Telomere position effect is regulated by heterochromatin-associated proteins and NkuA in Aspergillus nidulans

Abstract

Gene-silencing mechanisms are being shown to be associated with an increasing number of fungal developmental processes. Telomere position effect (TPE) is a eukaryotic phenomenon resulting in gene repression in areas immediately adjacent to telomere caps. Here, TPE is shown to regulate expression of transgenes on the left arm of chromosome III and the right arm of chromosome VI in Aspergillus nidulans. Phenotypes found to be associated with transgene repression included reduction in radial growth and the absence of sexual spores; however, these pleiotropic phenotypes were remedied when cultures were grown on media with appropriate supplementation. Simple radial growth and ascosporogenesis assays provided insights into the mechanism of TPE, including a means to determine its extent. These experiments revealed that the KU70 homologue (NkuA) and the heterochromatin-associated proteins HepA, ClrD and HdaA were partially required for transgene silencing. This study indicates that TPE extends at least 30 kb on chromosome III, suggesting that this phenomenon may be important for gene regulation in subtelomeric regions of A. nidulans.

Fig. 1.

(a) Gene replacement of AN5092 with the A. fumigatus pyrG gene resulted in a strain with visible growth defects on minimal medium compared to a control strain (TMM11). The observed phenotype was remedied when cultures were grown on medium containing the appropriate supplements, in this case uridine and uracil. (b) The growth reduction on minimal medium correlates with repression of the AfpyrG transgene located at AN5092. The control strain (TMM11) contains AfpyrG integrated at the nonessential gene AN5495. Semiquantitative RT-PCR revealed that relative AfpyrG expression levels were 0.4 for TDP1-1 and TDP1-2 and 0.5 for TDP2-7 and TDP2-12.

Fig. 2.

Schematic of gene disruption strategies (a) and Southern analysis of mutants constructed at the AN5092 locus (b). The AN5092 locus was replaced with three constructs: A. fumigatus pyrG, A. fumigatus pyrG in the opposite orientation (GrypfA), and the A. fumigatus pyroA gene. Due to the repetitive nature of TLH-like sequences, Southern blots produced a considerable amount of background. Strains used for Southern blots were: WT, RDIT9.32; ΔAN5092 : : AfpyrG, RJMP116.3; ΔAN5092 : : GrypfA, TSM18-3; and ΔAN5092 : : AfpyroA, RJMP121.7. Using a radiolabelled probe corresponding to a portion of the AN5092 locus and the 5′ flanking region, in a BamHI digestion the following bands were expected: WT, 5.6 kb; ΔAN5092 : : AfpyrG, 2.6 kb; ΔAN5092 : : GrypfA, 2.8 kb; and ΔAN5092 : : AfpyroA, >10 kb. An EcoRI digestion predicted WT, 3.0 kb; ΔAN5092 : : AfpyrG, 1.4 kb; ΔAN5092 : : GrypfA, 3.2 kb; and ΔAN5092 : : AfpyroA, 2.5 kb.

Fig. 3.

Scale schematic representation of the left telomere of chromosome III of A. nidulans. AN5092 is located ∼20 kb from the telomere of the left arm of chromosome III. Clutterbuck & Farman (2008) describe that 18.4 kb of DNA is missing between the telomere of chromosome III and the genome assembly. Bioinformatic analysis places the SpoC1 cluster approximately 30 kb from the telomere of chromosome III, spanning the current annotation from AN5091 to AN5081. Three putative open reading frames exist between the SpoC1 cluster and the end of the current annotation in the AspGD: AN5093 is unlikely to be a functional gene as it contains repetitive sequences, AN5092 is a putative TLH-like gene, and AN5091 encodes a methyltransferase with sequence homology to LaeA.

Fig. 4.

The AN5091 locus was replaced with the A. fumigatus pyrG gene as drawn schematically. Strains were confirmed by Southern analysis. Strains used were WT (RDIT9.32) and ΔAN5091 : : AfpyrG (RJMP115.3). A SacI digestion predicted bands of WT 4.5 kb and ΔAN5091 : : AfpyrG 2.2 kb; a SphI digestion predicted WT 3.4 kb and ΔAN5091 : : AfpyrG 4.0 kb; while an EcoRV digestion predicted WT 3.4 kb+2.1 kb and ΔAN5091 : : AfpyrG 4.0 kb+2.1 kb.

Fig. 5.

(a) Prototrophic strains exhibiting TPE display a quantifiable growth defect on minimal medium versus minimal medium with supplements. Integration of AfpyrG at either AN5092 or AN5091 results in the same phenotype, suggesting that TPE extends at least 30 kb from the telomere. WT, RDIT9.32; ΔAN5091, RJMP115.3; ΔAN5092, RJMP116.3. (b) Sexual development is aberrant in TPE strains. Macroscopic cleistothecial development is unaffected by integration of AfpyrG at either the AN5092 or the AN5091 locus; however, ascospore production is severely debilitated. Quantification of ascosporogenesis indicates that TPE strains produce three orders of magnitude fewer ascospores when grown on minimal medium compared to medium with appropriate supplements. WT, RDIT9.32; ΔAN5091, RJMP115.3; and ΔAN5092, RJMP116.3. Note that the y-axis scale is logarithmic. Means±sd are plotted; asterisks indicate statistically significant differences between wild-type and other strains (P<0.001 using Student's t-test).

Fig. 6.

Physiological experiments with mutants created by Shaaban et al. (2010) allowed for determination of TPE at chromosome VI. (a) A radial growth assay illustrates the quantifiable difference in growth of ΔPbII (A. parasiticus pyrG integrated ∼8.5 kb from the telomere of chromosome VI-R) when grown on GMM versus GMM+UU. The isogenic control strain (TMS8.2) contains the A. parasiticus pyrG gene located ectopically and also shows a slight reduction in growth on GMM versus GMM+UU compared to WT (RJMP1.19); however, radial growth of ΔPbII is significantly reduced compared to TMS8.2. (b) Ascosporogenesis assays mirror the radial growth assay, where ΔPbII produces significantly fewer ascospores than either the WT or TMS8.2. These data illustrate TPE repression of the A. parasiticus pyrG transgene when located near the telomere of chromosome VI. Means±sd are plotted; asterisks indicate statistically significant differences between wild-type and other strains at P<0.001 using Student's t-test.

Fig. 7.

In order to address the mechanism of TPE in A. nidulans, radial growth and ascosporogenesis assays were conducted on double mutants. (a) Radial growth assays illustrate the reduction in growth on GMM versus GMM+pyridoxine (P) of strains harbouring ΔAN5092 : : AfpyroA. (b) Quantification of the radial growth assay indicates that HepA, ClrD and HdaA derepress the AfpyroA transgene, while VeA1 and HstA have no effect on AfpyroA repression. The single mutants (veA1, ΔhepA, ΔclrD, ΔhstA and ΔhdaA) show no effects on radial growth in this assay. (c) Ascosporogenesis assays match the radial growth assays, providing further evidence for the involvement of HepA, ClrD and HdaA in TPE. The single mutants (veA1, ΔhepA, ΔclrD, ΔhstA and ΔhdaA) show no effect on ascospore production in this assay. (d) Ascospore production is partially rescued in a double mutant (ΔAN5091 ΔnkuA) compared to the single ΔAN5091 mutant; however, ascospore production does not reach wild-type levels. WT (veA⁺), RDIT9.32; ΔAN5092 : : AfpyroA, RJMP121.7; ΔAN5092 : : AfpyroA veA1, RJMP121.4; ΔAN5092 : : AfpyroA ΔhepA, RJMP122.6; ΔAN5092 : : AfpyroA ΔclrD, RJMP125.20; ΔAN5092 : : AfpyroA ΔhstA, RJMP127.4; ΔAN5092 : : AfpyroA ΔhdaA, RJMP123.3; veA1, RDIT2.3; ΔhepA, RJW110.4; ΔclrD, RJMP135.11; ΔhstA, RJMP131.7; ΔhdaA, RMS1.22; ΔnkuA, TJMP45.2; ΔAN5091, RJMP115.3; and ΔAN5091ΔnkuA, TJMP16.1. Note that the y-axis scale is logarithmic in panels (c) and (d). Means±sd are plotted in panels (b–d); asterisks indicate statistically significant differences between wild-type and other strains at P<0.001 using Student's t-test.