Redundant Roles of Rpn10 and Rpn13 in Recognition of Ubiquitinated Proteins and Cellular Homeostasis

Abstract

Intracellular proteins tagged with ubiquitin chains are targeted to the 26S proteasome for degradation. The two subunits, Rpn10 and Rpn13, function as ubiquitin receptors of the proteasome. However, differences in roles between Rpn10 and Rpn13 in mammals remains to be understood. We analyzed mice deficient for Rpn13 and Rpn10. Liver-specific deletion of either Rpn10 or Rpn13 showed only modest impairment, but simultaneous loss of both caused severe liver injury accompanied by massive accumulation of ubiquitin conjugates, which was recovered by re-expression of either Rpn10 or Rpn13. We also found that mHR23B and ubiquilin/Plic-1 and -4 failed to bind to the proteasome in the absence of both Rpn10 and Rpn13, suggesting that these two subunits are the main receptors for these UBL-UBA proteins that deliver ubiquitinated proteins to the proteasome. Our results indicate that Rpn13 mostly plays a redundant role with Rpn10 in recognition of ubiquitinated proteins and maintaining homeostasis in Mus musculus.

Fig 1. Loss of Rpn13 causes neonatal lethality in mice.

(A) Genotype frequencies of embryos produced from Adrm1 ^+/- (Rpn13 ^+/-) mouse intercrosses. Numbers in parenthesis indicate resorbed fetuses or dead newborns. E: Embryonic day, P: Postnatal day. (B) Gross appearance of control and Rpn13KO littermates shortly after birth. (C) Skeletal analysis of E18.5 littermates by Alzarin red (bone) and Alcian blue (cartilage) staining. (D) Immunohistochemical analysis of sagittally sectioned E18.5 littermates by Rpn13 antibody. (E) Survival curves of newborn mice. Control and Rpn13KO mice were delivered by cesarean section.

Fig 2. Rpn13 deficiency in the liver impairs degradation of ubiquitinated proteins.

(A) Immunoblot analysis of liver lysates from 8-week-old control and Rpn13^LKO mice with antibodies against the indicated proteins. Asterisk indicates a nonspecific band. (B) Lysates from control and Rpn13^LKO livers were fractionated by glycerol gradient centrifugation (8 to 32% glycerol from fraction 1 to 30) and an equal amount of each fraction was used for immunoblot analysis using antibodies against the indicated proteins. Asterisks indicate nonspecific bands. (C) Each fraction of (B) was assayed for chymotrypsin-like activity using Suc-LLVY-AMC as a substrate. (D) The 26S proteasome fractions of (C) (fractions 20–23) were subjected to the assay of chymotrypsin-like activity (left panel), and degradation of ³⁵S-labeled cIAP1 with or without ubiquitination was measured and normalized by chymotrypsin-like activity (right panel). Data are mean ±standard deviations from triplicate experiments. **p < 0.01. (E) The deubiquitinating activities of 26S proteasome fractions of (C) were measured using ubiquitin-AMC as a substrate. Data are mean ± standard deviations from triplicate experiments. **p < 0.01

Fig 3. Simultaneous deletion of Rpn10-UIM and Rpn13 causes severe liver injury.

(A) Representative macroscopic images of 2-week-old control and DKO livers. (B) H&E staining of 4-week-old Rpn10^LΔUIM, Rpn13^LKO and 2-week-old control and DKO livers. All scale bars (black lines), 300 μm. (C) Immunohistochemical analysis on representative liver paraffin sections from 2-week-old mice by using Rpn10 (C) and Rpn13 antibodies. Scale bars, 300 μm. (D and E) Representative H&E staining (D) and immunohistochemical analysis of ubiquitin (E) on liver sections from 2-week-old mice. Arrows in (D) indicate regions of sloughing hepatocytes. Arrowheads in (E) indicate hepatocytes with high accumulation of ubiquitin in cytosol. All scale bars (black lines), 50 μm. (F) Azan staining of liver sections from 5-week-old control and DKO mice (right panels). All scale bars (black lines), 200 μm. Real-time RT-PCR was used to measure the expression of transcripts encoding fibrosis markers (Col1a1 and Epcam) in the livers of 3–6-week-old control and DKO mice (left panels). Data represent levels of transcripts in each genotype liver relative to those in control liver and are expressed as means; error bars denote SEM. **p < 0.01 (n = 4 each genotype).

Fig 4. Redundant roles of Rpn10 and Rpn13 in degradation of ubiquitinated proteins.

(A) Immunoblot analysis of liver lysates from 3-week-old control and DKO mice with antibodies against the indicated proteins. (B) Real-time RT-PCR was performed to measure the mRNA expressions of the proteasome subunits α6 and Rpt4 in the liver of 2–4-week-old control and DKO mice. Data represent transcript levels in DKO livers relative to those in control livers and are expressed as means; error bars denote SEM. **p < 0.01 (n = 7 for each genotype). (C) Lysates from control and DKO livers were fractionated by glycerol gradient centrifugation (8 to 32% glycerol from fraction 1 to 30) and an equal amount of each fraction was used for immunoblot analysis using antibodies against the indicated proteins. Asterisks indicate nonspecific bands. (D) Each fraction of (C) was assayed for chymotrypsin-like activity using Suc-LLVY-AMC as a substrate. (E) The 26S proteasome fractions of (D) (fractions 20–23) were subjected to the assay of chymotrypsin-like activity (left panel). Degradation rates of ³⁵S-labeled cIAP1 with or without ubiquitination were measured and normalized by chymotrypsin-like activity (right panel). Data are mean ± standard deviations from triplicate experiments. **p < 0.01. (F) Lysates from HeLa cells transfected with indicated siRNAs were assayed for Suc-LLVY-AMC hydrolyzing activity (left panel) and degradation of ³⁵S-labeled cIAP1 with or without ubiquitination. cIAP degradation rates are normalized by Suc-LLVY-AMC hydrolyzing activity (right panel). Data are mean ± standard deviations from three experiments. *p < 0.05; **p < 0.01.

Fig 5. Defective binding of ubiquitinated and UBL-UBA proteins to Rpn10ΔUIM/ΔRpn13 proteasomes.

(A) Homogenates from mouse livers were immunoprecipitated with an anti-Rpt6 antibody and subjected to immunoblotting with the indicated antibodies. Values for the relative band intensities of ubiquitin normalized by tubulin (input) or Rpt6 (IP) are shown as A and B, with the control being set to one. Values for B/A indicate the relative amount of bound ubiquitinated proteins to the amount of input ubiquitinated proteins. (B) HEK293T cells were transfected with siRNA against Rpn10, Rpn13, or Uch37. Where indicated, cells were transfected with a mixture of siRNAs. After 96h, cell extracts were subjected to SDS-PAGE, followed by immunoblotting with the indicated antibodies.

Fig 6. Synthetic effect of Rpn10-UIM and Rpn13 deletion on degradation of ubiquitinated proteins and cellular stress.

(A) Immunoblot analysis of whole-cell extracts of livers from indicated genotypes of mice (2-month-old for Rpn10^LΔUIM and Rpn13^LKO, and 2–4-week-old DKO) with antibodies against indicated proteins. (B and C) Real-time RT-PCR was used to measure the expression of transcripts encoding Nrf1, β-catenin, redox pathway (p62 [18412] and Ho-1 [15368]), and UPR pathway (Bip [14828] and Chop [13198]) genes in the livers of 3–6-week-old control and DKO mice. Data represent levels of transcripts in each genotype liver relative to those in control liver and are expressed as means; error bars denote SEM. *p < 0.05; **p < 0.01 (n = 3 for Rpn10^LΔUIM and Rpn13^LKO, and n = 4 for DKO). (D) HeLa cells were transfected with siRNA against Rpn10 and Rpn13. Where indicated, cells were transfected with a mixture of siRNAs. After 96h, whole-cell extracts were subjected to SDS-PAGE, followed by immunoblotting with the indicated antibodies.

Fig 7. Spontaneous liver injury, fibrosis and regeneration in DKO mice.

(A) Real-time RT-PCR was used to measure the expressions of p53 target pro-apoptotic related genes Noxa and Puma (left panel) and proliferation marker genes Gpc3 and Afp (right panel) in the livers of 3–6-week-old control and DKO mice. Data represent levels of transcripts in each genotype liver relative to those in control liver and are expressed as means; error bars denote SEM. *p < 0.05; **p < 0.01 (n = 4 each genotype). (B) Representative liver sections of 2- (top panels), 8-week (middle panels), and 5-month-old (bottom panels) control and DKO mice were stained for Ki-67 followed by quantification of the percentage of Ki-67 positive proliferating hepatocytes calculated from three high-power-fields (HPF) analysis. Results are shown as means, and error bars indicate SEM (n = 3–7 mice for each genotype). All scale bars (black lines), 100 μm. (C) Immunohistochemical analyses on representative liver paraffin sections from 4-week-old and 4-month-old mice by using Rpn10 C-terminal antibodies, refer to Rpn10 (C), and Rpn13 antibodies. All scale bars (black lines), 300 μm. (D) Immunofluorescent analysis of liver sections from 4-week-old DKO mice by using Rpn10 (C) and Rpn13 antibodies. 4’, 6-Diamidino-2-phenylindole (DAPI) was used for nuclear counterstaining. The dashed line region represents Rpn13 (+)/Rpn10 (-) hepatocytes, while the lined region represents Rpn13 (-)/Rpn10 (+) hepatocytes. All scale bars (white lines), 75 μm.