**2. Results**

#### *2.1. Ricin-Induced Death Is Primed by TRAIL, TNF-*<sup>α</sup>*, and FasL*

To test the hypothesis that extrinsic cytokines cause an increase in ricin-induced cell death, A549 cells were treated with increasing doses of ricin in the absence or presence of 100 ng/mL TRAIL, TNF-<sup>α</sup>, or FasL for 24 h at 37 ◦C. Over a range of toxin concentrations, addition of each of the cytokines resulted in a significant increase in ricin-induced cell death (Figure 1A). Our previous work indicated that ricin/TRAIL induces apoptosis of Calu-3 human lung epithelial cells [20]. Indeed, when we used the pan-caspase inhibitor zVAD-fmk, A549 cell death induced by ricin combined with TRAIL, TNF-<sup>α</sup>, or FasL was prevented (Figure 1B−D). Collectively, the results of Figure 1 indicate that TRAIL, TNF-<sup>α</sup>, and FasL enhance ricin-induced cell death in a manner that is likely caspase-dependent apoptosis.

**Figure 1.** Extrinsic cytokines enhance ricin-induced death of A549 cells in a zVAD inhibitable manner. A549 lung epithelial cells were treated with ricin alone or in combination with 100 ng/mL TRAIL, TNF-<sup>α</sup>, or FasL for 24h at 37 ◦C followed by measurement of cell death via WST-1 assay. (**A**) All 3 cytokines/ligands resulted in a significant increase in cell death relative to treatment with ricin alone. A549 cell death induced by ricin combined with (**B**) TRAIL, (**C**) TNF-<sup>α</sup>, or (**D**) FasL is prevented by the pan-caspase inhibitor, zVAD-fmk (50 μM). Results are the average of 3 independent experiments. Error bars = standard deviation. Two-way analysis of variance (ANOVA), \*\*\* *p* < 0.001.

#### *2.2. Cell Death Induced by Ricin*/*TRAIL Is Associated with Caspase Activation while Death by Ricin*/*TNF-*α *and Ricin*/*FasL Is Not*

To ge<sup>t</sup> a clear view of the involvement of caspases, the effectors of apoptosis [21], in cell death by ricin combined with TRAIL, TNF-<sup>α</sup>, or FasL, A549 cells were treated with 1 ng/mL ricin combined with 100 ng/mL TRAIL, TNF-<sup>α</sup>, or FasL for 4 h at 37 ◦C followed by cell lysis and western blot. Treatment with ricin or any of the cytokines alone did not result in caspase cleavage/activation (Figure 2A–C). When combined with TRAIL, ricin induced cleavage/activation of caspases-3, -7, -8, and -9 but not caspase-6 (Figure 2A and Figure S1). However, the combination of ricin and TNF-α or ricin and FasL did not cause cleavage/activation of any caspase tested (Figure 2B,C). To determine if caspases were cleaved with slower kinetics, we measured caspase cleavage in response to ricin/TNF-α or ricin/FasL after 8 h of treatment. However, caspases were not cleaved/activated in A549 cells at this time point (Figure S2). As a positive control for TRAIL-, TNF-, and FasL-induced apoptosis, A549 cells were treated with 250 ng/mL cycloheximide (CHX) combined with 100 ng/mL TRAIL, TNF-<sup>α</sup>, or FasL [30–37]. Apoptosis induced by CHX combined with any of the cytokines resulted in cleavage/activation of caspases-3, -6, -7, -8, and -9 (Figure 2D and Figure S3). These results clearly demonstrate that caspases are activated following attack by ricin/TRAIL but are not affected by ricin/TNF-α and ricin/FasL.

**Figure 2.** The combination of ricin and TRAIL induces caspase activation, while addition of TNF-α or FasL with ricin does not. A549 lung epithelial cells were treated with 1 ng/mL ricin alone or in combination with 100 ng/mL TRAIL, TNF-<sup>α</sup>, or FasL for 4 h at 37 ◦C followed by cell lysis and western blot. (**A**) The combination of ricin and TRAIL results in cleavage/activation of caspases-3, -7, -8, and -9. Caspase-6 is not cleaved/activated in response to ricin/TRAIL (**B**,**C**) The combination of ricin/TNF-α or ricin/FasL does not result in the cleavage/activation of caspases. (**D**) A549 cells were treated with 250 ng/mL cycloheximide (CHX) combined with TNF-<sup>α</sup>, FasL, or TRAIL as a positive control for TRAIL-, TNF-, and FasL-induced apoptosis. As expected, when cycloheximide is combined with TNF-<sup>α</sup>, FasL, or TRAIL it results in the cleavage/activation of caspases-3, -6, -7, -8, and -9. Shown are representative blots from 3 independent experiments.

#### *2.3. The Combination of Ricin and TRAIL Induces Caspase-Dependent Apoptosis*

The results of Figure 2 sugges<sup>t</sup> that ricin/TRAIL causes activation of caspases, and thus apoptosis, while ricin/TNF-α and ricin/FasL do not. This finding was unexpected for ricin/TNF-α and ricin/FasL. Therefore, we decided to perform a detailed characterization of A549 cell death caused by ricin combined with TRAIL, TNF-<sup>α</sup>, or FasL. A549 cells were treated with increasing doses of ricin and 100 ng/mL TRAIL for 24 h at 37 ◦C in the absence or presence of specific caspase inhibitors. We determined that cell death induced by ricin combined with TRAIL depends on executioner caspases-3 and -7 (Figure 3A and Figure S4) as well as initiator caspases-8 and -9 (Figure 3C,D and Figure S4) but not caspase-6 (Figure 3B). These results are in agreemen<sup>t</sup> with our caspase activation/cleavage results of Figure 2. As a positive control for TRAIL-induced apoptosis, A549 cells were treated with 250 ng/mL CHX combined with increasing concentrations of TRAIL [30] (Figure 3E). Cell death induced by CHX/TRAIL was prevented by inhibition of caspases-3, -7, -8, and -9 as the combination of CHX and TRAIL induces apoptosis [30,38,39]. These results indicate that ricin induces caspase-dependent apoptosis of human lung epithelial cells when combined with TRAIL similar to the combination of CHX and TRAIL.

**Figure 3.** A549 cell death induced by ricin/TRAIL depends on caspases-3, -7, -8, and -9. A549 lung epithelial cells were treated with ricin alone or in combination with 100 ng/mL TRAIL in the presence or absence of caspase inhibitors for 24 h at 37 ◦C followed by measurement of cell death via WST-1 assay. Cell death induced by the combination of ricin and TRAIL was prevented by inhibition of (**A**) caspases-3 and -7 with zDEVD-fmk (30 μM) but not (**B**) caspase-6 with zVEID-fmk (30 μM). Moreover, cell death by ricin/TRAIL was prevented by inhibitions of (**C**) caspase-8 with zIETD-fmk (30 μM) and (**D**) caspase-9 with zLEHD-fmk (10 μM). (**E**) A549 cells were treated with the combination of 250 ng/mL cycloheximide (CHX) and TRAIL as a positive control for TRAIL-induced apoptosis. As expected, cycloheximide/TRAIL-induced apoptosis is prevented by all caspase inhibitors tested. Results are the average of 3 independent experiments. Error bars = standard deviation. Two-way ANOVA, \*\*\* *p* < 0.001, \*\* *p* < 0.01.

#### *2.4. Ricin Combined with TNF-*α *or FasL Induces Caspase-Independent Cell Death*

We next characterized A549 cell death by ricin/TNF-α or ricin/FasL with respect to the findings of Figure 2. A549 cells were treated with increasing doses of ricin and 100 ng/mL TNF-α or FasL for 24 h at 37 ◦C. Cell death induced by ricin/TNF-α or ricin/FasL was not prevented by specific inhibition of caspases-3, -6, -7, -8, or -9 (Figure 4A–F and Figure S5). In contrast to the results with zVAD-fmk (Figure 1C,D), these results are in agreemen<sup>t</sup> with those of Figure 2. As positive controls for TNFand FasL-induced apoptosis, A549 cells were treated with 250 ng/mL CHX combined with increasing concentrations of TNF-α [30–37] (Figure 4G) or FasL [30,34,36] (Figure 4H). TNF- and FasL-induced apoptosis were prevented by inhibition of caspases-3, -7, -8, and -9 (Figure 4G,H). These results indicate

that ricin/TNF-α and ricin/FasL induce caspase-independent cell death which is distinct from the caspase-dependent apoptosis induced by CHX/TNF-α and CHX/FasL.

**Figure 4.** Ricin/TNF-α and ricin/FasL both induce caspase-independent death of A549 cells. A549 lung epithelial cells were treated with ricin alone or in combination with 100 ng/mL TNF-α or FasL in the presence or absence of caspase inhibitors for 24 h at 37 ◦C followed by measurement of cell death via WST-1 assay. Cell death induced by ricin/TNF-α and ricin/FasL was not prevented by inhibition of (**A**,**B**) caspases-3 and -7 with zDEVD-fmk (30 μM), (**C**,**D**) caspase-8 with zIETD-fmk (30 μM), or (**E**,**F**) caspase-9 with zLEHD-fmk (10 μM) (**G**,**H**) A549 cells were treated with the combination of 250 ng/mL cycloheximide (CHX) and TNF-α or cycloheximide and FasL as positive controls for TNF- and FasL-induced apoptosis. As expected, cycloheximide/TNF- and cycloheximide/FasL-induced apoptosis is prevented by all caspase inhibitors tested. Results are the average of 3 independent experiments. Error bars = standard deviation. Two-way ANOVA, \*\*\* *p* < 0.001.

Since cell death induced by ricin/TNF-α and ricin/FasL was prevented by the pan-caspase inhibitor, zVAD-fmk but not inhibition of specific apoptotic caspases (Figures 1C,D and 4 and Figure S5), we investigated the involvement of alternative caspases and apoptosis effectors in both instances of cell death. Inhibition of caspase-1, the central effector caspase of pyroptosis [40,41], did not prevent cell death by ricin/TNF-α or ricin/FasL (Figure 5A,B). Caspase-2 is thought to have initiator and effector roles in some forms of apoptosis [42] ye<sup>t</sup> its inhibition did not prevent cell death by ricin combined

with either cytokine (Figure 5C,D). During intrinsic apoptosis, Bax is critical for mitochondrial outer membrane pore formation and cytochrome *c* release [21]. However, when Bax was inhibited there was no effect on the cell death induced by ricin/TNF-α or ricin/FasL (Figure 5E,F) in contrast to its effect on apoptosis induced by CHX/TNF-α or CHX/FasL (Figure S6). Furthermore, cytochrome *c* was not observed in the cytoplasm of cells treated with ricin/TNF-α or ricin/FasL (Figure S7).

**Figure 5.** A549 cell death induced by ricin in combination with TNF-α or FasL does not depend on caspase-1, -2, or Bax. A549 lung epithelial cells were treated with ricin alone or in combination with 100 ng/mL TNF-α or FasL in the presence or absence of various pharmacologic inhibitors. Cell death induced by the combination of ricin with TNF-α or FasL was not prevented by inhibition of (**A**,**B**) caspase-1 with zYVAD-fmk (10 μM), (**C**,**D**) caspase-2 with zVDVAD-fmk (50 μM), or (**E**,**F**) Bax with a peptide-based inhibitor (Bax-inhibiting peptide v5, 100 μM). Results are the average of 3 independent experiments. Error bars = standard deviation. Two-way ANOVA, \*\*\* *p* < 0.001.

Necroptosis is another major pathway of cell death in addition to apoptosis which may be induced by TNF-α and FasL [23,43]. Thus, we investigated the involvement of RIP1, the initiator kinase of necroptosis [23,43], in cell death by ricin combined with TNF-α or FasL. Cell death induced by ricin combined with either cytokine was not prevented by the RIP1 inhibitor, necrostatin-1s (Figure S8). Collectively, the results of Figure 5 and Figure S7 indicate that ricin/TNF-α and ricin/FasL do not induce one of the classical non-apoptotic cell death pathways. Additionally, the pattern of Mcl1 protein loss does not differ between ricin/TRAIL vs. ricin/TNF-α or ricin/FasL in A549 cells (Figure S9). Thus, we believe the distinct cell death responses to ricin combined with different cytokines is not the result of altered kinetics of protein synthesis inhibition.

#### *2.5. Cell Death Induced by Ricin*/*TNF-*α *and Ricin*/*FasL Depends on Cathepsins*

Treatment with zVAD-fmk (pan-caspase inhibitor) prevented cell death induced by ricin/TNF-α and ricin/FasL (Figure 1C,D) ye<sup>t</sup> it does not appear as though caspases are activated during these instances of cell death (Figures 2 and 4, Figure S2 and Figure S5). Importantly, zVAD-fmk may have off-target effects against cathepsins in addition to its specific effects on caspases [44,45]. Therefore, we wondered if the effects of this inhibitor were due to deactivation of cathepsins. Interestingly, zFA-fmk, an inhibitor of executioner caspases-2, -3, -6, and -7 [46] as well as cathepsins B, L and S [47], caused significant inhibition of cell death induced by ricin combined with either TNF-α or FasL (Figure 6A,B). In addition, E64d, an inhibitor of calpains as well as cathepsins B, H, and L [48], blunted A549 cell death by ricin/TNF-α or ricin/FasL (Figure 6C,D). However, calpeptin, an inhibitor of calpains as well as cathepsins L and K [49], had no effect on cell death by ricin/TNF-α or ricin/FasL (Figure S10). Cathepsin inhibitor 1 (CATI-1) has activity against cathepsins L and S with cathepsin B being its primary target [50]. Inhibition with CATI-1 prevented A549 cell death by ricin combined with TNF-α or FasL (Figure 6E,F). Importantly, CATI-1 had not effect on apoptosis induced by ricin/TRAIL (Figure S11). Cathepsin-dependent cell death is often associated with a dependence on reactive oxygen species (ROS) [51]. We investigated the involvement of ROS in cell death induced by ricin/TNF-α and ricin/FasL using the antioxidant, N-acetylcysteine. Scavenging of ROS by N-acetylcysteine (NAC) resulted in a significant prevention of cell death induced by ricin/TNF-α or ricin/FasL (Figure S12). Collectively, these results indicate that when combined with TNF-α or FasL, ricin induces cell death that depends on cathepsins with a likely role for ROS.

**Figure 6.** Cell death induced by ricin/TNF-α or ricin/FasL depends on cathepsins. A549 lung epithelial cells were treated with ricin alone or in combination with 100 ng/mL TNF-α or FasL for 24 h at 37 ◦C followed by measurement of cell death via WST-1 assay. Cell death induced by ricin/TNF-α and ricin/FasL was partially prevented by inhibition of cathepsins with (**A**,**B**) zFA-fmk (50 μM), (**C**,**D**) E64d (50 μM), or (**E**,**F**) cathepsin inhibitor 1 (CATI-1, 20 μM). Results are the average of 3 independent experiments. Error bars = standard deviation. ANOVA, \*\*\* *p* < 0.001, \*\* *p* < 0.01.

The dominant cell line used in this study was A549 human lung epithelial cells, a cell line derived from lung carcinoma. These cells were chosen as they represent a model of human alveolar, type II pneumocytes for drug and toxin metabolism [52,53]. To determine if our results were unique to A549 cells, we conducted experiments with Calu3 human lung epithelial cells. Previously, we showed that ricin/TRAIL induces caspase-dependent apoptosis in Calu3 cells [20]. Here we show that Calu3 cells appear to be insensitive to cell death by ricin/TNF-α or ricin/FasL (Figure S13). Therefore we tested another human cell line, U937 monocytes, as alveolar macrophages are another target cell of ricin upon inhalation. Indeed, ricin/TRAIL induced caspase-dependent apoptosis in U937 cells while ricin/TNF-α and ricin/FasL induced cathepsin-dependent cell death (Figure S14). While the results obtained in U937 cells supports our findings in A549 cells, these are also a cell line derived from cancerous tissue. Future work should focus on comparing these findings to those obtained in primary lung epithelial cells and macrophages.
