Healthy herds in the phytoplankton: the benefit of selective parasitism

The impact of selective predation of weaker individuals on the general health of prey populations is well-established in animal ecology. Analogous processes have not been considered at microbial scales despite the ubiquity of microbe-microbe interactions, such as parasitism. Here we present insights into the biotic interactions between a widespread marine thraustochytrid and a diatom from the ecologically important genus Chaetoceros. Physiological experiments show the thraustochytrid targets senescent diatom cells in a similar way to selective animal predation on weaker prey individuals. This physiology-selective targeting of ‘unhealthy’ cells appears to improve the overall health (i.e., increased photosynthetic quantum yield) of the diatom population without impacting density, providing support for ‘healthy herd’ dynamics in a protist–protist interaction, a phenomenon typically associated with animal predators and their prey. Thus, our study suggests caution against the assumption that protist–protist parasitism is always detrimental to the host population and highlights the complexity of microbial interactions.Subject terms: Microbial ecology, Water microbiology

Animal predators can exert overall positive effects on the health of prey populations by removing individuals with suboptimal health [1, 2] in a manner that has been termed ‘healthy herd’ dynamics [3]. While such top-down processes are well-established in animal ecology [1–3], they have largely been unconsidered in microbe-microbe interactions.Protist–protist parasitism is widespread in the marine environment [4] and is generally considered to be detrimental to host populations [5, 6]. However, despite their ubiquity, the ecophysiological impact of protist–protist parasitism remains poorly understood. An important case that necessitates investigation is protist parasitism of diatoms, which have limited representation with culture-dependent model systems despite the significance of diatoms in marine ecosystem functioning and global primary production [7].We observed and isolated a heterotrophic protist growing epibiotically on moribund and dead Chaetoceros sp. diatoms from a summer bloom at Station L4 in the Western English Channel off Plymouth (UK) (Fig. 1A, B; Supplementary Figs. 1 and 2; Supplementary Methods). Single-cell picking achieved diatom and parasite co-cultures and uninfected host diatoms. The 18 S rRNA gene V4 region of the protist (termed ‘ThrauL4’) identified the epibiont as a novel thraustochytrid (Stramenopila; Labyrinthulomycota; Thraustochytrida) (Supplementary Fig. 3). Searching for ThrauL4 18 S rRNA gene homologues in the Ocean Sampling Day dataset revealed that the parasite has a wide distribution in coastal temperate regions (Supplementary Fig. 4).Open in a separate windowFig. 1Growth experiments demonstrate that thraustochytrids preferentially target and grow on unhealthy diatom cells.A Differential interference contrast (DIC) image of Chaetoceros chain exhibiting different degrees of infection by ThrauL4. Uninfected cell (un), a lightly infected cell (li), heavily infected cells (hi) and a dead, empty frustule (d). Scale bar = 20 µm. B Scanning Electron Micrograph (SEM) of a Chaetoceros diatom swarmed by ThrauL4. Scale bar = 5 µm. C ThrauL4 growth dynamics on a selected range of diatoms and dinoflagellates (Alexandrium minutum and Prorocentrum minimum) (±SEM, n = 3). D Chaetoceros growth with ThrauL4 (±SEM, n = 5). Dashed lines demarcate the lag (1), exponential (2) and stationary (3) phases of Chaetoceros growth. E Time-lapse of Chaetoceros-ThrauL4 showing ThrauL4 colonising unhealthy cells. Asterisk = cytoplasmic bleb from unhealthy diatom. Arrowhead = initial thraustochytrid colonisation. Timestamp = HH:MM:SS. Difference in the abundance (F) and prevalence (G) of parasites in healthy (control), stressed and dead Chaetoceros populations (n = 5) inoculated with ThrauL4 following heat stress exposure. ANOVA Tukey’s HSD n.s p > 0.05 (not significant), *p < 0.05, **p < 0.01, ***p < 0.001. H Example diatom exposed to different laser powers used to generate individual Chaetoceros cells of varying health. Red channel overlay demarks chlorophyll autofluorescence. Scale bar = 5 µm. I Time taken for individual diatom cells (n = 15) exposed to varying laser treatments to be colonised by ThrauL4. J Diagrammatic representation of the proposed diatom-thraustochytrid interaction cycle based on time-lapse microscopy observations (see Supplementary Videos).Stable Chaetoceros-ThrauL4 co-cultures permitted the characterisation of ThrauL4 internal structures (Supplementary Figs. 5 and 6), epibiotic growth (Fig. 1A, B; Supplementary Figs. 7 and 8) and infection dynamics (Fig. 1C, D). ThrauL4 also attached to other diatoms (Odontella sinensis, Ditylum brightwellii and Coscindodiscus sp.) in a similar manner to Chaetoceros sp. but not dinoflagellates (Fig. 1C; Supplementary Fig. 9).The proportion of diatom cells with ThrauL4 attached increased when Chaetoceros sp. cells entered the stationary growth phase (Fig. 1D). Time-lapse microscopy revealed the dynamic nature of the ThrauL4-diatom interaction (Fig. 1E, Supplementary Movies 1–6), with the motile ThrauL4 apparently targeting physiologically ‘unhealthy’ cells identified by cytoplasmic blebbing prior to colonisation (Fig. 1E).We set out to test the hypothesis that ThrauL4 targeted unhealthy diatoms using population-level ecophysiology experiments. When introduced to heat-stressed diatom populations, ThrauL4 had a higher fitness (i.e. became more abundant) and infected more Chaetoceros sp. cells than when exposed to healthy un-stressed diatoms (Fig. 1F, G), confirming more optimal growth of the parasite amongst unhealthy diatom populations. Furthermore, selective targeting was also demonstrated at the single-cell level using laser-damaged individual cells and time-lapse microscopy (Fig. 1H, I). 80% of stressed cells and 60% of dead cells were colonised by ThrauL4 during the 30 min experimental period, whereas diatoms in healthy control populations were un-colonised.These results led us to investigate the physiological impact of thraustochytrid parasitism on host diatom populations by comparing the dynamics and health of parasite exposed and non-exposed Chaetoceros sp. populations (Fig. 2A–C). Based on the previous growth experiments showing ThrauL4 proliferation during the diatom stationary phase (Fig. 1D), Chaetoceros sp. cultures grown to their stationary phase after 7 d were chosen to mimic environmental bloom decline. Using the photosynthetic quantum yield (Fv/Fm) as a proxy for overall diatom health [8], after 8 d, the parasitized Chaetoceros sp. populations were consistently healthier than those in the control non-exposed populations (Fig. 2A). Diatom population density was similar in both treatments (Fig. 2B) and parasite prevalence peaked after 8 days (Fig. 2C). In a separate experiment to investigate the role of genotype specificity in ThrauL4 parasitism, we generated a clonal Chaetoceros sp. population by single-cell picking and exposed the population to ThrauL4 cultures growing independently from diatoms. Although the clonal population declined in health more rapidly overall, ThrauL4 parasitism also resulted in healthier populations (Fig. 2D–F) suggesting that these results are a not an artefact of genotype specificity and succession.Open in a separate windowFig. 2Selective targeting of unhealthy diatom cells by thraustochytrids improves the overall health of the diatom population.A–C Population dynamics of the Fv/Fm (A) and total number (B) of stationary Chaetoceros diatoms for control and parasitized diatom populations over the experimental period (±SEM, n = 5). Welch’s t-test *p < 0.05, **p < 0.01, ***p < 0.001. The parasite prevalence did not exceed about a third of the total population (C) (±SEM, n = 5). Parasites added at 0 d. In a separate experiment (D–F), a clonal Chaetoceros population was generated. Population dynamics of the Fv/Fm (D), total number (E) and infection prevalence (F) of stationary Chaetoceros diatoms for control and parasitized populations made clonal by single-cell picking (±SEM, n = 5). Significance values as above. Parasites added at 0 day. Taken together these results indicate that preferential thraustochytrid parasitism of unhealthy diatoms strengthens the overall health of the population therefore providing evidence for the ‘healthy herd’ hypothesis in a phytoplankton population, which is summarised diagrammatically in (G).By removing physiologically weaker individuals from the population, the remaining cells will constitute an overall healthier population. However, other mechanisms may also promote an overall healthier diatom population. It may be that selective parasitism relieves nutrient competition between unhealthy and healthy individuals. In the natural environment, diatom-diatom competition is a major growth limiting factor [9, 10] and removing the pressure exerted by weaker cells may allow the population to be more robust. It is also possible that the thraustochytrid could be ‘cleaning’ the population by preventing the build-up of toxic waste products or the proliferation of detrimental co-culture bacteria in an analogous way to how carrion removal by vultures prevents the spread of diseases to mammals [11]. In addition, thraustochytid parasitism could accelerate nutrient recycling by releasing nutrients from dying cells. The consequences of physiology-selective diatom parasitism should be assessed in the marine environment, including impacts at the community scale and in the context of ecosystem functioning.The proposed influence of thraustochytrid parasitism on diatom population health is summarised in Fig. 2G. We suggest that this thraustochytrid-diatom interaction provides evidence of ‘healthy herd’ dynamics in a protist–protist interaction, an ecological phenomenon typically associated with animal predator-prey interactions [3]. As we show here with ThrauL4, animal predators such as lions [12], cougars [13], African wild dogs [14], and wolves [15] have been shown to target prey with suboptimal health. The ‘healthy herd’ hypothesis states that by selective predation on unhealthy prey, predators increase the overall health of the prey population by increasing resource availability or by removing potential carriers of disease [3]. Evidence for ‘healthy herd’ dynamics where predation generates healthier prey populations has also been demonstrated in lobster-sea urchin [16], fish-Daphnia [17], and fox-grouse [18] predator–prey systems. Here, we provide analogous supportive evidence from a marine protist–protist system.‘Heathy herd’ dynamics between protists challenges the assumption that protist–protist parasitism is always detrimental to the host population and raises caution in this assumption in ecosystem modelling or inference from molecular ecology surveys (e.g., metabarcoding). Our results have demonstrated the potential complexity of protist–protist symbioses, highlighting the value of culture-based experimentation and the importance of developing model co-culture systems in resolving complex ecological interactions. The underpinning biology and ecological importance in natura of such interactions now require further investigation.     

Protists as catalyzers of microbial litter breakdown and carbon cycling at different temperature regimes

Soil bacteria and fungi are key drivers of carbon released from soils to the atmosphere through decomposition of plant-derived organic carbon sources. This process has important consequences for the global climate. While global change factors, such as increased temperature, are known to affect bacterial- and fungal-mediated decomposition rates, the role of trophic interactions in affecting decomposition remains largely unknown. We designed synthetic microbial communities consisting of eight bacterial and eight fungal species and tested the influence of predation by a model protist, Physarum polycephalum, on litter breakdown at 17 and 21 °C. Protists increased CO₂ release and litter mass loss by ~35% at 17 °C lower temperatures, while they only had minor effects on microbial-driven CO₂ release and mass loss at 21 °C. We found species-specific differences in predator–prey interactions, which may affect microbial community composition and functioning and thus underlie the impact of protists on litter breakdown. Our findings suggest that microbial predation by fast-growing protists is of under-appreciated functional importance, as it affects decomposition and, as such, may influence global carbon dynamics. Our results indicate that we need to better understand the role of trophic interactions within the microbiome in controlling decomposition processes and carbon cycling.Subject terms: Climate-change impacts, Soil microbiology, Microbial ecology

Soil microorganisms, mainly bacteria and fungi, are major drivers of soil carbon cycling through their decomposing activity of plant-derived carbon [1, 2] and their role in soil carbon stabilization [3, 4]. This has important consequences for atmospheric carbon concentrations and thereby, for ongoing climate change [5, 6]. It is well established that large-scale abiotic factors, such as climate, affect microbial activity and thereby, decomposition rates [7]. More recently it was shown that climate-independent variation in local-scale factors can drive broad-scale variation in decomposition rates [8]. Among these might be microbial predators that vary and affect microbial community composition and functioning at the local scale [9]. However, how microbial predators alter litter breakdown remains largely unknown.Protists are major microbial predators of soil bacteria and to some extent fungi [10]. Protists are the taxonomically most diverse eukaryotes and occupy all key functional roles in soil food webs [10]. Most soil protists are phagotrophic [11] and prey on bacteria and fungi, which leads to changes in microbial biomass, activity, and community structure [10]. This is likely to have important functional consequences, including impacts on litter decomposition processes and thereby, the global carbon cycle. However, there is little experimental evidence underpinning how protists impact decomposition. Moreover, both protist and microbial activity are affected by temperature [9, 12], but whether temperature also modifies protist-induced changes in microbial functioning remains unknown.To test the role of protist predation on microbial-driven decomposition we inoculated microcosms of synthetic microbial communities consisting of sixteen bacterial and fungal species (Tables S1 and S2) to sterilized oak litter (Quercus robur) at both 17 and 21 °C. After one week we added protists of the model species Physarum polycephalum at three different concentrations (no protists, and low, medium, and high concentration). This resulted in a full-factorial design with 16 treatments: 2 microbial inocula (yes/no) × 2 temperatures (17/21 °C) × 4 protist concentrations (Table S3) and we used six replicates per treatment. Microcosms without microbial inocula were established to test for successful establishment of the synthetic microbial community and were not used for further analyses as they did not remain sterile. For each microcosm, we measured CO₂ production, litter mass loss and litter nitrogen and carbon content of the remaining litter. See supplementary methods for further details.Before the addition of protists, microcosms with bacteria and fungi produced more CO₂ than microbial-free ones (F_1,92 = 431.16, p < 0.001), and this effect was not different between temperatures (F_1,92 = 0.04, p = 0.846; Fig. S1), indicating successful establishment of a synthetic microbial community after inoculation. After protistan addition, there was no interactive effect of protists and temperature on CO₂ production (F_3,40 = 1.48, p = 0.234). However, both increased temperature (F_1,40 = 14.96, p < 0.001) and presence of protists irrespective of their concentration (F_3,40 = 3.24, p = 0.032) increased CO₂ production (Fig. 1a). A posthoc analysis indicated that protist addition effects appeared stronger at lower than at higher temperatures (Fig. 1; please note that boxplots highlight medians while posthoc tests compare means). An interaction between the protist and temperature treatment affected litter mass loss (F_3,40 = 10.50, p < 0.001; Fig. 1b), indicating that the addition of protists at all concentrations increased litter mass loss at 17 °C by more than 35% on average, but not at 21 °C (Fig. 1b). The addition of protists did not affect litter carbon (C) (F_3,40 = 0.55, p = 0.653) and nitrogen (N) content (F_3,40 = 0.03, p = 0.993) and the litter C:N ratio (F_3,40 = 0.04, p = 0.990) at the end of the experiment (Fig. S2). Litter N content was higher at 21 than at 17 °C, indicating higher N loss during decomposition at lower temperatures (F_1,30 = 7.42, p = 0.010; Fig. S2b), resulting in higher C:N ratios at 17 °C than at 21 °C (F_1,40 = 8.08, p = 0.007).Open in a separate windowFig. 1Changes in microbial CO₂ production and litter decomposition rates as induced by protist predators.Boxplots showing (a) cumulative CO₂ respiration (measured from the addition of protists until the end of the experiment) and (b) litter mass loss for microcosms with no protists or low, medium (mid) or high concentrations of protists (x-axis) at 17° and 21 °C. Different letters above the boxes indicate significant differences (p < 0.05) between treatments, as was indicated in a Tukey HSD posthoc test. Tukey tests were carried out across the protists × temperature interactions, so letters can be compared across facets.Interaction-assays in split-petri dishes to test for volatile-induced microbial effects (Fig. S3) showed that protist growth (plasmodial length) was affected by bacterial (F_5,23 = 63.22, p < 0.001) and fungal volatiles (F_5,24 = 12.29, p < 0.001; Fig. 2). Presence of Collimonas pratensis T91, Pseudomonas sp. AD21 and Trichoderma citrinoviride reduced protist growth most strongly (Fig. 2). The overall negative effects of bacteria and fungi on protists likely through volatiles contradict with the variable effects of volatiles on other protist species which ranged from stimulation to inhibition [13]. But as inhibition differed between microbial species, some potentially efficient decomposers might benefit through a reduction of competition from more easily preyed microbes, which could explain the observed increased decomposition rates. Yet, other mechanisms are likely to contribute to increased decomposition in presence of predators, such as predation-induced increased microbial activity or alternative enzyme production- details to be explored in future studies.Open in a separate windowFig. 2Bacterial and fungal long-distance effects on protist growth.Boxplots showing plasmodial length of the model protist Physarum polycephalum in response to different (a) bacterial and (b) fungal taxa (x-axis) that were part of the microbial decomposer communities (Tables S1 and S2). C is the control with only nutrient agar without bacteria (left) or potato dextrose agar without fungi (right). Different letters above the bars indicate that protist responses differed significantly (p < 0.05) between the microbial species in a Tukey HSD test. Tukey HSD tests were carried out for bacteria and fungi separately, therefore letters should be compared within panels only.Our results support previous findings showing that predator–prey interactions within the microbiome affect microbial-derived CO₂ production [14], but we extend this knowledge and show that this effect tends to of lower importance at higher temperature. Furthermore, we now show that microbial predators alter litter decomposition in a temperature-dependent manner, with an increased importance at lower temperature. This result extends the known importance of larger-sized soil animals in increasing litter decomposition [15, 16] and contrasts previous findings that microscopic predators (mostly protists and nematodes) have a limited effect on litter breakdown [16]. Mechanistically, protists might increase decomposition via microbe-specific predator–prey interactions [10] that change microbial community composition and functioning [17]. Our interaction-assays suggests that microbial predator–prey interactions mediated by volatiles could differ, which might benefit some efficient microbial decomposers.The effect of protists on litter decomposition was strongest at lower temperatures, contradicting previous findings that larger soil animals have increased effects on decomposition at higher temperatures [18]. This discrepancy might be explained by the higher microbial diversity in our model communities compared to often single-decomposer model species used before, in which predation might favor metabolically active microorganisms [10]. The effect of predation on microbial-driven decomposition seems to differ between protists and soil animals, as soil animals were shown to have limited effects on decomposition rates [16]. The increased importance of protist predation on microbial decomposition at lower temperatures suggest a more profound role of predation on carbon cycling in colder, non-tropical climates that host most microbial biomass [19] and store most carbon [20]. If this pattern can be confirmed with a wider range of protists, and in natural soils rather than this simplified laboratory assay, these microbial predators may play a key role in accelerating the global carbon cycle. Further studies should test exactly those by using realistic climate scenarios, more diverse protists and microbial decomposers, and in natural settings to untangle the importance of protists on decomposition and the carbon cycle. In turn, even more detailed laboratory analyses are needed to unreliably determine the exact mechanisms of how protists affect decomposition.In summary, we reveal microbiome predation by protists as a key driver of microbial-driven decomposition with potential impacts on the global carbon cycle. Further integrated microbiome analyses are needed to investigate how and under which conditions microbial predation affects litter decomposition and if and how protists contribute to the global carbon cycle.     

Specific inactivation of an antifungal bacterial siderophore by a fungal plant pathogen

Bacteria and fungi secrete many natural products that inhibit each other’s growth and development. The dynamic changes in secreted metabolites that occur during interactions between bacteria and fungi are complicated. Pyochelin is a siderophore produced by many Pseudomonas and Burkholderia species that induces systemic resistance in plants and has been identified as an antifungal agent. Through imaging mass spectrometry and metabolomics analysis, we found that Phellinus noxius, a plant pathogen, can modify pyochelin and ent-pyochelin to an esterification product, resulting in reduced iron-chelation and loss of antifungal activity. We also observed that dehydroergosterol peroxide, the fungal metabolite, is only accumulated in the presence of pyochelin produced through bacteria–fungi interactions. For the first time, we show the fungal transformation of pyochelin in the microbial interaction. Our findings highlight the importance of understanding the dynamic changes of metabolites in microbial interactions and their influences on microbial communities.Subject terms: Microbial ecology, Metabolomics

Microorganisms use various strategies to establish themselves within an ecological niche while facing keen competition in the environment. Natural products such as antibiotics, quorum sensing molecules, and siderophores are crucial in microbial interactions [1–3]. Certain microorganisms are equipped with uptake systems that enable them to acquire siderophores, even by those that may not produce them [4]. For example, pyochelin is a siderophore produced by many Pseudomonas and Burkholderia strains. Such bacterial strains are commonly found in soils, as endophytes, and from the rhizosphere where they may inhibit plant pathogens [5, 6].Burkholderia cenocepacia 869T2 was isolated as an endophyte and showed beneficial abilities to control banana Fusarium wilt [7]. It harbors many biosynthetic gene clusters of secondary metabolites, such as pyochelin, pyrrolnitrin, and pyrroloquinoline quinone [8]. Recently, we found that this strain could temporarily inhibit the growth of P. noxius, a fungal pathogen of brown root rot disease, which is prevalent in tropical and subtropical regions and has a wide host range covering over 200 plant species [9]. However, in the competition between fungi and bacteria, P. noxius can resist this inhibition and overwhelm bacterial colonies after 1–2 weeks under dual-culture conditions (Fig. S1). These results imply that fungi might have resistance responses and undergo metabolic changes in bacteria–fungi interactions [10]. Here we unveiled metabolic changes in the competitive interaction between B. cenocepacia 869T2 and P. noxius 2252 using the matrix-assisted laser desorption ionization-time of flight imaging mass spectrometry (MALDI-TOF IMS) [11, 12].We specifically monitored the metabolites in the inhibition region of B. cenocepacia 869T2 and P. noxius 2252 dual-culture using MALDI-TOF IMS. Several induced or enzymatically modified metabolites were detected, including m/z 275, 362, 383, and 427 (Fig. 1A). In particular, pyochelin (m/z 325), surrounding the B. cenocepacia 869T2 colony, showed asymmetric distribution in dual-culture samples. Near the P. noxius 2252 mycelia, a new metabolite with m/z 383 was detected with a complementary distribution to pyochelin (Fig. 1A). In LC-MS/MS-based molecular networking analysis [13], we found that this new metabolite structure is an esterification product of pyochelin and glycolic acid, which we named pyochelin-GA (Fig. 1B). We then constructed a pchF-null mutant strain, ΔpchF, which cannot produce pyochelin, and then dual cultured it with P. noxius. Pyochelin and pyochelin-GA were not observed in the MALDI-TOF IMS and LC-MS analysis of dual-culture samples (Fig. 1A and Fig. S2). We further inoculated P. noxius 2252 with pyochelin-GA-free extract harvested from B. cenocepacia 869T2 single culture, and the complementary distribution of pyochelin and pyochelin-GA was observed by MALDI-TOF IMS again (Fig. S3). These results demonstrated that pyochelin-GA was transformed from pyochelin by P. noxius 2252, rather than produced by B. cenocepacia 869T2 under dual-culture conditions.Open in a separate windowFig. 1Metabolic changes in the bacteria–fungi interaction.A Spatial distribution of selected mass signals (m/z) in MALDI-TOF IMS analysis of Phellinus noxius 2252 (Pn2252) dual-cultured with Burkholderia cenocepacia 869T2 (869T2) and a pchF-null mutant strain (Δ pchF). B Molecular networking analysis of pyochelin and analogs from the dual-culture sample. The red node is pyochelin, and the green node is pyochelin-GA. The structures of pyochelin, pyochelin-GA, and dehydroergosterol peroxide (DHEP), together with their mass signals in MALDI-TOF IMS, are shown. C Iron-chelating abilities of pyochelin and pyochelin-GA were evaluated by Chrome Azurol S liquid assay using different concentrations (2.5, 1.25, 0.63, 0.31, and 0.16 mM, n = 3). Proportions of siderophore units are shown in Fig. S14. D Fungal transformation of pyochelin and ent-pyochelin by treating P. noxius 2252 with ethyl acetate crude extracts of B. cenocepacia 869T2, Pseudomonas aeruginosa PAO1, and P. protegens Pf-5 for 8 h. LC-MS was used to monitor the signals of pyochelin (red), ent-pyochelin (blue), and transformation product 383 (black).The chemical structure of pyochelin-GA was further confirmed via total synthesis, NMR, and LC-MS/MS analysis (Supplementary Material and Methods, and Figs. S4–7). The purified pyochelin and pyochelin-GA were also evaluated for their iron-chelating ability. Chrome Azurol S assay indicated that pyochelin had the dose-dependent iron-chelating ability, but pyochelin-GA had lower iron-binding efficiency (Fig. 1C, Fig. S8). Pyochelin chelates iron in the extracellular medium and transports it into cells via the specific outer membrane transporter FptA. The X-ray structure of FptA-pyochelin-Fe indicated that the terminal carboxylic acid of pyochelin plays an essential role in the iron uptake ability [14, 15]. Our docking analysis suggested that the glycolic ester moiety of pyochelin-GA would affect the binding pocket shape of FptA and result in different binding properties compared to FptA-pyochelin (Fig. S9).Pyochelin and ent-pyochelin are produced independently by different biosynthetic gene clusters in Pseudomonas species [16]. To determine whether P. noxius 2252 can transform both enantiomers via this esterification process, we treated P. noxius 2252 with the extracts of pyochelin producers (P. aeruginosa PAO1 and B. cenocepacia 869T2) and an ent-pyochelin producer (P. protegens Pf-5). After 8 h of treatment, both pyochelin and ent-pyochelin were converted to pyochelin-GA (or ent-pyochelin-GA) (Fig. 1D), demonstrating this is a non-stereospecific transformation.To better understand the iron-chelating ability of pyochelin, we used pyochelin and pyochelin-GA to treat P. noxius 2252 under iron-deficiency conditions, by adding the iron chelator deferoxamine, and iron-rich conditions by adding FeCl₃ (Fig. 2). Pyochelin-GA did not affect the growth of P. noxius 2252 under all conditions. However, P. noxius 2252 was more sensitive to pyochelin in iron-deficient conditions and more resistant to pyochelin in iron-rich conditions, demonstrating that iron availability directly affected the tolerance of P. noxius 2252 to pyochelin. A similar phenomenon was reported previously for Aspergillus fumigatus [17].Open in a separate windowFig. 2Pyochelin inhibition of mycelial growth of Phellinus noxius 2252 is inversely associated with iron concentration.Pyochelin-GA did not have an inhibition effect on P. noxius 2252. Potato dextrose agar (PDA) with deferoxamine (DFO; 200 and 400 µM) was used to mimic iron-deficiency conditions. Iron-rich conditions was prepared by adding FeCl₃ (200 and 400 µM) in PDA. P. noxius 2252 was treated with 0.03, 0.06, 0.12, and 0.24 µmol of pyochelin or pyochelin-GA at 30 °C for 24 h. The antifungal assay was performed in two biological replicates.Using MALDI-TOF IMS analysis of the dual-culture of B. cenocepacia 869T2 and P. noxius 2252, we observed that several metabolites (e.g., m/z 275, 362, and 427) were only observed in the boundary of fungal mycelia (Fig. 1A). Although those metabolites were not detected in the dual-culture of ΔpchF and P. noxius 2252 (Fig. 1A), they were present when we treated P. noxius 2252 with pyochelin (Fig. S10). We identified the metabolite associated with m/z 427 as dehydroergosterol peroxide (DHEP) (Fig. S11), which was initially oxidized from ergosterol and dehydroergosterol [18]. Pyochelin can enhance intercellular reactive oxygen species (ROS) and ultimately disrupts membrane integrity, leading to cell death [17, 19, 20]. To clarify whether ROS induced the accumulation of DHEP, we treated P. noxius 2252 with pyochelin, pyochelin-GA, and 2,2′-bipyridyl (an iron chelator). Pyochelin and 2,2′-bipyridyl showed antifungal effects on P. noxius 2252 and induced ROS production (Fig. S12). However, the accumulation of DHEP in P. noxius 2252 was only associated with pyochelin treatment (Fig. S13). The induction of ROS in P. noxius 2252 by pyochelin and pyochelin-GA was not significantly different (Fig. S14). Therefore, we predict that pyochelin-induced accumulation of DHEP in P. noxius 2252 is independent of ROS production and iron-deficiency.Overall, we demonstrate that pyochelin transformation by fungi, in the interaction between pyochelin-producing bacteria and the plant pathogen P. noxius transforms pyochelin and ent-pyochelin into pyochelin-GA (and ent-pyochelin-GA). This product no longer functions as an iron chelator and no longer shows antifungal activity. The production of a fungal metabolite, dehydroergosterol peroxide, was induced explicitly by pyochelin through an unknown mechanism. These results highlight the importance of monitoring dynamic changes of metabolites in situ to better understand the functions and influences of metabolites on microbial community interactions.     

Unexpectedly high mutation rate of a deep-sea hyperthermophilic anaerobic archaeon

Deep-sea hydrothermal vents resemble the early Earth, and thus the dominant Thermococcaceae inhabitants, which occupy an evolutionarily basal position of the archaeal tree and take an obligate anaerobic hyperthermophilic free-living lifestyle, are likely excellent models to study the evolution of early life. Here, we determined that unbiased mutation rate of a representative species, Thermococcus eurythermalis, exceeded that of all known free-living prokaryotes by 1-2 orders of magnitude, and thus rejected the long-standing hypothesis that low mutation rates were selectively favored in hyperthermophiles. We further sequenced multiple and diverse isolates of this species and calculated that T. eurythermalis has a lower effective population size than other free-living prokaryotes by 1-2 orders of magnitude. These data collectively indicate that the high mutation rate of this species is not selectively favored but instead driven by random genetic drift. The availability of these unusual data also helps explore mechanisms underlying microbial genome size evolution. We showed that genome size is negatively correlated with mutation rate and positively correlated with effective population size across 30 bacterial and archaeal lineages, suggesting that increased mutation rate and random genetic drift are likely two important mechanisms driving microbial genome reduction. Future determinations of the unbiased mutation rate of more representative lineages with highly reduced genomes such as Prochlorococcus and Pelagibacterales that dominate marine microbial communities are essential to test these hypotheses.Subject terms: Archaea, Population genetics

One theory for the origin of life is that the last universal common ancestor was an anaerobic hyperthermophilic organism inhabiting the deep-sea hydrothermal vents, as these environments display a few characteristics paralleling the early Earth [1]. While hydrothermal vents vary with chemical parameters, they all share a high temperature zone near the black chimney with anaerobic fluid from it. In the past decades, great efforts were made to understand the metabolic strategies deep-sea hyperthermophiles use to conserve energy and cope with physicochemical stresses, and to appreciate the molecular mechanisms leading to the stabilization of nucleic acids and proteins at exceedingly high temperatures [2, 3]. However, little is known whether they have a low or high intrinsic (i.e., not selected by environmental pressure) rate to change their genetic background information and whether this intrinsic potential itself is a result of selection shaped by these unique habitats.A previous population genomic analysis showed that protein sequences are under greater functional constraints in thermophiles than in mesophiles, suggesting that mutations are functionally more deleterious in thermophiles than in mesophiles [4]. This explanation is also supported by experimental assays showing nearly neutral mutations in temperate conditions become strongly deleterious at high temperature [5]. Furthermore, fluctuation tests on a hyperthermophilic archeaon Sulfolobus acidocaldarius [6] and a hyperthermophilic bacterium Thermus thermophilus [7] consistently showed that hyperthermophiles have much lower mutation rate compared to mesophiles. This appears to support the hypothesis that selection favors high replication fidelity at high temperature [5].Nevertheless, mutation rates measured using fluctuation experiments based on reporter loci are known to be biased, since the mutation rate of the organism is extrapolated from a few specific nonsynonymous mutations enabling survival in an appropriate selective medium, which renders the results susceptible to uncertainties associated with the representativeness of these loci and to inaccuracies of the assumptions made in extrapolation methods [8–10]. These limitations are avoided by the mutation accumulation (MA) experiment followed by whole-genome sequencing (WGS) of derived lines. In the MA part, multiple independent MA lines initiated from a single progenitor cell each regularly pass through a single-cell bottleneck, usually by transferring on solid medium. As the effective population size (N_e) becomes one, selection is unable to eliminate all but the lethal mutations, rendering the MA/WGS an approximately unbiased method to measure the spontaneous mutation rate [11].Members of the free-living anaerobic hyperthermophilic archaeal family Thermococcaceae are among the dominant microbial lineages in the black-smoker chimney at Guaymas Basin [12] and other deep-sea hydrothermal vents [13, 14]. This family only contains three genera: Thermococcus, Pyrococcus and Palaeococcus. In this study, the MA/WGS procedure was applied to determine the unbiased spontaneous mutation rate of a representative member Thermococcus eurythermalis A501, a conditional pizeophilic archaeon which can grow equally well from 0.1 MPa to 30 MPa at 85 °C [15, 16]. The MA lines were propagated at this optimal temperature on plates with gelrite which tolerates high temperature, and the experiment was performed under normal air pressure and in strictly anaerobic condition (Fig. 1A–D). To the best of our knowledge, this is the first report of unbiased mutation rate of a hyperthermophile and an obligate anaerobe.Open in a separate windowFig. 1Experimental determination of the unbiased mutation rate of the Thermococcus eurythermalis A501 is challenging because this archaeon has unusual physiology (i.e., obligate anaerobic and obligate hyperthermophilic).A The preparation of anaerobic high temperature tolerant gelrite plate. After sterilization and polysulfide addition via syringe, the plates are made in an anaerobic chamber. B The incubation of the strain T. eurythermalis A501 at 85 °C in liquid medium. C The initiation of mutation accumulation (MA) by spreading cells from a single founding colony to 100 lines. Plates are placed in an anaerobic jar for incubation in strictly anaerobic condition at 85 °C. D The MA process followed by whole-genome sequencing and data analysis. Single colony of each line is transferred to a new plate for N times (here N = 20). E Base-substitution mutations and insertion/deletion mutations across the whole genome of T. eurythermalis. The dashed vertical line separates the chromosome and plasmid. The height of each bar represents the number of base-substitution mutations across all MA lines within 10 kbp window. Green and red triangles denote insertion and deletion, respectively. The locus tags of the 14 genes with statistical enrichment of mutations are shown.Our MA experiment allowed accumulation of mutations over 314 cell divisions (after correcting the death rate (Table S1) [17]) in 100 independent lines initiated from a single founder colony and passed through a single cell bottleneck every day. By sequencing genomes of 96 survived lines at the end of the MA experiment, we identified 544 base-substitution mutations over these lines (Table S2), which translates to an average mutation rate (µ) of 85.01 × 10⁻¹⁰ per cell division per nucleotide site (see Supplementary information). The ratio of accumulated nonsynonymous to synonymous mutations (371 vs 107) did not differ from the ratio of nonsynonymous to synonymous sites (1,485,280 vs 403,070) in the A501 genome (χ² test; p > 0.05). Likewise, there was no difference of the accumulated mutations between intergenic (65) and protein-coding sites (478) (χ² test; p > 0.05). These are evidence for minimal selective elimination of deleterious mutations during the MA process. In general, the mutations were randomly distributed along the chromosome and the plasmid, though 86 base-substitution mutations fell into 14 genes which showed significant enrichment of mutations (bootstrap test; p < 0.05 for each gene) and 52 out of the 86 base-substitution mutations were found in five genes (TEU_RS04685 and TEU_RS08625-08640 gene cluster) (Fig. 1E, Table S3). A majority of mutations in these five genes may have inactivated these genes (38 out of 71 in the former gene and 33 out of 43 in the latter gene cluster) either by nonsense mutation or insertion-deletion (INDEL) mutation. The phenomenon of mutation clustering is not unique to this organism; it was reported in another MA study with the yeast Schizosaccharomyces pombe, and these genomic regions may represent either mutational hotspots or that mutations confer selective advantages under experimental conditions [18]. The TEU_RS04685 encodes the beta subunit of the sodium ion-translocating decarboxylase which is an auxiliary pathway for ATP synthesis by generating sodium motive force via decarboxylation [19], and the TEU_RS08625-08640 encodes a putative nucleoside ABC transporter. These genes appear to be important for energy conservation in the highly fluctuating deep-sea hydrothermal fluids. Under the culture conditions in which peptides and amino acids were stably and sufficiently supplied (see the TRM medium recipe in Supplementary information), however, these genes may be dispensable because peptides and amino acids are the preferred carbon and energy sources for T. eurythermalis [15]. On the other hand, some of these genes (e.g., TEU_RS08625) were shown to be upregulated under alkaline stress [16], and thus may be similarly induced under the culture condition in which pH is elevated compared to the vents. Besides, the laboratory condition differed from the vents in a number of other physicochemical features including hydrostatic pressure (0.1 MPa during the MA process versus 20 MPa in situ), temperature and salinity, which likely imposed additional selective pressures on the mutation accumulation processes. Taken together, deleting these genes were likely translated to a net fitness gain and were thus driven by selection. Removing these mutations led to a spontaneous mutation rate of 71.57 × 10⁻¹⁰ per cell division per site for T. eurythermalis A501. After removing the mutations in these 14 genes, both the accumulated mutations at nonsynonymous sites (288) relative to those (104) at synonymous sites (χ² test; p = 0.014) and the accumulated mutations at intergenic regions (65) relative to protein-coding regions (392) (χ² test; p = 0.013) showed marginally significant differences.To date, over 20 phylogenetically diverse free-living bacterial species and two archaeal species isolated from various environments have been assayed with MA/WGS, and their mutation rates vary from 0.79 × 10⁻¹⁰ to 97.80 × 10⁻¹⁰ per cell division per site [20]. The only prokaryote that displays a mutation rate (97.80 × 10⁻¹⁰ per cell division per site) comparable to A501 is Mesoplasma florum L1 [21], a host-dependent wall-less bacterium with highly reduced genome (~700 genes). Our PCR validation of randomly chosen 20 base-substitution mutations from two MA lines displaying highest mutation rates and of all nine INDEL mutations involving >10 bp changes across all lines (Table S2) indicates that the calculated high mutation rate did not result from false bioinformatics predictions.The extremely high mutation rate of T. eurythermalis is unexpected. One potential explanation in line with the “mutator theory” [22–24] is that high mutation rate may allow the organisms to gain beneficial mutations more rapidly and thus is selectively favored in deep-sea hydrothermal vents where physicochemical parameters are highly fluctuating. Alternatively, high mutation rate is the result of random genetic drift according to the “drift-barrier model” [21]. In this model, increased mutation rates are associated with increased load of deleterious mutations, so natural selection favors lower mutation rates. On the other hand, increased improvements of replication fidelity come at an increased cost of investments in DNA repair activities. Therefore, natural selection pushes the replication fidelity to a level that is set by genetic drift, and further improvements are expected to reduce the fitness advantages [11, 21]. These two explanations for the high mutation rate of T. eurythermalis are mutually exclusive, and resolving them requires the calculation of the power of genetic drift, which is inversely proportional to N_e.A common way to calculate N_e for a prokaryotic population is derived from the equation π_S = 2 × N_e × µ, where π_S represents the nucleotide diversity at synonymous (silent) sites among randomly sampled members of a panmictic population [25]. We therefore sequenced genomes of another eight T. eurythermalis isolates available in our culture collections. Like T. eurythermalis A501, these additional isolates were collected from the same cruise but varying at the water depth from 1987 m to 2009 m at Guaymas Basin. They differ by only up to 0.135% in the 16S rRNA gene sequence and share a minimum whole-genome average nucleotide identity (ANI) of 95.39% (Table S4), and thus fall within an operationally defined prokaryotic species typically delineated at 95% ANI [26]. Population structure analysis with PopCOGenT [27] showed that these isolates formed a panmictic population and that two of them were repetitive as a result of clonal descent (see Supplementary information). Using the median value of π_S = 0.083 across 1628 single-copy orthologous genes shared by the seven non-repetitive genomes, we calculated the N_e of T. eurythermalis to be 5.83 × 10⁶.Next, we collected the unbiased mutation rate of other prokaryotic species determined with the MA/WGS strategy from the literature [11, 28–30]. While the N_e data were also provided from those studies, the isolates used to calculate the N_e were identified based on their membership of either an operationally defined species (e.g., ANI at 95% cutoff) or a phenotypically characterized species (e.g., many pathogens), which often create a bias in calculating N_e [25]. We therefore again employed PopCOGenT to delineate panmictic populations from those datasets and re-calculated N_e accordingly. There was a significant negative linear relationship between µ and N_e on a logarithmic scale (dashed gray line in Fig. 2A [r² = 0.83, slope = −0.85, s.e.m. = 0.09, p < 0.001]) according to a generalized linear model (GLM) regression. This relationship cannot be explained by shared ancestry, as confirmed by phylogenetic generalized least square (PGLS) regression analysis (solid blue line in Fig. 2A [r² = 0.81, slope = −0.81, s.e.m. = 0.09, p < 0.001]). The nice fit of T. eurythermalis to the regression line validated the drift-barrier hypothesis. This is evidence that the high mutation rate of T. eurythermalis is driven by genetic drift rather than by natural selection.Open in a separate windowFig. 2The scaling relationship involving the base-substitution mutation rate per cell division per site (µ), the estimated effective population size (N_e), and genome size across 28 bacterial and two archaeal species.All three traits’ values were logarithmically transformed. The mutation rates of these species are all determined with the mutation accumulation experiment followed by whole-genome sequencing of the mutant lines. The mutation rate of species numbered 1–29 (blue) is collected from literature and that of the species 30 (red) is determined in the present study. Among the numbered species shown in the figure, the species #6 Haloferax volcanii is facultative anaerobic halophilic archaeon, and the species #30 is an obligate anaerobic hyperthermophilic archaeon. A The scaling relationship between µ and N_e. B The scaling relationship between µ and genome size. C The scaling relationship between genome size and N_e. Numbered data points 21–29 are not shown in A and C because of the lack of population dataset for estimation of N_e. The dashed gray lines and blue lines represent the generalized linear model (GLM) regression and the phylogenetic generalized least square (PGLS) regression, respectively. The Bonferroni adjusted outlier test for the GLM regression show that #7 Janthinobacterium lividum is an outlier in the scaling relationship between µ and N_e, and #9 Mesoplasma florum is an outlier in the scaling relationship between genome size and N_e. No outlier was identified in the PGLS regression results.As stated in the drift-barrier theory, high mutation rate is associated with a high load of deleterious mutations. In the absence of back mutations, recombination becomes an essential mechanism in eliminating deleterious mutations [31]. In support of this argument, the ClonalFrameML analysis [32] shows that members of the T. eurythermalis population recombine frequently, with a high ratio of the frequency of recombination to mutation (ρ/θ = 0.59) and a high ratio of the effect of recombination to mutation (r/m = 5.76). In fact, efficient DNA incorporation to Thermococcaceae genomes from external sources has been well documented experimentally [33, 34]. A second potentially important mechanism facilitating T. eurythermalis adaptation at high temperature is strong purifying selection at the protein sequence level, as protein sequences in thermophiles are generally subjected to stronger functional constraints compared to those in mesophiles [4, 35].Our result of the exceptionally high mutation rate of a free-living archaeon is a significant addition to the available collection of the MA/WGS data (Table S5), in which prokaryotic organisms with very high mutation rate have only been known for a host-dependent bacterium (Mesoplasma florum L1) with unusual biology (e.g., cell wall lacking). The availability of these two deeply branching (one archaeal versus the other bacterial) organisms adopting opposite lifestyles (one free-living versus the other host-restricted; one hyperthermophilic versus the other mesophilic; one obligate anaerobic versus the other facultative anaerobe), along with other phylogenetically and ecologically diverse prokaryotic organisms displaying low and intermediate mutation rates, provides an opportunity to help illustrate mechanisms potentially driving genome size evolution across prokaryotes. We found a negative linear relationship (dashed gray line in Fig. 2B [r² = 0.49, slope = −1.66, s.e.m. = 0.32, p < 0.001]) between genome size and base-substitution mutation rate, which is consistent with the hypothesis that increased mutation rate drives microbial genome reduction. We also showed a positive linear relationship (dashed gray line in Fig. 2C [r² = 0.47, slope = 0.24, s.e.m. = 0.06, p < 0.001]) between genome size and N_e, which suggests that random genetic drift drives genome reduction across prokaryotes. These correlations remain robust when the data were analyzed as phylogenetically independent contrasts (blue solid lines in Fig. 2B [r² = 0.47, slope = −1.75, s.e.m. = 0.34, p < 0.001] and in Fig. 2C [r² = 0.45, slope = 0.25, s.e.m. = 0.06, p < 0.001]). Our results are consistent with recent studies which employed mathematical modeling and/or comparative sequence analyses to show random genetic drift [36] and increased mutation rate [37] driving genome reduction across diverse bacterial lineages including both free-living and host-dependent bacteria. One benefit of the present study is that it directly measures µ and N_e, as compared to those recent advances which relied on proxies for these metrics (e.g., using the ratio of nonsynonymous substitution rate to synonymous substitution rate to represent N_e) to infer mechanisms of genome reduction.Despite this advantage, there are important caveats to our conclusions related to the mechanisms of genome reduction. The correlation analyses performed here are inspired by Lynch and colleagues’ work, who had great success explaining eukaryotic genome expansion with genetic drift [11, 38]. However, there are a few key differences of genomic features between prokaryotes and eukaryotes, which makes it more difficult to explain the correlation observed in prokaryotes. Importantly, genome sizes of eukaryotes can vary over several orders of magnitude, whereas those of free-living prokaryotes differ by only an order of magnitude [11], so there is much less variability to explain in prokaryotes. Moreover, eukaryotic genomes experience dramatic expansions of transposable elements which are often considered as genomic parasites, whereas prokaryotic genomes including those large ones are usually depleted with transposable elements and their genome size variations are largely driven by gene content [39]. Aside from these conceptual difficulties, the plots (Fig. 2B, C) are poorly populated with typical free-living species carrying small genomes such as the Prochlorococcus (mostly 1.6–8 Mb) and Pelagibacterales (1.3–1.5 Mb), which dominate the photosynthetic and heterotrophic microbial communities, respectively, in the ocean [40]. It has been generally postulated that bacterial species in these lineages have very large N_e [39–41], though there has been little direct evidence for it [42, 43]. If confirmed through the measurement of the unbiased mutation rate (µ) followed by the calculation of N_e based on µ, it might compromise the linear relationship between genome size and N_e observed here (Fig. 2C). It is also not necessarily appropriate to translate correlations to causal relationships. For example, the correlation between increased mutation rates and decreased genome sizes (Fig. 2B) does not necessarily mean that increased mutation rate drives genome reduction. This is because high mutation rates are observed in species with small N_e. Given that deletion bias is commonly found in prokaryotes [44, 45], genome reduction can be easily explained by increased fixation of deletional mutations in species with smaller N_e. High mutation rates in these species are simply the result of random genetic drift as explained by the drift-barrier theory, and they may have a limited role in driving genome reduction.Whereas our analysis based on the available data did not support natural selection as a universal mechanism driving genome reduction across prokaryotes (Fig. 2B, C), it does not mean that selection has no role in genome reduction of a particular taxon. In the case of thermophiles, proponents for selection acting to reduce genomes explained that genome size, due to its positive correlation with cell volume, may be an indirect target of selection which strongly favors smaller cell volume [35]. The underlying principle is that high temperature requires cells to increase the lipid content and change the lipid composition of the cell membranes, which consumes a large part of the cellular energy, and thus lower cell volume is selectively favored at high temperature [35]. Our calculations of a relatively small N_e in T. eurythermalis does not necessarily contradict with this selective argument, given that the fitness gained by decreasing cell volume and thus reducing genome size is large enough to overcome the power of random genetic drift. On the other hand, our data strongly indicate that neutral forces dictate the genome evolution of T. eurythermalis, and they are not negligible with regard to its genome reduction process. The significantly more deletion over insertion events (t test; 95 versus 37 events with p < 0.001 and 48 versus 20 events with p < 0.05 before and after removing the 14 genes enriched in mutations, respectively) and the significantly more nucleotides involved in deletions over insertions (t test; 433 versus 138 bp with p < 0.05 and 386 versus 121 bp with p < 0.001 before and after removing the 14 genes enriched in mutations, respectively) suggest that the deletion bias, combined with increased chance fixation of deletion mutants due to low N_e, is a potentially important neutral mechanism giving rise to the small genomes of T. eurythermalis (2.12 Mbp).The globally distributed deep-sea hydrothermal vents are microbe-driven ecosystems, with no known macroorganisms surviving at the vent fluids. Sample collections, microbial isolations, and laboratory propagations of mutation lines at hot and anoxic conditions are challenging. In the present study, we determined that T. eurythermalis, and perhaps Thermococcaceae in general, has a highly increased mutation rate and a highly decreased effective population size compared to all other known free-living prokaryotic lineages. While it remains to be tested whether this is a common feature among the vents’ microbes, the present study nevertheless opens a new avenue for investigating the hyperthemophile ecology and evolution in the deep sea.     

Loss of symbiont infectivity following thermal stress can be a factor limiting recovery from bleaching in cnidarians

Increases in seawater temperature can cause coral bleaching through loss of symbiotic algae (dinoflagellates of the family Symbiodiniaceae). Corals can recover from bleaching by recruiting algae into host cells from the residual symbiont population or from the external environment. However, the high coral mortality that often follows mass-bleaching events suggests that recovery is often limited in the wild. Here, we examine the effect of pre-exposure to heat stress on the capacity of symbiotic algae to infect cnidarian hosts using the Aiptasia (sea-anemone)-Symbiodiniaceae model system. We found that the symbiont strain Breviolum sp. CS-164 (ITS2 type B1), both free-living and in symbiosis, loses the capacity to infect the host following exposure to heat stress. This loss of infectivity is reversible, however, a longer exposure to heat stress increases the time taken for reversal. Under the same experimental conditions, the loss of infectivity was not observed in another strain Breviolum psygmophilum CCMP2459 (ITS2 type B2). Our results suggest that recovery from bleaching can be limited by the loss of symbiont infectivity following exposure to heat stress.Subject terms: Microbial ecology, Biodiversity

Cnidarians including reef-building corals harbor endosymbiotic dinoflagellates of the family Symbiodiniaceae, from which they derive the majority of their energy. Therefore, the breakdown of the symbiotic relationship, a process known as bleaching, can result in the host starving. However, bleaching is not always lethal because symbiont densities can recover [1, 2]. Recovery from bleaching is driven mainly by symbiotic algae that remain within the bleached corals (the residual population) dividing and spreading throughout the colony [3], and also possibly through the recruitment of free-living symbiotic algae from the external environment [4]. In the last few decades, coral cover has drastically decreased in many regions, due to frequent mass coral bleaching events caused by global warming [5], implying that recovery from bleaching is often limited by unknown factors. In the present study, we demonstrate that both free-living and residual symbiont cells lose their capacity to infect cnidarian host cells once they are exposed to high temperature stress, and present this mechanism as a limiting factor for the host’s recovery from bleaching.We first examined the effect of pre-exposure to high temperature on infectivity using aposymbiotic Exaiptasia pallida (or “Aiptasia”) polyps (Supplementary Fig. 1) and cultured strains of Breviolum sp. CS-164 (ITS2 type B1). Symbiotic algae and polyps were separately incubated at either 25 or 32 °C for 3 days. Following this initial treatment, polyps were inoculated with symbiotic algae at 25 °C for 3 days. Infectivity was then determined by counting the number of algae in the tentacles where algal colonization occurs quickly and individual symbiont cells are easily visualized [6]. When both symbiotic algae and hosts were pre-exposed to 25 °C, significant numbers of algae were seen in tentacles (Fig. 1a). In contrast, the number (Fig. 1a) and density (Fig. 1b) of algae in the tentacles were significantly lower when both the algae and the host were exposed simultaneously to 32 °C, and when the algae alone were exposed to 32 °C. Thus, symbiotic algae, but not host polyps, lose their capacity to form a symbiotic relationship once they are exposed to high temperature. Neither cell viability measured by Evans blue staining (Supplementary Figs. 2a, b and 3) nor cell density (Supplementary Fig. 2c) differed between CS-164 cells exposed to 25 and 32 °C, indicating that infectivity was not lost by the lethal damage to cells. We repeated this experiment with another strain, B. psygmophilum CCMP2459 (ITS2 type B2). In contrast to the results with CS-164, high temperature had no effect on infectivity of CCMP2459 (Supplementary Fig. 4a). These results demonstrate that symbiotic algae can lose their capacity to infect host cells following exposure to high temperature and that thermal sensitivity differs between these two algal strains (Fig. 1b and Supplementary Fig. 4a).Open in a separate windowFig. 1Loss of infectivity in Breviolum sp. CS-164 following exposure to elevated temperature.a Fluorescent photographs of Aiptasia polyps 3 days after culturing with symbiont cells in four different treatments (i) neither symbionts nor polyps exposed to high temperature (32 °C) for 3 days, (ii) only symbionts exposed to high temperature, (iii) only polyps exposed to high temperature, (iv) both symbionts and polyps exposed to high temperature. Red dots show chlorophyll fluorescence from algal symbionts. b The density of symbionts in tentacles was measured 3 days after culturing Aiptasia polyps (H) with symbiont cells (S) in four different treatments, as indicated below the panel and outlined in the text. c The density of symbionts was measured 3 days after culturing Aiptasia polyps with symbiotic algae in different treatments. In this experiment, symbiotic algae that had been expelled from Aiptasia polyps cultured at 25 or 32 °C for 3 days were used to infect Aiptasia. b, c Values are log₂ fold changes with respect to the samples without any temperature treatment. Each point represents an independent experiment. ns, not significant (with p > 0.05); **, p < 0.01.We next examined whether the loss of infectivity following pre-exposure to high temperature also occurs when symbiotic algae are in symbiosis with the host rather than free living (Fig. 1c and Supplementary Fig. 4b). We prepared symbiotic Aiptasia polyps with either CCMP2459 or CS-164 by separately inoculating them in aposymbiotic polyps, and then exposed each group to either 25 or 32 °C for 3 days. Algae expelled from the polyps during this treatment were collected and then used to inoculate aposymbiotic Aiptasia at 25 °C. In CS-164, after 3 days of inoculation, symbiont density in Aiptasia became lower with algae collected at 32 °C than 25 °C (Fig. 1c). However, in CCMP2459, there was no difference in the infectivity between algae collected at 25 and 32 °C (Supplementary Fig. 4b). Our results demonstrate that symbiont cells, both free-living and in symbiosis, can lose infectivity following exposure to high temperature and that thermal sensitivity differs between these two algal strains.We then tested whether or not the temperature-induced loss of infectivity was reversible (Fig. 2). Free-living CS-164 cells were pre-exposed to 25 or 32 °C for either 2 or 3 days after which they were allowed to recover for a maximum of 10 days at 25 °C. After these treatments, symbiotic algae were used to inoculate aposymbiotic Aiptasia polyps at 25 °C for 3 days. Cells with pre-exposure to 32 °C for 2 days had lower infectivity but regained the capacity to infect host cells after a 5-day recovery period (Fig. 2). However, after 3 days exposure, infectivity gradually recovered but remained lower than the controls even after 10 days (Fig. 2). Our results demonstrate that the loss of infectivity following temperature stress is reversible in algal cells but a longer exposure to heat stress increases the time taken to reverse the loss of infectivity.Open in a separate windowFig. 2Reversibility of the lost infectivity upon the exposure to elevated temperature in Breviolum sp. CS-164.The density of symbionts in tentacles was measured following treatments as described in the text and shown by the relative to control. Values are log₂ fold changes with respect to the control. The box and line represent the quartiles and median, respectively. Each point represents an independent experiment. *, p < 0.05.In the present study, infectivity was tested by introducing symbiotic algae directly into the host’s body cavity, suggesting that the loss of infectivity seen in our experiments is likely due to a failure of the host to take up the algal cells via phagocytosis or for symbiont cells to persist within host cells. Factors such as symbiont cell size [7] and the symbiont surface glycome [8, 9] are potentially important determinants of symbiont uptake and persistence, though we still know little about this topic (see review [10]). Furthermore, it is unknown how these various discriminatory factors are influenced by thermal stress.In coral larvae and juveniles, the initiation of symbiosis with algae is reduced at high temperatures, suggesting that global warming will complicate the relationship between host and symbiont [11–13]. However, the mechanism for this reduction in the rate of symbiosis establishment is not clear. Our results suggest that the loss of symbiont infectivity is one possible cause of this phenomenon.Recovery of symbiont densities following coral bleaching relies on a supply of symbiont cells either from within the host or from the external environment. Our results show that symbiotic algae, both free-living and symbiotic, lose the capacity to infect the host following exposure to high temperature stress (Fig. 1). This loss of infectivity is reversible but dependent on the duration of the thermal stress (Fig. 2). Thus, following coral bleaching events, especially those induced by thermal anomalies that can last for weeks, symbiont densities within the host are unlikely to recover in time to avoid the host starving due to physiological compromise of the symbionts, rather than the host. Nonetheless, given the differences in sensitivity between two strains tested (Fig. 1b and Supplementary Fig. 4a), if heat tolerant symbionts are available in the environment, this might provide a chance for recovery.     

Flow-through stable isotope probing (Flow-SIP) minimizes cross-feeding in complex microbial communities

Stable isotope probing (SIP) is a key tool for identifying the microorganisms catalyzing the turnover of specific substrates in the environment and to quantify their relative contributions to biogeochemical processes. However, SIP-based studies are subject to the uncertainties posed by cross-feeding, where microorganisms release isotopically labeled products, which are then used by other microorganisms, instead of incorporating the added tracer directly. Here, we introduce a SIP approach that has the potential to strongly reduce cross-feeding in complex microbial communities. In this approach, the microbial cells are exposed on a membrane filter to a continuous flow of medium containing isotopically labeled substrate. Thereby, metabolites and degradation products are constantly removed, preventing consumption of these secondary substrates. A nanoSIMS-based proof-of-concept experiment using nitrifiers in activated sludge and ¹³C-bicarbonate as an activity tracer showed that Flow-SIP significantly reduces cross-feeding and thus allows distinguishing primary consumers from other members of microbial food webs.Subject terms: Microbiology, Biological techniques, Metabolism, Biogeochemistry, Microbial ecology

Stable isotope probing (SIP) is widely applied to link specific microbial populations to metabolic processes in the environment and has greatly advanced our understanding of the role of microorganisms in biogeochemical cycling. SIP relies on tracing the incorporation of specific isotopically labeled substrates (e.g., ¹³C, ¹⁵N, ¹⁸O, ²H) into cellular biomarkers or bulk cellular biomass [e.g., 1–4]. SIP is considered a robust technique to identify microbial populations that assimilate a labeled substrate of interest in complex environmental communities. However, cross-feeding can occur when isotopically labeled metabolites are released from a primary consumer and then used by other microorganisms, which subsequently also become isotopically labeled. Likewise, when ¹³C-bicarbonate and unlabeled substrate are supplied to assess the activity of specific chemolithoautotrophs [e.g., 5–7], undesired ¹³C-incorporation can occur due to cross-feeding between chemolithoautotrophs whose activity depends on each other, for example in nitrifiers, where ammonia oxidizers provide nitrite oxidizers with their substrate, nitrite. The uncertainties associated with cross-feeding in SIP studies increase as the incubation time of microbial communities increases. While this phenomenon can be used to study microbial interactions and trophic networks [8–11], cross-feeding can lead to erroneous identification of organisms that are not directly responsible for the process of interest, but are rather connected to primary consumers via a microbial food web [2, 10, 12, 13].We developed an approach that significantly reduces the effect of cross-feeding in SIP studies. For this purpose, a thin layer of microbial cells is placed on a membrane filter, and isotopically labeled substrate is supplied at a fixed concentration by continuous flow, which constantly removes released metabolites and degradation products of primary substrate consumers. While previous SIP studies have employed a continuous flow of medium or substrate [e.g., 6, 14–16], in these studies, cross-feeding still occurred, as large amounts of biomass were placed in a 3D space, which allowed for the exchange of metabolites. Here, we present a proof-of-concept experiment with a nitrifying activated sludge microbial community, which converts ammonia to nitrite by the activity of ammonia-oxidizing bacteria (AOB), and subsequently oxidizes nitrite to nitrate by nitrite-oxidizing bacteria (NOB). In our experiments, the carbon source for both groups of autotrophic nitrifiers (the sludge contained no comammox bacteria [17, 18]) was isotopically labeled inorganic carbon (¹³C–NaHCO₃) and, as the sole electron donor, unlabeled ammonium was provided.In the flow-through approach, AOB, but not NOB, should be ¹³C-labeled because the substrate for NOB (nitrite), produced by AOB is continuously removed and thus the NOB should remain metabolically inactive (Fig. 1). In addition to a regular batch incubation, we included a control incubation, where the flow-through was recirculated to determine the impact of the experimental setup (continuous medium flow and retainment of biomass on a membrane filter) in Flow-SIP on the activity of the bacterial cells (in particular on the NOB as their autotrophic activity is used as a read out for cross-feeding in our experiments) in comparison to the batch experiment. Cross-feeding is expected to occur in both recirculated and batch control incubations, where nitrite is not removed and thus both AOB and NOB conserve energy to fix ¹³C–CO₂. After the experiments, fluorescence in situ hybridization (FISH) with rRNA-targeted oligonucleotide probes was used to identify AOB and NOB and combined with nanoscale secondary ion mass spectrometry (nanoSIMS) to quantify ¹³C-assimilation at the single-cell level for all setups.Open in a separate windowFig. 1Schematic representation of the experimental setup: (left) batch, (center) recirculated, and (right) flow-through incubation.In all incubations, the carbon source for both autotrophic nitrifiers (AOB, yellow and NOB, magenta) was isotopically labeled (i.e., CO₂ as ¹³C–NaHCO₃) and ammonia was provided as the only external energy source (as NH₄Cl). Cross-feeding is expected to occur in the batch and recirculated approaches, where NOB consume nitrite produced via ammonia oxidation by AOB and thus both AOB and NOB incorporate ¹³C–CO₂. In the flow-through approach, only AOB are expected to be ¹³C-labeled, as cross-feeding should be eliminated by the continuous removal of nitrite. Other, non-nitrifier cells are indicated in gray.For these experiments, activated sludge from a Danish municipal wastewater treatment plant was initially treated by sonication to disrupt large flocs. Cells were then either placed on a membrane filter for flow-through incubation and the recirculated control experiment, or incubated in a conventional batch experiment. All experiments were set up using the same amount of biomass, and the ratio of biomass to medium volume was the same in batch and recirculated control experiments. Incubations were done using mineral medium containing 250 µM NH₄Cl and 2 mM ¹³C–NaHCO₃ for 24 h. Medium flow was maintained at a rate of 26 ml h⁻¹. We did not select a higher flow rate in order to avoid excessive stress by the medium flow on the microbial cells and to minimize the required amounts of media containing isotopically labeled bicarbonate. Furthermore, modeling nitrite advection and diffusion at different flow rates showed that, for example, a tenfold higher flow rate would only marginally reduce nitrite concentrations surrounding the AOB colonies (Fig. S1). In contrast, in a purely diffusive system without continuous flow, our model showed that nitrite would accumulate to significantly higher concentrations around single AOB colonies (Fig. S2). For example, after 24 h, ~23 µM nitrite would accumulate at a distance of 0–100 µm (with no significant decrease over distance) around an AOB colony of 50 cells, which is 230- to 9200-fold higher (depending on the distance to the colony) than modeled nitrite concentrations at the flow rate used in our experiments. Most inorganic metabolites that can directly be taken up into cells behave similar as nitrite in a diffusive system, i.e., they have a similar diffusion coefficient. Larger molecules tend to have an even smaller diffusion coefficient, which would lead to slower diffusion, and thus even more efficient metabolite removal when a medium flow is applied.We monitored nitrification activity via concentration measurements of ammonium, nitrite and nitrate (Fig. S3) and conducted nanoSIMS analyses (Fig. 2) for two successive experiments using sludge collected from the same treatment plant on different days as replication of experimental results (referred to as E1 and E2). Additionally, we confirmed the reproducibility of the method with two further experiments, where nitrification activity was followed (Fig. S4). Details on the experimental setup are given in Fig. 1 and the Supplementary Text.Open in a separate windowFig. 2Single cell isotope probing of nitrifying activated sludge in batch, recirculated, and flow-through incubations.a–c Show representative FISH images of E2 (AOB in yellow; NOB in magenta; other cells counterstained by DAPI in gray) of batch, recirculated, and flow-through incubations, respectively, and d–f show the corresponding nanoSIMS images. Scale bar is 10 µm in all images. g–l Show ¹³C labeling of AOB, NOB, and other cells quantified by nanoSIMS at the single-cell level for E1 (g–i) and E2 (j–l) in batch, recirculated, and flow-through incubations, respectively. We used FISH probe sets targeting AOB (Nitrosomonas oligotropha cluster (Cl6a192), Nitrosomonas eutropha/europea/urea cluster (NEU)) and NOB (Nitrotoga (Ntoga122), Nitrospira Lineage 1 (Ntspa1431), and Nitrospira Lineage 2 (Ntspa1151)), respectively, for differential staining of the two nitrifier groups. In (g–l), dashed lines give the ¹³C natural abundance values of the filter surface. The number of cells analyzed per group is indicated below each boxplot. For each experiment, lower case letters indicate significant difference in ¹³C labeling between groups (AOB, NOB, other cells) within an incubation type and upper case letters indicate significant difference between incubation types for a given group (Kruskal–Wallis test followed by Dunn’s test; Statistics are given in Table S2). Boxplots depict the 25–75% quantile range, with the center line depicting the median (50% quantile) and whiskers encompass data points within 1.5× the interquartile range.In recirculated and batch control incubations, the consumption of ammonium, production of nitrite and nitrate (Fig. S3), and single cell ¹³C-incorporation (Fig. 2) indicated that both AOB and NOB were active. However, nitrification activity (i.e., nitrite and nitrate production) in recirculated incubations were reduced by 57% (E1) and 83% (E2) compared to batch incubations (Fig. S3). The reduced ammonia oxidation activity in the recirculated incubations compared to the batch incubations was also reflected by a 73–82% lower ¹³C-incorporation in AOB cells in the former incubations (Fig. 2, median AOB ¹³C-enrichment in recirculated setup was 3.7 and 3.9 ¹³C-atom%, in batch incubations 20.7 and 14.7 ¹³C-atom% for E1 and E2, respectively). AOB in the flow-through incubations also showed lower ¹³C-enrichment levels (8.2 and 8.5 atom% for E1 and E2, respectively) compared to batch incubations but higher enrichment than in the recirculated incubations. The lower enrichment of AOB in the recirculated compared to the flow-through incubations might be due to an accumulation of compounds leaching from the used tubing (PharMed^® Ismaprene, Table S1), which may negatively affect AOB. Indeed, nitrifiers have previously been reported to be sensitive to various organic compounds [19, 20]. Use of different rubber tubing or replacing rubber tubing by glass might alleviate these effects. AOB ¹³C-enrichment was highest in batch incubations, which could be due to both the lack of stress from the continuous medium flow and the observed reaggregation of the sonicated activated sludge into larger flocs—reminiscent of native activated sludge flocs.As expected, NOB were ¹³C-enriched in both the batch (13.3 and 4.9 atom% for E1 and E2, respectively) and recirculated incubations (7.2 and 4.7 atom% for E1 and E2, respectively). In contrast, as intended, the flow-through setup resulted in a substantial reduction in ¹³C-enrichment of NOB (2.0 atom% for both E1 and E2, respectively; with consistently low ¹³C-enrichment in all NOB cells measured). This demonstrates that Flow-SIP efficiently removed the secondary substrate nitrite released by the AOB primary substrate consumers, thereby strongly limiting cross-feeding between AOB and NOB. The low ¹³C-enrichment of NOB in the flow-through incubations was statistically not significantly different to the ¹³C-enrichment of non-nitrifier cells (Table S2). It is unlikely that this low background ¹³C-enrichment was due to ¹³C-bicarbonate adsorption, as all samples were treated with acid before nanoSIMS analysis. It is, however, possible that at least some of the observed ¹³C-enrichment in NOB and other bacteria is due to anaplerotic reactions leading to C-fixation by background cellular activity rather than substrate-induced autotrophic C-fixation [e.g., 21, 22]. Transfer of ¹³C-labeled metabolites from the autotrophic nitrifiers to non-nitrifier cells was negligible in all incubations, including batch and recirculated incubations (Fig. 2), which was likely due to the short incubation time (24 h). In contrast, other SIP studies using incubation times of several days reported significant C-isotope transfer from nitrifiers to non-nitrifiers [11, 23].Our results demonstrate that Flow-SIP is a promising approach to significantly reduce cross-feeding in complex microbial communities and can be even successfully applied to highly aggregated samples like activated sludge flocs when they are dispersed prior to the experiment. Flow-SIP and conventional SIP are complementary to each other in the analysis of such aggregated or biofilm communities. In such systems, Flow-SIP enables microbial ecologists studying microbial physiologies with drastically reduced cross-feeding, but destroys the spatial arrangement of cells, while conventional SIP retains the 3D architecture, but its results are strongly influenced by cross-feeding. We expect that Flow-SIP is ideally suited for oligotrophic fresh- or seawater samples, which predominantly harbor planktonic cells or small aggregates, and thus do not require any sonication pretreatment before incubation. Furthermore, for such samples, tracer can be directly added to sterile filtered water without the need for using artificial medium, as used for the presented proof-of-principle experiments.Flow-SIP may, after upscaling to label more biomass, also be used in combination with DNA-, RNA- or protein-SIP, which should in comparison to conventional SIP, where cross-feeding is not inhibited, allow microbial ecologists to more precisely identify both previously known and yet unknown primary consumers of a supplied substrate. For example, Flow-SIP with ¹³C-bicarbonate and unlabeled ammonium would allow distinguishing comammox organisms from canonical NOB, as comammox but not the canonical NOB would be active under these conditions together with the canonical ammonia oxidizers. In addition, Flow-SIP has the potential to study direct use of chemically unstable substrates by microorganisms, by distinguishing it from microbial consumption of their chemically formed decomposition products. For example, cyanate, which abiotically decays relatively fast to ammonium and carbon dioxide [24, 25], has previously been shown to serve as energy and nitrogen source for ammonia-oxidizing archaea [25, 26]. Using Flow-SIP, cyanate could be constantly supplied, thereby strongly reducing abiotic decay. At the same time, any abiotically formed ammonium (and ammonium produced by other organisms) would be constantly removed, which should allow identifying ammonia-oxidizing microorganisms that directly use cyanate as a substrate. Furthermore, the presented approach may be coupled to fluorescence-based activity markers, where a substrate of interest and bioorthogonal noncanonical amino acids are supplied and, subsequently, translationally active cells are visualized on an epifluorescence microscope (BONCAT) [27]. In conclusion, Flow-SIP expands the toolbox of microbial ecologists interested in structure–function analyses of microbial communities and will contribute to a more precise understanding of the ecophysiology of bacteria and archaea catalyzing key processes in their natural environments.     

Human-associated microbiota suppress invading bacteria even under disruption by antibiotics

In light of their adverse impacts on resident microbial communities, it is widely predicted that broad-spectrum antibiotics can promote the spread of resistance by releasing resistant strains from competition with other strains and species. We investigated the competitive suppression of a resistant strain of Escherichia coli inoculated into human-associated communities in the presence and absence of the broad and narrow spectrum antibiotics rifampicin and polymyxin B, respectively. We found strong evidence of community-level suppression of the resistant strain in the absence of antibiotics and, despite large changes in community composition and abundance following rifampicin exposure, suppression of the invading resistant strain was maintained in both antibiotic treatments. Instead, the strength of competitive suppression was more strongly associated with the source community (stool sample from individual human donor). This suggests microbiome composition strongly influences the competitive suppression of antibiotic-resistant strains, but at least some antibiotic-associated disruption can be tolerated before competitive release is observed. A deeper understanding of this association will aid the development of ecologically-aware strategies for managing antibiotic resistance.Subject terms: Microbial ecology, Community ecology, Antibiotics

The overuse of broad-spectrum antibiotics in clinical and agricultural settings is a key driver of the current antibiotic resistance crisis [1]. Research into antibiotic resistance has traditionally focused on the evolution of resistance in individual pathogens [2]. In the last decade, researchers have turned their attention to the collateral damage inflicted on commensal members of the microbiome, such as those belonging to the dense communities of the human gastrointestinal tract [3, 4]. Several studies have shown that antibiotics can leave gut communities vulnerable to colonisation by other pathogens [5–7], and, most recently, resistance evolution in invading strains can be facilitated by the absence of community suppression [8, 9]. Taken together, these two lines of enquiry appear to bear out conventional wisdom that relative to narrow-spectrum antibiotics or antibiotic-free conditions, broad spectrum antibiotics should increase the likelihood of communities being invaded by resistant strains [10, 11]. On the other hand, given evidence that community-level properties can sometimes be robust to changes in taxonomic composition [12], it is possible that some antibiotic-associated disruption can be tolerated before colonization resistance is affected. Despite the importance of these contrasting predictions, there have been few, if any, direct tests in human-associated microbiota.We investigated the effect of broad and narrow spectrum antibiotics on the strength of competitive suppression on a resistant variant (generated by in vitro selection for resistance mutations) of a focal strain (Escherichia coli K-12 MG1655) inoculated into gut microbiome communities collected from human faecal samples. The focal strain was jointly resistant to the broad-spectrum antibiotic rifampicin (targets Gram-positives and Gram-negatives via inhibition of the highly conserved bacterial RNA polymerase) and the narrow spectrum antibiotic polymyxin B (only targets Gram-negatives). The focal strain was inoculated alongside live or sterile slurry produced using a sample from one of three healthy human donors (described in [9]) into customized gut media without antibiotics or supplemented with 128 μg/ml rifampicin or 4 μg/ml polymyxin B (see Fig S1). Following 24 h incubation under anaerobic conditions, focal strain density and total biomass were measured via colony counting and flow cytometry, and community composition and diversity were analysed via 16S rRNA sequencing.In the absence of either antibiotic, focal strain density after 24 h was significantly lower in the presence of the three donor communities, indicative of strong competitive suppression (Fig. 1a). Surprisingly, we detected similarly strong competitive suppression in both the antibiotic treatments as we did in the antibiotic-free treatment. Instead, we found that focal strain performance was a stronger function of the specific donor community, irrespective of antibiotic treatment (Figs. 1b, and S2).Open in a separate windowFig. 1Effect of community, donor and antibiotic on focal strain abundance.a Violin plots showing the distribution of observed abundances of the focal strain in each antibiotic treatment. Blue denotes community free treatments; yellow denotes community treatment. Point shape denotes the individual human donor of live community or sterilized slurry: donor 1 = circles, donor 2 = squares, donor 3 = diamonds. b Treatment contrasts (posterior distributions of parameter estimates for a linear model with negative binomial errors) for focal strain abundance as a function of community (live vs sterile slurry), antibiotic (none, polymixin B or rifampicin), and donor (slurry prepared with samples from human donor 1, 2 or 3), and the interactions between community and antibiotic, and community and donor. Posterior parameter estimates in green have 95% credible intervals that do not overlap with 0 (i.e., there is less than 5% probability there is no effect of the variables/interactions captured by these coefficients). The reference level (vertical black line) = donor 1 in the no antibiotic treatment in the absence of the community (i.e., sterilized slurry).What makes these results particularly striking is that, consistent with previous studies [7, 10, 13], treatment with a broad-spectrum antibiotic was still associated with a marked shift in community composition (analysis of 16S amplicon data) (Fig. 2a). Based on OTU composition, all three donors in the rifampicin treatment cluster separately from the polymyxin B and antibiotic-free treatments, which cluster together (Fig. 2b). This divergence in composition appears to be largely driven by enrichment of both Enterobacteriaceae and Erysipelotrichaceae in the rifampicin treatment (Fig. 2a). In addition to strong shifts in composition, total bacterial abundance was significantly reduced in the rifampicin treatment (Figs. 2c and S3). Despite this, total richness and diversity (Shannon Index) after 24 h did not differ between the treatments (Fig. 2c). In contrast, diversity loss over time was more strongly associated with donor identity, with the donor community associated with the weakest competitive suppression (donor 3) also exhibiting the largest decline in richness and diversity across all treatments. This observation is consistent with previous work demonstrating that colonization resistance in the mouse gut is highly contingent on the complexity and composition of the resident microbiota [14].Open in a separate windowFig. 2Community response to antibiotic treatments.a Heatmap of relative abundance of the ten most abundant families of bacteria across treatments (derived from amplicon data). I = inoculum; AB free = Antibiotic free; Poly = polymyxin B; Rif = rifampicin. b NMDS ordination of family level composition in each treatment-donor combination. c Violin plots showing the abundance (top), species richness (middle) and diversity (Shannon Index) (bottom) distributions in each treatment. In b and c: circles = donor 1; squares = donor 2, diamonds = donor 3.A limitation of this study is that we only considered the effects of two antibiotics. Nevertheless, given the scale of community perturbation observed (Fig. 2), we can at least be sure our findings are not explained by a lack of antibiotic effects in our system. There must be some limit dictated by antibiotic concentration, combination, or duration of exposure, beyond which we would expect to observe stronger competitive release. Indeed, prior research has shown that antibiotics can greatly inhibit colonisation resistance [15, 16]. As such, characterizing where this limit lies (e.g., by investigating community-mediated suppression as a function of antibiotic concentration/duration) will be an important challenge for future work. Similarly, although we only considered a single focal strain, and other strains/species may have been more invasive (for example, those with fewer, different or less costly resistance mutations), key for our experiment was that the focal strain had a positive growth rate over the timescale of the experiment, despite exhibiting significant resistance costs in antibiotic-free assays (Fig. S1). This allowed us to test for sensitivity of competitive suppression to antibiotic treatment. We also note that in spite of a small boost in the focal strain’s performance in the presence of rifampicin independent of the community (a possible hormetic response [17] absent under aerobic growth in LB, Fig S1), we did not observe an increase in the magnitude of competitive release in the rifampicin treatment. Finally, the drop in diversity indicates, unsurprisingly, microcosms are a novel environment relative to the source environment. Despite this, key taxa in each community were stable over the course of the experiment, and previously over a longer timescale in the same set-up [9], demonstrating these conditions sustain diverse human-associated communities over relevant timescales.In conclusion, these results are consistent with prevailing wisdom that healthy gut communities can suppress invading strains and thereby reduce the likelihood of resistance emerging [8, 9, 18]. Nevertheless, the absence of a significant effect of broad, or even narrow, spectrum antibiotics on the degree of competitive suppression of our focal strain is much more surprising. Despite the limitations of scope discussed above, this shows that the functional diversity of gut communities may be more robust to disturbance by broad spectrum antibiotics than previously recognised. This is not to suggest that the use of broad-spectrum antibiotics does not drive marked changes in composition but rather that there is some degree of functional redundancy in diverse communities that facilitates the maintenance of competitive suppression [12, 19]. Notwithstanding the need to test how these findings translate to in vivo settings, this finding is relevant for optimizing personalised treatments that either account for disruption by antibiotics or that make microbiomes harder for pathogens to invade.     

A New Extracellular β-Galactosidase Producing Kluyveromyces sp. PCH397 from Yak Milk and Its Applications for Lactose Hydrolysis and Prebiotics Synthesis

β-Galactosidase is a crucial glycoside hydrolase enzyme with potential applications in the dairy, food, and pharmaceutical industries. The enzyme is produced in the intracellular environment by bacteria and yeast. The present study reports yeast Kluyveromyces sp. PCH397 isolated from yak milk, which has displayed extracellular β-galactosidase activity in cell-free supernatant through the growth phase. To investigate further, cell counting and methylene blue staining of culture collected at different growth stages were performed and suggested for possible autolysis or cell lysis, thereby releasing enzymes into the extracellular medium. The maximum enzyme production (9.94 ± 2.53U/ml) was achieved at 37 °C in a modified deMan, Rogosa, and Sharpe (MRS) medium supplemented with lactose (1.5%) as a carbon source. The enzyme showed activity at a wide temperature range (4–50 °C), maximum at 50 °C in neutral pH (7.0). In addition to the hydrolysis of lactose (5.0%), crude β-galactosidase also synthesized vital prebiotics (i.e., lactulose and galacto-oligosaccharides (GOS)). Additionally, β-fructofuranosidase (FFase) activity in the culture supernatant ensued the synthesis of a significant prebiotic, fructo-oligosaccharides (FOS). Hence, the unique features such as extracellular enzymes production, efficient lactose hydrolysis, and broad temperature functionality by yeast isolate PCH397 are of industrial relevance. In conclusion, the present study unrevealed for the first time, extracellular production of β-galactosidase from a new yeast source and its applications in milk lactose hydrolysis and synthesis of valuable prebiotics of industrial importance.Supplementary InformationThe online version contains supplementary material available at 10.1007/s12088-021-00955-1.Keyword: β-Galactosidase, Lactulose, Galacto-oligosaccharides, Fructo-oligosaccharides, Milk-microbes

β-Galactosidase (EC 3.2.1.23) hydrolyzes the glycosidic bond in β-galactosides and finds applications in the food industry [1, 2]. The trans-glycosylation property of β-galactosidase (β-gal) is widely used to produce various galactosylated products and prebiotics such as GOS and lactulose [3–7]. The β-gal enzyme is produced intracellularly by many bacteria and yeast, a major constraint for industrial production [1, 8]. Therefore, extracellular β-gal producing bacteria/yeast are of huge relevance. Hence, the present work revealed an efficient extracellular β-gal producing microbe from dairy products of the Indian Himalaya and evaluated its applications in lactose hydrolysis and prebiotics’ synthesis.In this study, twenty milk and four curd samples were collected from the Lahaul and Pangi valleys of Himachal Pradesh, India. The samples were plated on MRS and Elliker agar medium (Himedia, India) for 2–7 days at 28 °C and 37 °C until visible microbial growth. Morphologically distinct isolates were screened for β-gal activity using X-Gal and IPTG plate assay [6, 9]. The positive isolates were screened for β-gal production in liquid MRS medium. The β-gal activity was expressed as U/mg dcw (dry cell weight) for whole cells and U/ml for cell-free supernatant [10, 11]. Yeast isolate PCH397 showing the highest and extracellular enzymatic activity was selected. The culture and reaction conditions for maximum β-gal activity were optimized. FFase activity of whole cells and cell-free supernatant was estimated as described by Lincoln and More [12].The cell-free supernatant (β-gal) was employed for applications in lactose hydrolysis and prebiotic synthesis. The enzyme was incubated with lactose solution (5%, w/v) at 37 °C for lactose hydrolysis followed by thin layer chromatography (TLC) [13] analysis and quantification using the ImageJ program (http://rsbweb.nih.gov/ij/). Further, the cell-free supernatant was incubated with milk at 4 °C for milk lactose hydrolysis. Samples were withdrawn at different time intervals and analyzed for residual lactose concentration using ultra-high performance liquid chromatography-quadrupole-time of flight-ion mobility mass spectrometry (UHPLC-Q-TOF-IMS) [14]. Prebiotic production was carried out by mixing an equal volume of the enzyme with a sugar solution i.e., lactose (40%, w/v) for GOS, and lactose (20%, w/v) + fructose (20%, w/v) for lactulose and FOS production, respectively at 50 °C for 24 h [6]. Samples were analyzed by TLC for GOS, UHPLC-Q-TOF-IMS for FOS and lactulose synthesis.The study resulted in the isolation of 203 morphologically distinct microbes, 62 of which were tested positive for β-gal. Based on quantitative screening, eight isolates showing maximum β-gal activity were selected and examined for the intracellular and extracellular enzymatic activities (Table S1). Yeast isolate PCH397 exhibited maximum extracellular β-gal activity (9.94 ± 2.53 U/ml) along with FFase activity (0.59 ± 0.155) after 48 h of incubation. Isolate PCH397 was identified as Kluyveromyces marxianus by its morphological and molecular characterization (Fig. S1). Phylogenetic tree based on ITS DNA sequence showed similarity (99.63%) with Kluyveromyces marxianus CBS712. To the best of our knowledge, the genus Kluyveromyces has not been reported earlier for extracellular β-gal production. In the past, efforts were made to produce β-gal extracellularly through permeabilization or incorporation of signal peptide to β-gal gene in a fusion construct [15, 16]. The isolate PCH397 was selected due to its generally regarded as safe (GRAS) status and the novel feature of extracellular enzyme production.Highest β-gal activity in the extracellular environment was observed when PCH397 was grown in MRS medium supplemented with 1.5% (w/v) lactose as a substrate and incubated at 37 °C for 48 h (Fig. S2). PCH397 produced extracellular β-gal at lower lactose concentration (1.5%) as compared to various Kluyveromyces spp. [15] where 3% lactose has been used in the growth medium for intracellular β-gal production. Further, whether the extracellular enzyme activity is due to the secretion or cell lysis, the CFU count and cell viability were checked by the methylene blue test. The decreased cell count in the late stationary phase for live cells (Fig. S3) and increased number of methylene blue stained cells indicated cell death (Fig S4). These results suggested that cell lysis in the late stationary phase leads to the secretion of enzymes in extracellular medium. The extracellular production of enzyme would lead to a lower production costs of the enzyme.Cell-free supernatant showed the highest β-gal activity at pH 7.0 in 10 mM sodium phosphate buffer at 50 °C in 5 min (Fig S2). The β-gal enzyme from the current finding holds promise in the sweet whey and milk lactose hydrolysis [1] due to its neutral pH optima. Also, β-gal, which is functional at high temperatures, is used in the synthesis of oligosaccharides [1, 3]. High temperature increases the reaction rate as well as lactose solubility, thus, facilitating transgalactosylation reactions [17]. The β-gal activity (9 U/ml) in cell-free supernatant of PCH397 completely hydrolyzed 5.0% of lactose within 8 h at 37 °C (Fig. 1a, S5a). In a recent study, 5.0% lactose was also hydrolyzed by purified β-gal (5 U/ml) of Paenibacillus barengoltzii CAU904 within 8 h at 40 °C [13]. Under refrigerated conditions (4 °C), the cell free supernatant hydrolyzed ~ 50% milk lactose within 36 h and ~ 80% in 72 h (Fig. 1b, S5b). Since β-gal of PCH397 is active at 4 °C, the enzyme could be utilized to hydrolyze lactose in dairy products under refrigerated conditions. Lactose-free milk products or low-lactose milk products are important dietary constituents for lactose intolerant individuals and deliver essential nutrients to combat nutritional deficiencies [18]. Even with commercially purified enzymes, 100% milk-lactose hydrolysis could not be achieved at a low temperature [19]. However, the crude enzyme from the present investigation can efficiently hydrolyze milk lactose at ambient and refrigerated conditions, reducing the cost associated with enzyme purification. Additionally, the source of enzyme is Kluyveromyces sp. which has GRAS status, therefore, can be used in food applications.Open in a separate windowFig. 1Lactose hydrolysis by crude β-gal of PCH397. a Relative quantification of the hydrolysed products from lactose (5%, w/v) at 37 °C for 24 h. b Relative decrease in lactose concentration (%) at refrigerated conditions obtained by UHPLC-QTOF-IMSFurther, the enzyme was evaluated for its ability to catalyze transgalactosylation reactions at 50 °C. The crude enzyme was incubated with different substrate mixture viz. lactose and fructose. After 8 h of incubation, 50% of lactose was hydrolyzed into glucose, galactose, and GOS (Fig. S6a). Maximum GOS production was achieved after 12 h (Fig. 2a). The purified β-gal from Paenibacillus barengoltzii synthesized GOS from 350 g/L of lactose within 4 h [13]. Though GOS synthesis was faster in comparison to the current study, it is to be noted that we used a crude enzyme mixture instead of a purified enzyme. The crude enzyme has also shown FFase activity (Table S1), and was used for the synthesis of FOS from lactose and fructose mixture. UHPLC-Q-TOF-IMS analysis confirmed the formation of FOS (Fig. 2b). Multiple peaks were observed in the sample containing lactulose, one of which was identical with the peak of lactulose standard (Fig. 2c) as confirmed by HPAEC-PAD (Fig. S6b). The lactulose formation was maximum at 20 h of incubation (Fig. S6c).Open in a separate windowFig. 2Hydrolysis and transgalactosylation of lactose by crude enzyme from PCH397 having β-gal and FFase activity. a Relative quantification of the hydrolyzed and transgalactosylated products. UHPLC-QTOF-IMS detection of prebiotics b FOS and c lactulose with their respective standardIt is the first report of simultaneous co-synthesis of multiple prebiotics i.e., GOS, FOS, and lactulose using a yeast strain. Similar reports for GOS and FOS synthesis have been attempted by enzymatic means from fungal sources in the past [6]. The synthesis of multiple prebiotics is very advantageous. Numerous studies have shown that blended consumption of multiple prebiotics including GOS and FOS has many health benefits [20–24]. The combination of GOS, FOS, and lactulose can be of considerable importance for their prebiotic applications. In conclusion, our findings revealed a yeast source for the cost-effective production of β-galactosidase and a strategy for co-synthesis of valuable prebiotics, which is not reported in the past. The utilization of a yeast source with GRAS status for lactose hydrolysis and co-synthesis of prebiotics promises various health benefits and commercial relevance.     

Endoscopically visible steam pop during high-energy laser pulmonary vein ablation

A 58-year-old woman with atrial fibrillation underwent laser balloon ablation at our centre. During 12 W ablation in the left superior pulmonary vein, a sudden steam pop was witnessed with displacement of the balloon catheter. Visualisation of the pulmonary vein antrum showed a red discolouration at the last ablation site.The endoscopically assisted laser balloon ablation system (EAS) is a relatively novel technique that is being used to perform pulmonary vein isolation (PVI) in the treatment of atrial fibrillation [1]. The EAS consists of a flexible, compliant balloon for sustained wall contact and a power adjustable laser beam for ablation independent of tissue contact.A 58-year-old woman underwent PVI with the EAS due to drug-refractory, symptomatic and paroxysmal atrial fibrillation. During 12 W ablation at the antrum of the left superior pulmonary vein (LSPV), a sudden steam pop was witnessed, with displacement of the EAS catheter (Fig. 1). Visualisation of the LSPV antrum showed a red discolouration, most likely a haematoma in the antral wall of the LSPV, at the last ablation site. A successful PVI was performed; the red discolouration was still present after 1 h. The patient did not develop symptoms related to the steam pop and echocardiography did not reveal any abnormalities.Open in a separate windowFig. 1Witnessed steam pop during endoscopically assisted ablation. Panel a displays the fifth ablation site in the left superior pulmonary vein (LSPV) where the steam pop occurred. The white ring of exposed tissue is a sign of optimal catheter-wall contact. Panel b displays the LSPV antrum directly after the steam pop. Note the red discolouration which was not present in panel a Steam pops are caused by overheating of myocardial tissue, exceeding 100 ℃, and are preceded by a shift in impedance levels, which cannot be measured with the EAS. Higher energy settings and higher contact force are known to increase the risk of steam pops. Steam pops can lead to tissue disruption and cardiac perforation [2]. However, steam pops appear to be a rare complication with reduced EAS energy settings, which we mostly used in 50 EAS patients, in whom no steam pops were observed [3].     

Finding smORFs: getting closer

Millions of small open reading frames exist in eukaryotes. We do not know how many, or which are translated, but bioinformatics is getting us closer to the answer.See related Research article: http://www.genomebiology.com/2015/16/1/179DNA sequences encoding small open reading frames (smORFs) of fewer than 100 amino acids (aa) exist in each eukaryotic genome in numbers several orders of magnitude higher than the corresponding annotated protein-coding genes (Fig. 1). Due to difficulties with bioinformatic detection and experimental analysis, along with their sheer numbers, smORFs have been ignored for a long time by mainstream genomics. Thanks to recent advances in bioinformatic and experimental techniques, however, smORFs are receiving increasing attention. Extensive use of RNA-Seq has shown that thousands of smORFs are transcribed, in many cases, in putative noncoding RNAs, and high-throughput experimental techniques have detected translation of a few hundred of these. However, the possibility remains that many more smORFs are functional, but yet uncharacterized. Bioinformatic methods followed by targeted experimental verification are needed to improve the identification of putative functional smORFs. A new paper in Genome Biology [1] provides a significant step towards such a solution.Open in a separate windowFig. 1The number of small open reading frames (smORFs) in eukaryotic genomes (shown in log scale) greatly exceeds that of annotated protein-coding genes, and reaches 265,000 in yeast [4], 556,000 in the fruit fly Drosophila [2], and 40,700,000 in mouse [3]. Note that the current number of corroborated functional smORFs is but a small fraction of these (see text and [1] for details). The number of annotated protein-coding genes was obtained from the Saccharomyces Genome Database (yeast; http://www.yeastgenome.org/), FlyBase (fruit fly; http://flybase.org/), and Ensembl (mouse; http://www.ensembl.org/index.html) (accessed 12 August 2015)     

Ubiquitination of S4-RNase by S-LOCUS F-BOX LIKE2 Contributes to Self-Compatibility of Sweet Cherry ‘Lapins’

Recent studies have shown that loss of pollen-S function in S₄′ pollen from sweet cherry (Prunus avium) is associated with a mutation in an S haplotype-specific F-box4 (SFB4) gene. However, how this mutation leads to self-compatibility is unclear. Here, we examined this mechanism by analyzing several self-compatible sweet cherry varieties. We determined that mutated SFB4 (SFB4ʹ) in S4′ pollen (pollen harboring the SFB4ʹ gene) is approximately 6 kD shorter than wild-type SFB4 due to a premature termination caused by a four-nucleotide deletion. SFB4′ did not interact with S-RNase. However, a protein in S4′ pollen ubiquitinated S-RNase, resulting in its degradation via the 26S proteasome pathway, indicating that factors in S4′ pollen other than SFB4 participate in S-RNase recognition and degradation. To identify these factors, we used S₄-RNase as a bait to screen S4′ pollen proteins. Our screen identified the protein encoded by S₄-SLFL2, a low-polymorphic gene that is closely linked to the S-locus. Further investigations indicate that SLFL2 ubiquitinates S-RNase, leading to its degradation. Subcellular localization analysis showed that SFB4 is primarily localized to the pollen tube tip, whereas SLFL2 is not. When S₄-SLFL2 expression was suppressed by antisense oligonucleotide treatment in wild-type pollen tubes, pollen still had the capacity to ubiquitinate S-RNase; however, this ubiquitin-labeled S-RNase was not degraded via the 26S proteasome pathway, suggesting that SFB4 does not participate in the degradation of S-RNase. When SFB4 loses its function, S₄-SLFL2 might mediate the ubiquitination and degradation of S-RNase, which is consistent with the self-compatibility of S4′ pollen.

In sweet cherry (Prunus avium), self-incompatibility is mainly controlled by the S-locus, which is located at the end of chromosome 6 (Akagi et al., 2016; Shirasawa et al., 2017). Although the vast majority of sweet cherry varieties show self-incompatibility, some self-compatible varieties have been identified, most of which resulted from the use of x-ray mutagenesis and continuous cross-breeding (Ushijima et al., 2004; Sonneveld et al., 2005). At present, naturally occurring self-compatible varieties are rare (Marchese et al., 2007; Wünsch et al., 2010; Ono et al., 2018). X-ray-induced mutations that have given rise to self-compatibility include a 4-bp deletion (TTAT) in the gene encoding an SFB4′ (S-locus F-box 4′) protein, located in the S-locus and regarded as the dominant pollen factor in self-incompatibility. This mutation is present in the first identified self-compatible sweet cherry variety, ‘Stellar’, as well as in a series of its self-compatible descendants, including ‘Lapins’, ‘Yanyang’, and ‘Sweet heart’ (Lapins, 1971; Ushijima et al., 2004). Deletion of SFB3 and a large fragment insertion in SFB5 have also been identified in other self-compatible sweet cherry varieties (Sonneveld et al., 2005; Marchese et al., 2007). Additionally, a mutation not linked to the S-locus (linked instead to the M-locus) could also cause self-compatibility in sweet cherry and closely related species such as apricot (Prunus armeniaca; Wünsch et al., 2010; Zuriaga et al., 2013; Muñoz-Sanz et al., 2017; Ono et al., 2018). Much of the self-compatibility in Prunus species seems to be closely linked to mutation of SFB in the S-locus (Zhu et al., 2004; Muñoz-Espinoza et al., 2017); however, the mechanism of how this mutation of SFB causes self-compatibility is unknown.The gene composition of the S-locus in sweet cherry differs from that of other gametophytic self-incompatible species, such as apple (Malus domestica), pear (Pyrus spp.), and petunia (Petunia spp.). In sweet cherry, in addition to a single S-RNase gene, the S-locus contains one SFB gene, which has a high level of allelic polymorphism, and three SLFL (S-locus F-box-like) genes with low levels of, or no, allelic polymorphism (Ushijima et al., 2004; Matsumoto et al., 2008). By contrast, the apple, pear, and petunia S-locus usually contains one S-RNase and 16 to 20 F-box genes (Kakui et al., 2011; Okada et al., 2011, 2013; Minamikawa et al., 2014; Williams et al., 2014a; Yuan et al., 2014; Kubo et al., 2015; Pratas et al., 2018). The F-box gene, named SFBB (S-locus F-box brother) in apple and pear and SLF (S-locus F-box) in petunia, exhibits higher sequence similarity with SLFL than with SFB from sweet cherry (Matsumoto et al., 2008; Tao and Iezzoni, 2010). The protein encoded by SLF in the petunia S-locus is thought to be part of an SCF (Skp, Cullin, F-box)-containing complex that recognizes nonself S-RNase and degrades it through the ubiquitin pathway (Kubo et al., 2010; Zhao et al., 2010; Chen et al., 2012; Entani et al., 2014; Li et al., 2014, 2016, 2017; Sun et al., 2018). In sweet cherry, a number of reports have described the expression and protein interactions of SFB, SLFL, Skp1, and Cullin (Ushijima et al., 2004; Matsumoto et al., 2012); however, only a few reports have examined the relationship between SFB/SLFL and S-RNase (Matsumoto and Tao, 2016, 2019), and none has investigated whether the SFB/SLFL proteins participate in the ubiquitin labeling of S-RNase.Although the function of SFB4 and SLFL in self-compatibility is unknown, the observation that S4′ pollen tubes grow in sweet cherry pistils that harbor the same S alleles led us to speculate that S4′ pollen might inhibit the toxicity of self S-RNase. In petunia, the results of several studies have suggested that pollen tubes inhibit self S-RNase when an SLF gene from another S-locus haplotype is expressed (Sijacic et al., 2004; Kubo et al., 2010; Williams et al., 2014b; Sun et al., 2018). For example, when SLF2 from the S₇ haplotype is heterologously expressed in pollen harboring the S₉ or S₁₁ haplotype, the S₉ or S₁₁ pollen acquire the capacity to inhibit self S-RNase and break down self-incompatibility (Kubo et al., 2010). The SLF2 protein in petunia has been proposed to ubiquitinate S₉-RNase and S₁₁-RNase and lead to its degradation through the 26S proteasome pathway (Entani et al., 2014). If SFB/SLFL in sweet cherry have a similar function, the S4′ pollen would not be expected to inhibit self S₄-RNase, prompting the suggestion that the functions of SFB/SLFL in sweet cherry and SLF in petunia vary (Tao and Iezzoni, 2010; Matsumoto et al., 2012).In this study, we used sweet cherry to investigate how S4′ pollen inhibits S-RNase and causes self-compatibility, focusing on the question of whether the SFB/SLFL protein can ubiquitinate S-RNase, resulting in its degradation.     

Holliday Junction Resolvase MOC1 Maintains Plastid and Mitochondrial Genome Integrity in Algae and Bryophytes

When DNA double-strand breaks occur, four-stranded DNA structures called Holliday junctions (HJs) form during homologous recombination. Because HJs connect homologous DNA by a covalent link, resolution of HJ is crucial to terminate homologous recombination and segregate the pair of DNA molecules faithfully. We recently identified Monokaryotic Chloroplast1 (MOC1) as a plastid DNA HJ resolvase in algae and plants. Although Cruciform cutting endonuclease1 (CCE1) was identified as a mitochondrial DNA HJ resolvase in yeasts, homologs or other mitochondrial HJ resolvases have not been identified in other eukaryotes. Here, we demonstrate that MOC1 depletion in the green alga Chlamydomonas reinhardtii and the moss Physcomitrella patens induced ectopic recombination between short dispersed repeats in ptDNA. In addition, MOC1 depletion disorganized thylakoid membranes in plastids. In some land plant lineages, such as the moss P. patens, a liverwort and a fern, MOC1 dually targeted to plastids and mitochondria. Moreover, mitochondrial targeting of MOC1 was also predicted in charophyte algae and some land plant species. Besides causing instability of plastid DNA, MOC1 depletion in P. patens induced short dispersed repeat-mediated ectopic recombination in mitochondrial DNA and disorganized cristae in mitochondria. Similar phenotypes in plastids and mitochondria were previously observed in mutants of plastid-targeted (RECA2) and mitochondrion-targeted (RECA1) recombinases, respectively. These results suggest that MOC1 functions in the double-strand break repair in which a recombinase generates HJs and MOC1 resolves HJs in mitochondria of some lineages of algae and plants as well as in plastids in algae and plants.

Mitochondria and plastids were established in eukaryotic cells by endosymbiotic events of α-proteobacterial and cyanobacterial ancestors, respectively (Gray, 1992; Archibald, 2015). Reminiscent of their bacterial ancestors, both organelles possess their own genomes and proliferate by division of preexisting ones (Martin and Kowallik, 1999). Plastid DNA (ptDNA) and mitochondrial DNA (mtDNA) encode some components of the photosynthetic apparatus and respiratory chain, respectively (Allen, 2003). Thus, to maintain the functions of plastids and mitochondria, ptDNA and mtDNA must faithfully replicate and segregate during proliferation of the organelles.The mitochondrion and plastid possess multiple copies of DNA, which are organized with proteins into nucleoids (Kuroiwa, 1991; Pfalz and Pfannschmidt, 2013). Nucleoids, which can be visualized as dot-like or globular structures in mitochondria and plastids when stained with DNA-specific fluorochromes such as 4′, 6-diamidino-2-phenylindole (DAPI) or SYBR Green I, are ubiquitously observed in diverse lineages of algae and plants (Kuroiwa, 1991; Sato, 2001; Kobayashi et al., 2016). The morphology of nucleoids dynamically changes according to cell cycle progression and development (Powikrowska et al., 2014). For example, in the unicellular green alga Chlamydomonas reinhardtii, ∼80 copies of ptDNA are packaged into 5 to 8 globular nucleoids in a single cup-shaped plastid during the gap 1 (G1) phase (Armbrust, 1998). Prior to plastid division, during the synthesis (S) and mitosis (M) phases, plastid nucleoids change into filamentous structures and are scattered throughout the plastids. Then the nucleoids are inherited by two daughter plastids stochastically (Ehara et al., 1990; Kamimura et al., 2018). A similar morphological change of plastid nucleoids is also observed in the plant Arabidopsis (Arabidopsis thaliana; Terasawa and Sato, 2005), and thus, the mechanism of nucleoid segregation is apparently conserved in algae and plants. However, the molecular mechanisms underlying organelle DNA segregation and changes in nucleoid morphology have remained largely unknown.In a previous study using the green alga C. reinhardtii, mutants defective in nucleoid segregation were isolated (Misumi et al., 1999). One of the mutants possessed only a single enlarged nucleoid in a plastid (Fig. 1A), which was inherited by daughter plastids unevenly (Misumi et al., 1999). Later, the mutation responsible for this phenotype was identified in a previously uncharacterized gene, Monokaryotic Chloroplast1 (MOC1), which is conserved in eukaryotic algae and plants (Kobayashi et al., 2017). MOC1 exhibited endonuclease activity in vitro, where it specifically cleaved Holliday junctions (HJs), four-stranded DNA structures formed during homologous recombination (HR; Kobayashi et al., 2017). MOC1 symmetrically introduced nicks between consecutive cytosines (C↓C, where the arrow indicates the cleavage point) at the core of HJs (Kobayashi et al., 2017). Because HJs provide a covalent link between recombining DNA molecules and must be removed prior to genome segregation (Liu and West, 2004; West, 2009), it was suggested that HR affects the nucleoid structure and MOC1 segregates plastid nucleoids by cleaving HJs between ptDNA molecules prior to plastid division (Kobayashi et al., 2017).Open in a separate windowFigure 1.MOC1 depletion destabilizes ptDNA by increasing ectopic recombination between SDRs in C. reinhardtii. A, Differential interference contrast (DIC) and fluorescent images of SYBR Green I-stained wild-type (WT) and MOC1 KO cells. Green is SYBR Green I-stained DNA and magenta is plastid chlorophyll fluorescence (Chl). The arrow indicates a plastic nucleoid. N, Nucleus. Scale bars = 5 μm. B, qPCR comparing relative copy number of ptDNA between the wild type and CrMOC1 KO. The values of plastid rpl2, psbB, chlN, and psbD loci were normalized with that of the nuclear CBLP locus. For CrMOC1 KO, two independent clones (clones 1 and 2) were analyzed. The normalized value of the wild type was defined as 1.0. The error bar represents the mean ± sd (n = 3). Asterisks indicate significant difference by Student’s t test (**P < 0.01). C, Positions of SDRs CD5, CD5′, CI12, and CD15 in C. reinhardtii ptDNA (Odahara et al., 2016). Large inverted repeats (IRa and IRb) are shown by bold lines. SDRs are indicated by triangles. D to F, qPCR comparing relative copy numbers of ectopic recombinants of CD5 (D), CI12 (E), and CD15 (F). The recombinants were quantified with primers designed as shown in Supplemental Figure S2. Each value was normalized with that of the plastid psbB locus. The error bar represents the mean ± sd (n = 3). Asterisks indicate significant difference by Student’s t test (ns, P ≥ 0.05, *P < 0.05, and **P < 0.01).In general, HR follows either of two main pathways, the double-strand break repair (DSBR) or the synthesis-dependent strand annealing (SDSA) pathway (Supplemental Fig. S1; Holliday, 1964; Szostak et al., 1983; Pâques and Haber, 1999). The two pathways are similar in the initial step. After a double-strand break occurs, 5′ ends of the break are cut back to create 3′ overhangs of single-strand DNA (ssDNA). Recombinases bind the 3′ overhangs of ssDNA and search through vast quantities of DNA sequence to align and pair ssDNA with a homologous double-strand DNA template, facilitating the formation of a d-loop (Dunderdale et al., 1991; Murayama et al., 2008). In the DSBR pathway, the end of the invading 3′ strand is extended by a DNA polymerase and converted into a HJ. The other 3′ overhang strand also forms a HJ with the homologous DNA. After that, there are two pathways to convert the double HJs into recombination products (Sung and Klein, 2006). One pathway is mediated by HJ resolvases, which cleave HJs and produce either crossover or noncrossover products (Szostak et al., 1983). Various HJ resolvases have been found in archaea, bacteria, and eukaryotes (West, 2009). The eukaryotic nucleus also possesses another pathway, which is driven by the BTR complex consisting of the Bloom syndrome helicase, topoisomerase IIIα (TOP3A), and recombination-deficient Q-mediated genome instability subcomplex proteins (Wu and Hickson, 2003). The BTR complex does not cleave but dissolves the double HJs (Wu and Hickson, 2003). During the dissolution, the two HJ branches migrate toward each other until they form a hemicatenated intermediate, which is decatenated by TOP3A. Therefore, the dissolution pathway never produces crossover products (Supplemental Fig. S1). In contrast to DSBR, in the SDSA pathway, the invading 3′ ssDNA is utilized as a primer and extended along the recipient DNA duplex by a DNA polymerase without forming HJs. The newly synthesized strand dissociates from the template DNA and anneals with the other 3′ overhang strand through complementary base pairing. After the strands anneal, the remaining single-stranded gaps are filled by a DNA polymerase (Supplemental Fig. S1; Pâques and Haber, 1999).In nuclear, mitochondrial, and plastid genomes, numerous short dispersed repeats (SDRs) exist (Ottaviani et al., 2014). Thanks to the accuracy of recombinases in finding the homologous sequences despite the existence of myriad SDRs, the genomic sequence of the broken DNA is restored precisely via HR (McEntee et al., 1979; Shibata et al., 1979; Qi et al., 2015). However, in the absence of recombinases, SDRs anneal through complementary base pairing and produce recombinants between SDRs (Supplemental Fig. S1). Intramolecular recombination between SDRs results in deletion or inversion of the flanking region when they are oriented as direct or inverted repeats, respectively (Supplemental Fig. S2). This pathway is known as microhomology-mediated end joining (MMEJ; Supplemental Fig. S1; McVey and Lee, 2008).Recombinase genes, including Rad51 (eukaryotic type), radA (archaeal type), and recA (bacterial type), are believed to have evolved from a common ancestral gene (Lin et al., 2006). Two phylogenetically distinct RECA proteins are encoded in the nuclear genome of land plants, of which one is targeted to plastids (ptRECA) and the other to mitochondria (mtRECA; Lin et al., 2006). ptRECA and mtRECA are most closely related to cyanobacterial and proteobacterial counterparts, respectively, suggesting endosymbiotic origins of these proteins (Lin et al., 2006). Two observations should be made regarding the functions of these two plant recombinases: (1) Suppression or loss of function of ptRECA causes ectopic recombination between SDRs at a high frequency and destabilizes the plastid genome in the green alga C. reinhardtii and the moss Physcomitrella patens (Odahara et al., 2015a, 2016), suggesting that ptRECA maintains integrity of the plastid genome by promoting HR and thus suppressing MMEJ. Because MOC1 possesses HJ-resolving activity in vitro and is required for segregation of ptDNA in vivo (Kobayashi et al., 2017), RECA-mediated HR is likely accomplished with MOC1 through the DSBR pathway in plastids. However, how MOC1 functions in vivo has not been investigated. (2) Like ptRECA, mtRECA suppresses ectopic recombination between SDRs in the mitochondrial genome in P. patens and the seed plant Arabidopsis (Shedge et al., 2007; Odahara et al., 2009; Miller-Messmer et al., 2012). These results suggest that HJs are formed in plant mitochondria. However, except for CCE1, which is specific to yeasts (Saccharomyces cerevisiae and Schizosaccharomyces pombe; Kleff et al., 1992; Oram et al., 1998), mitochondrial HJ resolvases have not been identified in eukaryotes. Thus, it remains unclear whether HJs are formed and, if they are formed, how they are removed in mitochondria in plants.Regarding issues 1 and 2 described above, in this study, we show first that MOC1 suppresses ectopic recombination between SDRs in C. reinhardtii plastids, as does ptRECA, suggesting that MOC1 is involved in DSBR in the plastid. Next, we show that MOC1 dually targets plastids and mitochondria in the moss P. patens and maintains the integrity of ptDNA and mtDNA via suppression of ectopic recombination in both of these organelles. Putative dual-targeted transit peptides are also predicted in MOC1s of charophyte algae, a liverwort, a fern, and some seed plants, and we show that some of them are targeted to both plastids and mitochondria. Thus, it is suggested that HJs are formed during HR and removed by MOC1 in both plastids and mitochondria in some algae and plants.