Figures
Abstract
Candida glabrata is an opportunistic human pathogen, capable of causing severe systemic infections that are often resistant to standard antifungal treatments. To understand the importance of protein SUMOylation in the physiology and pathogenesis of C. glabrata, we earlier identified the components of SUMOylation pathway and demonstrated that the deSUMOylase CgUlp2 is essential for pathogenesis. In this work we show that the CgUlp2 is essential to maintain protein homeostasis via the SUMO-targeted ubiquitin ligase pathway. The dual loss of deSUMOylase and specific ubiquitin ligase, CgSlx8, results in heightened protein degradation, rendering the cells vulnerable to various stressors. This degradation affects crucial processes such as purine biosynthesis and compromises mitochondrial function in the mutants. Importantly, the absence of these ubiquitin ligases impedes the proliferation of C. glabrata in macrophages. These findings underscore the significance of SUMOylation and SUMO-mediated protein homeostasis as pivotal regulators of C. glabrata physiology and capacity to survive in host cells. Understanding these mechanisms could pave the way for the development of effective antifungal treatments.
Author summary
Opportunistic fungal infections are a major cause of mortality across the world. There are a limited number of antifungal drugs and many fungi are resistant to one or more of these. Therefore, there is an unmet need for new antifungals. As fungi are evolutionarily close to humans, identifying targets that are unique or sufficiently different from human counterparts is crucial. Our laboratory has been investigating if the post-translational modification pathway, SUMOylation, can be an effective target. To this end, we are studying the importance of this pathway in the pathogenic yeast, Candida glabrata, which is the second most common causative agent for candidiasis. We had earlier identified the components of this pathway and showed that this reversible protein modification was important for pathogenesis. Here, we investigated the mechanistic basis of the role of SUMOylation-deSUMOylation in C. glabrata infection and find that the balanced activity of this pathway is critical to maintain protein homeostasis in the cell. Perturbation of this pathway leads to unwarranted protein degradation rendering the cell incapable of combating stress including survival in host cells. Therefore, the SUMO-mediated protein homeostasis pathway in C. glabrata could be targeted for new antifungal development.
Citation: Gupta D, Shukla R, Mishra K (2024) SUMO-targeted Ubiquitin Ligases as crucial mediators of protein homeostasis in Candida glabrata. PLoS Pathog 20(12): e1012742. https://doi.org/10.1371/journal.ppat.1012742
Editor: Joachim Morschhäuser, University of Würzburg, GERMANY
Received: August 9, 2024; Accepted: November 11, 2024; Published: December 6, 2024
Copyright: © 2024 Gupta et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All DATA is in the manuscript or Supporting information.
Funding: KM acknowledges support from SERB (EMR/2017/003020) for this project. DG thanks CSIR for fellowship and RS was supported by DST-WOS-A-SR/LS/464/2018. KM lab is supported by funds from DBT, SERB and Institution of Eminence, UoH (RC3-21-060). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Fungal infections are a global health problem. Estimates say over 150 million severe cases of fungal infections occur worldwide, resulting in approximately 1.7 million deaths per year world-wide due to lack of effective antifungals [1]. Candida glabrata is listed as a high-priority pathogen by the WHO in the Fungal priority pathogen list (FFPL) as one of the key causative agents of candidiasis and due to its innate resistance to high levels of antifungals [2–7]. The increased antifungal resistance causes high level of mortality in immunocompromised individuals [7]. Therefore, there is an unmet need for new and effective antifungals. A key problem in antifungal research is the close evolutionary relatedness between animals and fungi, therefore, identifying effective targets that are not toxic to humans is challenging. One of the potential targets could be post-translational modifications, because they regulate gene expression in fungi under stress including infection of host cells [8]. Ubiquitin and SUMO (Small Ubiquitin-like Modifier) belong to the Ubiquitin-like modifier proteins (Ubls) that are conjugated to lysine residues of the target protein to regulate their function and localization. They are also found to be actively involved in multiple cellular activities like cell stress response and overall maintenance of cellular homeostasis [9,10].
The coupling of SUMO and Ubiquitin to their substrate is a multi-step process. In SUMOylation, the translated SUMO protein is proteolytically processed at the C-terminus to reveal a di-glycine motif in the mature SUMO. The C-terminal glycine is conjugated to the epsilon amino group of lysine in target proteins. First, in an ATP-dependent reaction, the mature SUMO moiety are attached to E1 activating enzyme (Aos1 and Uba2 in S. cerevisiae) and then transferred to E2 conjugating enzyme (Ubc9) through a transesterification reaction [11,12]. E2 finally binds to E3 ligases (Siz1, Siz2, Mms21, Zip3) to facilitate the transfer of SUMO moieties to the lysine residues of the target substrate [9,10,13–15]. SUMO can be removed from targets by the deSUMOylases, Ulp1 and Ulp2, thereby ensuring the reversibility of SUMO modification and also maintain a pool of free SUMO in the cell for the SUMOylation cycle [16–18]. Ulp1 is the enzyme that also catalyses the maturation of SUMO. Both Ulp1 and Ulp2 have several targets that they deSUMOylate and specificity is partly imposed by the differential localization of the two deSUMOylases. Ulp1 is associated with the nuclear pore complex while Ulp2 is in the nucleoplasm in S. cerevisiae. While yeast has two deSUMOylases, there are six known deSUMOylases of the SENP class in mammals. SENP-1, SENP-2, SENP-3, SENP-5 are evolutionarily closer to Ulp1 while SENP-6, and SENP-7 are more similar to Ulp2 [18].
Protein SUMOylation is conserved across eukaryotes. Monomer of SUMO can be attached to a single lysine residue of a target or to multiple lysine residues of a target protein generating either monoSUMOylated or multiSUMOylated proteins. In addition, polySUMOylation is a special form of SUMOylation where polymeric chains of SUMO are produced when another SUMO molecule is added to the internal lysine residues of the SUMO present on target proteins. In yeast, the E3 ligase Siz2 is thought to be the primary ligase that produces polySUMO and Ulp2 is the deSUMOylase that effectively removes polySUMO chains [19]. PolySUMO serves as a platform for the recruitment of SUMO-targeted ubiquitin ligases (STUbLs) that recognize polySUMO chains. STUbLs are E3-ubiquitin ligases, and in yeast, the well-studied STUbLs are the RING domain containing Slx5/Slx8 heterodimer and Uls1 [20]. In mammals, to date, two STUbLs have been identified, namely, RNF4 and RNF11. Thus the deSUMOylases, especially Ulp2, play a key role in balancing SUMOylation and ubiquitination of target proteins and as STUbL mediated ubiquitination usually leads to protein degradation, Ulp2 is critical to maintain protein homeostasis in several complexes including chromatin structure, kinetochore assembly, DNA repair in S. cerevisiae [21–24].
In our laboratory we have shown that protein SUMOylation is essential for Candida glabrata and the pathway is critical for pathogenesis [25]. Other laboratories have shown that SUMOylation plays a major role in pathogenesis of fungi like Candida albicans, Aspergillus sp. and Magnaportha sp [26–29]. After demonstrating that loss of SUMO pathway genes leads to reduced pathogenicity of C. glabrata, we also identified the genes involved in SUMOylation and the SUMO-mediated protein homeostasis pathway in C. glabrata and other fungi [30]. In this manuscript, we have characterized the SUMO-regulated protein homeostasis pathway of C. glabrata. We demonstrate that imbalance in STUbLs and the deSUMOylase leads to increased protein degradation, reducing the capability of C. glabrata to combat stress. We also show that several pathways, particularly purine biosynthesis, mitochondrial metabolism and protein homeostasis pathways are affected. Overall this results in reduced fitness and reduced ability of C. glabrata to infect and proliferate in host cells, thus providing a promising pathway that can be targeted for antifungal development.
Results
STUbLs are required for key cellular processes
In our previous work, we had identified the homologues of the SUMOylation and the SUMO-targeted ubiquitin ligase (STUbL) pathways in fungi [30]. We found that all fungi had at least one of the STUbLs suggesting that the SUMO-dependent ubiquitin pathway of protein homeostasis could be important for the physiology of fungi [30]. To understand the importance of STUbLs in C. glabrata, we used BLAST analysis to compare homologies between the S. cerevisiae and C. glabrata proteins. These proteins include Slx5, Slx8, and Uls1 and S1 Table displays their percent similarity across the complete protein sequence. Uls1 is the most conserved of them, having 60% similarity between C. glabrata and S. cerevisiae (S1 Table). Both the Slx5 and Slx8 share over 45% protein sequence similarity between C. glabrata and S. cerevisiae. Additionally, STUbLs have conserved regions of multiple SIM motifs and a RING motif (S1 Fig). SIM motifs are SUMO interacting motifs that bind to SUMO moiety while the RING domain is required for catalysis and protein-protein interaction [31]. CgUls1, like the S. cerevisiae counterpart, has a "SNF2-like translocase" domain (S1 Fig).
We generated STUbL-deficient strains of C. glabrata to study the impact of STUbL disruption on the physiology and infectivity of the organism. Using a fusion PCR and homologous recombination-based method, we were able to create deletion strains for the CgSLX5, CgSLX8, and CgULS1 genes. In order to better understand how STUbLs interact among themselves and with SUMO protease CgULP2 in the pathobiology of C. glabrata, we also created double deletion strains of STUbLs and STUbLs with CgULP2. We could not generate Cgslx8ΔCguls1Δ double mutant despite several attempts, possibly because loss of both the STUbLs is lethal in C. glabrata.
We first performed the phenotypic characterization of the STUbL mutants in C. glabrata to study the role of SUMO-targeted ubiquitin ligases in cell physiology. Growth analysis of single deletion strains of Cgulp2Δ, Cgslx5Δ, Cgslx8Δ, and Cguls1Δ, as well as double deletion strains of Cgslx5ΔCgslx8Δ, Cgslx5ΔCguls1Δ, Cgulp2ΔCgslx8Δ, and Cgulp2ΔCguls1Δ, showed that the mutant strains of Cgulp2Δ grew slower than the wild-type (WT) as expected [25]. In addition, Cgulp2ΔCgslx8Δ double mutant grew much slower compared to WT and the Cgulp2Δ in both YPD and RPMI even at conducive temperatures (Figs 1A and 2B).
Overnight grown cultures of indicated C. glabrata strains were inoculated in (A) YPD and (B) RPMI complete medium to an initial A600 of 0.1. Absorbance at 600nm was measured at regular time intervals. Data represent the mean of biological triplicates. C. Sensitivity of the STUbL deletion strains to various stress conditions. Two microlitres of ten-fold serial dilutions of exponentially growing cultures were spotted on indicated plates and images were taken after 2–4 days of incubation. D. Heat map illustrating growth of STUbL mutants in the presence of diverse stress-causing agents. The sensitivity of STUbL mutants of C. glabrata to various treatments is represented as a heat map with black color (not sensitive; full growth) and different shades of red color (dark red indicates a high level of sensitivity). E. Bar graph depicting the survival of wild type, Cgulp2Δ, Cgslx8Δ, Cguls1Δ, Cgulp2ΔCgslx8Δ, and Cgulp2ΔCguls1Δ mutants treated with 32 μg/ml fluconazole, 150 ng/ml caspofungin, and 10 μg/ml amphotericin B. Survival was calculated relative to untreated control samples. Data represent the average of three independent experiments, with error bars indicating standard deviation (SD). Statistical significance was assessed using a two-tailed student’s t-test (*- p ≤ 0.01; **- p ≤ 0.005; ***- p ≤ 0.0005; ****- p ≤ 0.0001). F. Expression studies of SUMO protease and STUbL genes under various stress conditions (42°C, 0.04% MMS, 100mM HU) by performing RT-PCR. Data was normalized with untreated wild type cell and the graphs show mean and SD of biological triplicates. G. Biofilm formation was recorded for C. glabrata mutants vs wild type using XTT assay. Data represent the mean and SD of three independent experiments. Significance was assessed by two-tailed unpaired Student’s t-test with ****- p ≤ 0.0001; ns- non-significant. H. Biofilm from the wild type cells were used for RNA isolation and cDNA conversion. RT-PCR was then performed to check the expression level of SUMO protease CgULP2 and STUbL genes. The fold change for each gene in the biofilm-forming wild type cells was calculated with respect to planktonic cells. The experiment was done in biological triplicate and mean was plotted. The error bar shows the standard deviation of the mean.
Earlier studies have shown that Cgulp2Δ mutant grew slowly at 42°C and was vulnerable to the DNA-alkylating agent methyl methanesulfonate (MMS), replication fork staller hydroxyurea (HU), and oxidative stress-inducing agent hydrogen peroxide (H2O2) [25]. Single mutant of the STUbLs grew similar to wild type under most conditions, the exceptions being mild temperature sensitivity of Cgslx8Δ at 42°C and slow growth of Cguls1Δ in low pH media (Figs 1C, 1D and S2). Contrary to Cgslx8Δ and Cguls1Δ, the growth of the Cgslx5Δ mutant remains unaffected under diverse stress conditions (Figs 1C, 1D and S2). Both Cgslx5ΔCguls1Δ and Cgulp2ΔCguls1Δ double mutants were temperature sensitive for growth at 42°C when compared with WT (Figs 1C, 1D and S2). Interestingly, Cgulp2ΔCguls1Δ double mutants suppressed the sensitivity of Cgulp2Δ single mutant to oxidative stress (20 mM H2O2), DNA damage (0.03% MMS) and low growth on non-fermentable carbon sources (2% ethanol, 3% glycerol, 2% lactate, and 2% sodium acetate). S. cerevisiae strains with slx5Δ and slx8Δ deletion are sensitive to cell membrane stress (0.03% SDS) [32]. However, neither the single nor the double mutants of C. glabrata were sensitive to SDS (Fig 1D).
The Cgulp2ΔCgslx8Δ strain was generally more sick and grew poorly at 30°C compared to either single mutants and it grew better at low temperature (22°C) than at 30°C (Figs 1C, 1D and S2). Cguls1Δ, Cgslx5ΔCgslx8Δ, Cgslx5ΔCguls1Δ, and Cgulp2ΔCgslx8Δ were sensitive to acidic pH as well (Fig 1D). However, Cgulp2ΔCgslx8Δ was more sensitive to all stress conditions compared to either single mutants i.e., Cgslx8Δ exacerberated the sensitvity of Cgulp2Δ single mutant to stress. The phenotypes of increased stress sensitivity in single and double deletion strains of SUMO-targeted ubiquitin ligases in various stress conditions were rescued by ectopic expression of CgULP2 and STUbL genes, indicating, indeed, the phenotypes observed are due to loss of STUbLs (S3 Fig). In summary, these data show that the loss of Cgslx5 has minimal effect on the various conditions tested here; Cgslx8Δ exacerbates multiple phenotypes of Cgulp2Δ and Cguls1Δ suppresses sensitivity of Cgulp2Δ to several stress conditions.
As the STUbLs appeared to contribute to stress resistance in C. glabrata, we also assessed antifungal susceptibility of these mutants. We first established that 50% growth inhibition of wild type C. glabrata was at 32 μg/ml fluconazole, 150 ng/ml caspofungin, and 10 μg/ml amphotericin B (Fig 1E). The Cgulp2Δ mutant showed heightened sensitivity to these antifungal agents, with survival of 31%, 42%, and 32%, respectively, indicating a reduced tolerance. In contrast, the Cgslx8Δ and Cguls1Δ mutants exhibited slightly higher resistance, maintaining survival between 56–58% across the treatments. However, the Cgulp2ΔCgslx8Δ double mutant showed the greatest sensitivity, with only 19%, 22%, and 21% survival under the same conditions, highlighting an increase in vulnerability when both the genes were deleted. Interestingly, the Cgulp2ΔCguls1Δ double mutant displayed a more robust antifungal tolerance, with survival of 73%, 66%, and 65%, suggesting that the loss of Cguls1 partially mitigates the stress experienced by the Cgulp2Δ single mutant (Fig 1E).
As the loss of STUbLs had differential sensitivities to various stress conditons tested here, it is possible that they are specifically required to combat these conditons. Therefore we analysed the expression of the SUMO protease CgULP2 and STUbL genes under thermal stress (42°C), 0.04% MMS and hydroxyurea treatment by quantitative RT-PCR (Fig 1F). The phenotypes observed largely correlated with the expression analyses of the SUMO protease CgULP2 and STUbL genes under thermal stress (42°C), MMS and hydroxyurea treatment. For instance, in the aforementioned stress conditions, the lack of CgULP2 caused significant growth anomalies and we see elevated expression of CgULP2 when exposed to these stress. However, CgSLX5, CgSLX8, and CgULS1 genes were downregulated in response to hydroxyurea treatment; and these strains do not display any sensitivity to hydroxyurea. Cgslx8Δ was sensitive to higher temperatures, and its expression increases when subjected to thermal stress while CgSLX5 and CgULS1 levels were lower. When treated with MMS, the CgULS1 gene upregulation was seen, whereas CgSLX5, and CgSLX8 were downregulated (Fig 1F). These findings collectively suggest that specific SUMO-targeted ubiquitin ligases are essential for resistance to heat and DNA damage.
Biofilm formation by C. glabrata on the biotic (body tissue) and abiotic (plastic and medical devices) surfaces are major virulence factors [33]. We therefore tested the ability of these mutants to form biofilm and compared them with wild type strains. Interestingly, all single mutants of STUbL had reduced capacity of form biofilms and the double mutants of Cgulp2ΔCgslx8Δ were more compromised than either single mutant (Fig 1G). Cgulp2ΔCguls1Δ, on the other hand, formed biofilm similar to that of WT. This follows the same trend that we have observed for stress conditions tested in that the double mutant of Cgulp2ΔCguls1Δ suppresses the biofilm formation defect of both the single mutants, Cgulp2Δ and Cguls1Δ. Further, to assess if these phenotypes indicated the requirement of these gene products for biofilm formation, we checked the expression of deSUMOylase CgULP2 and STUbL (CgSLX5, CgSLX8, and CgULS1) genes. CgULS1 was found to be highly upregulated which was followed by CgULP2 and CgSLX5, while CgSLX8 levels were similar to wild type (Fig 1H). Interestingly, while CgULS1 and CgULP2 expression is upregulated in biofilms, the double mutant Cgulp2ΔCguls1Δ is competent for biofilm formation, unlike the two single mutants. This indicates that there is a connection between Ulp2 substrates and Uls1-mediated protein degradation that leads to different outcomes for the single and double mutants.
STUbLs are essential for the survival of C. glabrata within the macrophage
We studied the ability of STUbL mutants to survive and proliferate in macrophages using the colony-forming unit (CFU) assay. For this, cultured human THP-1 macrophages were infected with equal numbers of wild type, Cgulp2Δ, and STUbL mutants. Internalized C. glabrata were then extracted from the macrophages at the indicated time points and plated on YPD solid medium to obtain the live cell count. Based on our previous studies, we know that after 8 hours of addition of C. glabrata to macrophages, the internalized C. glabrata cells start to multiply and this was captured in the WT infections in this experiment as well. However, we observed that the survival and replication of all the internalized STUbL mutants of C. glabrata were much less than the wild type (Fig 2A and 2B). Also, while growth in YPD was similar in all strains except Cgulp2Δ and the Cgulp2ΔCgslx8Δ as described above, growth in RPMI medium was 6–8 fold lower in Cgulp2Δ and all STUbL mutants as compared to the wild type (S4 Fig) suggesting that STUbLs are required for optimal growth in alternate culture media.
A. and B. 2.5x106 cells of the indicated C. glabrata strains were added to phorbol 12-myristate 13-acetate-differentiated THP-1 cells in a 24-well plate, and non-phagocytosed C. glabrata cells were removed after 2 hours by washing. Intracellular yeast cells were recovered at the indicated time intervals by lysing THP-1 cells in water and plating appropriate dilutions of lysates on the YPD agar. After 24–48 hours of incubation at 30°C, yeast colonies were counted. Mean of three independent experiments are shown alongwith SD. C. Expression levels of STUbL genes of the macrophage internalized C. glabrata cells. Mean of three independent experiments are shown alongwith SD (*- p ≤ 0.01; ns- non-significant, two-tailed unpaired Student’s t-test).
To test if deSUMOylase and SUMO-targeted ubiquitin ligases were induced upon infection in macrophages, we checked the expression of deSUMOylase CgULP2 along with STUbL genes CgSLX5, CgSLX8, and CgULS1 at different time points. Fig 2C shows the changes in expression during the time course of infection. The expression of CgULP2 was upregulated as early as 2 hours of infection, which remained high upto 24 hours. Importantly, expression of CgSLX5, CgSLX8, and CgULS1 was also found to be induced; while the CgSLX5 and CgSLX8 transcripts show a modest increase, CgULS1 had higher levels of expression (Fig 2C). Notably, these findings also corroborate with the CFU assay, which revealed that single mutants of Cgulp2Δ, Cgslx5Δ, Cgslx8Δ, and Cguls1Δ had reduced intracellular proliferation and had a low CFU count (Fig 2A and 2B). This supports the notion that CgULP2, CgSLX5, CgSLX8, and CgULS1 are required for the survival and proliferation of pathogen in the host.
Protein SUMOylation in STUbL mutants of C. glabrata
Loss of function of Ulp2 was earlier shown to reduce the amount of SUMOylated proteins [25]. We had reasoned that accumulation of polySUMOylated proteins led to increased degradation of proteins via the SUMO-targeted Ubiquitin ligases, resulting in lower levels of SUMOylated protein in Cgulp2Δ strains. Therefore, we hypothesized that double deletion strain of both Cgulp2 and STUbLs will lead to the retention of higher levels of polySUMOylated proteins. In order to determine the SUMO proteome in each of the STUbL mutants, we used an anti-FLAG antibody to perform western blots on whole cell lysates from the C. glabrata STUbL mutants augmented with dual tagged CgSmt3 protein (carrying His6 and FLAG epitopes at the N-terminus). The lack of higher molecular weight SUMOylated proteins in the Cgulp2 mutant was confirmed (Fig 3A). In the case of STUbL single mutants, only Cgslx5Δ was found to have higher molecular weight SUMOylated proteins (Fig 3A). In principle, STUbL deletion strains should accumulate higher molecular weight polySUMOylated proteins [34]. It is possible that the absence of higher molecular weight SUMO conjugates in the Cgslx8Δ and Cguls1Δ single mutants suggests that in the absence of one of the STUbLs, another might compensate and one STUbL gene may be sufficient to maintain homeostasis. In this context, it is interesting to note that the Cgslx8ΔCguls1Δ double mutant could not be generated, suggesting at least one of these STUbLs was essential for survival. Intriguingly, accumulation of higher molecular weight SUMOylated proteins was seen in the double mutants of Cgulp2ΔCguls1Δ, and Cgslx5ΔCguls1Δ. This suggests that Uls1 could be the primary ubiquitin-ligase for many of the SUMOylated proteins and in the absence of Uls1, these proteins are stabilized (Fig 3A, Lane 7 and 9). Of note, we have demonstrated that Cgulp2ΔCguls1Δ double mutant was able to restore Cgulp2Δ mutant growth under a variety of stress conditions (Fig 1C and 1D) and interpret the appearance of these SUMOylated proteins as a reason for the restoration of growth in the double mutant. Importantly, the Cgulp2ΔCgslx8Δ, which had exacerbated the Cgulp2Δ phenotypes, not only grew very slowly, it also has the least amount of SUMOylated proteins, suggesting enhanced degradation of proteins (Fig 3A-Lane 8 and Fig 3B).
A. Strains were grown in YPD agar and an equal number of cells were taken for whole-cell extracts. Tubulin was used as a loading control. The blot was stained with Ponceau S to show the level of protein degradation in Cgulp2ΔCgslx8Δ double mutant in comparison to WT and other mutants. B. The diagram illustrates the increased protein degradation in Cgulp2ΔCgslx8Δ and the restoration of protein levels in the Cgulp2ΔCguls1Δ double mutant.
Each of these enzymes affects growth phenotypes and SUMOylated products differently. Based on these observations, we reason that because the polySUMOylated proteins are not degraded when Cguls1 is lost in the Cgulp2Δ mutants, the Cgulp2Δ reduced growth phenotype is suppressed in the double mutant. On the other hand, Cgulp2Cgslx8Δ cells lose more protein as seen in the western blot, and get sicker (Fig 3A). Based on this and the observation that single mutants of STUbLs do not accumulate higher molecular weight SUMOylated proteins, we hypothesize that Cgslx8 deletion possibly upregulates Uls1 activity and leads to increased protein degradation making the Cgulp2Cgslx8Δ particularly vulnerable. An alternative explanation is that CgUls1 is the main STUbL, however, several essential proteins required for strong growth may be degraded in Cgslx8Δ mutant or accumulated as polySUMOylated proteins and induce toxicity leading to compromised growth.
CgUls1 protein is upregulated in STUbL mutants
To test whether CgUls1 is upregulated in Cgslx8 deletion, we created a construct of dual-tagged 6xHIS-3xFLAG-CgULS1 (HF-CgULS1) under its own promoter and confirmed the expression of CgUls1 protein in the Cguls1Δ mutant by western blot using an α-flag antibody. Cguls1Δ was sensitive to DNA-damaging agents at higher MMS (0.04%) concentration. We tested the complementation of HF-CgULS1 in the Cguls1Δ mutant in this condition and we found that HF-CgULS1 restored the phenotype of the Cguls1Δ mutant (S5A Fig).
We checked the level of CgUls1 protein in WT, Cgulp2Δ, Cgslx8Δ, and Cgulp2ΔCgslx8Δ deletion strain and found higher amounts of CgUls1 protein in Cgulp2ΔCgslx8Δ double deletion mutant in comparison with wild type and Cgulp2Δ, Cgslx8Δ single mutants (Fig 4A and 4B), suggesting that there was upregulation of CgUls1 in the Cgulp2ΔCgslx8Δ mutants. Further, to see if this was transcriptional upregulation, we compared the expression of CgULS1 gene in these deletion strains by quantitative RT-PCR. Indeed the transcripts were higher in the double mutant compared to Cgulp2Δ and Cgslx8Δ single mutant (Fig 4C). Of note, we detected higher than wild type levels of CgUls1 protein in Cgulp2ΔCgslx8Δ strains in the proteomics data as well (S5B Fig). Together, these findings support our hypothesis that, due to the increased potential activity of the CgUls1 protein in the Cgulp2ΔCgslx8Δ mutant, cells lose more protein and are compromised for growth even under conducive conditions.
A. Western blot analysis to show the expression level of CgUls1 protein in STUbL mutants using anti-FLAG antibody. Tubulin was used as a loading control. B. Quantification of the western blot. ***- p ≤ 0.0005; ns- non-significant; one-way ANOVA. C. YPD grown wild type, Cgulp2Δ, Cgslx8Δ, and Cgulp2ΔCgslx8Δ cells were used for RNA isolation and cDNA conversion. RT-PCR was then performed to check the expression level of CgULS1 gene in these mutants. Data represents the mean and standard deviation of three independent experiments. Statistical significance was calculated using one-way ANOVA (*, p < 0.05).
Loss of proteins and reduced growth of STUbL mutants is due to enhanced proteasomal degradation
We examined the SUMOylation pattern in Cgulp2Δ, Cgslx8Δ, and Cgulp2ΔCgslx8Δ cells treated with the proteasome inhibitor MG132 to test the possibility that an increase in the accumulation of polySUMOylated proteins in these mutants results in their degradation by the ubiquitin-proteasome pathway. Upon treatment with MG132, degradation of higher molecular weight SUMO-conjugated proteins was prevented, and higher molecular weight SUMOylated proteins could be detected in Cgulp2Δ single mutant (Fig 5B, Lane 2) as compared to the untreated mutant strain (Fig 5A, Lane 2). Fewer changes were seen in the Cgslx8Δ consistent with the phenotypes seen for this mutant. Most interestingly, in the double mutant of Cgulp2ΔCgslx8Δ, with MG132, higher molecular weight SUMO-conjugated proteins were now appearing suggesting prevention of degradation of proteins (Fig 5B-Lane 4 and Fig 5C) compared to the untreated mutant (Fig 5A, Lane 4). Furthermore, the reappearance of higher molecular weight proteins is accompanied by the concomitant restoration in the growth of these mutants (Fig 5D). Also, MG132 treatment could overcome the temperature and MMS sensitivity of the Cgulp2Δ, Cgslx8Δ, and Cgulp2ΔCgslx8Δ mutant (Fig 5D–5E). These findings strongly imply that the reason for slow growth at 30°C and sensitivity to stress inducers in Cgulp2ΔCgslx8Δ double mutant was due to excess proteasomal degradation. Together these results support our hypothesis that loss of Ulp2 leads to the increased accumulation of polySUMOylated proteins that were then degraded by proteasome via the STUbL pathway leading to deregulated protein homeostasis.
C. glabrata strains, both wild type and mutant, were grown overnight in YNB medium at 30°C and 180 rpm with 0.1% proline serving as the only nitrogen source. After cell harvesting, a fresh medium containing 0.003% SDS was used to adjust the A600 to 0.5, and the cells were then grown for 4 hours. A. Cells without MG132 treatment were used as a control. B. Cultures were placed in a 2 ml microcentrifuge tube with 1 ml of YNB-proline media and 200μM MG132, and were then incubated for two hours at 30°C and 180 rpm. Following cell pelleting and whole cell extract preparation, cells were analyzed by western blot using α-Flag antibody. Tubulin was used as a loading control. Blots stained with Ponceau S show protein degradation in Cgulp2ΔCgslx8Δ double mutant in untreated cells (A, lane 4) in comparison to WT and recovery in protein levels upon MG132 treatment (B, lane 4). C. Diagram shows that MG132 treatment prevents protein degradation in Cgulp2ΔCgslx8Δ mutant. D. MMS (0.045%) was added to the media and grown for 3 hours after MG132 treatment for 2 hours. Similarly, temperature treatment was given to the cultures for 3 hours. After that, cultures were serially diluted 10-fold, and a 2 μl volume was spotted on the YPD medium. Images of the plates were taken following 2–3 days of incubation at desired temperatures. E. Quantification of the growth of MGI32 treated STUbL mutants in various stress conditions. Data represents the mean and SD of at least three independent experiments. Statistical significance was assessed using a two-tailed paired student’s t-test (*- p ≤ 0.01; **- p ≤ 0.005; ***- p ≤ 0.0005).
Identification of global proteome of Cgulp2Δ, Cgslx8Δ, Cguls1Δ, Cgulp2ΔCgslx8Δ and Cgulp2ΔCguls1Δ deletion strains in comparison with wild type
The results shown in the above sections indicate that knock out of a single STUbL has minimal effect on the growth and capacity of the organism to combat stress when grown in rich media. However, when the deSUMOylase is also inactive, we observe contrasting outcomes. When Cguls1 is deleted in the Cgulp2Δ, the strains recover from some of the adverse effects of Cgulp2Δ. On the other hand, loss of Cgslx8Δ in the Cgulp2Δ, leads to exacerbation of phenotypes and rendering the double mutants more compromised for growth in most conditions. We speculate that both the exacerbation and suppression of Cgulp2Δ phenotypes are due to differences in the degradation of selected polySUMOylated proteins. In order to obtain an overall idea of the proteome in the absence of these STUbLs and Ulp2, and also to determine if proteins that are part of critical growth pathways are particularly degraded in the STUbL double mutant Cgulp2ΔCgslx8Δ and restored in Cgulp2ΔCguls1Δ, we conducted tandem mass tag (TMT) labeled quantitative mass spectrophotometry analyses of log-phase grown cells of WT, Cgulp2Δ, Cgslx8Δ, Cguls1Δ single mutants, Cgulp2ΔCgslx8Δ, and Cgulp2ΔCguls1Δ double mutants.
From the TMT-labeled mass spectrometry, we have acquired 3281 proteins in total. 2194 proteins were retained for analysis after Principal Component Analysis (PCA) and eliminating proteins that displayed differences between biological replicates. Based on the abundance value of the proteins, we have calculated the p-value for each mutant using MeV_4.9.0 software and took a threshold of 0.05 for significant p-value. We identified 107, 87, 148, 278, and 300 proteins that were significantly upregulated and 90, 130, 241, 184, and 134 proteins were significantly downregulated in Cgulp2Δ, Cgslx8Δ, Cguls1Δ single mutants, Cgulp2ΔCgslx8Δ, and Cgulp2ΔCguls1Δ double mutants respectively (S2–S6 Tables). Next, we used the DAVID database to perform Gene Ontology (GO) to identify the pathways of the upregulated and downregulated proteins for each mutant in comparison with WT (S7–S11 Tables). The pathways are represented as horizontal bar graphs using SR plot software (Fig 6).
The horizontal bar graph shows the pathways in A) Cgulp2Δ, B) Cgslx8Δ, C) Cguls1Δ, D) Cgulp2ΔCgslx8Δ, and E) Cgulp2ΔCguls1Δ mutant in comparison to WT. The red bar represents fraction of proteins in the indicated pathways with elevated protein levels and the green bar represents fraction of proteins in the indicated pathways with reduced protein levels.
We have seen multiple pathways that were differentially affected in these mutants. These include proteins that are involved in metabolic processes, biosynthetic processes, membrane transport, ribosomal biogenesis, translation, proteasomal protein mediated degradation and others. Details of the protein components of pathways identified for each mutant is in S7–11 Tables. Proteomics data analysis revealed that the proteins involved in transport, particularly mitochondrial electron transport, transmembrane transport, and other ion transports were reduced in Cgulp2Δ, Cgslx8Δ, Cguls1Δ and Cgulp2ΔCgslx8Δ, but were higher in Cgulp2ΔCguls1Δ double mutant (Figs 6 and S6).
We also observed proteins involved in ribosomal biogenesis and translation were reduced in the Cgulp2Δ and Cgslx8Δ single mutant. Similarly, a large group of mitochondrial and cytoplasmic ribosomal proteins comprising both large and small subunits were reduced in Cgulp2ΔCgslx8Δ double mutant (Figs 6 and S7). Thus, the Cgulp2ΔCgslx8Δ double mutant is likely to be compromised in translation of both cytosolic and mitochondrial proteins leading to growth defects even at conducive temperatures in rich medium. Although the Cgulp2ΔCguls1Δ double mutant had increased levels of mitochondrial ribosomal proteins of both large and small subunits, the majority of cytoplasmic ribosomal proteins were reduced (Figs 6 and S7). This indicates members of the ribosomal biogenesis and translation pathway are more likely to be impacted by SUMO-mediated protein homeostasis.
Another interesting observation was that in Cgulp2Δ single mutant, there was elevated levels of a small number of proteins that belongs to the proteasomal degradation pathway. The majority of the proteins in the 26S proteasomal complex and those involved in endoplasmic reticulum-associated protein degradation (ERAD) which include CgKar2, CgCdc48, CgHlj1, CgHrd3, CgJem1, CgUbr1 and CgUfd1 were also highly upregulated in the Cgulp2ΔCgslx8Δ double mutant (Figs 6 and S8), suggesting that this enhanced levels of the degradative pathways could be responsible for the elevated levels of protein degradation occurring in this double mutant. In contrast, the Cguls1Δ and Cgulp2ΔCguls1Δ mutants showed downregulation of the ubiquitin-mediated proteasomal degradation pathway, possibly resulting in reduced protein degradation (Figs 6 and S8). This differential regulation of proteasomal pathway suggests that the balance of protein turnover was strongly influenced by the specific genetic context, particularly when combining deletion of CgUlp2 with either CgSlx8 or CgUls1.
These findings correlate with the SUMOylation profiles observed in the western blots (Fig 5) with Cgulp2ΔCguls1Δ strains with lowered proteasomal components showing higher levels of SUMO conjugates. Conversely, the Cgulp2ΔCgslx8Δ double mutants exhibit fewer SUMOylated proteins, consistent with increased proteasomal activity leading to elevated protein degradation. Moreover, MG132 treatment of the Cgulp2ΔCgslx8Δ mutants inhibits proteasomal degradation, leading to the accumulation of higher molecular weight SUMOylated proteins. Together, these observations highlight the link between proteasomal activity and SUMOylated protein regulation across these mutants.
As these deletion strains displayed altered growth and ability to combat stress, we examined the biosynthetic processes that are critical for growth. We found that several biosynthetic pathways were also differentially impacted in Cgulp2Δ, Cgslx8Δ, Cguls1Δ single mutants, Cgulp2ΔCgslx8Δ, and Cgulp2ΔCguls1Δ double mutants (Figs 6 and S9). Moreover, Cgulp2ΔCgslx8Δ also had lower levels of proteins involved in aminoacid, lipid, and ATP biosynthesis (S9 Fig, first column). Among the numerous biosynthetic processes, proteins involved in purine biosynthesis were lower in the Cgulp2Δ and Cgulp2ΔCgslx8Δ double mutant and higher in Cgulp2ΔCguls1Δ (Fig 7A). Nucleotide biosynthesis pathways fall into two categories: de novo pathways and salvage pathways (35,36). Proteins of de novo purine biosynthesis pathways were reduced in the Cgulp2ΔCgslx8Δ double mutant (Fig 7A). Purine nucleotide biosynthesis, as shown in Fig 7B, begins with 5-phosphoribosyl-1-pyrophosphate (PRPP), which is then converted through a sequence of reactions into inosine 5’-monophosphate (IMP). After that, IMP may go through various processes that produce either GMP or AMP, which are subsequently transformed into ADP and GDP, respectively [35,36]. Based on the mass spectrometry data, the Cgulp2ΔCgslx8Δ double mutant had downregulated levels of several proteins involved in this pathway. Particularly, CgAde5,7 was reduced in the Cgulp2Δ and Cgulp2ΔCgslx8Δ double mutant but was increased in the Cgulp2ΔCguls1Δ double mutant (Fig 7A).
A. Bar graph shows the levels of the proteins (in comparion to wild type) involved in this pathway for Cgulp2Δ (red bar), Cgulp2ΔCgslx8Δ (blue bars), and Cgulp2ΔCguls1Δ (green bar). The error bar represents the log2 standard deviation. B. de novo pathway of purine nucleotide biosynthesis. Proteins labeled in green color were obtained in the mass spectrometry analysis. The site of action of azaserine, pemetrexed, and isatin in the purine biosynthesis pathway are indicated. Abbreviation: PRPP- 5-phosphoribosyl-1- pyrophosphate, PRAT- phosphoribosylamidotransferase, PRA- phosphoribosylamine, GAR- glycineamideribonucleotide, FGAR- N-formylglycinamide ribonucleotide, FGAM- N-formylglycinamidine ribonucleotide, AIR- 5-aminoimidazole ribonucleotide, CAIR- AIR carboxamide, SAICAR- N-succino-5-aminoimidazole-4-carboxamide ribonucleotide, AICAR- aminoimidazole-4-carboxamide ribonucleotide, FAICAR- 5-formyl-AICAR, IMP, SAMP- adenylosuccinate, AMP- adenosine monophosphate, ADP- adenosine diphosphate, XMP- xanthosine monophosphate, GMP- guanosine monophosphate, GDP- guanosine diphosphate, ADK1-adenylate kinase 1, GUK1- guanylate kinase 1. C. Overexpression of CgADE5,7 in STUbL mutants. pGRB2.2 plasmid encoding CgADE5,7 was transformed in both the wild type (WT) and mutant strains of C. glabrata and the empty vector was used as control. To assess the impact of CgADE5,7 overexpression on growth, 3 μl of 10-fold serial dilutions of exponential phase (1.0 A600) grown cultures of C. glabrata strains were spotted in medium lacking uracil. Plate images were taken after 1–3 days of incubation at 30°C and 42°C, respectively. D and E. Bar graphs show the quantification of cell growth for each mutant upon overexpressing the CgADE5,7 gene (blue bars), compared to the control strains (light green bars), at both 30°C and 42°C. Statistical significance was assessed using a two-tailed student’s t-test (*- p ≤ 0.01; **- p ≤ 0.005; ***-p ≤ 0.0005; ns-non-significant). F-H. Growth assay in STUbL mutants. Overnight grown cultures of indicated C. glabrata strains were diluted in fresh YPD broth to an initial A600 of 0.1. 100μl of diluted C. glabrata cell suspension were added to each well of a 96-well plate. Azaserine, pemetrexed, and isatin inhibitors were added in the concentration of 2mM, 6mM, and 4mM respectively to the cell suspension, and the final volume of 200μl for each well was maintained by adding YPD. YPD broth and cell suspension (100μl) without inhibitors were used as a control. Inoculum was incubated at 30°C for 48 hr with 300rpm agitation. Absorbance at 600nm was measured at 48 hours to assess the survival rates, calculated relative to untreated control samples. The data are expressed as a percentage, with the no-drug well (vehicle control) of each strain set as 100%. Statistical significance was assessed using a two-tailed student’s t-test (*- p ≤ 0.01; ***- p ≤ 0.0005; ns-non-significant). Data represent the mean of biological triplicates and the error bar represents the standard deviation. I. Growth of WT and mutants at 30°C when treated with 2mM azaserine, 6mM pemetrexed, and 4mM isatin is shown. 2 μl of 5-fold serial dilutions of logarithmic phase grown culture of WT and mutants were spotted on YPD plates. Plate images were captured after 48 hours of incubation at 30°C.
We then investigated whether CgADE5,7 overexpression could restore growth of the Cgulp2ΔCgslx8Δ double mutant at 30°C. In order to accomplish this, we created a construct with the CgADE5,7 gene in the pGRB2.2 C. glabrata expression vector (carrying 3xFLAG at the N-terminus), and we used an α-flag antibody to confirm the expression of CgAde5,7 protein in the WT, Cgulp2Δ, Cgslx8Δ, Cguls1Δ, Cgulp2ΔCgslx8Δ, and Cgulp2ΔCguls1Δ mutants through western blot. All these mutants showed similar levels of expression of the CgAde5,7 protein (S11A Fig). In comparison to controls, the overexpression of the CgADE5,7 gene resulted in some degree of growth restoration in the cells of the WT and mutants at 30°C and 42°C (Fig 7C). Significantly, we found that overexpressing the CgADE5,7 gene allowed the Cgulp2ΔCgslx8Δ double mutant to partially recover its growth at 30°C and 42°C (Fig 7C–7E). Interestingly, the temperature sensitivity of the Cgulp2ΔCguls1Δ mutant at 42°C was also partly rescued by overexpressing CgAde5,7 indicating that while the CgAde5,7 levels were elevated at 30°C, it maybe compromised at 42°C in Cgulp2ΔCguls1Δ.
If the Cgulp2ΔCgslx8Δ and the Cgulp2 were compromised for purine biosynthesis, they are likely to be more sensitive to purine biosynthesis inhibitors. We therefore, tested the sensitivity of the STUbL mutants to purine biosynthesis inhibitors, namely, azaserine, pemetrexed, and isatin. We examined the cell growth in WT and mutants. The initial, rate-limiting step involves the conversion of PPRP to PRA, which is catalyzed by PRPP amidotransferase. Azaserine, an antifungal agent, inhibits this step [35,36]. The anticancer drug, pemetrexed, prevents GAR transformylase from converting GAR to FGAR in the third step. The process by which the enzyme AIR carboxylase converts AIR to CAIR is inhibited by isatin, an antifungal agent.
In this study, we observed that the survival of C. glabrata WT cells decreased to approximately 50–60% when treated with azaserine, pemetrexed, and isatin, compared to the vehicle control. This reduction corresponded to a 40–50% growth inhibition, highlighting the significant impact of these purine biosynthesis inhibitors on WT cell growth (Fig 7F–7I). Specifically, survival for the Cgulp2Δ and Cgulp2ΔCgslx8Δ mutants dropped to 19% following treatment with 2 mM azaserine, indicating an 81% inhibition in these strains. Furthermore, survival was reduced to 40% with 6 mM pemetrexed and 42% with 4 mM isatin. In contrast to the WT, these mutants exhibited notably slower growth, especially under 2 mM azaserine treatment (S11B Fig). The Cgslx8Δ, Cguls1Δ, and Cgulp2ΔCguls1Δ mutants displayed growth and survival patterns similar to those of the WT (Fig 7F–7I). Altogether, these data suggest that compromised purine nucleotide biosynthesis was one of the key reasons for the slow growth of the Cgulp2ΔCgslx8Δ mutant at 30°C and the restoration of cell growth in Cgulp2ΔCguls1Δ double mutant. This suggests that particular components and pathways are more sensitive to the SUMO-mediated protein homeostasis.
Mitochondria function is compromised in Cgulp2Δ and Cgulp2ΔCgslx8Δ
Apart from the biosynthetic processes, proteins involved in mitochondrial protein synthesis and members of the electron transport chain were affected in the Cgulp2ΔCgslx8Δ double mutant. Complexes II, III, IV, and V were highly reduced in Cgulp2ΔCgslx8Δ, suggesting that mitochondrial metabolism was severely affected. In the case of Cgulp2Δ, there was no clear down regulation, with few of components being present in higher or lower levels than wild type. In addition, proteins associated with these complexes were either upregulated or up to the level of WT in the Cgulp2ΔCguls1Δ double mutant (Fig 8A). Dysregulation of mitochondria has several consequences such as ATP depletion, ROS generation, and oxidative stress [37,38]. Also, Cgulp2Δ and Cgulp2ΔCgslx8Δ have reduced growth in non-fermentable carbon sources like 2% ethanol, 3% glycerol, and 2% lactate, and Cgulp2ΔCguls1Δ appeared to suppress the defects seen in the single mutant of Cgulp2Δ (Fig 1D).
A. Bar graph shows the differential levels of the proteins involved in the mitochondrial complexes of the ETC pathway in Cgulp2Δ, Cgulp2ΔCgslx8Δ, and Cgulp2ΔCguls1Δ. The error bar represents the standard deviation. B. Cells were grown in YPD medium without 10mM NAC (- NAC) at 30°C and stained with MitoTracker Green and MitoSOX (magenta). Confocal microscopic images (left panel) of WT, Cgulp2Δ, Cgulp2ΔCgslx8Δ, and Cgulp2ΔCguls1Δ to check mitochondrial morphology and ROS levels. The scale bar represents 5 μm. The right panel displays the quantification of mitochondrial morphology in WT and mutants in absence of NAC. The red dotted box indicates the region enlarged for better visualization. C. Cells were grown in YPD medium with 10mM NAC (+ NAC) at 30°C for 12 hours and then stained with MitoTracker Green and MitoSOX. Images were captured in a confocal microscope (left panel). Examples of tubular and fragmented mitochondria selected for quantification are shown. The scale bar is 5 μm. The right panel shows the quantification of mitochondrial morphology in WT and mutants following NAC treatment. For each experiment, 100 cells were counted, and with three independent replicates, a total of 300 cells were analyzed. D-E. Overnight grown cultures of indicated C. glabrata strains were inoculated in YPD medium without NAC (- NAC) or with 10mM NAC (+ NAC) at an initial A600 of 0.1. Absorbance at 600nm was measured at regular intervals. Data represent the mean of biological triplicates. F-G. Doubling time was calculated for the C. glabrata strains with and without NAC and plotted. Statistical significance was assessed using a two-tailed student’s t-test (*- p ≤ 0.01; ***- p ≤ 0.0005; ****- p ≤ 0.0001; ns-non-significant).
We initially examined the mitochondrial morphology of the wild type and mutants of the C. glabrata. The wild type cells exhibited a tubular mitochondrial structure, while Cgulp2ΔCgslx8Δ displayed 90% fragmented mitochondria, followed by the Cgulp2Δ mutant which had 60% fragmentation. However, Cgulp2ΔCguls1Δ mutant had a mitochondrial morphology similar to that of the wild type (Fig 8B). To assess if the mitochondrial morphological changes in the mutants influenced reactive oxygen species (ROS) levels, we stained WT and C. glabrata mutants with MitoSOX, a mitochondria superoxide indicator. Cgulp2ΔCgslx8Δ cells showed elevated ROS levels, with a stronger fluorescent signal followed by Cgulp2Δ mutant. Cgulp2ΔCguls1Δ and WT had no or very little amount of ROS (Fig 8B). Further, to validate that the increased signal in Cgulp2Δ and Cgulp2ΔCgslx8Δ cells was due to enhanced mtROS production, we treated these cells with N-acetylcysteine (NAC), an antioxidant known to reduce ROS generation. NAC treatment significantly decreased fluorescent signals in both Cgulp2Δ and Cgulp2ΔCgslx8Δ cells compared to untreated cells (Fig 8C). These data together suggest a compromised mitochondrial function in Cgulp2ΔCgslx8Δ and Cgulp2Δ. To further confirm this data, we also grew cells in the presence of NAC and found that growth of the Cgulp2Δ single mutant improved on NAC medium, Cgslx8Δ single mutant displayed growth similar to WT under both conditions, and the Cgulp2ΔCgslx8Δ double mutant showed significantly improved growth with NAC relative to growth without NAC (Fig 8D–8E). These findings were further supported by the doubling time analysis, which showed a consistent trend with the growth curves, indicating a reduced doubling time for Cgulp2Δ and Cgulp2ΔCgslx8Δ in the presence of NAC (Fig 8F–8G). After NAC treatment, 29% of Cgulp2ΔCgslx8Δ cells displayed a restored tubular mitochondrial structure, and the Cgulp2Δ mutant showed minor improvement, with 65% of cells exhibiting tubular morphology (Fig 8C). These data establish that increased ROS levels could be one of the factors that leads to compromised growth of the Cgulp2ΔCgslx8Δ double mutant.
Discussion
In this work, we show that in Candida glabrata, protein homeostasis is regulated by protein deSUMOylase CgUlp2 and the STUbLs. A balance of the removal of polySUMOylation by CgUlp2 and degradation of polySUMOylated proteins by STUbLs is critical for growth, combating stress and for C. glabrata to survive in host cells (Fig 9). We had earlier identified the potential STUbLs in C. glabrata based on sequence similarity with STUbLs of S. cerevisiae. While the full length deletions of individual STUbLs renders the organism susceptible to several stress causing agents and inability to proliferate in macrophages, it is the combination of STUbL deletions with the deSUMOylase CgUlp2 that has profound phenotypes. The loss of Cguls1 along with the Cgulp2 suppresses some of the stress sensitivity phenotypes of single mutant Cgulp2. Conversely, the loss of Cgslx8 along with Cgulp2 exacerbates the growth defects of Cgulp2 even at conducive temperatures. This intriguing phenotype prompted us to investigate the connections between STUbLs and CgUlp2.
In the presence of CgUlp2, SUMO chains are removed from the target protein whereas, in its absence, polySUMOylated proteins are accumulated and targeted for protein degradation via STUbL, primarily CgUls1. This is a key contributor to the homeostasis of the target proteins. As more polySUMOylated protein accumulates in the absence of CgUlp2, in CgulsΔ, it is not degraded. The double mutant Cgulp2ΔCgslx8Δ exhibits an increased level of protein degradation possibly because of upregulated activity of CgUls1. Following MG132 treatment, degradation of proteins was prevented and SUMOylated proteins appeared in Cgulp2Δ and Cgulp2ΔCgslx8Δ mutants and so did the improved growth. Based on the experimental validation of quantitative mass spectrometry data, ubiquitin-proteosome pathway, purine nucleotide biosynthesis pathway, and mitochondrial function were found to be significantly impacted in Cgulp2ΔCgslx8Δ double mutant. This was followed by the Cgulp2Δ single mutant. We propose that these are the primary causes of the increased protein loss and slow growth of Cgulp2ΔCgslx8Δ cells at 30°C. The ubiquitin-proteosome pathway components were decreased whereas purine nucleotide biosynthesis pathway and mitochondrial proteins had increased in Cgulp2ΔCguls1Δ due to which restoration of growth and accumulation of polySUMOylated proteins was seen in this mutant compared to the Cgulp2Δ mutant.
When we examined the overall SUMO proteome of Cgulp2Δ, Cgslx8Δ, Cguls1Δ and the double mutants Cgulp2ΔCgslx8Δ and Cgulp2ΔCguls1Δ, there are very stark differences. Of particular importance is that the Cgulp2Δ cells have fewer SUMOylated proteins, especially the higher molecular weight proteins, which could be the polySUMOylated proteins. While this appears counter intuitive at first, as we conjectured earlier and demonstrate here, this is because the polySUMOylated proteins are targeted for degradation, leading to increased protein degradation in the absence of CgUlp2. When STUbLs were deleted, we expected to see restoration/accumulation of these polySUMOylated proteins. Interestingly, we find that this is partially restored in the Cgulp2ΔCguls1Δ but there is increased protein degradation in Cgulp2ΔCgslx8Δ double mutants (Fig 9). This suggests that CgUls1 could be a major STUbL in C. glabrata for the CgUlp2 targets. The observation that Cgulp2ΔCgslx8Δ is more susceptible to stress than the Cgulp2Δ or Cgslx8Δ single mutant suggested increased protein degradation in this mutant. We confirmed this by SUMO western blots, MG132 treatment and also uncovered that CgUls1 was upregulated in the absence of Cgslx8. However, this is not a global protein degradation as levels of several proteins in Cgulp2ΔCgslx8Δ are similar to wild type and some proteins, especially those involved in the metabolic processes, are even present at a higher level than wild type (S10 Fig). This indicates that targeting the SUMO-mediated protein homeostasis pathway will selectively incapcitate a few critical pathways of the pathogen.
The contrasting phenotypes of the two double mutants provided a handle to investigate the downstream pathways that are likely to be targets of these STUbLs. Our quantitative whole proteome analysis showed that several pathways were affected (Fig 6). Of note, we found that the components of protein degradation pathways including 23S proteasome and ERAD pathway were significantly higher in Cgulp2ΔCgslx8Δ and some of these were high in the Cgulp2Δ single mutant also. In contrast, these components were present at lower than wild type levels in the Cgulp2ΔCguls1Δ double mutant (S8 Fig). This could be the reason for the increased protein degradation observed in Cgulp2ΔCgslx8Δ and Cgulp2Δ, and also supports our observations with MG132 restoring growth to these mutants. Components of the cytosolic ribosome biogenesis are reduced in almost all mutants except for a few components that are higher in Cguls1Δ (Figs 6 and S7). This suggests that while there may be global reduction in translation when the SUMO pathway is perturbed, it is the increased levels of the protein degradation pathways in the Cgulp2ΔCgslx8Δ mutant that leads to reduced levels of proteins that were polySUMOylated.
We have also dissected the potential downstream pathways that are affected by altered proteostasis and establish that both purine biosynthesis and mitochondrial function are affected in the absence of CgSlx8. This correlates well with the phenotypes observed in the mutants. Metabolic processes especially nucleotide synthesis is critical for combating stresses like DNA damage. The Cgulp2ΔCgslx8Δ and Cgulp2Δ mutants are unable to effectively use non-fermentable carbon sources like lactate and glycerol. There is also increased ROS in these mutants even under unperturbed conditions (Fig 8B). This indicates poor mitochondrial health in these mutants. Mitochondria are important for fungal pathogenesis as the organism adapts to establish infection in mammalian cells [39,40]. The compromised mitochondrial function could be one of the factors for reduced capacity of these mutants to survive and multiply in macrophages.
Our studies confirm that SUMO-mediated ubiquitin ligases are crucial for protein homeostasis and even more so during infection. A significantly reduced survival in the macrophages by all the STUbL mutants, compared to Cgulp2 at 2 hours post infection is a clear indication of their inability to survive in the phagosomes. Probably, a reduced mitochondrial function, low pH inside the phagosome and compromised nucleotide synthesis together contribute to the inability to grow and survive by these mutants in the macrophages. Interestingly, mutants Cgulp2Δ and Cgulp2ΔCgslx8Δ grow well in YPD but not in RPMI media, which is closer to the mammalian cell milieu than YPD (S4 Fig). This shows that compromising the STUbL pathway may be more detrimental than estimated and testing YPD is not reflective of the requirements of the pathogen to establish a successful infection. Our macrophage infection studies clearly show that the STUbL mutants cannot survive and proliferate in macrophages. C. glabrata is one of the few pathogens that not only survives but also proliferates in macrophages; therefore, the inability of STUbL mutants to replicate in macrophages is a good indicator of the importance of STUbLs, and by extrapolation, SUMO mediated protein homeostasis, in establishing successful infection by Candida glabrata [41].
While several functions of the deSUMOylase and STUbLs are similar in C. glabrata compared to other model organisms studied, there are some key differences, even between C. glabrata and S. cerevisiae which are closely related. Ulp2 plays a critical role in regulating SUMO dynamics, genome stability, and protein homeostasis in many organisms. In C. glabrata, deletion of CgULP2 results in sensitivity to oxidative stress and DNA damage, as well as defects in virulence, adherence, and biofilm formation [25]. This mirrors the phenotype observed in S. cerevisiae, where ulp2Δ mutants display increased sensitivity to genotoxic agents such as methyl methanesulfonate (MMS) and hydroxyurea (HU) due to impaired DNA repair and protein homeostasis, and chromosomal instability [17,42,43]. Similarly, in other organisms including S. pombe, C. albicans and humans (SENP6 and SENP7), loss of Ulp2 leads to genome instability, reduced DNA repair and aneuploidy [44–47]. However, unlike S. cerevisiae, Cgulp2Δ doesnot accumulate higher molecular weight SUMO conjugates.
Previously, we identified SUMO-targeted ubiquitin ligases (STUbLs), including Slx5, Slx8, and Uls1, across various fungi, with varying degrees of conservation. Slx5 was found to be the least conserved, present only in the Saccharomycetes class of the Ascomycota phylum, whereas Slx8 and Uls1 showed broader conservation across fungi [30]. In S. cerevisiae, Slx5 and Slx8 form a heterodimeric complex that is critical for maintaining genome stability and deletion of SLX5 or SLX8 results in sensitivity to agents like MMS and HU [44,48–50]. However, in C. glabrata, deletion of these genes individually and in combination does not result in sensitivity to DNA damage-inducing agent and the individual deletions display distinct phenotypes. Cgslx5Δ are not temperature sensitive, and surprisingly, suppress the temperature sensitivity of Cgslx8Δ. In addition, the Cgslx5ΔCgslx8Δ strains are unable to grow on ethanol, and intriguingly, loss of Cgslx5 individually or in combination with Cgslx8Δ restores some of the higher molecular weight SUMO conjugates. Slx5 is not present in all organisms that contain Slx8, suggesting that Slx8 can independently carry out the STUbL activity [30]. Thus it appears that there is functional separation of Slx5 in C. glabrata (and perhaps in other fungi as well) and loss of Slx5 in other organisms. Uls1 plays a key role in regulating replication stress and maintaining rDNA stability in S. cerevisiae. Although uls1 mutant cells do not exhibit sensitivity to DNA damaging agents, they accumulate high molecular-weight SUMO conjugates [51–53]. In contrast, deletion of ULS1 in C. glabrata does not lead to any observable phenotypic response under stress conditions, nor does this strain accumulate high molecular-weight SUMO conjugates, indicating potential functional differences between the species.
A key difference between C. glabrata and S. cerevisiae that has come out of this study is the SUMO/STUbL/proteasome axis. In S. cerevisiae, Δulp2, Δslx5/slx8 and Δuls1 all accumulate polySUMO proteins [16,53]. While in many studies specific Ulp2 substrates have been identified that are ubiquitylated and degraded, in Δulp2 overall SUMO proteome is higher, especially, polySUMO proteome [20]. This suggests that they are not targeted to the proteasome for degradation effectively. However, in C. glabrata, in all these mutants, we see reduced higher molecular weight SUMO conjugated proteins and increased degradation of proteins. This suggests that, in C. glabrata, the SUMO-STUbL-proteasome axis is more tightly coupled and plays a key role in overall protein homoeostasis. Our mass spectrometry and MG132 experiments corroborate with this conclusion.
The study presented here suggests SUMO-meditated protein homeostasis pathway could be a good target for antifungal development. While STUbLs have not been studied in pathogenic fungi, the SUMOylation pathway is conserved across fungi. SUMO is either essential for survival (S. cerevisiae and C. glabrata) or for stress response and/or virulence (all other fungi studied) [26–30,47,54–61]. Understanding the SUMO-dependent protein homeostasis pathway in pathogenic fungi, will widen our understanding of fungal pathogenesis in general and also uncover new targets for antifungal therapies. Targeting this process provides a possibility of identifying several downstream targets as it strikes multiple pathways simultaneously. Our data also provide an overview of all potential pathways that are affected by perturbing the SUMO-mediated protein homeostasis in C. glabrata, offering an additional approach for designing combinatorial therapies. For instance, some of the purine biosynthesis inhibitors are already used as antifungals in the clinic [62–64]. If this can be combined with the inhibitors of Ulp2, they could augment the effect of these drugs. Therefore, further understanding the pathways that are regulated by SUMO-mediated protein homeostasis in establishment of successful infection would enable the design of better antifungals.
Methods
Bacterial and yeast culture conditions- The LB medium, which contains 1% tryptone, 0.5% yeast extract, and 1% NaCl, and the YPD medium, which contains 1% yeast extract, 2% peptone, and 2% dextrose, were used to routinely maintain the strains. C. glabrata cells were grown in either YPD or CAA medium (0.6% casamino acids, 0.67% yeast nitrogen base (YNB) without amino acids, and 2% dextrose) for uracil phototrophy. C. glabrata cells in the logarithmic phase were obtained by incubating overnight-grown cultures in fresh YPD/CAA media for roughly 4–6 hours at 30°C. Strains, and plasmids used in this study are listed in Table 1.
Gene disruption and cloning in Candida glabrata—In C. glabrata, ORFs were disrupted using homologous recombination with a nourseothricin resistance cassette. This cassette included the nourseothricin acetyltransferase (NAT1) gene, bordered by the 5’- and 3’-UTR regions of the target gene, each approximately 600 bp long, amplified from wild-type genomic DNA. The NAT1 gene was amplified from a plasmid and fused with parts of the UTRs, creating PCR products with overlapping 300–350 bp regions. These fused PCR products were co-transformed into the wild type strain, followed by plating on YPD medium. After 12–16 hours of incubation at 30°C to allow for homologous recombination, the cells were replica-plated onto YPD with 200 mg/ml nourseothricin and incubated for another 24 hours. Colonies that grew were screened via PCR to confirm gene disruption through homologous recombination [67].
The parental single-deletion strain was transformed with the FLP1 flippase plasmid and selecting transformants based on uracil prototrophy in the CAA medium in order to generate double-deletion strains. The NAT1 cassette was excised by the flippase enzyme at FRT (Flippase recognition target) sites flanking the NAT1 gene [68]. The transformants that were sensitive to nourseothricin were isolated and FLP1 flippase plasmid was removed after four to five passages in a nonselective YPD medium which was further verified in CAA plates. These cells were then transformed with a PCR-amplified linear segment from the genomic DNA of a single deletion strain and carried the sequence 5′-untranslated region (5′-UTR)-NAT1cassette-3′-untranslated region (3′-UTR). Colonies that were resistant to nourseothricin were isolated, and a PCR was used to check for gene disruption for double deletion strain [69].
To create complemented strains, the genes for CgSLX5 (1.5 kb), CgSLX8 (633bps), and CgULS1 (4.2 kb) were amplified from wild-type genomic DNA using the high-fidelity Vent polymerase. These genes were then inserted into the pGRB2.2 (CKM 364) plasmid at the BamHI-XhoI sites downstream of PGK1 promoter. Plasmids were sequenced for confirmation. For the complementation assay, the resulting recombinant plasmids were expressed in the appropriate deletion mutants.
Plate growth assay- C. glabrata wild type and deletion strains were grown in YPD/CAA medium at 30°C until reaching an O.D of 1.0 at A600. Then 10-fold serial dilutions were made in 96 well plates. 2–3μl of diluted cell suspensions were spotted onto appropriate plates and incubated for 2–4 days at the required temperatures to evaluate cell growth [25].
Antifungal susceptibility test- The C. glabrata wild type (WT) strain and mutants (Cgulp2Δ, Cgslx8Δ, Cguls1Δ, Cgulp2ΔCgslx8Δ, and Cgulp2ΔCguls1Δ) were cultured in YPD broth at 30°C. Overnight cultures were diluted to an OD600 of 0.1. Fluconazole, caspofungin, and amphotericin B were prepared at stock concentrations of 2 mg/ml, 5 μg/ml, and 250 μg/ml, respectively, using sterile distilled water. WT cells were treated with different concentrations of fluconazole (16, 32, and 64 μg/ml), caspofungin (75, 150, and 225 ng/ml), and amphotericin B (5, 10, and 20 μg/ml). Growth inhibition was assessed by measuring OD600 after 48 hours of drug exposure. The concentration at which 50% growth inhibition occurred (IC50) was determined for each antifungal agent and identified as 32 μg/ml for fluconazole, 150 ng/ml for caspofungin, and 10 μg/ml for amphotericin B. The mutants were exposed to 32 μg/ml fluconazole, 150 ng/ml caspofungin, and 10 μg/ml amphotericin B for 48 hours. OD600 was measured at 48 hours to assess the survival. All experiments were performed in triplicates. The survival was calculated relative to untreated control samples.
Western blot- Using trichloroacetic acid (TCA) precipitation, whole-cell extracts from the logarithmic phase (absorbance at A600-1) C. glabrata cultures were prepared. Protein extraction was conducted using equal number of cells as determined by absorbance. Protein was separated on SDS-PAGE was followed by transfer onto PVDF membrane. After blocking for 1 hour with 5% skim milk powder, the membrane was incubated with α-flag primary antibody (Sigma, 1:10,000) for 14–16 hours. Following washes with TBST, blots were incubated with mouse HRP-conjugated secondary antibody (Abcam, 1:20,000) for 1 hour. Blots were developed using G-Biosciences chemiluminal development solutions after washes with TBST. Blots were further reprobed with α-tubulin antibody as the loading control.
Preparation of THP-1 Macrophage monolayer- Human monocyte, THP-1 cells were seeded in a 25mm culture flask at 37°C, in a humidified incubator with 5% CO2, in RPMI-1640 culture medium, supplemented with 10% FBS, 2 mM glutamine, and antibiotic cocktail. Cells were allowed to grow until a monolayer was formed. Cell density was maintained at 2.5 x 105 cells/ml by counting with a hemocytometer and 1μl of 160mM PMA stock (phorbol 12-myristate 13-acetate solution prepared in DMSO) was added to 10ml THP-1 cell suspension (final concentration 16nM) to stop the cells from further division and to convert them into macrophages. After 12 hours, the old medium containing PMA was removed and a prewarmed fresh RPMI-1640 complete medium was added and allowed the cells to recover for another 12 hours at 37°C. The appearance of macrophage cells were checked under an inverted microscope for their flattened, spindle-shaped, and adherent morphology.
Infection of C. glabrata in THP-1 cells- C. glabrata strains were grown in liquid YPD media at 30°C with 180 rpm of agitation. A cell density of 2.5 x 106 cells/ml was used to infect the cells, maintaining a multiplicity of infection of 1:10. C. glabrata cells were co-cultured with THP-1 macrophages for 2 hours, followed by gently washing several times (3–6 times) using 1 ml of sterile PBS to ensure the complete elimination of the non-phagocytosed yeast pathogen. Cells were then incubated for 8, 12, and 24 hours to examine the C. glabrata cell survival andproliferation.
To quantify internalized yeast, infected macrophages were washed and lysed using 1ml of sterile water for 2–3 minutes in the culture plate. Cells were scraped softly and cell lysate was collected. The cells were then serially diluted to 100 fold and 100μl of the suspension was plated on a YPD agar followed by incubation at 30°C for 1–2 days. Colonies that appeared on YPD plates were counted manually, and CFU (colony forming units) was calculated.
Biofilm formation- Assay for biofilm was performed on flat-bottomed 24-well polystyrene plates. YPD-grown C. glabrata cells in the exponential phase were added to the plate and incubated at 37°C with mild rotation (75 rpm) for 90 minutes. Wells were then washed twice using PBS, followed by the addition of 1ml of RPMI-1640 complete media to each well. The plate was then incubated at 37°C with 75 rpm, for another 24 hours to promote biofilm formation. The next day, 500μl of used media was pipetted out and fresh 500μl RPMI containing 10% FBS was added. The plate was further incubated for 48 to 72 hours. During incubation, every day 500μl of fresh media was replaced to provide continuous nutrients for the growing biofilm. After 72 hours, the plate was washed with PBS to eliminate all the planktonic cells. Cells were then collected in PBS by scraping, pelleted down, and stored at -80°C for RNA isolation.
XTT assay- XTT assay was used to quantify biofilm formation in wild type and various deletion strains of C. glabrata. After 72 hours of biofilm formation, the plate was washed thrice with PBS to remove unattached cells. 100μl of 1x XTT solution was added, and the plate was incubated in the dark for one hour. Absorbance at 492nm was measured, RPMI-1640 complete medium serving as a negative control, and the absorbance of the Candida-grown wells was subtracted from this value [70,71].
Quantitative Real-Time PCR
C. glabrata cells were grown overnight in YPD at 30°C, it was sub-cultured till OD6001. Cells were treated with DNA damaging agent (0.04% MMS), and hydroxyurea (100mM) and allowed to grow for additional 2 hours. For temperature sensitivity analysis, C. glabrata cells were shifted to 42°C and incubated for 2 hours. The cultures were then pelleted down and stored at -80°C for RNA extraction. For RNA extraction, the Ribo Pure kit from Invitrogen was used, followed by cDNA conversion using the SuperScript III First-strand Synthesis System for RT-PCR (Invitrogen) with 1 μg RNA as a template. SYBR green master mix (Invitrogen) was employed for quantitative reverse transcriptase PCR. CgACT1 was used to normalize RNA levels and also served as endogenous control.
Tandem mass tag-based quantitative total proteomics analysis
Sample preparation- For proteome analysis, YPD grown C. glabrata cells in exponential phase were pelleted at 3000 rpm, washed once with MiliQ, and resuspended in a lysis buffer containing 50 mM HEPES pH 8.2, 8 M Urea, 50 mM NaCl, and protease inhibitors (complete mini, EDTA-free). Cells were lysed using the Beadbeater in microfuge tubes for six cycles of 30 s each, with 1 min pauses between cycles to avoid overheating of the lysates. Lysates were cleared by centrifugation at 15000 rpm for 30 min at 4°C. Protein concentration was then determined using the Lowry method and 200μg of proteins were run on a 10% SDS-PAGE to check the protein quality [72]. These proteins were then sent to Mass Spectrometry (MS) facilities, Institute of Bioinformatics, Bangalore, India. At the MS facilities, proteins were acetone precipitated, alkylated, and digested with trypsin.
For TMT labeling, the 12 channels from the TMT pro-16 plex kit (ThermoFisher Scientific) were used for labeling the respective samples. Each TMT channel reagent was dissolved in 200μl of anhydrous acetonitrile and incubated for 5 minutes. Then, 40 μl of respective TMT reagent was added to 80ug of each of the 12 sample peptide mix (dissolved in 40mM TEABC buffer), incubated for 60 mins at RT (after mixing) and the labeling reaction was stopped by adding 5% methylamine solution. An aliquot (5μl) of each of the labeled samples was quenched by adding 8 μl of 5% hydroxylamine, pooled, dried, reconstituted in 0.1% TFA, cleaned up using C18 stage tip, dried and re-dissolved in LCMS grade water (containing 0.1% TFA and 3% acetonitrile) for labeling efficiency check. After the C18 stage tip clean-up, the sample was dried and dissolved in water (containing 0.1% TFA and 3% acetonitrile) for LCMS analysis to check the labeling efficiency. The labeling efficiency (>95%) and reporter ion intensity normalization (median of TMT ratios for mixing equal amounts of TMT-labeled peptides) were performed. After the labelling efficiency check, reaction was quenched using 5% hydroxylamine (for 15 min, at RT). Based on the reporter ion intensity values, each of the labeled samples was pooled, dried, and fractionated using basic pH Reverse phase HPLC into 96 fractions and concatenated into 12 fractions. The final LCMS run was performed (in triplicates) after dissolving the (desalted) 12 fractions in water (containing 0.1% TFA and 3% acetonitrile) and loading 1μg equivalent peptide to the mass spectrometer [73,74].
The LCMS analysis was performed using Q Exactive HF-X (Thermo Scientific, Bremen, Germany) interfaced with Dionex Ultimate 3000 nanoflow liquid chromatography system. The acquired mass spectrometry data was analyzed against C. glabrata reference proteome (downloaded from the UniProt database) using the SEQUEST search algorithm on the Proteome Discoverer platform (version 2.4, Thermo Scientific). Data was filtered with a 0.01% false discovery rate (FDR) based on the decoy search.
Analysis of gene ontology
The corresponding C. glabrata systematic ORFs (CAGL IDs) were obtained from the UniProt Database (https://www.uniprot.org/) [75] using accession numbers of proteins that were determined by mass spectrometry analysis. Proteins were identified by using CAGL IDs from the Candida Genome Database (http://www.candidagenome.org/) [76]. Using the DAVID tools (https://david.ncifcrf.gov/tools.jsp) [77], functional enrichment analysis (biological processes, cellular components, and molecular functions) was carried out. Graphs were plotted by using GraphPad Prism and SRplot software.
Microscopy
To study the mitochondrial morphology and reactive oxygen species (mtROS) in wild type and deletion strains of C. glabrata, MitoTracker Green dye (Invitrogen, M7514) and MitoSOX Red (Invitrogen, M36008) were used respectively. Also, we used N-acetylcysteine (NAC), a known antioxidant to check the reduction in ROS level. Cells grown in YPD medium (1.0 A600) with or without 10mM NAC were harvested, and the pellet was washed thrice with 1x PBS to get rid of any remaining YPD medium [78]. After that, cells were treated with 200 nM MitoTracker Green and 1 μM MitoSOX Red in 250μl PBS and allowed to incubate for 20 minutes at 30°C. Following incubation, cells were rinsed with PBS and resuspended in PBS without MitoTracker Green and MitoSOX Red. Cells resuspended in PBS were placed on a 35mm cover glass bottom dish that had been coated with 0.1% solution of concanavalin A to facilitate imaging. Images were captured using a Leica TCS SP8 confocal microscope equipped with an HC PL APO CS2 63x/1.40 oil objective. The microscopy data was processed and analyzed using Fiji Software.
Supporting information
S1 Fig. Schematic representation of orthologs of the S. cerevisiae STUbL proteins Slx5, Slx8, Uls1 in Candida glabrata.
Maps of proteins with their domains were produced using IBS 1.0.
https://doi.org/10.1371/journal.ppat.1012742.s001
(PDF)
S2 Fig. Quantification of growth of STUbLs mutants.
The indicated strains were plated out on the specific treatment plates and the colonies that appeared were counted and normalized against counts obtained from untreated plates. Data represent the mean of biological triplicates and the error bars indicate standard deviation.
https://doi.org/10.1371/journal.ppat.1012742.s002
(TIF)
S3 Fig. Complementation of STUbL mutants.
To test growth under various stress conditions, a serial dilution-spotting assay was done. 3 μl of 10 fold serial dilutions of cultures at A600 1 of the indicated C. glabrata strains were spotted on different media. Images of paltes were taken after 2 days of incubation at 30°C and 37°C, 3 days of incubation with methylmethanesulfonate (MMS), and hydroxyurea (HU), at concentrations of 0.04%, 100mM, and at higher temperature (42°C), respectively.
https://doi.org/10.1371/journal.ppat.1012742.s003
(TIF)
S4 Fig. Growth of C. glabrata STUbL deletions in YPD and RPMI.
CFU to assess cell proliferation of the indicated C. glabrata strains in YPD (A, B) and RPMI-1640 medium containing 10% FBS (fetal bovine serum) (C, D). Yeast cells were collected at the indicated time points (2, 8, 12, and 24 hours), serially diluted, and then plated on YPD medium. Following 24–48 hours of incubation at 30°C, yeast colonies were counted. Data shows the mean and SD of three independent experiments.
https://doi.org/10.1371/journal.ppat.1012742.s004
(TIF)
S5 Fig. A. Complementation of Cguls1Δ mutant with plasmid encoding HF-CgUls1.
3μl of 10 fold serial dilutions of cultures at A600 1 of the indicated C. glabrata strains were spotted on CAA plates. Plate images were taken after 2–3 days of incubation. B. CgUls1p is elevated in STUbL deletion mutants. The bar graph illustrates the log2 fold change of CgUls1 protein levels in the mutants compared to the wild type as obtained in the mass spectrometry. Error bars representing the log2 standard deviation.
https://doi.org/10.1371/journal.ppat.1012742.s005
(TIF)
S6 Fig. Comparison of levels of proteins involved in various transport processes in STUbL deletion mutants.
Graph shows the change in levels of proteins involved in various transport pathways in the Cgulp2Δ, Cgulp2ΔCgslx8Δ, and Cgulp2ΔCguls1Δ mutants in comparison to wild type. The error bar represents the log2 standard deviation.
https://doi.org/10.1371/journal.ppat.1012742.s006
(TIF)
S7 Fig. Heat map illustrating the differences in protein levels of ribosomal proteins.
A. Heat map of mitochondrial ribosomal proteins and B. cytoplasmic ribosomal proteins of large and small subunits in Cgulp2Δ, Cgslx8Δ, Cguls1Δ single mutants, Cgulp2ΔCgslx8Δ and Cgulp2ΔCguls1Δ double mutants in comparison with WT. Heat maps were generated for the increased (shades of red color) and decreased (shades of green color) levels of proteins for each mutant with respect to wild type.
https://doi.org/10.1371/journal.ppat.1012742.s007
(TIF)
S8 Fig. Heat map illustrating the altered levels of the components of protein degradation pathways in Cgulp2Δ, Cgslx8Δ, Cguls1Δ single mutants, Cgulp2ΔCgslx8Δ and Cgulp2ΔCguls1Δ double mutant in comparison with WT.
A heat map of the elevated (shades of red color) and diminished (shades of green color) levels of the components of protein degradation pathways including the proteasome and ERAD pathway for each mutant.
https://doi.org/10.1371/journal.ppat.1012742.s008
(TIF)
S9 Fig. Comparison of levels of proteins involved in biosynthetic processes.
Heat map shows the changes in levels of proteins involved in biosynthetic processes in each mutant in comparison to wild type with increase indicated in red and decrease in green.
https://doi.org/10.1371/journal.ppat.1012742.s009
(TIF)
S10 Fig. Comparison of levels of proteins involved in metabolic processes.
Heat map shows the changes in levels of proteins involved in metabolic processes in each mutant in comparison to wild type with increase indicated in red and decrease in green.
https://doi.org/10.1371/journal.ppat.1012742.s010
(TIF)
S11 Fig.
A. Expression of FLAG tagged CgAde5,7. Western blot was performed to check the expression of CgAde5,7 protein in WT and mutants using α-flag antibody. The six lanes preceding the molecular weight marker are negative controls of WT and mutants with no tag and the lanes following the marker are strains carrying plasmid encoding FLAG tagged CgAde5,7. Tubulin is used as loading control. B. Effect of purine synthesis inhibitors on growth of C. glabrata. Doubling times were calculated based on absorbance A600 for 48 hours measured at regular intervals and are presented as a bar graph. Error bar represents the standard deviation.
https://doi.org/10.1371/journal.ppat.1012742.s011
(TIF)
S1 Table. Orthologs of the S. cerevisiae STUbL proteins Slx5, Slx8, Uls1 in Candida glabrata.
Protein sequences were collected from Saccharomyces Genome Database [37] and Blastp was used to determine C. glabrata orthologs. Domains were annotated using Pfam. The EMBOSS Stretcher (pairwise sequence alignment) tool was used to calculate the percentages of identity and similarity.
https://doi.org/10.1371/journal.ppat.1012742.s012
(TIF)
S2 Table. A list of 107 significantly elevated proteins and 90 significantly reduced proteins identified in Cgulp2Δ mutant in comparison with wild type.
https://doi.org/10.1371/journal.ppat.1012742.s013
(XLSX)
S3 Table. A list of 87 significantly elevated proteins and 130 significantly reduced proteins identified in Cgslx8Δ mutant in comparison with wild type.
https://doi.org/10.1371/journal.ppat.1012742.s014
(XLSX)
S4 Table. A list of 148 significantly elevated proteins and 241 significantly reduced proteins identified in Cguls1Δ mutant in comparison with wild type.
https://doi.org/10.1371/journal.ppat.1012742.s015
(XLSX)
S5 Table. A list of 278 significantly elevated proteins and 184 significantly reduced proteins identified in Cgulp2ΔCgslx8Δ mutant in comparison with wild type.
https://doi.org/10.1371/journal.ppat.1012742.s016
(XLSX)
S6 Table. A list of 300 significantly elevated proteins and 134 significantly reduced proteins identified in Cgulp2ΔCguls1Δ mutant in comparison with wild type.
https://doi.org/10.1371/journal.ppat.1012742.s017
(XLSX)
S7 Table. Enriched GO terms for biological process (BP), cellular component (CC) and molecular function (MF) categories, as determined by the DAVID tool, for the elevated and reduced proteins in Cgulp2Δ mutant.
https://doi.org/10.1371/journal.ppat.1012742.s018
(XLSX)
S8 Table. Enriched GO terms for biological process (BP), cellular component (CC) and molecular function (MF) categories, as determined by the DAVID tool, for the elevated and reduced proteins in Cgslx8Δ mutant.
https://doi.org/10.1371/journal.ppat.1012742.s019
(XLSX)
S9 Table. Enriched GO terms for biological process (BP), cellular component (CC) and molecular function (MF) categories, as determined by the DAVID tool, for the elevated and reduced proteins in Cguls1Δ mutant.
https://doi.org/10.1371/journal.ppat.1012742.s020
(XLSX)
S10 Table. Enriched GO terms for biological process (BP), cellular component (CC) and molecular function (MF) categories, as determined by the DAVID tool, for the elevated and reduced proteins in Cgulp2ΔCgslx8Δ mutant.
https://doi.org/10.1371/journal.ppat.1012742.s021
(XLSX)
S11 Table. Enriched GO terms for biological process (BP), cellular component (CC) and molecular function (MF) categories, as determined by the DAVID tool, for the upregulated and downregulated proteins in Cgulp2ΔCguls1Δ mutant.
https://doi.org/10.1371/journal.ppat.1012742.s022
(XLSX)
Acknowledgments
The authors thank Dr. Rupinder Kaur, CDFD, for plasmids and strains. We thank Institute of Bioinformatics, Bangalore for quantitative proteomics. Authors acknowledge infrastructure support from DST-FIST SR/FST/LS-11/2023/1172C and DBT-SAHAJ-Builder (BT/INF/22/SP41176/2020).
References
- 1. Bongomin F, Gago S, Oladele R, Denning D. Global and Multi-National Prevalence of Fungal Diseases—Estimate Precision. J Fungi. 2017 Oct 18;3(4):57. pmid:29371573
- 2. Parums DV. Editorial: The World Health Organization (WHO) Fungal Priority Pathogens List in Response to Emerging Fungal Pathogens During the COVID-19 Pandemic. Med Sci Monit Int Med J Exp Clin Res. 2022 Dec 1;28:e939088. pmid:36453055
- 3. Arendrup MC, Patterson TF. Multidrug-Resistant Candida: Epidemiology, Molecular Mechanisms, and Treatment. J Infect Dis. 2017 Aug 15;216(suppl_3):S445–51. pmid:28911043
- 4. Ademe M, Girma F. Candida auris: From Multidrug Resistance to Pan-Resistant Strains. Infect Drug Resist. 2020;13:1287–94. pmid:32440165
- 5. McDermott A. Drug-resistant fungi on the rise. Proc Natl Acad Sci. 2022 Nov 29;119(48):e2217948119. pmid:36413501
- 6. Spivak ES, Hanson KE. Candida auris: an Emerging Fungal Pathogen. Kraft CS, editor. J Clin Microbiol. 2018 Feb;56(2):e01588–17.
- 7. Pagano L, Caira M, Candoni A, Offidani M, Fianchi L, Martino B, et al. The epidemiology of fungal infections in patients with hematologic malignancies: the SEIFEM-2004 study. Haematologica. 2006 Aug;91(8):1068–75. pmid:16885047
- 8. Retanal C, Ball B, Geddes-McAlister J. Post-Translational Modifications Drive Success and Failure of Fungal–Host Interactions. J Fungi. 2021 Feb 9;7(2):124. pmid:33572187
- 9. Flotho A, Melchior F. Sumoylation: a regulatory protein modification in health and disease. Annu Rev Biochem. 2013;82:357–85. pmid:23746258
- 10. Gill G. SUMO and ubiquitin in the nucleus: different functions, similar mechanisms? Genes Dev. 2004 Sep 1;18(17):2046–59. pmid:15342487
- 11. Johnson ES, Schwienhorst I, Dohmen RJ, Blobel G. The ubiquitin-like protein Smt3p is activated for conjugation to other proteins by an Aos1p/Uba2p heterodimer. EMBO J. 1997 Sep 15;16(18):5509–19. pmid:9312010
- 12. Johnson ES, Blobel G. Ubc9p is the conjugating enzyme for the ubiquitin-like protein Smt3p. J Biol Chem. 1997 Oct 24;272(43):26799–802. pmid:9341106
- 13. Geiss-Friedlander R, Melchior F. Concepts in sumoylation: a decade on. Nat Rev Mol Cell Biol. 2007 Dec;8(12):947–56. pmid:18000527
- 14. Hochstrasser M. SP-RING for SUMO: new functions bloom for a ubiquitin-like protein. Cell. 2001 Oct 5;107(1):5–8. pmid:11595179
- 15. Zheng N, Shabek N. Ubiquitin Ligases: Structure, Function, and Regulation. Annu Rev Biochem. 2017 Jun 20;86:129–57. pmid:28375744
- 16. Bylebyl GR, Belichenko I, Johnson ES. The SUMO Isopeptidase Ulp2 Prevents Accumulation of SUMO Chains in Yeast. J Biol Chem. 2003 Nov;278(45):44113–20. pmid:12941945
- 17. Li SJ, Hochstrasser M. The Yeast ULP2 (SMT4) Gene Encodes a Novel Protease Specific for the Ubiquitin-Like Smt3 Protein. Mol Cell Biol. 2000 Apr 1;20(7):2367–77.
- 18. Mukhopadhyay D, Dasso M. Modification in reverse: the SUMO proteases. Trends Biochem Sci. 2007 Jun;32(6):286–95. pmid:17499995
- 19. Psakhye I, Branzei D. SMC complexes are guarded by the SUMO protease Ulp2 against SUMO-chain-mediated turnover. Cell Rep. 2021 Aug;36(5):109485. pmid:34348159
- 20. Chang YC, Oram MK, Bielinsky AK. SUMO-Targeted Ubiquitin Ligases and Their Functions in Maintaining Genome Stability. Int J Mol Sci. 2021 May 20;22(10):5391. pmid:34065507
- 21. D’Ambrosio LM, Lavoie BD. Pds5 Prevents the PolySUMO-Dependent Separation of Sister Chromatids. Curr Biol. 2014 Feb;24(4):361–71. pmid:24485833
- 22. Joseph J, Tan SH, Karpova TS, McNally JG, Dasso M. SUMO-1 targets RanGAP1 to kinetochores and mitotic spindles. J Cell Biol. 2002 Feb 18;156(4):595–602. pmid:11854305
- 23. Wagner K, Kunz K, Piller T, Tascher G, Hölper S, Stehmeier P, et al. The SUMO Isopeptidase SENP6 Functions as a Rheostat of Chromatin Residency in Genome Maintenance and Chromosome Dynamics. Cell Rep. 2019 Oct 8;29(2):480–494.e5. pmid:31597105
- 24. Yin Y, Seifert A, Chua JS, Maure JF, Golebiowski F, Hay RT. SUMO-targeted ubiquitin E3 ligase RNF4 is required for the response of human cells to DNA damage. Genes Dev. 2012 Jun 1;26(11):1196–208. pmid:22661230
- 25. Gujjula R, Veeraiah S, Kumar K, Thakur SS, Mishra K, Kaur R. Identification of Components of the SUMOylation Machinery in Candida glabrata: ROLE OF THE DESUMOYLATION PEPTIDASE CgUlp2 IN VIRULENCE. J Biol Chem. 2016 Sep 9;291(37):19573–89. pmid:27382059
- 26. Harting R, Bayram Ö, Laubinger K, Valerius O, Braus GH. Interplay of the fungal sumoylation network for control of multicellular development. Mol Microbiol. 2013 Dec;90(5):1125–45. pmid:24279728
- 27. Leach MD, Stead DA, Argo E, Brown AJP. Identification of sumoylation targets, combined with inactivation of SMT3, reveals the impact of sumoylation upon growth, morphology, and stress resistance in the pathogen Candida albicans. Mol Biol Cell. 2011 Mar 1;22(5):687–702. pmid:21209325
- 28. Lim YJ, Kim KT, Lee YH. SUMOylation is required for fungal development and pathogenicity in the rice blast fungus Magnaporthe oryzae. Mol Plant Pathol. 2018 Sep;19(9):2134–48. pmid:29633464
- 29. Wong KH, Todd RB, Oakley BR, Oakley CE, Hynes MJ, Davis MA. Sumoylation in Aspergillus nidulans: sumO inactivation, overexpression and live-cell imaging. Fungal Genet Biol FG B. 2008 May;45(5):728–37. pmid:18262811
- 30. Gupta D, Garapati HS, Kakumanu AVS, Shukla R, Mishra K. SUMOylation in fungi: A potential target for intervention. Comput Struct Biotechnol J. 2020;18:3484–93. pmid:33294142
- 31. Hickey CM, Wilson NR, Hochstrasser M. Function and regulation of SUMO proteases. Nat Rev Mol Cell Biol. 2012 Dec;13(12):755–66. pmid:23175280
- 32. Cao C, Cao Z, Yu P, Zhao Y. Genome-wide identification for genes involved in sodium dodecyl sulfate toxicity in Saccharomyces cerevisiae. BMC Microbiol. 2020 Feb 17;20(1):34. pmid:32066383
- 33. Harriott MM, Lilly EA, Rodriguez TE, Fidel PL, Noverr MC. Candida albicans forms biofilms on the vaginal mucosa. Microbiology. 2010 Dec 1;156(12):3635–44. pmid:20705667
- 34. Sriramachandran AM, Dohmen RJ. SUMO-targeted ubiquitin ligases. Biochim Biophys Acta BBA—Mol Cell Res. 2014 Jan;1843(1):75–85. pmid:24018209
- 35. Chandel NS. Nucleotide Metabolism. Cold Spring Harb Perspect Biol. 2021 Jul;13(7):a040592. pmid:34210662
- 36. Pedley AM, Benkovic SJ. A New View into the Regulation of Purine Metabolism: The Purinosome. Trends Biochem Sci. 2017 Feb;42(2):141–54. pmid:28029518
- 37. Guo Y, Guan T, Shafiq K, Yu Q, Jiao X, Na D, et al. Mitochondrial dysfunction in aging. Ageing Res Rev. 2023 Jul;88:101955. pmid:37196864
- 38. Srivastava S. The Mitochondrial Basis of Aging and Age-Related Disorders. Genes. 2017 Dec 19;8(12):398. pmid:29257072
- 39. Black B, Lee C, Horianopoulos LC, Jung WH, Kronstad JW. Respiring to infect: Emerging links between mitochondria, the electron transport chain, and fungal pathogenesis. PLoS Pathog. 2021 Jul;17(7):e1009661. pmid:34237096
- 40. Verma S, Shakya VPS, Idnurm A. Exploring and exploiting the connection between mitochondria and the virulence of human pathogenic fungi. Virulence. 2018 Jan 1;9(1):426–46. pmid:29261004
- 41. Kasper L, Seider K, Hube B. Intracellular survival of Candida glabrata in macrophages: immune evasion and persistence. Bolotin-Fukuhara M, editor. FEMS Yeast Res. 2015 Aug;15(5):fov042.
- 42. Suhandynata RT, Quan Y, Yang Y, Yuan WT, Albuquerque CP, Zhou H. Recruitment of the Ulp2 protease to the inner kinetochore prevents its hyper-sumoylation to ensure accurate chromosome segregation. Cheeseman I, editor. Genet PLOS. 2019 Nov 20;15(11):e1008477.
- 43. Ryu HY, Wilson NR, Mehta S, Hwang SS, Hochstrasser M. Loss of the SUMO protease Ulp2 triggers a specific multichromosome aneuploidy. Genes Dev. 2016 Aug 15;30(16):1881–94. pmid:27585592
- 44. Rizzo M, Soisangwan N, Vega-Estevez S, Price RJ, Uyl C, Iracane E, et al. Stress combined with loss of the Candida albicans SUMO protease Ulp2 triggers selection of aneuploidy via a two-step process. Moran GP, editor. PLOS Genet. 2022 Dec 27;18(12):e1010576. pmid:36574460
- 45. Pan X, Lei B, Zhou N, Feng B, Yao W, Zhao X, et al. Identification of novel genes involved in DNA damage response by screening a genome-wide Schizosaccharomyces pombe deletion library. BMC Genomics. 2012 Dec;13(1):662. pmid:23173672
- 46. Schonbrun M, Kolesnikov M, Kupiec M, Weisman R. TORC2 is required to maintain genome stability during S phase in fission yeast. J Biol Chem. 2013 Jul 5;288(27):19649–60. pmid:23703609
- 47. Claessens LA, Verlaan-de Vries M, de Graaf IJ, Vertegaal ACO. SENP6 regulates localization and nuclear condensation of DNA damage response proteins by group deSUMOylation. Nat Commun. 2023 Sep 22;14(1):5893. pmid:37735495
- 48. Thu YM, Van Riper SK, Higgins L, Zhang T, Becker JR, Markowski TW, et al. Slx5/Slx8 Promotes Replication Stress Tolerance by Facilitating Mitotic Progression. Cell Rep. 2016 May;15(6):1254–65. pmid:27134171
- 49. Burgess RC, Rahman S, Lisby M, Rothstein R, Zhao X. The Slx5-Slx8 Complex Affects Sumoylation of DNA Repair Proteins and Negatively Regulates Recombination. Mol Cell Biol. 2007 Sep 1;27(17):6153–62. pmid:17591698
- 50. Zhang C, Roberts TM, Yang J, Desai R, Brown GW. Suppression of genomic instability by SLX5 and SLX8 in Saccharomyces cerevisiae. DNA Repair. 2006 Mar;5(3):336–46. pmid:16325482
- 51. Cal-Bakowska M, Litwin I, Bocer T, Wysocki R, Dziadkowiec D. The Swi2-Snf2-like protein Uls1 is involved in replication stress response. Nucleic Acids Res. 2011 Nov 1;39(20):8765–77. pmid:21764775
- 52. Kramarz K, Litwin I, Cal-Bąkowska M, Szakal B, Branzei D, Wysocki R, et al. Swi2/Snf2-like protein Uls1 functions in the Sgs1-dependent pathway of maintenance of rDNA stability and alleviation of replication stress. DNA Repair. 2014 Sep;21:24–35. pmid:25091157
- 53. Uzunova K, Göttsche K, Miteva M, Weisshaar SR, Glanemann C, Schnellhardt M, et al. Ubiquitin-dependent Proteolytic Control of SUMO Conjugates. J Biol Chem. 2007 Nov;282(47):34167–75. pmid:17728242
- 54. Sahu MS, Patra S, Kumar K, Kaur R. SUMOylation in Human Pathogenic Fungi: Role in Physiology and Virulence. J Fungi Basel Switz. 2020 Mar 4;6(1):32.
- 55. Islam A, Tebbji F, Mallick J, Regan H, Dumeaux V, Omran RP, et al. Mms21: A Putative SUMO E3 Ligase in Candida albicans That Negatively Regulates Invasiveness and Filamentation, and Is Required for the Genotoxic and Cellular Stress Response. Genetics. 2019 Feb;211(2):579–95. pmid:30530734
- 56. Yan M, Nie X, Wang H, Gao N, Liu H, Chen J. SUMOylation of W or1 by a novel SUMO E 3 ligase controls cell fate in C andida albicans. Mol Microbiol. 2015 Oct;98(1):69–89.
- 57. Szewczyk E, Chiang YM, Oakley CE, Davidson AD, Wang CCC, Oakley BR. Identification and characterization of the asperthecin gene cluster of Aspergillus nidulans. Appl Environ Microbiol. 2008 Dec;74(24):7607–12. pmid:18978088
- 58. Nie X, Yu S, Qiu M, Wang X, Wang Y, Bai Y, et al. Aspergillus flavus SUMO Contributes to Fungal Virulence and Toxin Attributes. J Agric Food Chem. 2016 Sep 7;64(35):6772–82.
- 59. Aslanyan L, Sanchez DA, Valdebenito S, Eugenin EA, Ramos RL, Martinez LR. The Crucial Role of Biofilms in Cryptococcus neoformans Survival within Macrophages and Colonization of the Central Nervous System. J Fungi Basel Switz. 2017 Feb 24;3(1):10.
- 60. Mayer FL, Sánchez-León E, Kronstad JW. A chemical genetic screen reveals a role for proteostasis in capsule and biofilm formation by Cryptococcus neoformans. Microb Cell Graz Austria. 2018 Oct 31;5(11):495–510. pmid:30483521
- 61. Kim S, Park SY, Kim KS, Rho HS, Chi MH, Choi J, et al. Homeobox transcription factors are required for conidiation and appressorium development in the rice blast fungus Magnaporthe oryzae. PLoS Genet. 2009 Dec;5(12):e1000757. pmid:19997500
- 62. Rodriguez-Suarez R, Xu D, Veillette K, Davison J, Sillaots S, Kauffman S, et al. Mechanism-of-Action Determination of GMP Synthase Inhibitors and Target Validation in Candida albicans and Aspergillus fumigatus. Chem Biol. 2007 Oct;14(10):1163–75. pmid:17961828
- 63. Morrow CA, Valkov E, Stamp A, Chow EWL, Lee IR, Wronski A, et al. De novo GTP biosynthesis is critical for virulence of the fungal pathogen Cryptococcus neoformans. PLoS Pathog. 2012;8(10):e1002957. pmid:23071437
- 64. Köhler GA, Gong X, Bentink S, Theiss S, Pagani GM, Agabian N, et al. The Functional Basis of Mycophenolic Acid Resistance in Candida albicans IMP Dehydrogenase. J Biol Chem. 2005 Mar;280(12):11295–302. pmid:15665003
- 65. Battu A, Purushotham R, Dey P, Vamshi SS, Kaur R. An aspartyl protease-mediated cleavage regulates structure and function of a flavodoxin-like protein and aids oxidative stress survival. Edgerton M, editor. PLOS Pathog. 2021 Feb 25;17(2):e1009355. pmid:33630938
- 66. Srivastava VK, Suneetha KJ, Kaur R. The mitogen-activated protein kinase CgHog1 is required for iron homeostasis, adherence and virulence in Candida glabrata. FEBS J. 2015 Jun;282(11):2142–66.
- 67. Borah S, Shivarathri R, Kaur R. The Rho1 GTPase-activating protein CgBem2 is required for survival of azole stress in Candida glabrata. J Biol Chem. 2011 Sep 30;286(39):34311–24. pmid:21832071
- 68. Beese-Sims SE, Pan SJ, Lee J, Hwang-Wong E, Cormack BP, Levin DE. Mutants in the Candida glabrata glycerol channels are sensitized to cell wall stress. Eukaryot Cell. 2012 Dec;11(12):1512–9. pmid:23087370
- 69. Choudhary DK, Bhakt P, Kaur R. Essential Role for the Phosphatidylinositol 3,5-Bisphosphate Synthesis Complex in Caspofungin Tolerance and Virulence in Candida glabrata. Antimicrob Agents Chemother. 2019 Aug;63(8):e00886–19. pmid:31138567
- 70. Dhale RP. Comparison of Various Methods Used to Detect Biofilm Production of Candida Species. J Clin Diagn Res [Internet]. 2014 [cited 2024 Jul 8]; http://jcdr.net/article_fulltext.asp?issn=0973-709x&year=2014&volume=8&issue=11&page=DC18&issn=0973-709x&id=5147 pmid:25584219
- 71. Nett JE, Cain MT, Crawford K, Andes DR. Optimizing a Candida biofilm microtiter plate model for measurement of antifungal susceptibility by tetrazolium salt assay. J Clin Microbiol. 2011 Apr;49(4):1426–33. pmid:21227984
- 72. Isasa M, Rose CM, Elsasser S, Navarrete-Perea J, Paulo JA, Finley DJ, et al. Multiplexed, Proteome-Wide Protein Expression Profiling: Yeast Deubiquitylating Enzyme Knockout Strains. J Proteome Res. 2015 Dec 4;14(12):5306–17. pmid:26503604
- 73. Dey KK, Sun H, Wang Z, Niu M, Wang H, Jiao Y, et al. Proteomic Profiling of Cerebrospinal Fluid by 16-Plex TMT-Based Mass Spectrometry. Methods Mol Biol Clifton NJ. 2022;2420:21–37. pmid:34905163
- 74. Li J, Van Vranken JG, Pontano Vaites L, Schweppe DK, Huttlin EL, Etienne C, et al. TMTpro reagents: a set of isobaric labeling mass tags enables simultaneous proteome-wide measurements across 16 samples. Nat Methods. 2020 Apr;17(4):399–404. pmid:32203386
- 75. The UniProt Consortium, Bateman A, Martin MJ, Orchard S, Magrane M, Ahmad S, et al. UniProt: the Universal Protein Knowledgebase in 2023. Nucleic Acids Res. 2023 Jan 6;51(D1):D523–31. pmid:36408920
- 76. Skrzypek MS, Binkley J, Binkley G, Miyasato SR, Simison M, Sherlock G. The Candida Genome Database (CGD): incorporation of Assembly 22, systematic identifiers and visualization of high throughput sequencing data. Nucleic Acids Res. 2017 Jan 4;45(D1):D592–6.
- 77. Dennis G, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, et al. DAVID: Database for Annotation, Visualization, and Integrated Discovery. Genome Biol. 2003 May;4(5):P3.
- 78. Okamoto M, Nakano K, Takahashi-Nakaguchi A, Sasamoto K, Yamaguchi M, Teixeira MC, et al. In Candida glabrata, ERMES Component GEM1 Controls Mitochondrial Morphology, mtROS, and Drug Efflux Pump Expression, Resulting in Azole Susceptibility. J Fungi Basel Switz. 2023 Feb 10;9(2):240. pmid:36836353