Applied and Environmental Microbiology, April 2003, p . 2116-2125, Vol . 69, No . 4

Spatial and Temporal Analysis of the Microbial Community in Slow Sand Filters Used for Treating Horticultural Irrigation Water

Leo A . Calvo-Bado,¹ Tim R . Pettitt,¹ Nick Parsons,² Geoff M . Petch,¹ J . Alun W . Morgan,^1* and John M . Whipps¹

Plant Pathology and Microbiology Department,¹ Biometrics Department, Horticulture Research International, Wellesbourne, Warwickshire CV35 9EF, United Kingdom²

Received 29 October 2002/ Accepted 17 January 2003

The action of SSFs against many microorganisms, including some human pathogens and plant pathogens, is still not fully understood but is considered to be a combination of physicochemical and biological processes which remove and consume much of suspended particulate organic carbon from the water passing through the sand column (24, 35, 65, 66) . Using traditional microbial methods, the diversity and dynamics of bacteria in SSF used in the water industry have been widely studied (8, 13, 18) . However, much of this work has relied upon conventional plating and isolation techniques which do not study the nonculturable and fastidious species generally thought to dominate environmental samples (49) . Direct methods to study microbial populations overcome this limitation, and in recent years, a genetic fingerprinting technique, denaturing gradient gel electrophoresis (DGGE) (40, 41), has been widely applied to ecological studies of microbial populations (39) . This method, like any other, is not without its problems: differences in the efficiency of DNA extraction from different cell types are likely to exist, 16S rRNA copy number per cell is known to vary, and the kinetics of PCR amplification of the different molecules present in samples containing a diversity of bacteria may not be uniform (7, 64) . However, PCR amplification of DNA and DGGE analysis of the products has provided a useful means to directly characterize many bacterial populations within samples .

This paper describes the characterization of experimental SSF units shown to efficiently remove zoospores of Phytophthora cryptogea . Both spatial and temporal information on the bacterial community present within the SSF column were obtained over the critical early stages of the filter maturation process . For this, conventional plating, direct microscopic observation, and ATP analysis techniques were used in combination with PCR DGGE profiling of the bacterial population . To our knowledge, the results presented here are the first where DGGE has been applied to the study of the dynamic spatial-temporal community in SSFs .

 
Zoospore removal by SSF.
To test the efficacy of the SSF column, an inoculum of P . cryptogea zoospores was applied to each filter column and the filtrate (outflow water) was monitored . Zoospore suspensions (10⁴ spores ml^-1) were prepared as described by Pettitt and Pegg (44) and a 20-ml inoculum was introduced into the water immediately above the sand surface through a specially designed port . Following inoculation, five 1-liter samples of filtered water were collected from each of the SSF columns (20) . Each 1-liter sample was assayed for the presence of Phytophthora CFU by using the membrane filtration technique of Ali-Shtayeh et al . (2) . Filter pieces from this were placed in universal bottles containing 5 ml of resuspension medium and placed on a rotary arm flask shaker at medium speed for 5 min . The resuspension medium, containing antibiotics and fungicides at the same concentration as the modified BNPRA oomycete selective medium (44), consisted of 0.09% (wt/vol) agar dissolved in sterile distilled water and benlate (1 mg ml^-1), pimaricin (1 mg ml^-1), PCNB (2.5 mg ml^-1), rifamycin (1 mg ml^-1), ampicillin (50 mg ml^-1), and nystatin (2.5 mg ml^-1) . One-milliliter aliquots of resuspension medium were pipetted and spread onto modified BNPRA plates and incubated at 25°C for 24 to 36 h . Once visible colonies had formed, CFU per plate were counted, and from these, the mean numbers of Phytophthora (CFU liter^-1) were calculated .

Sampling from SSF columns.
Sand samples from different depths within the SSF column (top, 1 cm; middle layer, 50 cm; bottom layer, 80 cm) were taken 1, 2, and 4 weeks after starting each SSF run . A sterile metal corer was used to take samples of ca . 5 g (wet weight) of sand from the columns via sampling ports . For water samples, 1 liter of water was taken from the storage tank, headwater tank, and outflow water .

Viable bacterial plate counts.
Routinely, 1 g (wet weight) of sand was placed in 10 ml of sterile water and vortexed for 10 s every 2 to 3 min for 15 min . Water samples were used directly . A tenfold dilution series of these samples was made by using sterile water . The dilutions were plated in triplicate on appropriate media . Bacteria were enumerated on 1/10-strength tryptic soy agar (0.1 TSA; Oxoid Ltd., Basingstoke, Hampshire, United Kingdom); fungi were enumerated on potato dextrose agar (Oxoid Ltd.) containing 10 mg of aureomycin ml^-1; and Pseudomonas species was enumerated on P1 agar (29) . All plates were incubated at 20°C for 7 days, and plates with between 30 and 200 colonies were counted .

Direct count of fungal spores.
Sand samples (1 g [wet weight]) were taken 2 and 4 weeks after the start of two SSF runs from the top, middle, and bottom layers of the SSF column and mixed with 1 ml of sterile water . The samples were vortexed for 3 min and centrifuged at 4,000 x g for 5 min . Numbers of fungal spores in dilutions of the supernatant were determined by hemocytometer counts at a magnification of x40 .

ATP extraction and measurement.
To three replicate 0.5-g (wet weight) sand samples, 0.5 g of glass beads (0.1-mm diameter; BioSpect Products, Bartlesville, Okla.) and 500 µl of ATP extraction buffer (0.1 M Tris-acetate [pH 7.75]) were added . The cells within the sample were disrupted by bead beating for 3 min with a mini-bead beater (BioSpect Products) at homogenizing speed . The samples were cooled on ice and centrifuged at 5,000 x g for 5 min . ATP was measured by using the Enliten assay system (Promega, Southampton, United Kingdom) and a luminometer from LKB-Wallac (Uppsala, Sweden), model 1250 . Relative light units were converted into ATP concentrations (in micrograms gram^-1) by using an internal standard curve .

DNA isolation.
DNA was extracted from 1 liter of water and 0.5-g (wet weight) sand samples . The water samples were filtered (0.22-µm-pore-size filters; GV-Durapore Millipore Ltd., Watford, United Kingdom), and the cells were washed off the filter surface with 5 ml of sterile water . Each sample was then centrifuged at 13,000 x g for 20 min, and the pellet was saved . To the cell pellets or the sand samples, 1 ml of extraction buffer (0.12 M K₂HPO₄, pH 8.0) and 1/3 of a volume of 0.1-mm-diameter glass beads (BioSpect Products, Inc.) was added . Each sample was shaken vigorously for 3 min in a mini-bead beater (BioSpect Products, Inc.) . Immediately, 0.1% (wt/vol) sodium dodecyl sulfate was added and the sample was homogenized and placed on ice for 10 min . One milliliter of phenol (pH 8.0; Sigma, Cambridge, United Kingdom) was added, and each sample was centrifuged at 5,000 x g for 15 min . The supernatants containing the DNA were recovered, and 1 ml of chloroform-isoamyl alcohol (24:1) was added to the DNA suspension . After mixing, each sample was centrifuged at 5,000 x g for 15 min and the aqueous phase was saved . The samples were precipitated with a 0.6 volume of isopropanol and a 0.1 volume of 5 M NaCl and washed with 70% (vol/vol) ethanol . The DNA was further purified by using a Geneclean spin kit (Bio 101, Inc., Nottingham, United Kingdom) according to the manufacturer's recommendations . The DNA from each sample was finally eluted in 50 µl and analyzed by agarose gel electrophoresis to estimate the yield . Appropriate dilutions were made for PCR amplifications .

PCR amplification.
The V3 region (39) of the 16S rRNA-borne gene between positions 341 to 534 on the Escherichia coli numbering system was amplified by PCR . Routinely, 10 to 50 ng of DNA template, 25 pmol of each primer, 20 mM deoxynucleoside triphosphate mix, 1.25 U of thermostable DNA polymerase, 1x reaction buffer, and 1.5 mM MgCl₂ (Advance Biotechnologies, Surrey, United Kingdom) was used in a final reaction volume of 100 µl . The following cycling conditions were used: one cycle at 95°C for 1 min followed by 35 cycles of 95°C for 45 s, 55°C for 30 s, and 72°C for 45 s, and a final extension cycle at 72°C for 10 min . PCR products were analyzed by agarose gel electrophoresis to determine yield and purity .

DGGE.
DGGE analysis was performed with a DCode mutation detection system (Bio-Rad, Hemel Hempstead, Hertfordshire, United Kingdom) . Gels of 8% acrylamide (37:1 acrylamide-bisacrylamide) were formed between 20 and 70% denaturant, with 100% denaturant defined as 7 M urea and 40% (vol/vol) formamide (39) . Linear denaturant gradients were made by using a gradient maker (BDH, Lutterworth, Leicestershire, United Kingdom) in a 16-cm gel with a 1-mm gel width . Normally, 300 to 500 ng of PCR product was loaded onto each lane of the gel . Gels were run at 60 V for 16.5 h and maintained at a constant temperature of 60°C in 7 liters of 0.5x TAE buffer (40 mM Tris-acetate and 1 mM EDTA [pH 8.0]) . A set of seven laboratory strains (Agrobacterium rhizogenes, Arthrobacter polychromogenes, Bacillus subtilis, Burkholderia phenazium, Paenibacillus amyloyticus, Pseudomonas fluorescens, and Sphingomonas yanoikuyae) were used to construct a standard marker for DGGE analysis .

Sequencing of DGGE bands.
Selected DGGE bands were excised from the gel with a sterile blade . The PCR products were cloned into the pGEM-T easy vector system I (Promega) following the manufacturer's instructions, and plasmid DNA was extracted from clones (>3) with the Qiagen 8 Ultra Plasmid kit (Qiagen, West Sussex, United Kingdom) . Sequencing reaction mixtures (20-µl volume) with the ABI PRISM BigDye terminator cycle sequence ready reaction kit (Perkin Elmer-Applied Biosystems, Warrington, United Kingdom) and standard PCR sequencing reaction conditions were used according to the manufacturer's recommendations . Products were analyzed on an ABI PRISM 377 DNA cycle sequencer, and all sequences were edited and assembled by using the DNAstar SeqMan II sequence analysis package (Lasergene, Inc., Madison, Wis.) . Sequences were compared to those on the Ribosomal Database Project II (http://rdp.cme.msu.edu/html/) (32) by using sequence match and to EMBL DNA database sequences by using FASTA 3.0 (http://www.ebi.ac.uk/embl/index.html) (56) .

Statistical analysis.
For bacterial and Pseudomonas fungal populations, CFU data from all three depths for all three SSF runs and their replicate SSF columns were used in an analysis of variance . For ATP contents and spore counts, data from all three depths from only two SSF runs and their replicate SSF columns at weeks 2 and 4 were analyzed . The significance of differences among treatment means was determined by using Duncan's multiple range tests and a P value of 0.05 .

The DGGE banding patterns were scored manually, and a binary matrix was made based on the presence (1) or absence (0) of the bands . The binary data representing the banding patterns were used to generate a distance matrix by using the Dice coefficient (42) . The distance matrix was analyzed by using classical multidimensional scaling (11) and Sammon's nonlinear mapping (52) and used to construct a dendrogram by cluster analysis with the UPGMA (unweighted pair group mean average) linkage method (54) . Multidimensional scaling and Sammon's nonlinear mapping gave representations of the interband distance in a two-dimensional space, approximating the corresponding interband distance in the original space . The maps were used to visualize and interpret relative spatial and temporal changes in the bacterial community structure between groups . More-complex self-organized maps can also be used for interpreting the data generated from rapid profiling techniques, such as DGGE, and are particularly useful for large and variable data sets (15) . However, for the visualization of relatively small and structured data sets, such as those presented in this paper, the more-conventional mapping and clustering techniques were preferred . The statistical significance of differences in structure between groups, based on the original distance matrix, were assessed by comparing the mean distances between groups to the within-group distance distribution . Between-group distances were considered significant at the 5% level if they were greater than the mean within-group distance plus t_0.05[n] standard deviations of the within-group distance, where t_0.05[n] is the appropriate critical value of the t distribution on n degrees of freedom .

Nucleotide sequence accession number.
All sequences from excised DGGE bands have been deposited in the EMBL database under accession no . AJ536093 to AJ536099 .

 
Dynamics of the total microbial population.
The sizes of the bacterial populations of the storage tank and the outflow water samples during the study were significantly different from each other . The water from the storage tank showed 3 to 4 times higher bacterial populations (900 CFU of pseudomonads ml^-1 and 2.9 x 10⁵ CFU of total bacteria ml^-1) compared to the outflow water (300 CFU of pseudomonads ml^-1 and 7.2 x 10⁴ CFU of total bacteria ml^-1) . A significant difference in the total fungal population between the storage tank and outflow water samples was also observed . Fungi were detected in the water from the storage tank at a level of 3 x 10³ CFU ml^-1, and there was a 30-fold reduction in this fungal population in the outflow water to 100 CFU ml^-1 .

The sand had a low number of microorganisms as it was loaded into the SSF columns (8.8 x 10⁴ CFU of bacteria g^-1, 233 CFU of pseudomonads g^-1, and 123 CFU of fungi g^-1) . It was rapidly colonized by microorganisms, by the end of the first week, the sand appeared to be fully colonized, and these populations remained relatively constant during the 4-week study (Fig . 2) . The bacterial population in the top of the filter was significantly higher than that of the middle and bottom layers at week 1 (Fig . 2A) . This difference was small and not apparent at week 2 or 4 and may simply reflect the natural variation seen within the system . The number of colonies growing on Pseudomonas selective media was lower than that recorded on 1/10-strength TSA and showed significant differences between layers at weeks 1 and 2 but not at week 4 . Lower numbers of bacteria (with the same population pattern described previously for TSA) were observed by direct counting of bacteria with acridine orange staining (data not shown) . It is possible that most of the microbial communities present within the SSF are tightly attached to the sand grains within biofilms and that lower numbers exist as free-living bacteria, and these are counted . Previous studies, where scanning electron microscopy was used to examine the sand grains in a SSF, showed the existence of discrete colonies of bacteria on the sand grain surface (6, 22) .

 
The fungal population was significantly higher in the top of the filter compared to the middle and bottom samples on all weeks by plate counting (Fig . 2C) . No CFU of fungi were detected in the bottom layer during the first week in all three SSF runs . After 4 weeks, the fungal population in the top layer (1.4 x 10⁵ CFU g^-1 [dry weight]) of the filter was at least 40 times and 130 times higher than that observed at the middle (3.6 x 10³ CFU g^-1 [dry weight]) and in the bottom (1 x 10³ CFU g^-1 [dry weight]), respectively . Similar results were observed from direct counts of the fungal spores at different depths of the SSF columns (Fig . 2D) . The number of spores in the middle (3.5 x 10⁴ spores g^-1 [dry weight]) and bottom layers (2.1 x 10⁴ spores g^-1 [dry weight]) were significantly lower than those in the top layer (1.6 x 10⁶ spores g^-1 [dry weight]) .

These results show that large microbial populations develop in the top of the sand filter within 1 week of setting it up, and Pseudomonas species seem to dominate the culturable bacterial population . This higher population at the top of the filters may be related to the higher concentration of particulate organic carbon known to accumulate in the upper layers of SSFs (13) .

ATP content.
The ATP contents at different depths in the SSFs were determined as an indicator of microbial activity . Statistically significantly higher levels of ATP were detected at the top of the filter bed (0.1296 mg g^-1 [dry weight]) than in its middle (0.0036 mg g^-1 [dry weight]) or at the bottom (0.0038 mg g^-1 [dry weight]) (Fig . 3) . At each sampling location, no significant differences in the ATP content with time was detected between 2 and 4 weeks . Using these measurements, the top of the sand column had almost 30 times greater ATP levels than the other two layers . ATP has been used as a relevant biochemical indicator for estimation of microbial biomass and metabolic activity in soil for more than 2 decades (28) . If total bacterial and fungal counts on laboratory media are representative of the populations present, then these results would indicate that the cells in the top layer of the SSF have the greatest activity .

 
Since the greatest retention of what little organic matter there is within the reservoir water is likely to be in the top layer of the sand column, it is not surprising that this area has the greatest microbial activity . However, since nonculturable bacteria could dominate the top fraction to a greater extent, these results may just indicate a larger total microbial biomass within this area . Attempts have been made to evaluate microbial populations of SSFs by using methods such as particulate organic carbon, bacterial counts, nitrogen content of volatile solids, acriflavine direct cell count, growth on selective R2A medium, and bacterial respiration levels (13, 19, 21), and our results are in general agreement with these . However, in this study, we have demonstrated that the majority of biological activity occurred in the top layer of the sand column . Biochemical analyses which used BIOLOG (sole-carbon-source utilization pattern) and phospholipid fatty acid analysis of sand filters have also shown that the microbial community present in the top of the sand column utilized the biodegradable dissolved organic carbon more quickly than the communities present in the lower depths (37) . Nevertheless, the phospholipid fatty acid profiles showed that the communities were closely grouped taxonomically, regardless of filter depths (38) .

Bacterial community structure based on DGGE analysis.
Analysis of the V3 region of the 16S rRNA genes within the microbial population by PCR DGGE showed changes in the structure of the bacterial community . PCR amplification of DNA from different extractions, and samples from different replicate columns with different dilutions (1:10, 1:100, and 1:1,000), followed by DGGE separation showed a highly reproducible profile, indicating the accuracy of the molecular technique (data not shown) . Overall, the replicate samples showed a variation in the DGGE banding pattern between 2 to 4% . Major changes in the microbial structure were observed during the first week of the experimental work (Fig . 4) . During the experiment, the number of bands within the DGGE profiles was relatively constant . In week 1, between 31 and 35 bands were detected in the three depths, and with time, the bands tended to remain and increase slightly (faint DGGE bands difficult to detect) but some became more prominent within the profile . By week 4, 41 of the bands were from the top, 33 bands were from the middle, and 31 bands were from the bottom fraction of the sand column, of which only 24 bands were common within the sand column, with between 2 and 5 unique bands for each depth . With time, the top layer of the sand column showed the greatest changes in band intensity and diversity . If it can be assumed that the intensity of the DGGE band represents the relative abundance of a particular species in the population (17, 27), then the bands within the profile represent the dominance of particular bacteria within the system . Bruggemann et al . (7) support the contention that profiles of bacterial communities generated by PCR-based methods are a reasonable estimation of dominant in situ community structure . However, this may not always be the case, and some element of bias is likely (7, 64), as there will be with any method applied to such diverse samples . Since the PCR DGGE profile is obtained directly from the filter sample, it provides one of the best direct measurements of the total bacterial population . The sand used to fill the columns showed a DGGE profile of 24 bands . Some of these bands do not appear after 1 week of the SSF operation and may represent bands derived from DNA remaining from the killed bacterial population . The DGGE banding pattern within the sand itself (before filling the sand columns), the storage tank, and outflow water samples were less complex and different from the total sand column populations . A slightly higher number of bands (5 DGGE bands) was observed in the water from the storage tank compared to the outflow water (Fig . 5) . Similar results were found during the recycling of nutrient solutions through a sand filter in a glasshouse cropping situation (47) . However, when the bands from within the water from the storage tank and original sand samples were compared by addition, bands representing all those seen in the samples from the SSF columns could be detected . This is not surprising since it would be expected that colonization of the filter would be from bacteria present within these samples . The results therefore indicate that a diverse bacterial population develops within the sand column . The banding pattern also indicates little change down the sand column and, hence, relatively little stratification of the populations with depth . This may be expected as continual movement of water down the column will replenish nutrients and oxygen and remove waste products from the system (24) .

 
The DGGE profiles from the sand samples in all depths were different from the profiles taken from bacteria representing the culturable fraction of the bacterial population . The DGGE banding patterns obtained from community DNA from different depths and from culturable bacteria on 0.1 TSA and P1 agar media were compared with samples from the top, middle, and bottom layers at week 4 (Fig . 6) . The DGGE analysis showed great variation in the number of bands depending on the agar medium used for enrichment . Based on the similarity of the denaturing migration distances of the DGGE bands in the gel, most of the bands observed in DGGE banding patterns from SSF were not observed in the DGGE from the culturable bacteria on the media used . The DGGE bands that were strong in intensity from the culturable bacteria showed a faint DGGE band in the SSF profiles, indicating that they represent a minority species within the bacterial population in the sand column . It was estimated that less than 10 to 20% of the DGGE bands were common between the direct and culturable fractions from the SSF column . This indicates that the majority of bacteria grow slowly on the media and do not dominate the culturable biomass or that they may be unculturable on the media used . Four to six major bands were observed from all the depths on 0.1 TSA or P1 medium . The banding pattern on P1 medium was similar at all the depths except for one band missing in the top layer . This indicates that the culturable pseudomonad population from all areas of the sand filter was essentially the same . Greater variation in the DGGE patterns from 0.1 TSA media was observed with some similarity in DGGE banding patterns from the middle and bottom layers . These results confirm that PCR DGGE is a powerful culturability-independent technique for monitoring bacterial populations (25) .

 
Classical multidimensional scaling and Sammon's nonlinear mapping were applied to the complex DGGE banding patterns in order to provide two-dimensional visual representations of overall changes (spatial and temporal) in the structure of the microbial community . The map produced from classical multidimensional scaling proved to be a poor representation of the original distance data (stress = 0.16), whereas Sammon's nonlinear mapping (Fig . 7) proved to be a good representation of the original data (stress = 0.07) . In Fig . 7, replicate samples within an experimental treatment are grouped together, emphasizing the fact that variation between replicate samples within experimental treatment groups proved to be considerably lower than variation between treatment groups . The statistical significance of differences in structure between treatment groups, based on the original distance matrix, showed that the only groups that did not differ at the 5% level were the middle (M1) and bottom (B1) at week 1 . The Sammon map illustrates the proximity of the M1 and B1 groups and shows the other treatment groups as distinct clusters . As a comparative approach, a dendrogram (Fig . 8) was built from the UPGMA algorithm and showed similar clustering to the Sammon map . With time, populations from the sand column changed, indicating shifts within the microbial populations . Clearly, bacterial populations in the top, middle, and bottom of the columns did develop in different groups even though there were dominant bands common to all samples . To our knowledge, this experimental study is the first where DGGE has been applied to analyze spatial and temporal changes in the microbial community in SSFs .

 
DGGE band characterization.
Six dominant bands at different gradient levels were present within the bacterial DGGE profile from all the depths within the sand column at week 4 . These bands were selected for further studies (Fig . 4), as they would provide information on the types of bacteria dominating the sand column at all levels . A comparison of the sequence of bands to those within databases revealed homology to characterized strains . The results are given in Table 2 . Since only a part of the 16S rRNA gene is sequenced (ca . 194 bp), the data are only used as an indicator of the types of bacteria present rather than an accurate affiliation of the band to a species level with phylogenetic-based analysis software . The sequences presented were obtained from clones of extracted bands as a full-length double-stranded sequence could not be obtained from the product, since ca . 50 bp from each end is lost in direct sequencing in both directions, resulting in only 50% coverage of a 200-bp product . To determine if each band represented one product, a number of clones were investigated, and only one sequence was obtained, which is presented . Sequencing of the selected DGGE bands and a comparison to the database indicated that each band had the most similarity to bacterial 16S rRNA genes from groups such as Bacillus megaterium, Cytophaga, Desulfovibrio, Rhodococcus rhodochrous, Sphingomonas, and one environmental clone . Similar sequences to the microorganisms described here have been found in water from sea or lake and sediment environments . Cytophaga spp . are bacteria common in soil and fresh seawater environments and are well known for the production of cellulase, amylase, and chitinase (9, 30, 60) . Rhodococcus ruber is well known for the degradation of a wide range of organic pollutants and can catalyze many useful biotransformations (5, 10, 12, 14) . An additional DGGE PCR band within the sand column had sequence similarity to Legionella cherrii . A detailed study on this group of bacteria in the inflow water, outflow water, and sand column was carried out and is reported separately . Some of these microorganisms might exhibit antagonistic properties against fungal plant pathogens by the production of fungal cell wall degrading enzymes or extracellular compounds that influence the retention or immobilization of the fungal spores in the sand bed (62) . In addition, the enrichment of a pathogen such as Legionella within an SSF system is undesirable, and methods to induce a natural suppressive population without encouraging the growth of this pathogen would be an advantage . The introduction of suppressive bacteria into an SSF may provide good fungal pathogen control and prevent colonization by Legionella species . Already, several microbial antagonists have been tested as biological agents in horticulture crops including species of Bacillus, Pseudomonas, and Trichoderma (16, 31, 33, 43, 59, 68); however, these have had variable success (67, 68) . Bacteria isolated from an SSF, tested for suppressive activity, and then reintroduced into the system may provide the best option to this problem in the long term . Identifying bacterial groups associated with PCR DGGE bands could allow the development of selective media to isolate groups individually from the filter . This information could also be used to directly relate isolates back to the composition within the filter .

 
This study describes the characterization of the performance and microbial composition of SSFs used for the treatment of water for use in the horticultural industry . Although there is still uncertainty concerning the mechanisms involved in SSF functioning against fungal plant pathogens, the results presented here clearly show that fungal zoospores are efficiently removed during passage of water through the SSF and that this is concurrent with the development of a highly active microbial population . This study has also shown that the groups of bacteria that colonize the SSF and are detected by PCR DGGE are maintained after 1 week of operation . The bacteria detected may be associated with the removal of zoospores . Certainly, increasing pressure has been placed on the horticulture industry worldwide to minimize the use of chemicals to control pathogens and to recycle water . Utilization of a naturally suppressive population of microorganisms either directly or by manipulation of the environment in an SSF may provide a more reproducible control method . Further characterization of the bacterial, fungal, and protozoan populations in many different sand columns is desirable .

We thank the nursery and technical staff at HRI and Julie Jones for statistical analysis of the data .

1. Ali-Shtayeh, M . S., and J . D . MacDonald. 1991 . Occurrence of Phytophthora species in irrigation water in the Nablus area (West Bank of Jordan) . Phytopathol . Mediterr . 30:143-150.
2. Ali-Shtayeh, M . S., J . D . MacDonald, and J . Kabashima. 1991 . A method for using commercial ELISA tests to detect zoospores of Phytophthora and Pythium species in irrigation water . Plant Dis . 75:305-311.
3. Berkelmann, B., W . Wohanka, and G . Wolf. 1994 . Characterisation of the bacterial flora in recirculating nutrient solutions of a hydroponic system with rockwool . Acta Hortic . 361:372-381.
4. Bewley, W . F., and W . Buddin. 1921 . On the fungus flora of glasshouse water supplies in relation to plant disease . Ann . Appl . Biol . 8:10-19.
5. Bock, C., R . M . Kroppenstedt, and H . Diekmann. 1996 . Degradation and bioconversion of aliphatic and aromatic hydrocarbons by Rhodococcus ruber 219 . Appl . Microb . Biotechnol . 45:408-410.
6. Brand, T., and W . Wohanka. 2000 . Importance and characterization of the biological component in slow filters . Acta Hortic . 554:313-321.
7. Bruggemann, J., J . Stephen, Y.-J . Chang, S . J . Macnaughton, G . A . Kowalchuk, E . Kline, and D . C . White. 2000 . Competitive PCR-DGGE analysis of bacterial mixtures an internal standard and an appraisal of template enumeration accuracy . J . Microb . Methods 40:111-123.
8. Burman, N . P. 1978 . Slow sand filtration . H₂O 11:348-351.
9. Casamayor, E . O., H . Schäfer, L . Bañeras, C . Pedrós-Alió, and G . Muyzer. 2000 . Identification of and spatio-temporal differences between microbial assemblages from two neighboring sulfurous lakes: comparison by microscopy and denaturing gradient gel electrophoresis . Appl . Environ . Microbiol . 66:499-508.
10. Chaineau, C . H., J . Morel, J . Dupont, E . Bury, and J . Oudot. 1999 . Comparison of the fuel oil biodegradation potential of hydrocarbon-assimilating microorganisms isolated from a temperate agricultural soil . Sci . Total Environ . 277:237-247.
11. Chatfield, C., and A . J . Collins. 1996 . Introduction to multivariate analysis . Chapman and Hall, London, United Kingdom.
12. Chauvaux, S., F . Chevalier, C . Le Dantec, F . Fayolle, I . Miras, F . Kunst, and P . Beguin. 2001 . Cloning of a genetically unstable cytochrome P-450 gene cluster involved in degradation of the pollutant ethyl tert-butyl ether by Rhodococcus ruber . J . Bacteriol . 183:6551-6557.
13. Collins, M . R., T . T . Eighmy, J . M . Fenstermacher, and S . K . Spanos. 1992 . Removing natural organic matter by conventional slow sand filtration . J . Am . Water Works Assoc . 84:80-90.
14. Colquhoum, J . A., S . C . Heald, L . Li, J . Tamaoka, C . Kato, K . Horikoshi, and A . T . Bull. 1998 . Taxonomy and biotransformation activities of some deep-sea actinomycetes . Extremophiles 2:269-277.
15. Dollhopf, S . L., S . A . Hashsham, and J . M . Tiedje. 2001 . Interpreting 16S rDNA T-RFLP data: application of self-organizing maps and principal component analysis to describe community dynamics and convergence . Microb . Ecol . 42:495-505.
16. Duijff, B . J., P . A . H . M . Bakker, and B . Schippers. 1994 . Suppression of Fusarium-wilt of carnation by Pseudomonas putida WCS358 at different levels of disease incidence and iron availability . Biocontrol Sci . Technol . 4:279-288.
17. Duineveld, B., A . S . Rosado, J . D . van Elsas, and J . A . van Veen. 1998 . Analysis of the dynamics of bacterial communities in the rhizosphere of the chrysanthemum via denaturing gradient gel electrophoresis and substrate utilization patterns . Appl . Environ . Microbiol . 64:4950-4957.
18. Duncan, A. 1988 . The ecology of slow sand filters, p . 163-180 . In H . J . D . Graham (ed.), Slow sand filtration . Horwood, Chichester, United Kingdom.
19. Eighmy, T . T., M . R . Collins, S . K . Spanos, and J . Fenstermacher. 1992 . Microbial populations, activities and carbon metabolism in slow sand filters . Water Res . 26:1319-1328.
20. Ellis, K . V. 1986 . Slow sand filtration . Crit . Rev . Environ . Control 15:315-354.
21. Ellis, K . V., and M . E . Aydin. 1995 . Penetration of solids and biological activity into slow sand filters . Water Res . 29:1333-1341.
22. Esch, P., and A . Nehrkorn. 1988 . Rasterelektronenmikroskopische untersuchung über die mikrobielle besiedlung von langsamsandfiltern . Zentbl . Bakteriol . Hyg . B 185:569-579.
23. Goldberg, N . P., M . E . Stanghellini, and S . L . Rasmussen. 1992 . Filtration as a method of controlling Pythium root rot of hydroponically grown cucumbers . Plant Dis . 76:777-779.
24. Haarhoff, J., and J . L . Cleasby. 1991 . Biological and physical mechanisms in slow sand filtration, p . 19-68 . In G . S . Logsdon (ed.), Slow sand filtration, vol . 1 . American Society of Civil Engineers, New York, N.Y.
25. Hugenholtz, P., B . M . Goebel, and N . R . Pace. 1998 . Impact of culture-independent studies on the emerging phylogenetic view of bacterial diversity . J . Bacteriol . 180:4765-4774.
26. Huisman, L., and W . E . Wood. 1974 . Slow sand filtration . World Health Organisation, Geneva, Switzerland.
27. Jackson, C . R., E . E . Roden, and P . F . Churchill. 1998 . Changes in bacterial species composition in enrichment cultures with various dilutions of inoculum as monitored by denaturing gradient gel electrophoresis . Appl . Environ . Microbiol . 64:5046-5048.
28. Karl, D . M. 1980 . Cellular nucleotide measurements and applications in microbial ecology . Microbiol . Rev . 44:739-796.
29. Katoh, K., and I . Kunio. 1983 . New selective media for Pseudomonas strains producing fluorescent pigment . Soil Sci . Plant Nutr . 29:525-532.
30. Laetitia, B., H . Schäfer, F . Joux, C . Courties, G . Muyzer, and P . Lebaron. 2000 . Genetic diversity of total, active and culturable marine bacteria in coastal seawater . Aquat . Microb . Ecol . 23:1-11.
31. Leeman, M., F . M . DenOuden, J . A . van Pelt, C . Cornelissen, A . Matamala-Garros, P . A . H . M . Bakker, and B . Schippers. 1996 . Suppression of Fusarium wilt of radish by co-inoculation of fluorescent Pseudomonas spp . and root-colonizing fungi . Eur . J . Plant Pathol . 102:21-31.
32. Maidak, B . L., J . R . Cole, T . G . Lilburn, C . T . Parker, Jr., P . R . Saxman, R . J . Farri, G . M . Garrity, G . J . Olsen, T . M . Schmidt, and J . M . Tiedje. 2001 . The ribosomal database project (The RDP-II) . Nucleic Acids Res . 29:173-174.
33. McCullagh, M., R . Utkhede, J . G . Menzies, Z . K . Punja, and T . C . Paulitz. 1996 . Evaluation of plant growth-promoting rhizobacteria for biological control of Pythium root rot of cucumber grown in rockwool and effects on yield . Eur . J . Plant Pathol . 102:747-755.
34. McIntosh, D . L. 1966 . The occurrence of Phytophthora spp . in irrigation systems in British Columbia . Can . J . Bot . 44:1591-1598.
35. McNair, R . D., R . C . Sims, D . L . Sorensen, and M . Hulbert. 1987 . Schmutzdecke characterization of clinoptilolite-amended slow sand filtration . J . Am . Water Works Assoc . 79:74-81.
36. McPherson, G . M., M . R . Harriman, and D . Pattison. 1995 . The potential for spread of root diseases in recirculating hydroponic systems and their control with disinfection . Med . Fac . Landbouww . Univ . Gent 60/2b:371-379.
37. Moll, D . M., and R . S . Summers. 1996 . Performance, biomass and community structure profiles of biological rapid media filters, p . 159-166 . In N . Graham and R . Collins (ed.) . Advances in slow sand and alternative biological filtration . John Wiley and Sons, Chichester, United Kingdom.
38. Moll, D . M., R . S . Summers, and A . Breen. 1998 . Microbial characterization of biological filters used for drinking water treatment . Appl . Environ . Microbiol . 64:2755-2759.
39. Muyzer, G., E . C . De Waal, and A . G . Uitterlinden. 1993 . Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes encoding for 16S rRNA . Appl . Environ . Microbiol . 59:695-700.
40. Myers, R . M., S . G . Fisher, L . S . Lerman, and T . Maniatis. 1985 . Nearly all single base substitution in DNA fragments joined to a GC-clamp can be detected by denaturing gradient gel electrophoresis . Nucleic Acids Res . 13:3131-3145.
41. Myers, R . M., T . Maniatis, and L . S . Lerman. 1987 . Detection and localization of single base changes by denaturing gradient gel electrophoresis . Methods Enzymol . 155:501-527.
42. Nei, M., and W . H . Li. 1979 . Mathematical model for studying genetic variation in terms of restriction endonucleases . Proc . Natl . Acad . Sci . USA 76:5269-5273.
43. Ongena, M., F . Daayf, P . Jacques, P . Thonart, N . Benhamou, T . C . Paulitz, P . Cornélis, N . Koedam, and R . R . Bélanger. 1999 . Protection of cucumber against Pythium root rot by fluorescent pseudomonads: predominant role of induced resistance over siderophores and antibiosis . Plant Pathol . 48:66-76.
44. Pettitt, T . R., and G . F . Pegg. 1991 . The quantitative estimation of Phytophthora cactorum in infected strawberry tissue . Mycol . Res . 95:233-238.
45. Pettitt, T . R., A . R . Finlay, M . A . Scott, and E . M . Davies. 1998 . Development of a system simulating commercial production conditions for assessing the potential spread of Phytophthora cryptogea root rot of hardy nursery stock in recirculating irrigation water . Ann . Appl . Biol . 132:61-75.
46. Pittis, J . E., and J . Colhoun. 1984 . Isolation and identification of pythiaceous fungi from irrigation water and their pathogenicity to Antirrhinum, tomato and Chamaecyparis lawsoniana. Phytopathol . Z . 110:301-318.
47. Postma, J., E . A . van Os, and G . Kritzma. 1999 . Prevention of root diseases in closed soilless growing systems by microbial optimization . Med . Fac . Landbouww . Univ . Gent 64:431-440.
48. Pottorff, L . P., and K . L . Panter. 1997 . Survey of Pythium and Phytophthora spp . in irrigation water used by Colorado commercial greenhouses . Hortic . Technol . 7:153-155.
49. Roszak, D . B., and R . R . Colwell. 1987 . Survival strategies of bacteria in the natural environment . Microbiol . Rev . 51:365-379.
50. Runia, W . T. 1994 . Elimination of root-infection pathogens in recirculation water from closed cultivation systems by ultra-violet radiation . Acta Hortic . 361:361-371.
51. Runia, W . T. 1995 . A review of possibilities for disinfection of recirculating water from soilless culture . Acta Hortic . 382:221-229.
52. Sammon, J . W. 1969 . A nonlinear mapping for data structure analysis . IEEE Trans . Comput . C 18:401-409.
53. Shokes, F . M., and S . M . McCarter. 1979 . Occurrence, dissemination, and survival of plant pathogens in surface irrigation ponds in southern Georgia . Phytopathology 69:510-516.
54. Sneath, P . H . A., and R . R . Sokal. 1973 . Numerical taxonomy, p . 230-234 . W . H . Freeman and Company, San Francisco, Calif.
55. Stanghellini, M . E., L . J . Stowell, and M . L . Bates. 1984 . Control of root rot of spinach caused by Pythium aphanidermatum in a recirculating hydroponic system by ultraviolet irradiation . Plant Dis . 68:1075-1076.
56. Stoesser, G., W . Barker, A . van den Broek, E . Camon, M . Garcia-Pastor, C . Kanz, T . Kulikova, R . Leinonen, Q . Lin, V . Lombard, R . Lopez, N . Redaschi, P . Stoehr, M . A . Tuli, K . Tzouvara, and R . Vaughan. 2002 . The EMBL nucleotide sequence database . Nucleic Acids Res . 30:21-26.
57. Strong, S . S., B . K . Behe, C . F . Deeke, K . L . Bowen, and G . J . Keever. 1997 . Cultivar and spacing effects on transmission of Phytophthora parasitica in an ebb-and-flow subirrigation system . Plant Dis . 81:89-95.
58. Thomson, S . V., and R . M . Allen. 1974 . Occurrence of Phytophthora species and other potential plant pathogens in recycled irrigation water . Plant Dis . Rep . 58:945-949.
59. Utkhede, R . S., C . A . Koch, and J . G . Menzies. 1999 . Rhizobacterial growth and yield promotion of cucumber plants inoculated with Pythium aphanidermatum . Can . J . Plant Pathol . 21:265-271.
60. van Hannen, E., G . Zwart, M . P . van Agterveld, H . J . Gons, J . Ebert, and H . Laanbroek. 1999 . Changes in bacterial and eukaryotic community structure after mass lysis of filamentous cyanobacteria associated with viruses . Appl . Environ . Microbiol . 65:795-801.
61. van Os, E . A., J . J . Amsing, A . J . van Kuik, and H . Willers. 1999 . Slow sand filtration: a potential method for elimination of pathogens and nematodes in recirculating nutrient solution from glasshouse-grown crops . Acta Hortic . 481:519-526.
62. van Os, E . A., M . A . Bruins, J . van Buuren, D . J . van der Veer, and H . Willers. 1997 . Physical and chemical measurements in slow sand filters to disinfect recirculating nutrient solutions, p . 313-327 . In Proceedings of the Ninth International Congress on Soilless Culture, St . Helier, Jersey, Channel Islands, 12 to 19 April 1996.
63. Visscher, J . T., R . Paramasivam, A . Raman, and H . A . Heijnen. 1987 . Slow sand filtration for community water supply . Planning, design, construction and maintenance . Technical paper no . 24 . International Reference Centre for Community Water Supply and Sanitation, The Hague, The Netherlands.
64. Watanabe, K., Y . Kodama, and S . Harayama. 2001 . Design and evaluation of PCR primers to amplify bacterial 16S ribosomal DNA fragments used for community fingerprinting . J . Microbiol . Methods 44:253-262.
65. Weber-Shirk, M . L., and R . Dick. 1997 . Biological mechanisms in slow sand filters . J . Am . Water Works Assoc . 89:72-83.
66. Weber-Shirk, M . L., and R . Dick. 1997 . Physical-chemical mechanisms in slow sand filters . J . Am . Water Works Assoc . 89:87-100.
67. Whipps, J . M. 2001 . Microbial interactions and biocontrol in the rhizosphere . J . Exp . Bot . 52:487-511.
68. Whipps, J . M., and K . G . Davies. 2000 . Success in biological control of plant pathogens and nematodes by microorganisms, p . 231-269 . In G . Gurr and S . Wratten (ed.), Biological control: measures of success . Klumer Academic Publishers, The Netherlands.
69. Wohanka, W., and M . Helle. 1997 . Suitability of various filter media for slow filtration, p . 551-557 . In Proceedings of the Ninth International Congress on Soilless Culture, St . Helier, Jersey, Channel Islands, 12 to 19 April 1996.
70. Wohanka, W., H . Luebke, H . Ahlers, and M . Luebke. 1999 . Disinfection of recirculating nutrient solutions by slow sand filtration . Acta Hortic . 481:539-544.

Free Online Full-text Article

What Is Staphylococcus Aureus?, What Is Salmonella?, What Is Antibiotic?, What Is Molecular Biology?, What Is Listeria Monocytogenes?, n, Microbes, e, Bacterium, r, Microbiology, r, Microorganism, a, Bacteria, s, Fermentations, r, Agrobacterium, r, Gram negative, r, Salmonella, i, Antimicrobials, s, Listeriosis, c, Pseudomonas aeruginosa, s, Microorganisms, c, Yeasts, i, Microflora, n, Yeasts, s, Bactericidal, e, Escherichia coli, a, Burkholderia, e, Pichia, n, Escherichia coli, c, Escherichia coli, c, Denitrificans, a, Escherichia coli, r, Microorganisms, o, Staphylococcus aureus




 

© 2005 Transgalactic Ltd (manufacturer of Bioscreen C software) | Privacy Statement | P.O. Box 1393, 00101 Helsinki, Finland, phone: +358 9 85172920, fax: +358 9 8749481, e-mail: microbiology@bionewsonline.com
 

Last modified: May 25, 2005