Unique Regulatory Mechanism for d-Galactose Utilization in Aspergillus nidulans

Abstract

This study describes two novel regulators, GalX and GalR, that control d-galactose utilization in Aspergillus nidulans. This system is unique for A. nidulans since no GalR homologs were found in other ascomycetes. GalR shares significant sequence identity with the arabinanolytic and xylanolytic regulators AraR and XlnR, but GalX is more distantly related.

TEXT

d-Galactose is ubiquitous in nature, in particular in the plant cell wall polysaccharides, the hemicelluloses and pectin. Filamentous fungi release d-galactose using α-galactosidases, β-galactosidases, and endogalactanases (2). Intracellularly, the Leloir metabolic pathway converts d-galactose in three steps to d-glucose-6-phosphate, which enters glycolysis (Fig. 1 and Table 1) (6–8). This pathway has been studied in detail in Aspergillus nidulans and Trichoderma reesei (15–20). In addition, a second, alternative d-galactose utilization pathway has been described for these two species (Fig. 1) (4, 18). Some enzymes of this pathway were shown to also be involved in pentose catabolism (5).

Fig. 1.

A model for d-galactose catabolism in A. nidulans and T. reesei as modified from reference 11. The Leloir metabolic pathway converts d-galactose to d-glucose-6-phosphate, while the alternative pathway converts d-galactose to d-fructose-6-phosphate. d-Glucose-6-phosphate and d-fructose-6-phosphate then enter glycolysis. The gene products are as follows: galE, galactokinase (AN4957); galD, galactose-1-phosphate uridyl transferase (AN6182); galF, UTP-glucose-1-phosphate uridyl transferase (AN9148); galG, UDP-galactose-4-epimerase (AN4727); pgmB, phosphoglucomutase (AN2867); ladA, l-arabinitol-4-dehydrogenase (AN0942); lxrA, l-xylulose reductase (AN10169); xdhA, xylitol dehydrogenase (AN9064); hxkA, hexokinase (AN7459).

Table 1.

Strains used in this study

Strain	Genotype	Reference or source
FP-308.1	pyrG89 argB2	Gift from C. Scazzocchio
pantoB100 pyrG89 strain	pantoB100 pyrG89	This study
FGSC A211	galA1 bia1 bioA1 wA3	19
FP-309.1 (reference strain)	argB2 pyrG89::pyrG+	This study
FP-310.1	argB2 pyrG89 galRΔ::pyrG+	This study
FP-311.1	pantoB100 galA1	This study
FP-312.1	pantoB100::pantoB+	This study
FP-313.1	pantoB100 galX::pantoB+	This study
FP-314.1	pantoB100 gpd::galR::pantoB+	This study

The regulation of d-galactose metabolism has been extensively studied in yeast (3, 9, 10, 12, 14, 21), but little is known in filamentous fungi. Here, we describe the characterization of two novel regulators, GalR and GalX, involved in d-galactose release and catabolism in A. nidulans.

GalR was identified using BlastP analysis against the Broad Institute Aspergillus Comparative Database (http://www.broadinstitute.org/annotation/genome/aspergillus_group/MultiHome.html), using the A. nidulans ortholog of the A. niger xylanolytic regulator XlnR (22) as a query. Three proteins from A. nidulans with significant sequence identity were found: XlnR, AraR (1), and a novel putative regulator (AN10550). Disruption of this gene resulted in a strain which showed severely reduced growth on minimal medium (MM) with 25 mM d-galactose and slightly reduced growth on galactitol (Fig. 2). Growth was also tested on different d-galactose-containing substrates (guar gum, locust bean gum, arabic gum, and arabinogalactan) and d-glucose and l-arabinose. Slightly reduced growth was observed for ΔgalR on guar gum, possibly because it has the highest d-galactose content. No differences were observed on any of the other carbon sources.

Fig. 2.

Growth profile of the A. nidulans reference strain (ref) and galR disruptant (ΔgalR) strains. glc, 25 mM d-glucose; gal, 25 mM d-galactose; ara, 25 mM l-arabinose; g-ol, 25 mM galactitol; GG, 1% guar gum; LBG, 1% locust bean gum; AG, 1% arabic gum; ABG, 1% arabinogalactan.

Liquid cultures using MM with 10 mM d-galactose resulted in strongly reduced α-galactosidase activity for ΔgalR compared to that of the reference strain, while the β-galactosidase activity remained unaltered (data not shown). d-Galactose uptake in these cultures was also similar between the strains, demonstrating that the influence of this regulator is mainly on intracellular catabolism.

The phenotype of the ΔgalR strain resembled that of a previously described A. nidulans UV mutant (galA1) (17). Sequencing the galR gene from the galA1 mutant revealed no mutations in the coding or promoter region. However, a point mutation was detected 58 bp downstream of the ATG of a neighboring open reading frame (ORF) (AN10543, named GalX) which generated a TAA stop codon within its putative DNA binding domain (Fig. 3A). Transforming this mutant with galX under the control of its own promoter restored growth on d-galactose, confirming that the point mutation in the galX ORF is responsible for the phenotype of the mutant (data not shown).

Fig. 3.

(A) Schematic representation of the positioning of galR and galX on chromosome III, contig 75. The galX gene is located 477 bp downstream of galR. The point mutation (T → A) in the galA1 mutant is located in the galX ORF, 58 bp downstream of its ATG. White regions depict introns. (B) Alignment of the amino acid sequences within the DNA binding domain of Gal4 with GalX and GalR.

GalX and GalR consist of protein sequences of 679 and 798 amino acids, respectively, that both contain a Zn₂Cys₆ binuclear cluster domain and a fungal-specific transcription factor domain. The amino acid sequences of GalR and GalX were aligned using ClustalW (http://www.genome.jp/tools/clustalw/), showing that they only share 12% identity. In contrast, GalR has 43% and 32% identity to A. nidulans AraR and XlnR, respectively. This indicates that, although GalR and GalX are located next to each other on the A. nidulans genome, it is unlikely that galR originated from a local gene duplication. Sequence comparison of GalR and GalX with the Saccharomyces cerevisiae d-galactose regulator, Gal4, showed low overall identities, 11% and 20%, respectively, and high variation within their DNA binding domains (Fig. 3B). BlastP analysis of the A. nidulans genome using Gal4 as a query identified a closer homolog of Gal4 (AN4785) whose function remains to be studied. The synteny of galX and galR in the genomes of Aspergillus nidulans, Aspergillus niger, Aspergillus oryzae, Aspergillus flavus, Aspergillus clavatus, Neosartorya fischeri, Aspergillus terreus, and Aspergillus fumigatus was analyzed using the Sybil algorithm at www.aspgd.org. These data demonstrated that galR is only present in A. nidulans, while galX is present in most aspergilli (see Fig. S1A in the supplemental material). In addition, it is clear that this genomic region has undergone many changes during the evolution of these aspergilli (see Fig. S1B in the supplemental material).

Growth on d-galactose differs strongly between the aspergilli (www.fung-growth.org). The absence of growth on d-galactose for A. clavatus and A. terreus may be explained in part by the absence of both galR and galX in their genomes. However, galX is also absent in N. fischeri, which is able to grow on d-galactose, suggesting an alternative regulator for d-galactose catabolism in this species. This indicates that the regulation of growth on d-galactose may involve even more factors than the presence or absence of GalR and GalX.

To study the effect of GalR and GalX on the expression of d-galactose-related genes in A. nidulans, an expression study was performed in which the reference, ΔgalR, and galA1 strains were pregrown overnight in complete medium with 0.1% (wt/vol) d-fructose, after which the pregrown cultures were induced with 10 mM d-fructose, 10 mM d-galactose, or 10 mM galactitol and incubated for 2.5 h. In the reference strain, a gene from the alternative galactose utilization pathway (ladA), a putative alcohol dehydrogenase which is located upstream of galX and galR (ladB, AN4336), and a secreted α-galactosidase (aglC, AN8138) were expressed in the presence of d-galactose and galactitol (Fig. 4). A basal level of expression was observed for two genes of the Leloir pathway (galE and galD) on d-fructose, with increased levels on d-galactose and galactitol. The expression of a third gene of the Leloir pathway (galG) appears to be constitutive. In the ΔgalR strain, the expression of ladA and aglC is lost and the expression of galE and galD is reduced to basal levels on d-galactose and galactitol, suggesting that GalR controls their d-galactose- and galactitol-dependent expression. Despite the absence of expression of ladA in the ΔgalR strain on galactitol, only a small reduction in growth was observed compared to the growth of the wild type, indicating that galactitol can also be converted by other enzymes. A possible candidate for this is LadB, whose function remains to be determined.

Fig. 4.

(A and B) Expression analysis of the reference strain (ref), galR disruptant (ΔgalR), and galA1 mutant on d-fructose (Frc), d-galactose (Gal), and galactitol (G-ol). The 18S rRNA gene was used as a loading control.

In the galA1 mutant, the expression of galR and ladB is lost, suggesting that GalX regulates these genes either directly or indirectly. The expression of the galX gene is not dependent on GalR, since galX is expressed in the ΔgalR strain. Transforming the galA1 mutant with galR under the control of the constitutive gpdA promoter restored growth on d-galactose, showing that GalX influences the expression of d-galactose catabolic genes via galR. Furthermore, no expression of galE and further reduced expression of galD were observed in the galA1 mutant, suggesting that they may also to a small extent be regulated by GalX.

In summary, this study demonstrates that the regulation of d-galactose utilization in A. nidulans differs from that in the other aspergilli and even more from the well-described Gal4 system in S. cerevisiae. This is further supported by the fact that no sequences that strongly resembled the Gal4 binding site (CGG-N₁₁-CCG) (13) were found in the promoter regions of A. nidulans genes from the Leloir pathway and the alternative galactose pathway (data not shown).