To enable delivery of the desired electrophile directly to Keap1, Z-REX requires ectopic expression of Halo-Keap1 in larval fish (Figure 1A, Figure 1—figure supplement 1A). Since Halo-hKeap1 has been proven to be amenable to on-target LDE-signaling studies primarily in cultured cells (Parvez et al., 2016; Long et al., 2017b; Parvez et al., 2015; Lin et al., 2015; Fang et al., 2013) and C. elegans (Van Hall-Beauvais et al., 2018; Long et al., 2018), we first demonstrated functionality of this fusion protein in fish. Following mRNA injection directly to the yolk sac at the 1–4 cell stage, the resulting expression of Halo-hKeap1 at 24 hr post fertilization (hpf) was: (1) ubiquitous, and (2) similar to the overall level of the endogenous zKeap1, since global Keap1-signal doubled in transgene-injected fish, relative to controls (Figure 1—figure supplement 1C). Note: our Keap1-antibody detects hKeap1 with similar or higher efficiency to zKeap1a/b-paralogs-combined (Figure 1—figure supplement 1D). We next confirmed using two independent readouts (live-imaging and mRNA-analysis) that Halo-hKeap1 was functionally active. The imaging readout used the well-established transgenic AR-reporter strain, Tg(–3.5gstp1:GFP)/it416b National BioResource Project Zebrafish, 2020 [hereafter Tg(gstp1:GFP)] (Figure 1—figure supplement 2A). In this and all subsequent experiments with these reporter fish, adult Tg(gstp1:GFP) zebrafish were crossed with wild-type (wt) zebrafish to generate embryos for experimentation. This strategy ensures transgenic progeny maintain consistent zygosity (heterozygotes) and standardization of number of reporter alleles per transgenic fish (Figure 1B). Imaging data were further backed-up by qRT-PCR analysis of Nrf2/AR-driven endogenous downstream genes in Casper zebrafish (Figure 1—figure supplement 2B). In both instances, a drop in overall basal AR and AR-driven genes was observed selectively in zebrafish injected with Halo-hKeap1 mRNA, compared to controls injected with Halo-mRNA alone (Figure 1—figure supplement 2A-B).

We observed during these validations differential AR-responsivity between the head and tail of the fish. This intriguing disparity was first measured in Tg(gstp1:GFP) fish, wherein ubiquitous Halo-hKeap1-expression led to AR-downregulation selectively in the tail (Figure 1—figure supplement 2A). This spatially-selective response was also observed in Halo-hKeap1-mRNA-injected Casper embryos using qRT-PCR, where the tail showed a more prominent attenuation in the AR-driven endogenous zebrafish genes, than the head (Figure 1—figure supplement 2B). Thus, although the qRT-PCR and the reporter assays likely observe different pools of cells, at least in some instances, data from the reporter and qRT-PCR are in agreement.

To further demonstrate that hKeap1 functions in zebrafish in a zNrf2-dependent manner, we made use of the zNrf2a/b-MOs previously described by several independent laboratories Timme-Laragy et al., 2012; Kobayashi et al., 2002; Sant et al., 2017 and the resulting morphants in our hands also gave rise to the expected AR-suppression outcome in Tg(gstp1:GFP) fish tail (Figure 1—figure supplement 2C; left panel, 1st vs. 2^nd bars; and Figure 1—figure supplement 3A). This result is consistent with the data from Keap1-overexpression above, analyzed by both imaging and qRT-PCR analysis (Figure 1—figure supplement 2A-B), further validating the MOs. Upon Halo-hKeap1 overexpression in these zNrf2-depleted morphants, no further AR-suppression in the tail was noted (Figure 1—figure supplement 2C; left panel: 1^st-2^nd vs. 4^th-5^th bars). This epistasis is strong evidence that hKeap1 and zNrf2 reside on the same axis, as required by the current model of Keap1/Nrf2/AR-signaling (Figure 1A). Furthermore, since hKeap1-expression had no effect on AR in the head in fish treated with control morpholinos (MOs) (Figure 1—figure supplement 2C; right panel: 1^st vs. 4^th bar), as well as in non-MO-treated reporter fish (Figure 1—figure supplement 2A: 3^rd vs. 4^th bar), there may be a dominant suppressor of AR in the head.

We examined zNrf2 and zKeap1 in the head and tail of the fish separately by qRT-PCR. (Consequences of expression levels of both paralogs—present in teleost fish due to a whole genome duplication event that occurred during their evolution—are also further examined below by paralog-specific-knockdown in fish and -overexpression in cell culture). Note: these data are, as illustrated in Figure 1C (inset), for the whole tail and whole head, and thus, do not likely reflect expression levels in specific tissues in the tail, e.g., where GFP, or other genes subsequently investigated for that matter, are expressed. The qRT-PCR analysis (Figure 1C) revealed that levels of both zNrf2a and zNrf2b were overall higher in the tail than the head. zNrf2b was a minor contributor to both segments of the fish and was almost undetectable in the head. Levels of zKeap1a were overall similar in the head and the tail, but zKeap1b was overall slightly depleted in the tail.

To investigate the function of zKeap1 paralogs and their effects on AR, we initially examined effects of their targeted-knockdown in Tg(gstp1:GFP) heterozygous embryos. The zKeap1a/b-MOs were first validated: (i) by in vitro translation reporter assay (for translation-blocking ATG-MOs) (Figure 1—figure supplement 3B), (ii) by RT-PCR (for splice-blocking SPL-MOs) (Figure 1—figure supplement 4), and (iii) by measuring depletion of signal by whole-mount immunofluorescence (for both types of MOs) (Figure 1—figure supplement 5). With these validations in hand, we naively expected zKeap1a-knockdown to increase AR. However, in the tail, only depletion of both zKeap1-paralogs increased AR (single knockdown was not effective at modulating AR) (Figure 2A). More strikingly still, loss of zKeap1b decreased AR in the head, but knockdown of both zKeap1-paralogs increased AR as expected. These data indicate an intriguing, antagonistic interplay between the two zKeap1 alleles.

Figure 2 with 2 supplements see all

Download asset Open asset

Figure 2—source data 1

Quantification results.

https://cdn.elifesciences.org/articles/83373/elife-83373-fig2-data1-v4.xlsx

Download elife-83373-fig2-data1-v4.xlsx

Depletion of zNrf2a or zNrf2b (using the same MOs validated above Timme-Laragy et al., 2012; Kobayashi et al., 2002; Sant et al., 2017), decreased AR in the tail, confirming their expected AR-regulatory roles (Figure 2A left panel, Figure 1—figure supplement 2C, Figure 1—figure supplement 3A). There was little overall change in AR in the head when either Nrf2-paralog was knocked down (Figure 2A right panel, Figure 1—figure supplement 3A).

We thus investigated how AR-upregulation was affected in the different morphants by treatment with a representative bioactive LDE, 2-nonenal (NE) (Figure 1A inset). No effect was observed on the response in the head, upon depleting either zNrf2-paralog (Figure 2B right panel), but the head was unresponsive to AR regulation (Figure 2A right panel), and thus, this result is unsurprising. (Note: NE and other electrophiles can label proteins in the head and tail, as we describe below). Consistent with a previous report on Nrf2a-defective fish Mills et al., 2020, knockdown of zNrf2a suppressed the tail’s ability to mount AR in response to an electrophile challenge, whereas zNrf2b-knockdown led to a hyper-elevated AR-response in the tail (Figure 2B left panel). From the data in the tail, it is likely that overall zNrf2b countermands electrophile-induced AR-upregulation, whereas zNrf2a functions in mounting AR in response to electrophile stimulation. Similar data for some AR-related genes were previously reported for the zNrf2-paralogs Timme-Laragy et al., 2012.

Since knockdown of zNrf2-paralogs did little to affect regulation of AR in the head, both in electrophile-stimulated and non-stimulated conditions (Figure 2A–B, right panels), we focused our subsequent investigations on how zKeap1a/b-knockdown affects LDE-induced AR. Previous overexpression studies have implied that zKeap1a (unresponsive) and zKeap1b (responsive) respond differently to different electrophiles, although these studies were performed via overexpression at early embryonic stages Kobayashi et al., 2009. We found that NE-treatment of zKeap1a-depleted fish led to elevated AR in both head and tail (Figure 2B). This effect was also observed in 8-hr old embryos. In this manifold, injection of zKeap1a-mRNA rescued effects of the MO (Figure 2—figure supplement 1). Unfortunately, zKeap1a-mRNA injection had minimal effect on older embryos, likely due to the stability of zKeap1a-mRNA, or possibly due to secondary effects incurred through ectopic expression of zKeap1a. Nonetheless, the fact that zKeap1a-knockdown was sufficient to rescue response to electrophile treatment in cells expressing GFP in the head (Figure 2B, right panel) strongly implies that the presence of zKeap1a is responsible for ‘the head’s’ recalcitrance to mounting AR, following electrophile treatment. Similar, but less pronounced, effects were observed in the tail (Figure 2B, comparing the difference in the 1st and 4th bars in the left panel, vs. corresponding bars in the right panel), likely because intrinsic expression of zKeap1a within the GFP-positive cells in the tail is relatively lower than that in the head. The same posit also explains why in the absence of zKeap1a knockdown, NE induced higher fold AR-upregulation in the tail than in the head (Figure 2B, the 1st bars in the left vs. right panels). Note: the assessment of the mRNA-abundance of zKeap1a/b by qRT-PCR above (Figure 1D) represents global levels of allele expression in the two segments, but does not inform on levels in the GFP-positive cells used in the GFP-reporter assays. Thus, these data indicate that hypomorphism in NE-induced AR-upregulation in the tail is zKeap1a-dependent.

By contrast, in zKeap1b-knockdown fish, electrophile treatment elicited lower-fold AR-upregulation selectively in the tail (Figure 2B left panel). This effect could be rescued by injection of Keap1b mRNA, even though these procedures had little effect on the basal AR in this region (Figure 2—figure supplement 2). zKeap1b-knockdown elicited little or no effect on the head (Figure 2B right panel), compared to control fish. Note: in fish subjected to no knockdown, electrophile treatment resulted in little or no AR-upregulation (~1.1-fold) in the head, although the tail was responsive (~3.4-fold) (Figure 2B, 1st bars in both panels). Overall ~3-fold increase in the magnitude of AR following electrophile stimulation has been reported by us Parvez et al., 2016; Parvez et al., 2015; Lin et al., 2015; Long et al., 2018 and other laboratories Huang et al., 2012; Levonen et al., 2004; Chen et al., 2009; Zou et al., 2016; Kachadourian et al., 2011; Dev et al., 2015; Trott et al., 2008 in cultured cells and animals, using multiple orthogonal readouts including GFP- and luciferase-reporter assays and by western blot and qRT-PCR analyses assessing endogenous AR-driven genes. Electrophile treatment of zKeap1a/1b-double-knockdown fish also upregulated AR in both head and tail (Figure 2B, both panels), an observation that may be due to re-routing of electrophile response in these fish, or possibly hinting at a dominant-negative effect being at play. Notably, zebrafish Keap1 paralogs have been shown to form heterodimers post lysis and these paralogs can occur in the same tissues, indicating that particularly in some tissues, a heterodimer of zKeap1 paralogs could be formed Li et al., 2008. Overall, our data and previous reports Kobayashi et al., 2009 indicate that there is dominant-negative effect shown by one zKeap1-allele, but typical regulatory behavior shown by the other, at least under electrophile stimulation.

One possible alternative explanation for the observed differences in AR in the head vs. the tail, is that bolus electrophile dosing to stimulate AR is dominated by divergent off-target effects between the head and tail (for instance, due to varied levels of detoxifying enzymes Hayes and Dinkova-Kostova, 2014, or different expression levels of electrophile-sensors Parvez et al., 2018). We thus used fish transiently expressing Halo-hKeap1 to examine how Keap1 ‘on-target’ electrophile-modification affects AR signaling in the head and the tail.

Z-REX is built on proximity-directed delivery of an LDE in situ controlled by light (Figure 1A and Figure 1—figure supplement 1A). HaloTag–Ht-PreLDE complex serves as a latent source for rapid release of nascent LDE in an amount stoichiometric to Halo-POI Parvez et al., 2016; Long et al., 2017c. Because labeling occurs rapidly (t_1/2 < 1 min) Lin et al., 2015 and under what constitutes electrophile-limited (k_cat/K_m-type) conditions Liu et al., 2019, this scenario is in stark contrast to bolus LDE-administered from outside of animals, and does not significantly label proteins other than hKeap1 (see below for proof of this experimentally). This technique is not as subject to upstream factors extrinsic to hKeap1 (e.g. detoxification) as bolus dosing (Figure 1A), but is particularly effective at illuminating unexpected pathway intersections with Keap1-modification-specific changes in signaling Long et al., 2017b.

Immediately following mRNA-injection, Ht-PreLDE was introduced, and after 28.5 hr, excess Ht-PreLDE was removed by 30-min cycles of washing (x3) (see Appendix-Scheme 1). Our recent data establish that our Ht-PreHNE [LDE = 4-hydroxynonenal (HNE)] saturates Halo-binding site within Halo-POIs in fish embryos Long et al., 2017c. We found that Z-REX-targeted hKeap1-modification resulted in 1.5-fold AR-upregulation (~40% of what we observed with NE-bolus dosing) in the tail 34 hr-post-fertilization (hpf), following light-exposure 30 hpf (Figure 3A). The head was unresponsive.

Figure 3 with 5 supplements see all

Download asset Open asset

Figure 3—source data 1

Quantification results.

https://cdn.elifesciences.org/articles/83373/elife-83373-fig3-data1-v4.xlsx

Download elife-83373-fig3-data1-v4.xlsx

We next compared these data with results from fish bolus-treated with HNE. The conditions for whole-animal LDE exposure (25 µM) were chosen to: (1) closely mirror Z-REX conditions, and (2) avoid prolonged bolus LDE exposure that was deleterious to embryonic development (Figure 3—figure supplement 1), whereas exposure to photocaged precursors of similar or higher dose/longer duration was easily tolerated in these larval fish: Ht-PreHNE showed little effect on development, even when Z-REX was performed at 30 hpf and embryos were left to develop for another 3 days (Figure 3—figure supplement 1). Whole-fish treatment with HNE (25 µM) at 30 hpf, for 4 hr, elicited similar tail-specific AR-upregulation as was found under Z-REX (Figure 3A inset; the last two bars in left panel). Taken together, differences between head and tail response, and the fact that fold changes in AR are generally relatively small, collectively indicate that the measured response in the tail is indeed a biologically relevant change.

We next examined by qRT-PCR the extent of upregulation of endogenous AR-driven genes of conserved importance in higher eukaryotes, under both regimens, namely: global LDE-exposure and Z-REX-mediated Keap1-specific-modification. Representative genes associated with drug metabolism under control of Nrf2 (Gst-isoforms, Hmox1, and Abcb6) were activated to similar levels between Z-REX and bulk HNE-treatment, and AR modulation was most prominent in the tail (Figure 3B), particularly in the case of gsta.2, but also for abcd6a that only showed a significant upregulation relative to controls in the tail. gstpi1 showed hypomorphic responses in the head relative to the tail, but these were not significantly different, at this level of statistical power. We ascribe this modest AR-upregulation in the head [seen only by qRT-PCR analysis of several genes, and not by immunofluorescence(IF)-imaging of Tg(gstp1:GFP)] to increased sensitivity of qRT-PCR analysis compared to in vivo fluorescence-imaging, and the fact that the gstp1 locus (used in the GFP-reporter fish) is not the most responsive in the head. By contrast, bolus LDE-treatment yielded mixed responses in most cases (Figure 3B). We further note that mRNA extracted from head or tail covers cells that are not necessarily examined in Tg(gstp1:GFP), rendering data from head and tail in the qRT-PCR (e.g. Figure 3B) and IF-imaging (e.g. Figures 2 and 3A) assays not necessarily comparable.

To validate that AR-upregulation observed upon Z-REX is due to Keap1-specific HNEylation as opposed to off-target effects, we compared the extent of AR modulation in fish expressing either Halo-•-hKeap1, or the ‘non-fused’ Halo-P2A-•-hKeap1 (where ‘•’ designates TEV-protease-site that is cleaved post lysis). The P2A-sequence allows for slipping of the translation machinery Kim et al., 2011, thereby expressing Halo and hKeap1 proteins separately (i.e. non-fused), from the same mRNA, during the same translation step. Based on our previous cell-based studies Parvez et al., 2016; Long et al., 2017c; Parvez et al., 2015; Lin et al., 2015; Van Hall-Beauvais et al., 2018, the non-fused system wherein Halo and POI are expressed as two separate proteins serves as a robust negative control in our proximity-directed LDE-targeting platform (Figure 3C): when electrophile release is executed in the non-fused system, both labeling of POI and downstream ramifications are ablated. A similar level of ubiquitous expression of hKeap1 was achieved following injection of mRNA encoding the non-fused construct (Figure 3—figure supplement 2A). Similar to our previous cell-based data, the P2A-integrated construct yielded little or no AR-upregulation upon Z-REX, compared to that achieved with Halo-•-hKeap1, demonstrated by both IF-imaging of GFP-reporter upregulation (Figure 3D–E) and qRT-PCR analyses targeting 3 different endogenous AR-driven genes (Figure 3—figure supplement 2B). These data also show that minor upregulation of AR-specific genes due to treatment with light alone, or photocaged-compound alone, do not synergize significantly and hence do not contribute dramatically to responses observed during Z-REX. Importantly, upon bolus exposure to Tecfidera (Figure 1—figure supplement 1B), an approved pleiotropic electrophilic drug that stimulates AR as part of its pharmaceutical program Poganik et al., 2021; Poganik and Aye, 2020, both fused and non-fused systems gave twofold AR-upregulation, particularly in the tail region (Figure 3D–E).

Importantly, at the level of target labeling by Z-REX, we showed that the non-fused expression system strongly-diminished the extent of LDE-signal on hKeap1 compared to the fused construct (Figure 3—figure supplement 3A-B). We demonstrated this outcome using ‘biotin-Click-streptavidin pulldown’ of LDE-modified proteins in fish, subsequent to Z-REX in vivo. Briefly, lysates from either control fish or Z-REX-treated fish—expressing either Halo-•-hKeap1 or Halo-•-P2A- hKeap1—were all subjected to: (in the former instance) TeV-protease to separate Halo and Keap1; Click-coupling with biotin-azide; and streptavidin enrichment to evaluate LDE-modified proteins (see Appendix-Scheme 1).

Using the same enrichment protocol, we examined the extent of hKeap1 labeling subsequent to bolus HNE-treatment of embryos (Figure 3—figure supplement 4). We unexpectedly found little discernable hKeap1-modifcation. This result is surprising because bolus dosing promoted AR-upregulation to a level similar to that elicited by Z-REX targeted hKeap1-HNEylation (Figure 3A–B). This result underscores the importance of using Z-REX to parse on-target electrophile signaling, especially in organisms because phenotypic effects of on-target LDE-modification can be directly and unambiguously interrogated by Z-REX. However, because bolus dosing affects numerous proteins, and changes redox balance, such outcomes are overall not surprising. To further investigate this phenomenon, we compared how AR was induced as a function of time under bulk-exposure and Z-REX.

AR increased as the fish aged, such that the DMSO-treated group had significantly higher AR 48 hpf than 34 hpf (Figure 3—figure supplement 5A-B). Thus, even though Z-REX-mediated Keap1-specific LDEylation led to a 1.5–1.7-fold increase in AR 4 hr post light-exposure (ple) (34 hpf), there was no significant difference between Z-REX and controls 18 hr ple (48 hpf) (Figure 3—figure supplement 5C). To further examine the time-dependent AR-upregulation following Z-REX, we measured AR-upregulation at 34 hpf (Figure 3—figure supplement 5D). However, photouncaging was executed at different time points prior to harvesting (1.5, 4, and 8 h ple).

Interestingly, Z-REX gave a transient AR-upregulation, which peaked around 4 hr ple but returned to basal levels 8 hr ple (Figure 3—figure supplement 5D). This profile is consistent with the Z-REX model, wherein the transient release of LDE in an amount sub-stoichiometric to POI in vivo mimics signaling conditions Liu et al., 2019; Long et al., 2021, such that AR-upregulation is not sustained. Conversely, when we examined the effects of bolus HNE-treatment, a similar kinetic profile was observed (Figure 3—figure supplement 5E), although fish were constantly exposed to electrophile. It is likely that the off-target effects incurred during bolus-electrophile treatment cause insults that lead to severe negative effects on fish, which we showed above are not present in Z-REX (Figure 3—figure supplement 1). This finding makes precise comparisons between bolus dosing and Z-REX in terms of absolute outcomes difficult, and may explain the difficulty in correlating labeling of Keap1 with the extent of AR-upregulation under bolus dosing with that resulting from Z-REX. Regardless, the overall conclusions of the two methods in terms of head vs. tail AR are surprisingly similar.

The above results (primarily obtained with HNE, selected because it is the most well-known LDE) were intriguing, and we were keen to understand how Keap1 labeling contributed to overall AR for structurally-homologous native LDEs (Figure 1 inset). HDE (4-hydroxydodecenal, identified in human urine Florens et al., 2016 and heated oils Seppanen and Csallany, 2004), NE (nonenal, another endogenously-generated LDE exhibiting an age-dependent rise in production Haze et al., 2001), and DE (3-decen-2-one, an FDA-approved food additive with natural occurrence in certain fruits and mushrooms Knowles and Knowles, 2012) were chosen as representatives. As with HNE, the HaloTag-targetable photocaged precursors to HDE, NE, and DE have been successfully applied to target hKeap1 selectively in cells Parvez et al., 2016; Lin et al., 2015. The measured AR-upregulation in each case is of comparable efficiency to that obtained under whole-cell LDE flooding Lin et al., 2015. Using Tg(gstp1:GFP) fish, we found that these LDEs elicited similar AR-upregulation in fish tails upon Z-REX (Figure 4A), but the head was not responsive (Figure 4B).

Figure 4 with 1 supplement see all

Download asset Open asset

Figure 4—source data 1

Quantification results.

https://cdn.elifesciences.org/articles/83373/elife-83373-fig4-data1-v4.xlsx

Download elife-83373-fig4-data1-v4.xlsx

Independent qRT-PCR analysis of endogenous AR-driven genes following Z-REX in casper fish gave broadly-consistent results across all the LDEs examined (Figure 4C). Using Click-biotin-pulldown following Z-REX in vivo, we validated using DE as an example that hKeap1-labeling extent was similar to that achieved with HNE (Figure 4—figure supplement 1). Given that AR under Z-REX conditions stems only from hKeap1-modification, these findings explain the similar magnitude of AR-outputs observed with Z-REX. These outcomes were further substantiated by qRT-PCR analysis of downstream genes discussed above (Figure 3—figure supplement 2B), where we showed that Halo-P2A-•-hKeap1 was hypomorphic for Z-REX-assisted hKeap1-HNEylation-promoted AR-upregulation.

Having established that each LDE when targeted specifically to Keap1 could trigger tail-specific AR-upregulation, we compared effects on AR from whole-fish-LDE-exposure under otherwise identical conditions/timescales to Z-REX. Bolus dosing with these LDEs elicited widely different AR that was chemical-structure dependent. By both imaging of the AR-reporter fish (Figure 5A–B) and qRT-PCR analysis of endogenous AR-driven genes (Figure 5C), LDEs that did not contain a 4-hydroxyl group were best at eliciting AR. Interestingly, 4-dehydroxy species are intrinsically less electrophilic than their 4-hydroxylated counterparts. This result was markedly different from those seen under Z-REX targeted conditions for the same LDEs.

Figure 5 with 1 supplement see all

Download asset Open asset

Figure 5—source data 1

Quantification results.

https://cdn.elifesciences.org/articles/83373/elife-83373-fig5-data1-v4.xlsx

Download elife-83373-fig5-data1-v4.xlsx

We also assayed the effects of electrophiles of clinical relevance under bulk-exposure. Tecfidera gave outputs between those elicited by bolus HNE and NE (and of higher magnitude than the Z-REX responses) (Figure 5A–B; Cf. Figure 3D). Bardoxolone methyl (CDDO-Me; Phase II trials recently completed for pulmonary hypertension, Figure 1—figure supplement 1B) was severely toxic to the fish at this concentration, although AR was upregulated (Figure 5—figure supplement 1A). Sulforaphane (Figure 1—figure supplement 1B) did not elicit AR under these conditions (Figure 5A–B). However, regardless of the magnitude of the output in the tail, where responses were observed, there remained minimal, and indeed non-statistically-significant, increase in AR in the head (Figure 5B).

We next examined the ability of the LDEs to label the fish proteome following whole-fish treatment (Figure 6A–B, Figure 5—figure supplement 1B). Interestingly, this trend closely mirrored that of AR-upregulation elicited upon bolus dosing, with hydroxyl-bearing LDEs (HNE and HDE) manifesting a reduced level of proteome labeling and corresponding reduced AR-induction (Figure 5 vs. Figure 6A-B, Figure 5—figure supplement 1B). Thus, AR-upregulation upon bulk LDE-exposure is dominated by permeation/pharmacokinetic-effects over inherent ligand electrophilicity. However, the divergent AR-induction in the front and hind portions of the fish was retained.

Figure 6

Download asset Open asset

We thus predicted that zKeap1a is a negative regulator of electrophile-stimulated AR, and zKeap1b is the principal means through which electrophile-stimulated AR is mounted in zebrafish. Based on this hypothesis, we examined the phylogeny between the two zebrafish genes and the human protein, for clues as to which residue could lead to the proposed unusual behavior of zKeap1a. We reasoned that signaling could be affected through mutation of a cysteine in the human protein (sensing competent) to a bulky residue in Keap1a (mimicking the electrophile-modified state Poganik et al., 2018). In Keap1b, this residue would remain a cysteine. Based on this logic (Figure 7—figure supplement 1A), we identified C273 (human numbering); the analogous residue in zKeap1a is an isoleucine, but it remains a cysteine in zKeap1b (Figure 7A and Figure 7—figure supplement 1B). We thus cloned hKeap1(C273I) in a bid to find a humanized form of zKeap1a. This mutation has already been identified as a loss-of-function mutation to hKeap1 (Levonen et al., 2004; Saito et al., 2016), although it has not been extensively characterized, particularly in terms of how it interacts with wild-type(wt)-hKeap1. We thus progressed to evaluate hKeap1(C273I) more deeply. We also amplified zKeap1a and zKeap1b from cDNA and cloned these genes into mammalian expression vectors with a HaloTag.

Figure 7 with 2 supplements see all

Download asset Open asset

Figure 7—source data 1

Quantification results.

https://cdn.elifesciences.org/articles/83373/elife-83373-fig7-data1-v4.xlsx

Download elife-83373-fig7-data1-v4.xlsx

Behavior ascribed to zKeap1a could be due to an inability to sense LDE that leads to an overall muted AR, similar to what has been previously postulated Kobayashi et al., 2009. We thus first examined the sensing abilities of the four Keap1-variants to NE and HNE using T-REX in HEK293T cells ectopically expressing respective Halo-tagged proteins. All four constructs expressed similarly in HEK293T cells and at levels far above that of the endogenous Keap1-protein, allowing us to discard the influence of the endogenous hKeap1. hKeap1 was expressed at marginally higher levels than some of the other proteins (Figure 7—figure supplement 2A). Intriguingly, all four proteins sensed HNE equally well (Figure 7B and Figure 7—figure supplement 2B-C). hKeap1 and hKeap1(C273I) sensed NE marginally better than zKeap1a/b, both of which sensed NE to the same extent (Figure 7B). Thus, differences in sensing abilities of the different proteins cannot explain the data we, and indeed others, have observed with zKeap1a and zKeap1b.

There remains the possibility that the different zKeap1-paralogs have different modes of marshalling AR in response to electrophiles. We thus examined these behaviors, again using cell culture. When the drop in AR was measured following transfection of equal amounts of the different zKeap1 plasmids (which, based on the experiments above, gave roughly similar amounts of protein), we found that all zKeap1a/b and hKeap1 could all significantly suppress AR, again consistent with previous reports Li et al., 2008. zKeap1b appeared to be less efficient at suppressing basal AR than zKeap1a (Figure 7—figure supplement 2D), which does not completely agree with previous reports, although the previous reports are limited both in terms of statistical power, and direct comparability as they were carried out in early embryos where signal-to-noise can be low. zKeap1a was similar at suppressing AR relative to hKeap1. hKeap1(C273I) suppressed basal AR, but marginally (Figure 7C). Critically, none of these conditions were able to suppress basal AR to undetectable levels (Figure 7—figure supplement 2D). Intriguingly, when zKeap1b was co-transfected with sub-saturating amounts of zKeap1a, no decrease in basal AR was observed relative to zKeap1a alone (Figure 7D), implying that zKeap1a somehow affects the ability of zKeap1b to suppress AR. When we performed similar experiments mixing hKeap1 and hKeap1(C273I), a similar effect was observed (Figure 7D).

We progressed to examine response of the different Keap1-variants upon electrophile modification. For these purposes, we used NE as a bolus electrophile. We selected NE because in fish NE was the most permeable molecule giving the most robust AR and labeling (Figure 5A–C, Figure 6A, Figure 5Figure 1B). Intriguingly, although hKEAP1 and zKeap1b were significantly responsive to bolus NE treatment, significantly upregulating AR, zKeap1a, and hKeap1(C273I) were much less responsive (Figure 7E). In terms of AR-upregulation ability, these data fully recapitulate data from zebrafish embryos expressing zKeap1a and zKeap1b Kobayashi et al., 2009. These data further imply that the C273I mutation is a likely cause of the zKeap1a’s inability to mount AR. Furthermore, as the ability to suppress basal levels of AR were significantly different between zKeap1a and hKeap1(C273I) (Figure 7C), this suppressive effect is independent of fold basal-AR suppression. However, this finding does not fully explain what we found from the initial morphant data (Figure 2B); that is, why knockdown of zKeap1a could lead to hyper-elevated AR in zebrafish following bolus electrophile treatment. Such a mechanism could be explained if zKeap1a were a dominant-negative regulator of AR in the electrophile-stimulated states.

We next transfected equal amounts of zKeap1a/b plasmids, or 2:1 zKeap1a:zKeap1b and compared AR-levels upon bolus electrophile treatment to those found in cells transfected with zKeap1a- or zKeap1b-alone. Consistent with a dominant-negative effect on AR induction, NE caused zKeap1b-alone-expressing cells to induce AR significantly, whereas zKeap1a-alone or zKeap1a/Keap1b-expressing cells mounted AR significantly less efficiently upon NE treatment (Figure 7F). A similar effect was observed when hKeap1 and hKeap1(C273I) were mixed. These data explain why morpholinos targeting zKeap1a led to upregulation of electrophile-induced AR in both tail and especially the head and provide evidence for dominant-negative effects during Keap1/AR signaling.

To examine how these paralog-divergent electrophile responses arose, we next examined how NE treatment affected the amount of Nrf2 bound to zKeap1a and zKeap1b, or the heterozygotic state. For these experiments we used HEK293T cells transiently expressing Nrf2, co-transfected with empty vector, zKeap1a, zKeap1b, or equal amounts zKeap1a and zKeap1b. We found that zKeap1b accumulated Nrf2 in the basal (i.e. non-electrophile-stimulated) state, whereas relative to zKeap1b, zKeap1a accumulated less Nrf2, and zKeap1a/zKeap1b accumulated an amount of Nrf2 that was significantly more than Keap1a alone and less than zKeap1b alone (Figure 8—figure supplement 1). When treated with NE, zKeap1b-bound Nrf2 was diminished by approximately 40% (Figure 8A–B). Conversely for zKeap1a and zKeap1a/zKeap1b no release of Nrf2 was observed. These data agree with the paralog-specific regulation of electrophile signaling observed above, which showed that zKeap1a and the heterozygotic state were unable to upregulate AR effectively following electrophile treatment. Furthermore, given that significant amount of Nrf2 is bound to zKeap1b in both the heterozygous- and zKeap1b-only-expressing states, these data are consistent with a model in which electrophile engagement of zKeap1b can trigger AR through causing net release of Nrf2, whereas zKeap1a (either with or without zKeap1b) cannot lead to net release of Nrf2 upon electrophile labeling, and hence cannot upregulate AR. Thus, zKeap1a (either electrophile modified or unmodified) acts to suppress loss of Nrf2 binding to zKeap1b.

Figure 8 with 1 supplement see all

Download asset Open asset

Figure 8—source data 1

Full view blot image.

https://cdn.elifesciences.org/articles/83373/elife-83373-fig8-data1-v4.zip

Download elife-83373-fig8-data1-v4.zip

Figure 8—source data 2

Raw blot image.

https://cdn.elifesciences.org/articles/83373/elife-83373-fig8-data2-v4.zip

Download elife-83373-fig8-data2-v4.zip

Figure 8—source data 3

Quantification results.

https://cdn.elifesciences.org/articles/83373/elife-83373-fig8-data3-v4.xlsx

Download elife-83373-fig8-data3-v4.xlsx

Wherever applicable, figure legends contain information pertaining to SEM with associated P values, sample size (e.g. number of fish embryos analyzed, number of independent biological replicates). Representative raw images are included with accompanying quantitation where relevant. Figure legends contain description of independent biological replicates, vs. technical replicates. Outliers were maintained in all data sets with error bars designating SEM and p-values from application of two-tailed Students’ t-test included. Figure legends contain information pertaining to the following: the exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement; whether measurements were taken from distinct samples or whether the same sample was measured repeatedly; null hypothesis testing (two-tailed Students’ t-test) with exact p values noted whenever suitable. No statistical methods were used to pre-determine sample size. Size of datasets was chosen according to literature and based on our own experience, integrating similar methods of analysis. Number of technical replicates and biological replicates are reported in figure legends. Summary information related to sample allocation/handling is detailed in the supplementary text.Briefly, no masking was used. Prior to beginning each experiment, cells/embryos were allocated into groups randomly, for each sample group. When an experiment was commenced, groups of cells/embryos were allocated into treatment groups without pattern or bias. This ensured that each treatment group in an experiment was identical to account for any variation across cells/fish breeding. Cell counting was performed at each step of the experiment whenever relevant, to rigorously standardize conditions both within each experiment and across different experiments. Casper strain zebrafish, wild-type zebrafish, and previously-validated reporter strains were used for the experiments involving embryos and for the latter transgenic reporter strains, at consistent zygosity. As zebrafish are believed to exhibit polygenic sex determination, at the age at which experiments were undertaken, the sex of the fish was unable to be determined, but likely account for 50:50 male: female.

Plasmids to express (His₆)-Halo-TEV-Keap1-(2xHA) (hereafter, Halo-•-Keap1) and (His₆)-Halo-(2xHA)-P2A-TEV-Keap1-(2xHA) (hereafter, Halo-P2A-•-Keap1) in pCS2 +8 vectors were cloned using ligase-independent cloning method using primers specified in Appendix. Briefly, Halo-TEV-Keap1 was amplified from pFN21a Halo-TEV-Keap1 (Halo ORFeome, Promega Kazusa collection) using fwd1 and rev1 primers. The amplified product was then extended using fwd ext1 and rev ext1. An additional extension step was performed using the product from the first extension step as the template and using primers fwd ext2 and rev ext2 to generate the ‘megaprimer’. The ‘megaprimer’ was inserted in empty pCS2 +8 vector linearized using EcoR1.

To generate Halo-P2A-•-Keap1, Halo and Tev-Keap1 were amplified separately from pFN21a Halo-TEV-Keap1 using primers fwd1 and rev1, and fwd1’ and rev1’, respectively. The amplified products were then extended using fwd ext1 and rev ext1, and fwd ext1’ and rev ext1’. The two fragments were joined using PCR and amplified using another set of primers, fwd ext2 and rev ext2, to generate the ‘megaprimer’. The ‘megaprimer’ was inserted in empty pCS2 +8 vector linearized using EcoR1.

To generate zKeap1a and zKeap1b in mammalian expression vectors, RNA from wild-type zebrafish was isolated and used to generate cDNA using an oligo(dT) primer. From the cDNA, primers designed to overlap with the zKeap1a/b sequences were used to amplify zKeap1a and zKeap1b. The primers also contained an overlapping region with the pCS2 +8 backbone. The empty pCS2 +8 plasmid was cut with EcoRI-HF enzyme and the zKeap1a/b PCR product with overlapping pCS2 +8 backbone was inserted into the vector. To obtain HaloTagged zKeap1a/b, the zKeap1a/b plasmids were cut with SpeI and a ‘megaprimer’ PCR sequence encoding Halo-TEV-Flag (3 x) was inserted to form the full construct: Halo-TEV-Flag3-zKeap1a/b, referred to as Halo-•-(3xF)-zKeap1a/b. The amino acid sequence for zKeap1a and zKeap1b from our fish matched the sequences reported in uniprot. Note: additional synonymous mutations were introduced in pCS2 +8 Halo-•–3xFlag-zKeap1a to prevent knockdown interference in the course of rescue experiments where embryos were injected with both Keap1a-ATG-MO MO and Halo-•–3xFlag-zKeap1a mRNA. The stop codon was added separately by site-directed mutagenesis. All primers used are listed in Appendix.

Successful inserts for each plasmid were identified with colony PCR. ‘Hit’ colonies were grown overnight in LB-AMP medium and purified using a miniprep kit (Bio Basic, BS614). The insert was fully sequenced at the Cornell Biotechnology Resource Center Genomics facility and Microsynth AG (Switzerland).

All procedures performed at Cornell (2017–2018) and EPFL (2018-present) conform to the animal care, maintenance, and experimentation procedures followed by Cornell University’s and EPFL’s Institutional Animal Care and Use Committee (IACUC) guidelines and approved by the respective institutional committees. All experiments with zebrafish performed at EPFL (2018-present) have been performed in accordance with the Swiss regulations on Animal Experimentation (Animal Welfare Act SR 455 and Animal Welfare Ordinance SR 455.1), in the EPFL zebrafish unit, cantonal veterinary authorization VD-H23. Either Casper strain or Tg(–3.5gstp1:GFP)/it416b reporter [hereafter Tg(gstp1:GFP)] Danio rerio (zebrafish) were used for all experiments. Casper strain was a kind gift from the Fetcho Lab (Cornell University). Tg(gstp1:GFP) fish was from the Japanese National BioResource Project. Since zebrafish are believed to exhibit polygenic sex determination, at the age at which experiments were undertaken, sex of the fish was unable to be determined, although our samples are likely roughly equal mixtures of males and females. Animals were maintained and embryos were obtained according to standard fish husbandry procedures. For Tg(gstp1:GFP) fish, crosses were set between a homozygous transgenic parent and a WT parent such that the resulting progeny were all heterozygous for the reporter gene.

For injection in fish embryos, mRNA for Halo-•-Keap1 and Halo-P2A-•-Keap1 was generated. First, the desired genes were amplified using RNA-fwd and RNA-rev primers (Appendix). mRNA was generated using an mMessage mMachine SP6 in vitro transcription kit (Ambion, AM1340) as per manufacturer’s protocol except the reaction was scaled up for two preps.

Fertilized eggs at the one- to four-cell stage were injected with 2 nl mRNA (1.3–1.6 mg/ml) and/or morpholino (2.7 mg/mL, approximately 5 ng of MO per fish) into the yolk sac. Immediately after injection, embryos were pooled, and separated into two petri dishes (10 cm) filled with 30 mL 10% Hank’s salt solution with methylene blue and penicillin (100 U/ml) / streptomycin (100 µg/ml). To one set was added the photocaged precursor to designated LDE at a final concentration of 6 µM and to the other equal volume of DMSO in the dark. Embryos were maintained at 28 °C in the dark for 28 hr after which time fish larva were washed in 10% Hank’s solution with no methylene blue/antibiotic (3 times for 30 min each). Larvae were moved to six-well plates. Half of the larvae (Ht-PreLDE-treated or -untreated) were exposed to light for 4 min, and the other half of each set was not. For bolus dosing, treatment of larvae with LDEs was staggered such that the harvest time was the same for all samples including larvae that underwent Z-REX (34 hpf). Further experiments were performed as illustrated in Appendix-Scheme 1 and described elsewhere in the supplementary methods, and main and supporting figure legends.

All qPCR experiments were performed in casper strain. 2 hr post light illumination or bolus LDE treatment, 12–15 larvae per sample were euthanized by chilling, dechorionated, the head and tail separated using sharp forceps (11252–40 Dumont #5 Forceps - Biologie/Titanium), and transferred to separate Eppendorf tubes. The samples were washed twice with ice-cold PBS and homogenized in 1 mL Trizol (ThermoFisher Scientific, 15596018) together with vortexing with glass beads for 2 min. Total RNA was extracted per manufacturer’s protocol. Glycoblue (ThermoFisher Scientific, AM9516) was used for visualization of the RNA pellet. Around 600 ng of total RNA was treated with AMP grade DNaseI (ThermoFisher Scientific, 18068015), reverse transcribed using Superscript III reverse transcriptase (ThermoFisher Scientific, 18080085) per manufacturer’s instruction. qRT-PCR was performed for the indicated genes using primers specified in Appendix. All primers were validated as previously reported. qRT-PCR analysis was performed with iQ SYBR Green Supermix (Bio-Rad, 170–8880) on a Light Cycler 480 instrument (Roche). In a total volume of 10 µL the PCR reaction mix contained, in final concentrations, 1 X iQ SYBR Green Supermix, 0.30 µM each of the forward and reverse primers and 10–13 ng of template cDNA. The qPCR program was set for 3 min at 95 °C followed by 40-repeat cycles comprising heating at 95 °C for 10 s and at 55 °C for 10 s. The expected products were of ~100–130 bp in size. The data was analyzed using ΔΔC_t method and presented relative to zebrafish actin, β2.

To assess AR upregulation in Tg(gstp1:GFP) fish, larvae 4 hr post light illumination or bolus LDE treatment were dechorionated, washed twice in ice-cold PBS and fixed in 4% paraformaldehyde in 1 X PBS (Gibco 14190169) for at least overnight with gentle rocking at 4 °C. Fixed larvae were permeabilized with chilled methanol at –20 °C for 4 hr–overnight. Fish were then washed 2 times with PBS-0.1%Tween-1% DMSO for 30 min each with gentle rocking, then blocked in PBS-0.1%Tween containing 2% BSA and 10% FBS, then stained with anti-GFP FITC conjugated (Abcam, ab6662) primary antibody overnight at 4 °C in blocking buffer. Subsequently, the larvae were washed twice (30 min each wash), re-blocked for 1 hr at room temperature, and incubated with the AlexaFluor 568-conjugated fluorescent secondary antibodies (Abcam, ab175707) in blocking buffer for 1.5 hr at room temperature with gentle rocking, and then washed three times. Fish were imaged on 2% agarose plates on a Leica M205-FA equipped with a stereomicroscope. Quantitation of IF data was performed using ImageJ/FIJI (NIH).

To assess protein expression in zebrafish, larvae were fixed at 34 hours post fertilization (hpf) after dechorionation. Permeabilization and immunostaining protocols are as above except antibodies to the desired protein/tag and appropriate secondary antibodies were used.

Casper zebrafish expressing Halo-•-Keap1, and either treated with the photocaged precursor (6 μM, overnight) or bolus treatment (2 hr) with the indicated LDE, were dechorionated at 34 hpf and fixed in 4% PFA for at least overnight with gentle rocking at 4 °C. Where relevant, PF-4981517 (1 or 5 µM) was added 4 hr prior to LDE treatment. Fish were then permeated in methanol (100%) at −20 °C for at least 24 hr. Fish were then washed twice in PBS and two times in Hepes (50 mM, pH 7.6) for 30 min each (Note: Fish tend to float after MeOH wash. Allow them to settle prior to manipulation/washing). Fish were then exposed to a cocktail containing (in final concentration): 50 mM Hepes (pH 7.6), t-BuOH (5%), CuSO₄ (1.1 mM), sodium ascorbate (10 mM; made as a 100 mM stock in 500 mM Hepes (pH 7.6) with no further pH adjustment), and Cy5-azide (10 µM). This was shaken at room temperature for 1 h, then washed three times in 1 x PBS with 0.015% Tween-20. After third wash, fish were incubated overnight at 4 °C, then fresh 1 x PBS with 0.015% Tween-20 was added and fish were imaged. Fixed fish can stick to plastic. Tween helps to reduce this problem.

Casper zebrafish expressing Halo-•-Keap1 were treated with either the photocaged precursor or bolus treatment with the indicated LDEs (~120 per condition). Photocaged precursors were added to the fish water after injection of Halo-•-Keap1 mRNA at a final concentration of 6 μM and Z-REX was performed as specified above. Bolus dosing was done for 2 hr. Immediately after Z-REX, larvae were dechorionated and deyolked manually at 4 °C, washed twice with cold PBS to remove yolk proteins, and washed once with cold 50 mM Hepes (pH 7.6). The zebrafish pellet was flash frozen in liquid nitrogen and stored at −80 °C until lysis. Fish pellet was resuspended in 50 mM Hepes (pH 7.6), 1% Triton X-100, 0.1 mg/ml soybean trypsin inhibitor, and 2 X Roche protease inhibitor. Lysis was performed by vortexing with Zirconia beads for 20 s and subsequent 3 times freeze-thaw. Lysate protein was collected after centrifugation at 21,000×g for 10 min, and concentration determined using Bradford dye relative to BSA standard. 30–50 μg of the lysate protein was removed, quenched with Laemmli buffer and saved as input. The remaining lysate was diluted to 1 mg/ml with 50 mM Hepes (pH 7.6) and 0.2 mM TCEP, TEV protease was added at a final concentration of 0.2 mg/ml, and the sample incubated at 37 °C for 30 min. Next, 5% t-BuOH was added to the sample. A stock solution containing 10% SDS, 10 mM CuSO₄, 1 mM Cu-TBTA, 1 mM biotin-azide and 20 mM TCEP (made as a 100 mM stock in 500 mM HEPES pH 7.5) was prepared and added to the sample such that the final concentration are as follows: 1% SDS, 1 mM CuSO₄, 0.1 mM Cu-TBTA, 0.1 mM biotin-azide, and 2 mM TCEP. The mixture was mixed thoroughly and incubated at 37 °C for 15 min, after which another 1 mM TCEP was added, mixed, and the sample incubated for additional 15 min. Protein precipitation was performed by adding EtOH (prechilled at −20 °C) at a final concentration of 75% (v/v), vortexing the sample, and incubating at −80 °C for at least overnight. Precipitated protein was collected by centrifugation at 21,000×g at 4 °C for 2 hr, washed twice with prechilled EtOH (twice), once with 75% EtOH in water, and an additional wash with prechilled acetone. Precipitate was air-dried and subsequently redissolved in 8% LDS in 50 mM HEPES (pH 7.6), 1 mM EDTA by sonication at 50 °C and vortexing. The solubilized lysate protein was collected following centrifugation and diluted in 50 mM Hepes (pH 7.6) to give a final concentration of 0.5% LDS. The sample was added to pre-washed streptavidin high-capacity resin and incubated at 4–6 hr at rt. The supernatant was removed following a low-speed centrifugation (1000×g), and the beads washed thrice with 50 mM Hepes (pH 7.6) containing 0.5% LDS. Bound proteins were eluted by boiling beads in 2 x Laemmli buffer with 6% βME at 98 °C. Samples were analyzed using SDS-PAGE followed by western blot as specified below.

Up to 30 μl of input or elution samples were separated on a 10% polyacrylamide gel using SDS-PAGE. The gel was subsequently transferred to a PVDF membrane at 4 °C in ice-cold transfer buffer containing 25 mM Tris, 192 mM Glycine, and 15% Methanol (v/v). Membrane was blocked in 5% milk for 2 hr at rt, incubated with primary antibody in 1% milk for 5 hr at rt, washed three times with Tris Buffer Saline (100 mM Tris, pH 7.6, 150 mM NaCl) containing 0.2% Tween-20 (TBST). Where applicable, the membrane was incubated with secondary antibody in 1% milk for 5 hr at rt, washed twice with TBST, followed by an additional wash with TBS. Pierce ECL western blotting substrate was used for detection of the desired protein bands.

Imaging data was quantitated using ImageJ (NIH). For assessing AR upregulation in Tg(gstp1:GFP) fish, the area around the head (excluding the eyes), the tail (median fin fold), or the whole fish (excluding the yolk sac) were selected using freeform selection tool. Corresponding illustrations are included in each sub-figure for clarity. For IF, the mean red fluorescence intensity of the selected region was measured and subtracted from the mean background fluorescence intensity (region with no fish). For live fish imaging, the mean GFP fluorescence intensity of the selected region was measured and subtracted from the mean background fluorescence intensity (non-transgenic fish).

The mean value for the control group was calculated from the raw, background-subtracted, values within that control group. Then all raw values were divided by the mean for the control. n for imaging experiments represent the number of single cells or fish embryos quantified from at least 7–8 fields of view with controls (empty vector controls for ectopically-overexpressed proteins, shRNA knockdown cell controls for endogenous proteins) shown in the figures.

Unless specified, all t tests were two-tailed analysis. n for western blot/gels, qRT-PCR, and luciferase assays represents the number of lanes on western blots/gels under identical experimental conditions and each lane is from a separate individual replicate, no. of independent biological replicates as indicated in the figure legends.

HEK293T cells (obtained from ATCC) were cultured in complete 10% FBS medium (MEM Glutamax, 1 X sodium pyruvate, 1 X penicillin streptomycin, 1 X MEM NEAA, 10% FBS). Cells were grown at 37 °C with 5% CO₂ in a humidified incubator. Cell lines were verified to be free of mycoplasma contamination by Venor GeM Mycoplasma Detection Kit from Sigma.

HEK293T cells were seeded in 48 (or 96) well plates (cell density was 0.25×10⁶ cells/mL, 0.3 mL (or 0.1 mL) cells per well). Cells were transfected with 100 (or 33) ng pCDNA3 myc-Nrf2, 100 (or 33) ng luciferase plasmid mix (20:1 Firefly luciferase ARE promoter:Renilla CMV promoter), and 100 (or 33) ng total of either empty pCS2 +8, WT Keap1 (hKeap1, zKeap1a, or zKeap1b), or hKeap1 C273I. Transfection was carried out using Mirus TransIT-2020 transfection reagent. After 24 hr of transfection, the medium was removed and replaced with rinse medium (MEM glutamax, 1 X sodium pyruvate, 1 X penicillin streptomycin, 1 X MEM NEAA) containing 25 μM NE or HNE or corresponding volume of DMSO. Cells were incubated for a further 18 hr, at which point they were lysed in 65 (or 30) μL passive lysis buffer by shaking at room temperature for 25 min. 25 μL of sample was transferred to a 96-well opaque white plate for reading, and an additional 30 μL of sample was mixed with 10 μL of 4 X Laemmli dye with BME to run on a gel to confirm expression levels of the variants of Keap1. A BioTek Cytation3 or a Perkin Elmer 2300 microplate reader was used to perform Firefly and Renilla luciferase activity readings as previously reported Zhao et al., 2018; Surya et al., 2018.

Procedure was followed as above, except T-REX was performed as previously published after 24 hr of transfection and the cells were lysed 18 hr post light Zhao et al., 2018; Surya et al., 2018. Briefly, after 24 hr transfection, 10 μM Ht-PreHNE was added to the cells in rinse medium (or corresponding amount of DMSO). After 2 hr incubation in the dark, the medium was removed, and the cells rinsed three times with rinse medium, half an hour for each rinse. After the last rinse, the cells were exposed to UV light (~3 mW/cm²) for 5 min and returned to the incubator for 18 hr before being lysed and read as described above.

HEK293T cells were transfected to express Halo-TEV-hKeap1, Halo-TEV-hKeap1 (C273I), or Halo-TEV-zKeap1a/b. After 36 hr of transfection, 25 µM of PreHNE or PreNE was introduced. After 2 hr of incubation in the dark, the cells were rinsed three times, 30 min each. The cells were exposed to light (~5 mW/cm²) for 10 min, incubated for a further 5 min, and then harvested. Click reactions with Cy5-azide were performed as previously published Zhao et al., 2018; Long et al., 2019.

Plasmids were generated that contained the binding site of the zKeap1a or zKeap1b ATG morpholino (the start codon and approximately 20 following bases) upstream of the gene encoding Firefly Luciferase (with a Gly-Ser linker). mRNA was generated using mMessage mMachine SP6 in vitro transcription kit. To test the translation blocking ability of the MOs, 50 ng of mRNA (0.136 µM) was incubated with random control MO, standard control MO, water, or corresponding zKeap1 MO in indicated ratios for 5 min at room temperature. Then, each sample was added to Rabbit Reticulocyte Lysate (Promega) and incubated at 37 °C for 15 min. The sample was loaded onto a white plate (Corning 3912) and the Firefly Luciferase quantified using standard methods. A decrease in Firefly signal indicated blocked translation.

The yolk sac of Tg(gstp1:GFP) embryos at the one- to four-cell stage was microinjected with approximately 2 nL of splice-blocking (SPL) morpholino oligonucleotides [0.5 mM (GeneTools LLC; sequences in Appendix)]. Embryos were grown at 28.5 °C and euthanized at 30 hpf. Euthanized embryos were lysed by vortexing for approximately 30 s in Trizol (Life Technologies) with zirconia beads, and RNA was isolated following the manufacturer’s protocol. The quality of the RNA was assessed by Nanodrop spectrophotometry (A260/A280 ratio ~2) and integrity was assessed by agarose gel electrophoresis. 1 μg of total RNA was treated with amplification-grade DNase I (NEB) and reverse transcribed using Oligo(dT)₂₀ as a primer and Superscript III Reverse Transcriptase (Life Technologies) following the manufacturer’s protocols. The resulting cDNA was used as a template in PCR reactions with primers flanking the sites targeted by the splice-blocking MOs (Appendix). PCR products were resolved on an agarose gel and band intensity was quantitated using the Measure tool of Image-J(NIH).

The yolk sac of Tg(gstp1:GFP) embryos at the one- to four-cell stage was microinjected with approximately 2 nL of morpholino oligonucleotides, namely SPL-MOs targeting zKeap1a and zKeap1b [0.5 mM (GeneTools LLC; sequences in Appendix)]. Embryos were grown at 28.5 °C and euthanized at 30 hpf. Whole-mount IF was carried out as described above, using anti-Keap1.

Either Approach (1) or (2) below, was deployed as indicated in figure legends and text discussion: (1) The yolk sac of Tg(gstp1:GFP) embryos at the one- to four-cell stage was co-microinjected with 2 nL of 0.2 mM zKeap1a-ATG-MO and 250 ng/μL Halo-TEV-zKeap1a mRNA. Embryos were grown at 28.5 °C and treated with 10 μM NE alkyne at 4 hpf for 4 hr before being imaged. (2) The yolk sac of Tg(gstp1:GFP) embryos at the one- to four-cell stage was co-microinjected with 2 nL of 0.25 mM zKeap1b-ATG-MO and 250 ng/μL Halo-TEV-zKeap1b mRNA. Embryos were grown at 28.5 °C and treated with 20 μM NE alkyne at 30 hpf for 4 hr before image acquisition. Whole-mount IF was carried out as described above, using anti-GFP FITC-conjugated (Abcam, ab6662) primary antibody and AlexaFluor 568-conjugated fluorescent secondary antibody (Abcam, ab175707). NOTE: the bolus LDE dosage and MO concentrations deployed were reduced in Approach (1), since the blastula period (4 hpf) was found to be more sensitivite to high-LDE/MO-dosage-induced viability loss.

HEK293T cells (~5–6×10⁶) were seeded in a 10 cm adherent tissue culture plate. After the cells reached 70–80% confluence (~18–24 h), the media were replaced with fresh complete media (8 mL). Cells were transfected with 7.5 µg (total amount) of indicated plasmid(s) using TransIT-2020 transfection reagent (per the manufacturer’s recommendation, Mirus). Following 24–36 hr incubation period, the media were changed, and the cells were treated with fresh media containing either 25 µM NE or DMSO, and incubated for a further 18 h. Cells were harvested, pooled, washed twice with ice-chilled 1 X DPBS, and flash-frozen in liquid nitrogen. Cell lysis was performed by first resuspending the cell pellets in 100–200 µL (per 1.5×10⁶ cells) of lysis buffer [containing in final concentrations, 50 mM Hepes (pH 7.6), 150 mM NaCl, 1% Nonidet P-40 and 1 X Roche cOmplete, mini, EDTA-free protease inhibitor cocktail], then by subjecting the resulting cell suspension to rapid freeze-thaw cycles (x3). The lysate was clarified by centrifugation at 18,000 x g for 10 min at 4 °C. Total protein concentration was determined using Bradford assay using BSA as standard (triplicate measurements were made and average value was taken). The lysate was subsequently diluted to 2 mg/mL with binding buffer containing in final concentrations, 50 mM Hepes (pH 7.6), 150 mM NaCl, 1 X Roche cOmplete, mini, EDTA-free protease inhibitor cocktail, and 0.1% Tween-20. This diluted lysate was subjected to either 50–100 µL bed volume of Anti-Flag M2 affinity agarose gel (A2220, Sigma) that had been pre-equilibrated with the binding buffer above. The sample was incubated with beads overnight at 4 °C by end-over-end rotation, after which time the supernatant was removed post-centrifugation at 500 x g. The beads were washed three times at 4 °C with 500 µL wash buffer containing in final concentrations, 50 mM Hepes (pH 7.6), 150 mM NaCl, 1 X Roche cOmplete, mini, EDTA-free protease inhibitor cocktail, and 0.1% Tween-20, using end-over-end rotation over 10 min during each wash. The bound protein was eluted by incubating with 0.15 mg/mL 3 X Flag peptide for 2 h at 4 °C. The sample was subjected to SDS-PAGE and transferred to a PVDF membrane for western blot analysis using antibodies indicated in corresponding figure legends.

All plasmids generated by the authors are provided in Appendix. These plasmids are being deposited to Addgene upon manuscript publication. In the interim, they are available from the corresponding primary contact upon request, in the same way as for small-molecule reagents used in REX technologies. All other chemical and biological materials used are commercially available and their sources are indicated in Materials & Methods (Key Resources Table) and Appendix.

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Novartis FreeNovation Grant; European Research Council (ERC) grant (Project no. 101043303) funded by the State Secretairat for Education, Research and Innovation, Switzerland (SERI), and Swiss Federal Institute of Technology Lausanne (EPFL) (to YA). For zebrafish husbandry: Dr. Guillaume Valentin and Ms. Chloé C L Jollivet at EPFL; and Mr. Brian J Miller and Mrs. Nikki Gilbert, and Professor Joe Fetcho (NIH R01 NS026593, PI: JRF) at Cornell University. National BioResource Project Zebrafish (NBRP) grant funded by Japanese government for Tg(gstp1:GFP) fish line. NIH CBI training grant [NIH T32GM008500 (JRP as a trainee fellow)] and AHA predoctoral fellowship (17PRE33670395 to JRP); HHMI International Fellow (SP).

All procedures performed at Cornell (2017-2018) and EPFL (2018-present) conform to the animal care, maintenance, and experimentation procedures followed by Cornell University's and EPFL's Institutional Animal Care and Use Committee (IACUC) guidelines and approved by the respective institutional committees. All experiments with zebrafish performed at EPFL (2018-present) have been performed in accordance with the Swiss regulations on Animal Experimentation (Animal Welfare Act SR 455 and Animal Welfare Ordinance SR 455.1), in the EPFL zebrafish unit, cantonal veterinary authorization VD-H23.

© 2022, Van Hall-Beauvais, Poganik, Huang et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.