The C. elegans wild-type reference strains N2/Bristol (Stiernagle, 2006), CB4270 tra-1 (e2200) (Zarkower et al., 1994), CB3844 fem-3 (e2006) IV (Hodgkin, 1986), RA7 (rdEx1) (Mathies et al., 2004), and TJ356 daf-16 (zIs356) IV (Lee et al., 2001; Lin et al., 2001) were obtained from the Caenorhabditis Genetics Center—CGC (University of Minnesota, Minneapolis, MN, United States), and are summarized in Table 1. All worm strains were maintained at 20°C, except of the CB4270 tra-1 mutant (e2200), which was maintained at 15°C. According to the instructions of the CGC, this tra-1 mutant needs to be cultured at 15°C to obtain hermaphrodites, because this worm strain will become sterile and gonad-defective at 25°C, due to feminization by smg (NMD) mutations (compare Table 3).

C. elegans was cultured on nematode growth medium (NGM: 0.3% NaCl, 1.7% agar, 0.25% peptone, 5 mg/ml cholesterol, 1 mM KPO₄, and 1 mM MgSO₄) in plastic dishes (Greiner Bio-One GmbH, Frickenhausen, Germany), which were seeded with a lawn of E. coli OP50 food bacteria (CGC, University of Minnesota, Minneapolis, MN, United States). The E. coli OP50 bacteria were cultured overnight at 37°C and diluted to an OD₆₀₀ concentration of ∼1.0 as described previously (Ke et al., 2020). Afterward, 100 µL of the E. coli OP50 suspension was plated on 35 mm NGM plates for lifespan assays or 300 µL on 60 mm NGM plates to maintain C. elegans. The plates were dried overnight at room temperature. C. elegans was maintained at 20°C according to the standard protocol (Brenner, 1974), except for the CB4270 tra-1 (e2200) and CB3844 fem-3 (e2006) strains, which were maintained at 15°C.

Lifespan assays were performed as recently described (Qi et al., 2021). For synchronization, 30 gravid C. elegans worms were transferred to new NGM plates supplemented with E. coli OP50 food bacteria to lay eggs for 3 h. Next, the worms were picked and washed. After the hatching of early-stage L1 larvae from the eggs, the larvae developed within approximately 40 h to late larval stage L4 and were then transferred to new NGM plates containing OP50 food bacteria and 100 µM sulforaphane (DL-sulforaphane; ≥95%; S4441; Sigma–Aldrich, Manheim, Germany). The control plates contained OP50 bacteria without sulforaphane. Sulforaphane was dissolved in DMSO (AppliChem, Darmstadt, Germany) to a 100 mM stock solution. During the following 9 days, the worms were transferred daily to new NGM plates with OP50 food bacteria ± sulforaphane to avoid new larvae hatching from laid eggs on the old NGM plates to avoid falsifying the results of the lifespan experiments. From Day 10, the transfer of worms to new plates was performed every second day because older worms lay fewer eggs. The transfer was performed under the microscope: a platinum pick (Neolab, Heidelberg, Germany) was sterilized by passing it through the flame of a Bunsen burner (WLD-TEC, Arenshausen, Germany), and the tip was loaded with a glob of OP50 bacteria by gently scraping through the bacterial lawn of the LB agar plate. To select a worm, its body was gently tapped with the pick until the worm adhered to the pick and could be transferred to another plate. Here, the tip of the pick was gently touched to the agar until the worm crawled off the pick. Next, the pick was sterilized for the next round of worm transfer. A worm was considered dead if it stopped moving—even after touching. Worms that exhibited internal hatching or climbed up the wall of the dishes were excluded from the analysis.

Synchronized L4 larvae (n = 200/group) were transferred to fresh NGM plates seeded with a lawn of OP50 food bacteria ± sulforaphane (100 µM), and 24, 48, 72, or 120 h later, the worms were collected, washed 3× with M9 buffer (3 g of KH₂PO₄, 6 g of Na₂HPO₄, 5 g of NaCl, 1 ml of 1 M MgSO₄, in 1 L of H₂O_bidest) (Stiernagle, 2006), and crushed under liquid nitrogen. According to the manufacturer’s instructions, total RNA was extracted using the RNeasy Mini Kit (QIAGEN, Hilden, Germany). Next, the high-capacity RNA-to-cDNA^TM Kit (Thermo Fisher Scientific, Hilden, Germany) was applied for reverse transcription. PowerUp^TM SYBR^TM Green Master Mix (Thermo Fisher Scientific, Hilden, Germany) as used to perform RT–qPCR. A 10 µL reaction mixture contained 2 µL of cDNA, 500 nM of each primer, and 5 µL of PowerUp^TM SYBR^TM Green Master Mix (2×). PCR was performed using the Applied Biosystem^TM 7,500 Real-Time PCR System (Thermo Scientific, Schwerte, Germany) and included denaturation for 2 min at 95°C and 40 cycles of amplification (15 s, 95°C; 15 s, 60°C; 1 min, 72°C). C. elegans-specific primers (Table 2) were obtained ready-to-use from Eurofins Genomics (Mannheim, Germany).

Synchronized L4 larvae (n = 200/group) of the C. elegans reporter strain RA7 were incubated in NGM agar plates with OP50 food bacteria, ±100 µM sulforaphane. After 24, 48, 72, or 120 h, the worms were collected and washed 3× with M9 buffer. The worms were transferred to H₂O_bidest containing a protease inhibitor mix (Sigma–Aldrich, Manheim, Germany) in Laemmli buffer and boiled at 95°C for 5 min, followed by centrifugation for 1 min at 13,000 g. According to the standard protocol (Wueseke et al., 2014), the protein content was not measured, but the proteins were extracted from the same number of live worms per group (n = 200). Next, 50 µL of supernatant with protein was loaded per slot of a 10% SDS polyacrylamide gel, and electrophoresis was performed at 120 V for 2 h. The gels were blotted to Immobilon®-P transfer membranes (Millipore, Billerica, United States) using a wet transfer system (Bio–Rad, Hercules, United States), followed by incubation with rabbit polyclonal anti-GFP antibody (ab290; Abcam, Cambridge, United States), diluted 1:1,000, or mouse monoclonal anti-αTubulin antibody (T5168; Sigma–Aldrich, MO, United States), diluted 1:5,000. Anti-mouse IgG and anti-rabbit IgG secondary antibodies (IRDye® 800CW; LI-COR Biosciences, Lincoln, United States) were diluted at 1:5,000 before use. After washing, the membranes were imaged using an Odyssey® CLx Infrared Imaging System (Li-cor®, Lincoln, United States).

The C. elegans RA7 and TJ356 strains were placed on 2% agarose pads (n = 5/pad) and paralyzed by adding 50 µL of a 10 μM levamisole solution (Sigma–Aldrich, MO, United States). Images of the live worms (20/group) were taken using a Leica DMRB fluorescence microscope (Leica, Wetzlar, Germany) equipped with a Thorlabs CS895CU Compact Scientific 8.9 MP color camera (Thorlabs GmbH, Bergkirchen, Germany). The expression levels were quantified by measuring the GFP fluorescence intensity using ImageJ.

RNAi-mediated inhibition of C. elegans genes was performed as previously described (Kamath et al., 2001). The tra-1 RNAi and daf-16 RNAi clones were obtained from the C. elegans RNAi feeding library, distributed by Source BioScience (Nottingham, United Kingdom). An RNAi-transfected E. coli HT115 (DE3) clone or a nontransfected E. coli HT115 (DE3) clone was picked from the delivered agar plates and incubated overnight at 37°C in 5 ml of LB medium containing 50 µg/ml of ampicillin (Sigma–Aldrich, MO, United States). Feeding plates were prepared using standard NGM agar plates supplemented with 1 mM IPTG and 25 µg/ml of carbenicillin (both from Sigma–Aldrich, MO, United States). RNAi-transfected E. coli HT115 or nontransfected E. coli HT115 control bacteria, 100 µL each, at an OD₆₀₀ concentration of ∼1.0, were seeded onto 35 mm NGM feeding plates and dried overnight at room temperature. Next, synchronized L4 larvae (n = 30) were transferred to plates and kept at 20°C for 48 h to allow the intake of RNAi-transfected food bacteria. Afterward, the gravid adult worms were picked and seeded onto new feeding plates with the same bacteria and laid eggs for 3 h. After that, the adult worms were washed. After hatching, C. elegans larvae proceeded through four larval stages (L1-L4) until the late larval stage L4/young adult stage and were then transferred to new feeding NGM plates for new experiments.

The frequency of pharyngeal pumping as a measure of appetite and food intake was determined as previously described (Raizen et al., 2012). Synchronized L4 C. elegans larvae (n = 20/group) were maintained on NGM plates containing E. coli OP 50 bacteria, RNAi-transfected E. coli HT115 bacteria, or nontransfected E. coli HT115 control bacteria in the presence or absence of 100 µM sulforaphane. The pharyngeal pumping frequency of the terminal pharyngeal bulb of individual worms was counted 60  sec and 3, 6, 9, and 12 days later under an inverted microscope (Leica, Wetzlar, Germany). Additionally, a differential interference contrast microscope (Ni-E, Nikon, Germany) was used to photograph the pharyngeal structure of C. elegans.

Worm mobility was used to reflect the functional state of motor neurons and muscle cells (Mergoud Dit Lamarche et al., 2018), as recently described (Qi et al., 2021). Synchronized L4 C. elegans larvae (n = 50/group) were maintained on NGM plates containing E. coli OP 50 bacteria or nontransfected or RNAi-transfected E. coli HT115 bacteria in the presence or absence of 100 µM sulforaphane. The body bends of single worms were counted for 60 s and 3, 6, 9, and 12 days later, and the number of body bends was calculated as body bends/min. A change in the posterior bulb direction or regular swinging of the head was considered a body bend.

For Kaplan–Meier survival curves, the p values of differences between groups were tested using the log-rank test. Student’s t-test was used to assess differences between two groups. IBM SPSS Statistics 25 software was used to perform statistical analysis. p < 0.05 was considered statistically significant.

To investigate the impact of sulforaphane, which is chemically a sulfur-containing isothiocyanate (Figure 1A), we fed synchronized L4 wild-type C. elegans larvae with E. coli OP50 food bacteria in the presence or absence of 100 µM sulforaphane. We chose this sulforaphane concentration because it is considered optimum to treat C. elegans (Qi et al., 2021). The number of worms was recorded daily until no more live worms could be found from Day 33 onwards. Afterward, a Kaplan–Meier survival analysis was performed. Sulforaphane significantly increased the mean lifespan of wild-type worms to 21.4 days compared with 19.5 days of untreated control worms, indicating a 9.9% longer survival (Figure 1B; Table 3).

To examine the involvement of tra-1 signaling, we inhibited TRA-1 by feeding E. coli HT115 bacteria transfected with tra-1 RNAi constructs to the C. elegans RA7 strain, which expresses a TRA-1-GFP fusion protein (Mathies et al., 2004). After 48 h, the proteins were extracted (200 worms/group), and Western blot analysis was performed to ensure the effectiveness of RNAi-mediated inhibition of TRA-1. A representative Western blot image is shown along with a diagram of the pixel intensities of the TRA-1 band relative to the tubulin pixel intensity, which served as a control for equal conditions. This confirmed a significant inhibition of TRA-1-GFP expression of approximately 70% (Figure 1C).

Next, we set up a series of lifespan assays to examine the longevity of tra-1(−) mutant worms relative to the presence of sulforaphane. Kaplan–Meier survival analysis revealed that the mean lifespan of tra-1 RNAi-treated nematodes that received sulforaphane was 9.6 days compared with 9.9 days of control worms, with no significant difference (Figure 1D). To further confirm the involvement of tra-1 signaling in lifespan, the tra-1(−) mutant CB4270 worm strain was evaluated by Kaplan–Meier survival analysis, resulting in a survival of 18.7 days in the presence of sulforaphane and 19.7 days in untreated control worms, with no significant difference (Figure 1E). These data suggest that sulforaphane-induced prolongation of the lifespan in C. elegans is dependent on intact tra-1 signaling.

To determine the effect of sulforaphane on tra-1 expression, we fed synchronized wild-type worms in the presence or absence of sulforaphane. After 24, 48, 72, and 120 h, RNA was isolated, and tra-1 expression was examined by RT–qPCR analysis. The level of tra-1 was significantly upregulated at 48 h after sulforaphane feeding, whereas no obvious difference was observed at 24, 72, or 120 h (Figure 2A). Likewise, the C. elegans RA7 strain, which expresses a TRA-1-GFP fusion protein, was used to identify the effect of sulforaphane on TRA-1 protein expression. Synchronized RA7 C. elegans worms were fed with or without sulforaphane. Next, 24, 48, 72, and 120 h later, the proteins were extracted, and Western blot analysis was performed using a GFP-specific antibody. We observed that sulforaphane significantly induced TRA-1-GFP expression at each time point, as shown by a representative Western blot image and the evaluation of the mean pixel intensities by ImageJ and relation to the pixel intensities of the tubulin control (Figure 2B, Supplementary Figure S1).

Next, we examined native GFP fluorescence in untreated RA7 nematodes by fluorescence microscopy and counterstained the cell nuclei with DAPI. We observed spots with enhanced fluorescence intensity in intestinal nuclei, indicating the expression and nuclear accumulation of the TRA-1 transcription factor (Figure 2C). To further validate the sulforaphane-increased accumulation of the TRA-1-GFP fusion protein, we fed RA7 worms with or without sulforaphane. Thereafter, 24, 48, 72, and 120 h later, we counted the green fluorescent spots, indicating nuclear localization of the TRA-1-GFP fusion protein using fluorescence microscopy. The mean fluorescence intensities were assessed using ImageJ and revealed that sulforaphane further increased the nuclear accumulation of the TRA-1-GFP fusion protein at each time point (Figures 2D,E). These findings suggest that sulforaphane enhances TRA-1 expression.

To evaluate whether TRA-1 serves as a mediator of an increased healthspan after sulforaphane treatment, we analyzed the mobility and pharyngeal pumping rate of C. elegans because these features display age‐related declines (Collins et al., 2008). To assess appetite and food intake, the pumping frequency of the terminal pharyngeal bulb (Figure 3A) of each worm was documented by microscopy and video recordings at Days 3, 6, 9, and 12 after sulforaphane feeding. However, sulforaphane did not significantly alter the pharyngeal pumping frequency in younger worms at Days 3, 6, and 9; it significantly increased the frequency to 52 pumps/minute in older worms on Day 12 (Figure 3B). By contrast, untreated wild-type C. elegans worms exhibited 19 pumps/minute at Day 12, a time point at which the healthspan of worms was significantly decreased. Using tra-1 RNAi-treated nematodes and examining the pharyngeal pumping rate in the presence or absence of sulforaphane, we observed no significant differences between the groups at any time point examined (Figure 3C). We would like to mention, that we tried to measure the pharyngeal pumping rate also at Day 12, which was, however, not possible, because tra-1 RNAi transfection dramatically shortened the lifespan and at Day 12 almost all worms were dead, in both, the sulforaphane-treated and -untreated groups (compare Figure 1D). Interestingly, the healthspan of tra-1 RNAi-treated C. elegans worms significantly decreased at Day 9, and the worms had a lower pumping rate from the beginning and exhibited an almost quiescent state with little pharyngeal pumping, reflecting the shorter healthspan.

Next, mobility was assessed by counting the number of body bends/minute (Figure 3D). Compared with untreated wild-type worms, a significant increase from 7 to 10 body bends/minute was observed in older worms at Day 12 after sulforaphane treatment, whereas no significant difference occurred in younger worms at Days 3, 6, and 9 (Figure 3E). By contrast, sulforaphane was ineffective in C. elegans with RNAi-mediated inhibition of TRA-1 expression (Figure 3F). These results indicate that TRA-1 expression is necessary for an enhanced healthspan following sulforaphane treatment.

To further elucidate the sulforaphane-induced signaling cascade, which mediates longevity and an increased healthspan, we focused on the transcription factor DAF-16 because its binding by TRA-1 was recently detected (Hotzi et al., 2018), and DAF-16 signaling was associated with the longevity of C. elegans (Lin et al., 2018). We used the C. elegans reporter strain TJ356, which expresses a DAF-16-GFP fusion protein (Lin et al., 2001). Synchronized TJ356 L4 larvae were fed with or without sulforaphane. After 48 h, the expression and cellular localization of the DAF-16-GFP fusion protein were examined by fluorescence microscopy. Sulforaphane-fed worms exhibited strong green fluorescent spots, indicating enhanced DAF-16 nuclear translocation, whereas the signal in control worms was lower (Figure 4A). This observation was confirmed by quantification of the fluorescence intensity using ImageJ, and the mean fluorescence intensities with standard deviations and significance are shown (Figure 4B). We confirmed these results by RNAi-mediated inhibition of TRA-1 expression in TJ356 C. elegans (Figures 4C,D). These data suggest that TRA-1 mediates sulforaphane-induced nuclear translocation of DAF-16.

To examine the relationship between TRA-1 and DAF-16 in longevity and healthspan, we evaluated the survival of CB3844 fem-3 mutant worms that harbor a hyperactive TRA-1 protein (Hotzi et al., 2018). Compared with wild-type worms with an average lifespan of 20.5 days, the survival of CB3844 mutant worms was 22.3 days, corresponding to a longer survival of 8.9% (Figure 5A; Table 4). To determine whether DAF-16 is involved in TRA-1-mediated life prolongation, we inhibited daf-16 by RNAi in wild-type and CB3844 mutant worms and performed Kaplan-Meier analysis. With inhibited daf-16 expression, the CB3844 worms did not survive longer than the wild-type worms (Figure 5B), indicating that DAF-16 is required for TRA-1-induced longevity.

Finally, we assessed whether DAF-16 may be required for the previously observed TRA-1-mediated prolonged healthspan, the expression of daf-16 was inhibited by feeding E. coli HT115 bacteria transfected with daf-16 RNAi, which resulted in a significant inhibition of daf-16 RNA levels in N2 wild-type and CB3884 worms, as examined by RT–qPCR 48 h later (Supplementary Figure S1). The examination of pharyngeal pumping demonstrated that the pumping frequency significantly decreased from 28 to 7 pumps/minute at Day 12 in older wild-type worms, whereas no difference was observed in younger worms at Days 3, 6, and 9 (Figure 5C). This effect was even more pronounced in the C. elegans CB3844 strain following RNAi-mediated inhibition of daf-16 because the pumping frequency already decreased not only at Day 12 but also at Days 6 and 9 (Figure 5D). Likewise, the inhibition of daf-16 expression by RNAi in wild-type worms significantly reduced the body bends/minute at all time points examined (Figure 5E). The effects were even more pronounced following RNAi-mediated inhibition of daf-16 in CB3844 worms (Figure 5F), as expected. These results imply that DAF-16 is a crucial downstream effector in TRA-1-mediated increased longevity and healthspan in C. elegans (Figure 6), although this conclusion requires further confirmation.