PREreview of A fungal transcription factorBOT6facilitates the transition of a beneficial root fungus into an adapted anthracnose pathogen

Summary: 

This article by Ujimatsu et al. characterizes the cluster residing transcription factor (TF) CtBOT6 from the root-associated fungus Colletotricum tofieldiae (Ct). The presence of a putative abscisic acid (ABA) and botrydial biosynthesis gene cluster (ABA-BOT) known to produce intermediate metabolites of botrydial is highly expressed in the pathogenic strain C3, but not the beneficial strain C4. The authors hypothesize that the Zn₂Cys₆ TF CtBOT6 regulates this ABA-BOT cluster and possible plays a role in regulating virulence. 

Using overexpression (OX) lines of CtBOT6 in the C4 strain, the authors performed various metabolomic, phenotypic, transcriptomic assays to investigate the role of this TF. Their data revealed that CtBOT6 regulates many genes both within and outside of the cluster. Despite slower growth on rich media, the OX lines exhibit enhanced growth and appressoria production on roots and leaves thus transitioning this beneficial strain into a pathogenic strain. This came at a cost for the plant, with reduced shoot fresh weight and activation of defence genes. 

This study provides novel insights into the TF-mediates regulatory mechanisms underlying the transition from mutualist to pathogen. This research significantly enhances our understanding of the growth-defence/virulence trade-off that both plants and microbes must balance. Future work to unravel the CtBOT6 regulon during root or leaf colonization would be an exciting avenue for exploration. 

General comments: 

The manuscript is well written. Overall, the finding that fungal lifestyle (beneficial vs pathogenic) can be so dramatically altered by manipulating one transcription factor is very interesting. In particular, novel colonisation of Arabidopsis leaves and formation of appressoria on roots by the OX#1 strain via activation of ‘hidden virulence’ was striking. 

One general criticism – the connection between secondary metabolism changes and virulence in different fungal strains is key for the main conclusion of the paper, however, supporting evidence is a little weak. Metabolites discussed in Figure S1 and 2 (shown to be regulated by CtBOT6) were not referred to much in later figures or discussion. Although, we appreciate that it may be difficult to demonstrate direct links between specific metabolite levels and virulence, given the number of genes that CtBOT6 potentially regulates. 

In general, evidence for ‘enhanced virulence’ of CtBOT6 overexpression lines could be strengthened. In Figures 5 and 7, a reduction in shoot fresh weight is used as a proxy for virulence. However, growth inhibition could indicate activation of defence signalling, not necessarily enhanced virulence of a pathogen. In support of this alternative hypothesis, hyperactivation of defence marker genes by OX#1 vs Ct3 was reported in Figure 6. 

Section- and figure-specific comments: 

Intro: 

• Although discussed in more detail elsewhere in the literature, more information on the function of botrydial and mechanisms of non-host resistance could be beneficial for readers’ understanding. 

Figure 1: 

• A scale for schematic in panel (A) would be useful. 

• General layout of panels could be improved for clarity. Reading across the page, rather than down, would be preferable. 

• Which stats test was conducted? There is no mention in the figure legend. 

• Inclusion of stats for every comparison in panel (E) would be preferable. Were any stats conducted for the CtACTIN data? 

• In panel (D), there is significantly reduced expression of CtABA4 in OX lines, but this is not referred to in the text, which just says expression ‘is not induced’. This is true, but does CtBOT6 act as a transcriptional repressor in this case? 

Figure S1 and S2: 

• Would be nice to include a y-axis to show all plots are on the same scale. This will enhance our ability to discern any increase in metabolite production. 

• There is a typo in the text referring to this figure: ‘and pathogenic C. incanum (Ci) WT strains wherein the ABA-BOT has’ This makes the main message of the figure unclear. 

• If rice media is just a rich media for fungal growth, this could be justified more explicitly? 

• It would be nice if the names of chemical intermediates were given in the figure legend, or the figure itself. 

• If ABA could not be detected when the fungus was grown on rice media, were any other types of media tested? It would be good to include a positive control for this figure that is not just a standard. 

Figure S3: 

• Some more comments on the significance and structure of caryophyllene would be helpful for readers’ understanding – is this similar to ABA or hypothesised to function in virulence in any way? 

Figure 3: 

• All images of fungal colonies should ideally be the same size to make direct comparison of colony size easier for the reader. 

Figure 4: 

• The title of the figure mentions virulence-related genes, but these are not referred to in the figure itself. Either change the figure/results header title or include a figure to demonstrate this. 

• In panel (A), there are lots of down-regulated genes in OX#1 vs Ct4. As mentioned in comments for Figure 1, could BOT6 also act as a repressor of transcription? GO enrichment analysis for these down-regulated genes may also be interesting. 

• 500 bp seems quite short to use as a promoter region to search for motifs. Is it possible that additional information could be missed upstream? 

• Are the motifs identified in panel (B) similar to any known promoter motifs in fungi, or unique to this paper? 

• Are these motifs upstream of the virulence-associated genes mentioned in the text e.g. CAZymes and effector genes or the genes found within the ABA-BOT cluster? 

• Some clarity on the dot size scale for GO enrichment analysis would be useful – what is N^o Test Seqs? 

• To accompany panels (D) and (E), it would be nice to include an example of a gene that isn’t upregulated by CtBOT6. 

• Some technically inaccurate choice of wording in the text describing this figure - the assays reported in panels (D) and (E) do not assess ‘binding ability’ of CtBOT6 to promoter regions, only transcriptional activity.  

Figure 5: 

• In panel (A) the Ct3-treated root is longer than that the mock-treated root. One would expect a longer root when treated with the Ct4 growth-promoting strain, but not for the pathogenic Ct3 strain? Was this phenotype consistent or quantified at any point (the Ct3-treated root also looks longer in Figure 7B)? If not, the images could be replaced for clarity. 

• Some arrows or a zoomed-in image would be useful to highlight the appressoria on Ct3-infected roots in panel (D). 

• To accompany panel (E), would be nice to see examples of leaves treated with other OX lines, and perhaps a positive control (Botrytis pathogen?) and negative control (wild-type Ct3?) could be included. 

• Quantification of virulence phenotypes could bring more support to their arguments e.g. number/density of appressoria could be measured, as this is a hallmark of virulence. 

• Introduction of the abi1-1C in panel F feels a bit abrupt. No reference is included for this mutant in the results section but is found in the methods section. Some context leading up to this may be nice. 

• Please write out TD and WGA in the figure legend. 

Figure S5: 

• Effect size is an odd choice of analysis? Comparison to mock via ratios and statistical tests would be useful as was done in their previous publication (Hiruma et al. 2023, Figure 3g). 

• The impact of abi1C-1C mutation is not very obvious here. Data points for mock and OX#1 look very similar in Col-0 and abi1C-1C backgrounds. 

Figure S6: 

• Direct measurement of ABA would be clearer than quantifying MAPKKK18 marker gene expression. The authors were able to quantify this in a previous publication (Hiruma et al. 2023) 

• Statistics for all comparisons would be helpful such as a 1-way ANOVA. In fact, it may be more biologically relevant to compare to Mock for this figure. 

Figure 6: 

• A change in wording is suggested when referring this figure. According to the figure, FRK1 expression is significantly different in Col-0 and abi1C-1C backgrounds for OX#1. But in the text, it is stated that ‘hyperactivation of defence responses observed here are unaffected in abi1C-1C’. 

• Again, stats for comparisons would be useful here, especially comparisons to mock. 

• One would expect induction of ethylene defence markers in Ct3 vs mock as stated in the text ‘The expression levels of these genes are significantly higher in Ct3 compared with Ct4’, but this doesn’t seem to be the case for ETR2. Please revise this statement in the text. 

Figure S7: 

• Similar to Figure S5, effect size is an odd choice of analysis? Comparison to mock via ratios and statistical tests would be useful. 

• The effect of ein3 eil1 mutation is not very obvious. Data points for mock and OX#1 look very similar in Col-0 and ein3 eil1 backgrounds. 

• Couldn’t see a reference for the ein3 eil1 mutant anywhere in the paper. 

Figure S8: 

• Hard to tell if expression of GENE_00011871 is higher in late stages of infection, especially with no stats tests conducted here. 

• The error bars in this figure are difficult to see. 

Figure 7: 

• Are the statistics correct in panel (C)? Significant differences are reported, although some distributions have a lot of overlap e.g those labelled f, e and d. 

• Correlations between shoot fresh weight and CtBOT6 expression are reported in the text, but this is not represented in the figure. Please include a figure showing this with a Pearson’s correlation coefficient and p value. 

• It is a little confusing that the phos promoter was reported to be active in the late stages of infection, but this trend if difficult to see in panel (D) as these three figures are on different scales. It would benefit the reader to use the same scales for these three figures.   

• It would be good to include the  CtACTIN expression as well for panel (D) as done in Figure 1E to ensure that changes in CtBOT6 expression are not a result of changes in fungal biomass.  

• As mentioned in comments on Figure 5, quantification of appressoria formation could also be useful as a proxy for virulence. 

• It is possible that this figure could swap places with Figure 6, as Figure 5 is where the link between CtBOT6 expression and virulence is discussed. Currently, Figure 7 incrementally expands upon the results of Figure 5. Figure 6, however, addresses something different – how the plant responds to the strains with enhanced virulence.  

Discussion: 

• No major problems here, although discussion could be expanded to include more ‘big picture’ topics e.g. the growth/virulence trade-off for fungi. 

Competing Interests 

The authors declare that they have no competing interests. 

Competing interests

The authors declare that they have no competing interests.