Rpb1 sumoylation in response to UV radiation or transcriptional impairment in yeast

Abstract

Covalent modifications of proteins by ubiquitin and the Small Ubiquitin-like MOdifier (SUMO) have been revealed to be involved in a plethora of cellular processes, including transcription, DNA repair and DNA damage responses. It has been well known that in response to DNA damage that blocks transcription elongation, Rpb1, the largest subunit of RNA polymerase II (Pol II), is ubiquitylated and subsequently degraded in mammalian and yeast cells. However, it is still an enigma regarding how Pol II responds to damaged DNA and conveys signal(s) for DNA damage-related cellular processes. We found that Rpb1 is also sumoylated in yeast cells upon UV radiation or impairment of transcription elongation, and this modification is independent of DNA damage checkpoint activation. Ubc9, an E2 SUMO conjugase, and Siz1, an E3 SUMO ligase, play important roles in Rpb1 sumoylation. K1487, which is located in the acidic linker region between the C-terminal domain and the globular domain of Rpb1, is the major sumoylation site. Rpb1 sumoylation is not affected by its ubiquitylation, and vice versa, indicating that the two processes do not crosstalk. Abolishment of Rpb1 sumoylation at K1487 does not affect transcription elongation or transcription coupled repair (TCR) of UV-induced DNA damage. However, deficiency in TCR enhances UV-induced Rpb1 sumoylation, presumably due to the persistence of transcription-blocking DNA lesions in the transcribed strand of a gene. Remarkably, abolishment of Rpb1 sumoylation at K1487 causes enhanced and prolonged UV-induced phosphorylation of Rad53, especially in TCR-deficient cells, suggesting that the sumoylation plays a role in restraining the DNA damage checkpoint response caused by transcription-blocking lesions. Our results demonstrate a novel covalent modification of Rpb1 in response to UV induced DNA damage or transcriptional impairment, and unravel an important link between the modification and the DNA damage checkpoint response.

Figure 1. Western blots showing Rpb1 sumoylation in response to UV radiation or impairment of transcription elongation.

(A) Rpb1 was immunoprecipitated from the unirradiated and UV irradiated cells using antibody 8WG16 (anti-Rpb1) and probed with anti-SUMO and 8WG16 antibodies. (B) Sumoylated proteins were immunoprecipitated from the unirradiated and UV irradiated cells and probed with 8WG16 and anti-SUMO antibodies. (C) UV-induced Rpb1 sumoylation in wild type (JKM179), sml1 (YFD756) and sml1 mec1 (YAA25) cells. (D) Sumoylation of Rpb1 in response to UV or treatments of transcription inhibitors. (E) UV-induced Rpb1 sumoylation in cells expressing wild type (CX84) or K1487R mutant (CX79) Rpb1. Bars on the left of the blot indicate distinct bands formed by wild type Rpb1. Arrow heads on the right of the blot mark bands abolished by the K1487R mutation. (F) UV-induced Rpb1 sumoylation in cells expressing wild type (CX84) or K to R mutant (CX79, CX105, CX106, CX108, CX110 and CX110) Rpb1. Bars on the left of the blot indicate distinct bands formed by wild type Rpb1. Arrow heads on the right of the blot mark bands not shown by the mutant Rpb1. WT, wild type.

Figure 2. Western blots showing the roles of Ubc9 and Siz1 in UV-induced Rpb1 sumoylation.

(A) Degradation of degron-myc tagged Ubc9 upon shifting to nonpermissive temperature (37°C) in galactose containing medium (to induce the expression of plasmid pKL142 encoded Ubr1, a ubiquitin E3 ligase). Tubulin serves as an internal loading control. (B) Abolishment of UV-induced Rpb1 sumoylation when Ubc9 was depleted. Rpb1 was immunoprecipitated from the cells cultured at the indicated conditions using antibody 8WG16 and probed with anti-SUMO and 8WG16 antibodies. (C) The roles of Siz1 and Siz2 in UV-induced Rpb1 sumoylation. Rpb1 was immunoprecipited from the UV irradiated wild type (BY4741) and mutant (strains 4245 and 2412) cells using antibody 8WG16 and probed with anti-SUMO and 8WG16 antibodies. The control was a sample prepared from unirradiated wild type cells. WT, wild type.

Figure 3. Sumoylation of Rpb1 does not affect its degradation in response to UV radiation.

Whole cell extracts were prepared from the cells that had been incubated for different times following UV irradiation. Rpb1 in the whole cell extracts were probed with antibody 8WG16 on the Western blots. Tubulin serves as an internal loading control. (A) Levels of Rpb1 in isogenic wild type (BY4741) and siz1 (strain 4245) cells. (B) Levels of wild type and K1487R mutant Rpb1 expressed in isogenic cells (CX84 and CX79). (C) Levels of Rpb1 in wild type (Y452) and hex3 slx8 (MHY3861) cells. (D) Levels of Rpb1 in isogenic cells expressing wild type (JD74-13c) or K11,15,19R mutant Smt3 (YKU116). WT, wild type.

Figure 4. Ubiquitylation of Rpb1 does not affect its sumoylation.

Rpb1 was immunoprecipitated from the unirradiated (control, BJ5465) and UV irradiated isogenic wild type [BJ5465 (lane 2) and Y452 (lane 4)], elc1 (CR105) and def1 (SL128) cells and cells expressing wild type [CX84 (lane 6)], K330R (CR191) or K695R (CR192) mutant Rpb1 using antibody 8WG16 and probed with anti-SUMO and 8WG16 antibodies.

Figure 5. UV-induced sumoylation in wild type and NER-deficient cells.

Log phase cells were irradiated with UV and incubated in a rich medium at 30°C. Rpb1 was immunoprecipitated from the cells at different times of the post-UV incubation using antibody 8WG16 and probed with anti-SUMO and 8WG16 antibodies. (A) UV-induced Rpb1 sumoylation in wild type (BJ5465), rad7 (GGR-deficient) (SL212), rad26 rpb9 (TCR-deficient) (SL81), rad7 rad26 rpb9 (GGR- and TCR-deficient) (SL244) and rad14 (GGR- and TCR-deficient) (CR14) cells. As Rpb1 was gradually degraded during the post-UV incubation in RPB9 ⁺ (WT, rad7 and rad14) cells , the loadings of samples from these cells at the different time points were adjusted to approximately the same level of Rpb1 remaining. (B) UV-induced Rpb1 sumoylation in rad7 rpb9 (SL221), rad7 rad26 rpb9 (SL244) and rad7 rad26 rpb9 spt4 (SL243) cells.

Figure 6. Abolishment of Rpb1 sumoylation at K1487 does not affect overall TCR or Rad26-independent TCR.

(A) DNA sequencing gels showing repair of UV-induced cyclobutane pyrimidine dimers (CPDs) in the transcribed strand of the RPB2 gene in rad16 cells expressing wild type (CX85) or K1487R mutant (CX87) Rpb1. (B) DNA sequencing gels showing repair of CPDs in the transcribed strand of the RPB2 gene in rad16 rad26 cells expressing wild type (CX112) or K1487R mutant (CX113) Rpb1. Lanes U are unirradiated controls. Other lanes are samples from cells incubated for different times (min) following UV irradiation. The arrow on the left of the gels indicates the transcription start site of RPB2.

Figure 7. Effects of Rpb1 sumoylation at K1487 on UV-induced Rad53 phosphorylation.

(A–C) UV-induced Rad53 phosphorylation in log phase wild type (for NER genes) cells expressing wild type (CX84) or K1487R mutant (CX79) Rpb1. (D–F) UV-induced Rad53 phosphorylation in log phase rad16 cells expressing wild type (CX85) or K1487R mutant (CX87) Rpb1. (G–I) UV-induced Rad53 phosphorylation in log phase rad16 rad26 cells expressing wild type (CX112) or K1487R mutant (CX113) Rpb1. The cells were irradiated with UV and incubated in a rich medium at 30°C. Whole cell extracts were prepared from the cells at different times of the post-UV incubation. Rad53 in the whole cell extracts was probed with an anti-Rad53 antibody on Western blots. p and u on the left of the blots indicate bands of phosphorylated and unphosphorylated Rad53, respectively. Plots C, F and I show ratios of phosphorylated Rad53 (Rad53p) to unphosphorylated Rad53 (Rad53u) in the wild type, rad16 and rad16 rad26 cells, respectively. Error bars represents standard deviations.