All about biofilms

Our research



About us






Research Abstracts

Why are there so many microbial species? What are they all doing? And, what can they tell is about the environment in which they live?: Exploring the biodiversity and biogeography of bacterial communities in New Zealand's rivers, lakes and streams.

Gavin Lear 1, Vidya Washington2, Andrew Dopheide2 & Gillian Lewis2

1Lincoln University, New Zealand
2The University of Auckland, New Zealand

New Zealand Microbiological Society Conference, Auckland, December 2010

Do aquatic freshwater bacterial communities exhibit biogeographical patterns at a national scale, and are these patterns predictable? In 2010, we began exploring the biogeographical distribution of aquatic bacteria, comparing the community structure of over 1,500 stream biofilm samples, collected across New Zealand. This nationwide study revealed that, similar to the distribution of various macroorganisms, the structure of stream biofilm bacterial communities (assessed using ARISA of bacterial 16S rRNA genes) is by no means random, but varies predictably among New Zealand's regions. At the regional scale, comparison of the ARISA data with various physico-chemical parameters recorded at each site revealed differences in bacterial community structure could be largely explained by the nature and extent of catchment development (e.g., conversion to urban or rural land-uses).
To identify the potential impact of the observed changes in bacterial community structure on freshwater ecological health, a subset of these samples were further interrogated using (i) high throughput 454 DNA sequencing of bacterial 16S rRNA genes, and (ii) GeoChip functional gene microarrays. The DNA sequence data revealed surprisingly little variability in the total abundance and diversity of bacterial genera between sites. However, significant differences were detected in the abundance and diversity of key functional bacterial genes (e.g., encoding for C degradation, N fixation, denitrification and resistance to various metals). Our findings indicate that freshwater biofilm communities may contain a high level of 'taxonomic redundancy', such that in response to various physico-chemical parameters, similar groups of bacterial taxa can possess very different functional capabilities.

Expression of rpoS gene in naturalized and commensal Escherichia coli.

Anne-Marie Perchec-Merien & Gillian D. Lewis

The University of Auckland, Auckland, New Zealand

New Zealand Microbiological Society Conference, Auckland, December 2010

This research intends to characterise naturalized strains of Escherichia coli found in water environments. E. coli is used as a faecal indicator in water quality monitoring programmes worldwide, with the assumption that this bacterium is exclusively a commensal of the vertebrate gut. Recent findings, recording growth and multiplication of E. coli in water and soils, either in tropical and temperate climates in different parts of the world, shown this is not the case. Although many studies demonstrate the existence of naturalized populations of E. coli, there is actually no method available to distinguish them from the commensal E. coli.
The stress factor σS, encoded by the gene rpoS, regulates about 10% of the genes in E. coli. This gene plays an important role in adaptive metabolic pathways when the bacteria enter in the stationary phase and under specific stress conditions. Therefore this gene can play a particularly important role in the adaptation of E. coli to natural environmental conditions, and in the process of naturalization.
To explore this hypothesis, comparative real-time PCR experiments are performed to assess the expression levels of the rpoS gene, in presumed naturalized and commensal E. coli. Preliminary analysis carried out on three strains showed that specific expression levels can help to differentiate one category or one other.
Therefore, this research project can be a valuable contribution to the identification of naturalized populations of E. coli. To be able to differentiate between environmental 'naturalized' E. coli and commensal E. coli will give important indications about their origins and will be highly valuable for microbial source tracking.

Comparison of methods for the extraction of DNA from stream epilithic biofilms.

Yimin Dong1, Gavin Lear2 & Gillian Lewis1

1School of Biological Sciences, University of Auckland, Auckland, New Zealand
2Department of Soil and Physical Sciences, Lincoln University, Christchurch, New Zealand

New Zealand Microbiological Society Conference, Auckland, December 2010

A biofilm is a well-organized community of microorganisms that adheres to surfaces. So far no study has attempted to compare the efficacy of extraction techniques in terms of the quality and quantity of DNA yielded from freshwater stream biofilm samples, or their effect on interpretations of DNA-based microbial community analysis.
In this study, the efficiency of six different DNA extraction methods including three methods that are both published and widely cited in peer-reviewed literature and three commercial kits were compared. The quantity and quality of DNA extracted using each technique was assessed using spectrophotometric analysis and agarose gel electrophoresis. The impacts of each extraction technique on interpretations of biological variability were then assessed using a DNA-based bacterial community fingerprinting technique (Automated Ribosomal Intergenic Spacer Analysis &emdash; ARISA). Significant differences were found in DNA yields, ranging from 56600 ± 31 to 1400 ± 1 (ng of DNA ± standard error). Methods including chloroform extraction after enzymatic lysis and/or chemical treatment resulted in the greater DNA yields and purer DNA extracts than the filtration-based commercial extraction kits. The increased variability in bacterial community structure among samples extracted using commercial kits were also observed using ARISA.
In conclusion, the commercial kits used in this study are not recommended for the extraction of DNA from stream biofilms as they yielded low concentrations of sheared DNA. Both of these factors may have contributed to the increased variability in bacterial community structure, observed using ARISA. The findings of this study identify the method of Zhou et al. (1996) as capable of yielding high concentrations of 'intact' DNA from bacterial communities associated with stream biofilms.

Using visual media to encourage interest in microbial ecology.

Andrew Dopheide & Gillian Lewis

The University of Auckland, Auckland, New Zealand

New Zealand Microbiological Society Conference, Auckland, December 2010

Public understanding of science is limited, and non-scientists may regard the findings of scientific research as perplexing, boring or irrelevant. Microbiology may be particularly unfathomable because microbial phenomena are generally far too small to be observed and understood without specialised equipment and techniques. Although the microbial biofilms found in aquatic environments are widely recognised as 'slime', few non-scientists understand that these biofilms are complex, interesting communities of micro-organisms with important ecological functions.
Visual media such as posters, illustrations and animations can provide an effective way of conveying awareness and understanding of scientific concepts and research findings to non-specialist audiences. Well designed visualisations may have an accessibility and immediacy which can encourage popular interest in microbiology and other areas of science research.
In order to convey an appreciation of microbial ecology to a wider audience, we developed a visualisation of life at the microbial scale. Various microscopy techniques were used to gather representative images of the micro-organisms found in stream biofilms. These were assembled into an eye-catching, detailed and scientifically accurate visualisation of a microbial biofilm community, using digital imaging techniques. Details about different micro-organisms were provided as text. The resulting poster has been distributed to schools, councils, community groups and individuals in New Zealand and internationally, and has been rewarded in an international science visualisation competition. This has contributed to successful generation of interest in microbiology research, and has made insights into the nature of microbial ecology and biofilms available to a wide audience.

A biofilm-based bacterial community index to track changes in stream ecosystem health.

Gillian Lewis1, Gavin Lear1, Martin Neale2, Vidya Washington1 & Vicky Fan3

1School of Biological Sciences, University of Auckland New Zealand
2Auckland Regional Council, New Zealand
3Bioinformatics Centre, University of Auckland, New Zealand

North American Benthological Society Conference, Sante Fe, June 2010

Bacteria form a significant part of stream communities with important roles in nutrient cycling, contaminant processing, and the fixation, cycling and transfer of energy in aquatic food webs. The bacterial composition of resident biofilm communities is responsive to external conditions including temperature, pH, water chemistry, nutrient supply and contaminants.
This study describes the development and broad scale testing of a bacterial community index, based on the bacterial composition of stream biofilms, as a tool for tracking and comparing stream ecosystem health.
A PCR based DNA community fingerprinting approach is used to record the presence and frequency of occurrence of bacteria within a biofilm sample. The relative occurrence of a range of peaks in the fingerprint is used to calculate the bacterial community index. Extensive evaluation is underway encompassing 1500 samples from 300 sites throughout New Zealand, including urban, rural, intensive agriculture and forested catchments.
The responsive nature of the bacterial community to change suggests that such an index will become an important tool for water managers to detect ecosystem threats before significant damage occurs to macro-organism communities.

Using stream biofilm microbial communities as indicators of freshwater ecosystem health.

Pierre-Yves Ancion, Gavin Lear, Kelly Roberts, Vidya Washington & Gillian Lewis

School of Biological Sciences, University of Auckland, New Zealand

13th International Symposium on Microbial Ecology, ISME 13, Seattle, USA, August 2010

Stream biofilms are a complex aggregation of microorganisms embedded in a polymer matrix and cover almost every surface in freshwater environments. Because of their sedentary way of life, microorganisms associated with biofilms are affected by past and present environmental conditions and therefore constitute a potential integrative indicator of stream health.
A wide range of experiments was conducted in both flow chamber microcosms and natural stream environments to investigate the main drivers of microbial community structure and composition and evaluate the potential use of biofilms as a bio-indicator of freshwater ecosystem health. Using community fingerprinting techniques such as terminal-Restriction Fragment Length Polymorphism and Automated Ribosomal Intergenic Spacer Analysis as well as 16S rRNA gene clone libraries we investigated variations occurring in biofilm bacterial and ciliate protozoan communities. Initial experiments conducted in flow chamber microcosms showed that significant differences in microbial community structure could be detected within only a few days of exposure to common water contaminants and remained detectable weeks after transfer to uncontaminated water.
Further research investigating biofilm of more than 60 stream sites variously impacted by urbanization revealed a strong separation between rural and urban streams and confirmed the potential use of stream biofilm as a bio-indicator of stream health. Environmental monitoring techniques developed in this project were then successfully tested to investigate the efficacy of an enclosed stormwater treatment system, where traditional biological indicators such as macro-benthic invertebrates were not available. We are now extending our research to 300 different streams in order to define a general Bacterial Community Index characterising stream ecosystem health based on the structure of biofilm bacterial communities.

Are there naturalized Escherichia coli in the environment?

Anne-Marie Perchec

School of Biological Sciences, University of Auckland, New Zealand

MSc. Thesis, 2009

This research project investigated whether a naturalized clade of Escherichia coli exists in the aquatic environment. E. coli, a well described bacterial species, is also used as a faecal indicator in water quality monitoring programs worldwide, with the assumption that this bacterium is exclusively a commensal of the vertebrate gut. Recent findings, recording growth and multiplication of E. coli in water and soils, show this is not the case. Although many studies have attempted to define an environmental population, it has yet to be elucidated.
This study seeks to clarify the relationships between environmental and commensal E. coli strains by evaluating fundamental genetic differences using MLST (Multilocus Sequence Typing) and differences in gene expression focused on the rpoS gene. E. coli strains used were isolated from a wetland, freshwater streams, and animal hosts. While little genetic variability was demonstrated for each MLST gene, both environmental and commensal strains showed a high diversity of MLST profiles. Genetic analyses of linkage disequilibrium, index of association and rates of synonymous and non-synonymous substitutions were used to investigate sequence variability and nature of change. Phylogenetic trees based on the concatenated sequences of the seven MLST housekeeping genes displayed distinct clustering of environmental strains. The comparison of the New Zealand sequences with worldwide E. coli strains retrieved from the Shigatox MLST database online did not allow the identification of a clear environmental genotype. However the New Zealand strains showed some close relationships with strains from human and cow origins.
In order to infer the spatial and temporal stability of E. coli in the environment, the 16S rDNA sequences of these studied strains were compared with sequences of E. coli stored in a clonal library obtained from previous studies on biofilms but no patterns were identified. A preliminary study using a quantitative PCR method measuring the expression levels of rpoS, a gene coding for the stress factor σS and potentially involved in adaptation to environmental conditions, was performed on two environmental and one commensal strains. This very limited experiment showed significant differences between the environmental and commensal strains tested. Overall, no identification of an environmental genotype was demonstrated although indication of adaptive naturalization of E. coli occurred.

Are there naturalized Escherichia coli in water?

Anne-Marie Perchec & Gillian Lewis

School of Biological Sciences, University of Auckland.

New Zealand Hydrological and Freshwater Sciences Societies Joint Conference, Whangarei, November 2009

Background/Aim: This research project investigates the existence of a natural environmental clade of Escherichia coli. E. coli is also a common micro-organism used as a faecal indicator in water quality monitoring programs worldwide. Although it is thought to be exclusively commensal of the gut of vertebrates, this assumption is put in question by many observations of growth and multiplication of E. coli in water and soils. While many studies have been made to compare the different populations, there is still no clear picture of the relationships between environmental and host-related E. coli strains.
Methods: This study investigates differences in the core genome of E. coli from environmental and commensal origins, using a MLST (Multi Locus Sequence Type) method. The strains used in this work come from a dataset of isolates collected from a wetland environment, from stream biofilm samples taken in environment and of commensals in diverse hosts.
Results: As expected, MLST results show a high degree of conservation for the sequenced housekeeping genes of the studied strains. The strains studied exhibit a higher percentage of synonymus substitutions compared to the percentage of non-synonymous substitutions as recorded in other studies of environmental E. coli. Although there is an important diversity of genetic profiles, the phylogenetic analysis reveals a cluster of specific strains. This clustering indicates that those strains have replicated at a given time without recent contact with a host. Interestingly, some identical profiles can be retrieved at the same place at different dates, which signifies that some strains are able to persist across time.
Conclusion: Finally this study gives an insight into the evolution and divergences between demonstrated environmental population of E. coli and commensal E. coli characterized by genetic clusters based on housekeeping genes. It is not yet clear whether an environmental clade of E. coli exists or whether the environmental replication of persistent specific strains shown to occur indicates a process of adaption which may be termed naturalization.





Contact us

Stream Biofilm Research Group
School of Biological Sciences
The University of Auckland
3a Symonds Street
Auckland, New Zealand


About us






© Stream Biofilm Project, The University of Auckland 2008-2010