The STRIPAK component SipC is involved in morphology and cell-fate determination in the nematode-trapping fungus Duddingtonia flagrans

Abstract The striatin-interacting phosphatase and kinase (STRIPAK) complex is a highly conserved eukaryotic signaling hub involved in the regulation of many cellular processes. In filamentous fungi, STRIPAK controls multicellular development, hyphal fusion, septation, and pathogenicity. In this study, we analyzed the role of the STRIPAK complex in the nematode-trapping fungus Duddingtonia flagrans which forms three-dimensional, adhesive trapping networks to capture Caenorhabditis elegans. Trap networks consist of several hyphal loops which are morphologically and functionally different from vegetative hyphae. We show that lack of the STRIPAK component SipC (STRIP1/2/HAM-2/PRO22) results in incomplete loop formation and column-like trap structures with elongated compartments. The misshapen or incomplete traps lost their trap identity and continued growth as vegetative hyphae. The same effect was observed in the presence of the actin cytoskeleton drug cytochalasin A. These results could suggest a link between actin and STRIPAK complex functions.


Introduction
Polar growth, polarity establishment, and maintenance of polarity of cells are common themes in biology and largely depend on cytoskeletal elements and functions. A bona fide example of polar growth is filamentous fungi with their extremely elongated hyphae (Riquelme et al. 2018). Most fungal species grow apically by continuous cell wall and cell membrane extension. Secretion vesicles containing, e.g., cell wall biosynthesis enzymes are transported along microtubules (MTs) toward the tip, accumulate transiently in a so-called vesicle supply center or Spitzenkö rper from where they are transported toward the apical membrane. After fusion with the membrane, they deliver their protein content (Riquelme et al. 2007). For the last travel distance, the actin cytoskeleton is required (Bergs et al. 2016;Schultzhaus et al. 2016). The MT and the actin cytoskeletons are coordinated through the action of so-called cell-end marker proteins at the cortex (Mata and Nurse 1997;Takeshita et al. 2013Takeshita et al. , 2014Takeshita et al. , 2017. Any variation from the tube-like morphology of hyphae is probably preceded and accompanied by cytoskeletal rearrangements. One intriguing fungal hyphal morphology is the ring-like structure of traps of nematode-trapping fungi (NTF) . These fungi can switch from a saprotrophic to a predatory lifestyle under low-nutrient conditions and in the presence of nematodes. More than 200 NTF have been identified and many are found in the Orbiliaceae (Ascomycota). A wide range of morphologically different traps are produced by different species, ranging from adhesive knobs and columns, adhesive trap rings and networks, and constricting and non-constricting rings (Su et al. 2017). One of the best-studied NTF at the molecular level is Arthrobotrys oligospora, where also the signal exchange between fungus and nematodes and the control of trap formation were studied (Niu and Zhang 2011;Hsueh et al. 2013Hsueh et al. , 2017Yu et al. 2021). In the case of Duddingtonia flagrans, a close relative of A. oligospora, ring-like, adhesive traps are formed, in which Caenorhabditis elegans or other nematodes are captured (Youssar et al. 2019). The ring-like morphology of the traps is rather unique and requires a redirection of the polarity axis, cell-cell recognition, and hyphal fusion. This morphological differentiation goes along with physiological differentiation, the production of adhesives, and the preparation for the penetration of caught nematodes. Another fascinating aspect is the determination of the size of the traps. They should have a defined diameter to effectively catch nematodes. If they are too large, nematodes could slip through, and if the diameter is too small, nematodes will not be able to enter the traps. Size determination probably also requires a functional cytoskeleton. How these fascinating morphological changes are initiated and which signaling modules are required are largely unknown.
Although some trapping-deficient mutants have been identified in A. oligospora, the molecular basis and cell biology of trap morphogenesis are poorly understood (Kuo et al. 2020;Chen et al. 2021). Recently, a random mutagenesis analysis identified 15 mutants with deficiencies in trap morphogenesis. However, there are still difficulties in gene mapping because of heterokaryotic transformants and the lack of genetic crosses of NTF (Huang et al. 2021). In D. flagrans, it was shown that a cell dialog between the tip trap cell and the basal cell is required for ring closure (Youssar et al. 2019).
The early stages of trap formation in D. flagrans resemble the spiral growth of ascogonia or crozier formation of ascomycetes during sexual development (Read and Beckett 1996). An ascogenous hypha elongates and bends over to form a hook-shaped structure. In Sordaria macrospora, a genetic screen has identified several "pro" mutants with defects in sexual development, and some PRO proteins are components of the striatin-interacting phosphatase and kinase (STRIPAK) complex (Stein et al. 2020;Teichert et al. 2020). The STRIPAK complex is a highly conserved eukaryotic signaling module responsible for the regulation of numerous cellular and developmental processes. It was initially identified in mammals, where it was linked to human diseases like Alzheimer's, diabetes, and cardiac diseases (Goudreault et al. 2009;Hwang and Pallas 2014). Recently, the structure of the human STRIPAK core was solved by cryo-EM (Jeong et al. 2021). The serine-threonine protein phosphatase 2 (PP2A) core enzyme consists of a scaffold subunit A (PP2A), the catalytic subunit C (PP2AC), and a regulatory subunit of the B 000 (striatin) family. In addition, multiple subunits are associated with the core subunits of the phosphatase, including the striatin-interacting protein 1 or 2 (STRIP1 or STRIP2), the monopolar spindle one-binder (Mob) protein Mob3, cerebral cavernous malformation 3, the sarcolemmal membrane-associated protein (SLMAP), and the coiled-coil protein suppressor of IjB kinase-e. Furthermore, various germinal center kinases (GCKs) are associated with the STRIPAK complex for enhanced regulation.
The STRIPAK complex is involved in the regulation of key cellular processes including signaling, cytoskeletal organization, cell cycle control, cell migration, cell polarity, apoptosis, Golgi assembly, cell morphology, and vesicular trafficking (Hwang and Pallas 2014;Kü ck et al. 2019). In Drosophila melanogaster, it was shown that the STRIPAK complex is a negative regulator of the Hippo signaling pathway through its phosphatase activity (Ribeiro et al. 2010). RNAi screening identified the STRIP1/2 orthologs FAM40A and FAM40B to affect cell shape in Drosophila S2 cells . In an RNAi screening in PC3 prostate cancer cells, FAM40A-depleted cells were flattened with high levels of cortical F-actin compared to the control, whereas FAM40B-depleted cells were elongated with long thin protrusions (Bai et al. 2011). In Saccharomyces cerevisiae, the STRIPAK homolog Factor arrest protein complex (Far complex) is involved in pheromone-induced cell cycle arrest and affects signaling through the target of rapamycin complex 2 (Kemp and Sprague 2003;Pracheil et al. 2012). In Schizosaccharomyces pombe, the STRIPAK homologous Septation initiation network (SIN) Inhibitory PP2A complex (SIP) negatively regulates SIN and is important for coordinating mitosis and cytokinesis (Singh et al. 2011;Simanis 2015).
Functional analyses in filamentous fungi revealed that different components are required for cell-cell fusion, sexual fruiting body development, septation, secondary metabolite production, symbiotic interactions, and pathogenicity (Xiang et al. 2002;Pö ggeler and Kü ck 2004;Green et al. 2016;Zhang et al. 2018;Elramli et al. 2019). But only recent proteomic and phosphoproteomic studies in S. macrospora revealed an extended signaling network that is modulated by the STRIPAK complex (Mä rker et al. 2020).
Here, we studied the STRIPAK complex in the NTF D. flagrans, investigated the role of the STRIPAK complex component SipC/ STRIP1/2 during septation and the unique hyphal ring formation during trap morphogenesis, and found that SipC is involved in cell-fate determination.

Material and methods
Strains and culture conditions D. flagrans strains were cultivated at 28 C on potato dextrose agar (PDA; Carl Roth). All fungal strains used in the study are listed in Table 1.
To induce trap formation, the fungal strains were cultivated on thin low-nutrient agar (LNA) slides by adding around 1Â 10 4 D. flagrans spores and around 100-500 C. elegans individuals. Co-incubation was carried out at 28 C in darkness for 12-48 h. C. elegans strains were cultivated at 20 C on nematode-growth medium seeded with Escherichia coli strain OP50 as food source.
Targeted gene deletion of sipC sipC was deleted by homologous recombination. One kb flanks homologous to the 5' and 3' regions of sipC were amplified by PCR with primers with 25 bp overhangs homologous to the hygromycin-B resistance cassette as well as to the pJET1.2 plasmid backbone (Thermo Fisher), digested with EcoRV (Tables 2 and  3). The hygromycin-B resistance cassette hph was amplified using plasmid pFC332 as template, and all fragments were assembled into pJET1.2 using the NEBuilder HiFI DNA Assembly Cloning Kit [New England Biolabs (NEB), Frankfurt]. The fragments were amplified using Q5 polymerase (NEB) and the manufacturer's recommended protocol. Standard protocols were used for E. coli transformation and plasmid isolation. The gene deletion construct was introduced into D. flagrans protoplasts as described in Youssar et al. (2019). For selection of transformants, PDA plates supplemented with 100 mg/ml hygromycin-B were used. To reintroduce the full-length target gene into the deletion strains, the whole gene including its 1 kb upstream and 0.5 kb downstream regulatory regions was amplified by PCR. To select for positive

Verification of homologous recombination
To extract genomic DNA from D. flagrans, around 1Â 10 8 spores were inoculated in potato dextrose broth and incubated in a 6-cm diameter Petri dish at 28 C for 48 h. Afterward, the mycelium was collected, and DNA was extracted using the protocol described in Zhou et al. (2018). Southern blot analysis was done with digoxigenin (DIG)-labeled probes. Genomic DNA was digested with the appropriate restriction enzyme overnight, separated in a TAE agarose gel and transferred to a nylon membrane. Probes were synthesized using the PCR DIG Probe Synthesis Kit (Roche) according to the manufacturer's protocol. After probe hybridization, the membrane was hybridized with anti-DIG-AP Fab fragments (Roche) and development was performed using CDP-Star solution (Roche).

Localization of proteins with GFP and mCherry
mCherry-TubA: TubA was tagged with mCherry at the N-terminus expressed from the natural promoter. Using Gibson assembly, PCR-amplified fragments of the hygromycin resistance cassette hph, 1 kb tubA promoter, GFP, and tubA open reading frame plus 1 kb downstream regulatory region were cloned into a pJET1.2 plasmid backbone. To express the construct in D. flagrans mutant strains, the tubA(p)::mCherry::tubA::tubA(t) fragment was cloned into pJM16 containing the G418 resistance cassette. Plasmids and oligonucleotides are listed in Tables 1  and 2.

Transcription reporter assay
To localize the expression of sipC in D. flagrans during different developmental stages, the sipC promoter was fused to the fluorescent reporter gene construct h2b-mCherry. The promoter fragment was amplified by PCR and assembled into the plasmid backbone pVW56 containing the h2b::mCherry reporter and the hygromycin-B resistance cassette by Gibson assembly. The plasmid was transformed into D. flagrans sVW10 expressing h2b(p)::h2b::GFP.

Quantification of compartment length, conidiation, and trap morphogenesis
To quantify compartment length of vegetative mycelium, around 1000 spores of the respective D. flagrans strain were inoculated on LNA at 28 C for 24 h. The cell wall was stained with Calcofluor White and images were taken at 100Â magnification. For each strain, the length of ten compartments of five germinated spores was quantified and performed in three replicates.
To quantify the amount of conidia produced by the different D. flagrans strains, 8-mm diameter hyphal discs were punched from the edges of a 7-day-old cultured PDA plate, transferred into 500 ml dH 2 O, and vortexed for 5 min. Conidia were quantified using a Helber counting chamber. Each biological replicate was quantified four times. To quantify chlamydospore quantities and morphology in the DsipC strain, 1000 spores of wild type (WT), DsipC, and rescue strain were incubated on LNA at 28 C. After 72 h, chlamydospore number and morphology were assessed. To quantify the morphology of fungal traps, around 1000 spores of the respective D. flagrans strains were co-inoculated with around 200 C. elegans larvae on thin LNA pads (2Â 2 cm) at 28 C for 48 h. The traps were counted and quantified using 100Â and 400Â magnification. For each strain, three biological replicates were quantified.
For confocal imaging, a Zeiss LSM900 microscope with Plan Apochromat 63Â/1.4 Oil objective and Airyscan 2 detector in super resolution mode was used.

Fluorescent dyes and inhibitors
To visualize the cell wall of the fungus, the mycelium was stained using Calcofluor-white (CFW, fluorescent brightener 28; Sigma-Aldrich). The 1% (wt/vol) stock solution was prepared by dissolving CFW in 0.5% (wt/vol) KOH and 83% (vol/vol) glycerol. A 1:1000 dilution working solution (in dH 2 O) was added to the mycelium and incubated for 3 min to stain the cell wall before imaging.
A stock solution (1 mg/ml) of the anti-MT agent methyl 1-(butylcarbamoyl)-2-benzimidazolecarbamate (benomyl; Sigma) was prepared in dimethyl sulfoxide (DMSO) and used at a final working concentration of 5 mg/ml in dH 2 O. A stock solution (1 mg/ ml) of the F-actin depolymerizing drug Cytochalasin A was prepared in DMSO and used at a final working concentration of 5 mg/ ml in dH 2 O. For the inhibitors, a 0.5Â 0.5-cm agar pad with mycelium was incubated in 100 ml working solution of each inhibitor and incubated for up to 60 min prior to imaging.

Statistical analysis
P-values were calculated using the mean of three biological replicates and the t-test with the software GraphPad Prism 9. In the bar graphs, each biological replicate is color-coded (orange, blue, gray). While each data point is depicted as dot, the averages are shown as triangles (Lord et al. 2020).

Conservation of the STRIPAK signaling complex in Duddingtonia flagrans
To unravel a role of the STRIPAK complex in trap formation, we identified orthologs of S. macrospora PRO11, PP2AA, PP2Ac1, PRO22, SmMOB3, PRO45, SmKIN3, SmKIN24, and SCI1 in D. flagrans ( Figure 1). They all share conserved domains with homologs from other fungi, suggesting that D. flagrans harbors a functional STRIPAK complex. In A. nidulans, the STRIPAK complex consists of the striatin scaffold protein StrA and six striatin (StrA) interacting proteins (Sip), and the heptameric complex is involved in the regulation of light-dependent development, the production of secondary metabolites, and the stress response (Elramli et al. . Two putative STRIPAK-associated GCKs DFL_002427 and DFL_007266 were identified. No ortholog for SipB was found.

STRIPAK sipC expression
As a first step to characterize the functions of the STRIPAK complex in D. flagrans, we analyzed the sipC gene. The open reading frame comprises 3528 nucleotides and is disrupted by nine introns [confirmed by RNAseq data (Youssar et al. 2019)]. To test if the expression levels change in different hyphae, a microscopic reporter assay was used. To this end, we fused the sipC-promoter to the fluorescent reporter gene mCherry which was further linked to the histone h2b gene (h2b), resulting in red fluorescent nuclei when the promoter is active (Yu et al. 2021). As a control, the fluorescent reporter gene h2b-GFP was expressed with the h2b-promoter. The overall expression of sipC was weak. No fluorescence was observed in early stages, 4 h after germination of conidiospores, or in the first 10-30 mm of the hyphae ( Figure 1B). Fluorescence started to appear in nuclei when the first septum or hyphal branches became visible. Higher levels of expression were observed in older parts of the mycelium (16 h incubation). During trap formation, no significant changes in expression were detected, and the fluorescence of the reporter construct resembled the intensity of later-stage mycelium. These results suggest a function after germ-tube formation during hyphal differentiation.

sipC is required for fast vegetative growth and septation
Since sipC was expressed during septum formation, we next generated a sipC-deletion strain by replacing the gene with a hygromycin-B resistance cassette by homologous recombination. Twenty-nine hygromycin-B resistant transformants were obtained and screened by PCR and Southern blot analysis. One transformant showed the expected integration and was used for further studies ( Figure 3A).
Deletion of sipC resulted in slow-growing colonies. While the WT completely covered the medium on a 9-cm Petri dish in 7 days when center inoculated, the DsipC mutant needed around 21 days ( Figure 3B). The formation of aerial mycelium is a common feature of D. flagrans on PDA medium. In the DsipC-deletion strain, it was restricted to the first 1-2 cm of the initial inoculation site. Additionally, the DsipC mutant showed a 1-cm ring-like growth pattern on PDA. The mutant phenotype could be fully re-complemented with an ectopic copy of the sipC gene, including 1 kb upstream and 0.5 kb downstream regulatory elements.
Since we observed the start of expression of sipC during septum formation, we quantified the number of septa in the DsipCdeletion strain after inoculating spores on a low-nutrient medium and incubation for 24 h ( Figure 3C). Generally, growth of the DsipC strain was slower in comparison to the WT and the respective re-complementation strain. While hyphal compartments in WT and the re-complemented mutant strain had a length of 48.5 6 5 mm [mean standard deviation (SD)] and 42.6 6 0.3 mm, respectively (n ¼ 150), the compartment size was doubled in the DsipC-deletion strain with a length of 102 6 5 mm. The differences were still present after an additional 24 h of incubation, resulting in decreased numbers of septa compared to WT. In the DsipC-deletion strain, septa were mainly found at branching points. Interestingly, whereas the Dpro22 mutant in S. macrospora and Dham-2 mutant in Neurospora crassa failed to undergo cell-cell fusion (Xiang et al. 2002;Bloemendal et al. 2010), in the D. flagrans DsipC-deletion strain, the process appeared unaffected ( Figure 3D). The frequency of hyphal fusion in the DsipC-deletion strain was difficult to compare to WT, since growth speed was heavily reduced.

SipC is required for asexual development and determines chlamydospore morphology
Deletion of sipC heavily influenced the number of asexual conidia production, resulting in an 80% reduction of conidia of 7-day-old cultures compared to WT ( Figure 4A). Likewise, the amount of chlamydospores produced after 72 h on LNA was reduced by 88% compared to WT. In addition, the morphology of chlamydospores was drastically altered, with 71 6 3% highly elongated and expanded spores in the mutant (n ¼ 3) (Figure 4, C-E). All chlamydospores of the WT had a round shape with an average diameter of 22.7 6 0.5 mm compared to the chlamydospores of the DsipC-deletion strain which had a length of 33.9 6 0.7 mm. Yet, these spores were still able to germinate. Chlamydospores develop from differentiated vegetative hyphae, while conidia are formed at the tip of conidiophores. Chlamydospores are formed in older mycelium by condensing the cytoplasm within a compartment into a thick-walled spore. This formation is rather difficult in the DsipCdeletion strain since the compartment size is much larger than in WT.

SipC determines the fate of trap cells
To investigate trap morphogenesis of the DsipC-deletion strain, C. elegans was co-incubated with fungal spores on LNA slides for at least 48 h. While trap formation in WT and the re-complemented strain showed ring-like trap networks, this was heavily disturbed in the DsipC-deletion mutant ( Figure 5A). Trap loops were often incomplete and resulted in column-like structures. These column-like traps were either identified and distinguished from normal vegetative hyphae by comparing the hyphal diameter or by an increased fluorescence after CFW staining ( Figure 5B). An already formed trap can be a new starting point for another trap resulting in more complex trapping networks. Compared to WT, this phenomenon was restricted in the DsipC mutant ( Figure 5C). Trap morphology was restored in the re-complemented strain ( Figure 5D). The number of nuclei in trap cells in WT and the DsipC-deletion mutant strain showed no apparent difference ( Figure 5E).
Traps of the DsipC-deletion strain were divided into four groups according to the morphology of the trap, and the occurrence was quantified ( Figure 6A). Of all counted traps, 77 6 2% showed a column-like structure (I), 13 6 3% a ring-like structure without cell fusion (II), 8 6 1% a 90 degrees bending (III), and 1 6 1% a ring with cell fusion (IV) appearance (total trap number in three biological replicates were n ¼ 224, n ¼ 156, and n ¼ 276, respectively; Figure 6B). In contrast, mostly traps of group II and IV occurred in WT and the re-complemented strain. Despite the morphological differences, traps of the DsipC-deletion strain were still adhesive, and the strain was able to immobilize and digest nematodes ( Figure 7A).  Traps in WT usually consist of two to four compartments before fusing to the neighboring basal hypha. In contrast, compartments of the traps in the DsipC-deletion strain were highly elongated as in vegetative mycelium. Interestingly, the elongated trap cells often switched to a vegetative growth mode after one, or a maximum of three trap compartments, indicated by a thinner diameter of the hyphae. The change of cell fate from trap cells to normal vegetative hyphae was confirmed by expressing the ortholog of the trap-specific serine protease P12 (DFL_008096) tagged with mCherry in the WT and the DsipC mutant background. Serine proteases have been identified as crucial virulence factors upon exposure to nematodes, and the A. oligospora P12 was highly upregulated during nematode exposure (Yang et al. 2011;Zhen et al. 2018). P12-mCherry localized in dynamic spots inside the traps of the WT and the DsipC mutant ( Figure 7B). The change of cell fate from trap cells to normal vegetative hyphae in traps of the DsipC-deletion strain was indicated by the absence of fluorescent accumulations during trap morphogenesis. This change also coincided with a change in hyphal diameter ( Figure 7C). Additionally, we observed dynamic spots of P12-mCherry with weaker intensities in neighboring compartments. Therefore, we visualized the promoter activity of p12 as described above by fusing the p12 promoter to the fluorescent reporter construct of H2B-mCherry and transformed the plasmid into a strain expressing H2B-GFP under the control of the h2b promoter. Promoter activity was also present in neighboring cells but decreased in more distant vegetative hyphae ( Figure 7D). In the WT, the nuclei were labeled by H2B-mCherry (colored in yellow). In the DsipC mutant, the nuclei were labeled by H2B-GFP (colored in green). The cell wall was stained by CFW (colored in magenta). 3D reconstruction.
Since the morphology of the traps was drastically changed in the DsipC-deletion strain, we hypothesized that the deletion of sipC influences the dynamics of the cytoskeleton. The cytoskeleton has never been visualized in living traps of NTF. Therefore, we visualized MTs and F-actin by tagging the D. flagrans alphatubulin TubA with mCherry and the actin-binding peptide LifeAct with GFP. In filamentous fungi, the minus ends of MTs locate at spindle pole bodies which serve as MT-organizing centers (MTOC). Additionally, areas close to the septum can act as septal MTOC (sMTOC) (Zekert et al. 2010;Gao et al. 2019). In the tip compartments, we observed dynamic MTs reaching the hyphal tip ( Figure 8A). In trap cells, the most visible MTs nucleated from septal MTOCs ( Figure 8A). Interestingly, major bending of the trap cell was only visible after the first septum at the base ( Figure 8, B and C). Although MTs were dynamic, long MTs were observed at the inner side of the trap, reaching from the septum to the tip. In the DsipC mutant, MT localization appeared similar ( Figure 8D). To study the role of MTs during trap morphogenesis, we added cytoskeleton-destabilizing drugs during trap development. The presence of benomyl (5 mg/ml) led to hyperbranching of the mycelium ( Figure 8E). Hyphal growth during trap development stopped after the addition of benomyl ( Figure 8F). Depolymerization of MTs was confirmed by imaging mCherry-labeled MTs after the addition of benomyl ( Figure 8G).
Next, we visualized actin using the actin-binding peptide LifeAct-GFP (Riedl et al. 2008). Mostly, actin patches at the hyphal tip and subapical regions as well as the actomyosin ring during septum formation were visualized (Figure 9, A and B). Rarely, dynamic actin cables at the apex of hyphae were observed. Expression of Lifeact-GFP in the DsipC mutant revealed that septum formation was delayed during trap morphogenesis, resulting in elongated cells with decreasing hyphal diameter (Figure 9, C and D). The DsipC-deletion strain did not show any difference in the sensitivity toward cytochalasin A as compared to WT (data not shown).
To study the role of actin during trap morphogenesis, we added the F-actin-destabilizing drug cytochalasin A (5 mg/ml) during trap development, which led to swelling of hyphal tips of vegetative mycelium ( Figure 9D). This effect was less obvious in growing traps; however, they continued to grow as vegetative hyphae at the end of the reaction ( Figure 9E; Supplementary Movie S1). These results suggest a putative function of SipC in regulating cytoskeleton dynamics. However, we did not see apparent differences in the actin organization in the DsipC-deletion strain as compared to WT.

Discussion
The STRIPAK complex is involved in many cellular processes in eukaryotes. In filamentous fungi, it is important for hyphal fusion, sexual development, and hyphal growth (Kü ck et al. 2019). In this study, we show that the D. flagrans STRIP1/2 ortholog SipC plays a role in septation, conidiation, and trap morphogenesis  Table 4 for the quantification. The length of four compartments was measured. In the DsipC mutant, quantification of the most abundant groups I (stick-like traps) and II (ring-like traps) are displayed. Each biological replicate is color-coded (orange, blue, gray). While each data point is depicted as dot, the averages are shown as triangles. Error bars show the SD.  Visualization of the expression of p12 during trapping of C. elegans using a H2B-mCherry reporter construct. The expression of H2B-GFP under the h2b promoter was used as control. Arrows indicate transition points of reduced p12-mCherry expression in vegetative hyphae. (Sakuma et al. 2015). It was reported that the Strip-dependent regulation of the GCKs MST3 and MST4 modulates actomyosin contraction and therefore cell motility in cancer cells (Madsen et al. 2015). Our results raise the question of possible links among the STRIPAK signaling complex, the actin cytoskeleton, and cell-fate determination in the traps of D. flagrans.
In filamentous fungi, F-actin forms actin cables, rings, and patches (Berepiki et al. 2011). Actin cables are nucleated by formins and localize at the apex of hyphae, where they are involved in polar growth (Riquelme et al. 2018). Filamentous fungi usually only encode one formin and deletion of the gene in A. nidulans or N. crassa is lethal (Sharpless and Harris 2002;Justa-Schuch et al. 2010). Interestingly, the STRIPAK subunit PRO45 was found as an interaction partner of formin in S. macrospora, although the exact molecular function of the interaction has not yet been solved (Stein et al. 2020). It is conceivable that PRO45 influences the activity of formin. Additionally, a class II myosin was found as an interaction partner of the STRIPAK complex component PRO45 in S. macrospora (Nordzieke et al. 2015). The fact that D. flagrans sipC mutants grow very slow may reflect the role of SipC on the actin cytoskeleton in the apical dome. The effect is clearly different from the effect of cytochalasin, which destroys the actin cytoskeleton. In addition to an effect on polar growth, sipC mutation had a strong effect on septation. Comparable to the effect on hyphal growth, septation is still possible without SipC, but the timing appears to be different. Although this has no obvious effect on vegetative hyphae, the process appears to be crucial for trapcell memory. A similar effect was observed during sexual development of the S. macrospora mutant Dpro22, where aseptated ascogonia showed a developmental arrest at the protoperithecial stage (Bloemendal et al. 2010). This defect resulted in elongated ascogonial coils.
In addition, to a direct effect of SipC on the actomyosin system, there is good evidence in S. pombe that the SIN can be regulated by the STRIPAK complex. It was shown that SIP-mediated dephosphorylation of the Ste20 kinase Cdc7p initiates SIN assembly (Singh et al. 2011). Misregulation of SIP inhibits the assembly of SIN. In A. nidulans, the Cdc7p ortholog SepH is required for initiation of septation prior to actin ring assembly. While downregulation of sepH results in a loss of septation, overexpression results in hyper-septation (Bruno et al. 2001). A similar phenotype was observed in S. macrospora in a Smkin3 mutant. Smkin3 is the ortholog of A. nidulans SepL, and the downstream target of SepH (Radchenko et al. 2018). In this study, the deletion of the STRIPAK-complex component sipC resulted in a drastic reduction of septa and trap compartments of very different sizes.
Taken together, there are multiple links how SipC may affect the actin cytoskeleton and septation. Compartment size appears to be very important for cell-fate determination during trap formation.

Data availability
Strains and plasmids are available upon request. The authors affirm that all data necessary for confirming the conclusions of the article are present within the article, figures, and tables. Supplementary material is available at figshare: https://doi.org/ 10.25386/genetics.16617721.  The addition of the actin-depolymerizing drug cytochalasin A (5 mg/ml) leads to swollen hyphal tips. As solvent control, hyphae were treated with 0.5% DMSO. (F) Loss of trap identity after the addition of cytochalasin A (5 mg/ml). After some time, traps continued to grow as vegetative hyphae (see also Supplementary Movie S1).