A Dissertation
entitled
Role of Extracellular Polymeric Substances (EPS) on Biofilm Disinfection in a Model
Drinking Water Distribution System
by
Zheng Xue
Submitted to the Graduate Faculty as partial fulfillment of the requirements for the
Doctor of Philosophy Degree in Civil Engineering
______Dr. Youngwoo Seo, Committee Chair
______Dr. Ashok Kumar, Committee Member
______Dr. Cyndee L. Gruden, Committee Member
______Dr. Dong-Shik Kim, Committee Member
______Dr. Jeffrey G. Szabo, Committee Member
______Dr. Patricia Komuniecki, Dean College of Graduate Studies
The University of Toledo
December 2012
Copyright 2012, Zheng Xue
This document is copyrighted material. Under copyright law, no parts of this document
may be reproduced without the expressed permission of the author.
An Abstract of
Role of Extracellular Polymeric Substances (EPS) on Biofilm Disinfection in a Model
Drinking Water Distribution System
by
Zheng Xue
Submitted to the Graduate Faculty as partial fulfillment of the requirements for the
Doctor of Philosophy Degree in Civil Engineering
The University of Toledo
December 2012
Biofilms are undesirable but ubiquitous in drinking water systems. This study investigated the role of extracellular polymeric substances (EPS) in the biofilm life cycle, including planktonic cells, attached biofilm, detached biofilm clusters and redistributed biofilm, in a model distribution system with minimal disinfectant residuals. EPS contributed to bacterial surface properties, biofilm structural characteristics, disinfectant diffusion and reaction, organic matter retention and utilization, hence playing pivotal roles in bacterial resistance to disinfectants. Strains from an opportunistic pathogen,
Pseudomonas aeruginosa with different EPS secretion capabilities were tested. Two major components in P. aeruginosa EPS, polysaccharides and proteins, both reacted rapidly with chlorine while monochloramine reacted specifically with proteins. The impact of biofilm EPS reactivity with disinfectants on disinfection efficacy was evaluated by monitoring planktonic bacteria viability, disinfectant decay, biofilm viability, biofilm structure, and detached biofilm viability as well as their redistribution, systematically
iii
during the disinfection process. The obtained results suggested that the presence of EPS increased the resistance of planktonic bacteria, biofilm and detached biofilm to both chlorine and monochloramine. The EPS reactivity led to different protection approaches for bacterial cells, acting either as a disinfectant consumer (for chlorine inactivation) or limiting access to reactive sites on a cell membrane (for monochloramine inactivation).
The biofilm structure characterization using confocal laser scanning microscopy (CLSM) revealed that EPS production affected biofilm structure, specifically surface roughness, surface area to volume ratio and average diffusion distance. These structural characteristics were closely related to overall biofilm viability and the spatial distribution of viability within biofilms. Although the overall viable ratios were similar under the two disinfectants for each strain, monochloramine penetrated deeper into biofilm matrix than chlorine regardless the quantity of EPS content, showing a higher inactivation efficacy in the middle section of biofilms. However, chlorine was more efficient in controlling planktonic and detached cluster viability than monochloramine. The combined results suggested that different reactivity of biofilm EPS with disinfectants influenced the susceptibility of both biofilm and detached biofilm during the disinfection practices. This study provides valuable insight for both fundamental studies of biofilm life cycle and disinfection practices to optimize water quality maintenance in distribution systems.
iv
This dissertation is dedicated to my parents for their endless love, support and encouragement since the very beginning of my studies.
v
Acknowledgements
I would like to thank all of the people who helped make this dissertation possible.
First, I wish to thank my advisor, Dr. Youngwoo Seo for all his guidance, encouragement, support, and patience. His sincere interests in research and education have been a great inspiration to me. Also, I would like to thank my committee members,
Dr. Ashok Kumar, Dr. Cyndee Gruden, Dr. Dong-Shik Kim, and Dr. Jeffrey Szabo, for their very helpful insights, comments and suggestions. Additionally, I would like to acknowledge all of those people who assisted, advised, and supported my research over the years: Dr. Andrea Kalinoski, for training and assisting me with confocal laser scanning microscopy; Dr. Mau-yi Wu, for valuable advice and suggestions; Dr.
Yakov Lapitsky, for guidance on the Zetasizer analysis; Tammy Phares, for providing very helpful support in conducting experiments; and Sean Linkes, for technical support at the Flow Cytometry Core Facility. Finally, I would like to thank my colleagues and friends, Christopher Hessler, Varunraj Sendamangalam, Kimberly Coburn, Bin Li and
Rui Zheng, who all provided invaluable assistance, support and suggestions throughout this process.
Finally, I would like to thank my parents, Zaiquan and Ruiwei, for their support and encouragement. I could not have completed this effort without their love, tolerance, and enthusiasm.
vi
Contents
Abstract ...... iii
Acknowledgements ...... vi
Contents ...... vii
List of Tables ...... xii
List of Figures ...... xiii
List of Abbreviations ...... xvi
List of Symbols ...... xviii
Preface ...... xix
1 Introduction ...... 1
2 Objective and Significance ...... 7
3 Pseudomonas aeruginosa Inactivation Mechanism is Affected by Capsular
Extracellular Polymeric Substance Reactivity with Chlorine and Monochloramine ...... 12
Abstract ...... 12
3.1 Introduction ...... 13
3.2 Materials and Methods ...... 15
3.2.1 Bacterial Culture ...... 15
3.2.2 Preparation of Disinfectant Solutions ...... 16
3.2.3 Batch Disinfection Experiments ...... 16
vii
3.2.4 Modeling with GInaFiT ...... 17
3.2.5 Bacteria Cell Staining ...... 18
3.2.6 EPS Extraction and Characterization ...... 19
3.2.7 Disinfectant Decay by Extracted EPS ...... 20
3.2.8 FTIR Spectroscopy ...... 20
3.2.9 Statistical Analysis ...... 21
3.3 Results ...... 21
3.3.1 Culture and EPS Properties ...... 21
3.3.2 Inactivation Kinetics ...... 22
3.3.3 Model Fitting ...... 25
3.3.4 Viability Analysis by Fluorescent Staining ...... 27
3.3.5 Disinfectant Decay ...... 27
3.3.6 FTIR Analysis...... 30
3.4 Discussion ...... 35
3.5 Conclusion ...... 41
Acknowledgement ...... 42
4 Exopolysaccharide Production by Pseudomonas aeruginosa Contributes to Biofilm
Architecture and Chlorine Tolerance of Biofilm and Detached Clusters ...... 43
Abstract ...... 43
4.1 Introduction ...... 44
4.2 Materials and Methods ...... 46
4.2.1 Bacteria Culture ...... 46
4.2.2 Solution Preparation ...... 46
viii
4.2.3 Biofilm Cultivation and Disinfection in Flowcell Systems ...... 47
4.2.4 Confocal Laser Scanning Microscopy and Image Analysis ...... 48
4.2.5 Bacterial Enumeration and Viability ...... 49
4.2.6 Flow Cytometry Analysis ...... 50
4.2.7 Dissolved Organic Matter (DOM) Analysis ...... 51
4.2.8 Statistical Analysis ...... 51
4.3 Results ...... 51
4.3.1 Effect of EPS Content on Biofilm Structure ...... 51
4.3.2 DOM Retention by Biofilm ...... 52
4.3.3 Biofilm Susceptibility to Chlorine Disinfection ...... 54
4.3.4 Detached Cell Viability during Biofilm Disinfection...... 57
4.4 Discussion ...... 61
4.4.1 Influence of EPS on Biofilm Structure and Biofilm Susceptibility during
Chlorine Disinfection ...... 61
4.4.2 Influence of EPS on DOM Retention by Biofilm...... 63
4.4.3 Influence of DOM on Biofilm and Biofilm Detachment Susceptibility ...... 65
4.5 Conclusion ...... 67
Acknowledgements ...... 67
5 Impact of Chlorine Disinfection on Detachment and Redistribution of Cell Clusters from Biofilm ...... 69
Abstract ...... 69
5.1 Introduction ...... 70
5.2 Materials and Methods ...... 72
ix
5.2.1 Bacterial Strains and Solution Preparation ...... 72
5.2.2 Biofilm Reactor ...... 72
5.2.3 Effluent Sample Collection ...... 74
5.2.4 Size Distribution of Detached Biofilm ...... 74
5.2.5 Bacterial Viability Test ...... 75
5.2.6 Redistribution of Detached Biofilm ...... 75
5.2.7 Interaction Energy Analysis between Detached Biofilm and Glass Surfaces .. 76
5.2.8 Statistical Analysis ...... 76
5.3 Results ...... 77
5.3.1 Size Distribution of Detached Biofilm Clusters ...... 77
5.3.2 Detached Biofilm Susceptibility to Chlorine Disinfection ...... 79
5.3.3 Zeta Potential ...... 82
5.3.4 Redistribution of Detached Biofilm and Biofilm Structural Characterization . 85
5.4 Discussion ...... 88
5.5 Conclusion ...... 92
Acknowledgement ...... 93
6 Comparison of Chlorine and Monochloramine Inactivation in Biofilm and Detached
Biofilm in a Model Drinking Water Distribution System ...... 94
Abstract ...... 94
6.1 Introduction ...... 95
6.2 Materials and Methods ...... 97
6.2.1 Bacteria Culture ...... 97
6.2.2 Solution Preparation ...... 98
x
6.2.3 Biofilm Cultivation and Disinfection in Flow Cell System ...... 98
6.2.4 Confocal Laser Scanning Microscopy and Image Analysis ...... 99
6.2.5 Bacterial Enumeration and Viability ...... 100
6.2.6 Flow Cytometry Analysis ...... 101
6.2.7 Statistical Analysis ...... 101
6.3 Results ...... 102
6.3.1 Effect of EPS Content on Biofilm Structure ...... 102
6.3.2 Biofilm Susceptibility to Chlorine and Monochloramine ...... 103
6.3.3 Detached Cell Viability during Biofilm Disinfection...... 107
6.4 Discussion ...... 111
6.5 Conclusion ...... 116
Acknowledgements ...... 117
7 Conclusions ...... 118
8 Recommendations for Future Research ...... 121
Reference ...... 123
xi
List of Tables
3.1: Ct values by plate counting and viable ratio by fluorescent staining ...... 25
3.2: Disinfection kinetic parameters from inactivation model fitting ...... 26
3.3: Assignment of characteristic bands of FTIR spectra ...... 31
4.1: Biofilm structural parameters and biofilm EPS content ...... 52
5.1: Redistributed biofilm structural characteristics for long-term (6-day) tests...... 86
6.1: Biofilm structural parameters and biofilm EPS content...... 102
xii
List of Figures
3-1: P. aeruginosa strains on agar plates and under SEM. (a) algT(U); (b) PAO1; (c) mucA22...... 22
3-2: P. aeruginosa cells inactivation kinetics by (a) chlorine and (b) monochloramine.. 24
3-3: Disinfectant decay by P. aeruginosa cells. (a) chlorine; (b) monochloramine...... 28
3-4: Disinfectant decay by extracted EPS and BSA standard. (a) chlorine reaction with
EPS; (b) monochloramine reaction with EPS; (c) monochloramine reaction with BSA. 30
3-5: FTIR spectra of (a) untreated and chlorine treated bacterial cells...... 31
3-6: Transformed FTIR spectra of chlorine treated (a, b) and monochloramine treated (c, d) bacterial cells. (a, c) 1300-900 cm-1; (b, d) 1800-1300 cm-1...... 35
3-7: Inactivation kinetics of EPS removed P. aeruginosa cells by (a) chlorine and (b) monochloramine...... 36
3-8: LIVE/DEAD stained mucA22 cells after chlorine disinfection...... 39
4-1: Diagram of flowcell system set up……………………………………...... …... 48
4-2: DOM adsorption by biofilms...... 53
4-3: Selective removal of different DOM fractions by biofilms...... 54
4-4: Relationship between biofilm EPS content and biofilm overall viability ratio. The relationship was analyzed by linear regression with goodness of fit separating the presence and absence of DOM conditions...... 55
xiii
4-5: Spatial distribution of control and disinfected biofilm viability ratio in the absence of
DOM...... 56
4-6: Detached biofilm cluster viability ratio determined by HPC in the presence and absence of DOM...... 58
4-7: Detached biofilm viability quantified using flow cytometry. (a) flow cytomery data in dot plot. Cell viability percentages by strains for P. aeruginosa detached biofilm: (b) in the absence of DOM; (c) in the presence of DOM...... 61
4-8: Spatial distribution of biomass through horizontal cross-sections. Biomass content in each cross-section determined using COMSTAT...... 63
5-1: Diagram of flowcell system setup ...... 74
5-2: Size distribution of biofilm clusters detached from different stages during biofilm growth. (a) algT(U) biofilm; (b) PAO1 biofilm; (c) mucA22 biofilm...... 78
5-3: Chlorine inactivation kinetics of (a) planktonic cells; (b) cell clusters detached from
3-day biofilms; (c) cell clusters detached from 6-day biofilms...... 82
5-4: Zeta potential of planktonic bacteria suspension and resuspended biofilm of P. aeruginosa strains. (A) Planktonic bacterial suspension - control; (B) planktonic bacterial suspension – chlorine disinfected; (C) planktonic bacterial suspension – resuspended in buffer after chlorine disinfection; (D) biofilms resuspended in buffer – control...... 83
5-5: Zeta potential of cell clusters detached from biofilms on different stages during biofilm growth. Cell clusters detached from (a) the first flowcells; (b) the second flowcells with chlorine disinfection; (c) the second flowcells without chlorine disinfection...... 84
xiv
5-6: Coverage rate in the second flowcell for short term (24-hour) redistribution test of mucA22 strain...... 88
6-1: Overall viable ratio of biofilm determined by HPC under chlorine and monochloramine disinfection. …… ………………….....…...... 104
6-2: Biofilm overall viable ratio determined by CLSM image analysis with chlorine and monochloramine disinfection...... 105
6-3: Spatial distribution of control and monochloramine disinfected biofilm viable ratio within biofilms...... 107
6-4: Detached biofilm cluster viable ratio determined by HPC with chlorine and monochloramine disinfection...... 109
6-5: Detached biofilm viability quantified using flow cytometry with chlorine and monochloramine disinfection...... 111
xv
List of Abbreviations
ATR ……………………………………………...... attenuated total reflection ANOVA ………………………………………………………...... analysis of variance
BSA ……………………………………………………………….. bovine serum albumin
CDF ………………………………………………………………… chlorine demand free CFU ………………………………………………………………….. colony forming unit CLSM ……………………………………………….. confocal laser scanning microscope Ct ……………………………………………………. concentration of disinfectant × time
DBP ……………………………………………………………… disinfectant by-product DOM ……………………………………………………………. dissolved organic matter DLVO …………………………………………….. Derjaguin-Landau-Verwey-Overbeek DOC ……………………………………………………………. dissolved organic carbon DPD ……………………………………………………. N,N-diethly-p-phenylenediamine
EPS …………………………………………………... extracellular polymeric substances
FACS …………………………………………………... fluorescence activated cell sorter FTIR ……………………………………………. Fourier transform infrared spectroscopy
HPC ……………………………………………………………. heterotrophic plate count HRT …………………………………………………………….. horizontal retention time
LB ……………………………………………………………………………… luria broth
NOM ………………………………………………………………. natural organic matter pH ……………………………………………… potential for hydrogen ion concentration PI ……………………………………………………………………….. propidium iodide
Rpm ………………………………………………………………….. rotations per minute
SEM ……………………………………………………….. scanning electron microscope
UV-vis …………………………………………………………………. ultraviolet-visible
xvi
VBNC …………………………………………………………... viable but non culturable
xvii
List of Symbols
oC..…………………………………………………………………. celsius g ………….………………………………………………….………. gram K ………………………………………………………………...... kelvin mL ……………………………………………………………..… milliliter mg …………………………………………………………..…... milligram mm ……………………………………………………………... millimeter mM …………………………………………………………….. millimolar mol …………………………………………………….....…….....…. mole nm...……………………………………………………………. nanometer µm…………………………………………………………..…. micrometer µg…………………………………………………………..…... microgram
xviii
Preface
The biological stability of water in distribution systems is a major concern of drinking water quality control. During the distribution of drinking water, bacterial regrowth may lead to a deterioration of water quality, amplification of corrosion, ascetics, and proliferation of microorganisms. The protection of public water supplies against waterborne diseases greatly relies upon the use of disinfectants. However, reports from many water utilities in the US have shown that pathogens can survive in water distribution systems, despite the presence of residual disinfectants. The drinking water distribution pipelines offer a very large surface area for the adhesion of bacteria. At least
95% of the total bacterial biomass in drinking water is found as biofilm facilitated by the production of extracellular polymeric substances (EPS). The presence of EPS has been known to facilitate bacterial immobilization, proliferation, reattachment and regrowth in the biofilm life cycle. More importantly, the presence of EPS significantly impacts bacterial surface characteristics and biofilm structure features, hence contributes to the resistance of bacteria, especially those embedded in biofilm matrix, to a variety of antibiotics and disinfectants. Biofilm control is becoming an important part of the operation of drinking water plants and distribution systems. In order to prevent bacterial regrowth in drinking water distribution systems, there has been an increasing interest in alternative disinfectants, such as chloramines, to provide a more stable disinfectant residual and to reduce disinfectant by-product (DBP) formation than free chlorine. The xix
reactivity between different EPS components and disinfectants significantly influence disinfection efficacy for planktonic bacteria, biofilm, as well as detached biofilm clusters, exhibiting unique protection mechanisms by the presence of EPS.
xx
Chapter 1
Introduction
Biofilm is an aggregate of microorganisms in which cells adhere to each other on a surface (e.g., rocks, medical devices, pipe lines, skin, etc.). These adherent cells are frequently embedded within a self produced matrix of extracellular polymeric substances
(EPS). EPS is a comprehensive term for organic macromolecules including extracellular
DNA, proteins, lipids, and polysaccharides. EPS may vary in their physical and chemical properties, but there are primarily composed of polysaccharides and proteins. The production of EPS by microorganisms has been observed both in their planktonic state and biofilm state (1).
Increasing numbers of studies focused on the functions of EPS have been conducted during the last few decades. The presence of EPS influences the microenvironment in biofilm by affecting porosity, density, water content, charge, sorption properties, hydrophobicity and mechanical stability (2). In general, the most important functions of EPS are holding together and providing a dense and protected environment for biofilm microorganisms. Meanwhile, a variety of other possible functions have also been attributed to the presence of EPS. EPS have been observed to be involved in the initial colonization and accumulation of bacteria on inert and tissue
1
surfaces. Tsuneda et al. reported that EPS covering cell surfaces promoted bacterial adhesion to solid surfaces (3). EPS also facilitate cells aggregation to form bacterial flocs and biofilms by providing adhesion sites to immobilize bacterial cells (4). The EPS matrix promote water retention and nutrient accumulation to prevent desiccation or starvation under inhospitable conditions (1). Most recently, the role of EPS in genetic transfer has also been subject of intense investigation. Peterson et al. reported that the extracellular DNA was associated with the export of competence-stimulating peptide
(CSP), which plays an important role in cell-to-cell signal during biofilm formation (5).
Biofilm may form on living or non-living surfaces and can be ubiquitously found in natural, industrial, and hospital settings. The microbial cells growing in a biofilm are physiologically distinct from planktonic cells of the same organism. The development of a complex structure allows resistance to biocides that would destroy individual cells not part of a biofilm colony. The antibiotic resistance can be increased a thousand fold within the biofilm compared to the suspended planktonic cells (6).
The harmful consequences of biofilm establishment and proliferation impact many facets of civilization, such as the risk posed by hospital-based infections and contamination in the food service industry. Such examples illustrate a need to understand biofilm development and effective strategies for its control. On the other hand, there are other biofilms that are known to be beneficial and can be used to fix problems, such as wastewater treatment and remediation of underground oil contamination. Researchers in the medical, food, water, and environmental microbiology fields have begun to investigate microbiologic processes from a biofilm perspective. The Center for Disease
Control and Prevention (Atlanta, GA, USA) reported that 65% of human disease caused
2
by bacterial infections involved biofilms (7). Billions of dollars are spent every year on biofilm caused energy losses, equipment damage, product contamination, and medical infections (8). As biofilm related public health issues become more prominent in the pharmaceutical, food, and health-care industries, novel strategies for biofilm prevention and control are being developed and implemented.
In drinking water distribution systems, it is of paramount importance to minimize the risk of water contamination in the infrastructure used to deliver drinking water from the water treatment plants to the public. One of the reasons that water utilities became concerned with biofilms in drinking water systems was due to growth of coliform bacteria in the pipeline network. In 1993, nearly 4,400 United States water systems, affecting 21 million people, violated drinking water standards for total coliform bacteria.
Similar trends occurred in 1994 and 1995, with over 12,000 systems exceeding standard coliform levels during that period (9). Largely, these problems are due to regrowth of the bacteria in distribution systems.
The concern of biofilms came to the forefront of research because bacteria in biofilms are found everywhere and are more resistant to disinfection than their planktonic cells (10, 11). In the presence of biofilms, drinking water distribution systems could encounter operational problems such as pipe corrosion, water quality deterioration, or other undesirable events. The first recorded observation concerning biofilms comes from a 1933 paper by Henrici, which states “....it is quite evident that for the most part water bacteria are not free floating organisms, but grow upon submerged surfaces” (12). Sixty years later, it was determined that, more than 95% of the entire biomass of distribution pipelines in drinking water distribution systems is found to be located at the pipe walls
3
due to the large surface to volume ratio. This leaves less than 5% remaining in the water phase (13).
Multiple characteristics of biofilms contribute to its resistance to chemically diverse biocides implemented to combat biofouling in industrial processes. The efficacy and application of biocides are optimal for the disinfection of nonattached planktonic cells; however, bacterial cells in biofilm are embedded in the mucoid EPS matrix (14), whose presence hinders disinfection practices. There are multiple theories explaining biofilm resistance to antimicrobial activities. One commonly accepted characteristic of biofilm resistance is the failure of disinfectants to penetrate the full depth of the biofilm.
The EPS, that make up the matrix of a biofilm, have the ability to retard the diffusion and decrease the flow rate of disinfectants within biofilms compared to the rate in the liquid suspension (15). The poor penetration accounts for the inability of disinfectants to destroy microorganisms embedded in biofilm.
Many studies have been done on illustrating what makes the biofilm phenotype so different from the planktonic phenotype. While there is great concern about the resistance of biofilms against disinfectants; unfortunately, the resistance mechanisms of biofilm are not well understood (16). One reason for this lack of understanding could be the complexity of biofilm EPS structure and composition. More than 80% of biofilm is comprised of EPS (17) and it is believed that these structures provide protective barriers by sequestering biocides from microorganisms (18). The simultaneous interaction between disinfectants and EPS leads to a transport limitation and inactivation limitation of disinfectants used to control biofilms, thus allowing microorganisms embedded in EPS matrix to survive and multiply.
4
For now, there is still no single standard method for the study of biofilm susceptibility, due to the complexity of biofilm structures and formation. It is even impossible to compare the results obtained with biofilms of the same species cultured under vastly different conditions (19). Initially biofilms were studied by electron microscopy, while recent advances in confocal laser scanning microscopy (CLSM) have allowed the visual inspection of fully hydrated biofilms without physically slicing. For example, it has been studied that the structure of a mature P. aeruginosa biofilm is comprised by mushroom-shaped microcolonies of bacteria, which are surrounded by an extracellular polysaccharide matrix and separated by fluid-filled channels (19).
Lechevallier et al. demonstrated the correlation between microbial attachments to microbial survivability in the presence of high levels of disinfecting agents where chlorine residual does not eliminate all bacteria from a distribution system (20). Camper et al. found that the free residual concentration in a distribution system did not provide enough inactivation of pathogenic microorganisms from their water column and were concerned that this had led to a false sense of security regarding disinfection practices for a drinking water distribution system (21). Disinfectants may injure some bacteria; however, they would be able to recover and grow within the distribution system. Chlorine decay within a drinking water system is another factor that contributes to microbial survivability (22, 23). If pathogenic microorganisms become incorporated into a pre- existing biofilm then these organisms could become more resistant to disinfectants (24).
LeChevallier et al. showed that bacteria have a two to tenfold increase in their resistance to disinfectants with biofilms on ductile iron pipes than those attached on galvanized, copper, and PVC materials (20). They attributed this effect to reactions between ferrous
5
iron and free chlorine that resulted in insoluble ferric hydroxide thereby reducing available residual for disinfection. Organisms within biofilm matrix are less vulnerable to chlorine and other biocides than those floating freely in water; however, not enough quantitative studies have been done to illustrate the biological or physical mechanisms that cause the increased resistance of biofilm to disinfectants (25).
As an important phase on the biofilm life cycle, the susceptibility of detached biofilm is the least studied. Biofilm detachment is said to be due to at least four different mechanisms: 1) erosion: small particles from the surface of biofilm detaching into bulk fluid; 2) sloughing: large piece of biofilm detaching from inner part or base of the biofilm;
3) abrasion: detachment by collision of carrier medium in fluidized bed systems; and 4) grazing: removal of biomass due to the consumption by other organisms such as protozoa
(26). Generally, it is assumed that the detachment of biomass is caused by a combination of these mechanisms and several environmental factors, such that the distinction between erosion and sloughing may be arbitrary, when there is a broad distribution of detached particle sizes (27). Causes for erosion and sloughing reside both in forces created by the movement of fluid along the biofilm and in the biofilm structure itself (28). Results of this study show that the quantity of biofilm EPS affects biofilm structural properties.
Accordingly, the quantity of EPS present in biofilm may influence the size distribution of detached biofilm clusters, while its biochemical composition may promote increased viability of these clusters. Therefore, this study will assess the role of EPS on bacterial stability throughout the biofilm life cycle by comparing the viability of three different phases of bacteria cells: planktonic cell, biofilm, and biofilm detachment.
6
Chapter 2
Objective and Significance
Current mouthwashes, antibiotics, household cleaners, and most other treatment formulas were developed by testing their effect on target microorganisms in a planktonic state. However, the effect of these treatments is significantly limited when applied to real world environments where the target microorganisms are actually in the form of biofilm.
This feature is especially prevalent in aqueous environments where 95% of microorganisms are in biofilms instead of their planktonic state (13). In this case, the control of microorganisms present in the form of biofilm is currently being aggressively pursued by biofilm researchers. The structure, growth, resistance mechanisms, reproduction, and all other factors affecting biofilm proliferation must be understood so that optimal methods to control biofilm can be developed.
Biofilm formation is the result of several physical, chemical, and biological processes occurring simultaneously, including: i) cell and particle transport to the substratum; ii) cell adhesion and attachment; iii) confluent biofilm formation by cellular growth and extracellular polymer production; iv) biofilm detachment; and v) substrate and product transport to and from the biofilm (29). Previous researchers mostly focused on one state in the biofilm life cycle. In this study, biological stability of bacterial cells in
7
the entire biofilm life cycle, including planktonic cells, steady state biofilm, detachment biofilm and redistributed biofilm, were investigated to understand the role of EPS in disinfection processes.
Biofilm formation and bacterial regrowth are dependent on complex interactions of drinking water characteristics as well as engineering and operational parameters (30).
The efficiency of disinfectant residuals is important to reduce the total bacterial cell counts in bulk water and to avoid biological contamination in drinking water systems.
Also, natural organic matter (NOM), which is not retained by conventional water treatment processes, is known to be a factor that contributes to biofilm formation and bacterial regrowth (31). In this research, a flow cell system was used to cultivate biofilm to simulate water distribution systems. The flow cell system permits a controlled development of biofilm so that replicate trials produced biofilm samples with consistent features such as biomass, thickness, surface roughness, cell number and resistance to disinfectants.
Simulating regulatory disinfectant residuals and low carbon content substrate in water distribution systems, inactivation kinetics and disinfectant decay rates of the EPS producing variants were at low disinfectant concentrations. Changes in the EPS quantity and composition as a response to disinfectant exposure were monitored using Fourier transform infrared spectroscopy (FTIR) and chemical analysis, thus determining how bacterial EPS may provide resistance to disinfectants in both whole cells and extracted
EPS in batch experiments. The impact of capsular EPS for the Gram-negative bacteria P. aeruginosa on bacterial inactivation mechanisms by chlorine-based disinfectants was investigated using a combination of batch experiments, viability tests with LIVE/DEAD
8
staining, and FTIR. Both wild-type and isogenic mutant strains with different alginate
EPS production capabilities were used to evaluate their susceptibility to chlorine and monochloramine. Multiple analyses were combined to better understand the mechanistic role of capsular EPS against chlorine-based disinfectants.
Beyond planktonic bacteria, EPS also play significant roles in biofilm formation and resistance to disinfectants. Biofilm structural characteristics were quantitatively analyzed and compared considering the amount of EPS production. The influence of biofilm structure, EPS content, and interaction between biofilm and dissolved organic matter on viability of biofilm and detached biofilm to chlorine and monochloramine were studied.
The detachment of bacterial cells from biofilms in water distribution systems is of key importance to water quality and public health. Although many studies have been done on biofilm disinfection and disinfectant penetration efficiency in biofilm, detachment is the least studied biofilm process and remains poorly understood (32). The detachment of bacterial cells from biofilm in a distribution system is mainly a result of hydraulic erosion or sloughing. The detached bacterial cells may survive in aqueous solutions to interrupt water biological stability or reattach to solid surfaces to form new biofilm. However, the behavior of detached biofilm under the presence of disinfectant residual has not been fully studied. Information on biofilm detachment is usually inferred from monitoring the detachment rate and size distribution. These studies mostly explained the extent and frequency of detachment, but information on the susceptibility of the detached bacterial cells has not been systematically studied (33). Also, there is no sufficient information regarding the combined impact of EPS and DOM on detached
9
biofilm viability, despite that many researchers have discussed the interaction between organic carbon and biofilms. A minimum disinfectant residual dose to maintain a certain biological level may be underestimated if biofilm detachment is not taken into consideration. Studying the behavior of biofilm detached cells considering disinfectant demanding substrates provides important information regarding minimal disinfectant residuals and DOM removal efficacy to prevent bacterial regrowth in drinking water distribution systems.
These biofilm disinfection experiments were conducted to assess the impact of biofilm EPS on biological stability of water within model distribution systems (flow cell system) in the presence and absence of DOM. Biofilm detachment and detached biofilm redistribution was monitored within the time period from early biofilm formation to a steady state. The susceptibility of the detached biofilm was investigated and compared with their planktonic states. Size distribution of detached biofilm and its redistribution pattern was also studied. After applying disinfectant to steady state biofilm, the biofilm was visualized using confocal laser scanning microscopy (CLSM) followed by image analysis to quantify biofilm structural characteristics and viability spatial distribution.
The survival rate of detached cells from biofilms was analyzed by flow cytometry to differentiate live, dead and membrane compromised cells considering the presence or absence of disinfectant demanding substrate (DOM). Both biofilm and detached biofilm cluster viability were confirmed by heterogenic plate count method. These results provided quantitative assessment of i) the effects of biofilm EPS on biofilm structure; ii) the role of biofilm structural characteristics and EPS content on biofilm and detached biofilm susceptibility to model disinfectants (chlorine and monochloramine); iii) the
10
impacts of the presence of DOM on bacterial water quality (biofilm and detached biofilm viability); and iv) detached cluster redistribution influenced by the quantity of EPS in the presence of disinfectant.
The results obtained in this study can provide valuable insights regarding the concentration of residual disinfectant required to facilitate biofilm inactivation and prevent bacteria regrowth in water distribution systems.
11
Chapter 3
Pseudomonas aeruginosa Inactivation Mechanism is
Affected by Capsular Extracellular Polymeric
Substance Reactivity with Chlorine and
Monochloramine
Abstract
The reactivity of capsular extracellular polymeric substances (EPS) to chlorine and monochloramine was assessed and compared in this study. The impact of capsular
EPS on the Gram-negative bacteria Pseudomonas aeruginosa inactivation mechanisms was investigated both qualitatively and quantitatively using a combination of batch experiments, viability tests with LIVE/DEAD staining, and Fourier transform infrared spectroscopy (FTIR). Both wild-type and isogenic mutant strains with different alginate
EPS production capabilities were used to evaluate their susceptibility to chlorine and monochloramine. The mucA22 mutant strain, which overproduces the EPS composed largely of acidic polysaccharide alginate, exhibited high resistance and prolonged inactivation time to both chlorine and monochloramine relative to PAO1 (wild-type) and
12
algT(U) mutant strains (alginate EPS deficient). Multiple analyses were combined to better understand the mechanistic role of EPS against chlorine-based disinfectants. The extracted EPS exhibited high reactivity with chlorine and very low reactivity with monochloramine, suggesting different mechanisms of protection against disinfectants.
Moreover, capsular EPS on a cell membrane appeared to reduce membrane permeabilization by disinfectants as suggested by deformation of key functional groups in both the EPS and cell membrane (the C-O-C stretching of carbohydrate and the C=O stretching of ester group). The combined results supported that capsular EPS, acting either as a disinfectant consumer (for chlorine inactivation) or limiting access to reactive sites on cell membrane (for monochloramine inactivation), provide a protective role for bacterial cells against regulatory residual disinfectants by reducing membrane permeabilization.
3.1 Introduction
The protection of public water supplies against waterborne diseases greatly relies upon the use of disinfectants. However, reports from many water utilities in the US have shown that pathogens can survive in water distribution systems, despite the presence of residual disinfectants (34).
A growing number of studies have discussed the influence of extracellular polymeric substances (EPS) produced by microorganisms on bacterial regrowth and biofilm formation in water systems (16). The presence of EPS has been known to facilitate bacterial immobilization, proliferation, reattachment and regrowth. EPS have significant impact on bacterial surface characteristics, which may alter biochemical properties of bacterial cells (35, 36). 13
Additionally, it is well accepted that EPS provide cells with protection against inhospitable environments, such as antibiotics, desiccation, heavy metals, etc. (37).
Several researchers have reported that the presence of EPS enhanced bacterial viability against bactericidal activities (38, 39). It was also found that extracellular alginate has an intrinsic protective effect on suspended P. aeruginosa cells against antibiotics by a non- charge-related binding mechanism (40). In addition to physical attributes that aid in resistance of microorganisms to free chlorine, many bacterial species defend against oxidative stresses by adopting resistance mechanisms that increase EPS production (41).
It was previously hypothesized that extracellular material may sacrificially react with free chlorine and reduce the concentration of disinfectant at the cell membrane. Furthermore, a debate persists whether the microbial inactivation by free chlorine involves changes in membrane permeability or extensive oxidation of the microbial cell membrane, which results in leakage of intracellular material and cell death (42, 43). However, the reactivity of extracellular material on the cell membrane with disinfectants was not considered in the debated inactivation mechanisms.
Despite growing acceptance of protective role of EPS against the act of antimicrobial agents, previous studies predominantly focused on monitoring the bacterial survival rate and disinfectant decay, without consideration of the physiochemical properties of the microorganism. Specifically, the chemical composition or the functional groups of bacterial EPS involved in the reaction with disinfectants were not well characterized (44), although most bacterial strains produce EPS even as suspended cultures (37, 45, 46). Currently, no systematic study has been conducted to illustrate the reactivity of EPS with disinfectants or the role of EPS in microbial inactivation (38, 39).
14
Furthermore, there has been increasing interest in alternative disinfectants to free chlorine, such as chloramines, to reduce disinfectant by-product (DBP) formation. The role of capsular EPS in inactivation mechanisms of chloramines was not robustly evaluated, specifically considering the reactivity between EPS and chloramines (47).
In this study, we sought to assess and compare the reactivity of capsular EPS to chlorine and monochloramine using both qualitative and quantitative analyses. Three P. aeruginosa strains, both wild type and isogenic mutants with varied alginate EPS production, were evaluated to assess the role of EPS on bacterial inactivation. Simulating minimal regulatory disinfectant residuals in water distribution systems, inactivation kinetics and disinfectant deacay rates of the EPS producing variants were studied at low disinfectant concentrations. Examining both whole cells and extracted EPS in batch experiments, changes in the EPS quantity and composition as a response to disinfectant exposure was monitored.
3.2 Materials and Methods
3.2.1 Bacterial Culture
P. aeruginosa is an opportunistic human pathogen with cases of contamination in water distribution systems as well as food processing equipment (48-50). All isogenic mutant strains were constructed in wild-type P. aeruginosa strain PAO1, whose primary
EPS component is the acidic polysaccharide alginate. Two mutants, mucA22
(overproduction of alginate EPS) and algT(U) (minimal or no alginate EPS production), were constructed as described elsewhere (51, 52). All strains were grown in one-tenth strength LB broth (2.5 g/L, Difco Laboratory, Detroit, MI) at 37°C until the late- exponential phase and then harvested by centrifugation at 2,000 × g for 15 min (model 15
5804R centrifuge, Eppendorf, Hamburg, Germany), allowing for minimal removal of capsular alginate EPS (53), washed twice and resuspended in chlorine demand free (CDF) buffer (0.54 g Na2HPO4 and 0.88 g KH2PO4 per liter, pH=6.98) (54) as bacterial suspensions (~106 cells/mL). The cell density selected for batch disinfection experiments are consistent with those employed in other studies (55-57). Additionally, the cell density selected for batch disinfection experiments also reflects biofilm cell densities found in distribution systems (10, 58).
3.2.2 Preparation of Disinfectant Solutions
All disinfection experiments were conducted with CDF buffer. Chlorine solution
(0.5 mg/L) was prepared with Clorox bleach (The Clorox Co., Oakland, CA).
Monochloramine solution (2 mg/L) was prepared by combining solutions of sodium hypochlorite (5.65-6%, pH=8.3) and ammonium chloride (0.2 mM/L, pH=8.3) in a 4:1 chlorine-to-ammonia-nitrogen mass ratio, where monochloramine is the dominant species at this pH. The disinfectant solutions were prepared immediately preceding the inactivation experiments. The disinfectant concentrations were selected based on residual disinfectant concentration typically used in water distribution systems (59). Both the initial doses and residual concentrations were measured as free chlorine by N,N-diethly- p-phenylenediamine (DPD) method for chlorine and indophenol method for monochloramine using a DR/2700 spectrophotometer (HACH Company, Loveland, CO)
(60).
3.2.3 Batch Disinfection Experiments
All batch experiments were performed in 250 mL amber glass bottles (Fisher
Scientific) at room temperature (22oC ± 2). Amber glass bottles were used as parallel
16
reactors, where duplicate sample bottles containing culture suspension and disinfectant were compared to a control bottle. Inactivation tests with disinfectants were conducted simultaneously measuring disinfectant decay and bacterial survival. After quenching residual disinfectant with CDF buffer solution containing Na2S2O3 (1 mM/l), enumerations of viable microbial cells were performed using the plate count method with
R2A agar plates (Difco Laboratories, Detroit, MI). Batch experiments were repeated more than three times for each strain with each disinfectant and each sample was plated in duplicate.
3.2.4 Modeling with GInaFiT
The Geeraerd and Van Impe inactivation model fitting tool (GInaFiT, Version 1.5,
KULeuven, Belgium) was used to interpret non-log-linear microbial survival models on inactivation data (61). The best-fit model was chosen by comparing the values of the root mean sum of squared errors (RMSE) as described previously (62). The model applied was a log-linear model with shoulder and tailing for monochloramine and with tailing for chlorine disinfection (Equation 1) (61). Based on the model-fitted data, Ct values
(concentration of disinfectant × time) for 2-log inactivation (99% of the population is inactivated) were calculated.
N(t) = (N(0) – Nres) · exp(-kmaxt) · exp(kmaxS1) / {1 + [exp(kmaxS1) – 1] · exp(-kmaxt)} +
Nres [1]
Where, Kmax (1/time unit) is the specific inactivation rate.
S1 (time unit) is the lag time representing the shoulder (S1 = 0 for chlorine
inactivation).
Nres is the number of more resistant population representing the tailing.
17
3.2.5 Bacteria Cell Staining
The LIVE/DEAD Bacterial Viability Kit (BacLight, InVitrogen by Life
Technologies, Carlsbad, CA) was applied to enumerate both viable and total counts of bacteria investigating disinfectant efficacy. In the LIVE/DEAD stain kit, SYTO 9 penetrates all bacterial membranes and stains the cells green, while propidium iodide (PI) only penetrates cells with damaged membranes. The combination of the two stains stoichiometrically produces red fluorescing cells. When the BacLight viability stain is compared with traditional colony forming unit counts, potentially viable but non- culturable cells can be identified as live, as long as they have intact cell membrane.
Combining Live/Dead staining and plate count methods provided information about the enumeration of culturable and nonculturable bacteria, but also membrane compromised injured cells. This reflected different levels of damage to the cell membrane due to chemical disinfection. Reports from previous studies have shown that the staining of bacterial cells with SYTO 9 and PI did not always produce distinct “live” and “dead” populations (63, 64). The appearance of yellow or orange stained cells observed in this study indicated an intermediate state of membrane compromised cells and was summarily included with the red cell population. Cells which can only be observed in the green fluorescence channel were defined as viable.
Bacterial suspensions were exposed to disinfectants and quenched with sodium thiosulfate (1 mM/L) when ~90% inactivation was achieved, as determined from
GInaFiT curve fitting results. The 90% inactivation level was chosen for Live/Dead staining experiments, because it was difficult to properly differentiate viability of the three strains at 99% inactivation phase. Bacterial cells, particularly those were alive, were
18
more clearly stained at 90% inactivation. When 99% inactivation was examined, bacteria tended to be either red stained or orange and yellow stained, due to complete or partial cell membrane injury. Stained solution was filtered through black polycarbonate membrane filters (0.45 µm pore size, Millipore) for fluorescence microscopic imaging.
Fluorescent images were observed at 480/500 nm for SYTO 9 and 488/617 nm for PI, respectively. The ratios of viable cell counts (green) to total cell counts were derived from at least 20 images for each sample and experiments were repeated at least twice for each bacterial strain. A fluorescent microscope (IX51, Olympus, USA) equipped with a
100X oil immersion objective was used for image acquisition. ImageJ
(http://rsb.info.nih.gov/ij/) and CellAnalyst (AssaySoft Inc., Fountain Valley, CA) were used for image processing and cell counting.
3.2.6 EPS Extraction and Characterization
EPS extraction of each strain was performed in triplicate on bacterial suspensions prepared from an initial 200 mL bacterial culture harvested at the late-exponential phase
(~16 hours). The modified EDTA extraction method combined with high speed centrifugation was employed as described elsewhere (65). The supernatant from extraction was collected and quantified for total protein and polysaccharide content in
EPS.
Protein concentrations were determined using the modified Lowry Protein Assay
Kit using Comassie blue (Pierce Biotechnology, Rockford, IL) with bovine serum albumin (BSA) as standard. Total polysaccharide concentration was measured using the phenol-sulfuric acid method using glucose as standard (65). Alginate content of the
19
extracted EPS was specifically quantified using the H2SO4-Borate method with sodium alginate as a standard (66).
3.2.7 Disinfectant Decay by Extracted EPS
The extracted EPS was diluted to a series of concentrations ranging from 0 to 200
µg/mL based on the alginate concentration and reacted with chlorine (0.5 mg/L) and monochloramine (2 mg/L) in 1:1 volume ratio respectively. In addition, the same concentration range of bovine serum albumin (BSA) was prepared and interacted with monochloramine as described above for comparison. The initial disinfectant concentration and total exposure time was identical to batch inactivation experiments. At each sampling point, disinfectant concentrations were measured to determine the reaction kinetics using colorimetric methods described earlier. A two-phase decay model adapted from the EPA 1998 model (Equation 2) was used to evaluate reactivity of alginate and tested disinfectants (67). In this model, the rate constants k1 and k2 were assigned for decay rates of the two phases during disinfectant decay, where k1 dominates the fast decay phase and k2 dominates the slow decay phase.
-k1t -k2t C = C0 * [A * e + (1-A) * e ] [2]
3.2.8 FTIR Spectroscopy
Previous studies have shown that IR spectroscopy could be used in microbial disinfection research (68). In this study, Fourier Transform Infrared Spectroscopy (FTIR) was employed to probe both the chemical composition of bacterial cells and extracted
EPS, but also to monitor their alterations (infrared absorbance intensity) induced by the two disinfectants. Batch experiment procedures were followed to disinfect and quench samples. Bacterial cells were harvested by filtering 15 mL of each suspension through a
20
0.2 µm aluminum oxide membrane filter (Anodisc, Whatman Inc., NJ), which has no IR absorbance in the range of 4000-600 cm-1. In EPS reactivity experiments, EPS samples were prepared with an initial concentration of 50 µg alginate/mL and reacted with disinfectants following batch experiment exposure time corresponding to 2-log reduction.
The EPS samples were evenly spread on stainless steel, which contributes no IR spectra, and air dried overnight. The dried samples were analyzed using FTIR in micro attenuated total reflection (ATR) mode. The distinctive functional groups of capsular EPS and inner cell substances could be identified by subtracting the extracted EPS spectra from the bacterial cell spectra.
A FTS-4000 Varian Excalibur Series FTIR spectrometer (Varian, Palo Alto, CA) was used for collecting the infrared spectra. Spectra from 4000 to 400 cm-1 were collected with a resolution of 2 cm-1, and the ordinate was expressed as absorbance. Each spectrum was an average of 256 scans; over 6 spectrums were collected for each sample.
For further data analysis, collected spectra were smoothed with a Gaussian function to eliminated system noise, followed by a second-derivative transformation to amplify the compositional and structural changes in bacterial cellular constituents (69).
3.2.9 Statistical Analysis
Data were presented as mean ± standard deviation. Differences were analyzed with unpaired t-test or one-way ANOVA test at 0.05 significance level using
SigmaPlot (Jandel Scientific, Sausalito, CA).
3.3 Results
3.3.1 Culture and EPS Properties
21
The analysis of the two major components of extracted EPS, protein and polysaccharide, provided evidence of varied EPS production capacities of the three strains and confirmed with previous report (70). The total carbohydrate content was
6.71×10-11, 1.62×10-8, and 2.46×10-8 µg per cell for algT(U), PAO1 and mucA22 strain respectively. The protein content was in a lower magnitude compared to the carbohydrate content, which was 2.18×10-12, 4.38×10-9, and 6.62×10-9 µg per cell for the three strains respectively. The SEM imaging of the tested strains clearly demonstrated a difference in surface morphology, which is attributed to capsular EPS (Figure 3-1).
Figure 3-1: P. aeruginosa strains on agar plates and under SEM. (a) algT(U); (b) PAO1;
(c) mucA22.
3.3.2 Inactivation Kinetics
22
Batch experimental results were interpreted as the fraction of culturable bacteria
[Log (N/N0)] versus Ct values, which were used to assess the susceptibility of cells to disinfectants when a 2-log (99%) inactivation occurred. Repeated tests showed that all three strains were immediately sensitive to chlorine exposure but lagged in response to monochloramine. The inactivation of the three P. aeruginosa strains by 0.5 mg/L of chlorine was rapid (<1 s) for 99% reduction [Figure 3-2(a)]. However, more than 30 minutes was required to reach the same level of inactivation for all three strains studied upon reaction with 2 mg/L of monochloramine [Figure 3-2(b)].
23
Figure 3-2: P. aeruginosa cells inactivation kinetics by (a) chlorine and (b)
monochloramine. Error bars = ±1 standard deviation (N ≥ 12).
Batch experiments with the three P. aeruginosa strains revealed inactivation in the presence of disinfectant occurred at a lower rate for the alginate EPS overproducing strain compared to the wild-type and alginate EPS under-producing strain, given by average Ct values summarized in Table 3-1. The 99% inactivation Ct values of the three
24
strains were compared, using wild-type PAO1 strain with normal EPS production as a baseline. The Ct value required for 99% chlorine inactivation in algT(U) was 20% lower and mucA22 was three times higher than that of PAO1. For monochloramine inactivation, the comparative result was 10% lower [algT(U)] and 18% higher (mucA22).
Table 3.1: Ct values by plate counting and viable ratio by fluorescent staining
Ct (mg·min/l)1 Viable Bacteria Ratio2 Bacterial Strain Chlorine Monochloramine Chlorine Monochloramine
algT(U) 0.0073 68.40 0.01 ± 0.01 0.27 ± 0.05
PAO1 0.0091 75.95 0.13 ± 0.07 0.44 ± 0.10
mucA22 0.038 89.83 0.69 ± 0.08 0.49 ± 0.05
1Ct values required for 99% inactivation according to inactivation kinetic model fitting.
2Ratio of live cell versus total cell at the 90% inactivation time obtained from plate counting. The results are expressed as mean ± standard deviation of two independent experiments in duplicate (N≥80). All pairwise comparisons among the three tested strains were statistically significant in each group (P < 0.05, Tukey test).
3.3.3 Model Fitting
Table 3-2 shows the summary of disinfection kinetic parameters derived from inactivation curve fitting conducted with GInaFit. Monochloramine inactivation tests revealed a lag time at the initial period of inactivation followed by rapid inactivation and then a tail for all strains. The shoulder length (SL) was calculated by multiplying lag time with monochloramine concentration. The algT(U) strain showed shortest lag time (13 min) followed by PAO1 (25 min) and mucA22 (32 min). Inactivation rate could be
25
evaluated using the Kmax value calculated from the model (Equation 1). A higher Kmax indicates a more rapid inactivation rate. The modeling fitting results showed that the algT(U) strain exhibited highest Kmax during inactivation and least survival in tail, followed by PAO1 and mucA22 strain. In chlorine inactivation experiments, which exhibited rapid inactivation rate followed by a tail with no lag time, mucA22 was comparatively more resistant, signified by the lowest Kmax [74% lower than Kmax for
PAO1]. In comparison, algT(U) exhibited the highest Kmax [10% higher than Kmax for
PAO1] representing the highest inactivation rate and most susceptibility to disinfectants.
Table 3.2: Disinfection kinetic parameters from inactivation model fitting
algT(U) PAO1 mucA22
mean SE1 mean SE1 mean SE1
Chlorine 2 -1 a a b Kmax (min ) 118.80 24.43 107.60 20.68 27.84 5.18 disinfection
Monochloramine SL3 (mg·min/L) 26.75a 6.48 50.22b 3.68 63.70b 4.74
2 -1 a a a disinfection Kmax (min ) 0.25 0.05 0.17 0.02 0.15 0.04
1SE: standard error. The results are expressed as mean ± standard deviation of three independent experiments in duplicate (N ≥ 12).
2 Kmax: inactivation rate.
3SL: shoulder length. Units correspond to Ct value. Division of SL by concentration provides inactivation lag time.
Values not followed by a common letter are statistically different from each other (P <
0.05, Tukey test).
26
3.3.4 Viability Analysis by Fluorescent Staining
Bacterial viable ratio was expressed in the form of the mean and standard deviation outlined in Table 3-1. Based on the one-way ANOVA test, there was a difference in viable ratio among these three strains (P<0.05). In chlorine inactivation, viable ratio of mucA22 strain was 5.3 times higher than that for PAO1, while algT(U) survival was not significant. For monochloramine inactivation, the viable ratio of PAO1 was 39% lower than mucA22 and 11% higher than algT(U) strain.
3.3.5 Disinfectant Decay
Figure 3-3 and Figure 3-4 illustrate disinfectant decay for the three P. aeruginosa strains and extracted EPS. For the whole cell disinfectant decay test (Figure 3-3), chlorine was rapidly consumed during the initial period of inactivation. The residual concentrations after 99% of inactivation were similar for all three strains. Limited by short inactivation times and high reactivity, significant differences in disinfectant residual among strains were not observed. Monochloramine, known to be a slow-reacting disinfectant, was consumed differently by bacterial cells with varying EPS amount
(P<0.05). The algT(U) mutant consumed the greatest amount of monochloramine, while higher levels of residual monochloramine were observed in the mucA22 inactivation tests.
For the disinfectant decay test with extracted alginate-based EPS, rapid reaction with chlorine was evident within short time periods [Figure 3-4(a)]. However, EPS reactions with monochloramine were slower, even under an extended exposure time [Figure 3-
4(b)]. Compared to the extracted EPS, BSA reacted much faster with monochloramine
27
[Figure 3-4(c)], where monochloramine concentration declined quickly within the first 20 minutes of interaction then decay slowed for the remaining exposure.
Figure 3-3: Disinfectant decay by P. aeruginosa cells. (a) chlorine; (b) monochloramine.
Error bars = ±1 standard deviation (N ≥ 12).
28
29
Figure 3-4: Disinfectant decay by extracted EPS and BSA standard. (a) chlorine reaction
with EPS; (b) monochloramine reaction with EPS; (c) monochloramine reaction with
BSA. Symbols: alginate concentration in the extracted EPS or BSA standard solution
concentration (µg/mL)
3.3.6 FTIR Analysis
Infrared absorbance intensities of both the bacterial cells and the extracted EPS spectra were monitored to evaluate alterations and reaction of capsular EPS (functional group changes) with disinfectants. Figure 3-5 depicts the representative FTIR spectra
(4000-400 cm-1) of the untreated and disinfectant treated (2-log inactivation) P. aeruginosa bacteria cells [Figure 3-5(a)] and extracted EPS [Figure 3-5(b)]. The spectral bands primarily impacted by disinfectant exposure were located at 3700-2700 cm-1 and
1800-900 cm-1. Characteristic band assignment for FTIR spectra are based on previous studies (Table 3-3) (69, 71).
30
Figure 3-5: FTIR spectra of (a) untreated and chlorine treated bacterial cells. Symbols:
, , ; (b) untreated and chlorine treated
extracted EPS. Symbols: , ,
. The spectra have been vertically shifted for clarity.
Table 3.3: Assignment of characteristic bands of FTIR spectra 31
Wave number (cm-1) Associated bands
1745-1680 Amide I (C=O stretching vibration)
1650-1580 Amide I and Amide II (N-H deformation)
1620-1590 Amide II (N-H deformation and C-N stretching vibration)
1550-1510 Amide II (N-H deformation and C-N stretching vibration)
1460-1200 Carbohydrates (C-H and O-H deformation vibration)
1160-1000 Carbohydrates (C-O stretching vibration)
960-730 Carbohydrates (C-H deformation vibration)
Figure 3-5(b) shows functional group contributions of extracted EPS and their response to disinfectant exposure. The polysaccharide alginate has been linked to biofilm bacteria resistance against antimicrobial agents that include many front-line antibiotics
(72). Previous reports indicate that the extracellular alginate is composed of β-1, 4-linked
D-mannuronic acid and its C-5 epimer, L-guluronic acid (37). The presence of these acids in the extracted EPS can be identified from their characteristic bands: the C-O-C stretching of carbohydrate at 1025 cm-1, the C=O stretching of ester group at 1600 cm-1,
-1 the CH2/CH3 vibration at 1425 cm .
Gaussian smoothed and second derivative transformed spectra for chlorine and monochloramine treated bacterial samples are included in Figure 3-6. There were clear differences in the spectral intensity variation among three strains in the region between
1500 and 900 cm-1, indicating the extent of O-H and C-H stretch deformation in carbohydrate bands. Overall, the mucA22 mutant showed the largest absorbance
32
reduction after disinfection. There was an obvious peak decrease in the 1460-1200 and
1160-1000 cm-1 range, indicating the associated reduction of alginate and other polysaccharide band intensities. Additionally, there were peak changes associated with proteins in the amide I and amide II regions. The mucA22 mutant strain showed significant decrease in amide band vibration followed by the wild-type strain PAO1 and the algT(U) strain. In general, chlorine was more reactive than monochloramine, thus greater absorbance intensity reductions occurred for all three strains after exposure. The mucA22 mutant had an absorbance intensity reduction up to 60-70% on average, compared to 40-55% for wild-type strain PAO1 and 20-35% for the algT(U) mutant.
33
34
Figure 3-6: Transformed FTIR spectra of chlorine treated (a, b) and monochloramine
treated (c, d) bacterial cells. (a, c) 1300-900 cm-1; (b, d) 1800-1300 cm-1. Symbols:
, , ; Bold-faced lines: untreated samples;
normal-type lines: disinfectant treated samples.
3.4 Discussion 35
In this study, the EPS overproducing strain consistently yielded a proportionally greater number of culturable cells and higher viability rate, compared to wild-type and
EPS non-producing strains against tested disinfectants. Meanwhile, removal of the EPS greatly increased the susceptibility of the mucA22 strain to both chlorine and monochloramine (Figure 3-7).
Figure 3-7: Inactivation kinetics of EPS removed P. aeruginosa cells by (a) chlorine and
(b) monochloramine. Error bars = ±1 standard deviation (N≥12). 36
A unique observation in this study was disinfectant decay with bacterial cells and
EPS components. It has been suggested that chlorine reacts differently than monochloramine as disinfectant (73). Chlorine is nonselective reactants, rapidly reacting with both organics and inorganics. In comparison, monochloramine is found to be more stable in the water distribution system and provides better control of bacterial regrowth than chlorine. Monochloramine deactivates bacteria by penetration of the cell wall, reacting with amino acids in the bacterial DNA and disrupting the bacterial metabolism
(73). Jacangelo et al. reported that inhibition of typical protein-associated biological activities, e.g., bacterial transport, respiration, and substrate dehydrogeneration, were observed at monochloramine concentrations normally used for disinfection (74). Based on the two-phase disinfectant decay model, the rate constant (k1) dominating the fast decay phase of chlorine-alginate [Figure 3-3(a)] and monochloramine-BSA (protein surrogate) [Figure 3-3(c)] was found to be as high as 0.2-1.8 (min-1), whereas the same parameter for chlorine decay with natural organic matter (NOM) was previously reported around 0.02-0.04 (min-1) (75).
In monochloramine disinfection, alginate-based EPS, a polysaccharide with low observed monochloramine demand [Figure 3-4(c)], could limit access to reactive sites on the bacterial cell membrane, which is more reactive with monochloramine than the alginate component of EPS. When monochloramine in the aqueous solution reaches the bacterial cell surface, the slow reaction with alginate-based EPS on the cell surface may limit monochloramine interaction with the cell membrane (76). The lag time in the inactivation curves and non-severe cell membrane damage in the FTIR results both
37
support the hypothesis that monochloramine inactivation may be initiated through reactions at several sensitive sites on the bacterial surface. The presence of alginate EPS reduces the available reaction sites for monochloramine inactivation on the cell surface, and the slow penetration of monochloramine through the alginate EPS retards the protein inhibition. The decreased monochloramine decay witnessed for the high alginate EPS producing mucA22 mutant indicates that the protective effect of capsular EPS against monochloramine is not attributed to EPS associated disinfectant decay, but instead it limits and delays accessibility to available reaction sites on the cell membrane for monochloramine. In contrast to the monochloramine inactivation, chlorine performs disinfection by fast oxidation of cellular material, which has been reported by many researchers (47). The high EPS producing mucA22 excretes copious materials on the cell membrane that non-selectively react with these disinfectants, thus reducing available disinfectant before cell surface reactions occur. Considering these two different inactivation mechanisms, alginate EPS could limit microbial deactivation in two different ways, either acting as a disinfectant demanding substrate or as a reaction limiting material
(reduced monochloramine penetration).
The results from fluorescent staining offered further support that capsular EPS provide a protective role for bacteria in the inactivation process, contributing to a viable but non-culturable (VBNC) state found in disinfected microorganisms (77). The presence of capsular EPS necessitated longer times for bacterial inactivation by disinfectants.
Limited PI penetration of the cells increases the presence of yellow or orange labeled cells (42, 78), occurring more frequently for the mucA22 mutant strain than the other two strains (Figure 3-8). Both a higher viable rate and more intermediate state for the mucA22
38
indicates a greater possibility that this organism suffered from partial membrane injury and not from lethal membrane damage, when compared to the other two strains. An interruption of membrane permeability aided by capsular EPS could be a possible reason to explain the appearance of the intermediate state organisms, which may maintain their capacity to regrow and cause public health problems.
Figure 3-8: LIVE/DEAD stained mucA22 cells after chlorine disinfection. From left to
right: green channel, red channel, combined channel.
To monitor the occurrence of membrane permeabilization and membrane damage, possible leakage of the intercellular material released from cells was measured at 260 nm
(79). Virto et al. reported that exposure of bacterial cells to chlorine in distilled water caused extensive damage to the cell membrane (42). However, the chlorine concentrations used were much higher (as high as 50 mg/l) than residual disinfectant concentrations in the water distribution systems. Our results (data not shown) suggest that leakage of intercellular material was not remarkable for the low disinfectant concentration used in this study. The results support that severe membrane damage or rupture is not a key event in the inactivation by chlorine and monochloramine at residual concentrations. Observations from other researchers hypothesized that more subtle events, 39
such as enzyme activity or functional group deformation in the cell membrane, are involved in the bactericidal mechanism of free chlorine (80). Functional group deformation in the cell membrane may alter membrane permeability and increase the probability of disinfectant entry into bacteria cells. To confirm the functional groups deformation, FTIR analysis was employed using both bacterial cells and extracted EPS.
FTIR results revealed a decrease in IR absorbance for the C-O-C stretching vibrations of polysaccharides (1200-900 cm-1), indicative of the cell wall peptidoglycan layer and lipopolysaccharide outer leaflet was observed (71); however, the associated peak was not eliminated, potentially indicating non-severe membrane damage.
Further analysis of FTIR provided evidence indicating the changes of chemical composition and concentration for both bacterial cells and extracted EPS occurred after exposure to disinfectants (Figure 3-5). A higher reduction in absorbance intensity of protein and polysaccharide groups for mucA22 was observed, indicating that greater amounts of these components interacted with disinfectants. Comparing the three strains at the same inactivation threshold, infrared spectra show more polysaccharide and protein content react with the disinfectant for the mucA22 strain, which is attributed to reactions with capsular EPS. Macromolecules of bacterial EPS on the outer membrane possess active functional groups, including proteins and polysaccharides, which play a significant role in the interaction of the cell wall with the surrounding environment (81). Similarities in infrared spectrum when comparing bacterial cells and extracted EPS likely indicate that the spectra of the intact bacteria cells largely reflected the properties of the cell wall and cell wall associated EPS. Therefore, the distinctive functional groups of capsular EPS and inner cell substances could be identified by subtracting the extracted EPS spectra
40
from the bacterial cell spectra. In a study considering only whole cell spectra, a decrease of infrared absorbance intensities for corresponding wave number ranges were observed after disinfection tests, providing information about deformation and reduction of characteristic functional groups in the cell membrane of E. coli (68).
In comparison of functional group interaction with the two tested disinfectants, it is clear that these disinfectants react very differently with cellular biomolecules (Figure
3-6). Peak area reductions reveal that protein band reductions for each disinfectant are similar (~80% for all disinfectants). However, reductions in polysaccharide bands for monochloramine disinfectant are much less significant as compared to chlorine, whose polysaccharide reduction is comparable to protein reduction. This data confirms results obtained from disinfectant decay and the extracted EPS spectra [Figure 3-4(b)].
3.5 Conclusion
In summary, the results from this study provide a mechanistic understanding regarding the protective role of capsular EPS in response to chlorine and monochloramine. In depth evaluation of the capsular EPS from both whole cell and extracted EPS analyses revealed that capsular EPS may enhance bacterial survival in two distinct ways. Increased amount of EPS on cell surface was associated with accelerated decay of chlorine, thus reducing availability and efficacy of chlorine residual for microbial inactivation. Alternatively, alginate EPS had minimal monochloramine demand, although the increased presence of alginate EPS was shown to reduce the efficacy of monochloramine, which showed high reactivity with proteins. Thus, the alginate EPS obscured disinfectant reactive sites on cell surface and retarded bacterial membrane interaction with monochloramine. In addition to EPS reactivity with different 41
disinfectants, capsular EPS on cell membrane appeared to reduce membrane permeabilization by oxidative disinfectants, which was observed by functional group deformation. Functional moieties comprising the bacterial cell and capsular EPS were not eliminated, but were instead reduced in magnitude upon exposure to disinfectants. This suggests that the extensive membrane damage does not occur given low concentrations of chlorine-based disinfectant residuals in water distribution systems. The combined results supported that capsular EPS, either by consuming disinfectant or limiting direct cell membrane access, provide a protective role for bacterial cells against regulatory residual disinfectants by reducing membrane permeabilization.
Acknowledgement
This research was supported by the US Geological Survey (Project Number:
2009OH89B) and the National Science Foundation (CBET0933288). And we greatly appreciate Dr. Mau-yi Wu for his input.
42
Chapter 4
Exopolysaccharide Production by Pseudomonas aeruginosa Contributes to Biofilm Architecture and
Chlorine Tolerance of Biofilm and Detached Clusters
Abstract
In this study, multiple roles of biofilm EPS were assessed in respect to susceptibility of biofilm and biofilm detachment upon chlorine disinfection. Strains from an opportunistic pathogen, Pseudomonas aeruginosa (both wild type and mutant strains) with different extracellular polymeric substance (EPS) secretion capabilities were tested.
The impact of biofilm EPS quantity on disinfection was evaluated by monitoring biofilm viability, biofilm structure, biosorption of dissolved organic matter (DOM), and viability of detached biofilm simultaneously during chlorine disinfection. The obtained results suggested that the presence of EPS increased biofilm and detached biofilm resistance to chlorine in both presence and absence of DOM. The biofilm structure characterization using confocal laser scanning microscopy revealed that EPS production had an effect on biofilm architecture. These structural characteristics were related to both overall biofilm viability and the spatial distribution of viability within the biofilm. Additionally, the 43
increased amount of EPS influenced biosorption of DOM. However the DOM adsorption did not show significant impact on biofilm viability against chlorine disinfection.
Meanwhile, viability of detached biofilm clusters, particularly the EPS overproducing strain, was significantly increased in the presence of DOM. The combined results suggested that biofilm EPS played multiple roles toward influencing the susceptibility of both biofilm and detached biofilm by increasing resistance to disinfectant.
4.1 Introduction
The biological stability of water in distribution systems is a major concern for drinking water quality control. During drinking water distribution, bacterial regrowth may lead to a deterioration of water quality and proliferation of microorganisms, despite the presence of residual disinfectants (82). Facilitated by the large surface area of the distribution pipeline and the production of extracellular polymeric substances (EPS), at least 95% of the total bacterial biomass in water distribution systems is found as biofilm
(83). Therefore, studying the efficiency of disinfectant residuals is of great importance to control biofilm growth in drinking water distribution systems (30).
Known as the major structural components of biofilms, EPS enhance bacterial adhesion, promote structural development of biofilm, provide a protective barrier, as well as adsorb and store nutrients for biofilm growth. Among these roles, the protective role of
EPS has been reported to significantly increase biofilm tolerance to antibiotics and disinfectants (84, 85). Mechanisms of EPS protection previously reported include transport limitation of biocides through the EPS matrix, sacrificial reaction of EPS with biocides, and formation of a nutrient gradient within the biofilm resulting in complex phenotypes of cells (18). However, EPS may affect other characteristics of biofilm which 44
may also influence its susceptibility against disinfectants. Specifically, EPS excretion strongly influences development of biofilm structure and its interface with the fluid environment (1). Furthermore, as result of their composition and influence on biofilm structure, EPS also provide absorption sites for unremoved dissolved organic matter
(DOM) entering into the distribution system (31, 86). However, the influence of biofilm structure and biosorption of disinfectant demanding substances (such as DOM) on biofilm resistance have not been quantitatively studied, specifically in regard to the EPS roles.
Beyond their significant influence on the susceptibility against disinfectant, EPS may affect fate and redistribution of detached biofilm clusters, which is an essential component in biofilm life cycle (87). Biofilm detachment returns sessile cells to the bulk solution. These detached cells are redistributed and colonize on new surfaces, which may adversely influence water quality and increase public health risks. While most previous studies have focused on biofilm disinfection and disinfectant penetration in biofilm (88), susceptibility of detached biofilm against disinfectants remains elusive.
Previous studies investigating biofilm detachment have quantified the frequency of detachment and size of cell clusters (89, 90); however, resistance of detached biofilm clusters considering disinfection process dynamics has not been explored. Among few available studies, Fux et al. (2004) found that the antibiotic resistance of detached biofilm was size dependent, attributed to nutrient and diffusion limitations. Behnke et al. (2011) compared chlorine tolerance of planktonic culture, attached biofilm and detached biofilm clusters, drawing the same size dependency conclusion. However, these studies tested detached biofilm susceptibility by reacting biofilm clusters collected from effluent
45
samples with antibiotics or disinfectants, without considering competitive disinfectant consumption exerted by organic components in biofilm and other disinfectant demanding substances simultaneously under the flowing conditions. Accordingly, examining combined roles of EPS on susceptibility of both biofilm and detached biofilm clusters can provide important information regarding minimal disinfectant residuals to control biofilm and prevent bacterial redistribution in drinking water distribution systems.
To address the current knowledge gap, this study assessed multiple roles of biofilm EPS in respect to biofilm and detached biofilm susceptibility to a common disinfectant (chlorine). Specifically, this study investigated i) the effect of biofilm EPS on biofilm structure, disinfectant transport and DOM adsorption; ii) the role of biofilm structure on the spatial distribution of viable cells in biofilm; iii) the influence of DOM on susceptibility of biofilm and detached biofilm clusters during chlorine disinfection.
4.2 Materials and Methods
4.2.1 Bacteria Culture
P. aeruginosa is an opportunistic pathogen which has been identified in water distribution systems (48). Wild-type PAO1 and two mutant strains, algT(U) (inhibited alginate EPS production) and mucA22 (overproduction of alginate EPS), were selected to construct confluent biofilms (Figure 3-1). All strains were grown in one-tenth strength
LB broth (2.5 g/L, Difco Laboratory, Detroit, MI) at 37°C and then harvested during the late-exponential phase by centrifugation at 2,000 × g for 15 min. The cells were diluted in chlorine demand free (CDF) buffer (0.54 g Na2HPO4 and 0.88 g KH2PO4 per liter, pH=6.98) as a bacterial suspension (OD600 = 0.5 ± 0.02) (54).
4.2.2 Solution Preparation 46
Biofilms were cultivated in 0.02 strength LB broth to create nutrient limited growth conditions (91, 92). Biofilm interaction with DOM was conducted using
Suwannee River NOM (SR-NOM; IHSS, MN) filtered through a 0.45 µm membrane and applied at a final concentration of 2 ± 0.2 mg/L (93). Chlorine solutions were prepared by adding Clorox bleach (The Clorox Co., Oakland, CA) to autoclaved deionized water.
The chlorine concentration was determined with the N, N-diethyl-p-phenylenediamine
(DPD) method using a DR/2700 spectrophotometer (HACH Company, Loveland, CO)
(94).
4.2.3 Biofilm Cultivation and Disinfection in Flowcell Systems
Biofilms were cultivated in two channel flowcell systems (BioSurface
Technologies Corp., Bozeman, MT) fitted with a glass microscope slide opposing a glass cover slip (channel dimensions, 1.6 by 12.7 by 47.5 mm; flow rate = 0.2 mL/min) at room temperature (22 ± 2 oC) (Figure 4-1) (95). Two carboys were used as nutrient media and chlorine supply reservoirs, respectively. All feeds were delivered using a multichannel peristaltic pump (ISMATEC, Glattbrugg, Switzerland). Flow cells, tubing and solutions were sterilized prior to each experiment. The flow cell channels were aseptically inoculated with bacterial suspension and incubated 2 hours without flow for initial bacterial attachment. Then, flow was introduced and gradually increased to 0.2 mL/min. Biofilm was grown six days to reach structural maturity (89). For each flowcell, one channel was used as a control, while disinfectant was applied to the other. Chlorine concentration was maintained at 0.5 mg/L at the flow cell inlet throughout the disinfection process. For DOM experiments, media containing DOM was mixed with chlorine in a bubble trap (HRT = 25 min) before entering the flow cells. Chlorine feed
47
concentration was adjusted to compensate for the rapid chlorine demand exerted by DOM at the point of mixing (96), so inlet chlorine concentration could be maintained at 0.5 mg/L. Flowcell effluent was collected every 30 minutes for 2 hours during the disinfection process and quenched with 0.1 M sodium thiosulfate before further analysis.
The total disinfection time was chosen based on the chlorine decay kinetic with biomass in water distribution systems (97). For each bacterial strain, the experiment was repeated more than three times.
Figure 4-1: Diagram of flowcell system set up.
4.2.4 Confocal Laser Scanning Microscopy and Image Analysis
The biofilm content on glass slides was discriminated using BacLight
LIVE/DEAD staining kit (Molecular Probes Inc.) to differentiate live and dead cells.
Extracellular polysaccharides content in the biofilm were visualized by applying Alexa
633 conjugated concanavalin A (ConA-Alexa 633; 1 mg/mL in stock solution). This stain specifically targets the polysaccharides (D-glucose and D-mannose residues) that are
48
major EPS components of P. aeruginosa (98). A mixture of stains was prepared in CDF buffer with a final concentration of 1.5 µl/mL each for SYTO 9 and propidium iodide
(PI), as well as 20 µl/mL of EPS stain. The mixture was then injected into flow channels and incubated in the dark for 15 min. Live SYTO 9-stained cells, dead PI-stained cells, and ConA-Alexa 633 stained EPS were visualized with a Leica SP5 Confocal Laser
Scanning Microscope (CLSM) equipped with a 63× oil immersed objective and a 20× dry objective. An argon laser, at 488, 561 and 633 nm, was used as the excitation source for the fluorescent probes. For each biofilm sample slide, at least five positions were randomly selected for image acquisition and further image analysis.
CLSM images were further processed using the image processing program
COMSTAT to determine total biomass, EPS content, and biofilm structural parameters, as defined in detail elsewhere (99). Important structural parameters discussed in this study include: i) the roughness coefficient - a measure of how the thickness of biofilm varies; ii) surface area to volume ratio - representing the spatial complexity of biofilm structure; iii) diffusion distance - defined as the shortest distance from a pixel containing biomass to a pixel not containing biomass. Average diffusion distance is the average of diffusion distances for all pixels containing biomass. This parameter indicates the extent of void spaces in the biofilm structure. EPS content per unit area was generated from image analysis to evaluate the role of EPS on biofilm structure and biofilm susceptibility to disinfectant.
4.2.5 Bacterial Enumeration and Viability
After disinfection and microscopic analysis, biofilms from both control and disinfected channels were scraped from glass slides and resuspended in CDF buffer.
49
Viable cells were enumerated using plate count method on R2A agar plates (Difco
Laboratories, Detroit, MI) in duplicate. The chlorine disinfection efficacy was evaluated using the ratio of viable bacteria in disinfected biofilm over the control biofilm value.
Enumerations of viable bacterial cells in the effluent samples followed the same procedure.
4.2.6 Flow Cytometry Analysis
While plate count enumeration yields counts of culturable cells, flow cytometry analysis of the detached biofilm was used to differentiate and quantify dead, injured and live cells as a function of fluorescence intensity, based on the extent of membrane damage. Data was acquired in “log” mode using a FACScalibur flow cytometer (BD
Biosciences, San Jose, CA). The flow cytometer was equipped with an argon laser set at
15 mV and an excitation wavelength of 488 nm. PI and SYTO 9 were used in combination to determine membrane compromised cells and intact cells, respectively.
Stains were simultaneously added at concentrations of 0.15 µl PI and 0.1 µl of SYTO 9 to
1 mL of sample and incubated as described above. Cell concentration was determined by comparing cell events to events from a microsphere standard of known concentration
(InVitrogen, Carlsbad, CA).
On the basis of negative and positive controls, flow cytometry analysis was performed on two fluorescent channels (PI and SYTO 9) to evaluate the cellular viability.
Each acquired data plot was analyzed using WinMDI (J. Trotter 1993-1998) in four quadrants (100): (A) PI positively stained dead cells with a permeabilized cell membrane;
(B) both PI and SYTO 9 positively stained membrane compromised cells; (C) SYTO 9 positively stained live cells with intact cell membrane; (D) negative signals.
50
4.2.7 Dissolved Organic Matter (DOM) Analysis
During the six day biofilm cultivation, effluent samples from biofilm and abiotic flow cell channels were collected for DOM analysis. Five samples per day were collected from each flow channel. Collected effluent samples were filtered through 0.45 µm membrane filters to remove bacterial cells and other impurities. The filtered samples were analyzed in triplicate using a UV spectrophotometer at 220-350 nm (UV-1800,
Shimadzu) to estimate DOM retention by biofilms.
4.2.8 Statistical Analysis
Data are presented as mean ± standard error. Differences were analyzed using unpaired t-test or one-way ANOVA test. The association between biofilm structural parameters, EPS content and biofilm viability was analyzed by linear regression analysis. P<0.05 was accepted as statistically significant. All calculations were performed by using SigmaPlot (Jandel Scientific, Sausalito, CA).
4.3 Results
4.3.1 Effect of EPS Content on Biofilm Structure
After six days growth, biofilm composition and structure was quantitatively analyzed using CLSM (Table 4-1). Image analysis revealed the EPS content in biofilm was 0.12 ± 0.03 µm3/µm2 for algT(U) biofilm, 0.87 ± 0.32 µm3/µm2 for PAO1 biofilm, and 1.4 ± 1.5 µm3/µm2 for mucA22 biofilm. Despite clear differences in biofilm EPS content, biofilm from all strains possessed a similar level of bacterial cell biomass (18 ± 3
µm3/µm2). However, EPS production was determined to significantly affect biofilm structural parameters, such as roughness coefficient, surface area to volume ratio, average diffusion distance. These three parameters were found to be negatively related to the EPS 51
amount (P<0.001). No significant changes in biofilm structural parameters, cell biomass and EPS content were observed when comparing control and disinfected biofilm for all strains.
Table 4.1: Biofilm structural parameters and biofilm EPS content
algT(U) PAO1 mucA22
Roughness Coefficient 1.00 ± 0.26 0.69 ± 0.26 0.41 ± 0.41
Surface area to volume ratio 2.31 ± 1.28 1.99 ± 0.94 1.15 ± 0.36 (µm2/µm3)
Average diffusion distance (µm) 0.80 ± 0.20 0.58 ± 0.06 0.42 ± 0.13
EPS amount (µm3/µm2) 0.12 ± 0.03 0.87 ± 0.32 1.40 ± 1.50
Total biomass (µm3/µm2) 16.83 ± 6.26 15.43 ± 2.44 21.54 ± 5.82
Values represent average ± standard error (N≥15).
4.3.2 DOM Retention by Biofilm
The spectrospic study of DOM revealed both overall and specific adsorption of different fractions of DOM by biofilms. The area under the spectra between 250 and 350 nm (A250-350)was used to measure the overall DOM concentration (Figure 4-2). Around
20% reduction in DOM concentration was observed from the second day through the sixth day of biofilm growth compared to the concentration at the beginning of experiment.
In general, there was no significant difference in DOM adsorption among the three biofilms during the 6-days biofilm growth, although DOM concentration was slightly
52
lower in the mucA22 bioiflm effluent than the concentration in PAO1 biofilm effluent than the algT(U) biofilm.
Figure 4-2: DOM adsorption by biofilms.
In order to characterize the selective adsorption of different fractions of the DOM, area under the spectra between 220-230 nm (A220-230) was compared to the area between
240-260 nm (A240-260). The 220-230 nm band is within the benzenoid (Bz) band range representing aromatic rings substituted predominant with aliphatic functional groups. The
240-260 nm band assigned as the electron transfer (ET) band is related to the presence of polar functional groups such as hydroxyl, carboxyl and ester groups on the ring. In Figure
4-3, A220-230/A240-260 ratio was increased differently by the three biofilms from day 2 to day 6 when compared to the ratio at the beginning of the experiment. The increase in
A220-230/A240-260 ratio suggested a tendency of selective adsorption of aromatic rings substituted with polar functional groups. Furthermore, the ratio was statistical higher for 53
EPS producing strains than the EPS deficient strain, indicating that the presence of EPS may be related to the selective removal of DOM molecules with hydroxyl, carboxyl and ester groups.
Figure 4-3: Selective removal of different DOM fractions by biofilms.
4.3.3 Biofilm Susceptibility to Chlorine Disinfection
To better understand EPS influence on the biofilm disinfection process, both heterotrophic plate count (HPC) and CLSM image analysis from control and disinfected channels were interpreted to determine both overall and the spatial distribution of biofilm viability upon disinfectant exposure.
The HPC results of resuspended biofilm showed the mucA22 biofilm had the highest viable ratio (34.33 ± 0.15%), followed by the PAO1 biofilm (19.78 ± 0.18%), and the algT(U) biofilm (0.67 ± 0.26%), where no significant difference in overall biofilm viability was determined considering the presence or absence of DOM (P = 0.613). More 54
importantly, the overall viability ratio was greatly enhanced by higher EPS content in biofilm (P<0.001) (Figure 4-4). In addition, the overall biofilm viability ratio was negatively correlated to the roughness coefficient and average diffusion distance (P<0.05).
Figure 4-4: Relationship between biofilm EPS content and biofilm overall viability ratio.
The relationship was analyzed by linear regression with goodness of fit separating the
presence and absence of DOM conditions.
Beyond the overall biofilm survival, a spatial distribution of biofilm viability was analyzed to investigate disinfectant efficacy among strains using CLSM image analysis.
To evaluate the spatial distribution, data was acquired by imaging in 1 µm depth increments throughout the image acquisition process, then calculating a viability ratio of live over total cells. Data was plotted as the distance from the substratum to the biofilm
55
surface versus the viability ratio determined in each slice, where each curve represents an individual strain and condition (Figure 4-5).
Figure 4-5: Spatial distribution of control and disinfected biofilm viability ratio in the absence of DOM. The vertical axis represents the distance from the substratum increasing
to the biofilm surface, where the viability ratio (horizontal axis) was determined at each
cross-section.
A similar pattern in viability ratio distribution was observed for all strains in control channels, ranging from 70 to 100% viability with minimum viability located in the middle biofilm depth. For disinfected biofilm, the maximum viability ratio determined for algT(U), PAO1 and mucA22 biofilms was less than 10%, 10-15% and up to 30% respectively. The disinfectant impact depth, defined as the distance from the top surface of biofilm to the point where maximum viability ratio was observed, was 20-30
56
µm for algT(U) biofilm, 15-20 µm for PAO1 biofilm, and about 10 µm for mucA22 biofilm. Compared to the other strains, significant viability ratio decrease in mucA22 biofilm was observed only at the surface layer, and the viability ratio remained constant as depth increased into the middle section of biofilm. Among all strains, a decline in stratified viability was observed approaching the substratum.
4.3.4 Detached Cell Viability during Biofilm Disinfection
As an element of biofilm lifecycle, the detached biofilm viability was investigated using flow cytometry and HPC. HPC results indicated an increased viability ratio for detached biofilm from the EPS overproducing strain (Figure 4-6), while the deficient strain was most susceptible to disinfection. In the absence of DOM, only clusters detached from algT(U) biofilm reached and maintained greater than 2-log reduction (99% inactivation) at all sampling points. Clusters detached from PAO1 biofilm reached 2-log reduction after 60 minutes. However, the mucA22 clusters were seen to maintain a much higher viability ratio, reaching only 1-log reduction after 2 hours biofilm disinfection. In the presence of DOM, the viability ratio increased for all strains, but was most pronounced for the mucA22 strain. Even the highly susceptible algT(U) detached biofilm could not reach 2-log reduction. PAO1 and mucA22 detached biofilm showed less than 1- log reduction. The viability ratio for PAO1 biofilm clusters was around 3%. The viability ratio of mucA22 biofilm clusters was as high as 30% at the first sampling point and was reduced to a constant value of approximately 12% after 1 hour biofilm disinfection.
57
Figure 4-6: Detached biofilm cluster viability ratio determined by HPC in the presence
and absence of DOM. Data points represent mean ± 1 standard error.
In addition to HPC, flow cytometry provided a measure of both cell concentration and cell viability. The detached cell concentration was found consistent for all tested strains and conditions (~106 cells/mL). To analyze cell viability, the flow cytometry acquired a data plot of PI versus SYTO 9 fluorescence intensity and analyzed in four quadrants [Figure 4-7 (a)]. Reports from previous studies have shown that the staining of bacterial cells with SYTO 9 and PI did not always produce distinct “live” and “dead” populations (63, 64). The appearance of yellow or orange stained cells observed in previous study indicated an intermediate state of membrane compromised cells. In this study, the intermediate state cells were observed in quadrant B (upper right), indicating both green and red positive signal. The percentage of live, dead and membrane
58
compromised cells were plotted in a bar graph [Figure 4-7 (b, c)], where a comparison of flow cytometry to HPC results revealed a similar trend in viability ratio related to EPS content of tested strains. No significant difference in viability ratio was found for the control samples when comparing the conditions of with and without DOM. However, when the disinfected samples were examined, more SYTO 9 positive signals were observed (rightward in the plot) under the DOM condition, indicating higher viability ratio for cells during disinfection. Considering the presence of DOM, the portion of dead cells in detached biofilm was reduced by 8%, 24% and 38% for algT(U), PAO1, and mucA22 strains, respectively. Consequently, the live cell portion increased significantly for algT(U) and PAO1 detached biofilm; however, this increase was not as substantial for the mucA22 as the other strains. Instead, with the addition of DOM, a noticeable increase in the injured cell percentage was observed for the mucA22 detached biofilm, which was 4.7 times higher compared to the absence of DOM.
59
60
Figure 4-7: Detached biofilm viability quantified using flow cytometry. (a) flow
cytomery data in dot plot: Quadrant A) PI positive signal (dead cells); Quadrant B) PI &
SYTO 9 positive signal (injured cells); Quadrant C) SYTO 9 positive signal (live cells);
Quadrant D) negative signal. Cell viability percentages by strains for P. aeruginosa
detached biofilm: (b) in the absence of DOM; (c) in the presence of DOM.
4.4 Discussion
4.4.1 Influence of EPS on Biofilm Structure and Biofilm Susceptibility during
Chlorine Disinfection
In this study we observed a relationship among alginate EPS content, biofilm structure and biofilm susceptibility to chlorine disinfection. Alginate EPS overproduction has been reported as a primary cause of differences in biofilm architecture (72). Biofilms formed by an alginate overproducing P. aeruginosa strain exhibited enhanced microcolony formation and greater structural maturity (101). In this study, overall biofilm thickness was around 30-40 µm, which was similar to what Hentzer, et al. (2001) reported. The EPS overproducing mucA22 biofilm was found to have a lower biofilm roughness coefficient, surface area to volume ratio and average diffusion distance when compared to the wild type PAO1 biofilm and EPS deficient algT(U) biofilm. Overall biofilm viability was greatly enhanced by increased EPS content in biofilm structure, which conforms to a previous study showing that a mucoid strain produces a biofilm with high antibiotic resistance (88).
The biofilm structural characteristics induced by EPS also appeared to impact biofilm viability. After initial cell attachment, microcolony formation with EPS 61
overproduction may result from daughter cells originating from the attached bacterium enveloped in the EPS matrix and held in closer proximity to the parent cell (72). The presence of EPS reduced the void area between adjacent cells, which results in lower surface area to volume ratio and lower diffusion distance, hence reduced disinfectant transport into biofilm.
Previous study showed that the gel-like EPS matrix may protect the deeper layers of cells from antibiotics or disinfectants by permitting limited diffusion of these chemicals into biofilm (102). Besides the slow penetration of biocides, other hypotheses for biofilm resistance were discussed, such as altered microenvironment with nutrient and oxygen depletion or the emergence of resistant phenotypes (103). These hypotheses predicted increased resistance and survival of bacterial cells near the substratum.
However, in contrary to previous prediction, a unique spatial distribution pattern of viability ratio was observed in this study. Excluding the algT(U) biofilm, which was almost fully disinfected throughout the thickness, the viability ratio distribution within
PAO1 and mucA22 biofilm begins near 0% at the biofilm surface then gradually increased with depth inside biofilm. In the middle section, a maximum viability ratio was reached; however, the viability ratio in the bottom section of biofilm began to decrease when approaching the substratum. The pattern of this viability ratio distribution was found similar to the distribution of biomass occupying each horizontal cross-section of biofilm (Figure 4-8). Similar in all three strains, the middle section of biofilm had a more compact structure, possessing a higher density of biomass, when compared to the top and bottom sections. This more compacted interior structure may retard disinfectant transport in biofilm and may also be associated with low metabolic activity, oxygen and nutrient
62
limitation, which promote tolerance against disinfectants (1). The bottom section of all three biofilms had a lower potion occupied by biomass when compared to their middle section. Overall, this distribution pattern could be related to the emergence of a mushroom shaped biofilm structure, which was reported by previous studies (104, 105).
The reduced biomass density in the bottom section, resulting from greater void space near the substratum, promotes disinfectant transport and inactivation efficiency.
Figure 4-8: Spatial distribution of biomass through horizontal cross-sections. Biomass
content in each cross-section determined using COMSTAT. Biofilm thickness is
normalized for strain comparison. Error bars represent one standard error.
4.4.2 Influence of EPS on DOM Retention by Biofilm
The presence of EPS promotes retention of dissolved or colloidal organic substances by providing absorption sites. Dissolved organic matter is one of the organic substances found in the water distribution system and known to exert disinfectant demand.
63
DOM, which is not completely removed by conventional water treatment processes, is known to contribute to biofilm formation and bacterial regrowth in the water distribution systems (82, 93). However, reduction of organic carbon levels to inhibit regrowth requires very high level of treatment before water would enter the distribution system.
With conventional water treatment processes, 2-5 mg/L effluent DOC concentration has been commonly reported by water utilities (106). In this study, a moderate DOM concentration (TOC = 2 ± 0.2 mg/L) was chosen to simulate distribution system.
The UV254 measurement revealed that the DOM removal by biofilm was approximately 2-5% compared to a control channel without biofilm. Our previous study investigating DOM biosorption by UV inactivated planktonic Pseudomonas strains indicated that adsorption equilibrium was reached within 30 minutes and EPS quantity influenced DOM biosorption (107), suggesting DOM biosorption is limited by the available absorption sites on cell biomass. However, the difference in DOM retention by biofilm among the three strains was not significant, except on the third and the fourth days (expansion stage) of biofilm growth (104). Possible explanation could be largely related to DOM biosorption as the biodegradable fraction constitutes a small portion of
DOM (93). During the initial phase of biofilm formation, due to limited adsorption sites, a difference in biosorption among strains was not observed. In the biofilm expansion phase, cell growth and division is accelerated, producing more EPS, which provide absorption sites to retain DOM, as well as increasing the complexity of biofilm structure.
Consequently, the EPS over-producing mucA22 strain had a higher DOM retention than the other strains. Upon formation of ample cell and EPS biomass, the biofilm matrix is built up, structural complexity is formed, and, the biofilm enters into the maturation stage.
64
During the mature stage (~ 5 to 6 days), the fast biosorption kinetic likely controls DOM retention, due to the saturation of DOM onto available biomass adsorption sites as DOM was continuously fed into the system. Thus, a difference in retention among strains disappeared.
4.4.3 Influence of DOM on Biofilm and Biofilm Detachment Susceptibility
Cell biomass detached from biofilm is thought to display a transitional phenotype between attached and planktonic cells. The detached biofilm encompass a wide size distribution, including single cells and clusters comprised of hundreds of cells (87). These large clusters may provide protection for embedded cells and shield them from disinfectant exposure (108).
Another aspect providing greater tolerance for bacteria in larger clusters is the presence of EPS, which have been known to sacrificially react with oxidative disinfectants, such as chlorine (1). In our previous study, P. aeruginosa EPS showed very rapid reactions with chlorine, yielding a reaction rate nearly ten-fold greater than that of
DOM-chlorine reaction. When more chlorine is consumed by the EPS, less will be available to inactivate cells both in the biofilm and in the detached clusters. It can be assumed that EPS content in the detached clusters reflects the EPS content in biofilm matrix, thus the biofilm detached from the mucA22 biofilm will possess more EPS than biofilm detached from the EPS deficient strain [algT(U)]. In addition to the intrinsic protective role of EPS against disinfectant, the DOM both retained in biomass and suspended in the bulk solution exert demand on available chlorine in the system, thus these reactions further reduced available chlorine for biomass disinfection (93).
65
No significant effect of DOM was observed on overall biofilm viability during chlorine disinfection for all strains. However, the detached cell tolerance was significantly enhanced under the presence of DOM, especially the mucA22 strain. The low retention of DOM in the biofilm in comparison to the biomass quantity indicates that adsorbed DOM did not significantly influence biofilm structure and composition within strains. During the disinfection process, chlorine reaches the biofilm surface, gradually being consumed by organic components of biofilm (bacterial cells, EPS and adsorbed organic substances) during penetration into deeper biofilm layers. Even though the mucA22 biofilm accumulated more DOM after six day cultivation, the effect of DOM during disinfection was distributed through the entire biofilm matrix, which limits its effect on total chlorine demand when compared to the chlorine demand of total biomass in biofilm.
On the other hand, when compared to biofilm, the detached clusters have a much higher exposure area to volume ratio without the protective effect provided by the biofilm structure. For the detached biofilm clusters, chlorine may react and be consumed by both retained and bulk DOM immediately with much smaller penetration distance as compared to biofilm, which makes the disinfectant demand of DOM more significant.
Noticeably, the proportion of membrane injured cells in detached clusters greatly increased, especially for the mucA22 strain. These injured cells had partially compromised cell membrane after disinfection, as opposed to dead cells with complete cell membrane damage. Under hospitable conditions, injured cells may recover and recolonize downstream. Considering the possibility of cell revival, the viability ratio of the detached clusters could become even higher than what was been measured.
66
4.5 Conclusion
In this study, multiple roles of biofilm EPS were assessed in respect to both biofilm and detached cluster susceptibility to chlorine disinfection by monitoring biofilm viability, biofilm structure, biosorption of DOM, and viability of detached biofilm simultaneously. The obtained results suggested that: i) the presence of EPS increased both biofilm and detached cluster resistance to chlorine in the presence and absence of
DOM; ii) EPS production had an influence on biofilm architecture; iii) Some structural characteristics, such as surface roughness, surface area to volume ratio and average diffusion distance, were closely related to both overall biofilm viability and the spatial distribution of viability within biofilm; iv) the DOM adsorption did not show significant impact on biofilm viability against chlorine disinfection; however, the viability of detached biofilm significantly increased in the presence of DOM, especially for the EPS overproducing strain. The combined results revealed that multiple roles of biofilm EPS synergistically influence susceptibility of both biofilm and biofilm detachment leading to a higher resistance against disinfection practices. The results obtained in this study can provide valuable information regarding the concentration of residual chlorine required to facilitate biofilm inactivation and prevent regrowth in the distribution system, where current levels do not provide sufficient disinfection for biofilm control and further suggest that accelerated chlorine decay may necessitate chlorine “boosting” in the distribution system to maintain biological stability.
Acknowledgements
67
We would like to acknowledge Dr. Daniel Hassett for providing bacterial culture and Christopher Hessler for manuscript preparation. This project is supported by National
Science Foundation (Award Number: CBET0933288) and US Geological Survey
(Project Number: 6HQGR0113).
68
Chapter 5
Impact of Chlorine Disinfection on Detachment and
Redistribution of Cell Clusters from Biofilm
Abstract
The detachment and redistribution are essential stages in biofilm life cycle; however they are the least studied especially under the presence of disinfectant. In this study, both physical and biological properties of cells detached from different growth stages of biofilm and the ability of these detached cells to reattach on new surfaces were analyzed and compared. Three Pseudomonas aeruginosa strains with different extracellular polymeric substance (EPS) production capacity were used to cultivate three single strain biofilms in flowcell systems. Results indicate that the promoted secretion of biofilm EPS increase size and surface charge of detached clusters as well as their resistance to chlorine disinfection. Even under the presence of chlorine, simulating a residual chlorine concentration in drinking water distribution system, the detached clusters from all tested strains were still able to survive and immobilize themselves downstream, forming new biofilms with relatively high viability. The redistributed biofilms of EPS producing strains had higher amount of total biomass, larger biofilm
69
thickness, more complex structural properties for both with and without chlorine conditions, and elevated viability under the presence of chlorine when compared to the counterpart EPS deficient strain. The results suggest that the transport and fate of detached biofilm may have considerable impact on water conveyance as it may continually detach, reattach, and initiate regrowth, thus perpetuating biofilm contamination. This study can be valuable for both practical treatment processes and fundamental explorations of biofilm life cycle.
5.1 Introduction
Biofilms are the result of bacteria attachment and multiplication of cells on a surface which is facilitated by the production of extracellular polymeric substances (EPS).
The formation of biofilm creates a microenvironment for community living compared to free-living planktonic cells due to the unique structure and properties of EPS matrix. The life cycle progression of biofilm started with the transition from planktonic free-flowing cells to attached multicellular biofilm community. Then, by not being restricted to the biofilm environment, cells are able to disperse from biofilm into the aqueous environment (109). The detached cells may survive as dispersed cells to deteriorate water quality or reattach to solid surfaces to initiate new biofilm. In water distribution systems, biofilm formation is difficult to detect and has been causing continuous problems in maintaining biological stability of water quality. Many water utilities have reported bacterial survival despite the regulatory presence of residual disinfectants.
Although increasing recognition that detachment and redistribution are essential stages in biofilm life cycle, enabling biofilm to spread and colonize on new surfaces, currently no systematic study has been done to characterize properties of these detached 70
cells and the redistributed biofilm formed by these cells. In previous publications dealing with biofilm detachment, many factors, such as liquid shear stress (28, 110), nutrient availability (111, 112), temperature (113), cell signaling (114), biofilm aging (115), and molecular genetic expression (116-118), have been found to influence biofilm detachment events, including size distribution and detachment rate. Furthermore, detached cells were defined as a transitional stage between sessile and planktonic states, displaying an increased resistance to antibiotics or disinfectants (87). In spite of the unique properties of detached cells, their redistribution ability is not well understood.
Specifically, quantitative investigation on the reattachment pattern of detached biofilm clusters as well as survival of reattached biofilm under the presence of disinfectant has not been conducted. In addition, although EPS are well known to influence biofilm architecture and provide protective barrier to cells within biofilm matrix (119), the impact of biofilm EPS on biofilm detachment, redistribution, and detached cell susceptibility remains elusive.
This study aimed to investigate the influence of a model disinfectant, chlorine, on the detachment and redistribution processes in biofilm life cycle. The physical properties, including size and surface charge, susceptibility to disinfectant, and ability to immobilize on new surfaces of the detached biofilm were studied using flowcell systems with and without chlorine disinfection. The detached biofilm redistribution was monitored for both short-term (24-hour) and long-term (6-day) with and without continuous chlorine exposure. The 6-day redistributed biofilms were analyzed for their structural properties and viability. On the other hand, the 24-hour redistribution analysis provided information about the influence of chlorine as well as the physiological properties of detached clusters
71
on their initial immobilization and redistribution. Three single-strain Pseudomonas aeruginosa biofilms with varied EPS production capacity were applied and compared to elucidate the role of EPS during the detachment and redistribution processes. The results presented in this study will provide direct insight into the survival and redistribution of detached biofilm clusters in aqueous environment, thus conveying important information regarding minimal disinfectant residuals necessary to prevent bacterial regrowth and maintain biological stability in water distribution systems.
5.2 Materials and Methods
5.2.1 Bacterial Strains and Solution Preparation
Pseudomonas aeruginosa strain PAO1 and two mutant strains algT(U) (EPS deficient) and mucA22 (EPS overproduction) were evaluated in this study. All strains were grown in one-tenth strength LB broth at 37°C with shaking and harvested during their late-exponential phase. The bacterial cells were collected by centrifugation at 2,000
× g for 15 min, allowing for minimal removal of capsular EPS (53). The cells were diluted in chlorine demand free (CDF) buffer as a bacterial suspension (54). Chlorine solutions were prepared with Clorox bleach (The Clorox Co., Oakland, CA) immediately preceding the inactivation experiments. The free chlorine concentration was adjusted to
0.5 mg/L at the inlet point of the flowcell system, representing recommended minimal residual disinfectant concentrations in water distribution systems (59). The chlorine concentration was determined by the N, N-diethyl-p-phenylenediamine (DPD) method using a DR/2700 spectrophotometer (HACH Company, Loveland, CO) (120).
5.2.2 Biofilm Reactor
72
Two square glass capillary flow cells (10 cm of 1 by 1 mm tubing, 0.15 mm thickness; Friedrich & Dimmock Inc.,NJ) were connected in series with silicon tubing
(Masterflex, Vernon Hills, IL) and a “T” shape connector as one flow channel (Figure 5-
1). Six flow channels were run in parallel at room temperature, with duplicate channels for each bacterial strain. The first flowcell in the sequence was used to cultivate biofilms, while the redistribution of detached biofilm was monitored in the second flowcell.
Effluent samples from the first flowcell were collected at the “T” junction. A 0.02 strength LB broth was used as a nutrient source to create nutrient limited growth conditions (91, 92). The first flowcell in the sequence was inoculated with bacterial
10 suspension (OD600=0.5, ~10 cells/mL) and incubated without flow for 2 hours at room temperature to permit initial bacterial attachment. After 2 hours, the nutrient flow commenced and was gradually increased to a flow rate of 1 mL/min by a multichannel peristaltic pump (ISMATEC, Glattbrugg, Switzerland) (115). As the flow residence time within the flow channels was below the doubling time for the microorganism, any cells found in the effluent of the first flowcell were assumed to be detached from the biofilm
(121). Flow cells, tubing, and solutions were autoclaved prior to the beginning of each experiment. Operation and sampling of the flowcells followed aseptic techniques throughout the experiments. Experiments were repeated more than three times for validation of results.
73
Figure 5-1: Diagram of flowcell system setup
5.2.3 Effluent Sample Collection
Effluent samples were collected daily from the “T” junction between the first and second flowcells of each flow channel as described previously. Effluent bottles were handled to avoid disruption of detached biofilm clusters. Orifice size of pipette tips was increased in order to reduce shear and prevent physical disruption of the clumps. The collected effluent samples were analyzed for particle size distribution, Zeta potential and chlorine susceptibility. In order to compare planktonic bacteria and detached biofilm cluster susceptibility to chlorine, the planktonic cells suspensions were adjusted to the same magnitude of cell concentration (~106 cells/mL) as that in the effluent samples for viability test (see below).
5.2.4 Size Distribution of Detached Biofilm
The size distribution of detached biofilm was determined by epifluorescent microscopic analysis. One milliliter of the effluent was stained with 0.5 µg/mL DAPI stain (4,6-diamidino-2-phenylindole dihydrochloride) for 10 min in dark and vacuum filtered onto a black polycarbonate membrane (0.22 µm/ 25 mm). Three membranes were
74
prepared for each effluent sample. Membranes were analyzed on a fluorescent microscope (IX51, Olympus, USA) equipped with a 40X objective. Fifty images were captured using Qcapture software (QImage, Canada) for each membrane. ImageJ
(http://rsb.info.nih.gov/ij/) was used for image processing and particle analysis.
5.2.5 Bacterial Viability Test
All bacterial viability tests were performed in 150 mL amber glass bottles (Fisher
Scientific) at room temperature. The susceptibility of planktonic cell samples and each effluent sample of detached biofilm clusters to chlorine was tested in triplicate. Collected effluent samples were mixed by gentle inversion, then 10 mL of the effluent samples was added into each amber glass bottle containing 90 mL chlorine solution (0.5 mg/L). At each sampling point, both disinfectant residual and viable cell counts were measured.
After quenching residual chlorine with Na2S2O3 (1 mM/L) and vortexing at high speed for 10 seconds, samples from each reactor bottle were diluted in series before cultivating on R2A agar plates (Difco Laboratories, Detroit, MI) in duplicate.
5.2.6 Redistribution of Detached Biofilm
The redistributed biofilm in the second flowcell was visualized by fluorescent staining with BacLight LIVE/DEAD bacterial viability staining kit (Molecular Probes
Inc.) to differentiate live and dead cells. Extracellular polysaccharides in the biofilm formed by P. aeruginosa were visualized with Alexa 633 conjugated concanavalin A
(ConA-Alexa 633; 1mg/mL). Two stock solutions of stain (SYTO 9 and propidium iodide) were each diluted to a concentration of 1.5 µL/mL in CDF buffer. Equal volumes of the two stock solutions were injected into the flow channels and then incubated for 15 min in the dark. The stained biofilm was visualized using a Leica confocal laser scanning
75
microscope (CLSM) equipped with a 63X oil immersed objective. The CLSM images were further processed using the biofilm analysis software COMSTAT to determine surface coverage, total biomass, EPS content, and other biofilm structural parameters (99).
All image analyses were based on at least 6 imaging spots from one sample.
Initial attachment patterns of detached cell clusters from different stages of biofilm growth were compared under the presence and absence of chlorine. After bacterial culture was inoculated to the first flowcells, freshly prepared flowcells were connected to the first flowcells on the first, second, and fifth day for short-term biofilm redistribution tests. Detached clusters from the first flowcells were allowed to recolonize on the second flowcells for 24 hours prior to microscopic image analysis for redistributed biofilm. For disinfected channels, the chlorine solution was blended with nutrient solution and applied at the “T” junction while the control channels only received the nutrient solution. The total flow rate was maintained equally among all flow channels.
5.2.7 Interaction Energy Analysis between Detached Biofilm and Glass Surfaces
The electrophoretic measurements of detached biofilm were measured by a
Zetasizer (Nano – ZS90, Malvern, U.K.) (122). The van deer Waals attraction between the detached biofilm and glass surface was acquired by measuring the contact angles of diiodomethane on the bacteria coated slide using a Goniometer (Tantec Model CAM-
MICRO Contact Angle Meter). The bacterial coating on glass surface was done as follows: A few drops of the highly concentrated bacterial culture were placed on a glass slide in triplicate and the culture was allowed to dry by incubating the glass slide inside a laminar flow hood.
5.2.8 Statistical Analysis
76
The data were presented as mean ± standard deviation. Significance of differences was analyzed by t-test or one-way ANOVA test. P < 0.05 was accepted as statistically significant. All calculations were performed by using SigmaPlot statistics package
(Jandel Scientific, Sausalito, CA).
5.3 Results
5.3.1 Size Distribution of Detached Biofilm Clusters
The detached biofilm clusters were analyzed for their size distribution as shown in
Figure 5-2. The size distribution results revealed that 30-50% of the total detached clusters were below 30 µm2 and the overall trend of size distribution was similar for the three strains. However, the occurance of detached clusters from PAO1 and mucA biofilm larger than 300 µm2 was approximately twice the number of similarly sized detached clusters from algT biofilm on average. Numerically, detached cell clusters greater than
300 µm2 accounted for less than 10% of the total detachment events druing biofilm formation. However, the total detached biomass from large clusters contributed about 45% of the total detached biomass from algT(U) biofilm, and more than 60% for PAO1 and mucA22 biofilm. These findings suggested that, even though large clusters appeared less frequently during biofilm detachment, they provided a disportionally high amount of the total detached biomass availible for redistribuition.
77
Figure 5-2: Size distribution of biofilm clusters detached from different stages during
biofilm growth. (a) algT(U) biofilm; (b) PAO1 biofilm; (c) mucA22 biofilm.
78
5.3.2 Detached Biofilm Susceptibility to Chlorine Disinfection
Batch experiment results were interpreted as the fraction of culturable bacteria
[Log (N/N0)] versus Ct values as shown in Figure 5-3. For planktonic bacteria disinfection, a beyond 3-log reduction was reached within 10 seconds for all three strains.
For an extended exposure time, the survival rate continually declined. However, the cell clusters detached from both 3-day and 6-day old biofilm of all three strains exhibited a much lower viability reduction, when compared to the planktonic cells reacted with the same disinfectant concentration. A comparison between planktonic cell and detached biofilm susceptibility revealed that the chlorine inactivation rate was significantly reduced for cells detached from biofilms when compared to planktonic cells of the same strain (P < 0.05). The 0.5 mg/L chlorine disinfectant was sufficient to maintain all three planktonic bacteria below a 3-log reduction after 1 min exposure. However, at the same
Ct value, for the detached clusters, none of the three strains could reach the 2-log inactivation level. Even the EPS deficient strain algT(U) only reached a 1.5-log inactivation. The other two EPS producing strains achieved near 1-log reduction after extended reaction time with chlorine.
When comparing differences among strains, there was a trend that the high EPS producing strain was more resistant to chlorine disinfection for both the planktonic and detached cells as expected. A significant difference was observed among stains for these three sampling conditions (planktonic cell, cell cluster from 3-day and 6-day biofilm; P <
0.05). However, pairwise comparison showed that significant difference was only between algT(U) and mucA22 strains for the planktonic strains while two pairs (algT(U)
79
and PAO1, algT(U) and mucA22) were statistically different for cell cluster detached from 3-day biofilm. For cell clusters detached from 6-day biofilm, all pairs were significantly different. The comparison result suggested that the impact of EPS on bacterial susceptibility to chlorine in much significant in biofilm clusters than their planktonic cells. Our data suggested that detached biofilm cell clusters were more disinfectant resistant than their counterpart planktonic cells. In the presence of EPS, the survival rate was greatly increased for all strains. In this case, it appeared that a regulatory disinfectant residual concentration may not be sufficient to inactivate the detached biofilm clusters to prevent bacterial regrowth and proliferation.
80
81
Figure 5-3: Chlorine inactivation kinetics of (a) planktonic cells; (b) cell clusters
detached from 3-day biofilms; (c) cell clusters detached from 6-day biofilms.
5.3.3 Zeta Potential
Effect of chlorination on surface charge of both planktonic bacteria (Figure 5-4) and detached clusters (Figure 5-5) was analyzed by measuring Zeta potential. For the planktonic cells, Zeta potential was significantly different among the tested three strains, showing a higher negative value with the increased EPS content. For the detached clusters from the first flowcells [Figure 5-5(a)], clusters detached from 1-day old biofilms maintained similar Zeta potential as their planktonic cells. An increased negative value in
Zeta potential was observed for cells detached from biofilms after 2-day growth for all three strains. As the same trend as in their planktonic cells, Zeta potential was higher in cell clusters detached from EPS producing strains than EPS deficient strain. Interestingly, after these cell clusters detached from the first flowcell passing through the second flowcell [Figure 5-5(c)], the cell surface charge was significantly reduced for the EPS producing strains while stayed consistent for the EPS deficient strain.
82
Figure 5-4: Zeta potential of planktonic bacteria suspension and resuspended biofilm of P. aeruginosa strains. (A) Planktonic bacterial suspension - control; (B) planktonic bacterial
suspension – chlorine disinfected; (C) planktonic bacterial suspension – resuspended in
buffer after chlorine disinfection; (D) biofilms resuspended in buffer – control.
83
Figure 5-5: Zeta potential of cell clusters detached from biofilms on different stages during biofilm growth. Cell clusters detached from (a) the first flowcells; (b) the second
flowcells with chlorine disinfection; (c) the second flowcells without chlorine
disinfection. 84
To investigate the effect of chlorination on bacterial surface charge, the same concentration of chlorine was applied to both planktonic cells and cell clusters prior to entering the second flowcell. For planktonic suspensions, chlorine exposure increased
Zeta potential for the three strains (Figure 5-4). Although there was a slight increase in surface charge between detached clusters coming out from the first and the second flowcell for all three strains, no significant change in Zeta potential was observed under the same concentration of chlorine [Figure 5-5(a & b)]. Also, under chlorine disinfection, cell clusters coming out from the second flowcell had no significant difference in surface charge among strains. The effect of residual chlorine on Zeta potential was tested and found to be minimal. Statistical analysis was performed by comparing the Zeta potential of detached cells from the first flowcell, from the second flowcell with and without application of chlorine. A significant difference was found among the three sampling conditions for all three strains. The pairwise comparison showed that, difference between cells detached from the first flowcell and cells detached from the second flowcell without chlorine was not significant for the algT(U) strain. On the contrary, for the other two strains, the differences were between cells detached from the first flowcell and cells detached from second flowcell without chlorine, and between cells detached from the second flowcell with chlorine and without chlorine.
5.3.4 Redistribution of Detached Biofilm and Biofilm Structural Characterization
The influence of chlorine disinfection on the redistribution of detached clusters was evaluated by monitoring the formation of biofilm in the second flowcell under the presence and absence of chlorine. For long-term tests, the second flowcells were
85
connected to the first flowcells from the beginning of the experiment and disconnected on the sixth day for image analysis. The structural characteristics of redistributed biofilms were summarized in Table 5-1. The structural analyses generally showed an increasing amount of total biomass with increasing amount of EPS content within the biofilms for both control and disinfected conditions. Accordingly, the same trend was found in biofilm thickness for both conditions. The EPS producing strains with larger quantity of biomass and higher thickness created a more complex biofilm structure than the counterpart EPS deficient strain, indicated by the structural characteristics such as surface roughness, surface area to volume ratio, and diffusion distance. Besides physical differences in biofilm structure, the viable ratio was different among strains under chlorine disinfection condition, showing an increased viable ratio with promoted EPS production. The mucA22 biofilm exhibited close to two fold higher viable ratio than the algT(U) biofilm. When comparing the control and disinfection condition for each tested strain, the total biomass and maximum thickness of biofilms were reduced under the presence of chlorine. However, viable ratios remained relatively high for the EPS producing strains, especially the mucA22 biofilm, which had no statistically significant difference in viable ratio when compared to the control condition.
Table 5.1: Redistributed biofilm structural characteristics for long-term (6-day) tests.
Under chlorine disinfection No chlorine disinfection (0.5 mg/L)
algT(U) PAO1 mucA22 algT(U) PAO1 mucA22
Roughness coefficient 0.85 ± 0.98 ± 0.52 ± 1.80 ± 1.33 ± 0.91 ±
86
0.46 0.65 0.30 0.10 0.42 0.16
Maximum diffusion 13.77 ± 18.30 ± 37.64 ± 8.68 ± 6.70 ± 14.15 ±
distance (µm) 6.65 9.24 14.86 1.72 2.77 3.58
Average diffusion 0.07 ± 0.36 ± 1.21 ± 0.01 ± 0.00 ± 0.02 ±
distance (µm) 0.06 0.48 0.60 0.00 0.01 0.01
Surface area to volume 1.65 ± 1.50 ± 0.81 ± 1.63 ± 3.96 ± 2.79 ±
ratio (µm2/ µm3) 0.44 0.47 0.24 0.46 0.29 0.21
Maximum thickness 158.97 ± 161.95 ± 163.52 ± 43.93 ± 84.51 ± 148.76 ±
(µm) 33.61 78.12 24.96 24.48 23.59 41.37
Viable ratio (%) 94 ± 17 95 ± 25 97 ± 9 54 ± 14 79 ± 15 95 ± 2
0.79 ± 3.43 ± 4.38 ± 0.46 ± 2.63 ± 6.34 ± EPS (µm3/ µm2) 0.25 0.74 0.58 0.05 1.61 1.35
Total biomass (µm3/ 23.96 ± 24.57 ± 46.55 ± 1.27 ± 3.79 ± 9.93 ±
µm2) 15.58 15.44 19.67 0.84 2.35 2.71
To further investigate the impact of chlorine on the initial reattachment of detached clusters, a short-term (24 hours) redistribution analysis was conducted. The mucA22 strain, which had a higher reattachment rate among the tested three strains, was chosen for the short term redistribution study to elucidate the initial redistribution pattern under with and without chlorine disinfection conditions. The second flowcells were connected to the first flowcells on day 1, day 3 and day 6. Then after 24 hour continuously receiving detached clusters from the first flowcell, these second flowcells were disconnected for image analysis. Without disinfectant, the detached cells were able to form a flat layer largely covering surfaces in the second flowcells. A major reduction
87
in detached cell redistribution was monitored when applying 0.5 mg/L chlorine; however, small portion (2% on day 3 and 5% on day 6) of these detached cells were still able to survive and anchor themselves onto new surfaces. As shown in Figure 5-6, the coverage rate was relatively higher when the biofilm, where the cell clusters were detached from, tended to reach a mature stage for both control and disinfected conditions.
Figure 5-6: Coverage rate in the second flowcell for short term (24-hour) redistribution
test of mucA22 strain.
5.4 Discussion
Cells detached from biofilm are thought to display a transitional phenotype between attached cells and planktonic cells (87). The detached biofilm clusters cover a wide range of size distribution, including single cells and clusters comprised of hundreds of cells. In this study, the detached cluster size distribution appeared to have a similar
88
trend for all three strains (Figure 5-2). Mostly, small biofilm clusters (< 2 µm2) were dominant and large clusters (>300 µm2) appeared less frequently in collected effluent samples. However, considering biomass of the detached clusters, large clusters provided a disproportionally high amount biomass in the total detached biomass. The increased viable ratio of cells detached from EPS producing strains could be directly linked to the higher proportion of large clusters, where embedded cells are shielded from exposure to disinfectant (108).
The algT(U) and mucA22 are isogenic variants of the wild type strain PAO1, and differ only in their EPS secretion capability. The Zeta potential results in this study was mucA22 > PAO1 > algT(U) in net negative values and the differences among strains were larger for detached cells than the planktonic cells, indicating that detached cells may carry a larger EPS content than their planktonic state. These results confirm previous reports that primary carbohydrate constituents of PAO1 EPS are neutral sugars (35) and the mucoid mucA22 strain secretes negatively charged acidic polysaccharide alginate
(123). Under without chlorine disinfection condition, a significant decrease in Zeta potential of cells coming out from the second flowcells was observed only for EPS producing strains, PAO1 and mucA22, hence an inference can be made that surface charge of detached cells influenced their redistribution. Meanwhile, the significant variation in the overall cell reattachability among the three strains demonstrates that interactions between bacterial surface polymers and glass surfaces played important roles in controlling detached cell redistribution in the flowcell system. Although Wozniak et al. pointed out that alginate is not a key sugar component of the extracellular polysaccharide matrix of P. aeruginosa biofilms and also alginate is not necessary during the formation
89
of nonmucoid P. aeruginosa biofilms (124), our results suggest that the promoted EPS secretion in general increased cell adhesion under the presence or absence of chlorine.
Multiple factors have known to be involved in cell attachment, such as surface conditioning, mass transport, surface charge, hydrophobicity, surface roughness, growth medium and surface microtopography (125). Previous studies reported conflicting opinions regarding the attachment theories. It is difficult to characterize the role and the overall importance of each factor on bacterial attachment. In this study, impact of multiple factors were weighted and compared based on the structural characteristics of the redistributed biofilm.
Redistribution of detached biofilm clusters started with initial attachment process.
There are several hypotheses discussed on bacterial initial attachment. Among them, the two step attachment process is one of the generally accepted theories (125). The first step involves the bacteria being transported close enough to allow initial attachment to take place, where the involved forces are van der Waals forces, electrostatic forces and hydrophobic interactions. At this step, cell attachment is reversible subject to hydraulic conditions and other chemical conditions in the bulk solution. The next crucial step in the attachment process is the irreversible attachment of cells to the surface, as bacteria locking on to the surface by the production of exopolysaccharides and/or specific ligands, such as pili or fimbriae that may enhance surface adhesion (126). Various short range forces are involved in the transition from reversible attachment to irreversible attachment, including covalent and hydrogen bonding as well as hydrophobic interactions (127). The presence of EPS could act as bonding material during this transitional phase to promote bacterial cells anchor themselves on glass surfaces permanently.
90
During biofilm detachment, especially the initial two days, the detached events were predominantly single cells. The attachment of these single cells can be described using the DLVO theory as reported previously (128). One critical aspect of the DLVO theory is ionic strength of the solution. Rijnaarts et al. (1994) described that at low ionic strength solution, for example <0.001 M, long range electrostatic repulsion dominates bacterial attachment, but at high ionic strength (>0.1 M) other factors such as steric interactions (hydrophobicity) dominate (129). In this study, the ionic strength of the nutrient solution was 0.006 M, suggesting electrostatic repulsion, even though may not be dominant, may play a more important role than other interactions. The total interaction energy profiling using DLVO calculation showed that the energy barrier between algT(U) cells and glass surface was below 1000 kT, while the energy barrier for PAO1 and mucA22 strains was over 2000 kT. These data suggested that it is easier for algT(U) cells to overcome the energy barriers to attach on the glass surface. However, as noticed by other researchers, predicting bacterial attachment by DLVO theory is often contradictory with experimental data under many conditions (130). One limitation of applying DLVO theory to describe bacterial attachment is measuring the overall surface charge at the macroscopic level without considering the localized surface properties. The bacterial surface is not a model colloidal particle but a delicate and complex array of carbohydrates, proteins and other components. These complicated surface structures may exert their own localized cell surface charge at a microscopic cell surface level that could possibility mediate attachment through local electrostatic attraction despite the overall electrostatic repulsion between cells and surfaces. Although there is debate whether EPS play an important role on bacterial attachment (124), the presence of membrane associated
91
proteins and polysaccharides have been reported to have strong affinity with many inert surfaces including glass surface (131, 132). In this case, even though there is a higher energy barrier for PAO1 and mucA22 cells to overcome, the presence of EPS may benefit their attachment through polymeric interaction. The DLVO theory is also limited by the size of detached clusters. Other than cells in the single form, cells were also detached in aggregated clusters. As discussed in previously paragraph, cells in large clusters possessed a disproportional biomass even though they appeared less frequently. The attachment of these large clusters cannot be simplified as ideal colloidal particle to use the DLVO theory. Besides the influence of the forces between bacterial cells and glass surfaces, the redistribution is also controlled by other physiochemical factors. The initially attached cells, especially the production of EPS, provide a modified surface to receive floating cells to attach. Furthermore, when bacteria with EPS production capacity get detached, they may leave “footprints”, which provide surface conditioning to benefit bacterial attachment (35).
5.5 Conclusion
Results in this study indicate that the promoted secretion of biofilm EPS increase size and surface charge of detached clusters as well as their resistance to chlorine disinfection. Even under the presence of chlorine, at a minimal residual chlorine concentration required in drinking water distribution systems, the detached clusters from all tested strains were still able to survive and immobilize themselves downstream, forming new biofilms with relatively high viability. The redistributed biofilms of EPS producing strains had higher amount of total biomass, larger biofilm thickness, more complex structural properties for both with and without chlorine conditions, and elevated 92
viability under the presence of chlorine when compared to the counterpart EPS deficient strain. The results suggest that the transport and fate of detached biofilm may have considerable impact on water conveyance as it may continually detach, reattach, and initiate regrowth, thus perpetuating biofilm contamination. This study can be valuable for both practical treatment processes and fundamental explorations of biofilm life cycle.
Acknowledgement
This study was supported in part by US Geological Survey (Project Number:
2010OH160B) and National Science Foundation (Award Number: CBET0933288). We thank Dr. Daniel Hassett (University of Cincinnati, Cincinnati, OH) for providing the P. aeruginosa strains.
93
Chapter 6
Comparison of Chlorine and Monochloramine
Inactivation in Biofilm and Detached Biofilm in a
Model Drinking Water Distribution System
Abstract
Biofilms are undesirable but ubiquitous in drinking water systems. This study compared influence of biofilm extracellular polymeric substances (EPS) reactivity with two commonly used secondary drinking water disinfectants, chlorine and monochloramine, on the susceptibility of biofilm and detached biofilm clusters. Strains from an opportunistic pathogen, Pseudomonas aeruginosa with different EPS secretion capabilities were tested. Two major components in biofilm EPS, polysaccharides and proteins, both reacted rapidly with chlorine while monochloramine specifically reacted with proteins. The impact of biofilm EPS reactivity with disinfectants on disinfection efficacy was evaluated by monitoring biofilm viability, biofilm structure, and detached biofilm viability simultaneously during disinfection process by minimal disinfectant residual concentrations. The obtained results suggested that the presence of EPS
94
increased biofilm and detached biofilm resistance to both chlorine and monochloramine.
The biofilm structure characterization using confocal laser scanning microscopy (CLSM) analysis revealed that EPS production affected biofilm architecture, specifically surface roughness, surface area to volume ratio and average diffusion distance. These structural characteristics were closely related to overall biofilm viability and the spatial distribution of viability within biofilms. Although the overall viable ratios were similar under the two disinfectants for each strain, monochloramine penetrated deeper into biofilm matrix than chlorine regardless the quantity of EPS content, showing a higher inactivation efficacy in the middle section of biofilms. However, chlorine was more efficient in controlling detached cluster viability than monochloramine. The combined results suggested that different reactivity of biofilm EPS with disinfectants influenced the susceptibility of both biofilm and detached biofilm during disinfection practices. This study will provide valuable information regarding how disinfection practices can be optimized to sufficiently maintain water quality by promoting biofilm inactivation and reducing viable detached clusters.
6.1 Introduction
Biofilm is ubiquitous in water distribution systems and it is becoming evident that the presence of biofilm can greatly impact water quality and safety (133, 134). In addition to the persistence of biofilm formation, the detachment of biofilm from the parent structure may greatly contribute to biomass redistribution and cell re-colonization downstream (135). Therefore, inactivating preformed biofilm as well as preventing biofilm redistribution and proliferation have been great challenges for water utilities to deliver high-quality water. Moreover, the increasing attention on drinking water safety 95
and increasing stringency of disinfectant by-product guidelines requires disinfectant application in drinking water distribution systems to be more carefully conducted and more tightly controlled.
Chlorine has been used in distribution systems for decades to control bacterial regrowth; however, there has been a strong trend toward replacing chlorine with monochloramine in order to meet the more strict rules on disinfectant by-products formation (136). In addition, chlorine residual concentration has been reported to be greatly influenced by organic matter concentrations in water systems and the extracellular polymeric substances (EPS) secreted by bacteria. As an alternative disinfectant to chlorine, many water utilities are adapting chloramines for water distribution system residual. In response, growing number of studies have been investigating monochloramine disinfection, comparing its disinfection efficacy to that of chlorine (10,
137, 138). Both laboratory studies (20, 139, 140) and field studies (141, 142) reported that monochloramine maintained a more stable residual and showed higher inactivation efficacy for biofilms than chlorine. Based on the direct microelectrode measurement for chloramines transport in biofilm, the advantage of monochloramine over chlorine for biofilm inactivation was related to its lower reactivity with biofilm components resulting in better biofilm penetration (143, 144). However, the spatial distribution of biofilm viability caused by disinfectant transport gradient is still illusive. In addition, previous comparisons between chlorine and monochloramine for biofilm inactivation were conducted with relatively high Ct values for monochloramine (144), which were not representative for drinking water distribution systems especially the slow-flow sections and dead ends of the systems. More importantly, even though detachment and
96
redistribution have been recognized as essential stages in the biofilm life cycle, relatively little is known about the disinfection efficacy of chlorine or monochloramine on these detached clusters during the detachment process in a dynamic flow system.
Monochloramine was reported to be less effective for planktonic bacterial inactivation than chlorine but no experiment has been conducted for the detached clusters (20, 145).
In addition, production of EPS has been linked to biofilm resistance against disinfection, but they have also contributed to biofilm structure and are involved in biofilm detachment and redistribution processes (146). However, there is a paucity of data on the influence of EPS reactivity with disinfectants on both biofilm and detached biofilm susceptibility to disinfectants.
The aim of this study was to determine the disinfection efficacy of monochloramine on both biofilm and detached biofilm clusters when compared to chlorine. Strains from an opportunistic pathogen, Pseudomonas aeruginosa, with varied
EPS secretion capabilities were tested and compared to elucidate the role of biofilm EPS during the disinfection process. The impact of biofilm EPS reactivity on disinfection efficacy was evaluated by monitoring biofilm viability, biofilm structure, and the viability of detached biofilm clusters, simultaneously during the disinfection process.
6.2 Materials and Methods
6.2.1 Bacteria Culture
P. aeruginosa is an opportunistic pathogen which has been identified in water distribution systems (48). Wild-type PAO1 and two mutant strains, algT(U) (inhibited alginate EPS production) and mucA22 (overproduction of alginate EPS), were selected to construct confluent biofilms. All strains were grown in one-tenth strength LB broth (2.5 97
g/L, Difco Laboratory, Detroit, MI) at 37°C and then harvested during the late- exponential phase by centrifugation at 2,000 × g for 15 min. The cells were diluted in chlorine demand free (CDF) buffer (0.54 g Na2HPO4 and 0.88 g KH2PO4 per liter, pH=6.98) as a bacterial suspension (OD600 = 0.5 ± 0.02) (54).
6.2.2 Solution Preparation
Biofilms were cultivated in 0.02 strength LB broth to create nutrient limited growth conditions (91, 92). Chlorine solutions were prepared by adding Clorox bleach
(The Clorox Co., Oakland, CA) to autoclaved deionized water. Monochloramine solution was prepared by combining solutions of sodium hypochlorite (6%, pH=8.3) and ammonium chloride (0.2 mM/l, pH=8.3) in a 4:1 chlorine-to-ammonia-nitrogen mass ratio, where monochloramine is the dominant species at this pH. The chlorine concentration and monochloramine concentration as free chlorine was determined with the N, N-diethyl-p-phenylenediamine (DPD) method and the indophenols method using a
DR/2700 spectrophotometer (HACH Company, Loveland, CO) (94).
6.2.3 Biofilm Cultivation and Disinfection in Flow Cell System
Biofilms were cultivated in two channel flowcell systems (BioSurface
Technologies Corp., Bozeman, MT) fitted with a glass microscope slide opposing a glass cover slip (channel dimensions, 1.6 by 12.7 by 47.5 mm; flow rate = 0.2 mL/min) at room temperature (22 ± 2 oC) (Figure 4-1) (95). Two carboys were used as nutrient media and disinfectant supply reservoirs, respectively. All feeds were delivered using a multichannel peristaltic pump (ISMATEC, Glattbrugg, Switzerland). Flow cells, tubing and solutions were sterilized prior to each experiment. The flow cell channels were aseptically inoculated with bacterial suspension and incubated 2 hours without flow for
98
initial bacterial attachment. Then, flow was introduced and gradually increased to 0.2 mL/min. Biofilm was grown six days to reach structural maturity (89). For each flowcell, one channel was used as a control, while disinfectant was applied to the other. For chlorine and monochloramine disinfection, media and disinfectant solutions were mixed in a bubble trap (HRT = 25 min) before entering the flow cells. The disinfectant concentrations were maintained at 0.5 mg/L for chlorine and 2 mg/L for monochloramine at the flow cell inlet throughout the disinfection process. Flowcell effluent was collected every 30 minutes for 2 hours during the disinfection process and quenched with 0.1 M sodium thiosulfate before further analysis. The total disinfection time was chosen based on the chlorine decay kinetic with biomass in water distribution systems (97). For each bacterial strain and each disinfectant, the experiment was repeated more than three times.
The use of flowcell reactors provided accessibility to monitor biofilm development and inactivation as well as behavior of detached clusters in situ.
6.2.4 Confocal Laser Scanning Microscopy and Image Analysis
The biofilm content on glass slides was discriminated using BacLight
LIVE/DEAD staining kit (Molecular Probes Inc.) to differentiate live and dead cells.
Extracellular polysaccharides content in the biofilm were visualized by applying Alexa
633 conjugated concanavalin A (ConA-Alexa 633; 1 mg/mL in stock solution). This stain specifically targets the polysaccharides (D-glucose and D-mannose residues) that are major EPS components of P. aeruginosa (98). A mixture of stains was prepared in CDF buffer with a final concentration of 1.5 µl/mL each for SYTO 9 and propidium iodide
(PI), as well as 20 µl/mL of EPS stain. The mixture was then injected into flow channels and incubated in the dark for 15 min. Live SYTO 9-stained cells, dead PI-stained cells,
99
and ConA-Alexa 633 stained EPS were visualized with a Leica SP5 Confocal Laser
Scanning Microscope (CLSM) equipped with a 63× oil immersed objective and a 20× dry objective. An argon laser, at 488, 561 and 633 nm, was used as the excitation source for the fluorescent probes. For each biofilm sample slide, at least five positions were randomly selected for image acquisition and further image analysis.
CLSM images were further processed using the image processing program
COMSTAT to determine total biomass, EPS content, and biofilm structural parameters, as defined in detail elsewhere (99). Important structural parameters discussed in this study include: i) the roughness coefficient - a measure of how the thickness of biofilm varies; ii) surface area to volume ratio - representing the spatial complexity of biofilm structure; iii) diffusion distance - defined as the shortest distance from a pixel containing biomass to a pixel not containing biomass. Average diffusion distance is the average of diffusion distances for all pixels containing biomass. This parameter indicates the extent of void spaces in the biofilm structure. EPS content per unit area was generated from image analysis to evaluate the role of EPS on biofilm structure and biofilm susceptibility to disinfectant.
6.2.5 Bacterial Enumeration and Viability
After disinfection and microscopic analysis, biofilms from both control and disinfected channels were scraped from glass slides and resuspended in CDF buffer.
Viable cells were enumerated using plate count method on R2A agar plates (Difco
Laboratories, Detroit, MI) in duplicate. The disinfection efficacy was evaluated using the ratio of viable bacteria in disinfected biofilm over the control biofilm value.
100
Enumerations of viable bacterial cells in the effluent samples followed the same procedure.
6.2.6 Flow Cytometry Analysis
While plate count enumeration yields counts of culturable cells, flow cytometry analysis of the detached biofilm was used to differentiate and quantify dead, injured and live cells as a function of fluorescence intensity, based on the extent of membrane damage. Data was acquired in “log” mode using a FACScalibur flow cytometer (BD
Biosciences, San Jose, CA). The flow cytometer was equipped with an argon laser set at
15 mV and an excitation wavelength of 488 nm. PI and SYTO 9 were used in combination to determine membrane compromised cells and intact cells, respectively.
Stains were simultaneously added at concentrations of 0.15 µl PI and 0.1 µl of SYTO 9 to
1 mL of sample and incubated as described above. Cell concentration was determined by comparing cell events to events from a microsphere standard of known concentration
(InVitrogen, Carlsbad, CA).
On the basis of negative and positive controls, flow cytometry analysis was performed on two fluorescent channels (PI and SYTO 9) to evaluate the cellular viability.
Each acquired data plot was analyzed using WinMDI (J. Trotter 1993-1998) in four quadrants (100): (A) PI positively stained dead cells with a permeabilized cell membrane;
(B) both PI and SYTO 9 positively stained membrane compromised cells; (C) SYTO 9 positively stained live cells with intact cell membrane; (D) negative signals.
6.2.7 Statistical Analysis
Data are presented as mean ± standard error. Differences were analyzed using unpaired t-test or one-way ANOVA test. The association between biofilm structural
101
parameters, EPS content and biofilm viability was analyzed by linear regression analysis. P<0.05 was accepted as statistically significant. All calculations were performed by using SigmaPlot (Jandel Scientific, Sausalito, CA).
6.3 Results
6.3.1 Effect of EPS Content on Biofilm Structure
After six days growth, biofilm composition and structure was quantitatively analyzed using CLSM (Table 6-1). Image analysis revealed the EPS content in biofilm was 0.19 ± 0.02 µm3/µm2 for algT(U) biofilm, 1.46 ± 0.48 µm3/µm2 for PAO1 biofilm, and 2.24 ± 0.40 µm3/µm2 for mucA22 biofilm. Despite clear differences in biofilm EPS content, biofilm from all strains possessed a similar level of bacterial cell biomass (25 ± 3
µm3/µm2). However, EPS production was determined to significantly affect biofilm structural parameters, such as roughness coefficient, surface area to volume ratio, and the average diffusion distance. These three parameters were found to be negatively related to the EPS amount (P<0.01). No significant changes in biofilm structural parameters, cell biomass, and EPS content were observed when comparing control and disinfected biofilm for all strains under two disinfectants.
Table 6.1: Biofilm structural parameters and biofilm EPS content.
algT(U) PAO1 mucA22
Roughness coefficient 0.97 ± 0.34 1.05 ± 0.29 0.69 ± 0.24
Maximum diffusion distance (µm) 10.36 ± 4.45 13.69 ± 3.73 9.70 ± 3.32
Average diffusion distance (µm) 0.43 ± 0.37 0.29 ± 0.15 0.25 ± 0.19
102
Surface area to volume ratio (µm2/ µm3) 1.57 ± 0.66 1.39 ± 0.39 1.17 ± 0.28
Maximum thickness (µm) 37.49 ± 5.13 46.91 ± 9.17 45.85 ± 7.19
EPS (µm3/ µm2) 0.19 ± 0.02 1.46 ± 0.48 2.24 ± 0.40
Total biomass(µm3/ µm2) 23.14 ± 12.92 28.86 ± 13.06 22.33 ± 4.23
Data are presented as mean ± 1 standard deviation (N ≥ 36).
6.3.2 Biofilm Susceptibility to Chlorine and Monochloramine
To better understand EPS influence on the biofilm disinfection process, both heterotrophic plate count (HPC) and CLSM image analysis from the control and disinfected channels were interpreted to determine both overall and the spatial distribution of biofilm viability upon disinfectant exposure.
The HPC results of resuspended biofilm showed the mucA22 biofilm had the highest viable ratio, followed by the PAO1 biofilm, and the algT(U) biofilm for both chlorine and monochloramine disinfection (Figure 6-1). There was no significant difference in overall viable ratio between these two disinfectants for each strain. The
CLSM image analysis revealed the same trend among strains for both chlorine and monochloramine disinfection (Figure 6-2). The overall viable ratios were similar for all three strains under the control condition, which was around 65% on average. Upon disinfectant exposure, the viable ratios were influenced differently among strains for both disinfectants, showing an enhanced resistance with increased EPS content in the biofilm.
Although there was no statistical difference between the two disinfectants for each strain, the viable ratio of mucA22 biofilm was 31% lower for monochloramine disinfection compared to chlorine disinfection while the other two strains exhibited very similar
103
viability for both disinfectants. Both the HPC results and CLSM analysis suggested that the overall viable ratio was greatly promoted by higher EPS content in biofilm (P<0.001).
In addition, the overall biofilm viability ratio was negatively correlated to the roughness coefficient and average diffusion distance (P<0.05).
Figure 6-1: Overall viable ratio of biofilm determined by HPC under chlorine and
monochloramine disinfection. Error bars represent ± 1 standard deviation (N ≥ 18).
104
Figure 6-2: Biofilm overall viable ratio determined by CLSM image analysis with
chlorine and monochloramine disinfection. Error bars represent ± 1 standard deviation
(N ≥ 36).
Beyond the overall biofilm survival, a spatial distribution of biofilm viability was analyzed to investigate disinfectant efficacy among strains using CLSM image analysis.
To evaluate the spatial distribution, images were acquired in 1 µm depth increments and then obtained images were analyzed to calculate viable ratios of live over total cells. The average viable ratios determined in each depth within biofilm were plotted against the normalized distance from the substratum to biofilm surface, where each curve represents an individual strain and condition averaged from more than 36 imaging spots (Figure 6-3).
A similar pattern in viable ratio distribution was observed for all strains in the control channels, ranging from 50 to 70% viability on average. For monochloramine disinfected biofilms, the maximum viability ratio determined for algT(U), PAO1 and mucA22
105
biofilms was less than 10%, about 17%, and up to 25%, respectively. Meanwhile, for chlorine disinfected biofilms, the maximum viable ratio was similar to the monochloramine results for the three strains. However, the disinfectant impact depth, defined as the distance from the surface of biofilm to the point where maximum viability ratio was observed, was different under the two disinfectants. For monochloramine disinfection, the impact depth appeared at the middle section for all three biofilms, which was approximately 20 µm. On the other hand, the impact depth by chlorine was different among the three strains. The chlorine impact depth was limited by the presence of EPS within the biofilm matrix: 20-30 µm for algT(U) biofilm, 15-20 µm for PAO1 biofilm, and about 10 µm for mucA22 biofilm. For chlorine disinfection, significant viable ratio decrease in mucA22 biofilm was observed only at the surface layer, and the viable ratio remained constant as depth increased into the middle section of biofilm. Compared to chlorine disinfection, monochloramine was able to achieve a deeper penetration for EPS producing strains with a similar inactivation efficacy as chlorine.
106
Figure 6-3: Spatial distribution of control and monochloramine disinfected biofilm viable
ratio within biofilms. The horizontal axis represents the normalized distance from the
substratum increasing to the biofilm surface, where the viability ratio (vertical axis) was
determined at each cross-section. Error bars represent ± 1 standard deviation (N ≥ 36).
6.3.3 Detached Cell Viability during Biofilm Disinfection
As an element of biofilm lifecycle, the detached biofilm viability was investigated using HPC and flow cytometry. HPC results indicated an increased viable ratio for detached biofilm from the EPS overproducing strain (Figure 6-4), while the deficient strain was the most susceptible to disinfection. For chlorine disinfection, only clusters detached from algT(U) biofilm reached and maintained greater than 2-log reduction (99% 107
inactivation) at all sampling points. Clusters detached from PAO1 biofilm reached 2-log reduction after 60 minutes. However, the mucA22 clusters were observed to maintain a much higher viable ratio, reaching only 1-log reduction after two hours of biofilm disinfection. Compared to chlorine disinfection, the viable ratio was higher under monochloramine disinfection for all strains. Even the highly susceptible algT(U) detached biofilm could not reach 2-log reduction. The viable ratio for PAO1 and mucA22 biofilm clusters was around 40%. These results suggested that monochloramine (2 mg/L) was less efficient than chlorine (0.5 mg/L) in controlling detached biofilm viability at the same exposure time.
108
Figure 6-4: Detached biofilm cluster viable ratio determined by HPC with chlorine and
monochloramine disinfection. Data points represent mean ± 1 standard deviation (N ≥
36).
In addition to HPC, flow cytometry provided a measure of both cell concentration and cell viability. The detached cell concentration was found consistent for all tested strains and conditions. To analyze cell viability, a data plot of PI versus SYTO 9 fluorescence intensity was acquired by flow cytometry and analyzed in four quadrants
[Figure 4-7(a)]. Previous studies have shown that the staining of bacterial cells with
SYTO 9 and PI did not always produce distinct “live” and “dead” populations (63, 64).
The appearance of yellow or orange stained cells observed in previous studies indicated 109
an intermediate state of membrane compromised cells. In this study, the intermediate state cells were observed in quadrant B (upper right), indicating both green and red positive signal. The percentage of live, dead and membrane compromised cells were plotted in a bar graph (Figure 6-5), where flow cytometry results revealed a similar trend to HPC results in viable ratio corresponding to EPS content of tested strains under both chlorine and monochloramine disinfection. No significant difference in viable ratio was found for the control samples when comparing cells detached from the three biofilms.
However, when the disinfected samples were examined, more SYTO 9 positive signals were observed (rightward in the plot) under the monochloramine disinfection, indicating higher survival of detached cells during monochloramine disinfection compared to chlorine disinfection. The portion of dead cells in detached biofilm was 17% lower for algT(U) under monochloramine disinfection than chlorine disinfection, while it was 20% and 36% lower for PAO1 and mucA22, respectively. Consequently, the live cell portion, as well as the membrane injured cell portion, was significantly higher under monochloramine disinfection than chlorine disinfection for cells detached from the three biofilms. The flow cytometry results suggested that detached cells of all three tested strains had a higher survival rate under monochloramine disinfection than chlorine disinfection.
110
Figure 6-5: Detached biofilm viability quantified using flow cytometry with chlorine and
monochloramine disinfection. Notations in x-axis: letters represent bacteria strains (A:
algT(U); P: PAO1; M: mucA22); numbers represent different treatments (1: control; 2:
monochloramine disinfection; 3: chlorine disinfection).
6.4 Discussion
In this study we observed a relationship among EPS content, biofilm structure, biofilm and detached biofilm clusters susceptibility to chlorine and monochloramine disinfection. The EPS overproducing mucA22 biofilm was found to have a lower biofilm roughness coefficient, surface area to volume ratio and average diffusion distance when compared to the wild type PAO1 biofilm and EPS deficient algT(U) biofilm. The overall biofilm viability was significantly enhanced by increased EPS content in biofilm structure under both chlorine and monochloramine disinfection, which conforms a
111
previous study showing that a mucoid strain produced biofilms with high antibiotic resistance (88).
The biofilm structural characteristics induced by EPS appeared to impact biofilm viability as discussed in previous chapter. The presence of EPS reduced void spaces in between adjacent bacterial cells within biofilm matrix, resulting retarded disinfectant transport. Furthermore, the interaction between EPS and disinfectants caused disinfectant depletion along the penetration. In this case, EPS reactivity with disinfectants may greatly contribute to biofilm disinfection efficacy. As described in previous chapter, different components of EPS have different reactivity with disinfectants. Chlorine, as a non- selective disinfectant, reacted with both polysaccharides and proteins very rapidly. On the other hand, monochloramine reacts specifically with proteins while barely reacts with polysaccharides (Figure 3-4). Since monochloramine is less reactive than chlorine, many researchers have discussed its advantages and disadvantages over chlorine when applied as residual disinfectant in the distribution systems (137). Although monochloramine requires higher Ct values to achieve similar inactivation levels to those obtained with chlorine, monochloramine has been reported to be persistent in the distribution system and produce less disinfection by-products. More importantly, it has been reported that monochloramine is more effective than chlorine for biofilm disinfection due to its limited reactivity with and deeper penetration into the biofilm matrix.
Many studies have compared chlorine and monochloramine transport into biofilm; however their results varied largely depending on disinfectant concentration, exposure time, and more specifically influenced by local hydrodynamics as well as biomass heterogeneity for microelectrode applications. The chlorine penetration is known to be
112
limited by the reaction and diffusion mechanism due to high reactivity of chlorine with biofilm matrix. Chen and Stewart (1996) explored the reaction-diffusion interaction by applying chlorine (18.6 mg/L) into artificial biofilms and found that chlorine concentration at the substratum (526 µm thickness) was 10% of the bulk solution concentration after 3 hours exposure (147). Beer et al. (1994) observed the chlorine transfer boundary layer of a thickness of 100 to 400 µm at very high chlorine concentrations (0.28 to 0.42 mM) (148). Lee et al. (2011) reported that, with equivalent free chlorine concentration (2.8 mg/L) and exposure time, monochloramine penetrated deeper into biofilm than chlorine and by increasing disinfectant concentration and exposure time a deeper penetration could be achieved for both disinfectants (143).
However, the biofilm they used was too thick to be fully visualized using CLSM; so that the biofilm inactivation efficacy resulted from disinfectant penetration gradient was not proved in their study. In our study, biofilm thickness was relatively thinner compared to previous penetration studies, approximately 40 µm for all tested strains, but was more representative for drinking water biofilms (149). Moreover, the disinfectant concentrations applied in this study were chosen based on the minimum disinfectant residual guideline of EPA, in order to simulate vulnerable points in the distribution systems, especially the slow-flow sections and dead ends of the pipe network where conditions are known to be favorable for biofilm proliferation.
Although direct measurement of disinfectant gradient within biofilms was not conducted in this study, CLSM observation of biofilm viability spatial distribution clearly indicated disinfectant penetration. Unlike predictions in previous studies that increased resistance and survival of bacterial cells appeared near the substratum (102, 103), a
113
unique spatial distribution pattern of viability ratio was observed in this study. The viability ratio distribution begins near 0% at the biofilm surface then gradually increased with depth inside biofilm. In the middle section, a maximum viability ratio was reached; however, the viability ratio in the bottom section of biofilm began to decrease when approaching the substratum. The pattern of this viability ratio distribution was found similar to the distribution of biomass occupying each horizontal cross-section of biofilm
(Figure 4-8). Similar in all three strains, the middle section of biofilm had a more compact structure, possessing a higher density of biomass, when compared to the top and bottom sections. This more compacted interior structure may retard disinfectant transport in biofilm and may also be associated with low metabolic activity, oxygen and nutrient limitation, which promote tolerance against disinfectants (1). A difference in the middle section viability of EPS producing biofilms was observed between chlorine and monochloramine disinfection. The maximum viability appeared closer to biofilm surfaces
(defined as impact depth) and maintained constant in the middle section for EPS overproducing biofilm under chlorine disinfection, while maximum viability was only found in the midpoint of all three biofilms under monochloramine disinfection. This finding could be related to different reactivity of EPS with these two disinfectants within the more compact middle section structure. Previous studies pointed out that greater penetration capacity of monochloramine compensated for its disinfection efficacy, resulting in a similar inactivation level as chlorine in overall biofilm viability (10).
However, the advantage of monochloramine penetration could be clearly observed by the viability distribution within the middle biofilm section in this study.
114
Furthermore, the comparison of disinfection efficacy between chlorine and monochloramine could be made in response to their equivalent Ct values to a certain inactivation level. Our previous planktonic disinfection results showed that chlorine was much more effective than monochloramine (over 1,000 fold more effective), comparing the Ct values required for 99% inactivation level (Table 3-1). For biofilm disinfection, at the same exposure time for chlorine and monochloramine, the overall biofilm viability turned out to be similar under these two disinfectants. That is, monochloramine achieved the same biofilm inactivation level as chlorine with only 4 times of the Ct value than that of chlorine. The data indicated that, relative to the effectiveness for planktonic bacteria disinfection, monochloramine is hundreds fold more effective for biofilm disinfection than chlorine. It was previously observed that monochloramine Ct values for homogenized biofilms (145), as well as planktonic bacteria (20), were higher than those for intact biofilms to achieve near 100% inactivation. On the contrary, the required chlorine Ct values were higher for intact biofilms (145). Our finding confirmed with these reports that chlorine disinfect unattached bacteria more rapidly than monochloramine; while monochloramine is more effective than chlorine for biofilm inactivation. However, these previous studies were conducted with much higher disinfectant concentrations, which were close to or over the maximum disinfectant residual limits. Biofilm sloughing and detaching were observed under high disinfectant concentrations, especially for chlorine disinfection (143, 145). Therefore, further viability analysis of the detached clusters would not be reliable under relatively high disinfectant concentrations. In this study, with the minimum disinfectant residual concentrations applied, no significant biofilm detachment was observed under chlorine (0.5 mg/L) or
115
monochloramine (2 mg/L) disinfection for all tested strains according to the HPC results and flow cytometry results. Consequently, all detached cells were considered coming from natural detachment process, so that the disinfection efficacy of chlorine and monochloramine were able to be compared based on a non-interfered baseline of detached biomass quantity. Furthermore, disinfectant concentration in the collected flowcell effluent was monitored continuously during disinfection course. Both disinfectants were not completely depleted showing residual concentration of approximately 0.16 mg/L for chlorine and 1.2 mg/L for monochloramine. However, cell clusters detached from EPS producing biofilms were able to survive with these disinfectant residuals; the survival was prominent especially under monochloramine disinfection. Comparing effectiveness of chlorine and monochloramine for detached cluster inactivation, the equivalent Ct value of monochloramine was approximately 10 to
40 times over that of chlorine. The detached cluster viability data indicate that the detached clusters possess a transitional stage between planktonic and biofilm bacteria, thus exhibiting the susceptibility in between these two stages of biofilm life cycle.
6.5 Conclusion
The impact of biofilm EPS reactivity with disinfectants on disinfection efficacy was evaluated by monitoring biofilm viability, biofilm structure, and detached biofilm viability simultaneously during disinfection process. The obtained results suggested that: i) the production of EPS increased biofilm and detached biofilm resistance to both chlorine and monochloramine; ii) EPS production affected biofilm structure, specifically surface roughness, surface area to volume ratio and average diffusion distance; iii) these structural characteristics were closely related to both overall biofilm viability and the 116
spatial distribution of viability within biofilm; iv) although the overall viable rate was similar under the two disinfectants for each strain, monochloramine penetrated deeper into biofilm matrix than chlorine regardless the quantity of EPS content, showing a higher inactivation rate in the middle section of biofilms; v) chlorine was more efficient in controlling detached cluster viability than monochloramine. The combined results suggested that different reactivity of biofilm EPS with disinfectants influenced the susceptibility of both biofilm and detached biofilm during disinfection practices. This study will provide valuable information regarding how disinfection practices can be optimized to sufficiently maintain water quality by promoting biofilm inactivation and controlling viable detached clusters.
Acknowledgements
This project is supported by National Science Foundation (Award Number:
CBET0933288). We would like to acknowledge Dr. Daniel Hassett for providing the
Pseudomonas aeruginosa cultures.
117
Chapter 7
Conclusions
The results from this study provide a mechanistic understanding regarding the multiple roles of P. aeruginosa EPS in the biofilm life cycle, including planktonic bacteria, biofilm, detached biofilm clusters and redistributed biofilm. In depth evaluation of the capsular EPS from both whole cell and extracted EPS analyses revealed that capsular EPS may enhance bacterial survival in two distinct ways. Increased amount of
EPS on cell surface was associated with accelerated decay of chlorine, thus reducing availability and efficacy of residual chlorine for microbial inactivation. Alternatively, alginate EPS had minimal monochloramine demand, although the increased presence of alginate EPS was shown to reduce the efficacy of monochloramine, which showed high reactivity with proteins. Thus, the alginate EPS obscured disinfectant reactive sites on cell surface and retarded bacterial membrane interaction with monochloramine. In addition to EPS reactivity with different disinfectants, capsular EPS on cell membrane appeared to reduce membrane permeabilization by oxidative disinfectants, which was observed by functional group deformation. Functional moieties comprising the bacterial cell and capsular EPS were not eliminated, but were instead reduced in magnitude upon exposure to disinfectants. This suggests that the extensive membrane damage does not
118
occur given low concentrations of chlorine-based disinfectant residuals in water distribution systems. The planktonic bacteria results supported that capsular EPS, either by consuming disinfectant or limiting direct cell membrane access, provide a protective role for bacterial cells against regulatory residual disinfectants by reducing membrane permeabilization.
In addition, multiple roles of biofilm EPS were assessed in respect to both biofilm and detached cluster susceptibility upon chlorine and monochloramine disinfection by monitoring biofilm viability, biofilm structure, biosorption of DOM, and viability of detached biofilm clusters simultaneously. The obtained results suggested that: i) the presence of EPS increased both biofilm and detached cluster resistance to disinfection; ii)
EPS production had an influence on biofilm architecture; iii) some structural characteristics, such as surface roughness, surface area to volume ratio and average diffusion distance, were closely related to both overall biofilm viability and the spatial distribution of viability within biofilm; iv) the DOM adsorption did not show significant impact on biofilm viability against chlorine disinfection; however, the viability of detached biofilm clusters significantly increased in the presence of DOM, especially for the EPS overproducing strain; v) monochloramine was more effective to inactivate biofilm but less efficient in controlling detached cluster viability than chlorine; vi) even under the presence of chlorine, at a minimal residual concentration required in drinking water distribution systems, the detached clusters from all tested strains were still able to survive and recolonizd themselves downstream, forming new biofilms with relatively high viability; and vii) the redistributed biofilms of EPS producing strains had higher amount of total biomass, larger biofilm thickness, more complex structural properties for
119
both with and without chlorine conditions, and elevated viability under the presence of chlorine when compared to the counterpart EPS deficient strain.. The combined results revealed that multiple roles of biofilm EPS synergistically influence susceptibility of both biofilm and detached biofilm cluster leading to a higher resistance against disinfection practices. More importantly, the presence of EPS enhances detached cluster viability and recolonization, which may result in continuous deterioration of water quality. The results obtained in this study can provide valuable insights regarding the concentration of residual disinfectant required to facilitate biofilm inactivation and prevent bacteria regrowth in water distribution systems, where current levels do not provide sufficient disinfection for biofilm control and further suggest that accelerated chlorine decay may necessitate chlorine “boosting” in the distribution system to maintain biological stability.
120
Chapter 8
Recommendations for Future Research
Recommended topics for future studies based upon this work were included:
i) The impact of nutrient condition and hydrology condition on biofilm
formation and susceptibility can be further explored.
ii) Other bacteria strains or mix-culture biofilm can be explored to investigate the
influence of different EPS compositions and community communications in
the biofilm life cycle.
iii) To better understand EPS reactivity with disinfectants, longer exposure time
or higher disinfectant concentrations can be applied.
iv) Further investigations of detached biofilm redistribution under other
alternative disinfectants can be conducted to evaluate monochloramine
efficacy on detached cluster recolonization.
v) EPS composition and quantity in detached clusters can be assessed to closely
evaluate the role of EPS during detachment and redistribution processes.
vi) Direct measurement of disinfectant penetration using microelectrodes can be
considered for future study to correlate spatial distribution of biofilm viability
to Ct values.
121
vii) Reaction mechanisms and kinetics of EPS and disinfectants.
122
Reference
1. Wingender, J.; Neu, R. T.; Flemming, H.-C., Microbial extracellular polymeric
substances: characterization, structure and function. 1 edition ed.; Springer:
Berlin, 1999.
2. Flemming, H. C.; Neu, T. R.; Wozniak, D. J., The EPS matrix: the “house of
biofilm cells”. Journal of Bacteriology 2007, 189, (22), 7945-7947.
3. Tsuneda, S.; Aikawa, H.; Hayashi, H.; Yuasa, A.; Hirata, A., Extracellular
polymeric substances responsible for bacterial adhesion onto solid surface. FEMS
microbiology letters 2003, 223, (2), 287-292.
4. Tago, Y.; Aida, K., Exocellular mucopolysaccharide closely related to bacterial
floc formation. Applied and Environmental Microbiology 1977, 34, (3), 308-314.
5. Petersen, F. C.; Tao, L.; Scheie, A. A., DNA binding-uptake system: a link
between cell-to-cell communication and biofilm formation. Journal of
Bacteriology 2005, 187, (13), 4392-4400.
6. Jost Wingender, T. R. N., Hans-Curt Flemming, Microbial extracellular
polymeric substances: characterization, structure and function. 1 edition ed.;
Springer: Berlin, 1999.
7. Chiras, D. D., Human biology : health, homeostasis, and the environment. West
Pub. Co.: St. Paul, 1991; p xxiv, 610, [54] p.
123
8. Burke, J. P., Infection control — A problem for patient safety. New England
Journal of Medicine 2003, 348, (7), 651-656.
9. National-Research-Council, Identifying future drinking water contaminants. 1998
Workshop on Emerging Drinking Water Contaminants 1999, (10), 206.
10. Lechevallier, M. W.; Cawthon, C. D.; Lee, R. G., Inactivation of biofilm bacteria.
Applied and Environmental Microbiology 1988, 54, (10), 2492-2499.
11. Koskinen, R.; Ali-Vehmas, T.; Kampfer, P.; Laurikkala, M.; Tsitko, I.; Kostyal,
E.; Atroshi, F.; Salkinoja-Salonen, M., Characterization of Sphingomonas isolates
from Finnish and Swedish drinking water distribution systems. Journal of Applied
Microbiology 2000, 89, (4), 687-696.
12. O'Toole, G.; Kaplan, H. B.; Kolter, R., Biofilm formation as microbial
development. Annual Review of Microbiology 2000, 54, 49-79.
13. Flemming, H. C.; Tamachkiarowa, A.; Klahre, J.; Schmitt, J. In Monitoring of
fouling and biofouling in technical systems, 1998; Pergamon-Elsevier Science Ltd:
1998; pp 291-298.
14. Costerton, J. W.; Stewart, P. S.; Greenberg, E. P., Bacterial biofilms: A common
cause of persistent infections. Science 1999, 284, (5418), 1318-1322.
15. Chambless, J. D.; Hunt, S. M.; Stewart, P. S., A three-dimensional computer
model of four hypothetical mechanisms protecting biofilms from antimicrobials.
Applied and Environmental Microbiology 2006, 72, (3), 2005-2013.
16. Momba, M. N. B.; Kfir, R.; Venter, S. N.; Cloete, T. E., An overview of biofilm
formation in distribution systems and its impact on the deterioration of water
quality. Water Sa 2000, 26, (1), 59-66.
124
17. Characklis W.G., M. K. C., Biofilm. Wiley: New York, 1990; p 585-633.
18. Stewart, P. S., Mechanisms of antibiotic resistance in bacterial biofilms.
International Journal of Medical Microbiology 2002, 292, (2), 107-113.
19. Christensen, B. B.; Sternberg, C.; Andersen, J. B.; Palmer, R. J.; Nielsen, A. T.;
Givskov, M.; Molin, S., Molecular tools for study of biofilm physiology. In
Biofilms, Academic Press Inc: San Diego, 1999; Vol. 310, pp 20-42.
20. Lechevallier, M. W.; Lowry, C. D.; Lee, R. G., Disinfecting biofilms in a model
distribution system. Journal American Water Works Association 1990, 82, (7),
87-99.
21. Camper, A. K.; McFeters, G. A.; Characklis, W. G.; Jones, W. L., Growth
kinetics of coliform bacteria under conditions relevant to drinking water
distribution systems. Applied and Environmental Microbiology 1991, 57, (8),
2233-2239.
22. Maul, A.; Elshaarawi, A. H.; Block, J. C., Heterotrophic bacteria in water
distribution systems. I. Spatial and temporal variation. Science of the Total
Environment 1985a, 44, (3), 201-214.
23. Maul, A.; Elshaarawi, A. H.; Block, J. C., Heterotrophic bacteria in water
distribution systems. II. Sampling design for monitoring. Science of the Total
Environment 1985b, 44, (3), 215-224.
24. Brooks, J. D.; Flint, S. H., Biofilms in the food industry: problems and potential
solutions. International Journal of Food Science and Technology 2008, 43, (12),
2163-2176.
125
25. Xu, X. M.; Stewart, P. S.; Chen, X., Transport limitation of chlorine disinfection
of Pseudomonas aeruginosa entrapped in alginate beads. Biotechnology and
Bioengineering 1996, 49, (1), 93-100.
26. Bryers, J. D., Effects of cell recycle on microbial cell viability and metabolism.
PDA Journal of Pharmaceutical Science and Technology 1988, 42, (2), 40-46.
27. Stewart, P. S., A model of biofilm detachment. Biotechnology and Bioengineering
1993, 41, (1), 111-117.
28. Picioreanu, C.; van Loosdrecht, M. C. M.; Heijnen, J. J., Two-dimensional model
of biofilm detachment caused by internal stress from liquid flow. Biotechnology
and Bioengineering 2001, 72, (2), 205-218.
29. Characklis, W. G.; Trulear, M. G.; Bryers, J. D.; Zelver, N., Dynamics of biofilm
processes: methods. Water Research 1982, 16, (7), 1207-1216.
30. LeChevallier, M. W.; Welch, N. J.; Smith, D. B., Full-scale studies of factors
related to coliform regrowth in drinking water. Applied and Environmental
Microbiology 1996, 62, (7), 2201-2211.
31. Lechevallier, M. W.; Schulz, W.; Lee, R. G., Bacterial nutrients in drinking water.
Applied and Environmental Microbiology 1991, 57, (3), 857-862.
32. Thormann, K. M.; Saville, R. M.; Shukla, S.; Spormann, A. M., Induction of rapid
detachment in Shewanella oneidensis MR-1 biofilms. Journal of Bacteriology
2005, 187, (3), 1014-1021.
33. Wilson, S.; Hamilton, M. A.; Hamilton, G. C.; Schumann, M. R.; Stoodley, P.,
Statistical quantification of detachment rates and size distributions of cell cumps
126
from wild-type (PAO1) and cell signaling mutant (JP1) Pseudomonas aeruginosa
biofilms. Applied and Environmental Microbiology 2004, 70, (10), 5847-5852.
34. Jjemba, P. K.; Weinrich, L. A.; Cheng, W.; Giraldo, E.; LeChevallier, M. W.,
Regrowth of potential opportunistic pathogens and algae in reclaimed-water
distribution systems. Applied and Environmental Microbiology 2010, 76, (13),
4169-4178.
35. Liu, Y.; Yang, C. H.; Li, J., Influence of extracellular polymeric substances on
Pseudomonas aeruginosa transport and deposition profiles in porous media.
Environmental Science & Technology 2007, 41, (1), 198-205.
36. Madigan, M. T.; Martinko, J. M., Brock biology of microorganisms. Prentice Hall:
New Jersey, 2006.
37. Wingender, J.; Neu, T. R.; Flemming, H.-C., Microbial extracellular polymeric
substances: characterization, structure and function. 1 edition ed.; Springer:
Berlin, 1999.
38. Liu, Y.; Li, J.; Qiu, X. F.; Burda, C., Bactericidal activity of nitrogen-doped metal
oxide nanocatalysts and the influence of bacterial extracellular polymeric
substances (EPS). Journal of Photochemistry and Photobiology a-Chemistry 2007,
190, (1), 94-100.
39. Mitchell, A. C., Phillips AJ, Hamilton MA, Gerlach R, Hollis WK, Kaszuba JP,
Cunningham AB., Resilience of planktonic and biofilm cultures to supercritical
CO2. Journal of Supercritical Fluids 2008, 47, (2), 318-325.
127
40. Hodges, N. A.; Gordon, C. A., Protection of Pseudomonas aeruginosa against
ciprofloxacin and β-lactams by homologous alginate. Antimicrobial Agents and
Chemotherapy 1991, 35, (11), 2450-2452.
41. Helbling, D. E.; VanBriesen, J. M., Free chlorine demand and cell survival of
microbial suspensions. Water Research 2007, 41, (19), 4424-4434.
42. Virto, R.; Manas, P.; Alvarez, I.; Condon, S.; Raso, J., Membrane damage and
microbial inactivation by chlorine in the absence and presence of a chlorine-
demanding substrate. Applied and Environmental Microbiology 2005, 71, (9),
5022-5028.
43. Hurst, J. K.; Barrette, W. C.; Michel, B. R.; Rosen, H., Hypochlorous acid and
myeloperoxidase-catalyzed oxidation of iron-sulfur clusters in bacterial
respiratory dehydrogenases. European Journal of Biochemistry 1991, 202, (3),
1275-1282.
44. Chhabra, S.; Philipp, B.; Eberl, L.; Givskov, M.; Williams, P.; Cámara, M.,
Extracellular communication in bacteria. In The chemistry of pheromones and
other semiochemicals II, Schulz, S., Ed. Springer Berlin / Heidelberg: 2005; Vol.
240, pp 117-128.
45. Laspidou, C. S.; Rittmann, B. E., A unified theory for extracellular polymeric
substances, soluble microbial products, and active and inert biomass. Water
Research 2002, 36, (11), 2711-2720.
46. Beech, I.; Hanjagsit, L.; Kalaji, M.; Neal, A. L.; Zinkevich, V., Chemical and
structural characterization of exopolymers produced by Pseudomonas sp NCIMB
2021 in continuous culture. Microbiology-Uk 1999, 145, 1491-1497.
128
47. Gagnon, G. A.; O'Leary, K. C.; Volk, C. J.; Chauret, C.; Stover, L.; Andrews, R.
C., Comparative analysis of chlorine dioxide, free chlorine and chloramines on
bacterial water quality in model distribution systems. Journal of Environmental
Engineering-Asce 2004, 130, (11), 1269-1279.
48. Hardalo, C.; Edberg, S. C., Pseudomonas aeruginosa: Assessment of risk from
drinking water. Critical Reviews in Microbiology 1997, 23, (1), 47-75.
49. DeQueiroz, G. A.; Day, D. F., Antimicrobial activity and effectiveness of a
combination of sodium hypochlorite and hydrogen peroxide in killing and
removing Pseudomonas aeruginosa biofilms from surfaces. Journal of Applied
Microbiology 2007, 103, (4), 794-802.
50. Geldreich, E. E., Drinking-water microbiology - new directions toward water-
quality enhancement. International Journal of Food Microbiology 1989, 9, (4),
295-312.
51. Schweizer, H. P., Two plasmids, X1918 and Z1918, for easy recovery of the xylE
and lacZ reporter genes. Gene 1993, 134, (1), 89-91.
52. Schweizer, H. P.; Hoang, T. T., An improved system for gene replacement and
xylE fusion analysis in Pseudomonas aeruginosa. Gene 1995, 158, (1), 15-22.
53. Eboigbodin, K. E.; Biggs, C. A., Characterization of the extracellular polymeric
substances produced by Escherichia coli using infrared spectroscopic, proteomic,
and aggregation studies. Biomacromolecules 2008, 9, (2), 686-695.
54. Thurston-Enriquez, J. A.; Haas, C. N.; Jacangelo, J.; Gerba, C. P., Chlorine
inactivation of adenovirus type 40 and feline calicivirus. Applied and
Environmental Microbiology 2003, 69, (7), 3979-3985.
129
55. Ridgway, H. F.; Olson, B. H., Chlorine resistance patterns of bacteria from two
drinking water distribution systems. Applied and Environmental Microbiology
1982, 44, (4), 972-987.
56. Luh, J.; Marinas, B. J., Inactivation of Mycobacterium avium with free chlorine.
Environmental Science & Technology 2007, 41, (14), 5096-5102.
57. Vicuna-Reyes, J. P.; Luh, J.; Marinas, B. J., Inactivation of Mycobacterium avium
with chlorine dioxide. Water Research 2008, 42, (6-7), 1531-1538.
58. Carter, J. T.; Rice, E. W.; Buchberger, S. G.; Lee, Y., Relationships between
levels of heterotrophic bacteria and water quality parameters in a drinking water
distribution system. Water Research 2000, 34, (5), 1495-1502.
59. USEPA, Microbial and disinfection byproduct rules simultaneous compliance
guidance manual. In 1999; Vol. EPA 815-R-99-015.
60. Qiang, Z.; Macauley, J. J.; Mormile, M. R.; Surampalli, R.; Adams, C. D.,
Treatment of antibiotics and antibiotic resistant bacteria in swine wastewater with
free chlorine. Journal of Agricultural and Food Chemistry 2006, 54, (21), 8144-
8154.
61. Geeraerd, A. H.; Valdramidis, V. P.; Van Impe, J. F., GInaFiT, a freeware tool to
assess non-log-linear microbial survivor curves. International Journal of Food
Microbiology 2005, 102, (1), 95-105.
62. Berney, M.; Weilenmann, H. U.; Ihssen, J.; Bassin, C.; Egli, T., Specific growth
rate determines the sensitivity of Escherichia coli to thermal, UVA, and solar
disinfection. Applied and Environmental Microbiology 2006, 72, (4), 2586-2593.
130
63. Berney, M.; Hammes, F.; Bosshard, F.; Weilenmann, H. U.; Egli, T., Assessment
and interpretation of bacterial viability by using the LIVE/DEAD BacLight kit in
combination with flow cytometry. Applied and Environmental Microbiology 2007,
73, (10), 3283-3290.
64. Stocks, S. M., Mechanism and use of the commercially available viability stain,
BacLight. Cytometry Part A 2004, 61A, (2), 189-195.
65. Brown, M. J.; Lester, J. N., Comparison of bacterial extracellular polymer
extraction methods. Applied and Environmental Microbiology 1980, 40, (2), 179-
185.
66. Knutson, C. A.; Jeanes, A., A new modification of the carbazole analysis:
application to heteropolysaccharide. Analytical Biochemistry 1968, 24, (3), 470-&.
67. Sohn, J.; Amy, G.; Cho, J. W.; Lee, Y.; Yoon, Y., Disinfectant decay and
disinfection by-products formation model development: chlorination and
ozonation by-products. Water Research 2004, 38, (10), 2461-2478.
68. Al-Qadiri, H. M.; Al-Alami, N. I.; Al-Holy, M. A.; Rasco, B. A., Using Fourier
transform infrared (FT-IR) absorbance spectroscopy and multivariate analysis to
study the effect of chlorine-induced bacterial injury in water. Journal of
Agricultural and Food Chemistry 2008, 56, (19), 8992-8997.
69. Holt, C.; Hirst, D.; Sutherland, A.; Macdonald, F., Discrimination of species in
the genus listeria by fourier-transform infrared-spectroscopy and canonical variate
analysis. Applied and Environmental Microbiology 1995, 61, (1), 377-378.
131
70. Tielen, P.; Rosenau, F.; Wilhelm, S.; Jaeger, K.-E.; Flemming, H.-C.; Wingender,
J., Extracellular enzymes affect biofilm formation of mucoid Pseudomonas
aeruginosa. Microbiology 2010, 156, (7), 2239-2252.
71. Kansiz, M.; Heraud, P.; Wood, B.; Burden, F.; Beardall, J.; McNaughton, D.,
Fourier transform infrared microspectroscopy and chemometrics as a tool for the
discrimination of cyanobacterial strains. Phytochemistry 1999, 52, (3), 407-417.
72. Hentzer, M.; Teitzel, G. M.; Balzer, G. J.; Heydorn, A.; Molin, S.; Givskov, M.;
Parsek, M. R., Alginate overproduction affects Pseudomonas aeruginosa biofilm
structure and function. Journal of Bacteriology 2001, 183, (18), 5395-5401.
73. Connell, G. F., The chlorination/chloramination handbook. American Water
Works Association: 1996.
74. Jacangelo, J. G.; Olivieri, V. P.; Kawata, K., Investigating the mechanism of
inactivation of Escherichia coli B by monochloramine. Journal American Water
Works Association 1991, 83, (5), 80-87.
75. Gang, D. C.; Clevenger, T. E.; Banerji, S. K., Relationship of chlorine decay and
THMs formation to NOM size. Jounal of Hazadous Materials 2003, 96, (1), 1-12.
76. Kouame, Y.; Haas, C. N., Inactivation of Escherichia coli by combined action of
free chlorine and monochloramine. Water Research 1991, 25, (9), 1027-1032.
77. Rowan, N. J., Viable but nonculturable forms of food and waterborne bacteria:
Quo Vadis? Trends in Food Science & Technology 2004, 15, (9), 462-467.
78. Phe, M. H.; Dossot, M.; Guilloteau, H.; Block, J. C., Nucleic acid fluorochromes
and flow cytometry prove useful in assessing the effect of chlorination on
drinking water bacteria. Water Research 2005, 39, (15), 3618-3628.
132
79. Samrakandi, M. M.; Roques, C.; Michel, G., Influence of trophic conditions on
exopolysaccharide production: bacterial biofilm susceptibility to chlorine and
monochloramine. Canadian Journal of Microbiology 1997, 43, (8), 751-758.
80. Venkobachar, C.; Iyengar, L.; Rao, A. V. S. P., Mechanism of disinfection: effect
of chlorine on cell membrane functions. Water Research 1977, 11, (8), 727-729.
81. Hong, Y.; Brown, D. G., Cell surface acid-base properties of Escherichia coli and
Bacillus brevis and variation as a function of growth phase, nitrogen source and
C : N ratio. Colloids and Surfaces B-Biointerfaces 2006, 50, (2), 112-119.
82. Camper, A. K., Coliform regrowth and biofilm accumulation in drinking water
systems - a review. Biofouling and Biocorrosion in Industrial Water Systems 1994,
91-105.
83. Schwartz, T.; Hoffmann, S.; Obst, U., Formation and bacterial composition of
young, natural biofilms obtained from public bank-filtered drinking water systems.
Water Research 1998, 32, (9), 2787-2797.
84. Gilbert, P.; Das, J.; Foley, I., Biofilm susceptibility to antimicrobials. Advances in
Dental Research 1997, 11, (1), 160-7.
85. Haas, C. N.; Joffe, J.; Heath, M.; Jacangelo, J.; Anmangandla, U., Predicting
disinfection performance in continuous flow systems from batch disinfection
kinetics. Water Science and Technology 1998, 38, (6), 171-179.
86. Herzberg, M.; Kang, S.; Elimelech, M., Role of extracellular polymeric
substances (EPS) in biofouling of reverse osmosis membranes. Environmental
Science & Technology 2009, 43, (12), 4393-4398.
133
87. Behnke, S.; Parker, A. E.; Woodall, D.; Camper, A. K., Comparing the chlorine
disinfection of detached biofilm clusters with those of sessile biofilms and
planktonic cells in single and dual species cultures. Applied and Environmental
Microbiology 2011, 77, (20), 7176-7184.
88. Walters, M. C.; Roe, F.; Bugnicourt, A.; Franklin, M. J.; Stewart, P. S.,
Contributions of antibiotic penetration, oxygen limitation, and low metabolic
activity to tolerance of Pseudomonas aeruginosa biofilms to ciprofloxacin and
tobramycin. Antimicrobial Agents and Chemotherapy 2003, 47, (1), 317-323.
89. Sauer, K.; Camper, A. K.; Ehrlich, G. D.; Costerton, J. W.; Davies, D. G.,
Pseudomonas aeruginosa displays multiple phenotypes during development as a
biofilm. Journal of Bacteriology 2002, 184, (4), 1140-1154.
90. Simões, M.; Simões, L. C.; Pereira, M. O.; Vieira, M. J., Antagonism between
Bacillus cereus and Pseudomonas fluorescens in planktonic systems and in
biofilms. Biofouling 2008, 24, (5), 339-349.
91. Purevdorj, B.; Costerton, J. W.; Stoodley, P., Influence of hydrodynamics and cell
signaling on the structure and behavior of Pseudomonas aeruginosa biofilms.
Applied and Environmental Microbiology 2002, 68, (9), 4457-4464.
92. Delille, A.; Quilès, F.; Humbert, F., In situ monitoring of the nascent
Pseudomonas fluorescens biofilm response to variations in the dissolved organic
carbon level in low-nutrient water by attenuated total reflectance-fourier
transform infrared spectroscopy. Applied and Environmental Microbiology 2007,
73, (18), 5782-5788.
134
93. Croue, J. P.; Korshin, G. V.; Benjamin, M. M., Characterization of natural
organic matter in drinking water. American Water Works Association Research
Foundation: Denver, 1999.
94. Engelbrecht, R. S.; Weber, M. J.; Salter, B. L.; Schmidt, C. A., Comparative
inactivation of viruses by chlorine. Applied and Environmental Microbiology
1980, 40, (2), 249-256.
95. Rogers, S. S.; van der Walle, C.; Waigh, T. A., Microrheology of bacterial
biofilms in vitro: Staphylococcus aureus and Pseudomonas aeruginosa. Langmuir
2008, 24, (23), 13549-13555.
96. Gang, D.; Clevenger, T. E.; Banerji, S. K., Relationship of chlorine decay and
THMs formation to NOM size. Journal of Hazardous Materials 2003, 96, (1), 1-
12.
97. Lu, W.; Kiéné, L.; Lévi, Y., Chlorine demand of biofilms in water distribution
systems. Water Research 1999, 33, (3), 827-835.
98. Strathmann, M.; Wingender, J.; Flemming, H.-C., Application of fluoroscently
labelled lectins for the visualization and biochemical characterization of
polysaccharides in biofilms of Pseudomonas aeruginosa. Journal of
Microbiological Methods 2002, 50, 237-248.
99. Heydorn, A.; Nielsen, T. A.; Hentzer, M.; Sternberg, C.; Givskov, M.; Ersboll, K.
B.; Molin, S., Quantification of biofilm structures by the novel program
COMSTAT. Microbiology 2000, 146, 2395-2407.
135
100. Gregori, J. G.; Ragheb, K.; Robinson, J. P., A tutorial of WINMDI version 2.8.
22nd Congress of International Society for Analytical Cytology 2004, 59A, (1),
156-156.
101. Davies, D. G.; Parsek, M. R.; Pearson, J. P.; Iglewski, B. H.; Costerton, J. W.;
Greenberg, E. P., The involvement of cell-to-cell signals in the development of a
bacterial biofilm. Science 1998, 280, (5361), 295-298.
102. Gordon, C. A.; Hodges, N. A.; Marriott, C., Antibiotic interaction and diffusion
through alginate and exopolysaccharide of cystic fibrosis-derived Pseudomonas
aeruginosa. Journal of Antimicrobial Chemotherapy 1988, 22, (5), 667-674.
103. Stewart, P. S.; William Costerton, J., Antibiotic resistance of bacteria in biofilms.
The Lancet 2001, 358, (9276), 135-138.
104. Stoodley, P.; Sauer, K.; Davies, D. G.; Costerton, J. W., Biofilm as complex
differentiated communities. Annual Review of Microbiology 2002, 56, (1), 187-
209.
105. Klausen, M.; Aaes-Jørgensen, A.; Molin, S.; Tolker-Nielsen, T., Involvement of
bacterial migration in the development of complex multicellular structures in
Pseudomonas aeruginosa biofilms. Molecular Microbiology 2003, 50, (1), 61-68.
106. USEPA, Control of biofilm growth in drinking water distribution systems. 1992.
107. Wang, Z.; Hessler, C. M.; Xue, Z.; Seo, Y., The role of extracellular polymeric
substances on the sorption of natural organic matter. Water Research 2012, 46,
(4), 1052-1060.
136
108. Fux, C. A.; Wilson, S.; Stoodley, P., Detachment characteristics and oxacillin
resistance of Staphyloccocus aureus biofilm emboli in an in vitro catheter
infection model. Journal of Bacteriology 2004, 186, (14), 4486-4491.
109. Romeo, T., When the party is over: A signal for dispersal of Pseudomonas
aeruginosa biofilms. Journal of Bacteriology 2006, 188, (21), 7325-7327.
110. Salek, M. M.; Jones, S. M.; Martinuzzi, R. J., The influence of flow cell geometry
related shear stresses on the distribution, structure and susceptibility of
Pseudomonas aeruginosa 01 biofilms. Biofouling 2009, 25, (8), 711-725.
111. Gjermansen, M.; Nilsson, M.; Yang, L.; Tolker-Nielsen, T., Characterization of
starvation-induced dispersion in Pseudomonas putida biofilms: genetic elements
and molecular mechanisms. Molecular Microbiology 2010, 75, (4), 815-826.
112. Sauer, K.; Cullen, M. C.; Rickard, A. H.; Zeef, L. A. H.; Davies, D. G.; Gilbert, P.,
Characterization of nutrient-induced dispersion in Pseudomonas aeruginosa
PAO1 biofilm. Journal of Bacteriology 2004, 186, (21), 7312-7326.
113. Thormann, K. M.; Duttler, S.; Saville, R. M.; Hyodo, M.; Shukla, S.; Hayakawa,
Y.; Spormann, A. M., Control of formation and cellular detachment from
Shewanella oneidensis MR-1 biofilms by cyclic di-GMP. Journal of Bacteriology
2006, 188, (7), 2681-2691.
114. Wilson, S.; Hamilton, M. A.; Hamilton, G. C.; Schumann, M. R.; Stoodley, P.,
Statistical quantification of detachment rates and size distributions of cell clumps
from wild-type (PAO1) and cell signaling mutant (JP1) Pseudomonas aeruginosa
biofilms. Applied and Environmental Microbiology 2004, 70, (10), 5847-5852.
137
115. Purevdorj-Gage, B.; Costerton, W. J.; Stoodley, P., Phenotypic differentiation and
seeding dispersal in non-mucoid and mucoid Pseudomonas aeruginosa biofilms.
Microbiology-Sgm 2005, 151, 1569-1576.
116. Itoh, Y.; Wang, X.; Hinnebusch, B. J.; Preston, J. F.; Romeo, T.,
Depolymerization of beta-1,6-N-acetyl-D-glucosamine disrupts the integrity of
diverse bacterial biofilms. Journal of Bacteriology 2005, 187, (1), 382-387.
117. Kaplan, J. B.; Ragunath, C.; Ramasubbu, N.; Fine, D. H., Detachment of
Actinobacillus actinomycetemcomitans biofilm cells by an endogenous beta-
hexosaminidase activity. Journal of Bacteriology 2003, 185, (16), 4693-4698.
118. Morgan, R.; Kohn, S.; Hwang, S. H.; Hassett, D. J.; Sauer, K., BdlA, a
chemotaxis regulator essential for biofilm dispersion in Pseudomonas aeruginosa.
J Bacteriol 2006, 188, (21), 7335-43.
119. Ramsey, D. M.; Wozniak, D. J., Understanding the control of Pseudomonas
aeruginosa alginate synthesis and the prospects for management of chronic
infections in cystic fibrosis. Molecular Microbiology 2005, 56, (2), 309-322.
120. Besner, M. C.; Servais, P.; Prevost, M., Efficacy of disinfectant residual on
microbial intrusion: A review of experiments. Journal American Water Works
Association 2008, 100, (10), 116-+.
121. Fux, C. A.; Wilson, S.; Stoodley, P., Detachment characteristics and oxacillin
resistance of Staphyloccocus aureus biofilm emboli in an in vitro catheter
infection model. J Bacteriol 2004, 186, (14), 4486-91.
122. Hunter, R. J., Zeta potential in colloid science: principles and applications. 1981.
138
123. Gacesa, P., Bacterial alginate biosynthesis - recent progress and future prospects.
Microbiology 1998, 144, (5), 1133-1143.
124. Wozniak, D. J.; Wyckoff, T. J. O.; Starkey, M.; Keyser, R.; Azadi, P.; O'Toole, G.
A.; Parsek, M. R., Alginate is not a significant component of the extracellular
polysaccharide matrix of PA14 and PAO1 Pseudomonas aeruginosa biofilms.
Proceedings of the National Academy of Sciences 2003, 100, (13), 7907-7912.
125. Palmer, J.; Flint, S.; Brooks, J., Bacterial cell attachment, the beginning of a
biofilm. Journal of industrial microbiology & biotechnology 2007, 34, (9), 577-
588.
126. Dunne Jr, W. M., Bacterial adhesion: seen any good biofilms lately? Clinical
microbiology reviews 2002, 15, (2), 155-166.
127. Kumar, C. G.; Anand, S., Significance of microbial biofilms in food industry: a
review. International Journal of Food Microbiology 1998, 42, (1), 9-27.
128. Hermansson, M., The DLVO theory in microbial adhesion. Colloids and Surfaces
B: Biointerfaces 1999, 14, (1-4), 105-119.
129. Rijnaarts, H. H. M.; Norde, W.; Bouwer, E. J.; Lyklema, J.; Zehnder, A. J. B.,
Reversibility and mechanism of bacterial adhesion. Colloids and Surfaces B:
Biointerfaces 1995, 4, (1), 5-22.
130. Jacobs, A.; Lafolie, F.; Herry, J.; Debroux, M., Kinetic adhesion of bacterial cells
to sand: cell surface properties and adhesion rate. Colloids and Surfaces B:
Biointerfaces 2007, 59, (1), 35-45.
131. Li, B.; Logan, B. E., Bacterial adhesion to glass and metal-oxide surfaces.
Colloids and Surfaces B: Biointerfaces 2004, 36, (2), 81-90.
139
132. Jucker, B. A.; Zehnder, A. J. B.; Harms, H., Quantification of polymer
interactions in bacterial adhesion. Environmental Science & Technology 1998, 32,
(19), 2909-2915.
133. Block, J.; Haudidier, K.; Paquin, J.; Miazga, J.; Levi, Y., Biofilm accumulation in
drinking water distribution systems. Biofouling 1993, 6, (4), 333-343.
134. Block, J., Biofilms in drinking water distribution systems. Biofilms—Science and
Technology 1992, 469.
135. Stoodley, P.; Sauer, K.; Davies, D.; Costerton, J., Biofilms as complex
differentiated communities. Annual Reviews in Microbiology 2002, 56, (1), 187-
209.
136. Boorman, G. A., Drinking water disinfection byproducts: review and approach to
toxicity evaluation. Environmental Health Perspectives 1999, 107, (Suppl 1), 207.
137. Sadiq, R.; Rodriguez, M. J., Disinfection by-products (DBPs) in drinking water
and predictive models for their occurrence: a review. Science of the Total
Environment 2004, 321, (1), 21-46.
138. LeBel, G. L.; Benoit, F. M.; Williams, D. T., A one-year survey of halogenated
disinfection by-products in the distribution system of treatment plants using three
different disinfection processes. Chemosphere 1997, 34, (11), 2301-2317.
139. Neden, D. G.; Jones, R. J.; Smith, J. R.; Kirmeyer, G. J.; Foust, G. W., Comparing
chlorination and chloramination for controlling bacterial regrowth. Journal
American Water Works Association 1992, 84, (7), 80-88.
140
140. Momba, M. N. B.; Cloete, T. E.; Venter, S. N.; Kfir, R., Evaluation of the impact
of disinfection processes on the formation of biofilms in potable surface water
distribution systems. Water Science and Technology 1998, 38, (8-9), 283-289.
141. Norton, C. D.; LeChevallier, M. W., Chloramination: Its effect on distribution
system water quality. Journal American Water Works Association 1997, 89, (7),
66-77.
142. Zhang, W.; DiGiano, F. A., Comparison of bacterial regrowth in distribution
systems using free chlorine and chloramine: a statistical study of causative factors.
Water Research 2002, 36, (6), 1469-1482.
143. Lee, W. H.; Wahman, D. G.; Bishop, P. L.; Pressman, J. G., Free chlorine and
monochloramine application to nitrifying biofilm: comparison of biofilm
penetration, activity, and viability. Environmental Science & Technology 2011, 45,
(4), 1412-1419.
144. Chandy, J.; Angles, M., Determination of nutrients limiting biofilm formation and
the subsequent impact on disinfectant decay. Water Research 2001, 35, (11),
2677-2682.
145. Griebe, T.; Chen, C. I.; Srinivasan, R.; Stewart, P. S., Analysis of biofilm
disinfection by monochloramine and free chlorine. 1994; p 151-160.
146. Cogan, N.; Keener, J. P., The role of the biofilm matrix in structural development.
Mathematical Medicine and Biology 2004, 21, (2), 147-166.
147. Chen, X.; Stewart, P. S., Chlorine penetration into artificial biofilm is limited by a
reaction-diffusion interaction. Environmental Science & Technology 1996, 30, (6),
2078-2083.
141
148. De Beer, D.; Srinivasan, R.; Stewart, P. S., Direct measurement of chlorine
penetration into biofilms during disinfection. Applied and Environmental
Microbiology 1994, 60, (12), 4339-4344.
149. Batté, M.; Appenzeller, B.; Grandjean, D.; Fass, S.; Gauthier, V.; Jorand, F.;
Mathieu, L.; Boualam, M.; Saby, S.; Block, J., Biofilms in drinking water
distribution systems. Reviews in Environmental Science and Biotechnology 2003,
2, (2), 147-168.
142