Heterochromatic marks are associated with the repression of secondary metabolism clusters in Aspergillus nidulans

Fungal secondary metabolites are important bioactive compounds but the conditions leading to expression of most of the putative secondary metabolism (SM) genes predicted by fungal genomics are unknown. Here we describe a novel mechanism involved in SM-gene regulation based on the finding that, in Aspergillus nidulans, mutants lacking components involved in heterochromatin formation show de-repression of genes involved in biosynthesis of sterigmatocystin (ST), penicillin and terrequinone A. During the active growth phase, the silent ST gene cluster is marked by histone H3 lysine 9 trimethylation and contains high levels of the heterochromatin protein-1 (HepA). Upon growth arrest and activation of SM, HepA and trimethylated H3K9 levels decrease concomitantly with increasing levels of acetylated histone H3. SM-specific chromatin modifications are restricted to genes located inside the ST cluster, and constitutive heterochromatic marks persist at loci immediately outside the cluster. LaeA, a global activator of SM clusters in fungi, counteracts the establishment of heterochromatic marks. Thus, one level of regulation of the A. nidulans ST cluster employs epigenetic control by H3K9 methylation and HepA binding to establish a repressive chromatin structure and LaeA is involved in reversal of this heterochromatic signature inside the cluster, but not in that of flanking genes.

tdiB 12 PCR product (amplified by primers NAIf1 and NAIr1), a 859 bp hepA PCR product (amplified by HpFwd and HpRev), a 864 bp stcO PCR product (amplified by stcOF and stcOR), a 991 bp AN7826.3 PCR product (amplified by corAF and corAR), a 2917 bp AN7801.3 PCR product (amplified by 7801F and 7801R), a 478 bp actin PCR product (amplified by ActFW and ActRV) and a 1 Kb clrD PCR product (amplified by crlDFWD and crlDREV). For Reverse transcription quantitative real-time PCR, cDNA was synthesized using the SuperScript™ First-Strand Synthesis System (Invitrogen) following the instructions of the provider. The platinum SYBR green qPCR SuperMix-UDG (Invitrogen) was used for amplification and detection of DNA in qPCR using a thousand fold dilution of the cDNA. aflR transcription was assessed using the primers aflRorfF and aflRorfR, hepA with primers HP3F and HP3R and clrD with primers clrDorf 2 F and clrDorf 2 R. All signals were normalized to the constitutively transcribed actin gene (acnA) amplified with primers acnf and acnR. The Bio-Rad (Hercules, CA) MyiQ cycler was used as device.

Construction of deletion/fusion cassettes and strains.
A deletion cassette for hepA was constructed were pyrGf was flanked by 3' and 5' sequences of the hepA ORF. The 5'-flanking hepA sequence was amplified from A. nidulans wild type genomic DNA using primers HP1840F and HP5270R, and the 3'flanking sequence was amplified from A. nidulans wild type genomic DNA with HP6090F and HP9600R. pyrGf was amplified from A. fumigatus wild type genomic DNA with primers pyrGHPF and pyrGhHPR. Nested primers HP2240 and HP9029 were used to amplify the complete assembled molecule. The hepA deletion cassette was used to transform a pyrG89 pantoB100 riboB2 argB2 yA2 strain. Transformants were selected on minimal media with appropriate supplements omitting uracil and uridine.
The deletion of clrD ORF was carried out by replacing it by the pyrGf gene (pyrGf). A deletion cassette was constructed were pyrGf was flanked by 3' and 5' sequences of the clrD ORF. The 5´-flanking clrD sequence was amplified from A. nidulans wild type genomic DNA with the primers clrD161F and clrD3113R and the 3'-flanking sequence was amplified with primers clrD4758F and clrD7120R. pyrGf was amplified from A. fumigatus wild type genomic DNA with primers clrDpyrF and clrDpyrR. Nested primers clrDNestF and clrDNestR were used to amplify the complete assembled molecule. The clrD deletion cassette was used to transform a pyrG89 pyroA4 argB2 yA2 strain.
Transformants were selected on minimal media with appropriate supplements omitting uracil and uridine.

Complementation of hepA in a hepAΔstrain.
In order to introduce hepA in a hepAΔ::pyrG89, pantoB100, riboB2, argB2 strain, a fusion hepA-argB molecule was constructed. hepA was amplified from A. nidulans wild type genomic DNA with primers HpmF and HpmR and fused with the 2.16 Kb PCR product amplified from plasmid pFB39 (carrying the argB gene) with primers HpmArgF and argBdownR. The assembled molecule was amplified with primers hpmNesF and argBbglIIR and used to transform hepAΔ::pyrG89, argB2, pantoB100, riboB2.
Transformants were selected on minimal media with appropriate supplements in the absence of arginine.

HepA
In order to over-express hepA, a transcriptional fusion between the S-tagged hepA coding sequence and the alcA promoter was generated. 952 bp of the HepA coding sequence was amplified with HPF and HPR primers, in which the HPF primer contains additional bases coding the for the S-tag and a BamHI site. The HPR primer contains a KpnI restriction site. The amplified fragment was introduced into KpnI-BamHI sites of pMT-mRFP (Toews et al. 2004), which contains the alcA promoter and argB. The terminator of trpC was amplified from plasmid phER-trpC (Pachlinger et al, 2005) as 590 bp KpnI fragment and introduced into the plasmid at the 3´ end of the HepA coding region with a KpnI restriction site. The resulting plasmid was sequenced to verify the correctness of the HepA-ORF and used to transform YR257 ΔhepA: pyrG argB2 pantoB100 riboB2 yA2 veA1. Transformants were selected on minimal media with appropriate supplements without arginine and confirmed by PCR and Southern hybridization (not shown). For Southern analysis genomic DNA was digested with BamHI and hybridized with a 32 P labelled probe derived from a 520 bp PCR product amplified by HP5580F and Kpn1-5´R. Western blot: Proteins were extracted using a modified TCA protein extraction protocol.
Mycelia were homogenized in liquid nitrogen to a fine powder. 0.1 grams of powder was suspended in 1ml of TCA buffer (10mM Tris pH 8, 20mM KCl, 1mM EDTA and 12% TCA), vortexed vigorously and centrifuged for 15 minutes at 13.000 rpm at 4°C and the supernatant was discarded. The pellet was resuspended and washed in 1ml Tris Base (1M pH unadjusted) to remove remaining TCA from the pellet and the supernatant discarded.
The pellet was resuspended in 200 µl of sample buffer (0, 3% SDS, 0,1M Tris.HCl pH 7,0) , heated to 95°C for 5 minutes and centrifuged for 15 minutes at 13000 rpm and 4°C to remove cell debris. Protein concentration in the supernatant was quantified using the BCA protein quantification kit (Pierce, USA). Roughly 20 µg of proteins were used for SDS-PAGE and Western blotting, and the membrane was probed with HP1 (1:200) antibody and S-tag antibody from Abcam (#19321) in a 1:10.000 dilution.

Complementation of clrD in a clrDΔ strain.
In order to introduce clrD in a clrDΔ::pyrG89, pyroA4, argB2, yA2 strain, a fusion clrD-argB molecule was constructed. clrD was amplified from A. nidulans wild type genomic DNA with primers clrDMinF and clrDMinR and fused with the 2.16 Kb PCR product amplified from pFB39 plasmid (carrying the argB gene) with primers dimMRargBF and argBdownR. The assembled molecule was amplified with primers clrDmNestF and argBbglIIR and used to transform clrDΔ::pyrG89, argB2, pyroA4, yA2. Transformants were selected on minimal media with the appropriate supplement in the absence of arginine.

Secondary metabolites analysis.
For solid media, a cork borer with diameter of 1.2 cm was used to collect samples from each plate after 72 hr growth. Each sample was homogenized and mixed well with 3 ml of double distilled H 2 O. Then 3 ml of CHCl 3 was added and agitated with a vortex.
After centrifugation for 5 min at 1,000 rpm, the separated organic phase was transferred to an 15 ml conical tube. All samples were dried down and resuspended in 100 μl For liquid shake cultures, A. nidulans strains were inoculated from conidia suspensions into 50 ml of liquid GMM in 125 ml flasks to a final concentration of 10 6 conidia/ml. Flasks were then incubated at 37°C for 48 or 72 hours with shaking at 250 rpm. 25 ml acetone was added to liquid culture, mixed well with a vortex and filtrated to remove mycelia. Acetone was evaporated over night in fume hood. The next day 50 ml chloroform was added to the filtrate, mixed well and then centrifuged for 10 min at 2,500 rpm. Chloroform layers (40 ml) were transferred to new tubes, dried and 100 μl chloroform was added to resuspend extracts in each tube. Thirty μl of each extract was separated on a TLC plate for ST analysis. Quantification of ST and NOR from A. nidulans extracts was accomplished using a CAMAG II densitometer, according to manufacturer's instructions. A wavelength of 245 nm was used for all analyses.

Quantification of radial growth.
Aspergillus nidulans strains wild type (RJW84.5), hepAΔ ( RJW63.1) and clrDΔ (YR31.1) were grown on solid GMM. Twenty μl of a spore suspension containing 10E+4 spores per ml (in 0.1% aqueous Tween80) were inoculated at the centre of a plate. The diameter of the colony was measured at a time period of every 24 hours.

Quantification of biomass.
Aspergillus nidulans strains wild type (RJW84.5), hepAΔ ( RJW63.1) and clrDΔ (YR31.1) were grown on liquid GMM. 2.5 E+8 conidia were inoculated in 250ml Erlenmeyer flasks containing 50ml GMM. The strains were incubated at 37°C with shaking at 180 rpm for five different time period of analysis i.e. 0, 4, 8, 16, 24, 48 and 72 hours. Mycelia was harvested by filtering, washed with sterile deionised water and squeezed in paper towel to get rid of all the liquid, packed in pre weighted aluminium foil and dried in a 65°C oven for three days. Weight of dry mass was measured for two consecutive days.

Quantification of conidial production.
Aspergillus nidulans strains wild type (RJW84.5), hepAΔ ( RJW63.1) and clrDΔ (YR31.1) were grown on solid GMM media which were inoculated at the centre of the plate with 20μl of a spore suspension containing 10E+4 spores per ml (in 0.1% aqueous Tween80 solution). The conidia from the entire colony were collected after 5 days of incubation in 0.01% Tween20 and counted with a haemocytometer.   Reyes et al., Supporting Figure S3 Supporting Figure S3. The expression level of laeA is only slightly altered in heterochromatin mutants. Comparison of mRNA steady-state levels of the laeA gene in strains grown in liquid GMM for for 24h (primary metabolism) or 48h (secondary metabolism). Strains are laeA + hepA + clrD + (wild type, WT) or are deleted for laeA, hepA or clrD,as single genes or in combinations. laeA expression is shown along with the actin (acnA) and rRNA loading controls. Numbers below each lane are actin-normalized relative expression levels of the mutant strains in relation to the actin-normalized wild type control, for which the expression level at 24 hours has been arbitrarily set to 1.
Supporting Figure S4.  Figure S5 TLC-plate analysis of secondary metabolites produced on solid GMM. In panel A, NOR production was quantified in a hepAΔ strain and compared to wild type, laeA Δ and hepAΔ laeAΔ strains. In panel B, NOR amount was compared between a clrDΔ strain and wild type, laeAΔ and clrDΔ laeAΔ strains Bar-graphs below the TLC plates summarize the results from three independent GMM solid media cultures grown for 5 days, extracted and analyzed separately for their production of norsolorinic acid (NOR). NOR amount was determined by densitometry and subjected to statistical analysis as described in the Experimental Procedures section. WT production levels were assigned a value of 1, and all other production levels are presented relative to WT. Different letters above columns represent statistical differences at p=0.01. Error bars represent +/-one standard deviation.