Abstract
Objective
UV irradiation of the skin induces photo damage and generates cytotoxic intracellular reactive oxygen species (ROS), activating the unfolded protein response (UPR) to adapt or reduce these UVB-mediated damages. This study was designed to understand the role of the UPR mediator IRE1α in the antioxidant response following UVB irradiation of mouse skin and keratinocytes.
Methods
We used mice with an epidermal deletion of IRE1α and primary mouse keratinocytes to examine effects of UV on different parameters of the antioxidant response in the presence and absence of functional IRE1α.
Results
In the absence of IRE1α, PERK activity and protein levels are significantly compromised following UVB irradiation. Additionally, the loss of IRE1α suppressed phosphorylation of the PERK target, nuclear factor erythroid-2-related factor 2 (NRF2), and NRF2-dependent antioxidant gene expression after UVB irradiation. Interestingly, IRE1α-deficient keratinocytes exhibit elevated basal ROS levels, while a robust ROS induction upon UVB exposure is abolished. Because UVB-induced ROS plays an essential role in regulating skin inflammation, we analyzed recruited immune cell populations and the expression of pro-inflammatory cytokines, Il-6 and Tnfα, in mice with epidermally targeted deletion of Ire1α. Following UVB irradiation, there was significantly less recruitment of neutrophils and leukocytes and reduced expression of pro-inflammatory cytokine genes in the skin of mice lacking IRE1α. Furthermore, keratinocyte proliferation was also significantly reduced after chronic UVB exposure in the skin of these mice.
Conclusion
Collectively, our findings indicate that IRE1α is essential for basal and UVB-induced oxidative stress response, UV-induced skin immune responses, and keratinocyte proliferation.
Significance statement
These findings shed new light on the protective function of IRE1α in the response to UV. IRE1α plays an important role in the regulation of ROS, PERK stability, and antioxidant gene expression in response to UVB in mouse keratinocytes and epidermis.
Introduction
Reactive oxygen species (ROS) caused by UVA and B irradiation play a significant role in cellular damage caused by UV (Masakil et al. 1995, Jurkiewicz & Buettner 1996, Narayanapillai et al. 2012). Since maintenance of the proper oxidative state in the ER is essential for proper protein folding (Kannan & Jaiswal 2006), accumulation of intracellular ROS after UV irradiation can induce an imbalance of oxidoreductase activities, leading to ER stress and activation of the unfolded protein response (Rezvani et al. 2006, Zima & Blatter 2006, Masaki et al. 2009).
Among the ER stress sensory molecules, inositol-requiring enzyme 1α (IRE1α) is an evolutionarily well-conserved dual-functional protein which contains kinase and endoribonuclease (RNase) domains. The RNase domain mediates unconventional splicing of XBP1 mRNA encoding a transcription factor essential for the UPR and can degrade select mRNAs or microRNAs (miRNAs). This process, termed regulated IRE1α-dependent decay (RIDD), reduces ER stress by decreasing translation load in the ER (Chalmers et al. 2019). Activation of the RNase domain is mediated by auto- and cross-phosphorylation of the IRE1α kinase domain (Oslowski et al. 2012, Blazanin et al. 2017). IRE1α acts as both a pro- and anti-apoptotic UPR protein under ROS-mediated stress conditions. As a pro-apoptotic protein, IRE1α suppresses expression of the key ROS regulatory proteins thioredoxin (TRX) 1 and 2 through RIDD degradation of miR-1 that targets Trx 1 and 2 mRNA (Huang et al. 2015). In addition, activation of JNK through the IRE1α–TRAF2–ASK1 complex in response to ROS inhibits mitochondrial activities and promotes further ROS generation and apoptosis (Win et al. 2014). In cirrhotic hepatocytes, XBP1-dependent upregulation of the E3-ubiquitin ligase HRD1 is linked to downregulation of the NRF2-mediated antioxidant response (Wu et al. 2014). Conversely, as an anti-apoptotic protein, IRE1α protects cells from ROS leaks by maintaining ER membrane integrity (Kanekura et al. 2015). Oxidative stress can also directly modify IRE1α’s cysteine 663 residue in the kinase activation loop through sulfenylation (Hourihan et al. 2016), attenuating its IRE1α UPR sensory activity and promoting the SKN-1/Nrf2 antioxidant response.
PERK involvement in the ROS response is more direct compared to IRE1α (Ma, 2013). Under severe ROS-mediated ER stress conditions, the PERK–ATF4–CHOP pathway is activated which intensifies ROS accumulation and induces apoptosis (Marciniak et al. 2004). However, PERK can also directly phosphorylate NRF2 enabling nuclear translocation and activation of ARE-dependent gene expression (Cullinan et al. 2003, Wu et al. 2014).
While both IRE1α and PERK regulate oxidative stress through NRF2 activity (Cullinan et al. 2003, Wu et al. 2014), it remains unclear whether this effect is mediated by interactions between these two UPR proteins. To evaluate crosstalk between these two UPR pathways in the regulation of the oxidative stress response, we examined NRF2 activation in the absence of IRE1α in keratinocytes in vitro and in vivo. Here, we show that IRE1α functions to stabilize PERK and to regulate the NRF2 ROS response after UVB irradiation. In the absence of IRE1α, keratinocytes exhibit reduced ARE gene expression, inflammatory and proliferative responses after UVB-mediated skin damage, leading to elevated cell apoptosis in vitro and in vivo. These results offer new insights into the significance of IRE1α in UV-mediated skin damage, with implications for cancer therapy and prevention.
Materials and methods
Animal husbandry
Mice were housed in microisolator cages and fed a standard chow diet. Seven-week-old age-matched littermate C57/BL6 background mice were used for in vivo UVB irradiation. All mice in this study were treated based on an approved protocol from the Pennsylvania State University Institutional Animal Care and Use Committee (IACUC).
Cell culture
Ire1αΔep primary keratinocytes were obtained from crosses of either the K14Cre (Jackson Laboratory #018964) (Jackson Laboratory) or K14CreERT2 (gift from Dr Arup Indra) (Indra et al. 1999) transgenic line with Ire1αf/f mice (Iwawaki et al. 2009) (gift from Dr Laurie Glimcher lab). All lines were on a C57BL/6 background. Newborns were genotyped with specific primers (see below) to identify control and experimental animals. Keratinocytes were isolated and cultured as described (Blazanin et al. 2017). Conditional deletion of Ire1α in primary keratinocytes (K14CreERT2 × Ire1αf/f) was done using 5 µM 4-hydroxy tamoxifen for 3 days on day 1 of culture.
Immunohistochemistry
At indicated times after UVB irradiation, dorsal skin was isolated and fixed in 4% paraformaldehyde (Sigma-Aldrich). Tissue sections were blocked with 5% normal goat serum (NGS) (VectorLabs, Burlingame, CA, USA) in PBS for 1 h, 37°C, and incubated overnight at 4°C with primary antibody (CD45 (Ebioscience, San Diego, CA, USA): 1:250 dilution, Ki67 (Cell Signaling Technology): 1:100 dilution, Gr-1 (BioLegend, San Diego, CA, USA): 1:250 dilution in 1% NGS). After washing samples were incubated with biotinylated secondary antibody (anti-rabbit 1:500, VectorLabs) for 1 h at room temperature followed by Streptavidin-HRP (1:500, Jackson ImmunoResearch) at room temperature for 1 h. DAB (VectorLabs) was used as an HRP substrate and hematoxylin (VectorLabs) was used for nuclear counterstain. Staining was visualized with Olympus BX43 light microscope.
UVB irradiation
Seven-week-old age-matched littermates were shaved with a hair clipper and depilated with Nair 48 h prior to exposure to 360 mJ/cm2 UVB with a UVC-cutout cellulose triacetate film (Kodak) covered UV lamp (American Ultraviolet Light Co., Lebanon, IN, USA). This device was fabricated based on a previous study (Ravindran et al. 2014) to produce 280–320 nm wavelength UVB light. Irradiated UVB energy was measured with a UVX radiometer (Analytik Jena US LLC, Upland, CA, USA). During UVB irradiation, mice were sedated using an intraperitoneal injection of avertin solution (EMD Millipore, 25 mg/mL). Keratinocytes were cultured for 4 days before UVB irradiation. Primary keratinocytes were irradiated with UVB using a Cl-1000 ultraviolet crosslinker (Analytik Jena). Cell culture medium was removed and replaced with PBS during irradiation. Removed media was centrifuged to get rid of cell debris and reapplied to cells.
Epidermal protein and RNA extraction
Mouse dorsal skin was shaved with hair clippers, depilated with Nair, and removed with scissors and forceps. Excised mouse dorsal skin was flattened on an ice-chilled glass plate. A razor blade equipped scraper (Stanley Black & Decker) was used for scraping off the epidermal layer of mouse skin. Five to eight strokes with the scraper on skin successfully separated out the epidermal layer of skin. Scraped epidermis was transferred with clean fine-tip spatula from the razor blade to a microfuge containing RIPA buffer for total protein extraction and into Ribozol (Amresco) for total RNA extraction after UVB irradiation.
Whole-cell protein lysates and immunoblotting
Sample harvest and immunoblotting were performed as described (Son et al. 2021). Primary antibodies were from Cell Signaling Technology (phosphorylated and total PERK, phosphorylated and total eIF2α, IRE1α, actin, cleaved caspase 3, tubulin, and lamin A/C) or GeneTex (phosphorylated and total Nrf2). Actin was used as a loading control for whole cell lysate. For nuclear fractionation samples, tubulin and lamin A/C were used as a loading control for cytosolic and nuclear fractions. Most proteins were transferred on nitrocellulose membrane (Bio-Rad Laboratories).
Quantitative real-time polymerase chain reaction (qRT-PCR)
qRT-PCR was performed with PerfeCTaTM SYBR® Green SuperMix using an iQ5 (Bio-Rad) as described (Son et al. 2021). The calculated starting amount of each gene was normalized to starting amount of β-actin RNA levels.
The following primers were used for qPCR analysis:
Ho-1 forward primer: 5′-GCCGAGAATGCTGAGTTCATG-3′
Ho-1 reverse primer: 5′-TGGTACAAGGAAGCCATCACC-3′
Gclm forward primer: 5′-GCCACCAGATTTGACTGCCTTTG-3′
Gclm reverse primer: 5′-TGCTCTCGATGACCGAGTACC-3′
Nqo1 forward primer: 5′-AGGGTTCGGTATTACGATCC-3′
Nqo1 reverse primer: 5′-AGTACAATCAGGGCTCTTCTCG-3′
Tnfα forward primer: 5′-GATTATGGCTCAGGGTCCAA-3′
Tnfα reverse primer: 5′-GAGACAGAGGCAACCTGACC-3′
Il-6 forward primer: 5′-AACCGCTATGAAGTTCCTCTCTGC-3′
Il-6 reverse primer: 5′-TAAGCCTCCGACTTGTGAAGTGGT-3′
β-Actin forward primer: 5′-ACCAACTGGGACGATATGGAGAAGA-3′
β-Actin reverse primer: 5′-TACGACCAGAGGCATACAGGGACAA-3′
Flow cytometry
Skin leukocytes were isolated and flow cytometry was performed as described previously (Podolsky et al. 2017). Skin leukocytes were isolated from shaved mice back skin. Cells were incubated with Fc blocker α-CD16/32 (BioLegend) and stained with desired fluorescence-conjugated antibodies. The amples were analyzed with the LSR Fortessa Cytometer (BD Biosciences). Fluorescence-conjugated antibodies used for flow cytometry were rat α-CD3 (17A2, BioLegend), rat α-CD11b (M1/70, BioLegend), α-rat CD45 (30-F11, Ebioscience), rat α-Ly6C (AL-21, BD Biosciences), rat α-Ly6G (1A8, BioLegend), and rat α-F4/80 (BM8, BioLegend).
Nuclear fractionation
Cytosolic and nuclear protein fractionation was performed as described (Hogan et al. 2013). Cells were scrape-harvested on ice with hypotonic lysis method. Cells were harvested with buffer A (0.33 M sucrose, 10 mM HEPES pH 7.4, 1 mM MgCl2, 0.1% Triton X-100). The samples were centrifuged, and the separated supernatants were the cytosolic fraction. The pellets were resuspended with buffer B (0.45 M NaCl, 10 mM HEPES pH 7.4, 1 mM MgCl2) and centrifuged to remove debris. The separated supernatants were the nuclear fraction.
Quantitation and statistical analysis
Qupath software (Peter et al. 2017) was used to calculate positive cells from immunohistochemistry (H-DAB positive cell analysis tab) and mean fluorescence intensity measurements (fluorescence image cell analysis tab-cell: green mean value) from 10× and 20× magnification images. Cell size and boundary were automatically determined by the software. The threshold for detection was determined with No UV control images (threshold values: 5–10). More than three fields per sample or section were quantitated and at least 150 cells per field were analyzed in duplicates for each group at the same UVB exposure time. Fluorescence IIF images were selected with the rectangular annotation tool and analyzed by the cell detection function in the analysis tab with the desired analysis color. The software determines the total positive cell number and mean fluorescence intensity in the annotated area. The skin IHC pictures were selected with the brush annotation tool to select desired skin area and selected analysis tab-H-DAB positive cell detection. The software determines the total DAB positive cell numbers in the annotated area. Statistical significance was determined using a Student’s t-test or two-way ANOVA, with significance determined as P ≤ 0.05. The asterisks (*) indicate significant differences compared to the same timepoint of control group observance (*P ≤ 0.05, **P ≤ 0.005, ***P ≤ 0.0005).
Results
IRE1α regulates PERK protein stability under stress conditions through the proteasome pathway
We generated mice with epidermally deleted IRE1α (Ire1αΔep) by crossing K14Cre and Ire1αf/f mice. We previously showed that IRE1α deficiency sensitized the epidermis to UVB irradiation and caused elevated basal ROS levels in keratinocytes (Son et al. 2021). Since PERK is important in the oxidative stress response, we determined the effect of IRE1α deficiency on UV-induced PERK activation in the skin of adult Ire1αΔep and control mice as well as primary newborn keratinocytes. 48 h following UVB irradiation, there was a significant reduction in levels of total PERK in scraped epidermal protein from adult Ire1αΔep mice but not control mice (Fig. 1A and B). In primary cultures of Ire1αf/f and Ire1αΔep keratinocytes, no significant difference in basal levels of PERK were observed. However, following UVB irradiation, Ire1αΔep keratinocytes exhibited a 40% greater reduction in PERK protein compared to Ire1αf/f(Fig. 1C and D). Additionally, there was a significant reduction in the phosphorylation of eIF2α, a direct target of PERK (Owen et al. 2005) (Fig. 1C). Treatment with the proteasome inhibitor MG132 partially rescued the UVB-induced reduction in PERK protein in the Ire1αΔep keratinocytes, indicating the involvement of 26S-mediated proteasomal degradation (Fig. 1C and D). Moreover, MG132 treatment increased levels of p-eIF2α after UVB irradiation in the Ire1αΔep keratinocytes (Fig. 1C). To determine whether PERK instability in the absence of IRE1α was a UVB-specific response or general ER stress related response, we induced ER stress with a low concentration (5 nM) of the SERCA inhibitor thapsigargin (TG). Similar to UVB, TG caused a reduction in total PERK protein levels in Ire1αΔep keratinocytes, such that it was nearly undetectable after 4 h increasing at 6 h post-UVB (Supplementary Fig. 2A and B, see section on supplementary materials given at the end of this article). While MG32 had little effect on PERK levels in control keratinocytes, levels of total PERK were increased by MG132 at all timepoints in the Ire1αΔep keratinocytes although the reduction was not prevented. As ER-associated 26s proteasomal degradation is mediated by the ERAD pathway (Hegde & Ploegh 2010, Scheper & Hoozemans 2015), we investigated the impact of the ERAD inhibitor eeyarestatin (Eer1) on the TG-induced reduction of PERK. As expected, TG activated ER stress in both Ire1αΔep and control cells as evidenced by induction of the ER chaperone BiP (Supplementary Fig. 2C). Unlike Mg132, Eer1 did not increase the baseline levels of total PERK in Ire1αΔep keratinocytes or prevent the TG-induced reduction in total PERK levels. Instead, the Tg-induced reduction was enhanced (Supplementary Fig. 2C). Furthermore, in Ire1αΔep keratinocytes treated with Eer1 alone the mobility of PERK, presumably reflecting levels of phosphorylation, was lower compared to control keratinocytes although this increased as expected with Tg treatment. The mechanism underlying this altered mobility of PERK is not clear. Nevertheless, the lack of effect of Eer1 indicates that PERK degradation in IRE1α deficient keratinocytes is not mediated by the ERAD pathway. We previously showed that in keratinocytes IRE1α deficiency causes elevated [Ca2+]i and ROS through InsP3R-mediated calcium release (Son et al. 2021). To determine if elevated [Ca2+]i or ROS contributed to PERK degradation in the absence of IRE1α, we pretreated keratinocytes with 2ABP (10 μM), an InsP3R inhibitor, or NAC (2.5 mM) to quench ROS prior to exposure to either UVB or TG. However, neither inhibition of elevated [Ca2+]i nor ROS quenching reversed the enhanced degradation of PERK in the absence of IRE1α (Supplementary Fig. 2D). Together these results show that IRE1α is important in stabilizing PERK protein and maintaining its eIF2α kinase activity from 26S proteasomal degradation under distinct stress conditions.
Control (Ire1αf/f) and Ire1αΔep primary keratinocytes or mice were irradiated with UVB at 30 mJ/cm2 and 360 mJ/cm2, respectively, and harvested at the indicated timepoints. (A) PERK protein expression was analyzed in epidermal scrapings using immunoblotting with beta-actin used as a loading control. Each lane represents one mouse. (B) Quantification of PERK expression in 48-h post-UVB samples by densitometric analysis. The statistical analysis was done using the Student’s t-test. Values = mean ± s.d. *Indicates a significant difference (*** P ≤ 0.0005). (C) Immunoblot analysis of indicated UPR proteins in primary keratinocytes after UVB irradiation. Primary keratinocytes were cultured 4 days before irradiation and then harvested at the indicated times. Cells were pre-treated with DMSO as a vehicle (0.1%, v/v) or MG132 (30 μM) 2 h prior to UVB irradiation. Beta-actin was used as a loading control. (D) Quantitation of PERK expression by immunoblotting. Densitometric signal from western blots with three biological repeats (n = 3). The PERK expression values were normalized with beta-actin expression values. The calculated PERK/Actin values were normalized with UVB-treated control cell values to convert to fold change. All calculated values were graphed with GraphPad Prism 5.0 software.
Citation: Redox Experimental Medicine 2024, 1; 10.1530/REM-23-0030
Loss of IRE1α suppresses PERK mediated NRF2 activation and oxidative stress response
We used CellROX Red to measure ROS levels in control and Ire1αΔep keratinocytes. In the absence of IRE1a there was a significant elevation in basal ROS levels (Fig. 2A and B). As expected UVB caused a significant increase in intracellular levels in wildtype primary keratinocytes. However, in keratinocytes lacking IRE1α, the elevated basal ROS levels were not further increased by UVB and were less than the induced level in wildtype cells (Fig. 2A and B). Since NRF2 is a substrate of the PERK kinase, we examined whether the oxidative stress response was defective in Ire1αΔep keratinocytes. In control keratinocytes, NRF2 was induced 3 h post-UVB, followed by a gradual decline in protein levels. However, in IRE1α-deficient keratinocytes, NRF2 induction was lower than in control cells, and the decrease was more pronounced (Fig. 2C and D). In tamoxifen inducible IRE1α knockout keratinocytes (Ire1αERT2Δep), there was no induction of NRF2 6 h after UVB, and by 16 h, NRF2 levels were lower than untreated keratinocytes without Ire1α deletion (Fig. 2E). NRF2 is phosphorylated by PERK, a mechanism that enhances NRF2 nuclear translocation and can activate anti-oxidative stress genes, such as NQO1, HO1,and GCLM (Ma, 2013). Although some nonspecific bands were detected in the cytoplasmic fraction, in control keratinocytes there was little change in total NRF2 in the cytoplasmic or nuclear fractions after UVB (Fig. 2F). p-NRF2 was not detected in the cytoplasmic fraction but 16 h post-UVB, nuclear p-NRF2 increased in the control Ire1αf/f keratinocytes. Surprisingly, under basal conditions nuclear p-NRF2 was higher in the Ire1αΔep keratinocytes possibly reflecting the elevation of ROS levels but the levels of both total and p-NRF2 in the nuclear fraction decreased after UVB (Fig. 2F). These data suggest that NRF2 protein stability and nuclear translocation after UVB is dependent on IRE1α. To assess the functional consequences of the difference in nuclear p-NRF2, we examined expression of the NRF2 target genes Ho-1, Gclm, Gclc, and NQO1, which are crucial for the oxidative stress response. Unexpectedly there was no difference in basal expression of Gclm, Gclc, and NQO1 between Ire1αf/f and Ire1αΔep keratinocytes despite elevated levels of nuclear p-NRF2 in the Ire1αΔep keratinocytes. However, in control keratinocytes, UVB induced expression of each from 3- to 12-fold, but this induction was dampened significantly in Ire1αΔep keratinocytes (Fig. 2G). To test if IRE1α deficiency altered the oxidative stress response in vivo, protein and RNA were isolated from epidermal scrapings from 7/8-week-old Ire1αf/f to Ire1αΔep mice that were exposed to 360 mJ/cm2 UVB. In the absence of IRE1α, NRF2 was nearly undetectable 6 h after UVB and remained two-fold lower than control by 24 h (Fig. 3A and B). Paralleling the reduction in NRF2, the expression of Ho-1, Nqo1, and Gclm was also significantly suppressed (two-fold) in the epidermis of Ire1αΔep mice 24 h after UVB irradiation (Fig. 3C). Together, these data show that IRE1α deficiency in keratinocytes impacts the oxidative stress response both in vitro and in vivo.
IRE1α is important for NRF2 and ARE gene regulation after UVB irradiation in primary keratinocytes. Control and Ire1αΔep primary keratinocytes were irradiated with UVB (30 mJ/cm2) 4 days after culturing and analyzed at the indicated timepoints. (A) ROS levels after UVB irradiation were determined by CellROX Redtm staining, and (B) DAPI counterstaining was used to determine the total cell number. Percent ROS-positive cells were calculated as CellROX Redtm-positive cells over total cells and graphed. A Student’s t-test was used for statistical analysis between control and UVB-treated cells. (C) Western blot analysis of IRE1α and NRF2 expression in control and Ire1αΔep primary keratinocytes after UVB irradiation. Beta-actin was used as a loading control. (D) The NRF2 expression values were normalized with Actin expression values. The calculated NRF2/Actin values were normalized with UVB-treated control cell values to convert to a fold change. (E) NRF2 levels in Ire1αERT2Δep primary keratinocytes pretreated with or without 4-OHT (TAM, 100 μM) for 4 days prior to UVB irradiation. (F) NRF2 and p-NRF2 levels in the nuclear and cytoplasmic fractions of control and Ire1αΔep primary keratinocytes at the indicated times after UVB irradiation. *Indicates NRF2 and p-NRF2. Tubulin is used as a loading control for the cytoplasmic fraction, and Lamin A/C is used as a nuclear fraction loading control. (G) Quantitative RT-PCR analysis for indicated oxidative stress response genes was performed with three biological replicates (n = 3) of sample cDNA and two technical repeats. Values = mean ± s.e.m. (n = 3). All calculated values were graphed with GraphPad Prism 5.0 software. *Indicates a significant difference (***P ≤ 0.0005).
Citation: Redox Experimental Medicine 2024, 1; 10.1530/REM-23-0030
IRE1α is important for in vivo NRF2 and ARE gene regulation after UVB irradiation. (A) Immunoblot analysis of NRF2 and IRE1α expression in epidermal scrapings of UVB-irradiated mice. Each lane represents one mouse. (B) The densitometric signal from NRF2 western blots was quantitated with ImageJ (NIH). The NRF2 expression values were normalized with beta-actin expression values. (C) Expression of the indicated oxidative stress response genes was determined by quantitative RT-PCR of total RNA isolated from the scraped epidermis of UVB-irradiated mice. Quantitative RT-PCR values of the indicated genes were normalized to beta-actin expression values. All calculated values were graphed with GraphPad Prism 5.0 software. The statistical analysis was done by a two-way ANOVA (P < 0.05) with a Bonferroni post-test. Values = mean ± s.e.m. (n = 4). *Indicates a significant difference (*P ≤ 0.05, **P ≤ 0.005, ***P ≤ 0.0005).
Citation: Redox Experimental Medicine 2024, 1; 10.1530/REM-23-0030
IRE1α is required for immune cell recruitment to UVB-mediated skin after acute UVB irradiation
UVB-mediated ROS can induce the expression and secretion of proinflammatory cytokines such as IFNγ, IL-8, and IL-6 through several signaling pathways (Davis et al. 2011, Quist et al. 2016, Piipponen et al. 2020, Fitsiou et al. 2021, Frommeyer et al. 2022). This, in turn, contributes to recruitment of neutrophils and other inflammatory cells to the skin in response to UVB. We tested whether the defective oxidative stress response in IRE1α-deficient keratinocytes affected the inflammatory response to UVB. As expected, the expression of Tnfa was elevated in the epidermis 48 h after UVB irradiation, but there was no difference in expression between Ire1αΔepand Ire1αf/fmice. Surprisingly, after UVB exposure, the expression of Il-6 was significantly repressed in Ire1αΔep compared to the Ire1αf/fcontrol. Consistent with these results, immunohistochemistry with a pan-leukocyte marker (CD-45) revealed a two-fold reduction in CD45+ cells throughout all time points after UV irradiation (Fig. 4B and C). GR-1+ cells were also lower in Ire1αΔep skin compared to control at 72 h after UVB irradiation (Fig. 4B). This was further quantitated by flow cytometry of leukocytes isolated from skin of control and irradiated mice. Although there was an expected increase in the percentages of GR-1+ leukocytes in both genotypes after UVB, the increase in Ire1αΔep skin was 2.5-fold lower compared to controls (Fig. 4D).
IRE1α is important for immune cell recruitment and cytokine expression. (A) Quantitative RT-PCR analysis of cytokine gene expression in total RNA isolated from scraped epidermis of UVB-irradiated back skin. Each quantitative RT-PCR value of the indicated genes was normalized with a beta-actin expression value. (B) Immunohistochemical staining of CD45+ and GR-1+ cells in the skin of mice after UVB irradiation (360 mJ/cm2). Sections were counterstained with hematoxylin. Bar = 50 μm (C) Quantitation of CD45+ cells in skin using QuPath software H-DAB-positive cell analysis. (D) Percent neutrophils in skin after UVB irradiation determined using flow cytometry; each histogram represents the average of three mice. (E) Epidermal proliferation measured with Ki67 immunohistochemistry at indicated timepoint after UVB (360 mJ/cm2). Bar = 20 μm (F) Quantitation of Ki67 staining using QuPath software H-DAB-positive cell analysis. (G) Immunoblot analysis of cleaved caspase 3 in epidermal scrapings of chronic UVB-exposed skin from control (Ire1αf/f) and Ire1αΔep UVB-irradiated mice beta-actin was used as loading control. Each lane represents one mouse (no UV: Ire1αf/fn = 3, Ire1αΔepn = 3; UVB 6 weeks: Ire1αf/fn = 6, Ire1αΔepn = 4). All calculated values were graphed with GraphPad Prism 5.0 software. The statistical analysis was done using the Student’s t-test. Values = mean ± s.d. *Indicates significant difference (P ≤ 0.05).
Citation: Redox Experimental Medicine 2024, 1; 10.1530/REM-23-0030
We previously demonstrated that epidermal deletion of IREα led to increased keratinocyte apoptosis after UVB irradiation, both in vivo and in vitro, although there was no significant difference in cell proliferation in the absence of IRE1α (Son et al. 2021) (Supplementary Fig. 1). Following chronic UVB irradiation (6 weeks), there were significantly more epidermal Ki67 positive cells in control skin compared to Ire1αΔep skin (Fig. 4E and F). In contrast, there were higher levels of cleaved caspase 3 in the epidermal samples of Ire1αΔep mice compared to control mice (Fig. 4G). These data suggest that IRE1α is necessary for both proliferation and survival of keratinocytes after exposure to chronic UV irradiation.
Discussion
It has become increasingly clear that there is significant crosstalk both positive and negative between antioxidant response and ER stress response pathways with important implications for cell health and chronic disease (Nakajima & Kitamura 2013, Wu et al. 2014, Ahmed et al. 2017). The UVB-mediated ROS are involved in more than half of UVB-mediated damage events in keratinocytes (Heck et al. 2003, Dinkova-Kostova & Talalay 2008). The generation of ROS in the skin by UVB irradiation has a significant impact on cellular damage, stress signaling pathways, and expression of inflammatory cytokines that can ultimately contribute to skin cancer. We previously showed that IRE1α through its ability to control calcium release from the ER by the Ins3PR is essential for maintenance of cellular ROS homeostasis (Son et al. 2021). In the absence of IRE1α, both intracellular calcium and basal ROS levels are elevated (Son et al. 2021). Since NRF2 is a central regulator of the oxidative stress response, we examined the functional consequences of IRE1α loss on events upstream and downstream of NRF2 activation. In the absence of IRE1α, basal levels of nuclear p-NRF2 were increased, likely linked to elevated basal ROS although surprisingly there was no increased expression of ARE genes, suggesting additional mechanisms important for regulation of these genes. However, UVB-induced NRF2 phosphorylation and nuclear accumulation are impaired in vitro, while NRF2 stability is reduced in vivo. In both settings, the induction of ARE target genes by UVB is significantly diminished in the absence of IRE1α. These results support the concept that IRE1α is essential for the NRF2-mediated antioxidant response initiated by UVB exposure and are in line with a previous study showing that ROS-mediated IRE1α sulfenylation causes NRF2 stabilization and activation of ARE gene expression (Hourihan et al. 2016). In contrast, studies in the liver have shown that IRE1α promotes NRF2 ERAD degradation through XBP1-induced HRD1 ubiquitin ligase expression (Wu et al. 2014). Under normal conditions, NRF2 forms a complex with its negative regulator, the E3 ubiquitin ligase KEAP1, promoting ubiquitination and proteasomal degradation. However, oxidative stress causes dissociation of KEAP1, leading to the activation of NRF2 transcriptional activity (Ahmed et al. 2017). NRF2 is also a direct target of the PERK kinase, and phosphorylation by PERK and other kinases leads to nuclear localization and activation of ARE target genes. We found that UV activation of PERK is compromised in the absence of IRE1α, partly due to decreased PERK protein stability through 26S proteasome-mediated degradation. This response is not unique to UV, as the ER stress inducer thapsigargin also reduces PERK stability in the absence of IRE1α (Supplementary Fig. 2). These results suggest that IRE1α is crucial in maintaining PERK stability in response to various cellular stressors. Although the mechanism is unclear, it is independent of ER calcium release, elevated ROS, or the ERAD pathway. In the case of UV, our data suggest that this crosstalk is important for PERK-mediated phosphorylation of NRF2 and activation of the antioxidant response.
Inflammation is a well-documented response of the skin to UVB irradiation (Hruza & Pentland 1993, D’orazio et al. 2013, Ansary et al. 2021, Salminen et al. 2022), and UVB can induce the release of pro-inflammatory cytokines, such as IL-1β, IL-6, IL-8, and TNFα from HaCaT cells, a human keratinocyte cell line (Yoshizumi et al. 2008, Quist et al. 2016). Our results show that acute induction of the pro-inflammatory cytokine Il-6 and recruitment of CD45+ leukocytes and neutrophils into the skin in response to UVB is dependent on epidermal IRE1α. Furthermore, we found that chronic UV-induced epidermal proliferation is suppressed, and apoptosis is increased in the absence of IRE1α. Both PERK and IRE1α have been linked to inflammatory signaling through NF-κB. Phosphorylation of eIF2α by PERK, with subsequent translational suppression of IκBα synthesis, was shown to be necessary and sufficient to activate NF-κB signaling (Deng et al. 2004). Conversely, in response to ER stress, IRE1α, in a complex with TRAF2, recruits IkKa, leading to the phosphorylation and degradation of IκBα and induction of TNFa (Hu et al. 2006). Other studies have implicated IRE1α in ER-calcium-dependent cytokine secretion during airway epithelial inflammation (Martino et al. 2009). ROS has also been linked to inflammation through the activation of NFkB (Nakajima & Kitamura 2013), and conversely, NRF2 inhibits NFκB through induction of ARE target genes such as HO-1 (Ahmed et al. 2017). Together, these results suggest that the impairment of pro-inflammatory cytokine induction and recruitment of neutrophils after UV exposure in the absence of IRE1α may be independent of IRE1α effects on oxidative stress signaling although more studies are needed.
The IRE1α-dependent regulation in UV-damaged skin might affect skin repair and the initiation of UV-mediated skin cancer. Our data, showing increased cleaved caspase 3 and reduced Ki67 staining in Ire1αΔep mice as compared to control mice epidermis, imply defective cellular proliferation for damage repair in Ire1αΔep skin. To further understand the function of IRE1α in UV-damaged skin repair, it is necessary to examine the wound repair process, including immuno-histological cell proliferation assay and pro-angiogenic factor expression profiling, after chronic UVB irradiation in both control and Ire1αΔep skin.
Together, our findings suggest that IRE1α can modulate UVB-mediated skin damage and damage repair through regulating intracellular ROS, skin immune responses, and pro-angiogenic responses. Although further studies are required to reveal the detailed mechanism of IRE1α-dependent UV response, these data shed new light on the potential function of IRE1α in UVB-mediated skin carcinogenesis and repair.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/REM-23-0030.
Data availability statement
No datasets were generated or analyzed during the current study.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the study reported.
Funding
This work was supported by R01 CA197942 to ABG, and the USDA National Institute of Food and Agriculture Federal Appropriations Project PEN4772, Accession number: 7000371.
Author contribution statement
JS and ABG designed the study; JS, JTB, and SW performed data collection; JTB and SW contributed to procuring new reagents/analytic tools and supported data analysis; and JS and ABG wrote the manuscript
Acknowledgements
The authors thank the the Microscopy Facility of the Huck Institutes of the Life Sciences and Flow Cytometry Core. The authors acknowledge Dr Laurie Glimcher and Dr Arup Indra for providing animals and Dr Stuart Yuspa for his continued support.
References
Ahmed SMU, Luo L, Namani A, Wang XJ & & Tang X 2017 Nrf2 signaling pathway: pivotal roles in inflammation. Biochimica et Biophysica Acta. Molecular Basis of Disease 1863 585–597. (https://doi.org/10.1016/j.bbadis.2016.11.005)
Ansary TM, Hossain MR, Kamiya K, Komine M & & Ohtsuki M 2021 Inflammatory molecules associated with ultraviolet radiation-mediated skin aging. International Journal of Molecular Sciences 22. (https://doi.org/10.3390/ijms22083974)
Blazanin N, Son J, Craig-Lucas AB, John CL, Breech KJ, Podolsky MA & & Glick AB 2017 ER stress and distinct outputs of the IRE1alpha RNase control proliferation and senescence in response to oncogenic Ras. PNAS 114 9900–9905. (https://doi.org/10.1073/pnas.1701757114)
Chalmers F, Mogre S, Son J, Blazanin N & & Glick AB 2019 The multiple roles of the unfolded protein response regulator IRE1α in cancer. Molecular Carcinogenesis 58 1623–1630. (https://doi.org/10.1002/mc.23031)
Cullinan SB, Zhang D, Hannink M, Arvisais E, Kaufman RJ & & Diehl JA 2003 Nrf2 is a direct PERK substrate and effector of PERK-dependent cell survival. Molecular and Cellular Biology 23 7198–7209. (https://doi.org/10.1128/MCB.23.20.7198-7209.2003)
Davis BK, Wen H & & Ting JPY 2011 The inflammasome NLRs in immunity, inflammation, and associated diseases. Annual Review of Immunology 29 707–735. (https://doi.org/10.1146/annurev-immunol-031210-101405)
Deng J, Lu PD, Zhang Y, Scheuner D, Kaufman RJ, Sonenberg N, Harding HP & & Ron D 2004 Translational repression mediates activation of nuclear factor kappa B by phosphorylated translation initiation factor 2. Molecular and Cellular Biology 24 10161–10168. (https://doi.org/10.1128/MCB.24.23.10161-10168.2004)
Dinkova-Kostova AT & & Talalay P 2008 Direct and indirect antioxidant properties of inducers of cytoprotective proteins. Molecular Nutrition and Food Research 52(Supplement 1) S128–S138. (https://doi.org/10.1002/mnfr.200700195)
D’orazio J, Jarrett S, Amaro-Ortiz A & & Scott T 2013 UV radiation and the skin. International Journal of Molecular Sciences 14 12222–12248. (https://doi.org/10.3390/ijms140612222)
Fitsiou E, Pulido T, Campisi J, Alimirah F & & Demaria M 2021 Cellular senescence and the senescence-associated secretory phenotype as drivers of skin photoaging. Journal of Investigative Dermatology 141 1119–1126. (https://doi.org/10.1016/j.jid.2020.09.031)
Frommeyer TC, Gilbert MM, Brittain GV, Wu T, Nguyen TQ, Rohan CA & & Travers JB 2022 UVB-induced microvesicle particle release and its effects on the cutaneous microenvironment. Frontiers in Immunology 13 880850. (https://doi.org/10.3389/fimmu.2022.880850)
Heck DE, Vetrano AM, Mariano TM & & Laskin JD 2003 UVB light stimulates production of reactive oxygen species: unexpected role for catalase. Journal of Biological Chemistry 278 22432–22436. (https://doi.org/10.1074/jbc.C300048200)
Hegde R & & Ploegh H 2010 Quality and quantity control at the endoplasmic reticulum. Current Dermatology Reports 20 437–446. . (https://doi.org/10.1016/j.ceb.2010.05.005)
Hogan KA, Ravindran A, Podolsky MA & & Glick AB 2013 The TGFβ1 pathway is required for NFκB dependent gene expression in mouse keratinocytes. Cytokine 64 652–659. (https://doi.org/10.1016/j.cyto.2013.09.004)
Hourihan JM, Mazzeo LEM, Fernández-Cárdenas LP & & Blackwell TK 2016 Cysteine sulfenylation directs IRE-1 to activate the SKN-1/Nrf2 antioxidant response. Molecular Cell 63 553–566. (https://doi.org/10.1016/j.molcel.2016.07.019)
Hruza LL & & Pentland AP 1993 Mechanisms of UV-induced inflammation. Journal of Investigative Dermatology 100 35S–41S. (https://doi.org/10.1111/1523-1747.ep12355240)
Hu P, Han Z, Couvillon AD, Kaufman RJ & & Exton JH 2006 Autocrine tumor necrosis factor alpha links endoplasmic reticulum stress to the membrane death receptor pathway through IRE1α-mediated NF-κB activation and down-regulation of TRAF2 expression. Molecular and Cellular Biology 26 3071–3084. (https://doi.org/10.1128/MCB.26.8.3071-3084.2006)
Huang Q, Zhou HJ, Zhang H, Huang Y, Hinojosa-Kirschenbaum F, Fan P, Yao L, Belardinelli L, Tellides G, Giordano FJ, et al.2015 Thioredoxin-2 inhibits mitochondrial reactive oxygen species generation and apoptosis stress kinase-1 activity to maintain cardiac function. Circulation 131 1082–1097. (https://doi.org/10.1161/CIRCULATIONAHA.114.012725)
Indra AK, Warot X, Brocard J, Bornert JM, Xiao JH, Chambon P & & Metzger D 1999 Temporally-controlled site-specific mutagenesis in the basal layer of the epidermis: comparison of the recombinase activity of the tamoxifen-inducible Cre-ER(T) and Cre-ER(T2) recombinases. Nucleic Acids Research 27 4324–4327. (https://doi.org/10.1093/nar/27.22.4324)
Iwawaki T, Akai R, Yamanaka S & & Kohno K 2009 Function of IRE1 alpha in the placenta is essential for placental development and embryonic viability. PNAS106 16657–62. (https://doi.org/10.1073/pnas.0903775106)
Jurkiewicz BA & & Buettner GR 1996 EPR detection of free radicals in UV-lrradiated skin: Mouse versus Human. Photochemistry and Photobiology 64 918–922. (https://doi.org/10.1111/j.1751-1097.1996.tb01856.x)
Kanekura K, Ma X, Murphy JT, Zhu LJ, Diwan A & & Urano F 2015 IRE1 prevents endoplasmic reticulum membrane permeabilization and cell death under pathological conditions. Science Signaling 8 ra62. (https://doi.org/10.1126/scisignal.aaa0341)
Kannan S & & Jaiswal AK 2006 Low and high dose UVB regulation of transcription factor NF-E2-related factor 2. Cancer Research 66 8421–8429. (https://doi.org/10.1158/0008-5472.CAN-06-1181)
Ma Q 2013 Role of Nrf2 in oxidative stress and toxicity. Annual Review of Pharmacology and Toxicology 53 401–426. (https://doi.org/10.1146/annurev-pharmtox-011112-140320)
Marciniak SJ, Yun CY, Oyadomari S, Novoa I, Zhang Y, Jungreis R, Nagata K, Harding HP & & Ron D 2004 CHOP induces death by promoting protein synthesis and oxidation in the stressed endoplasmic reticulum. Genes and Development 18 3066–3077. (https://doi.org/10.1101/gad.1250704)
Martino MEB, Olsen JC, Fulcher NB, Wolfgang MC, O’neal WK & & Ribeiro CMP 2009 Airway epithelial inflammation -induced endoplasmic reticulum Ca2+ store expansion is mediated by X-box binding protein-1. Journal of Biological Chemistry 284 14904–14913. (https://doi.org/10.1074/jbc.M809180200)
Masaki H, Izutsu Y, Yahagi S & & Okano Y 2009 Reactive oxygen species in HaCaT keratinocytes after UVB irradiation are triggered by intracellular Ca 2 þ levels. Journal of Investigative Dermatology. Symposium Proceedings 14 50–52. (https://doi.org/10.1038/jidsymp.2009.12)
Masakil H, Atsumi T & & Sakurai H 1995 Detection of hydrogen peroxide and hydroxyl radicals in murine skin fibroblasts under UVB irradiation. Biochemical and Biophysical Research Communications 206 474–479. (https://doi.org/10.1006/bbrc.1995.1067)
Nakajima S & & Kitamura M 2013 Bidirectional regulation of NF-kappaB by reactive oxygen species: a role of unfolded protein response. Free Radical Biology and Medicine 65 162–174. (https://doi.org/10.1016/j.freeradbiomed.2013.06.020)
Narayanapillai S, Agarwal C, Tilley C & & Agarwal R 2012 Silibinin is a potent sensitizer of UVA radiation-induced oxidative stress and apoptosis in human keratinocyte HaCaT cells. Photochemistry and Photobiology 88 1135–1140. (https://doi.org/10.1111/j.1751-1097.2011.01050.x)
Oslowski CM, Hara T, O’sullivan-Murphy B, Kanekura K, Lu S, Hara M, Ishigaki S, Lihua H, Hayashi E, Hui ST, et al.2012. Thioredoxin-interacting protein mediates ER stress-induced β cell death through initiation of the inflammasome. Cell Metabolism 16 265–273. (https://doi.org/10.1016/j.cmet.2012.07.005)
Owen CR, Kumar R, Zhang P, Mcgrath BC, Cavener DR & & Krause GS 2005 PERK is responsible for the increased phosphorylation of eIF2alpha and the severe inhibition of protein synthesis after transient global brain ischemia. Journal of Neurochemistry 94 1235–1242. (https://doi.org/10.1111/j.1471-4159.2005.03276.x)
Peter B, Maurice BL, José AF, Yvonne D, Darragh GM, Philip DD, Stephen M, Ronan TG, Liam JM & Helen GC et al.2017 QuPath: open source software for digital pathology image analysis. Scientific Reports 7 1.
Piipponen M, Li D & & Landen NX 2020 The immune functions of keratinocytes in skin wound healing. International Journal of Molecular Sciences 21. (https://doi.org/10.3390/ijms21228790)
Podolsky MA, Bailey JT, Gunderson AJ, Oakes CJ, Breech K & & Glick AB 2017 Differentiated state of initiating tumor cells is key to distinctive immune responses seen in H-RasG12V–Induced squamous tumors. Cancer Immunology Research 5 198–210. (https://doi.org/10.1158/2326-6066.CIR-16-0304)
Quist SR, Wiswedel I, Quist J & & Gollnick HP 2016 Kinetic profile of inflammation markers in human skin in vivo following exposure to ultraviolet B indicates synchronic release of cytokines and prostanoids. Acta Dermato-Venereologica 96 910–916. (https://doi.org/10.2340/00015555-2406)
Ravindran A, Mohammed J, Gunderson AJ, Cui X & & Glick AB 2014 Tumor-promoting role of TGFß1 signaling in ultraviolet B-induced skin carcinogenesis is associated with cutaneous inflammation and lymph node migration of dermal dendritic cells. Carcinogenesis 35 959–966. (https://doi.org/10.1093/carcin/bgt486)
Rezvani HR, Mazurier F, Cario-André M, Pain C, Ged C, Taïeb A & & Verneuil HD 2006 Protective effects of catalase overexpression on UVB-induced apoptosis in normal human keratinocytes. Journal of Biological Chemistry 281 17999–18007. (https://doi.org/10.1074/jbc.M600536200)
Salminen A, Kaarniranta K & & Kauppinen A 2022 Photoaging: UV radiation-induced inflammation and immunosuppression accelerate the aging process in the skin. Inflammation Research 71 817–831. (https://doi.org/10.1007/s00011-022-01598-8)
Scheper W & & Hoozemans JJM 2015 The unfolded protein response in neurodegenerative diseases: a neuropathological perspective. Acta Neuropathologica 130 315–331. (https://doi.org/10.1007/s00401-015-1462-8)
Son J, Mogre S, Chalmers FE, Ibinson J, Worrell S & & Glick AB 2021 The ER stress sensor IRE1α regulates the UV DNA repair response through control of intracellular calcium homeostasis. Journal of Investigative Dermatology 142 1682–1691. (https://doi.org/10.1007/s00401-015-1462-8)
Win S, Than TA, Fernandez-Checa JC & & Kaplowitz N 2014 JNK interaction with Sab mediates ER stress induced inhibition of mitochondrial respiration and cell death. Cell Death and Disease 5 e989–e989. (https://doi.org/10.1038/cddis.2013.522)
Wu T, Zhao F, Gao B, Tan C, Yagishita N, Nakajima T, Wong PK, Chapman E, Fang D & & Zhang DD 2014 Hrd1 suppresses Nrf2-mediated cellular protection during liver cirrhosis. Genes and Development 28 708–722. (https://doi.org/10.1101/gad.238246.114)
Yoshizumi M, Nakamura T, Kato M, Ishioka T, Kozawa K, Wakamashu K & & Kimura H 2008 Release of cytokines/chemokines and cell death in UVB-irradiated human keratinocytes, HaCaT. Cell Biology International 32 1405–1411. (https://doi.org/10.1016/j.cellbi.2008.08.011)
Zima AV & & Blatter LA 2006 Redox regulation of cardiac calcium channels and transporters. Cardiovascular Research 71 310–321. (https://doi.org/10.1016/j.cardiores.2006.02.019)
