Fungal Genetics and Biology 49 (2012) 1033–1043

Contents lists available at SciVerse ScienceDirect

Fungal Genetics and Biology

journal homepage: www.elsevier .com/locate /yfgbi

Aspergillus nidulans galactofuranose biosynthesis affects antifungal drug sensitivity

Md Kausar Alam a,1, Amira M. El-Ganiny a,b,1, Sharmin Afroz a, David A.R. Sanders c, Juxin Liu d,Susan G.W. Kaminskyj a,⇑a Department of Biology, University of Saskatchewan, 112 Science Place, Saskatoon, SK, Canada S7N 5E2b Department of Microbiology, Faculty of Pharmacy, Zagazig University, Egyptc Department of Chemistry, University of Saskatchewan, 110 Science Place, Saskatoon, SK, Canada S7N 5C9d Department of Mathematics and Statistics, University of Saskatchewan, 106 Wiggins Road, Saskatoon, SK, Canada S7N 5E6

a r t i c l e i n f o

Article history:Received 15 May 2012Accepted 14 August 2012Available online 16 October 2012

Keywords:Aspergillus nidulansGalactofuranoseUDP-galactopyranose mutaseUDP-glucose/galactose-4-epimeraseUDP-galactofuranose transporteralcA promoter

1087-1845/$ - see front matter � 2012 Elsevier Inc. Ahttp://dx.doi.org/10.1016/j.fgb.2012.08.010

Abbreviations: Af, Aspergillus fumigatus; alcA(ppromoter; CM, complete medium; CM3G, completeCMT, complete medium with 100 mM threonine; Ggenomic DNA; mAb, monoclonal antibody; niiA(p),MM, minimal medium; MMA, minimal medium amedium nitrate; qPCR, quantitative real time PCR; UDUDP-glucose/galactose-4-epimerase; UGM, UDP-galUDP-galactofuranose transporter.⇑ Corresponding author.

E-mail address: susan.kaminskyj@usask.ca (S.G.W.1 These authors contributed equally to the manuscri

a b s t r a c t

The cell wall is essential for fungal survival in natural environments. Many fungal wall carbohydrates areabsent from humans, so they are a promising source of antifungal drug targets. Galactofuranose (Galf) is asugar that decorates certain carbohydrates and lipids. It comprises about 5% of the Aspergillus fumigatuscell wall, and may play a role in systemic aspergillosis. We are studying Aspergillus wall formation in thetractable model system, A. nidulans. Previously we showed single-gene deletions of three sequentialA. nidulans Galf biosynthesis proteins each caused similar hyphal morphogenesis defects and 500-foldreduced colony growth and sporulation. Here, we generated ugeA, ugmA and ugtA strains controlled bythe alcA(p) or niiA(p) regulatable promoters. For repression and expression, alcA(p)-regulated strainswere grown on complete medium with glucose or threonine, whereas niiA(p)-regulated strains weregrown on minimal medium with ammonium or nitrate. Expression was assessed by qPCR and colonyphenotype. The alcA(p) and niiA(p) strains produced similar effects: colonies resembling wild type forgene expression, and resembling deletion strains for gene repression. Galf immunolocalization usingthe L10 monoclonal antibody showed that ugmA deletion and repression phenotypes correlated with lossof hyphal wall Galf. None of the gene manipulations affected itraconazole sensitivity, as expected. Dele-tion of any of ugmA, ugeA, ugtA, their repression by alcA(p) or niiA(p), OR, ugmA overexpression by alcA(p),increased sensitivity to Caspofungin. Strains with alcA(p)-mediated overexpression of ugeA and ugtA hadlower caspofungin sensitivity. Galf appears to play an important role in A. nidulans growth and vigor.

� 2012 Elsevier Inc. All rights reserved.

1. Introduction

Human systemic fungal infections are an increasing threat(Fisher et al., 2012), particularly due to improved medical technol-ogy leading to larger populations of immune-compromised people.Systemic infections from a dozen or more fungi lead to morbiditydespite aggressive anti-fungal drug treatment (Lass-Florl, 2009).Overall, Aspergillus is second only to Candida as a cause of invasivefungal infections (Erjavec et al., 2009). Humans have few anti-fun-

ll rights reserved.

), alcohol dehydrogenase Imedium with 3% glucose;

alf, galactofuranose; gDNA,nitrite utilization promoter;mmonium; MMN, minimalP, uridine diphosphate; UGE,

actopyranose mutase; UGT,

Kaminskyj).pt.

gal drug options because fungi share many biochemical pathwayswith animals. A major difference between organisms in these king-doms is that fungal cells have a carbohydrate wall. This wall pro-tects the fungal cell from damage, and mediates the interactionbetween the fungus and its environment. If the wall is removedor weakened, the fungus cannot survive in natural environments(Bennett et al., 1985). About 90% of the fungal wall is composedof polysaccharides that are not found in human hosts (Latgé,2007). Echinocandins are the only antifungal drugs in clinical usethat block an aspect of fungal wall synthesis, formation of b-1,3-glucans that comprise about half of fungal cell wall polysaccha-rides in some species (Espinel-Ingroff, 2009). However, echinocan-dins are clinically effective only against Candida and Aspergillus;they must be used intravenously (Denning, 2003); and drug-resis-tant strains have already been identified (Walker et al., 2010).There is a pressing need to develop additional antifungal drugs(Mircus et al., 2009) particularly against novel aspects of fungalphysiology. Gauwerky et al. (2009) suggest that minor wall compo-nents as well as virulence factors have the potential to be usefulanti-fungal drug targets.

1034 M.K. Alam et al. / Fungal Genetics and Biology 49 (2012) 1033–1043

We are exploring an aspect of wall biosynthesis in Aspergillusnidulans, the galactofuranose (Galf) biosynthesis pathway. Galf isthe five-membered ring form of galactose found in the cell wallof many microorganisms including fungi, but only the six-memberring form called galactopyranose is found in mammals. In microor-ganisms, Galf residues form the side chains of glycoconjugatesincluding galactomannan. Galf biosynthesis genes encode UDP-glucose/galactose-4-epimerase (UgeA) (El-Ganiny et al., 2010),UDP-galactopyranose mutase (UgmA) (El-Ganiny et al., 2008),and the UDP-galactofuranose transporter (UgtA) (Afroz et al.,2011). Although none of these genes was essential for A. nidulansin culture, these deletion strains had aberrant wall maturation(Paul et al., 2011) and 500-fold reduced hyphal growth and sporu-lation rate.

Galf residues in human serum can be used to track a patient’sresponse to therapy for systemic Aspergillus infections (Bennettet al., 1985; Stynen et al., 1992; Shibata et al., 2009). Galf has beenshown to be essential for the virulence of the pathogenic proto-zoan, Leishmania (Kleckza et al., 2007). Galf could be a virulencedeterminant in Aspergillus fumigatus based on studies with a UgmA(also known as GlfA) deletion strain in a murine model (Schmal-horst et al., 2008). Recent evidence to the contrary (Heesemannet al., 2011) showed that A. fumigatus hyphae pretreated withmonoclonal antibodies (mAbs) specific to hyphal Galf residueshad similar virulence to untreated hyphae. Nevertheless, Galfmetabolism could still be a useful target for anti-fungal drug devel-opment (Pederson and Turco, 2003) because it plays importantalthough not fully-understood roles in fungal growth.

Regulated promoters can be used to explore the role (Waringet al., 1989) and requirement (Hu et al., 2007) for Aspergillus genes,typically using the alcA (Waring et al., 1989; Monteiro and DeLu-cas, 2010) and niiA (Monteiro and DeLucas, 2010) promoters whoseexpression is controlled by carbon source or nitrogen source,respectively. For example, repression of A. nidulans chitin synthasechsB in an alcA(p)–chsB strain caused slow growth and highlybranched hyphae (Ichinomiya et al., 2002). Repression of alcA(p)-regulated A. fumigatus O-mannosyltransferase 2 (Afpmt2) causedgrowth retardation, abnormal cell polarity, defective cell wallintegrity, and reduced conidiation (Fang et al., 2010). Repressionof alcA(p)-regulated protein kinase C (pkcA) reduced germinationrate, hyphal growth and conidiation, as well as increasing hyphalwall thickness and sensitivity to wall-selective agents: Caspofun-gin, Calcofluor White, and Congo Red (Ronen et al., 2007). As ex-pected, this strain was also more sensitive to the protein kinaseinhibitor staurosporine although it responded like wild type toamphotericin B and voriconazole (Ronen et al., 2007). Binderet al. (2010) showed that an alcA(p)–pkaA strain grown on repres-sion medium was hypersensitive to PAF, an antifungal protein. Ta-ken together, expression-regulated strains have been usedsuccessfully to study cell wall formation in Aspergillus species.

Expression-regulated strains have also been used to identifynew cell wall destabilizing compounds (Mircus et al., 2009) andto study cell wall integrity in the presence of anti-fungal agents.Fortwendel et al. (2009) used alcA(p)-regulated A. fumigatus strainsdefective in cell wall integrity pathway genes (rasA, cnaA and crzAstrains) to show that low cell wall integrity correlated with echino-candin sensitivity. Jiang et al. (2008) created A. fumigatus strainsconditionally expressing alcA(p)-regulated GDP-mannose pyro-phosphorylase. These strains had defective cell walls, impairedpolarity maintenance, reduced conidiation, and increased sensitiv-ity to wall targeting chemicals. Regulating gene expression is apowerful adjunct to deletion analysis for exploring gene function.

Here, we describe studies on the function of Galf biosynthesisgenes using strains whose endogenous promoter was replaced bythe alcohol dehydrogenase promoter alcA(p) (Ichinomiya et al.,2002; Waring et al., 1989) which is highly induced by alcohols,

especially threonine, and is repressed by glucose (Romero et al.,2003; Tribus et al., 2010). As a precaution against potential compli-cations from using a sugar-repressed promoter to study the carbo-hydrate-rich wall, we repeated key experiments with these genesregulated by niiA(p), which is regulated by nitrogen source(Monteiro and deLucas, 2010). Under gene repression conditionswith both promoters, the regulated strain colonies phenocopiedtheir respective deletion strains. Notably, all the deletion strainswere significantly more sensitive to the wall-targeting drug Caspo-fungin, as were the repressed strains. Our results suggest that therole of Galf biosynthesis plays multiple roles in fungal growth, wallarchitecture, and drug sensitivity, and as such may be a useful drugtarget.

2. Materials and methods

2.1. Strains, plasmids and culture conditions

Strains, primers and plasmids are listed in Suppl. Table A.A. nidulans strains were maintained on complete medium (CM)with supplements for nutritional markers as described in Kamin-skyj (2001). To manipulate alcA(p)-regulated gene expression, CMthat otherwise contained 1% glucose was modified as follows:maximum repression used CM containing 3% glucose (CM3G);overexpression used CM lacking glucose but containing 100 mMthreonine (CMT). For niiA(p)-regulated gene expression, minimalmedium (MM) containing 70 mM NaNO3 (MMN) was used forinduction, and MM containing 10 mM ammonium tartrate(MMA) was used for repression. Media used for niiA(p)-regulationwere supplemented with vitamin solution and pyridoxine. Wewere not able to find a medium for overexpressing niiA(p)-regu-lated genes.

2.2. Construction of regulatable promoter strains

A promoter exchange module consisting of AfpyrG and alcA(p)was amplified from the plasmid, palcA(p) (a gift of Loretta Jack-son-Hayes, Rhodes College). This module was inserted immedi-ately upstream of each Galf biosynthesis gene to create alcA(p)-regulated strains. A 1.2-kb DNA fragment containing AfniiA(p)was amplified by PCR from genomic DNA of wild type A. fumigatus.This was used to create a niiA(p)-regulatable promoter module,also with AfpyrG as the selectable marker. This construct was in-serted immediately upstream of the coding sequence for each Galfbiosynthesis gene to create niiA(p)-regulated strains (Suppl. Fig. A).

2.3. Transformation

Constructs were transformed into A1149 protoplasts. Protop-lasting and transformation followed procedures in Osmani et al.(2006) and Szewczyk et al. (2007). Long-term storage of competentprotoplasts is described in El-Ganiny et al. (2010). Primary trans-formants were grown on selective medium containing 1 M sorbitolor 1 M sucrose as osmostabilizer. Spores from these colonies werestreaked three times on selective medium prior to gDNA extrac-tion. Genomic DNA from transformant spores (see below) was usedas template for PCR to confirm the each manipulation (Suppl.Fig. A).

2.4. Isolation of genomic DNA from A. nidulans spores

Approximately 1 � 107 spores from individual 2 d old colonieswere harvested in sterile distilled water, pelleted in 1.5 mL micro-fuge tubes, and aspirated to dampness. Tubes containing dampspore samples were sealed with lid-locks, floated in 100 mL water,

M.K. Alam et al. / Fungal Genetics and Biology 49 (2012) 1033–1043 1035

and microwaved at full power for 6 min. Then 35 lL of TE buffer(10 mM Tris–Cl, 1 mM EDTA, pH 8) was added to each tube. Tubeswere vortexed briefly and put on wet ice for 5 min. The spore-DNAsuspension was centrifuged at 16,000g for 5 min to pellet debris.One microliter of supernatant was used as PCR template for ampli-cons up to 4 kb long.

2.5. RT-PCR and qPCR

Gene expression studies compared levels of Galf biosynthesisgenes in wild type, deleted, and alcA(p)-, and niiA(p)-regulatedugeA, ugmA and ugtA strains. Spores were inoculated in CM orinduction or repression liquid media, and incubated at 28 �C withshaking at 250 rpm for 16 h. Mycelia were collected by centrifuga-tion, frozen in liquid nitrogen, and lyophilized. Total RNA was ex-tracted using an RNeasy plant kit (Qiagen) followingmanufacturer’s instructions. RNA concentration was measuredusing a Nanodrop�, then diluted to 400 ng/lL. Genomic DNA elim-ination and reverse transcription used a QuaniTect reverse tran-scription kit (Qiagen) following the manufacturer instructions(Cuero et al., 2003).

Quantitative real time PCR (qPCR) was performed in 96-welloptical plates in an iQ5 real-time PCR detection system (Bio-Rad).Gene expression was assayed in triplicate in total volume 20 lLreactions containing cDNA at an appropriate dilution and SYBRgreen fluorescein (Qiagen). A no-template control was used foreach gene. Actin was used as a reference gene (Bohle et al.,2007). Primers actF and actR (Suppl. Table A) were designed to am-plify regions that included an intron in the actA genomic sequence.These primers generated a �200 bp band from cDNA and �400 bpband from gDNA, to detect genomic DNA contamination in thetemplate. The qPCR amplification used the following conditions:95 �C/15 min for one cycle, 95 �C/15 s, 55 �C/40 s and 72 �C/30 sfor 40 cycles and final extension cycle of 72 �C/2 min. Meltingcurve analysis was done using the following cycle: 15 s at 65 �Cwith an increase of 0.5 �C each cycle to 95 �C. For FKS and ags1 geneexpression analysis, we used wild type (AAE1), ugmAD, andAfniiA(p)–ugmA strains. The relative expression was normalizedto actA (Upadhyay and Shaw, 2008) and calculated using the DDCtmethod (Livak and Schmittgen, 2001). Three independent biologi-cal replicates were performed for each strain/culture condition.

2.6. Colony growth and sporulation

Colony characters were examined as described in El-Ganinyet al. (2008). Strains were grown for two generations on inducingor repressing media to control for possible dowry effects, thenspores of wild type and regulated strains were streaked on induc-tion or repression media and incubated for 3 d at 28 �C to give iso-lated colonies. The diameter of 10 colonies/strain was measured tothe nearest millimeter using a dissection microscope. The numberof spores produced per colony was counted for four colonies/strain.Isolated colonies were vortexed in a microfuge tube containing1 mL water, then samples of spores were counted with ahemocytometer.

2.7. Microscopy

Samples were prepared for confocal microscopy as described inEl-Ganiny et al. (2008). Freshly harvested spores were grown oncoverslips at 28 �C for 16 h in induction or repression media. Hy-phae were fixed and stained with Hoechst 33258 (for nuclei) andCalcofluor (for cell walls). Samples were imaged using a ZeissMETA510 as described previously (El-Ganiny et al., 2008). LSM im-age browser software was used for measurements of hyphal width(at septa) and basal cell length (distance between adjacent septa).

Galf content of wild type, ugmA-GFP, ugmAD, and niiA(p)-ugmAhyphae was assessed using immunolocalization. Monoclonal anti-Galf antibodies L10 and L99 were a generous gift of Prof. Frank Ebel(Univ Munich). Primary antibodies were used at full strength;TRITC-conjugated goat-anti-mouse was used at 1:10 dilution.Immunofluorescence procedures followed El-Ganiny et al. (2008)and samples were examined using confocal epifluorescencemicroscopy.

Transmission electron microscopy was used to measure hyphalwall thickness for alcA(p)-regulated colonies grown on dialysistubing overlying CMT or CM3G. Samples were prepared as de-scribed in Kaminskyj (2000). Hyphal wall thickness was measuredwhere the cell membrane bilayer was crisply in focus. Typicallythree sites were measured for each of ten hyphae per strain/med-ium combination.

Scanning electron microscopy was used to examine conidiatingwild type and alcA(p)-regulated colonies following the procedurein El-Ganiny et al. (2008). Strains were grown on dialysis tubinglaid on CMT and CM3G for 3 d at 28 �C. Colonies were fixed byimmersion in 1% glutaraldehyde, dehydrated in acetone, criticalpoint dried (Polaron E3000, Series II), and gold sputter coated (Ed-wards model S150B). Samples were imaged with a JEOL 840A scan-ning electron microscope.

2.8. Drug sensitivity testing

Antifungal agents and solvents were obtained from the follow-ing manufacturers: amphotericin B, terbinafine, itraconazole, di-methyl sulphoxide (DMSO) (Sigma Chemicals); ethanol (VWR);caspofungin (a gift from Merck, now Schering-Plough). Stock solu-tions were prepared as follows: amphotericin B (20 mg/mL inDMSO), itraconazole (1.6 mg/mL in DMSO), terbinafine (1.6 mg/mL in 50% ethanol), and caspofungin (20 mg/mL in sterile water).Aliquots of stock solutions were stored at �80 �C.

Wild type and Galf-biosynthesis gene deletion strains weregrown on CM; gene-regulated strains were grown on repressionand expression media. Spore suspensions collected from 4 d oldcolonies were filtered through VWR 413 filter paper (average poresize 5 lm) to remove hyphae and conidiophores. Spores in the fil-tered suspension were counted using a hemocytometer and ad-justed to 50,000 spores/lL.

Antifungal drug sensitivity testing used a disk-diffusion methodmodified from (Kontoyiannis et al., 2003; Lass-Florl, 2009), as de-scribed in Afroz et al. (2011). Strains were grown for two roundson the test media before being used for drug sensitivity studies,to compensate for potential spore dowry effects. For the drug sen-sitivity test, 1 � 107 spores were mixed into 20 mL of 50 �C expres-sion or repression agar and immediately poured in 9 cm diameterPetri plates. After the medium had solidified, sterile 6 mm paperdisks were placed on the agar surface. Anti-fungal drug stock solu-tions (described above) were micro-pipetted onto individual disks:10 lL of terbinafine and itraconazole, and 20 lL of amphotericin Band caspofungin. Solvent control studies with DMSO and 50% eth-anol showed no zone of inhibition. The plates were incubated at28 �C and assessed after 48 h (Kiraz et al., 2009).

The radius of the zone of inhibition was calculated as

½diameter ðmmÞ with no visible growth

� disk diameter ðmmÞ�=2

Two measurements at orthogonal axes were taken for each disk oneach plate; four plates were assessed for each strain and mediumcombination, and each experiment was repeated. These data wereused for statistical analysis. Data are presented as indexed valueswith respect to wild type on CM (for deletion strains) and with re-spect to wild type on the test medium (for regulated strains).

1036 M.K. Alam et al. / Fungal Genetics and Biology 49 (2012) 1033–1043

2.9. Data processing and analysis

Confocal images were processed using LSM examiner. Imagesare presented using Adobe Photoshop 7.1. Statistical analysis usedKruskal–Wallis and associated post hoc tests with R software: R2.13 for kruskal.test, and R 2.9 for kruskalmc.test.

3. Results

We examined the roles of Galf-biosynthesis gene products foranti-fungal drug sensitivity using conditional strains whose geneexpression was under the control of the regulatable promoter,alcA(p) (Romero et al., 2003; Tribus et al., 2010). Previously wehad deleted each of these genes: ugmA (El-Ganiny et al., 2008);ugeA (El-Ganiny et al., 2010); ugtA (Afroz et al., 2011), to studythe effects on A. nidulans morphogensis, conidiation, and wallultrastructure. We expected that gene repression would have sim-ilar effects to gene deletion, and that over-expression would havelittle effect. We used these strains to assess the role of Galf biosyn-thesis in sensitivity to anti-fungal drugs, especially Caspofunginthat targets cell wall synthesis. For each section, we present ugmAresults in detail, followed by comparison with results for ugeA andugtA. The alcA(p) data were confirmed using niiA(p)-regulation(Gauwerky et al., 2009; Monteiro and DeLucas, 2010).

3.1. Construction and validation of regulated strains

We replaced each endogenous Galf biosynthesis gene promoterwith the alcA(p) and with the niiA(p) conditional promoters, usingAfpyrG as the selectable marker (Szewczyk et al., 2007). Primarytransformants were grown on CM or MM, respectively, containing1 M sorbitol or 1 M sucrose as an osmotic stabilizer. Strain con-struction and confirmation is shown in Suppl. Fig. A.

To select media appropriate for regulation of Galf biosynthesisgenes in A. nidulans, the alcA(p)–ugmA strain was grown on CM con-taining a variety of carbon sources, and the colony morphologieswere compared qualitatively. Media included CM with 100 mMthreonine (CMT), 1% fructose plus 100 mM threonine, 1% ethanol,1% glycerol, 1% glucose (CM), or 3% glucose (CM3G), as well as YEPD

Table 1Colony and hyphal characteristicsa of wild type and deletion strains grown on CM, and wigrown on over-expression (CMT) and repression (CM3G) media. First row of each column

Character Expression (medium) Wild

Colony diameter (mm)b Neutral (CM; wild type, D) 17 ±Overexpression (CMT) 22 ±Repression (CM3G) 16 ±

Spores/colony (�106)c Neutral (CM; wild type, D) 107 ±Overexpression (CMT) 12.4 ±Repression (CM3G) 38 ±

Hyphal width (lm)d Neutral (CM; wild type, D) 2.7 ±Overexpression (CMT) 2.1 ±Repression (CM3G) 3.1 ±

Basal cell length (lm)d Neutral (CM; wild type, D) 26 ±Overexpression (CMT) 27 ±Repression (CM3G) 30 ±

Cell wall thickness (nm)e Neutral (CM; wild type, D) 54 ±Overexpression (CMT) 36 ±Repression (CM3G) 50 ±

a Values are presented as mean ± standard error. For colonies grown on CM, values fohyphae from El-Ganiny et al. (2008); for ugtAD hyphae from Afroz et al. (2011). Cellulcharacteristic to be significantly different at P < 0.05. Differences for strains grown on Ceach row, values followed by different letters (f–i) are significantly different at P < 0.05.

b Ten colonies per strain/medium combination.c At least three colonies per strain/medium combination.d Fifty measurements per strain/medium combination. Hyphal width was measured ae Measured from transmission electron micrographs for at least seven cross-sectioned

the cell membrane bilayer was crisply in focus.

(1% yeast extract, 2% peptone, 2% dextrose [=glucose]) that was pre-viously used in Zarrin et al. (2005). Growth of the alcA(p)–ugmAstrain on YEPD resembled the ugmAD strain, however the sporula-tion of the wild type strain on YEPD was greatly reduced (Suppl.Fig. B) so it was deemed unsuitable for our purposes. In comparison,growing the wild type strain on CM vs CM3G vs CMT caused substan-tially less variation in wild type colony characteristics (Table 1).

Growth of the niiA(p)-regulated ugmA strain was compared onMM containing nitrate salts (MMN), nitrite salts, ammonium tar-trate (MMA), and ammonium nitrate as nitrogen sources. Colonygrowth of the niiA(p)–ugmA strain resembled wild type on MMcontaining nitrate, nitrite, and ammonium nitrate, but resembledthe deletion strain on MMA.

The expression level of ugeA, ugmA and ugtA were assessed inwild type and respective regulated strains using qRT-PCR, withactA expression as the comparator. Expression of ugmA and actAin the wild type transformation parent of the regulated strainswas comparable on all media used (upper panels in Fig. 1A andB). For alcA(p)-ugmA regulation, growth on CMT showed roughlycomparable ugmA and actA expression levels, whereas growth onCM3G caused a dramatic decrease on ugmA but not actA expression(lower panels in Fig. 2A). We found substantially similar resultswith niiA(p)–ugmA expression levels (with respect to actA) whengrown on inducting MMN vs repressing MMA media (Fig. 1B).We also found that regulation of ugeA (upper panel in Fig. 1C)and ugtA (lower panel in Fig. 1C) by these promoters was compa-rable to ugmA. The effect of alc(p)–ugeA, alc(p)–ugmA, and alc(p)–ugtA regulation with respect to the wild type strain were also com-pared using qPCR, again with actA as the comparator (Fig. 1D). Con-sistent with our RT-PCR results for alcA(p)-regulation,overexpression was associated with 8, 4, and 6-fold increase inugeA, ugmA and ugtA expressions respectively, and repression with100–150-fold decrease in all the alcA(p)-regulated strains (Fig. 1D).

3.2. Repression of Galf biosynthesis genes causes compact colonygrowth and reduced sporulation

Growing the alcA(p)–ugmA strain on CM3G to repress ugmA pro-duced growth and sporulation phenotypes similar to the ugmAD

ld type and alcA(p)-regulated Aspergillus nidulans galactofuranose biosynthesis strainsrepresents wild type and respective deleted strains grown on CM.

type ugmA ugeA ugtA

0.4f 4 ± 0.2g 5 ± 0.2g 4 ± 0.2g

0.5f 17 ± 0.6g 21 ± 0.5f 16 ± 0.2g

0.4f 4 ± 0.2g 5 ± 0.2g 4 ± 0.1g

15f 0.2 ± 0.1g 0.4 ± 0.1g 1.3 ± 0.2g

1f 8.2 ± 1g 10.8 ± 1fg 13.8 ± 4f

11f 0.6 ± 0.1g 3.0 ± 0.6g 3.8 ± 0.1g

0.4f 3.6 ± 0.6g 3.6 ± 0.1g 3.5 ± 0.1g

0.03f 3.0 ± 0.0g 2.2 ± 0.04f 1.9 ± 0.05h

0.05f 3.7 ± 0.1g 3.6 ± 0.1g 3.4 ± 0.0g

1f 15 ± 1g 14 ± 1g 16 ± 1g

1.2f 14 ± 0.6g 22 ± 1.2h 25 ± 0.5h

1.7f 17 ± 1.0g 24 ± 1.1f 16 ± 0.5g

2f 204 ± 10g 123 ± 4h 183 ± 1i

2f 42 ± 1g 30 ± 1h 40 ± 1g

2f 82 ± 3g 59 ± 2g 83 ± 5g

r wild type and ugmAD hyphae were taken from El-Ganiny et al. (2008); for ugeADar characteristics were compared using a Kruskal–Wallis test, and shown for eachMT or CM3G were compared using a Kruskal–Wallis multiple comparison test. For

t septa. Basal cell length was between adjacent septa.hyphae per strain/medium combination, typically at three places per hypha where

Fig. 1. Expression of ugmA, ugeA, and ugtA in wild type and regulated strains. (A and B) RT-PCR showing actA and ugmA expression in wild type and ugmA-regulated strains.(A) alcA(p)–ugmA grown on overexpression medium CMT, and repression medium CM3G. (B) niiA(p)–ugmA grown on expression medium MMN, and repression mediumMMA. (C) RT-PCR showing the ugeA (upper panel) and ugtA (lower panel) expression in regulated strains grown on overexpression medium CMT, and repression mediumCM3G. (D) Relative expression using qPCR of ugeA, ugmA, and ugtA with respect to actA for wild type grown on CM, and alcA(p)-regulated strains grown on CM3G (repression)and CMT (over-expression).

M.K. Alam et al. / Fungal Genetics and Biology 49 (2012) 1033–1043 1037

strain, regarding colony diameter, spores per isolated colony, hy-phal width, and basal cell length (Table 1; Fig. 2). Scanning electronmicrographs showed reduced sporulation was due to a decrease inconidiophore number as well as aberrant conidiophore formation(Fig. 3). For the alcA(p)–ugeA and alcA(p)–ugtA strains, growth onCM3G resulted in compact colonies with wide, branched hyphaethat had substantially reduced sporulation. Strains grown onCMT resembled wild type (Table 1; Figs. 2 and 3).

The ugmAD and alcA(p)–ugmA repressed strains differed in hy-phal wall thickness (Table 1). The walls of the alcA(p)-ugmA strainon CM3G were twice as thick as wild type on the same medium,but only half as thick as the ugmAD strain on CM (Table 1; El-Gan-iny et al., 2008). We repeated the alcA(p)–ugmA CM3G fixation ontwo separate occasions with similar results. The alcA(p)–ugeA andalcA(p)–ugtA wall thicknesses when grown on CM3G were alsothinner than the comparable deletion strains (Table 1). Thus, dele-tion and alcA(p)-repression phenotypes for Galf biosynthesis genesdiffered in some quantitative aspects.

3.3. Deletion and repression of ugmA cause hyphal morphogenesisdefects that correlate with loss of immunolocalizable wall Galf

We expected repression of Galf biosynthesis would reduce wallGalf content, as we had previously shown for ugmAD (El-Ganinyet al., 2008). Prof. Frank Ebel’s group (Univ. Munich) recently gen-erated the L10 and L99 anti-Galf mAbs, (Heesemann et al., 2011)that have high affinity for hyphal wall Galf. They generously pro-vided us with samples of each mAb. We had strong Galf immuno-localization in wild type phenotype A. nidulans hyphal walls usingL10 (Fig. 4A, B and D), but less intense signal with L99 (not shown).

In addition to wall staining, the cytoplasm also contained L10-reac-tive material, consistent with UgmA-GFP localization being cyto-plasmic (Fig. 4B0 and 4B00) and UDP-Galf being synthesized in thecytoplasm prior to Galf incorporation into wall compounds (El-Ganiny et al., 2010). Both ugmAD (Fig. 4C), and niiA(p)–ugmAstrains grown on MMA repression medium (Fig. 4E), had wide,highly-branched hyphae. As expected, the ugmA deletion andrepression strain hyphal walls lacked detectable wall Galf (Fig. 4Cand E), as did their cytoplasm.

We attempted to localize alpha-glucan using the monoclonalantibody, MOPC-104E, which had been described by Futagamiet al. (2011) for dot-blot studies in A. nidulans. Our intent was todetermine if alpha-glucan content or distribution was affectedwhen Galf synthesis was reduced or absent. However MOPC-104E did not react for immunofluorescence localization. It is possi-ble that the extractions required for cell wall ghost preparation ex-posed an epitope that is not accessible in whole cells.

3.4. Galf biosynthesis affects a-glucan synthase and b-glucan synthaseexpression

Both deletion and repression of Galf biosynthesis in A. nidulansare associated with increased hyphal wall thickness (Table 1; El-Ganiny et al., 2008), albeit to different levels. Damveld et al.(2008) had previously shown that A. niger strains with defectiveugmA homologues had increased wall a-glucan content. We com-pared the expression level of a-glucan synthase (ags1) in wild type,ugmAD, and AfniiA(p)–ugmA strains growing on MMN and MMA(Fig. 5A). As expected, AfniiA(p)–ugmA grown on MMN (induction)had similar expression to wild type. Compared to the wild type

Fig. 2. Colony morphology of wild type and conditional promoter-regulated strains of ugmA, ugeA, ugtA. In each of A and B, the colonies are the same age. (A) Wild type,alcA(p)-regulated, and ugmAD strains grown on complete medium containing 100 mM threonine (CMT) or 3% glucose (CM3G) as sole carbon sources, for gene over-expression or repression, respectively. (B) Wild type and niiA(p)-regulated strains grown on minimal medium containing nitrate (MMN) or ammonium (MMA) for geneexpression or repression, respectively. (For interpretation to colours in this figure, the reader is referred to the web version of this paper.)

Fig. 3. Scanning electron micrographs showing: colony (A), and conidiophore (B) phenotype for wild type and alcA(p)-regulated ugmA, ugeA, and ugtA strains grown on CMTfor overexpression and CM3G for repression. (C) Colony and conidiophores of ugmAD on CMT and CM3G. On CMT media, the alcA(p) strains have abundant conidiophoreswith long chains of spores similar to wild type strain. On CM3G media, alcA(p)–ugmA, alcA(p)–ugeA, alcA(p)–ugtA produced many fewer conidiophores similar to ugmADstrain. Scale bar is 100 lm for A and C (upper panel) and 10 lm for B and C (lower panel).

1038 M.K. Alam et al. / Fungal Genetics and Biology 49 (2012) 1033–1043

Fig. 4. Galactofuranose immunolocalization in Aspergillus nidulans hyphae using the L10 monoclonal antibody. (A) Wild type, (B) UgmA-GFP, (C) ugmAD, (D) niiA(p)–ugmAgrown on induction medium, (E) niiA(p)–ugmA grown on repression medium. Lower case letters are corresponding transmission images. B0 shows UgmA-GFP distribution,which is cytoplasmic, as expected. B00 shows a merge of B and B0 . Bars = 10 lm.

M.K. Alam et al. / Fungal Genetics and Biology 49 (2012) 1033–1043 1039

parental strain, ags1 expression was more than 4-fold higher in augmAD strain, or in the AfniiA(p)–ugmA on MMA (repression).We also compared the expression level of b-glucan synthase(FKS) in wild type, ugmAD, and AfniiA(p)–ugmA strains growingon MMN and MMA (Fig. 5B). AfniiA(p)–ugmA grown on MMN(induction) had almost similar expression to wild type. Comparedto the wild type parental strain, fksA expression was about 2.5-folddecreased in a ugmAD strain, or in the AfniiA(p)–ugmA on MMA(repression).

3.5. Repression of Galf biosynthesis increases sensitivity toCaspofungin

Sensitivity to anti-fungal drugs of wild type and gene deletionstrains on CM, alcA(p)-regulated strains on CMT and CM3G, andniiA(p)-regulated strains on MMN and MMA, was tested using adisk-diffusion drug-sensitivity assay. We compared Caspofungin(blocks wall b-1,3-glucan synthesis) with drugs that target ergos-terol (Amphotericin B) and ergosterol biosynthesis (Itraconazole

Fig. 5. The expression of a-glucan synthase (ags1) and b-glucan synthase (fksA) isaffected by galactofuranose biosynthesis. Quantitative PCR results for ags1 (A) andfksA (B) expression (as fold changes with respect to actA ± SE) for wild type, ugmAdeletion, and ugmA-regulated strains.

Fig. 6. Response of wild type and alcA(p)-regulated ugmA, ugeA, and ugtA strains to(A) Itraconazole and (B) Caspofungin. The alcA(p)-regulated genes were overex-pressed by growth complete medium containing 100 mM threonine (CMT), andrepressed by growth on complete medium containing 3% glucose (CM3G). OnCaspofungin, sensitivity is measured at the innermost clear zone (arrows).

1040 M.K. Alam et al. / Fungal Genetics and Biology 49 (2012) 1033–1043

and Terbinafine). In addition, both 5-fluocytosine and Nikkomy-cin Z were assessed for their effect on wild type and Galf-biosyn-thesis gene deletion strains. Neither of these latter drugsinhibited colony growth, but Nikkomycin Z delayed germinationby several hours, causing spores to swell considerably before germtube establishment (data not shown).

For clarity, Table 2 shows drug sensitivity values for Itraconaz-ole and Caspofungin only, since in this assay Amphotericin B and

Table 2Antifungal drug sensitivitya indexb comparing wild type, deletion, and alcA(p)-regulated Galf biosynthesis strains.

Expression (medium) Wild type ugmA ugeA ugtA

Itraconazolec

Neutral (CM); wild type and D strains 1.0 1.1 1.1 1.1alcA(p) Overexpression (CMT) 0.8 1.1 0.9 1.1

Repression (CM3G) 0.8 1.1 1.0 1.1niiA(p) Expression (MMN) 1.0 1.0 1.1 1.0

Repression (MMA) 1.0 1.2 1.2 1.2

Caspofunginc

Neutral (CM); wild type and D strains 1.0 1.8 1.3 1.7alcA(p) Overexpression (CMT) 1.0 1.3 0.5 0.7

Repression (CM3G) 1.0 1.9 1.5 1.8niiA(p) Expression (MMN) 1.0 1.0 0.9 1.0

Repression (MMA) 1.0 1.9 1.6 1.7

a Drug sensitivity was measured using a disk diffusion assay (see Fig. 6). ForCaspofungin, sensitivity was the radius (mm) of the clear zone with no visiblegrowth (see arrows on Fig. 6B). Mean ± SE of two measurements for each of fourbiological replicates were used for statistical analysis (not shown).

b Index sensitivity values are for each drug/strain combination compared to wildtype on that medium. Wild type values are compared to CM with 1% glucose. Genedeletion strains are compared to wild type on CM. Regulated strains are comparedto wild type on the same medium. Index values >1.0 are more sensitive than wildtype. Index values that differed by >0.2 were based on data that were significantlydifferent.

c See Section 2 for details regarding drug dosage and medium formulation.

Terbinafine results were comparable to Itraconazole. Sensitivitiesof deletion and regulated expression strains are presented as an in-dex compared to wild type on the same medium. Values greaterthan 1.0 indicate strains and conditions associated with drughypersensitivity. Statistical analysis of the raw data and compari-son to sensitivity indices showed that an index difference of >0.2correlated with statistical significance. Fig. 6 shows representativeresults related to those in Table 2, where gene transcription levelwas controlled by alcA(p).

Expression levels of the Galf biosynthesis genes did not signifi-cantly affect sensitivity to Itraconazole. Caspofungin sensitivitywas increased for all three Galf biosynthesis gene-deletion strainscompared to wild type. Similarly, the alcA(p)- and niiA(p)-regulatedstrains were hypersensitive to Caspofungin when grown on theirrespective repression media (Table 2, Fig. 6). When these geneswere over-expressed on CMT, the alcA(p)–ugmA strain was moresensitive to Caspofungin, but the regulated ugeA and ugtA strainswere less sensitive. Growth of these strains on MMN (expression,but not overexpression) did not affect drug sensitivity. Taken to-gether, the hyphal wall Galf content (Fig. 4) was consistent withour expectations for the Caspofungin sensitivity (Table 2) of ugmAdeleted and repressed strains.

4. Discussion

Deletion of any of three sequential-acting genes in the Galf bio-synthesis pathway causes similar defects in A. nidulans hyphalmorphogenesis and colony development (El-Ganiny et al., 2008,2010; Afroz et al., 2011) and in wall surface structure, force com-pliance, and adhesion (Paul et al., 2011). Galf forms the side chainsof many glycoconjugates (Latgé, 2007, 2009) but its role(s) in wallformation and maturation are still far from understood. We ex-pected that wild type levels of Galf biosynthesis would be impor-tant for Caspofungin resistance because we had previously foundthat the walls Galf deletion strains were structurally compromised(Paul et al., 2011). To test this we compared the drug sensitivity ofgene deletion and wild type with regulated Galf-biosynthesisstrains. This is the first study to explore the roles of Galf biosynthe-sis proteins on drug sensitivity of A. nidulans strains using the alcAand niiA promoters, and the first to examine the effect of regulatinggenes in the Galf biosynthesis pathway. We expected that the

M.K. Alam et al. / Fungal Genetics and Biology 49 (2012) 1033–1043 1041

products of these genes would have similar effects on sensitivity toantifungal compounds that targeted cell walls but not to those thattargeted other aspects of fungal physiology. Consistent with this,we found increased echinocandin sensitivity for the deletionstrains and repressed strains, but no significant change for com-pounds affecting ergosterol or its biosynthesis.

4.1. Altering Galf biosynthesis gene expression level affects wallformation, colony growth, and development

The phenotypes of strains deleted or repressed for Galf biosyn-thesis pathway genes were generally comparable. Repression ofalcA(p)–ugmA with 3% glucose (CM3G) and of niiA(p)–ugmA withammonium (MMA) both caused colony and hyphal defects compa-rable to the ugmAD strain (El-Ganiny et al., 2008). Consistent withour RT-PCR and qPCR results, this is strong evidence that our reg-ulatable promoter strains were functioning as expected. Similar re-sults have been found for regulated histone deacetylase (rpdA)controlled by alcA(p) and xylP(p) (Tribus et al., 2010), and repres-sion of protein kinase C (pkcA) on hyphal growth, wall thickness,and antifungal compound sensitivity (Ronen et al., 2007).

Gene-deletion and gene-repression phenotypes were qualita-tively similar in most respects. However, for each of the Galf biosyn-thesis genes considered in this study, alcA(p)-repressed strainssporulated better than the comparable deletion strains, and their hy-phal walls were half the thickness of the respective deletion strains.These anomalies prompted us to repeat our studies on all three genesusing niiA(p)-regulation which is repressed by ammonium. Resultsfor colony growth and development regulated by the niiA(p) werecomparable to those for alcA(p), consistent with the phenotypeswere caused by changes in gene expression level. Overexpressionof alcA(p)–ugmA on CMT also caused a hyper-branching phenotypethat slightly reduced colony growth rate. Taken together, repressionvs deletion morphology, and (over)expression vs wild type pheno-types were generally similar.

In A. niger, mutation of ugm1 is associated with an increase in wallalpha-glucan content (Damveld et al., 2008). We tested whetherthere was a comparable relationship between Galf and alpha-glucanin A. nidulans using qPCR to comparea-glucan synthase (ags1) levels.Both the ugmAD and alcA(p)-ugmA strain grown on repression med-ium had about fourfold increases in ags1 expression. The MOPC104EmAb has previously been used for an a-glucan dot blot assay, how-ever to date we have been unable to replicate these results at leastwith immunofluorescence. The function of Aspergillus wall a-glucanis not yet well understood (He and Kaminskyj, in preparation). Nota-bly, a triple AGS mutant in A. fumigatus was found to lack a-1,3-glu-cans (Henry et al., 2011). This triple deletion strain had a wild typephenotype on solid medium, suggesting that a-1,3-glucans may bedispensable for A. fumigatus vegetative growth (Henry et al., 2011)at least under these conditions.

4.2. Altering Galf biosynthesis-gene expression level affectsCaspofungin sensitivity

We compared sensitivity of wild type, Galf biosynthesis-genedeletion strains, and Galf biosynthesis-regulated strains for theirsensitivity to Caspofungin compared to drugs that target fungalmembranes via ergosterol or ergosterol-biosynthesis. In this study,sensitivity to Itraconazole was typical of other membrane-targetingdrugs, and was not related to Galf biosynthesis gene presence orfunction. Therapeutically, Aspergillus species are reported to be resis-tant to Itraconazole (Erjavec et al., 2009), making it a useful controltreatment. Statistical analysis compared average radii of growthinhibition, presented as an index of sensitivity with respect to wildtype.

All of the gene deletion strains were significantly more sensitivethan wild type to Caspofungin, consistent with our previous AFMstudies on the deletion strains that showed defects in wall archi-tecture and strength (Paul et al., 2011). For ugmA strains regulatedby alcA(p) and niiA(p), colonies grown in repression conditionswere at least as Caspofungin-sensitive as the knockout strains.Since we showed that deletion or repression of ugmA decreasedfksA expression, this might contribute to Caspofugin sensitivity.Unexpectedly, colonies with over-expressed alcA(p)–ugmA werealso significantly more sensitive to Caspofungin than wild type.This is consistent with the abnormal width and branching of over-expressed alcA(p)–ugmA hyphae, but the cause is not known.

Unlike ugmA, although both ugeA and ugtA, alcA(p)- and niiA(p)-repression caused Caspofungin hypersensitivity, the alcA(p)-over-expressed ugeA and ugtA strains were significantly less sensitiveto Caspofungin. Consistent with this, increasing MycobacteriumugmA expression was reported to increase resistance to isoniazid(Richards et al., 2009). Galf biosynthesis genes have both sharedand divergent functions in cell wall metabolism.

4.3. Effect of gene deletion versus gene repression

We had expected that gene deletion and repression phenotypeswould be similar for most aspects of colony and cell morphometry,since net UgmA and UgtA activity should be substrate-limited andto our current knowledge the only role of Galf is in the cell wall. Con-trast enhancement of the RT-PCR results in Fig. 1 showed that tran-scription was not abolished on gene repression media. Similarlygene over-expression was expected to have little effect due to sub-strate limitations, but found that Caspofungin sensitivity decreasedfor alcA(p)-overexpression of ugeA and ugtA. Transcript stability andtranslation of ugmA mRNA has not been explored in this study.

Galactose metabolism genes are clustered in Saccharomyces cere-visiae (Slot and Rokas, 2010). In A. nidulans the ugmA and ugtA codingsequences are AN3112 and AN3113, respectively, but the othergenes in its Leloir pathway are not clustered. Gene product interac-tion is important for many pathways, e.g. signal transduction, re-viewed in Keshet and Seger (2010). Surface topography and chargemediate protein–protein interactions. The surface of a selectablemarker product will be unlike the gene product it replaces, so it can-not participate in wild type pathway interactions. In contrast, sup-pression of gene expression, even to 1% of wild type level asestimated for alcA(p)- and niiA(p)–ugmA repression in this study,could lead to a small amount of wild type protein. Consistent withthis, some aspects of the repression morphometry phenotypes wereless severe than the deletion strains. The alcA(p)-overexpressiondrug sensitivity phenotypes also suggest that modulating wild typegene product abundance might have subtle effects.

In sum we have shown that the alcA(p)- and niiA(p)-regulatedcontrol of Galf biosynthesis provides novel information about therole of this pathway in A. nidulans. In general, the gene repressionand gene deletion phenotypes were comparable for colony growthand morphology, and overexpressed strains resembled wild type.That this was not always so, suggests possible interactions withother aspects of cell wall synthesis or maintenance. Notably, dele-tion and repression phenotypes were significantly more sensitiveto Caspofungin, suggesting that as proposed in Paul et al. (2011),their walls are significantly less robust. We remain hopeful thatan inhibitor of this pathway may be therapeutically beneficial.

Note added in proof

He and Kaminskyj (in preparation) used a modified the MOPC-104E immunostaining protocol and have demonstrated robusta-glucan localization in A. nidulans hyphal walls.

1042 M.K. Alam et al. / Fungal Genetics and Biology 49 (2012) 1033–1043

Acknowledgments

S.G.W.K. and D.A.R.S. are pleased to acknowledge shared fund-ing from the Canadian Institutes of Health Research Regional Part-nership Program and individual funding from the Natural Sciencesand Engineering Research Council of Canada Discovery Grant pro-gram. A.M.E. thanks the Egyptian Ministry of Higher Education forher Fellowship. K.A. thanks the University of Saskatchewan (UofS)for a UGS postgraduate scholarship. S.A. thanks UofS for a GraduateEquity scholarship and a UGS. We thank Merck (now Schering-Plough) for their gift of Caspofungin. Fungal Genetics Stock Centerarchives fungal strains and related resources. We thank Prof. FrankEbel, University of Munich for the gift of L10 and L99 monoclonalanti-Galf antibodies, and Dr. Loretta Jackson-Hayes, Rhodes Col-lege, Memphis TN for the alcA(p) plasmid. We thank Dr. Peta Bon-ham-Smith and Chad Stewart (UofS Biology) for assistance withthe qPCR, and Tom Bonli, UofS Geological Sciences, for SEM techni-cal assistance.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.fgb.2012.08.010.

References

Afroz, S., El-Ganiny, A.M., Sanders, D.A.R., Kaminskyj, S.G.W., 2011. The UDP-galactofuranose transporter, UgtA in Aspergillus nidulans contributes to wildtype hyphal morphogenesis and conidiation. Fungal Genet. Biol. 48, 896–903.

Bennett, J.E., Bhattacharjee, A.K., Glaudemans, C.P., 1985. Galactofuranosyl groupsare immunodominant in Aspergillus fumigatus galactomannan. Mol. Immunol.22, 251–254.

Binder, U., Oberparleiter, C., Meyer, V., Marx, F., 2010. The antifungal protein PAFinterferes with PKC/MPK and cAMP/PKA signalling of Aspergillus nidulans. Mol.Microbiol. 75, 294–307.

Bohle, K., Jungebloud, A., Göcke, Y., Dalpiaz, A., Cordes, C., Horn, H., Hempel, D.C.,2007. Selection of reference genes for normalisation of specific genequantification data of Aspergillus niger. J. Biotechnol. 132, 353–358.

Cuero, R., Ouellet, T., Yu, J., Mogongwa, N., 2003. Metal ion enhancement of fungalgrowth, gene expression and aflatoxin synthesis in Aspergillus flavus: RT-PCRcharacterization. Appl. Microbiol., 953–956.

Damveld, R.A., Franken, A., Arentshorst, M., Punt, P.J., Klis, F.M., van den Hondel,C.A.M.J.J., Ram, A.F.J., 2008. A novel screening method for cell wall mutants inAspergillus niger identifies UDP-galactopyranose mutase as an importantprotein in fungal cell wall biosynthesis. Genetics 178, 873–881.

Denning, D.W., 2003. Echinocandin antifungal drugs. Lancet 362, 1142–1151.El-Ganiny, A.M., Sanders, D.A.R., Kaminskyj, S.G.W., 2008. Aspergillus nidulans UDP-

galactopyranose mutase (ugmA) plays key roles in colony growth, hyphalmorphogensis, and conidiation. Fungal Genet. Biol. 45, 1533–1542.

El-Ganiny, A., Sheoran, M.I., Sanders, D.A.R., Kaminskyj, S.G.W., 2010. Aspergillusnidulans UDP-glucose-4-epimerase UgeA has multiple roles in wall architecture,hyphal morphogenesis, and asexual development. Fungal Genet. Biol. 47, 629–635.

Erjavec, Z., Kluin-Nelemans, H., Verweij, P.E., 2009. Trends in invasive fungalinfections, with emphasis on invasive aspergillosis. Clin. Microbiol. Infect. 15,625–633.

Espinel-Ingroff, A., 2009. Novel antifungal agents, targets or therapeutic strategiesfor the treatment of invasive fungal diseases: a review of the literature (2005–2009). Rev. Ibero Micol. 26, 15–22.

Fang, W., Ding, W., Wang, B., Zhou, H., Ouyang, H., Ming, J., Jin, C., 2010. Reducedexpression of the O-mannosyltransferase 2 (AfPmt2) leads to deficient cell walland abnormal polarity in Aspergillus fumigatus. Glycobiology 2010, 542–552.

Fisher, M.C., Henk, D.A., Briggs, C.J., Bronstein, J.S., Madoff, L.C., McCraw, S.L., Gurr,S.J., 2012. Emerging fungal threats to animal, plant, and ecosystem health.Nature 484, 186–194. http://dx.doi.org/10.1038/nature1094.

Fortwendel, J., Juvvadi, P.R., Pinchai, N., Perfect, B.Z., Alspaugh, J.A., Perfect, J.R.,Steinbach, W.J., 2009. Differential effects of inhibiting chitin and 1,3-beta-D-glucan synthesis in Ras and calcineurin mutants of Aspergillus fumigatus.Antimicrob. Agents Chemother. 53, 476–482.

Futagami, T., Nakao, S., Kido, Y., Oka, T., Kajiwara, Y., Takashita, H., Omori, T.,Furukawa, K., Goto, T., 2011. Putative stress sensors WscA and WscB areinvolved in hypo-osmoticand acidic pH stress tolerance in Aspergillus nidulans.Eukaryotic Cell 11, 1504–1515.

Gauwerky, K., Borelli, C., Korting, H.C., 2009. Targeting virulence: a new paradigmfor antifungals. Drug Discovery Today 14, 214–222.

Heesemann, L., Kotz, A., Echtenacher, B., Broniszewska, M., Routier, F., Hoffmann, P.,Ebel, F., 2011. Studies on galactofuranose-containing microstructures of thepathogenic mold Aspergillus fumigatus. Int. J. Med. Microbiol. 301, 523–530.

Henry, C., Latgé, J.P., Beauvais, A., 2011. 1,3 Glucans are dispensable in Aspergillusfumigatus. Eukaryotic Cell 11, 26–29.

Hu, W., Sillaots, S., Lemieux, S., Davison, J., Kauffman, S., Breton, A., Linteau, A., Xin,C., Bowman, J., Becker, J., Jiang, B., Roemer, T., 2007. Essential gene identificationand drug target prioritization in Aspergillus fumigatus. PLoS Pathol. 3, e24.http://dx.doi.org/10.1371/journal.ppat.0030024.

Ichinomiya, M., Motoyama, T., Fujiwara, M., Takagi, M., Horuichi, H., Ohta, A., 2002.Repression of chsB expression reveals the functional importance of class IVchitin synthase gene chsD in hyphal growth and conidiation of Aspergillusnidulans. Microbiology (Reading) 148, 1335–1347.

Jiang, H., Ouyang, H., Zhou, H., Jin, C., 2008. GDP-mannose pyrophosphorylase isessential for cell wall integrity, morphogenesis and viability of Aspergillusfumigatus. Microbiology (Reading) 154, 2730–2739.

Kaminskyj, S.G.W., 2000. Septum position is marked at the tip of Aspergillus nidulanshyphae. Fungal Genet. Biol. 31, 105–113.

Kaminskyj, S.G.W., 2001. Fundamentals of growth, storage, genetics and microscopyof Aspergillus nidulans. Fungal Genet Newsl. 48, 25–31.

Keshet, Y., Seger, R., 2010. The MAP kinase signaling cascades: a system of hundredsof components regulates a diverse array of physiological functions. MethodsMol. Biol. 661, 3–38.

Kiraz, N., Dag, I., Yamac, M., Kiremitci, A., Kasifoglu, N., Akgun, Y., 2009. Antifungalactivity of Caspofungin in combination with Amphotericin B against Candidaglabrata: comparison of disk diffusion, Etest, and time-kill methods. Antimicrob.Agents Chemother. 53, 788–790.

Kleckza, B., Lamerz, A.C., Zandbergen, G., Wenzel, A., Gerardy-Schahn, R., Wiese, M.,Routier, F.H., 2007. Targeted gene deletion of Leishmania major UDP-galactopyranose mutase leads to attenuated virulence. J. Biol. Chem. 282,10498–10505.

Kontoyiannis, D.P., Lewis, R.E., Osherov, N., Albert, N.D., May, G.S., 2003.Combination of caspofungin with inhibitors of the calcineurin pathwayattenuates growth in vitro in Aspergillus species. Antimicrob. AgentsChemother. 51, 313–316.

Lass-Florl, C., 2009. The changing face of epidemiology of invasive fungal disease inEurope. Mycoses 52, 197–205.

Latgé, J.P., 2007. The cell wall: a carbohydrate armour for the fungal cell. Mol.Microbiol. 66, 279–290.

Latgé, J.P., 2009. Galactofuranose containing molecules in Aspergillus fumigatus.Med. Mycol. 47, S104–S109, doi: http://dx.doi.org/10.1080/13693780802258832.

Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression datausing real-time quantitative PCR and the 2-DDCT method. Methods 25,402–408.

Mircus, G., Hagag, S., Levdansky, E., Sharon, H., Shadkchan, Y., Shalit, I., Osherov, N.,2009. Identification of novel cell wall destabilizing antifungal compounds usinga conditional Aspergillus nidulans protein kinase C mutant. J. Antimicrob.Chemother. 64, 755–763.

Monteiro, M.C., DeLucas, J.R., 2010. Study of the essentiality of the Aspergillusfumigatus triA gene, encoding RNA triphosphatase, using the heterokaryonrescue technique and the conditional gene expression driven by the alcA andniiA promoters. Fungal Genet. Biol. 47, 66–79.

Osmani, A.H., Oakley, B.R., Osmani, S.A., 2006. Identification and analysis ofessential Aspergillus nidulans genes using the heterokaryon rescue technique.Nat. Protoc. 1, 2517–2526.

Paul, B.C., El-Ganiny, A.M., Abbas, M., Kaminskyj, S.G.W., Dahms, T.E.S., 2011.Quantifying the importance of galactofuranose in Aspergillus nidulans hyphalwall surface organization by atomic force microscopy. Eukaryotic Cell 10, 646–653.

Pederson, L.L., Turco, S.J., 2003. Galactofuranose metabolism: a potential target forantimicrobial chemotherapy. Cell. Mol. Life Sci. 60, 259–266.

Richards, M.R.L., Lowary, T.L., 2009. Chemistry and biology of galactofuranose-containing polysaccharides. ChemBioChem 10, 1920–1938.

Romero, B., Turner, G., Olivas, I., Laborda, F., Roberts, E.M.D., Lucas, J.R.D., 2003. TheAspergillus nidulans alcA promoter drives tightly regulated conditional geneexpression in Aspergillus fumigatus permitting validation of essential genes inthis human pathogen. Fungal Genet. Biol. 40, 103–114.

Ronen, R., Sharon, H., Levdansky, E., Romano, J., Shadkchan, Y., Osherov, N., 2007.The Aspergillus nidulans pkcA gene is involved in polarized growth,morphogenesis and maintenance of cell wall integrity. Curr. Genet. 51, 321–329.

Schmalhorst, P.S., Krappmann, S., Vercecken, S.W., Rohde, M., Muller, M., Braus,G.H., Contreras, R., Braun, A., Bakker, H., Routier, F.H., 2008. Contribution ofgalactopyranose to the virulence of the opportunistic pathogen Aspergillusfumigatus. Eukaryotic Cell 7, 1268–1277.

Shibata, N., Saitoh, T., Tadokoro, Y., Okawa, Y., 2009. The cell wall galactomannanantigen from Malassezia furfur and Malassezia pachydermatis contains b-1,6-linked linear galactofuranosyl residues and its detection has diagnosticpotential. Microbiology (Reading) 155, 3420–3429.

Slot, J.C., Rokas, A., 2010. Multiple GAL pathway gene clusters evolvedindependently and by different mechanisms in fungi. Proc. Natl. Acad. Sci.USA 107, 10136–10141.

Stynen, D., Sarafati, J., Goris, A., Prevost, A.M., Lesourd, M., Kamphuis, H., Darras, V.,Latge, J.P., 1992. Rat monoclonal antibodies against Aspergillus galactomannan.Infect Immun. 60, 2237–2245.

Szewczyk, E., Nayak, T., Oakley, C.E., Edgerton, H., Xiong, Y., Taheri-Talesh, N.,Osmani, S.A., Oakley, B.R., 2007. Fusion PCR and gene targeting in Aspergillusnidulans. Nat. Protoc. 1, 3111–3120.

M.K. Alam et al. / Fungal Genetics and Biology 49 (2012) 1033–1043 1043

Tribus, M., Bauer, I., Galehr, J., Rieser, G., Trojer, P., Brosch, G., Loidl, P., Haas, H.,Graessle, S., 2010. A novel motif in fungal class 1 histone deacetylases isessential for growth and development of Aspergillus. Mol. Biol. Cell 21, 345–353.

Upadhyay, S., Shaw, B.D., 2008. The role of actin, fimbrin and endocytosis in growthof hyphae in Aspergillus nidulans. Mol. Microbiol. 68, 690–705.

Walker, L.A., Gow, N.A.R., Munro, C.A., 2010. Fungal echinocandin resistance. FungalGenet. Biol. 47, 117–126.

Waring, R.B., May, G.S., Morris, N.R., 1989. Characterization of an inducibleexpression system in Aspergillus nidulans using alcA and tubulin-coding genes.Gene 79, 119–130.

Zarrin, M., Leeder, A.C., Turner, G., 2005. A rapid method for promoter exchange inAspergillus nidulans using recombinant PCR. Fungal Genet. Biol. 42, 1–8.