Temporal transcriptional response of Candida glabrata during macrophage infection reveals a multifaceted transcriptional regulator CgXbp1 important for macrophage response and drug resistance
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
The paper by Rai and colleagues examines the transcriptional response of Candida glabrata, a common human fungal pathogen, during interaction with macrophages. They use RNA PolII profiling to identify not just the total transcripts but instead focus on the actively transcribing genes. By examining the profile over time, they identify particular transcripts that are enriched at each time point, building a hierarchical model for how a transcription factor, CgXbp1, may regulate part of this response. While the authors have generated a large and potentially impactful dataset, along with several interesting observations, it is important to be cautious as the direct targets of CgXbp1 were characterized under one particular condition and the transcriptional analyses were obtained in another condition, one shown to be highly dynamic as during macrophage infection.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors.)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
Abstract
Candida glabrata can thrive inside macrophages and tolerate high levels of azole antifungals. These innate abilities render infections by this human pathogen a clinical challenge. How C. glabrata reacts inside macrophages and what is the molecular basis of its drug tolerance are not well understood. Here, we mapped genome-wide RNA polymerase II (RNAPII) occupancy in C. glabrata to delineate its transcriptional responses during macrophage infection in high temporal resolution. RNAPII profiles revealed dynamic C. glabrata responses to macrophage with genes of specialized pathways activated chronologically at different times of infection. We identified an uncharacterized transcription factor (CgXbp1) important for the chronological macrophage response, survival in macrophages, and virulence. Genome-wide mapping of CgXbp1 direct targets further revealed its multi-faceted functions, regulating not only virulence-related genes but also genes associated with drug resistance. Finally, we showed that CgXbp1 indeed also affects azole resistance. Overall, this work presents a powerful approach for examining host-pathogen interaction and uncovers a novel transcription factor important for C. glabrata ’s survival in macrophages and drug tolerance.
Article activity feed
-
Author response:
Reviewer #3 (Public Review):
The paper by Rai and colleagues examines the transcriptional response of Candida glabrata, a common human fungal pathogen, during interaction with macrophages. They use RNA PolII profiling to identify not just the total transcripts but instead focus on the actively transcribing genes. By examining the profile over time, they identify particular transcripts that are enriched at each timepoint, and build a hierarchical model for how a transcription factor, Xbp1, may regulate this response. Due to technical difficulties in identifying direct targets of Xbp1 during infection, the authors then turn to the targets of Xbp1 during cellular quiescence.
The authors have generated a large and potentially impactful dataset, examining the responses of C. glabrata during an important host-pathogen …
Author response:
Reviewer #3 (Public Review):
The paper by Rai and colleagues examines the transcriptional response of Candida glabrata, a common human fungal pathogen, during interaction with macrophages. They use RNA PolII profiling to identify not just the total transcripts but instead focus on the actively transcribing genes. By examining the profile over time, they identify particular transcripts that are enriched at each timepoint, and build a hierarchical model for how a transcription factor, Xbp1, may regulate this response. Due to technical difficulties in identifying direct targets of Xbp1 during infection, the authors then turn to the targets of Xbp1 during cellular quiescence.
The authors have generated a large and potentially impactful dataset, examining the responses of C. glabrata during an important host-pathogen interface. However, the conclusions that the authors make are not well supported by the data. The ChIP-seq is interesting, but the authors make conclusions about the biological processes that are differentially regulated without testing them experimentally. Because Candida glabrata has a significant percent of the genome without GO term annotation, the GO term enrichment analysis is less useful than in a model organism. To support these claims, the authors should test the specific phenotypes, and validate that the transcriptional signature is observed at the protein level.
Additionally, the authors should also include images of the infections, along with measurements of phagocytosis, to show that the time points are the appropriate. At 30 minutes, are C. glabrata actually internalized or just associated? This may explain the difference in adherence genes at the early timepoint. For example, in Lines 123-132, the authors could measure the timing of ROS production by macrophages to determine when these attacks are deployed, instead of speculating based on the increased transcription of DNA damage response genes. Potentially, other factors could be influencing the expression of these proteins. At the late stage of infection, the authors should measure whether the C. glabrata cells are proliferating, or if they have escaped the macrophage, as other fungi can during infection. This may explain some of the increase in transcription of genes related to proliferation.
An additional limitation to the interpretation of the data is that the authors should put their work in the context of the existing literature on C. albicans temporal adaptation to macrophages, including recent work from Munoz (doi: 10.1038/s41467-019-09599-8), Tucey (doi: 10.1016/j.cmet.2018.03.019), and Tierney (doi: 10.3389/fmicb.2012.00085), among others.
When comparing the transcriptional profile between WT and xbp1 mutant, it is not clear whether the authors compared the strains under non-stress conditions. The authors should include an analysis of the wild-type to xbp1 mutants in the absence of macrophage stress, as the authors claims of precocious transcription may be a function of overall decreased transcriptional repression, even in the absence of the macrophage stress. The different cut-offs used to call peaks in the two strain backgrounds is also somewhat concerning-it is not clear to me whether that will obscure the transcriptional signature of each of the strains. Additionally, the authors go on to show that the xbp1 mutant has a significant proliferation defect in macrophages, so potentially this could confound the PolII binding sites if the cells are dying.
In the section on hierarchical analysis of transcription factors, at least one epistasis experiment should have been performed to validate the functional interaction between Xbp1 and a particular transcription factor. If the authors propose a specific motif, they should test this experimentally through EMSA assays to fully test that the motif is functional.
The jump from macrophages to quiescent culture is also not well justified. If the transcriptional program is so dynamic during a timecourse of macrophage infection, it is hard to translate the findings from a quiescent culture to this host environment.
Overall, there is a strong beginning and the focus on active transcription in the macrophage is an exciting approach. However, the conclusions need additional experimental evidence.
We thank this reviewer’s critical analysis of our manuscript and the comments.
We fully agree that the jump from macrophages to quiescent culture is also not well justified. We have successfully performed CgXbp1 ChIP-seq during macrophage infection and have rewritten the manuscript according to the new results. With the CgXbp1 ChIP-seq data during macrophage infection added, we have removed the data related to quiescence to focus the paper on the macrophage response. Because of this, we have also removed the DNA binding motif analysis from this work and will report the findings in a separate manuscript comparing CgXbp1 bindings between macrophage response and quiescence.
As mentioned above, the RNAPII ChIP-seq time course experiment compared RNAP occupancies at different times during infection to the first infection time point. We did not calculate relative to the data in the absence of stress (e.g. before infection), because Xbp1 was expressed at a low level and induced by stresses. Hence its role under no stress conditions is expected to be less than inside macrophages. In addition, up-regulation of its target genes depends on the presence of their transcriptional activators under the experimental conditions, which is going to be very different in normal growth media (RPMI or YPD; i.e. before infection) versus inside macrophages. Hence, comparing to normal growth media would not show the real CgXbp1 effects and/or the CgXbp1 effect might be different. In fact, this can be seen from the new RNAseq analysis of wildtype and Cgxbp1∆ C. glabrata cells in the presence and absence of fluconazole (which are added to the revised manuscript to study CgXbp1’s role on fluconazole resistance). The result shows that CgXbp1 (which was expressed at a low level) has a very small effect on global expression and the up-regulated genes are mainly related to transmembrane transport. More importantly, the effect of the Cgxbp1∆ mutant on TCA cycle and amino acid biosynthesis genes’ expression during macrophage infection is not observed when the mutant is grown under normal growth conditions (YPD without fluconazole). Therefore, the results show that CgXbp1 has condition-specific effects on global gene expression, which is also dependent on the transcriptional activators present in the cell. The result of the new RNAseq analysis of wildtype and Cgxbp1∆ C. glabrata cells in the absence of fluconazole is described in lines 329-339 as follows: “On the other hand, 135 genes were differentially expressed in the Cgxbp1∆ mutant during normal exponential growth (i.e. no fluconazole treatment) (Figure 6c) with up-regulated genes highly enriched with the “transmembrane transport” function and down- regulated genes associated with different metabolic processes (e.g. carbohydrate, glycogen and trehalose) (e.g. carbon metabolism, nucleotide metabolism, and transmembrane transport, etc.) (Supplementary Table 12). Interesting, the TCA cycle and amino acid biosynthesis genes, whose expressions were accelerated in the Cgxbp1∆ mutant during macrophage (Figure 3C, 3D), were not affected by the loss of CgXbp1 function under normal growth conditions (i.e. in YPD media without fluconazole) (Supplementary Figure 11, Supplementary Table 11), suggesting that the overall (direct and indirect) effects of CgXbp1 are condition-specific.”
For the comment about RNAPII bindings affected by dying cells, our observation of reduced proliferation does not mean that the cells were dying, because we did observe increase in cell numbers over time (i.e. the cells were proliferating) but the rate of proliferation was slower in the Cgxbp1∆ mutant comparing to wildtype. Presumably, the reduced proliferation and/or growth within macrophages is due to poorer adaptation in and compromised response to macrophages.
We have also discussed our findings in the context of the suggested (and other) literatures in various parts of the Discussion.
Reviewer #4 (Public Review):
Macrophages are the first line of defense against invading pathogens. C. glabrata must interact with these cells as do all pathogens seeking to establish an infection. Here, a ChIP-seq approach is used to measure levels of RNA polymerase II levels across Cg genes in a macrophage infection assay. Differential gene expression is analyzed with increasing time of infection. These differentially expressed genes are compared at the promoter level to identify potential transcription factors that may be involved in their regulation. A factor called CgXbp1 on the basis of its similar with the S. cerevisiae Xbp1 protein is characterized. ChIP-seq is done on CgXbp1 using in vitro grown cells and a potential binding site identified. Evidence is provided that CgXbp1 affects virulence in a Galleria system and that this factor might impact azole resistance.
As the authors point out, candidiasis associated with C. glabrata has dramatically increased in the recent past. Understanding the unique aspects of this Candida species would be a great value in trying to unravel the basis of the increasing fungal disease caused by C. glabrata. The use of ChIP-seq analysis to assess the time-dependent association of RNA polymerase II with Cg genes is a nice approach. Identification of CgXbp1 as a potential participant in the control of this gene expression program is also interesting. Unfortunately, this work suffers by comparison to a significant amount of previous effort that renders the progress detailed here incremental at best.
I agree that their ChIP-seq time course of RNA polymerase II distribution across the Cg genome is both elegant and an improvement on previous microarray experiments. However, these microarray experiments were carried out 14 years ago and while the current work is certainly at higher resolution, little more can be gleaned from the current work. The authors argue that standard transcriptional analysis is compromised by transcript stability effects. I would suggest that, while no approach is without issues, quite a bit has been learned from approaches like RNA-seq and there are recent developments to this technique that allow for a focus on newly synthesized mRNA (thiouridine labeling).
The CgXbp1 characterization relies heavily on work from S. cerevisiae. This is disappointing as conservation of functional links between C. glabrata and S. cerevisiae is not always predictable.
The effects caused by loss of CgXBP1 on virulence (Figure 4) may be statistically significant but are modest. No comparison is shown for another gene that has already been accepted to have a role in virulence to allow determination of the biological importance of this effect.
The phenotypic effects of the loss of XBP1 on azole resistance look rather odd (Figure 6). The appearance of fluconazole resistant colonies in the xbp1 null strain occurs at a very low frequency and seems to resemble the appearance of rho0 cells in the population. The vast majority of xbp1 null cells do not exhibit increased growth compared to wild-type in the presence of fluconazole.
Irrespective of the precise explanation, more analysis should be performed to confirm that CgXbp1 is negatively regulating the genes suggested in Figure 6A to be responsible for the increased fluconazole resistance.
Additionally, the entire analysis of CgXbp1 is based on ChIP-seq performed using cells grown under very different conditions that the RNA polymerase II study. Evidence should be provided that the presumptive CgXbp1 target genes actually impact the expression profiles established earlier.
We thank this reviewer’s critical analysis of our manuscript. We have done the following to address the comments. As a result, the manuscript is significantly improved.
• The ChIP-seq data of Xbp1 in macrophage has been successfully generated and the result is now presented in Figure 2C-2F, and lines 182-227 of the revised manuscript. With the addition, we have removed the ChIPseq data related to quiescent from the revised manuscript and re-written the manuscript focusing on the role of Xbp1 in macrophage.
• We agree that the conservation of functional links between C. glabrata and S. cerevisiae is not always predictable. That’s the reason why we did not solely rely on the S. cerevisiae network for inferring Xbp1’s functions, and had undertaken several different ways (e.g. ChIP-seq of Xbp1 and characterization of the Cgxbp1∆ mutant) to delineate its functions.
• We also agree that the virulence effect is modest, but it is, nevertheless, an effect that may contribute to the overall virulence of C. glabrata. Since virulence is a pleiotropic trait involving many genes and every gene affects different aspects of the complex process, we feel that it is not fair to penalize a given gene based on its (weaker) effect relative to another gene. Therefore, we respectfully disagree that another gene should be included for benchmarking the effect.
• We have measured C. glabrata cell numbers in a time course experiment. The result (presented in Figure 4A) showed that there was an increase in cell number at the end of the macrophage infection time course experiment (e.g. 8 hr). We have highlighted this information on lines 278-283.
• Additional analysis of the fluconazole resistance phenotype of the Cgxbp1∆ mutant has been added, including standard MIC assays. The results are presented in Figure 5C-5E.
• As suggested and to understand the role of CgXbp1 on fluconazole resistance, we have now carried out RNAseq analysis of WT and the Cgxbp1∆ mutant in the presence and absence of fluconazole. The genes differentially controlled in the Cgxbp1∆ mutant have been identified and a proposed model on how CgXbp1 affects fluconazole resistance is added to Figure 7 in the revised manuscript.
-
Evaluation Summary:
The paper by Rai and colleagues examines the transcriptional response of Candida glabrata, a common human fungal pathogen, during interaction with macrophages. They use RNA PolII profiling to identify not just the total transcripts but instead focus on the actively transcribing genes. By examining the profile over time, they identify particular transcripts that are enriched at each time point, building a hierarchical model for how a transcription factor, CgXbp1, may regulate part of this response. While the authors have generated a large and potentially impactful dataset, along with several interesting observations, it is important to be cautious as the direct targets of CgXbp1 were characterized under one particular condition and the transcriptional analyses were obtained in another condition, one shown to be highly …
Evaluation Summary:
The paper by Rai and colleagues examines the transcriptional response of Candida glabrata, a common human fungal pathogen, during interaction with macrophages. They use RNA PolII profiling to identify not just the total transcripts but instead focus on the actively transcribing genes. By examining the profile over time, they identify particular transcripts that are enriched at each time point, building a hierarchical model for how a transcription factor, CgXbp1, may regulate part of this response. While the authors have generated a large and potentially impactful dataset, along with several interesting observations, it is important to be cautious as the direct targets of CgXbp1 were characterized under one particular condition and the transcriptional analyses were obtained in another condition, one shown to be highly dynamic as during macrophage infection.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors.)
-
Reviewer #1 (Public Review):
The manuscript by Rai et al., presents a straightforward approach to identify key transcriptional changes occurring as Candida glabrata infects macrophages. This is based on the premise that the changes occurring early on, as part of the pathogen response, are key in determining the progression of the infection process. Candida species are important opportunistic fungal pathogens, posing a relevant problem among immunocompromised populations. While for decades C. albicans has been responsible for most candidiasis infections, in recent years reports have indicated an upsurge in infections caused by Candida glabrata. The capacity of the latter to survive and divide within immune cells, and its increased resistance to drugs like fluconazole makes of this pathogen an organism of interest. Therefore, new …
Reviewer #1 (Public Review):
The manuscript by Rai et al., presents a straightforward approach to identify key transcriptional changes occurring as Candida glabrata infects macrophages. This is based on the premise that the changes occurring early on, as part of the pathogen response, are key in determining the progression of the infection process. Candida species are important opportunistic fungal pathogens, posing a relevant problem among immunocompromised populations. While for decades C. albicans has been responsible for most candidiasis infections, in recent years reports have indicated an upsurge in infections caused by Candida glabrata. The capacity of the latter to survive and divide within immune cells, and its increased resistance to drugs like fluconazole makes of this pathogen an organism of interest. Therefore, new information that can help to molecularly dissect aspects related to its infectious process is relevant both from the clinical and scientific points of view.
In this study, based on CHiP-seq assays directed to elongating polymerase the authors identified a series of DEGs displaying different expression profiles over a time course during macrophage infection. The authors identify several hundred of genes that show distinct profiles, from increased expression at early times, to ones becoming more active later on in the process. Based on GO analyses several correlations are drawn regarding key physiological changes that may be key for survival and virulence. Such chronological study of transcriptional changes (with a good resolution) over the first hours of macrophage infection represents an important dataset from where different testable hypotheses can emerge. The authors paid special attention to several transcription factors encoding genes which expression was high during early time points. Among them, they focused on a homolog of the S. cerevisiae transcriptional repressor Xbp1. Thus, they generated a KO of CgXBp1 and interrogated the resulting strain regarding its gene expression profile, through an equivalent time-course. The RNAPol II-Chipseq analysis showed a series of genes which expression was accelerated relative to WT, which can be interpreted as many of them being directly repressed by CgXBp1. To assess the latter, they attempt to conduct a Chip-seq of a tagged version of CgXBp1 as C. glabrata infects macrophages, nevertheless, the correlation between replicates was low and further analyses were not conducted (data not shown). Therefore, the conditions of the assay were changed and CgXbp1 Chip-seq was performed in quiescent cells, a condition where Xbp1 is known to play important roles in S. cerevisiae. This data indicated that among direct targets there are several genes encoding TFs, which suggest an important transcriptional cascade where CgXbp1 plays an important role. Such data are correlated with the RNAPolII data obtained early on in the study, and a mechanistic model is proposed. Importantly, CgXbp1 appears to recognize different types of cis-elements in the bound promoters: one similar to the reported one in yeast and another one displaying a quite different DNA logo. Additional analyses focus on determining the consequences on growth, virulence, and fluconazole resistance of the CgXbp1 (and complemented strain). Three aspects stand out: increased resistance to fluconazole, decreased proliferation in macrophages, and decreased virulence. Such phenotypes are not discussed in extenso, since most of that section focuses on the transcriptional aspects of the work.
While the datasets are valuable and several observations are interesting, it is important to be cautious as the direct targets of CgXbp1 were characterized under one particular condition and the transcriptional analyses were obtained in another condition, one shown to be highly dynamic. Therefore, several inferred targets may or may not be under CgXbp1 control during macrophage infection. Most importantly, as it is, the study does not provide a clear parallel between one list of genes and the other one, to get a glimpse of such concepts. Since CgXbp1 shows to recognize distinct binding motifs, it becomes relevant to understand whether one group behaves differently from the other one in the absence of CgXbp1.
-
Reviewer #2 (Public Review):
This manuscript describes the temporal transcriptional response of Candida glabrata during macrophage infection and characterizes the role of the transcriptional repressor CgXbp1 the process. The manuscript is well written, the experiments were well conducted and the subject is very interesting.
However, a few issues should be addressed to improve the quality of the manuscript. Particularly, it will be important to: 1) Either repeat the experiment or discuss further the unexpected failure to obtain reliable ChIP-seq results for Xbp1 within the macrophage microenvironment. during macrophage infection". The option for defined media makes it difficult to compare with the RNA PolII dataset; 2) Validate experimentaly the proposed consensus sequence recognized by Xbp1; 3) Use standard MIC determination, to have a …
Reviewer #2 (Public Review):
This manuscript describes the temporal transcriptional response of Candida glabrata during macrophage infection and characterizes the role of the transcriptional repressor CgXbp1 the process. The manuscript is well written, the experiments were well conducted and the subject is very interesting.
However, a few issues should be addressed to improve the quality of the manuscript. Particularly, it will be important to: 1) Either repeat the experiment or discuss further the unexpected failure to obtain reliable ChIP-seq results for Xbp1 within the macrophage microenvironment. during macrophage infection". The option for defined media makes it difficult to compare with the RNA PolII dataset; 2) Validate experimentaly the proposed consensus sequence recognized by Xbp1; 3) Use standard MIC determination, to have a clear notion on the impact of Xbp1 on fluconazole resistance.
These extra experiments will provide a stronger basis for the author's claims and increase the foreseen impact of this work.
-
Reviewer #3 (Public Review):
The paper by Rai and colleagues examines the transcriptional response of Candida glabrata, a common human fungal pathogen, during interaction with macrophages. They use RNA PolII profiling to identify not just the total transcripts but instead focus on the actively transcribing genes. By examining the profile over time, they identify particular transcripts that are enriched at each timepoint, and build a hierarchical model for how a transcription factor, Xbp1, may regulate this response. Due to technical difficulties in identifying direct targets of Xbp1 during infection, the authors then turn to the targets of Xbp1 during cellular quiescence.
The authors have generated a large and potentially impactful dataset, examining the responses of C. glabrata during an important host-pathogen interface. However, the …
Reviewer #3 (Public Review):
The paper by Rai and colleagues examines the transcriptional response of Candida glabrata, a common human fungal pathogen, during interaction with macrophages. They use RNA PolII profiling to identify not just the total transcripts but instead focus on the actively transcribing genes. By examining the profile over time, they identify particular transcripts that are enriched at each timepoint, and build a hierarchical model for how a transcription factor, Xbp1, may regulate this response. Due to technical difficulties in identifying direct targets of Xbp1 during infection, the authors then turn to the targets of Xbp1 during cellular quiescence.
The authors have generated a large and potentially impactful dataset, examining the responses of C. glabrata during an important host-pathogen interface. However, the conclusions that the authors make are not well supported by the data. The ChIP-seq is interesting, but the authors make conclusions about the biological processes that are differentially regulated without testing them experimentally. Because Candida glabrata has a significant percent of the genome without GO term annotation, the GO term enrichment analysis is less useful than in a model organism. To support these claims, the authors should test the specific phenotypes, and validate that the transcriptional signature is observed at the protein level.
Additionally, the authors should also include images of the infections, along with measurements of phagocytosis, to show that the time points are the appropriate. At 30 minutes, are C. glabrata actually internalized or just associated? This may explain the difference in adherence genes at the early timepoint. For example, in Lines 123-132, the authors could measure the timing of ROS production by macrophages to determine when these attacks are deployed, instead of speculating based on the increased transcription of DNA damage response genes. Potentially, other factors could be influencing the expression of these proteins. At the late stage of infection, the authors should measure whether the C. glabrata cells are proliferating, or if they have escaped the macrophage, as other fungi can during infection. This may explain some of the increase in transcription of genes related to proliferation.
An additional limitation to the interpretation of the data is that the authors should put their work in the context of the existing literature on C. albicans temporal adaptation to macrophages, including recent work from Munoz (doi: 10.1038/s41467-019-09599-8), Tucey (doi: 10.1016/j.cmet.2018.03.019), and Tierney (doi: 10.3389/fmicb.2012.00085), among others.
When comparing the transcriptional profile between WT and xbp1 mutant, it is not clear whether the authors compared the strains under non-stress conditions. The authors should include an analysis of the wild-type to xbp1 mutants in the absence of macrophage stress, as the authors claims of precocious transcription may be a function of overall decreased transcriptional repression, even in the absence of the macrophage stress. The different cut-offs used to call peaks in the two strain backgrounds is also somewhat concerning-it is not clear to me whether that will obscure the transcriptional signature of each of the strains. Additionally, the authors go on to show that the xbp1 mutant has a significant proliferation defect in macrophages, so potentially this could confound the PolII binding sites if the cells are dying.
In the section on hierarchical analysis of transcription factors, at least one epistasis experiment should have been performed to validate the functional interaction between Xbp1 and a particular transcription factor. If the authors propose a specific motif, they should test this experimentally through EMSA assays to fully test that the motif is functional.
The jump from macrophages to quiescent culture is also not well justified. If the transcriptional program is so dynamic during a timecourse of macrophage infection, it is hard to translate the findings from a quiescent culture to this host environment.
Overall, there is a strong beginning and the focus on active transcription in the macrophage is an exciting approach. However, the conclusions need additional experimental evidence.
-
Reviewer #4 (Public Review):
Macrophages are the first line of defense against invading pathogens. C. glabrata must interact with these cells as do all pathogens seeking to establish an infection. Here, a ChIP-seq approach is used to measure levels of RNA polymerase II levels across Cg genes in a macrophage infection assay. Differential gene expression is analyzed with increasing time of infection. These differentially expressed genes are compared at the promoter level to identify potential transcription factors that may be involved in their regulation. A factor called CgXbp1 on the basis of its similar with the S. cerevisiae Xbp1 protein is characterized. ChIP-seq is done on CgXbp1 using in vitro grown cells and a potential binding site identified. Evidence is provided that CgXbp1 affects virulence in a Galleria system and that this …
Reviewer #4 (Public Review):
Macrophages are the first line of defense against invading pathogens. C. glabrata must interact with these cells as do all pathogens seeking to establish an infection. Here, a ChIP-seq approach is used to measure levels of RNA polymerase II levels across Cg genes in a macrophage infection assay. Differential gene expression is analyzed with increasing time of infection. These differentially expressed genes are compared at the promoter level to identify potential transcription factors that may be involved in their regulation. A factor called CgXbp1 on the basis of its similar with the S. cerevisiae Xbp1 protein is characterized. ChIP-seq is done on CgXbp1 using in vitro grown cells and a potential binding site identified. Evidence is provided that CgXbp1 affects virulence in a Galleria system and that this factor might impact azole resistance.
As the authors point out, candidiasis associated with C. glabrata has dramatically increased in the recent past. Understanding the unique aspects of this Candida species would be a great value in trying to unravel the basis of the increasing fungal disease caused by C. glabrata. The use of ChIP-seq analysis to assess the time-dependent association of RNA polymerase II with Cg genes is a nice approach. Identification of CgXbp1 as a potential participant in the control of this gene expression program is also interesting. Unfortunately, this work suffers by comparison to a significant amount of previous effort that renders the progress detailed here incremental at best.
I agree that their ChIP-seq time course of RNA polymerase II distribution across the Cg genome is both elegant and an improvement on previous microarray experiments. However, these microarray experiments were carried out 14 years ago and while the current work is certainly at higher resolution, little more can be gleaned from the current work. The authors argue that standard transcriptional analysis is compromised by transcript stability effects. I would suggest that, while no approach is without issues, quite a bit has been learned from approaches like RNA-seq and there are recent developments to this technique that allow for a focus on newly synthesized mRNA (thiouridine labeling).
The CgXbp1 characterization relies heavily on work from S. cerevisiae. This is disappointing as conservation of functional links between C. glabrata and S. cerevisiae is not always predictable. The effects caused by loss of CgXBP1 on virulence (Figure 4) may be statistically significant but are modest. No comparison is shown for another gene that has already been accepted to have a role in virulence to allow determination of the biological importance of this effect. The phenotypic effects of the loss of XBP1 on azole resistance look rather odd (Figure 6). The appearance of fluconazole resistant colonies in the xbp1 null strain occurs at a very low frequency and seems to resemble the appearance of rho0 cells in the population. The vast majority of xbp1 null cells do not exhibit increased growth compared to wild-type in the presence of fluconazole. Irrespective of the precise explanation, more analysis should be performed to confirm that CgXbp1 is negatively regulating the genes suggested in Figure 6A to be responsible for the increased fluconazole resistance. Additionally, the entire analysis of CgXbp1 is based on ChIP-seq performed using cells grown under very different conditions that the RNA polymerase II study. Evidence should be provided that the presumptive CgXbp1 target genes actually impact the expression profiles established earlier.
-
