Septins promote caspase activity and coordinate mitochondrial apoptosis

Abstract Apoptosis is a form of regulated cell death essential for tissue homeostasis and embryonic development. Apoptosis also plays a key role during bacterial infection, yet some intracellular bacterial pathogens (such as Shigella flexneri, whose lipopolysaccharide can block apoptosis) can manipulate cell death programs as an important survival strategy. Septins are a component of the cytoskeleton essential for mitochondrial dynamics and host defense, however, the role of septins in regulated cell death is mostly unknown. Here, we discover that septins promote mitochondrial (i.e., intrinsic) apoptosis in response to treatment with staurosporine (a pan‐kinase inhibitor) or etoposide (a DNA topoisomerase inhibitor). Consistent with a role for septins in mitochondrial dynamics, septins promote the release of mitochondrial protein cytochrome c in apoptotic cells and are required for the proteolytic activation of caspase‐3, caspase‐7, and caspase‐9 (core components of the apoptotic machinery). Apoptosis of HeLa cells induced in response to infection by S. flexneri ΔgalU (a lipopolysaccharide mutant unable to block apoptosis) is also septin‐dependent. In vivo, zebrafish larvae are significantly more susceptible to infection with S. flexneri ΔgalU (as compared to infection with wildtype S. flexneri), yet septin deficient larvae are equally susceptible to infection with S. flexneri ΔgalU and wildtype S. flexneri. These data provide a new molecular framework to understand the complexity of mitochondrial apoptosis and its ability to combat bacterial infection.

Following activation, initiator caspases trigger proteolytic stimulation of effector caspase-3 and caspase-7 zymogens that enable the execution phase of the apoptotic cell death program by cleaving functionally important proteins within a cell (Ashkenazi & Salvesen, 2014).
Apoptosis is widely recognized as important for immune defense against bacterial invasion, yet pathogens deploy a variety of mechanisms to manipulate cell death pathways and promote infection (Naderer & Fulcher, 2018).In the case of S. flexneri, both inhibition and induction of apoptosis have been described.S. flexneri can inhibit apoptosis in epithelial cells by inhibiting the activation of caspase-3 (Clark & Maurelli, 2007), and recent work identified that S. flexneri lipopolysaccharide (LPS) can directly bind to caspase-3/-7 via its Oantigen moiety, blocking apoptosis signaling (Günther et al., 2020).On the other hand, work has shown that S. flexneri can induce apoptosis of infected cells via mitochondrial depolarization and caspase-9 activation (Lembo-Fazio et al., 2011).Taken together, the precise role of apoptosis in control of S. flexneri infection is complex and, in terms of host and pathogen survival, is the subject of great interest (Ashida, Suzuki, & Sasakawa, 2021).
In this study, we investigate a role for septins in mitochondrial apoptosis induced by pharmacological treatment and S. flexneri infection.We discover that septins (SEPT7 and SEPT2) are required for apoptotic cell death.Consistent with this, we show that septins promote the release of cytochrome c from mitochondria, as well as the activation of caspase-3, caspase-7, and caspase-9.Finally, we highlight the role of septin-mediated apoptosis during S. flexneri infection in vitro using human epithelial cells and in vivo using zebrafish larvae.

| Septins are required for apoptotic cell death
To test a role for septins in apoptosis of human epithelial cells (HeLa), small interfering RNA (siRNA) was used to deplete SEPT7, and staurosporine (STS, a pan-kinase inhibitor) was used to induce mitochondrial apoptosis (Figure 1a).We observed that 24 hr incubation of 1 μM STS dramatically reduced the viability of control cells (9.2 ± 1.9% cells survived), but only partially reduced the viability of SEPT7-depleted cells (23.1 ± 2.0% cells survived) (Figure 1b).Similar results are observed for SEPT2-depleted cells (Figure 1c, d).To visualize apoptosis in STS-treated cells, samples were fluorescently labeled with Annexin V (to identify phosphatidylserine externalization, a hallmark of apoptotic cells) and Hoechst (to visualize nuclei) (Figure 1e).
In agreement with cell viability results, the proportion of Annexin Vpositive cells is significantly reduced (4.2 ± 0.1 fold) in SEPT7-depleted cells, as compared to control cells (Figure 1f).
We next used etoposide, a drug that inhibits topoisomerase II causing errors in DNA replication, to induce mitochondrial apoptosis.Consistent with results obtained for STS-treated cells, depletion of SEPT7 significantly reduced (4.2 ± 0.8 fold) the proportion of Annexin V-positive cells, as compared to control cells (Figure 1e, f).Together, these results show that septins promote mitochondrial apoptosis of HeLa cells.

| Septins are required for cytochrome c release
Cytochrome c is located in the mitochondrial intermembrane space under normal conditions, but under apoptotic conditions is released to the cytosol for the activation of initiator caspases (Bock & Tait, 2020;Dorstyn, Akey, & Kumar, 2018).Considering that septins are well known to play a key role in mitochondrial dynamics (Pagliuso et al., 2016;Sirianni et al., 2016), we hypothesized that septins can also promote the release of cytochrome c when mitochondrial apoptosis is stimulated.We treated control and SEPT7-depleted cells with STS and examined cytochrome c release using confocal microscopy and immunostaining of cytochrome c (as well as Tom20, a mitochondrial marker) (Figure 2a).In this case, the number of SEPT7-depleted cells releasing cytochrome c is significantly reduced (3.2 ± 0.4 fold), as compared to control cells (Figure 2b).
Previous work has shown that ARTS, a variant of SEPT4, promotes mitochondrial apoptosis (Larisch et al., 2000).To investigate if SEPT7 depletion compromises the activation of apoptosis via ARTS, we quantified protein levels of ARTS and other septins in SEPT7-depleted conditions.Consistent with previous work, we showed that the depletion of SEPT7 significantly reduced SEPT2, SEPT7, and SEPT9 protein levels (Estey et al., 2010;Lobato-Marquez et al., 2019), but ARTS protein levels are not affected by SEPT7 depletion (Figure 2c).These results indicate that the requirement of SEPT7 in mitochondrial apoptosis is not due to the regulation of ARTS.

| Septins are required for activation of caspase-3/ -7/, and -9
Activation of effector caspases (caspase-3 and caspase-7) occurs in two steps, beginning with initial cleavage by active caspase-9 to form the p20 subunit.After this cleavage, caspase-3 or caspase-7 removes its own prodomain to generate the active p17 subunit.To understand how septins promote mitochondrial apoptosis, we measured the effect of septin depletion on STS-or etoposide-induced activation of caspase-9 by Western blotting (Figure 3a, b).In both cases, SEPT7 depletion caused a significant decrease in levels of active caspase-9, as compared to control cells.Consistent with a role for septins in mitochondrial apoptosis, SEPT7 depletion caused a significant decrease in levels of active caspase-3 and caspase-7, as compared to control cells (Figure 3a, b).Similar results are observed for SEPT2-depleted cells (Figure 3c).Importantly, SEPT7 depletion had no significant effect on levels of caspase-4, an inflammatory caspase with no role in STS-mediated mitochondrial apoptosis (Figure 3d).
Next, we imaged HeLa cells labeled for active caspase-3 using confocal microscopy (Figure 4a).Consistent with results obtained by Western blot, SEPT7-depleted cells showed a significant decrease (3.0 ± 0.6 fold) in the number of apoptotic cells labeled for active caspase-3, as compared to control cells (Figure 4b).Similar results are obtained when we labeled for active caspase-3 and performed flow cytometry analysis on siRNA-treated cells (Figure 4c).Together, these results highlight an important role for septins in the activation of caspase-3/ -7/, and -9 during mitochondrial apoptosis.

| Testing the role of septin-mediated apoptosis during S. flexneri infection
New work has shown that S. flexneri LPS can bind to caspase-3/ -7 via its O-antigen moiety, blocking apoptosis signaling (Günther et al., 2020).Considering this, we used a S. flexneri mutant lacking the O-antigen and outer core components of LPS (ΔgalU) (Lobato-Márquez et al., 2021) to test for caspase activation during infection of HeLa cells.Consistent with a role for bacterial LPS in blocking apoptosis, S. flexneri ΔgalU fails to inhibit the activation of caspase-3/ -7/ and -9 as efficiently as wildtype S. flexneri (Figure 5a).Strikingly, the activation of caspase-3/ -7/ -9 in response to S. flexneri ΔgalU infection is dependent on SEPT7 (Figure 5a).
To investigate the role of septin-mediated apoptosis during infection in vivo, we used the well-established S. flexneri-zebrafish infection model (Duggan & Mostowy, 2018;Gomes & Mostowy, 2020;Torraca & Mostowy, 2018).Zebrafish septins are highly homologous to human septins and all septin groups (SEPT2, SEPT3, SEPT6, SEPT7) are represented in zebrafish, though some septins are duplicated in zebrafish (Figure S1A).For example, three homologs of human SEPT7 are present in zebrafish (sept7a, sept7b, sept15).In contrast, only one zebrafish sept2 gene has been identified, and it has high homology to human SEPT2 (Figure S1A).Moreover, RNAseq data from our lab (Torraca et al., 2019) suggested that sept2 is the most expressed septin in zebrafish at the whole animal level (Figure S1B, Table S1).Considering this, to investigate the role of septin-mediated apoptosis in vivo, we targeted zebrafish Sept2 using a highly effective CRISPR-Cas9 method (Kroll et al., 2021) resulting in F0 knockout (Figure 5b, Figure S1C).Van Ngo & Mostowy, 2019).Here, we discover that septins are required for mitochondrial apoptosis because they promote cytochrome c release and activation of caspase-3, caspase-7, and caspase-9.We highlight the role of septin-mediated apoptosis during S. flexneri infection in vitro using human epithelial cells and in vivo using zebrafish larvae.Further investigations into septin biology will help to decipher the molecular mechanisms underlying human diseases associated with mitochondrial apoptosis, including susceptibility to infection.
The similarity of results obtained with SEPT7 and SEPT2 strongly suggests that septin hetero-oligomers (and not septin monomers or dimers) are required for their role in apoptosis.This is in contrast to the apoptotic role of ARTS which is viewed to act independently of other septins.ARTS is localized to the mitochondrial outer membrane (Edison et al., 2012).In response to apoptotic stimuli, ARTS translocates to the cytosol and directly binds and antagonizes anti-apoptotic XIAP, leading to the release of non-lethal active caspases (e.g., caspase 9) from XIAP.In turn, these caspases cleave substrates (e.g., Bid), resulting in mitochondrial outer membrane permeabilization (MOMP) and apoptosis.Work has shown that ARTS acts as a scaffold to bring XIAP into close proximity with anti-apoptotic Bcl-2 and promote its degradation (Edison et al., 2017).Considering work from our lab and others (Krokowski et al., 2018;Lobato-Márquez et al., 2021;Mageswaran et al., 2021;Mostowy et al., 2010;Pagliuso et al., 2016;Sirianni et al., 2016) it is tempting to speculate that septins (including SEPT7 and SEPT2, but not ARTS) are recruited to mitochondrial constriction sites that present micron-scale curvature.Here, septins may function as a scaffold to recruit Drp1 to mitochondrial fission sites (in healthy cells) or apoptotic foci (in apoptotic cells), and promote interactions of Drp1 with other proteins involved in MOMP (e.g., Bax).Consistent with this, new work has shown that Drp1 interacts with Bax at apoptotic foci resulting in MOMP and apoptosis (Jenner et al., 2022).
To test the role of septins in apoptosis in vivo, we took advantage of the Shigella-zebrafish infection model (Duggan & Mostowy, 2018;Gomes & Mostowy, 2020;Torraca & Mostowy, 2018).We observe that, as compared to control larvae, sept2 crispants are not more susceptible to S. flexneri ΔgalU (i.e., bacteria that do induce apoptosis) as compared to wildtype S. flexneri (i.e., bacteria that do not induce apoptosis).In contrast, control larvae are more susceptible to S. flexneri ΔgalU as compared to wildtype S. flexneri.Considering that survival of sept2 crispants is not affected in this case, these data are in agreement with our hypothesis that septins are required for S. flexneri ΔgalU to induce host cell death via apoptosis.However, given the pleiotropic role of septins in vivo, these results involving zebrafish infection are likely not solely because of Sept2's role in apoptosis.As compared to control larvae, sept2 crispants are significantly more susceptible to wildtype S. flexneri.In agreement with previous work from our lab studying septins in zebrafish (Mazon-Moya et al., 2017;Mostowy et al., 2013), these data suggest that the role of Sept2 in restricting bacterial infection may also be acting through septin cages, bacterial autophagy, and inflammation control.Relatively little is known about the role of apoptosis during S. flexneri infection in vivo.
It will thus be of great interest to further study the link between septins and apoptosis using a zebrafish model of Shigella infection.
S. flexneri M90T (Mostowy et al., 2010) or S. flexneri ΔgalU (lacking O-antigen and outer core components of LPS) (Lobato-Márquez et al., 2021) were grown in trypticase soy (TCS) agar containing 0.01% (w/v) congo red to select for red colonies, indicative of a functional type III secretion system (T3SS).TCS liquid cultures were inoculated with individual red colonies of S. flexneri M90T or S. flexneri ΔgalU and were grown overnight at 37 C with shaking.The following day, bacterial cultures were diluted in fresh prewarmed TCS (1:50 v/v) and cultured until an optical density (OD 600 ) of 0.6.

| siRNA transfections and chemical treatments
HeLa cells (7 Â 10 4 ) were plated in 6-well plates (Thermo Scientific) for 24 hr and then transfected with selected siRNAs as previously described (Lobato-Márquez et al., 2019;Mostowy et al., 2010;Sirianni et al., 2016).siRNA transfection was performed in DMEM with oligofectamine (Invitrogen) according to the manufacturer's instructions.Cells were tested 72 hr after siRNA transfection.Control siRNA (ID#14614) and predesigned siRNA for SEPT2 (ID#14709 and F I G U R E 2 Cytochrome c release is significantly reduced in SEPT7-depleted cells.HeLa cells were transfected with control or SEPT7 siRNA, treated with DMSO or 1 μM STS for 10 hr before being fixed and labeled with an anti-cytochrome c and anti-Tom20 antibodies and DAPI.assays (Kumar, Nagarajan, & Uchil, 2018).After DMSO or STS treatments, cells were washed with pre-warmed PBS and placed in serum-free DMEM.MTT was added to each well to a final concentration of 1 mg/mL and plates were incubated for 45 min at 37 C in 5% CO 2 .DMSO was added to each well (final volume 50%) and plates were agitated gently for 10 min at room temperature to dissolve the precipitate of the blue dyes.Samples were then diluted 1:1 with DMSO and measured the absorbance at the wavelength of 600 nm.

| Apoptosis assays
For apoptosis assays (Figure 1c), approximately 2.5 Â 10 4 cells were seeded into Lab-Tek 8-well glass-bottom chamber slides At 3 days post-fertilization (dpf), sept2 crispants displayed some growth impairment, as shown by a reduced body length (Figure5c-d); only larvae capable of normal hatching and free from major developmental abnormalities were selected for infection studies.sept2 crispants are significantly more susceptible to wildtype S. flexneri infection as compared to control larvae (Figure 5e-f), highlighting an in vivo role for septins in host defense.Control larvae are significantly more susceptible to S. flexneri ΔgalU as compared to control larvae infected with wildtype S. flexneri, but sept2 crispants (i.e., where mitochondrial apoptosis is compromised) are not significantly more susceptible to S. flexneri ΔgalU as compared to sept2 crispants infected with wildtype S. flexneri (Figure 5e-f).3| CONCLUDING REMARKSInvestigations using cellular and animal models have shown that septins are crucial for sensing bacteria and for promoting effector mechanisms to eliminate them(Mostowy & Shenoy, 2015;Robertin & Mostowy, 2020;    F I G U R E 1 Septin-depleted cells undergo significantly less apoptosis.(a) HeLa cells were transfected for 72 hr with control (CTRL) siRNA or a siRNA targeting SEPT7.Effect of siRNA on SEPT7 expression.A representative Western blot of HeLa cell lysates is shown.GAPDH was used as a loading control.(b) Impact of siRNA targeting SEPT7 on viability of HeLa cells in (i) untreated or (ii) STS-treated conditions.Cell viability is measured using MTT assays as described in Materials and Methods.Data are mean ± SEM from three independent experiments.**p = .0069by two-tailed Student's t test.(c) HeLa cells were transfected for 72 hr with control (CTRL) siRNA or a siRNA targeting SEPT2.Effect of siRNA on SEPT2 expression.A representative Western blot of HeLa cell lysates is shown.(d) Impact of siRNA targeting SEPT2 on viability of HeLa cells in (i) untreated or (ii) STS-treated conditions.Cell viability is measured using MTT assays as described in Materials and Methods.Data are mean ± SEM from three independent experiments.***p = .0004by two-tailed Student's t test.(e) Control and SEPT7-depleted cells were treated with DMSO or 1 μM STS or 500 μM ETP for 10 hr and labeled with Alexa488-Annexin V (Annexin V) and Hoechst.Representative widefield microscopy images showing Annexin V in green and Hoechst in blue.Scale bar = 100 μm.(f) The % of Annexin V positive cells is calculated in Fiji by dividing the number of cells that displayed Annexin V labeling by the total number of cells identified by nuclear Hoechst labeling, then multiplying by 100.Each bar represents the mean % ± SEM from three independent experiments (a minimum of 1,000 cells were counted per condition separated in three independent experiments).****p < .0001bytwo-way ANOVA.
(a) Representative confocal microscopy images showing cytochrome c in green, Tom20 in red and DAPI in blue.Arrowheads indicate cells releasing cytochrome c.Scale bar = 10 μm.(b) The percentage of cells releasing cytochrome c is calculated in ImageJ by counting cells that exhibited cytosolic (rather than mitochondrial) cytochrome c labeling, and dividing by the total number of cells, then multiplying by 100.Each bar represents the mean % ± SEM from three independent experiments (a minimum of 1,000 cells were counted per condition separated in three independent experiments).****p < .0001by two-way ANOVA.(c) SEPT7 and ARTS act in different pathways.(i) HeLa cells were treated with control or SEPT7 siRNA for 72 hr.Whole-cell lysates were immunoblotted for SEPT7, SEPT2, SEPT9, and ARTS.GAPDH was used as a loading control.Data are representative Western blots of three independent experiments.(ii) Densitometry analysis of the bands was performed using ImageJ.Data are mean ± SEM from three independent experiments.**p < .01 by two-tailed Student's t test ID#14614), SEPT4 (ID#142770), SEPT7 (ID#10323) or SEPT9 (ID#18228) are all from Ambion.Cells were treated with 1 μM staurosporine (STS, #ALX-380-014-C250, Enzo) or 500 μM etoposide (ETP, #E1383-25MG, Sigma) for 10 hr.Equivalent volumes of Dimethyl sulfoxide (DMSO) were used as controls.
Viability of HeLa cells (Figures 1b and S1B) was measured by MTT (3-[4,5-deimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) Initiator and executioner caspase cleavage is significantly reduced when septins are depleted.(a), (b) HeLa cells were transfected with control or SEPT7 siRNA for 72 hr.Cells were then treated with DMSO or (a) 1 μM STS or (b) 500 μM ETP for 10 hr.Whole-cell lysates were immunoblotted against caspase-3, cleaved caspase-3, caspase-7, caspase-9, and GAPDH (used as loading control).Data are representative of three independent experiments.(c) SEPT2 is required for caspase activation.HeLa cells were transfected with control or SEPT2 siRNA for 72 hr.Cells were then treated with DMSO or 1 μM STS for 10 hr.Whole-cell lysates were immunoblotted with antibodies to caspase-7, caspase-9, and GAPDH.Data are representative of three independent experiments.(d) HeLa cells were transfected with control or SEPT7 siRNA for 72 hr.Cells were then treated with DMSO or 1 μM STS for 10 hr.Whole-cell lysates were immunoblotted with antibodies to caspase-4, caspase-7, caspase-9, and GAPDH.Data are representative of three independent experiments (#177445, Thermo Scientific) and transfected with siRNA 72 hr prior to DMSO, STS or ETP treatments.Alexa488-Annexin V (#A13201, Invitrogen), and Hoechst(#62249, Thermo Scientific)    were added directly to the media and incubated for 15 min at 37 C in 5% CO 2 .All live imaging was performed at 37 C in 5% CO 2 and captured using a 10x lens on a Zeiss Axio Observer Z1 widefield epifluorescence microscope driven by ZEN Blue 2.3 software (Carl Zeiss).
U R E 4 SEPT7 is required for the activation of caspase-3.(a) Control and SEPT7-depleted cells were treated with DMSO or STS, and stained with cleaved anti-caspase-3 antibody and DAPI.Representative confocal microscopy images showing cleaved caspase-3 in red and DAPI in blue.Scale bar = 100 μm.(b) The percentage of cleaved caspase-3-stained cells was calculated in Fiji by dividing the number of cells that displayed cleaved caspase-3 staining by the total number of cells, then multiplying by 100.Each bar represents the mean % ± SEM from three independent experiments (a minimum of 1,000 cells were counted per condition separated in three independent experiments).**p < .01,****p < .0001by two-way ANOVA.(c) HeLa cells (CTRL or SEPT7 siRNA transfected) were treated with 1 μM staurosporine for 5 hr, followed by staining of active caspase-3 and analyzed by flow cytometry (a minimum of cells were counted per condition separated in three independent experiments).*p < 0.05, ***p < 0.001 by two-way ANOVA