Fig 1.
Sem1 is required for the maturation of cleistothecia in A. nidulans.
(A) The Δsem1 strain is delayed in formation of green asexual spores on minimal media in light in comparison to A. nidulans wildtype strain (sem1) or the complementation strain (Δsem1::sem1-gfp) with a functional Sem1-GFP. (B) Δsem1 grows slower compared to sem1 or complementation strains. Upper panel- the respective strains were point-inoculated with 10,000 spores and incubated in light for the indicated time points. Lower panel- the colony diameter was plotted. The wildtype grows 1.3±0.03 cm per day and Δsem1 grows 0.81±0.03 cm per day. Significance of differences was calculated with t-test compared to sem1, **p<0.01, n = 4. (C) Deletion of sem1 impairs asexual spore formation. Spores were visible in all strains only after 2 days of asexual growth. Wildtype and complementation produced ≈ 150x106 spores/ml, this number remained similar over the time course. Increased conidia were measured in the deletion strain over time, with maxima of ≈ 94x106 spores/ml after 5 days of asexual growth (64% compared to wildtype). For the quantification of asexual spore, plates were inoculated with 10,000 spores and harvested at the indicated time points. The values shown are the average ± SD from 2 independent experiments. Significance of differences was calculated with t-test compared to sem1, *p<0.05, **p<0.01. (D) Δsem1 mutant strain shows delayed formation of conidiophores (red circles). All asexual growth impairments were restored in the complementation strain. (E) Sem1 is required for maturation of sexual fruiting bodies (cleistothecia) in darkness. (F) Δsem1 mutant strain is block in early sexual development and develop only Hülle cells and primordia (p) after 10 days of sexual growth. Wildtype and complementation strains produced pigmented cleistothecia (Cl) after 7 days. Cleistothecia were squeezed to check for the presence of ascospores. Scale bars: 50 μm for full size images, 20 μm for squeezed cleistothecia and 200 μm for the panel comparing the size of the sexual structures in the respective strains. (G) Similar numbers of cleistothecia per cm2 were observed in wildtype or complementation strains. The average size of the cleistothecia from Δsem1 was 5850±823 μm2. 15% of cleistothecia from wildtype and complementation had similar size and contained ascospores (Fig 1F, squeezed panel). Columns represent average number of cleistothecia ± SD per cm2, n = 2.
Fig 2.
Accumulation of 20S proteasome complexes in Δsem1.
(A) Electron micrographs of negatively stained proteasome complexes derived from fungal cell extracts. The total numbers of proteasome complexes and the respective complexes observed (%) in Δsem1, wildtype (sem1) or complemented (Δsem1::sem1-gfp) strains are indicated. (B) Δsem1 proteasomes showed similar peptidase activities regardless to the presence or absence of ATP and KCl. Activities were measured in assay buffer containing 0.125mM ATP, 10mM KCl and 2.5μM of the proteasome inhibitor MG132. The assay buffer also contained 12.5 mM Tris HCl pH = 7.5 +1.25 mM MgCl2 +0.25 mM DTT+0.0125 mg/ml BSA. Blue, red and green curves represent peptidase activity of the indicated strains in the presence of ATP+KCl (26S activity), in the absence of ATP and KCl (-ATP–KCl, 20S) and in the presence of MG132 (proteasome inhibitor).
Fig 3.
Deletion of sem1 promotes decrease in the total ubiquitin-conjugates and destabilization of the Rpn11 deubiquitinating enzyme.
(A) Significant decrease in polyubiquitylated substrates upon deleting sem1. Polyubiqutinated substrates were detected with α-ubiquitin and α-GAPDH served as loading control. T-test of Δsem1 vs sem1, ***p<0.001 and sem1 vs Δsem1::sem1-gfp, *p<0.05, n = 2. (B) Transcription levels of 19S RP subunits and genes involved in maintaining intracellular ubiquitin pools. RT-PCR results are shown as relative expression compared to sem1, n = 4, ***p<0.001. (C) Reduced amount of CulA/Cul1 protein in Δsem1. Neddylated substrates were detected with α-CulA (n = 3) and α-Nedd8 (n = 2), respectively, *p<0.05, **p<0.01. (D) Stability of Rpn10-GFP and Rpn11-GFP proteins in Δsem1. Western blots were probed with α-GFP to determine the relative amounts of full-length and truncated (37KDa) proteins. Mean intensities from three biological replicates were normalized to GAPDH (see also S2 and S3 Figs).
Fig 4.
Functional GFP-tagged 19S RP subunits consist of a prominent nuclear and a smaller cytoplasmic subpopulation.
(A) Indicated 19S RP subunits fused to GFP are functional and support asexual development. (B) Subcellular localization of GFP-tagged 19S RP strains determined by fluorescence microscopy. OE-overexprssion strain. The images show green GFP fluorescence (left) for 19S RP subunits, MitoTracker Red (second from left) for mitochondria, DAPI for nuclei (third from the left), and an overlay (most right). The fluorescence intensities in the cytoplasm and nucleus for the indicated GFP strains are shown as mean fluorescence intensity. Significance of differences was calculated with t-test compared to sem1-gfp, ***p<0.001.
Fig 5.
Proteins co-identified with Sem1 and 19S RP in the presence or absence of Sem1.
(A) 34 proteins associated with 19S RP proteins fused to GFP in strains carrying Sem1. Proteins were identified in three biological replicates plotted as heat map representing LFQ intensities. HypoP1 and HypoP2 are conserved hypothetical proteins, with orthologs only among Aspergillus-related species. HypoP1 (AN2234) has no known conserved domain, whereas HypoP2 (AN4931) contains an Acyl-coenzyme A synthetase/AMP-(fatty) acid ligase domain (accession number COGO365). Lower panel: schematic representation of 30 proteins plotted in a heat map. Nas6, Hsp90, Uch24 and HypoP1 are not represented. (B) 33 proteins associated with 19S RP strains lacking sem1 (marked withΔ) identified in two biological replicates plotted as heat map representing LFQ intensities. Lower panel: representation of 20 proteins identified with all RP. 9 proteins are not represented including Nas6, Hsp90, Mpp and 6 mitochondria related proteins. Proteins identified with only one tagged RP protein and/or with MSMS counts ≥5 were not represented (total 4). Heat maps representing MS/MS counts were plotted in S5 Fig. Overexpressing GFP strain served as control. In the area of the heat map were LFQ intensities and MS/MS counts are low, proteins were considered identified only when both criteria were fulfilled: LFQ>22, MS/MS counts >4. (see also S4 and S5 Figs and S1 and S2 Tables).
Fig 6.
Sem1 interacts in vivo with Rpn10 and Rpn11.
(A) Schematic model of the orientation of N- and C-termini of Sem1 (yellow) within the lid associated to the base (light blue) and the 20 core particle of the 26S proteasome. The red circle represents the last N-terminal amino acid residue modelled in Sem1. Zoomed in positions of the C-terminal region of human (Dss1 in magenta) and yeast (yellow) Sem1 are depicted in frame. (B) Bimolecular fluorescence complementation studies (BiFC) of Sem1 fusion proteins with the C-terminal half of yellow fluorescent protein co-expressed with either Rpn10 or Rpn11 fusions to N-terminal YFP result in specific signals corroborating physical interaction in fungal cells (A. nidulans strains: sem1-yfpc+rpn10-yfpn or sem1-yfpc+rpn11-yfpn). Nuclei were visualized in red by expression of rfp-h2A. The fluorescence intensities both in the cytoplasm and nucleus for the indicated strains are shown as mean fluorescence intensity. Significance of differences was calculated with t-test compared to the control strains expressing either sem1-yfpc, rpn10-yfpn or rpn11-yfpn, ***p<0.001, n = 10. Significantly higher fluorescence was observed in sem1-yfpc+rpn10-yfpn or sem1-yfpc+rpn11-yfpn compared to the control strains. (C) Homology model of the 26S proteasome from A. nidulans based on the human proteasome cryoEM structure (EMDB-4002, PDBs: 5L4K and 5L46). The model depicts two surface representations of the 19S RP with C-terminal region of Sem1 located at the interface formed by the N-terminal domain of Rpn3 and Rpn7 (yellow surface). The model of Sem1 is missing the N-terminally fragment of approximately 30 residues. 120° rotation depicts the last N-terminal amino acid residue modelled, located in a cleft formed by Rpn3 (red circle). It is conceivable that the missing N-terminal region of Sem1 can pass through this cleft and form interactions with Rpn10 and Rpn11 within the lid (dashed yellow line).
Fig 7.
Sem1 is required for an appropriate oxidative stress response in A. nidulans.
(A) 19S RP without Sem1 associates with TCA cycle and respiratory chain related proteins. Interacting enzymes are marked with orange circles and the respective 19S regulatory particle subunits associated with them are indicated with diamonds. (B) Mitochondria morphology is compromised in Δsem1 mutant strain as observed by time-lapse microscopy with the fluorescence marker MitoTracker. Hypha from wildtype, complementation and deletion strain were observed 5 min, 15 min and 35 min after the addition of the MitoTracker. Scale bar: 10 μm. (C) The total cellular NADH production is reduced in the Δsem1 mutant strain. n = 5, ***p<0.001. (D) Oxidative stress inducing compounds inhibit Δsem1 colony growth. Respective strains were spotted on minimal medium plates (MM, control) and MM plates supplemented as indicated, n = 4. Two complementation strains served as internal biological replications. (E) Δsem1 mutant strain induced the transcription of the antioxidants encoding genes catA and sodB. Results are shown as relative expression compared to sem1, n = 3, ***p<0.001. (F) sem1 and rpn11 are induced as response to oxidative stress in sem1+1.27mM H2O2. Bars represent mean values of four independent experiments, ***p<0.001 (see also S6 Fig).
Fig 8.
Molecular Sem1 function and A. nidulans multicellular development.
Sem1 links cellular redox state, mitochondria integrity, efficient assembly of several complexes including the 26S proteasome to multicellular differentiation. Oxidative stress induces the oxidative stress response. Multicellular development requires an internal reactive oxygen species (ROS) as a signal as well as protection against oxidative stress. Oxidative stress induced transcription of genes for proteasome subunits sem1 and rpn11 (indicated with +). Capped proteasomes, which can be formed in the presence or absence of Sem1, differ in composition and quantity. Sufficient amounts of Sem1 and Rpn11 are required during oxidative stress to provide correctly assembled 26S proteasomes including the ubiquitin receptor Rpn10, the tethering factor Ecm29 and stabilized full-length deubiquitinating enzyme Rpn11. Consequently, these stable capped 26S proteasomes are fully functional and can selectively catalyse ubiquitinated proteins (indicated as diamonds) in an ATP/ubiquitin-dependent manner. Decreased amounts of Sem1 result in damaged mitochondria, unstable capped 26S proteasomes and increased levels of activated oxidation-driven 20S core proteasomes, which can efficiently hydrolyse proteins in an ATP/ubiquitin-independent manner.