Search

Additional file 1: Supplementary Table 1. Number of OTUs significantly differentially abundant among all treatments as estimated using a generalized linear mixed model, among the 515 most abundant 16S rRNA OTUs and the 439 most abundant 18S rRNA OTUs. Supplementary Figure 1. Abundances of total bacteria and total fungi. Quantification of 16S rRNA (a and c) and ITS (b and d) gene copy numbers in the original soil, the removal treatments and the control (Step 1; a and b) and in the coalescence treatment, the self-mixed removal treatment and the control samples (Step 2; c and d) (mean ± s.e. of log10-transformed data expressed as gene copy g-1 dry soil). Letters indicate significantly different statistical groups (Tukey’s test, p-value ≤ 0.05). Supplementary Figure 2. Diversity of prokaryotic and eukaryotic communities after Step 1. The Faith’s phylogenetic diversity (a and c) and Simpson’s reciprocal (b and d) indices are shown (mean ± s.e.) in the original soil, the removal treatments and the control (Step 1) for the 16S rRNA OTUs (a and b) and the 18S rRNA OTUs (c and d). Letters indicate significantly different statistical groups (Tukey’s test, p-value ≤ 0.05). Supplementary Figure 3. Structure and composition of the eukaryotic communities in the original soil and after Step 1. (a) Number of observed species (mean ± s.e.). Letters indicate significantly different statistical groups (Tukey’s test, p-value ≤ 0.05). (b) Relative abundances of the fourteen most abundant classes of eukaryotic community. (c) Similarity between the control samples and between the control and either the original soil or the removal treatments, based on the Weighted UniFrac distances (mean ± s.e.). Letters indicate significantly different statistical groups (Adonis pairwise comparison, Benjamini-Hochberg corrected p-value ≤ 0.05). Supplementary Figure 4. Principal coordinate analysis (PCoA) of the prokaryotic (a) and eukaryotic (b) communities, based on the weighted UniFrac distance matrix showing the original soil, the removal treatment and the control samples and the 95% joint confidence ellipse for the control samples. Supplementary Figure 5. Taxonomic relationships and distribution of OTUs significantly affected by the Step 1 removal treatments compared to the Step 1 control. The outer rings show the effect of each removal treatment on the relative abundances of the 515 most abundant 16S rRNA OTUs (a) and the 439 most abundant 18S rRNA OTUs (b) compared to the control (R vs C), as estimated using a generalized linear mixed model. The blue and red boxes in the outer rings indicate OTUs with increasing and decreasing fitness respectively, while white boxes indicate OTUs that are not affected by the treatment. The OTU class level is indicated by different colors on the innermost ring. Supplementary Figure 6. Structure and composition of the eukaryotic communities after Step 2. (a) Number of observed species (mean ± s.e.). Letters indicate significantly different statistical groups (Tukey’s test, p-value ≤ 0.05). (b) Relative abundances of the fourteen most abundant class of eukaryotic community. (c) Similarity between the control samples and between the control and either the self-mixed removal treatment or the coalescence treatments based on the Weighted UniFrac distances (mean ± s.e.). Letters indicate significantly different statistical groups (Adonis pairwise comparison, Benjamini-Hochberg corrected p-value ≤ 0.05). Supplementary Figure 7. Diversity levels of prokaryotic and eukaryotic communities after Step 2. The Faith’s phylogenetic diversity (a and b) and Simpson’s reciprocal (c and d) indices are shown (mean ± s.e.) in the coalescence treatment, the self-mixed removal treatments and the control samples (Step 2) for the 16S rRNA OTUs (a and c) and the 18S rRNA OTUs (b and d). Letters indicate significantly different statistical groups (Tukey’s test, p-value ≤ 0.05). Supplementary Figure 8. Principal coordinate analysis (PCoA) of the prokaryotic (a) and eukaryotic (b) communities, based on the weighted UniFrac distance matrix showing the self-mixed removal treatment, the coalescence treatment and the control samples and the 95% joint confidence ellipse for the control samples. Supplementary Figure 9. Taxonomic relationships and distribution of OTUs significantly affected by the Step 2 removal treatments compared to the Step 2 control. The outer rings show the effect of each self-mixed removal treatment on the relative abundance of the 515 most abundant 16S rRNA OTUs (a) and the 439 most abundant 18S rRNA OTUs (b) compared to the self-mixed control (R+R vs C+C), as estimated using a generalized linear mixed model. The blue and red boxes in the outer rings indicate OTUs with increasing and decreasing fitness respectively, while white boxes indicate OTUs that are not affected by the treatment. The OTU class level is indicated by different colors on the innermost ring. Supplementary Figure 10. Taxonomic relationships and distribution of OTUs significantly affected by the Step 2 coalescence treatments compared to the Step 2 control. The outer rings show the effect of each coalescence treatment on the relative abundance of the 515 most abundant 16S rRNA OTUs (a) and the 439 most abundant 18S rRNA OTUs (b) compared to the self-mixed control (R+C vs C+C), as estimated using a generalized linear mixed model. The blue and red boxes in the outer rings indicate OTUs with increasing and decreasing fitness respectively, while white boxes indicate OTUs that are not affected by the treatment. The OTU class level is indicated by different colors on the innermost ring. Supplementary Figure 11. Taxonomic relationships and distribution of OTUs significantly affected by the Step 2 coalescence treatments compared to the Step 2 removal treatments. The outer rings show the effect of each coalescence treatment on the relative abundance of the 515 most abundant 16S rRNA OTUs (a) and the 439 most abundant 18S rRNA OTUs (b) compared to its corresponding self-mixed removal treatment (R+C vs R+R), as estimated using a generalized linear mixed model. The blue and red boxes in the outer rings indicate OTUs with increasing and decreasing fitness respectively, while white boxes indicate OTUs that are not affected by the treatment. The OTU class level is indicated by different colors on the innermost ring. Supplementary Figure 12. Taxonomic relationships and distribution of significantly differentially abundant OTUs across treatments. Outer rings show prokaryotic (a) and eukaryotic (b) OTUs exhibiting relative abundances significantly higher or lower in the coalescence treatments than in the self-mixed source community separately (R+C>C+C ∩ R+C>R+R and R+C 0 and red if ρ < 0. Edge width is proportional to |ρ|. Supplementary Figure 15. Global inter-domain network inferred from all samples across both experimental steps. Nodes represent OTUs and they are colored according to the OTU taxonomic class. Edges represent partial correlations ρ and they are colored blue if ρ > 0 and red if ρ < 0. Edge width is proportional to |ρ|. Supplementary Figure 16. Treatment-induced changes in soil respiration rate as proxies of C-cycling. Substrate-induced respiration was measured by the MicroResp™ method using the substrates alanine (a and d), fructose (b and e) and gallic acid (c and f) in the original soil, the removal treatments and the control (Step 1; a, b and c) or in the coalescence treatments and the self-mixed treatments (Step 2; d, e and f) (mean ± s.e.). Letters indicate significantly different statistical groups (Tukey’s test, p-value ≤ 0.05). Supplementary Figure 17. Treatment-induced changes in abundance of N-cycle microbial guilds as proxies of N-cycling. Abundances of ammonia-oxidizing bacteria (AOB in a and e), bacterial denitrifiers (nirK in b and f; nirS in c and g) and nitrogen-fixing bacteria (nifH in d and h) in the original soil, the removal treatments and the control (Step 1; a, b, c and d) or in the coalescence treatments and the self-mixed treatments (Step 2; e, f, g and h) (mean ± s.e. of log10-transformed data expressed as gene copy g-1 dry soil). Letters indicate significantly different statistical groups (Tukey’s test, p-value ≤ 0.05).

Microbial communities play a key role in ecosystem functioning as well as in plant, animal, and human welfare. Their assembly relies on different processes commonly referred to as abiotic and biotic filters. Despite the widespread emphasis on abiotic filters in terrestrial ecosystems, a growing body of evidence suggests that interactions between microorganisms play a critical role in community assembly. However, the contribution of these interactions to microbial community assembly and the factors influencing them have not been clearly established. In most cases, biotic interactions between microorganisms have been investigated based on cocultures of a handful of strains, completely overlooking the high complexity of microbial communities in nature.This thesis objectives were to decipher biotic interactions occurring between microbial community members and to assess the role of these interactions in the assembly of complex communities. Thereby, we used three top-down experiments based on the manipulation of native soil microbial communities and their reinoculation in sterile soil microcosms allowing them to reassemble during soil recolonisation.Our results suggest that between 15% and 28% of the bacterial community could be subjected to negative interactions during soil recolonization. Proteobacteria and Firmicutes exhibited a recurrent negative correlation, indicating that competition between them might be a rule governing microbial community assembly. Our approach also uncovered the importance of density-dependent interactions in microbial community assembly, the competitiveness of individuals strongly depending on the density and identity of surrounding neighbours. Furthermore, we showed that coalescence (mixing of manipulated and control communities) could also restore -at least partly- community diversity and soil functions. While the diversity of the coalesced communities was related to that of the source communities, their structure was mostly influenced by the density and the composition of the source communities.Altogether, we bring here additional evidence that community manipulation could become a standard tool to decipher microbial interactions in complex communities, their contribution to community assembly, and the links between community structure and function. Furthermore, our work highlights the potential of using community manipulation to design new microbial engineering strategies allowing to steer microbial communities in situ, without adding non-endemic species, by using the interactions occurring between the native community members., Les communautés microbiennes sont primordiales dans le fonctionnement des écosystèmes ainsi que pour la santé des plantes, des animaux et des humains. Leur assemblage repose sur différents processus communément appelés filtres abiotiques et biotiques. Malgré l'accent mis sur les filtres abiotiques dans les écosystèmes terrestres, de nouvelles preuves suggèrent que les interactions entre les microorganismes jouent un rôle clé dans l'assemblage des communautés. Cependant, la contribution de ces interactions à l'assemblage des communautés microbiennes et des facteurs qui les influencent demeure floues. Traditionnellement, les interactions biotiques entre microorganismes ont été étudiées par coculture d'une poignée de souches, négligeant complètement la complexité des communautés microbiennes naturelles.Les objectifs de cette thèse étaient d’identifier les interactions biotiques se produisant entre les membres de communautés microbiennes et d'évaluer le rôle de ces interactions dans l'assemblage de communautés complexes. Pour cela, nous avons utilisé trois expériences basées sur la manipulation de communautés microbiennes natives de sol et sur leur ré-inoculation dans des microcosmes de sol stérile leur permettant de se réassembler lors de la recolonisation du sol.Nos résultats suggèrent qu'entre 15% et 28% des communautés bactériennes seraient soumis à des interactions négatives pendant la recolonisation du sol. La corrélation négative récurrente entre Proteobacteria et Firmicutes indique que la compétition entre eux pourrait être une règle régissant l'assemblage des communautés microbiennes. Notre approche a également mis en évidence l'importance de la densité dans l'assemblage des communautés microbiennes, la compétitivité des individus dépendant fortement de la densité et de l'identité de leurs voisins. En outre, nous avons montré que la coalescence (c’est-à-dire le mélange de communautés manipulées et contrôle) pourrait restaurer - au moins partiellement - la diversité et les fonctions des communautés. Alors que la diversité des communautés coalescées dépend de celle des communautés sources, leur structure était principalement influencée par la densité et la composition des communautés sources.Nos travaux apportent des preuves supplémentaires que la manipulation des communautés peut devenir un outil de référence pour identifier les interactions microbiennes dans des communautés complexes, leur contribution à l'assemblage de ces communautés, et les liens entre structure et fonction des communautés. De plus, notre travail met en évidence le potentiel de la manipulation de communautés pour concevoir de nouvelles stratégies d'ingénierie écologique permettant de remanier les communautés microbiennes in situ, sans ajouter d'espèces non endémiques, en utilisant les interactions qui se produisent entre les membres de la communauté native.

Microbial communities play a key role in ecosystem functioning as well as in plant, animal, and human welfare. Their assembly relies on different processes commonly referred to as abiotic and biotic filters. Despite the widespread emphasis on abiotic filters in terrestrial ecosystems, a growing body of evidence suggests that interactions between microorganisms play a critical role in community assembly. However, the contribution of these interactions to microbial community assembly and the factors influencing them have not been clearly established. In most cases, biotic interactions between microorganisms have been investigated based on cocultures of a handful of strains, completely overlooking the high complexity of microbial communities in nature. This thesis objectives were to decipher biotic interactions occurring between microbial community members and to assess the role of these interactions in the assembly of complex communities. Thereby, we used three top-down experiments based on the manipulation of native soil microbial communities and their reinoculation in sterile soil microcosms allowing them to reassemble during soil recolonisation. Our results suggest that between 15% and 28% of the bacterial community could be subjected to negative interactions during soil recolonization. Proteobacteria and Firmicutes exhibited a recurrent negative correlation, indicating that competition between them might be a rule governing microbial community assembly. Our ap- proach also uncovered the importance of density-dependent interactions in microbial community assembly, the competitiveness of individuals strongly depending on the density and identity of surrounding neighbours. Fur- thermore, we showed that coalescence (mixing of manipulated and control communities) could also restore -at least partly- community diversity and soil functions. While the diversity of the coalesced communities was related to that of the source communities, their structure was mostly influenced by the density and the composition of the source communities. Altogether, we bring here additional evidence that community manipulation could become a standard tool to decipher microbial interactions in complex communities, their contribution to community assembly, and the links between community structure and function. Furthermore, our work highlights the potential of using commu- nity manipulation to design new microbial engineering strategies allowing to steer microbial communities in situ, without adding non-endemic species, by using the interactions occurring between the native community members., Les communautés microbiennes sont primordiales dans le fonctionnement des écosystèmes ainsi que pour la santé des plantes, des animaux et des humains. Leur assemblage repose sur différents processus communé- ment appelés filtres abiotiques et biotiques. Malgré l’accent mis sur les filtres abiotiques dans les écosystèmes terrestres, de nouvelles preuves suggèrent que les interactions entre les microorganismes jouent un rôle clé dans l’assemblage des communautés. Cependant, la contribution de ces interactions à l’assemblage des communautés microbiennes et des facteurs qui les influencent demeure floues. Traditionnellement, les interactions biotiques entre microorganismes ont été étudiées par coculture d’une poignée de souches, négligeant complètement la com- plexité des communautés microbiennes naturelles. Les objectifs de cette thèse étaient d’identifier les interactions biotiques se produisant entre les membres de communautés microbiennes et d’évaluer le rôle de ces interactions dans l’assemblage de communautés complexes. Pour cela, nous avons utilisé trois expériences basées sur la manipulation de communautés microbiennes natives de sol et sur leur ré-inoculation dans des microcosmes de sol stérile leur permettant de se réassembler lors de la recolonisation du sol. Nos résultats suggèrent qu’entre 15% et 28% des communautés bactériennes seraient soumis à des interactions négatives pendant la recolonisation du sol. La corrélation négative récurrente entre Proteobacteria et Firmicutes in- dique que la compétition entre eux pourrait être une règle régissant l’assemblage des communautés microbiennes. Notre approche a également mis en évidence l’importance de la densité dans l’assemblage des communautés mi- crobiennes, la compétitivité des individus dépendant fortement de la densité et de l’identité de leurs voisins. En outre, nous avons montré que la coalescence (c’est-à-dire le mélange de communautés manipulées et contrôle) pourrait restaurer - au moins partiellement - la diversité et les fonctions des communautés. Alors que la diver- sité des communautés coalescées dépend de celle des communautés sources, leur structure était principalement influencée par la densité et la composition des communautés sources. Nos travaux apportent des preuves supplémentaires que la manipulation des communautés peut devenir un outil de référence pour identifier les interactions microbiennes dans des communautés complexes, leur contribu- tion à l’assemblage de ces communautés, et les liens entre structure et fonction des communautés. De plus, notre travail met en évidence le potentiel de la manipulation de communautés pour concevoir de nouvelles stratégies d’ingénierie écologique permettant de remanier les communautés microbiennes in situ, sans ajouter d’espèces non endémiques, en utilisant les interactions qui se produisent entre les membres de la communauté native.

National audience; Soil microbial communities play key roles in ecosystem functioning. Yet, little is known aboutthe importance of microbe-microbe interactions in soil microbial community assembly andfunctions. To address this knowledge gap, the objective of my thesis is to assess the role ofinteractions between micro-organisms in soil microbial communities. For this purpose, we willmanipulate the interactions between microorganisms within complex soil microbialcommunities by (i) removing different members of the soil community, (ii) adding microbialtaxa to the soil community and (iii) increasing the physical distance between members of thesoil community. How shifts in microbial interactions affect community assembly will bedetermined after recolonization of sterile soil microcosms by the inoculated communitiesduring several weeks. The resulting communities will be analyzed by 16S rRNA and 18S rRNAmetabarcoding and bioinformatics tools such as network inference to identify the interactingcommunity members. The consequences on community functions will be assessed for C and Ncycles. This thesis project should allow unravelling soil microbial community assembly rulesand the assessment of community manipulation impact on soil functions, therefore paving theway to steer soil microbial communities in agroecosystems.

Splenic diffuse red pulp lymphoma is an indolent small B-cell lymphoma recognized as a provisional entity in the World Health Organization 2008 classification. Its precise relationship to other related splenic B-cell lymphomas with frequent leukemic involvement or other lymphoproliferative disorders remains undetermined. We performed whole-exome sequencing to explore the genetic landscape of ten cases of splenic diffuse red pulp lymphoma using paired tumor and normal samples. A selection of 109 somatic mutations was then evaluated in a cohort including 42 samples of splenic diffuse red pulp lymphoma and compared to those identified in 46 samples of splenic marginal zone lymphoma and eight samples of hairy-cell leukemia. Recurrent mutations or losses in BCOR (the gene encoding the BCL6 corepressor) – frameshift (n=3), nonsense (n=2), splicing site (n=1), and copy number loss (n=4) – were identified in 10/42 samples of splenic diffuse red pulp lymphoma (24%), whereas only one frameshift mutation was identified in 46 cases of splenic marginal zone lymphoma (2%). Inversely, KLF2, TNFAIP3 and MYD88, common mutations in splenic marginal zone lymphoma, were rare (one KLF2 mutant in 42 samples; 2%) or absent (TNFAIP3 and MYD88) in splenic diffuse red pulp lymphoma. These findings define an original genetic profile of splenic diffuse red pulp lymphoma and suggest that the mechanisms of pathogenesis of this lymphoma are distinct from those of splenic marginal zone lymphoma and hairy-cell leukemia.