The proteasome activity reporter GFP-Cl1 is degraded by autophagy in the aging model Podospora anserina

P. anserina strains and cultivation

P. anserina was grown on plates with M2 medium (0.25 g/l KH₂PO₄ (Merck Cat# 5099.1000), 0.3 g/l K₂HPO₄ (Roth Cat# P749.1), 0.25 g/l MgSO₄ × 7 H₂O (Merck Cat# 1.05886.0500), 0.5 g/l urea (Merck Cat# 1.08487.0500) and 10 g/l yellow dextrin (Roth Cat# 6777.1), supplemented with 2.5 mg/l biotin (Serva Cat# 15060), 50 mg/l thiamine (Serva Cat# 36020), 5 mg/l citric acid × 1 H₂O (Sigma-Aldrich Cat# C-0759), 5 mg/l ZnSO₄ × 7 H₂O (Merck Cat# Z-0625), 1 mg/l Fe(NH₄)₂(SO₄)₂ × 6 H₂O (Merck Cat# 1.03861.0250), 2.5 mg/l CuSO₄ × 5 H₂O (Merck Cat# 2790.1000), 25 mg/l MnSO₄ × 1 H₂O (Serva Cat# 28405), 50 mg/l Na₂MoO₄ × 2 H₂O (Serva Cat# 30207) and 50 mg/l H₃BO₃ (Merck Cat# 100165.5000) after sterilization of the basal medium) or in shaking Erlenmeyer flasks with CM-Medium (70 mM NH₄Cl (Merck Cat# 1.01145.5000), 7.3 mM KH₂PO₄, (Merck Cat#1.04873.100), 6.7 mM KCl (Merck Cat# 1.04936.1000), 2 mM MgSO₄ (Merck Cat# 1.05886.0500), 1% glucose (Sigma Cat# G-5400), 0.2% tryptone (Otto Nordwald Cat# 211701), 0.2% yeast extract (DIFCO Cat# 0127-07), 5 mM FeCl₂ × 7 H₂O (Merck Cat# 13478-10-9), 3.5 mM ZnSO₄, (Merck Cat# 108883), 6.2 mM MnCl₂, (Merck Cat# 5934.0100), pH 6.5) under constant light at 27°C. For germination, spores were incubated for two days in the dark on standard cornmeal agar supplemented with 60 mM ammonium acetate (Merck Cat# 1116.1000)²⁵. Pieces of the mycelium derived from germinated spores were transferred on M2 medium to obtain cultures of specific age.

Quantitative Real-time PCR (qRT-PCR)

After germination of spores, pieces of the mycelium were directly used or grown at 27° and constant light on M2 medium for 13 – 16 days (middle-aged) or 21 – 24 days (senescent), depending on the lifespan of the specific individual, to obtain cultures of specific age. A piece of the growth front was subsequently spread on a fresh M2 plate covered with cellophane (BioRad Cat# 1650963) and grown for 3 days. RNA was extracted with RNA-Plant kit (Machery-Nagel Cat# 740.949.250) and cDNA synthesis was performed using iScript kit (BioRad Cat# 170-8891). After dilution of cDNA to a concentration of 10 ng/µl, 20 ng was used per qRT-PCR reaction (IQ SybrGreen SuperMix, BioRad cat# 170-8882). The primers summarized in Table 1 were used to perform the qRT-PCR with three technical replicates per sample. A specific culture was compared to the mean CP of the juvenile cultures. Relative expression was normalized to PaPorin with the following formula²⁶.

E = PCR-Efficiency; CP = crossing point

Western blot analysis

To obtain total protein extracts, fungi of specific age were spread on a cellophane foil covered M2 surface for 3 days. Proteins were extracted as described in²⁴. Briefly, the mycelia were harvested, ground under liquid nitrogen, mixed with extraction buffer and centrifuged at 14,000 g at 4°C for 10 min. The supernatant was recovered and used for the experiments. The protein extracts were fractionated by 2-phase SDS-PAGE (14% separating gels) according to standard protocol²⁷. Proteins were subsequently transferred to PVDF membranes (Millipore Cat# IPFL00010). Blocking, antibody incubation and washing was performed according to western blot analysis handbook (LIC-OR Bioscience, Bad Homburg, Germany). The following primary antibodies were used: anti-PaPRE2 (rabbit, 1:500 dilution, raised against the specific peptide [H]-WKTKLEKGEFSNVT-[OH]; Sigma), anti-PaPRE3 (1:2500 dilution, raised against the specific synthetic peptide [H]-LYLPDTDYKVRHEN-[OH]; Sigma), anti-PaPUP1 (rabbit, 1:5000 dilution, raised against the specific peptide Ac-CLKRNYIKPNERT-amid, NEP), anti-HSP60 (mouse, 1:4000 dilution, Biomol, Cat# SPA-807), Anti-GFP (mouse, 1:10000 dilution, Sigma-Aldrich Cat# G6795 RRID:AB_563117). Secondary antibodies conjugated with IRDye 680 (1:15000 dilution, goat anti-mouse 680RD: LI-COR Biosciences Cat# 926-68070 RRID:AB_10956588) or IRDye CW 800 (1:15000 dilution, goat anti-rabbit 800: LIC-OR Biosciences, Cat# 926-3221) were used. After western transfer, the polyacrylamid gels were stained 1 h with coomassie blue as additional loading control.

Cloning procedures and generation of P. anserina mutants

The vector pExMthph²⁸ was used as backbone for the generation of PaPre2, PaPre3 and PaUmp1 overexpression plasmids. For the assembly of pPaPre2Ex1, pExMthph was cut with BamHI and XbaI. The PaPre2 gene and terminator were amplified with the primers Pre2ExpFor (AAGGATCCATGGACACCCTCGTTGCG; restriction sites are underlined) and Pre2ExpXbaRev (AAAGATCTTGGCCCTCCTTACTAGAC), cut with BamHI and XbaI and ligated with the backbone. For the generation of pPaPre3Ex1, the PaPre3 gene and terminator were amplified with the primers PaPre3FwdBam (TTGGATCCATGGAATTCGGTACATCGGG) and PaPre3RevPst (TTCTGCAGCCCACAACCAGAACCTTTCAC) cut with BamHI and PstI and ligated with the similarly restricted vector pExMthph. pPaUmp1Ex1 was generated by amplification of PaUmp1, including the terminator, with the primers PaUmp1FwdBamHI (TTGGATCCATGGTAAGTTGCAGCCAACC) and PaUmp1RevPst1 (TTCTGCAGGCTCCCGTGAGGGCAGGAC), restriction of the product with BamHI and PstI, and ligation into the similarly cut vector pExMthph. The generation of a Gfp-Cl1 overexpression plasmid was performed by 3 fragment ligation. The Gpd promotor from Aspergillus nidulans, the eGfp gene and the first part of the Cl1-sequence were amplified by PCR with the plasmid pSM5 (based on pSM2²⁹) as template and with primers eGfp-Pgpd-cl1for (CCTCGAGGTCGACGGTATCGATAAGCTTGATATCGAATT) containing a HindIII restriction site and eGfp-Pgpd-cl1rev (GTGGCTAGC GCTGCTGAACCAGTTCTTGCAGGC CTTGTACAGCTCGTCCAT), containing half of the Cl1 sequence (italic) optimized for codon usage of P. anserina and a Eco47III restriction site. The second half of the Cl1-sequence was amplified with the TrpC terminator in a similar manner with the primers Ttrpc-cl1for (CTTCAGCGAG CTCAGCCACTTCGTCATCCACCTCTA ATCCACTTAACGTTACTGA) containing an Eco53kI restriction sites (underlined) and Ttrpc-cl1rev (CCACCGCGGTGGCGGCCGCTCTAGAAAGAAGGATTACCTC) containing a XbaI restriction site. The PCR products were ligated into the purified backbone of pSM5, previously cut with HindIII and XbaI. The plasmids were used to transform P. anserina wild-type spheroplasts according to²⁴. pSM5 was used to generate a strain expressing Gfp without degron sequence. Briefly, mycelium of wild type “s” was blended and the cell wall digested with an enzyme solution. After filtration and concentration of sphaeroblasts by centrifugation, the sphaeroblasts were mixed with 10 µg plasmid DNA. Subsequently, polyethylene glycole (Serva Cat# 33136) was added to the sphaeroblasts. Transformants were selected for hygromycin B (Calbiochem Cat# 400051) resistance and the number of integrations was verified by Southern blot analysis.

Fluorescence microscopy

A piece of 2 day old mycelium was grown on a glass slide with a piece of PASM-medium²⁷ covered with a coverslip and incubated at 27°C for 2 days. Heat stressed samples were incubated at 27°C for 24 h followed by an incubation at 37°C for 24 h. The cover slip with hyphae on it was visualized using a fluorescence microscope (DM LB, Leica, Wetzlar, Germany) with the appropriate excitation and emission filter to detect the GFP signal and a digital camera system (DC500, Leica, Wetzlar, Germany).

Regulation of proteasome components during aging and oxidative stress

In order to address the role of the UPS on aging of P. anserina, we investigated the expression of the genes coding for proteolytic subunits PaPre2 (β5) and PaPre3 (β1) and of the proteasome assembly factor PaUmp1. First, we determined the abundance of transcript by using total RNA of juvenile, middle-aged and senescent cultures (Figure 1A). No significant changes in mRNA levels were observed in cultures of different age although mRNA abundance of all three genes was slightly reduced in senescent cultures. Next we analyzed protein levels of proteasome subunits in cultures of different age grown in standard growth medium and in medium to which paraquat was added as an inducer of oxidative stress. An increased abundance of mitochondrial HSP60 verified an increase in oxidative stress in senescent and in paraquat treated cultures (Figure 1B). However, no changes in the abundance of subunits PaPRE2, PaPRE3 and PaPUP1 (β3) of the proteasome were observed in the corresponding P. anserina cultures. We thus were unable to demonstrate a role of the ubiquitin proteasome system in counteracting adverse effects on cellular proteins in aged cultures and in cultures challenged with exogenous oxidative stress.

Figure 1. Gene expression and protein abundance of proteasome subunits during aging and PQ-stress.

(A) The expression of PaPre2, PaPre3 and PaUmp1 transcripts in juvenile, middle-aged and senescent samples is depicted relative to the juvenile wild type as mean ± SEM (5 – 7 biological replicates). (B) Western blot analysis of 50 µg total protein extracts of 5 d, 13 d and 21 d old wild type cultures grown on medium with and without the addition of 5 µM paraquat. Used antibodies are indicated on the right. The polyacrylamid gel stained with coomassie after blotting is used as loading control.

Overexpression of PaPre2 or PaPre3 does not increase the total amount of proteasome

High activity of the proteasome has been linked to increased health and lifespan^30–33. In human cell cultures, the overexpression of the subunits β5 and β1 was found to increase the overall abundance of the proteasome, as well as its activity and resistance to oxidative stress³⁴. To investigate whether or not such an effect is also observed in P. anserina, strains overexpressing the homolog subunits PaPre2 (β5) and PaPre3 (β1) were generated. First, plasmids conveying PaPre2 and PaPre3 overexpression were constructed and transformed into P. anserina spheroplasts (Table 2). Subsequently, overexpression of the genes was verified by qRT-PCR. PaPre3 expression was increased by factor 140 to 380 in the respective overexpression strain compared to wild type (Figure 2A). The PaPre2 overexpression strain exhibited a 94 times higher PaPre2 expression than the wild type (Figure 2B). In the next step, we evaluated protein levels in the overexpression strains by western blot immunodetection. The analysis of three independent PaPre3 overexpressors revealed two strains with unchanged protein abundance and one strain (PaPre3_OEx2) in which increased PaPRE3 signals occurred (Figure 2C). However, the detected signals are larger or smaller than expected for processed PaPRE3 and probably represent unprocessed PaPRE3 and a degradation product. A strain overexpressing PaPre2 showed no increase in PaPRE2 abundance compared to the wild type (Figure 2D). Thus, despite the strong increase in mRNA abundance, no substantial change in protein levels of the two investigated proteasome subunits was observed in the generated strains.

Figure 2. Overexpression of catalytic 20S subunits PaPre2 or PaPre3 does not alter the level of processed protein.

The expression of PaPre3 (A) and PaPre2 (B) in the respective overexpression strain was examined by qRT-PCR. (C, D) Total protein extracts of PaPre3 (50 µg) and PaPre2 (60 µg) overexpression strains were probed with α-PaPRE3 and α-PaPRE2 for the amount of processed proteasome subunits. The polyacrylamid gel stained with coomassie after blotting is displayed as loading control.

The proteasome activity reporter GFP-CL1 is degraded by autophagy

The activity of the UPS is not exclusively defined by the abundance of proteasome subunits but influenced by various factors like ubiquitin ligases, deubiquitin ligases, ATP-level, the regulatory particle, oxidative stress and post-translational modifications. In order to evaluate the efficiency of the UPS during aging of P. anserina, we generated two Gfp-cl1 strains with similar properties and a Gfp strain. Successful transformation of wild type was verified by Southern blot analysis (Table 2). The introduced genes are under the control of the constitutive Gpd promoter of Aspergillus nidulans. Gfp-cl1 codes for a protein containing the CL1 degron sequence fused to GFP. The CL1 sequence is a part of the Saccharomyces cerevisiae genome. It was first described in a ScURA3-CL1 fusion protein, which is unstable in wild type, but stable in strains lacking the ubiquitin ligases ScUbc6 and ScUbc7. Due to these characteristics the Cl1 degron has been used to monitor proteasome activity in various species including fly³⁵, mouse³⁶, rat^37,38 and human cell cultures^39–44.

In our work, we investigated the degradation of the CL1 degron fused to GFP. Fluorescence microscopy revealed diffuse fluorescence in whole cells of strains expressing Gfp-cl1 and Gfp, respectively, indicating a cytoplasmic localization (Figure 3A). Significantly, after applying heat stress, the Gfp-cl1 strain revealed a vacuolar localization of the GFP signal in some parts of the mycelium (Figure 3A). Western-blot analysis revealed two distinct GFP signals in Gfp-cl1-strains (Figure 3B). One signal corresponds to a protein with a size of 28.8 kDa expected for GFP-CL1 fusion protein while the other has the size of free GFP (26.9 KDa). This result was surprising, because proteasomal degradation should result in total decomposition of GFP-CL1 and provided a first clue for the degradation of the CL1 degron sequence by autophagy since the GFP part remains and is not, or only slowly, degraded by vacuolar proteases^17,18. To verify the degradation of the CL1 degron by autophagy, we generated a P. anserina strain lacking PaAtg1⁴⁵ and expressing Gfp-cl1 by crossing of single mutants and selection of the double mutant. PaATG1 is necessary for autophagy and ΔPaAtg1-strains are not able to transport proteins to the vacuole for degradation⁴⁵. Western blot analysis revealed that the double mutant contains only GFP-CL1 and no free GFP (Figure 3C), demonstrating that GFP-CL1 is at least partially degraded via autophagy in the wild type of P. anserina. This conclusion is supported by the accumulation of green fluorescence in the vacuoles of Gfp-cl1 overexpressing strains after the induction of autophagy by heat stress (Figure 3A).

Figure 3. GFP-Cl1 is degraded by autophagy.

(A) Fluorescence microscopy analysis of Gfp and Gfp-cl1-1 mutant. Fungi were grown for 2d at 27°. Heat stressed samples were incubated at 37°C for the last 24 h. (B) Western-blot analysis of 60 µg total protein extracts from Gfp-Cl1-1 and Gfp strains. Proteins were detected by western-blot-analysis with α-GFP antibody. The corresponding polyacrylamid gel was stained with coomassie after blotting as loading control. (C) Western-blot analysis of 45 µg total protein extracts from Gfp-cl1-2 strains and from a Gfp-cl1-2/ΔPaAtg1 double mutant. Proteins were detected by western-blot-analysis with α-GFP antibody. The polyacrylamid gel was stained with coomassie after blotting.