• heavy metal pollution;
  • soil particle-size fractions;
  • microbial diversity;
  • DGGE ;
  • qPCR


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. Authors’ contribution
  10. References

Chemical and microbial characterisations of particle-size fractions (PSFs) from a rice paddy soil subjected to long-term heavy metal pollution (P) and nonpolluted (NP) soil were performed to investigate whether the distribution of heavy metals (Cd, Cu, Pb and Zn) regulates microbial community activity, abundance and diversity at the microenvironment scale. The soils were physically fractionated into coarse sand, fine sand, silt and clay fractions. Long-term heavy metal pollution notably decreased soil basal respiration (a measurement of the total activity of the soil microbial community) and microbial biomass carbon (MBC) across the fractions by 3–45% and 21–53%, respectively. The coarse sand fraction was more affected by pollution than the clay fraction and displayed a significantly lower MBC content and respiration and dehydrogenase activity compared with the nonpolluted soils. The abundances and diversities of bacteria were less affected within the PSFs under pollution. However, significant decreases in the abundances and diversities of fungi were noted, which may have strongly contributed to the decrease in MBC. Sequencing of denaturing gradient gel electrophoresis bands revealed that the groups Acidobacteria, Ascomycota and Chytridiomycota were clearly inhibited under pollution. Our findings suggest that long-term heavy metal pollution decreased the microbial biomass, activity and diversity in PSFs, particularly in the large-size fractions.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. Authors’ contribution
  10. References

Heavy metal pollution is increasingly threatening to crop production in agricultural soils (Weber et al., 2001). Paddy soils are considered to be unique anthropogenic soils in China and are responsible for one quarter of China's cereal production (Pan et al., 2003). However, heavy metal pollution has been widely reported in the paddies of South China in recent years due to the rapid development of mining activities, the metallurgical industry, sewage irrigation and the application of pesticides and fertilisers (Tan et al., 2005; Liu et al., 2006). The growth and function of microorganisms as well as the community composition and diversity in soil ecosystems can be severely affected by pollution (Bååth et al., 1998; Pérez-de-Mora et al., 2006). Although a number of studies have confirmed the negative influences of heavy metal pollution on soil microorganisms, few studies performed to date have fully integrated knowledge of the effects of long-term heavy metal pollution on the soil microbial abundance and community structure as well as total activity within microhabitats under field conditions. For example, changes in the microbial populations within particle-size fractions (PSFs) were assessed by Kandeler et al. (2000), who detected the microbial community based on analysis of phospholipids fatty acids (PLFAs) together with denaturing gradient gel electrophoresis (DGGE) following the addition of heavy metals to the soils.

PSFs of soils obtained via low-energy sonication have shed light on the complex relationships between soil diversity and organic carbon dynamics (Stemmer et al., 1998). The physical protection of soil organic carbon (SOC) in aggregates is considered a putative mechanism affecting SOC stability and turnover (Six et al., 2000; Paul et al., 2001). Acting as microhabitats for soil microorganisms, different sizes of soil particles provide spatially heterogeneous living conditions with varying nutrient levels, oxygen concentrations and water regimes (Ladd et al., 1996) and could therefore potentially influence the activity of microorganisms and dynamics of SOC.

In topsoil ecosystems, bacteria and fungi generally constitute more than 90% of the total soil microbial biomass and are key regulators of soil organic matter dynamics and nutrient availability (Six et al., 2006). Additionally, the production of mucilage by bacteria and fungi enhances the formation of aggregates (Oades, 1993; Six et al., 2004). Moreover, soil bacteria and fungi have been reported to exhibit different responses to heavy metal pollution (Rajapaksha et al., 2004), which could potentially lead to differences in the distribution and diversity of microbial communities within both bulk soils and PSFs. PSFs with higher SOC contents, particularly those containing more labile substrates, are usually accompanied by higher MBC accumulation and microbial diversity (Six et al., 2000; Roy & Singh, 1994). Recently, knowledge regarding the microscale spatial distribution of soil microbial communities has improved our understanding of the underlying mechanisms affecting the stability of organic carbon (Grundmann, 2004). Several studies have observed a higher biomass content and community diversity of bacteria in silt and clay fractions, whereas higher values of these parameters for fungi were mainly found in coarse sand fractions (Kandeler et al., 2000; Poll et al., 2003; Sessitsch et al., 2001). Zhang et al. (2007) also reported that the highest bacterial diversity, but lowest microbial activity, occurred in the < 2-μm PSF under different long-term fertilisation treatments and noted out that the organic carbon detected in the smallest PSFs was more stable. Different distributions of organic carbon within aggregates could, in turn, drive microbial activity and functional diversity (Gude et al., 2012; Lagomarsino et al., 2012). Based on the examination of new-generation pyrosequencing data coupled with mid-infrared spectroscopy, Davinic et al. (2012) suggested that the relative abundances of soil bacteria within soil aggregate fractions were driven more by shifts in the chemical composition of SOC, rather than in its quantity. Studies involving the separation of PSFs provide great insight into the microniches of microorganisms. However, it remains unclear whether the composition and abundance of the microbial community within different PSFs exhibit consistent changes under environmental stress, particularly for long-term heavy metal pollution disturbances.

We hypothesise that long-term heavy metal pollution may affect the abundance, diversity and activity of the soil microbial community inconsistently due to the varying availability of SOC among different sizes of soil particles. PSFs from soils collected from polluted and nonpolluted rice paddies were obtained through a combination of low-energy sonication, wet sieving and repeated centrifugation. The total and bioavailable concentrations of heavy metals were measured, as were the SOC, microbial biomass and respiration. Microbiological analyses were performed to reveal the changes in bacterial and fungal community abundance and diversity via qPCR and DGGE, respectively.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. Authors’ contribution
  10. References


The experiment was performed within a rice paddy field that has been contaminated with multiple heavy metals (mainly Pb, Cd, Zn and Cu) from a metallurgy plant since the 1970s, located in Yixing (31°24′N, 119°41′E), Jiangsu province, China. The area has been previously described by Liu et al. (2006) and Cui et al. (2011). For comparison, an adjacent nonpolluted rice paddy under consistent rice cultivation and crop management practices was also selected to represent background conditions. The soils were classified as Hydragric Anthrosols (IUSS Working Group WRB, 2007) and typical Ferric-accumulic Stagnic Anthrosols based on the Chinese Soil Taxonomical System (Gong, 1999). The local climate conditions are governed by a subtropical monsoon, showing a mean annual temperature of 15.7 °C and annual precipitation of 1177 mm. The farm has been traditionally cultivated under a rice and winter wheat rotation, with crop residues being returned to the fields. The physicochemical properties of the topsoil measured in the polluted and nonpolluted soils were as follows: pH (H2O), 5.8 and 6.0; SOC, 23.3 and 24.0 g kg−1; total nitrogen, 2.4 and 2.6 g kg−1; bulk density, 1.57 and 1.54 g cm−3; clay content, 399.9 and 311.6 g kg−1; and CEC, 22.7 and 23.63 cmol kg−1, respectively.

Soil sampling and treatment

To minimise the influence of plants and tillage on soil aggregation and microbiological properties, bulk topsoil (0–20 cm) was collected from each site in March 2011. At the time of sampling, the plots were planted with winter wheat, and the topsoil moisture was c. 40%. Three composite topsoil samples from each site were randomly collected at a distance of c. 5 m from each other. Each composite sample comprised twelve soil cores (c. 5 cm in diameter) and was transferred to a sterile plastic bag, sealed and placed on ice to be transported to the laboratory. After removing plant residues and gravels, the samples were gently and thoroughly homogenised and divided into two portions. One portion was subjected to particle-size fraction separation and subsequent aerobic incubation as fresh material, and the second portion was further air-dried at room temperature and passed through a 0.25-mm sieve for chemical analysis.

Particle-size fraction separation

Soil particle fractions of different sizes were obtained via low-energy sonication and a combination of wet sieving and centrifugation as described by Stemmer et al. (1998) and Sessitsch et al. (2001), with minor modifications. Fresh soil was dispersed in distilled water to allow the PSFs to become saturated. The soil–water suspension was then dispersed via low-energy sonication (output energy of 170 J g−1 dry soil). The coarse sand fraction (2000–200 μm) was separated via wet sieving. The 200–20-μm fraction was subsequently obtained through siphonage and sedimentation and was considered the fine sand fraction. The remainder of the sample was centrifuged to collect the 20–2-μm fraction (silt), and the supernatant was centrifuged again to collect the < 2-μm fraction (clay). The bulk soil and PSFs (triplicates) were freeze-dried and stored at −70 °C prior to nucleic acid extraction. The particle-size distributions for the examined soils are shown in Table 1.

Table 1. Particle-size distribution in the studied soils (means ± SD, n = 3)
Studied soilParticle-size distribution (%)Recovery (%)
Coarse sand (2000–200 μm)Fine sand (200–20 μm)Silt (20–2 μm)Clay (< 2 μm)
NP18.2 ± 5.933.2 ± 2.333.8 ± 1.414.9 ± 2.792.7
P20.5 ± 3.033.6 ± 2.431.9 ± 0.414.0 ± 0.893.8

Analysis of the soil properties and metal contents

Analyses of the basic properties and metal contents of the samples were performed following the protocols described by Lu (2000). Briefly, the soil pH was determined at a soil-to-water ratio of 1 : 2.5 using a precision pH meter (Mettler Toledo Seveneasy, Switzerland). The SOC and total nitrogen contents were determined via wet digestion through K2Cr2O7 oxidation and the Kjeldahl method, respectively. The topsoil bulk density was measured for each site using 100-cm3 cylinders. The gravimetric moisture content was measured by drying a portion of a finely ground soil sample at 105 °C for 24 h to a constant weight. The total heavy metal content was determined by digesting the samples with a mixed solution of HF/HClO4/HNO3 (10/2.5/2.5, v/v/v), and the bioavailable pool of heavy metals (the metal fractions in chemical forms that can be taken up by different soil organisms and by plants) (Mench et al., 2000) was extracted with a 0.01 M CaCl2 solution, using a soil/solution ratio of 1/5 (m/m). The Cd content was determined based on graphite furnace atomic absorption spectrometry (SpectrAA 220Z, Varian), while the levels of Pb, Cu and Zn were measured with a flame atomic adsorption spectrophotometer (TAS-986, Persee, China). The Nemerow pollution index (Nemerow, 1991) was used to further evaluate the degree of overall heavy metal pollution. Using the Standards of Soil Environmental Quality (GB15618-1995) as soil quality assessment criteria, the Nemerow pollution index was calculated according to the equation described by Liu et al. (2012). The total contents of heavy metals and the basic physicochemical properties of the bulk soil and PSFs are listed in Tables 2 and 3, respectively.

Table 2. Content of heavy metals and Nemerow indexes in the bulk soils and PSFs (means ± SD, n = 3)
Heavy metals (mg kg−1)TreatmentBulk soilParticle sizeRecovery (%)
Coarse sand (2000–200 μm)Fine sand (200–20 μm)Silt (20–2 μm)Clay (< 2 μm)
  1. a

    Bioavailable pool of heavy metal extracted with a 0.01 M CaCl2 solution.

Total PbNP61.2 ± 1.976.2 ± 7.854.3 ± 2.058.2 ± 5.368.0 ± 8.8100.7
P708.6 ± 27.7895.1 ± 38.7671.8 ± 24.5525.1 ± 35.9667.6 ± 28.194.6
Total CdNP0.5 ± 0.00.6 ± 0.10.4 ± 0.00.5 ± 0.00.6 ± 0.086.8
P24.2 ± 0.928.5 ± 1.117.0 ± 0.818.2 ± 1.026.7 ± 0.787.3
Total CuNP34.4 ± 0.240.4 ± 2.523.7 ± 0.931.3 ± 2.349.3 ± 1.996.1
P66.3 ± 3.8104.0 ± 8.559.7 ± 7.857.6 ± 2.5102.2 ± 8.3111.7
Total ZnNP78.3 ± 1.353.7 ± 6.137.0 ± 5.076.2 ± 0.8116.5 ± 5.983.1
P185.3 ± 1.1225.8 ± 26.492.9 ± 2.8109.1 ± 8.2148.8 ± 1.171.8
Nemerow IndexNP0. 
Bioavailable PbaNP0.1 ± 0.00.0 ± 0.00.1 ± 0.00.1 ± 0.00.1 ± 0.077.3
P0.7 ± 0.10.6 ± 0.00.5 ± 0.00.6 ± 0.00.6 ± 0.084.9
Bioavailable CdaNP0.0 ± 0.00.1 ± 0.00.1 ± 0.00.1 ± 0.00.0 ± 0.0137.9
P3.7 ± 0.04.2 ± 0.03.3 ± 0.13.3 ± 0.03.7 ± 0.096.3
Bioavailable CuaNP0.2 ± 0.00.2 ± 0.00.2 ± 0.00.1 ± 0.00.1 ± 0.072.8
P0.2 ± 0.00.3 ± 0.00.3 ± 0.10.2 ± 0.00.2 ± 0.0105.9
Bioavailable ZnaNP0.3 ± 0.00.5 ± 0.10.6 ± 0.00.6 ± 0.00.6 ± 0.2200.1
P3.0 ± 0.03.1 ± 0.02.9 ± 0.13.0 ± 0.13.4 ± 0.1101.4
Table 3. Soil properties of the bulk soils and PSFs (means ± SD, n = 3)
Soil propertyTreatmentBulk soilParticle size
Coarse sand (2000–200 μm)Fine sand (200–20 μm)Silt (20–2 μm)Clay (< 2 μm)
SOC (mg g−1)NP24.0 ± 0.240.1 ± 1.919.4 ± 0.421.8 ± 0.525.5 ± 1.9
P23.3 ± 0.539.1 ± 2.222.3 ± 0.320.3 ± 1.525.5 ± 2.2
Total N (mg g−1)NP2.6 ± 0.13.9 ± 0.22.0 ± 0.02.4 ± 0.23.2 ± 0.1
P2.4 ± 0.13.3 ± 0.42.0 ± 0.12.2 ± 0.12.7 ± 0.1
C/NNP9.2 ± 0.210.2 ± 0.29.6 ± 0.18.9 ± 0.78.1 ± 0.4
P9.9 ± 0.312.1 ± 0.810.9 ± 0.69.3 ± 0.49.2 ± 0.4
DNA yields (× 103 ng g−1 dry soil)NP7.8 ± 0.77.7 ± 0.33.8 ± 0.35.8 ± 0.65.2 ± 0.1
P7.6 ± 0.48.2 ± 0.62.7 ± 0.33.3 ± 0.45.0 ± 0.4

Analysis of soil microbial biomass C and enzyme activity

The MBC contents in the bulk soils and PSFs were measured using a fumigation–extraction method, according to the protocol described by Vance et al. (1987). The MBC contents were determined with a Multi N/C 2100 analyser (Analytik Jena, Germany), using an extraction efficiency coefficient of 0.45 (the portion of MBC extracted in a K2SO4 solution during this procedure) to convert the measured C to MBC. Dehydrogenase activity (Dehy) was determined based on the reduction in triphenyltetrazolium chloride to triphenylformazan following the protocol described by Serra-Wittling et al. (1995).

Soil basal respiration

Soil basal respiration was determined according to the protocol described by Zheng et al. (2007), with minor modifications. Briefly, a 20-g (fresh weight) sample was moistened to 60% of its water-holding capacity and incubated under aerobic conditions at 25 °C in the dark for 7.5 day in a 125-mL jar. A 0.25-mL gas sample was collected from the head space of the jar using a gas-tight syringe at 0.5, 1, 1.5, 2, 3, 4, 5.5 and 7.5 day after incubation was initiated. The CO2 concentration was analysed with a gas chromatography (Agilent 4890D) equipped with a flame ionisation detector. The basal respiration rate was expressed on a dry soil basis. The metabolic quotient (the respiration per unit biomass) was used as an indicator of the C utilisation efficiency in the soil.

DNA extraction and real-time PCR

Soil DNA was extracted using the PowerSoil® DNA isolation kit (Mo Bio Laboratories Inc., CA) according to the manufacturer's protocol in a MoBio vortex adapter tube holder.

Real-time PCR was performed in an iCycler iQ 5 thermocycler (Bio-Rad) via fluorometric monitoring with SYBR Green 1 dye. The primer pair 338F/518R (Fierer et al., 2005) was used to quantify bacterial 16S rRNA genes, while the primer pair NS1-F/FungR (May et al., 2001) was used to quantify fungal 18S rRNA genes. The DNA extracts were diluted 10-fold and used as a template, with a final content of c.1–10 ng in each reaction mixture. The reactions were performed in a 25 μL volume containing 10 ng of DNA, 0.2 mg mL−1 BSA, 0.2 μM each primer and 12.5 μL of SYBR premix EX Taq (Takara Shuzo, Shiga, Japan). A melting curve analysis was conducted following each assay to confirm specific amplification.

A standard curve was constructed using plasmids from cloned rRNA genes separately for bacterial and fungal genes, according to the protocol described by Liu et al. (2012). Briefly, the amplified PCR products were purified and ligated into the pEASY-T3 cloning vector (Promega, Madison, WI) and cloned into Escherichia coli DH5α. The plasmid DNA in the positive clones was isolated using a plasmid extraction kit (Takara, Japan). The number of copies of the standards was calculated from the concentration of the extracted plasmid DNA. A 10-fold dilution series of the plasmid DNA was produced to generate a standard curve covering seven orders of magnitude, from 102 to 108 copies of the template. Real-time PCR was conducted in triplicate, and amplification efficiencies of 99.1–103.2% were obtained, with R2 values of 0.993–0.997. The final copy numbers of the bacterial 16S rRNA genes and fungal 18S rRNA genes were obtained by calibrating the total DNA concentration and soil water content. The relative ratio of fungi/bacteria was calculated as the ratio of the fungal gene copy numbers to bacterial gene copy numbers (Fierer et al., 2005).

PCR-DGGE assay

The bacterial and fungal community structures were determined via DGGE following a previously described protocol (Liu et al., 2012). Briefly, PCR amplification was performed with the community-specific primer pairs F968/R1401 (Heuer et al., 1997) and NS1-F/FungR (May et al., 2001) for the bacterial and fungal communities, respectively, but with a GC clamp (Muyzer et al., 1993) attached to the 5′ end of each reverse primer. The PCR amplification was performed in an Eppendorf gradient cycler (Eppendorf, Hamburg, Germany) using a 25-μL reaction mixture containing 1 μL of a 10 μM solution of each primer, 1 μL of DNA template and 12.5 μL of Go Taq® Green Master Mix (Promega). DGGE was performed using the DCode universal mutation detection system (Bio-Rad Laboratories) according to the instruction manual. The obtained PCR products were separated in 8% (w/v) polyacrylamide [acrylamide–bisacrylamide (37.5 : 1)] gels with denaturing gradients of 40–60% for bacteria and 20–40% for fungi (with a 100% denaturant containing 7 M urea and 40% formamide). The gels were subjected to silver staining prior to UV transillumination with the Gel Doc-2000 Image Analysis System (Bio-Rad). The band intensity in each lane was digitised using Quantity One image analysis software (version 4.6.2, Bio-Rad). The Shannon diversity index, (H′), was also calculated based on the relative intensity of each DGGE band according to the formula described by Hedrick et al. (2000).

Sequencing and phylogenetic analysis

Some important bands corresponding to bacterial 16S rRNA gene and 18S rRNA genes were detected in the DGGE gels and numbered on the basis of their specific positions or their relative intensity between the polluted and nonpolluted soils across different PSFs and bulk soils. The specific bands were carefully excised from the DGGE gels, and the gel slices were crushed and resuspended overnight at 4 °C in 25 μL of sterile water to elute the DNA. A 2-μL aliquot of eluted DNA was re-amplified and cloned into E. coli as described above in the real-time PCR assay section. White clones were selected and subjected to DGGE again to confirm their identity. Finally, the clones showing the same mobilities as the bands excised from the DGGE gels were sequenced. After sequencing, we found that the bands showing the same mobilities in the DGGE gels exhibited the same nucleotide sequences. Therefore, we used one sequence of each numbered band for phylogenetic analysis. The sequences were checked for chimeras using Bellerophon (Huber et al., 2004), and the putative chimeras were excluded from further analysis. The closest relatives of each sequence were checked via blast (Basic Local Alignment Search Tool) searches within GenBank [National Center for Biotechnology Information NCBI)]. The sequences of the DGGE bands have been deposited in GenBank under accession numbers KC412070-KC412089 (bacterial 16S rRNA genes) and KC412090-KC412098 (fungal 18S rRNA genes).

Data analysis

The presentation and treatment of the data were processed with Microsoft Excel 2003, and the results were expressed as the means plus or minus one standard deviation from three replicates. The statistical procedures were conducted with the software package spss 16.0 for Windows. An independent samples t-test was used to check the differences between the polluted and nonpolluted samples. Principal component analysis (PCA) of the DGGE profiles was conducted using Minitab 15 software to elucidate microbial community structures based on relative band intensity and positions.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. Authors’ contribution
  10. References

Distribution of heavy metals in the topsoil layers of the bulk soils and PSFs

Physical fractionation procedures according to particle size yielded high recovery rates for heavy metals (71.8–111.7%, Table 2). While the total Cu and Zn contents were c. 2 times higher in the polluted PSFs than in the nonpolluted bulk soils and PSFs, the total Pb and Cd contents were c. 11 and 46 times higher in the polluted PSFs than in the nonpolluted soils, respectively. Different PSFs (coarse sand, fine sand, silt and clay) were characterised by consistent differences among the four detected heavy metals in terms of their total contents, with the highest content being observed in the coarse sand fraction, followed by the clay, silt and fine sand fractions. As shown by the values of the Nemerow pollution index for the polluted soil, the degree to which the PSFs were polluted with heavy metals fell in the order coarse sand > clay > silt > fine sand. Similar to the total heavy metal distribution, the average bioavailable Pb, Cd, Cu and Zn contents in the polluted PSFs were c. 13, 72, 2 and 6 times higher than those in the nonpolluted soils, respectively. Nevertheless, the contents of bioavailable metals varied little between the PSFs.

SOC, total N, C/N and DNA yields in the bulk soils and PSFs

As listed in Table 3, there were generally few differences in soil properties detected between the polluted and the nonpolluted bulk soils and PSFs, with the exception of total N and C/N. The highest contents of SOC and total N (TN) and highest DNA yields in the nonpolluted soils were detected in the coarse sand fraction, followed by the clay, silt and fine sand fractions. In contrast, the C/N ratio markedly decreased with decreasing particle size. The soils subjected to heavy metal pollution showed a similar trend, but displayed a significant decrease in TN within the coarse sand and clay fractions as well as an increase in the C/N ratio across all fractions. The coarse sand fraction from the polluted soils exhibited the highest degree of pollution (Nemerow index), along with the highest SOC, TN and DNA yields.

Soil MBC, respiration and enzyme activity

As shown in Table 4, similar to the tendency of SOC in the PSFs, MBC and soil basal respiration were shown to be highest in the coarse sand fraction and lowest in the clay fraction, indicating that although both the coarse sand and clay fractions exhibited higher SOC, they presented different biochemical characteristics. The PSFs subjected to heavy metal pollution showed decreases in MBC contents and basal respiration of 21–53% and 3–45%, respectively, resulting in generally higher metabolic quotients compared with the nonpolluted PSFs, suggesting a severe impact of pollution on the mineralisation of SOC and microbial activity. A consistently lower dehydrogenase activity across the polluted PSFs confirmed that there was low biological activity in the polluted soils.

Table 4. Soil MBC, basal respiration and dehydrogenase activity in the studied soils (means ± SD, n = 3)
Soil propertyTreatmentBulk soilParticle size
Coarse sand (2000–200 μm)Fine sand (200–20 μm)Silt (20–2 μm)Clay (< 2 μm)
MBC (mg kg−1)NP940.9 ± 60.5733.2 ± 53.5607.8 ± 34.7721.8 ± 56.6297.8 ± 17.8
P566.0 ± 40.6462.8 ± 19.0288.4 ± 14.1421.1 ± 67.0235.9 ± 41.4
MBC/SOC (%)NP3.9 ± 0.21.5 ± 0.12.6 ± 0.12.8 ± 0.11.0 ± 0.1
P2.4 ± 0.21.0 ± 0.11.1 ± 0.01.8 ± 0.10.8 ± 0.1
Basal respiration (mg CO2-C kg−1 soil)NP217.5 ± 5.8284.7 ± 12.1150.6 ± 6.6163.2 ± 4.080.3 ± 6.5
P111.1 ± 5.0182.4 ± 4.996.4 ± 0.3110.7 ± 5.579.2 ± 3.0
Metabolic quotient (mg CO2-C g−1 MBC h−1)NP1.3 ± 0.12.6 ± 0.11.6 ± 0.11.5 ± 0.01.8 ± 0.1
P1.1 ± 0.12.6 ± 0.12.2 ± 0.01.7 ± 0.12.2 ± 0.1
Dehydrogenase activity (μg triphenylformazan g−1 24 h−1)NP27.4 ± 0.719.2 ± 1.227.6 ± 1.528.5 ± 1.84.5 ± 1.6
P18.9 ± 1.512.7 ± 3.119.3 ± 1.219.1 ± 1.32.4 ± 0.5

Abundance of bacterial and fungal genes in the PSFs

The largest bacterial population size in the nonpolluted soil was detected in the coarse sand fraction, followed by the clay, silt and fine sand fractions in a significantly decreasing order (Fig. 1a). Few differences were observed between the polluted and nonpolluted soil fractions, indicating that heavy metal pollution had less of an effect on bacterial abundance. Regarding fungal abundance, the population size decreased with particle size in both the polluted and nonpolluted soils (Fig. 1b). In particular, the fungal copy numbers in the clay fraction were decreased by 61% compared with the coarse sand fraction in the nonpolluted soils. Overall, the fungal 18S rRNA gene copy numbers detected in the polluted soils were lower than those in the nonpolluted soils, with the exception of the fine sand fraction. The relative fungal/bacterial ratio was clearly decreased due to the diminished fungal population size. Heavy metal pollution resulted in a much lower fungal/bacterial ratio in the bulk soils and silt and clay fractions than was observed in the nonpolluted soils (Fig. 1c). Compared with the bacterial population size, the fungal population size was more sensitive to heavy metal pollution stress.


Figure 1. Abundance of total bacteria (a), total fungi (b) and the relative fungal/bacterial ratio (c) based on qPCR analysis across bulk soils and PSFs from nonpolluted (NP, white bar) and polluted (P, shaded bar) soils. The asterisks over the error bars indicate a significant difference between nonpolluted and polluted PSFs at the same particle size based on a t-test at < 0.05 (n = 3; error bars are + SD).

Download figure to PowerPoint

Pearson's correlations

The correlation coefficients between soil heavy metals and chemical and biochemical properties are shown in Table 5. No correlations were observed between soil heavy metals and SOC or TN, whereas positive correlations were found between heavy metals and C/N. In general, both total and bioavailable heavy metals were negatively correlated with MBC, MBC/SOC, basal respiration and dehydrogenase activity, which indicated that pollution had a significant impact on the biochemical characteristics of the soil. No significant correlations were noted between heavy metals and bacterial abundance, but significantly negative correlations were found between bacterial abundance/fungal abundance (F/B) and total Cu, and bioavailable Pb and Zn.

Table 5. Correlation coefficients between soil heavy metals, chemical and biochemical properties
 SOCTNC/NMBCMBC/SOCRespirationMetabolic quotientBacterial abundanceFungal abundanceF/BDehyH′ of BacteriaH′ of Fungi
  1. n = 30 (number of samples used to calculate correlation coefficients).

  2. **Correlation is significant at the 0.01 level. *Correlation is significant at the 0.05 level.

Total Pb0.165−0.1340.614**−0.501**−0.517**−0.624**0.2920.029−0.177−0.259−0.423*−0.142−0.486**
Total Cd0.137−0.1240.524**−0.497**−0.502**−0.634**0.2480.077−0.219−0.349−0.478**−0.242−0.487**
Total Cu0.3550.1690.465**−0.567**−0.662**−0.723**0.461*0.299−0.124−0.384*−0.708**−0.271−0.515**
Total Zn0.3180.1480.405*−0.327−0.410*−0.618**0.1330.2870.000−0.182−0.548**0.013−0.210
Bioavailable Pb−0.025−0.2660.441*−0.507**−0.453*−0.624**0.136−0.071−0.315−0.375*−0.402*−0.288−0.503**
Bioavailable Cd0.076−0.2010.538**−0.520**−0.501**−0.627**0.237−0.019−0.259−0.342−0.420*−0.252−0.528**
Bioavailable Cu0.063−0.2080.549**−0.356−0.360−0.456*0.2360.000−0.141−0.181−0.304−0.109−0.372*
Bioavailable Zn−0.007−0.2540.458*−0.586**−0.529**−0.657**0.224−0.085−0.365*−0.443*−0.443*−0.367*−0.603**
Nemerow index0.138−0.1240.525**−0.497**−0.502**−0.635**0.2490.077−0.219−0.349−0.478**−0.241−0.487**

Bacterial community analysis

DGGE analyses of 16S rRNA gene sequences revealed shifts in the bacterial community structure in the bulk soils and PSFs under pollution compared with the nonpolluted soils (Fig. 2). The analysis of three replicates for each treatment showed good reproducibility of the DGGE banding patterns. The numbered bands in the DGGE gels were excised for sequencing. Bands B11, B13-B17, B19 and B20, which were dominant in all of the fractions and bulk soils, exhibited similar intensities across all of the treatments, regardless of pollution, indicating that all of the fractions exhibited a similar predominant bacterial community. However, the bulk soils and coarse sand fraction exhibited more diverse banding patterns, with the additional bands B4, B6, B9, B12 and B18 being observed compared with the other fractions, suggesting a distinct, higher bacterial diversity, which was reflected in the higher Shannon's diversity index H′ (Fig. 3). The intensities of bands B5, B6 and B8 decreased with decreasing particle size, which indicated that some bacterial species preferred the macroscale habitats. Heavy metal pollution had less of an impact on the bacterial community structure than the particle size. The intensities of bands B1–B4 were clearly decreased under pollution. In contrast, bands B7 and B10 were unique and were enhanced by pollution. Nevertheless, the H′ diversity index for the silt and clay fractions derived from the DGGE profiles was significantly lower than in the corresponding nonpolluted soils (Fig. 3).


Figure 2. DGGE profiles of the bacterial communities from bulk soils and PSFs from polluted (P) and nonpolluted soils (NP). Sample designations are indicated above each DGGE lane. Arrows indicate the bands (B1-B20) excised for sequencing.

Download figure to PowerPoint


Figure 3. The H′ community diversity index derived from the DGGE profiles of amplified bacterial 16S rRNA genes (a) and fungal 18S rRNA genes (b) from bulk soils and PSFs from polluted (P) and nonpolluted soils (NP). The asterisks over the error bars indicate a significant difference between the polluted and nonpolluted soils at the same particle size based on a t-test at < 0.05 (n = 3; error bars are + SD).

Download figure to PowerPoint

The PCA of the bacterial DGGE profiles from the bulk soils and PSFs yielded a good separation between the different PSFs as well as the polluted and nonpolluted samples, although the first two components explained only 47.8% of the total variability (Fig. 4a). The bacterial community structures in the bulk soils and coarse sand fraction formed a cluster that was separate from the other fractions in PCA-1 regardless of pollution. Moreover, although PCA-2 explained only 13.5% of the total variability, all the fractions subjected to pollution were well separated from those of the nonpolluted soils in PCA-2.


Figure 4. PCA of the amplified 16S rRNA gene fragments from the bacterial community (a) and 18S rRNA gene fragments from the fungal community (b) in bulk soils and PSFs from polluted (P) and nonpolluted soils (NP).

Download figure to PowerPoint

Fungal community analysis

DGGE analyses of 18S rRNA gene sequences revealed more changes in the fungal community than in the bacterial community (Fig. 5). Bands F2–F8 were dominant in the soils; however, among these bands, only bands F4–F6 were unchanged across the PSFs, indicating that the distribution of the fungal community was strongly influenced by size fractionation. Bands F1 and F9 were unique to the larger fractions or bulk soils. Heavy metal pollution seemed to have a much stronger impact on the fungal community structure, resulting in less diverse banding patterns, with lower intensities or a lack of bands F2, F3 and F8 observed under pollution. The H′ diversity index of the fungal community was much lower than that of the bacterial community (Fig. 3). In addition, the H′ indexes of the PSFs decreased with a decreasing particle size and were notably lower than in the bulk soils, which may implicate a disturbance or destruction of fungal hyphae during the separation procedure. Heavy metal pollution markedly reduced the fungal diversity index across the PSFs, as reflected in the decreasing number of bands observed in the DGGE patterns. Significantly negative correlations were found between all of the heavy metals and the fungal H′ diversity index (Table 5). The PCA of the fungal DGGE profiles also showed strong separation of community structures between the polluted and nonpolluted soils, which further supported the results regarding decreased fungal community diversity under pollution (Fig. 4b). Meanwhile, the community structures obtained for the coarse sand fraction were greatly separated from those of the other PSFs, which indicated that the separation of PSFs may have an impact on the fungal community structure.


Figure 5. DGGE profiles of the fungal communities in bulk soils and PSFs from polluted (P) and nonpolluted soils (NP). Sample designations are indicated above each DGGE lane. Arrows indicate the bands (F1–F9) excised for sequencing.

Download figure to PowerPoint

Phylogenetic affiliations

To examine the differences in the microbial community compositions in the PSFs in greater detail, specific bands were excised and sequenced from the DGGE gels. Phylogenetic assignment was performed via blast searches of the NCBI database, and the results are listed in Table 6 (for bacterial community 16S rRNA genes) and Table 7 (for fungal community 18S rRNA genes). The sequencing results showed that the bacterial community (bands B1–B3, B11, B13-B17, B19 and B20) in the studied soils was dominated by Acidobacteria, Betaproteobacteria, Gammaproteobacteria and Chloroflexi. The additional bands B4, B6, B9, B12 and B18 found in the coarse sand fraction mostly belonged to Acidobacteria, which suggested that the Acidobacteria might survive well in the macrohabitats. Bands B5, B6 and B8, which decreased with decreasing particle size, were affiliated to Acidobacteria and Gammaproteobacteria. The bands inhibited by pollution were assigned to Chlamydiae (band B1) and Acidobacteria (bands B2-B4), while bands B7 and B10, which was enhanced in the coarse and fine sand fractions, belonged to Chlamydiae and TM7.

Table 6. blast results for the nucleotide sequences of the DGGE bands generated for bacterial 16S rRNA genes
BandsAccession numberClosest NCBI database matchIdentity%Phylogenetic affiliationsAccession numberLocation in soil
  1. +/− indicates that bands are enhanced or decreased by pollution.

B1 KC412070 Uncultured bacterium clone H-LN-B3-Min_551696 Chlamydiae JQ051743 All fractions (−)
B2 KC412071 Uncultured Acidobacteria bacterium clone SNNP_2012-8399 Acidobacteria JX114416 All fractions (−)
B3 KC412072 Uncultured soil bacterium clone C09498 Acidobacteria AF507414 All fractions (−)
B4 KC412073 Uncultured Acidobacteria bacterium clone AEG_08_05399 Acidobacteria HQ597068 Bulk soil and coarse sand (−)
B5 KC412074 Uncultured Acidobacteria bacterium clone KBS_T1_R5_149261_f299 Acidobacteria HM062256 Mainly coarse sand
B6 KC412075 Uncultured Acidobacterium sp. clone MBT197 Acidobacteria FJ538114 Bulk soil
B7 KC412076 Uncultured bacterium clone H-LN-B3-Org_22598598 Chlamydiae JQ059622 Bulk soil and coarse sand (+)
B8 KC412077 Uncultured gammaproteobacterium clone P299 Gammaproteobacteria GU169060 All fractions
B9 KC412078 Uncultured bacterium clone SymTree1-8298TM7 JF416143 Bulk soil
B10 KC412079 Uncultured bacterium clone FCPN69594TM7 EF516216 Bulk soil and coarse sand (+)
B11 KC412080 Uncultured bacterium clone p7i09ok99 Acidobacteria FJ479036 All fractions
B12 KC412081 Uncultured bacterium clone AKAU412499 Acidobacteria DQ125874 Mainly bulk and coarse sand
B13 KC412082 Uncultured bacterium clone sdm12799 Gammaproteobacteria JQ798530 All fractions
B14 KC412083 Uncultured bacterium clone spz599 Chloroflexi JF708709 All fractions
B15 KC412084 Geobacillus sp. GAM23100 Firmicutes JX263000 All fractions
B16 KC412085 Uncultured archaeon clone R2094 Chloroflexi FJ184664 All fractions
B17 KC412086 Uncultured Pseudomonadales bacterium clone Plot18-E0295 Gammaproteobacteria FJ889295 All fractions
B18 KC412087 Uncultured Acidobacteria bacterium clone SEG_08_13599 Acidobacteria HQ597798 Bulk soil and coarse sand
B19 KC412088 Uncultured betaproteobacterium gene99 Betaproteobacteria AB288552 All fractions
B20 KC412089 Uncultured bacterium clone ncd2491b08c199 Betaproteobacteria JF215000 All fractions
Table 7. blast results for the nucleotide sequences of the DGGE bands generated from fungal 18S rRNA genes
BandsAccession numberClosest NCBI database matchIdentity%Phylogenetic affiliationsAccession numberLocation in soil
  1. − indicates that bands are decreased by pollution.

F1 KC412090 Uncultured Ichthyophonida isolate DGGE gel band 1496Eukaryota (Ichthyosporea) FJ236952 Mainly bulk, coarse and fine sand
F2 KC412091 Cladosporium sp. 6027100Ascomycota (Dothideomycetes) FJ235525 All fractions (−)
F3 KC412092 Uncultured Fusarium clone MPJ-B3-5100Ascomycota (Sordariomycetes) JN166418 All fractions (−)
F4 KC412093 Uncultured eukaryote clone Litter_25899 Fungi incertae sedis HQ199575 All fractions
F5 KC412094 Tilletia iowensis AFTOL-ID 171299Basidiomycota (Exobasidiomycetes) DQ832252 All fractions
F6 KC412095 Sarcoleotia turficola strain H25339798Ascomycota (Geoglossomycetes) AY789276 All fractions
F7 KC412096 Rhizophagus irregularis partial 18S rRNA gene, strain MUCL4319596Glomeromycota (Glomeromycetes) FR750228 All fractions (−)
F8 KC412097 Uncultured Chytridiomycota clone T5P2AeD0295 Chytridiomycota GQ995333 All fractions (−)
F9 KC412098 Uncultured fungus clone 3-498 Fungi incertae sedis JX126760 Bulk soil

Ascomycota and Basidiomycota were dominant in the fungal community (bands F2, F3, F5 and F6). Band F7 (Glomeromycota), which was also predominant in the majority of the fractions, nearly disappeared in the clay fraction. Heavy metal pollution had a strong effect on bands F2, F3 and F8, which belonged to Ascomycota and Chytridiomycota. The unique thin bands, F1 and F9, found in the bulk and coarse sand fractions were affiliated to Ichthyosporea and unclassified groups. The results suggested that fungi with hyphae larger than 2 μm may not be present in the clay fraction, and heavy metal pollution has a strong impact on the fungal community.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. Authors’ contribution
  10. References

SOC, MBC and soil respiration in PSFs under heavy metal pollution

A number of studies have shown that agricultural land management practices, such as crop planting, tillage and the irrigation state of the field, could affect the formation of aggregates via altering soil physical, chemical and biological properties (Kandeler & Murer, 1993; Six et al., 2000; Jiang et al., 2011; Gude et al., 2012). Such practices are also important drivers of microbial community dynamics and soil chemical processes. However, based on the objectives of this study, we chose to focus only on the effect of particle size under heavy metal pollution in field conditions. Because it is difficult to find two or more representative field plots under a similar state of long-term contamination, we finally selected two field plots under consistent crop cultivation and management practices as research sites, one of which had been polluted for c. 40 years, while the other was nonpolluted. Soil composite sampling was performed before the plots were ploughed and flooded for rice transplantation to minimise the influence of tillage. We acknowledge that although composite sampling likely underestimates the variability between the plots, it can also reduce heterogeneity that may exist in the field, and this methodology is commonly adopted and well accepted, particularly in ecological investigations of, for example, soil, water and sediment samples (Pennanen et al., 1996; Sekiguchi et al., 2002; Chiu et al., 2006; Farkas et al., 2007; Liu et al., 2012). Despite the limitations of the applied sampling design, the findings of the present study shed light on the impacts of variations in heavy metal pollution concerning decreasing microbial activity and diversity between PSFs.

Heavy metals, and particularly their bioavailable concentrations, can potentially affect soil quality-related parameters such as enzyme activities or related processes. Although decreases in MBC and in the culturable microbial population size from homogenised bulk soil samples under long-term heavy metal pollution are well documented in the literature (Vásquez-Murrieta et al., 2006; Li et al., 2009; Liu et al., 2012), the information regarding microbial populations at the microenvironment scale under pollution is scant. Separation of soil particles provides insight into the association between the distribution of organic C, heavy metals and microbial biomass and activity at a relatively small scale. In the present study, the total heavy metal contents were found to be markedly higher in both polluted bulk soils and PSFs than in the background soils. However, the contents of bioavailable heavy metals showed few differences between the PSFs, which may indicate an equal toxicity to microorganisms in these fractions. Nevertheless, heavy metal pollution resulted in a significantly lower MBC contents and soil respiration across the PSFs, regardless of the SOC quantity they contained. Significantly negative correlations between total and bioavailable heavy metals and the respiration, MBC contents and dehydrogenase activity were also observed, which indicated the effects of heavy metal pollution on the biological characteristics across the PSFs. Our results were in agreement with the findings of Li et al. (2009), who observed decreases in microbial biomass, enzymes and function under long-term acid metal stress.

At the microenvironment scale, coarse sand fractions are characterised by high concentrations of labile C and N originating predominately from plant residues, whereas silt and clay fractions are usually characterised by high concentrations of relatively stable organic C and N (Elliott, 1986; Six et al., 2000). Moreover, high SOC contents are usually found in clay and silt fractions, whereas low contents are observed in the sand fractions of most soils, such as in Calcaric Phaeozems (Kandeler et al., 2000), Calcic Chernozems and Cambisols (Stemmer et al., 1998) and Humic Dystrudepts (Chiu et al., 2006). In the present study, however, we detected the highest SOC concentrations in the coarse sand fractions, followed by the clay fractions, which generally agrees with our previous findings in rice paddy soils (Zhang et al., 2007; Zheng et al., 2007). Indeed, a similarly higher SOC content in the coarse sand fraction was also noted under rape/rice rotation in a Hydragric Anthrosol in which crop residues were returned to the soil (Jiang et al., 2011). It has been demonstrated that the organic C (fresh or labile) derived from crop residues is first incorporated into the coarse sand fraction during the initial decomposition period and subsequently accumulates and becomes stable in silt or clay soils (Angers et al., 1997; Six et al., 2000). Consequently, it is possible that higher fresh SOC contents could first accumulate in the coarse sand fractions, particularly in soils receiving large amounts of crop residues. In the present study, according to the observed SOC content and PSFs distribution (%) in the coarse sand fraction, we detected a large contribution of coarse sand to the SOC content of the bulk soil. Another reason for the higher SOC content observed in the coarse sand than in the clay fraction may be the high contribution of finer minerals (i.e. particulate organic matter) and clay to the coarse sand fraction during the formation of macroaggregates, as microaggregates are formed within macroaggregates (Six et al., 2000).

The variations in microbial biomass and activity observed between particles of different sizes under heavy metal pollution could likely be derived from differences in their available SOC content. The higher bioavailable SOC content in the coarse sand fraction could lead to increases in MBC because the bioavailable SOC derived from fresh residue is an important C source for microbial activity (Six et al., 2000). Singh & Singh (1995) reported that MBC contents were higher in the coarse sand fraction than in the clay fraction across forest, savannah and crop soils. However, the coarse sand fractions subjected to heavy metal pollution showed greater decreases in MBC and soil basal respiration than the clay fractions (Table 4), which could be attributed to the significant reduction in both bacterial and fungal gene copy numbers in the former fractions. This type of response also suggested a persistent toxicity or stress effect on soil microbial activity associated with long-term field conditions. The sharply decreased dehydrogenase activity observed in the coarse and fine sand fractions under pollution further supported the inhibitory effect of heavy metals on microbial activity or regarding low SOC availability. Dehydrogenase is an intracellular enzyme that participates in oxidative phosphorylation in microorganisms. This enzyme has often been correlated with the availability of organic carbon in soils (Serra-Wittling et al., 1995) and is assumed to be linked to microbial respiratory processes (Insam, 2001). Thus, it is likely that the toxicity of metals has a much greater impact on soil respiration and MBC in the coarse sand fraction than in the clay fraction. In contrast, the smaller decrease in respiration observed in the clay fraction under pollution compared with nonpolluted conditions in the present study could be attributed to the lower bioavailability of organic C in this fraction. Previous studies have confirmed that the clay fraction contains higher contents of carbon compounds (e.g. polyethylene and lipids) that are recalcitrant to decomposition by microorganisms and are generally more stable (Spaccini et al., 2001; Chen & Chiu, 2003). The clay fraction from rice paddies has also been found to show higher contents of humus and humic acid than the other fractions (Ding et al., 2006). In a recent study, Davinic et al. (2012) further revealed that although the silt and clay fractions exhibit a high SOC content, it consists of older and more stable forms of C. The much lower dehydrogenase activity observed in the clay fraction therefore revealed a lower availability of C and microbial activity. Alternatively, microorganisms might be more resistant to environmental stress due to the accessibility of substrates and protective agents in the clay fraction (Sessitsch et al., 2001). We further found a significantly higher metabolic quotient across the PSFs under heavy metal pollution, which has frequently been reported in studies in response to heavy metal contamination (Brookes, 1995). Positive correlations were observed between the total Pb and Cu contents and the metabolic quotient, which indicated less efficient C utilisation or higher maintenance energy requirements under adverse conditions (Giller et al., 1998). In fact, the differences in respiration observed in response to heavy metal stress may indicate that there were shifts in microbial community diversity within the PSFs.

Bacterial and fungal abundance in the PSFs under heavy metal pollution

To determine the differences in microbial abundance that contributed to the decreased microbial biomass observed under pollution, a real-time PCR assay targeting the bacterial 16S rRNA genes and fungal 18S rRNA genes was applied. Compared with the nonpolluted soils, a lower bacterial abundance was only found in the coarse sand fraction, which indicated that heavy metal pollution had few effects on the bacterial populations within the PSFs. Fresh residues or relatively labile constituents that are abundant in the coarse sand fraction could be beneficial for the bacterial population (Denef et al., 2001). Moreover, the smaller pore size and high resistance to desiccation and protection against predation found in the clay fraction could be beneficial for bacterial inhabitation (Rutherford & Juma, 1992). Most groups of soil bacteria could survive well, and a few could even be promoted under pollution, as observed in the bacterial DGGE profiles, suggesting tolerance of the bacterial community to heavy metals. An unchanged bacterial abundance was also observed by Liu et al. (2012), who reported no differences in bacterial abundance at two sites among four rice paddies subjected to long-term heavy metal pollution.

Previous studies have demonstrated that fungi prefer to feed on decomposable SOC and predominantly colonise labile and light fractions (Frey et al., 2003; Simpson et al., 2004). Fungal hyphae are also considered to improve aggregate stability by binding particles with extracellular polysaccharides and enmeshing microaggregates to form macroaggregates (Haynes & Beare, 1997; Bronick & Lal, 2005). Thus, it is not surprising that fungal abundance was found to be highest in the coarse sand fraction and lowest in the clay fraction in the present study. A heterogeneous tolerance of microorganisms to heavy metals has been found between short-term laboratory studies and long-term field experiments, and this tolerance also varies between sites (Giller et al., 2009). As reviewed by Giller et al. (2009), there is a range of sensitivity to heavy metals among various types of microorganisms, and a threshold for sensitivity likely exists. In the present study, a significantly lower fungi abundance was observed across the PSFs under pollution compared with that of bacteria, with the exception of the fine sand fraction, suggesting that heavy metal pollution had more long-term effects on the fungal community. However, a few studies have found contrasting results, and positive effects of heavy metal addition on the fungal community have even been observed in short-term laboratory experiments (Frostegård et al., 1993; Rajapaksha et al., 2004). One explanation for these divergent results could be that compared with fungi, bacteria can likely be well adapted to chronic toxicity or stress in long-term-contaminated soils due to their wide substrate utilisation profile and high metabolic activity. Additionally, bacteria dominate in rice paddies and may compete with fungi for substrates, resulting in greater stress on the fungal community. The reduced fungal community diversity shown by the DGGE profiles in the present study further substantiated the effects of heavy metals on the fungi under long-term heavy metal stress. Decreases in the levels of fungal PLFAs and a lower tolerance of the fungal community to heavy metal pollution along two different gradients in Scandinavian coniferous forest soils were previously described by Pennanen et al. (1996), who attributed these changes to decreases in the numbers of ectomycorrhizal fungi under pollution. In addition, Liu et al. (2012) also reported a reduction in the culturable fungal population as well as fungal PLFAs and gene copy numbers in four rice paddies under long-term heavy metal pollution compared with background soils. Decreases in fungal population PLFAs were also observed by Kandeler et al. (2000), who found that fungal PLFAs contents were reduced in the coarse and fine sand fractions following the addition of heavy metals (Zn, Cu and Cd) to a Calcaric Phaeozems under field conditions after 10 years of exposure. In this study, a significantly negative correlation was observed between extractable Zn and the fungal abundance (Table 5), which may indicate that bioavailable Zn might have an effect on the fungal population. Our results, derived from c. 40 years of heavy metal pollution, further supported the notion that the microbial responses produced under long-term field conditions (chronic toxicity or stress) differed from those in short-term laboratory assays (acute toxicity or disturbance) (Giller et al., 2009).

As bacteria and fungi are two principal groups in rice paddy soils, a decrease in fungal abundance could contribute to the lower MBC contents observed under pollution compared with nonpolluted areas. Archaea may be as abundant as fungi in terms of their rRNA genes. However, due to the low contribution of this group to basal respiration under aerobic incubation, the archaeal community was not addressed in the present study. In addition, a significantly positive correlation between fungal abundance and basal respiration (= 0.513, < 0.01) was noted across the PSFs, which indicated that the lower fungal abundance detected under pollution could be responsible for the lower soil respiration compared with nonpolluted conditions, as C respired from fungi constituted a higher proportion of total CO2 (Six et al., 2006).

Bacterial and fungal community structures in PSFs under heavy metal pollution

In the literature, coarse sand fractions are reported to exhibit lower bacterial species richness and diversity (Kandeler et al., 2000; Poll et al., 2003; Sessitsch et al., 2001) or higher diversity (Marhan et al., 2007) or few differences (Jackson & Weeks, 2008) regarding the bacterial composition in PSFs. In the present study, a greater number of specific DGGE bands and a higher H′ diversity index (Fig. 3) were observed in the coarse sand fraction than in the others fractions, regardless of the existence of heavy metal pollution. One possible explanation for this finding is that the coarse sand fractions most likely provide multiple microhabitats and more labile SOC for bacterial groups. Alternatively, microaggregates may be formed within macroaggregates and then released upon breakdown of the macroaggregates (Six et al., 2000). The bacterial community could be driven more by shifts in the chemical composition than in the quantity of soil organic matter, as reported by Davinic et al. (2012), who found higher levels of diversity in macroaggregates than in the microaggregates via pyrosequencing. The majority of the band types observed in the present study (B1-B3, B11 and B13-B17), particularly the dominant bands, were shared by all of the fractions, indicating that the bacterial community structure was highly stable. Sequencing of these bands revealed that Acidobacteria, Betaproteobacteria, Gammaproteobacteria and Chloroflexi were the predominant groups (Table 6), and these genera are commonly found in rice paddies, as reported in previous studies by our group (Hussain et al., 2011; Chen et al., 2013). These findings indicate that these bacteria are widely distributed and survive well across all fractions and may play an important role in the turnover and stability of SOC. The cultured representatives of Chloroflexi, such as the Anaerolineae lineage, are a group of bacteria consisting of filamentous, slow-growing, anaerobic heterotrophs that decompose carbohydrates and amino acids (Yamada & Sekiguchi, 2009) and are found at high abundance in anoxic soils. These species are assumed to potentially act as degraders of relatively recalcitrant carbon compounds such as phenol (Fang et al., 2006) and 4-methylbenzoate (Wu et al., 2001), which may further contribute to their dominance in the silt and clay fractions, where there are lower labile SOC and oxygen concentrations. The additional bands (B4, B6, B9, B12 and B18) that were relatively enhanced in the coarse sand fraction were assigned to Acidobacteria and TM7, in agreement with the findings of Mummey & Stahl (2004), who observed that Acidobacteria predominated in macroaggregates and the outer fractions of microaggregates. Although a predominance of Acidobacteria has been associated with low-C soils (Fierer et al., 2007) and is assumed to be of ecological importance in rice paddies (Kielak et al., 2008), our understanding of their function is still limited due to the scarcity of cultivated species. Several species (i.e. B1-B4, belonging to Chlamydiae and Acidobacteria) were clearly inhibited under heavy metal pollution, whereas other species (B7 and B10, belonging to Chlamydiae and TM7) were enhanced, resulting in an unchanged bacterial diversity. This last result is in accord with the findings of Zhou et al. (2002), who reported that chromium contamination in high-organic-matter soils did not greatly reduce diversity due to the heterogeneity of microorganisms both spatially and in terms of carbon resources.

Contrary to the variation in bacterial diversity observed under pollution, the fungal community structure exhibited a different DGGE pattern, with a decreased diversity index being observed across the PSFs compared with the nonpolluted soils. The smaller number of fungal DGGE bands and decreased diversity provided more evidence of decreased fungal abundance under pollution. PCA of fungal 18S rRNA gene DGGE profiles revealed clear separation of the polluted PSFs compared with the nonpolluted fractions. Additionally, the significantly negative correlations detected between all of the heavy metals and the H′ fungal diversity index further supported the inhibitory effect of heavy metals on the fungal community, which is primarily responsible for the decomposition of organic matter in the coarse and fine sand fractions (Marhan et al., 2007). Ascomycota and Basidiomycota are important fungal groups in most soils (Carlile et al., 2001). Hussain et al. (2011) also reported that Ascomycota dominated the fungal community in the rhizosphere of rice plants. In the present study, Ascomycota and Basidiomycota were found to be predominant across all of the fractions, which suggested a ubiquity of these species and an important role in agroecosystems. Caesar-Tonthat (2002) noted that polysaccharides produced by Basidiomycetes are involved in soil aggregation and suggested that fungal-derived material plays an important role in the stability of the soil structure. Glomeromycota, which are known to form hyphae and improve aggregate stability, were ubiquitous among the fractions, although present in very small numbers in the clay fractions. Additionally, although the method described by Stemmer et al. (1998) is widely adopted for the separation of PSFs, it may not be completely appropriate for determination of the microbial community, especially for fungi. The combination of low-energy sonication, wet sieving and repeated centrifugation procedures might have caused the destruction of community assemblages and even cell structure. It may also not be completely suitable for the efficient collection of fungi, which exhibit a diameter of 3–8 μm (Killham, 1994), associated with particle sizes below 2 μm. Furthermore, despite its importance in studies of microbial ecology, the PCR-DGGE technique has some drawbacks that must be taken into account. A single DGGE band, despite being obtained after careful consideration, might not represent a single phylotype or species. Thus, new methods and techniques are required to better explore the relationships between changes in soil organic dynamics and the microbiological status.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. Authors’ contribution
  10. References

This study made progress towards providing insight into the distribution and activity of the microbial community in the soil matrix under long-term heavy metal pollution. Our results suggest that a high heavy metals content decreases microbial biomass and activity via inhibiting microbial community diversity, particularly that of fungal groups (i.e. Ascomycota and Chytridiomycota) in the large-size fractions, which mainly depends on heterogeneous SOC availability across the PSFs. Due to the limitations of PSF separation procedures and the DGGE technique, further monitoring using new fingerprinting methods (i.e. pyrosequencing) might be necessary to evaluate the changes in the soil microbial population and its relationships with SOC dynamics. Furthermore, caution should be taken in interpreting our results, as the experimental design was not optimal for making statistical inferences.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. Authors’ contribution
  10. References

This work was supported by grants from the China National Natural Science Foundation (numbers 41001365 and 41101269).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. Authors’ contribution
  10. References
  • Angers DA, Recous S & Aita C (1997) Fate of carbon and nitrogen in water-stable aggregates during decomposition of 13C15N-labelled wheat straw in situ. Eur J Soil Sci 48: 295-300.
  • Bååth E, Díaz-Raviña M, Frostegård Å & Campbell CD (1998) Effect of metal-rich sludge amendments on the soil microbial community. Appl Environ Microbiol 64: 238245.
  • Bronick CJ & Lal R (2005) Soil structure and management: a review. Geoderma 124: 322.
  • Brookes PC (1995) The use of microbial parameters in monitoring soil pollution by heavy metals. Biol Fert Soils 19: 269275.
  • Caesar-Tonthat TC (2002) Soil binding properties of mucilage produced by a basidiomycete fungus in a model system. Mycol Res 106: 930937.
  • Carlile MJ, Watkinson SC & Gooday GW (2001) The Fungi. Academic Press, San Diego, CA.
  • Chen J-S & Chiu C-Y (2003) Characterization of soil organic matter in different particle-size fractions in humid subalpine soils by CP/MAS 13C NMR. Geoderma 117: 129141.
  • Chen J, Liu X, Zheng J et al. (2013) Biochar soil amendment increased bacterial but decreased fungal gene abundance with shifts in community structure in a slightly acid rice paddy from Southwest China. Appl Soil Ecol 71: 3344.
  • Chiu CY, Chen TH, Imberger K & Tian G (2006) Particle size fractionation of fungal and bacterial biomass in subalpine grassland and forest soils. Geoderma 130: 265271.
  • Cui L, Li L, Zhang A, Pan G, Bao D & Chang A (2011) Biochar amendment greatly reduces rice Cd uptake in a contaminated paddy soil: a two-year field experiment. BioResources 6: 26052618.
  • Davinic M, Fultz LM, Acosta-Martinez V, Calderón FJ, Cox SB, Dowd SE, Allen VG, Zak JC & Moore-Kucera J (2012) Pyrosequencing and mid-infrared spectroscopy reveal distinct aggregate stratification of soil bacterial communities and organic matter composition. Soil Biol Biochem 46: 6372.
  • Denef K, Six J, Bossuyt H, Frey SD, Elliott ET, Merckx R & Paustian K (2001) Influence of dry-wet cycles on the interrelationship between aggregate, particulate organic matter, and microbial community dynamics. Soil Biol Biochem 33: 15991611.
  • Ding AF, Pan GX & Li LQ (2006) Distribution characters and the environmental significance of PAHs in particle size fractions of selected paddy soils from the Tai Lake region, China. Acta Scientiae Circumstantiae 26: 293299 (in Chinese with English summary).
  • Elliott ET (1986) Aggregate structure and carbon, nitrogen, and phosphorus in native and cultivated soils. Soil Sci Soc Am J 50: 627633.
  • Fang H, Liang D, Zhang T & Liu Y (2006) Anaerobic treatment of phenol in wastewater under thermophilic condition. Water Res 40: 427434.
  • Farkas A, Erratico C & Viganò L (2007) Assessment of the environmental significance of heavy metal pollution in surficial sediments of the River Po. Chemosphere 68: 761768.
  • Fierer N, Jackson JA, Vilgalys R & Jackson RB (2005) Assessment of soil microbial community structure by use of taxon-specific quantitative PCR assays. Appl Environ Microbiol 71: 41174120.
  • Fierer N, Bradford MA & Jackson RB (2007) Toward an ecological classification of soil bacteria. Ecology 88: 13541364.
  • Frey S, Six J & Elliott E (2003) Reciprocal transfer of carbon and nitrogen by decomposer fungi at the soil-litter interface. Soil Biol Biochem 35: 10011004.
  • Frostegård Å, Tunlid A & Bååth E (1993) Phospholipid fatty acid composition, biomass, and activity of microbial communities from two soil types experimentally exposed to different heavy metals. Appl Environ Microbiol 59: 36053617.
  • Giller KE, Witter E & McGrath SP (1998) Toxicity of heavy metals to microorganisms and microbial processes in agricultural soils: a review. Soil Biol Biochem 30: 13891414.
  • Giller KE, Witter E & McGrath SP (2009) Heavy metals and soil microbes. Soil Biol Biochem 41: 20312037.
  • Gong ZT (1999) Chinese Soil Taxonomic Classification. pp. 5215. China Science Press, Beijing (in Chinese).
  • Grundmann GL (2004) Spatial scales of soil bacterial diversity – the size of a clone. FEMS Microbiol Ecol 48: 119127.
  • Gude A, Kandeler E & Gleixner G (2012) Input related microbial carbon dynamic of soil organic matter in particle size fractions. Soil Biol Biochem 47: 209219.
  • Haynes R & Beare M (1997) Influence of six crop species on aggregate stability and some labile organic matter fractions. Soil Biol Biochem 29: 16471653.
  • Hedrick DB, Peacock A, Stephen JR, Macnaughton SJ, Brüggemann J & White DC (2000) Measuring soil microbial community diversity using polar lipid fatty acid and denaturing gradient gel electrophoresis data. J Microbiol Methods 41: 235248.
  • Heuer H, Krsek M, Baker P, Smalla K & Wellington E (1997) Analysis of actinomycete communities by specific amplification of genes encoding 16S rRNA and gel-electrophoretic separation in denaturing gradients. Appl Environ Microbiol 63: 32333241.
  • Huber T, Faulkner G & Hugenholtz P (2004) Bellerophon: a program to detect chimeric sequences in multiple sequence alignments. Bioinformatics 20: 23172319.
  • Hussain Q, Liu Y, Zhang A, Pan G, Li L, Zhang X, Song X, Cui L & Jin Z (2011) Variation of bacterial and fungal community structures in the rhizosphere of hybrid and standard rice cultivars and linkage to CO2 flux. FEMS Microbiol Ecol 78: 116128.
  • Insam H (2001) Developments in soil microbiology since the mid 1960s. Geoderma 100: 389402.
  • IUSS Working Group WRB (2007) World Reference Base for Soil Resources 2006, first update 2007. World Soil Resources Reports No. 103. FAO, Rome.
  • Jackson CR & Weeks AQ (2008) Influence of particle size on bacterial community structure in aquatic sediments as revealed by 16S rRNA gene sequence analysis. Appl Environ Microbiol 74: 52375240.
  • Jiang X, Wright AL, Wang X & Liang F (2011) Tillage-induced changes in fungal and bacterial biomass associated with soil aggregates: a long-term field study in a subtropical rice soil in China. Appl Soil Ecol 48: 168173.
  • Kandeler E & Murer E (1993) Aggregate stability and soil microbial processes in a soil with different cultivation. Geoderma 56: 503513.
  • Kandeler E, Tscherko D, Bruce KD, Stemmer M, Hobbs PJ, Bardgett RD & Amelung W (2000) Structure and function of the soil microbial community in microhabitats of a heavy metal polluted soil. Biol Fert Soils 32: 390400.
  • Kielak A, Pijl AS, van Veen JA & Kowalchuk GA (2008) Phylogenetic diversity of Acidobacteria in a former agricultural soil. ISME J 3: 378382.
  • Killham K (1994) Soil Ecology, pp. 4052. Cambridge University Press, Cambridge, UK.
  • Ladd JN, Foster RC, Nannipieri P & Oades J (1996) Soil structure and biological activity. Soil Biochem 9: 2378.
  • Lagomarsino A, Grego S & Kandeler E (2012) Soil organic carbon distribution drives microbial activity and functional diversity in particle and aggregate-size fractions. Pedobiologia 55: 101110.
  • Li YT, Rouland C, Benedetti M, Li F, Pando A, Lavelle P & Dai J (2009) Microbial biomass, enzyme and mineralization activity in relation to soil organic C, N and P turnover influenced by acid metal stress. Soil Biol Biochem 41: 969977.
  • Liu H, Li Y, Li L, Jin L & Pan G (2006) Pollution and risk evaluation of heavy metals in soil and agro-products from an area in the Tai Lake region. J Safety Environ 6: 6163 (in Chinese with English summary).
  • Liu Y, Zhou T, Crowley D et al. (2012) Decline in topsoil microbial quotient, fungal abundance and C utilization efficiency of rice paddies under heavy metal pollution across South China. PLoS ONE 7: e38858.
  • Lu RK (2000) Methods of Soil and Agro-Chemical Analysis. China Agricultural Science and Technology Press, Beijing (in Chinese).
  • Marhan S, Kandeler E & Scheu S (2007) Phospholipid fatty acid profiles and xylanase activity in particle size fractions of forest soil and casts of Lumbricus terrestris L. (Oligochaeta, Lumbricidae). Appl Soil Ecol 35: 412422.
  • May LA, Smiley B & Schmidt MG (2001) Comparative denaturing gradient gel electrophoresis analysis of fungal communities associated with whole plant corn silage. Can J Microbiol 47: 829841.
  • Mench M, Vangronsveld J, Clijsters H, Lepp NW & Edwards R (2000) In situ metal immobilization and phytostabilization of contaminated soils. Phytoremediation of Contaminated Soil and Water (Terry N, & Bañuelos G, eds), pp. 323358. Lewis Publication, Boca Raton, FL.
  • Mummey DL & Stahl PD (2004) Analysis of soil whole- and inner-microaggregate bacterial communities. Microb Ecol 48: 4150.
  • Muyzer G, De Waal E & Uitterlinden A (1993) Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl Environ Microbiol 59: 695.
  • Nemerow NL (1991) Stream, Lake, Estuary and Ocean Pollution. Van Nostrand Reinhold Publishing Co, New York, NY.
  • Oades J (1993) The role of biology in the formation, stabilization and degradation of soil structure. Geoderma 56: 377400.
  • Pan G, Li L, Wu L & Zhang X (2003) Storage and sequestration potential of topsoil organic carbon in China's paddy soils. Glob Change Biol 10: 7992.
  • Paul EA, Collins HP & Leavitt SW (2001) Dynamics of resistant soil carbon of Midwestern agricultural soils measured by naturally occurring 14C abundance. Geoderma 104: 239256.
  • Pennanen T, Frostegard A, Fritze H & Baath E (1996) Phospholipid fatty acid composition and heavy metal tolerance of soil microbial communities along two heavy metal-polluted gradients in coniferous forests. Appl Environ Microbiol 62: 420428.
  • Pérez-de-Mora A, Burgos P, Madejón E, Cabrera F, Jaeckel P & Schloter M (2006) Microbial community structure and function in a soil contaminated by heavy metals: effects of plant growth and different amendments. Soil Biol Biochem 38: 327341.
  • Poll C, Thiede A, Wermbter N, Sessitsch A & Kandeler E (2003) Micro-scale distribution of microorganisms and microbial enzyme activities in a soil with long-term organic amendment. Eur J Soil Sci 54: 715724.
  • Rajapaksha RMCP, Tobor-Kapłon MA & Bååth E (2004) Metal toxicity affects fungal and bacterial activities in soil differently. Appl Environ Microbiol 5: 29662973.
  • Roy S & Singh J (1994) Consequences of habitat heterogeneity for availability of nutrients in a dry tropical forest. J Ecol 82: 503509.
  • Rutherford P & Juma N (1992) Influence of soil texture on protozoa-induced mineralization of bacterial carbon and nitrogen. Can J Soil Sci 72: 183200.
  • Sekiguchi H, Watanabe M, Nakahara T, Xu B & Uchiyama H (2002) Succession of bacterial community structure along the Changjiang River determined by denaturing gradient gel electrophoresis and clone library analysis. Appl Environ Microbiol 68: 51425150.
  • Serra-Wittling C, Houot S & Barriuso E (1995) Soil enzymatic response to addition of municipal solid-waste compost. Biol Fert Soils 20: 226236.
  • Sessitsch A, Weilharter A, Gerzabek MH, Kirchmann H & Kandeler E (2001) Microbial population structures in soil particle size fractions of a long-term fertilizer field experiment. Appl Environ Microbiol 67: 42154224.
  • Simpson R, Frey S, Six J & Thiet R (2004) Preferential accumulation of microbial carbon in aggregate structures of no-tillage soils. Soil Sci Soc Am J 68: 12491255.
  • Singh S & Singh JS (1995) Microbial biomass associated with water-stable aggregates in forest, savanna and cropland soils of a seasonally dry tropical region, India. Soil Biol Biochem 27: 10271033.
  • Six J, Elliott ET & Paustian K (2000) Soil macroaggregate turnover and microaggregate formation: a mechanism for C sequestration under no-tillage agriculture. Soil Biol Biochem 32: 20992103.
  • Six J, Bossuyt H, Degryze S & Denef K (2004) A history of research on the link between (micro)aggregates, soil biota, and soil organic matter dynamics. Soil Till Res 79: 731.
  • Six J, Frey SD, Thiet RK & Batten KM (2006) Bacterial and fungal contributions to carbon sequestration in agroecosystems. Soil Sci Soc Am J 70: 555569.
  • Spaccini R, Zena A, Igwe C, Mbagwu J & Piccolo A (2001) Carbohydrates in water-stable aggregates and particle size fractions of forested and cultivated soils in two contrasting tropical ecosystems. Biogeochemistry 53: 122.
  • Stemmer M, Gerzabek MH & Kandeler E (1998) Organic matter and enzyme activity in particle-size fractions of soils obtained after low-energy sonication. Soil Biol Biochem 30: 917.
  • Tan T, Wang C, Li B, He X & Zhang L (2005) Pollution and evaluation of Pb in soil in Chengdu Plain. Res Environ Yangtze Basin 14: 7175 (in Chinese with English summary).
  • Vance E, Brookes P & Jenkinson D (1987) An extraction method for measuring soil microbial biomass C. Soil Biol Biochem 19: 703707.
  • Vásquez-Murrieta M, Migueles-Garduño I, Franco-Hernández O, Govaerts B & Dendooven L (2006) C and N mineralization and microbial biomass in heavy-metal contaminated soil. Eur J Soil Biol 42: 8998.
  • Weber O, Scholz RW, Bühlmann R & Grasmück D (2001) Risk perception of heavy metal soil contamination and attitudes toward decontamination strategies. Risk Anal 21: 967977.
  • Wu JH, Liu WT, Tseng IC & Cheng SS (2001) Characterization of microbial consortia in a terephthalate-degrading anaerobic granular sludge system. Microbiology 147: 373382.
  • Yamada T & Sekiguchi Y (2009) Cultivation of uncultured chloroflexi subphyla: significance and ecophysiology of formerly uncultured chloroflexi ‘Subphylum I’ with natural and biotechnological relevance. Microbes Environ 24: 205216.
  • Zhang P, Zheng J, Pan G, Zhang X, Li L & Rolf T (2007) Changes in microbial community structure and function within particle size fractions of a paddy soil under different long-term fertilization treatments from the Tai Lake region, China. Colloids Surf B Biointerfaces 58: 264270.
  • Zheng J, Zhang X, Li L, Zhang P & Pan G (2007) Effect of long-term fertilization on C mineralization and production of CH4 and CO2 under anaerobic incubation from bulk samples and particle size fractions of a typical paddy soil. Agr Ecosyst Environ 120: 129138.
  • Zhou J, Xia B, Treves DS, Wu L-Y, Marsh TL, O'Neill RV, Palumbo AV & Tiedje JM (2002) Spatial and resource factors influencing high microbial diversity in soil. Appl Environ Microbiol 68: 326334.