Distribution of aerobic motile and non-motile bacteria within the capillary fringe of silica sand

water research 44 (2010) 1279–1287

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/watres

Distribution of aerobic motile and non-motile bacteria within the capillary fringe of silica sand Daniel Jost, Josef Winter*, Claudia Gallert Universita¨t Karlsruhe (KIT), Institut fu¨r Ingenieurbiologie und Biotechnologie des Abwassers, Am Fasanengarten Geb. 50.31, 76131 Karlsruhe, Germany

article info

abstract

Article history:

Retention of bacterial cells as ‘‘particles’’ by silica sand during formation of a capillary

Received 23 May 2009

fringe (CF) and the influence of motility was examined with motile Pseudomonas putida and

Received in revised form

non-motile Corynebacterium glutamicum suspensions in the absence of nutrients. The

16 December 2009

fractional retention of C. glutamicum cells at all regions of the CF was higher than for

Accepted 6 January 2010

P. putida cells, most probably due to the motility of P. putida. Only about 5% of P. putida cells

Available online 14 January 2010

and almost no C. glutamicum cells reached the upper end of a CF of 10 cm height.

Keywords:

a CF in silica sand fractions of 355–710 mm and 710–1000 mm respectively, was finished after

Capillary fringe

about 6 h. Growth of cells proceeded for about 6 days. P. putida formed a biofilm on silica

Pseudomonas putida

grains, whereas no attachment of C. glutamicum on silica sand occurred. Relative cell

Corynebacterium glutamicum

densities of C. glutamicum on the bottom and in the upper regions of the CF were always

With cell suspensions of P. putida and C. glutamicum in nutrient broth the development of

Distribution

lower than those of P. putida and were also lower than those reached in suspended cultures

Motility

with the same medium. In coarse sand the motile P. putida cells reached significantly

Fluorescein diacetate

higher cell densities in upper CF regions than in fine sand. Growth of C. glutamicum in the CF apparently was slower and a higher proportion of the energy was required for maintenance. Whereas cell densities of P. putida, in CFs of both sand fractions, varied less than one order of magnitude, those of C. glutamicum varied in a wider range from the basis to the top of the CF. Analyses of the esterase activity of P. putida and C. glutamicum with fluorescein diacetate (FDA) revealed that the cells in higher CF regions were significantly more active than those at the bottom of the CF. Furthermore, a significant correlation (r ¼ 0.66, p < 0.01) between cells ml1 and the FDA conversion to fluorescein was found. ª 2010 Elsevier Ltd. All rights reserved.

1.

Introduction

The capillary fringe (CF) is a highly variable, commonly oligotrophic natural ecosystem at the transition of vadose zone and groundwater. Structure and extension of CFs are mainly determined by the grain size distribution, wetting of mineral surfaces and surface tension of the aquatic solution (Ronen

et al., 1997, 2000). These determine transport phenomena of gases (e.g. Affek et al., 1998) or solutes and colloids (e.g. Mc Carthy and Johnson, 1993; Abit et al., 2008) through the CF in all directions. Berkowitz et al. (2004) reported that the CF significantly affected water flow and chemical transport from the vadose zone into groundwater, whereas Silliman et al. (2002) demonstrated in laboratory experiments with porous media

* Corresponding author. Tel.: þ49 721 608 2297; fax: þ49 721 608 7704. E-mail address: [email protected] (J. Winter). 0043-1354/$ – see front matter ª 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2010.01.001

1280

water research 44 (2010) 1279–1287

that water flow and solute transport occurred regularly in the CF in vertical and horizontal direction. Although many soil scientists and hydrologists have investigated the CF of the vadose zone, the complex interaction of geophysical, geochemical, hydrological and in particular microbiological parameters, such as particle transport or movement and growth of bacteria in the CF has been reviewed by Holden and Fierer (2005), but is not well understood. For investigation of vertical transport of bacteria in water through unsaturated porous media bacterial suspensions were trickled through soil columns (Corapcioglu and Haridas, 1984, 1985; Scha¨fer et al., 1998a,b; Jewett et al., 1999; Hua et al., 2003; Gargiulo et al., 2007), whereas Trevors et al. (1994) inoculated soil with bacteria and investigated the effect of trickling water on their distribution. Long distance transport of bacteria in sand columns filled with quartz sand under saturated conditions was investigated by Lutterodt et al. (2009). Results of their study indicated a dependency of transport distance from motility, surface hydrophobicity, outer surface potential (Gram negative bacteria) and sticking efficiency. Bacterial transport from the vadose zone into the groundwater is a major concern with regard to public health (Hua et al., 2003; Paul et al., 2004; Xu et al., 2007). Vice versa, migration of Escherichia coli JM109 from the groundwater through the CF into the unsaturated underground was also demonstrated, using a box with porous medium and horizontal water flow and thus simulating the CF at groundwater flow conditions (Dunn et al., 2005). In the studies of Rockhold et al. (2007) a decrease of height and water saturation of a CF in translucent quartz sand by 7–9%, due to colonization by Pseudomonas fluorescens was observed. In own experiments with Hele-Shaw Cells bacteria were transported by capillary forces into the CF of silica sand, but without a superimposed horizontal water flow. After transport by capillary forces, bacteria need nutrients for growth and biofilm formation in the different zones of the CF. Adsorption coefficients have presumably a major influence on the distribution of bacteria within the sand grain structure of a CF (Camper et al., 1993; Chen et al., 2003). A negative influence on chemotaxis of motile P. fluorescens cells during starvation at low nutrient supply was described by Singh and Arora (2001). Nutrient supply by retention of colloids at air–water–solid interfaces and in zones of immobile water for bacteria to overcome starving conditions was observed by Morales et al. (2009). The aim of our work was to determine the moisture profile in CFs with different grain sizes, the distribution of aerobic motile and non-motile bacteria during formation of CFs and cell density and activity development with time after the CF was formed. A detailed understanding of activities within the CF could help to improve bioremediation methods and biological filter constructions.

2.

Materials and methods

2.1.

Bacterial strains and media

Two typical soil bacteria, Pseudomonas putida (DSM 291, type strain) and Corynebacterium glutamicum (DSM, 20300, Kinoshita et al., 1958) were obtained from Deutsche Sammlung von

Mikroorganismen und Zellkulturen (DSMZ GmbH 2004) and were used in this study. The motile P. putida was cultured in LB medium that contained in g L1: NaCl, 10; yeast extract, 5 and tryptone, 10; pH 7.0–7.2. The non-motile C. glutamicum was cultured in medium 53 of DSMZ, containing in g L1: NaCl, 5; yeast extract, 5; glucose, 5; casein peptone, 10; pH 7.2–7.4. Both strains were cultured in 100 ml cotton-plugged Erlenmeyer flasks, containing 40 ml medium, at 27  C and 110 U/min on a rotary shaker. After an incubation time of one day a cell density of >109 cells ml1 (optical density > 3.0) was obtained. Grown-up culture suspensions were either immediately used for experiments or were stored at 6  C in a cold room as stock cultures. Stock cultures were reactivated in fresh medium every week. Cells of P. putida were 2–3 mm long and 1 mm in diameter, those of C. glutamicum were 2–5 mm long and 2 mm in diameter. Thus, cell size and shape should not have a significant effect on transport processes within the capillary fringe (CF) (Fontes et al., 1991; Gannon et al., 1991). For CF experiments 50 ml of sterile medium (LB medium for P. putida and medium 53 for C. glutamicum) were diluted with 50 ml deionized and autoclaved water, inoculated with 1–3 ml of one of the bacterial stock cultures to reach a cell density of 5–6  107 cells ml1 and filled into the sterilized aluminum trays of 21 cm (length)  2.5 cm (width)  3 cm (depth) of HeleShaw cells (see Section 2.2). To avoid sedimentation of nonmotile C. glutamicum cells during the formation of a CF the suspensions were stirred, whereas no stirring was necessary of P. putida suspensions due to their motility. Concerning adsorption it could be assumed that P. putida, known as a strong biofilm forming microorganism (e.g. Klausen et al., 2006) and then being no longer motile, has a higher affinity to attach on the sand grains than C. glutamicum, due to physical/chemical interactions of bacterial and sand surfaces.

2.2.

Experimental procedure and sampling

For CF experiments, Hele-Shaw Cells were assembled with two 150  150  3 mm glass plates. Silicone-greased glass Tspacer on each side guaranteed 2 mm distance. At the bottom of the glass plates a stainless steel reticule (0.25 mm mesh size) was inserted as a carrier layer for the sand (Fig. 1 A). Glass plates and spacers were decontaminated with 95% ethanol (Merck, Germany). After assembly the Hele-Shaw Cells were heat sterilized at 180  C for 3 h. Thereafter they were filled with heat sterilized silica sand that contained less than 1.6 mg TOC per 100 g sand Two grain fractions of the silica sand were used: Fine sand with particles ranging from 355 to 710 mm in size and sand with particles ranging from 710 to 1000 mm in size. Then the Hele-Shaw Cells were inserted about 1 cm below liquid level into trays, which contained 100 ml of either 0.9% NaCl solution or diluted nutrient solution inoculated with P. putida or C. glutamicum cultures, respectively. To avoid contamination during the formation of the CF the trays were covered with an aluminum foil. The bacterial suspension in the trays was removed after 24 h and replaced by 100 ml of 0.9% NaCl solution. The development of CFs took about 6 h. In Hele-Shaw Cells with both grain sizes the CF spanned from minimally 9 cm (Fig. 1 B) to maximally 10 cm in different experiments. The behavior of microorganisms in CFs was checked for up to 6 days.

water research 44 (2010) 1279–1287

1281

Fig. 1 – Experimental setup: A: Hele-Shaw glass cell (15 3 15 cm) filled with silica sand (Ø: 355–710 mm), B: Sample regions at a disassembled cell with a fully established capillary fringe.

At the end of each test the glass plates were carefully disassembled so that the sand structure and the CF were not destroyed. Sand samples (1 cm2 area, z0.2 cm3, Fig. 1 B) from different heights of the CF were then transferred into 2 ml Eppendorf cups. The sample regions were defined for each test depending on the maximum height of the CF and on different water saturation: One sample was taken at the bottom of the CF at 1–2 cm height (>98% water saturation), one at 6–7 cm height (around 70% water saturation) and another sample at 8–9 cm (upper end of CF, around 30% water saturation; Fig. 2). Then sterile 0.9% NaCl solution was added to the sand samples to get a liquid volume of 1 ml in each cup. The bacteria were detached from the sand by vortexing for 15 s on an IK MS1mini shaker (Wilmington, USA). After sand sedimentation and withdrawal of the suspension, another portion of saline solution was added and the suspension procedure repeated to test the efficiency of bacterial removal from the sand.

2.3.

Analyses and methods

Water saturation at different heights of the CFs was determined gravimetrically by scratching off 1 cm layers of the CF

Fig. 2 – Water saturation curves of capillary fringes with sand grain fractions of 355–710 mm and 710–1000 mm (values were calculated after drying the sand).

in Hele-Shaw Cells over the total width of the cell and determining the weight loss after drying at 105  C. NaCl at 0.9% concentration did not influence the water saturation (data not shown). The porosity of the sand was measured in two steps: First the density was determined according to DIN 18 125-1 (DIN, 2009) in a 100 ml glass cylinder filled partially with water. After addition of a pre-weighed amount of sand the displaced water volume was read. Then the pore volume of dry sand was determined by substraction of the calculated sand volume (from density measurement) from the real bulk volume in the vessel. Different grain size fractions were obtained by using 3 stainless steel sieves (Retsch, Haan, Germany) with mesh sizes 355 mm, 710 mm and 1 mm. Total bacterial cell numbers from sand samples of capillary fringes in different heights were determined by counting cells after suspension with a phase contrast microscope (Zeiss Axioskop, Oberkochen, Germany) according to Taylor et al. (2002). To obtain the total bacterial cell count at a certain sampling height the sand was washed twice and all volumes were united. Counts in the united volumes were standardized to 1 ml pore water. After washing, the sand was subjected to electron microscopy for residual cells. Electron microphotographs were taken with a reflection electron microscope Type LEO Gemini 1530 (Zeiss. Oberkochen, Germany). Samples were air-dried over night and then dry silica grains were platin-sputtered. For REM images, the resolution obtained is commonly about 1 nm (Cowley and Liu, 1993). For bacterial detection on rough sand grains this is probably the best method. Fluorescein diacetate (FDA) conversion was used for measurement of esterase activity. Twenty ml of an FDA stock solution (20 mg FDA in 10 ml acetone; Sigma–Aldrich, Taufkirchen, Germany) was added to the 1 ml-sample suspensions of each CF height to reach a final concentration of 40 mg FDA per liter. The suspensions were incubated for 30 min at 30  C and the reaction was stopped by addition of 500 ml acetone. Esterases hydrolyzed the FDA and green fluorescent fluorescein was formed (Schnu¨rer and Rosswall, 1982). After centrifugation for 3 min (12000 U/min) the amount of fluorescein

1282

water research 44 (2010) 1279–1287

was determined by measuring the extinction at 490 nm with a photometer and calculation with a calibration curve (Table 1). FDA conversion [%] was calculated by comparing the measured fluorescein concentration with the maximal concentration of 32 mg L1 of introduced FDA. Statistical analysis (One or Two-way ANOVA), curve fittings, regressions and correlations (two-tailed by Pearson) were performed with the statistic program Graph Pad Prism 4.0 (GraphPad Software, Inc., USA).

3.

Results

3.1. Water saturation and distribution of bacteria within the capillary fringe The height of capillary fringes (CFs) in Hele-Shaw Cells with silica sand fractions was clearly visible (Fig. 1 B) and varied for silica sand fractions of 355–710 mm and 710–1000 mm only between 9 and 10 cm. Whereas in the first 5 cm above the water level the water saturation decreased only slightly to about 80% there was a more drastic decrease of the water saturation to finally 30–35% at 9 cm height for both grain fractions (Fig. 2). The transport of resting cells of P. putida and C. glutamicum during the establishment of a CF in Hele-Shaw Cells was investigated by using cell suspensions in 0.9% NaCl solution. Thus, growth of the bacteria during formation of CFs was prevented. When the CFs were formed after 5–6 h only 60% of the initial cell density of P. putida and 40% of C. glutamicum, respectively had reached 2 cm height in the CF, whereas at the upper end of the CFs at 10 cm height only about 5% of the original cell density of P. putida and only <1% of the original cell density of C. glutamicum cells were detected (Fig. 3). This indicated a clear retardation of bacteria, either by their own specific weight or by collision and/or interaction with the silica sand grains. Motility of P. putida cells may have been the reason for relatively more cells of P. putida at all zones of the CF than of C. glutamicum (Fig. 3).

3.2. fringe

Growth and activity of bacteria in the capillary

In Hele-Shaw Cells with the 355–710 mm silica sand fraction CFs were developed with suspensions of P. putida and C. glutamicum in nutrient solutions. Cell densities of P. putida in the saturated zone at 2 cm height of the CF after 1, 3 and 6 days were always slightly lower than in the transition zone at 6 or 9 cm height (Fig. 4). At a height of 6 cm or higher, sufficient oxygen for unlimited respiration and an optimal nutrient supply were apparently available for good growth. Whereas

Table 1 – Calibration curve of fluorescein (l [ 490 nm) with 0.9% NaCl solution. Fluorescein concentration [mg/L] Extinction at 490 nm

0.105 0.21 0.42 0.84 1.7

3.4

6.7

13.4

0.011 0.020 0.041 0.080 0.164 0.308 0.611 1.22

Fig. 3 – Percent recovery of cells (±Standard Error, n [ 2) of P. putida and C. glutamicum, suspended in 0.9% NaCl, at different CF heights with silica sand (Ø: 355–710 mm) after 6 h of test duration, as related to the initial cell number (100% means 6.7 3 107 cells mlL1 for P. putida and 5.1 3 107 for C. glutamicum). * [ p < 0.05, *** [ p < 0.001; Two-way ANOVA with Bonferroni post test.

after 1 and 3 days of growth the cell concentration of P. putida was statistically significant ( p < 0.01) higher in the upper regions of the CF than in the tray and in the saturated zone at 2 cm height, cell concentrations after an incubation time of 6 days were no longer significantly different at all heights of the CF (Fig. 4). After 6 days, cell densities at 2, 6 and 9 cm height were slightly lower than after 3 days. This may have been due to the formation of a shear-force-stable biofilm, embedded in extracellular polymeric substances, that resisted quantitative wash-off (Fig. 5 A). The growth behavior of C. glutamicum in the CF differed from that of P. putida. For C. glutamicum, growth at 2 cm and 9 cm height in the CF was retarded. At 2 cm the oxygen requirement for growth may have been suboptimal, whereas only a few cells reached the CF at 9 cm, requiring a longer incubation to reach a dense suspension of C. glutamicum cells. In the transition zone at 6 cm CF height, C. glutamicum cell densities were significantly higher ( p < 0.01) than in the saturated zone. No biofilm was apparently formed by C. glutamicum (Fig. 5 B). The loosely attached cells could easily be washed off and only a few single cells were remaining in pores and at the surface of silica sand grains. P. putida cells from the CF, taken at 6, 7 and 9 cm height after 3 and 6 days, had a similar esterase activity, whereas the esterase activity was significantly lower in the almost watersaturated zone at 2 cm height of the CF (Fig. 6). The esterase activity of C. glutamicum cells from the water-saturated CF at 2 cm height was slightly lower than that of P. putida cells taken from the same height, whereas the esterase activity of C. glutamicum cells from 6 cm height of the CF was as high as

water research 44 (2010) 1279–1287

1283

Fig. 4 – Mean cell counts mlL1 (±Standard Error, n [ 4) of P. putida and C. glutamicum cells in tray (0) and at different CF heights (2, 6 and 9 cm), for different test durations (1, 3 and 6 days) with silica sand (Ø: 355–710 mm). * [ p < 0.05, ** [ p < 0.01, *** [ p < 0.001; One-way ANOVA (vs mean value from 2 cm respectively). The initial cell concentration was 5– 6 3 107 cells mlL1 for both strains.

that of P. putida cells. At this height both microorganisms were present at similar numbers in the CF during the observation time of 6 days (Fig. 4). The esterase activity of C. glutamicum at the upper end of the CF was much lower than at 6 cm height, where the maximal cell density was found. This corroborated with the lower cell numbers at 9 cm height of the CF and indicated active cells. Furthermore, C. glutamicum grew better in the water-saturated CF than at highly reduced moisture content at the upper end of the CF.

3.3. Influence of the average grain size on population densities in the capillary fringe Cell densities in Hele-Shaw Cells with the two silica grain fractions were higher for P. putida than for C. glutamicum at any height of the CF (Fig. 7). In addition, P. putida seemed to grow to significantly higher cell densities in CFs of the 710–1000 mm grain fraction at the three highest sample regions than in the CF of the 355–710 mm grain fraction. Cell densities in the upper 30% height of the CF (6 cm and higher) of the 710–1000 mm grain fraction reached those of aerated suspended cultures (approximately 3  109 ml1; data not shown). The difference may, however, be insignificant if the biofilm on the surface and in the pores of the smaller grains could not be washed off with the same efficiency as that on the surface and in the pores of the larger silica sand grains, due to the higher surface–volume ratio. For C. glutamicum no such differences were found in CFs of the two silica grain fractions (Fig. 7). In this case the highest final cell densities were found between 40 and 80% water saturation. As observed before (Fig. 4), C. glutamicum reached a significantly lower cell density in the CF with the smaller sand fraction at 30% water saturation or 9 cm height (Fig. 7). Either there is only a lower proportion of the carbon sources of the medium available for respiration and growth, which is unlikely, or the C. glutamicum cells in general required more

energy for maintenance metabolism and thus would not reach the same cell densities as P. putida.

4.

Discussion

4.1. Distribution of motile and non-motile bacteria within the capillary fringe The water saturation in the CF of the smaller silica sand fraction is somewhat higher at the same height than in the CF of the larger sand fraction, but almost identical at the top end. Apparently mainly due to their motility and not directed by nutrient concentrations or chemotactic responses, motile P. putida cells had an advantage to reach the upper end of the CF more numerously than C. glutamicum cells, as judged from the percentage of cells of both organisms that were transported upwards with the capillary water during the formation of the CF (Fig. 3). Since the formation of the CF took maximally 6 h and the initial suction of water was very fast, one might expect about the same cell numbers of motile and non-motile bacteria. However, collision with sand grains, entrapment and adsorption in pores may have led to the observed higher retention of C. glutamicum cells on silica sand than of motile P. putida cells. In addition the higher surface hydrophobicity of C. glutamicum cells, which enhances the adsorption at air–water interfaces (Scha¨fer et al., 1998a), probably had the strongest effect on their vertical retardation above the transition zone. Retention by straining in unsaturated silica sand packed columns was also shown for non-motile Rhodococcus rhodochrous and was more efficient in the finer sand fractions than in coarse sand (Gargiulo et al., 2007). By their motility P. putida cells were apparently more versatile to compensate for entrapment and collisions. Camesano and Logan (1998) introduced motile P. fluorescens P17 into saturated soil columns and found that the fractional retention and the collision efficiency were reduced when the pore velocity was reduced from 120 to

1284

water research 44 (2010) 1279–1287

substances produced by genetically different organisms of the same species (e.g. E. coli isolates from different environmental samples) may significantly influence the transport behavior of bacteria through a sandy soil (Bolster et al., 2009).

4.2. Growth of bacteria during and after formation of a capillary fringe

Fig. 5 – REM photomicrograph (EHT [ 10 kV) from sand grain surfaces. Samples were taken at 6 cm CF height after 6 days from the Hele-Shaw Cell with 355–710 mm silica sand. A: P. putida cells, B: C. glutamicum cells.

0.56 m per day. A reduction of the flow velocity may also occur during formation of a CF in the final stage. A dependency of the sorption efficiency of bacteria on a low water flow rate was also seen by Bengtsson and Lindqvist (1995). Whereas non-motile bacteria had a high adsorption capability on soil grains, motile bacteria were apparently able to avoid sticking to soil grains at low fluid velocity. On the other hand adsorption is dependent on the surface properties. Camper et al. (1993) for instance did experiments with 3 different P. fluorescens strains with glass spheres and found that the motile cells had a higher adsorption rate on glass than the non-motile cells. A more intensive motility of exponentially growing P. fluorescens cells may also have been the reason for the higher collision efficiency observed by Smets et al. (1999). However, despite of some of the factors mentioned above (motility, hydrophobicity, chemotactic behavior) other factors such as electrophoretic motility, cell size and shape, surface charge and extracellular polymeric

After rapid formation of a CF, bacterial movement by water suction ceased and population changes were then mainly dependent on substrate and oxygen supply for growth and metabolism. P. putida was apparently able to use the available substrate faster and more efficiently than C. glutamicum (Fig. 4). In addition P. putida apparently changed its metabolism towards production of EPS and lost its motility. By this it was able to attach tightly to the sand grains in a hydrated matrix (Fig. 5 A), whereas C. glutamicum only was retained in the pores of the sand (Fig. 5 B). This may be consistent with high kinetic adsorption rates that were reported for P. putida (Chen et al., 2003) and a fast progression of biofilm formation (Cunningham et al., 1991). The higher fractional cell densities of P. putida as compared to C. glutamicum above 6 cm of height in the CF (Fig. 3) may be a synergistic effect of motility, adsorption efficiency and chemotactic response to substrates, as found for P. putida F1 by Wang et al. (2008) in sand filters. Porosity of the sand may have led to a better retention of motile and non-motile bacteria. Morales et al. (2009) have shown that smooth grain surfaces retain fewer colloids than rough surfaces under both, saturated and unsaturated conditions. By comparing the distribution of microspheres and of bacteria, Heise and Gust (1999) reported a distribution related mainly to advective flow of the microspheres while the bacteria were spatially more homogeneously distributed. During the formation of a CF in Hele-Shaw Cells the nonmotile C. glutamicum cells were apparently transported by capillary forces into the transition zone and reached there almost the same cell densities as the motile P. putida cells. However, much higher numbers of motile P. putida cells seemed to reach the upper end of the CF.

4.3.

Esterase activity of cells in different zones of the CF

FDA hydrolysis can be used for many research aspects, for instance to test the toxicity of chemicals in soil (Haigh and Rennie, 1994) or as a sensitive estimator of total esterase activity in stream sediments and biofilms (Battin, 1997). In this study it was used to test cell activity in different regions of a CF in silica sand. The test revealed that aerobic bacteria had only a low esterase activity at the basis of the CF with still high water saturation, whereas cells taken from upper regions of a CF (6–9 cm height) after 3 and 6 days had a high esterase activity and converted around 25% of the FDA (Fig. 6). A doubling of the incubation time led to more than 95% FDA conversion (data not shown). These results indicate that P. putida cells in the CF at 2 cm height lost their esterase activity, presumably due to cell death by a change of the conditions towards anaerobiosis. The apparent slight decrease of the cell numbers of P. putida in the upper regions of the CF after 6 days was not consistent with the esterase activity tests. This might be explained by a stronger biofilm

water research 44 (2010) 1279–1287

1285

Fig. 6 – Mean FDA conversion [%] (±Standard Error, n [ 8) of P. putida and C. glutamicum cells at different CF heights after a test duration of 3 and 6 days. * [ p < 0.05, ** [ p < 0.01; One-way ANOVA with Dunnett’s test.

formation of P. putida cells and a reduced re-suspension efficiency. Cell counts of Fig. 4 revealed a relatively high positive correlation with their esterase activity (Pearson r ¼ 0.66, p < 0.001 for 3 days and r ¼ 0.6, p < 0.01 for 6 days), although the cell activity was very low at 2 cm height of the CF. Other researchers, such as Sa´nchez-Monedero et al. (2008) or Aseri and Tarafdar (2006), who used the FDA method as a biological indicator in arid soils, also found high significant positive correlations (r ¼ 0.8–0.96, p < 0.01) between the FDA hydrolysis and microbial biomass. Thus it can be assumed that the FDA method is suitable for cell activity investigations in the CF, even though the coefficient r was not as high as normally expected. This shows that there are many independent growth conditions in CFs, which might influence cell activities. It can be assumed that the transition zone at 6–7 cm height above saturation offers the best conditions for aerobic bacteria such as P. putida.

4.4. Influence of the average grain size and of physical/ chemical properties Fractional cell densities of P. putida and C. glutamicum in CFs of the two sand fractions were compared for the same water saturation (Fig. 7). The highest population density of P. putida was found at the top of the fringe formed in the larger sand fraction. Cell densities in fine sand were apparently lower at

all water saturations (Fig. 7 A), presumably due to a stronger formation of a biofilm on the surface and in the pores of the sand grains, which could not be washed off quantitatively. Additionally in experiments conducted by Dunn et al. (2005) it was shown, that E. coli cells where better transported in coarse sand fractions into the CF from below the water table than in fine sand. C. glutamicum reached maximum cell densities at 50–70% water saturation in the CF of both sand fractions (Fig. 7 B). Less cells were formed at higher water saturation and at a water saturation below 50% towards the top of the CF (Fig. 2), indicating that optimal growth conditions were fairly below the top of the CF. Several factors have been discussed in literature that influence the distribution of microorganisms within a CF. For motile P. putida cells it seems that the grain size has an influence on its distribution across the vertical profile of a CF, especially at lower water contents. Berkowitz et al. (2009) found that tracers migrating from fine medium to coarse medium arrived significantly faster than those which migrated from coarse to fine medium. This indicates that advection could be reduced if the pore diameters get smaller (which occurs in fine sand) and that motile bacteria have an advantages over nonmotile microorganisms if the pore diameters get bigger, especially at such low fluid velocities in the higher CF regions. Attachment of a Lactobacillus strain to sandy soil was for instance attributed to hydrophobicity and electrostatic

Fig. 7 – Mean cell counts mlL1 (±Standard Error, n [ 4) of P. putida and C. glutamicum cells in capillary fringes of different grain size fractions of silica sand after 3 days. * [ p < 0.05, ** [ p < 0.01; Two-way ANOVA with Bonferroni post test.

1286

water research 44 (2010) 1279–1287

interaction (Huysman and Verstraete, 1993), whereas the cell size was found to influence transport of 19 bacterial strains through Kendaia loam most significantly but no correlation of hydrophobicity, net surface charge or even of flagella on the transport behavior was observed (Gannon et al., 1991).

5.

Conclusions

The following conclusions can be drawn from the capillary fringe (CF) studies:  In the silica sand fraction of 355–710 mm grain size the CF reached a height of maximally 10 cm, in the fraction of 710–1000 mm the height was 1–2 cm lower.  The water saturation in the CF decreased exponentially with height to 20–30% at the top.  In the coarse sand fraction the cell concentration of P. putida was higher at all water saturation levels than in the fine sand faction, whereas the cell concentration of C. glutamicum did not differ significantly.  Bacterial cells were mainly transported by capillary forces into the CF: Motile P. putida cells were found in relatively higher numbers in the upper regions of the CF than nonmotile C. glutamicum cells, presumably due to their motility.  At higher water concentrations at the basis of the CF less growth occurred and less bacteria were found. This could be due to limited oxygen supply.  After 3 days P. putida had formed a biofilm on the sand grains, whereas apparently no biofilm was formed by C. glutamicum cells.  Cells were still active in the top regions of the CF after 6 days and much less active at the basis at almost water saturation, as judged from FDA hydrolysis. Activity correlated with the cell concentration.

Acknowledgement This work is part of the DFG research unit DyCap (dynamic capillary fringes) and was financed by DFG, Deutsche Forschungsgemeinschaft, Bonn-Bad Godesberg under FOR 831, Wi 524/18-1. We thank D. Bonefas for excellent technical assistance and P. Pfundstein, Laboratorium fu¨r Elektronenmikroskopie of the University of Karlsruhe for preparing the SEM microphotographs.

references

Abit, S.M., Amoozegar, A., Vepraskas, M.J., Niewoehner, C.P., 2008. Solute transport in the capillary fringe and shallow groundwater: field evaluation. Vadose Zone Journal 7, 890–898. Affek, H.P., Ronen, D., Yakir, D., 1998. Production of CO2 in the capillary fringe of a deep phreatic aquifer. Water Resources Research 34, 989–996. Aseri, G.K., Tarafdar, J.C., 2006. Fluorescein diacetate: a potential biological indicator for arid soils. Arid Land Research and Management 20, 87–99.

Battin, Tom J., 1997. Assessment of fluorescein diacetate hydrolysis as a measure of total esterase activity in natural stream sediment biofilms. The Science of the Total Environment 198, 51–60. Bengtsson, G., Lindqvist, R., 1995. Transport of soil bacteria controlled by density-dependent sorption kinetics. Water Resources Research 31, 1247–1256. Berkowitz, B., Silliman, S.E., Dunn, A.M., 2004. Impact of the capillary fringe on local flow, chemical migration and microbiology. Vadose Zone Journal 3, 534–548. Berkowitz, B., Cortis, A., Dror, I., Scher, H., 2009. Laboratory experiments on dispersive transport across interfaces: the role of flow direction. Water Resources Research 45, W02201. doi:10.1029/2008WR007342. Bolster, C.H., Haznedaroglu, B.Z., Walker, S.L., 2009. Diversity in cell properties and transport behavior among 12 different environmental Escherichia coli isolates. Journal of Environmental Quality 34, 465–472. Camesano, T.A., Logan, B.E., 1998. Influence of fluid velocity and cell concentration on the transport of motile and nonmotile bacteria in porous media. Environmental Science and Technology 32, 1699–1708. Camper, A.K., Hayes, J.T., Sturman, P.J., Jones, W.L., Cunningham, A. B., 1993. Effects of motility and adsorption rate coefficient on transport of bacteria through saturated porous media. Applied and Environmental Microbiology 59, 3455–3462. Chen, G., Rockhold, M., Strevett, K.A., 2003. Equilibrium and kinetic adsorption of bacteria on alluvial sand and surface thermodynamic interpretation. Research in Microbiology 154, 175–181. Corapcioglu, M.Y., Haridas, A., 1984. Transport and fate of microorganisms in porous media: a theoretical investigation. Journal of Hydrology 72, 149–169. Corapcioglu, M.Y., Haridas, A., 1985. Microbial transport in soils and groundwater: a numerical model. Advances in Water Resources 8, 188–200. Cowley, J.M., Liu, J., 1993. Contrast and resolution in REM, SEM and SAM. Surface Science 298, 456–467. Cunningham, A.B., Characklis, W.G., Abedeen, F., Crawford, D., 1991. Influence of biofilm accumulation on porous media hydrodynamics. Environmental Science and Technology 25, 1305–1311. DIN, 2009. Untersuchung von Bodenproben: Bestimmung der Dichte des Bodens – Teil 1, Laborversuche. DIN Deutsches Institut fu¨r Normung e.V., Beuth Verlag GmbH, Berlin. Dunn, A.M., Silliman, S.E., Dhamwichukorn, S., Kulpa, C.F., 2005. Demonstration of microbial transport into the capillary fringe via advection from below the water table. Journal of Hydrology 306, 50–58. Fontes, D.E., Mills, A.L., Hornberger, G.M., Herman, J.S., 1991. Physical and chemical factors influencing transport of microorganisms through porous media. Applied and Environmental Microbiology 57, 2473–2481. Gannon, J.T., Manibal, V.B., Alexander, M., 1991. Relationship between cell surface properties and transport of bacteria through soil. Applied and Environmental Microbiology 57, 190–193. Gargiulo, G., Bradford, S., Simunek, J., Ustohal, P., Vereecken, H., Klumpp, E., 2007. Bacteria transport and deposition under saturated conditions: the role of the matrix grain size and the bacteria surface protein. Journal of Contaminant Hydrology 92, 255–273. Haigh, S.D., Rennie, A.F.K., 1994. Rapid methods to assess the effects of chemicals on microbial activity in soil. Environmental Toxicology and Water Quality 9, 347–354. Huysman, F., Verstraete, W., 1993. Effect of cell surface characteristics on the adhesion of bacteria to soil particles. Biol. Fertil. Soils 16, 21–26.

water research 44 (2010) 1279–1287

Heise, S., Gust, G., 1999. Influence of the physiological status of bacteria on their transport into permeable sediments. Marine Ecology Progress Series 190, 141–153. Holden, P.A., Fierer, N., 2005. Microbial processes in the vadose zone. Vadose Zone Journal 4, 1–21. Hua, J., An, P., Winter, J., Gallert, C., 2003. Elimination of COD, microorganisms and pharmaceuticals from sewage by trickling through sandy soil below leaking sewers. Water Research 37, 4395–4404. Jewett, D.G., Logan, B.E., Arnold, R.G., Bales, R.C., 1999. Transport of Pseudomonas fluorescens strain P17 through quartz sand columns as a function of water content. Journal of Contaminant Hydrology 36, 73–89. Kinoshita, S., Nakayama, K., Akita, S., 1958. Taxonomical study of glutamic acid accumulating bacteria, Micrococcus glutamicus nov. spec. Bulletin of the Agricultural Chemical Society of Japan 22, 176–185. Klausen, M., Gjermansen, M., Kreft, J.U., Tolker-Nielsen, T., 2006. Dynamics of development and dispersal in sessile microbial communities: examples from Pseudomonas aeruginosa and Pseudomonas putida model biofilms. FEMS Microbiology Letters 261, 1–11. Lutterodt, G., Basnet, M., Foppen, J.W.A., Uhlenbrook, S., 2009. The effect of surface characteristics on the transport of multiple Escherichia coli isolates in large scale columns of quartz sand. Water Research 43, 595–604. Mc Carthy, K.A., Johnson, R.L., 1993. Transport of volatile organic compounds across the capillary fringe. Water Resources Research 29, 1675–1683. Morales, V.L., Gao, B., Steenhuis, T.S., 2009. Grain surfaceroughness effects on colloidal retention in vadose zone. Vadose Zone Journal 8, 11–20. Paul, M., Wolf, L., Fund, K., Held, I., Winter, J., Eiswirth, M., Gallert, C., Ho¨tzl, H., 2004. Microbiological condition of urban groundwater in the vicinity of leaky sewer systems. Acta Hydrochimica et Hydrobiologica 32, 351–360. Rockhold, M.L., Yarwood, R.R., Niemet, M.R., Bottomley, P.J., Brockman, F.L., Selker, J.S., 2007. Visualization and modeling of the colonization dynamics of a bioluminescent bacterium in variably saturated, translucent quartz sand. Advances in Water Resources 30, 1593–1607. Ronen, D., Scher, H., Blunt, M., 1997. On the structure and flow processes in the capillary fringe of phreatic aquifers. Transport in Porous Media 28, 159–180.

1287

Ronen, D., Scher, H., Blunt, M., 2000. Field observations of a capillary fringe before and after a rainy season. Journal of Contaminant Hydrology 44, 103–118. Sa´nchez-Monedero, M.A., Mondini, C., Cayuela, M.L., Roig, A., Contin, M., De Nobili, M., 2008. Fluorescein diacetate hydrolysis, respiration and microbial biomass in freshly amended soils. Biology and Fertility of Soils 44, 885–890. Scha¨fer, A., Harms, H., Zehnder, A.J.B., 1998a. Bacterial accumulation at the air–water interface. Environmental Science and Technology 32, 3704–3712. Scha¨fer, A., Ustohal, P., Harms, H., Stauffer, F., Dracos, T., Zehnder, A.J.B., 1998b. Transport of bacteria in unsaturated porous media. Journal of Contaminant Hydrology 33, 149–169. Schnu¨rer, J., Rosswall, T., 1982. Fluorescein diacetate hydrolysis as a measure of total microbial activity in soil and litter. Applied and Environmental Microbiology 43, 1256–1261. Silliman, S.E., Berkowitz, B., Simunek, J., van Genuchten, M.Th., 2002. Fluid flow and solute migration within the capillary fringe. Ground Water 40, 76–84. Singh, T., Arora, D.K., 2001. Motility and chemotactic response of Pseudomonas fluorescens toward chemoattractants present in the exudates of Macrophomina phaseolina. Microbiological Research 156, 343–351. Smets, B.F., Grasso, D., Engwall, M.A., Machinist, B.J., 1999. Surface physicochemical properties of Pseudomonas fluorescens and impact on adhesion and transport through porous media. Colloids and Surfaces B: Biointerfaces 14, 121–139. Taylor, J.P., Wilson, B., Mills, M.S., Burns, R.G., 2002. Comparison of microbial numbers and enzymatic activities in surface soils and subsoils using various techniques. Soil Biology & Biochemistry 34, 387–401. Trevors, J.T., Kuikman, P., van Elsas, J.D., 1994. Release of bacteria into soil: cell numbers and distribution. Journal of Microbiological Methods 19, 247–259. Wang, M., Ford, R.M., Harvey, R.W., 2008. Coupled effect of chemotaxis and growth on microbial distributions in organicamended aquifer sediments: observations from laboratory and field studies. Environmental Science and Technology 42, 3556–3562. Xu, J., Gallert, C., Winter, J., 2007. Multiple antibiotic resistances of Enterococcus isolates from raw or sand filtered sewage. Applied Microbiology and Biotechnology 74, 493–500.