- Email: [email protected]
The role of peroxiredoxin 4 in inflammatory response and aging
Biochimica et Biophysica Acta 1862 (2016) 265–273
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbadis
The role of peroxiredoxin 4 in inflammatory response and aging Vladimir I. Klichko, William C. Orr, Svetlana N. Radyuk ⁎ Department of Biological Sciences, Southern Methodist University, Dallas, TX, USA
a r t i c l e
i n f o
Article history: Received 13 October 2015 Received in revised form 25 November 2015 Accepted 8 December 2015 Available online 9 December 2015 Keywords: Peroxiredoxin Immunity Inflammation Aging Apoptosis Drosophila
a b s t r a c t In prior studies, we determined that the moderate overexpression of the Drosophila endoplasmic reticulum (ER)localized peroxiredoxin (Prx), dPrx4, reduced oxidative damage and conferred beneficial effects on life span, while a high-level expression increased the incidence of tissue-specific apoptosis and dramatically shortened longevity. The detrimental pro-apoptotic and life-shortening effects were attributed to aberrant localization of dPrx4 and the apparent ER stress elicited by dPrx4 overexpression. In addition, the activation of both the NF-κB- and the JAK/STAT-mediated stress responses was detected, although it was not clear whether these served as functional alarm signals. Here we extend these findings to show that the activation of the NF-κB-dependent immunity-related/inflammatory genes, associated with life span shortening effects, is dependent on the activity of a Drosophila NF-κB ortholog, Relish. In the absence of Relish, the pro-inflammatory effects typically elicited by dPrx4 overexpression were not detected. The absence of Relish not only prevented the hyperactivation of the immunity-related genes but also significantly rescued the severe shortening of life span normally observed in dPrx4 overexpressors. The overactivation of the immune/inflammatory responses was also lessened by JAK/STAT signaling. In addition, we found that cellular immune/ pro-inflammatory responses provoked by the oxidant paraquat but not bacteria are mediated via dPrx4 activity in the ER, as the upregulation of the immune-related genes was eliminated in flies underexpressing dPrx4, whereas immune responses triggered by bacteria were unaffected. Finally, efforts to reveal critical tissues where dPrx4 modulates longevity showed that broad targeting of dPrx4 to neuronal tissue had strong beneficial effects, while targeting expression to the fat body had deleterious effects. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Changes in redox-sensitive signaling due to fluctuations in the cellular redox milieu lead to functional consequences in a variety of biological processes, including stress response, inflammation, and apoptosis. While such responses are potentially adaptive in nature, permitting cells to overcome various exogenous and endogenous challenges, chronic, or inadequate responses may have a deleterious impact. Consequently, a robust redox homeostatic response is considered essential in maintaining healthy cell function, whereas its disruption may inform/promote a multitude of disease states. A pivotal role in maintaining cellular redox homeostasis and proper redox-sensitive signaling belongs to the peroxiredoxins, a family of thiol-dependent peroxidases that possess the capacity to act as redox sensors. Peroxiredoxins (Prx) not only have the capacity to control local hydrogen peroxide fluxes but have also been shown to transmit
Abbreviations: ER, endoplasmic reticulum; Prx, peroxiredoxin; AMP, antimicrobial peptide; UPR, unfolded protein response; Act, actin; Da, daughterless; AttD, attacin D; Dipt, diptericin; Drs, drosomycin; TotA, Turandot A; ROS, reactive oxygen species. ⁎ Corresponding author at: 6501 Airline Rd, Room 113, Dallas, TX 75275, USA. E-mail address: [email protected] (S.N. Radyuk).
http://dx.doi.org/10.1016/j.bbadis.2015.12.008 0925-4439/© 2015 Elsevier B.V. All rights reserved.
H2O2-mediated protein sulfhydryl modifications to specific target proteins [1]. Endoplasmic reticulum (ER) is one of the major sources of cellular H2O2, where it is produced during the oxidative disulfide formation of newly synthesized proteins [2]. Recently, we reported that Prx4, which resides in the ER, is involved in modulating Drosophila longevity under normal conditions or in the presence of various environmental stressors [3]. Prx4 serves a dual function in the ER. It both maintains proper H2O2 concentrations, generated during protein oxidative folding [4], and also uses hydrogen peroxide to catalyze the formation of disulfide bonds in protein disulfide isomerase [5,6]. The moderate ectopic overexpression of Prx4 was found to have beneficial effects on fly physiology, including a significant extension in life span. By contrast, the global high-level overexpression of dPrx4 (N 5-fold) elicited increased incidence of tissue-specific apoptosis and shortened life span concomitant with overproduction of antimicrobial peptides (AMP), a signature of the NF-κB signaling in the immune and inflammatory responses [3], and is reminiscent of the alarm signal triggered by ER stress due to accumulation of unfolded proteins, the so-called unfolded protein response (UPR) [7,8]. The activation of the NF-κB signaling pathway is presumed to safeguard cellular viability and function, but when chronically induced, it may become disruptive, leading to cell suicide as a last resort to dispose of dysfunctional cells where the accumulation of misfolded proteins has overwhelmed repair capacity [9].
266
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273
The modulation of the ER redox domain by manipulating Prx4 levels provides the opportunity to experimentally establish a state of chronic inflammation and identify specific pathway elements that mediate this ER-derived stress and thus facilitate the design of interventions to manage inflammatory responses and apoptosis.
2. Materials and methods 2.1. Fly strains and procedures The UAS-dPrx4 transgenic lines, RNAi-dPrx4 mutants, RNAi mutants for the genes representing the JAK/STAT pathway (hopscotch, Stat92E, and domeless), and driver fly lines were those described by Radyuk et al. [3]. The S106-inducible GeneSwitch-GAL4 fat body-specific driver was kindly supplied by Blanka Rogina (University of Connecticut Health Science Center). A null mutant strain for the NF-κB-like transcription factor Relish, relE20, was obtained from the Bloomington Stock Center. All fly lines were backcrossed into our reference y w genetic background. Fly collection and husbandry were as described by Radyuk et al. [3]. Briefly, flies overexpressing dPrx4 at high levels and flies overexpressing dPrx4 in combination with domeless RNAi-mediated underexpression were developed at 18 °C because of the lethal effects on development at 25 °C. Adults were reared at 25 °C after hatching. The underexpression of Hopscotch and Stat92E by RNAi was achieved with high-level global daughterless (Da) or Actin (Act) GAL4 drivers, and the overexpression of dPrx4 with S106-Switch driver was achieved by feeding flies food supplemented with 100 μg/ml mifepristone (RU486). Controls included the RNAi transgene target alone and the driver alone except for the S106 GeneSwitch experiment where genetically identical flies fed food containing the mifepristone solvent, ethanol, served as control. Responses to oxidants and infection were studied by feeding flies with 1% sucrose solution containing 10 mM paraquat or septically injuring flies by pricking with a needle dipped in an Escherichia coli suspension, as detailed by Radyuk et al. [3]. Survivorship studies were conducted as described previously [10].
2.2. RT-PCR Primers specific for AMP genes and Turandot A (TotA) were those indicated in publications [3,11]. Real-time PCR analysis was performed as described previously [11], where PCR reactions were performed with SYBR Green fluorescent dye (Molecular Probes), and the signals obtained for each gene were standardized against signals obtained for the rp49 housekeeping gene.
2.3. TUNEL labeling The assessment of apoptosis-induced DNA fragmentation was made in cryosections prepared from whole flies using the In Situ Cell Death Detection Kit, TMR red (Roche, Indianapolis, IN, USA), as detailed previously [3]. Images were acquired by fluorescence microscopy (Zeiss), using AxioVisionLE4_3 software.
2.4. Statistical analysis All statistics were calculated using Prism for Macintosh version 4.0a software (GraphPad Software, Inc.). Differences in mRNA levels were compared between groups by analysis of variance. In studies of life span, the mean survivorship time and differences between survivorship curves were assessed using the log-rank test.
3. Results 3.1. Induction of AMPs and TotA in response to Prx4 overexpression requires Relish As reported by us previously [3], dPrx4 overexpression at high levels (N5-fold) caused an increase in the expression of both the NF-κBdependent immune response/pro-inflammatory genes, AMPs, and the cytokine-like protein Tot A, a target of JAK/STAT signaling. Given that the NF-κB transcription factor Relish has been identified as a common regulator of the expression of AMPs and Turandot proteins [12] as well as a mediator of responses elicited due to ER stress [8], we postulated that the stress response due to Prx4 overload is mediated via Relish. To test this idea, we measured the Prx4-driven stress response in the absence of Relish using the rel20 null mutant. The results of this analysis, as shown in Fig. 1, revealed that the activity of Relish is indeed required for induction of AMPs and TotA in response to the apparent ER stress caused by dPrx4 overload. The upregulation of the AMPs attacin D (AttD) and diptericin (Dipt), as well as TotA, normally observed in flies overexpressing dPrx4, was completely eliminated in the relish mutant. We also observed lower levels of drosomycin (Drs) in flies underexpressing Relish. Notably the basal levels of AMPs were lower in 10 days old rel20 mutant flies compared to controls. It is tempting to conjecture that this transcription factor is also critical for the age-dependent increase in AMP transcripts levels that occurs in normal flies. 3.2. JAK/STAT signaling mitigates the activation of NF-κB-dependent immunity genes JAK/STAT signaling has been previously shown to repress NF-κBdependent immune signaling. In a series of SL2 cell culture studies, RNAi was used to reduce levels of the JAK/STAT signaling factor Stat92E, giving rise to a significant increase in the expression of Relish-mediated immunity genes and providing support for the notion that Stat92E serves as a Relish antagonist [13]. To determine whether a similar relationship could be discerned in response to stress elicited by dPrx4 overexpression, we analyzed the expression of AMP genes in RNAi mutant flies underexpressing three distinct elements of the JAK/STAT pathway (Stat92E, Hopscotch, and Domeless) in the presence or absence of Prx4 overexpression. Consistent with previous findings ((3) and Fig. 1), the overexpression of dPrx4 alone at high levels resulted in AMP induction, particularly evident for AttD and Dipt (Fig. 2). The underexpression of the domeless and Stat92E genes of the JAK/STAT pathway resulted in increases in AMPs similar to those observed for Prx4 overexpression alone, while the underexpression of hopscotch, the Drosophila JAK homologue, resulted in a particularly strong increase in AMP levels relative to those elicited by Prx4 overexpression and comparable to the increase during aging (55-day-old control). The AMP activation effects were not diminished in response to a combination treatment (Prx4 overexpression + JAK/STAT knockdown), and in fact some synergy was evident in three cases (See arrows, Fig. 2). Overall, these results support the notion that JAK/STAT signaling plays an inhibitory role in the Relishdependent activation of AMPs. 3.3. dPrx4 is required for induction of Relish-dependent AMPs in response to PQ but not infection There is evidence that exposure to oxidants or infection (particularly viral) may provoke ER stress and, as a consequence, lead to proinflammatory responses mediated via the NF-κB pathways [8,9]. We reasoned that if the ER-specific Prx4 mediates AMP activation in response to oxidative stress and septic injury, the knockdown of Prx4 would abolish this activation effect. Consequently, levels of Prx4 were knocked down N90% by RNAi, and flies were subsequently subjected
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273
267
Fig. 1. Effects of Relish on induction of the immune and stress response genes in flies overexpressing dPrx4. All flies were 10 days old. C—driver control (Actin-GAL4/+); ActNdPrx4—flies overexpressing dPrx4 using Act-GAL4 driver; rel—rel20 mutant; ActNdPrx4; rel—flies overexpressing dPrx4 in rel20 mutant background. RT-PCR analysis was conducted with two independent cohorts of flies and replicated for each cohort. Shown are means ± SEM (n = 4). Asterisks denote statistically significant differences (*P b 0.05; **P b 0.005; ***P b 0.0005, and ****P b 0.00005).
to paraquat treatment (oxidative stress) or septic injury with E. coli, after which AMP activation was assessed. RT-PCR analysis of diptericin and attacin D genes, targets of Relish, showed that dPrx4 activity is required for the activation of Relish signaling in response to paraquat, as the activation of Dipt and AttD, observed in the control, was absent in the dPrx4 RNAi mutants (Fig. 3). No differences in expression were observed in septically injured flies, suggesting that the upregulation of AMPs in response to bacteria does not depend on dPrx4-mediated signaling in the ER and is instead regulated by the canonical Imd/Relish signaling. There were also no effects of dPrx4 on the upregulation of the Toll/Dif-dependent drosomycin (Drs) in flies stressed with either paraquat or E. coli (Fig. 3), suggesting that the signals mediated via ER do not affect this branch of immunity. Thus, Relish activation appears to be a specific outcome of exposure to oxidants mediated via Prx4/ER. 3.4. A rapid death phenotype caused by the dPrx4 overexpression is partially rescued in relish mutant Previously, we reported that dPrx4 significantly shortened life span when overexpressed globally at high levels and that these lifeshortening effects were accompanied by tissue-specific apoptosis [3]. Here we tested whether this reduced longevity could be mediated by NF-κB signaling triggered by ER stress. In a relish null background, the reduction in longevity caused by dPrx4 overload was significantly attenuated (Fig. 4 and Table 1), supporting the idea that Relish mediates at least in part the inflammatory responses originating in the ER. These inflammatory responses can be destructive and further exacerbate ER stress by contributing to a greater pro-oxidative environment. Previously, we noted that dPrx4 overexpression also had deleterious temperature
effects on development, displaying a recessive lethal phenotype when cultured at 25 °C, but viable when cultured at 18 °C. This recessive lethal phenotype was partially reversed in the absence of Relish, in that a significant number of female flies and to a lesser degree male flies overexpressing dPrx4 in a relish null background (ActinNdPrx4; rel) eclosed at 25 °C (data not shown). Consistent with previous observations [3], no male or female flies expressing high levels of dPrx4 (Act NdPrx4) in a rel+ background emerged at 25 °C. 3.5. Relish modulates cell death triggered by the overexpression of dPrx4 The increase of tissue-specific apoptosis in response to dPrx4 overexpression has been ascribed to the release of dPrx4 from the ER to the cytosol, where it is thought to antagonize IAP1 and thereby liberate pro-apoptotic caspases [14]. NF-κB signaling involving Relish has been implicated in other models of cell death under investigation in Drosophila [15], and consequently we set out to establish the significance of Relish-dependent signaling on apoptosis in our dPrx4 model. As we found previously, the overexpression of dPrx4 resulted in increased occurrence of apoptotic cells in fly muscles and fat body tissue, as was determined by the TUNEL assay [3]. To examine if Relish plays a role in mediating the observed apoptotic changes caused by Prx4, we examined dPrx4 and dPrx4;rel flies for tissue-specific apoptosis (Fig. 5). In agreement with previous observations, a higher number of cells displaying DNA fragmentation was observed in thoracic muscle and abdominal fat body tissues. These patterns of apoptosis were not observed in similar regions in cryosections made from flies overexpressing dPrx4 in a relish null background, suggesting that tissue-specific apoptosis is mediated, at least in part, via the Relish pathway. Consistent
268
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273
with previous observations [3], we were unable to detect apoptotic changes in the heads of flies overexpressing dPrx4. 3.6. Prx4 overexpression in neuronal tissue extends life span while overexpression in the fat body had negative effects on fly survivorship The negative effects of dPrx4 overexpression on life span observed when driven by a high-level constitutive driver such as Da can be contrasted with the relatively beneficial effects observed when driven by a weak constitutive driver (armadillo), suggesting a dosage effect. Nevertheless, it cannot be ruled out that these differences in longevity may be ascribed to tissue-specific expression patterns. To explore this, we expressed dPrx4 ectopically in neuronal tissue, which has been shown to be particularly sensitive to changes in redox state, and in fat body tissue, the main site of the production of immune-related and stress response genes regulated by Imd/ and JAK/STAT signaling [10,16]. Survivorship was assessed in flies in which dPrx4 was ectopically expressed either in neuronal tissue using the pan-neuronal APPLGAL4 driver or in the fat body using the S106-GeneSwitch-GAL4 driver. The overexpression of dPrx4 neuronally resulted in significant positive effects on survivorship under normal conditions (Fig. 6 A and Table 2 A). In contrast, when dPrx4 production was enhanced specifically in the fat body, the life spans of flies were shortened by 4-8 % relative to controls (Fig. 6 B and Table 2 B). 4. Discussion A major finding of this study is that the immune/inflammatory and cell death responses, triggered by disturbances in ER function or exposure to exogenous reactive oxygen species (ROS), depend on the activity of the ER-specific peroxiredoxin dPrx4. Previously, we showed that the high-level overexpression of dPrx4 elicits a state of ER stress leading to aberrant immune response and negative functional consequences [3]. Here, we establish that the effects of dPrx4 on the responses derived from the ER are mediated via NF-κB (Relish)-dependent signaling and, furthermore, that these elevated Relish-dependent immune/inflammatory responses can account for a significant portion of the adverse effects on Drosophila physiology, including the shortening of life span. In fruit flies, the management of immune function and pro-inflammatory response relies on overlapping strategies, involving the recruitment of immune cells (hemocytes), the burst of ROS, and the release of other compounds intended for elimination of infectious agents, including various AMPs, whose production is regulated via the immune and stress response pathways (reviewed by Shaukat et al. [17]). These pathways, including JNK and NF-κB, are also utilized to elicit a state of inflammation in response to sterile triggers, such as molecular damage, and the ER stress described in the present study. The first evidence for the involvement of Prx4 in the TNF-αdependent activation of the NF-κB signaling cascades came from studies of the Rhee group, conducted with cultured HeLa cells [18]. Proinflammatory TNF-α/NF-κB signaling in mammals is analogous to the Imd/Relish signaling in Drosophila. To place these results in a more physiological context, we overexpressed dPrx4 in flies at elevated levels and evoked a UPR-like inflammatory response [3]. Indeed, we observed increased transcription levels of the AMPs in our dPrx4 overexpressing Fig. 2. Effects of the silencing of JAK/STAT signaling on AMP expression in flies expressing normal and high levels of Prx4. All flies were 10 days old. Control flies were 10 days old and 55 days old (C, 55 days). C—driver control (Da-GAL4/+); dPrx4Nda—flies overexpressing dPrx4 by Da-GAL4 driver. dome, hop, and stat—RNAi mutants underexpressing Domeless, Hopscotch, and Stat92E [3]. DaNdPrx4; dome, hop, or stat—flies overexpressing dPrx4 and underexpressing Domeless, Hopscotch, and Stat92E. Two different RNAi lines were used for each gene, and two individual dPrx4 transgenic lines, giving comparable results; hence, average values are presented. Shown are means ± SEM. Analyses were conducted in triplicate for each fly line and reproduced with 2 independent cohorts of flies (n = 6). Asterisks denote statistically significant differences between control (C) and experimental lines (*P b 0.05). Arrows indicate synergistic effects of dPrx4 overexpression and underexpression of the components of JAK/STAT signaling.
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273
269
Fig. 4. Effects of dPrx4 overexpression on fly life span in the presence and absence of Relish. ActNdPrx4—flies overexpressing dPrx4 by Actin-GAL4 driver; ActNdPrx4;rel—flies overexpressing dPrx4 in the absence of Relish. Controls included flies doubly heterozygous for the Act driver and the rel20 mutation (Act/+; rel/+) as well as flies heterozygous for the driver and homozygous for the rel20 mutation (Act/+; rel/rel), all backcrossed into the y w reference background. Median ages of flies (50% survivorship) are indicated in Table 1. Statistically significant differences were observed between ActNdPrx4 vs. control (Pb0.05) and relish mutant, and ActNdPrx4 vs. ActNdPrx4;rel as determined by log-rank test.
found to be mediated via Relish, as the activation of AMPs elicited by dPrx4 overexpression does not occur in the absence of Relish activity (Fig. 1). Furthermore, the absence of Relish not only prevented the activation of AMPs observed in Prx4 overexpressors but also reversed the life span shortening effects in these flies along with the accompanying pro-apoptotic changes (Figs. 4 and 5). Such results are consistent with a causal connection between Relish-mediated AMP activation and longevity effects. While we clearly established a role for Relish in Prx4/ER-mediated pro-inflammatory/pro-apoptotic signaling, the exact mechanisms that culminate in the life span shortening effects are still not clear. On one hand, such effects could be due to apoptosis, observed predominantly in the fly muscle and fat body of Prx4 overexpressors [3]. On the other hand, the activation of Relish-dependent AMPs, markers of inflammation, could themselves contribute to life span reduction. Support for this latter idea is the observation that the overexpression of AMPs, such as drosocin, attacin, defensin, and drosomycin, in young flies induced neurodegeneration comparable to that observed in old flies [19]. Furthermore, since the deleterious effects of these AMPs were not suppressed in Relish knockdowns, it allowed the authors to conclude that the prolonged exposure of the host cells to high levels of AMPs is sufficient to cause neurodegeneration [19]. In agreement with these findings, we determined that the global high-level overexpression of AMPs, in particular attacin and defensin, or overexpression targeted to the fat body had significant life-shortening effects and these effects correlated with a marked increase in tissue-specific cell death (unpublished observations). It is not clear whether the cell death observed in Prx4 overexpressors, is a direct result of the cytotoxic effects of AMPs or rather is elicited by inappropriate Relish signaling that mediates its effects by an alternative pathway, as has been documented for a case of apoptosis-related retinal tissue degeneration [15]. The possibility that pro-apoptotic changes are caused by a combination of factors, Fig. 3. Effects of dPrx4 underexpression on the activation of Imd/Relish-dependent AMPs. All flies were 10 days old. C—driver control (Da-GAL4/+); DabRNAi-dPrx4—RNAi mutants underexpressing dPrx4 using Da-GAL4 driver. Flies were untreated, exposed to paraquat, or septically injured with a suspension of Escherichia coli, as specified in Section 2. Shown are means ± SEM. Analyses were conducted in triplicate with two different RNAi fly lines and reproduced with 2 independent cohorts of flies (n = 6). Asterisks denote statistically significant differences (*P b 0.05).
flies ([3] and Figs. 1 and 2), a known signature of ER stress, and reminiscent of a response caused by a mutation in the transmembrane protein, a constituent of ER [8]. These responses caused by dPrx4 overload were
Table 1 Median age (days) of flies overexpressing dPrx4 and relish mutants. Genotype
Actin/+,dPrx4/+
rel
ActinNdPrx4
ActinNdPrx4;rel
Experiment 1 Experiment 2
49.5 51.5
53.0 52.0
24.0 19.0
34.5 32.5
Differences in median age between flies overexpressing dPrx4 in a normal background (ActinNdPrx4) or relish mutant background (ActinNdPrx4;rel) and heterozygote driver and transgene control (Actin/+,dPrx4/+), as well as relish (rel) mutant were statistically significant. The rescue effect of the relish mutant is statistically significant, as determined by log-rank test (P b 0.05).
270
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273
Fig. 5. TUNEL assay of cell death in flies overexpressing dPrx4 in normal and relish mutant background. Cryosections were made from 10-day-old flies overexpressing dPrx4 by Actin-GAL4 driver (ActNdPrx4) and flies overexpressing dPrx4 in rel20 null mutant background (ActNdPrx4;rel). Representative images of the abdominal and thoracic regions show a reduction of apoptosis in abdominal fat body tissue and gut epithelia, as well as in thoracic muscles in samples made from flies overexpressing dPrx4 in the absence of Relish. ms—muscle; g—gut lumen; fb—fat body. No apoptotic cells were observed in the heads of flies overexpressing dPrx4 in normal or rel mutant background.
including missexpression of AMPs, aberrant Relish signaling and IAP1dependent signaling, triggered by cytosolic Prx4 [14] cannot be ruled out, as summarized in Fig. 7. Another important finding of the present study is that the inflammatory response to the oxidant paraquat is mediated via ER signaling that is dependent on dPrx4 activity, while the response to infection seems mainly regulated by classical immune pathways, independent of ER. This is based on the observation that the induction of the Relish/NFκB-dependent AMPs failed in RNAi dPrx4 mutants exposed to paraquat but occurred normally in mutants infected with bacteria (Fig. 3). There is precedent for oxidant-induced inflammation mediated through ER stress [8,9]. A current model for this “sterile” inflammation envisions
mitochondria as a primary target, where functional disruption leads to redox state fluctuations that elicit ER stress, which in turn promotes NF-κB-mediated inflammation [9,20]. This study is a first report that the ER-specific Prx4 is a critical control factor for ER inflammatory signaling and is required for alarm and, perhaps, adaptation responses. Furthermore, the data obtained suggest that these responses do not depend simply on antioxidant function of dPrx4 and changes in ROS generation, as the higher ROS/peroxide levels observed in the dPrx4 knockdown were accompanied by only a minimal increase in AMPs and had no effect on life span ([3] and Fig. 3). One possible explanation for this seemingly contradictory finding is that dPrx4 functions primarily as a relay in transmitting stress signals via interaction with yet to be
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273
Fig. 6. The effects of tissue-specific dPrx4 overexpression on fly LS. (A) Life spans of flies overexpressing dPrx4 in neuronal tissues with the Appl-GAL4 driver. Controls are Appl-GAL4 driver (+/Appl) and three different UAS-dPrx4 transgenic lines (A, B, or C/+). Experimental flies overexpressing dPrx4 are ApplNdPrx4. Data are representative of two replicate experiments, summarized in Table 1 A. (B) Life spans of flies overexpressing dPrx4 in the fat bodies with the S106 pSwitch-inducible driver. Controls were pSwitch106NdPrx4 flies fed ethanol; experimentals were flies fed with mifepristone (+RU486). Data are representative of two replicate experiments, summarized in Table 1 B.
determined targets. Such a relay system involving Prx2 and JAK/STAT signaling has been reported recently [21]. Components of the three known UPR pathways (ATF6, PERK, or Ire1) are potential targets for dPrx4-mediated effects.
271
In this study, we also further clarified the function of the secreted form of Prx4 and found that it may play a role in mitigation of inflammatory response. Previously, we have reported that various stresses or high-level Prx4 overexpression causes subcellular re-distribution of dPrx4, which triggers the activation of the JAK/STAT pathway as inferred from the upregulation of the stress response protein Turandot A [3]. The JAK/STAT pathway controls several biological processes in Drosophila during development and in adults and directly contributes to immune and stress response by activating infection-induced genes [16,22]. In response to septic injury or other stressors, such as heat-shock or dehydration, hemocytes produce the Upd3 ligand, which initiates the expression of several immune-related proteins, including cytokines and stress response proteins [12]. Our results suggest that in the case of Prx4 overexpression, the role of JAK/STAT signaling is to suppress Relish-mediated proinflammatory responses, consistent with the established interplay between NF-κB and STAT transcription factors [13,23]. Perhaps, one of the functions of the secreted dPrx4 is to suppress excessive inflammatory response via a regulatory loop where the activity of the ER-localized dPrx4 in eliciting the NF-κB responses is antagonized by the same enzyme, re-localized to a different milieu (Fig. 7). Finally, we have found that the in vivo effects of Prx4 on life span vary not only with respect to different levels of global overexpression, as reported earlier [3], but are also sensitive to overexpression in specific tissues (Fig. 6). Similar to a previous finding, where life span extension was achieved by enhancing the pro-reducing capacity of Drosophila neuronal tissues by increasing glutathione production [10], pan-neuronal overexpression of dPrx4 had a strong beneficial effect on longevity (Fig. 6 A). It is tempting to ascribe these life span-extending effects to increased antioxidant function as well as enhancement of oxidative folding capacity. The brain tissues of Drosophila, being aerobic, are the most vulnerable to oxidative stress and exposure to ROS, such as H2O2, which is mainly produced by mitochondria and ER during respiration and in the process of protein oxidative folding [24,25]. To date there have been no report of longevity effects for other animal models, including transgenic mice overexpressing Prx4. On the other hand elevated Prx4 levels in specific tissues have been shown to play a preventative role in the development of diseases, elicited by various stressors. Studies in mice transgenics expressing the human Prx4 transgene displayed a significant reduction in atherosclerotic lesions, which was attributed to its ROS-scavenging capacity in extracellular space, such as vascular vessels [25]. Transgene-expressed Prx4 might also prevent the development of the drug-induced diabetes caused by destruction of beta cells in pancreatic islets, where the effects of Prx4 were ascribed to ROS-scavenging and thus anti-inflammatory and anti-apoptotic function of this enzyme [26]. On the other hand, the
Table 2 Mean age of flies overexpressing dPrx4 in pan-neuronal tissue (A) and the fat body (B) A Genotype
Mean (days)
% vs. transgene/+
% vs. driver/+
+/dPrx4-A +/dPrx4-B +/dPrx4-C Appl/+ ApplNdPrx4-A ApplNdPrx4-B ApplNdPrx4-C
57.6; 61.2 54.1; 57.8 45.8; 59.1 51.2; 56.8 67.3; 62.3 71.3; 55.1 71.0; 64.2
116.8; 101.8 131.8; 95.3 155.0; 108.6
131.4; 109.7 139.3; 97.0 138.7; 113.1
B Genotype
Mean (days) RU486-
Mean (days) RU486+
% RU486/+ vs. RU486/-
S106NdPrx4-A S106NdPrx4-B S106NdPrx4-C
35.5; 38.1 27.1; 35.5 29.1; 35.3
34.0; 36.5 28.8; 32.6 27.7; 33.1
95.8; 95.8 106.3; 91.8 95.2; 93.8
Values obtained in two independent experiments are listed in column 1 (A) and columns 1 and 2 (B). The overall percentage of the mean age changes between experimentals vs. the transgene/+ and driver/+ controls or experimental flies fed mifepristone are indicated in columns 2 and 3 (A) and in column 3 (B). Statistically significant differences between control and experimental flies, determined by log-rank test (P b 0.05), are indicated by bold.
272
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273
Fig. 7. Proposed model of the effects of dPrx4 on cell function. The scheme summarizes our previous [3] and current findings. High-level overexpression of the ER dPrx4 causes release of this protein into the cytosol and its secretion into extracellular space. The extracellular form of dPrx4 activates the JAK/STAT pathway and downstream response genes. The cytosolic form of dPrx4 promotes apoptotic cell death. The ER-localized dPrx4 is involved in the regulation of the NF-κB-dependent immune/inflammatory responses and apoptosis. The complex interactions between pathways modulated by dPrx4 are depicted. The links indicated by dotted lines, including potential cross talk between ER and mitochondrion (MT), have yet to be established.
overexpression of Prx4 improved the function of insulin-secreting INS1E cells, which was interpreted as an ability of this Prx to enhance the ER protein oxidative folding capacity [27]. In contrast, the beneficial antioxidant effects of Prx4 overexpression were apparently offset by pro-inflammatory changes observed in flies with elevated global or fat body-specific Prx overexpression ([3] and Fig. 6 B). Perhaps then the biological function of Prx4 is largely determined by cellular context. Thus, the overexpression of dPrx4 in the fat body, a major site of AMP production, would be likely to enhance AMP-mediated inflammation, whereas the overexpression of dPrx4 in brain tissue does not promote AMP production and the beneficial antioxidant effects of dPrx4 predominate. These differential effects may be due in part to differential expression, and perhaps, the activation of Relish in the brain and fat body tissues. Furthermore, we found no evidence of apoptosis in the brains of flies overexpressing dPrx4 while pro-apoptotic changes were clearly promoted in other tissues, such as thoracic muscles and fat bodies, as determined in the current study (Fig. 5) and reported previously [3]. Thus, higher levels of dPrx4 in the brain tissue do not appear to induce neuronal apoptosis and consequently do not have negative effects on fly physiology. In agreement with these findings, apoptotic pathways are not always activated in the nervous system in response to mutations that lead to neuroinflammation [19] or in old flies, although other tissues display pro-apoptotic changes [28]. To conclude, we found that dPrx4 plays a critical role in controlling inflammatory response in Drosophila and these effects vary according to tissue context. The data are summarized in the following scheme (Fig. 7). As reported earlier [3], the overexpression of dPrx4 or stressors, such as oxidants or bacteria, causes re-localization of dPrx4 to different subcellular compartments where this protein has differential effects on biological processes. The extracellular form of dPrx4 activates JAK/STAT signaling, resulting in the upregulation of the stress response Turandot proteins. The activation of JAK/STAT signaling also antagonizes the
activation of the Relish-dependent AMPs, as their levels are significantly higher in the JAK/STAT pathway mutants (Fig. 2). Unlike the extracellular dPrx4, which seems to serve a pro-survivorship function, the cytosolic dPrx4 has been reported to promote apoptosis due to interaction with IAP1 [14]. Apoptosis can also be elicited via Relish-dependent signaling, although it is not clear whether cell death is a direct effect of Relish or a secondary response to Relish-dependent inflammation. Besides the apoptosis and activation of JAK/STAT signaling, the overexpression of dPrx4 also activates induction of AMPs and this induction is dependent on the activity of the NF-κB factor Relish. The activation of the immune response (AMPs) by bacteria is independent of the ER-localized dPrx4 and is mediated through a canonical immune pathway (Imd), unlike the upregulation of the immune/inflammatory responses to the ROS generator paraquat, which is mediated via ER signaling and requires dPrx4 activity. Thus, it may be concluded that the inflammatory and pro-apoptotic effects of dPrx4 in response to sterile stimuli are mediated via the NF-κB analog Relish by the activation of the ER/Relish signaling branch, independent of the canonical Imd/Relish immune pathway. Transparency document The Transparency document associated with this article can be found, in online version. Acknowledgments This work was supported by the grant R01 AG032342 from the National Institute on Aging/National Institutes of Health. References [1] S.G. Rhee, H.A. Woo, I.S. Kil, S.H. Bae, Peroxiredoxin functions as a peroxidase and a regulator and sensor of local peroxides, J. Biol. Chem. 287 (2012) 4403–4410.
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273 [2] B.P. Tu, J.S. Weissman, The FAD- and O(2)-dependent reaction cycle of Ero1mediated oxidative protein folding in the endoplasmic reticulum, Mol. Cell 10 (2002) 983–994. [3] S.N. Radyuk, V.I. Klichko, K. Michalak, W.C. Orr, The effect of peroxiredoxin 4 on fly physiology is a complex interplay of antioxidant and signaling functions, FASEB J. 27 (2013) 1426–1438. [4] T.J. Tavender, N.J. Bulleid, Peroxiredoxin IV protects cells from oxidative stress by removing H2O2 produced during disulphide formation, J. Cell Sci. 123 (2010) 2672–2679. [5] Z. Cao, T.J. Tavender, A.W. Roszak, R.J. Cogdell, N.J. Bulleid, Crystal structure of reduced and of oxidized peroxiredoxin IV enzyme reveals a stable oxidized decamer and a non-disulfide-bonded intermediate in the catalytic cycle, J. Biolumin. Chemilumin. 286 (2011) 42257–42266. [6] L. Wang, L. Zhu, C.C. Wang, The endoplasmic reticulum sulfhydryl oxidase Ero1beta drives efficient oxidative protein folding with loose regulation, Biochem. J. 434 (2011) 113–121. [7] C. Xu, B. Bailly-Maitre, J.C. Reed, Endoplasmic reticulum stress: cell life and death decisions, J. Clin. Invest. 115 (2005) 2656–2664. [8] K.A. Boltz, G.E. Carney, Loss of p24 function in Drosophila melanogaster causes a stress response and increased levels of NF-kappaB-regulated gene products, BMC Genomics 9 (2008) 212. [9] N. Chaudhari, P. Talwar, A. Parimisetty, C. Lefebvre d'Hellencourt, P. Ravanan, A molecular web: endoplasmic reticulum stress, inflammation, and oxidative stress, Front. Cell. Neurosci. 8 (2014) 213. [10] W.C. Orr, S.N. Radyuk, L. Prabhudesai, D. Toroser, J.J. Benes, J.M. Luchak, R.J. Mockett, I. Rebrin, J.G. Hubbard, R.S. Sohal, Overexpression of glutamate-cysteine ligase extends life span in Drosophila melanogaster, J. Biolumin. Chemilumin. 280 (2005) 37331–37338. [11] S.N. Radyuk, K. Michalak, V.I. Klichko, J. Benes, W.C. Orr, Peroxiredoxin 5 modulates immune response in Drosophila, Biochim. Biophys. Acta 1153-1163 (1800). [12] H. Agaisse, U.M. Petersen, M. Boutros, B. Mathey-Prevot, N. Perrimon, Signaling role of hemocytes in Drosophila JAK/STAT-dependent response to septic injury, Dev. Cell 5 (2003) 441–450. [13] L.K. Kim, U.Y. Choi, H.S. Cho, J.S. Lee, W.B. Lee, J. Kim, K. Jeong, J. Shim, J. Kim-Ha, Y.J. Kim, Down-regulation of NF-kappaB target genes by the AP-1 and STAT complex during the innate immune response in Drosophila, PLoS Biol. 5 (2007), e238. [14] T. Tenev, A. Zachariou, R. Wilson, A. Paul, P. Meier, Jafrac2 is an IAP antagonist that promotes cell death by liberating Dronc from DIAP1, Embo J. 21 (2002) 5118–5129. [15] Y. Chinchore, G.F. Gerber, Dolph, P, J. Alternative pathway of cell death in Drosophila mediated by NF-kappaB transcription factor Relish. Proc. Natl. Acad. Sci. U. S. A. 109 (2012) E605–E612.
273
[16] H. Agaisse, N. Perrimon, The roles of JAK/STAT signaling in Drosophila immune responses, Immunol. Rev. 198 (2004) 72–82. [17] Z. Shaukat, D. Liu, S. Gregory, Sterile inflammation in Drosophila, Mediat. Inflamm. 2015 (2015) 369286. [18] D.Y. Jin, H.Z. Chae, S.G. Rhee, K.T. Jeang, Regulatory role for a novel human thioredoxin peroxidase in NF-kappaB activation, J. Biolumin. Chemilumin. 272 (1997) 30952–30961. [19] Y. Cao, S. Chtarbanova, A.J. Petersen, B. Ganetzky, Dnr1 mutations cause neurodegeneration in Drosophila by activating the innate immune response in the brain, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) E1752–E1760. [20] S.Z. Hasnain, R. Lourie, I. Das, A.C. Chen, M.A. McGuckin, The interplay between endoplasmic reticulum stress and inflammation, Immunol. Cell Biol. 90 (2012) 260–270. [21] M.C. Sobotta, W. Liou, S. Stocker, D. Talwar, M. Oehler, T. Ruppert, A.N. Scharf, T.P. Dick, Peroxiredoxin-2 and STAT3 form a redox relay for H2O2 signaling, Nat. Chem. Biol. 11 (2015) 64–70. [22] M. Boutros, H. Agaisse, N. Perrimon, Sequential activation of signaling pathways during innate immune responses in Drosophila, Dev. Cell 3 (2002) 711–722. [23] T. Kim, J. Yoon, H. Cho, W.B. Lee, J. Kim, Y.H. Song, S.N. Kim, J.H. Yoon, J. Kim-Ha, Y.J. Kim, Downregulation of lipopolysaccharide response in Drosophila by negative crosstalk between the AP1 and NF-kappaB signaling modules, Nat. Immunol. 6 (2005) 211–218. [24] B.P. Tu, J.S. Weissman, Oxidative protein folding in eukaryotes: mechanisms and consequences, J. Cell Biol. 164 (2004) 341–346. [25] X. Guo, S. Yamada, A. Tanimoto, Y. Ding, K.Y. Wang, S. Shimajiri, Y. Murata, S. Kimura, T. Tasaki, A. Nabeshima, T. Watanabe, K. Kohno, Y. Sasaguri, Overexpression of peroxiredoxin 4 attenuates atherosclerosis in apolipoprotein E knockout mice, Antioxid. Redox Signal. 17 (2012) 1362–1375. [26] Y. Ding, S. Yamada, K.Y. Wang, S. Shimajiri, X. Guo, A. Tanimoto, Y. Murata, S. Kitajima, T. Watanabe, H. Izumi, K. Kohno, Y. Sasaguri, Overexpression of peroxiredoxin 4 protects against high-dose streptozotocin-induced diabetes by suppressing oxidative stress and cytokines in transgenic mice, Antioxid. Redox Signal. 13 (2010) 1477–1490. [27] I. Mehmeti, S. Lortz, M. Elsner, S. Lenzen, Peroxiredoxin 4 improves insulin biosynthesis and glucose-induced insulin secretion in insulin-secreting INS-1E cells, J. Biolumin. Chemilumin. 289 (2014) 26904–26913. [28] J. Zheng, S.W. Edelman, G. Tharmarajah, D.W. Walker, S.D. Pletcher, L. Seroude, Differential patterns of apoptosis in response to aging in Drosophila, Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 12083–12088.
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbadis
The role of peroxiredoxin 4 in inflammatory response and aging Vladimir I. Klichko, William C. Orr, Svetlana N. Radyuk ⁎ Department of Biological Sciences, Southern Methodist University, Dallas, TX, USA
a r t i c l e
i n f o
Article history: Received 13 October 2015 Received in revised form 25 November 2015 Accepted 8 December 2015 Available online 9 December 2015 Keywords: Peroxiredoxin Immunity Inflammation Aging Apoptosis Drosophila
a b s t r a c t In prior studies, we determined that the moderate overexpression of the Drosophila endoplasmic reticulum (ER)localized peroxiredoxin (Prx), dPrx4, reduced oxidative damage and conferred beneficial effects on life span, while a high-level expression increased the incidence of tissue-specific apoptosis and dramatically shortened longevity. The detrimental pro-apoptotic and life-shortening effects were attributed to aberrant localization of dPrx4 and the apparent ER stress elicited by dPrx4 overexpression. In addition, the activation of both the NF-κB- and the JAK/STAT-mediated stress responses was detected, although it was not clear whether these served as functional alarm signals. Here we extend these findings to show that the activation of the NF-κB-dependent immunity-related/inflammatory genes, associated with life span shortening effects, is dependent on the activity of a Drosophila NF-κB ortholog, Relish. In the absence of Relish, the pro-inflammatory effects typically elicited by dPrx4 overexpression were not detected. The absence of Relish not only prevented the hyperactivation of the immunity-related genes but also significantly rescued the severe shortening of life span normally observed in dPrx4 overexpressors. The overactivation of the immune/inflammatory responses was also lessened by JAK/STAT signaling. In addition, we found that cellular immune/ pro-inflammatory responses provoked by the oxidant paraquat but not bacteria are mediated via dPrx4 activity in the ER, as the upregulation of the immune-related genes was eliminated in flies underexpressing dPrx4, whereas immune responses triggered by bacteria were unaffected. Finally, efforts to reveal critical tissues where dPrx4 modulates longevity showed that broad targeting of dPrx4 to neuronal tissue had strong beneficial effects, while targeting expression to the fat body had deleterious effects. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Changes in redox-sensitive signaling due to fluctuations in the cellular redox milieu lead to functional consequences in a variety of biological processes, including stress response, inflammation, and apoptosis. While such responses are potentially adaptive in nature, permitting cells to overcome various exogenous and endogenous challenges, chronic, or inadequate responses may have a deleterious impact. Consequently, a robust redox homeostatic response is considered essential in maintaining healthy cell function, whereas its disruption may inform/promote a multitude of disease states. A pivotal role in maintaining cellular redox homeostasis and proper redox-sensitive signaling belongs to the peroxiredoxins, a family of thiol-dependent peroxidases that possess the capacity to act as redox sensors. Peroxiredoxins (Prx) not only have the capacity to control local hydrogen peroxide fluxes but have also been shown to transmit
Abbreviations: ER, endoplasmic reticulum; Prx, peroxiredoxin; AMP, antimicrobial peptide; UPR, unfolded protein response; Act, actin; Da, daughterless; AttD, attacin D; Dipt, diptericin; Drs, drosomycin; TotA, Turandot A; ROS, reactive oxygen species. ⁎ Corresponding author at: 6501 Airline Rd, Room 113, Dallas, TX 75275, USA. E-mail address: [email protected] (S.N. Radyuk).
http://dx.doi.org/10.1016/j.bbadis.2015.12.008 0925-4439/© 2015 Elsevier B.V. All rights reserved.
H2O2-mediated protein sulfhydryl modifications to specific target proteins [1]. Endoplasmic reticulum (ER) is one of the major sources of cellular H2O2, where it is produced during the oxidative disulfide formation of newly synthesized proteins [2]. Recently, we reported that Prx4, which resides in the ER, is involved in modulating Drosophila longevity under normal conditions or in the presence of various environmental stressors [3]. Prx4 serves a dual function in the ER. It both maintains proper H2O2 concentrations, generated during protein oxidative folding [4], and also uses hydrogen peroxide to catalyze the formation of disulfide bonds in protein disulfide isomerase [5,6]. The moderate ectopic overexpression of Prx4 was found to have beneficial effects on fly physiology, including a significant extension in life span. By contrast, the global high-level overexpression of dPrx4 (N 5-fold) elicited increased incidence of tissue-specific apoptosis and shortened life span concomitant with overproduction of antimicrobial peptides (AMP), a signature of the NF-κB signaling in the immune and inflammatory responses [3], and is reminiscent of the alarm signal triggered by ER stress due to accumulation of unfolded proteins, the so-called unfolded protein response (UPR) [7,8]. The activation of the NF-κB signaling pathway is presumed to safeguard cellular viability and function, but when chronically induced, it may become disruptive, leading to cell suicide as a last resort to dispose of dysfunctional cells where the accumulation of misfolded proteins has overwhelmed repair capacity [9].
266
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273
The modulation of the ER redox domain by manipulating Prx4 levels provides the opportunity to experimentally establish a state of chronic inflammation and identify specific pathway elements that mediate this ER-derived stress and thus facilitate the design of interventions to manage inflammatory responses and apoptosis.
2. Materials and methods 2.1. Fly strains and procedures The UAS-dPrx4 transgenic lines, RNAi-dPrx4 mutants, RNAi mutants for the genes representing the JAK/STAT pathway (hopscotch, Stat92E, and domeless), and driver fly lines were those described by Radyuk et al. [3]. The S106-inducible GeneSwitch-GAL4 fat body-specific driver was kindly supplied by Blanka Rogina (University of Connecticut Health Science Center). A null mutant strain for the NF-κB-like transcription factor Relish, relE20, was obtained from the Bloomington Stock Center. All fly lines were backcrossed into our reference y w genetic background. Fly collection and husbandry were as described by Radyuk et al. [3]. Briefly, flies overexpressing dPrx4 at high levels and flies overexpressing dPrx4 in combination with domeless RNAi-mediated underexpression were developed at 18 °C because of the lethal effects on development at 25 °C. Adults were reared at 25 °C after hatching. The underexpression of Hopscotch and Stat92E by RNAi was achieved with high-level global daughterless (Da) or Actin (Act) GAL4 drivers, and the overexpression of dPrx4 with S106-Switch driver was achieved by feeding flies food supplemented with 100 μg/ml mifepristone (RU486). Controls included the RNAi transgene target alone and the driver alone except for the S106 GeneSwitch experiment where genetically identical flies fed food containing the mifepristone solvent, ethanol, served as control. Responses to oxidants and infection were studied by feeding flies with 1% sucrose solution containing 10 mM paraquat or septically injuring flies by pricking with a needle dipped in an Escherichia coli suspension, as detailed by Radyuk et al. [3]. Survivorship studies were conducted as described previously [10].
2.2. RT-PCR Primers specific for AMP genes and Turandot A (TotA) were those indicated in publications [3,11]. Real-time PCR analysis was performed as described previously [11], where PCR reactions were performed with SYBR Green fluorescent dye (Molecular Probes), and the signals obtained for each gene were standardized against signals obtained for the rp49 housekeeping gene.
2.3. TUNEL labeling The assessment of apoptosis-induced DNA fragmentation was made in cryosections prepared from whole flies using the In Situ Cell Death Detection Kit, TMR red (Roche, Indianapolis, IN, USA), as detailed previously [3]. Images were acquired by fluorescence microscopy (Zeiss), using AxioVisionLE4_3 software.
2.4. Statistical analysis All statistics were calculated using Prism for Macintosh version 4.0a software (GraphPad Software, Inc.). Differences in mRNA levels were compared between groups by analysis of variance. In studies of life span, the mean survivorship time and differences between survivorship curves were assessed using the log-rank test.
3. Results 3.1. Induction of AMPs and TotA in response to Prx4 overexpression requires Relish As reported by us previously [3], dPrx4 overexpression at high levels (N5-fold) caused an increase in the expression of both the NF-κBdependent immune response/pro-inflammatory genes, AMPs, and the cytokine-like protein Tot A, a target of JAK/STAT signaling. Given that the NF-κB transcription factor Relish has been identified as a common regulator of the expression of AMPs and Turandot proteins [12] as well as a mediator of responses elicited due to ER stress [8], we postulated that the stress response due to Prx4 overload is mediated via Relish. To test this idea, we measured the Prx4-driven stress response in the absence of Relish using the rel20 null mutant. The results of this analysis, as shown in Fig. 1, revealed that the activity of Relish is indeed required for induction of AMPs and TotA in response to the apparent ER stress caused by dPrx4 overload. The upregulation of the AMPs attacin D (AttD) and diptericin (Dipt), as well as TotA, normally observed in flies overexpressing dPrx4, was completely eliminated in the relish mutant. We also observed lower levels of drosomycin (Drs) in flies underexpressing Relish. Notably the basal levels of AMPs were lower in 10 days old rel20 mutant flies compared to controls. It is tempting to conjecture that this transcription factor is also critical for the age-dependent increase in AMP transcripts levels that occurs in normal flies. 3.2. JAK/STAT signaling mitigates the activation of NF-κB-dependent immunity genes JAK/STAT signaling has been previously shown to repress NF-κBdependent immune signaling. In a series of SL2 cell culture studies, RNAi was used to reduce levels of the JAK/STAT signaling factor Stat92E, giving rise to a significant increase in the expression of Relish-mediated immunity genes and providing support for the notion that Stat92E serves as a Relish antagonist [13]. To determine whether a similar relationship could be discerned in response to stress elicited by dPrx4 overexpression, we analyzed the expression of AMP genes in RNAi mutant flies underexpressing three distinct elements of the JAK/STAT pathway (Stat92E, Hopscotch, and Domeless) in the presence or absence of Prx4 overexpression. Consistent with previous findings ((3) and Fig. 1), the overexpression of dPrx4 alone at high levels resulted in AMP induction, particularly evident for AttD and Dipt (Fig. 2). The underexpression of the domeless and Stat92E genes of the JAK/STAT pathway resulted in increases in AMPs similar to those observed for Prx4 overexpression alone, while the underexpression of hopscotch, the Drosophila JAK homologue, resulted in a particularly strong increase in AMP levels relative to those elicited by Prx4 overexpression and comparable to the increase during aging (55-day-old control). The AMP activation effects were not diminished in response to a combination treatment (Prx4 overexpression + JAK/STAT knockdown), and in fact some synergy was evident in three cases (See arrows, Fig. 2). Overall, these results support the notion that JAK/STAT signaling plays an inhibitory role in the Relishdependent activation of AMPs. 3.3. dPrx4 is required for induction of Relish-dependent AMPs in response to PQ but not infection There is evidence that exposure to oxidants or infection (particularly viral) may provoke ER stress and, as a consequence, lead to proinflammatory responses mediated via the NF-κB pathways [8,9]. We reasoned that if the ER-specific Prx4 mediates AMP activation in response to oxidative stress and septic injury, the knockdown of Prx4 would abolish this activation effect. Consequently, levels of Prx4 were knocked down N90% by RNAi, and flies were subsequently subjected
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273
267
Fig. 1. Effects of Relish on induction of the immune and stress response genes in flies overexpressing dPrx4. All flies were 10 days old. C—driver control (Actin-GAL4/+); ActNdPrx4—flies overexpressing dPrx4 using Act-GAL4 driver; rel—rel20 mutant; ActNdPrx4; rel—flies overexpressing dPrx4 in rel20 mutant background. RT-PCR analysis was conducted with two independent cohorts of flies and replicated for each cohort. Shown are means ± SEM (n = 4). Asterisks denote statistically significant differences (*P b 0.05; **P b 0.005; ***P b 0.0005, and ****P b 0.00005).
to paraquat treatment (oxidative stress) or septic injury with E. coli, after which AMP activation was assessed. RT-PCR analysis of diptericin and attacin D genes, targets of Relish, showed that dPrx4 activity is required for the activation of Relish signaling in response to paraquat, as the activation of Dipt and AttD, observed in the control, was absent in the dPrx4 RNAi mutants (Fig. 3). No differences in expression were observed in septically injured flies, suggesting that the upregulation of AMPs in response to bacteria does not depend on dPrx4-mediated signaling in the ER and is instead regulated by the canonical Imd/Relish signaling. There were also no effects of dPrx4 on the upregulation of the Toll/Dif-dependent drosomycin (Drs) in flies stressed with either paraquat or E. coli (Fig. 3), suggesting that the signals mediated via ER do not affect this branch of immunity. Thus, Relish activation appears to be a specific outcome of exposure to oxidants mediated via Prx4/ER. 3.4. A rapid death phenotype caused by the dPrx4 overexpression is partially rescued in relish mutant Previously, we reported that dPrx4 significantly shortened life span when overexpressed globally at high levels and that these lifeshortening effects were accompanied by tissue-specific apoptosis [3]. Here we tested whether this reduced longevity could be mediated by NF-κB signaling triggered by ER stress. In a relish null background, the reduction in longevity caused by dPrx4 overload was significantly attenuated (Fig. 4 and Table 1), supporting the idea that Relish mediates at least in part the inflammatory responses originating in the ER. These inflammatory responses can be destructive and further exacerbate ER stress by contributing to a greater pro-oxidative environment. Previously, we noted that dPrx4 overexpression also had deleterious temperature
effects on development, displaying a recessive lethal phenotype when cultured at 25 °C, but viable when cultured at 18 °C. This recessive lethal phenotype was partially reversed in the absence of Relish, in that a significant number of female flies and to a lesser degree male flies overexpressing dPrx4 in a relish null background (ActinNdPrx4; rel) eclosed at 25 °C (data not shown). Consistent with previous observations [3], no male or female flies expressing high levels of dPrx4 (Act NdPrx4) in a rel+ background emerged at 25 °C. 3.5. Relish modulates cell death triggered by the overexpression of dPrx4 The increase of tissue-specific apoptosis in response to dPrx4 overexpression has been ascribed to the release of dPrx4 from the ER to the cytosol, where it is thought to antagonize IAP1 and thereby liberate pro-apoptotic caspases [14]. NF-κB signaling involving Relish has been implicated in other models of cell death under investigation in Drosophila [15], and consequently we set out to establish the significance of Relish-dependent signaling on apoptosis in our dPrx4 model. As we found previously, the overexpression of dPrx4 resulted in increased occurrence of apoptotic cells in fly muscles and fat body tissue, as was determined by the TUNEL assay [3]. To examine if Relish plays a role in mediating the observed apoptotic changes caused by Prx4, we examined dPrx4 and dPrx4;rel flies for tissue-specific apoptosis (Fig. 5). In agreement with previous observations, a higher number of cells displaying DNA fragmentation was observed in thoracic muscle and abdominal fat body tissues. These patterns of apoptosis were not observed in similar regions in cryosections made from flies overexpressing dPrx4 in a relish null background, suggesting that tissue-specific apoptosis is mediated, at least in part, via the Relish pathway. Consistent
268
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273
with previous observations [3], we were unable to detect apoptotic changes in the heads of flies overexpressing dPrx4. 3.6. Prx4 overexpression in neuronal tissue extends life span while overexpression in the fat body had negative effects on fly survivorship The negative effects of dPrx4 overexpression on life span observed when driven by a high-level constitutive driver such as Da can be contrasted with the relatively beneficial effects observed when driven by a weak constitutive driver (armadillo), suggesting a dosage effect. Nevertheless, it cannot be ruled out that these differences in longevity may be ascribed to tissue-specific expression patterns. To explore this, we expressed dPrx4 ectopically in neuronal tissue, which has been shown to be particularly sensitive to changes in redox state, and in fat body tissue, the main site of the production of immune-related and stress response genes regulated by Imd/ and JAK/STAT signaling [10,16]. Survivorship was assessed in flies in which dPrx4 was ectopically expressed either in neuronal tissue using the pan-neuronal APPLGAL4 driver or in the fat body using the S106-GeneSwitch-GAL4 driver. The overexpression of dPrx4 neuronally resulted in significant positive effects on survivorship under normal conditions (Fig. 6 A and Table 2 A). In contrast, when dPrx4 production was enhanced specifically in the fat body, the life spans of flies were shortened by 4-8 % relative to controls (Fig. 6 B and Table 2 B). 4. Discussion A major finding of this study is that the immune/inflammatory and cell death responses, triggered by disturbances in ER function or exposure to exogenous reactive oxygen species (ROS), depend on the activity of the ER-specific peroxiredoxin dPrx4. Previously, we showed that the high-level overexpression of dPrx4 elicits a state of ER stress leading to aberrant immune response and negative functional consequences [3]. Here, we establish that the effects of dPrx4 on the responses derived from the ER are mediated via NF-κB (Relish)-dependent signaling and, furthermore, that these elevated Relish-dependent immune/inflammatory responses can account for a significant portion of the adverse effects on Drosophila physiology, including the shortening of life span. In fruit flies, the management of immune function and pro-inflammatory response relies on overlapping strategies, involving the recruitment of immune cells (hemocytes), the burst of ROS, and the release of other compounds intended for elimination of infectious agents, including various AMPs, whose production is regulated via the immune and stress response pathways (reviewed by Shaukat et al. [17]). These pathways, including JNK and NF-κB, are also utilized to elicit a state of inflammation in response to sterile triggers, such as molecular damage, and the ER stress described in the present study. The first evidence for the involvement of Prx4 in the TNF-αdependent activation of the NF-κB signaling cascades came from studies of the Rhee group, conducted with cultured HeLa cells [18]. Proinflammatory TNF-α/NF-κB signaling in mammals is analogous to the Imd/Relish signaling in Drosophila. To place these results in a more physiological context, we overexpressed dPrx4 in flies at elevated levels and evoked a UPR-like inflammatory response [3]. Indeed, we observed increased transcription levels of the AMPs in our dPrx4 overexpressing Fig. 2. Effects of the silencing of JAK/STAT signaling on AMP expression in flies expressing normal and high levels of Prx4. All flies were 10 days old. Control flies were 10 days old and 55 days old (C, 55 days). C—driver control (Da-GAL4/+); dPrx4Nda—flies overexpressing dPrx4 by Da-GAL4 driver. dome, hop, and stat—RNAi mutants underexpressing Domeless, Hopscotch, and Stat92E [3]. DaNdPrx4; dome, hop, or stat—flies overexpressing dPrx4 and underexpressing Domeless, Hopscotch, and Stat92E. Two different RNAi lines were used for each gene, and two individual dPrx4 transgenic lines, giving comparable results; hence, average values are presented. Shown are means ± SEM. Analyses were conducted in triplicate for each fly line and reproduced with 2 independent cohorts of flies (n = 6). Asterisks denote statistically significant differences between control (C) and experimental lines (*P b 0.05). Arrows indicate synergistic effects of dPrx4 overexpression and underexpression of the components of JAK/STAT signaling.
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273
269
Fig. 4. Effects of dPrx4 overexpression on fly life span in the presence and absence of Relish. ActNdPrx4—flies overexpressing dPrx4 by Actin-GAL4 driver; ActNdPrx4;rel—flies overexpressing dPrx4 in the absence of Relish. Controls included flies doubly heterozygous for the Act driver and the rel20 mutation (Act/+; rel/+) as well as flies heterozygous for the driver and homozygous for the rel20 mutation (Act/+; rel/rel), all backcrossed into the y w reference background. Median ages of flies (50% survivorship) are indicated in Table 1. Statistically significant differences were observed between ActNdPrx4 vs. control (Pb0.05) and relish mutant, and ActNdPrx4 vs. ActNdPrx4;rel as determined by log-rank test.
found to be mediated via Relish, as the activation of AMPs elicited by dPrx4 overexpression does not occur in the absence of Relish activity (Fig. 1). Furthermore, the absence of Relish not only prevented the activation of AMPs observed in Prx4 overexpressors but also reversed the life span shortening effects in these flies along with the accompanying pro-apoptotic changes (Figs. 4 and 5). Such results are consistent with a causal connection between Relish-mediated AMP activation and longevity effects. While we clearly established a role for Relish in Prx4/ER-mediated pro-inflammatory/pro-apoptotic signaling, the exact mechanisms that culminate in the life span shortening effects are still not clear. On one hand, such effects could be due to apoptosis, observed predominantly in the fly muscle and fat body of Prx4 overexpressors [3]. On the other hand, the activation of Relish-dependent AMPs, markers of inflammation, could themselves contribute to life span reduction. Support for this latter idea is the observation that the overexpression of AMPs, such as drosocin, attacin, defensin, and drosomycin, in young flies induced neurodegeneration comparable to that observed in old flies [19]. Furthermore, since the deleterious effects of these AMPs were not suppressed in Relish knockdowns, it allowed the authors to conclude that the prolonged exposure of the host cells to high levels of AMPs is sufficient to cause neurodegeneration [19]. In agreement with these findings, we determined that the global high-level overexpression of AMPs, in particular attacin and defensin, or overexpression targeted to the fat body had significant life-shortening effects and these effects correlated with a marked increase in tissue-specific cell death (unpublished observations). It is not clear whether the cell death observed in Prx4 overexpressors, is a direct result of the cytotoxic effects of AMPs or rather is elicited by inappropriate Relish signaling that mediates its effects by an alternative pathway, as has been documented for a case of apoptosis-related retinal tissue degeneration [15]. The possibility that pro-apoptotic changes are caused by a combination of factors, Fig. 3. Effects of dPrx4 underexpression on the activation of Imd/Relish-dependent AMPs. All flies were 10 days old. C—driver control (Da-GAL4/+); DabRNAi-dPrx4—RNAi mutants underexpressing dPrx4 using Da-GAL4 driver. Flies were untreated, exposed to paraquat, or septically injured with a suspension of Escherichia coli, as specified in Section 2. Shown are means ± SEM. Analyses were conducted in triplicate with two different RNAi fly lines and reproduced with 2 independent cohorts of flies (n = 6). Asterisks denote statistically significant differences (*P b 0.05).
flies ([3] and Figs. 1 and 2), a known signature of ER stress, and reminiscent of a response caused by a mutation in the transmembrane protein, a constituent of ER [8]. These responses caused by dPrx4 overload were
Table 1 Median age (days) of flies overexpressing dPrx4 and relish mutants. Genotype
Actin/+,dPrx4/+
rel
ActinNdPrx4
ActinNdPrx4;rel
Experiment 1 Experiment 2
49.5 51.5
53.0 52.0
24.0 19.0
34.5 32.5
Differences in median age between flies overexpressing dPrx4 in a normal background (ActinNdPrx4) or relish mutant background (ActinNdPrx4;rel) and heterozygote driver and transgene control (Actin/+,dPrx4/+), as well as relish (rel) mutant were statistically significant. The rescue effect of the relish mutant is statistically significant, as determined by log-rank test (P b 0.05).
270
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273
Fig. 5. TUNEL assay of cell death in flies overexpressing dPrx4 in normal and relish mutant background. Cryosections were made from 10-day-old flies overexpressing dPrx4 by Actin-GAL4 driver (ActNdPrx4) and flies overexpressing dPrx4 in rel20 null mutant background (ActNdPrx4;rel). Representative images of the abdominal and thoracic regions show a reduction of apoptosis in abdominal fat body tissue and gut epithelia, as well as in thoracic muscles in samples made from flies overexpressing dPrx4 in the absence of Relish. ms—muscle; g—gut lumen; fb—fat body. No apoptotic cells were observed in the heads of flies overexpressing dPrx4 in normal or rel mutant background.
including missexpression of AMPs, aberrant Relish signaling and IAP1dependent signaling, triggered by cytosolic Prx4 [14] cannot be ruled out, as summarized in Fig. 7. Another important finding of the present study is that the inflammatory response to the oxidant paraquat is mediated via ER signaling that is dependent on dPrx4 activity, while the response to infection seems mainly regulated by classical immune pathways, independent of ER. This is based on the observation that the induction of the Relish/NFκB-dependent AMPs failed in RNAi dPrx4 mutants exposed to paraquat but occurred normally in mutants infected with bacteria (Fig. 3). There is precedent for oxidant-induced inflammation mediated through ER stress [8,9]. A current model for this “sterile” inflammation envisions
mitochondria as a primary target, where functional disruption leads to redox state fluctuations that elicit ER stress, which in turn promotes NF-κB-mediated inflammation [9,20]. This study is a first report that the ER-specific Prx4 is a critical control factor for ER inflammatory signaling and is required for alarm and, perhaps, adaptation responses. Furthermore, the data obtained suggest that these responses do not depend simply on antioxidant function of dPrx4 and changes in ROS generation, as the higher ROS/peroxide levels observed in the dPrx4 knockdown were accompanied by only a minimal increase in AMPs and had no effect on life span ([3] and Fig. 3). One possible explanation for this seemingly contradictory finding is that dPrx4 functions primarily as a relay in transmitting stress signals via interaction with yet to be
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273
Fig. 6. The effects of tissue-specific dPrx4 overexpression on fly LS. (A) Life spans of flies overexpressing dPrx4 in neuronal tissues with the Appl-GAL4 driver. Controls are Appl-GAL4 driver (+/Appl) and three different UAS-dPrx4 transgenic lines (A, B, or C/+). Experimental flies overexpressing dPrx4 are ApplNdPrx4. Data are representative of two replicate experiments, summarized in Table 1 A. (B) Life spans of flies overexpressing dPrx4 in the fat bodies with the S106 pSwitch-inducible driver. Controls were pSwitch106NdPrx4 flies fed ethanol; experimentals were flies fed with mifepristone (+RU486). Data are representative of two replicate experiments, summarized in Table 1 B.
determined targets. Such a relay system involving Prx2 and JAK/STAT signaling has been reported recently [21]. Components of the three known UPR pathways (ATF6, PERK, or Ire1) are potential targets for dPrx4-mediated effects.
271
In this study, we also further clarified the function of the secreted form of Prx4 and found that it may play a role in mitigation of inflammatory response. Previously, we have reported that various stresses or high-level Prx4 overexpression causes subcellular re-distribution of dPrx4, which triggers the activation of the JAK/STAT pathway as inferred from the upregulation of the stress response protein Turandot A [3]. The JAK/STAT pathway controls several biological processes in Drosophila during development and in adults and directly contributes to immune and stress response by activating infection-induced genes [16,22]. In response to septic injury or other stressors, such as heat-shock or dehydration, hemocytes produce the Upd3 ligand, which initiates the expression of several immune-related proteins, including cytokines and stress response proteins [12]. Our results suggest that in the case of Prx4 overexpression, the role of JAK/STAT signaling is to suppress Relish-mediated proinflammatory responses, consistent with the established interplay between NF-κB and STAT transcription factors [13,23]. Perhaps, one of the functions of the secreted dPrx4 is to suppress excessive inflammatory response via a regulatory loop where the activity of the ER-localized dPrx4 in eliciting the NF-κB responses is antagonized by the same enzyme, re-localized to a different milieu (Fig. 7). Finally, we have found that the in vivo effects of Prx4 on life span vary not only with respect to different levels of global overexpression, as reported earlier [3], but are also sensitive to overexpression in specific tissues (Fig. 6). Similar to a previous finding, where life span extension was achieved by enhancing the pro-reducing capacity of Drosophila neuronal tissues by increasing glutathione production [10], pan-neuronal overexpression of dPrx4 had a strong beneficial effect on longevity (Fig. 6 A). It is tempting to ascribe these life span-extending effects to increased antioxidant function as well as enhancement of oxidative folding capacity. The brain tissues of Drosophila, being aerobic, are the most vulnerable to oxidative stress and exposure to ROS, such as H2O2, which is mainly produced by mitochondria and ER during respiration and in the process of protein oxidative folding [24,25]. To date there have been no report of longevity effects for other animal models, including transgenic mice overexpressing Prx4. On the other hand elevated Prx4 levels in specific tissues have been shown to play a preventative role in the development of diseases, elicited by various stressors. Studies in mice transgenics expressing the human Prx4 transgene displayed a significant reduction in atherosclerotic lesions, which was attributed to its ROS-scavenging capacity in extracellular space, such as vascular vessels [25]. Transgene-expressed Prx4 might also prevent the development of the drug-induced diabetes caused by destruction of beta cells in pancreatic islets, where the effects of Prx4 were ascribed to ROS-scavenging and thus anti-inflammatory and anti-apoptotic function of this enzyme [26]. On the other hand, the
Table 2 Mean age of flies overexpressing dPrx4 in pan-neuronal tissue (A) and the fat body (B) A Genotype
Mean (days)
% vs. transgene/+
% vs. driver/+
+/dPrx4-A +/dPrx4-B +/dPrx4-C Appl/+ ApplNdPrx4-A ApplNdPrx4-B ApplNdPrx4-C
57.6; 61.2 54.1; 57.8 45.8; 59.1 51.2; 56.8 67.3; 62.3 71.3; 55.1 71.0; 64.2
116.8; 101.8 131.8; 95.3 155.0; 108.6
131.4; 109.7 139.3; 97.0 138.7; 113.1
B Genotype
Mean (days) RU486-
Mean (days) RU486+
% RU486/+ vs. RU486/-
S106NdPrx4-A S106NdPrx4-B S106NdPrx4-C
35.5; 38.1 27.1; 35.5 29.1; 35.3
34.0; 36.5 28.8; 32.6 27.7; 33.1
95.8; 95.8 106.3; 91.8 95.2; 93.8
Values obtained in two independent experiments are listed in column 1 (A) and columns 1 and 2 (B). The overall percentage of the mean age changes between experimentals vs. the transgene/+ and driver/+ controls or experimental flies fed mifepristone are indicated in columns 2 and 3 (A) and in column 3 (B). Statistically significant differences between control and experimental flies, determined by log-rank test (P b 0.05), are indicated by bold.
272
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273
Fig. 7. Proposed model of the effects of dPrx4 on cell function. The scheme summarizes our previous [3] and current findings. High-level overexpression of the ER dPrx4 causes release of this protein into the cytosol and its secretion into extracellular space. The extracellular form of dPrx4 activates the JAK/STAT pathway and downstream response genes. The cytosolic form of dPrx4 promotes apoptotic cell death. The ER-localized dPrx4 is involved in the regulation of the NF-κB-dependent immune/inflammatory responses and apoptosis. The complex interactions between pathways modulated by dPrx4 are depicted. The links indicated by dotted lines, including potential cross talk between ER and mitochondrion (MT), have yet to be established.
overexpression of Prx4 improved the function of insulin-secreting INS1E cells, which was interpreted as an ability of this Prx to enhance the ER protein oxidative folding capacity [27]. In contrast, the beneficial antioxidant effects of Prx4 overexpression were apparently offset by pro-inflammatory changes observed in flies with elevated global or fat body-specific Prx overexpression ([3] and Fig. 6 B). Perhaps then the biological function of Prx4 is largely determined by cellular context. Thus, the overexpression of dPrx4 in the fat body, a major site of AMP production, would be likely to enhance AMP-mediated inflammation, whereas the overexpression of dPrx4 in brain tissue does not promote AMP production and the beneficial antioxidant effects of dPrx4 predominate. These differential effects may be due in part to differential expression, and perhaps, the activation of Relish in the brain and fat body tissues. Furthermore, we found no evidence of apoptosis in the brains of flies overexpressing dPrx4 while pro-apoptotic changes were clearly promoted in other tissues, such as thoracic muscles and fat bodies, as determined in the current study (Fig. 5) and reported previously [3]. Thus, higher levels of dPrx4 in the brain tissue do not appear to induce neuronal apoptosis and consequently do not have negative effects on fly physiology. In agreement with these findings, apoptotic pathways are not always activated in the nervous system in response to mutations that lead to neuroinflammation [19] or in old flies, although other tissues display pro-apoptotic changes [28]. To conclude, we found that dPrx4 plays a critical role in controlling inflammatory response in Drosophila and these effects vary according to tissue context. The data are summarized in the following scheme (Fig. 7). As reported earlier [3], the overexpression of dPrx4 or stressors, such as oxidants or bacteria, causes re-localization of dPrx4 to different subcellular compartments where this protein has differential effects on biological processes. The extracellular form of dPrx4 activates JAK/STAT signaling, resulting in the upregulation of the stress response Turandot proteins. The activation of JAK/STAT signaling also antagonizes the
activation of the Relish-dependent AMPs, as their levels are significantly higher in the JAK/STAT pathway mutants (Fig. 2). Unlike the extracellular dPrx4, which seems to serve a pro-survivorship function, the cytosolic dPrx4 has been reported to promote apoptosis due to interaction with IAP1 [14]. Apoptosis can also be elicited via Relish-dependent signaling, although it is not clear whether cell death is a direct effect of Relish or a secondary response to Relish-dependent inflammation. Besides the apoptosis and activation of JAK/STAT signaling, the overexpression of dPrx4 also activates induction of AMPs and this induction is dependent on the activity of the NF-κB factor Relish. The activation of the immune response (AMPs) by bacteria is independent of the ER-localized dPrx4 and is mediated through a canonical immune pathway (Imd), unlike the upregulation of the immune/inflammatory responses to the ROS generator paraquat, which is mediated via ER signaling and requires dPrx4 activity. Thus, it may be concluded that the inflammatory and pro-apoptotic effects of dPrx4 in response to sterile stimuli are mediated via the NF-κB analog Relish by the activation of the ER/Relish signaling branch, independent of the canonical Imd/Relish immune pathway. Transparency document The Transparency document associated with this article can be found, in online version. Acknowledgments This work was supported by the grant R01 AG032342 from the National Institute on Aging/National Institutes of Health. References [1] S.G. Rhee, H.A. Woo, I.S. Kil, S.H. Bae, Peroxiredoxin functions as a peroxidase and a regulator and sensor of local peroxides, J. Biol. Chem. 287 (2012) 4403–4410.
V.I. Klichko et al. / Biochimica et Biophysica Acta 1862 (2016) 265–273 [2] B.P. Tu, J.S. Weissman, The FAD- and O(2)-dependent reaction cycle of Ero1mediated oxidative protein folding in the endoplasmic reticulum, Mol. Cell 10 (2002) 983–994. [3] S.N. Radyuk, V.I. Klichko, K. Michalak, W.C. Orr, The effect of peroxiredoxin 4 on fly physiology is a complex interplay of antioxidant and signaling functions, FASEB J. 27 (2013) 1426–1438. [4] T.J. Tavender, N.J. Bulleid, Peroxiredoxin IV protects cells from oxidative stress by removing H2O2 produced during disulphide formation, J. Cell Sci. 123 (2010) 2672–2679. [5] Z. Cao, T.J. Tavender, A.W. Roszak, R.J. Cogdell, N.J. Bulleid, Crystal structure of reduced and of oxidized peroxiredoxin IV enzyme reveals a stable oxidized decamer and a non-disulfide-bonded intermediate in the catalytic cycle, J. Biolumin. Chemilumin. 286 (2011) 42257–42266. [6] L. Wang, L. Zhu, C.C. Wang, The endoplasmic reticulum sulfhydryl oxidase Ero1beta drives efficient oxidative protein folding with loose regulation, Biochem. J. 434 (2011) 113–121. [7] C. Xu, B. Bailly-Maitre, J.C. Reed, Endoplasmic reticulum stress: cell life and death decisions, J. Clin. Invest. 115 (2005) 2656–2664. [8] K.A. Boltz, G.E. Carney, Loss of p24 function in Drosophila melanogaster causes a stress response and increased levels of NF-kappaB-regulated gene products, BMC Genomics 9 (2008) 212. [9] N. Chaudhari, P. Talwar, A. Parimisetty, C. Lefebvre d'Hellencourt, P. Ravanan, A molecular web: endoplasmic reticulum stress, inflammation, and oxidative stress, Front. Cell. Neurosci. 8 (2014) 213. [10] W.C. Orr, S.N. Radyuk, L. Prabhudesai, D. Toroser, J.J. Benes, J.M. Luchak, R.J. Mockett, I. Rebrin, J.G. Hubbard, R.S. Sohal, Overexpression of glutamate-cysteine ligase extends life span in Drosophila melanogaster, J. Biolumin. Chemilumin. 280 (2005) 37331–37338. [11] S.N. Radyuk, K. Michalak, V.I. Klichko, J. Benes, W.C. Orr, Peroxiredoxin 5 modulates immune response in Drosophila, Biochim. Biophys. Acta 1153-1163 (1800). [12] H. Agaisse, U.M. Petersen, M. Boutros, B. Mathey-Prevot, N. Perrimon, Signaling role of hemocytes in Drosophila JAK/STAT-dependent response to septic injury, Dev. Cell 5 (2003) 441–450. [13] L.K. Kim, U.Y. Choi, H.S. Cho, J.S. Lee, W.B. Lee, J. Kim, K. Jeong, J. Shim, J. Kim-Ha, Y.J. Kim, Down-regulation of NF-kappaB target genes by the AP-1 and STAT complex during the innate immune response in Drosophila, PLoS Biol. 5 (2007), e238. [14] T. Tenev, A. Zachariou, R. Wilson, A. Paul, P. Meier, Jafrac2 is an IAP antagonist that promotes cell death by liberating Dronc from DIAP1, Embo J. 21 (2002) 5118–5129. [15] Y. Chinchore, G.F. Gerber, Dolph, P, J. Alternative pathway of cell death in Drosophila mediated by NF-kappaB transcription factor Relish. Proc. Natl. Acad. Sci. U. S. A. 109 (2012) E605–E612.
273
[16] H. Agaisse, N. Perrimon, The roles of JAK/STAT signaling in Drosophila immune responses, Immunol. Rev. 198 (2004) 72–82. [17] Z. Shaukat, D. Liu, S. Gregory, Sterile inflammation in Drosophila, Mediat. Inflamm. 2015 (2015) 369286. [18] D.Y. Jin, H.Z. Chae, S.G. Rhee, K.T. Jeang, Regulatory role for a novel human thioredoxin peroxidase in NF-kappaB activation, J. Biolumin. Chemilumin. 272 (1997) 30952–30961. [19] Y. Cao, S. Chtarbanova, A.J. Petersen, B. Ganetzky, Dnr1 mutations cause neurodegeneration in Drosophila by activating the innate immune response in the brain, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) E1752–E1760. [20] S.Z. Hasnain, R. Lourie, I. Das, A.C. Chen, M.A. McGuckin, The interplay between endoplasmic reticulum stress and inflammation, Immunol. Cell Biol. 90 (2012) 260–270. [21] M.C. Sobotta, W. Liou, S. Stocker, D. Talwar, M. Oehler, T. Ruppert, A.N. Scharf, T.P. Dick, Peroxiredoxin-2 and STAT3 form a redox relay for H2O2 signaling, Nat. Chem. Biol. 11 (2015) 64–70. [22] M. Boutros, H. Agaisse, N. Perrimon, Sequential activation of signaling pathways during innate immune responses in Drosophila, Dev. Cell 3 (2002) 711–722. [23] T. Kim, J. Yoon, H. Cho, W.B. Lee, J. Kim, Y.H. Song, S.N. Kim, J.H. Yoon, J. Kim-Ha, Y.J. Kim, Downregulation of lipopolysaccharide response in Drosophila by negative crosstalk between the AP1 and NF-kappaB signaling modules, Nat. Immunol. 6 (2005) 211–218. [24] B.P. Tu, J.S. Weissman, Oxidative protein folding in eukaryotes: mechanisms and consequences, J. Cell Biol. 164 (2004) 341–346. [25] X. Guo, S. Yamada, A. Tanimoto, Y. Ding, K.Y. Wang, S. Shimajiri, Y. Murata, S. Kimura, T. Tasaki, A. Nabeshima, T. Watanabe, K. Kohno, Y. Sasaguri, Overexpression of peroxiredoxin 4 attenuates atherosclerosis in apolipoprotein E knockout mice, Antioxid. Redox Signal. 17 (2012) 1362–1375. [26] Y. Ding, S. Yamada, K.Y. Wang, S. Shimajiri, X. Guo, A. Tanimoto, Y. Murata, S. Kitajima, T. Watanabe, H. Izumi, K. Kohno, Y. Sasaguri, Overexpression of peroxiredoxin 4 protects against high-dose streptozotocin-induced diabetes by suppressing oxidative stress and cytokines in transgenic mice, Antioxid. Redox Signal. 13 (2010) 1477–1490. [27] I. Mehmeti, S. Lortz, M. Elsner, S. Lenzen, Peroxiredoxin 4 improves insulin biosynthesis and glucose-induced insulin secretion in insulin-secreting INS-1E cells, J. Biolumin. Chemilumin. 289 (2014) 26904–26913. [28] J. Zheng, S.W. Edelman, G. Tharmarajah, D.W. Walker, S.D. Pletcher, L. Seroude, Differential patterns of apoptosis in response to aging in Drosophila, Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 12083–12088.