Spore-forming bacteria as a proxy for the reconstruction of past environment and possible use for the detection of antibiotic resistance genes in the environment
Abstract Paleoecology is the study of past environments based on the analysis of sedimentary records and their chemical, isotopic and biological composition. Paleoecology aims to establish the relationship between organisms and their physical and chemical environments, in order to assess the causes and rate of ecological change and to reconstruct ecosystem history. Such studies allow for a better understanding of climate variability, range of natural fluctuations, extreme events frequency and/or anthropogenic impact over time, at various time scales. Paleoecological indicators include a wide variety of compounds and structures including mineralogical, chemical and isotopic composition of sediment particles, as well as biological structures such as pollen grain or microfossils (diatoms, ostracods). Environmental DNA has also been proposed as a biological proxy, but DNA-based methods are dependent on DNA conservation. To be applicable, such proxies must be preserved in sediments for long-time periods. That must be the case for bacterial endospores or other spore-like structures. These highly specialized cellular forms allow the organisms able to produce them to resist and survive harsh environmental conditions by entering a dormant state. Such structures have long been proposed as paleoecological proxies, but only the recent development of molecular methods allowing their study results in their potential widespread application for paleoecology. The new-developed methods include notably a DNA extraction protocol adapted to highly resistant structures, and a treatment for the enrichment of spores (spore-separation method). The nature of the spore-separation treatment, which consists in the enrichment of cells able to withstand a harsh lysis method, does not provide direct evidence for the formation of specialized cells, we will use the term “lysis-resistant” instead of spores when appropriate.

The first part of this thesis aimed to evaluate the potential of using bacterial DNA, from both the total and the lysis-resistant community, as a proxy for the reconstruction of past environments. First, the efficiency of the newly developed method for the isolation of spores was assessed by comparing the total and the lysis-resistant community. The community composition was compared in samples that have or not been submitted to the spore-separation treatment. The results showed that the lysisresistant community possess a unique signature compared to the total community. Interestingly, various genera hitherto considered as non-sporulating were found in high abundance in the lysisresistant community but were not among the most abundant genera in the total community. This demonstrates a capacity to resist the spore-separation treatment and strongly suggests the ability in these groups of the production of a lysis-resistant structure. Second, the separation method was used to investigate the diversity, distribution, and community structure of potential spore-forming organisms in the environment. To that purpose, data was obtained from four contrasting study sites. The results of our analysis revealed an unsuspected diversity of organisms in the lysis-resistant community. The lysis-resistant community also showed a geographic distribution pattern, challenging a hypothetical cosmopolitan distribution of spore-formers. Results of this study suggest that the ability to form spores or similar resting and durable cell structures is more widespread than previously suspected.

In a second study, the application of these novel methods was tested for the reconstruction of the history of the ephemeral Lake Liambezi (Namibia). Using a multidisciplinary approach including the use of bacterial DNA in complement to geochemical and sedimentological analyses, the climate evolution over the past 5500 years was reconstructed. This highlighted an alternation of dry and wet periods, and changes in the hydrological lake regime (from fen to lake). DNA was isolated from both the total and the lysis-resistant fraction of the community, both of which reflected changes in the environmental conditions, demonstrating their relevancy for paleoecological studies. Interestingly, in some cases, putative environmental conditions, biological processes, or extreme events were only seen in one fraction of the community, highlighting the complementarity of investigating both fractions of the community. In addition, the analysis of the bacterial community and specific populations helped to elaborate a coherent age model relating the three sediment cores studied. This analysis suggested hydrothermal activity and sulfur cycling within the lake. Results of this study demonstrated the potential of using bacterial DNA (total and/or spores) to identify changes and variability in the environmental conditions, and the strengths of a multidisciplinary approach to reconstruct ecosystem history. The second part of this thesis aimed to evaluate the possible use of these new-developed methods for studying the prevalence of antibiotic resistance genes (ARG) in the environment. Since the discovery of penicillin by Fleming in 1928 and the development of antibiotics for medical purpose from the 1940s’, the extensive use of antibiotics has created a selection pressure that has contributed to the emergence of antibiotic resistant bacteria (ARB) and multi-resistant bacteria (MRB). The multiplication of such organisms represents a real threat for human health in the future. Long ignored, this problem is now considered as urgent. Although the distribution and frequency of ARG in the environment is widely unknown, it is clear that human activities have an impact in prevalence. In this part of the thesis, the use of DNA extracted from lysis-resistant bacterial cells for tracking ARG in the environment was

In a first study, the accumulation of ARG over time was investigated, in both the total and the lysisresistant community, in sediments from Lake Geneva (CH). The results showed that two selected ARG (tet(W) and sul1) could be detected in both the lysis-resistant and the total fraction of the community. Both genes were found in higher frequency (copies/ng DNA) in the lysis-resistant community of the community compared to the total community, suggesting the lysis-resistant fraction was enriched in ARG. Accumulation patterns of these two ARG showed to be correlated to the historical use of their related antibiotics. When investigating the relationship between the ARG accumulation and the bacterial community composition, each gene appeared to be correlated to different taxonomic group. While tet(W) was mainly correlated to a change in the relative abundance in the Firmicutes, sul1 was correlated to a more diverse group of organisms, suggesting both ARG are differently distributed across bacterial taxa. These results show that spore-like structures can be used to trace back the effect of historical usage of antibiotics on resistance prevalence. In a second study, the impact of wastewater release on the environmental spread of ARG associated to the lysis-resistant community was investigated in wastewater-impacted sediments from the Vidy Bay (CH). The wastewater treatment plant (WWTP) was identified as a source of ARG. The two studied ARG (tet(W) and sul1) were detected in all samples, and their abundance/frequency decreased with increasing distance to the WWTP outlet. ARG levels were correlated to other indicators of wastewater discharge, such as Corg, Ntot and DNA. Both ARG showed to be differently enriched in the lysis-resistant community. Similarly to what had been observed in the first study, tet(W) frequency mainly correlated with the relative abundance of genera belonging to the Firmicutes (Clostridium and Ruminococcus), and sul1 frequency correlated with a large taxonomic spectrum of organisms. The high relative abundance of Clostridium spp., coupled to its correlation with tet(W) frequency, suggested that members of this genus might be a potential vector for tet(W) dissemination.

These two studies constituted the first evidence of the possible detection of ARG in spores or lysisresistant structures. DNA extracted from these resilient structures appeared to be a good proxy for assessing the dispersal and the accumulation of ARG over time in environmental samples. Given their high survival ability and propensity for dispersion, this lysis-resistant fraction of the community might receive more attention in the future, for a better understanding of its role in the fate of ARG, and to help implementing appropriate usage management strategies to halt antibiotic resistance and improve their removal during waste treatment.
Keywords spores, lysis-resistant, paleoecology, antibiotic resistance, bacterial community, sediment
Citation Paul, C. (2019). Spore-forming bacteria as a proxy for the reconstruction of past environment and possible use for the detection of antibiotic resistance genes in the environment. Doctorat, Neuchâtel, Neuchâtel.
Type Thesis (English)
Year 2019
Academic department Faculté des sciences, Institut de biologie
University Neuchâtel (Neuchâtel)
Degree Doctorat