1 Introduction

Nitrification is ubiquitous in the environment and plays a central role in the global nitrogen (N) cycle (Francis et al. 2005; Leininger et al. 2006; Prosser and Embley 2002). Ammonia oxidation, the first step in nitrification, is the reaction of NH +4 to NH2OH catalyzed by ammonia monooxygenase encoded by the amoA gene of ammonia-oxidizing bacteria (AOB) (Kowalchuk et al. 1999; Kowalchuk and Stephen 2001; Rotthauwe et al. 1997). Recently, studies have shown that ammonia-oxidizing archaea (AOA) are also involved in nitrification and often show higher abundance than AOB (Abell et al. 2010; Leininger et al. 2006). AOA may play an important role in N cycling on Earth (Könneke et al. 2005; Zhang et al. 2010). The amoA gene is commonly used as a molecular marker of both AOA and AOB for abundance and community diversity studies (Abell et al. 2010; Francis et al. 2007).

Nitrification is of particular significance in estuarine sediments because coupled nitrification/denitrification can remove a substantial percentage (10% to 80%) of anthropogenic N input into estuaries (Beman and Francis 2006; Seitzinger 1988). Studies have shown that the AOA and AOB communities are sensitive to many environmental factors, such as pH, carbon, nitrogen, sulfur, oxygen, temperature, and sediment depth. Vegetation is also an important factor controlling N removal in wetlands directly and indirectly (Ruiz-Rueda et al. 2009) and close interactions are observed between plants and belowground microbial communities (Burke et al. 2002; Kowalchuk et al. 2002).

Exotic plants threaten the integrity of agricultural and natural systems throughout the world (Callaway and Aschehoug 2000). Invasive plants can change the interaction between soil and plants (Callaway and Ridenour 2004; Klironomos 2002; Kourtev et al. 2002) and affect nutrient cycling through differences in rates of rhizodeposition and litter production (Dollhopf et al. 2005; Ehrenfeld 2003; Van der Krift et al. 2001). Soil microbes are considered to be one of the drivers of successful exotic plant invasion (Callaway et al. 2004; Kourtev et al. 2002), playing a central role between nutrient cycling and plant invasion (Hawkes et al. 2005). Previous work has shown that exotic plant invasion could clearly impact nitrifying bacteria (Hawkes et al. 2005), denitrifiers, and total bacterial communities (Angeloni et al. 2006). Spartina alterniflora is a typical exotic invasive plant in the coastal wetland of China (Liao et al. 2008a; Liao et al. 2008b). S. alterniflora rapidly dominates ecosystems and forms a monoculture by excluding native plants, which subsequently change the carbon, nitrogen (Liao et al. 2008a; Liao et al. 2008b), and sulfur (Zhou et al. 2009) cycles in the invaded ecosystem. However, to our knowledge, it is not clear yet whether the presence of S. alterniflora would result in changes of abundance and composition of AOA and AOB communities. Specifically, no previous studies have addressed the effects of S. alterniflora invasion on the AOA community composition.

We hypothesize that the abundance and composition of AOA or AOB could be changed in S. alterniflora-invaded coastal wetlands. A typical S. alterniflora-invaded coastal wetland was selected in Jiulong River estuary in Fujian Province, China. The monoculture of S. alterniflora was formed after successful invasion in 2007. This study focused on characterization of the abundance and composition of AOB and AOA at the spread stage of S. alterniflora invasion, as well as analyzing the total bacterial community, sediment properties, and potential nitrification rate.

2 Materials and methods

2.1 Sediment sampling

Sampling sites were located in the Mangrove Nature Reserve of the Jiulong River estuary (24°26′N, 117°54′ E) in Fujian, China. In 2007, S. alterniflora invaded the native mangrove habitat in the reserve and formed monoculture in the low tidal wetland. Three different habitats were selected to sample, i.e., unvegetated bare mudflat zone (B), S. alterniflora-invaded zone (S), and native mangrove zone (M). Sediment sample cores were collected with a PVC pipe (5 cm diameter, 20 cm length) on June 26, 2009, with five replicates in each zone. The sediment cores were placed in an ice box and transported to the laboratory within 2 h. Each core was then sliced into two layers, i.e., a shallow oxic layer (0–5 cm, O) and a lower reductive layer (5–20 cm, R). Therefore, it became 30 samples in total. Each core was homogenized and divided into four portions. One portion was used to determine moisture content, pH, total carbon, total nitrogen, and total sulfur; one fresh portion was used for analysis of potential nitrification rate (PNR), and the last two portions were stored at −80°C for DNA extraction after being freeze-dried.

2.2 Sediment properties and PNR analyses

Sediment pH was determined in a 1:2.5 sediment/water mixture. Total nitrogen (TN), total carbon (TC), and total sulfur (TS) were measured using C/N/H/S analyzer (Elementar Analysensysteme, Vario Max, Germany). PNR was measured using the chlorate inhibition method with minor modifications as previously described (Xia et al. 2007). KClO3 was used to inhibit NO -2 oxidation to NO -3 . Briefly, 5 g of fresh sediment was added to 50-ml centrifuge tubes containing 20 ml 1 mM phosphate-buffered saline (NaCl, 8.0 g L−1; KCl, 0.2 g L−1; Na2HPO4, 0.2 g L−1; NaH2PO4, 0.2 g L−1; pH 7.1) containing 1 mM of (NH4)2SO4; KClO3 was added to the tubes with final concentration of 100 mg L−1 to inhibit nitrite oxidation. The slurries were incubated at room temperature in the dark for 24 h after a 10 min shaking at 160 rpm. Five milliliters of 2 M KCl was added to the incubated tubes for extracting NO -2 -N, which was then determined by a flow injection analyzer (Lachat-QC8500, Lachat Instruments, Loveland, CO) at 540 nm with N-(1-naphthyl) ethylenediamine dihydrocholride.

2.3 DNA extraction

DNA was extracted from 0.5 g freeze-dried sediment samples using the FastDNA Spin Kit for Soil (Qbiogene, Carlsbad, CA). DNA extracts were suspended in 70 μl of DES solution (Qbiogene, USA) and quantified by Nanodrop spectrophotometer. A test of replicate extractions was performed to ensure reproducible yields from DNA extractions (data not shown).

2.4 PCR amplification for DGGE

For bacterial 16S rRNA gene polymerase chain reaction and denaturing gradient gel electrophoresis (PCR–DGGE) analysis, the 341fGC/534r primer was used (Muyzer et al. 1993; Table 1). PCR amplification was performed in a 50-μl reaction mixture, including 1 × PCR buffer, 3.0 mM MgCl2, 400 μM each dNTP, 2.5 U Taq DNA polymerase (Fermentas, Canada), 0.2 mg ml−1 bovine serum albumin (BSA) plus 0.2 mM of each primer, and 6 ng sediment genomic DNA as template. The thermal profile was conducted according to the previous method (Clark and Hirsch 2008) (see Table 1).

Table 1 Primers, oligonucleotide probes, qPCR, and PCR–DGGE conditions in this study
Full size table

For amoA gene denaturing gradient gel electrophoresis (DGGE), amoA1F-GC/amoA2R (Rotthauwe et al. 1997) and arch-amoAF-GC/arch-amoAR (Francis et al. 2005) were primers for bacterial and archaeal amoA gene amplification, respectively (see Table 1). Amplifications were performed for AOA and AOB in 50-μl reaction mixtures including 1 × PCR buffer, 3.0 mM MgCl2, 400 μM each dNTP, 2.5 U Taq DNA polymerase (Fermentas, Canada), and 0.2 mg ml−1 BSA plus 0.2 mM of each primer; 120 ng sediment genomic DNA was added as template.

2.5 DGGE analysis

DGGE analysis of bacterial 16S rRNA gene fragments and amoA PCR products were performed with Dcode Universal Mutation Detection System (Bio-Rad Laboratories, Hercules, USA). Eight microliters and 15 μl PCR products of 16S rRNA and both amoA genes were loaded into polyacrylamide gel wells, respectively. Denaturing gradient and electrophoresis conditions are given in Table 1. The gels were stained with 1:10,000 SYBR-Green I (Sigma, USA) for 30 min and documented using a GelDoc XR (BIO-RAD, USA). Unweighted pair group method with arithmetic mean (UPGMA) algorithms were used to cluster the DGGE banding patterns of bacterial 16S rRNA gene using the software Quantity One (Bio-Rad Laboratories, Hercules, USA).

2.6 Cloning and sequencing of amoA gene

Selected DGGE bands were excised for cloning and sequencing. Excised bands were crushed and incubated overnight at 4°C in 30 μl of sterilized water to elute the DNA and then reamplified with the primers amoA1F/amoA2R and arch-amoAF/arch-amoAR. Reamplified PCR products were purified and ligated into the PMD™ 19-T Vector (TaKaRa Code: D102A, Japan). The resulting ligation mixture transformed into Escherichia coli DH5α competent cells following the instructions of the manufacturer. Positive clones were amplified using the above primers with GC clamp and cross-checked by DGGE for migration behavior.

2.7 Real-time PCR assay for bacterial 16S rRNA and amoA genes

Quantitative real-time polymerase chain reaction (qPCR) assay methods were slightly modified according to previous studies (Dang et al. 2010; Shen et al. 2008; Peng et al. 2010; see Table 1). Gene copy numbers were determined for all samples in triplicates using Applied Biosystems 7500 Real-Time PCR System (ABI, USA). Total bacteria population size was quantified using primers BACT1369F/PROK1492R and oligonucleotide probe TM1389F (see Table 1, Suzuki et al. 2000; Peng et al. 2010). Each reaction was performed in a 25-μl volume containing 1 μl of DNA extract (containing 6 ng sediment genomic DNA) as template, 0.2 mg ml−1 BSA, 0.4 mM of each primer and probe (Takara, Japan), and 12.5 μl Premix Ex taq™ (Takara, Japan). The quantification of amoA gene was based on the fluorescent dye SYBR-Green I. A total of 40 cycles were run with annealing temperatures using primers listed in Table 1. Each reaction was performed in a 25-μl volume containing 1 μl of DNA extract as template (containing 60 ng of sediment genomic DNA), 0.2 mg ml−1 BSA, 0.2 mM of each primer, and 12.5 μl of FastStart Universal SYBR Green Master (Rox; Roche, Switzerland). Product specificity was confirmed by melting curve analysis and visualization in 1.2% agarose gels. Specific product was seen at the expected size for the bacterial and archaeal amoA genes ca. 491 and 635 bp, respectively.

Standard curves for qPCR were developed as previously described (Dang et al. 2010; He et al. 2007) Bacterial 16S rRNA gene fragments were obtained from extracted DNA with primers 27F/1492R (Lane 1991). Archaeal and bacterial amoA and bacterial 16S rRNA gene fragments were cloned as described in section 2.6, and clones that had the right gene inserts were chosen as standards for real-time PCR. Plasmid DNA was extracted with Plasmid Kit (TaKaRa, Japan), and the plasmid concentration was measured with a spectrophotometer (Nanodrop). As the sequences of the vector and PCR inserts were known, the copy numbers of amoA were calculated directly from the concentration of extracted plasmid DNA (Okano et al. 2004). Tenfold serial dilutions of a known copy number of plasmid of the gene clone from sediments were generated to produce the standard curve over seven orders of magnitude (4.97 × 103 to 4.97 × 109 copies of template for archaeal amoA, 4.83 × 103 to 4.83 × 109 copies of template for bacterial amoA, and 1.17 × 104 to 1.17 × 1010 copies of template for bacterial 16S rRNA gene) per assay, respectively. High amplification efficiencies of 96.8–115.4% were obtained for archaeal and bacterial amoA quantification with R 2 values 0.99 and slopes from −3.0 to −3.4. No template DNA negative control was present in each qPCR assay procedure. The inhibitor of DNA extract was tested by the method of previous study (Dumonceaux et al. 2006); no significant inhibition at any dilution (1–120 ng μl−1 DNA extract) was found. Data analysis was carried out with the 7500 System software (ABI, USA).

2.8 Phylogenetic analysis

Phylogenetic analysis was carried out using methods previously described (Chen et al. 2008; Shen et al. 2008). Sequences were compared with GenBank database sequences using BLAST (http://www.ncbi.nlm.nih.gov/BLAST/), and the closest matches were included in the alignment. The neighbor-joining trees were constructed using MEGA 4 [Molecular Evolutionary Genetics Analysis (http://megasoftware.net)] and bootstrapped 1,000 times to calculate linear distances. The Shannon diversity index H and Evenness index E were calculated for the band pattern diversity of AOB and AOA based on the following equations:

$$ \begin{array}{*{20}{c}} {H = - \sum\limits_{{i = 1}}^s {Pi\ln Pi} = - \sum\limits_{{i = 1}}^S {\left( {Ni/N} \right)\ln \left( {Ni/N} \right)} } \\ {E = H/{ \ln }S} \\ \end{array} $$

where Ni is the abundance of the ith ribotype, N is the total abundance of all ribotypes in the sample (lane of DGGE gels), and S is the number of ribotypes.

The definition of the amoA sequence clusters was based on the previous study (Hunter et al. 2006).

2.9 Data statistical analysis

All statistical analyses were performed using SPSS version 16.0 and one-way analysis of variance followed by Duncan test which was used to check for quantitative differences between treatments. P < 0.05 was considered to be statistically significant.

2.10 Sequence accession numbers

All amoA gene sequences from DGGE bands have been deposited in the GenBank nucleotide sequence database under accession numbers HM235899 to HM235913 for AOB and HM235889 to HM235898 for AOA.

3 Results

3.1 Sediment properties

Sediment characteristics were altered by S. alterniflora invasion. Sediment pH ranged from 6.12 to 6.99 with some variations among the different sediment samples (Table 2). The highest pH was found in the 5–20 cm sample of the unvegetated zone (BR), followed by the S. alterniflora-invaded and unvegetated zones in the shallow layer. Sediment pH in the S. alterniflora-invaded zone was significantly higher than in the native mangrove zone (P < 0.05).

Table 2 Sediment properties and potential nitrification rates (PNR) in the studied samples
Full size table

TC and TN in the unvegetated and S. alterniflora zones were significantly lower than those in native mangrove zones. The lowest ratio of TC to TN was found in S. alterniflora zones, and significant differences of the C/N ratio were found between the native mangrove and S. alterniflora-invaded zones. Conversely, the lowest TS was found in native mangrove zones. Significant differences in TS were also found between the native and S. alterniflora-invaded zones. In addition, positive correlations were observed between TC and TN (r = 0.92, n = 30, P < 0.01) and between TS and pH (r = 0.86, n = 18, P < 0.01), respectively.

PNR provides an independent estimate of the abundance of ammonia oxidizers (see Table 2). PNR was lower in the S. alterniflora-invaded zones than in the native mangrove zones at both depths.

3.2 Bacterial 16S rRNA gene abundance in sediments

Bacterial 16S rRNA gene abundance was affected by S. alterniflora invasion (Table 3). The real-time PCR assays showed that bacterial 16S rRNA gene copy numbers per gram of dry sediment ranged from 3.49 × 1011 to 1.46 × 1012. In the mangrove zone, 16S rRNA gene was significantly higher than in both invaded and unvegetated zones at two layers. Bacterial abundance positively related to TC (r = 0.84, n = 18, P < 0.01) and TN (r = 0.74, n = 18, P < 0.01), respectively, but negatively related with pH (r = −0.63, n = 18, P < 0.01) and TS (r = −0.57, n = 18, P < 0.01), respectively. Bacterial 16S rRNA gene abundance positively correlated with AOA abundance (r = 0.79, n = 18, P < 0.01), but no evident correlation was found between bacterial 16S rRNA gene and AOB abundance.

Table 3 Copy number of archaeal amoA, bacterial amoA, and bacterial 16S rRNA genes in sediment with different vegetations
Full size table

3.3 amoA gene abundance in sediments

Average abundance of AOA genes in the S. alterniflora-invaded zones were 3.23 × 108 (O) and 4.01 × 107 (R) copy numbers per gram dry sediment and were reduced by 11.3-fold (O) and 46.1-fold (R) in the S. alterniflora-invaded zones, compared with native mangrove zone (3.72 × 109 (O) and 1.85 × 109 (R) copy numbers per gram dry sediment; see Table 3). However, AOB abundance in S. alterniflora zone was 1.61 × 108 (O) and 5.14 × 106 (R) copy numbers per gram dry sediment, higher than in the native mangrove zone (9.8-fold (O) and 1.8-fold (R), respectively). AOB abundance in S. alterniflora zone was significantly higher than in the native and unvegetated zones in the 0–5-cm layer, while no evident difference was observed among zones in the 5–20-cm layer.

Similar to total bacteria, positive correlations were obtained between AOA abundance, TC (r = 0.82, n = 18, P < 0.01), and TN (r = 0.77, n = 18, P < 0.01), respectively, but negative correlations were obtained with pH (r = −0.83, n = 18, P < 0.01) and TS (r = −0.83, n = 18, P < 0.01), respectively. No significant correlation was found between PNR and AOA or AOB abundance.

AOA was more abundant than AOB in all sediments (see Table 3). The ratios especially AOA/AOB in mangrove zones were 227 (O) and 639 (R), much higher than the others. The ratio of AOA/bacterial 16S rRNA gene in native zone (MO, 0.254%; MR, 0.284%) was also higher than those in the invaded (SO, 0.0561%; SR, 0.00968%) and unvegetated zones. However, the ratio AOB/bacterial 16S rRNA gene in native zone (MO, 0.00112%; MR, 0.000445%) was lower than that of invaded zones (SO, 0.0279%; SR, 0.00124%).

3.4 Genetic profiling of bacterial 16S rRNA and amoA genes

Bacterial 16S rRNA gene, AOB, and AOA composition were analyzed by PCR–DGGE (Figs. 1a, 2, and 3). Reproducibility of triplicates DGGE profiles from sediment DNA extracts was found to be high (data not shown). Duplicate results are shown in this study. For bacterial 16S rRNA gene, profiles from the native mangrove zones clustered separately from the unvegetated or S. alterniflora-invaded zones, and it exhibited 64% similarity between mangrove zone and others (see Fig. 1b).

Fig. 1
figure 1

a Denaturing gradient gel electrophoresis profiles of bacterial 16S rRNA gene in sediments with different overlying vegetation. Duplicate sample results using 1 and 2 were shown in each treatment. b Cluster analysis of the bacterial DGGE profiles based on 16S rRNA gene. The dendrogram was calculated on the basis of the UPGMA. The analysis was based on the presence or absence of bands at positions in each lane of the gel. Duplicate sample results using 1 and 2 were shown in each treatment

Full size image
Fig. 2
figure 2

Denaturing gradient gel electrophoresis profiles of bacterial amoA gene in sediments of different vegetations. Bands used for sequencing and phylogenetic analysis are highlighted. Duplicate sample results using 1 and 2 were shown in each treatment

Full size image
Fig. 3
figure 3

Denaturing gradient gel electrophoresis profiles of archaeal amoA gene in sediments of different vegetations. Bands used for sequencing and phylogenetic analysis are highlighted. Duplicate sample results using 1 and 2 were shown in each treatment

Full size image
Fig. 4
figure 4

Phylogenetic relationship among bacterial amoA gene sequences retrieved from the Jiulong River estuary sediments. Designation of the clones in bold includes the following information: accession number in the GenBank with DGGE band label. Bootstrap values (≥50%) for 1,000 replicates are shown at the branch points. The scale bar represents 5% estimated sequence divergence

Full size image

Shannon index and Evenness were used to analyze microbial community diversity (Table 4). The Shannon index of bacterial 16S rRNA gene in mangrove zone was higher than that in other zones. Shannon diversity was significantly negatively correlated with pH (r = −0.91, n = 18, P < 0.01) and TS (r = −0.75, n = 18, P < 0.01) but was positively correlated with TC (r = 0.81, n = 18, P < 0.01) and TN (r = 0.69, n = 18, P < 0.01). Conversely, diversity of AOA in the S. alterniflora-invaded and unvegetated zones was higher than in the native mangrove zone. AOA abundance negatively correlated with AOA Shannon index (r = −0.59, n = 18, P < 0.01) and Evenness (r = −0.47, n = 18, P < 0.05), respectively. AOA Shannon index also negatively correlated with TC (r = −0.56, n = 18, P < 0.05) and TN (r = −0.61, n = 18, P < 0.01), but positively correlated with pH (r = 0.85, n = 18, P < 0.05) and TS (r = 0.81, n = 18, P < 0.01). For AOB, the diversity in upper layer (0–5 cm) was higher than those of deeper layer (5–20 cm). However, no difference in AOB diversity was found among different zones.

Table 4 Diversity properties of AOB, AOA and 16S rRNA gene calculated from DGGE band pattern data under different sediment samples
Full size table

Fifteen bands of AOB were sequenced and used to construct a phylogenetic tree (Fig. 4). AOB sequences were grouped into the environmental cluster, distinct from previously identified Nitrosomonas and Nitrosospira amoA clades. Ten AOA bands were sequenced and grouped into cluster S (soil/sediment) of Crenarchaeota (Fig. 5).

Fig. 5
figure 5

Phylogenetic relationships among archaeal amoA gene sequences retrieved from the Jiulong River estuary sediments. Designation of the clones in bold includes the following information: accession number in the GenBank with DGGE band label. Bootstrap values (≥50%) for 1,000 replicates are shown at the branch points. The scale bar represents 10% estimated sequence divergence

Full size image

4 Discussion

This study shows that exotic S. alterniflora invasion affects the abundance and composition of AOA and AOB and provides some evidence on the linkage between exotic plants invasion and ecosystem function in Jiulong River estuarine sediment. In this study, changes in sediment properties and potential nitrification rate were explained in part by changes to AOA and AOB communities caused by S. alterniflora establishment and spread, and these alterations may directly affect nitrification in estuarine sediment.

Our experiments showed that pH was increased during the invasion and had a negative impact on AOA abundance, which was consistent with a previous study in acid soil (Nicol et al. 2008). Also, total sulfur concentration was found to be higher in S. alterniflora-invaded zones than in those with native plant and unvegetated zones which is consistent with previous studies (Stribling 1997; Zhou et al. 2009). AOA abundance increased with decreasing concentration of sulfur; this may due to AOA abundance being negatively correlated to sulfur concentration as demonstrated by other researchers (Caffrey et al. 2007; Dollhopf et al. 2005). Moreover, AOA abundance was found to be positively correlated with total carbon and total nitrogen which is in agreement with previous studies (Abell et al. 2010; Mosier and Francis 2008). PNR has been shown to be related to AOA abundance (Caffrey et al. 2007; Beman and Francis 2006) or AOB abundance (Nicol et al. 2008; Shen et al. 2008; Ying et al. 2010). In this study, both PNR and AOA abundance significantly decreased in S. alterniflora zone, while AOB abundance increased. However, in contrast to previous studies, our results show that PNR was not strongly correlated to the amoA gene copy numbers of AOB or AOA; this might be due to the number of samples analyzed in this study. From our results, it appears that S. alterniflora invasion have an impact on the sediment properties. However, it was difficult to attribute these changes to S. alterniflora only, as ecosystems are complex systems, and further studies are required.

The amoA and bacterial 16S rRNA gene abundance in sediments revealed important patterns (see Table 3). Firstly, discrepancy responding to exotic plant S. alterniflora was found between AOA and AOB. To our best knowledge, this is the first study revealing that AOA was significantly reduced in the presence of the invasive plant species. Of particular interest is the fact that the abundance of AOB was one to two orders of magnitude higher after S. alterniflora invasion (in 0–5 cm sediment layer), compared with other zones in this study, which was consistent with previous work (Hawkes et al. 2005), although A more recent study in paddy soil showed that AOA and AOB abundance were mainly determined by the soil types and not related to aboveground rice (Chen et al. 2010). Secondly, amoA and bacterial 16S rRNA gene abundance at shallow layer (0–5 cm) was higher than those at deeper layer (5–20 cm). This different vertical distribution of microbes among sediment layers may be due to bioavailability of nitrogen and organic carbon (Leininger et al. 2006). Finally, our results showed that archaeal amoA copies were more abundant than bacterial amoA copies in sediment for both depths, which was similar to other previous studies (Abell et al. 2010; Jia and Conrad 2009; Leininger et al. 2006). The ratios, especially, of AOA/AOB and AOA/total bacteria with the native mangrove zone were much higher than others. Conversely, the ratio of AOB/total bacteria was increased by the S. alterniflora invasion. In addition, the ratio in the deeper layer was lower than in the shallow layer, suggesting that archaeal amoA dominated over bacterial amoA in sediments, while high ratios of AOA to AOB and AOA to total bacteria in the native mangrove zone were dramatically decreased by the S. alterniflora invasion. A previous study demonstrated that AOA played an important role in nitrification using stable isotope probing method (Zhang et al. 2010). However, other studies have shown that the AOA population sizes were much higher than AOB; AOB rather than AOA functionally dominate ammonia oxidation in grassland soils (Di et al. 2009, 2010) and agricultural soil (Jia and Conrad 2009).

No significant difference in AOB composition was found among different zones (see Fig. 2 and Table 4), demonstrating that AOB was quite stable in relation to plant community composition. Some previous studies also showed that there was little response in AOB composition to long-term fertilization treatments (He et al. 2007), vegetations in terrestrial ecosystem (Ying et al. 2010), ammonium, and temperature (Avrahami and Conrad 2003), and the same type flooded paddy soils (Chen et al. 2010). High variation in overall bacterial composition in relation to a stable ammonia oxidizer community was also found in forest soil (Laverman et al. 2001) and freshwater sediments (Speksnijder et al. 1998). The slow-growing and specialized ammonia oxidizers are frequently more tolerant to these environmental changes than the fast-growing and changing bacterial community (Hawkes et al. 2005). However, the cDNA DGGE profile of AOB communities without rice plants were clearly different from those with planted rice (Chen et al. 2008). The AOB sequences recovered from DGGE bands were highly similar to the known sequences from the previous studies with a variety of marine-related or estuarine sediment environments, such as Jiaozhou Bay (Dang et al. 2010), Pearl River Estuary, marine sponge (Mohamed et al. 2010), San Francisco Bay (Mosier and Francis 2008), and Chesapeake Bay (Francis et al. 2003). These provide further evidence that this AOB group is ubiquitous in environments, suggesting that AOB found here are not specific in Jiulong River estuary of this study. These amoA sequences were not closely affiliated with any amoA gene from known cultured ammonia-oxidizing bacteria strains, but all phylotypes were affiliated with Nitrosospira spp.

AOA abundance was reduced by S. alterniflora invasion and positively related to bacterial abundance, while AOA diversity was higher in the invaded and unvegetated zones (see Fig. 3 and Table 4). It might imply that the change in AOA abundance could be partly due to the shifts of the community composition. More taxa of ammonia-oxidizing archaea may play functional role in the unvegetated and invaded zones than those in the native mangrove zone. The composition and population sizes of AOA were changed by plant invasion, which was similar to land utilization patterns (Ying et al. 2010). All AOA sequences fell in cluster S (soil/sediment) and are closely related to those found in the estuarine sediment from Australia (Abell et al. 2010) and China (He et al. 2007; Zhou et al. 2009), suggesting that these organisms are widely distributed in similar environments. Sequences originating from terrestrial sources suggested that some of AOA may be a result of the influx of particulate matter from land runoff, consistent with the high rainfall in this catchment (Abell et al. 2010).

The sediment overall bacterial composition and gene abundance in the sediments were impacted by S. alterniflora invasion. Interestingly, bacteria had no relationship with AOB, while it had a close correlation with AOA community. This may be due to the exotic plants′ root exudates (Callaway and Aschehoug 2000) inducing changes in sediment environmental parameters, such as pH, carbon, nitrogen, and sulfur, in turn, significantly shaping rhizospheric bacterial community structure (Haichar et al. 2008; Hunter et al. 2006; Berg and Smalla 2009). Our results demonstrated sediment properties had similar impacts on both total bacteria and AOA but not on AOB. Specific microbial community favors for specific environment. Therefore, this indicated that exotic plant–microbe interaction in sediment has a complex influence on ecosystem function.

The process of biological invasion has been artificially divided into four stages as transport, colonization, establishment, and spread (Theoharides and Dukes 2007). Indeed, knowledge of temporal patterns of ammonia oxidizers is important in the S. alterniflora invasion succession (Hawkes et al. 2005). To answer what the response is of ammonia oxidizers to exotic S. alterniflora invasions, the spread stage of S. alterniflora invasion was selected in this study. The samples were collected in the end of June at the growing peak of S. alterniflora. The environmental factor, such as temperature, was chosen to favor ammonia oxidizers and enhance potential nitrification rate (Urakawa et al. 2008; Tourna et al. 2008). Root exudation at the growing peak strongly influence sediment microenvironment. As we learned in this study, AOB composition was quite stable in sediments from the three different zones. Our results could be representative for growing season with a maximum response of ammonia oxidizers to the S. alterniflora invasion, although multiple time point studies may be ideal (Hawkes et al. 2005).

This study revealed the significant effects of S. alterniflora invasion on ammonia oxidizers in estuarine sediment. Since ammonia oxidizers play a critical role in N cycle and transformation in estuarine sediment, these shifts caused by S. alterniflora invasion may be of great ecological importance.