The Neurospora RNA polymerase II kinase CTK negatively regulates catalase expression in a chromatin context-dependent manner

Summary

Clearance and adaptation to reactive oxygen species (ROS) are crucial for cell survival. As in other eukaryotes, the Neurospora catalases are the main enzymes responsible for ROS clearance and their expression are tightly regulated by the growth and environmental conditions. The RNA polymerase II carboxyl terminal domain (RNAPII CTD) kinase complex (CTK complex) is known as a positive elongation factor for many inducible genes by releasing paused RNAPII near the transcription start site and promoting transcription elongation. However, here we show that deletion of CTK complex components in Neurospora led to high CAT-3 expression level and resistance to H₂O₂₋induced ROS stress. The catalytic activity of CTK-1 is required for such a response. On the other hand, CTK-1 overexpression led to decreased expression of CAT-3. ChIP assays shows that CTK-1 phosphorylates the RNAPII CTD at Ser2 residues in the cat-3 ORF region during transcription elongation and deletion of CTK-1 led to dramatic decreases of SET-2 recruitment and H3K36me3 modification. As a result, histones at the cat-3 locus become hyperacetylated to promote its transcription. Together, these results demonstrate that the CTK complex is negative regulator of cat-3 expression by affecting its chromatin structure.

Introduction

Molecular oxygen (O₂), as a premiere biological electron acceptor, plays an essential role in fundamental cellular functions of aerobic organisms (Scandalios, 2005). It is beneficial to the cellular aerobic metabolism and provides much higher energy yield compared with anaerobic respiration (Scandalios, 1997). However, reactive oxygen species (ROS) are generated during this physiological processes, such as the superoxide radical (O₂^•−), hydroxyl radical (OH^•), hydrogen peroxide (H₂O₂) and sometimes reactive singlet form of oxygen (¹O₂) (Ames et al., 1993; Scandalios, 2002). These intermediates can cause harm to the macromolecules in cells, such as extensive oxidation damage to DNA, unsaturated fatty acids, and protein amino acid residues (Scandalios, 2005; Schumacker, 2006; Waris and Ahsan, 2006; Glorieux et al., 2015). Furthermore, the oxidative stress induced by excessive ROS also acts as a mediator of apoptosis and causes various chronic conditions (Buttke and Sandstrom, 1994; Waris and Ahsan, 2006).

To relieve oxidative stress, organisms developed clearance systems which convert harmful ROS into harmless constituents. Three main types of enzymes are involved in the ROS clearance systems, superoxide dismutases, catalases (CAT) and peroxidases (McCord and Fridovich, 1969; Fridovich, 1995; Scandalios, 1997, 2002). These clearance mechanisms are highly conserved from bacteria to mammals (Zamocky et al., 2008). In the filamentous fungus Neurospora crassa, there are four catalase genes, cat-1, cat-2, cat-3 and cct-1/cat-4 (Chary and Natvig, 1989; Borkovich et al., 2004; Schliebs et al., 2006; Yamashita et al., 2008). CAT-1 and CAT-3 are two large monofunctional CAT, which mainly exert their catalase activities on the conidia and mycelia respectively (Hansberg et al., 1993; Michán et al., 2002). CAT-3 is expressed during late exponential growth and is rigorously regulated by the growth conditions and environment stimulations. The cat-3 gene is located in the right arms of chromosomes III (Chary and Natvig, 1989). Our previous study showed that 5-kb upstream region of cat-3 gene formed a heterochromatic domain which plays a repressive role in cat-3 expression (Wang et al., 2016). Furthermore, CPC1, the homologue of Saccharomyces cerevisiae GCN4, is an activator of cat-3 transcription, via directly binding to the cat-3 locus in response to developmental and oxidative stress signals (Dong et al., 2018; Qi et al., 2018). These results suggested that chromatin structure and histone modifications play major roles in regulating the inducible expression of the cat-3 gene.

A key element of transcription regulation by RNA polymerase II is the modification on Y₁S₂P₃T₄S₅P₆S₇ repeats of the carboxyl terminal domain (CTD) of Rpb1, the largest subunit of RNAPII (Eick and Geyer, 2013; Stump and Ostrozhynska, 2013; Bowman and Kelly, 2014). The serine 5 and serine 2 phosphorylation (Ser5-P and Ser2-P) of the CTD are the best studied and the most conserved marks of transcription in eukaryotes. The Ser5-P is stimulated by Kin28/CDK7 (Kin28 in S. cerevisiae and CDK7 in mammals) (Srivastava and Ahn, 2015). There are two Ser2-P kinases, Bur1/CDK9 (Bur1 in S. cerevisiae and CDK9 in mammals) and Ctk1/CDK12 (Ctk1 in S. cerevisiae and CDK12 in mammals) (Bowman and Kelly, 2014). During transcriptional initiation, general transcription factors and a mediator complex are recruited to the promoter of genes by transcription activators; these events are responsible for RNAPII recruitment (Thomas and Chiang, 2006; Srivastava and Ahn, 2015). Furthermore, Ser5 is phosphorylated by Kin 28/CDK7, which stimulates RNAPII for promoter clearance and functions in pre-mRNA 5^′-capping and initiation–elongation transition (Rodriguez et al., 2000; Sogaard and Svejstrup, 2007; Viladevall et al., 2009; Srivastava and Ahn, 2015). Afterwards, the Bur1 kinase and its cyclin Bur2 (S. cerevisiae) could be recruited to the CTD of RNAPII and stimulate the phosphorylation of Ser2 of CTD at the 5^′ end of a gene. In S. cerevisiae, the CTK complex consists of a catalytic subunit Ctk1, a cyclin subunit Ctk2 and a co-cyclin factor Ctk3, and is proposed to interact with RNAPII through the Bur1-phosphorylated CTD to phosphorylate Ser2 of CTD at the 3^′ end of a gene (Ahn et al., 2004; Qiu et al., 2009; Bowman and Kelly, 2014; Srivastava and Ahn, 2015). In mammals, the catalytic subunit is CDK12 and the cyclin subunit is CCNK, which is proposed to be the central element in releasing paused RNAPII and moving RNAPII into productive elongation phase (Bowman and Kelly, 2014). Besides, Ctk1 acts as a positive elongation factor, which promotes the transcription elongation. The Ctk1-mediated Ser2 phosphorylation is necessary for efficient recruitment of the histone H3K36 methyltransferase Set2 to the CTD of RNAPII (Xiao et al., 2003).

In this study, we characterized the conserved kinase, CTK complex, which phosphorylates Ser2 of RNAPII CTD to regulate cat-3 expression. In Neurospora crassa, CTK complex consists of three subunits: CTK-1 (NCU06685), CTK-2 (NCU04495) and CTK-3 (NCU06470) (Sterner et al., 1995; Hautbergue and Goguel, 2001). We revealed that the CTK complex negatively regulates the expression of cat-3 and cat-1 genes. Deletion of ctk-1 resulted in decreased H3K36me3 levels and increased H3ac and H4ac levels at the cat-3 gene locus, indicating that the CTK complex participates in the transcription elongation of cat-3.

Results

CTK mutants are resistant to H₂O₂-induced ROS stress

To identify the factors that contribute to resistance to H₂O₂-induced ROS stress, we performed H₂O₂ sensitivity assays to screen the Neurospora knock-out mutants. We found that the ctk-1^KO strain was resistant to H₂O₂−induced ROS stress in constant light condition on plate (Fig. 1A and B). CTK-1 is the catalytic subunit of the RNA Pol II CTD kinase, which has two other subunits, cyclin CTK-2 and co-cyclin factor CTK-3. Similar results were obtained in the ctk-2^KO and ctk-3^KO mutants (Fig. 1A and B). Since CAT-3 is the major catalase in mycelia, these genetic results suggest that the CTK complex may participate in the regulation of cat-3 expression in N. crassa. To test this possibility, we performed western blot assay to analyse the protein levels of CAT-3 in wild-type (WT), ctk-1^KO, ctk-2^KO and ctk-3^KO strains. As shown in Fig. 1C and D, the protein levels of CAT-3 in the ctk deletion strains were increased compared with those in the WT strain with or without H₂O₂ treatment. To test whether ctk genes deletion affects the activity of CAT-3, we examined the catalase zymograms in the WT and ctk deletion strains treated with or without H₂O₂ by an in-gel assay. Proteins separated by PAGE under nondenatured conditions exhibited different bands of catalase activity. As shown in Fig. 1E, the stained bands corresponding to CAT-3 activity were brighter in the ctk mutants compared with those in the WT with or without H₂O₂ treatment. In addition, CAT-1 activities were also increased in these mutants (Fig. 1E), suggesting that the CTK complex can negatively regulate the expression of cat-1 gene in mycelia. Western blot analysis showed that the protein levels of CAT-1 in ctk mutant strains were increased compared with those in the WT with or without H₂O₂ treatment (Fig. 1F). Consistent with the protein and catalase activity results, the mRNA levels of cat-3 and cat-1 in the ctk mutants were also higher than those in the WT (Fig. 1G and H). These results suggest that the elevated expression of cat-3 and cat-1 in the ctk mutants are responsible for their resistance to H₂O₂-induced ROS stress.

Fig. 1.

The CTK complex subunits mutants are resistant to H₂O₂-induced oxidative stress and have high levels of catalases.

A. Relative values of mycelial growth rate with 0, 5, 10 or 20 mM H₂O₂ as indicated. Culturing was performed in 9 cm diameter petri dishes at 25C under constant light. The relative rate of growth was determined.

B. Relative values were calculated using the values obtained for controls of WT, ctk-1^KO, ctk-2^KO and ctk-3^KO mutants in (A).

C. Western blot analysis showing the levels of CAT-3 protein in the WT, ctk-1^KO, ctk-2^KO and ctk-3^KO mutants with or without H₂O₂. The membranes stained by Coomassie blue represent the total protein in each sample and act as loading control for western blot.

D. Quantification of CAT-3 protein expression levels in (C).

E. Catalase activity assay. Crude extracts from the WT, ctk-1^KO, ctk-2^KO and ctk-3^KO mutants with or without H₂O₂ were subjected to Native-PAGE, and the catalase activity was determined by the in-gel assay.

F. Western blot analysis showing the levels of CAT-1 protein in the WT, ctk-1^KO, ctk-2^KO and ctk-3^KO mutants with or without H₂O₂. The membranes stained by Coomassie blue represent the total protein in each sample and act as loading control for western blot.

G. RT-qPCR analysis showing the levels of cat-3 mRNA in the WT, ctk-1^KO, ctk-2^KO and ctk-3^KO mutants without H₂O₂.

H. RT-qPCR analysis showing the levels of cat-1 mRNA in the WT, ctk-1^KO, ctk-2^KO and ctk-3^KO mutants without H₂O₂. Error bars indicate SD (n = 3). *p < 0.05; ^**p < 0.01; ^***p < 0.001. Student’s t test was used. [Color figure can be viewed at wileyonlinelibrary.com]

CTK complex negatively regulates the expression of cat3 and cat-1 genes

To further confirm the effect of CTK on the catalase levels, three constructs carrying the sequences encoding Myc-tagged CTK-1, CTK-2 or CTK-3 driven by the quinic acid (QA)-inducible promoter were transformed into each corresponding ctk^KO strain. The expression of Myc-tagged CTK-1, CTK-2 or CTK-3 induced by QA restored the sensitivity to H₂O₂ of each CTK mutant to those of WT on plate assays (Fig. 2A and B), indicating that the observed phenotype of each mutant was due to the deletion of CTK complex subunits. The leaky expression of Myc-tagged-CTK-1 could be detected in the medium without QA in the complementary strains (Fig. 2C). In 2% glucose medium without QA, ectopic expression of Myc-tagged CTK-1, CTK-2 or CTK-3 in ctk mutants restored the activities of CAT-3 to levels similar to those of WT strains (Fig. 2D). Additionally, the intensity of the bands corresponding to CAT-1 activity was also reduced in the transformants (Fig. 2D). Ectopic expression of each CTK subunit efficiently suppressed expression of CAT-3 and CAT-1 in each ctk mutants to WT levels (Fig. 2E and F).

Fig. 2.

Ectopic expression of Myc-tagged CTK-1, CTK-2 or CTK-3 in CTK mutants can rescue the resistant phenotype.

A. Growth of WT, ctk-1^KO, ctk-2^KO or ctk-3^KO and CTK-1, CTK-2 or CTK-3 transformants in plates under the treatments of 0, 5, 10 or 20 mM H₂O₂. Quinic acid (QA) was used to induce the qa-2 promoter.

B. Relative values were calculated using the values obtained for controls of WT, ctk-1^KO, ctk-2^KO or ctk-3^KO and CTK-1, CTK-2 or CTK-3 transformants in (A).

C. The expression of Myc-tagged CTK-1 in the WT and ctk-1^KO, CTK-1. Different concentrations of QA were used to induce the qa-2 promoter.

D. Catalase activity assay. Crude extracts from the WT, ctk-1^KO, ctk-2^KO or ctk-3^KO and CTK-1, CTK-2 or CTK-3 transformants were subjected to Native-PAGE and catalase activities were determined by the in-gel assay.

E. Western blot analysis showing the levels of CAT-3 protein in the WT, ctk-1^KO, ctk-2^KO or ctk-3^KO and CTK-1, CTK-2 or CTK-3 transformants. The membranes stained by Coomassie blue represent the total protein in each sample and act as loading control for western blot.

F. Western blot analysis showing the levels of CAT-1 protein in the WT, ctk-1^KO, ctk-2^KO or ctk-3^KO and CTK-1, CTK-2 or CTK-3 transformants. The membranes stained by Coomassie blue represent the total protein in each sample and act as loading control for western blot. [Color figure can be viewed at wileyonlinelibrary.com]

Next, we examined whether CTK complex suppresses cat-3 expression in a dose-dependent manner. Plate assays were performed to examine the growth rates of CTK overexpression strains in medium containing different concentrations of QA following 10 mM H₂O₂ treatment. As shown in Fig. 3A and B, the growth rates of CTK-1 overexpression strain on plates with H₂O₂ decreased significantly, whereas the changes were modest in WT strains, wt, pqa-Myc-CTK-2 strain and wt, pqa-Myc-CTK-3 strain. The leaky expression of Myc-tagged-CTK-1 in the overexpressing strain could also be detected in the medium without QA (Fig. 3C). Western blot analysis revealed that the CAT-3 protein levels were decreased in wt, pqa-Myc-CTK-1 strain compared with that of the WT strains (Fig. 3D). Similarly, levels of CAT-3 activity and cat-3 mRNA were also reduced in wt, pqaMyc-CTK-1 transformants compared with those in WT strains (Fig. 3E and F). We noticed that the growth rate and CAT-3 protein levels of the mutant strains did not change gradually with increasing QA concentration. This may be due to the QA acting as carbon resource and influence growth rate, somehow causing stress or stimulate the expression of CAT-3. Hence QA concentration increments may introduce a combined effect to the growth of the overexpression strain. Anyways, CAT-3 level was reduced in wt, pqa-Myc-CTK-1 strain compared with WT strains in the presence of same concentration of QA, suggesting that the catalytic subunit plays a key role in the regulation of CAT-3 transcription. Taken together, these results indicate that CTK complex negatively regulates cat-3 gene expression.

Fig. 3.

The CTK-1 overexpressing strains exhibit decreased cat-3 expression and are sensitive to H₂O₂.

A. Growth of WT and CTK overexpressing strains in plates under the treatment of 10 mM H₂O₂. Culturing was performed in petri dishes at 25°C under constant light. Different concentrations of QA were used to induce the qa-2 promoter.

B. Relative values were calculated using the values obtained for controls of WT and CTK overexpressing strains in (A).

C. The expression of Myc-tagged CTK-1 in the WT and wt, CTK-1. Different concentrations of QA were used to induce the qa-2 promoter.

D. Western blot analysis showing the levels of CAT-3 protein in the WT and wt, CTK-1 strain under the different concentrations of QA. The membranes stained by Coomassie blue represent the total protein in each sample and act as loading control for western blot.

E. Catalase activity assay. Crude extracts from the WT, wt and CTK-1 strain were subjected to Native-PAGE and the catalase activities were determined by the in-gel assay. Different concentrations of QA were used to induce the qa-2 promoter.

F. RT-qPCR analyses showing the levels of cat-3 mRNA in the WT and wt, CTK-1 strain. 10⁻² M QA was used to induce the qa-2 promoter. Error bars indicate SD (n = 3). *p < 0.05; ^**p < 0.01; ^***p < 0.001. Student’s t test was used. [Color figure can be viewed at wileyonlinelibrary.com]

The kinase activity of CTK-1 is required for the regulation of the cat-3 and cat-1

CTK complex and its catalytic activity is conserved among species. Previous study shows that CTK-1 contributes to the Ser2 phosphorylation of RNAPII CTD in yeast (Bowman and Kelly, 2014). T338 residue of Ctk1 is the phosphorylation site by Cak1 (Cdk-activating kinases 1), which stabilizes cyclin binding and is essential for kinase activation in yeast (Ostapenko and Solomon, 2005). T338A mutation has been shown to cause reduced Ctk1 kinase activity in yeast (Ross et al., 2000). In addition, a mutation of the Ctk1catalytic site (D324N) abolishes all kinase activities of Ctk1 in yeast (Ostapenko and Solomon, 2005). When the protein sequence of Neurospora CTK-1 was used in a BLAST search against protein databases, its homologues were found to be highly conserved in Homo sapiens CDK12, Saccharomyces cerevisiae Ctk1 and Drosophila melanogaster CDK12 (Fig. 4A). The protein sequence of Neurospora CTK-1 displayed high similarity with CDK12 protein of human. To determine whether the kinase activity or structural integrity of CTK-1 protein is required for the suppression of cat-3 and cat-1, we generated a series of CTK-1 mutants by introducing T942A (T338A in yeast), D926N (D324N in yeast) or CDK12Δ (from G787 to H1068) substitution to the qa-Myc-CTK-1 construct. Plate assays showed that the expression of Myc-CTK-1 or Myc-CTK-1^T942A, but not Myc-CTK-1^D926N or Myc-CTK-1^CDK12Δ rescued the growth defective and the H₂O₂ resistant phenotype of ctk-1^KO strain (Fig. 4B and C), suggesting that the kinase activity of CTK-1 plays an important role in the regulation of ROS resistance. Consistent with the genetic phenotypes of transformants on the plates, the in-gel assays showed that the expression of Myc-CTK-1 and Myc-CTK-1^T942A but not Myc-CTK-1^D926N or MycCTK-1^CDK12Δ significantly suppressed the activities and protein levels of CAT-3 and CAT-1 of ctk-1^KO strain (Fig. 4D-G). Furthermore, RT-qPCR results revealed that the elevated levels of cat-3 and cat-1 mRNA in ctk-1^KO strain were suppressed by Myc-CTK-1 and Myc-CTK-1^T942A but not Myc-CTK-1^D926N or Myc-CTK-1^CDK12Δ proteins (Fig. 4H and I). Taken together, these data demonstrated that the catalytic activity of CTK-1 protein was required for its role in the regulation of catalase expression.

Fig. 4.

The kinase activity of CTK-1 contributes to the suppression of cat-3 expression and the resistance to H₂O₂-induced oxidative stress.

A. Amino acid sequence alignment of the CDK12 domains of CTK-1 from Neurospora crassa, Homo sapiens, Saccharomyces cerevisiae and Drosophila melanogaster.

B. Growth of WT, ctk-1^KO and CTK-1 transformants in plates under the treatment of 0, 5, 10 or 20 mM H₂O₂. 10⁻³ M QA was used to induce the qa-2 promoter.

C. Relative values were calculated using the values obtained for controls of WT, ctk-1^KO and CTK-1 transformants in (B).

D. Catalase activity assay. Crude extracts from the WT, ctk-1^KO and CTK-1 transformants were subjected to Native-PAGE, and the catalase activities were determined by the in-gel assay. 10⁻² M QA was used to induce the qa-2 promoter.

E. Western blot analysis showing the levels of CAT-3 protein in the WT, ctk-1^KO and CTK-1 transformants. The membranes stained by Coomassie blue represent the total protein in each sample and act as loading control for western blot. 10⁻² M QA was used to induce the qa-2 promoter.

F. Quantification of CAT-3 protein expression levels in (E).

G. Western blot analysis showing the levels of CAT-1 protein in the WT, ctk-1^KO and CTK-1 transformants. The membranes stained by Coomassie blue represent the total protein in each sample and act as loading control for western blot. 10⁻² M QA was used to induce the qa-2 promoter.

H. RT-qPCR analysis showing the levels of cat-3 mRNA in the WT, ctk-1^KO and CTK-1 transformants. 10⁻² M QA was used to induce the qa-2 promoter.

I. RT-qPCR analysis showing the levels of cat-1 mRNA in the WT, ctk-1^KO and CTK-1 transformants. 10⁻² M QA was used to induce the qa-2 promoter.

Error bars indicate SD (n = 3). *p < 0.05; ^**p < 0.01; ^***p < 0.001. Student’s t test was used. [Color figure can be viewed at wileyonlinelibrary.com]

CTK-1 phosphorylates Ser2 residues of RNAPII CTD at cat-3 locus

Previous studies showed that the CTK complex is responsible for Ser2 phosphorylation of RNAPII CTD (Sterner et al., 1995; Hautbergue and Goguel, 2001; Bowman and Kelly, 2014). To investigate the role of CTK-1 on cat-3 transcription, we measured the phosphorylation levels of Ser2 on RNAPII CTD using the specific antibody for CTD Ser2p in the WT, ctk-1^KO and different ctk-1^KO transformant strains. The levels of CTD Ser2 phosphorylation were very low in the ctk-1^KO strain compared with that in WT strains (Fig. 5A). However, the expression of Myc-CTK-1 and Myc-CTK-1^T942A but not Myc-CTK-1^D926N or Myc-CTK-1^CDK12Δ restored the Ser2 phosphorylation of RNAPII CTD (Fig. 5A), indicating that the CTK-1 is the major kinase for CTD Ser2 residues in Neurospora, and the Myc-CTK-1^D926N and Myc-CTK1^CDK12Δ are catalytic dead proteins. To assess the role of CTK-1-mediated Ser2 phosphorylation of RNAPII in the regulation of cat-3 gene, we performed ChIP assay using the CTD Ser2p specific antibody to check the recruitment of RNAPII Ser2p at cat-3 locus in WT and ctk-1^KO strain. The oligo nucleotide primer pairs were designed to target an approximately 8.5-kb region from the proximal 6-kb upstream region of cat-3 transcription start site (TSS) to the cat-3 ORF 3^′ region in the WT strains (primer pairs 1–8; Fig. 5B). ChIP-qPCR results showed that during cat-3 gene transcription, the total level of RNAPII CTD Ser2 phosphorylation was increased at the cat-3 locus (primer pairs 6–8) in WT strains (Fig. 5C), whereas the total level of Ser2 phosphorylation remains low in ctk-1^KO strain (Fig. 5C), confirming that the Ser2 phosphorylation of RNAPII CTD was also abolished at cat-3 locus in ctk-1^KO strain. To determine that the high expression of cat-3 is due to the increased recruitment of RNAPII at cat-3 locus in ctk-1^KO strain, we carried out ChIP assays using available commercial CTD specific antibody in WT and ctk-1^KO strains. Consistent with high expression of cat3 in ctk-1^KO strains, qPCR results revealed that the binding of RNAPII CTD at cat-3 locus in ctk-1^KO strain was increased compared with those in WT strains (Fig. 5D). To confirm this possibility, we generated an antiserum against Neurospora RPB-1 and repeated ChIP assays using this RPB-1 specific antibody. As shown in Fig. 5E, the recruitment levels of RPB-1 at cat-3 locus in ctk-1^KO strain also increased compared with those in the WT strains. These results demonstrated that loss of CTK-1 catalytic function led to increased recruitment of RNAPII at cat-3 gene and elevated cat-3 expression.

Fig. 5.

CTK-1 is required for Ser2 phosphorylation of RNAPII CTD.

A. Western blot analysis showing the levels of RNAPII CTD Ser2p in WT, ctk-1^KO and CTK-1 transformants. The membranes stained by Coomassie blue represent the total protein in each sample and act as loading control for western blot. 10⁻² M QA was used to induce the qa-2 promoter.

B. Schematic depiction of a 5-kb AT-rich DNA region located between cat-3 (NCU00355) and NCU00354 on linkage group III of Neurospora genome. Short black lines (primer pairs 1–8) under the schematic indicate the regions tested by ChIP-qPCR. TSS, transcription start site; ORF, open reading frame.

C. ChIP analysis showing the recruitment of RNAPII CTD Ser2p at different regions of the cat-3 locus in the WT and ctk-1^KO strain (primer pairs 5–8).

D. ChIP analysis showing the enrichment of CTD at different regions of the cat-3 locus in the WT and ctk-1^KO strain (primer pairs 5–8).

E. ChIP analysis showing the enrichment of RPB-1 at different regions of the cat-3 locus in the WT and ctk-1^KO strain (primer pairs 5–8).

Error bars indicate SD (n = 3). *p < 0.05; ^**p < 0.01; ^***p < 0.001. Student’s t test was used. [Color figure can be viewed at wileyonlinelibrary.com]

CTK-1-mediated Ser2 phosphorylation recruits SET-2 to cat-3 locus which maintains proper acetylation status at the cat-3 locus

During transcription elongation, the phosphorylated RNAPII Ser2 also acts as a platform for recruiting SET-2, a histone methyltransferase for H3K36 trimethylation (Li et al., 2002; Krogan et al., 2003; Li et al., 2003). To test whether CTK-1-mediated Ser2p is required for SET-2 recruitment at cat-3 locus, we performed ChIP assay using a specific anti-SET-2 antibody. As shown in Fig. 6A, the SET-2 enrichment was greatly diminished in ctk-1^KO strain compared with those in the WT strains. Furthermore, ChIP assay using H3K36me3 antibody revealed that the deletion of CTK-1 led to a dramatic decrease of H3K36me3 levels at cat-3 locus compared with levels in the WT strains (Fig. 6B). These results indicated that CTK-1-mediated Ser2 phosphorylation of RNAPII CTD leads to SET-2 recruitment and H3K36me3 at cat-3 locus during transcription elongation. The dramatic decrease of SET-2 and H3K36me3 levels at the cat-3 locus in the ctk-1^KO strain prompted us to examine whether the SET-2 functions in the regulation of cat-3 gene expression in Neurospora. As shown in Fig. 6C and D, the set-2^KO strain was resistant to H₂O₂-induced ROS stress in constant light condition on plate compared with those of WT strains. Consistent with the plate assay results, deletion of set-2 also elevated both CAT-3 protein and activity levels (Fig. 6E–G). Similarly, cat-3 mRNA levels in set-2^KO mutants were dramatically increased compared with those in the WT strains (Fig. 6H). These results suggest that CTK-1 regulates cat-3 transcription, in part, through SET-2-mediated H3K36me3 at the cat-3 locus.

Fig. 6.

CTK-1 regulates cat-3 transcription through SET-2-mediated H3K36me3 at the cat-3 locus.

A. ChIP analysis showing the enrichment of SET-2 at different regions of the cat-3 locus in the WT and ctk-1^KO strain (primer pairs 5–8).

B. ChIP analysis showing the enrichment of H3K36me3 at different regions of the cat-3 locus in the WT and ctk-1^KO strain (primer pairs 5–8).

C. Growth of WT, set-2^KO strains in plates under the treatments of 0, 5, 10 or 20 mM H₂O₂.

D. Relative values were calculated using the values obtained for controls of WT, set-2^KO mutants in (C).

E. Western blot analysis showing the levels of CAT-3 protein in the WT, set-2^KO mutants. The membranes stained by Coomassie blue represent the total protein in each sample and act as loading control for western blot.

F. Quantification of CAT-3 protein expression levels in (E).

G. Catalase activity assay. Crude extracts from the WT, set-2^KO mutants were subjected to Native-PAGE and the catalase activities were determined by the in-gel assay.

H. RT-qPCR analyses showing the levels of cat-3 mRNA in the WT and set-2^KO mutants.

Error bars indicate SD (n = 3). *p < 0.05; ^**p < 0.01; ^***p < 0.001. Student’s t test was used. [Color figure can be viewed at wileyonlinelibrary.com]

In S. cerevisiae, transcribed genes maintain states of hypoacetylation due to Set2-mediated H3K36me3 (Keogh et al., 2005; Li et al., 2007; Smolle et al., 2012; Venkatesh et al., 2012). Our recent study also showed that deletion of SET-2 resulted in histone hyperacetylation and thus loosens chromatin structure at frq locus (Sun et al., 2016). We performed ChIP assays using anti-H3 and H4 acetylation antibody to check the histone acetylation statuses of nucleosomes at cat-3 gene body in WT and ctk-1^KO strains. ChIP-qPCR results revealed that the relative levels of H3ac and H4ac increased at the cat-3 gene body in ctk-1^KO strain (Fig. 7A and B). Our previous research showed that CPC1 is a key activator for cat-3 expression via directly binding to the promoter of cat-3 (Dong et al., 2018; Qi et al., 2018). To test whether elevated histone acetylation caused the increased binding of CPC1 proteins at cat-3 locus for its expression in ctk-1^KO strain, we performed ChIP assay using the specific anti-CPC1 antibody. As shown in Fig. 7C, the enrichment of CPC1 was increased at the promoter and ORF of the cat-3 locus (primer pairs 5–8) in ctk-1^KO strain compared with those in WT strain. Interestingly, RT-qPCR analysis revealed that the transcription levels of cpc-1 elevated in ctk-1^KO strain compared with those in the WT strain (Fig. 7D), implying that the enhanced CPC1 expression results in its elevated recruitment at cat-3 locus. Taken together, these results indicated that deletion of CTK-1 led to decreased H3K36me3 levels and loosen chromatin structure at the cat-3 gene body by histone hyperacetylation, which facilitated the access of transcriptional machinery and promotes cat-3 gene expression.

Fig. 7.

CTK-1-mediated Ser2 phosphorylation affects proper histone acetylation status at the cat-3 locus and cpc-1 expression.

A. ChIP analysis showing the acetylation of H3 at cat-3 locus in the WT and ctk-1^KO strain (primer pairs 5–8).

B. ChIP analysis showing the acetylation of H4 at cat-3 locus in the WT and ctk-1^KO strain (primer pairs 5–8).

C. ChIP analysis showing the enrichment of CPC1 at different regions of the cat-3 locus in the WT and ctk-1^KO strain (primer pairs 5–8).

D. RT-qPCR analyses showing the levels of cpc-1 mRNA in the WT and ctk-1^KO mutants.

Error bars indicate SD (n = 3). *p < 0.05; ^**p < 0.01; ^***p < 0.001. Student’s t test was used. [Color figure can be viewed at wileyonlinelibrary.com]

Discussion

Transcriptional regulation of inducible gene elements is an important step in responding to environmental or developmental signals. Our previous studies have demonstrated that a bZIP transcription factor, CPC1/GCN4 is responsible for activating catalase-3 transcription (Dong et al., 2018; Qi et al., 2018). In this study, we found that the transcriptional regulator complex CTK, which is responsible for the phosphorylation of the Ser2 residues of RNAPII CTD, is also involved in the regulation of cat-3 transcription in N. crassa. Deletion of each subunit gene of CTK complex in N. crassa genome leads to robust CAT-3 and CAT-1 activation and high resistance to H₂O₂ treatment. CTK-1 is the catalytic subunit of CTK complex, and overexpression of CTK-1 results in higher sensitivity to H₂O₂ and decreased levels of CAT-3 protein and cat-3 mRNA. Furthermore, elevated expression of cat-3 and cat-1 is dependent on the kinase activity of CTK-1. Biochemical data demonstrated that N. crassa CTK-1 is responsible for the phosphorylation of Ser2 residues of RNAPII CTD at cat-3 locus. According to previous studies and our experiments, we assume that mutation of CTK-1 alters chromatin accessibility and increases cat-3 expression. Additionally, CAT-3 induction is impeded in ctk^KO strains treated with H₂O₂, indicating that functional CTK complex serves as a regulatory target for environmental or developmental stimuli. Taken together, our results demonstrated that CTK complex is critical for proper cat-3 and cat-1 transcription.

CTK-1 negatively regulates cat-3 transcription, and the role of CTK-1 on gene expression has been reported in previous studies. To identify the underlying molecular mechanism, we performed a series of experiments and found that Ser2 phosphorylation of RNAPII CTD at cat-3 locus in ctk-1^KO strain was extremely low (Fig. 5C). Several studies in multiple systems support that the histone methyltransferase Set2, which methylates histone H3K36, is associated with CTD phosphorylation. In this study, we found that deletion of CTK-1 led to a dramatic decrease of SET-2 and H3K36me3 levels at the cat-3 locus (Fig. 6A and B). These results suggest that the Ctk1-mediated Ser2 phosphorylation is necessary for efficient recruitment of the histone H3K36 methyltransferase Set2 to the CTD of RNAPII.

Nucleosomes generally act as barriers to transcription, by hampering RNAPII recruitment, movement and related transcription machineries. Histone modifications directly affect the chromatin structures, which play important roles in regulating gene transcription. Loss of H3K36me3 results in increased histone acetylation at the open reading frame (ORF) of genes and leads to cryptic transcription due to promoters within ORFs. Previous research shows that cat-3 gene responds to H₂O₂ stress; its expression level and H3 acetylation at its locus are significantly increased after H₂O₂ treatment. To further test how the chromatin structure is loosened in ctk^KO strains, we performed ChIP assay and found that the levels of H3 and H4 acetylation increases in ctk^KO strains (Fig. 7A and B), which facilitates the access of CPC1 at cat-3 gene locus (Fig. 7C). Our finding suggests that CTK-1 complex regulates the transcription elongation of cat-3 gene and the histone hyperacetylation affects gene expression efficiency in a context-dependent manner.

In yeast, the role of phosphorylating CTD Ser2 is divided between Ctk1 and Bur1. In N. crassa, the CTK-1 is the major kinase for CTD Ser2 residues. We tried to knockout bur-1 gene, but failed to get a homozygous bur-1 knockout strain, suggesting that although BUR-1 is not the main CTD Ser2 phosphorylation kinase in Neurospora, it is required for cell survival. It is worth noting that BUR-1 and CTK-1 are unable to functionally substitute even when overexpressed in yeast, suggesting that they have distinct functions (Keogh et al., 2003).

Many studies in yeast and humans demonstrate the role of H3K36 methylation in transcriptional activation. However, H3K36 methylation has also been implicated in transcriptional repression, alternative splicing, dosage compensation, DNA replication, DNA repair and DNA methylation. In higher eukaryotes, the abnormal levels of H3K36 methylation cause developmental defects and diseases (Wagner and Carpenter, 2012). In this study, we demonstrate that the CTK-1 play essential roles in the phosphorylation of CTD Ser2, which repress the expression of cat-3 through the SET-2 and H3K36me3 pathway. We also find that cat-1 is negatively regulated by the CTK complex. As the catalase and the ROS clearance system are highly conserved from prokaryotes to eukaryotes, it appears to be important to unravel the oxidant stress adaption pathways to better understand and implement relevant measures for the proper treatment of stress and disease.

Materials and methods

Strains and culture conditions

The 87–3 (bd, a) strain was used as the wild-type strain in this study (Belden et al., 2007). The ku70^RIP (bd, a) strain, generated previously (He et al., 2006), was used as the host strain for creating the ctk-1 (NCU06685), ctk2 (NCU04495) or ctk-3 (NCU06470) knock-out mutant by deleting the entire ORF through homologous recombination using a protocol described previously (Colot et al., 2006). The set-2^KO strain was generated previously (Zhou et al., 2013; Sun et al., 2016).The plasmid containing qa-2 promoter driven the CTK-1 ORF and its 3^′UTR (pqa-5Myc-6His-CTK-1) was used as the template for mutagenesis, and three mutations of CTK-1 (CTK1^T942A, CTK-1^D926N or CTK-1^CDK12Δ) were generated. Afterwards, the mutated plasmids (pqa-5Myc-6His-CTK1^T942A, pqa-5Myc-6His-CTK-1^D926N or pqa-5Myc-6HisCTK-1^CDK12Δ) were transformed into ctk-1^KO (bd, his-3⁻) strains to get ctk-1^KO, Myc-CTK-1^T942A, ctk-1^KO, Myc-CTK-1^D926N and ctk-1^KO, Myc-CTK-1^CDK12Δ transformants. The wt, pqa-Myc-CTK-1 were created by transferring pqa-5Myc-6His-CTK-1 constructs into the his-3 locus of 301–6 (bd, his-3⁻) host strain.

Liquid cultures were grown in minimal medium (1× Vogel’s, 2% glucose) with different concentration of H₂O₂. When QA was used, liquid cultures were grown in low-glucose medium (1× Vogel’s, 0.1% glucose, 0.17% arginine). Liquid cultures were grown at 25°C with shaking for 18 h in constant light (LL).

For protein and RNA analysis, the 5– or 7–day-old conidia were inoculated in petri dishes with 50 ml liquid medium containing 1× Vogel’s and 2% glucose under static culture condition at 25°C for 1–2 days. The mycelium mat (adhered hyphae) was cut into disks for quantification. For each strain, equal amounts of mycelium disks were transferred into flasks with 50 ml fresh liquid medium and then incubated with agitation at 130 rpm for 18 h in constant light (LL) at 25°C. All cell extracts from the adhered mycelium were used for performing protein and RNA analysis.

Protein analysis

Protein extraction, quantification and western blot analysis were performed as described previously (Yang et al., 2014). Equal amounts of total protein (40 μg) were loaded into each lane. After electrophoresis, proteins were transferred onto PVDF membrane. Western blot analysis was performed using antibodies against the proteins of interest. Densitometric analysis from three independent experiments was calculated with Quantity One® 1-D analysis software made by Bio-Rad Laboratories.

RNA analysis

For quantitative real-time reverse transcriptase-PCR, total RNA was isolated with Trizol agent and treated with DNase I to remove genomic DNA according to the manufacturer’s protocol. Each RNA sample (total RNA, 5 μg) was subjected to reverse transcription with M-MLV reverse transcriptase (Promega) and then amplified by real-time PCR. The primers used for qPCR were shown in Table S1. The relative values of gene expression were calculated using the 2^−ΔΔCT method (Livak and Schmittgen, 2001) by comparing the cycle number for each sample with that for the untreated control. The results were normalized to the expression levels of β-tubulin gene.

In-gel assay for catalase

Cell extracts of mycelium disks cultured 18 h in liquid medium were used for the zymogram. Grinded tissues were mixed with ice-cold extraction buffer containing 50 mM HEPES (pH 7.4), 137 mM NaCl, 10% glycerol and protease inhibitors, Pepstatin A (1 μg/ml), Leupeptin (1 μg/ml), PMSF (1 mM) and centrifuged at 10,000g for 10 min at 4°C. The protein concentration was measured by Bio-Rad protein assay dye at 595 nm.

For the in-gel assay, catalase activity was determined as described previously (Lledías et al., 1998). Equal amounts of total protein (40 μg) were loaded into a 7.5% native polyacrylamide slab gel. After electrophoresis, the gel was immersed in 10 mM H₂O₂ with shaking for 10 min and then in a 1:1 mixture of freshly prepared 1% potassium hexacyanoferrate (III) and 1% iron (III) chloride hexahydrate. Catalase activity was visualized as a band where H₂O₂ was decomposed by catalase.

Plate assay

The medium for plate assays contained 1× Vogel’s salts, 0.1% glucose, 0.17% arginine, 50 ng/ml biotin and 1.5% agar with different concentration of H₂O₂. In the plate medium containing QA, 0.1% glucose was replaced with the 10⁻³ M QA.

The 5- or 7-day-old conidia were inoculated in petri dishes with 50 ml liquid medium containing 1× Vogel’s and 2% glucose under static culture condition at 25°C for 1–2 days. The disks of mycelium mat (adhered hyphae) were cut with a cork borer for quantification. A piece of mycelium mat of WT or mutant strain was inoculated at the centre of the petri dishes and was grown under constant light (LL) with medium containing 0, 5, 10 or 20 mM H₂O₂. When the WT strain grew and covered the medium without H₂O₂, all plates were scanned and the average growth rate of each strain relative to that in medium without H₂O₂ (0 mM) was calculated. Each experiment was performed at least three times independently.

Generation of antiserum against RPB-1

GST-RPB1 (containing RPB-1 amino acids D804-T1098) fusion proteins were expressed in BL21 cells, and soluble recombinant proteins were purified and used as the antigens to generate rabbit polyclonal antiserum, as described previously (Xu et al., 2010; Zhao et al., 2010).

ChIP analysis

Chromatin immunoprecipitation (ChIP) assays were performed as described previously (Zhou et al., 2013). Briefly, N. crassa tissues were fixed with 1% formaldehyde for 15 min at 25°C with shaking. Glycine was added at a final concentration of 125 mM, and samples were incubated for another 5 min. The cross-linked tissues were grinded and re-suspended at 0.5 g in 6 ml lysis buffer containing protease inhibitors (1 mM PMSF, 1 μg/ml leupeptin and 1 μg/ml pepstatin A). Chromatin was sheared by sonication to approximately 500–1000 bp fragments. A 1 ml aliquot of protein solution (2 mg/ml) was used for each immunoprecipitation reaction, and 10 μl was kept as the input DNA. The ChIP was carried out with 10 μl of anti-RNA polymerase II CTD repeat YSPTSPS (phospho S2) antibody [H5] (ab24758; Abcam), 10 μl of anti-RNA polymerase II CTD repeat YSPTSPS antibody [8WG16] (ab817; Abcam), 3 μl of anti-H3K36me3 antibody (ab9050; Abcam), 3 μl of antiH3 antibody (2650; CST), 3 μl of anti-H3ac antibody (07–473; Millipore), 10 μl of anti-SET-2 antibody, 10 μl of anti-RPB-1 antibody and 10 μl of anti-CPC-1 antibody. Immunoprecipitated DNA was quantified by using real-time polymerase chain reaction with primer pairs. The primer pairs used are listed in Table S1. ChIP-quantitative PCR data were normalized by the input DNA and presented as a percentage of input DNA. Each experiment was independently performed at least three times.