039) LYY243 LYY247 54 Uncultured Gamma proteobacterium clone AKYG1795 (AY921681) 50 50 Methylobacter tundripaludum SV96 (NR 042107) 52 LXW40 73 Uncultured Methylobacter sp. clone GASP -0KA -565 -E11 (EU043582) (A) 0.02

Figure 2A. Neighbor-joining phylegenetic tree for phylum Proteobacteria detected in this study. Numbers at nodes represent the percentages of bootstrap resamplings based on 1000 replicates; only the values higher than 50 are presented. Sequences obtained in this study were in color words, with blue, red and green labels for sequences recovered from sediment of Lake Xuanwu, Lake Yueya and Lake Pipa, respectively.

Zhao et al. 2935

56 LXW18 95 Uncultured Chloroflexi bacterium clone AKYH1368 (AY921734) LYY249 98 62 Uncultured Anaerolineae bacterium clone AMBA6 (AM935711) LXW9 53 92 Uncultured Chloroflexi bacterium clone bac719 (JF727746) 19 LXW7 100 Uncultured bacterium clone (CU466709) LYY271 55 100 Uncultured Chloroflexi bacterium clone 4.68 (GQ183395) 50 LYY254 Uncultured bacterium clone d6-9 (AM409974) 95 LYY268 Uncultured Chloroflexi bacterium clone D15_47 (EU266883) 58 LXW19 61 94 Uncultured bacterium clone U37-6 (DQ137919) LXW4 100 Uncultured bacterium clone d1-43 (AM409919) LYY262 69 100 Uncultured Chloroflexi bacterium clone NLS4.15 (HQ397215) 100 Uncultured Chloroflexi bacterium clone MP-R77 (JN038747) LYY238 100 LYY226 90 Uncultured prokaryote clone 1Ab4-22 (GU208211) 100 LYY261 39 Uncultured prokaryote clone 1Ab4-22 (GU208211) Chloroflexi 68 LYY250 91 LYY229 35 Uncultured Chlorobi bacterium clone Sh765B-TzT/AG-5 (AJ519641) LPP129 97 74 100 Uncultured Chloroflexi bacterium clone 4.26 (GQ183434) 72 LYY270 89 Chloroflexi bacterium ET1 (EU875524) 100 Uncultured Chloroflexi bacterium clone AKYG630 (AY922045) 58 LYY253 LYY263 100 Uncultured Chloroflexi bacterium clone ELC_30_7_bac (EF464632) 99 LYY272 56 Uncultured Chloroflexi bacterium clone Z53M14B (FJ484501) 51 LXW35 76 Uncultured bacterium clone NK2_211 (EU376188) LXW43 79 98 59 Uncultured Chloroflexi bacterium clone TK-SH18 (DQ463738) 100 LXW45 65 Uncultured bacterium clone B0618R002_E03 (AB658206) 18 LYY264 Nitrospira 100 Uncultured bacterium clone B0423R001_G20 (AB656338) 100 LPP158 Uncultured Actinobacteridae bacterium clone HG150 (FN582328) Actinobacteria 100 LXW24 Staphylococcus capitis strain BQEP2-01d (FJ380956) Firmicutes 96 LPP128 100 Uncultured bacterium clone Zeebrugge_B67 (HM598609) 100 LXW14 71 73 Uncultured Gemmatimonadetes bacterium clone AKYG666 (AY921771) Gemmatimonadetes LXW27 100 Uncultured bacterium clone ORSFAB_h03 (EF393220) 100 LYY228 96 Uncultured Acidobacteria bacterium clone MVS-33 (DQ676329) LXW17 66 84 Uncultured Acidobacteria bacterium clone LH-53 (AB265876) Acidobacteria LPP142 52 Uncultured Acidobacteria bacterium clone P-R85 (JN038868) 100 100 LPP138 56 Uncultured Acidobacterium sp. clone Z53M89B (FJ484557) 53 51 100 LYY269 Uncultured bacterium clone B6_243 (HM228781) Lentisphaerae 99 LPP117 Uncultured organism clone Msed36 (GQ994747) 100 LPP161 100 Uncultured bacterium clone BSTSN-54 (JN105045) 84 LXW13 100 LPP145 95 Uncultured Verrucomicrobium DEV114 (AJ401132) Verrucomicrobia 57 LXW1 53 50 100 Uncultured Verrucomicrobia bacterium clone GASP-MA2W2_G06 (EF663260) 90 100 LPP144 Uncultured Verrucomicrobium DEV009 (AJ401108) 100 LPP155 Uncultured Verrucomicrobiaceae bacterium clone B05-12B (FJ543009) 100 LPP130 58 Uncultured bacterium clone HWB2224-1-89 (HM243910) 100 LXW5 54 Uncultured Planctomycete clone LC1-32 (DQ289903) 90 LXW16 Uncultured Planctomycete clone C8 (AY360083) 100 Planctomycetes 100 LXW26 51 Uncultured Planctomycete clone B2 (AY266448) 96 LPP131 98 Uncultured bacterium clone B1001R001_O06 (AB660286) 99 LYY227 100 Uncultured Planctomycetes bacterium QEDN7BG08 (CU926663) 100 LXW33 Bacterium TG141 gene (AB308367) 100 LPP149 100 Flavobacterium sp. WB2.3-15 (AM934646) LXW15 90 100 Uncultured Bacteriodetes bacterium clone AS-45-6 (GQ406176) 100 LXW11 53 Uncultured prokaryote clone Td2-13 (GU208382) 82 LYY236 92 54 Uncultured bacterium clone C53 (EU234264) LXW34 98 Uncultured Bacteroidetes bacterium clone QEDR1CC09 (CU922346) 100 LPP124 72 Uncultured bacterium clone FGL12_B4 (FJ437881) 57 99 Uncultured Cytophagales bacterium clone TDNP_Wbc97_200_1_88 (FJ517044) Bacteroidetes LPP127 100 LYY258 52 Uncultured Bacteroidetes bacterium clone QEDN9BC03 (CU926652) 63 LYY265 89 Uncultured Bacteroidetes bacterium clone QEDQ1CG08 (CU923071) Uncultured bacterium WCHB1-53 (AF050539) 88 83 LYY255 Uncultured Bacteroidetes bacterium clone Z273MB41 (FJ484674) 50 LYY232 LPP136 LXW41 Uncultured Bacteriodetes bacterium clone AS-39-10 (GQ406167) 84 74 Uncultured Bacteroidetes bacterium clone AKYG1128 (AY921953) 67 LYY237 Bacterium enrichment culture clone B160(2011) (JF830226) Cyanobacteria 100 LPP154 89 Uncultured Cyanobacterium clone TDNP_USbc97_2_47_65 (FJ516953) 63 LYY245 100 LYY230 Uncultured candidate division OD1 bacterium clone RODAS-138 (JF344116) 100 LXW23 OD1 96 Uncultured bacterium clone PT19 (HQ330568) LXW20 98 100 Uncultured bacterium clone B0618R001_O04 (AB659208)

0.05 (B)

Figure 2B. Neighbor-joining phylegenetic tree for all other phyla detected in this study. Numbers at nodes represent the percentages of bootstrap resamplings based on 1000 replicates; only the values higher than 50 are presented. Sequences obtained in this study were in color words, with blue, red and green labels for sequences recovered from sediment of Lake Xuanwu, Lake Yueya and Lake Pipa, respectively.

2936 Afr. J. Microbiol. Res.

Table 3. LIBSHUFF comparisons of the bacterial community of the three clone libraries.

Comparison (X vs Y) P-value Significantly different Lake Xuanwu vs Lake Yueya No XY 0.5845 YX 0.1201 Lake Xuanwu vs Lake Pipa No XY 0.9262 YX 0.8493 Lake Yueya vs Lake Pipa Yes XY 0.8354 YX 0.0036*

Libraries were considered significantly different when the P value is less than 0.0083. (* indicates significant difference between the two clone libraries.)

Table 4. Phylogenetic affiliations of bacterial 16S rRNA sequences retrieved in clone libraries constructed from the sediment samples of three lakes.

T-RFs (bp) Phylogenetic group X1 X2 X3 X4 X5 Y1 Y2 Y3 P1 P2 Acidobacteria 370 370 370 370 370 Bacteroidetes 103 Chloroflexi 61,62 62 61,62 61 62 62 61,62 62,568 62 62 Firmicutes 230 230 230 230 230 Gemmatimonadetes 221 221 Proteobacteria (Beta-) Burkholderiales 214 214 214 214 214 214 Proteobacteria (Delta-) Desulfuromonadales 81 Desulfobacterales 81 Syntrophobacterales 83* 83* 83* Myxococcales 78 Unclassified 98 98 359 98 98 Proteobacteria (Gamma-) Methylococcales 87 87 87 87 87,212 87,212 87 87,212 87 87 Xanthomonadales 212 212 212 Pseudomonadales 210 210 210 210 210 Unclassified 214,86 86 214,86 214,86 214,86 214 86 214,86 86 86 Verrucomicrobia 93 83* 203 93,99*,203 83*,93 93,203 83*,93,99*,203 Unidentified 96 96 96 96 96,99* 96 96 96,99*

T-RFs with relative abundance of more than 4% are indicated in bold and T-RFs detected in more than one phylogenetic group are marked with an asterisk.

could be assigned to more than one phylogenetic group. Meanwhile, the T-RFs of 86 (assigned to Unclassified T-RFs obtained from the sediment of Lake Xuanwu could Gammaproteobacteria ) and 87 bp (assigned to be assigned to Acidobacteria, Chloroflexi, Firmicutes, Methylococcales ) accounted for 4.5-10.3% and 3.8-17.1% Gemmatimonadetes,Verrucomicrobia, Betaproteobacteria of the total bacterial communities in sediment of Lake (Burkholderiales) and Gammaproteobacteria Xuanwu. In sediment of Lake Yueya, the T-RFs were (Methylococcales and Pseudomonadales) (Table 4). The mainly assigned to Acidobacteria, Bacteroidetes, T-RFs of 230 bp (assigned to Firmicutes ) and 221 bp Chloroflexi, Betaproteobacteria (Burkholderiales) and (assigned to Gemmatimonadetes ) were detected only in Gammaproteobacteria (Methylococcales, Lake Xuanwu. Xanthomonadales and Pseudomonadales). Specific

Zhao et al. 2937

T-RFs of 103 bp (affiliated with Bacteroidetes ) and 78 bp Lake Pipa was significant. Lake Xuanwu is the most (affiliated with Myxococcales ) were only found in Lake important urban lake in Nanjing with the surface area of Yueya, while the other specific T-RFs were not 3.7 km 2, which is significantly larger than that of Lake Pipa represented by any of the clone sequences and therefore (2.66×10 4 m2) and Lake Yueya (0.17 km 2). In this study, could not be assigned to any phylogenetic group. five sediment samples were collected from three different Chloroflexi, Verrucomicrobia, Gammaproteobacteria lake zones of Lake Xuanwu, which would also explain the (Methylococcales ) and Deltaproteobacteria diversity of clone library from this lake. (Desulfuromonadales, Desulfobacterales and CCA analysis revealed that OM was a significant Syntrophobacterales ) were the dominant bacterial environmental variable which driving the bacterial populations in sediment of Lake Pipa. The specific T-RF community composition in sediment of the three lakes. (81 bp) were affiliated with Desulfuromonadales and The effect of OM on the bacterial community Desulfobacterales. compositions has been reported previously in both water column and sediment (Li et al., 2011; Macalady et al., 2000; Zeng et al., 2009). Macalady et al. (2000) found that CCA analysis bacterial community structure was strongly related to sediment organic carbon content in a mercury-polluted To explain the relationship between the bacterial lake. Autochthonous organic matters in the lake community composition and environmental factors, ecosystem mainly come from the phytoplankton and its canonical correspondence analysis was carried out, and exudates, which would affect the growth of macrophytes the results are shown in Figure 3. Arrows represent the in the lake (Rooney-Varga et al., 2005). The growth status environmental factors and the sampling stations are of macrophytes may have different effects on indicated by upward triangles. The first and second axes environmental conditions (that is, chemical composition combined explained 40.4% of the species-environment and dissolved organic matter composition) (Zeng et al., relationships. Forward selection and Monte Carlo testing 2012), which may affect the bacterial community indicated that OM significantly ( P < 0.05) accounted for compositions indirectly. Furthermore, a positive the variability in the bacterial community composition. correlation was observed between the number of T-RFs and the OM concentrations in sediment, suggesting OM may promote the growth of heterotrophic bacteria. DISCUSSION Previous studies indicated that the soil bacterial communities were normally comprised of the nine major The eutrophication status and the diversity of bacterial bacterial phyla: the Proteobacteria (mainly the Alpha , community in large shallow eutrophic lake such as, Lake Beta and Delta subdivisions), Actinobacteria , Taihu in China have been well documented (Zeng et al., Acidobacteria , Chloroflexi , Bacteroidetes , Firmicutes , 2009). However, the bacterial communities which take Planctomycetes , Verrucomicrobia and part in the important nutrient cyclings in sediment of small Gemmatimonadetes (Janssen, 2006). There are totally urban lakes were overlooked for a long time. In the eleven bacterial phyla were observed in this study present study, the bacterial community compositions in including, Proteobacteria, Acidobacteria, Actinobacteria, three small urban lakes in Nanjing were compared based Bacteroidetes, Chloroflexi, Cyanobacteria, Firmicutes, on two culture-independent methods, T-RFLP and cloning Gemmatimonadetes, OD1, Planctomycete and library. Additionally, multivariate analysis was carried out Verrucomicrobia. Among these bacterial phyla, one that to explain the relationship between the sediment should be paid more attention to is the Chloroflexi . properties and microbial community structure. Previous studies indicated that Chloroflexi was one of The result of T-RFLP analysis indicated that there were the major bacteria group in deep subsurface sediments of several specific T-RFs representing the specific bacterial the ocean floor (Huber et al., 2006). At the same time, it is taxa in sediment of each lake. At the same time, the also the cosmopolitan members in the activated sludge of bacterial communities in different sampling stations within various wastewater treatment systems (Björnsson et al., the same lake were also not the same, which would be 2002). In the present study, Chloroflexi covers 26.0% of attributed to the different sediment properties within the the total clone number of the library from Lake Yueya, + same lake. For example, the concentrations of TN, NH 4 which was significantly higher than those of the other two and OM in sampling station X1 of Lake Xuanwu were lakes. Most of the 13 Chloroflexi clones were defined as significantly higher than other sampling stations of Lake Anaerolineae and the others were affiliated with Xuanwu (Table 1). LIBSHUFF analysis of the clone Caldilineae. Anaerolineae are the most abundant group in libraries constructing from three different lakes indicated the Chloroflexi -specific 16S rRNA gene libraries of that there was no significant difference between the activated sludge (Juretschko et al., 2002). Caldilineae are bacterial community from Lake Xuanwu and the other two known to prevent membrane fouling in wastewater lakes, however, the difference between Lake Yueya and treatment plant (Miura et al., 2007).

2938 Afr. J. Microbiol. Res.

Figure 3. Canonical correspondence analysis (CCA) of the relationship between bacterial community compositions and environmental variables in the sediment samples of three lakes. Arrows indicated the environmental factors. ( ▲), ( ●) and ( ■) represent samples collected from Lake Xuanwu, Lake Yueya and Lake Pipa, respectively.

The Chloroflexi sequences obtained in this study was In summary, this study shows the bacterial community affiliated with previous clones isolated from compositions in surface sediment of three urban lakes. hydrocarbon-contaminated soil (AM935711) (Militon et al., Chloroflexi were the most dominant bacterial group in the 2010), petroleum-contaminated saline-alkali soils clone library from Lake Yueya, whereas (JF727746), tar-oil contaminated aquifer sediments Betaproteobacteria appeared to be dominated colonizers (EU266883) (Winderl et al., 2008), wetland mesocosm in sediment of Lake Pipa. Multivariate statistical analysis (GQ183395) or soil (JN038747) and activated sludge indicated that OM had significant effect on bacterial (EU875524). Winderl et al. (2008) found that Chloroflexi community structure in lake sediments. This study points was one of the most abundant groups in the polycyclic to the interactions between bacterial community aromatic hydrocarbons (PAHs) contaminated soil zone, composition and sediment environmental characteristics, which further confirmed that the Chloroflexi was the which could shed light on the roles of microorganisms dominated group in hydrocarbon-contaminated soil and involved in the biogeochemical cycling of nutrient may involved in the biodegradation process of elements in the lake ecosystem. hydrocarbon pollutants. In this study, the sediment samples collected from Lake Yueya exhibited higher concentrations of OM and TN, suggesting the sediment ACKNOWLEDGEMENTS may be polluted by organic pollutants which partly explained the high abundance of Chloroflexi in this lake. This work was supported by the National Basic Research Further isolation of bacteria in this group and Program of China (2008CB418104), the National Natural investigation of their eco-physiological characteristics Science Foundation of China (41001044, 41101052), would be helpful for revealing their exact role in the Natural Science Foundation of Province, China decomposition of organic matters in freshwater lakes. (BK2010522, BK2011876), Open Foundation of State Key

Zhao et al. 2939

Laboratory of Hydrology-Water Resources and Hydraulic community composition dynamics in response to accumulation and Engineering (NO. 2010490311), the Project of Nanjing breakdown of Microcystis blooms. Water Res., 45: 1702-1710. Liu WT, Marsh TL, Cheng H, Forney LJ (1997). Characterization of Institute of Geography and Limnology microbial diversity by determining terminal restriction fragment length (NIGLAS2010QD10), the Fundamental Research Funds polymorphisms of genes encoding 16S rRNA. Appl. Environ. for the Central Universities of Hohai University Microbiol., 63: 4516-4522. (2009B01714), Ministry of Water Resources' Special Macalady JL, Mack EE, Nelson DC, Scow KM (2000). Sediment microbial community structure and mercury methylation in Funds for Scientific Research on Public Causes mercury-polluted Clear Lake, California. Appl. Environ. Microbiol., 66: (201201026) and a Project Funded by the Priority 1479-1488. Academic Program Development of Jiangsu Higher Militon C, Boucher D, Vachelard C, Perchet G, Barra V, Troquet J, Education Institutions. Peyretaillade E, Peyret P (2010). Bacterial community changes during bioremediation of aliphatic hydrocarbon-contaminated soil. FEMS Microbiol. Ecol., 74: 669-681. Miura Y, Watanabe Y, Okabe S (2007). Significance of Chloroflexi in REFERENCES performance of submerged membrane bioreactors (MBR) treating municipal wastewater. Environ. Sci. Technol., 41: 7787-7794. Ashelford KE, Chuzhanova NA, Fry JC, Jones AJ, Weightman AJ (2006). Muyzer G, de Waal EC, Uitterlinden AG (1993). Profiling of complex New screening software shows that most recent large 16S rRNA gene microbial populations by denaturing gradient gel electrophoresis clone libraries contain chimeras. Appl. Environ. Microbiol., 72: analysis of polymerase chain reaction-amplified genes coding for 16S 5734-5741. rRNA. Appl. Environ. Microbiol., 59: 695-700. Birch S, McCaskie J (1999). Shallow urban lakes: A challenge for lake Nealson KH (1997). Sediment bacteria: who's there, what are they doing, management. Hydrobiologia ,395: 365-378. and what's new? Annu. Rev. Earth Pl. Sc., 25: 403-434. Bissett A, Burke C, Cook PLM, Bowman JP (2007). Bacterial community Nixdorf B, Jander J (2003). Bacterial activities in shallow lakes–A shifts in organically perturbed sediments. Environ. Microbiol., 9: comparison between extremely acidic and alkaline eutrophic hard 46-60. water lakes. Hydrobiologia, 506: 697-705. Björnsson L, Hugenholtz P, Tyson GW, Blackall LL (2002). Filamentous Rooney-Varga JN, Giewat MW, Savin MC, Sood S, LeGresley M, Martin Chloroflexi (green non-sulfur bacteria) are abundant in wastewater JL (2005). Links between phytoplankton and bacterial community treatment processes with biological nutrient removal. Microbiology, dynamics in a coastal marine environment. Microbial Ecol., 49: 148: 2309-2318. 163-175. Carini S, Bano N, LeCleir G, Joye SB (2005). Aerobic methane oxidation Rubin MA, Leff LG (2007). Nutrients and other abiotic factors affecting and methanotroph community composition during seasonal bacterial communities in an Ohio River (USA). Microbial Ecol., 54: stratification in Mono Lake, California (USA). Environ. Microbiol., 7: 374-383. 1127-1138. Salles JF, Van Elsas JD, Van Veen JA (2006). Effect of agricultural da C Jesus E, Marsh TL, Tiedje JM, de S Moreira FM (2009). Changes management regime on Burkholderia community structure in soil. in land use alter the structure of bacterial communities in Western Microbial. Ecol., 52: 267-279. Amazon soils. ISME J., 3: 1004-1011. Schloss PD, Larget BR, Handelsman J (2004). Integration of microbial Huber JA, Johnson HP, Butterfield DA, Baross JA (2006). Microbial life in ecology and statistics: a test to compare gene libraries. Appl. Environ. ridge flank crustal fluids. Environ. Microbiol., 8: 88-99. Microbiol., 70: 5485-5492. Ikenaga M, Guevara R, Dean AL, Pisani C, Boyer JN (2010). Changes in Stepanauskas R, Moran MA, Bergamaschi BA, Hollibaugh JT (2003). community structure of sediment bacteria along the Florida coastal Covariance of bacterioplankton composition and environmental everglades marsh-mangrove-seagrass salinity gradient. Microbial variables in a temperate delta system. Aquat. Microb. Ecol., 31: 85-98. Ecol., 59: 284-295. Tamura K, Dudley J, Nei M, Kumar S (2007). MEGA4: molecular Iwamoto T, Tani K, Nakamura K, Suzuki Y, Kitagawa M, Eguchi M, Nasu evolutionary genetics analysis (MEGA) software version 4.0. Mol. Biol. M (2000). Monitoring impact of in situ biostimulation treatment on Evol., 24: 1596-1599. groundwater bacterial community by DGGE. FEMS Microbiol. Ecol., Ter Braak CJF (1987). The analysis of vegetation-environment 32: 129-141. relationships by canonical correspondence analysis. Plant Ecol., 69: Janssen PH (2006). Identifying the dominant soil bacterial taxa in 69-77. libraries of 16S rRNA and 16S rRNA genes. Appl. Environ. Microbiol., Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG 72: 1719-1728. (1997). The CLUSTAL_X windows interface: flexible strategies for Jensen SI, Kühl M, Prieme A (2007). Different bacterial communities multiple sequence alignment aided by quality analysis tools. Nucleic associated with the roots and bulk sediment of the seagrass Zostera Acids Res., 25: 4876-4882. marina . FEMS Microbiol. Ecol., 62: 108-117. Wang Q, Garrity GM, Tiedje JM, Cole JR (2007). Naïve Bayesian Jin XC, Tu QY (1990). Survey specification for Lake eutrophication. classifier for rapid assignment of rRNA sequences into the new Environmental Science Press, Beijing, pp. 211-230. bacterial taxonomy. Appl. Environ. Microbiol., 73: 5261-5267. Jukes TH, Cantor CR (1969). Evolution of protein molecules. In: Munro Winderl C, Anneser B, Griebler C, Meckenstock RU, Lueders T (2008). HH (ed) Mammalian protein metabolism. Academic Press, New York, Depth-resolved quantification of anaerobic toluene degraders and pp. 21-132. aquifer microbial community patterns in distinct redox zones of a tar Juretschko S, Loy A, Lehner A, Wagner M (2002). The microbial oil contaminant plume. Appl. Environ. Microbiol., 74: 792-801. community composition of a nitrifying-denitrifying activated sludge Wu L, Kellogg L, Devol AH, Tiedje JM, Zhou J (2008). Microarray-based from an industrial sewage treatment plant analyzed by the full-cycle characterization of microbial community functional structure and rRNA approach. Syst. Appl. Microbiol., 25: 84-99. heterogeneity in marine sediments from the Gulf of Mexico. Appl. Koizumi Y, Kojima H, Oguri K, Kitazato H, Fukui M (2004). Vertical and Environ. Microbiol., 74: 4516-4529. temporal shifts in microbial communities in the water column and Wu Y, Xiang Y, Wang J, Zhong J, He J, Wu QL (2010). Heterogeneity of sediment of saline meromictic Lake Kaiike (Japan), as determined by archaeal and bacterial ammonia-oxidizing communities in Lake Taihu, a 16S rDNA-based analysis, and related to physicochemical gradients. China. Environ. Microbiol. Rep., 2: 569-576. Environ. Microbiol., 6: 622-637. Xue LQ, Lv XW, Wu L (2004). Simulationg study on the inner source Lepš J, Šmilauer P (2003). Multivariate analysis of ecological data using nutrient exchange between water-sediment interface in the Xuanwu CANOCO. Cambridge University Press, Cambridge, pp. 43-60. Lake. Environ. Pollt. Contr., 26: 163-167. Li HB, Xing P, Chen MJ, Bian YQ, Wu QL (2011). Short-term bacterial Yang W, Ji L, Tan LR, Li SM, Wang Y, Liu HX, Luo YM (2011). Sensitive

2940 Afr. J. Microbiol. Res.

and specific detection of Agrobacterium tumefaciens in soil using a Zeng J, Yang LY, Liang Y, Li JY, Xiao L, Jiang LJ, Zhao DY (2008). rapid polymerase chain reaction (PCR). Afr. J. Microbiol. Res., 5: Spatial distribution of bacterial communities in sediment of a eutrophic 708-713. lake revealed by denaturing gradient gel electrophoresis and Yao M, Liu Q, Li YL, Yang XD (2009). Seasonal change of planktonic multivariate analysis. Can. J. Microbiol., 54: 1053-1063. diatom of two small shallow eutrophic lakes in Nanjing. J. Lake Sci., Zhao DY, Ma T, Zeng J, Yan WM, Wang JQ, Huang R, Wang M (2011). 21: 693-699. Effects of lake water chemistry on bacterioplankton community Zeng J, Bian Y, Xing P, Wu QL (2012). Macrophyte species drive the structures of three lakes. Afr. J. Microbiol. Res., 5: 4820-4831. variation of bacterioplankton community composition in a shallow Zhou J, Bruns MA, Tiedje JM (1996). DNA recovery from soils of diverse freshwater lake. Appl. Environ. Microbiol., 78: 177-184. composition. Appl. Environ. Microbiol., 62: 316-322. Zeng J, Yang L, Li J, Liang Y, Xiao L, Jiang L, Zhao D (2009). Vertical distribution of bacterial community structure in the sediments of two eutrophic lakes revealed by denaturing gradient gel electrophoresis (DGGE) and multivariate analysis techniques. World J. Microb. Biot., 25: 225-233.