Panel A. Schematic of the subtelomeric region of chromosome VI illustrates the penicillin gene cluster as well as the boundary elements PbI and PbII. For reference, Shwab et al. 2007 used AN2647 as a gene that was located outside of the penicillin biosynthetic gene cluster region. Automated gene calls from AspGD are depicted from the PbI region as well as transposable elements located in PbIa and PbIb regions. Both of these regions contain low GC content compared to the average of the A. nidulans genome. Diagram is drawn to scale.
Panel B. In order to investigate the roles of the repetitive elements Mariner-4, DNA-3, Mariner-1 and DNA-2 identified in PbI, two areas (PbIa and PbIb) were deleted using a gene replacement cassette harboring the A. parasiticus pyrG gene (depicted in gray). In order to rule out the phenotype was due to incorporation of the A. parasiticus pyrG gene into this genomic region, additional mutants were made by inserting A. parasiticus pyrG either proximally or distally to the PbIa region (PbIap and PbIad). Diagram is drawn to scale.

Panel A. To determine deletion of PbI, genomic DNAs were digested with XbaI and NcoI where the expected hybridization band patterns by XbaI digestion using the 5′ PbI flank as a probe are a 3.2 kb band for the wild type strain and a 2.7 kb band for ΔPbI transformants. When A. parasiticus pyrG was used as a probe, there is no expected hybridization band pattern by NcoI digestion of the wild type strain and 4.7 and 1.1 kb fragments for ΔPbI mutants.
Panel B. To determine deletion of PbII, expected hybridization band patterns by XbaI digestion using the 5′ PbII flank as a probe are 5.2 and 1.9 kb bands for the wild type strain and a 8.3 kb band for ΔPbII strains. When A. parasiticus pyrG was used as a probe, the expected hybridization band patterns by HindIII digestion are no band for the wild type strain and 5.6 and 2 kb bands for ΔPbII mutants. It was not possible to find unique DNA sequences in the region of the genome and hence the presence of multiple hybridization fragments. PCR analysis supported the deletion of this region (data not shown).
Panel C. To determine deletion of PbIa, genomic DNAs were digested with XbaI where the expected hybridization band patterns using the 5′ PbIa flank as a probe are 7.4 and 2.6 kb bands for the wild type strain and 7.4 kb and 1.2 kb bands for ΔPbIa strains. When the 3′ flank was used as a probe, the expected hybridization band patterns by XbaI digestion are 7.7 kb for the wild type strain and 10.3 kb for ΔPbIa mutants.
Panel D. To determine deletion of PbIb, expected hybridization band patterns by BamHI digestion using the 5′ PbIb flank as a probe are a 6 kb band for the wild type strain and a 4.7 kb band for ΔPbIb strains. When the 3′ PbIb flank was used as a probe, the expected hybridization band patterns by BamHI digestion are 9.7 kb band for the wild type strain and 5.4 kb for ΔPbIb mutants.
For all panels the probes are indicated by horizontal lines. Marker ladder is in the right most lane of most figures but left most in the second figure in Panel D and not shown in Panel C. WT = wild type, pyrG = Aspergillus parasiticus pyrG.

Panel A. Histogram showing relative production of penicillin by isogenic wild type, PbI and ΔPbII mutants (RJMP1.19, TMSI2 and TMSII2.4 respectively). Penicillin was extracted from 72 hrs liquid shake cultures in triplicate according to standard cultural conditions, with or without addition of uracil and uridine and quantified using a bacterial growth inhibition assay. Wild type production levels were assigned a value of 100%, and all other production levels are presented relative to the wild type. The error bars represent the standard error. Different letters above the bars represent statistical differences at P ≤ 0.05 according to Tukey-Kramer.
Panel B. Wild type, ΔPbI and ΔPbII mutants (RJMP1.19, TMSI2 and TMSII2.4 respectively) were grown on GMM supplemented with pyridoxine with and without addition of uracil and uridine. Addition of uracil and uridine partially remediates growth and developmental deficiencies of Delta;PbII. MM = glucose minimal medium without addition of uracil and uridine, UU+ = addition of uracil and uridine.
Panel C. Histogram showing relative production of penicillin by wild type, ΔPbI, ΔPbII, ΔPbIa and ΔPbIb mutants (RJMP1.19, TMSI2, TMSII2.4, TMSIa1 and TMSIb1 respectively). Penicillin was extracted from 72-hrs liquid shake cultures in triplicate and quantified using a bacterial growth inhibition assay. For the histograms, wild type production levels were assigned a value of 100%, and all other production levels are presented relative to the wild type. The error bars represent the standard error. Different letters above the bars represent statistical differences at P ≤ 0.05 according to Tukey-Kramer.
Panel D. Northern blots showing expression of the penicillin cluster genes ipnA (wild type = RDIT9.32 and ΔPbIa = RMS1) and penDE (WT = RDIT9.32 and ΔPbIa = RMS1). RNA was extracted after 24, 48, and/or 72 hrs of growth in liquid shake cultures. Only the 48 hr time point is shown for penDE expression. Ethidium bromide-stained rRNA is shown as a loading control.

Panel A. To determine the insertion of pyrG telomere proximal of PbIa (PbIap mutants), genomic DNAs were digested with XbaI where the expected hybridization band patterns using A. parasiticus pyrG as a probe; no band is expected in the wild type strain and a 4.5 kb band for PbIap mutants. Upon using probe 1, the wild type shows 7.5 and 2.65 kb bands and the PbIap transformants shows 7.5 and 1.2 kb bands.
Panel B. For insertion of pyrG distal of PbIa (PbIad mutants), the expected hybridization band patterns by a NcoI digestion using a 794 bp fragment (probe 2) are a 11.3 kb band for the wild type strain and a 7.1 kb band for PbIad strains. When A. parasiticus pyrG was used as a probe, no band is expected in the wild type strain and 5.9 and 1.1 kb bands for PbIad mutants.
Panel C. Penicillin was extracted from 72 hrs liquid shake cultures in triplicate and quantified using a bacterial growth inhibition assay. For the histograms, wild type production levels were assigned a value of 100%, and all other production levels are presented relative to the wild type. The error bars represent the standard error. Different letters above the bars represent statistical differences at P ≤ 0.05 according to Tukey-Kramer multiple comparison test.
For panels A and B the probes are indicated by horizontal lines. Marker ladder is in the right most lane of figures in panels A and B. WT = wild type, pyrG = Aspergillus parasiticus pyrG.

Panel A. Schematic of the terrequinone A cluster located on chromosome V. Insertion of the PbIa fragment (3.7 kb) was done at three locations: the flanking region before the cluster (A), in the middle of the cluster (B), and the flanking region after the cluster (C). This was accomplished by fusing the A. fumigatus pyroA gene to the PbIa fragment in addition to 5′ and 3′ flanking regions to target the construct to the desired location. Control strains were constructed that only contained the A. fumiagtus pyroA gene inserted at the same locations. Insertional complementation mutants were confirmed by Southern analysis by using wild type DNA as a probe. Expected banding patterns of a NcoI digestion for the A insertion were: WT – 2.7 kb, ΔPbIa – 2.7 kb, TJMP42 – 3.3 kb and 2.2 kb, and TJMP46 – 6.0 kb and 3.3 kb. A NcoI digestion for insertion B resulted in the following expected bands: WT – 2.3 kb, ΔPbIa 2.3 kb, TJMP43 – 3.0 kb and 2.2 kb, TJMP47 – 6.0 kb and 3.0 kb. Expected banding patters of a BamHI digestion for insertion C were: WT – 3.7 kb, ΔPbIa – 3.7 kb, TJMP44 – 4.0 kb and 2.4 kb, TJMP48 – 7.8 kb and 2.4 kb.
Panel B. A bioassay for penicillin production was done with extracts from 72 hr liquid shake cultures in GMM supplemented with pyridoxine and uracil/uridine. Histograms represent the zone of inhibition normalized to wild type levels. None of the complementation mutants restored penicillin production. (WT = TJMP45.2, ΔPbIa = TJMP41.1)
Panel C. Terrequinone A production was assayed by a chloroform extract 4-day cultures grown on Champs solid media. Subsequent extracts were analyzed via think layer chromatography using hexane:ethyl acetate (4:1) as a solvent system. A ΔtdiB strain (TJW65.7) was used to determine the spot corresponding to terrequinone A.

Panel A. Northern blots showing expression of the penicillin cluster genes ipnA and penDE by wild type (WT), ΔlaeA and ΔPbIa;ΔlaeA mutants (RDIT9.32, RJW41.A and RMS1.2 respectively) at 48 hrs of growth in liquid shake cultures. Ethidium bromide-stained rRNA is shown as a loading control.
Panel B. Northern blots showing expression of the penicillin cluster genes ipnA and penDE by wild type (WT), ΔPbIa, ΔhdaA and ΔhdaA mutants (RDIT9.32, RMS1.1 and RMS1.22 respectively). RNA was extracted after 24, 48, and/or 72 hrs of growth in liquid shake cultures. Only the 48 hr period is shown for the penDE gene. Ethidium bromide-stained rRNA is shown as a loading control.
Panel C. Penicillin was extracted from 72-hrs liquid shake cultures in triplicate and quantified using a bacterial growth inhibition assay. For the histograms, WT production levels were assigned a value of 100%, and all other production levels are presented relative to the WT. The error bars represent the standard error. Different letters (a, b, c and d) above the bars represent statistical differences at P ≤ 0.05 according to Tukey-Kramer multiple comparison test.
Panel D. Probing with the 3.7 kb PbIa fragment shows no expression in WT and the ΔPbIa mutant strain (RDIT9.32 and RMS1 respectively) in liquid shake culture for 48 hr. However, a transcript is expressed in ΔlaeA and ΔPbIa;ΔlaeA mutants. Ethidium bromide-stained rRNA is shown as a loading control.