Ubiquitin-like protein 5 is a novel player in the UPR-PERK arm and ER stress-induced cell death

Abstract

Biological functions of the highly conserved ubiquitin-like protein 5 (UBL5) are not well understood. In Caenorhabditis elegans, UBL5 is induced under mitochondrial stress to mount the mitochondrial unfolded protein response (UPR). However, the role of UBL5 in the more prevalent endoplasmic reticulum (ER) stress-UPR in the mammalian system is unknown. In the present work, we demonstrated that UBL5 was an ER stress-responsive protein, undergoing rapid depletion in mammalian cells and livers of mice. The ER stress-induced UBL5 depletion was mediated by proteasome-dependent yet ubiquitin-independent proteolysis. Activation of the protein kinase R-like ER kinase arm of the UPR was essential and sufficient for inducing UBL5 degradation. RNA-Seq analysis of UBL5-regulated transcriptome revealed that multiple death pathways were activated in UBL5-silenced cells. In agreement with this, UBL5 knockdown induced severe apoptosis in culture and suppressed tumorigenicity of cancer cells in vivo. Furthermore, overexpression of UBL5 protected specifically against ER stress-induced apoptosis. These results identify UBL5 as a physiologically relevant survival regulator that is proteolytically depleted by the UPR-protein kinase R-like ER kinase pathway, linking ER stress to cell death.

Keywords: ER stress; PERK; UBL5; UPR; apoptosis; cell survival; proteasome degradation; ubiquitin-independent proteasome system.

Figure 1

Loss of UBL5 is a general response to ER stress.A, HepG2, Huh7, and primary mouse hepatocytes were treated with TM (2, 0.2, 0.5 μg/ml, respectively), TG (1, 0.1, and 0.25 μM, respectively) for indicated hours or with DTT (1 mM, 3 h in all cell types). In this and the following panels, expression of UBL5 protein was analyzed with tricine protein gel. CHOP was included as a UPR activation marker and β-actin as loading control. B, HepG2 and Huh7 cells were treated with DOX (0.4 μM, 16 h), cisplatin (7.5 μM, 16 h), H₂O₂ (0.4 mM, 24 h), or ethanol (ETOH) (1.5% or 3%, 24 h) before immunoblotting analysis of indicated proteins. C, adult C57/BL6 male mice were i.p. injected with vehicle or with TM (2 mg/kg) (n = 3) to induce pharmacological ER stress. After 26 h, the control and TM-injected mice were sacrificed and liver tissues were collected for protein extraction. The effects of TM treatment on expression of UBL5 protein and multiple UPR activation markers (ATF6, BiP, elF2α-p, elF2α, CHOP, IRE1α-p, and IRE1α) were assessed by immunoblotting. ATF6, activating transcription factor 6; BiP, binding immunoglobulin protein; CHOP, C/EBP homologous protein; DOX, doxorubicin; ER, endoplasmic reticulum; H₂O₂, hydrogen peroxide; IRE1α, inositol-requiring enzyme 1α; TG, thapsigargin; TM, tunicamycin; UBL5, ubiquitin-like protein 5.

Figure 2

ER stress–induced UBL5 degradation is mediated by the UIPS independent of transcriptional regulation.A, HepG2 and Huh7 cells were treated with TM or TG for 6 or 16 h or DTT for 1 or 3 h at doses as for Figure 1A. The effects on UBL5 mRNA were analyzed by RT–qPCR. The results were normalized on that of β-actin and presented as fold change relative to vehicle-treated control cells (defined as 1). The statistical significances were indicated with black and red asterisks to indicate upregulation and downregulation, respectively. B, HepG2 and Huh7 cells were treated with TM or TG for 16 h at doses as for Figure 1A in the presence or the absence of MG132 (1 μM), 3-MA (5 mM), or Baf-A1 (25 nM). The effects of these treatments on UBL5 protein were examined by immunoblotting. C, tag-free wildtype UBL5 (UBL5 wt) or various lysine/arginine (A/R) mutants were transiently transfected into HepG2 or Huh7 cells. Transfected cells were treated with TM for 20 h at doses as for Figure 1A. The TM-induced degradation of wt and mutant UBL5 was examined by immunoblotting. 3-MA, 3-methyladenine; Baf-A1, bafilomycin A1; ER, endoplasmic reticulum; qPCR, quantitative PCR; TG, thapsigargin; TM, tunicamycin; UBL5, ubiquitin-like protein 5; UIPS, ubiquitin-independent proteasome system.

Figure 3

UBL5 degradation lies downstream of the PERK arm of UPR.A, HepG2 and Huh7 cells were treated for 18 h with TM at doses as for Figure 1A in the presence or the absence of the PERK inhibitor GSK2656157 (0.12 μM and 0.25 μM, respectively), the IRE1/XBP1 pathway inhibitor STF083010 (80 μM and 50 μM, respectively), or the ATF6 blocker ceapin-A7 (12 μM for both cell lines). The effects of these treatments on UBL5 and UPR markers were examined by immunoblotting. CHOP, XBP1s (spliced form), and ATF6 cleavage were included to show the efficacies of the respective inhibitors. B, PERK was knocked down by shRNA (left) or knocked out by CRISPR–Cas9 (PERK-KO) (right). The cells were treated with TM for 16 h at doses as for Figure 1A before immunoblotting for the indicated proteins. C, HepG2 and Huh7 cells were treated with the PERK activator CCT020312 (8 μM, 16 h) or vehicle before immunoblotting for the indicated proteins (left). Right, Myc-tagged wt PERK (PERK-WT) or its kinase dead mutant (PERK-K618A) was transfected into HepG2 and Huh7 cells. Cell lysates were prepared 48 h post-transfection and immunoblotted for the indicated proteins. ATF6, activating transcription factor 6; CHOP, C/EBP homologous protein; PERK, protein kinase R–like ER kinase; TM, tunicamycin; UBL5, ubiquitin-like protein 5; UPR, unfolded protein response; XBP1, X-box-binding protein 1.

Figure 4

UBL5 KD activates multiple death pathways and induces severe apoptosis.A, HepG2 and Huh7 cells were costained with an anti-UBL5 antibody and the blue-fluorescent DNA stain DAPI (left). Right, cytoplasmic and unclear proteins were fractionated. Total cellular lysates (total), cytosolic or nuclear fraction isolated from the same number of cells, were analyzed by immunoblotting with GAPDH and Lamin A/C as cytosolic and nuclear markers, respectively. B, HepG2 and Huh7 cells were transfected with Ctrl-siRNA or UBL5 siRNA. After 48 h, RNA was isolated for RNA-Seq analysis. RNA-Seq data from these cells were analyzed with IPA. Statistically significantly upregulated or downregulated protein-coding transcripts (padj < 0.05) were included for analysis and categorized into functionally related groups. Listed were functional groups that were significantly enriched in order of their p value ranges (low to high). The total numbers of affected transcripts in each group are listed in the right column. C, death and survival-related transcripts significantly altered in UBL5 siRNA KD cells were analyzed with the DAVID functional annotation and IPA enrichment programs. The specific death pathways activated or suppressed by siRNA KD in each cell line were shown with heatmaps in order of their Z scores (high to low). The statistical significances of enriched pathways are provided in right. D, HepG2 and Huh7 cells were transfected with UBL5 siRNA or Ctrl-siRNA. The transfected cells were analyzed 3 days post-transfection for viability by MTT staining (upper), quantification of viable cells with trypan blue exclusion (middle), or percentages of apoptosis with Annexin V staining (lower). E, the apoptotic markers (PARP and Cas-3 cleavage) and the effects of UBL5 siRNA on expression of ATF4 and CHOP were examined by immunoblotting. ATF4, activating transcription factor 4; CHOP, C/EBP homologous protein; DAPI, 4′,6-diamidino-2-phenylindole; IPA, ingenuity pathway analysis; KD, knockdown; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; PARP, poly(ADP-ribose) polymerase; UBL5, ubiquitin-like protein 5.

Figure 5

UBL5 stable KD inhibits cell survival and tumorigenicity of cancer cells in vivo.A, expression of UBL5 in HepG2 and Huh7 cells was silenced by UBL5 shRNA. The pooled colonies of puromycin-resistant cells were established as continuous culture. The viability of the cells was analyzed by MTT. The apoptosis of the cells in culture for 2 days was then determined by Annexin V staining. The stable KD of UBL5 protein expression in these cells was confirmed by immunoblotting. B, the ability of the UBL5 shRNA and Ctrl-shRNA transduced cells to form colonies in anchorage-independent conditions was assessed by growing them in soft agar. The colonies formed in soft agar were stained, photographed, quantified, and presented as numbers of colonies/well. C, the UBL5 shRNA and Ctrl-shRNA-transduced HepG2 and Huh7 cells were s.c. injected in male NSG mice (n = 5). Tumor growth curves were constructed from tumor volumes measured at indicated times (days) postinoculation (left). Tumors of each group were excised and photographed at the endpoint when the animals were sacrificed (middle). Shown in right were comparisons of endpoint tumor weights between the UBL5 KD and the corresponding control group. KD, knockdown; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; NSG, NOD scid gamma; UBL5, ubiquitin-like protein 5.

Figure 6

Depletion of UBL5 underlies ER stress–induced apoptosis.A, HepG2 and Huh7 clones expressing exogenous UBL5 and vector control clones were treated for 3 days with TM at doses described for Figure 1A. Overexpression of UBL5, partial depletion of UBL5 by TM treatment, and prevention of PARP and Cas-3 cleavages in UBL5-overexpressing cells were confirmed by immunoblotting (upper). The percentages of apoptosis in these cells were analyzed by Annexin V staining (lower). B, UBL5-overexpressing and control cells were treated with TM for two (HepG2) or three cycles (Huh7). Remaining cells in culture plates were stained with crystal violet, and the staining intensities were quantified. C, UBL5-overexpressing and control cells were incubated with the indicated doses of TM as in (A), and cell viabilities were measured with MTT and presented as percent of activity relative to vehicle control (defined as 100%). As control experiments, these cells were also treated with other death stimuli DOX and H₂O₂ at the indicated concentrations to monitor cell viabilities with MTT assay. DOX, doxorubicin; ER, endoplasmic reticulum; H₂O₂, hydrogen peroxide; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; PARP, poly(ADP-ribose) polymerase; TM, tunicamycin; UBL5, ubiquitin-like protein 5.

Figure 7

Summary of the signaling network from ER stress to inhibition of UBL5 protein stability and prosurvival function. In unstressed state, UBL5 protein is stable and involved in suppression of multiple death pathways contributing to cell survival. During ER stress, the UPR–PERK axis induces proteolytic degradation of UBL5 protein via UIPS, leading to the loss of UBL5 death-suppressive function, a proapoptotic mechanism in parallel with the ER stress–ATF4–CHOP death pathway. ATF4, activating transcription factor 4; CHOP, C/EBP homologous protein; ER, endoplasmic reticulum; UBL5, ubiquitin-like protein 5; UIPS, ubiquitin-independent proteasome system.