Open Access Repository

Experimental models of bacteria-phytoplankton interactions and bacterial influence on growth of toxic dinoflagellate, Gymnodinium catenatum

Downloads

Downloads per month over past year

Anantha Subramanian, T (2008) Experimental models of bacteria-phytoplankton interactions and bacterial influence on growth of toxic dinoflagellate, Gymnodinium catenatum. PhD thesis, University of Tasmania.

[img] PDF (Whole thesis)
Thesis_Anantha_...pdf | Request a copy
Full text restricted until 13 December 2018.
Available under University of Tasmania Standard License.

Abstract

Interactions between phytoplankton and their associated marine microbial community are believed to have major effects on the growth dynamics of both bacteria and phytoplankton. Phytoplankton are associated with bacterial communities composed of I0's to I00's of bacterial types with the potential for millions of potentially confounding interactions. This complexity currently hampers research to identify key bacteria and the mechanisms of interaction. To overcome these problems, this work investigates the behaviour and dynamics of simplified experimental models of bacteria-phytoplankton to examine the influence of these microbial interactions on the dynamics of phytoplankton growth. The basis of the experimental system was the toxic dinoflagellate Gymnodium catenatum, a well known causative organism of paralytic shellfish poisoning associated with frequent blooms around the globe. Simplified microbial communities in cultures were generated from surface-sterilised resting cysts germinated in the presence of bacteria isolated from non-axenic G. catenatum cultures (Brachybacterium sp., Alcanivorax sp. DG881, Marinobacter sp. DG879 and Roseobacter sp. DG874) in either unibacterial or mixed-bacterial experimental model systems.

Using the simplified experimental model, a range of experimental manipulations of the bacterial community of G. catenatum were undertaken to examine specific hypotheses. The first experiment examined the hypothesis that G. catenatum has an obligate requirement for bacteria, using antibiotic-resistant and antibiotic- sensitive strains of marine bacteria to provide important negative and positive controls. Addition of antibiotics to cultures with sensitive bacteria resulted in a significant decline in dinoflagellate cell concentration, where as control cultures grown with antibiotic resistant bacteria continued to grow throughout the experiment. Importantly, this experiment demonstrated that removal of bacteria, rather than the action of antibiotic, caused the decline and death of the dinoflagellate culture.

The influence of bacterial community composition on G. catenatum batch culture dynamics was examined using uni-bacterial, and mixtures containing the y-proteobacteria Alcanivorax sp. or Marinobacter sp. and/or the a-proteobacterium Roseobacter sp. Exponential growth rate, death rate, maximum cell concentration and batch culture dynamics were all influenced by the bacterial community composition in the experimental cultures, demonstrating that the bacterial community is a significant factor influencing the growth dynamics of G. catenatum. Uni-bacterial G. catenatum models showed that dinoflagellate grew significantly faster in the presence of Alcanivorax sp. or Marinobacter sp. than when grown with the a-proteobacterium Roseobacter sp. Pair-wise mixtures and tri-bacterial treatments displayed batch growth patterns intermediate or combined features of the respective uni-bacterial bacterial patterns, suggesting that bacterial community effects on dinoflagellate growth are additive. Epifluorescence microscopy and DAPI staining of cultures showed that uni-bacterial model cultures were not attached or associated with the dinoflagellate cell during logarithmic or stationary phase dinoflagellate growth but an increase in the proportion of bacteria associated with the cells wall was noted in late-stationary to death phase. During death phase, cultures grown with Roseobacter sp. showed a significantly higher proportion (18%) of
bacteria associated with the dinoflagellate cell, than Alcanivorax sp. (7.9%) or Marinobacter sp. (11.8 %).

Uni-bacterial G. catenatum cultures maintained for more than 6 months exhibited reduced exponential growth rates and low maximum dinoflagellate cell concentrations, and early onset of death phase. Uni-bacterial model cultures using antibiotic-resistant and sensitive-bacteria were used to investigate whether replacement or addition of "new" bacteria (Brachybacterium sp. or Marinobacter sp.) to cultures could restore or improve dinoflagellate growth. Cultures where the bacterial community was replaced showed improved dinoflagellate growth but long term survival and culture maintenance was not possible, indicating that the dinoflagellate-bacteria relationship in uni-bacterial model cultures may be unstable over long periods.

The G. catenatum model system was also used to examine whether dinoflagellate genotype or bacterial community composition was the dominant factor influencing G. catenatum growth dynamics. Two clonal parent cultures (GCDE08 and GCHUll) with markedly different batch growth dynamics exponential growth rates and microbial communities were compared with equivalent mixtures of non-clonal progeny established in the presence of microbial communities (8 11m filtrates) from each of the parent cultures. All non-clonal progeny treatments showed similar growth patterns that were different to either parent culture suggesting that genotype is the dominant influence on growth. However, tRFLP analysis showed that, regardless of the bacterial community added at germination, a consistent but different bacterial community was established in all nonclonal progeny cultures. This indicates that the growth dynamics are influenced primarily by the bacterial community composition rather than dinoflagellate genotype, and that the bacterial community may be selected or modified by factors associated with the dinoflagellate genotype.

Item Type: Thesis (PhD)
Copyright Information:

Copyright 2008 the author - The University is continuing to endeavour to trace the copyright
owner(s) and in the meantime this item has been reproduced here in good faith. We
would be pleased to hear from the copyright owner(s).

Date Deposited: 24 Mar 2014 01:56
Last Modified: 15 Sep 2017 01:06
Item Statistics: View statistics for this item

Actions (login required)

Item Control Page Item Control Page
TOP