Plaques, a zone of clearing on the bacterial lawn, were construed as evidence for the presence of phage. Overall, we found very little evidence for the presence of either lytic or lysogenic phages that attacked any of our bacterial isolates. We did find some evidence for lysogens: 20% of bacterial isolates showed evidence for the presence of lysogenic phages in the B only lines. There was evidence for lysogeny in only 7% of isolates in the BP lines where both types of phages were passaged. With regard to lytic phages in the original phage fraction, we also did not find evidence for phage predation of any bacterial isolate using this starting inoculum. The results are reported in Table 2.Overall, disruption of co-passaging of bacteria and phage on leaves over time and between plants was found to have an impact on both the composition and diversity of the resultant epiphytic bacterial communities. The bacterial communities resulting from passaging bacteria with their ancestral phage appeared to be the most dissimilar as compared to the other passaging treatments. This microbiome also had lower alpha diversity than that of both only bacteria passaged only and bacteria passaged along with any potentially evolving phage . BPa lines also have lower beta diversity than all other treatments, and Pseudomonaceae dominated the communities. All of this taken together suggests that the original phage present in the inoculum is capable of having the largest impact on the community, drying and curing buds even to a bacterial community that may have changed in composition by passaging on plants.
The most parsimonious explanation for this finding is that the phages present in the initial inoculum have low persistence in the phyllosphere, at least in a growth chamber, and as such, there were both more, and perhaps more diverse phage present in the initial inoculum than what remained after attrition on the leaf surface during passaging in the growth chamber. This is, however, not fully supported by our phage-isolation attempts, as we were unable to isolate phages from the original phage-fraction on any bacterial isolates. However, this may be explained by the apparent decay of phages during refrigeration as observed by others, and it does not preclude the possibility that there were lytic phages present at the time of the passaging experiments . In treatments in which lytic phages were passaged for three weeks, it is likely that the phage fraction contained very little, if any, active phage particles by the end of the experiment. This is supported by the fact that we were unable to recover any phage isolates from the final time point of the experiment, except for one isolate from plants exposed to a BPa microbiome. Poor persistence of phages in the phyllosphere is a finding supported also supported by the work of others. Phages, in general, are found in very low incidence on the surface of leaves compared to that in endophytic compartments. This may be due to phage’s sensitivity to UV on the surface of leaves, or may be due to low replication of bacteria in the phyllosphere, which is a nutrient-deplete environment that may limit bacterial growth, and hence lytic phage replication. The strong effect of the ancestral phage fraction on the bacterial community is consistent, in some regards, to the findings described in Chapter 4.
There, we found that after one week, plants that received bacteria and phage together had lower beta diversity than plants receiving only bacteria . Plants receiving passaged bacteria and ancestral phage have significantly lower beta diversity than both B and BP treatments. In both cases, it may be that the impact the original phage fraction had on the bacterial community shaped it in a way that made all the microbiomes similar to one another. Interestingly, alpha diversity was also the lowest in this treatment, yet there are no differences in alpha diversity between the B and BP treatments. The alpha diversity finding may be explained by the taxonomy of the bacteria that was most impacted by the phage fraction. Here, Pseudomonaceae was in highest relative abundance in the BPa treatment- the treatment with the lowest alpha diversity. In previous work, the Bacteria-only treatment had a significantly higher relative abundance of Pseudomonaceae as well. Again, this treatment had the lowest alpha diversity. Thus, it is possible that the phage fraction has an initial impact on a dominant bacterial family in the phyllosphere, and this lethality subsequently has a ripple effect on alpha diversity of the rest of the community. Future work using rationally designed synthetic communities of bacteria and phages could address with hypothesis with more clarity. One of the primary goals of this work was to disentangle the effects of lytic and lysogenic phages on bacterial communities in the phyllosphere. Through our experimental design, we attempted to include a variety of lytic versus lysogenic phage challenges to the bacteria. Visually, it appears that the bacterial communities in which only lysogenic phages were passaged are different from those in which lytic phages were passaged. Furthermore, 20% of bacterial isolates showed evidence for the presence of lysogenic phages in the B only lines in which only lysogenic phages were passaged.
In contrast, there was evidence for lysogeny in only 7% of isolates in the BP lines where both types of phages were passaged. However, these findings are not statistically significant, and thus no conclusions can be drawn. This work highlights the importance of time-scale when studying the effects of phages on the phyllosphere bacterial community. Predictions about alpha diversity based on results after only a one week experiment are not entirely consistent with our findings from the study of communities that were passaged between plants for three weeks. Specifically, we did not find an increase in alpha diversity in the lines in which both bacteria and phage were passaged together compared to bacteria-only lines. This may indicate that temperate phages are able to mediate long-term bacterial diversity as well as lytic phage, or it may be unrelated to the presence of phages and reflect other microbial dynamics occurring in the community. Future work that involves more rigorous identification of lysogenic phages, such as bacterial genome sequencing, may help address the question of their importance in maintaining diversity. We also found that the ancestral phage fraction has the strongest impact on both composition and diversity of the bacterial communities – probably because it was more abundant. Finally, our lack of ability to culture phages after passaging on plants suggests strongly that lytic phage particles do not persist well in the phyllosphere of plants grown in the growth chamber This supports work by others indicating that the feasibility of using phages as bio-control agents in agriculture may be largely dependent on their ability to persist in the phyllosphere. Overall, these findings are an important extension of previous work , and they underscore many of the unanswered questions that remain regarding the abundance, persistence, growing tray and importance of bacteriophages in the phyllosphere.A diverse field inoculum was generated using field-grown tomato plants. Above ground tomato plant material was collected from two fields from the UC Davis Student Organic Farm in June 2018. The material was stored on ice for transportation to the lab. One hundred grams of plant material was submerged in 1.5L of 10mM MgCl2 and sonicated for 5 minutes, vortexed for 30 seconds, and sonicated again. This was repeated with an additional 500 grams of plant material, 100 grams at a time. The leaf wash from above was filtered using 8 µm filter paper to remove large pieces of plant debris. The flow-through containing all microbes was then filtered using .22µm filter units. Microbes collected on the filters, which should be most bacteria, were sonicated off the filter paper into sterile buffer for 10 minutes. To concentrate the phage fraction of the microbiome, .22 µm flow-through was then concentrated using 100Kda MWCO Millipore filter units. Both the bacterial and phage fractions were split into 8 aliquots to account for the 3 weeks of inoculation and the need for ancestral phage and ancestral bacteria for weeks 2 and 3. Bacterial fractions were stored at -80°C in 1:1 KB glycerol, and phage fractions were stored in the dark at 4°C. On each day of inoculation, bacterial aliquots were re-suspended in 3mL of MgCl2 without the addition of any other fractions for the “B only” treatment. A bacterial aliquot was combined with a phage fraction aliquot for “B and P”. For the first week, three treatment groups received B and P. Inoculum was spray inoculated onto each plant individually. After 1 week, entire plants were harvested individually. Bacterial and phage fractions were recovered from the plants as described previously . The bacterial and phage fractions were re-combined and inoculated onto the plants. For evolved bacteria and ancestral phage, the bacterial fraction from the end of the passage was combined with an ancestral phage aliquot from frozen storage. For the ancestral bacteria and evolved phage treatment, the phage fraction from the end of the passage was combined with an aliquot of the ancestral bacterial treatment from frozen storage.
The experiment was continued for three passages in total, each consisting of one week.My thesis work began by demonstrating that vertically transmitted bacteria on the surface of tomato seeds are capable of protecting seedlings against a common bacterial pathogen. Vertical transmission is a well-studied process in other systems such as termites and aphids [192]. However in plants, it has been primarily limited to the transmission of pathogens and endophytes. Both are important fields of study- especially for the prevention of plant diseases. However, I felt that there was a significant lack of knowledge as to the functional importance of vertically transmitted commensals or mutualists in plants. Intuitively, vertical transmission of a microbiome or symbiont would allow for maintenance of key members of the microbial community across generations. Beneficial microbes would have primary access to both spatial niches and nutrients provided by seedlings. Interestingly, plants have a differential onset of resistance to pathogens throughout their life-stages, something described as age-related resistance or developmental resistance. However, much of the work on ARR investigates exposure and resistance to specific pathogens throughout the developmental stage of the plant and does not address if there is a crucial window of exposure to commensals, and whether these commensals ae contributing to ARR, as observed in other systems. To my knowledge, there are no studies to date that test the importance of timing of commensal microbial exposure on microbiome establishment or immune function in plants, although there is a wealth of literature on establishing biological control agents. For example, would a seedling exposed to beneficial microbes mount as strong of a response as an older plant? And would exposure of otherwise sterile adults result in the same successional dynamics of microbiome establishment as often observed in seedlings? Given that we know that resistance to pathogens can change throughout the life cycle of a plant, research focused on age-related tolerance and recruitment of commensals and plant-growth promoting bacteria has large implications in seed treatment and agricultural practices. One fundamental question that I was unable to answer is: are the types of bacteria transmitted on seeds merely a reflection of the parental plant from which the progeny was generated? If so, do transmitted microbes vary based on adult-plant microbiomes? If so, which anatomical portion of the plant is most influential in shaping the seed microbiome? In this workwe found that some seed microbiomes were more protective than others. This suggests that they may have differed in their composition, but we never confirmed that this was the case. Furthermore in my work on adult plants, I found that plant host genotype influences the phyllosphere microbiome. Are these differences between microbiomes heritable through vertical transmission? In order to address these outstanding questions, I would conduct a common garden experiment wherein multiple host genotypes would be planted in replicated field sites. Upon fruiting, I would not only collect tomato fruits, but I would also collect leaf, flower, and soil from each site. I would collect seeds from fruits using the same approach as described in Chapter 2. I would carry out Gyrase B amplicon sequencing in order to describe the bacterial communities of the seeds in addition to the adult plants from which they came.