1

.5...
.wnj....§ . ‘
a... l.

$.44 o
4 a“ fixéﬁ.
«v ‘

 

 

 

M313

9“ IIIIUIHIHH"HIHHHIHIIHII.WWWIIUHWI

31293 01772 128 7

This is to certify that the

thesis entitled

INFLUENCES OF SESSILE PRIMARY PRODUCTIVITY AND METI-IANOL ON
MICROBIAL COMMUNITY STRUCTURE AND SECONDARY METABOLISM IN
AN INDUSTRIAL GROUNDWATER RECHARGE POND

presented by

Timothy P. Laatsch

has been accepted towards fulﬁllment
of the requirements for

MS. degree in Crop and Sgil Sciences

//

 

Date 8,0345

0-7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

 

 

LIBRARY

Mlchlgan State

University

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1m WMpQ-pu

INFLUENCES OF SESSILE PRIMARY PRODUCTIVITY AND METHANOL ON
MICROBIAL COMMUNITY STRUCTURE AND SECONDARY METABOLISM IN
AN INDUSTRIAL GROUNDWATER RECHARGE POND
By

Timothy P. Laatsch

A THESIS

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Crop and Soil Sciences

1998

ABSTRACT

INFLUENCES OF SESSILE PRIMARY PRODUCTIVITY AND METHANOL ON
MICROBIAL COMMUNITY STRUCTURE AND SECONDARY METABOLISM IN
AN INDUSTRIAL GROUNDWATER RECHARGE POND

By

Timothy P. Laatsch

The Pharmacia/Upjohn Pond receives 3x107 L day'l of non-contact cooling water containing mg
L'I concentrations of methanol and ug L'I concentrations of several organic solvents. Methanol
was previously shown to be biologically oxidized by bioﬁlm communities in the system;
however, metabolism of other solvents is less well understood. The objective of this study was
to determine the inﬂuences of sessile primary productivity and methanol on metabolism of other
solvents. Attached microbial communities were allowed to develop on PVC surfaces in situ
under LIGHT and DARK treatment regimes. Mature bioﬁlms were harvested for microbial
community structural analyses and aerobic metabolic studies with methanol as a primary growth
substrate and several secondary non-growth substrates: dichloromethane (DCM), toluene (TOL),
and chlorobenzene (CB). Direct microscopic counts did not vary signiﬁcantly between
treatments, but fatty acid methyl ester (FAME) proﬁles revealed differences in microbial
community structure. TOL and CB were secondarily transformed at a signiﬁcantly higher rate in
communities colonized in the dark. Secondary transformation of DCM was observed only in
bioﬁlms that were pre-incubated with a constant source of methanol. Sessile photosynthesis did
not stimulate heterotrophic bioremediation activity on a primary or secondary metabolic level.
Primary metabolism of methanol enhanced secondary metabolism of DCM, but had no effect on

secondary metabolism of other solvents.

To my beautiful wife Jane, whose love, support, and self-sacriﬁce have allowed me to
realize my dreams

iii

e7

ACKNOWLEDGEMENTS

Foremost, I must thank God for the multitude of blessings in my life and for
providing the sense of purpose to help carry me through the darkest moments.

I am deeply indebted to my loving wife Jane for standing so earnestly by my side
through all the trials and tribulations of the past four years. Her abiding love and patience
were at times my only source of inspiration to continue toward completion of this work.
Jane sacriﬁced, postponed, or compromised many of her aspirations, so that I could fully
pursue my own and for this I will be forever grateful.

Special thanks to my advisor, Dr. Michael J. Klug, for sharing his experience and
insight with me and for guiding me down the sometimes treacherous path of scientiﬁc
research. Many thanks are also due him for helping to arrange the partnership with
Pharmacia/Upjohn that made this work possible and for providing excellent laboratory
facilities and equipment. Further thanks is owed for the guidance provided by my
committee members: Drs. James Tiedje, Boyd Ellis, Craig Criddle, and Larry F orney.

I would like to acknowledge the United States Department of Agriculture for
providing ﬁnancial assistance in the form of a 3-year National Needs Water Science
Fellowship. Drs. Frank D’itri and Boyd Ellis were particularly instrumental in securing
this honorable award on my behalf. The National Science Foundation Center for
Microbial Ecology provided ﬁnancial support for equipment and supplies, as well as one

year of stipend assistance. Finally, the Pharmacia/Upjohn Company made this research

iv

possible by generously granting ﬁnancial assistance for equipment and supplies and by
providing necessary access to their facilities and data.

This work would never have been completed without the assistance of my
extended laboratory family: Helen Garchow, Ann O’Neill, Dr. Hal Collins, Dr. Mike
Kaufman, Nina Consolatti, Michel Cavigelli, and particularly Sandy Marsh. In addition
to being a good friend, Sandy worked very hard under extreme conditions at the most
inconvenient hours of the day to help me complete this research and is to be commended
for her efforts. Thanks also to Carolyn Hammarskjold for her efforts in helping me ﬁnd
relevant literature, and to Sally Shaw and Alice Gillespie for administrative support.

Lastly, I would like to thank all those friends and family members that contributed
to my overall physical, mental, and ﬁnancial health in a number of ways: my parents, Bill
and Carol Laatsch; my siblings, Doug, Jill, and Joe Laatsch; Jane’s parents, John and
Lillian Yagow; Jane’s extended family; my net buddies, Jeff Mayhaus, Jim Markwiese,
Ken Schwartz, and the Homebrew Digest community; and last but not least, my scaled

and furry ﬁiends, Ziggy, Winifred, Zoe, Bean, Brambles, and my best ﬁ'iend, Luther.

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... viii
LIST OF FIGURES ............................................................................................................ ix
CHAPTER 1 1
LITERATURE REVIEW ..................................................................................................... 1
CHAPTER 2 14
OLIGOTROPHIC COMPARISON STUDY ..................................................................... 14
2.1 Introduction ............................................................................................................. 14
2.2 Experimental Design and Approach ....................................................................... 15

2. 2.1 Sampling Locations .......................................................................................... 15

2. 2. 2 Water Sampling and Analyses .......................................................................... 15

2. 2. 3 Biological Sampling and Processing ............................................................... 16
2.2.4 Direct Microscopic Counts .............................................................................. 18

2.2. 5 Methanol Oxidizer Counts ............................................................................... 19

2. 2. 6 Fatty Acid Methyl Ester (FAME) Processing and Extraction ......................... 21

2. 2. 7 FAME Data Analysis ....................................................................................... 23

2.3 Results and Discussion ........................................................................................... 24
2.3.1 Water Chemistry .............................................................................................. 24
2.3.2 Microbial Count Data ...................................................................................... 25
2.3.3 Microbial Community Structure (FAME) ........................................................ 27
CHAPTER 3 31
SECONDARY METABOLISM STUDIES ...................................................................... 31
3.1 Introduction ............................................................................................................. 31
3.2 Field Experimental Approach ................................................................................. 32
3.3 Sessile Primary Productivity ................................................................................... 35
3.4 Methanol Oxidation/Assimilation ........................................................................... 37
3.5 Secondary Metabolism Shakeﬂask Experiments .................................................... 41
3.5.1 General Experimental Approach ..................................................................... 41

3. 5. 2 Secondary Metabolism Experiment #1: Aromatic/A liphatic Screen ............... 44
3.5.3 Secondary Metabolism Experiment #2: Time-Intensive Subsampling ............ 49

3. 5. 4 Secondary Metabolism Experiment #3: 5 Simultaneous Substrates ............... 53
3.5.5 Secondary Metabolism Experiment #4: Primary/Secondary Link .................. 60

3. 5. 6 Secondary Metabolism Experiment #5: Anaerobic Sediment Study ............... 69

3. 5. 7 Secondary Metabolism Experiment #6: Methanol Exclusion ......................... 75
3.5.8 Secondary Metabolism Results Summary ........................................................ 84

3.6 Microbial Community Structure (FAME Analysis) ............................................... 94
3.7 Discussion ............................................................................................................... 98

3. 7.1 Microbial Community Structure ....................................................................... 98

vi

3. 7.2 Primary Methanol Metabolism ...................................................................... 100

3. 7. 3 Secondary Metabolism....‘ .......................................................... . .................... 101

3. 7. 4 Anaerobic Metabolism ................................................................................... 103

3. 7. 5 Link Between Methanol and Secondary Metabolism ..................................... 105

3.8 Summary and Conclusions .................................................................................... 106
LIST OF REFERENCES ................................................................................................. 111

vii

LIST OF TABLES

Table 1.1 Summary of literature references to biodegradation of compounds of

interest in the Pharmacia/Upjohn Pond .......................................................... 13
Table 2.1 Inorganic water chemistry parameters for three natural lake systems and the

Pharmacia/Upjohn Pond. ............................................................................... 26
Table 2.2 Microbiological count data and diversity indices for three natural lake

systems and the Pharmacia/Upjohn Pond. ..................................................... 26

Table 3.1 Summary of direct microscopic count data for all secondary metabolism
experiments with attached microbial communities from the
Pharmacia/Upjohn Pond. ............................................................................... 84

Table 3.2 Summary of ANOVA statistics and interpretation for direct microscopic
counts for all secondary metabolism experiments with attached microbial
communities from the Pharmacia/Upjohn Pond. ........................................... 85

Table 3.3 Summary of methanol transformation rates for all secondary metabolism
experiments with attached microbial communities from the
Pharmacia/Upjohn Pond. ............................................................................... 86

Table 3.4 Summary of ANCOVA statistics and interpretation for methanol
transformation rates for all secondary metabolism experiments with
attached microbial communities from the Pharmacia/Upjohn Pond. ............ 87

Table 3.5 Summary of dichloromethane transformation rates for all secondary
metabolism experiments with attached microbial communities from the
Pharmacia/Upjohn Pond. ............................................................................... 88

Table 3.6 Summary of AN COVA statistics and interpretation for dichloromethane
transformation rates for all secondary metabolism experiments with
attached microbial communities from the Pharmacia/Upjohn Pond. ............ 89

Table 3.7 Summary of toluene transformation rates for all secondary metabolism
experiments with attached microbial communities from the
Pharmacia/Upjohn Pond. ............................................................................... 90

Table 3.8 Summary of ANCOVA statistics and interpretation for toluene
transformation rates for all secondary metabolism experiments with
attached microbial communities from the Pharmacia/Upjohn Pond. ............ 91

Table 3.9 Summary of chlorobenzene transformation rates for all secondary
metabolism experiments with attached microbial communities from the
Pharmacia/Upjohn Pond. ............................................................................... 92

Table 3.10 Summary of ANCOVA statistics and interpretation for chlorobenzene
transformation rates for all secondary metabolism experiments with
attached microbial communities from the Pharmacia/Upjohn Pond. ............ 93

viii

LIST OF FIGURES

Figure 1.1 Map of Pharmacia/Upjohn Pond (September 1992) ........................................... 7
Figure 2.1 Experimental ﬂow diagram for microbiological analyses in the
oligotrophic comparison study. ...................................................................... 17

Figure 2.2 Fatty acid methyl ester (FAME) biplots for epiphytic microbial

communities from three natural oligotrophic systems and the

Pharmacia/Upjohn Pond. ............................................................................... 30
Figure 3.1 Primary productivity and respiration rates (mg 02 In2 d") for attached

microbial communities from three light treatment regimes in the

Pharmacia/Upj ohn Pond. .. ............................................................................. 36
Figure 3.2 Methanol oxidation and assimilation rates (nmol [1x106]cells'l hr'l) for

attached microbial communities from three light treatment regimes in the

Pharmacia/Upjohn Pond. ............................................................................... 40
Figure 3.3 Methanol transformation plot for laboratory incubations of attached

microbial communities from two light treatment regimes in the

Pharmacia/Upjohn Pond, Experiment #1 (7/13/95). ...................................... 46
Figure 3.4 Dichloromethane (DCM) transformation plot for laboratory incubations of

attached microbial communities from two light treatment regimes in the

Pharmacia/Upjohn Pond, Experiment #1 (7/ 13/95). ...................................... 47
Figure 3.5 Toluene (TOL) transformation plot for laboratory incubations of attached

microbial communities from two light treatment regimes in the

Pharmacia/Upjohn Pond, Experiment #1 (7/ 13/95). ...................................... 48
Figure 3.6 Methanol transformation plot for laboratory incubations of attached

microbial communities from two light treatment regimes in the

Pharmacia/Upjohn Pond, Experiment #2 (7/26/95). ...................................... 50
Figure 3.7 Dichloromethane (DCM) transformation plot for laboratory incubations of

attached microbial communities from two light treatment regimes in the

Pharmacia/Upjohn Pond, Experiment #2 (7/26/95). ...................................... 51
Figure 3.8 Toluene (TOL) transformation plot for laboratory incubations of attached

microbial communities from two light treatment regimes in the

Pharmacia/Upjohn Pond, Experiment #2 (7/26/95). ...................................... 52
Figure 3.9 Methanol transformation plot for laboratory incubations of attached

microbial communities from two light treatment regimes in the

Pharmacia/Upjohn Pond, Experiment #3 (8/11/95). ...................................... 54
Figure 3.10 Dichloromethane (DCM) transformation plot for laboratory incubations

of attached microbial communities from two light treatment regimes in

the Pharmacia/Upjohn Pond, Experiment #3 (8/11/95). ................................ 55
Figure 3.11 l,2-dichloroethane (DCA) transformation plot for laboratory incubations

of attached microbial communities from two light treatment regimes in

the Pharmacia/Upjohn Pond, Experiment #3 (8/11/95). ................................ 56

ix

Figure 3.12 Bromodichloromethane (BDCM) transformation plot for laboratory

incubations of attached microbial communities from two light treatment

regimes in the Pharmacia/Upjohn Pond, Experiment #3 (8/11/95). .............. 57
Figure 3.13 Chlorobenzene (CB) transformation plot for laboratory incubations of

attached microbial communities from two light treatment regimes in the

Pharmacia/Upjohn Pond, Experiment #3 (8/11/95). ...................................... 58
Figure 3.14 Toluene (ITOL) transformation plot for laboratory incubations of attached

microbial communities from two light treatment regimes in the

Pharmacia/Upjohn Pond, Experiment #3 (8/11/95). ...................................... 59
Figure 3.15 Methanol transformation plot for laboratory incubations of attached

microbial communities from two light treatment regimes in the

Pharmacia/Upjohn Pond, Experiment #4 (9/13/95). ...................................... 62
Figure 3.16 Dichloromethane (DCM) transformation plot for laboratory incubations

of attached microbial communities from two light treatment regimes in

the Pharmacia/Upjohn Pond, Experiment #4 (9/ 13/95). ................................ 63
Figure 3.17 Bromodichloromethane (BDCM) transformation plot for laboratory

incubations of attached microbial communities from two light treatment

regimes in the Pharmacia/Upjohn Pond, Experiment #4 (9/13/95). .............. 64
Figure 3.18 1,2-dichloroethane (DCA) transformation plot for attached microbial

communities from two light treatment regimes in the Pharmacia/Upjohn

Pond, incubated in the labOratory in the presence or absence of methanol,

Experiment #4 (9/13/95). ............................................................................... 65
Figure 3.19 Chlorobenzene (CB) transformation plot for attached microbial

communities from two light treatment regimes in the Pharmacia/Upjohn

Pond, incubated in the laboratory in the presence or absence of methanol,

Experiment #4 (9/13/95). ............................................................................... 66
Figure 3.20 Toluene (TOL) transformation plot for attached microbial communities

from two light treatment regimes in the Pharmacia/Upjohn Pond,

incubated in the laboratory in the presence or absence of methanol,

Experiment #4 (9/ 13/95). ............................................................................... 67
Figure 3.21 Multiple compound transformation plots for laboratory incubations of

sediment slurry from the Pharmacia/Upjohn Pond, Experiment #4

(9/13/95) ......................................................................................................... 68
Figure 3.22 Dichloromethane (DCM) transformation plot for laboratory incubations

of sediment slurry from the Pharmacia/Upjohn Pond in the presence or

absence of a sulfate reduction inhibitor, Experiment #5 ................................ 71
Figure 3.23 1,2-dichloroethane (DCA) transformation plot for laboratory incubations

of sediment slurry from the Pharmacia/Upjohn Pond in the presence or

absence of a sulfate reduction inhibitor, Experiment #5. ............................... 72
Figure 3.24 Chlorobenzene (CB) transformation plot for laboratory incubations of

sediment slurry from the Pharmacia/Upjohn Pond in the presence or

absence of sulfate reduction inhibitors, Experiment #5. ................................ 73
Figure 3.25 Toluene (TOL) transformation plot for laboratory incubations of sediment

slurry from the Pharmacia/Upj ohn Pond in the presence or absence of

sulfate reduction inhibitors, Experiment #5. .................................................. 74

Figure 3.26 Diagram of in vitro system for pre-incubation of attached microbial

communities from the Pharmacia/Upjohn Pond, Experiment #6. ................. 79
Figure 3.27 Methanol transformation plot for laboratory incubations of attached

microbial communities developed under dark conditions in the

Pharmacia/Upjohn Pond and pre-incubated for 3 weeks in the presence or

absence of methanol, Experiment #6 (12/18/95). ......................................... 80
Figure 3.28 Dichloromethane (DCM) transformation plot for laboratory incubations

of attached microbial communities developed under dark conditions in

the Pharmacia/Upj ohn Pond and pre-incubated for 3 weeks in the

presence or absence of methanol, Experiment #6 (12/18/95). ...................... 81
Figure 3.29 Chlorobenzene (CB) transformation plot for laboratory incubations of

attached microbial communities developed under dark conditions in the

Pharmacia/Upjohn Pond and pre-incubated for 3 weeks in the presence or

absence of methanol, Experiment #6 (12/18/95). ......................................... 82
Figure 3.30 Toluene (TOL) transformation plot for laboratory incubations of attached

microbial communities developed under dark conditions in the

Pharmacia/Upj ohn Pond and pre-incubated for 3 weeks in the presence or

absence of methanol, Experiment #6 (12/18/95). ......................................... 83
Figure 3.31 Fatty acid methyl ester (FAME) biplot for attached microbial

communities from two light treatment regimes in the Pharmacia/Upjohn

Pond, Experiment #2. ..................................................................................... 95
Figure 3.32 Fatty acid methyl ester (FAME) biplot for attached microbial

communities from two light treatment regimes in the Pharmacia/Upjohn

Pond, Experiment #3 ...................................................................................... 96
Figure 3.33 Fatty acid methyl ester (FAME) biplot for attached microbial

communities from two light treatment regimes in the Pharmacia/Upjohn

Pond, Experiment #4 ...................................................................................... 97

xi

Chapter 1

LITERATURE REVIEW

Large quantities of a variety of xenobiotic compounds are synthesized annually
for use as solvents, refrigerants, cleaning agents, ﬁre retardants, chemical synthesis
intermediates, and pesticides (Janssen et al., 1990). These compounds have become
pervasive environmental contaminants through point-source and dispersed discharge in
industrial and municipal wastewater, agricultural runoff, and landﬁll leachate (Chaudry
and Chapalamadugu, 1991; United States Evironmental Protection Agency, 1983).
Contamination of surface and subsurface drinking water resources presents a particularly
complex environmental problem and poses a substantial public health concern.

Xenobiotic compounds are generally toxic and many are suspected carcinogens,
mutagens, or terratogens having potentially long-term and adverse health effects. These
compounds are often persistent in the environment due to the existence of few natural
analogs or naturally-evolved mechanisms for degradation. Halogenation renders these
contaminants even more recalcitrant, because removing multiple halogen molecules can
be energetically costly and the large physical nature of the halogen molecule can block
the active sites of degradation enzymes (Sayler et al., 1984; Neilson et al., 1985; Boyle,
1989; Hardman, 1991 ). Moreover, wastewater discharges often consist of a complex
mixture of compounds of various types and concentrations, rendering a single

remediation technology ineffective (Bitzi et al., 1990). Remediation of contaminated

water resources can also be extremely expensive and time-consuming (Amman and Koch,
1994). As such, considerable research effort has been dedicated to discovery and
understanding of technology and mechanisms for cost-effective removal of complex
mixtures of trace-level contaminants from water resources (Rittman et al., 1980;

Bouwer and McCarty, 1985; Henson et al., 1989; Hwang et al., 1989a; Hwang et al.,
l989b; Bitzi et al., 1990).

Bioremediation represents one particularly promising approach to treatment of
contaminated water resources. Bioremediation exploits the immensely diverse metabolic
capabilities of microorganisms to effect a biological transformation of environmental
contaminants into harmless end-products. During the course of evolutionary history,
microorganisms have developed a wide array of enzymatic pathways for degradation of
naturally-occurring compounds (Alexander, 1965), the structures of which are oﬁen
analagous to many modern xenobiotic compounds. Some of these naturally-evolved
enzymes are capable of breaking the chemical bonds within synthetic analog compounds,
resulting in biodegradation. Ideally, the microorgansisms completely metabolize the
contaminants into the harmless end-products of carbon dioxide and water, releasing
energy for growth and generating microbial biomass.

Contaminants that can be completely oxidized by the microorganism and are
present in sufﬁcient concentrations to support microbial growth are referred to as primary
growth substrates (Alexander, 1981; Rittman and McCarty, 1980). When concentrations
fall below the threshhold to support microbial growth, the compounds are known as

secondary growth substrates. Secondary substrates can be metabolized in the presence of

a primary growth substrate or when the combined concentrations of multiple secondary
substrates is sufﬁciently high to support microbial growth (McCarty et al., 1981; Bouwer
and McCarty 1982; Bouwer and McCarty, 1983). Cometabolism is the apparently
fortuitous, partial metabolism of a compound through non-speciﬁc enzyme activity
(Jensen, 1963). Cometabolism results in no energy gain for the organism and generates
potentially harmful intermediate compounds, which can accumulate to toxic levels or be
used for energy by other opportunistic microorganisms in a consortium. Throughout this
current study, true secondary metabolism is not distinguished from cometabolism and
both are collectively referred to as secondary transformation. Primary and secondary
metabolic transformations are the principal mechanisms underlying bioremediation as a
treatment technology.

Bioremediation is often based upon a consortia of microorganisms growing within
a polysaccharide matrix attached to some physical substrate, such as soil, sediment,
plants, glass, or plastic. The importance of attached growth has long been recognized
(Zobell, 1943) and attached microorganisms are numerically predominant in natural
environments, particularly oligotrophic aquatic systems (Geesey et al., 1978; Costerton
and Geesey, 1979; Ladd et al., 1979). These attached microbial communities are termed
biofilms and presumably provide the organisms living within a number of system-speciﬁc
advantages (Breznak et al., 1984; Hamilton, 1987). Attachment prevents the organism
from being washed out of lotic systems and thereby maintains proximity to resources for
growth (Jones et al., 1969; Fletcher and Floodgate, 1973; Paerl, 1975; Geesey et al.,

1978). The polysaccharide matrix may act as a nutrient exchange reservoir, concentrating

critical nutrients and carbon sources in the proximity of the cell (Costerton and Lashen,
1984). In the case of epiphytic (plant-associated) bioﬁlms, the organism may beneﬁt
from substrates exuded from the living attachment surface. Attached growth may prevent
or retard bacterial predation by protozoa (Costerton et al., 1985). Perhaps most
importantly, attached grth fosters beneﬁcial microbial interactions that contribute to
maintaining a favorable microenvironment around the cell (Haack and McFeters, 1982;
Wimpenny et al., 1983; Hamilton, 1985; Jeffrey and Paul, 1986; Murray et al., 1986;
Hamilton, 1987; Characklis et al. , 1990). Microbial bioﬁlms are commonly used among
the wide array of bioremediation technologies in practice.

Biodegradation is dependent upon the presence of microorganisms with
appropriate metabolic pathways and environmental conditions that favor the desired
microbial activity. Bioaugmentation attempts to correct for a deﬁciency in metabolic
pathways by directly inoculating the contaminated site with microorganisms possessing
the desired metabolic properties. However, transplanted and engineered microorganisms
frequently do not survive in the environment because they lack the necessary adaptations
to make them competitively ﬁt (Acea and Alexander, 1988; Goldstein et al., 1985; Pahm
and Alexander, 1993; Zaidi, 1988; Zaidi, 1989; Pritchard et al., 1992). In fact,
bioaugmentation has never been successfully demonstrated on a signiﬁcant scale in the
scientiﬁc literature (Pritchard et al., 1996).

In situ bioremediation relies upon the adapted metabolic activity of
microorganisms inherent to the contaminated site. During the course of several years of

exposure to the contaminants, natural selection fosters development of a microbial

community with the ability to exploit these compounds as growth and energy subtrates.
The high diversity associated with natural environments increases the likelihood of
functional redundancy, promotes interspecies interactions, and allows the microbial
community to tolerate a broad range of environmental conditions (Atlas and Bartha,
1993). The inherent organisms are also more well-adapted to the conditions of the site
and are more likely to survive in highly deﬁned niches than are introduced organisms.
Multiple food-chain levels in natural environments may help to preserve balance and
maintain active microbial communities. Community diversity is often associated with
improved community stability and increases the chances of maintaining desired
bioremediation activity under a more broad subset of environmental conditions (Atlas and
Bartha, 1993).

Intrinsic or passive bioremediation, also known as natural attenuation, involves
allowing the contaminated site to self-remediate while carefully monitoring the water
quality and environmental conditions at the site (Hinchee et al. , 1995). Passive
bioremediation is becoming increasingly accepted by regulatory agencies as a low-cost
remediation alternative, particularly groundwater contamination. However, the
applicability of passive bioremediation is site-speciﬁc, depending upon the nature of the
contamination, the proximity to drinking water sources, and the existing environmental
conditions. Site-speciﬁc applied research is necessary to evaluate the potential
applicability of passive bioremediation and to completely understand the mechanisms of

remediation therein (Pritchard et al., 1996).

Even if the necessary metabolic pathways may be present, environmental
conditions within a contaminated site oﬁen need to be manipulated to elicit the desired
remediation in a timely fashion. Nutrients may be limiting, electron acceptors may be
deﬁcient, a primary growth substrate may be lacking, pH may be too acidic or alkaline, or
surface area for attachment may be insufficient. Manipulation of the existing microbiota
and/or environmental conditions to optimize bioremediation activity will be referred to in
this paper as biomanagement. Biomanagement generally involves extensive site-speciﬁc
research to identify the important system parameters and to determine the optimal
conﬁguration of those parameters. Only through a holistic and multidisciplinary research
approach can each system be manipulated in a manner that optimizes remediation activity
and thereby minimizes public health risk. This thesis will detail the research efforts to
determine the mechanistic basis of in situ bioremediation and the feasibility of

biomanagement in the Pharmacia/Upjohn Pond system in Portage, Michigan.

The Pond is a 34-ha industrial groundwater recharge basin at the
Pharmacia/Upjohn main pharmaceutical manufacturing facility (Figure 1.1). The Pond
receives approximately 30 million L day'l non-contact cooling water pumped directly
from groundwater resources at the site. This groundwater has been historically
contaminated with a host of secondary compounds, which sporadically appear in the Pond
influent at pg L" concentrations. Water entering the Pond is also chronically
contaminated with mg L'1 concentrations of methanol, due to the use of methanol brines
within the cooling system. Approximately 25-40% of water leaves the system as surface
discharge, while the remainder recharges local shallow aquifers. All organic

contaminants fall below detection limits in water discharged from the system, suggesting

Cooling Water Oxfall
Binary Retertim Basin D

W1

Figure 1.1 Map of Pharmacia/Upjohn Pond (September 1992).

that the pond is functioning as an effective large-scale natural bioreactor. One thrust of
this research was to determine the biological mechanisms responsible for contaminant
removal and the factors inﬂuencing those mechanisms.

Methanol comprises a signiﬁcant fraction of the dissolved organic carbon entering
the highly oligotrophic Pharmacia/Upjohn Pond system (Klug et al., 1996) and is likely
serving as a growth substrate for bioﬁlm bacteria in the Pond. Because one-carbon (C1)
compounds such as methane and methanol occur abundantly throughout nature (Hanson,
1992), organisms that utlize C1 compounds as sole carbon and energy sources
(methylotrophs) are also ubiquitously distributed and have been isolated from nearly all
types of environmental samples (Corpe, 1985; Hanson, 1980; Heyer, 1977; Hutton and
Zobell, 1949; Whittenbury et al., 1970; Sieburth et al., 1987; Strand and Lidstrom, 1984).
Expectedly, previous research (Klug et al., 1996) and the present Pond studies have
revealed substantial methanol oxidation activity in epiphytic bioﬁlrns and sediments.
Epiphytic methanol-oxidizing bacteria have also been isolated from Pond samples.

Methylotrophs are a metabolically diverse group of organisms, including both the
methane and methanol utilizers. However, the term methylotroph is more commonly
used to distinguish the subset of methanol utilizers unable to use methane from the
methanotrophs and will be used as such throughout this paper. During primary methanol
metabolism, all methylotrophs express the enzyme methanol dehydrogenase to oxidize
methanol to formaldehyde, a key intermediate which is then either respired to carbon
dioxide to produce energy or assimilated into cellular biomass (Bratina and Hanson,

1992). However, the methylotrophs are a physiologically diverse group or organisms and

include a number of cosmopolitan heterotrophs with a wide array of ancillary enzymatic
pathways for metabolism of other carbon compounds (Green, 1992). A vast number of
methylotrophs, other bacteria, and various mixed cultures are reported in the literature as
being capable of growth on many of the secondary contaminants entering the
Pharmacia/Upjohn Pond, a brief summary of which is presented in Table 1.1.
Methylotrophs have diverse bioremediation potential, but would be particularly
well-suited to metabolize the spectrum of contaminants entering the Pharmacia/Upjohn
Pond. The nature of the contaminant load, the conﬁrmed presence of methanol oxidizers,
and the metabolic diversity of methylotrophs as a group led to development of my ﬁrst
series of hypotheses:
Hypothesis la: Methanol is serving as a primary growth substrate in the bioﬁlms,
fueling secondary metabolism of the trace-level organic compounds entering the system.
Hypothesis 1b: Methanol is serving as a primary growth substrate in the bioﬁlm,
naturally selecting methylotrophs and driving changes in microbial community structure.
The second aspect of this research focuses upon determining the feasibility of
managing the bioﬁlm communities to optimize bioremediation activity. The predominant
macrophyte in the Pharmacia/Upj ohn Pond is a submerged, rhizoid-rooted algal
macrostructure of a species of Chara vulgaris (Klug et al., 1996). C. vulgaris is common
to hardwater aquatic systems and exhibits a highly dissected growth habit that offers a
large surface area for epiphyte colonization (Sculthorpe, 1985). During the warm

summer season, C. vulgaris provides more than adequate surface area for bioﬁlm

colonization, but senesces in the fall and completely disappears during the cold winter
months before reappearing in the spring.

Because C. vulgaris is seasonally variable, surface area for attachment represents
one important control point in the system. Attachment surfaces in aquatic bioremediation
systems can be artiﬁcially augmented through the installation of commercially available
polyvinylchloride (PVC) biological media. However, artiﬁcial substrata raise questions
about the inﬂuence of the attachment surface itself on microbial community structure and
remediative function, especially given that the natural physical substrate is alive. The
interior surfaces of the artiﬁcial media are also completely devoid of natural light,
inspiring questions about the inﬂuence of sessile primary productivity on bioremediation
activity.

The inherent variability in natural substrata often presents difﬁculties with
uniform sampling and quantiﬁcation for studies of attached microbial growth. Artiﬁcial
substrata simplify sampling and processing, reduce heterogeneity, and allow standardized
comparisons across habitats or sites. From the earliest work on attached growth (Zobell,
1943) to the present, many bioﬁlm studies have employed artiﬁcial attachment surfaces.
Artiﬁcial materials have included glass, ceramic, brick, aluminum, activated carbon,
nylon, plexiglass, polyethylene, polystyrene, and others (Tuchman and Blinn, 1979;
Roemer et al., 1984; La Motta and Hickey, 1980; Aloi, 1990; Shreve et al., 1990;
Cattaneo and Amireault, 1992; Liu et al., 1993; Haack, 1995). Accordingly, little
consensus has been reached on the most appropriate materials, but plastics in many forms

now seem to be the preferred substrata for bioﬁlm research (Cattaneo and Amireault,

10

1992). Hydrophobic plastics have been shown to strongly support bioﬁlm accumulation
in pure culture experiments (Shreve et al., 1990). Because most commercially available
biological media is constructed of polyvinylchloride and due to supporting evidence in
the literature, polyvinylchloride was deemed a suitable prototype substrata for the
purposes of this research.

Periphyton communities are the predominant source of primary productivity in
many aquatic systems (Kevem et al., 1966; McMahon et al., 1974; Roemer et al., 1984;
Geesey et al. , 1978). Sessile primary productivity is known to stimulate heterotrophic
activity within bioﬁlm communities. Geesey et al. (1978) observed that concentrations of
sessile bacteria on the upper surfaces of submerged rocks in a small mountain stream
coincided with ﬂuctuations in epilithic algae, and that some bacteria grew directly
atttached to sessile algal cells. Haack and McFeters (1982) demonstrated a direct ﬂux of
algal exudates to sessile bacteria in oligotrophic stream bioﬁlms, with little to no
heterotrophic utilization of other sources of dissolved organic carbon. Murray et al.
(1986) found that a simple bioﬁlm consisting of a diatom and a bacterium expressed
greater metabolic activity than did either organism individually.

The precise nature of the stimulated heterotrophic activity and the organisms
affected are not well understood. Generic bacterial activity in the experiments above is
often estimated by thymidine incorporation, which fails to identify speciﬁc ecologically
signiﬁcant changes in function. Haack et al. (1994) observed light-induced shiﬁs in
bacterial community structure and carbon metabolism proﬁles in simulated stream

bioﬁlms, but distinguishing between changes in species, metabolic function, or both is

11

difﬁcult. Very few researchers have examined the effect of sessile primary productivity
on bioremediation. Bender et al. (1995) have examined the biodegradation of chlorinated
organics by microbial mats, consortia of bacteria and photosynthetic cyanobacteria.
Photosynthetic and heterotrophic activity within the mat leads to adjacent oxic and anoxic
microenvironments, which are thought to sequentially facilitate reductive dechlorination
and aerobic cleavage of aromatic rings. These results again suggest a mutualistic or
synergistic relationship between photosynthetic and heterotrophic attached organisms.
The second series of hypotheses is based on this attached mutualism:

Hypothesis 2a: Sessile primary productivity will generate oxygen and carbon exudates,
supporting a metabolically active heterotrophic community and stimulating the

metabolism of organic contaminants.

Hypothesis 2b: Sessile primary productivity will generate oxygen and carbon exudates,
increasing bacterial numbers and altering heterotrophic community structure.

The remainder of this thesis will detail the research work done to test the
aforementioned hypotheses. Chapter 2 explores the background research necessary to
further deﬁne and characterize the naturally-occurring epiphytic bioﬁlms in the
Pharmacianpjohn Pond. This initial study provided a framework for examining primary
and secondary bioremediation activity in the context of biomanagement of naturally-

selected bioﬁlms, as detailed in Chapter '3.

12

Table 1.1 Summary of literature references to biodegradation of compounds of interest

 

in the Pharmacia/Upjohn Pond.
Contaminant Organisms References
methanol Multiple genera: Multiple references:
Merhy/obacillus, Methylophi/us, Komogata, I990.
Merhy/obacterium, A ncylobacter, Bratina and Hanson. 1992.
Hyphomicrobimrr, Xanthobacter,
Acidomonas

Paracoccus denitriﬁcans
T hiobacillus novel/us
Paracoccus alcaliphilus

 

 

Methyhlophaga
Nocurdia_sp. 239 deBoer et al., 1990.
dichloromethane Pseudomonas sp. DMl Brunner et al., 1980.
Xanthobacrer autotrophicus
Methylococcus capsulatas Colby et al., 1977.
Methylosinus trichosporium OB3b Oldenhuis et al., 1989.
Nitrosomonas europea Vanelli et al., 1990.
mixed soil consortia Davis and Madsen, 1991.
Henson et al., 1988.
chloroform Nitrosomonas europea Vanelli et al., 1990.

Methylosinus trichosporium OB3b Alvarez-Cohen et al., 1992.
Oldenhuis et al., 1989.
Speitel and Leonard, 1992.

 

mixed soil consortia Strand and Shippert, 1986.

mixed sludge/activated sludge Fathepure and Vogel, 1991.
l,2- mixed aerobic consortia Lanzarone and McCarty, 1990.
dichloroethane Ancylobacter aquaticus Van Den Wijngaard, 1992.

Methylosinus trichosporium OB3b Oldenhuis et al., 1989.
Riebeth et al., 1992.

 

 

 

 

 

 

Arthrobacter Scholtz et al., 1987.
Pseudomonasﬂuorescens Vandenbergh and Kunka, 1988.
mixed methane-utilizing consortia Yokata er al., 1986.
Xanrhobacter autotrophicus Janssen et al.. 1985.
Xanthobacter autotrophicus GI 10 Tardif et al., 1991.
methanogenic consortia Bouwer and McCargr, 1983.

acetone Mycobacterium spp. Lukins and Foster, 1963.
unknown soil isolates, possibly Taylor et al., 1980.

Corynebacterium spp.

toluene Pseudomonas sp. JS6 Pettigrew et al., 1991.
Xanthobacter autotrophicus GJ 10 Janssen et al., 1985.

chlorobenzene Pseudomonas sp. J S6 Pettigrew et al., 1991.
Alca/igenes spp. Oltrnanns et al., 1988.
Pseudomonas spp. Reineke and Knackmuss, 1984.

Reineke, 1988.

 

 

 

 

 

Chapter 2

OLIGOTROPHIC COMPARISON STUDY

2.1 Introduction

Previous research on the Pharmacia/Upjohn Pond established baseline
information about the system (Klug et al., 1996). Among the ﬁndings were that methanol
and secondary contaminant concentrations decreased along a gradient from Pond inlet to
outlet. A corresponding change in methanol metabolism rate was similarly noted, with
bioﬁlm communities near the pond inlet showing a greater capacity for methanol
degradation than those near the pond outlet. This evidence suggested that the methanol
was driving some change in the microbial community function and/or structure, as put
forth in Hypotheses 1a and 1b above. A preliminary study was designed to further
characterize the Pond bioﬁlm communities and to test Hypothesis 1b, by comparing
epiphytic microbial communities in the Pond against those found in a regional class of
similar natural, hard-water, oligotrophic systems. The rationale for conducting such a
study is expressed in the following research objectives:

1. Determine if the epiphytic communities near the pond inlet are uniquely

selected and structured to degrade methanol.

2. Determine if the region of the pond near the outlet is converging
chemically and biologically toward regional oligotrophic systems.

3. Collect and characterize methanol-degrading isolates that may serve

important functional roles in the epiphytic communities near the pond
inlet.

14

2.2 Experimental Design and Approach

2.2.1 Sampling Locations

In addition to three sampling locations within the pond, three regional lakes were
selected for sampling, the selection criteria being that they closely resemble the pond in
trophic status and in the predominance of Chara among submerged macrophytes. The
chosen sites were Lawrence Lake (designated L; 42° 28’ latitude, 85° 19’ longitude),
Hamilton Lake (designated H; 42° 24’ latitude, 85° 21’ longitude), and Palmetier Lake
(designated P; 42° 35’ latitude, 85° 25’ longitude). Sites within the Pharmacia/Upjohn
Pond (42° 12’ latitude, 85° 34’ longitude) were chosen to reﬂect the concentration
gradient from system inlet in the north to the outlet in the south. These were designated
south (S), north (N), and retention basin (RB). Duplicate water samples and triplicate
samples of Chara with intact bioﬁlms were collected at each site. Due to the difﬁculty of
randomly sampling lake macrophytes, three equidistant sampling locations were selected
along the periphery of each site and representative macrophyte samples harvested at

random within the chosen sampling areas.

2.2.2 Water Sampling and Analyses

All chemicals used in the present study, unless otherwise noted, were Obtained
from Fisher Scientiﬁc (Pittsburgh, PA). Water samples were collected in proximity to the
biological samples with a Van Dom water sampler deployed to a depth of 0.5 m. Opaque
brown l-L acid-washed Nalgene bottles (N alge Co., Rochester, NY) were triple-rinsed

with sample, ﬁlled to overﬂowing, capped, and stored on ice for transport back to the lab.

15

Water samples were immediately analyzed for pH, conductivity, alkalinity, and total
phosphorus as surrogate indicators of trophic status. Water pH was measured on an
Orion model number SA520 pH meter (Orion Research, Inc., Boston, MA) equipped with
a combined single probe electrode and 2-point calibrated with buffers at pH 7 and pH 10.
Speciﬁc conductivity was measured with a YSI model 32 conductance cell (Yellow
Springs Instrument Co., Yellow Springs, OH). Alkalinity was determined by titration of
a 50—mL sample to a purple endpoint with 0.02 N sulfuric acid in the presence of 4 drops
of methyl purple indicator.

Total phosphorus was determined by digesting a 50-mL subsample with 0.5 g
potassium persulfate in an autoclave for 30 minutes (APHA, 1992a). The digest was
allowed to cool and treated with 5 mL of combined reagent (100 mL ammonium
molybdate, 250 ml 2.42 M sulfuric acid, 100 mL ascorbic acid, 50 mL potassium
anitmonyl-tartrate). Color development proceded for a period of one hour, after which
absorbance at 885 nm was measured for each sample using a Perkin/Elmer Lambda 6
UV/V IS spectrophotometer (Perkin/Elmer Corp., San Jose, CA). Concentration was

determined by linear interpolation of a standard curve.

2.2.3 Biological Sampling and Processing

A ﬂow diagram of biological sample processing and analysis is presented in
Figure 2.1. Biological samples consisted of whole Chara shoots with attached bioﬁlms.

Biological analyses of the bioﬁlm communities included direct microscopic counts, total

16

 

Bioﬁlm 4, Serial > DMC

 

 

 

 

Homogenate Dilution & FAME
Total Plate Nitrocellulose
Counts C) Master Fi'ter
/ Replication
Q Starvation & C)
Treatment Filter I“C”bat'°" Control Filter
Deﬁned Media Deﬁned Media
+ 835 + Carbon + S35
. ll
Enumerate Autoradlography Locate target colonies
target colonies < :8; isolate from Master

 

Figure 2.1 Experimental ﬂow diagram for microbiological analyses in the oligotrophic
comparison study.

17

plate counts, methanol degrader counts, colony diversity assessment, whole community
fatty acid methyl ester (FAME) proﬁles, and isolation and preliminary characterization of
methanol degrading microbial populations.

Samples were manually harvested with a 4-prong rake. Rootlets and sediments
were removed with ethanol-sanitized scissors. Trimmed samples were placed in sterile
Whirlpak bags, stored on ice for transport, and processed in triplicate for analytical
replication immediately upon return to the lab. A IO-g wet weight subsample was added
to 90 mL of sterile 0.01 M NaHPO4 buffer (pH 7.2). The mixture was homogenized in a
stainless steel Eberbach tissue grinder (Eberbach Corp., Ann Arbor, M1) on a Waring
industrial blender base (Waring Products Division, New Hartford, CT). Sample was
homogenized on the low setting for 30 seconds, cooled in an ice bath for 30 seconds to
prevent cell damage due to overheating, and homogenized for an additional 30-second
interval. Bioﬁlm homogenate was transferred to a sterile screw-cap vial. The tissue
grinder was rinsed with an additional 100 mL of sterile buffer. Grinder rinsate was
consolidated with the bioﬁlm homogenate, which was collectively subsampled for a

series of analyses and serially diluted in sterile buffer.

2.2.4 Direct Microscopic Counts

Triplicate l—mL subsamples of bioﬁlm homogenate were transferred to screwcap
test tubes, each containing 9 mL of sterile phosphate buffer treated with 1% formaldehyde
to preserve samples for direct microscopic observation. Bacteria were enumerated by
acridine orange (AO) direct microscopic counts (ASTM, 1987). Preserved samples were

further diluted on a sample-dependent basis with ﬁlter-sterilized phosphate buffer to

18

achieve a countable number of cells per ﬁeld. Residual bioﬁlm matrix resulted in
clumping on the slide, necessitating sonication of each sample with a Branson Sonic
Power model W14OD Soniﬁer Cell Disruptor (Branson Instruments, Inc., Danbury, CT)
equipped with a microprobe and operating at 30% power. A 0.9-mL aliquot of diluted
sampled was added to 0.1 mL of 0.1% acridine orange solution in a 13x100 mm test tube.
The subsample was stained for a period of 1 minute and then vacuum-ﬁltered onto a
Nucleopore 0.2 pm polycarbonate ﬁlter (Costar Corp., Cambridge, MA). The ﬁlter was
mounted on a microscope slide with cover slip for microscopic observation on a Zeiss
epiﬂuorescent microscope (Carl Zeiss, Oberkochen, West Germany). Twenty-two ﬁelds
were counted per slide, the highest and lowest ﬁeld for each slide was discarded, and the
mean calculated for the remaining twenty ﬁelds. Numbers of bacteria in the original
sample were back-calculated according to the ﬁeld size, mean number of cells per ﬁeld,

and the cumulative dilution factor.

2.2.5 Methanol Oxidizer Counts

Methanol degraders were enumerated and isolated using direct-plating
autoradiography (Dunbar et al., 1996). The technique employs radiolabelled sulfur-35 as
an indirect indicator of growth on a target compound, such as methanol. Serially diluted
cell suspensions of C. vulgaris homogenate were inoculated onto Millipore HATF
nitrocellulose ﬁlters (Millipore, Bedford, MA) placed directly atop general-purpose BBL
R2A media (18.1 g L"; Becton Dickinson Microbiology Systems, Cockeysville, MD).
After ﬁve days incubation, the ﬁlters were replicated, producing two identical mirror-

image copies of the original master ﬁlter. The master ﬁlters were photographed and

I9

stored at 4° C for later use. The replicate ﬁlters were incubated on fresh R2A media
overnight and then transferred to deﬁned agarose basal (DAB) starvation media (per L of
media: 2.68 g NazHPO4'7H20, 1.74 g KH2P04, 1.06 g NH4CI, 140 mg NaZSO4, 102 mg
MgClz'6HzO, 87 mg CaClz, 5 mg NazEDTA'2HzO, 2 mg FeSO4'7HzO, 1.86 mg
MnClz'ZHzO, 1.53 mg NazMoO4'2HzO, 0.3 mg H3BO3, 0.32 mg CaC12'6HzO, 0.16 mg
ZnSO4'7HzO, 0.02 mg NiC12°6H20, 0.01 mg CuC12'2H20, 1.5% high-strength
electrophoresis-grade agarose [Bio-Rad Laboratories, Hercules, CA]) to deplete colonies
of intracellular carbon reserves.

Following the starvation period, the “treatment” ﬁlter was placed upon DAB
media containing methanol at 1000 mg L'1 as the sole source of carbon and energy, as
well as 350 uCl radiolabelled sulfur-35 L". Stock radiolabelled sulfur-35 was obtained
from New England Nuclear (Boston, MA) in the form of sodium sulfate with a speciﬁc
activity of 1000 mCi/mmol. The “control” ﬁlter was placed atop a similar DAB media,
except without added methanol. Both ﬁlters were incubated 7 days and then processed
for autoradiography. Excess radioisotope was washed from each ﬁlter by a l-hour
incubation on wash media (per L of media: 28.408 g NazSO4, 10 g Bacto agar [Difco
Laboratories, Detroit, MI]). Filters were removed from the agar to air-dry for 10 minutes
prior to mounting with transparent tape on a sheet of 8x10 plain white paper. The paper
was covered with a single layer of plastic wrap and placed in an autoradiography
exposure cassette. In a darkroom under red light, each cassette was ﬁtted with an
unexposed sheet of Kodak X5 X-ray ﬁlm (Eastman Kodak Co., Rochester, NY).

Cassettes were placed in a press to ensure good contact with ﬁlm and the press was

20

wrapped in a black plastic bag. The exposure period extended 12 hours at -70° C, after
which the ﬁlms were processed manually with Kodak chemicals according to the
recommended Kodak protocol for x-ray development.

Those colonies on the treatment ﬁlter that utilized the target compound
presumably also incorporated S-35, resulting in dark spots on X-ray ﬁlm which had been
exposed to the colonies. Degraders of the target compound were enumerated by pseudo-
plate counts of x-ray ﬁlms and isolated from the non-radiolabelled master ﬁlter. The
technique also facilitated total plate counts and assessment of colony morphological

diversity (Shannon-Weaver index).

2.2.6 Fatty Acid Methyl Ester (FAME) Processing and Extraction

Undiluted bioﬁlm homogenate was subsampled in triplicate 10-mL aliquots for
microbial community structure analysis by FAME proﬁles. Each subsample was placed
in a 15 x 100 mm ashed glass test tube with teﬂon-lined screwcap. Subsamples were
centrifuged at approximately 3400 rpm (setting 6) for 10 minutes on an IEC model CL
clinical centrifuge (International Equipment Corp., Needham, MA). The overlaying
liquid layer was decanted and the cell pellets were stored at -20° C until extraction and
analysis. Bacterial isolates were cultured on non-selective R2A media, harvested directly
into the test tubes described above using ﬂamed Pasteur pipette tips, and stored at -20° C
until extraction and analysis.

Extraction was performed according to commercial guidelines for preparing pure
culture extracts as published in the Microbial Identiﬁcation System operating manual

(Microbial ID, Inc., 1991). Cell pellets were saponiﬁed by adding 1.0 mL of a strong

21

methanolic base solution (45 g NaOH pellets, 150 mL methanol, 150 mL deionized
water) to each test tube, vortexing the tube 10 sec, holding the tube in a 100° C water bath
for 5 min, vortexing an additional 10 sec, and holding an additional 25 min at 100° C.
Fatty acids were methylated by adding 2.0 mL of methylation reagent (325 ml 6.0 _N_ HCl,
275 mL methanol) to each tube, vortexing the tube contents for 10 seconds, holding the
samples at 80° C for 10 min, and rapidly cooling the tubes in a water bath. FAMEs were
extracted from the acidic aqueous phase by adding 1.25 mL of extraction solvent (1:1 v/v
hexane/methyl-t-butyl-ether) to each tube, mixing the tubes end-over-end for 10 min, and
removing the upper organic phase to a fresh tube. The extracts were washed by adding
3.0 mL of a weakly basic solution (10.8 g NaOH, 900 mL deionized water) to each tube,
rotating the tubes end-over-end for 5 additional min, and removing the upper organic
phase to a fresh tube. The washed extracts were dried under a stream of nitrogen, sealed,
and stored at -20° C prior to analysis, at which time the dried FAMES were reconstituted
in 200 uL of extraction reagent and transferred to autosampler vials ﬁtted with ﬂat-
bottomed glass inserts and sealed with teﬂon-lined crimp caps (vials, insert, and crimp
caps obtained from Sun Brokers, Inc., Wilmington, NC).

FAME analysis by gas chromatography immediately followed reconstitution of
dried extracts. FAME extracts were analyzed on a Hewlett Packard (HP) 5890 Series 11
gas chromatograph equipped with a 7673A autosampling device and ﬂame ionization
detector (Hewlett Packard, Wilmington, DE). The GC was ﬁtted with an HP Ultra 2
fused silica capillary column (25 m length, 0.2 mm ID, 0.33 pm layer thickness) with a

cross-linked phenyl-methyl-siloxane active matrix. The GC was operated under the

22

following conditions: detector temperature, 300° C; injector temperature, 250° C; initial
column temperature, 170° C; temperature ramp, 5° C min"; ﬁnal temperature, 270° C; H2
combustion gas 30 mL min"; dry air combustion gas 400 mL min"; N2 auxiliary gas, 30
mL min"; split ﬂow, 50 mL min"; purge ﬂow, 5 mL min"; split ratio, 50:1. Raw
analytical data was processed by the Microbial Identiﬁcation System proprietary software

to generate FAME proﬁles for each sample.

2.2. 7 FAME Data Analysis

FAME datasets are typically analyzed by a complex multivariate technique known
as principal components analysis (PCA; SAS Institute, 1991; Cavigelli et al., 1995). The
FAME variables are orthogonally transformed into a set of uncorrelated variables known
as principal components (PCs), which are each linear combinations of all the variables;
i.e. every variable is represented in each PC. Eigenvector loadings are the extent to which
the variables affect each PC. Each principal component in order explains a decreasing
percentage of the total variance in the dataset, such that the ﬁrst few PCs typically
account for greater than 80°/o of the variance in the original dataset.

Because of the constrained nature of the dataset, the covariance matrix was used
in the PCA (Aitchison, 1986). Data transformation was required to force the dataset to
meet the criteria of SAS-PCA with the covariance matrix. Resultant FAME peak areas
were hand-entered into spreadsheet format, eliminating all unknowns and non-fatty acid
peaks. All samples were normalized to the lowest total peak area for biomass correction.
All those peaks were eliminated that totaled less than 10% of the most abundant peak

across all samples. A negligibly small value (less than detection limit) was added to each

23

peak to ensure a complete non-zero dataset, new total peak areas were calculated, and
samples were re-normalized to the new lowest total peak area. Peak area was then
converted to relative abundance and logratio transformed to generate a normally
distributed dataset meeting the PCA criteria (Aitchison, 1986).

The transformed dataset was subjected to SAS-PCA and the resultant ﬁrst two
PCs of each sample were plotted on opposing axes. Scaled eigenvectors for the ten most
heavily weighted fatty acids were then superimposed on the original PC plot. These
“biplots” allow a researcher to examine separation between samples while simultaneously
determining which acids were most prevalent in determining resolution between samples.

All FAME data analysis was standardized for the experiments presented in this work.

2.3 Results and Discussion

2. 3.1 Water Chemistry

Table 2.1 presents the combined results of all water chemistry analyses. Water
pH ranged between 7.5 and 8.5 for all sites, which is typical for the hard-water, calcerous,
alkaline class of lakes. Alkalinity was 3-5 meq L", quite high for even this class of lakes,
but not substantially different between lakes. Conductivity ranged between 350 and 450
umhos cm'1 for the natural systems, and between 600 and 650 umhos cm'1 in the pond
sites, the latter of which is possibly a reﬂection of iron concentrations entering the pond.
Total phosphorus ranged from 8-15 pg L" in the natural sites and the south-end pond site
to 22-26 pg L" in the ponds north end and retention basin. Natural surface waters
typically fall in the range of 10-50 pg L" for total phosphorus, with concentrations

between 10-30 pg L" indicative of meso-eutrophic status. Less than 5% of total

24

phosphorus is normally available as orthophosphate. Thus, although water near the
system inlet seems to have elevated total phosphorus levels, the concentrations are not
beyond the expected range for this class of lakes. Overall, the water chemistry data

veriﬁes that all the sites are generally comparable with respect to trophic status.

2.3.2 Microbial Count Data

Means of microbial count data are presented in Table 2.2, with different letters
designating signiﬁcant differences (SAS t-test, alpha=0.05). Total direct microscopic
counts were in the range of 1x10lo cells g" dry weight Chara at all sites, with slightly
elevated levels near the system inlet (p=0.0077). Culturable plate counts reﬂected the
same general pattern, but about 2 orders of magnitude less in abundance (p=0.0048).
Density of methanol degraders was very signiﬁcantly higher near the pond inlet, with
both the north-end and retention basin samples having nearly 3 orders of magnitude more
methanol degraders than the south-end or natural samples (p=0.0001). A SAS overall
multivariate linear model of count data was highly signiﬁcant (p=0.0001) for both
Hotelling’s T and Wilks’ Lambda tests. Methanol degraders comprised approximately
15% of total culturable organisms in the north-end and retention basin communities. This
data clearly supports the hypothesis that methanol is exerting selective pressure and
driving changes in microbial community structure near the system inlet along the most
concentrated portion of the contaminant gradient.

The calculated Shannon-Weaver morphological diversity index revealed
depressed diversity near the system inlet. Lowered diversity is common at sites of high

selective pressure (Atlas, 1991). Diversity in the south end of the pond is comparable to

25

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 2.1 Inorganic water chemistry parameters for three natural lake systems and the
Pharmacia/Upjohn Pond.
Sampling Conductivityzs Alkalinity Total Phosphorus
Location pH (umhos cm") (meq L") (pg L")
Palmetier 8.15 351 3.40 14.3
Hamilton 8.43 388 3.80 11.5
Lawrence 8.19 460 4.18 9.0
U1 South 8.00 602 3.92 9.7
UJ North 7.64 648 4.74 22.7
UJ Basin 7.57 653 4.62 25.8
Table 2.2 Microbiological count data and diversity indices for three natural
lake systems and the Pharmacia/Upjohn Pond.
Sampling Direct Counts Plate Counts Methanol Oxidizers Shannon-Weaver
Location (cells g") (cfu g") (ch1 g") Diversity Index
Palmetier B 9.17 x 109 B 7.20 x 107 B 1.08 x 105 1.45
Hamilton B 8.38 x 109 B 8.06 x 107 B 9.67 x 104 1.64
Lawrence B 9.37 x 10° B 2.87 x 107 B 1.26 x 105 1.35
U] South B 1.47x10lo B 7.80x107 B 2.18x105 1.34
UJ North A 3.78 x 1010 A 2.59 x 108 A 4.07 x 107 0.79
UJ Basin B 2.07x10lo A 2.19x108 A 3.51 x107 0.83

 

 

 

 

 

26

 

that calculated for the natural sites, further supporting the hypothesis of south-end

convergence toward typical systems.

2.3.3 Microbial Community Structure (FAME)

Fatty acids extracted from cell wall lipid components can be used as phenotypic
"ﬁngerprints" for isolates and for entire microbial communities. Figure 2.2 shows the
FAME biplot for the means of all ﬁeld replicates in the Oligotrophic Comparison Study.
The resulting pattern is quite difﬁcult to interpret, although a few conclusions can be
drawn. First, the absolute magnitudes of the PC axes are small, indicating small
perceptible differences between samples. The Lawrence Lake communities are greatly
separated from all other samples along the PCI axis, indicating the least similarity to
other samples. The fatty acids corresponding to the separation of the Lawrence Lake
samples are the fully saturated acids, 14:0, 16:0, 18:0. These fatty acids are ubiquitously
distributed in nature and provide little insight into the practical reason for the Lawrence
Lake samples being so highly resolved from other samples.

The remaining samples demonstrate a slight gradient separation along axis PC2
from the region of highest contamination in the pond (retention basin) to the region of
lowest contamination (south), indicative of a convergence of communities in the south
end of the pond toward those associated with natural systems. However, these patterns
are not strong and little conclusive information can be gleaned from the corresponding
fatty acid weightings.

As an analytical tool, whole community FAME proﬁles present distinct

limitations. Because of the redundant and ubiquitous distribution of many fatty acids

27

among even widely varying individual microbial species, whole community FAME
proﬁles appear to be useful primarily as a preliminary screening tool to detect large
differences between complex microbial communities. Even when large-scale differences
are detected, interpretation often requires further investigation using more targeted
metabolic assays. The broad-level resolution inherent to whole community FAME
proﬁles severely limits detection and interpretation of small differences between similar
microbial communities.

Even those organisms with unique signature fatty acids are often non-detectable in
whole community FAME proﬁles unless they represent a large proportion of the biomass.
Many other pure culture strains, such as Methylotrophs, exhibit ﬁngerprint proﬁles only
because of uniquely proportioned ubiquitous fatty acids. The fatty acid ratios become
distorted in a complex whole community signal with potentially hundreds of microbial
species contributing the same fatty acids but in different proportions. Therefore, even
when Methylotrophs are present in large numbers, they become undetectable in a
community proﬁle. As an illustration, Methylotrophs were not evident in whole
community FAME proﬁles from the pond, but were detected, isolated, and identiﬁed by
coupling direct-plating autoradiography and pure culture FAME proﬁles.

FAME proﬁles were also generated for the methanol degrading isolates collected
during preliminary experiments from communities near the system inlet. The proﬁles
were run against the Microbial Identiﬁcation System searchable library of proﬁles for
known organisms. All methanol-utilizing isolates appear to be of 3 main types: relatives
ofXanthobacter, relatives of Methylobacterium, and members of an unidentiﬁed group of

organisms. Both Xanthobacter and Methylobacterium are known methanol-degrading

28

organisms capable of growth on a variety of substrates including several common organic
contaminants. Such opportunistic and cosmopolitan organisms are often associated with
complex trace-level waste streams. The presence of these groups of bacteria holds
promise for secondary contaminant transformation in the Pharmacia/Upjohn Pond

system.

29

1.2

 

 

 

 

 

 

 

9 e
p
(10
9
PC2 0 L N“ 0
9 O
(130
R
1
-1.2
-l.2 0 1.2
PCI
H Hamilton 6) 20:5 cis (W3)
L Lawrence 9 16:0 anteiso
P Palmetier 6 15:0 iso
S UJ South (3 Summed Feature 3
N UJ North 0 16:0 20H
R UJ Retention Basin 6) '18:0
0 16:1cis9(w7) G) '16:0
9 '14:0 (ﬂ) 20:4 cis (w6)

 

 

 

Figure 2.2 Fatty acid methyl ester (FAME) biplots for epiphytic microbial communities
from three natural oligotrophic systems and the Pharmacia/Upjohn Pond.

30

Chapter 3

SECONDARY METABOLISM STUDIES

3.1 Introduction

Methanol comprises a signiﬁcant proportion of the dissolved organic carbon
(DOC) entering the Pharmacia/Upjohn Pond system (Klug et al., 1996), is a labile grth
substrate for a diverse array of heterotrophic organisms, and is present in sufﬁcient
concentrations (mg L'1 range) to support microbial growth. Therefore, methanol is likely
to be a primary growth substrate for attached microorganisms near the system inlet.
Furthermore, some methylotrophic organisms, such as those identiﬁed in the Oligotrophic
Comparison Study, have been shown to express alternate and auxiliary enzymatic
pathways for the degradation of other carbon compounds, such as toluene and
dichloromethane, which are among the suite of “secondary” contaminants periodically
entering the system. However, these secondary compounds are typically present in the
system at concentrations too low (pg L'1 range) to support microbial growth or induce
enzymatic pathways. In this case, the secondary compounds are likely to be metabolized
in conjunction with primary metabolism of a growth substrate, such as methanol.

Sessile primary productivity is thought to be important for stimulating bacterial
secondary productivity in naturally-occurring aquatic bioﬁlms such as those found in the
Pharmacia/Upjohn Pond. A small amount of evidence has been collected to suggest that

primary producers can have a synergistic relationship with the desired metabolic

31

remediation activity of heterotrophs in microbial mats (Bender et al., 1995). However,
questions remain regarding how applicable these concepts are to aquatic bioﬁlms and the
extent to which sessile primary productivity inﬂuences biodegradation of organics.

Potential management of the microbiota through manipulation of surface area for

attachment mandates answering fundamental questions about the bioremediation
capabilities of the bioﬁlm communities and how these communities will be affected by
artiﬁcial media and light depravation. Thus, the objectives of the secondary metabolism
experiments were as follows:

1. Determine the extent of primary methanol utilization and secondary
substrate utilization by attached microorganisms near the inlet of the
Pharmacia/Upjohn Pond

2. Determine if methanol is linked to or driving secondary substrate
utilization by attached microorganisms near the inlet of the
Pharmacia/Upjohn Pond

3. Determine the inﬂuence of sessile primary productivity on primary
methanol utilization and secondary substrate utilization by attached

microorganisms in the Pharmacia/Upjohn Pond

4. Determine the feasibility of surface area augmentation

3.2 Field Experimental Approach

In order to examine the feasibility of surface area augmentation, we established a
light exclusion experiment in the far north end of the pond near the system inlet.
Removeable Sintra PVC attachment surfaces (Commercial Plastics and Supply Corp.,
Kalamazoo, M1) were installed on 1/16/95 in treatment cells designed to manipulate light
intensity. The installation allowed microbial communities to develop in situ and

simulated the various light regimes that these communities would encounter if surface

32

area in the pond were augmented by the implementation of a product such as Biomedia
(Brentwood Industries, Inc., Reading, PA). Treatment cells consisted of plastic crates
suspended in the water column and wrapped in screening materials to manipulate light
intensity. Flow vents were placed in the coverings to allow normal water movement
within the crates.

The three treatment designations of LIGHT, MEDIUM, and DARK correspond to
natural unattenuated diurnal light, moderately attenuated light (approximately 50%), and
fully-attenuated light, respectively. However, in response to results from methanol
oxidation experiments and periphyton productivity estimates, the MEDIUM treatment
was later eliminated in lieu of additional ﬁeld replications for the DARK and LIGHT
treatments. Greater statistical rigor and conﬁrmational analyses were warranted in
subsequent experiments due to high variability among ﬁeld replications.

Sintra PVC was the chosen material for an attachment surface due to its dark
color, hydrophobic surface, ready availability, and low price. Sintra PVC tiles of uniform
surface area (2.54 cmz) were attached to unglazed ceramic tiles (anchors) and placed in
each crate for colonization. Although effective, this tiling system was later modiﬁed to
0.64 cm thick x 7.6 cm wide x 17.8 cm long rectangular-shaped Sintra PVC tiles mounted
within the crate interior using nylon hardware.

Periphyton productivity was measured in situ as an indicator of the effectiveness
of light manipulation treatments on controlling primary productivity. After communities
had fully developed, two tiles were harvested from each treatment cell on each of several

dates and returned to the lab for a series of analyses that included the following: direct

33

microscopic counts, fatty acid methyl ester (FAME) community proﬁles, methanol
oxidation and assimilation rate, and secondary metabolism shakeﬂask experiments.

Tiles were harvested by removing the nylon fastening hardware and carefully
manuevering the tiles into sterile Whirlpak bags (Nasco, Fort Atkinson, WI), which were
stored on ice for transport to the lab. Each bag received 100 mL of 0.22-11m ﬁlter-
sterilized pond water for processing purposes. Manual agitation and rubbing was applied
to the tiles and sterile pond water in the transport bags to remove bioﬁlm material from
the surface of the tiles. Homogenization of the bioﬁlm suspension was necessary to
disperse the polysaccharide matrix within bioﬁlms and to distribute the cellular material
for uniform subsampling. Although homogenization perturbed the bioﬁlm community,
diminished the mass transfer limitation presented by the polysaccharide matrix, and
potentially overestimated in situ transformation rates, the processing was deemed
necessary for accomodation of the destructive subsampling regime established in the
laboratory experimental design. The polysaccharide matrix was most prominent in the
DARK treatments. Bioﬁlm suspensions were homogenized for 30 seconds on the low
setting in a sterile stainless steel Eberbach tissue grinder on a Waring industrial blender
base. The tissue grinder was rinsed with an experiment-dependent volume of sterile pond
water, the rinsate from which was consolidated with the homogenized suspension. The
tissue grinder was sanitized between samples with a triple rinse of 10% sodium-
hypochlorite (Clorox Co., Oakland, CA), followed by liberal rinsing with tap water, then

distilled water, and ﬁnally a triple-rinse with sterile pond water.

34

The consolidated bioﬁlm suspension was subsampled for a variety of analyses,
depending upon the nature of the experiment. Triplicate 10-mL subsamples for FAME
analysis and triplicate l-mL subsamples for direct microscopic counts were collected and
analyzed according to the protocols described in Chapter 1. Additional subsamples of
experiment-dependent volume were collected and distributed to 20 mm, 135-mL serum
vials sealed with teﬂon-lined septa for an array of metabolic experiments. Processing

variations are noted for each individual experiment below.

3.3 Sessile Primary Productivity

Standing water sessile primary productivity was determined according to the
standard protocol as deﬁned by Standard Methods for the Examination of Water and
Wastewater (APHA, 1992b). PVC tiles with intact communities were placed in light and
dark bottles and were incubated for one half the photoperiod. Bottles were subsampled
for dissolved oxygen (DO) concentration before and after the incubation. DO samples
were chemically ﬁxed in the ﬁeld for preservation, followed by iodometric titration in the
laboratory. Sessile productivity and respiration rates were also corrected for suspended
phytoplankton activity by establishing similar bottles without attached communities.

Non-replicated 12-hour sessile productivity rates and 24-hour respiration rates for
the three treaments are presented in Figure 3.1. Productivity rates conﬁrmed that the
treatments were effective at controlling primary productivity within the bioﬁlms.
Interestingly, the respiration rates were nearly the same for all three treatments, a result

that will be discussed with the methanol oxidation data. The presented rates of primary

35

 

I productivity

 

 

 

 

3000
3‘ D respiration
E 2500
i3
39’ 2000
E
g. 1500
E
901000
is”
on 500
E
O
—500
dark medium light
treatment

Figure 3.1 Primary productivity and respiration rates (mg 02 In2 d") for attached
microbial communities from three light treatment regimes in the
Pharmacia/Upjohn Pond.

36

productivity and respiration are similar to those found in the periphyton literature

(Wetzel, 1983).

3.4 Methanol Oxidation/Assimilation

Methanol oxidation rates were determined by incubating cell suspensions in
sealed serum vials with l4C-radiolabeled methanol and subsequently trapping MC-labeled
carbon diox1de. Triplicate 10-mL aliquots of cell suspension per sample were placed in
37-ml serum vials sealed with butyl .ubber septa and aluminum crimp caps. Two
additional ﬂasks per sample were established for a t==0 estimate and biological control.
Biological controls were autoclaved for 25 minutes and cooled prior to incubation. A
concentrated stock solution of l4C-methanol with speciﬁc activity of 5 mCi/mmol was
obtained from New England Nuclear (Boston, MA). Each ﬂask received an injection of
diluted l4C-methanol via syringe to a ﬁnal concentration of 0.028 uCi mL" (4 mg
methanol L" suspension). Injections were staggered to allow immediate ﬂushing and
trapping of headspace gases. Samples were incubated at 25° C for 3 hours, except t=0
samples, which were halted immediater upon injection of label.

Methanol oxidation was halted in all ﬂasks by injecting 300 11L of 6 N HCI, which
also forces 14CO2 in solution to be evolved into the headspace. (Upon acidiﬁcation, each
ﬂask was then placed into a separate channel on a closed circuit N2 headspace ﬂushing
apparatus. Each channel is ﬁtted with a series of 3 scintillation vials, each containing 4
mL of 3:5 ethanolaminezmethanol trapping reagent for trapping l4CO2. Each vial was
ﬂushed for 10 mimites at a rate of approximately 60 m1 N2 min", after which time the

scintillation vials were removed from the apparatus, treated with 15 mL oi‘Safety-Solve

37

liquid scintillation cocktail (Research Products International Corp., Mount Prospect, IL),
and capped. Radioactivity in each vial was analyzed with a Wallac model 1409 Liquid
Scintillation Counter (Wallac Inc., Gaithersburg, MD).

Remaining cell suspensions were ﬁltered through Whatman 0.2 11m nitrocellulose
ﬁlters (Whatrnan Ltd., Maidstone, England) and the ﬁltrate collected. Biomass recovered
on the ﬁlter and 10 mL ﬁltrate were each transferred to separate scintillation vials with 15
mL Safety Solve each and counted with the LSC to estimate bacterial assimilation and
residual l4C-methanol in solution, respectively.

Methanol oxidation and assimilation rates are presented in Figure 3.2. Rates have
been adjusted to a per-million-cell basis by direct microscopic counts, which accurately
reﬂects the community-wide oxidation and assimilation rates. Although the relationship
between direct microscopic counts and the number of culturable methanol oxidizers
remains to be detennined, approximately 10% of the total culturable population were
identiﬁed as methanol oxidizers by direct-plating autoradiography. Therefore, oxidation
and assimilation rates for the populations of interest are most likely underestimated.

Error bars throughout the present study represent plus or minus one standard error
of the mean. Although a trend is apparent toward higher oxidation rates in the DARK
treatment, an AN COVA (SAS, alpha=0.05) revealed no signiﬁcant differences in
methanol oxidation or assimilation rates between treatments; i.e. given the high level of
inherent variability, our experiment was not designed with sufficient statistical power to

recognize the apparent trends in the data.

38

Despite the inherent variability in ﬁeld samples, the DARK treatment exhibited a
higher per-cell oxidation rate than the other two treatments. Thus, bioﬁlms growing in
the dark and without access to a diverse photosynthetically-produced carbon pool appear
to be more well-adapted to using methanol as a primary carbon source. In conjunction
with the gross respiration rate data, these results also suggest that methanol oxidation may
be a larger component of total respiration in the DARK treatment. However, the ratio of
mineralization to assimilation was substantially higher in the DARK treatment
(approximately 2: 1) than in the other treatments (approximately 1:1), suggesting that
communities growing without light incorporate less methanol into biomass and divert
more to energy production. The practical implication is that Biomedia augmentation
would not only be feasible, but may be preferable, since the majority of colonization on

artiﬁcial media would be in the dark.

39

 

 

0.00 I 2
- ﬂ oxidation

 

1:] assimilation

 

 

 

0.001 -

  
  
    
  
   

 

0.0008 -

 

1
0.0006 1
l

nmol/1E6 cells/hr

0.0004 -

0.0002 -

 

 

 

 

 

 

0-

 

 

dark medium light

Figure 3.2 Methanol oxidation and assimilation rates (nmol [1x106]cells" hr") for
attached microbial communities from three light treatment regimes in the
Pharmacia/Upjohn Pond.

40

3.5 Secondary Metabolism Shakeﬂask Experiments

3. 5.1 General Experimental Approach

The general experimental approach to studying secondary metabolism involved
harvesting bioﬁlms from the artiﬁcial attachment surfaces in the light exclusion chambers
and preparing bioﬁlm suspensions in the laboratory, as explained in section 2.2 above. In
sediment experiments, surﬁcial sediments were Eckman dredged to a depth of 10 cm and
a similar suspension prepared under micro-aerophilic conditions. Bioﬁlm suspensions
were subsampled in 50-mL aliquots, which were placed in 20 mm, 135-mL acid-washed
and sterile serum vials. Each vial was sealed with a teﬂon-lined septum and aluminum
crimp cap. Bioﬁlm suspensions were pre-incubated overnight to deplete any labile DOC
prior to initiating secondary metabolism incubations. All bioﬁlm subsamples from both
LIGHT and DARK ﬁeld treatments were incubated in the absence of light to strictly
isolate heterotrophic activity and eliminate the confounding inﬂuence of simultaneous
photosynthetic activity. All laboratory incubations and pre-incubations throughout the
present study were conducted at 25° C.

Biological controls (designated KILL) were introduced to establish a cause and
effect relationship between biologically active samples and transformation activity. The
biological control also allowed isolation and quantiﬁcation of biological matrix effects
(such as sorption) on transformation activity. Biological control vials were prepared as
described for biologically active samples, then autoclaved for 20 min and allowed to cool
prior to incubation. Experiments #1 through #3 revealed that a 20-min autoclave period

was not sufﬁcient to control activity beyond 24 hours, necessitating an extension of the

41

autoclave sterilization time to 45 min in subsequent experiments. The protective
properties of the bioﬁlm matrix may have allowed bacterial spores to survive the 20-min
sterilization period, after which spore germination could have led to biological activity in
the sterilized controls. Altemately, biological controls may have been cross-contaminated
during sampling.

A reagent control (designated RBLK) was introduced to isolate non—biological
effects (such as volatilization and chemical interactions) on apparent transformation
activity. Reagent control vials were established as described previously, but substituting
50 mL of ﬁlter-sterilized Pond water for the bioﬁlm suspension. Reagent controls were
treated exactly as biologically active samples throughout the remainder of the experiment.

Following pre-incubation, the sealed suspensions were spiked with 100 mg L"
methanol (or no methanol) and a mixture of secondary compounds at pg L"
concentrations and varying composition, depending upon the nature of the experiment.
Vials were destructively sampled in 6-mL aliquots through time to establish rates of loss
for methanol and secondary compounds. During experiments #1 through #3, each
subsample was collected in a single 10-mL gastight syringe that was rinsed 3 times with
sterile water and ﬁtted with a new sterile 18 ga needle between samples. This technique
may have resulted in some contamination of the reagent blanks or biological controls.
Subsamples in remaining experiments were each collected with a separate sterile syringe
and needle. An equal volume of atmospheric air was injected into each vial following

subsampling to replace the lost volume and to replenish oxygen to maintain aerobic

42

conditions. Anaerobic studies received an equal volume of oxygen-free high purity
nitrogen.

Methanol subsamples were ﬁltered through 0.2 pm Millex disposable ﬁltration
units (Millipore Products Division, Bedford, MA) to remove active bacterial biomass and
thereby stop biological transformation of methanol. Filtrate was collected in GC
autosampler vials ﬁtted with glass inserts, which were then sealed with teﬂon-lined crimp
caps and stored at 4°C prior to analysis. Methanol subsamples were analyzed by direct 2-
pL aqueous injection and ﬂame-ionization detection on a Hewlett Packard 5890 Series II
ﬁtted with a 7673A autosampler and a Restek RTX-200 capillary column (Restek
Corporation, Bellefonte, PA) with a crossbonded triﬂuoropropylmethly polysiloxane
active matrix (30 m x 0.53 mm ID, 1.0 pm ﬁlm thickness) ﬁtted with a Restek 5 m x
0.53 mm ID inert guard column. Concentrations were determined by applying the
resultant formula from a linear regression of the methanol external standard curve.

Subsamples for secondary compounds (volatile organic compounds, or VOCs)
were biologically stabilized by adjusting to pH 2.0 with 1.0 N HCl. Subsamples were
collected in Micro-mate 10 mL glass-barreled/glass-plungered syringes (Popper and Sons,
Inc., New Hyde Park, NY), sealed with custom-fabricated teﬂon-lined Luer caps, and
stored at 4°C for no more than 4 days prior to analysis. Although this sample storage
protocol is not recommended in EPA Method 8260A (USEPA, 1994), a series of
preliminary tests were executed to determine the suitability of the storage protocol and no
signiﬁcant deterioration was detected for the analytes of interest. Samples were then

analyzed according to EPA Method 8260A (USEPA, 1994) using a Hewlett Packard

43

purge-and-trap/gas chromatograph/mass spectrometer system. The HP purge-and-trap
concentrator was equipped with a Tekmar VOCARB 3000 trap (Tekmar Co., Cincinatti,
OH). The HP 5890 Series II EPC gas chromatograph was equipped with an HP—VOC
column (30 m x 0.2mm ID, 1.1 pm ﬁlm thickness) and a 5972 mass selective detector.
Reconstructed ion chromatographs and mass spectra were analyzed with HP Chemstation
and Enviroquant software to determine analyte concentrations based upon calibration
curves and internal standards.

Plots were prepared of analyte concentration against time and, when applicable,
rates were calculated based on linear regression of the transformation plots. Treatment
effects were then statistically evaluated using a SAS analysis of covariance for rates and
direct microscopic count data. Speciﬁc methods and resulting raw transformation plots
for individual secondary metabolism experiments are presented in sections 3.5.2 through
3.5.7. Tabular summaries of count data, rate estimates, statistical tests, and experiment-
wide means for all secondary metabolism experiments are presented in section 3.5.8.

Discussion of secondary metabolism results follows in section 3.6.

3. 5.2 Secondary Metabolism Experiment #1 : Aromatic/Aliphatic Screen

The objective of experiment #1 was to survey bioﬁlm secondary metabolic
activity on dichloromethane (DCM) and toluene (TOL). These compounds were selected
as representative secondary contaminants from the aliphatic and aromatic classes of
compounds and are both reported to be degraded by methylotrophs (see Table 1.1).
Methanol served as the primary growth substrate at 100 mg L", while DCM and TOL

were spiked at initial concentrations of 480 pg L" and 260 pg L", respectively. Because

44

this experiment was initiated before the subsample storage protocol was developed,
subsamples were collected at incubation times 0.5, 8.0, and 28.0 hours. These time
intervals allowed samples to be placed immediately on the purge-and-trap autosampler
for analysis.

Primary and secondary transformation plots for experiment #1 are presented in
Figures 3.3-3.5. Rapid transformation rates of methanol and toluene were observed in all
biologically active bioﬁlm samples, with dark communities exhibiting higher per-cell
rates. However, differences in transformation rates were not statistically signiﬁcant
(a=0.05) due to high variance within treatments. Direct microscopic counts were not
signiﬁcantly different between treatments, as shown in Tables 3.1 and 3.2.

Unexpectedly, no bioﬁlm secondary activity was observed with DCM. Some
methylotrophs are known to express dichloromethane dehalogenase, an enzyme that
dechlorinates DCM to form formaldehyde, which is then shunted directly into the
methanol oxidation and assimilation pathways. The failure to express this aerobic
enzymatic pathway could be due to absence of the appropriate genetic code, the sporadic
nature of DCM contamination, and/or concentrations below the threshhold level for
expression. Interestingly, DCM transformation was later observed in aerobic bioﬁlms
after extended periods of methanol exposure, as well as in sediment studies under

anaerobic conditions (Sections 3.5.6 and 3.7.4).

45

 

100 L

 

-1 s ' N ‘_‘ >
90 L 77 Al‘

80:

 

 

1
70_

 

60:

 

50;

 

mg methanol/L

40,

 

30:

20d

 

 

 

 

 

O 5 1 0 1 5 20
time (hours)

.1
O IIIJIIIITTITIIIFTIIIIITIITIl-r-J

25

 

30

 

 

I DARK
__. LIGHT
— Ar - RBLK
—+— KILL

 

Figure 3.3 Methanol transformation plot for laboratory incubations of attached
microbial communities from two light treatment regimes in the

Pharmacia/Upjohn Pond, Experiment #1 (7/13/95).

46

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

300
250; l DARK
’E : 4» LIGHT
33200;
g0 . ~A~ RBLK
150‘ 99+ KILL
100_
50:
()4 I I I I I I I I I IfTTI I I I II I IITITI ITI
0 5 10 15 20 25 30

time (hours)

Figure 3.4 Dichloromethane (DCM) transformation plot for laboratory incubations of
attached microbial communities from two light treatment regimes in the
Pharmacia/Upjohn Pond, Experiment #1 (7/13/95).

47

 

200

 

180‘

160%

140‘ ~ L...

 

 

EI-

 

 

 

I “”7"!
, I DARK
T L.. LIGHT

120:

 

.9100:
99 j A RB

80 j air KILL

 

 

 

 

 

60'

 

40‘

 

20’

 

II l——o—~—I

 

Oll‘lllllllTIIT
0 5 10 15 20 25 30

time (hours)

IITf TUTTI II
I I I

 

Figure 3.5 Toluene (TOL) transformation plot for laboratory incubations of attached
microbial communities from two light treatment regimes in the
Pharmacia/Upjohn Pond, Experiment #1 (7/13/95).

48

3. 5.3 Secondary Metabolism Experiment #2: T ime-Intensive Subsampling

Experiment #2 was designed as a follow-up study to experiment #1 , but with
freshly harvested bioﬁlms and a more intensive subsampling schedule (0.5, 9.0, 22.0, and
50.0 hours) for more accurate estimation of transformation rates. Initial concentrations of
DCM and TOL were reduced to 300 pg L" and 170 pg L" respectively to closer emulate
in situ concentrations (see Table 3.12). The subsample storage protocol was ﬁrst
implemented during this experiment. Subsamples for direct microscopic counts were not
collected, necessitating the use of counts from experiment #1 for rate calculation
comparisons.

Primary and secondary transformation plots for experiment #2 are presented in
Figures 3.6-3.8. Results were similar to experiment #1, but with lower variance.
Methanol and TOL were rapidly transformed in both light treatments. DCM was not
aerobically transformed throughout the duration of this experiment. No signiﬁcant
differences in primary or secondary transformation rates were observed between
treatments. The biological control (KILL) for TOL transformation exhibited substantial
activity at the t=50 h sampling, indicating that the 20-min autoclave sterilization period
was not sufﬁcient to control biological activity in these samples. The activity in the
biological control at 50 h suggests the presence of spore-forming bacteria or possible

cross-contamination during sampling.

49

100

 

LL41

90

 

 

80, _,

 

60

 

50

 

as: f i
I
l

 

mg methanol/L

40j

 

30j

 

20;

10‘

 

 

 

 

lﬁTIFTTfI

0 5 10 15

time (hours)

T F T

I I T If I I

20

 

25

 

 

trail

DARK
LIGHT
RBLK
KILL

 

Figure 3.6 Methanol transformation plot for laboratory incubations of attached
microbial communities from two light treatment regimes in the

Pharmacia/Upjohn Pond, Experiment #2 (7/26/95).

50

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I . -- g; ‘
250   * e 2.;
.2'I.
I DARK
200
E 1 o LIGHT
U
,, .
also 1‘“ RBLK
-+— KILL
1004
50:
0 TIT—TITWjIIII—ITrIrIITTITTWIerﬂFIjIIIIIIIIITIII

 

 

 

0 5101520253035404550
time(hours)

Figure 3.7 Dichloromethane (DCM) transformation plot for laboratory incubations of
attached microbial communities from two light treatment regimes in the
Pharmacia/Upjohn Pond, Experiment #2 (7/26/95).

51

200

 

180

 

 

 

160‘

 

 

_. r—t
N A
O 0

mg methanol/L
S
o

00
O

L l l

 

 

 

 

 

 

 

 

 

 

1 0
60: ‘i
404 [‘2‘
20: ,4? 5 "
0‘ TIYTTTITIIIlfTTTlTTTTllll[[TrﬁerTlllllllll‘lNI‘;
0 5 10 15 20 25 30 35 40 45 50

time (hours)

 

 

fl DARK

0 ~ LIGHT
f RB
4* KILL

 

Figure 3.8 Toluene (TOL) transformation plot for laboratory incubations of attached
microbial communities from two light treatment regimes in the

Pharmacia/Upjohn Pond, Experiment #2 (7/26/95).

52

 

3.5.4 Secondary Metabolism Experiment #3: 5 Simultaneous Substrates

Experiment #3 was designed to simultaneously evaluate transformation rates for
ﬁve secondary substrates, all of which are occasional pond contaminants. Methanol was
added at an initial concentration of 68 mg L". The following secondary substrates were
added at the approximate initial equilibrium concentrations indicated in parentheses:
DCM (70 pg L"), DCA (50 pg L"), BDCM (40 pg L"), CB (45 pg L"), and TOL (30 pg
L'I). Subsamples were collected at 0.5, 14, and 72 hours.

Direct microsc0pic counts were comparable to experiment #1 , but the decreased
variance in experiment #3 allowed statistical detection of signiﬁcantly higher cell
numbers in the DARK treatment. The decreased variance may have been a function of a
fully-matured developmental state within the bioﬁlm. Primary and secondary
transformation plots are presented in Figures 3.9 through 3.13. Results for the biological
controls were comparable to experiments #1 and #2, but they were omitted from
transformation plots due to the previously discussed problem of ineffective autoclave
sterilization over long term experiments. Methanol was transformed rapidly and at rates
comparable to experiments #1 and #2 (see Table 3.3). Methanol transformation rates
were again not signiﬁcantly different between light treatments (see Table 3.4).

None of the aliphatic secondary compounds were biologically transformed in this
experiment. Although all vials exhibited loss of DCM, DCA, and BDCM over time, the
rates of loss did not differ from those of the reagent controls (RBLK), indicating that non-
biological factors were responsible for these losses. TOL and CB were both secondarily

transformed at comparably rapid rates, with DARK communities exhibiting signiﬁcantly

53

100,

90: %
1

80:

 

 

 

 

.+.

70%

‘1H

 

 

60 f.
. -I DARK

 

50 j
~ 0 LIGHT

mg methanol/L

40 j
i a RBLK

 

 

3O

 

 

 

 

2O

10

 

 

 

O T T T I I T .T T I T 7 T T I T I T T I T TﬁT T I T T T T I T TKTTTTTIT T T
O 10 20 3O 4O 50 60 7O 80
time (hours)

 

Figure 3.9 Methanol transformation plot for laboratory incubations of attached
microbial communities from two light treatment regimes in the
Pharmacia/Upjohn Pond, Experiment #3 (8/11/95).

54

 

100,

 

90:

so? 9
70; A;

 

 

 

 

 

 

 

 

 

 

 

 

 

 

60; T
E 3 $ «1 DARK
go : wt LIGHT
40‘ ,-
3 9 A. RBLK
30 a?
201 I f
4 ,
10',
Oi“‘I"‘l“'I"'I“'I“'l"‘l'"I"

 

 

O 20 4O 6O 80 100 120 140 160 180
time (hours)

Figure 3.10 Dichloromethane (DCM) transformation plot for laboratory incubations of
attached microbial communities from two light treatment regimes in the
Pharmacia/Upjohn Pond, Experiment #3 (8/11/95).

55

 

100,

 

90;
j

80‘

j
.l
70‘
l
1

 

 

 

 

60:
J—- DARK

50;:
l «G LIGHT

_l

40‘,

j i— A RBLK
30: *4 _- ,

 

 

ug dca/L

 

 

 

 

 

20‘

 

 

10:

 

 

 

041"IM'ILDLITWI’D'I'HI'"1*"TT'H
0 20 40 60 80 100 120 140 160 180
time(hours)

Figure 3.11 1,2-dichloroethane (DCA) transformation plot for laboratory incubations of
attached microbial communities from two light treatment regimes in the
Pharmacia/Upjohn Pond, Experiment #3 (8/11/95).

56

 

 

 

 

 

 

 

$ I DARK

 

~G LIGHT

 

ug bdcm/L
{2’
T
r
f.
/
F—d‘—-—d

A RBLK

 

 

 

 

 

 

 

 

 

0 TITITTTITTTITTTITTTITTTITTTIT‘rTITT

0 20 40 60 80 100 120 140 160 180
time (hours)

Figure 3.12 Bromodichloromethane (BDCM) transformation plot for laboratory
incubations of attached microbial communities from two light treatment
regimes in the Pharmacia/Upjohn Pond, Experiment #3 (8/11/95).

57

 

 

 

 

 

 

 

1: I DARK

 

 

3 ‘ -o LIGHT

 

*‘V’ RBLK

 

 

 

 

 

 

!

0 A W—T T I T T T I T T T I T I I T T T I T T T I T T T I T T T ITI'T‘
0 20 40 60 80 100 120 140 160 180
time (hours)

 

 

 

Figure 3.13 Chlorobenzene (CB) transformation plot for laboratory incubations of
attached microbial communities from two light treatment regimes in the
Pharmacia/Upjohn Pond, Experiment #3 (8/11/95).

58

30

 

 

 

 

 

 

 

 

 

 

 

 

 

I
251
I
20 A.
I
4 —I LIGHT
5‘. I
.915 ..
go i o DARK
: l. l
104 i —-—A RBLK
5
— g.
0 TTTITTTITTTI_T!:ITTTITT_%ITT——TTITTTITTT

 

O 20 4O 6O 80 100 120 140 160 180
time (hours)

Figure 3.14 Toluene (ITOL) transformation plot for laboratory incubations of attached

microbial communities from two light treatment regimes in the
Pharmacia/Upjohn Pond, Experiment #3 (8/11/95).

59

higher CB transformation. Again, the reagent control (RBLK) exhibited losses, but the
loss rate was signiﬁcantly greater in the biologically active samples. Changes in the vial
headspace volume due to destructive sampling likely disturbed the equilibrium
concentration of volatile organics in solution, resulting in apparent losses in the reagent

controls.

3. 5.5 Secondary Metabolism Experiment #4: Primary/Secondary Link

Experiment #4 was designed to determine the link between primary methanol
metabolism and secondary substrate utilization. Incubation ﬂasks in the lab were divided
between two treatments: no methanol and 100 mg L" methanol. Flasks receiving
methanol (designated +MeOH) were injected with a mixture of secondary substrates
dissolved in methanol to the following initial concentrations: DCM (50 pg L'l), DCA (60
pg L'l), BDCM (50 pg L'I), CB (40 pg L'l), and TOL (40 ug L'l). Flasks not receiving
methanol (designated -MeOH) were treated with a water-based solution of the same
secondary substrates. DCM and BDCM were apparently immediately volatilized from
the water-based stock solution and were not found in any subsamples. Final initial
concentrations for the remaining secondary compounds in the water-based solution were
as follows: DCA (80 pg L"), CB (40 pg L"), and TOL (40 pg L"). A unreplicated
sediment sample was collected and processed according to the method described in
section 3.5.6 and injected with the methanol-based stock.

Direct microscopic counts are summarized in Tables 3.1 and 3.2. Total numbers
of bacteria in DARK bioﬁlms were comparable to results from experiments #1-#3.

However, LIGHT bioﬁlms exhibited approximately 5 times higher total numbers of

60

bacteria than were observed in previous experiments. Bacteria were signiﬁcantly more
numerous in the LIGHT treatment, perhaps a reﬂection of the increased temperature and
light intensity of late summer.

Primary and secondary transformation plots are presented in Figures 3.15 through
3.20. Figure 3.15 demonstrates that methanol transformation rates (in those ﬂasks
receiving methanol) were again comparable to previous experiments and were not
signiﬁcantly different between light treatments. The presence of methanol also had no
signiﬁcant effect on the transformation of any secondary compound.

DCM, DCA, and BDCM were not biologically transformed in the course of this
experiment. No data is presented in Figures 3.16 or 3.17 for the methanol-negative vials,
due to volatilization and solubility problems with DCM and BDCM in the water-based
stock. Although no direct comparison could be made between methanol-positive and
methanol-negative transformation rates for DCM or BDCM, methanol clearly did not
stimulate secondary transformation of these compounds. Figure 3.17 demonstrates the
lack of aerobic DCA transformation activity in either the presence of absence of
methanol. Transformation rates of TOL and CB were again comparable to previous
experiments and were signiﬁcantly higher in the DARK treatments. The presence of
methanol did not stimulate secondary transformation of either TOL or CB.

As shown in Figure 3.21, the anaerobic sediment slurry exhibited rapid
transformation of methanol, DCM, and TOL. Because this was the ﬁrst evidence of
secondary DCM transformation, a replicated sediment experiment was designed and

implemented next. BDCM was immediately undetectable in the sediment slurry,

61

 

 

 

 

 

 

 

 

 

 

120 _
. 1. "
I I. -
I _
lOO——? I} f
i I
80
Li; I
s . it
5 60
Q)
5 i
w
e « l
40
, T :F
20 I
I
O THTIYTTIITITTITTTIIITTTTTTI—TIIIIIIIIITIIIIIITHII
o 1 2 3 4 5 6 7 8 9 I0

time (hours)

 

 

I DARK
& LIGHT
A— RBLK
4* KILL

 

Figure 3.15 Methanol transformation plot for laboratory incubations of attached
microbial communities from two light treatment regimes in the

Pharmacia/Upjohn Pond, Experiment #4 (9/13/95).

62

 

 

 

 

 

 

 

 

 

 

 

 

 

60
I
50 - +
1 i
40
E I I DARK
,
_830 C LIGHT
w .
= 4‘ RBLK
20‘ 4% KILL
10
0 ”HI"I'Trrﬂlﬂ'llﬂﬁrlfﬂjlﬁﬂrTTﬂ'l””I‘Trt

 

 

0 1 2 3 4 5 6 7 8 9 10
time (hours)

Figure 3.16 Dichloromethane (DCM) transformation plot for laboratory incubations of
attached microbial communities from two light treatment regimes in the
Pharmacianpjohn Pond, Experiment #4 (9/13/95).

63

 

60

 

 

 

 

 

 

 

 

 

40 q f
E: i if “7 —I— DARK
f2: 30 + LIGHT
‘2‘? j A — RBLK
20 A 41¢ KILL
)
'l
I
'l
10
0 TITIIrTTTIjTrTIrTTTIIIITIITTIIIITTIITTTITTTTITTTT

 

 

 

0 l 2 3 4 5 6 7 8 9 10
time (hours)

Figure 3.17 Bromodichloromethane (BDCM) transformation plot for laboratory
incubations of attached microbial communities from two light treatment
regimes in the Pharmacia/Upjohn Pond, Experiment #4 (9/13/95).

64

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

90 ,
: L - _MA
80-§—§-~ 1133’4' “ -
3 B‘ " i
70
: ~l DARK (+)MeOH
ooj k/ i I} DARK(-)MeOH
E 50—2 I I ’ o-h‘ LIGHT (+)MeOl-I
u 3 0A LIGHT (-)MeOH
2,040
I + RBLK(+)MeOH
30 j + RBLK(-)Me0H
4
3 _+_
20 . KILL
l0 ;
I
o ....,...,,....T....T....T..TTFT...,....I....,e...

 

 

0 l 2 3 4 5 6 7 8 9 10
time (hours)

Figure 3.18 l,2-dichloroethane (DCA) transformation plot for attached microbial
communities from two light treatment regimes in the Pharmacia/Upjohn
Pond, incubated in the laboratory in the presence or absence of methanol,
Experiment #4 (9/13/95).

65

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

50 ,
45 j
40j
3 I! DARK (+)MeOH
35 ,
; D DARK (-)MeOH
" 1:
d 301 T 3 74¢ LIGHT (+)MeOH
1 I:
325 I ~94 LIGHT (—)MeOH
:3 4 P
20 j l + RBLK(+)MeOH
15 i —A— RBLK(-)MeOI-I
—+— KILL
10 ,
5 j
o ‘ WWW,..,-T4...r....,m.,.44.,,mlwmﬁ
o I 2 3 4 5 6 7 8 9 10

time (hours)

Figure 3.19 Chlorobenzene (CB) transformation plot for attached microbial communities
from two light treatment regimes in the Pharmacia/Upjohn Pond, incubated
in the laboratory in the presence or absence of methanol, Experiment #4
(9/13/95).

66

 

 

 

 

 

DARK (+)MeOH

 

DARK (-)MeOI-I
LIGHT (+)MCOH
LIGHT (-)MeOH
RBLK (+)MeOH
RBLK (-)MeOH
KILL

 

 

 

 

+++¢9m'

 

 

 

 

 

 

 

 

0 T'IT'YTTVIIYVYITTYYIYYTTITTYTITﬁT'YTTTITYTTITTT—T

0 l 2 3 4 5 6 7 8 9 10
time (hours)

Figure 3.20 Toluene (TOL) transformation plot for attached microbial communities from
two light treatment regimes in the Pharmacia/Upjohn Pond, incubated in the
laboratory in the presence or absence of methanol, Experiment #4 (9/13/95).

67

 

DCM
DCM-kill
DCA
DCA-kill
BDCM
BDCM-kill
CB

CB-kill
TOL

iii-litteri-

TOL-ki ll

 

 

 

 

0 10 20 30 40 50 60 70 80
time (hours)

Figure 3.21 Multiple compound transformation plots for laboratory incubations of
sediment slurry from the Pharmacia/Upjohn Pond, Experiment #4 (9/13/95)

68

indicating either very rapid uptake of BDCM by the sediment microorganisms or a strong

sorption afﬁnity for the compounds by sediment particles.

3. 5. 6 Secondary Metabolism Experiment #5: Anaerobic Sediment Study

Experiment #5 was designed to develop rate estimates for sediment-associated
anaerobic secondary activity and to determine the associated role of sulfate reduction.
Surﬁcial sediments were sampled in triplicate from the north end of the Pond with an
Eckman dredge operated to a depth of 10 cm. Sediment slurries were prepared in the lab
by diluting 1:1 (vzv) with ﬁlter-sterilized and pre-boiled Pond water while working under
a stream of oxygen-free high purity nitrogen. Slurry subsamples of SO-mL volume were
placed in 20 mm, 135-mL serum vials, which were sealed with teﬂon-lined septa and
purged for 10 minutes prior to incubation with high purity nitrogen gas passed through a
metallic copper oxygen-stripping column operating at 250 C. Slurry was also subsampled
in lO-ml aliquots for sediment dry-weight determination. Dry weights were measured
after oven-drying the slurry 10 days at 105° C. Other incubation methods were similar to
experiments #3 and #4, with the exception that each subsample was replaced with an
equal volume of oxygen-free high purity nitrogen gas to prevent oxygen inﬁltration into
the ﬂasks.

Sulfate reduction was shown to have a signiﬁcant role in anaerobic respiration of
methanol in the Pharmacianpjohn Pond sediments (Klug et al., 1996). Because sulfate
reduction signiﬁcantly impacts primary methanol metabolism, we hypothesized that
sulfate reduction would have a similar impact on secondary metabolism, particularly if

the secondary activity is linked to primary methanol metabolism. Half the samples were

69

treated with a 50 mM sodium molybdate solution to block sulfate reduction. A stock
solution of substrates was injected in each ﬂask to achieve the initial equilibrium
concentrations indicated in parentheses for each compound: MeOH (100 mg L'l), DCM
(40 pg L"), DCA (50 pg L"), CB (30 pg L"), and TOL (25 pg L"). Subsamples were
collected at 0.2, 6.5, 12.5, and 18.0 hours.

Several subsamples for methanol analysis were inadvertently destroyed prior to
analysis. Therefore, methanol transformation for sediment incubations was not plotted
and rate statistics were not calculated. Secondary transformation plots are presented in
Figures 3.22-3.25. DCM and TOL transformation (Figures 3.22 and 3.25, respectively)
was extremely rapid under sulfate-reducing conditions and was nearly 100% inhibited in
the presence of the sulfate reduction block. DCA and CB (Figures 3.23 and 3.24,
respectively) were not anaerobically transformed in the course of this experiment and

showed no differing response when sulfate reduction was blocked.

70

 

45;

35:

 

 

 

 

 

 

Lo)
O
1"“!
L]
. l .4
I—ﬁ-ﬂIF—j J —-I—" ——-|

 

 

 

 

 

 

 

 

E 25 3 ” I— UNBLOCKED
g : {L BLOCKED
m .
3 20; A- BLANK

15 j «4* KILL

1

10 ,

5 i +

0‘ T 1 T I T I TIT I I I I T I Iﬁ—T T IT [IAIN-T T

 

 

 

0 2 4 6 8 10 12 14
time (hours)

Figure 3.22 Dichloromethane (DCM) transformation plot for laboratory incubations of

sediment slurry from the Pharmacia/Upjohn Pond in the presence or absence
of a sulfate reduction inhibitor, Experiment #5.

71

 

60

50

 

 

 

 

10‘

 

 

 

O.

O

TTTIjT—TITTTITTTIfTﬁTTTIITTT

2 4 6 8 10 12
time (hours)

14

 

 

++mh

UNBLOCKED
BLOCKED
BLANK

KILL

 

 

Figure 3.23 1,2-dichloroethane (DCA) transformation plot for laboratory incubations of
sediment slurry from the Pharmacia/Upjohn Pond in the presence or absence
of a sulfate reduction inhibitor, Experiment #5.

72

 

 

 

 

 

 

 

 

 

 

 

 

 

40
353} I f f T
30 f - 1., ~
: i . xx
25: .
<1 . wl~ UNBLOCKED
220: E1 BLOCKED
3 f A BLANK
15,
1 ~+ KILL
10:
SI
1
O _T T T I T T TIT T TIT T TIT T T I T T T

 

 

 

T I T T
0 2 4 6 8 1 0 12 14
time (hours)

Figure 3.24 Chlorobenzene (CB) transformation plot for laboratory incubations of
sediment slurry from the Pharmacia/Upjohn Pond in the presence or absence
of sulfate reduction inhibitors, Experiment #5.

73

 

 

 

 

 

 

 

 

 

 

 

35k
30 ‘ ~ ii
13 ,
25‘ T‘
20‘ _ I UNBLOCKED
i 1 II
3 . £1 BLOCKED
00 4
”15‘ t BLANK
; "F 4L KILL
10‘
;
54
I
0 I ‘ITITI‘I'IH'IHTITW

 

 

 

0 2 4 6 8 10 1 2 14
time (hours)

Figure 3.25 Toluene (TOL) transformation plot for laboratory incubations of sediment

slurry from the Pharmacia/Upjohn Pond in the presence or absence of sulfate
reduction inhibitors, Experiment #5.

74

hl‘l

3. 5. 7 Secondary Metabolism Experiment #6: Methanol Exclusion

Experiment #6 was designed to test the hypothesis that primary methanol
metabolism supports secondary substrate utilization. Bioﬁlm communities were pre-
incubated 3 weeks in vitro in the presence or absence of methanol as the primary growth
substrate. Assuming that primary metabolism of methanol is linked to secondary
compound transformation, continuous exposure to methanol would hypothetically induce
or enhance enzyme systems required for metabolism of secondary non-growth substrates.
Conversely, bioﬁlms growing in the absence of methanol would hypothetically express
these secondary enzyme systems to a lesser degree. Subsequent secondary-level
incubations (as described in previous experiments) following the pre-incubation period
facilitated detection of any relationship between primary methanol metabolism and
induction Of secondary enzyme systems.

Bioﬁlms from DARK treatments only (4 replications) were harvested and slurries
prepared as in previous experiments. Subsamples Of the pre-incubation slurry were
collected for direct microscopic count and FAME analysis. Aliquots (100-mL) of bioﬁlm
slurry were transferred to sterile 500-m1 sidearm Erlenmeyer ﬂasks in a continuous ﬂow
in vitro incubation system consisting of two treatments: 100 mg L" methanol (designated
+MeOH) and 0 mg L" methanol (designated -MeOH). Each treatment consisted Of a
total of 6 ﬂasks: 5 ﬂasks with biologically active bioﬁlms (designated BIOFILM) and 1
groundwater reagent control ﬂask (designated GW).

The majority of cooling water entering the Pond originates from groundwater on

the Pharmacia/Upjohn site. The development Of microbial communities in the Pond is

75

likely to have been impacted by continuous inoculation with groundwater
microorgansisms. Furthermore, the inorganic chemistry Of this groundwater is
comparable to the Pond water, but the groundwater is free of methanol. For these
reasons, groundwater from the Pharmacia/Upjohn site was deemed the most appropriate
media for continuous ﬂow incubations. The basal media for the in vitro system consisted
of unﬁltered groundwater collected directly from Pharmacia/Upjohn wellhead #23, a
major source Of cooling water for the pharmaceutical plant.

A diagram of the in vitro system is presented in Figure 3.26. Fresh basal media
was added twice daily to each Of 2 bulk media tanks, one for each laboratory treatment.
The bulk tank for the +MeOH treatment was treated with methanol following each
infusion Of fresh groundwater to achieve a ﬁnal concentration of 100 mg methanol L".
Media was delivered to each treatment ﬂask by Tee] 1/170 hp magnetic drive pumps
(Dayton Electric Manufacturing Co., Chicago, IL) and a manual screw-clamp ﬂow-
control system. Outﬂow from each ﬂask was regulated by gravity overﬂow through the
ﬂask sidearm. Flow rates were controlled at 3.33 L ﬂask] day", which equates to a
residence time in each ﬂask of 3.6 hours. The low residence time ensured that those
bioﬁlm communities receiving methanol did not deplete the carbon resources in any
given residence cycle.

Following a 3-week incubation, bioﬁlms were destructively subsampled for direct
microscopic counts, FAMEs, and secondary metabolism incubations according to the
protocols described for previous experiments. One ﬂask per treatment was sacriﬁced for

an autoclave-sterilized biological control (designated KILL) in the secondary incubations.

76

Reagent control vials (designated RBLK) for secondary incubations were also prepared
for each treatment.

All secondary incubation serum vials were spiked with a water-based stock
substrate solution to achieve the initial equilibrium concentrations indicated in
parentheses for each of the following compounds: DCM (60 pg L"), TOL (25 pg LI),
and CB (35 pg L"). Methanol was added as a neat solution at a ﬁnal concentration Of 100
mg L" only to those bioﬁlms that had previously received methanol. Incubations were
conducted as in previous experiments with subsampling at 0.5, 5.0, 10.0, 15.0, 20.0, and
25.0 hours.

Direct microscopic counts prior to incubation were comparable to previous
DARK bioﬁlms from the Pond, expectedly increasing after pre-incubation with methanol
and decreasing after methanol starvation. These differences in post-incubation direct
microscopic counts were statistically signiﬁcant.

A methanol transformation plot for the +MeOH bioﬁlms is presented in ﬁgure
3.27. Methanol transformation rates were comparable to and slightly higher than those
Observed with DARK bioﬁlms in previous experiments. The groundwater-only bioﬁlms
(GW) also transformed methanol, but at about 20% Of the rate Observed with DARK
bioﬁlms.

Secondary transformation plots are presented in Figures 3.28-3.30. Although
DCM transformation (Figure 3.28) had never been Observed in previous aerobic
experiments, the +MeOH samples exhibited rapid DCM transformation and at a highly

signiﬁcantly greater rate than that Observed in the -MeOH samples.

77

The DCM transformation plot for +MeOH bioﬁlms exhibits a shape classically
associated with ﬁrst order kinetics. However, closer examination of the data reveals that
the deﬂection point in the curve at 15 hours coincides with depletion of the primary
substrate methanol. Therefore, conclusive determination of zero- versus ﬁrst-order
secondary kinetics would only have been possible by mid-incubation replenishment Of the
primary substrate and subsequent continued monitoring of secondary transformation.
Because primary substrate was not replenished and the order of kinetics for secondary
transformation could not be deciphered, calculation of DCM transformation rates was
based strictly on the zero-order portion of the curve.

CB (Figure 3.29) and TOL (Figure 3.30) were rapidly transformed in both
treatments and did not show signiﬁcant differences between treatments. Neither
transformation plot shows a deﬂection at the time point coinciding with methanol
depletion, suggesting that transformations Of CB and TOL are not directly associated with

primary methanol metabolism.

78

 

 

 

 

GW GW

without with
Methanol 100 ppm
Methanol

 

 

 

 

 

 

magnetic-drive
pumps

 

 

 

 

 

 

valve assembly valve assembly

111111 111111
incubation ﬂasks incubationﬂasks
|| | | || 111111

overﬂow waste overﬂow waste

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 3.26 Diagram Of in vitro system for pre-incubation Of attached microbial
communities from the Pharmacia/Upjohn Pond, Experiment #6.

79

120

 

 

 

 

 

 

 

 

 

100
1 '1" 1-
804
g .
0
g _
a 60
O
E _
d) .1
a I .
4o
4 .
.I
20 .
:i \\
o . , 41%.. y y _
0 10 15 20 25

time (hours)

 

 

- I BIOFILM

4

i_

4.1..

G WATER
RBLK
KILL

 

Figure 3.27 Methanol transformation plot for laboratory incubations Of attached

80

microbial communities developed under dark conditions in the
Pharmacianpjohn Pond and pre-incubated for 3 weeks in the presence or
absence Of methanol, Experiment #6 (12/18/95).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.101
{111 = I (+)BIOFILM
. C1 _
4c ( )BIOFILM
: O (+)GW
. O - w
go + (+)RBLK
- —A— (-)RBLK
20
I —-I-— (+) KILL
103 + 4+— () KILL
: i I l
0 TI ' I' T I I I ' T
5 10 15 20 25

time (hours)

Figure 3.28 Dichloromethane (DCM) transformation plot for laboratory incubations of

attached microbial communities developed under dark conditions in the
Pharmacia/Upj ohn Pond and pre-incubated for 3 weeks in the presence or
absence Of methanol, Experiment #6 (12/18/95).

81

 

 

 

 

(+) BIOFILM

 

(-) BIOFILM

 

(+) GW

 

(-) GW

 

 

(+) RBLK

 

(-) RBLK
(+) KILL
(-) KILL

 

H
o
i

++++OODI

 

 

 

 

LII

 

 

 

01111
0 5 10 15 20 25
time(hours)

Figure 3.29 Chlorobenzene (CB) transformation plot for laboratory incubations of
attached microbial communities developed under dark conditions in the
Pharmacia/Upjohn Pond and pre-incubated for 3 weeks in the presence or
absence of methanol. Experiment #6 (12/18/95).

82

 

200

 

 

180

 

160‘

1.1 4 g,

 

140‘

J_J.L

 

 

120,

J LIGHT

 

8100:
00 1

—Ar-RB

 

80‘

.1

+ KILL

 

 

 

 

60*

 

40‘

_I

 

 

‘1

 

20:

—I

 

TTIITTTITTTTITTTTITTTTITTTTIITTTITTTTITTrTITTTI:

 

0

0

51015 20 25 30 35 40 45 50
time(hours)

Figure 3.30 Toluene (TOL) transformation plot for laboratory incubations Of attached

microbial communities developed under dark conditions in the
Pharmacia/Upj ohn Pond and pre-incubated for 3 weeks in the presence or
absence of methanol, Experiment #6 (12/18/95).

83

3.5.8 Secondary Metabolism Results Summary

 

 

 

 

 

 

 

 

 

Table 3.1 Summary Of direct microscopic count data for all secondary metabolism
experiments with attached microbial communities from the
Pharmacia/Upjohn Pond.

Expt. # Treatment cells ml" Standard Error

1 Light 1.15x108 2.61 x107

Dark 7.20 x 107 2.72 x 107
2 Light no data no
Dark no data no

3 Light 1.45 x 108 1.62 x 107
Dark 2.65 x 108 2.07 x 107

4 Light 5.57x108 1.14x108
Dark 2.84 x 108 5.85 x 107

5 (sed) 8042' Block no data no
Unblocked no data no

6 pre-incubation 9.43 x 107 8.03 x 106
post-inc (+MeOH) 3.18 x 108 1.62 x 107
post-inc (- MeOH) 5.94 x 107 1.11 x 107

mean Light 2.72 x108 1 1.79 x108
Dark 1.79 x 108 6.42 x 107

 

 

 

 

84

 

Table 3.2 Summary Of AN OVA statistics and interpretation for direct microscopic
counts for all secondary metabolism experiments with attached microbial
communities from the Pharmacia/Upjohn Pond.

 

 

 

 

 

 

 

 

Expt. # Effect Test ANOVA p-value Interpretation

1 Light/Dark 0.2369 Light = Dark

2 no data na na

3 Light/Dark 0.0018 Light < Dark

4 Light/Dark 0.0502 Light > Dark

5 (sed) no data na na

6 +MeOH/-MeOH 0.0001 + MeOH > - MeOH

 

 

 

 

 

85

Table 3.3

Summary of methanol transformation rates for all secondary metabolism

experiments with attached microbial communities from the
Pharmacia/Upjohn Pond.

 

 

 

 

 

 

 

 

 

mg MeOH

Expt. # Treatment (1x106 cells)'l hr" Standard Error

1 Light 1.70 x 10'5 1.29 x 10'5
Dark 1.06 x 10" 8.98 x 10'5

2 Light 2.80 x 10-5 6.57 x 10*
Dark 3.70x10'5 1.13 x10'5

3 Light 2.45 x 10'5 6.44 x 10'6
Dark 1.94 x 10'5 2.80 x 10‘6

4 Light (+ MeOH) 7.36 x 10* 2.08 x 10'6
Dark (+ MeOH) 3.69 x 10'5 1.19 x 10'5
Light (NO MeOH) na na
Dark (NO MeOH) na na

5 (sed) SO42' Block no data no
Unblocked no data no

6 +MeOH 5.14x10“1 6.04x10'5
- MeOH na no

mean Light 1.92 x 10'5 5.28 x 10‘6
Dark 4.73 x 10'5 2.30 x 10'5

 

 

 

 

 

86

 

Table 3.4 Summary of AN COVA statistics and interpretation for methanol
transformation rates for all secondary metabolism experiments with attached
microbial communities from the Pharmacia/Upj ohn Pond.

 

 

 

 

 

 

 

 

Expt. # Effect ANCOVA p-value Interpretation
1 Light/Dark 0.3308 NO Difference
2 Light/Dark 0.0705 NO Difference
3 Light/Dark 0.1062 NO Difference
4 Light/Dark 0.1563 NO Difference
5 (sed) 8042' Block no data no

6 +/- MeOH no no

 

 

 

 

 

87

111

E1]

ils

"T /

 

—o-
c__/)

 

 

 

 

 

 

 

 

Table 3.5 Summary of dichloromethane transformation rates for all secondary
metabolism experiments with attached microbial communities from the
Pharmacia/Upjohn Pond.

pg DCM

Expt. # Treatment (1x106 cells)" hr" Standard Error

1 Light no transformation na
Dark no transformation na

2 Light no transformation na
Dark no transformation na

3 Light no transformation na
Dark no transformation na

4 Light (+ MeOH) no transformation na
Dark (+ MeOH) no transformation na
Light (NO MeOH) no transformation na
Dark (No MeOH) no transformation na

5 (sed) 8042' Block 9.55 x 10'3 I 4.71 x 10'3
Unblocked 5.63 x 10'2 1' 5.66 x 10'3

6 +MeOH 1.17x10‘S 1.25x10'6
No MeOH 6.72 x 10'6 2.95 x 10'6

 

 

 

 

‘1' Sediment transformation rates expressed as pg DCM (g dry weight)" hr"

88

 

Table 3.6 Summary Of ANCOVA statistics and interpretation for dichloromethane
transformation rates for all secondary metabolism experiments with attached
microbial communities from the Pharmacia/Upjohn Pond.

 

 

 

 

 

 

 

 

Expt. # Effect ANCOVA p-value Interpretation

1 Light/Dark no transformation na

2 Light/Dark no transformation na

3 Light/Dark no transformation na

4 Light/Dark no transformation na

+/- MeOH no transformation na

5 (sed) SO42' Block 0.0020 unblocked > blocked
(link to SO42' reduction)

6 +/- MeOH 0.0003 + MeOH > - MeOH
(link to MeOH metab.)

 

 

 

 

 

89

 

 

 

 

 

 

 

 

 

 

 

 

Table 3.7 Summary Of toluene transformation rates for all secondary metabolism
experiments with attached microbial communities from the
Pharmacia/Upjohn Pond.

pg TOL

Expt. # Treatment (1x106 cells)" hr" Standard Error
1 Light 3.40 x 10'5 1.41 x 10'5
Dark 1.41x10" 1.22x10"
2 Light 5.81 x 10'5 1.24 x 10'5
Dark 1.42 x 10" 1.28 x 10"
3 Light 8.42 x 10"6 2.47 x 10*
Dark 6.62 x 10'6 3.30 x 10'7
4 Light (+ MeOH) 6.57 x 10'7 1.75 x 10‘7
Dark (+ MeOH) 6.21 x 10'6 3.24 x 10'7
Light (NO MeOH) 1.32 x 10'6 3.46 x 10'7
Dark (No MeOH) 7.13 x 10"5 5.42 x 10'7
5 (sed) 8042' Block 1.76 x 10'3 f 5.01 x 10"
Unblocked 9.90 x 10'3 ‘1 6.08 x 10"
6 + MeOH 1.38 x 10"3 2.26 x 10'7
- MeOH 7.45 x 10'6 1.07 x 10"5
mean Light 2.05 x 10'5 1.25 x 10‘5
U Dark 6.06 x 10'5 3.69 x 10'5

 

1' Sediment transformation rates expressed as pg TOL (g dry weight)" hr"

90

 

Table 3.8 Summary of ANCOVA statistics and interpretation for toluene
transformation rates for all secondary metabolism experiments with attached
microbial communities from the Pharmacia/Upjohn Pond.

 

 

 

 

 

 

 

 

Expt. # Effect ANCOVA p-value Interpretation
1 Light/Dark 0.6386 NO Difference
2 Light/Dark 0.1653 NO Difference
3 Light/Dark 0.1022 NO Difference
4 Light/Dark 0.0001 Light < Dark
+/- MeOH 0.4984 NO Difference
5 (sed) SO42' Block 0.0007 unblocked > blocked
(link to SO42' reduction)
6 +/- MeOH 0.8249 No Difference

 

 

 

 

 

91

 

Table 3.9 Summary Of chlorobenzene transformation rates for all secondary

metabolism experiments with attached microbial communities from the

 

 

 

 

 

 

 

 

 

Pharmacia/Upjohn Pond.
ugCB
Expt. # Treatment (1x106 cells)" hr" Standard Error
1 Light na na
Dark na na
2 Light na na
Dark na na
3 Light 7.90 x 10'6 3.76 x 10'6
Dark 1.09 x 10'5 6.91 x 10‘7
4 Light (+ MeOH) 1.28 x 10'7 8.30 x 10‘8
Dark (+ MeOH) 2.77 x 10‘6 5.29 x 10'7
Light (NO MeOH) 5.94 x 10‘7 2.62 x 10‘6
Dark (NO MeOH) 4.17x10'6 1.10x10'6
5 8042' Block no transformation na
Unblocked no transformation na
6 + MeOH 1.45 x 10'6 6.48 x 10'7
No MeOH 4.00x10"3 1.10x10’6
mean Light 2.87 x 10'6 2.18 x 10"6
Dark 5.95 x 10'6 2.17 x 10"5

 

 

 

 

92

 

 

Table 3.10 Summary Of ANCOVA statistics and interpretation for chlorobenzene
transformation rates for all secondary metabolism experiments with attached
microbial communities from the Pharmacia/Upjohn Pond.

 

 

 

 

 

 

 

 

Expt. # Effect ANCOVA p-value Interpretation

1 Li ght/Dark na na

2 Light/Dark na na

3 Light/Dark 0.0202 Light < Dark

4 Light/Dark 0.0054 Light < Dark
+/- MeOH 0.6144 No Difference

5 (sed) 8042' Block no transformation na

6 +/- MeOH 0.2009 No Difference

 

 

 

 

 

93

 

3.6 Microbial Community Structure (FAME Analysis)

Microbial community structure was analyzed for secondary metabolism
experiments #2 through #4 according to the FAME technique described in section 2.2.
Resulting FAME biplots (see section 2.2.7) are presented in Figures 3.31-3.33. LIGHT
and DARK treatments are clearly well-resolved in all experiments, with approximately
80% Of variance explained in the ﬁrst 3 dimensions in each experiment and a low degree
of variability between replications. Resolution of treatments is based primarily on fatty
acids of known bacterial origin; i.e. the heterotrophic portion Of the microbial community.

Interpretation Of which fatty acids are most important in each treatment is more
difﬁcult. Temporal variation was large and no consistent or distinguishable patterns are
evident in degree Of saturation or chain-length with respect to treatment or bioﬁlm
developmental state. The single exception may be longer chain fatty acids, which tend to
be found in greater abundance in the LIGHT treatment, expectedly indicating a stronger

presence Of eukaryotes.

94

 

 

 

 

 

 

 

0
t 0
1111) 63
D L
PC2 O D @ W L
9
D
-5 4 L T
-5 0
PC1
D Dark 6 iso 17:1 at 5
L Light (3 ’15:0
0 18:1 cis 9 (w 9) 0 Summed Feature 9
9 16:0 anteiso 6) 15:0 iso
6) '14:0 Q) 22:1 cis 13 (w 9)
0 18:0 iso (ﬂ) '18:0

 

 

 

 

Figure 3.31 Fatty acid methyl ester (FAME) biplot for attached microbial communities
from two light treatment regimes in the Pharmacia/Upjohn Pond,
Experiment #2.

95

 

 

 

 

 

 

(B
0
L
L |_ 0 D
PC2 0 0 ﬁ
111 80
8 L
-4
-4 0
PC1
D Dark 6 iso 17:1 at 5
L Light (3 19:0 anteiso
0 '21 :0 0 '24:0
9 22:1cisl3(w9) 6) 15:1 isoG
6) 20:2 1 1,14 (w 6) G) Summed Feature 3
0 16:0 anteiso (E) 10:0 30H

 

 

Figure 3.32 Fatty acid methyl ester (FAME) biplot for attached microbial communities
from two light treatment regimes in the Pharmacia/Upjohn Pond,

Experiment #3.

96

 

 

 

 

 

 

 

 

 

 

[3
1D D D
O 110 L
PC2 0 0 a a '1:
9 L
G)
(B
D
-3 Q
-3 0 3
PC1
D Dark 6 '14:0
L Light (9 18:1 cis 9 (w 9)
0 16:0 anteiso 0 16:1 cis 11 (w 5)
9 22:1 cis 13 (w 9) 6) Summed Feature 3
6) Summed Feature 9 6) 10:0 30H
0 18:01so ([0 15:0 iso

 

 

 

Figure 3.33 Fatty acid methyl ester (FAME) biplot for attached microbial communities
from two light treatment regimes in the Pharmacia/Upjohn Pond,
Experiment #4.

97

 

3.7 Discussion

3. 7.1 Microbial Community Structure

Results for direct microscopic counts in the light manipulation experiment were
highly temporally variable, with statistical interpretation ranging the extremes: no
difference between treatments, DARK signiﬁcantly more populated than LIGHT, and
LIGHT signiﬁcantly more populated than DARK. Examining the trends in the data from
experiments #1 through #4 reveals that direct microscopic counts increase in both
treatments with increasing time of biOﬁlm development. The heterotrophic populations
of both bioﬁlms increased at an equal rate during early development. As the bioﬁlms
matured through early summer, the DARK bacterial populations increased more quickly
than the LIGHT populations. As the summer light intensity increased and temperatures
became elevated, the productivity in the LIGHT bioﬁlms resulted in rapid increases in
heterotrophic population numbers.

Although the trends in the data suggest that sessile primary productivity increases
heterotrophic population numbers, the effect is likely seasonal and dependent upon
biOﬁlm developmental state. Thus, the overall interpretation on an experiment-wide basis
would be a lack Of signiﬁcant difference between treatments, and no support for the
hypothesis that sessile primary productivity stimulates heterotrophic population numbers.

Even though the treatments did not vary signiﬁcantly in overall bacterial numbers,
the FAME analyses demonstrated that the heterotrophic communities are indeed

structurally or phenotypically different between treatments. Although FAME analysis is

98

 

Often associated with community structure, interpretation is actually based on a set of
phenotypic markers. These phenotypic markers are produced as part of the basic
physiology of the microbial species in a given sample. Differences between whole-
community microbial FAME proﬁles may be due to either a true structural difference in
species distribution or a difference in physiology among similar species distributions, or
some combination of the two. Therein lies a paradox in FAME interpretation; purely
structural differences are confounded by physiological differences induced by the very
treatments being resolved.

Therefore, the strong resolution between LIGHT and DARK treatments may be
due to differences in species distribution, physiology, or both. The secondary metabolism
studies demonstrated physiological differences in bioremediation activity between LIGHT
and DARK bioﬁlm communities, supporting the hypothesis that sessile primary

productivity inﬂuences bioremediation activity (physiology). However, differences in
eLlkzaryotic species distribution (the presence Of algae and diatoms in the LIGHT
cOl'l'lrnunities) were also veriﬁed by visual microscopic Observation and demonstrated by
the strong weighting of several eukaryotic marker acids in the LIGHT samples. Several
heaVily weighted fatty acids are known prokaryotic markers as well, but interpretation is
confounded regarding whether these differences among heterotrophs are due to changes
in physiology or species distribution. Whole-community FAME interpretation is
inherently restricted to concluding that either the communities are different or similar,
necessitating the use Of complementary physiological, microscopic, and genetic analyses

to draw more detailed conclusions.

99

3. 7.2 Primary Methanol Metabolism

LIGHT and DARK communities were not signiﬁcantly different in their ability to
oxidize methanol in the ”C-methanol experiment or in any of the four non-radiolabelled
methanol metabolism studies. Because the communities were shown to be structurally
different by FAME analysis, one conclusion is that some degree Of functional redundancy
exists with respect to methanol oxidation. Since the prevailing contaminant loading
conditions dictate that methanol comprises a signiﬁcant fraction of total DOC in the
Pond, ﬁlnctional redundancy would be expected for methanol oxidation activity.

A wide range Of cosmopolitan and specialized methanol degraders are reported in
the literature (see Table 1.1) and would be expected to occur in a heterogeneous system
Such as the Pharmacia/Upjohn Pond. Functional redundancy is thought to be a stabilizing
fstature among microbial communities, in this case ensuring that methanol-oxidizing
populations are present under a host Of temporally variable environmental conditions.
The current study resulted in isolation of several different types of methanol-oxidizing
baeterial isolates from epiphytic bioﬁlms in the Pond, demonstrated a signiﬁcantly greater
abundance Of methylotrophic organisms in these bioﬁlms, and recorded intense methanol
o"‘iiClation activity in the bioﬁlms and sediment. These results collectively suggest that

methanol is a dominant primary growth substrate.

100

 

(r.

3. 7.3 Secondary Metabolism

Bioﬁlms from both light manipulation treatments rapidly transformed the
aromatic secondary compounds toluene (TOL) and chlorobenzene (CB). Because several
cosmopolitan organisms are known to metabolize TOL, secondary transformation activity
was expected in both light treatments, but the original hypothesis predicted that sessile
primary productivity would stimulate heterotrophic activity and result in more rapid
secondary transformation of TOL in the LIGHT treatment. Contrary to the original
hypothesis, no statistically signiﬁcant differences were detected between LIGHT and
DARK treatments for secondary transformation of TOL, leading to several possible
conclusions.

One or more cosmopolitan microbial species may be ubiquitously and evenly
distributed between treatments and functioning equally well for TOL transformation in
either light regime, in which case the differences detected by FAME proﬁles are due to
some other fraction of the community. Altemately, the communities in each treatment
may have different microbial species distributions, but the various species are functioning
equally in total, in which case the differences detected by FAME proﬁles are due in part
to differences in structure among the TOL-oxidizing communities. In either case,
deﬁnitive conclusions cannot be made without additional research, perhaps involving the
use of genetic probes for different groups of TOL oxidizers.

CB transformation was consistently and signiﬁcantly higher in the DARK
communities, again contradicting the hypothesis that sessile primary productivity

promotes bioremediation activity. Given that halogenation generally renders a compound

101

 

more recalictrant, only specialized microbial populations have the enzymatic capability tO
compensate for the chlorO-substituent on the aromatic ring. Generalist heterotrophs
stimulated by sessile primary productivity in the LIGHT treatments may be less likely to
express energetically unfavorable dechlorination activity in lieu of large pools of readily
labile carbon exudates. Conversely, microorganisms growing in the DARK treatment
without the luxury Of labile carbon exudates may be forced to express more exotic
enzymatic pathways to extract energy for growth from less labile substrates, such as CB.

Three aliphatic secondary compounds known to sporadically appear in the pond
inﬂuent (dichloromethane [DCM], 1,2-dichloroethane [DCA], and
bromodichloromethane [BDCM]) were not transformed under aerobic conditions during
the course of the initial studies. In fact, DCA could potentially present a surface or
groundwater quality problem due to a complete lack Of aerobic transformation and only
very slow anaerobic transformation. BDCM was not transformed, but its sorption afﬁnity
to sediments and bioﬁlm matrixes made analyses problematic. Further experimentation is
warranted to determine the sorption afﬁnities for BDCM, the adsorption capacities Of the
matrixes, and the availability Of adsorbed BDCM. Sorption of organics is thought to
increase their availability to microorganisms and lead to accumulation in the matrixes,
which can elevate concentrations from the level Of secondary substrates to above the
primary substrate threshhold.

The lack of secondary aerobic DCM transformation was unexpected, because
some methylotrophs are known to express dichloromethane dehalogenase acitivity. This

enzyme dechlorinates DCM to form formaldehyde, which is then shunted directly into the

102

methanol oxidation and assimilation pathways. DCM transformation was only Observed

in pre-incubated DARK bioﬁlms that had been exposed to constant methanol

concentrations Of 100 mg L" for a 3-week period. Several explanations are possible for
this phenomenon, but selection or induction are the most likely scenarios.

The selective pressure of the laboratory incubation conditions may have increased
the numbers Of a speciﬁc population to a level where activity was measureable.
Altemately, the constant inﬂux of methanol could have induced auxiliary enzymatic
pathways that are not normally expressed under the pulse-loading conditions Of the
Pharmacia/Upjohn Pond, resulting in a functional shift in the microbial community rather
than a structural shift. Poor FAME data for the methanol exculsion experiment precluded
us from conclusively determining whether the cause of the induced DCM activity was
due to structural changes in the communities during incubation. Further explanation of
the relationship between primary methanol metabolism and secondary DCM

transformation is given in section 3.7.5

3. 7.4 Anaerobic Metabolism

The energetically preferred order of electron acceptors for microbial respiration is
oxygen, nitrate, iron/manganese, sulfate, and carbon dioxide. Oxygen is depleted rapidly
at or above the sediment water interface in the Pharmacianpjohn Pond, necessitating the
use of alternate electron acceptors for microbial respiration in the surﬁcial sediments.
Nitrate concentrations in the Pond inﬂuent are too low (less than 10 pg L") to contribute
substantially to the pool Of alternate electron acceptors. Iron concentrations are elevated,

but the fraction Of iron in the Pond that is microbially reducible has not been conclusively

103

 

determined and appears to be relatively low. Thus, the availability of alternate electron

acceptors in the sediment is restricted primarily to sulfate.

Concentrations of sulfate in the Pond inﬂuent (60 mg L") are sufﬁcient to
contribute signiﬁcantly to the pool Of alternate electron acceptors in the Pond. The high
volume of ﬂow through the system constantly replenishes the sulfate pool and results in
advective ﬂow Of sulfate into the sediments via high inﬁltration rates. New research also
suggests that sulfates may be regenerated in sediments by oxidation Of reduced sulﬁdes,
further contributing to the available sulfate pool. Iron is hypothesized to synergistically
contribute to the rate Of sulfate reduction via the intermediary binding Of iron to reduced
sulﬁdes, pulling equilibrium towards sulfate reduction. The elevated iron concentrations
in the Pond and the release of hydrogen sulﬁde from sediments treated with 6 N
hydrochloric acid both provide evidence of this intermediary binding phenomenon. For
these reasons, sulfate is hypothesized tO be the principal alternate electron accepting
process in the Pharmacia/Upjohn Pond sediments. Previous research has conclusively
validated this hypothesis for anaerobic oxidation of methanol (Klug et al., 1996).

Methanol, dichloromethane (DCM) and toluene (TOL) were rapidly transformed
anaerobically in the present studies. Anaerobic secondary transformation of DCM and
TOL was signiﬁcantly linked to sulfate reduction, suggesting that secondary metabolic
processes may also be governed by sulfate reduction in the sediments. Very slow
anaerobic transformation was noted for 1,2-dichloroethane (DCA) and no activity was
observed for chlorobenzene (CB). Sorption to sediment made analysis of

bromodichloromethane (BDCM) problematic. The aerobic and anaerobic biOﬁlms act in

104

 

concert to provide complete transformation of the methanol, DCM, CB, and TOL

entering the system.

3. 7.5 Link Between Methanol and Secondary Metabolism

Methanol is a principal component Of inﬂuent DOC in the Pharmacia/Upjohn
Pond system. Evidence from previous research (Klug et al., 1996) and the present studies
suggests that methanol is an important primary growth substrate for aerobic bioﬁlms in
the Pond. A key hypothesis was that primary metabolism of methanol was sustaining
microbial growth in the bioﬁlms and driving secondary substrate utilization. The
inability to detect a link between primary methanol metabolism and secondary substrate
utilization in experiment #4 was likely due to the confounding inﬂuence of in situ enzyme
expression; i.e. active bioﬁlms continued to express secondary metabolic pathways even
after methanol was removed in the laboratory. Thus, the methanol exclusion pre-
incubation experiment was devised to force deactivation and/or induction of methanol-
linked enzymatic pathways.

The results from experiment #6 support the hypothesis Of primary methanol
oxidation driving secondary transformation of TOL and DCM. The induction of DCM
activity by pre-incubating with a constant source Of methanol suggests that the sporadic
nature of methanol loading in the Pond may negatively impact DCM secondary activity.
A constant source Of methanol may have led to preferable population selection, induction
of desired secondary metabolic pathways, or a combination of both. Unfortunately, poor
FAME data prevented a decisive conclusion regarding whether the induced DCM activity

was associated with a shift in bacterial species distribution. Again, genetic probes may be

105

 

a more valuable tool for detecting changes in microbial community structure, particularly
in the case Of methylotrophs, for which an extensive library Of probes has been
developed.

Secondary transformation of CB was not shown to be linked tO methanol
oxidation, indicating that either an unrelated (non-methylotrophic) population is
responsible for the activity or that the enzymatic pathways for CB transformation are

constituitively induced in cosmopolitan heterotrophs also capable of growth on methanol.

3.8 Summary and Conclusions

The Oligotrophic comparison study demonstrated that the basic inorganic water
chemistry of the Pharmacia/Upjohn Pond is comparable to other natural Oligotrophic lake
systems in the region. The surface water outlet of the Pond system seems to be
converging chemically and biologically toward natural lake systems. The high proportion
of methanol in the inﬂuent DOC has impacted microbial community structure Of
epiphytic bioﬁlms near the pond inlet by increasing methylotrophic organisms by more
than two orders of magnitude above natural lake systems. However, FAME proﬁles
revealed that the specialized epiphytic bioﬁlms near the Pond inlet are similar in overall
microbial community structure to those bioﬁlms found in natural oligotrophic lakes.

High methanol oxidation rates were found in aerobic plant-associated bioﬁlms
and in anaerobic sediments. Methanol metabolism in sediments was also linked to sulfate
reduction. The present study strengthened the rate estimates for methanol metabolism
(Klug et al., 1996), surveyed the extent of secondary substrate transformation, examined

the link between primary methanol utilization and secondary transformation, developed

106

 

rate estimates for secondary transformations under aerobic and anaerobic conditions, and
explored the feasibility of artiﬁcially augmenting surface area for attached microbial
growth.

Bioﬁlms that developed in the absence of light on artiﬁcial surfaces were equally
adept at transforming methanol and several secondary substrates (dichloromethane,
toluene, and chlorobenzene) as those bioﬁlms grown under natural diurnal light. DARK
bioﬁlms exhibited signiﬁcantly higher secondary transformation rates for chlorobenzene.
These results suggest that surface area augmentation, in which most interior surfaces are
devoid of light, may be a feasible option for improving the bioremediation capacity of the
system.

Dichloromethane was aerobically transformed only after a 3-week pre-incubation
with methanol, either due to stimulation of enzymatic pathways or selection of desired
microbial populations. Primary methanol metabolism was signiﬁcantly linked to
secondary transformation of dichloromethane, but did not directly impact secondary
transformation Of toluene or chlorobenzene. Secondary transformation of
dichloromethane gradually diminished upon depletion of the primary substrate. Future
experiments should examine this relationship in greater detail through a series of
methanol additions and depletions to determine if the pattern is repeatable.

Direct microscopic counts of bacterial numbers varied seasonally and with bioﬁlm
developmental state, but were not signiﬁcantly different overall between treatments.
Microbial community structure as determined by fatty acid methyl ester (FAME) proﬁles

was well-resolved between treatments, but the precise nature Of the differences remains to

107

 

be determined. More detailed microbial community structural analysis is warranted,
given the lack of sensitivity and inconclusive results with FAME proﬁles. Genetic probes
may provide greater insights into the various types of methylotrophic organisms
responsible for primary and secondary bioremediation activity in the system.

Previous and current data collectively suggest that the Pharmacia/Upjohn Pond
system is a feasible site for in situ bioremediation. Attached microbial communities
rapidly metabolize methanol and several secondary contaminants in the cooling water
discharge to the pond. Bioﬁlm communities near the pond inlet are clearly well-adapted
to primary methanol metabolism, as evidenced by signiﬁcantly elevated numbers of
methylotrophic organisms and rapid methanol oxidation rates. Secondary transformation
has been linked to primary methanol metabolism, suggesting that a continuous inﬂux Of
methanol at part-per-million levels may actually be beneﬁcial to the system by
maintaining elevated numbers of adapted organisms while simultaneously stimulating
secondary enzyme activity. Anaerobic pond foster extensive attached methylotrophic
populations that further contribute to overall system transformation capacity. Anaerobic
dechlorination activity may synergistically promote further aerobic transformation of
speciﬁc halogenated secondary compounds such as CB among spatially heterogeneous
bioﬁlm and sediment communities. The capability of methylotrophic populations to
function effectively under non-oxygenic conditions on plastic surfaces suggests that
artiﬁcial surface area augmentation may be feasible as a means tO increase system
capacity.

Future research to further parameterize the system would be a critical step toward

developing a mathematical dispersion model. Considerations in such a model would be

108

 

seasonal variation in temperature and oxygen, mass transfer to air and sediment,
advective forces, and veriﬁcation Of the kinetic order and rates Of metabolic reactions.
When a complete array of system parameters has been established, elemental mass
balance equations would allow prediction Of system capacities under virtually any loading
and environmental scenario. Only then would more deﬁnitive conclusions regarding

system efﬁcacy be possible.

109

 

REFERENCES

110

REFERENCES

Acea, M. J ., and M. Alexander. 1988. Growth and survival Of bacteria introduced into
carbon-amended soils. Soil Biology and Biochemistry 20(4):703-709.

Aitchison, J. 1986. The Statistical Analysis of Compositional Data, Chapman and Hall,
New York.

Alexander, M. 1965. Biodegradation: Problems Of molecular recalcitrance and microbial
fallibility. Advances in Applied Microbiology 7:35-80.

Alexander, M. 1981. Biodegradation of chemicals of environmental concern. Science
211:132-138.

Aloi, J. E. 1990. A critical review of recent freshwater periphyton ﬁeld methods.
Canadian Journal of Fisheries and Aquatic Sciences 47:656-670.

Alvarez-Cohen, L., P. L. McCarty, E. Boulygina, R. S. Hanson, G. A. Brusseau, and H.
C. Tsien. 1992. Characterization of a methane-utilizing bacterium from a

bacterial consortium that rapidly degrades trichloroethylene and chloroform.
Applied and Environmental Microbiology 58(6): 1 886-1 893.

American Public Health Association. 1992a. 4500 P. Total phosphorus. In Standard
Methods for the Examination of Water and Wastewater, 18th edition, American
Public Health Association, Washington, DC.

American Public Health Association. 1992b. 10300 D. Periphyton/primary productivity.
In Standard Methods for the Examination of Water and Wastewater, 18th edition,
American Public Health Association, Washington, DC.

Amman, P. R., and G. S. Koch. 1993. Technical and economic analyses in the
development of bioremediation processes. Remediation 4(1):115-128.

ASTM. 1987. Standard test method for enumeration Of aquatic bacteria by
epiﬂuorescence microscopy counting procedure. ASTM D4455-85. Annual Book
of ASTM Standards, Vol. 11.02, Water, American Society for Testing and
Materials, Philadelphia, PA.

Atlas, R. M., A. Horowitz, MI. I. Krichevsky, and A. K. Bej. 1991. Response Of

microbial populations to environmental disturbance. Microbial Ecology 22:249-
256.

111

 

Atlas, R. M., and R. Bartha. 1993. Microbial Ecology: Fundamentals and Applications,
3rd edition. The Benjamin/Cummings Publishing Company, Inc., Redwood City.

Bender, J ., P. Phillips, R. Lee, S. Rodriguez-Eaton, G. Saha, B. Loganathan, and L.
Sonnenberg. 1995. Degradation of chlorinated organic compounds by microbial
mats. In Hinchee, R. E., G. D. Sayles, and R. S. Skeen (eds.) Biological Unit
Processes for Hazardous Waste Treatment, Vol. 9, Proceedings: Third
International In Situ and On-Site Bioreclamation and Bioremediation Symposium,
April 1995, San Diego, California. Battelle Press, Columbus.

Bitzi, U., T. Egli, and G. Hamer. 1991. The biodegradation Of mixtures Of organic
solvents by mixed and monocultures of bacteria. Biotechnology and
Bioengineering 37: 1037-1042.

Bouwer, E. J., and P. L. McCarty. 1982. Removal of trace chlorinated organic

compounds by activated carbon and ﬁxed-ﬁlm bacteria. Environmental Science
and Technology 16(12):836-843.

Bouwer, E. J ., and P. L. McCarty. 1983. Transformation of 1- and 2-carbon halogenated
aliphatic organic compounds under methanogenic conditions. Applied and
Environmental Microbiology 45(4): 1286-1294.

Bouwer, E. J ., and P. L. McCarty. 1985. Utilization rates of trace halogenated organic

compounds in acetate-grown bioﬁlms. Biotechnology and Bioengineering
27:1564-1571.

Boyle, M. 1989. The environmental microbiology Of chlorinated aromatic
decomposition. Journal of Environmental Quality 18(4):395-402.

Bratina, B., and R. S. Hanson. 1992. Methylotrophy. In Encyclopedia of Microbiology,
vol. 3, pp. 121-127, Academic Press, Inc.

Breznak, J. A., K. E. Cooksey, F. E. W. Echkardt, Z. Filip, M. Fletcher, R. J. Gibbons, H.
Gude, W. A. Hamilton, T. Hattori, H.-G. Hoppe, A. G. Matthysse, D. C. Savage,
and M. Shilo. 1984. Activity on surfaces, group report. In Microbial Adhesion
and Aggregation, pp. 203 -222, Springer-Verlag, New York.

Brunner, W., D. Staub, and T. Leisinger. 1980. Bacterial degradation of
dichloromethane. Applied and Environmental Microbiology 40(5):950-958.

Cattaneo, A., and M. C. Amireault. 1992. How artiﬁcial are substrata for periphyton?
Journal of the North American Benthological Society ll(2):244-256.

112

 

Cavigelli, M. A., G. P. Robertson, and M. J. Klug. 1995. Fatty acid methyl ester
(FAME) proﬁles as measures Of soil microbial community structure. In Collins,
H. P., G. P. Robertson, and M. J. King (eds.) The Significance and Regulation of
Soil Biodiversity, pp. 99-113, Kluwer Academic Publishers, Netherlands.

Characklis, W. G., G. A. McFeters, and K. C. Marshall. 1990. Physiological ecology in
bioﬁlm systems. In W.G. Characklis and KC. Marshall (eds.), Bioﬁlms, pp. 341-
394, John Wiley and Sons, Inc., New York.

Chaudry, G. S., and S. Chapalamadugu. 1991. Biodegradation of halogenated organic
compounds. Microbiological Reviews 55(1):59-79.

Colby, J ., D. I. Stirling, and H. Dalton. 1977. The soluble methane monooxygenase of
Methylococcus capsulatas (bath). Biochemistry Journal 165: 395-402.

Corpe, W. A. 1985. A method for detecting methylotrophic bacteria on solid surfaces.
Journal of Microbiological Methods 3(3-4):215-223.

Costerton, J. W., and G. G. Geesey. 1979. Microbial contamination of surfaces. In
Mittal, K. L. (ed.) Surface Contamination, pp. 211-221, Plenum Press, New York.

Costerton, J. W., and E. S. Lashen. 1984. The inherent biocide resistance of corrosion-
causing bioﬁlm bacteria. Materials Performance 23(2):]3-16.

Costerton, J. W., T. J. Marrie, and K.-J. Cheng. 1985. Phenomenon of bacterial
adhesion. In Savage, D. C. and M. Fletcher (eds.) Bacterial Adhesion:
Mechanisms and Physiological Signﬁcance, pp. 3-43, Plenum Press, New York.

de Boer, L., G. J. Euverink, J. van der Vlag, and L. Dijkhuizen. 1990. Regulation of
methanol metabolism in the facultative methylotroph Nocardia sp. 239 during

grth on mixed substrates in batch and continuous cultures. Archives of
Microbiology 153 :337-343.

Davis, J. W., and S. S. Madsen. 1991. The biodegradation of methylene chloride in
soils. Environmental Toxicology and Chemistry 10:463-474.

Dunbar, J ., D. C. L. Wong, M. Yarus, and L. J. Fomey. 1996. Unpublished data.

Fathepure, B. Z., and T. M. Vogel. 1991. Complete degradation of polychlorinated
hydrocarbons by a two-stage bioﬁlm reactor. Applied and Environmental
Microbiolog) 57(12):34l 8-3422.

Fletcher, M., and G. D. F loodgate. 1973. An electron-microscopic demonstration of acid
polysaccharide involved in the adhesion of marine bacterium to solid surfaces.
Journal of General Microbiology 74:325-334.

113

Geesey, G. G., R. Mutch, J. W. Costerton, and R. B. Green. 1978. Sessile bacteria: An
important component of the microbial populations in small mountain streams.
Limnology and Oceanography 23(6):1214-1223.

Goldstein, R. M., L. M. Mallory, and M. Alexander. 1985. Reasons for possible failure
of inoculation to enhance biodegradation. Applied and Environmental
Microbiology 50:977-983.

Green, P. N. 1992. Taxonomy of methylotrophic bacteria. Chapter 2 in Murrell, J. C.,
and H. Dalton (eds.) Methane and Methanol Utilizers, Biotechnology Handbooks
Series, vol. 5. Plenum Press, New York.

Haack, S. 1995. Personal communication.

Haack, T. K., and G. A. McFeters. 1982. Nutritional relationships among

microorganisms in an epilithic bioﬁlm community. Microbiology and Ecology
8:115-126.

Hamilton, WA. 1987. Bioﬁlms: Microbial interactions and metabolic activities. In
Fletcher, M., T. R. G. Gray, and J. G. Jones (eds.) Ecology of Microbial
Communities, pp. 361-3 85, Publications for Society for General Microbiology.

Hanson, R. S. 1992. Introduction. Chapter 1 in Murrell, J. C., and H. Dalton (eds.)
Methane and Methanol Utilizers, Biotechnology Handbooks Series, vol. 5.
Plenum Press, New York.

Hanson, R. S. 1980. Ecology and diversity of methylotrophic organisms. Advances in
Applied Microbiology 26:3-39.

Hardman, D. J. 1991. Biotransforrnation of halogenated compounds. Critical Reviews
in Biotechnology 1 1(1): 1-40.

Henson, J. M., M. V. Yates, J. W. Cochran, and D. L. Shackleford. 1988. Microbial
removal of halogenated methanes, ethanes, and ethylenes in an aerobic soil
exposed to methane. FEMS Microbiology Ecology 53:193-201.

Henson, J. M., M. V. Yates, and J. W. Cochran. 1989. Metabolism of chlorinated
methanes, ethanes, and ethylenes by a mixed bacterial culture growing on
methane. Journal of Industrial Microbiology 4:29-35.

Heyer, J. 1977. Results Of enrichment experiments with methane-assimilating organisms
from an ecological point of view. In Skyrabin, G. A., M. B. Ivanov, E. N.
Kondratjeva, G. A. Zavarzin, Y. A. Trotsenko, and A. I. Netrosev (eds.) Microbial
Growth on C1 Compounds, pp. 19-21, USSR Academy of Sciences, Rusching.

114

 

Hinchee, R. E., J. T. Wilson, and D. C. Downey, eds. 1995. Intrinsic Bioremediation,
Vol. 1, Proceedings: Third International In Situ and On-Site Bioreclamation and
Bioremediation Symposium, April 1995, San Diego, California. Battelle Press,
Columbus.

Hutton, W. E., and C. E. Zobell. 1949. The occurrence and characteristics of methane-
oxidizing bacteria in marine sediment. Journal of Bacteriology 58:463-473.

Hwang, H.-M., R. E. Hodson, and D. L Lewis. 1989. Microbial degradation kinetics of
toxic organic chemicals over a wide range of concentrations in natural aquatic
systems. Environmental Toxicology and Chemistry 8:65-74.

Hwang, H.-M, R. E. Hodson, and D. L. Lewis. 1989. Assessing interactions of organic
compounds during biodegradation of complex waste mixtures by naturally

occurring bacterial assemblages. Environmental Toxicology and Chemistry
8:209-214.

Janssen, D. B., R. Oldenhuis, and A. J. van den Wijngaard. 1990. Hydrolytic and
oxidative degradation of chlorinated aliphatic compounds by aerobic
microorganisms. Advances in Applied Biotechnology 4:105-125.

Janssen, D. B., A. Scheper, L. Dijkhuizen, and B. Witholt. 1985. Degradation of
halogenated aliphatic compounds by Xanthobacter autotrophicus GJ 1 0. Applied
and Environmental Microbiology 49(3):673-677.

Jeffrey, W. H., and J. H. Paul. 1986. Activity measurements of planktonic microbial and

microfouling communities in a eutrophic estuary. Applied and Evironmental
Microbiology 51 : 157-162.

Jensen, H. L. 1963. Carbon nutrition of some microorganisms decomposing halogen-
substituted aliphatic acids. Acta Agriculturae Scandinavica 13:404-412.

Jones, H. C., I. L. Roth, and W. M. Sanders III. 1969. Electron microscopic study of a
slime layer. Journal of Bacteriology 99:316-325.

Kevem, N. R., J. L. Wilhm, and G. M. Van Dyne. 1966. Use of artiﬁcial substrata to
estimate the productivity of periphyton. Limnology and Oceanography 11:499-
502.

Klug, M. J., M. G. Kaufman, T. E. Patt, and K. M. Walsh. 1996. Remediation properties
of the Pharmacia/Upjohn Pond. Chapter 3 in Hickey, R. F., and G. Smith (eds.)
Biotechnology in Industrial Waste Treatment and Bioremediation. A collection of
papers presented at the International Symposium on the Implementation of

115

 

Biotechnology in Industrial Waste Treatment and Bioremediation, Sept 15-16,
1992, Grand Rapids, MI. CRC Press, Inc., Boca Raton, FL.

Komogata, K. 1990. Systematics of methanol-utilizing bacteria. FEMS Microbiological
Reviews 87:291-296.

Ladd, T. I., J. W. Costerton, and G. G. Geesey. 1979. Determination of heterotrophic
activity of epilithic microbial populations. In Costerton, J. W. and R. R. Colwell
(eds.) Native Aquatic Bacteria: Enumeration, Activity, and Ecologl, pp. 180-195,
American Society of Testing Materials, Philadelphia.

La Motta, E. J ., and R. F. Hickey. 1990. Factors affecting attachment and development
of biological ﬁlms on solid media. In Smith, E. D., R. D. Miller, and Y. C. Wu
(eds.) Proceedings: First National Symposium/Workshop on Rotating Biological
Contactor Technology, February 4-6, 1980, Champion, Pennsylvania.

Lanzarone, N. A., and P. L. McCarty. 1990. Column studies on methanotrophic

degradation of trichloroethylene and 1,2-dichloroethane. Ground Water
28(6):910-919.

Liu, D., Y. L. Lau, Y. K. Chau, and G. J. Pacepavicius. 1993. Characterization of
bioﬁlm development on artiﬁcial substratum in natural water. Water Research
27(3):361-367.

Lukins, H. B., and J. W. Foster. 1963. Methyl ketone metabolism in hydrocarbon-
utilizing mycobacteria. Journal of Bacteriology 85: 1074-1087.

McCarty, P. L., M. Reinhard, and B. E. Rittman. 1981. Trace organics in ground water.
Environmental Science and Technology 15(1):40-51.

McMahon, R. F., R. D. Hunter, and W. D. Russell-Hunter. 1974. Variation in Aufwuchs
at six freshwater habitats in terms of carbon biomass and of carbonznitrogen ratio.
Hydrobiologia 45:391-404.

Microbial ID, Inc. 1991. Microbial Identiﬁcation System Operating Manual, Version 4.
Newark, Delaware.

Murray, R. E., K. E. Cooksey, and J. C. Priscu. 1986. Stimulation of bacterial DNA
synthesis by algal exudates in attached algal-bacterial consortia. Applied and
Environmental Microbiology 52(5): 1 177-1 182.

Nielson, A. H., A. S. Allard, and M. Remberger. 1985. Biodegradation and
transformation of recalcitrant compounds. In Hutzinger, 0. (ed.) The Handbook
of Environmental Chemistry, Vol. 2, part C, pp. 29-86, Springer-Verlag KG,
Berlin.

116

 

Oldenhuis, R., R. L. J. M. Vink, D. B. Janssen, and B. Witholt. 1989. Degradation of
chlorinated aliphatic hydrocarbons by Methylosinus trichosporium OB3b

expressing soluble methane monooxygenase. Applied and Environmental
Microbiology 55(1 l):28 l 9-2826.

Oltmanns, R. H., H. G. Rast, and W. Reineke. 1988. Degradation of 1,4-
dichlorobenzene by enriched and constructed bacteria. Applied Microbiology and
Biotechnology 28: 609-616.

Paerl, H. W. 1975. Microbial attachment to particles in marine and freshwater
ecosystems. Microbial Ecology 2:73-83.

Pahm, A. M., and M. Alexander. 1993. Selecting inocula for the bioremediation of
organic compounds at low concentrations. Microbial Ecology 25:275-286.

Pettigrew, C. A., B. E. Haigler, and J. C. Spain. 1991. Simultaneous biodegradation of

chlorobenzene and toluene by a Pseudomonas strain. Applied and Environmental
Microbiology 57(1): 157-162.

 

Pritchard, P. H. 1992. Use of inoculation in bioremediation. Current Opinions in
Biotechnology 3:232-243.

Pritchard, P. H., J. B. Lin, J. G. Mueller, and M. S. Shields. 1996. Bioremediation
research in EPA: An overview of needs, directions, and potentials. Chapter 1 in
Hickey, R. F., and G. Smith (eds.) Biotechnology in Industrial Waste Treatment
and Bioremediation. A collection of papers presented at the International
Symposium on the Implementation of Biotechnology in Industrial Waste
Treatment and Bioremediation, Sept 15-16, 1992, Grand Rapids, MI. CRC Press,
Inc., Boca Raton.

Reineke, W., and H. Knackmuss. 1984. Microbial metabolism of haloaromatics:
isolation and properties of a chlorobenzene-degrading bacterium. Applied and
Environmental Microbiology 47(2):395-402.

Reineke, W. 1988. Microbial degradation of haloaromatics. Annual Reviews in
Microbiology 42:263-287.

Riebeth, D. M., N. O. Belser, and C. E. Castro. 1992. Partial rapid metabolism of 1,2-
dichloroethane by Methylosinus trichosporium OB3b. Environmental Toxicology
and Chemistry 1 1:497-501.

Rittman, B. E., P. L. McCarty, and P. V. Roberts. 1980. Trace-organics biodegradation
in aquifer recharge. Ground Water 18(3):236-243.

117

Rittman, B. E., and P. L. McCarty. 1980. Model of steady-state-bioﬁlm kinetics.
Biotechnology and Bioengineering 22:2343-2357.

Roemer, S. C., K. D. Hoagland, and J. R. Rosowski. 1984. Development of a freshwater
periphyton community as inﬂuenced by diatom mucilages. Canadian Journal of
Botany 62:1799-1813.

SAS Institute. 1991. SAS User’s Guide, Version 6.03. SAS Institute, Cary, NC.

Sayler, G. S., H. L. Kong, and M. S. Shields. 1984. Plasmid-mediated biodegradative
fate of monohalogenated biphenyls in facultative anaerobic sediments. In
Ommen, G. S., and A. Holaender (eds.) Genetic Control of Environmental
Pollutants, Basic Life Sciences, Vol. 28, pp. 117-135, Plenum Press, New York.

Scholtz, R., A. Schmuckle, A. M. Cook, and T. Leisinger. 1987. Degradation of
eighteen l-monohaloalkanes by Arthrobacter sp. strain HA1. Journal of General
Microbiology 133 :267-274.

Shreve, G. S., R. H. Olsen, and T. M. Vogel. 1991. Development of pure culture

bioﬁlms of P. putida on solid supports. Biotechnology and Bioengineering
37:512-518.

Sieburth, J. Mc., P. W. Johnson, M. A. Eberhardt, M. E. Sieracki, and M. Lidstrom.
1987. The ﬁrst methane-oxidizing bacterium from the upper mixing layer of the
deep ocean: Methylomonas pelagica sp. nov. Current Microbiology 14:285-293.

Speitel, G. E., and J. M. Leonard. 1992. A sequencing bioﬁlm reactor for the treatment

of chlorinated solvents using methanotrophs. Water Environment Research
64:712-719.

Strand, S. E., and M. E. Lidstrom. 1984. Characterization of a new marine
methylotroph. FEMS Microbiology Letters 21:247-251.

Strand, S. E., and L. Shippert. 1986. Oxidation of chloroform in an aerobic soil exposed
to natural gas. Applied and Environmental Microbiology 52(1):203-205.

Tardif, G., C. W. Greer, D. Labbé, and P. C. Lau. 1991. Involvement of a large plasmid
in the degradation of 1,2-dichloroethane by Xanthobacter autotrophicus. Applied
and Environmental Microbiology 57(6): 1 853-1 857.

Taylor, D. G., P. W. Trudgill, R. E. Cripps, and P. R. Harris. 1980. The microbial
metabolism of acetone. Journal of General Microbiology 118:159—170.

118

 

Tuchman, M., and D. W. Blinn. 1979. Comparison of attached algal communities on
natural and artiﬁcial substrata along a thermal gradient. British Phycolog/
Journal 14:243-254.

United States Environmental Protection Agency. 1983. Organic chemicals and plastics
and synthetic ﬁbers (OCPSF) category efﬂuent limitations guidelines,
pretreatment standards, and new source performance standards. Federal Register
WH-FRL-2305-7, 11828.

United States Environmental Protection Agency. 1994. Method 8260A, volatile organic
compounds by gas chromatography/mass spectrometry (GC/MS): capillary
column technique. In Test Methods for Evaluating Solid Waste
Physical/Chemical Methods, 3rd edition, Final Update I. National Technical
Information Service, Washington, DC.

van den Wijngaard, A. J ., K. W. H. J. van der Kamp, J. van der Ploeg, F. Pries, B.
Kazemier, and D. B. Janssen. 1992. Degradation of 1,2-dichloroethane by
Ancylobacter aquaticus and other facultative methylotrophs. Applied and
Environmental Microbiology 58(3):976-983.

Vandenbergh, P. A., and B. S. Kunka. 1988. Metabolism of volatile chlorinated aliphatic
hydrocarbons by Pseudomonasﬂuorescens. Applied and Environmental
Microbiology 54(10):2578-2579.

Vannelli, T., M. Logan, D. M. Arciero, and A. B. Hooper. 1990. Degradation of
halogenated aliphatic compounds by the ammonia-oxidizing bacterium

Nitrosomonas europea. Applied and Environmental Microbiology 56(4):1169-
1171.

Wetzel, R. G. 1983. Littoral communities: Algae and zooplankton. Chapter 19in
Limnology, 2nd edition. Harcourt, Brace, and Jovanovich, Inc., Orlando, FL.

Whittenbury, R., K. C. Phillips, and J. F. Wilkinson. 1970. Enrichment, isolation, and
some properties of methane utilizing bacteria. Journal of General Microbiology
61 :205-218.

Wimpenny, J. W. T., R. W. Lovitt, and J. P. Coombs. 1983. Laboratory model systems
for the investigation of spatially and temporally organized microbial ecosystems.
In Slater, J. H., R. Whittenbury, and J. W. T. Wimpenny (eds.) Microbes in Their
Natural Environments, pp. 67-117, Cambridge University Press, Cambridge.

Yokata, T., H. Fuse, T. Omori, and Y. Minoda. 1986. Microbial dehalogenation of

haloalkanes mediated by oxygenase or halidohydrolase. Agricultural Biology and
Chemistry 50(2): 453-460.

119

 

Zaidi, B. R., G. Stucki, and M. Alexander. 1988. Low chemical concentrations and pH
as factors limiting the success of inoculation in bioremediation. Environmental
Toxicology and Chemistry 7: 143-151.

Zaidi, B. R., Y. Murakami, and M. Alexander. 1989. Predation and inhibitors in lake
water affect the success of inoculation to enhance biodegradation of organic
materials. Environmental Science and Technology 23:859-863.

Zobell, C. E. 1943. The effect of solid surfaces upon bacterial activity. Journal of
Bacteriology 43:39-56.

 

 

120

"‘illlllllllllllilli