Effect of Different Bacterial-Feeding Nematode Species on Soil

Pedosphere 24(1): 116–124, 2014 ISSN 1002-0160/CN 32-1315/P c 2014 Soil Science Society of China Published by Elsevier B.V. and Science Press

Eﬀect of Diﬀerent Bacterial-Feeding Nematode Species on Soil Bacterial Numbers, Activity, and Community Composition∗1

XIAO Hai-Feng1,2,LIGen2,LIDa-Ming2,HUFeng2 and LI Hui-Xin2,∗2 1Key Laboratory of Tropical Forest Ecology, Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences, Mengla 666303 (China) 2College of Resources and Environmental Sciences, Nanjing Agricultural University, Nanjing 210095 (China) (Received July 31, 2013; revised December 7, 2013)

ABSTRACT The effects of bacterial-feeding nematodes on bacterial number, activity, and community composition were studied through a microcosm experiment using sterilized soil inoculated with soil bacteria (soil suspension) and with bacteria and three species of bacterial-feeding nematodes (Cephalobus persegnis, Protorhabditis filiformis,andCaenorhabditis elegans). Catalyzed reporter deposition-fluorescence in situ hybridization, CO2 evolution, and denaturing gradient gel electrophoresis (DGGE) of PCR amplified 16S rRNA gene fragments were used to investigate bacterial numbers, activity, and community composition, respectively. Our results showed that bacterial numbers and activity significantly increased in the presence of bacterial-feeding nematodes, which indicated that bacterial-feeding nematodes had a significant positive effect on soil bacteria. The different nematode species had different effects on bacterial numbers and activity. C. persegnis and P. filiformis, isolated from native soil, increased the bacterial number and activity more than C. elegans. The DGGE analysis results showed that dominant bacterial species significantly differed among the treatments, which suggested that bacterial-feeding nematode species modified the bacterial community composition in soil. Further gene sequence analysis results showed that the dominant bacterial species in this study were gram-negative bacteria. Given the com- pletely same conditions except nematode species, the varied selective feeding behavior of different nematode species was the most likely reason for the altered bacterial community composition. Overall, the alteration of bacterial numbers, activity and community composition resulting from the bacterial-feeding nematodes may ultimately affect soil ecological functioning and processes.

Key Words: CARD-FISH, CO2 evolution, DGGE, gene sequence, gram-negative bacteria

Citation: Xiao, H. F., Li, G., Li, D. M., Hu, F. and Li, H. X. 2014. Eﬀect of diﬀerent bacterial-feeding nematode species on soil bacterial numbers, activity, and community composition. Pedosphere. 24(1): 116–124.

Bacterial community structure and activity are grazing and conclusions regarding the changes in bac- central factors that influence terrestrial ecosystem terial numbers caused by nematode grazing have been functions (Kennedy and Gewin, 1997) and are changed inconsistent. Researchers found that bacteria increase by bacterivorous predators such as bacterial-feeding when predators graze on them (Abrams and Mitchell, nematodes (Griffiths et al., 1999; Rønn et al., 2002; 1980; Traunspurger et al., 1997; Bardgett et al., 1998), Djigal et al., 2004) that graze in soils. Bacterial-feeding whereas others obtained the contrasting results (Cole- nematodes have been recognized as a main bacterivo- man et al., 1977; Anderson et al., 1983). rous predator (Griffiths, 1994; Li and Hu, 2001) be- Numerous studies have also reported bacterial- cause of their greater abundance and consumption in feeding nematodes that graze on bacteria, thereby in- soils. For example, Bernard (1992) and Liang et al. fluencing bacterial activity (usually by detecting soil (2000) found that nematodes are the most abundant respiration and enzyme activity) (Anderson and Cole- metazoans in the soil, ranging from 7.6 × 105 m−2 in man, 1977; Trofymow et al., 1983; Djigal et al., 2004; deserts to 2.9 × 107 m−2 in mixed deciduous forests. Fu et al., 2005). Most of these studies showed in- Furthermore, Venette and Ferris (1998) found that an creased bacterial activity in the presence of nema- adult bacterial-feeding nematode consumes 1 × 106 todes. For example, Fu et al. (2005) found that mi- cells daily. Bacterial-feeding nematodes have a huge crocosms with nematodes produced significantly more grazing potential on soil bacteria. However, this does CO2 than those without nematodes. Although the al- not mean that the number of bacteria will be substan- terations of bacterial numbers and activity in the pre- tially reduced. Bacteria respond variably to nematode sence of bacterial-feeding nematodes were reported by

∗1Supported by the National Natural Science Foundation of China (Nos. 41271270 and 31200409). ∗2Corresponding author. E-mail: [email protected]. NEMATODE EFFECT ON SOIL BACTERIA 117 many previous studies, related reasons were deficiently isolated and identified from the soil used in this study, discussed and need to be explored more deeply. namely, Cephalobus persegnis and Protorhabditis fili- Aside from affecting bacterial abundance and ac- formis,andCaenorhabditis elegans were used to com- tivity, bacterial-feeding nematodes that graze on bac- pare the effect of different bacterial-feeding nematode teria even modify the bacterial community composi- species on the soil bacterial number, activity, and com- tion (Griffiths et al., 1999; Djigal et al., 2004). Howe- munity composition. Altered bacterial species were also ver, previous studies did not examine which bacte- examined in this study. rial species changed, and the possible mechanisms MATERIALS AND METHODS have been rarely reported and need more investiga- tion. Generally, nematodes are selective and display Soil specific preferences when grazing. In turn, different bacterial resources may be appropriate for different ne- The soil used in this study was a sandy soil (57.5% matode species and affect nematode growth and fecun- sand, 26.6% silt, and 15.9% clay) collected from Nan- dity. Venette and Ferris (1998) found that six different tong City, Jiangsu Province, China. The soil contained −1 −1 bacterial-feeding nematode species have different re- 6.03 g kg organic C, 0.70 g kg total N, and 0.68 −1 production rates according to the ingested bacterium. gkg total P, and the pH (H2O) was 6.85. Before Selective feeding behavior may induce competition be- use, the fresh soil was passed through a 2-mm mesh to tween microbes, thereby altering the community com- remove stones, macrofauna, and some broken roots. A position and distribution in soil (Fu et al., 2005; Blanc portion of the soil was used to isolate nematodes and et al., 2006). prepare bacterial suspensions, and the remainder was ◦ Considering the complexity of environmental con- sterilized by heating at 121 C (102.9 kPa) for 30 min ditions and the limitation of research tools, obser- (Xiao et al., 2010a). ving the activity of microfauna and bacteria directly Nematode isolation and incubation in the soil is difficult. Recently, however, catalyzed reporter deposition-fluorescent in situ hybridization Two native nematode species were isolated from (CARD-FISH) with horseradish peroxidase (HRP)- the soil via a modified cotton wool filter method (Liang labeled oligonucleotide probes and tyramide signal am- et al., 2009) and identified as Cephalobus persegnis and plification and DNA fingerprinting (such as denaturing Protorhabditis filiformis, both of which are bacterial- gradient gel electrophoresis, DGGE) have been used to feeding nematodes that belong to the family Rhabditi- investigate the numbers and the community structure dae. A single individual of each was allowed to multiply of bacteria, respectively. CARD-FISH, instead of con- on an agar plate with Tryptic soy broth culture con- ventional monolabeled FISH (oligonucleotide probes taining mixed soil bacteria (soil bacterial suspension labeled with Cy3 fluorochrome), was suitable for soil being coated onto the plates) as the food source. C. microorganism analysis (Pernthaler et al., 2002; Eick- elegans maintained in long-term cultures in our labora- horst and Tippkötter, 2008). Both the detection sensi- tory was selected because it also belongs to the family tivity and specificity of CARD-FISH were higher than Rhabditidae and considered a model organism. All those of monolabeled FISH (Haugland, 2005; Eickhorst three nematode species were reared on agar plates to and Tippkötter, 2008). Muyzer et al. (1993) and Pern- generate enough for inoculation. These three nematode thaler et al. (2002) successfully used CARD-FISH and species had similar generation times of 4 to 5 d when ◦ DGGE to investigate the bacterial distribution and incubated at 28 C. community structure. Soil preparation Alterations in bacterial abundance, activity, and community composition are related to soil ecological To obtain nematode-free soil abundant with bac- functions because bacteria are the drivers of many bio- teria, the soil was sterilized and then treated with a chemical reactions in soil. Considering that different bacterial suspension to increase the bacterial count. nematode species likely have different effects on bacte- The bacterial suspension was filtered through two filter rial community composition because of their selective membranes with 5 μm pores to eliminate nematodes. feeding behavior, we hypothesize that bacterial-feeding Each pot contains 1 000 g of sterilized soil. Up to 40 nematodes modify the bacterial community composi- mL of nematode-free inoculum of the mixed soil bac- tion because of their preference for different bacterial teria (about 2 × 106 g−1 dry soil) was inoculated into species. Two native bacterial-feeding nematode species each pot. 118 H. F. XIAO et al.

Nematode inoculation fication (64 and 112 μm2), and the counts were ex- trapolated onto 1 g of soil (dry soil). To prevent the Each bacterial-feeding nematode species was ino- weakening of the fluorochrome under excitation light, −1 culated into a pot at 20 individuals g dry soil. Be- the samples were treated with Citifluor AF1 (Citifluor fore the nematodes were inoculated into the soil, they Ltd., London) at the surface. Excitation was performed were surface disinfected for 20 min with a mixture of under blue light before UV light because high-energy −1 −1 1.0 g L of streptomycin sulfate and 0.02 g L of UV may destroy the weak fluorescence signals. cycloheximide and then centrifuged (3 000 × g) for 3 min, after which the supernatant was discarded. The Soil respiration analysis nematodes were then washed 5 to 6 times with sterile CO2 evolution from the soils was measured using water to minimize bacterial interference during their the alkali absorption method (Anderson, 1982). Briefly, transfer into the soils. The control pots were not ino- 20 g of soil was weighed and transferred into 300-mL culated with nematodes. Each nematode species and jars, and small vials containing 5 mL of 0.05 mol L−1 the control treatment were performed with three repli- NaOH were placed inside the jars. The jars were sealed cates, with a total of 12 pots. Then, the soil moisture and left for 24 h. After incubation, the excess NaOH content was adjusted to 25% using distilled water. Fi- was titrated using 0.025 mol L−1 of HCl with phe- nally, all the pots were placed in an incubator at 28 ◦ nolphthalein as an indicator and soil respiration rate C in darkness for 28 d. Soils weighing 30, 0.5 and was calculated. 20 g were sampled non-destructively weekly from the beginning of experiment to determine nematode num- Extraction and purification of DNA from soil samples bers, bacterial abundance, and soil respiration. At the After 28 d of incubation, DNA was extracted from end of the incubation, DNA was extracted from 5 g of 5 g of the soil samples by mixing them with 13.5 mL soil to analyze the bacterial community structure. − of DNA extraction buffer (100 mmol L 1 of Tris-HCl, −1 −1 CARD-FISH 100 mol L of EDTA-Na2, 100 mmol L of Na3PO4, 1.5 mol L−1 of NaCl, and 10 g L−1 cetyltrimethylam- The samples were prepared on glass slides accor- monium bromide (CTAB, pH 8.0)). The samples were ◦ ding to the method in Eickhorst and Tippkötter then incubated for 30 min at 37 Cinahorizontal − (2008). Briefly, 0.5 g of fresh soil was weighed and shaking bath at 225 r min 1. After three rounds of ◦ transferred to 2 mL centrifuge tubes. Then, 320 μL freezing in liquid nitrogen and thawing in a 65 Cwa- − of 25% (w/v) particle-free paraformaldehyde solution ter bath, 1.5 mL of 200 g L 1 SDS was added and the ◦ (4% (w/v) final concentration) was added, filled with samples were further incubated for 2 h at 65 Cwith 1 × phosphate-buffered saline (PBS) (137 mmol L−1 agitation every 15 min. The samples were then cen- −1 −1 × NaCl, 2.7 mmol L KCl, 10 mmol L Na2HPO4,and trifuged (6 000 g) at room temperature for 10 min −1 ◦ 2 mmol L KH2PO4), mixed, and stored at 4 C. Af- to collect the supernatant. The supernatant was trans- ter 6 h, the suspension was centrifuged at 10 000 × g ferred into a 50-mL centrifuge tube, extracted with for 5 min at 4 ◦C, washed twice with 1 × PBS, cen- phenol and then purified with chloroform-isoamyl al- trifuged again under the same condition, and stored in cohol (24:1, v:v). The aqueous phase was transferred PBS/ethanol (1:1, v:v) at 20 ◦C for further processing. into 50-mL centrifuge tubes. Then, 0.6 volumes of iso- Then, 100 μL of the fixed sample was diluted with 900 propanol were added and the mixture was incubated μL of PBS/ethanol. Up to 30 μL of the diluted sample, at room temperature for 1 h. The samples were then 60 μLof1× PBS and 10 μL of 0.01% (v/v) sodium centrifuged (6 000 × g) at room temperature for 20 dodecyl sulfate (SDS) were placed on a glass slide. The min. After centrifugation, nucleic acid was collected CARD-FISH procedure of Pernthaler et al. (2002) was and washed with cold 70% (v/v) ethanol, dissolved in followed with slight modification. The samples on the 300 μL of sterile ultrapure water, and the crude DNA slides were hybridized with the probes EUB338 (5- product was purified using an OMEGA purification kit GCT GCC TCC CGT AGG AGT-3) (Eickhorst and (Omega Bio-Tek, USA) (Xiao et al., 2010a). Tippkötter, 2008). HRP-conjugated oligonucleotides PCR and DGGE probes were purchased from TaKaRa, Japan. Cy3- labeled tyramide was purchased from China Isotope Universal bacterial primers that amplified a 194 bp Corporation. Automated cell counting was performed fragment of the 16S rRNA gene, including the third on 10 to 15 randomly selected visual fields, the selected variable (V3) region, were used in this experiment. A visual fields were photographed under 40 × magni- 40 bp GC clamp was added at the 5 end of the for- NEMATODE EFFECT ON SOIL BACTERIA 119 ward primer. The PCR protocol included 5 min of ini- at 20 individuals g−1 dry soil and the growth of the tial denaturation at 94 ◦C, 30 cycles at 94 ◦C for 30 nematodes is shown in Fig. 1. The number of nema- s, at 61 ◦C for 30 s, and at 72 ◦C for 30 s, followed todes significantly increased over time (F = 1 443, by a final extension at 72 ◦C for 5 min. The reaction P<0.01). C. persegnis and P. filiformis were signifi- mixtures (50 μL) contained 1 × PCR reaction buffer cantly more than C. elegans after Day 14, but the num- (TaKaRa, Japan), 100 ng of DNA template, 10 pmol ber of C. persegnis was not significantly different from L−1 of the forward and reverse primers, 200 μmol L−1 that of P. filiformis. In addition, time significantly in- of deoxynucleotide (dNTP) mix, and 2.5 units of Ex teracted with the nematode treatment (F = 279.448, Taq DNA Polymerase (TaKaRa, Japan). P<0.01). The PCR-DGGE was performed with a DCode mutation detection system (Bio-Rad, USA). Poly- acrylamide gels (8% (v/v) of a 37.5:1 acrylamide- bisacrylamide mixture in 1 × Tris-acetate-EDTA (TAE) buffer) with a gradient of 30% to 70% denaturant (100% denaturant containing 7 mol L−1 urea and 40% (v/v) formamide) (Muyzer et al., 1993). Approxi- mately 200 ng of each PCR product was loaded, and the gels were electrophoresed for 5 h at 200 V and 60 ◦C. The gels were silver stained and fixed for image analysis using Quantity One gel analysis software ver- sion 4.62 (Bio-Rad, USA).

Cloning and sequencing

PCR amplifications without GC clamp were con- Fig. 1 Bacterial-feeding nematode numbers of the soils ino- ducted using the OMEGA quick PCR purification culated with three nematode species (C. persegnis, P. filiformis, kit prior to cloning. Approximately 4 μL of puri- and C. elegans). Error bars represent standard deviations of the means (n = 3). Different letters denoted statistically signifi- fied products were ligated into the PMD-19T vec- cant differences between the treatment groups according to least tor (TaKaRa Cloning Kit) (TaKaRa, Japan) and significant difference (LSD) test (P<0.05). then further transformed into Escherichia coli compe- Soil bacterial numbers tent cells DH5α (TaKaRa, Japan). The white colonies were randomly selected from each cloned sample, grew The bacterial numbers significantly increased du- overnight, and then sequentially reacted on an ABI ring the experiment (F =92.605, P<0.05), with 377 apparatus (BGI Company, China). The nucleotide those in the nematode treatments about twice that sequences were deposited in the GenBank database of the control (Fig. 2). The number of bacteria in the and assigned with accession numbers FJ911506 to C. persegnis and P. filiformis treatments were signifi- FJ911520. cantly higher than those with C. elegans on Day 14, with no significant difference between the C. persegnis Statistical analysis and P. filiformis treatments. By Day 28, the number A repeated measures analysis of variance (ANO- of bacteria in the C. persegnis treatment was signifi- VA) was used to determine significant differences cantly higher than those in the P. filiformis and C. among treatments at different dates. Differences a- elegans treatments. mong treatments were tested by a least significant di- Soil respiration fference (LSD) test (P<0.05). For DGGE analysis, Quantity One software was used to test the band in- Soil respiration was significantly higher in the pre- tensity. All the bands were normalized and principal sence of bacterial-feeding nematodes than the control component analysis (PCA) was performed using SPSS (F = 344.675, P<0.01) (Fig. 3). Soil respiration 16.0 software. showed no significant difference over time (F =1.765, P>0.05), but time significantly interacted with the RESULTS treatments (F =8.053, P<0.05), which indicated that the bacterial-feeding nematodes significantly in- Growth of bacterial-feeding nematodes in soil creased the soil respiration. The nematode treatments The soil was initially inoculated with nematodes did not significantly differ except on Day 14, wherein 120 H. F. XIAO et al.

cated that the bacterial diversity increased in the presence of the bacterial-feeding nematodes. The intensity of some bands (e.g.,a,b,d,f,i,andj)increased;however, the intensity of band c decreased in the presence of C. persegnis and P. filiformis. Additionally, bands g and h were associated with C. persegnis. The intensity of band k was sharply increased in the treatment with P. filiformis and intensity of bands l and m increased in the presence of C. elegans. The PCA of the DGGE profile differentiated the four treatments, especially the nematode treatments from the control (Fig. 5).

Fig. 2 Effect of bacterial-feeding nematodes on bacterial numbers of the control soil and the soils inoculated with three nematode species (C. persegnis, P. filiformis,andC. elegans). The control consisted of sterilized soil only inoculated with soil bacteria (without the nematodes). Error bars represent standard deviations of the means (n = 3). Different letters denote statistically significant differences between the treatment groups according to the least significant difference (LSD) test (P<0.05).

Fig. 4 Denaturing gradient gel electrophoresis (DGGE) analysis of 16S rRNA gene sequences ampliﬁed from DNA extracted from three replicate samples of the control soil and the soils inoculated with three nematode species (C. persegnis, P. ﬁliformis, and C. elegans).

Fig. 3 Effect of bacterial-feeding nematodes on soil respiration of the control soil and the soils inoculated with three nematode species (C. persegnis, P. filiformis,andC. elegans). The control consisted of sterilized soil only inoculated with soil bacteria (without the nematodes). Error bars represent standard deviations of the means (n = 3). Different letters denote statistically significant differences between treatment groups according to the least significant difference (LSD) test (P<0.05). the soil respiration in the C. persegnis and P. filiformis Fig. 5 Principal component analysis of the denaturing gradient treatments was significantly higher than that in the C. gel electrophoresis (DGGE) band patterns affecting the bacteria elegans treatment. community structure. The error bars represent the least significant difference; hence, the coordinates are significantly different (P<0.05) when the error bars do not overlap with the mean Soil bacterial community structure from another treatment.

The differences among the four treatments were To further analyze which bacterial species was al- clearly discernible from the bands in the DGGE gel tered by the different nematode species, we cloned and (Fig. 4). The number of bands increased in the nema- sequenced all the labeled bands (Fig. 4). The sequences tode treatments compared with the control, which indi- were aligned with previously published sequences using NEMATODE EFFECT ON SOIL BACTERIA 121 the Basic Local Alignment Search Tool (BLAST) in the 28 ◦C. Therefore, we consider that the differences be- NCBI database, and the results are shown in Table I. tween the number of C. elegans and those of the two native nematode species may have resulted from diffe- DISCUSSION rences in their adaptability to soil. Our results also showed that the number of soil bac- Effects of bacterial-feeding nematodes on bacterial teria significantly increased in the presence of bacterial- numbers and activity feeding nematodes (Fig. 2), which is consistent with many previous studies (Traunspurger et al., 1997; Ba- The numbers of nematodes gradually increased rdgett et al., 1998; Fu et al., 2005). There are two pos- over time, peaked on Day 14, and gradually decreased sible explanations for these results. First, according to thereafter (Fig. 1). The bacterial generation time was the nutritional dynamic hypothesis proposed by Car- much shorter than that of the nematodes. Thus, the penter et al. (1985), reproduction rates of one tro- bacterial growth rate was faster than the nematode phic level are maximal under moderate predation pres- growth rate in soil. This is why the bacteria rapidly sure by higher trophic level animals. Therefore, mode- increased and almost peaked on Day 7, whereas the rate nematode grazing may keep the bacterial number number of nematodes peaked on Day 14. With the in- growth rapidly. Fu et al. (2005) considered that the crease in the number of bacteria, the number of nema- bacteria-to-nematode ratio is a critical index for the todes increased gradually because of the increased bac- predator and prey relationship in soil, which deter- terial food resource. However, with the consumption mines bacteria and nematode growth. This surmise was of the substrate, both the bacteria and the nematodes confirmed by our previous study (Xiao et al., 2010b), decreased after two weeks. Therefore, sufficient sub- which we designed with different nematode concentra- strate for bacterial growth is an important factor for tions and found a significant “density regulating effect” stabilizing predator-prey fluctuation. In barren soil or between the nematodes and the bacteria. Second, the in closed experimental systems with limited resources, nematodes also excrete inorganic nitrogen and other the number of bacteria might decrease with the defi- organic matter, which provide nutrition for bacteria cient substrate under the pressure of grazing bacterial- and stimulate their growth (Anderson et al., 1983; In- feeding nematodes. Additionally, the lag times associ- gham et al., 1985; Griffiths and Bardgett, 1997). An- ated with different nematode and bacterial generation derson et al. (1983) reported that nematodes excreted times were important factors that affect the numbers significant amounts of amino acids into soil. Further- of nematodes and bacteria (Neher, 2010). However, more, most bacteria maintain activity during passage the generation times of the three different nematode through the nematode alimentary canal (Yeates, 1969; species in this study were similar, around 4 to 5 d at Smerda et al., 1971; Ingham et al., 1985; Bird and Ry-

TABLE I Alignment of 16S rRNA sequences of bacterial species from the control soil and the soils inoculated with three nematode species (C. persegnis, P. ﬁliformis,andC. elegans)

Banda) Sequenceb) Accession no. Gram staining characteristic Similarity Nearest relative % a FJ911506 Negative 100 Sphingobacterium sp. b FJ911507 Negative 100 Uncultured Flavobacterium sp. c FJ911508 Negative 98 Uncultured Sphingobacterium sp. d FJ911509 Negative 99 Uncultured Pseudomonas sp. e FJ911510 Negative 99 Flavobacterium sp. f FJ911511 Negative 96 Chitinophaga sp. g FJ911512 Negative 98 Uncultured Bacteroides sp. h 1 FJ911513 Negative 96 Uncultured Fluviicola sp. 2 FJ911514 Negative 96 Uncultured Fluviicola sp. i 1 FJ911515 Negative 100 Sphingobacterium sp. 2 FJ911516 Negative 98 Sphingobacterium sp. j FJ911517 Negative 93 Uncultured Solitalea sp. k FJ911518 Negative 100 Uncultured Flavobacterium sp. l FJ911519 Negative 96 Uncultured Stenotrophomonas sp. m FJ911520 Negative 98 Sphingobacterium sp. a)See Fig. 4 for the band code. b)From some bands, two separate bands were obtained after rerunning the excised band. 122 H. F. XIAO et al. der, 1993), and these bacteria possibly obtained hor- the microbial community structure than Acrobeloides mones and nutrients to help them grow. Although some nanus andCephalobus pseudoparvus. A. nanus resulted studies indicated that nematodes negatively affect bac- in the greatest change in the structure of the microbial terial numbers (Coleman et al., 1977; Anderson et al., community. Size-selective feeding by nematodes may 1983), these inconsistencies may be nematode species- alter the bacterial community composition because ne- specific (Ingham et al., 1985). Although all the nema- matodes can utilize probolae to restrict the size of food tode species may be capable of stimulating bacterial entering the buccal cavity (Lee and Atkinson, 1977). In growth, the additional production may be consumed the current study, P. filiformis wasabout0.4to0.5mm before a net increase in numbers can be observed (An- long, with a 16- to 18-μm long buccal cavity. C. perseg- derson et al., 1983). nis was about 0.7 to 0.8 mm long, with a 10- to 12-μm Although nematode grazing increased the number long buccal cavity (Wu, 1999). Adult C. elegans was of bacteria in our study, the different nematode species about 1 to 1.5 mm long (Wood, 1988), with a 20-μm affected the number of bacteria differently. As shown long buccal cavity. These different oral morphologies in Fig. 2, the numbers of bacteria in the C. persegnis probably preferred bacteria with different sizes. There- and P. filiformis treatments were significantly higher fore, the morphology and size differences of the nema- than that in the C. elegans treatment, except on Day todes may have led to differences in bacterial commu- 7, with no significant difference in the number of bac- nity composition. teria between the C. persegnis and P. filiformis treat- In addition, bacterial-feeding nematodes may be ments. This result indicated that the two native nema- capable of distinguishing the structure of bacterial cell tode species were more effective in increasing bacterial walls and selecting the food edibility while rapidly ex- number than C. elegans. creting the unsuitable bacteria. Bacterial-feeding ne- Respiration is an important index for bacterial matodes generally favor Gram-negative bacteria over activity in soils (Coleman et al., 1983). Our study Gram-positive bacteria because their thinner cell walls showed that compared with the control, soil respi- are easier to digest (Tortora et al., 2000), which is ration was significantly increased by both the native supported by Salinas et al. (2005), who found that bacterial-feeding nematode species and C. elegans after Cephalobus brevicauda preferred Gram-negative bac- 7 d (Fig. 3). This result indicated that nematode gra- teria. In this study, whether the different nematode zing increased bacterial activity (Woods et al., 1982; species preferred the dominant bacterial species in Coleman et al., 1983; Djiagal et al., 2004; Fu et al., each treatment was unclear. However, all the examined 2005). Results of soil respiration also showed that the dominant bacterial species were Gram-negative bacte- CO2 evolution in all the treatments peaked on Day ria (Table I). 7, which indicated that the bacteria exhibited higher Moreover, nematodes can identify the smell of activity under the stimulation of the bacterial-feeding different bacteria to distinguish their favorite foods nematodes when the bacteria were actively growing. (Zhang et al., 2005). Newsham et al. (2004) also found Similar to the bacterial numbers, the effects of native that different nematode species (Geomonhystera vil- nematode species on soil respiration were stronger than losa, Plectus spp., and Teratocephalus spp.) have dif- that of C. elegans (Fig. 3). ferent preferences for microbes (two microalgae, three microfungi, and six heterotrophic bacteria) in Antarc- Effect of bacterial-feeding nematodes on bacterial com- tic soil. Other indirect factors affect the composition of munity composition bacterial communities, such as nutrient and substrate In the DGGE profile analysis, we assumed that the availability, may also be operating. number and intensity of the bands in the DGGE pro- CONCLUSIONS file reflected bacterial diversity and abundance. The number and intensity of the bands were significantly The presence of bacterial-feeding nematodes in- different among the four treatments (Fig. 5), which increased the bacterial numbers and activity and changes dicated that the bacterial community was significantly the microbial community composition through a va- changed by the bacterial-feeding nematodes, consistent riety of selective feeding behavior. Native nematodes with the previous studies by Griffiths et al. (1999), Dji- had higher reproduction rates than C. elegans,with gal et al. (2004), and De Mesel et al. (2004). Djigal et the effect of the former stronger than that of the latter. al. (2004) found that different nematode species in- These results help us better understand the interaction fluence the soil microbial community differently. For between microfauna and bacteria as well as provide example, Zeldia punctata clearly has less impact on a basis for understanding the influence of microfauna NEMATODE EFFECT ON SOIL BACTERIA 123 grazing on microbial communities and functioning in population size? Soil Biol. Biochem. 37: 1979–1987. soil ecology. Griffiths, B. S. 1994. Microbial-feeding nematodes and protozoa in soil: Their effects on microbial activity and nitro- ACKNOWLEDGEMENTS gen mineralization in decomposition hotspots and the rhizosphere. Plant Soil. 164: 25–33. The manuscript was greatly improved by the in- Griffiths, B. S. and Bardgett, R. D. 1997. Interactions between sightful suggestions of the anonymous reviewers. We micro-feeding invertebrates and soil microorganisms. In van also thank the Public Technology Service Center, Xi- Elsas, J. D., Trevors, J. T. and Wellington, E. M. H. (eds.) Modern Soil Microbiolgy. Marcell Dekker, New York. pp. shuangbanna Tropical Botanical Garden, Chinese Aca- 165–182. demy of Sciences, for soil analyses. Griffiths, B. S., Bonkowski, M., Dobson, G. and Caul, S. 1999. Changes in soil microbial community structure in the presence of microbial-feeding nematodes and protozoa. Pedobi- REFERENCES ologia. 43: 297–304. Abrams, B. I. and Mitchell, M. J. 1980. Role of nematode-bacte- Haugland, R. P. 2005. The Handbook. A Guide to Fluorescent rial interactions in heterotrophic systems with emphasis on Probes and Labelling Technologies. 10th Ed. Molecular Pro- sewage sludge decomposition. Oikos. 35: 404–410. bes Inc., Eugene. Anderson, J. P. E. 1982. Soil respiration. In Page, A. L. (ed.) Ingham, R. E., Trofymow, J. A., Ingham, E. R. and Coleman, D. Methods of Soil Analysis. Part 2. Chemical and Microbio- C. 1985. Interactions of bacteria, fungi, and their nematode logical Properties. SSSAJ, Madison. pp. 831–871. grazers: effects on nutrient cycling and plant growth. Ecol. Anderson, R. V. and Coleman, D. C. 1977. The use of glass mi- Monogr. 55: 119–140. crobeads in ecological experiments with bacteriophagic ne- Kennedy, A. C. and Gewin, V. L. 1997. Soil microbial diversity: matodes. J. Nematol. 9: 319–322. present and future considerations. Soil Sci. 162: 607–617. Anderson, R. V., Gould, W. D., Woods, L. E., Cambardella, C., Lee, D. L. and Atkinson H. J. 1977. Physiology of Nematodes. Ingham, R. E. and Coleman, D. C. 1983. Organic and inor- Columbia University Press, New York. ganic nitrogenous losses by microbivorous nematodes in soil. Li, H. X. and Hu, F. 2001. Effect of bacterial-feeding nematode Oikos. 40: 75–80. inoculation on wheat growth and N and P uptake. Pedo- Bardgett, R. D., Keilier, S., Cook, R. and Gilburn, A. S. 1998. sphere. 11: 57–62. Dynamic interactions between soil animals and microorga- Liang, W. J., Lou, Y. L., Li, Q., Zhong, S., Zhang, X. K. nisms in upland grassland soils amended with sheep dung: a and Wang, J. K. 2009. Nematode faunal response to long- microcosm experiment. Soil Biol. Biochem. 30: 531–539. term application of nitrogen fertilizer and organic manure in Bernard, E. C. 1992. Soil nematode biodiversity. Biol. Fert. Northeast China. Soil Biol. Biochem. 41: 883–890. Soils. 14: 99–103. Liang, W. J., Shi, Y. and Steinberger, Y. 2000. Research pro- Bird, A. F. and Ryder, M. H. 1993. Feeding of the nematode gress of nematode diversity in agroecosystems. Chin. J. Acrobeloides nanus on bacteria. J. Nematol. 25: 493–499. Appl. Ecol. (in Chinese). 11(Suppl.): 113–116. Blanc, C., Sy, M., Djigal, D., Brauman, A., Normand, P. and Muyzer, G., de Waal, E. C. and Uitterlinden, A. G. 1993. Pro- Villenave, C. 2006. Nutrition on bacteria by bacterial-feeding filing of complex microbial populations by denaturing gradi- nematodes and consequences on the structure of soil bacterial ent gel electrophoresis analysis of polymerase chain reaction- community. Eur. J. Soil. Biol. 42: S70–S78. amplified genes coding for 16S rRNA. Appl. Environ. Mi- Carpenter, S. R., Kitchell, J. F. and Hodgson, J. R. 1985. Ca- crobiol. 59: 695–700. scading trophic interactions and lake productivity. Bio- Neher, D. A. 2010. Ecology of plant and free-living nematodes Science. 35: 634–639. in natural and agricultural soil. Annu. Rev. Phytopathol. 48: Coleman, D. C., Cole, C. V., Anderson, R. V., Blaha, M., Cam- 371–394. pion, M. K., Clarholm, M., Elliott, E. T., Hunt, H. W., Newsham, K. K., Rolf, J., Pearce, D. A. and Strachan, R. J. Shaefer, B. and Sinclair, J. 1977. An analysis of rhizosphere- 2004. Differing preferences of Antarctic soil nematodes for saprophage interactions in terrestrial ecosystems. Ecol. Bull. microbial prey. Eur. J. Soil. Sci. 40: 1–8. 25: 299–309. Pernthaler, A., Pernthaler, J. and Amann, R. 2002. Fluorescence Coleman, D. C., Reid, C. P. P. and Cole, C. V. 1983. Biological in situ hybridization and catalyzed reporter deposition for strategies of nutrient cycling in soil systems. Adv. Ecol. Res. the identification of marine bacteria. Appl. Environ. Micro- 13: 1–55. biol. 68: 3094–3101. De Mesel, I., Derycke, S., Moens, T., van der Gucht, K., Vincx, Rønn, R., McCaig, A. E., Griffiths, B. S. and Prosser, J. I. 2002. M. and Swings, J. 2004. Top-down impact of bacterivorous Impact of protozoan grazing on bacterial community struc- nematodes on the bacterial community structure: a micro- ture in soil microcosms. Appl. Environ. Microbiol. 68: cosm study. Environ. Microbiol. 6: 733–744. 6094–6105. Djigal, D., Brauman, A., Diop, T. A., Chotte, J. L. and Vil- Salinas, K. A., Edenborn, S. L., Sexstone, A. J. and Kotcon, lenave, C. 2004. Influence of bacterial-feeding nematodes J. B. 2005. Bacterial preferences of the bacterivorous soil (Cephalobidae) on soil microbial communities during maize nematode Cephalobus brevicauda (Cephalobidae): Effect of growth. Soil Biol. Biochem. 36: 323–331. bacterial type and size. Pedobiologia. 51: 55–64. Eickhorst, T. and Tippkötter, R. 2008. Improved detection Smerda, S. M., Jensen, H. J. and Anderson, A. W. 1971. Es- of soil microorganisms using fluorescence in situ hybridiza- cape of salmonellae from chlorination during ingestion by tion (FISH) and catalyzed reporter deposition (CARD- Pristionchus lheritieri (Nematoda: Diplogasterinae). J. Ne- FISH). Soil Biol. Biochem. 40: 1883–1891. matol. 3: 201–204. Fu, S. L., Ferris, H., Brown, D. and Plant, R. 2005. Does the Tortora, G. J., Funke, B. R. and Case, C. L. 2000. Microbiolo- positive feedback effect of nematodes on the biomass and ac- gy: An Introduction. 7th Ed. Pearson Benjamin Cummings, tivity of their bacteria prey vary with nematode species and London. 124 H. F. XIAO et al.

Traunspurger, W., Bergtold, M. and Goedkoop, W. 1997. The nese Academy of Sciences. effects of nematodes on bacterial activity and abundance in Xiao, H. F., Griffiths, B., Chen, X. Y., Liu, M. Q., Jiao, J. G., freshwater sediment. Oecologia. 112: 118–122. Hu, F. and Li, H. X. 2010a. Influence of bacterial-feeding Trofymow, J. A., Morley, C. R., Coleman, D. C. and Anderson, nematodes on nitrification and the ammonia-oxidizing bac- R. V. 1983. Mineralization of cellulose in the presence of teria (AOB) community composition. Appl. Soil Ecol. 45: chitin and assemblages of microflora and fauna in soil. Oe- 131–137. cologia. 60: 103–110. Xiao, H. F., Li, D. M., Chen, X. Y., Liu, M. Q., Zheng, J. W., Venette, R. C. and Ferris, H. 1998. Influence of bacterial type Jiao, J. G., Hu, F. and Li, H. X. 2010b. Effects of bacterial- and density on population growth of bacterial-feeding ne- feeding nematodes on the amount of ammonia oxidizing bac- matodes. Soil Biol. Biochem. 30: 949–960. teria colony in soils using CARD-FISH. Acta Ecol. Sin. (in Wood, W. B. 1988. The Nematode Caenorhabditis elegans. Chinese). 30: 5413–5421. Cold Spring Harbor Laboratory Press, New York. Yeates, G. W. 1969. Predation by Mononchoides poto- Woods, L. E., Cole, C. V., Elliot, E. T., Anderson, R.V. hikus (Nematoda: Diplogasteridae) in laboratory culture. and Coleman, D. C. 1982. Nitrogen transformations in soil Nematologica. 15: 1–9. as affected by bacterial-microfaunal interactions. Soil Biol. Zhang, Y., Lu, H. and Bargmann, C. I. 2005. Pathogenic bac- Biochem. 14: 93–98. teria induce aversive olfactory learning in Caenorhabditis Wu, J. H. 1999. Studies of freshwater and soil nematodes of elegans. Nature. 438: 179–184. China. Ph.D. Dissertation, Institute of Hydrobiology, Chi-