BACTERIAL COMMUNITY COMPOSITION IN STREAM BIOFILMS IS
INFLUENCED BY ALGAL RESPONSE TO VARYING LIGHT AND PHOSPHORUS
RATIOS
BY
YU-RUI CHANG
THESIS
Submitted in partial fulfillment of the requirements
for the degree of Master of Science in Natural Resources and Environmental Sciences
in the Graduate College of the
University of Illinois at Urbana-Champaign, 2010
Urbana, Illinois
Adviser:
Assistant Professor Angela D. Kent
ABSTRACT
Strong correlations between bacterial communities and algal seasonal
succession have been previously observed. In aquatic systems, dissolved
organic carbon derived by algae is an important resource for bacteria. Light and
phosphorus availability are two factors that influence biomass and abundance of
algae, and the changes will be reflected in the bacterial portion of the microbial
community. Glycolate is an algal-specific exudate produced under excess light
conditions. Glycolate uptake by bacteria has been shown to correlate with algal
primary productivity. Bacterial populations that utilize glycolate possess the gene,
glycolate oxidase subunit D (glcD). This gene was used as a marker to identify
changes in specific bacterial populations that respond to algal exudates. In this
study, development of periphyton biofilms in an experimental stream system was
monitored across different light and phosphorus levels. Samples were collected
every two days for community and chemistry analyses. Bacterial communities
were monitored using DNA fingerprinting techniques based on ribosomal RNA
genes and the glcD gene. We demonstrated that bacterial community
composition changed significantly over the course of biofilm development, and
light and phosphorus availability contributed to those differences in community
composition. Our results suggest that a strong coupling between carbon flow and
bacterial community composition. These results increase our understanding of
the ecological drivers that impact benthic biofilm communities that carry out
transformation of nutrients in streams.
ii
ACKNOWLEDGMENTS
I gratefully thank to my adviser, Angela Kent for her insights and suggestions that
assisted to shape my research skills. Without her, this research project and
thesis would not have been possible. I am grateful to my committee members,
Walter Hill for experiment setup and sample collection, and Tony Yannarell for
valuable comments and thoughts. I also thank to my lab coworkers, Sara Paver
and Ariane Peralta for helpful discussions and encouragements. This research
project was supported by North Temperate Lakes Microbial Observatory, National
Science Foundation and Department of Natural Resources and Environmental
Sciences at University of Illinois at Champaign-Urbana.
iii
TABLE OF CONTENTS
CHAPTER 1: INTRODUCTION
1.1 Microbial functions and algal-bacterial interactions
in aquatic systems………………………………………………………… 1
1.2 Effects of light and nutrients on microbial community………………… 2
1.3 Microbial community in stream biofilms………………………………… 8
1.4 Algal exudates supply growth substrates to
heterotrophic bacteria…………………………………………………….. 11
1.5 Purpose of this study……………………………………………………... 13
CHAPTER 2: METHODS
2.1 Experimental design……………………………………………………... 16
2.2 Sample collection………………………………………………………… 17
2.3 Laboratory analysis………………………………………………………. 18
2.4 Statistical analysis………………………………………………………... 22
CHAPTER 3: RESULTS
3.1 Experimental treatments…………………………………………………. 25
3.2 Total bacterial community composition (ARISA)………………………. 25
3.3 Glycolate-utilizing bacterial community composition
(glcD T-RFLP)..................................................................................... 29
CHAPTER 4: DISCUSSION
4.1 Experimental streams and treatments………………………………….. 31
4.2 Molecular fingerprinting techniques…………………………………….. 32
iv
4.3 Bacterial community dynamics………………………………………….. 32
4.4 Light and phosphorus effects on algal assemblages and
bacterial community composition……………………………………….. 34
4.5 Effects of light and phosphorus on heterotrophic bacterial
community composition………………………………………………….. 36
CHAPTER 5: CONCLUSIONS…………………………………………………….. 38
TABLES AND FIGURES……………………………………………………………. 39
REFERENCES………………………………………………………………………. 51
AUTHOR’S BIOGRAPHY…………………………………………………………... 57
v
CHAPTER 1
INTRODUCTION
1.1 Microbial functions and algal-bacterial interactions in aquatic systems
In aquatic ecosystems, microorganisms play important roles in nutrient cycling and
organic matter decomposition (Azam et al. 1983, Cotner and Biddanda 2002, Kent et al.
2004). The composition and activity of microbial populations involved in biogeochemical
transformations can be influenced by physical (e.g. weather and water temperature)
(White et al. 1991, Ducklow and Carlson 1992, Felip et al. 1996), chemical (e.g. pH,
availability of N, P, and quality and quantity of DOM) (Methe et al. 1998, Stepanauskas
et al. 1999, Fisher et al. 2001, Crump et al. 2003) and biological (e.g. grazer pressure,
competition for resources, or symbiotic interactions) (Jurgens et al. 1999, Hahn and
Hofle 2001) factors. Biological interactions are thought to be a major factor determining
bacterial community composition structure in aquatic systems (Kent et al. 2006). Strong
correlations between composition of phytoplankton communities and bacterioplankton
communities in lakes and rivers have been observed (Crump and Hobbie 2005, Kent et
al. 2007). Previous studies have demonstrated that correlation between bacterial and
algal communities is due to bacterial responses to dissolved organic carbon (DOC)
derived by algae (Azam et al. 1983, van Hannen et al. 1999, Arrieta and Herndl 2002,
Rier and Stevenson 2002, Crump et al. 2003, Crump and Hobbie 2005, Stets and
Cotner 2008). In the microbial food web, algae provide DOC to support bacterial growth
while bacteria remineralize organic compounds and nutrients to support algal production
(Cole 1982). It is estimated that 10-50% of fixed carbon is released by algae to the
1
aquatic DOC pool (Fuhrman and Azam 1982), and while DOC enters aquatic systems
from both allochthonous and autochthonous sources, the algal-derived DOC is
considered to be more readily available for use by heterotrophic bacteria (Wetzel 1992).
1.2 Effects of light and nutrients on microbial community
Despite bacterial growth and production respond to algal-released DOC, it appears that
bacterial secondary production is likely to be limited by the supply of nutrients, such as
nitrogen or phosphorus (Morris and Lewis 1992, Vadeboncoeur et al. 2001, Elser et al.
2007). In addition, nitrogen and phosphorus supply can have a great influence on
community structure (Makino and Cotner 2004, Carpenter et al. 2005). Nitrogen fixation
(Howarth 1988), nitrification and denitrification (Seitzinger 1988) carried out within lakes
and streams (autochthonous) are important biogeochemical transformations in the
nitrogen cycle that influence the availability of nutrients for microbial populations.
Additionally, allochthonous nutrient loading from runoff, groundwater, and rainfall is also
an important component of nutrient availability. Heterotrophic bacteria are responsible
for a large proportion of both N and P uptake in freshwater and marine ecosystems
(Currie and Kalff 1984, Bentzen and Taylor 1991, Kirchman 1994, Roberts and Howarth
2006). However, results from a number of studies suggest that heterotrophic bacteria
account for a greater uptake of phosphorus than nitrogen (Vadstein et al. 1988,
Kirchman 1994). There are two explanations for relatively high P uptake by bacterial
populations. (1) Aquatic microorganisms contain phosphorus in the membrane
(phospholipids) and nucleic acids. Compared to algae, bacterial cells are smaller with
higher surface area: volume ratios. Bacterial P content should be relatively high
2
compared to eukaryotic cells (Hessen et al. 2002), thus, there is a need for higher
phosphorus input in bacterial cells relative to other taxa. (2) The DNA and RNA content
of bacteria are as much as 20% of cellular carbon, which is much higher than algae
(Makino and Cotner 2004, Hessen et al. 2008). The requirement and uptake of P in
bacteria imply that P may be a limiting factor for the bacterial community.
The algal community provides a carbon source for heterotrophic organisms in aquatic
systems through primary production, and this carbon source is essential and important
for supporting the microbial food web. Nutrients (N and P) and light are two factors
limiting algal growth and primary production in freshwater ecosystems. Since algae
strongly respond to the availability of nutrients and light, algae exhibit a wide range of
variation in tissue nutrient stoichiometry such as C: N: P ratios and cellular
concentration (Sterner and Elser 2002). As the stoichoimetric composition of algae
changes with resource availability, primary production and resource quality of the algal
biomass not only directly affects herbivore growth but also strongly impacts the food
web, community structure and nutrient cycling (Vannote et al. 1980, Smith 2003).
The research efforts on the relationship between phosphorus and algal production in
aquatic systems are largely conducted in lakes (Vollenweider 1976, Hill et al. 2009).
Carpenter and colleagues demonstrated phytoplankton and periphyton response to
phosphorus addition by conducting whole-lake fertilization in three lakes in Michigan
(Carpenter et al. 1996, Vadeboncoeur et al. 2001). They found that total primary
production always increased with fertilization but contribution of primary production
3
switched from periphyton to phytoplankton as light penetration decreased with increase
in phytoplankton biomass and abundance.
Most studies of nutrient effects on periphytic algae in streams have only assessed the
presence or absence of nutrient limitation in the streams (2006). Few studies have
examined the response of periphyton across experimentally manipulated phosphorus
gradients (Hill et al. 2009). Bothwell et al. (1989) examined both cellular growth rates
and peak areal biomass of periphyton, and demonstrated that the amount of P needed
to saturate cellular growth rates in a thin biofilm was two orders of magnitude lower than
the concentration needed to saturate a well-developed biofilm at peak areal biomass.
The difference in nutrient concentrations might be caused by the increase in thickness
over time as the biofilm develops, which could decrease nutrient diffusion and nutrient
availability within the mat. Rier and Stevenson (2006) manipulated nitrogen and
phosphorus concentrations in stream mesocosms to develop quantitative relationships
between benthic algal growth rates and peak biomass with N and P concentrations.
They suggested that saturating concentrations for growth rates were 3-5 times lower
than concentrations needed to produce maximum biomass. Hill (2009) investigated the
individual and combined effects of light and P on the growth and P content of periphytic
algae, and demonstrated that periphyton P content was strongly and nonlinearly related
with soluble reactive phosphorus (SRP), reaching a maximum at 82 µgL-1. They also
demonstrated that periphytic algal growth reached saturation above 22 µgL-1 SRP.
4
Light is one of the most important factors affecting algal growth and primary production
in freshwater systems. Especially in stream environments, light is a determinant of
benthic algal community structure and function that influences biomass, productivity,
and taxonomic composition (Rier et al. 2006). Several studies have shown strong
evidence for high correlation between algal growth, primary production and light. Hill
(2009) examined the quantitative relationship between two abiotic factors (light and P)
and the growth of benthic algae, suggesting that the individual effect of light significantly
influences algal biomass accrual. Algal biomass increased linearly with increasing light
until biomass reached a saturation level. Algal growth and photosynthesis can be limited
by resource availability if light is available in excess relative to other resources. Also,
early studies by Hill et al. (Hill et al. 1995, Hill et al. 2001) demonstrated that irradiance
of less than 10 µmol photon m-2s-1 in streams may strongly constrain algal
photosynthesis, growth and biomass production.
There have been many studies addressing the individual effects of light or nutrients on
algae and heterotrophs. However, the combined effects of light and nutrients cannot be
disregarded when examining interactions between algal and bacterial communities in
nature. The effects of light and phosphorus interact strongly and play important roles in
freshwater phytoplankton stoichiometric composition (Dickman et al. 2006). Phosphorus
is a critical structural and functional component in algae, while light is a necessary
energy source for algae to carry out photosynthesis and primary productivity. The
interactive effects of light and phosphorus are often discussed with respect to algal
stoichiometric response. The light: nutrient hypothesis (LNH) predicts that algal carbon:
5
nutrient ratios are driven by the ratio of available light and nutrients (Sterner et al. 1997).
Algal nutrient content is negatively correlated with light intensity (Dickman et al. 2006).
For example, high light (or low P) can lead to high C: P ratios but low light (or high P)
can lead to low C: P ratios. At high light levels, algal carbon fixation is high, and unless
the total phosphorus supply is also high, the fixation of carbon and uptake of
phosphorus may be unbalanced, causing increases in C: P ratios in algal cells (BermanFrank and Dubinsky 1999). Low C: P ratios in nutrient-saturated environments indicate
luxury consumption of P or the ability of algae to store nutrients in excess of that
required for immediate growth (Rhee 1973).
At high light levels, algal cells tend to release a relatively high percentage (as much as
50%) of fixed carbon as DOC, particularly when cells are nutrient limited (Elser et al.
2002). A C-rich environment results from the physiological responses of algae to high
light and nutrient ratios, and the tendency toward excess exudate production with high
C: P ratio (Sterner et al. 1997). A relatively large fraction of this algal DOC can route into
microbial processes. Bacterial growth and community structure are dependent on
quality and quantity of organic carbon (Kent et al. 2006). A few studies have
demonstrated that the labile DOC derived from pelagic algal community is readily
utilized as high quality substrates for bacterial growth (van Hannen et al. 1999), more so
than DOC derived from allochthonous sources.
A comprehensive review of the literature on the flux of carbon from algae to bacteria in
marine and freshwater systems (Baines and Pace 1991) found a wide range in the
6
percentage of primary production released from algae as DOC. In individual measures,
algal exudates can range from 0 to 100% of DOC with typical values of 3-40%. In a
whole-lake 13C addition experiment exploring importance of allochthonous and
autochthonous carbon sources for support of pelagic bacteria in lakes of northern
Michigan, (Kritzberg et al. 2004) suggested that autochthonous carbon derived from
phytoplankton primary production supported 30-65% of bacterial production, which
implies that algal exudates are essential and high quality growth substrates for aquatic
bacteria. Variation in algal production due to light and/or nutrient limitation may alter
DOC supply for bacteria and influence bacterial growth and abundance (Elser et al.
2002). While the effects of light and nutrients on primary production and herbivores are
reasonably well studied (Urabe et al. 2002, Elser et al. 2003, Hill and Fanta 2008),
influences of algal responses to light and P on bacterial community composition are
poorly understood. We expect that under varying light: P ratios, the physiological
responses of algae and bacteria will be linked. Specifically, we expect changes in algal
and bacterial community under four contrasting conditions of light and phosphorus (Fig.
1a): (1) under high light and low P availability, algal biomass and productivity would be
high. Excess light results in algal release of large amounts of DOC due to low P
availability. Algal cells release these excess carbon compounds as exudates which are
available to bacterial populations. In this case, variability of bacterial community might
increase with diversity and abundance of exudates. (2) Under low P, low light
conditions, signs of algal light limitation might be observed, including growth limitation or
decline in algal biomass and organic carbon exudation. However, periphyton and
bacteria would be P-rich by responding to relatively high phosphorus (relative to
7
available DOC). (3) While at high light and high P, the abundance of algae and bacteria
might be unchanged compared to enriched conditions of either light or phosphorus, but
algal exudate production would be reduced due to low C: P ratio in algal cells. In this
case, primary production is expected to be routed to algal biomass rather than DOC
production. (4) At low light but high P levels, algal biomass would be decreased as
algae experience strong light limitation. Effects of light on bacteria are likely to be
indirect if any. As a result only responding to P enrichment (not limited by light), bacteria
will tend to dominate communities that develop with less light availability. Measurement
of algal biomass (Hill and Fanta 2008) under four extreme light: P ratios provides
evidence to support predictions of algal responses to varying light and phosphorus
treatments (Fig. 1B).
1.3 Microbial community in stream biofilms
Stream biofilms are comprised of multiple species of algae, heterotrophic bacteria,
cyanobacteria, chemoautotrophic bacteria and other microorganisms within a selfdeveloped polysaccharide matrix built on solid surfaces (Lock et al. 1984). Those
periphyton communities play a key role in the carbon and nutrient dynamics of stream
ecosystems. They are involved in various ecological processes such as primary
production, nutrient uptake and transformation, and organic compound decomposition
(Armstrong and Barlocher 1989, Meyer 1994, Mulholland et al. 1995). It has been
shown that planktonic heterotrophs favor use of algal-released DOC as carbon sources
(Wetzel 1992). A similar relationship between algae and bacteria has been found in
benthic environments (Haack and McFeters 1982, Kaplan and Bott 1989). In stream
8
systems, dissolved organic carbon from allochthonous sources, such as leachate from
litter fall accounts for 70-90% of total DOC (Munster 1993). However, leaf leachate or
other recalcitrant carbon sources cannot be utilized by all bacterial populations
(McNamara and Leff 2004). Instead, several studies have been shown that bacteria
preferentially use labile DOC released by benthic algae in biofilms (Haack and McFeters
1982, Kaplan and Bott 1989, Sinsabaugh et al. 1991, Jones and Lock 1993, Wetzel
1993, Romani and Sabater 2000, Ylla et al. 2009). The availability of light to stream
biofilms influences the relationship between algae and heterotrophic bacteria through
impacts on the abundance and diversity of labile extracellular organic compounds
released by photosynthetic microorganisms (Haack and McFeters 1982, Kaplan and
Bott 1989, Sinsabaugh et al. 1991, Jones and Lock 1993, Wetzel 1993, Romani and
Sabater 2000, Ylla et al. 2009). In contrast, the presence of relatively high phosphorus
(compared to light) will disrupt algal-bacterial interaction because primary production is
routed into biomass accumulation rather than organic carbon excretion. Under these
conditions, bacteria would be expected to utilize recalcitrant organic matter as a major
carbon source.
Bacterial growth, biomass and productivity in biofilms have been reported to be
significantly influenced by algal-released DOC (Espeland and Wetzel 2001). A positive
relationship between benthic algae and bacterial community has been observed:
biofilm-bacterial production was coupled to algal production as a result of organic
compounds excreted during photosynthesis (Haack and McFeters 1982, Carr et al.
2005). Nutrient enrichment of stream waters can have a range of effects on biofilm
9
microbial interactions, in part through impacting the release of DOC (Fogg 1983). More
specifically, nutrient enrichment tends to disrupt the positive relationship between algal
and bacterial production (Scott et al. 2008). Lyon et al. (2009) investigated the effects of
elevated nutrient concentrations on DOC release and C cycling within benthic biofilm
communities in four headwater streams in Arkansas. The results suggest that in the
most nutrient-rich streams, the majority of DOC was exported directly from algal cells to
downstream waters without the opportunity for heterotrophic microbes to take up and
incorporate algal-derived carbon compounds. In contrast, in relatively oligotrophic
streams, exported DOC (originally released from algae) was mostly derived from older
organic matter stored within the biofilm, while a higher proportion of newly synthesized
algal carbon was incorporated by heterotrophic bacteria. This result indicates a tighter
coupling between algae and bacteria under low available nutrient conditions in benthic
biofilms (Lyon and Ziegler 2009).
The effects of light on algal biomass and primary production in pelagic systems were
discussed previously. A similar relationship has been reported for stream biofilms. The
release of DOC from algal cells increases with photosynthetic production, and is
correlated with increasing light availability (Espeland and Wetzel 2001). However,
benthic algae are attached to solid surfaces and may grow in thick periphyton mats,
resulting in self-shading by these stream biofilms. These attached algae capable of
shade adaptation (Hill et al. 2001), which is the ability of algae to increase
photosynthetic efficiency when self-shading occurs during overgrowth. Even though
shade adaptation may allow algae to survive in streams where light availability is low,
10
shade adaptation cannot compensate completely for the effects of low light (Rier et al.
2006).
1.4 Algal exudates supply growth substrates to heterotrophic bacteria
1.4.1 Photorespiration and glycolate production
Light and nutrient ratios affect the quantity and quality of DOC released by algae.
Excess light and photosynthesis can shift the balance between carboxylation and
oxygenation of ribulose-1,5-bisphosphate (RuBp) (both processes carried out by
ribulose-1,5-bisphosphate carboxylase/oxygenase (RUBISCO)). Fixation of O2 rather
than CO2 produces 3-phosphoglycerate and 2-P-glycolate, the former re-enters Calvin
Cycle, but the latter cannot enter Calvin Cycle. Instead, 2-P-glycolate is
dephosphorylated and converted into glycolate. Consequently, glycolate can be
oxidized in the peroxisome, resulting in the formation of 3-P-glycerate, glycine, and
serine. Ultimately, 3-P-glycerate enters Calvin cycle (Parker and Armbrust 2005). During
this metabolic process, CO2 and NH3 are produced with consumption of ATP and
reducing equivalents. Photorespiration is essential for protection of algae exposed to
high light by serving as an energy sink that prevents the over-reduction of the
photosynthetic electron transport chain and photoinhibition (Wingler et al. 2000).
Glycolate produced through the oxygenase activity of RUBISCO can be oxidized by
algal cells, completing the process of “photorespiration”, or released from the algal cell
(Lau and Armbrust 2006). Glycolate production and excretion is thought to be a means
of modulating excess light energy (Leboulanger et al. 1998, Lau and Armbrust 2006,
Lau et al. 2007). Glycolate, a two-carbon compound, represents a potentially important
11
energy and carbon resource for heterotrophic aquatic bacteria.
Previous studies on photorespiration and glycolate exudation have been conducted in
marine systems (Parker et al. 2004, Parker and Armbrust 2005, Lau et al. 2007). These
previous studies assessed the occurrence and prevalence of photorespiration in the
marine environment and investigated photorespiration as a mechanism for DOC entry
into the marine food web. In the marine environment, glycolate can comprise a
substantial proportion (10-50%) of phytoplankton-excreted DOC (Wright and Shah
1977, Edenborn and Litchfield 1987). (Lau and Armbrust 2006) examined the influence
of phytoplankton production of glycolate on bacterial community structure during a
phytoplankton bloom in a marine environment. This study detected the coupling of
glycolate-utilizing bacteria to phytoplankton photorespiration on a diel cycle and
observed correlation in dynamics of phytoplankton and bacteria at the phylotype level.
The results suggested that bacterial utilization of carbon substrates such as glycolate
might be a critical determinant in bacterial interactions with the phytoplankton
communities.
1.4.2 Glycolate-utilizing bacteria possess glcD gene
Bacterial populations can utilize algal-derived glycolate using the enzyme glycolate
oxidase. Synthesis of this enzyme is encoded by the glc locus, containing five open
reading frames. All the open reading frames, glcB, glcD, glcE, glcF and glcG are in the
same orientation; glcC is in the opposite orientation. The glcD gene encodes a protein
involved in the oxidation of glycolate while glcC gene encodes regulatory protein, and
glcF is required for encoding the subunit that is required for catalyzing the subunits
12
encoded by glcD-glcE products (Pellicer et al. 1996). The glcD gene is only detected
and activated in glycolate-utilizing bacteria, and has been previously used as a
functional marker to study bacterial populations capable of glycolate oxidation (Lau and
Armbrust 2006). Study of bacterial populations that possess the glycolate oxidase
subunit D (glcD) gene may provide us with a better understanding of algal- bacterial
interactions. We propose that bacterial glcD genes provide a marker to identify bacteria
that are directly influenced by algal-derived carbon, essentially a functional gene for
heterotrophy (Lau et al. 2007).
1.5 Purpose of this study
In this study, we aimed to demonstrate the influence of algal responses to manipulation
of light and phosphorus ratios on bacterial community composition in stream biofilms.
We hypothesized that varying light: P ratios would result in differences in production of
glycolate and other algal exduates. Those differences would be reflected in the bacterial
portion of biofilm community, particularly in the diversity and abundance of those
bacterial populations that were able to use benthic algal exudates as carbon sources.
We expected that responses of glycolate-utilizing bacteria to differing light and
phosphorus should be more significant and direct (compared to the total bacterial
community) because glycolate production and excretion is sensitive to light available to
algal cells.
We created community DNA “fingerprints” based on length heterogeneity in the
intergenic spacers region between 16S rRNA and 23S rRNA gene to investigate the
13
dynamics and variability of the total bacterial community associated with stream biofilms
grown under various light and phosphorus treatments. In addition, we also generated
community fingerprints based on glcD genes possessed by potential glycolate-utilizing
bacteria to assess individual and combined effects of light and phosphorus on bacterial
community composition through changes in algal-released exudates. Specifically, we
addressed the following questions: (1) are there predictable patterns of succession in
bacterial communities associated with stream biofilms? (2) How does development of
the periphyton-associated bacterial community in artificial stream biofilms differ among
treatments of different light and nutrient levels? (3) How does the development of the
heterotrophic bacterial community (assessed by the glcD gene) differ with respect to
light: nutrient levels and algal assemblages in stream biofilms?
Differing levels of light and phosphorus were applied to experimental stream reaches to
assess the influence of these factors on the growth and composition of the periphyton
community (Hill et al. 2009), and on interactions between algal and bacterial
communities. To the best of our knowledge, this is the first research study manipulating
light and nutrient levels simultaneously (Hill et al. 2009). Most studies examining the
effects of light and nutrients on microbial community within biofilms have focused on the
algal portion of the community (Hill and Fanta 2008, Hill et al. 2009). Furthermore,
studies exploring algal-bacterial interaction have primarily focused on the individual
effect of either light or nutrient levels (Carr et al. 2005, Rier and Stevenson 2006, Rier et
al. 2006, Ylla et al. 2009). This study is unique in that we manipulated gradients of light
and phosphorus and examined combined effects of these two factors on algal-bacterial
14
interactions.
In natural streams, a limitation on examining effects of phosphorus concentrations on
periphyton-associated algal and bacterial growth is that most oligotrophic streams
require excess nutrients to stimulate early development of biofilms and saturate cellular
growth rates (Rier and Stevenson 2006). In this study, the range of phosphorus
concentration and light intensity is reasonable. The phosphorus concentration of First
Creek (Oak Ridge, TN) fell into the range of most US streams (Hill et al. 2009).
Previous studies on the dynamics of glycolate-utilizing bacteria have focused on marine
systems (Lau and Armbrust 2006, Lau et al. 2007). No information on freshwater
glycolate-utilizing bacterial communities is available. We used molecular methods to
study bacterial glcD genes that encode glycolate oxidase enzymes in freshwater
systems. Degenerate polymerase chain reaction (PCR) primers have been designed to
amplify the glcD gene, which encodes the D-subunit of the enzyme glycolate oxidase
(Lau and Armbrust 2006). Sequence diversity of glcD and correlations between
glycolate oxidizers and phytoplankton were explored by cloning and DNA sequencing
(Paver and Kent 20__). This database of freshwater glcD sequences can also help to
characterize bacterial community and dynamics in response to primary producers in an
algal biofilm. By studying the diversity and dynamics of bacteria that can use
phytoplankton-derived glycolate (based on the glcD gene), we can examine the
biological interactions that influence bacterial community structure in benthic biofilms.
15
CHAPTER 2
METHODS
2.1 Experimental design
2.1.1 Experimental streams
Phosphorus and light were manipulated in three replicate experiments at the Oak Ridge
National Laboratory indoor stream facility. Experimental designs described elsewhere
(Hill and Fanta 2008). Briefly, six U-shaped, flow-through streams (22m long, 0.3m
wide) were supplied with water from First Creek, a first-order unpolluted stream on the
Oak Ridge Reservation, Oak Ridge, TN. A continuous mat of unglazed, white ceramic
tiles (2.4 x 2.4 x 0.6 cm3) was placed on the bottom of the experimental streams.
Illumination was provided by eight metal halide lamps (400W) positioned above each
stream. A 14:10 light: dark cycle was maintained throughout the experiment. Water flow
in the streams was initiated three days before experimental treatments were supplied.
Cobbles from First Creek were placed at the head of the streams to provide a source of
algae and bacteria. The water flowing into the experimental stream reaches also
contained additional algal and microbial colonists. Periphyton was allowed to develop
on ceramic tiles.
2.1.2 Experimental treatments
Six different phosphorus concentrations were applied to the six streams in each of the
three replicate experiments by pumping stock solutions of dissolved Na2HPO4 at the
head of each stream. Target concentrations of soluble reactive phosphorus were 6, 12,
16
25, 75, 150, and 300 µg L-1 in the first two experiments and 6, 12, 25, 50, 75, and 150
µg L-1 in the third experiment. Sodium nitrate (NaNO3) was also added to reach over
300 µg L-1 in the streams to ensure nitrogen did not limit periphyton growth. A gradient
of light intensity ranging from photosaturating to highly shaded was established in each
stream in all three experiments. There were five treatment levels in the first two
experiments, designated as “+,” “0,” “1,” ”2,” ”3.” The “+” treatment was created by
mounting a 300-W halogen work light close (ca. 40 cm) to the tile substrata, producing
irradiances of ca. 375 µmol photons m-2s-1. Overhead lamps provided background
irradiances intensity approximately at 100 µmol photons m-2s-1. The “1,” “2,” and “3”
treatments were created by shading 1-m-long sections of the streams with 1, 2, or 3
layers of nylon window screening; these layers created irradiances of approximately 60,
40, and 20 µmol photons m-2s-1, respectively. All light treatments were applied to the
lower half of each experimental stream reach to allow for complete mixing of nutrient
stock solutions by the time the flowing water reached these sections.
2.2 Sample collection
Sample collection in each experiment began 3-4 days after the application of
phosphorus and light treatments after color was observed on tiles. Samples of both
dissolved nutrients and periphyton were collected every two days. Three replicate
experiments were blocked temporally because of facility limitation. The dates of the
three experiments were 25 January 2006- 08 February 2006, 15 February 2006- 27
February 2006, and 04 March 2006- 27 March 2006. Due to lack of samples from early
dates of the third replicate experiment, we excluded this experiment from the analyses
17
of community dynamics. In addition, because the 300 µg L-1 of phosphorus was only
applied in first two replicate experiments, samples from the third replicate experiment
were not included in any analyses involved phosphorus. Biofilm tiles collected for
bacterial analyses were stored at -80 ºC until samples were processed for DNA
extraction. Experimental design, sample collection and chemical analyses (algal
biomass and chrolophyll a, etc.) used as environmental variables in this study were
conducted by Walter Hill at the Illinois Natural History Survey. The detailed experimental
design was described previously (Hill et al. 2009). Results of periphyton responses to
the light and phosphorus manipulations have been previously reported (Hill and Fanta
2008).
2.3 Laboratory analysis
2.3.1 DNA extraction
Biofilm biomass was scraped from the tiles with a sterile razor blade and rinsed with
sterile PBS-Tween 80 (0.15%). Biomass from each tile was collected on a 0.2µm
polyethersulfone filter (Pall Supor-200) (Pall Corporation, Port Washington, NY, USA).
Genomic DNA was extracted from the filters using the Fast DNA kit (MP Biomedicals,
Solon, OH, USA).
2.3.2 Total bacterial community analysis based on 16S rRNA genes
The total bacterial community was characterized using automated ribosomal intergenic
spacer analysis (ARISA). ARISA uses the variation in length of the 16S-23S RNA
intergenic spacer regions to distinguish populations within a bacterial community. The
18
primers for ARISA are 1406f, 5’- TGYACACACCGCCCGT-3' (universal, 16S rRNA
gene), and 23Sr, 5'-GGGTTBCCCCATTCRG-3' (bacteria-specific, 23S rRNA gene). The
1406f primer was labeled at the 5' end with the phosphoramidite dye 6-FAM. The
polymerase chain reaction included a buffer containing 10mM Tris (pH 8.3), 0.25 mg/mL
BSA and 3.0 mM MgCl2, 0.25 mM deoxynucloside triphosphates, 0.4 µM of each primer,
and 0.05U µl-1 GoTaq DNA polymerase (Promega, Madison, WI). Amplification was
carried out using a 2 min initial denaturation at 94°C, followed by 30 cycles of 94°C for
35s, 55°C for 45s, 72°C for 2 min and a final extension of 72°C for 2 min. Denaturing
capillary electrophoresis was carried out for each PCR reaction using an ABI 3730XL
Genetic Analyzer (Applied Biosystems Inc., Foster City, CA, USA). Electrophoresis
conditions were 63°C and 15 kV with a run time of 120 min using POP-7 polymer. A
custom 100- to 2000-bp Rhodamine X-labeled size standard (BioVentures Inc.,
Murfreesboro, TN, USA) was used as the internal size standard for each sample. The
ABI GeneScan ROX 2000 size standard was used as the internal size standard for the
ARISA fingerprints.
2.3.3 glcD clone libraries and DNA sequencing
Little information is available on glycolate-utilizing bacteria in freshwater systems.
Therefore, we generated glcD clone libraries to identify the diversity of this gene in
freshwater bacterial populations. Environmental samples for glcD cloning and DNA
sequencing were harvested from six humic lakes (Crystal Bog, Forestry Bog, North
Sparkling Bog, South Sparkling Bog, Trout Bog, and Why Not Bog) in northern
Wisconsin that have been previously used to study bacterial-algal interactions (Kent et
19
al. 2007). We used these samples as a source of freshwater glcD genes for establishing
a database to characterize bacterial members in stream biofilm (Paver and Kent 20__).
The primer sequences (with 3‘ degenerate region underlined) used to amplify glcD
genes were 5’-GACCCAGACAATCGGAGTGCCGTGGTTSARCCNGGNGT-3’ (GlcD-1f)
and 5’-CATAAAATTGCGCTTCATAACGCCGATGCCRTGYTCNCC-3’ (GlcD-1r) (Lau
and Armbrust 2006). Predicted amplicon size is 980 bp. PCR reactions contained
1.5mM MgCl2 (Idaho Technology, Salt Lake City, Utah, USA), 200uM of each dNTP,
1uM of each primer, Taq DNA polymerase (Promega, Madison, WI, USA) and consisted
of a 15 min initial denaturation at 94°C for 2 min, followed by 40 cycles of 94°C for 30s,
60°C for 30s, 72°C for 90s, and a final extension of 72°C for 2 min. PCR products were
collected and extracted with QIAquick PCR Purification Kit (QIAGEN, Valencia, CA,
USA). In order to improve the efficiency of ligation into the pGEM-T Easy vector
(Promega, Madison, WI, USA), a single base A-overhanged was added to the PCR
products using dATP and Taq polymerase.
DNA templates for sequencing were generated by PCR amplification of plasmid inserts
using M13 forward and reverse primers, 5’-GTTTTCCCAGTCACGAC-3’ (M13F) and 5’CAGGAAACAGCTATGAC-3’ (M13R). Unincorporated primers, nucleotides, and salts
were removed using the AMPure system (Agencourt Bioscience Corp., Beverly, MA,
USA). Cleaned M13 PCR products were sent to the Roy J. Carver Biotechnology
Center at the University of Illinois at Urbana-Champaign for sequencing. DNA
sequences were edited with Geneious Pro 4.6.4. The sequences were aligned using
20
glcD-1f and glcD-1r primer. Edited sequence length was about 980bp. We compared
the freshwater glcD sequences to previously described sequences using Basic Local
Alignment Search Tool (BLAST) (NCBI). This library of glcD sequences was used to
identify appropriate primers and enzymes for a terminal restriction fragment length
polymorphism (T-RFLP) assay to examine the community of glycolate oxidizers
associated with the biofilms (Paver and Kent 20__).
2.3.4 GlcD T-RFLP community fingerprints
T-RFLP fingerprints based on the glcD gene (Paver and Kent 20__) were used to
investigate how bacterial populations interact with periphyton in stream biofilms. The
primer sequences (with 3’ degenerate region underlined) used to amplify glcD were 5’GTTGCCCCGCACGGCTTCTACTACGCNCCNGAYCC-3’ (GlcD-2f) and 5’CATAAAATTGCGCTTCATAACGCCGATGCCRTGYTCNCC-3’ (GlcD-1r). The GlcD-2f
primer was labeled at the 5’ end with the phosphoramidite dye HEX while the GlcD-1r
primer was labeled at the 5’ end with 6-FAM. The polymerase chain reaction contained
0.8 deoxynucloside triphosphates, 0.3 µM of each primer, 1x buffer with 2.0 mM MgCl2
(Idaho Technology, Salt Lake City, UT, USA) and 0.05U µl-1 GoTaq DNA polymerase
(Promega, Madison, WI, USA). PCR cycles consisted of a 2 min initial denaturation at
94°C, followed by 32 cycles of 94°C for 30s, 67°C for 30s, 72°C for 90s and a final
extension of 72°C for 2 min. PCR was carried out in an Eppendorf MasterCycler
(Eppendorf AG, Hamburg, Germany). Excess primers, nucleotides, and salts were
removed from PCR products using the Qiagen MinElute PCR purification kit. Cleaned
PCR products were digested with restriction enzymes, AluI and RsaI (New England
21
Biolabs, Ipswich, MA), to discriminate amplicons that differ in sequence (and thus
represent different populations of glycolate-ultilizing bacteria). Digested fragments were
analyzed by denaturing capillary electrophoresis using an ABI 3730XL Genetic Analyzer
(PE Biosystems). Electrophoresis conditions were 63°C and 15 kV with a run time of
120 min using POP-7 polymer. The ABI GeneScan ROX 1000 size standard was used
as the internal size standard for the T-RFLP fingerprints.
Size-calling and alignment of ARISA and glcD T-RFLP profiles were carried out using
GeneMarker version 1.75 (SoftGenetics). Each ARISA fragment represents a microbial
population, and the combination of all ARISA fragments within a given sample
corresponds to the microbial community composition. The signal strength (e.g. peak
area) of each ARISA peak or T-RF was normalized to account for run-to-run variations
in signal detection by dividing the area of individual peaks by the total fluorescence
detected in each profile, expressing each peak as a proportion of the observed
community (Yannarell and Triplett 2005). Peaks with relative fluorescence units greater
than 200 were included in ARISA analyses, 100 fluorescence units was used as the
detection threshold for glcD T-RFLP analyses. Normalized T-RFLP profiles generated
from different digests were concatenated to generate a single profile based on glcD to
represent each sample.
2.4 Statistical Analysis
For each ARISA and T-RFLP data set, the Bray-Curtis similarity coefficient (Legendre
and Legendre 1998) was used to assess the degree of similarity between bacterial
22
communities obtained from all samples. A similarity matrix was generated for all
possible pairs of samples, and these similarity values were used to produce analysis of
similarity (ANOSIM) statistics to calculate the degree of separation between bacterial
communities among groups of samples (Clarke and Green 1988). ANOSIM generates a
test statistic, R. The magnitude of R indicates the degree of separation between groups
of samples, with a score of 1 indicating complete separation and 0 indicating no
separation.
Serial correlation was carried out to determine whether the observed patterns of
succession in bacterial communities were similar among the replicate experiments. The
similarity matrices for all samples were compared among experiments using the
Spearman rank correlation coefficient (ρ). High ρ-value would suggest that patterns of
change in bacterial community composition in each replicate experiment were very
similar.
Calculation of similarity coefficients, analysis of similarity and comparison between
observed patterns were carried out using PRIMER 6 for Windows v. 6.1.10 (PRIMER-E
Ltd, Plymouth, UK). The relevant subroutines were ANOSIM and RELATE.
Correspondence analysis (CA) was conducted using bacterial community composition
data generated by ARISA and glcD T-RFLP to search for patterns in bacterial
community composition by varying light and phosphorus treatments. Partial
correspondence analyses were carried out to explore patterns in bacterial community
23
composition following removal of variance attributed to select factors. All
correspondence analyses were carried out using the CANOCO for Windows software
package 4.5.1 (Biometris-Plant Research International, Wageningen, The Netherlands)
(ter Braak and Smilauer 2002).
Mean centroid distance based on CA is a measure of variation in a cluster of points.
Sample scores from the first and second CA axes were used to calculate the distance
from each point to the center of a cloud. Lower mean centroid distance represents less
dispersion (greater community similarity) within a classified group. Bacterial community
fingerprints were used to generate CAs, and mean centroid distance was calculated
within each group of light and phosphorus treatment levels.
24
CHAPTER 3
RESULTS
3.1 Experimental treatments
Gradients of light and phosphorus were manipulated in all three experiments (Table 1).
Measured soluble reactive phosphorus (SRP) concentrations were very close to target
phosphorus concentrations at lower treatment levels, but variability increased with
treatment levels. Light treatments generated a range of irradiances representing a
gradient from highly shaded stream environments (less than 20 µmol photons m-2s-1) to
above photosaturation (more than 350 µmol photons m-2s-1). Distinct levels of both
phosphorus and light were successfully created in all three experiments (Hill et al. 2009).
3.2 Total bacterial community composition (ARISA)
3.2.1 Bacterial community dynamics
Patterns of succession in periphyton-associated bacterial communities were observed
during biofilm development regardless of community variation among treatments.
Correspondence analysis showed that bacterial composition changed over time. A
significant difference was detected among community composition for each sampling
date (ANOSIM R=0.32, P<0.001). Difference among sample dates was more
pronounced when only samples within a single replicate experiments were considered
(replicate 1: ANOSIM R=0.56, P<0.001; replicate 2: ANOSIM R=0.48, P<0.001) (Fig 2).
The third replicate experiment was excluded from analyses of bacterial community
dynamics because tile samples were not collected until day 10 during experimental
25
period. Bray-Curtis similarity coefficients indicated that bacterial communities observed
on the final samples were 29- 35% similar to those observed in the initial samples. In
comparison, bacterial communities harvested on the first day were 51- 55% similar to
each other from the same sampling date. At the end of the sampling period, there was
46-50% similarity among communities collected on the final sample date. Spearman
correlation coefficient (ρ) indicated that patterns of succession were significantly similar
among replicate (ρ=0.47, P<0.001).
3.2.2 Succession of bacterial community across light levels
Correspondence analysis plots showed that biofilm bacterial communities exhibited
different patterns of succession across different light levels (Fig 3). At the initial stage of
biofilm development, bacterial communities under high light and low light levels were
quite similar, but the communities in these contrasting light treatments diverged over
time, and were significantly different at the end of the sampling period (ANOSIM
R=0.597, P<0.001). On the first sample date, the average similarity between biofilm
bacterial communities under high and low light levels was 61%; community similarity
between treatments was reduced to 36% of similarity on the final day of the experiment.
Plots of centroids representing samples collected on the same day from the same light
level (Fig 3B) demonstrate that communities grown under contrasting light levels
followed different trajectories. Biofilm bacterial communities that develop under high
light conditions showed more community variability among all treatments than bacterial
communities grown in low light conditions. Distance between centroids of high-light or
low-light samples were generated for each time point. Differences in microbial
26
community composition between high-light and low-light treatments increased over the
sampling period (Fig 4).
3.2.3 Succession of bacterial community across phosphorus levels
In contrast to patterns observed under light treatments, communities from different P
treatments tended to become more similar over time (Fig 5A). The dispersion of
communities was estimated by measuring maximum distances in the CA plot between
samples from all light treatments and the centroid for the group of samples collected
from high or low P levels at each time point (Fig 5B). We observed that bacterial
communities exposed to the same phosphorus concentration were more dispersed at
the beginning of biofilm development and became more similar over time, regardless of
P concentrations. Not only do the differences among samples experiencing the same P
concentration decline over study period, but community differences between
phosphorus concentrations also decrease. While the low-phosphorus communities were
more variable across the range of light treatments included in the experiment, the
patterns of succession across phosphorus treatments followed a similar trajectory.
Decreasing distances were observed between community centroids from contrasting
phosphorus concentrations (Fig 6).
3.2.4 Light and phosphorus effects on total bacterial community
Composition and variability of periphyton-associated bacterial communities were
influenced by light availability. Correspondence analysis indicates distinct bacterial
communities developed under contrasting light intensity (Fig 7A, Fig 3A). To explore
27
whether algal response to light availability was linked to variation in bacterial
communities, we used algal biomass as a supplemental variable in our CAs. Light and
algal biomass were highly correlated with axis 1, indicating an indirect influence of light
availability on the bacterial communities through the algal responses. Bacterial
community samples at high light levels tend to be more variable across P levels
(particularly lower P concentrations) (Fig 7A). In order to determine bacterial community
variability among light treatments, mean centroid distance was calculated for
communities collected from each light level across all P levels. Variability in bacterial
communities increased with increasing light intensity. The community variability across
all P levels, indicated by mean centroid distances within each group of samples was
logarithmically correlated with light intensity (R2=0.98) (Fig 7B).
Compared to light treatments, the influence of phosphorus concentrations on variability
in bacterial community was less pronounced (Fig 8A). Distinct clusters of bacterial
communities were observed between contrasting phosphorus treatment levels,
suggesting that composition of bacterial community can be influenced by phosphorus
concentrations, but variability in community composition is similar among the highest
and lowest P levels. Mean centroid distance among bacterial communities decreased
slightly with increasing P concentrations (R2=0.35), indicating highest variability among
microbial communities occurred at low phosphorus concentrations (Fig 8B).
3.2.5 Effects of four extreme light and P ratios on bacterial community
28
The effects of light and phosphorus were highly interactive and consideration of the
combined effects of the two variables was important for examining the responses of
bacterial communities to both light and phosphorus. Differences in community
composition were observed under four extreme light: P ratios (Fig 9). In the CA plot,
differences in community composition were observed along axis 1, across contrasting
light levels. Some distinction in community composition was also apparent phosphorus
concentrations (correlated with Axis 2). Light and algal biomass are highly correlated
with axis 1 indicating light is a stronger factor than phosphorus for shaping biofilm
community structure. Greater variation in bacterial community was observed in high
light treatments, no patterns in community variation related to phosphorus treatments
were apparent.
3.3 Glycolate-utilizing bacterial community composition (glcD T-RFLP)
After observing differences in bacterial community composition as a result of light and
phosphorus levels and patterns of community succession during biofilm development,
we conducted correspondence analysis to assess the impact of experimental
treatments on specific periphyton-associated bacterial groups across varying light: P
ratios. T-RFLP profiles based on the glcD gene were used to compare development of
heterotrophic bacterial communities among different light: P treatments.
3.3.1 Individual and combined effects of light and phosphorus on glycolate-utilizing
bacterial community
29
Correspondence analysis showed differences in potential glycolate-utilizing bacteria at
contrasting light intensity and phosphorus concentrations (Fig 10). To explore whether
the algal responses to light availability were linked to variation in the heterotrophic
bacterial communities, we used algal biomass and chlorophyll a as supplemental
variables in our correspondence analysis. We observed a strong correlation between
algal biomass and light while chlorophyll a was highly correlated with phosphorus (as
previously reported (Hill et al. 2009). Compared to the individual effects of light and
phosphorus, we showed that the combined effects of light and phosphorus produced a
more distinct difference in glycolate-utilizing bacterial populations (Fig 11). Two extreme
light: P ratios indeed resulted in distinctly different communities of glycolate-utilizing
bacteria. This result indicates that changes in the light and nutrients available to algal
biofilms can influence the bacterial portion of the community.
30
CHAPTER 4
DISCUSSION
4.1 Experimental streams and treatments
Differences in bacterial responses to environmental conditions would be expected to
impact microbial community composition and consequently influence nutrient cycling
and biogeochemical transformation processes. Strong biological interactions between
algae and bacteria influence microbial community composition. Stream biofilms provide
an interesting system with which to examine algal-bacterial interactions. Such systems
are characterized by highly efficient nutrient retention and carbon transformation carried
out within the biofilm community (Romani et al. 2004, Olapade and Leff 2006),
indicating a strong interaction between algal and bacterial portions of the community.
However, there is an inevitable tradeoff between precision and natural context when
using experimental streams and ceramic tiles as mats for colonization of benthic algae
and bacterial populations. There are potential limitations of experimental manipulations
of light and phosphorus treatments within the same stream reach. Previous work has
shown that activities such as nutrient cycling and DOC release within headwater-stream
biofilm communities can significantly affect nutrient supply to downstream communities
(Lyon and Ziegler 2009). In the current study, phosphorus was applied to the
experimental system from the head of each reach. Differences in organic carbon export
and nutrient release from upstream biofilm communities may influence uptake of
organic carbon and nutrients by downstream communities, potentially masking effects
of the experimental treatments. As a result, bacterial community composition may be
31
correlated with surrounding algae and also respond to composition or activity of the
upstream biofilm communities.
4.2 Molecular fingerprinting techniques
Our analysis of bacterial community composition was based on relative abundance of
individual populations derived from PCR-amplified community fingerprints. As with all
molecular studies reliant on amplification of nucleic acids with polymerase chain
reaction (PCR), microbial community fingerprinting methods such as ARISA and T-RFLP
may be biased by the primer set chosen with different amplification efficiency (Forney et
al. 2004). However, ARISA profiles based on proportional fluorescence have recently
been used to successfully describe bacterial community composition over time and
space in a variety of aquatic habitats (Yannarell and Triplett 2005, Kent et al. 2007,
Shade et al. 2007). The high-throughput nature of the technique makes it a valuable tool
despite its limitations (Jones et al. 2008).
4.3 Bacterial community dynamics
Previous results (Hill et al. 2009) showed chlorophyll a and biomass of benthic algae
accrued steadily over time in each replicate experiment. In our study, bacterial
community composition changes during development of biofilms in all treatments,
suggesting that the dynamics in bacterial community composition are correlated with
changes in composition or abundance of benthic algal assemblages in stream biofilms.
We observed similar patterns of succession in bacterial community among replicate
32
experiments, and the patterns of succession in bacterial community composition were
reproducible.
4.3.1 Patterns of succession at contrasting light and phosphorus treatments
Bacterial community composition was highly variable over study period in both light and
phosphorus treatments. Under contrasting light treatments, distinct patterns of
succession in bacterial community composition were detected (Fig 3). Correspondence
analysis revealed distinct trajectories of succession in bacterial community composition
under low and high light conditions. In addition, greater variation in bacterial community
composition was observed under high light (Fig 4). During the study period, benthic
algal biomass increased with time and accumulated rapidly under high light intensity
(Hill et al. 2009). The greater temporal changes in bacterial community composition
under high light may reflect the rapid accumulation of algal community. In contrast,
bacterial community composition showed similar temporal patterns at contrasting
phosphorus concentrations (Fig 5) and variation in community structure among
treatments decreased through time at both high and low P concentrations (Fig 6). Our
study supports the hypothesis that available light plays an important role structuring
bacterial community composition; phosphorus had a less significant impact on bacterial
community structure than light. Interestingly, bacterial community composition tends to
become similar at the end of study period even for communities exposed to extremely
different phosphorus treatments. A model for the development of bacterial biofilm
assemblage suggested by (Jackson et al. 2001) provides a possible explanation for our
observation. At the initial stage of biofilm development (2-7 days), bacterial
33
assemblages are likely to be dynamic with no orderly arrangement of community
structure. Following the early stage, community structure may simplify as superior
competitors begin to dominate (15-30 days). One possibility for the decreasing diversity
in our study is that specific bacterial populations were able to consume major part of
resources, and then dominated the community during biofilm development. Indeed, we
observe that bacterial community variability decreases over time. However, the study
period only lasted for a maximum of 18 days. We have insufficient data to understand
long-term patterns of changes in bacterial assemblages associated with stream biofilms.
4.4 Light and phosphorus effects on algal assemblages and bacterial community
composition
The simultaneous manipulation of phosphorus and light provided to the experimental
biofilms offered an opportunity to explore the individual and combined effects of these
two factors. We observed high variability in community composition at high light
intensity and distinct community structure among light treatments (Fig 7), implying
available light can release constraints on the periphyton-associated microbial
community. A potential mechanism for this result is presented in Fig 1, showing that
differences in algal physiology and exudate production between phosphorus levels are
most apparent at high light levels, resulting in large variability in bacterial communities
across all phosphorus treatments under high light (Fig 3B). Previous studies have
demonstrated that the effects of a range of phosphorus concentrations on benthic algal
biovolume is more pronounced at higher light intensity (Hill et al. 2009). Under relative
high light intensity (or low phosphorus concentration), algae tend to release greater
34
amounts of organic carbon compounds (van Hannen et al. 1999) due to an imbalance of
light and phosphorus in algal cells. However, when high light and high phosphorus
levels are available to periphyton communities, primary production is routed into algal
biomass because there is more phosphorus available (compared to high light and low
P) to support accumulation of algal biomass. Differences in algal responses to different
phosphorus levels, especially under high light levels, result in contrasts in exudate
production and can contribute to the variability of bacterial community composition.
Distinct clusters of bacterial communities between contrasting phosphorus
concentrations, and higher variability in bacterial communities occurring at low
phosphorus suggests that the influence of the range of light values on bacterial
community composition is most marked at low P levels. This result illustrates algal and
bacterial responses to varying light: P ratios. At low phosphorus and high light (Fig 1),
algal biomass and exudates production are expected to be greater, compared to low
light and low phosphorus conditions. Higher diversity of C sources and increased biofilm
area to accommodate bacteria can result in higher variability in composition and
structure of bacterial community in stream biofilms. Our results also support the
hypothesis that nutrient enrichment tends to disrupt the tight relationship between
primary and secondary production, where bacterial cells favor use of algal-released
DOC as a labile carbon sources when nutrients are limited (Scott et al. 2008). At high
phosphorus levels, the fate of excess phosphorus is uptake and storage by algal cells
and/or uptake directly by the bacterial portion of the biofilm community.
35
The observed differences in community composition among four extreme ratios of light
and phosphorus levels also provided evidence that the effects of light were greater than
phosphorus (Fig 9). The bacterial community was significantly different among light
treatments, and less significantly separated by phosphorus treatments, indicating that
differences in bacterial community composition in stream biofilms were more influenced
by light.
4.5 Effects of light and phosphorus on heterotrophic bacterial community composition
We hypothesized that varying light: P ratios would result in differences in glycolate
production. Exudate production is sensitive to light and nutrient availability (BermanFrank and Dubinsky 1999, Hessen et al. 2002) (Fig 1). Heterotrophic bacterial
populations are not able to use light directly as an energy source. Rather, trophic
connections among heterotrophs and primary producers are responsible for the
changes in bacterial community structure across treatments. Abundance and quality of
exudates produced by algal communities is thought to strongly influence bacterial
community composition (Perez and Sommaruga 2006). Particularly, because glycolate
production is sensitive to light (Leboulanger et al. 1998, Lau and Armbrust 2006, Lau et
al. 2007), excess available light relative to phosphorus may promote algal glycolate
production, and indirectly influence heterotrophic bacterial community composition via
exudate consumption. Indeed, we observed distinct glycolate-utilizing bacterial
community composition across varying light and phosphorus treatments (Fig 10).
ANOSIM results showed that R=0.131 (P=0.002) at contrasting light levels, and
R=0.122 (P=0.006) at contrasting phosphorus levels. Both light and phosphorus are
36
important factors governing exudate production. The light: nutrient hypothesis proposes
that algal carbon: nutrient ratios are driven by the ratio of available light and nutrients
(Sterner et al. 1997). Our results demonstrated that the differences in algal community
resulting from light and phosphorus levels are reflected in heterotrophic bacterial
community composition (Fig 10). In nutrient-poor (relatively high light) streams, algalderived organic compounds (including glycolate) are more likely to be consumed by
heterotrophic bacteria (Lyon and Ziegler 2009). The distinct differences in heterotrophic
bacterial assemblages observed between the highest and lowest light: P ratios provide
strong evidence for combined effects of light and P on bacterial community composition
(Fig 11).
37
CHAPTER 5
CONCLUSIONS
This study has demonstrated that bacterial communities respond to changes in biofilm
community composition or biomass of periphyton, and distinct bacterial communities
were observed across varying light and phosphorus treatments. Our results suggest
that benthic algae may contribute to those differences in bacterial community
composition by responding differentially to light and phosphorus. We also observed
variation in a specific group of heterotrophic bacteria, glycolate-utilizing bacteria
responding to algal exudates, across light and phosphorus treatments. Despite the lack
of direct observation on bacterial uptake of algal-released glycolate, differences in
bacterial community composition due to treatments of light and phosphorus support the
hypothesis that algal response to environmental conditions indeed influences bacterial
community via food web connections. Stable isotopes of carbon (13C) may provide a
powerful tool to track carbon flow through food web from primary producers to
consumers (Post 2002), allowing us to evaluate interactions between algae and
heterotrophic bacteria more directly. Identification of the drivers influencing interactions
between bacteria and algae is necessary for understanding dynamics and composition
of microbial communities, which is a powerful and important implication for predicting
ecosystem responses to global changes.
38
TABLES AND FIGURES
A
!"#$%&"#$'%
+,#-.%
+,#-,%
./01-'.2%
3-4'.5"-%
&()%&"#$'%
. *- ( ,& + *) ! ( ' ' & %! $ #" ! / 0
&()%*%
!"#$%*%
/122
*%
&
3142
31/2
2125
2126
&&&&&2
!"#$#!% !"#$#"% ""#$#!% ""#$#"%
Figure 1. (A) Predicted responses of algal biomass, exudates production under four extreme
light: P ratios. (B) Algal biomass (mg/cm2) under four extreme light: P ratios (HL: high light, HP:
high phosphorus, LL: low light, LP: low phosphorus). Fig 1B were adapted from Hill and Fanta
(2008) Freshwater Biology.
39
Table 1. Soluble reactive phosphorus (SRP) and light treatments in the three stream
experiments. Values listed are mean ± standard deviations. Light data are photosynthetically
active radiation (PAR).
Parameter
Treatment
Experiment 1
Experiment 2
Experiment 3
Mean
Phosphorus (µg L-1)
6
12
25
75
150
300
4±2
10±1
24±2
76±6
173±10
329±44
6±2
12±1
24±3
68±2
178±34
292±27
6±2
12±2
26±2
66±10
134±11

5
11
24
69
162
310
+
0
1
2
3
384±17
100±11
58±7
31±6
18±3
367±24
109±14
58±4
34±6
20±2
365±17
108±16
70±9
42±6
21±2
372
106
62
36
20
2 -1
Light (µmol m- s )
Reprinted from Hill and Fanta et al. (2009) Limnology and Oceanography
40
1.3
Axis 2 (4.9%)
Day
4
6
8
10
12
14
-0.5
16
18
-1.2
Axis 1 (8.9%)
1.5
Figure 2. ARISA community fingerprints were generated for all samples in the first two replicate
experiments from periphyton biofilm. Correspondence analysis shows bacterial communities
change over time. Points represent periphyton-associated bacterial community samples
harvested every two days started on the 4th day of algal growth after treatment application.
41
"#$
!"#.
%&'()/)*$#0,-
%
%&'()")*+#+,-
"#$
"#$
!"#$
2
%&'()/)*$#0,-
High Low Day
0
1
+
".
"/
"0
!"#.
"1
"+
!"#$
%&'()")*+#+,-
"#$
Figure 3. Patterns of succession in bacterial communities over time at highest (375 µmol m-2s-1)
and lowest (20 µmol m-2s-1) light treatment levels. (A) Partial correspondence analysis (pCA,
distinction between replicate experiments was removed) shows bacterial communities change
over time, but bacterial community under highest and lowest light have different trajectories. (B)
Correspondence analysis plot with centroids of samples from high and low light levels collected
on each sample date. The bars show the maximum distance from samples to the centroid.
Bacterial communities at the highest light levels are symbolized by solid circles, at the lowest
levels are by open circles. The color indicates the date on which the sample was collected after
treatment application.
42
5174382."9.4:..8"2.84-;1<7
,%(
-./01234.",
-./01234."*
,%+
! "#"$%()
,%'
,%*
! "#"$%&'
,
$%(
$%+
$%'
$%*
$
$
'
+
(
,$
,*
,'
,+
,(
536
Figure 4. Comparison of bacterial communities over time between contrasting light treatments
in replicate experiment 1 ( ) and replicate experiment 2 ( ). Centroid distances were
calculated between high and low treatment levels for each day. These centroids were calculated
from the correspondence analyses shown in Fig 3.
43
"#%
!"#$
&'()*-*+%#,./
&
&'()*"*+,#-./
"#%
!"#$
"#%
&'()*-*+%#,./
3
High Low Day
0
1
2
"$
""0
!"#$
"1
!"#$
"2
&'()*"*+,#-./
"#%
Figure 5. Patterns of succession in bacterial communities over time at highest (300 µg L-1) and
lowest (6 µg L-1) phosphorus concentrations. (A) Partial correspondence analysis (distinction
between replicates was removed in analysis) shows bacterial communities change over time.
(B) Correspondence analysis plot with centroids of samples collected from the same day of
phosphorus levels. The bars show the maximum and minimum distance from samples to the
centroid. Bacterial communities at the highest light levels are symbolized by solid circles, at the
lowest levels are by open circles. The color indicates the date on which the sample was
collected after treatment application.
44
2-30/4.*15*06**41.*40)7-83
'"#
)*+,-./0*1'
)*+,-./0*1#
'
!"&
!"%
:#;!"<'
!"$
:#;!"'%
!"#
!
!
$
%
&
'!
'#
'$
'%
'(
2/9
Figure 6. Comparison of bacterial communities over time between contrasting phosphorus
concentrations in experiment 1 ( ) and 2 ( ). Centroid distances were calculated between high
and low treatment levels for each day. These centroids were calculated from the
correspondence analyses shown in Fig 5.
45
%#$
&
4(567*
+!89:*;6979<)*8!%)!"/
%$
&'()*%*+0#1./
23-
&:5?:
F(98?))
!"#-
4(567
!"#$
&'()*"*+,#-./
=>?<*@><7A9(B*C()7?<D>*
$%,"
%#$
E
! "#"$%&'"
$%+"
$%*"
$%)"
$%("
$"
$"
,$"
($$"
(,$"
)$$"
),$"
*$$"
*,$"
+$$"
4(567*+!89:*;6979<)*8!%)!"/
Figure 7. Light effects on bacterial community composition (ARISA). (A) Samples under highest
(375 µmol m-2s-1) and lowest (20 µmol m-2s-1) light intensity. Partial correspondence analysis
(pCA, after removing effects of experiment and day) shows distinct bacterial communities
develop under contrasting light levels. A solid circle represents community at high light, and an
open circle represents a community at low light. Bubble size indicates relative phosphorus
concentrations. (B) Mean centroid distance was calculated to assess bacterial community
variability among light treatments (20, 40, 60, 100, and 375 µmol m-2s-1). Bacterial community
variability increased with increasing light intensity (variability of bacterial community was
logarithmically related to light intensity, R2=0.98).
46
"#%
&
3*+!4*5!"/
,
2$$
&'()*0*+%#1./
&H49H*
I(>J9))
!"#$
3E>)FE>=G)
789:*;8:<=>(?*@()<9:A8
!"#$
&'()*"*+,#-./
"#%
D
$#%
$#6
B0C*$#2%
$#2
$#0
$#"
$
$
%$
"$$
"%$
0$$
0%$
2$$
2%$
3*+!4*5!"/
Figure 8. Phosphorus effects on bacterial community composition (ARISA). (A) Partial
correspondence analysis (experiment and time effects removed in the analysis) shows distinct
bacterial communities develop under contrasting P levels. Samples at highest P (300 µg L-1)
and lowest P (6 µg L-1) A solid circle represents community at high P concentration, and an open
circle represents a community at low P concentration. Bubble size indicates relative light
intensity. (B) Mean centroid distance was calculated to assess bacterial community variability
among phosphorus treatments (6, 12, 25, 75, 150, and 300 µg L-1). Bacterial community
variability had a negative logarithmical correlation with P concentration (R2=0.35).
47
"#%
8-98&
8-9-&
--98&
1:.)7<7;"$#@>?
--9-&
-./'0
12/32
4.(53))
6'273
!"#$
&'()*'(+,)
!"#$
1:.)7"7;"<#=>?
"#$
Figure 9. Partial correspondence analysis (pCA, after removing the effects of experiment and
day) of bacterial community composition with varying light: P ratios. Samples were collected on
the last two days of biofilm development. Arrows for the supplemental variables included in this
analysis show the direction of increase for each variable, and the length of each arrow indicates
the degree of correlation with the ordination axes. High Light (HL, 375 µmol m-2s-1), Low light
(LL, 20 µmol m-2s-1), High P (HP, 300 µg L-1), low P (LP, 6 µg L-1). Abbreviation is shown in
Figure 1.
48
"#$
%
Light (!874)8!.(!")
0'123
.:
%4154
6'785((
%&'().)*/#/,-
;<$
!"#$
924)5
%&'()")*+#",-
"#$
C
=)*!1A0B
300
%&'().)*+#:,-
"#$
!"#$
924)5
!"#:
%4154)6'785((
=27(>27?@(
!"#:
"#$
%&'()")*".#$,-
Figure 10. Light and phosphorus effects on community composition of glycolate-utilizing
bacteria in the last two days of replicate experiment 1 and 2. (A) Partial correspondence
analysis (pCA, after removing effects of experiment) of glycolate-utilizing community
composition under high light (375 µmol m-2s-1) and low light (20 µmol m-2s-1) intensity. (B) Partial
correspondence analysis (experiment as covariable) at high P (300 µg L-1) and low P (6 µg L-1)
concentrations. pCA plots show distinct communities under contrasting light levels (A) and
phosphorus (B).
49
"#%
-(842./'0439:4'./'4&
-./'0
&'()*'(+,)
12/3245.(63))
7'243
!"#$
1<.)4A4=A%#A?@
;./'42./'0439:42(84&
!"#$
1<.)4"4=>"#%?@
"#%
Figure 11. Partial correspondence analysis (pCA, after removing the effects of experiment and
day) of glycolate-utilizing community composition at high (Light: 375 µmol m-2s-1; P: 6 µg L-1)
and low (light: 20 µmol m-2s-1; P: 300 µg L-1) light: P ratios for biofilm samples collected on the
last days of biofilm development in each replicate experiment. Algal biomass is highly correlated
with light while chlorophyll a correlated with phosphorus.
50
REFERENCES
Armstrong, S. M., and F. Barlocher. 1989. Adsorption and release of amino acids from
epilithic biofilms in streams. Freshwater Biology 22:153-159.
Arrieta, J., and G. Herndl. 2002. Changes in bacterial β-glucosidase diversity during a
coastal phytoplankton bloom. Limnology and Oceanography 47:594-599.
Azam, F., T. Fenchel, J. G. Field, J. S. Gray, L. A. Meyerreil, and F. Thingstad. 1983.
The ecological role of water-column microbes in the sea. Marine EcologyProgress Series 10:257-263.
Baines, S. B., and M. L. Pace. 1991. The production of dissolved organic matter by
phytoplankton ant its importance to bacteria-patterns across marine and
freshwater systems. Limnology and Oceanography 36:1078-1090.
Bentzen, E., and W. D. Taylor. 1991. Estimating organic phosphorus utilization by
freshwater plankton using (P-32)ATP. Journal of Plankton Research 13:12231238.
Berman-Frank, I., and Z. Dubinsky. 1999. Balanced growth in aquatic plants: Myth or
reality? Phytoplankton use the imbalance between carbon assimilation and
biomass production to their strategic advantage. Bioscience 49:29-37.
Carpenter, S. R., J. J. Cole, M. L. Pace, M. Van de Bogert, D. L. Bade, D. Bastviken, C.
M. Gille, J. R. Hodgson, J. F. Kitchell, and E. S. Kritzberg. 2005. Ecosystem
subsidies: Terrestrial support of aquatic food webs from C-13 addition to
contrasting lakes. Ecology 86:2737-2750.
Carpenter, S. R., J. F. Kitchell, K. L. Cottingham, D. E. Schindler, D. L. Christensen, D.
M. Post, and N. Voichick. 1996. Chlorophyll variability, nutrient input, and grazing:
Evidence from whole-lake experiments. Ecology 77:725-735.
Carr, G. M., A. Morin, and P. A. Chambers. 2005. Bacteria and algae in stream
periphyton along a nutrient gradient. Freshwater Biology 50:1337-1350.
Clarke, K. R., and R. H. Green. 1988. Statistical design and analysis for a 'biological
effects' study. Mar Ecol Prog Ser:213-226.
Cole, J. J. 1982. Interactions between Bacteria and Algae in Aquatic Ecosystems.
Annual Review of Ecology and Systematics 13:291-314.
Cotner, J. B., and B. A. Biddanda. 2002. Small players, large role: Microbial influence
on biogeochemical processes in pelagic aquatic ecosystems. Ecosystems 5:105121.
Crump, B. C., and J. E. Hobbie. 2005. Synchrony and seasonality in bacterioplankton
communities of two temperate rivers. Limnology and Oceanography 50:17181729.
Crump, B. C., G. W. Kling, M. Bahr, and J. E. Hobbie. 2003. Bacterioplankton
community shifts in an arctic lake correlate with seasonal changes in organic
matter source. Applied and Environmental Microbiology 69:2253-2268.
Currie, D. J., and J. Kalff. 1984. A comparison of the abilities of freshwater algae and
bacteria to acquire and retain phosphorus. Limnology and Oceanography
29:298-310.
Dickman, E. M., M. J. Vanni, and M. J. Horgan. 2006. Interactive effects of light and
51
nutrients on phytoplankton stoichiometry. Oecologia 149:676-689.
Ducklow, H. W., and C. A. Carlson. 1992. Oceanic bacterial production. Advances in
Microbial Ecology 12:113-181.
Edenborn, H. M., and C. D. Litchfield. 1987. Glycolate turnover in the water column of
the New York Bight Apex. Marine Biology 95:459-467.
Elser, J. J., M. E. S. Bracken, E. E. Cleland, D. S. Gruner, W. S. Harpole, H. Hillebrand,
J. T. Ngai, E. W. Seabloom, J. B. Shurin, and J. E. Smith. 2007. Global analysis
of nitrogen and phosphorus limitation of primary producers in freshwater, marine
and terrestrial ecosystems. Ecology Letters 10:1135-1142.
Elser, J. J., P. Frost, M. Kyle, J. Urabe, and T. Andersen. 2002. Effects of light and
nutrients on plankton stoichiometry and biomass in a P-limited lake.
Hydrobiologia 481:101-112.
Elser, J. J., M. Kyle, W. Makino, T. Yoshida, and J. Urabe. 2003. Ecological
stoichiometry in the microbial food web: a test of the light : nutrient hypothesis.
Aquatic Microbial Ecology 31:49-65.
Espeland, E. M., and R. G. Wetzel. 2001. Effects of photosynthesis on bacterial
phosphatase production in biofilms. Microbial Ecology 42:328-337.
Felip, M., M. L. Pace, and J. J. Cole. 1996. Regulation of planktonic bacterial growth
rates: The effects of temperature and resources. Microbial Ecology 31:15-28.
Fisher, M. M., J. L. Klug, G. Lauster, M. Newton, and E. W. Triplett. 2001. Effects of
resources and trophic interactions on freshwater bacterioplankton. Microbial
Ecology 41:82-82.
Fogg, G. E. 1983. The ecological significance of extracellular products of phytoplankton
photosynthesis. Botanica Marina 26:3-14.
Forney, L. J., X. Zhou, and C. J. Brown. 2004. Molecular microbial ecology: land of the
one-eyed king. Current Opinion in Microbiology 7:210-220.
Fuhrman, J. A., and F. Azam. 1982. Thymidine incorporation as a measure of
heterotrophic bacterioplankton production in marine surface waters - evaluation
and field results. Marine Biology 66:109-120.
Haack, T. K., and G. A. McFeters. 1982. Nutritional relationships among
microorganisms in an epilithic biofilm community. Microbial Ecology 8:115-126.
Hahn, M. W., and M. G. Hofle. 2001. Grazing of protozoa and its effect on populations
of aquatic bacteria. Fems Microbiology Ecology 35:113-121.
Hessen, D. O., P. J. Faerovig, and T. Andersen. 2002. Light, nutrients, and P: C ratios
in algae: Grazer performance related to food quality and quantity. Ecology
83:1886-1898.
Hessen, D. O., E. Leu, P. J. Faerovig, and S. F. Petersen. 2008. Light and spectral
properties as determinants of C:N:P ratios in phytoplankton. Deep-Sea Research
Part Ii-Topical Studies in Oceanography 55:2169-2175.
Hill, W. R., and S. E. Fanta. 2008. Phosphorus and light colimit periphyton growth at
subsaturating irradiances. Freshwater Biology 53:215-225.
Hill, W. R., S. E. Fanta, and B. J. Roberts. 2009. Quantifying phosphorus and light
effects in stream algae. Limnology and Oceanography 54:368-380.
Hill, W. R., P. J. Mulholland, and E. R. Marzolf. 2001. Stream ecosystem responses to
forest leaf emergence in spring. Ecology 82:2306-2319.
Hill, W. R., M. G. Ryon, and E. M. Schilling. 1995. Light limitation in a stream ecosystem
52
- responses by primary producers and consumers. Ecology 76:1297-1309.
Howarth, R. W. 1988. Nutrient limitation of net primary production in marine ecosystems.
Annual Review of Ecology and Systematics 19:89-110.
Jackson, C. R., P. F. Churchill, and E. E. Roden. 2001. Successional changes in
bacterial assemblage structure during epilithic biofilm development. Ecology
82:555-566.
Jones, S. E., and M. A. Lock. 1993. Seasonal determinationss of extracellular hydrolytic
activities in heterotrophic and mixed heterotrophic autotrophtic biofilms from two
contrasting rivers. Hydrobiologia 257:1-16.
Jones, S. E., R. J. Newton, and K. D. McMahon. 2008. Potential for atmospheric
deposition of bacteria to influence bacterioplankton communities. Fems
Microbiology Ecology 64:388-394.
Jurgens, K., J. Pernthaler, S. Schalla, and R. Amann. 1999. Morphological and
compositional changes in a planktonic bacterial community in response to
enhanced protozoan grazing. Applied and Environmental Microbiology 65:12411250.
Kaplan, L. A., and T. L. Bott. 1989. Diel pluctuations in bacterial activity on streambed
during vernal algal blooms - Effects of temperature, water chemistry, and habitat.
Limnology and Oceanography 34:718-733.
Kent, A. D., S. E. Jones, G. H. Lauster, J. M. Graham, R. J. Newton, and K. D.
McMahon. 2006. Experimental manipulations of microbial food web interactions
in a humic lake: shifting biological drivers of bacterial community structure.
Environmental Microbiology 8:1448-1459.
Kent, A. D., S. E. Jones, A. C. Yannarell, G. H. Lauster, J. H. Graham, T. K. Kratz, and
E. W. Triplett. 2004. Annual patterns in bacterioplankton community variability in
a humic lake. Microbial Ecology 48:550-560.
Kent, A. D., A. C. Yannarell, J. A. Rusak, E. W. Triplett, and K. D. McMahon. 2007.
Synchrony in aquatic microbial community dynamics. Isme Journal 1:38-47.
Kirchman, D. L. 1994. The uptake of inorganic nutrients by heterotrophic bacteria.
Microbial Ecology 28:255-271.
Kritzberg, E. S., J. J. Cole, M. L. Pace, W. Graneli, and D. L. Bade. 2004.
Autochthonous versus allochthonous carbon sources of bacteria: Results from
whole-lake C-13 addition experiments. Limnology and Oceanography 49:588-596.
Lau, W. W. Y., and E. V. Armbrust. 2006. Detection of glycolate oxidase gene glcD
diversity among cultured and environmental marine bacteria. Environmental
Microbiology 8:1688-1702.
Lau, W. W. Y., R. G. Keil, and E. V. Armbrust. 2007. Succession and diel transcriptional
response of the glycolate-utilizing component of the bacterial community during a
spring phytoplankton bloom. Applied and Environmental Microbiology 73:24402450.
Leboulanger, C., V. Martin-Jezequel, C. Descolas-Gros, A. Sciandra, and H. J. Jupin.
1998. Photorespiration in continuous culture of Dunaliella tertiolecta
(Chlorophyta): Relationships between serine, glycine, and extracellular glycolate.
Journal Of Phycology 34:651-654.
Legendre, P., and L. Legendre. 1998. Numerical ecology. Second English edition
edition.
53
Lock, M. A., R. R. Wallace, J. W. Costerton, R. M. Ventullo, and S. E. Charlton. 1984.
River epilithon - Toward a structural-functional model. Oikos 42:10-22.
Lyon, D. R., and S. E. Ziegler. 2009. Carbon cycling within epilithic biofilm communities
across a nutrient gradient of headwater streams. Limnology and Oceanography
54:439-449.
Makino, W., and J. B. Cotner. 2004. Elemental stoichiometry of a heterotrophic bacterial
community in a freshwater lake: implications for growth- and resource-dependent
variations. Aquatic Microbial Ecology 34:33-41.
McNamara, C. J., and L. G. Leff. 2004. Response of biofilm bacteria to dissolved
organic matter from decomposing maple leaves. Microbial Ecology 48:324-330.
Methe, B. A., W. D. Hiorns, and J. P. Zehr. 1998. Contrasts between marine and
freshwater bacterial community composition: Analyses of communities in Lake
George and six other Adirondack lakes. Limnology and Oceanography 43:368374.
Meyer, J. L. 1994. The microbial loop in flowing waters. Microbial Ecology 28:195-199.
Morris, D. P., and W. M. Lewis. 1992. Nutrient limitation of bacterioplankton growth in
Lake Dillon, Colorado. Limnology and Oceanography 37:1179-1192.
Mulholland, P. J., E. R. Marzolf, S. P. Hendricks, R. V. Wilkerson, and A. K. Baybayan.
1995. Longitudinal patterns of nutrient cycling and periphyton characteristics in
streams - A test of upstream-downstream linkage. Journal of the North American
Benthological Society 14:357-370.
Munster, U. 1993. Concentrations and fluxes of organic-carbon substrates in the aquatic
environment. Antonie Van Leeuwenhoek International Journal of General and
Molecular Microbiology 63:243-274.
Olapade, O. A., and L. G. Leff. 2006. Influence of dissolved organic matter and
inorganic nutrients on the biofilm bacterial community on artificial substrates in a
northeastern Ohio, USA, stream. Canadian Journal of Microbiology 52:540-549.
Parker, M. S., and E. V. Armbrust. 2005. Synergistic effects of light, temperature, and
nitrogen source on transcription of genes for carbon and nitrogen metabolism in
the centric diatom Thalassiosira pseudonana (Bacillariophyceae). Journal Of
Phycology 41:1142-1153.
Parker, M. S., E. V. Armbrust, J. Piovia-Scott, and R. G. Keil. 2004. Induction of
photorespiration by light in the centric diatom Thalassiosira weissflogii
(Bacillariophyceae): Molecular characterization and physiological consequences.
Journal Of Phycology 40:557-567.
Paver, S. F., and A. D. Kent. 20__. Temporal dynamics of glycolate-utilizing bacteria
driven by phytoplankon communities in lakes. (submitted).
Pellicer, M. T., J. Badia, J. Aguilar, and L. Baldoma. 1996. glc locus of Escherichia coli:
Characterization of genes encoding the subunits of glycolate oxidase and the glc
regulator protein. Journal of Bacteriology 178:2051-2059.
Perez, M. T., and R. Sommaruga. 2006. Differential effect of algal- and soil-derived
dissolved organic matter on alpine lake bacterial community composition and
activity. Limnology and Oceanography 51:2527-2537.
Post, D. M. 2002. Using stable isotopes to estimate trophic position: Models, methods,
and assumptions. Ecology 83:703-718.
Rhee, G. Y. 1973. Continuous culture study of phosphate uptake, growth rate and
54
polyphosphate in Scenedesmus sp. Journal Of Phycology 9:495-506.
Rier, S. T., and R. J. Stevenson. 2002. Effects of light, dissolved organic carbon, and
inorganic nutrients on the relationship between algae and heterotrophic bacteria
in stream periphyton. Hydrobiologia 489:179-184.
Rier, S. T., and R. J. Stevenson. 2006. Response of periphytic algae to gradients in
nitrogen and phosphorus in streamside mesocosms. Hydrobiologia 561:131-147.
Rier, S. T., R. J. Stevenson, and G. D. LaLiberte. 2006. Photo-acclimation response of
benthic stream algae across experimentally manipulated light gradients: A
comparison of growth rates and net primary productivity. Journal Of Phycology
42:560-567.
Roberts, B. J., and R. W. Howarth. 2006. Nutrient and light availability regulate the
relative contribution of autotrophs and heterotrophs to respiration in freshwater
pelagic ecosystems. Limnology and Oceanography 51:288-298.
Romani, A. M., H. Guasch, I. Munoz, J. Ruana, E. Vilalta, T. Schwartz, F. Emtiazi, and
S. Sabater. 2004. Biofilm structure and function and possible implications for
riverine DOC dynamics. Microbial Ecology 47:316-328.
Romani, A. M., and S. Sabater. 2000. Influence of algal biomass on extracellular
enzyme activity in river biofilms. Microbial Ecology 40:16-24.
Scott, J. T., J. A. Back, J. M. Taylo, and R. S. King. 2008. Does nutrient enrichment
decouple algal-bacterial production in periphyton? Journal of the North American
Benthological Society 27:332-344.
Seitzinger, S. P. 1988. Denitrification in freshwater and coastal marine ecosystems ecological and geochemical significance. Limnology and Oceanography 33:702724.
Shade, A., A. D. Kent, S. E. Jones, R. J. Newton, E. W. Triplett, and K. D. McMahon.
2007. Interannual dynamics and phenology of bacterial communities in a
eutrophic lake. Limnology and Oceanography 52:487-494.
Sinsabaugh, R. L., D. Repert, T. Weiland, S. W. Golladay, and A. E. Linkins. 1991.
Exoenzyme accumulation in epilithic biofilms. Hydrobiologia 222:29-37.
Smith, V. H. 2003. Eutrophication of freshwater and coastal marine ecosystems - A
global problem. Environmental Science and Pollution Research 10:126-139.
Stepanauskas, R., H. Edling, and L. J. Tranvik. 1999. Differential dissolved organic
nitrogen availability and bacterial aminopeptidase activity in limnic and marine
waters. Microbial Ecology 38:264-272.
Sterner, R. W., and J. J. Elser. 2002. Ecological stoichiometry: the biology of elements
from molecules to the biosphere. Princeton University Press.
Sterner, R. W., J. J. Elser, E. J. Fee, S. J. Guildford, and T. H. Chrzanowski. 1997. The
light:nutrient ratio in lakes: The balance of energy and materials affects
ecosystem structure and process. American Naturalist 150:663-684.
Stets, E. G., and J. B. Cotner. 2008. The influence of dissolved organic carbon on
bacterial phosphorus uptake and bacteria phytoplankton dynamics in two
Minnesota lakes. Limnology and Oceanography 53:137-147.
ter Braak, C. J. F., and P. Smilauer. 2002. CANOCO reference manual and CanoDraw
for Windows user's guide: Software for canonical community ordination (version
4.5). Microcomputer Power, Ithaca, NY.
Urabe, J., M. Kyle, W. Makino, T. Yoshida, T. Andersen, and J. J. Elser. 2002. Reduced
55
light increases herbivore production due to stoichiometric effects of light/nutrient
balance. Ecology 83:619-627.
Vadeboncoeur, Y., D. M. Lodge, and S. R. Carpenter. 2001. Whole-lake fertilization
effects on distribution of primary production between benthic and pelagic habitats.
Ecology 82:1065-1077.
Vadstein, O., A. Jensen, Y. Olsen, and H. Reinertsen. 1988. Growth and phosphorus
status of limnetic phytoplankton and bacteria. Limnology and Oceanography
33:489-503.
van Hannen, E. J., W. Mooij, M. P. van Agterveld, H. J. Gons, and H. J. Laanbroek.
1999. Detritus-dependent development of the microbial community in an
experimental system: Qualitative analysis by denaturing gradient gel
electrophoresis. Applied and Environmental Microbiology 65:2478-2484.
Vannote, R. L., G. W. Minshall, K. W. Cummins, J. R. Sedell, and C. E. Cushing. 1980.
River continuum concept. Canadian Journal of Fisheries and Aquatic Sciences
37:130-137.
Vollenweider, R. A. 1976. Advance in defining critical load levels for phosphorus in lake
eutrophication. Mem. Ist. Ital. Idrobiol. 33:53-83.
Wetzel, R. G. 1992. Gradient-dominated ecosystems - Sources and regulatory functions
of dissolved organic matter in freshwater ecosystems. Hydrobiologia 229:181198.
Wetzel, R. G. 1993. Microcommunities and microgradients: Linking nutrient
regeneration, microbial mutualism, and high sustained aquatic primary
production Netherlands Journal of Aquatic Ecology 27:9.
White, P. A., J. Kalff, J. B. Rasmussen, and J. M. Gasol. 1991. The effect of
temperature and algal biomass on bacterial production and specific growth rate
in freshwater and marine habitats. Microbial Ecology 21:99-118.
Wingler, A., P. J. Lea, W. P. Quick, and R. C. Leegood. 2000. Photorespiration:
metabolic pathways and their role in stress protection. Philosophical
Transactions of the Royal Society B-Biological Sciences 355:1517-1529.
Wright, R. T., and N. M. Shah. 1977. The trophic role of glycolic acid In coastal
seawater Part 2. Seasonal changes in concentration and heterotrophic use in
Ipswich Bay, Massachusetts, USA. Marine Biology 43:257-263.
Yannarell, A. C., and E. W. Triplett. 2005. Geographic and environmental sources of
variation in lake bacterial community composition. Applied and Environmental
Microbiology 71:227-239.
Ylla, I., C. Borrego, A. M. Roman, and S. Sabater. 2009. Availability of glucose and light
modulates the structure and function of a microbial biofilm. Fems Microbiology
Ecology 69:42.
56
AUTHOR’S BIOGRAPHY
Yu-rui Chang graduated from National Sun Yat-Sen University at Kaohsiung, Taiwan in
2006 with a Bachelor of Science degree in Marine Sciences. Then, Chang relocated to
Urbana, Illinois to pursue graduate study in microbial ecology and worked as a research
assistant in Dr. Angela Kent’s Laboratory.
57