Understanding the pathways of contaminant transfer within trophic networks is of paramount importance to weigh up forces driving ecological changes and to plan focused intervention strategies targeted to environmental conservation. The subject developed in this PhD dissertation aligns with this goal and encompasses the combination of different modelling approaches with experimental data to interpret the effects of bioaccumulation and bioremediation phenomena on species. Toxic agents adversely influence substance and energy fluxes at the ecosystem level affecting in turn the number of inter- and intra-specific interactions both at the community and population levels. Furthermore, the environmental stochasticity and the diversity of food webs along with the specificity of action of different bioaccumulative compounds greatly increase the complexity of this field of research. Current challenges in ecotoxicology focus on the need to find reliable predictive tools able to turn toxicity data of biota into powerful estimation methods and to asses the long-term effects of chemical exposure on species. In this context, predator-prey relationships are crucial to characterize the contamination patterns and to predict how chemicals transfer and accumulate within food webs. Food web members exhibit different levels of bioaccumulation in function of their trophic role, in a way that trophic links cannot be considered equal for all species in bioaccumulation phenomena. However, marine ecosystems are not just ensembles of macro-species, but complex multiscale networks. Microbial marine communities are metabolically involved in bioremediation processes and also represent an active compartment in the lower trophic levels of food webs. Thus, microbial degradation of persistent organic chemicals may play a key factor in changing the fate of these compounds within ecosystems and in reducing the contaminant uptake that leads to bioaccumulation in marine species. The work of research carried out during my PhD studies has been centred on the bioaccumulation and bioremediation modelling problems with specific interest on polychlorinated biphenyls (PCBs) contamination of the Adriatic food web. Notwithstanding a number of specific experimental studies on PCBs concentrations in different Adriatic species have been carried out over the last decades, to the best of our knowledge, a comprehensive PCBs bioaccumulation model for the Adriatic food web was still missing prior to the work presented in this thesis. The contributions of this PhD dissertation are structured by publication. Chapter 1 introduces the main content and includes an unpublished review of experimental PCBs concentration data in Adriatic species over the last two decades, along with an overview of the most important modelling approaches for bioaccumulation and bioremediation. In Chapter 2, we present a computational framework to model the bioaccumulation of organic chemicals in aquatic food webs, and to discover the toxic keystones, i.e. the species with a key role in the trophic transfer of contaminants. The approach is applied to reconstruct the first PCBs bioaccumulation model of the Adriatic food web, parametrized with a subset of the concentration data reviewed in Chapter 1. The framework integrates different modelling and analysis techniques, the first being the reconstruction of a trophic network from biomass data. Then, we use the estimated biomass flows and concentration data to derive the PCBs bioaccumulation network. Network reconstruction is performed using linear inverse modelling (LIM), an efficient technique for estimating food webs from empirical data. This step allows us to infer concentration values and contaminant flows for all species and remarkably, also for species with no input data associated. The estimated concentrations highlight the occurrence of PCBs biomagnification, which we show depending mainly on the trophic structure. The second main part of the framework is dedicated to the problem of identifying the toxic keystones, for which we propose the application of network analysis tools, typically employed in the trophic context. To this aim, we define a new network index, sensitivity centrality, able to capture not just direct and indirect effects in the PCBs network, but also the dynamics of bioaccumulation. Indeed, the index is based on the sensitivity analysis of a differential equation model derived from the bioaccumulation network. We compare sensitivity centrality with established network centrality indices, by evaluating the impact of successive species extinctions on global network properties, where such extinctions are performed following the importance ordering of the different indices. This analysis demonstrates that the introduced index can better identify the species with the highest impact on the total contaminant flows and on the efficiency of contaminant transport within the food web. In Chapter 3, we propose a novel computational framework of analysis to investigate multiscale effects of bioremediation processes at the ecosystem level. We integrate the bioaccumulation model presented in Chapter 2 with the genome-scale metabolic network of Pseudomonas putida KT2440 (iJN746), which we extend to simulate the aerobic PCBs degradation under arbitrary scenarios of contaminant removal. We use a reaction-based ecological/microbial network representation by combining ecological and metabolic modelling techniques, namely LIM and flux balance analysis. In this way, we describe in a unique framework PCBs flows among species, metabolites concentrations and reactions fluxes in the microbial metabolism. We investigate the tradeoff between PCBs uptake and growth of P. putida at different oxygen levels, by using a bi-level flux balance analysis approach. We study the interdependence between PCBs and toluene uptake, which is a natural degradation pathways in P. putida, by performing a phenotypic phase plane analysis. We apply this framework to study how different bioremediation strategies can impact PCBs concentration in species, thus enabling an ecosystem level analysis. Finally, we evaluate also the effect of bioremediation on indices of species centrality in the PCBs bioaccumulation network. To sum up, the aims and contributions of this PhD thesis are: Provide a review of PCBs concentration data in the Adriatic ecosystem and of modelling methods for bioaccumulation and bioremediation. Reconstruct the first PCBs bioaccumulation model of the Adriatic food web and investigate species having a central role in the trophic transfer of contaminants. Integrate the Adriatic bioaccumulation model with a genome-scale metabolic model for PCBs biodegradation in order to enable multiscale ecosystem analyses. References chapter 2: M. Taffi, N. Paoletti, P. Lia'², S. Pucciarelli, and M. Marini. Bioaccumulation modelling and sensitivity analysis for discovering key players in contaminated food webs: The case study of PCBs in the Adriatic Sea. Ecological Modelling, 2014. chapter 3: M. Taffi, N. Paoletti, C. Angione, S. Pucciarelli, M. Marini, and P. Lia'². Bioremediation in marine ecosystems: a computational study combining ecological modelling and flux balance analysis. Frontiers in Genetics, 5(319), 2014. v
Bioaccumulation and bioremediation modelling in Aquatic ecosystems: the Adriatic Sea case study
TAFFI, MARIANNA
2015-03-17
Abstract
Understanding the pathways of contaminant transfer within trophic networks is of paramount importance to weigh up forces driving ecological changes and to plan focused intervention strategies targeted to environmental conservation. The subject developed in this PhD dissertation aligns with this goal and encompasses the combination of different modelling approaches with experimental data to interpret the effects of bioaccumulation and bioremediation phenomena on species. Toxic agents adversely influence substance and energy fluxes at the ecosystem level affecting in turn the number of inter- and intra-specific interactions both at the community and population levels. Furthermore, the environmental stochasticity and the diversity of food webs along with the specificity of action of different bioaccumulative compounds greatly increase the complexity of this field of research. Current challenges in ecotoxicology focus on the need to find reliable predictive tools able to turn toxicity data of biota into powerful estimation methods and to asses the long-term effects of chemical exposure on species. In this context, predator-prey relationships are crucial to characterize the contamination patterns and to predict how chemicals transfer and accumulate within food webs. Food web members exhibit different levels of bioaccumulation in function of their trophic role, in a way that trophic links cannot be considered equal for all species in bioaccumulation phenomena. However, marine ecosystems are not just ensembles of macro-species, but complex multiscale networks. Microbial marine communities are metabolically involved in bioremediation processes and also represent an active compartment in the lower trophic levels of food webs. Thus, microbial degradation of persistent organic chemicals may play a key factor in changing the fate of these compounds within ecosystems and in reducing the contaminant uptake that leads to bioaccumulation in marine species. The work of research carried out during my PhD studies has been centred on the bioaccumulation and bioremediation modelling problems with specific interest on polychlorinated biphenyls (PCBs) contamination of the Adriatic food web. Notwithstanding a number of specific experimental studies on PCBs concentrations in different Adriatic species have been carried out over the last decades, to the best of our knowledge, a comprehensive PCBs bioaccumulation model for the Adriatic food web was still missing prior to the work presented in this thesis. The contributions of this PhD dissertation are structured by publication. Chapter 1 introduces the main content and includes an unpublished review of experimental PCBs concentration data in Adriatic species over the last two decades, along with an overview of the most important modelling approaches for bioaccumulation and bioremediation. In Chapter 2, we present a computational framework to model the bioaccumulation of organic chemicals in aquatic food webs, and to discover the toxic keystones, i.e. the species with a key role in the trophic transfer of contaminants. The approach is applied to reconstruct the first PCBs bioaccumulation model of the Adriatic food web, parametrized with a subset of the concentration data reviewed in Chapter 1. The framework integrates different modelling and analysis techniques, the first being the reconstruction of a trophic network from biomass data. Then, we use the estimated biomass flows and concentration data to derive the PCBs bioaccumulation network. Network reconstruction is performed using linear inverse modelling (LIM), an efficient technique for estimating food webs from empirical data. This step allows us to infer concentration values and contaminant flows for all species and remarkably, also for species with no input data associated. The estimated concentrations highlight the occurrence of PCBs biomagnification, which we show depending mainly on the trophic structure. The second main part of the framework is dedicated to the problem of identifying the toxic keystones, for which we propose the application of network analysis tools, typically employed in the trophic context. To this aim, we define a new network index, sensitivity centrality, able to capture not just direct and indirect effects in the PCBs network, but also the dynamics of bioaccumulation. Indeed, the index is based on the sensitivity analysis of a differential equation model derived from the bioaccumulation network. We compare sensitivity centrality with established network centrality indices, by evaluating the impact of successive species extinctions on global network properties, where such extinctions are performed following the importance ordering of the different indices. This analysis demonstrates that the introduced index can better identify the species with the highest impact on the total contaminant flows and on the efficiency of contaminant transport within the food web. In Chapter 3, we propose a novel computational framework of analysis to investigate multiscale effects of bioremediation processes at the ecosystem level. We integrate the bioaccumulation model presented in Chapter 2 with the genome-scale metabolic network of Pseudomonas putida KT2440 (iJN746), which we extend to simulate the aerobic PCBs degradation under arbitrary scenarios of contaminant removal. We use a reaction-based ecological/microbial network representation by combining ecological and metabolic modelling techniques, namely LIM and flux balance analysis. In this way, we describe in a unique framework PCBs flows among species, metabolites concentrations and reactions fluxes in the microbial metabolism. We investigate the tradeoff between PCBs uptake and growth of P. putida at different oxygen levels, by using a bi-level flux balance analysis approach. We study the interdependence between PCBs and toluene uptake, which is a natural degradation pathways in P. putida, by performing a phenotypic phase plane analysis. We apply this framework to study how different bioremediation strategies can impact PCBs concentration in species, thus enabling an ecosystem level analysis. Finally, we evaluate also the effect of bioremediation on indices of species centrality in the PCBs bioaccumulation network. To sum up, the aims and contributions of this PhD thesis are: Provide a review of PCBs concentration data in the Adriatic ecosystem and of modelling methods for bioaccumulation and bioremediation. Reconstruct the first PCBs bioaccumulation model of the Adriatic food web and investigate species having a central role in the trophic transfer of contaminants. Integrate the Adriatic bioaccumulation model with a genome-scale metabolic model for PCBs biodegradation in order to enable multiscale ecosystem analyses. References chapter 2: M. Taffi, N. Paoletti, P. Lia'², S. Pucciarelli, and M. Marini. Bioaccumulation modelling and sensitivity analysis for discovering key players in contaminated food webs: The case study of PCBs in the Adriatic Sea. Ecological Modelling, 2014. chapter 3: M. Taffi, N. Paoletti, C. Angione, S. Pucciarelli, M. Marini, and P. Lia'². Bioremediation in marine ecosystems: a computational study combining ecological modelling and flux balance analysis. Frontiers in Genetics, 5(319), 2014. vI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
