Treatment wetlands (TWs) efficiently remove many pollutants including a several log order reduction of pathogens from influent to effluent; however, there is evidence to suggest that pathogen cells are sequestered in a subsurface wetland and may remain viable months after inoculation. Escherichia coli is a common pathogen in domestic and agricultural wastewater and the O157:H7 strain causes most environmental outbreaks in the United States. To assess attachment of E. coli to the TW rhizosphere, direct measurements of E. coli levels were taken. Experiments were performed in chemostats containing either Teflon nylon as an abiotic control or roots of Carex utriculata or Schoenoplectus acutus. Flow of simulated wastewater through the chemostat was set to maintain a 2 hour residence time. The influent was inoculated with E. coli O157:H7 containing DsRed fluorescent protein. Root samples were excised and analyzed via epifluorescent microscopy. E. coli O157:H7 was detected on the root surface at 2 hours after inoculation, and were visible as single cells. Microcolonies began forming at 24 hours post-inoculation and were detected for up to 1 week post-inoculation. Image analysis determined that the number of microcolonies with >100 cells increased 1 week post-inoculation, confirming that E. coli O157:H7 is capable of growth within biofilms surrounding wetland plant roots.

INTRODUCTION

Treatment wetlands (TWs) have been used as an alternative method for treatment of domestic wastewater for many years. The general structure of TWs has an impermeable membrane layer encompassing the wetland to protect the surrounding subsurface and aquifers from potential wastewater leaching. This bed is filled with some medium such as sand or gravel in which plants are established. Many substrates have been researched previously for their effect on removal rates of both chemical and bacterial contaminants (Baskar et al. 2014). Plant species have also been a point of interest and have been found to greatly influence TW remediation ability (Brisson & Chazarenc 2009; Taylor et al. 2011). Although the effects of plants are varied, there is consensus that increased removal of pollutants occurs in planted wetlands than in unplanted media or filter beds (Vymazal 2011). This effect is likely due in part to the interaction between the plant and the microbial community within the rhizosphere. Plants have been shown to harbor increased microbial density and diversity in close connection to the root systems (Vymazal 2011; Berendsen et al. 2012). They release oxygen and exudates, such as sugars and amino acids, into the rhizosphere, which provide a rich nutrient source to the microbial community as compared to the nutrient-starved soil matrix (Bhattacharyya & Jha 2012). The associated microbial community provides a large amount of the wastewater remediation reactions including nitrification, denitrification and phosphorus uptake (Jasper et al. 2013).

Several studies have found that bacterial pathogens, either specific organisms or indicator organisms, decrease significantly from TW influent to effluent (Reinoso et al. 2008). The mechanisms of removal are little understood, however, and long-term studies have indicated that bacterial contaminants can be retained within the wetlands and be released months after inoculation by wastewater. These methods have also been limited to detection by plate count alone, which is deceiving given many pathogens' ability to enter a viable but non-culturable state (VBNC) (Barcina & Arana 2009; Liu et al. 2009).

Escherichia coli is a Gram-negative bacterium in the phylum Proteobacteria. They are facultative anaerobes that are found most commonly in the intestinal tract of mammalian species (Griffin & Tauxe 1991). Although many strains are commensal organisms, there are a number of serotypes that are pathogenic (Kaper et al. 2004). E. coli O157:H7 is an enterohemorrhagic strain that is highly infectious, with the level of infection detected to be as few as 10 cells. It is a common pathogen found in non-treated water sources. A main reservoir is in cattle, which releases E. coli O157:H7 into the environment to contaminate water sources and subsequent crop production either through use of non-treated water irrigation or manure fertilizer (Hancock et al. 1998; Licence et al. 2001). It has been shown that E. coli O157:H7 is capable of long-term survival in river water as well as on the surfaces of plants, namely agricultural crops (Liu et al. 2009; Dinu & Bach 2011). These studies have been performed using direct counting or molecular methods due to the VBNC ability of E. coli O157:H7.

This study attempts to gain a better mechanistic view of the survival and persistence of E. coli O157:H7 within a TW to determine the risk of subsequent release of pathogens to the environment. Given the inaccuracy of plate counts, microscopy was employed as a direct cell counting method for the presence of E. coli O157:H7 within the root rhizosphere. The complexity of a full- or pilot-scale TW was scaled down and the medium removed as a potential variable in this study, instead focusing on E. coli O157:H7 interaction with the roots of plants. Roots have been shown to provide an ideal environment for biofilm growth and sustainability with the release of oxygen and exudates to the subsurface (Allen et al. 2002). It was hypothesized that the added benefit of a nutrient-rich environment would aid in attachment and survival of E. coli O157:H7 on a root as opposed to an inert surface within the TW.

MATERIALS AND METHODS

E. coli O157:H7 strain and growth

E. coli O157:H7 was originally isolated from an environmental outbreak by Dr Barry Pyle of Montana State University (Pyle et al. 1995). The DsRed fluorescent protein plasmid was previously inserted into the wild-type strain and constitutively expressed for all subsequent microscopy. A loopful of E. coli O157:H7 frozen stock was inoculated into 80 mL of custom tryptic soy broth (TSB) + 250 mg/L carbenicillin and grown overnight. The TSB media consisted of 24 g/L non-casein enzymatic digest of protein, 5 g/L NaCl, and 2.5 g/L K2HPO4. Three replicates of 20 mL overnight culture were spun down at 4,816 g for 20 minutes, washed with nanopure H2O, and concentrated in 1 mL nanopure H2O for inoculation into replicate reactors.

Plant strains and collection

Plant material was sourced from mesocosms containing monocultures of Carex utriculata and Schoenoplectus acutus that had been established for over 10 years in the greenhouse laboratory facilities at Montana State University. The plants were kept in a controlled environment and fed with simulated wastewater media (Table 1) diluted with treated tap water (city of Bozeman, MT, USA) to maintain non-pathogenic bacterial populations in the mesocosms. Shoots and roots were harvested with substantial root base and washed with tap water to remove excess dirt and sloughed biofilm cells. The extracted plants were transferred to 300 mL chemostat reactors open to the air in a laboratory environment under 24 hour plant growth lighting and allowed to grow hydroponically for 3–7 days in a simulated wastewater media (Table 1). Plants were established under constant flow conditions (3 mL/min), which continued during the reactor series experiments. The amount of plant material placed in each reactor was determined by the mass of displaced water (49.3 mL ± 26.4 mL).

Table 1

Properties of the growth media

Nutrient Chemical compound Concentration (g/L) 
Organic carbon C12H22O11 0.0997 
Iron FeCl3 0.0004 
Magnesium MgSO3 7 H20.062 
Ammonia NH4Cl 0.0191 
Copper CuSO4 0.0008 
Manganese MnSO4 0.0078 
Zinc ZnSO4 0.0078 
Calcium CaCl2 0.0019 
Phosphorus K2HPO4 0.044 
Nitrate NaNO3 0.1214 
Boron H3BO3 0.01 
Potassium KI 0.0019 
Molybdenum Na2MoO4 0.004 
Primatone complex 0.111 
Nutrient Chemical compound Concentration (g/L) 
Organic carbon C12H22O11 0.0997 
Iron FeCl3 0.0004 
Magnesium MgSO3 7 H20.062 
Ammonia NH4Cl 0.0191 
Copper CuSO4 0.0008 
Manganese MnSO4 0.0078 
Zinc ZnSO4 0.0078 
Calcium CaCl2 0.0019 
Phosphorus K2HPO4 0.044 
Nitrate NaNO3 0.1214 
Boron H3BO3 0.01 
Potassium KI 0.0019 
Molybdenum Na2MoO4 0.004 
Primatone complex 0.111 

An abiotic treatment was also used to determine the effect of a system of living roots as opposed to an inert surface of similar dimensions. Teflon nylon strings, used to simulate roots, were added to chemostats with simulated wastewater flow already established immediately prior to E. coli O157:H7 inoculation. Total weight of the nylon strings, cut into 14.5 cm lengths, was 2.94 g, and the ranged from 0.79 to 1.01 mm.

Reactor series

After plant establishment, 1 mL of concentrated overnight inoculum of E. coli O157:H7 was added to the open reactor system in three replicates of the three treatments (Carex, Schoenoplectus and nylon) and mixed via pipetting and rotating the plant or nylon ‘roots’. One additional reactor per treatment was not inoculated as a control for a total of four reactors per treatment. The control was kept on simulated wastewater media and Bozeman city tap water free of any pathogenic bacterial strains. The experiment was run in 300 mL chemostat reactors with a flow of 3 mL/min for the course of 1 week under 24 hour growth lighting and an ambient air temperature of approximately 22 °C. The experiment was repeated three times. An initial water sample, considered time point zero, was taken from each reactor immediately after inoculation and analyzed for E. coli O157:H7 concentration. Effluent samples were collected from the three inoculated reactors at 45-minute intervals over the first 6 hours unless the replicate was destructively sampled within 6 hours. The uninoculated control was sampled at only 0 and 6 hours to verify no E. coli O157:H7 was contaminated in the system. All water samples were immediately frozen until further analysis. After 2 hours, corresponding to approximately one residence time of the reactor, one of the triplicate inoculated reactors was destructively sampled. Root and nylon samples were cut in half and frozen for subsequent DNA extraction. A representative root length was pulled through a 2 mm square glass capillary (Wale Apparatus, PA) and observed using epifluorescent microscopy to view initial attachment. Another water sample was taken at 24 hours and a second replicate was extracted and destructively sampled as previously described. The third replicate and uninoculated control were destructively sampled at 1 week.

Epifluorescence microscopy

The root excision within the glass capillary was rinsed with water and viewed under 600× magnification with water optical focus. The root was viewed under 558 nm wavelength absorption for the DsRed plasmid excitation. Images were taken along the entire length of the root, focusing on the root tip, mid-root, and top root portions. The photosystem II chlorophyll enzymes within the root naturally fluoresce under a range of wavelengths, allowing the surface to be viewed under a contrasting condition to the E. coli O157:H7 DsRed plasmid. The images were semi-quantitatively analyzed using MetaMorph® Image Analysis Software to determine amount and colony size using average E. coli O157:H7 single cell fluorescence.

RESULTS AND DISCUSSION

Microscopy

E. coli O157:H7 was inoculated successfully into the triplicate reactors. It was verified that E. coli O157:H7 is able to persist but not proliferate within the simulated wastewater media designed for this experiment (Figure 1), by tracking the survival in sterile wastewater media batch cultures. Given this observation, it was assumed that E. coli O157:H7 cell counts and cell concentration would not be affected by cell division of the original inoculum. In addition, due to the low replication rate, the 100-minute residence time limits the retention of planktonic cells in the chemostat and encourages attachment to the root biofilm. One replicate was sampled in its entirety at 2 hours to determine initial attachment, and the other two at 24 hours and 1 week, respectively, to assess the ability of E. coli O157:H7 to persist over a duration typical of a TW residence time.

Figure 1

E.coli O157:H7 survival in sterile simulated wastewater media. Plate counts were tracked on two nutrient agar types. Triplicate technical replicates in two experimental replicates are represented. WW1 = first replicate in sterile simulated wastewater; WW2 = second replicate; LB = Luria broth agar plates used in plate count detection; R2A = Reasoner's 2A agar plates.

Figure 1

E.coli O157:H7 survival in sterile simulated wastewater media. Plate counts were tracked on two nutrient agar types. Triplicate technical replicates in two experimental replicates are represented. WW1 = first replicate in sterile simulated wastewater; WW2 = second replicate; LB = Luria broth agar plates used in plate count detection; R2A = Reasoner's 2A agar plates.

At 2 hours, one residence time for washout, one replicate each of the plants and nylon string was destructively sampled. A control root or nylon sample was also taken and verified that there was no contamination with E. coli O:157:H7-DsRed. Images were taken of the root and string under epifluorescence microscopy (Figure 2(a)). C. utriculata, S. acutus and the nylon all had visible single cells of E. coli O157:H7 attached on the surface. There was no significant difference between each plant species and the abiotic control, suggesting that initial attachment is based primarily on attachment surface availability and is less dependent on any effect of a living root. The nylon ‘roots’ were added just prior to E. coli O157:H7 inoculation; therefore no initial biofilm was present, also providing support that an established bacterial community on the surface does not have a strong effect on initial pathogen attachment.

Figure 2

Microscopy images of C. utriculata, S. acutus and Teflon nylon at 2 hours (left), 24 hours (middle), and 1 week (right) post-inoculation of E. coli O157:H7. Images were taken at ×600 optical zoom and 558 nm wavelength. Scale bar represents 20 μm.

Figure 2

Microscopy images of C. utriculata, S. acutus and Teflon nylon at 2 hours (left), 24 hours (middle), and 1 week (right) post-inoculation of E. coli O157:H7. Images were taken at ×600 optical zoom and 558 nm wavelength. Scale bar represents 20 μm.

Another sampling point was taken at 24 hours post-inoculation (Figure 2(b)). On all three experimental surfaces, E. coli O157:H7 cell abundance was seen to increase. Microcolonies were seen forming in addition to a high abundance of single cells retained on the surface. A pattern of increased attachment and active growth continued at the 1 week sampling point (Figure 2(c)). Much larger microcolonies were formed and three-dimensionality could be detected in portions. There was no difference in the level of persisting cells between the biotic and abiotic surfaces.

Cell clump analysis

Using MetaMorph Image Analysis Software, colony size was determined as a percentage of total cells counted over each sampling point (Figure 3). In all surfaces, the majority of cells remained as single cells. In C. utriculata, the total counts increased and colony sizes >100 cell counts were found at 24 hour and 1 week post-inoculation. S. acutus showed a smaller increase in cell counts and colony size, and viable cells and colonies were detectable on nylon roots at sampling times. The data suggest that for all surfaces, E. coli O157:H7 was able to remain persistent over the course of 1 week under constant flow conditions and possibly allowed for the proliferation of attached cells into larger colony structures.

Figure 3

Analysis of cell population per microcolony over the course of three sampling points for C. utriculata, S. acutus and nylon control, respectively. Populations are normalized as percentage of total cells counted.

Figure 3

Analysis of cell population per microcolony over the course of three sampling points for C. utriculata, S. acutus and nylon control, respectively. Populations are normalized as percentage of total cells counted.

CONCLUSION

Initial experimental results show no significant difference in attachment of E. coli O157:H7 between the biological surfaces, C. utriculata and S. acutus root systems, and the abiotic control surface, Teflon nylon string. E. coli O157:H7 was able to attach to all surfaces and was detected up to a week after inoculation in a continuous flow reactor system. In most cases, viable colony growth was observed. These data suggest the mechanism for survival and growth may not be influenced significantly by the added benefit of oxygen and/or exudates provided by plant roots or by the existence of an established biofilm on the attachment site.

Future studies are planned to determine E. coli O157:H7 presence in pilot-scale full TW systems, including the gravel matrix. Molecular studies will also be employed to determine quantitatively the percentage of E. coli O157:H7 that is able to persist and if there is significant proliferation of cells attached in biofilms.

REFERENCES

REFERENCES
Allen
W. C.
Hook
P. B.
Biederman
J. A.
Stein
O. R.
2002
Temperature and wetland plant species effects on wastewater treatment and root zone oxidation
.
Journal of Environmental Quality
31
,
1010
1016
.
Barcina
I.
Arana
I. S.
2009
The viable but nonculturable phenotype: a crossroads in the life-cycle of non-differentiating bacteria?
Review of Environmental Science and Biotechnology
8
,
245
255
.
Baskar
G.
Deeptha
V. T.
Annadurai
R.
2014
Comparison of treatment performance between constructed wetlands with different substrates
.
International Journal of Research in Engineering & Advanced Technology
2
(
3
),
1
5
.
Berendsen
R. L.
Pieterse
C. M.
Bakker
P. A.
2012
The rhizosphere microbiome and plant health
.
Trends in Plant Science
17
(
8
),
478
486
.
Bhattacharyya
P. N.
Jha
D. K.
2012
Plant growth-promoting rhizobacteria (PGPR): emergence in agriculture
.
World Journal of Microbiology and Biotechnology
28
(
4
),
1327
1350
.
Griffin
P. M.
Tauxe
R. V.
1991
The epidemiology of infections caused by Escherichia coli O157:H7, other enterohemorrhagic E. coli, and the associated hemolytic uremic syndrome
.
Epidemiologic Reviews
13
,
60
98
.
Hancock
D. D.
Besser
T. E.
Rice
D. H.
Ebel
E. D.
Herriott
D. E.
Carpenter
L. V.
1998
Multiple sources of Escherichia coli O157 in feedlots and dairy farms in the Northwestern USA
.
Preventive Veterinary Medicine
35
(
1
),
11
19
.
Jasper
J. T.
Nguyen
M. T.
Jones
Z. L.
Ismail
N. S.
Sedlak
D. L.
Sharp
J. O.
Luthy
R. G.
Horne
A. J.
Nelson
K. L.
2013
Unit process wetlands for removal of trace organic contaminants and pathogens from municipal wastewater effluents
.
Environmental Engineering Science
30
(
8
),
421
436
.
Kaper
J. B.
Nataro
J. P.
Mobley
H. L. T.
2004
Pathogenic Escherichia coli
.
Nature Reviews
2
,
123
140
.
Licence
K.
Oates
K. R.
Synge
B. A.
Reid
T. M. S.
2001
An outbreak of E. coli 0157 infection with evidence of spread from animals to man through contamination of a private water supply
.
Epidemiology and Infection
126
(
1
),
135
138
.
Pyle
B. H.
Broadaway
S. C.
McFeters
G. A.
1995
A rapid, direct method for enumerating respiring enterohemorrhagic Escherichia coli O157:H7 in water
.
Applied and Environmental Microbiology
61
(
7
),
2614
2619
.
Taylor
C. R.
Hook
P. B.
Stein
O. R.
Zabinski
C. A.
2011
Seasonal effects of 19 plant species on COD removal in subsurface treatment wetland microcosms
.
Ecological Engineering
35
(
5
),
703
710
.