Fungal Cil'lWlics illlli Biolog,· 21. 1,-2'1 11 qq"

:\ni('h' :'\(L FGH,< )~),7( I

Supplied by U.S. Dept. of Agric., National Centerfor Agricultural Utilization Research, Peoria, IL

REVIEW

Metabolic Pathway Gene Clustersin Filamentous Fungi

Nancy P. Keller* and Thomas M. Hohnt

*Department of Plant Pathology and Microbiology. Texas A&M University. College Station. Texas 77843-2132: andtMycotoxin Research Unit. National Center for Agricultural Utilization Research. USDAIARS. Peoria. Illinois 61604

Accepted for publication February 7. 1997

7711

Keller, N. P., and Hohn, T. M. 1997. Metabolic pathwaygene clusters in filamentous fungi. Fungal Geneticsand Biology 21 ,17-29. ;: 1997 Academic Press

Index Descriptors: Aspergillus nidulans; filamentousfungi; gene clusters; metabolic.

Clusters of functionally' related genes are a generalfeature of prokaryotic gene organization but are much lessprevalent in eukaryotes. The discovery that genes forcertain types of metabolic pathways are clustered infilamentous fungi is relatively recent. Only 8 years ago in adiscussion of the proline utilization pathway genes, Hull et

al. (1989) commented that "The organization and regula­tion of the genes involved in L-proline catabolism in theascomycete fungus Aspergillus nidulalls are of particularinterest because, rather unusually for functionally relatedeukaryotic genes. they are all clustered." This view wassupported by studies of metabolic pathway gene linkagerelationships from a variety of eukaryotes. Although ge­netic e\idence was available suggesting that some meta­bolic pathway genes were closely linked in A Ilidulalls (seeClutterbuck 1992), recognition of gene clustering as animportant feature of fungal metabolic pathways had toawait the molecular characterization of specific pathways.

Fungal gene clusters can be broadly defined as the closelinkage of two or more genes that participate in a commonmetabolic or developmental pathway. Fungi possess numer­ous pathways for what can be described as "dispensable"metabolic functions. and research over the last 5 veal's has

1087,1845/97 $25.00Copyright 1997 by Academic PressAll rights of reproduction in any form reserved.

shO\\11 that the genes for these dispensable pathways areoften organized in gene clusters. The term dispensablemetabolic pathways is used here to describe pathways thateither are not required for gro\\th or are onl;' reqUired forgro\\th under a limited range of conditions. Dispensablemetabolic pathways are typically e:-.pressed under subopti­mal grO\\th conditions and most likeI;' function to enhancefungal survival in response to nutrient deprivation orcompeting organisms. In this re\iew. two major t;1)es ofdispensable metabolic pathways \\ill be discussed. cata­bolic pathways for the utilization of low-molecular-weightnutrients such as proline or quinate (nutrient utilizationpathways) and biosYllthetic pathways for low-molecular­weight compounds which include antibiotics and mycotox­ins (natural product pathways).

Nutrient utilization pathways increase the metabolicversatilit;, of filamentous fungi. enabling them to utilize avariety of complex compounds as alternative sources ofnutrients. E:-.pression of the appropriate catabolic path­ways can be critically important for sunival under limitingnutrient gro\\th conditions. However, many of these nutri­ents are not commonly' encountered by fungi. so there islittle benefit derived from e:-.pressingpathway genes consti­tutively To ensure that the reqUired catabolic pathways areexpressed appropriately in response to changing nutri­tional conditions while Simultaneously limiting the loss ofcellular resources due to unnecessary pathway gene e:-.pres­sion. fungi have developed complex regulatory systems. Todate. enzymatic and regulatory genes for four well-studieddispensable catabolic pathways (quinate, ethanol, proline.

17

18

anclnitrate utilization i have been found to be clustered innuious Hlamentous fungi. All of these pathwa:'s are knO\\11

to be clustered in A. Ilidll!alls. but \\ith onl:' a few exceptions.little inforn1ation is available conceming their presence in otherfungi. In A. Ilidll!alls, patterns have emerged concerning theregulation of nutrient utilization pathways. Pathway regulationoften involves at least two levels of control that have beenreferred to ac; ~narrow" and "broad" domain regulation (Scazzoc­cllio. 1992), Narrow domain regulation is typically mediated bya positive acting pathway-specific regulatory protein of theCX~CXr,CXr,CX~CXr,CX~ zinc binuclear cluster type. whilebroad domain regulation employs the positive actingnitrogen regulator:' gene. areA (Kudla et al., 1990), and/orthe negative acting carbon catabolite repressor gene, ereA(Dowzer and Kelly. 1991). This multilevel regulationensures that nutrient utilization pathways can respond toboth the demands of general cellular metabolism and thepresence of specific pathway inducers. However, it re­mains to be seen whether similar regulation strategies areused by other fungi.

Natural product pathways result in the production ofstructurallv diverse metabolites also referred to as second­ar:' metabolites. In contrast to nutrient utilization path­ways, the function of natural product pathways is in manyinstances obscure. Recent progress in the molecular char­acterization of natural product pathways indicates that

maintaining these pathwa:'s requires a substantial im'est­ment of fungal genetic resources. It is not uncommon forfungi to produce a number of different natural productsand a single natural product pathwa:' can consist of as manyas 25 different genes and occup:' up to 60 kb of DNA (Fig.1. Brown et al.. 199.5b)' One proposed function tt)r somenatural products is that the:' contribute to some aspect oforganism ecology. Evidence supporting this function hasrecently come from a number of studies implicatingnatural products in specific fungal-plant interactions (Des­jardins and Holm, 1997; Walton. 1996). Regulating theexpression of natural product pathwa:'s would appear topresent many of the same problems to cells as theregulation of nutrient utilization patl1\vays. Currently. stud­ies of natural product pathwa:' regulation are not asadvanced as those for nutrient utilization pathwa:'s. E\i­dence for both "narrow" and "broad" domain regulationis not available for a single natural product pathway.although some pathwa:'s have been reported to have eitherone or the other. Positive acting pathwa:'-speciHc regula­tors are known for the trichothecene (Proctoret af.. 1995b)and aflatoxinlsterigmatocystin (Pa~me et al., 1993; Woloshuket al., 1994: I'u et al., 1996) patl1\vays, while generalmetabolic pathway regulators participate in the regulationof penicillin pathways (Haas and Marzluf. 1995; Suarezand Penalva, 1996; Tilburn et al., 1995). Preliminar:' data

kb 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75

A B CDE alIR FGB I J K LMN o P Q R STU VW x

ST ... .. ~....... .......... , I'" l ................ ~l -AF .... I .......... ..... ... fiJi" .. ... ... ... •

stcA =polyketide synthasesteB =P450 monooxygenasesteC =oxidasesteD = unknownstcE =dehydrogenase/reductaseaflR =zinc cluster transcription factorsteF =P450 monooxygenasesteG =dehydrogenasestcH =unknown

stcI =esterasestc] =fatty acid synthase asteK =fatty acid synthase ~stcL =P450 monooxygenasesteM =unknownsteN =dehydrogenase/reductasestcO =A. parasiticus cluster genesteP =O-methyltransferase

steQ =A. parasiticus cluster genesteR =unknownsteS =P450 monooxygenasesteT = elongation factor gammasteD =ketoreductasesteV =dehydrogenase/reductasesteW= monooxygenasesteX = unknown

FIG. 1. Order and direction of transcription of homologous genes in the sterigmatocystin and aflatoxin gene clusters. Homologous genes in thesterigmatocystin (ST) and aflatOxin (AF) gene clusters are indicated by the same color. Black bars represent ST and AF genes which have not yet beencompared. Likely transcription direction has been determined for those genes depicted with an arrow. Data are based on A. llidlllallS ST (Brown ct al..

1996) and A parasiticus AF cluster sequence and cosmidltnmscript maps (D. Bhatnagar, personal communication. and Trail ct al., 1995: Yu ct al., 1995a.

The relative spacing between AF cluster genes is not accurate as complete sequence and transcript mapping is currently in progress). ST cluster transcriptsare desig;natedstc (sterigmatocystin cluster) A to X with the exception of the sLxth gene in the cluster, afiK which is the conserved ST/AF regulatory gene.Putative stc gene acti\ities are as indicated. The figure is derived from Fig. 1. reprinted \\ith pennission of the publisher. copyright 1996, National Academy

of Sciences. USA, and unpublished data. courtesy of D. Bhatnagar.

Copyright ~ 1997 by Academic PressAll rights of reproduction in any form reserved.

Rt'view

indicate that the regulation of natural product pathwa:'sappears similar in man:' respects, to that observed fornutrient utilization pathwa:'s, but may also include develop­mental regulatory controls (Kale et aI" 1996; Keller et al.,1995a).

Progress in the characterization of dispensable pathwaysis now sufficient to allow meaningful generalizations con­cerning gene structure and gene organization within path­way gene clusters (Table 1). For example, it appears thatclustered genes are typicall:' indistinguishable from otherfungal genes. Genes within pathway gene clusters usuall:'display little variance in codon usage patterns and oftencontain intervening sequences typical of filamentous fun­gal genes. Analysis of gene arrangements in dispensablepathway gene clusters has not revealed specific biases ingene organizational patterns \vithin clusters. The func­tional categories of genes found in dispensable metabolicpathway gene clusters include genes encoding enzymes,transcription factors, and transporters. While some path­way gene clusters contain all three functional gene types, inothers it is unclear whether pathway-specific transcriptionfactors and transporters are closely linked or if they evenexist. The functional organization of fungal dispensablepathways invites comparisons with related pathways inprokaryotes. However, there is little evidence that fungalpathways have prokaryotic counterparts or that the cluster­ing of genes has some intrinsic importance for pathwaygene regulation.

TABLE I

Phvsical Characteristics of Selected Fungal Gene Clusters

19

The purpose of this review is to briefly summarize ourcurrent J..."1lowledge of selected dispensable metabolic geneclusters and to discuss the possible significance of geneclustering for pathway regulation and evolution. Pathwa:'gene clusters are described below with respect to availableinformation on their distribution, size, indhidual genefunctions, and regulation.

I. NUTRIENT UTILIZATION PATH\VAYCLUSTERS

lao QuinarelSllikimate Pathway

Quinate and shikimate can be metabolized by a variet:·of fungi as alternative carbon sources. Metabolism of thesecompounds leads first to protocatechuic acid and then tosuccinate and acetate (Griffin, 1994). Shikimic acid is alsoreqUired for aromatic amino acid biosynthesis. the sourceof numerous natural products including the ergot alkaloids(Griffin, 1994). Both pathways are glucose repressed andshare two common metabolic intennediates. 3-dehvdro­quinic acid and dehydroshikimic acid. These two com­pounds, as well as quinate itself, stimulate transcription ofquinate and shikimate pathway genes.

The genes encoding the regulatory and structural pro­teins for quinate catabolism are clustered in both NClIro-

Gene functional tIpes

Cluster Cluster size (kb) Number of genes Regulatory Enzvrne Transport Unl·.oIO\\11

Nutrient clustersqa 17.3 7 2 3 1 1Nitrate >12a 3 0 2 1 0Proline 13 5 1 2 1 1Ethanol -15a 7 1 0 5

Natural product clustersAflatoxin -75a >170 1 >l1a 0 >5°Sterigmatocystin -60" 25 1 -16" 0 -8a

Trichothecene -25° 9 1 6 1 1Penicillin -20" 3 0 3 0 0Melanin -30" 3 0 3 0 0

Note. Data are summarized from the following published deSCriptions: melanin cluster/Altemaria altemata (Kimura and Tsuge. 1993); nitrate, proline,ethanoL sterigmatocystin, and penicillin clusters/Aspergillus nidulam (Bro\\11 et al.. 1996b; Fillinger and Felenbok, 1996; Hull et al., 1989; Johnstone et al.,1990; Martin and Gutierrez, 1995): aflatoxin cluster/Aspergillus parasitiClls (D. Bhatnagar, personal communication); trichothecene cluster/Fusariu11lsporotrichioides (Hohn et al., 1995b, and unpublished); qa clusterlNeurospora crassa (Giles et al., 1989); penicillin cluster/Penicillium chnJSogellu11l (Diezet al.. 1990). Additional references can be found in the text.

a Cluster size (kb), number of genes/cluster, and function of each gene are not completely characterized for some clusters.

Copyright ~ 1997 by AcademiC PressAll rights of reproduction in any form reserved.

20

spora erassa (Giles et al.. 1989) and A. nidlilalls (Lamb ctal.. 1990). Gene function but not the order or direction oftranscription is conserved in these two clusters. The17.3-kb seven-gene cluster in N. erassa consists of tworegulatory genes encoding an autoregulated activator (Qa­IF. containing a binuclear zinc cluster motif) and arepressor (Qa-lS). three structural genes encoding quinate­inducible enZ:l11es \qa-2, qa-3, and qa-4), one quinatepermease encoding gene (qa-Y), and one gene. qa-X, ofunknO\\11 function. Qa-lF binds to a s:V111metrical 16-bpupstream activating site (GGRTAA RYRY ITATCC) foundone or more times ,5' to each of the seven genes.Biochemical evidence suggests that transcriptional controlis mediated through a complex interaction among Qa-1F,quinate. anel Qa-lS. The role of Qa-lS as a repressor issupported by isolation of Qa-1S constitutive mutants\qa-1SC

) (re\iewed by Geever et al., 1987). An analogoussystem exists in A. Ilidlilans (Hawkins et al., 1993, 1994)including a similar regulatory function ascribed to theQuWQutR association (Lamb et al., 1996). Repressorgenes have not been founel in other dispensable pathwayclusters, and their presence in the quinate pathways isreminiscent of metabolic pathway regulation in bacteria.

Interestingly, in A. nidlilans, sequences closely related tothe c{lItE gene product, a dehydroquinase, also constituteone domain of ARO.tvl, a pentafunctional protein involvedin quinate bios)TIthesis (Charles et al., 1986; Griffin, 1994,and references therein). This is one ofseveral observationslinking parallel nutrient utilization anel bios)TIthetic path­way gene products. Hawkins et al. (1993, 1994) havereported amino acid sequence similarities between por­tions of ARo.M and the quinate regulatory proteins andsuggested that qutA and qutR arose from duplication of aregion \vithin aTOm. Similar instances of amino acidhomologies between pathway-speCific regulatory proteinshave been suggested for the nitrate utilization pathway andanthranilate sVTIthase in IV. erassa and A. nidlilans as wellthe ethanol utilization regulator, AlcR, and the alcohol andaldehyde dehydrogenase proteins in A. nidlilans. This hasled to a controversial hypothesis that transcriptional regula­tory genes for some dispensable pathways evolved throughthe recruitment of functional domains from complemen­tary metabolic pathway genes (Hawkins et al., 1993,1994).

lb. Etl1anol Utilization

Like quinate, ethanol is an alternative carbon source thatis not utilized by all fungi. Early research on ethanolmetabolism focused on the industrial importance of itsproduction and utilization by yeasts, but several filamen-

Copyright 'C 1997 by Academic PressAll rights of reproduction in any form reserved.

tous fungi also metabolize alcohols. Ethanol can be oxi­dized to acetate via acetaldeh)·de and thereby enter variousintermediate metabolic pathwa)·s. In A. nidlilalls. ethanolcatabolism is dependent on two enZ\1nes: alcohol dehvdro­genase I (ADHI)l encoded b)' aleA 'and aldeh)'de deh;'dro­genase (AldDH) encoded b)' aldA. Both enZ:ll1es. along\vith the positive acting pathway regulator. AlcR. areinduced by the presence of ethanol and a number ofgratuitous inducers including eth)·lmeth)·1 ketone andthreonine (Felenbok. 1991, and references therein).

The aldA gene is not linked \vith the other ethanolutilization genes, aleA and aleR (Felenbok. 1991). HO\v'­ever, aleA and aleR are both located \vithin an approxi­mately 15-kb region on chromosome \11 that includes fiveother genes. Gene order for this region has been estab­lished as aleF, aleR, aleO, aleA, aleM, aleS followed bvaleU(Fillinger and Felenbok, 1996). Although onl)' aleR andaleA are necessary for ethanol utilization. all the pathwaygenes are subject to ethanol induction and CreA repression(\vith the possible exception of aleU). aleM, aleS, and/oraleU genes are reported to be necessary for grO\\th onethanolamine and for the oxidation of another alcohol.propan-1-01 (Fillinger and Felenbok, 1996).

Two transcription factors, AlcR and CreA, have beenshO\vTI to regulate the ex-pression of ethanol utilizationgenes. Both AlcR, another binuclear cluster transcriptionfactor, and CreA have specific binding sites in the aleR andaleA promoters (Kulmburg et al., 1992, 1993). In someinstances, these sites overlap or lie close together in bothpromoters, and it appears that CreA mediates repressionthrough binding site competition \vith AlcR (Kulmburg etal., 1993; Mathieu and Felenbok, 1994). AlcR representsone example of a pathway-speCific regulator that mediatesits O\\TI ex-pression. This phenomenon has also beenobserved \\ith QutA (Section Ia). Noteworthy too is thefact that the aleA promoter has become an important toolin fungal molecular biology, where it is frequently used toregulate the ex-pression of heterologous proteins (Felen­bok, 1991).

Ie. Proline Utilization

Proline differs from both quinate and ethanol because itcan serve as either a carbon or a nitrogen source. Thecellular management of proline metabolism is further

J Abbre\1ations used: ADHI. alcohol dehydrogenase: AldDH, alde­hyde dehydrogenase; PKS, polyketide S}llthase; FAS. fatty add synthase;DHN.1.8-dihydrox}llaphthalene.

R('vit'w

complicated b;' the fact that several of the intermediatesare either shared b;' or feed into·arginine and glutamatemetabolism. Proline utilization pathways are biochemicallyidentical in S. cereUsiae and A. Ilidlllalls (Brandriss and!vlagasanik. 1979); however. pathway gene organization isdispersed in S. cerer.:isiac and clustered in A. Ilidlllalls(Brandriss and \lagasanik. 1979: Hull et al., 1989).

The proline utilization cluster of A. llidlllallS is com­posed of five coregulated transcripts spread over a regionof 13 kb. Gene order has been established as pmA, pmX,pmD, pmB, and pmC (Hull et aZ., 1989: Scazzocchio,1992l. PmA. like all of the pathwa;'-specific transcriptionfactors described in this section. is a zinc binuclear clustertranscription factor and positively regulates ex-pression ofthe other four genes (Shanna and Arst. 1985). Characteriza­tion of the pathway gene cluster has shown that pmA isbordered by an uncharacterized gene not under pmAcontrol that may mark one end of the cluster (Scazzocchio,1992). PmD and PmC are enzymes (an oxidase anddehydrogenase, respectively) required for proline utiliza­tion and PmB is a pennease required for transport ofL-proline (Jones et aZ., 1981l. The function of PrnX has notbeen detennined.

Since proline can be utilized as either a nitrogen or acarbon source it is not surprising that the proline utilizationgenes are regulated by both CreA and AreA (Arst andCove. 1973; Scazzocchio. 1992, and references therein).This joint regulation is complicated by the conditionalrequirement for AreA regulation that is in tum dependenton whether active creA product is present. Detailedcharacterization of the shared pmDlpmB intergenic pro­moter region indicates that a region called ADA (forabsolute dependence on AreA) is essential for integrationof creA and O1'eA regulation; removal of ADA eliminatescreA regulation (c. Scazzocchio, personal communica­tion). Interestingly, deletion of a 290-bp region containingthe ADA element has been found to alter ex-pression of theneighboring gene, pmC (C. Scazzocchio, personal commu­nication). This observation may ex-plain the "at a distance"regulation of pmC previously described by Arst andco-workers, in which some pmDlpmB promoter deletionsexiending into pmB also affected pmC ex-pression (Arst etaZ., 1981; Sharma and Arst, 1985). The precise mechanismof how deletions in other parts of the proline clusterinfluence pmC ex-pression has yet to be elucidated.

ld. Nitrate Assimilation

Most fungi. although not S. cerevlszae, can readilyassimilate nitrate and reduce it to ammonium. This assimi-

21

lation pathwa;' has been intensivel;' studied in both A.nidllZalls and S. crassa and requires assembl;' of a Cof~lctor.

a pemlease to transport nitrate into the cell. and a nitrateand nitrite reductase. Standard genetic studies of A.nidllZans had earlier indicated that the permease andreductase genes for this pathwa;' were linked (for a reviewsee Arst and Scazzocchio. 19851. These linkage studieswere confirmed by a 1990 report describing the cloningand characterization of this three-gene cluster (Johnstoneet aZ., 1990). The three genes. ordered C1'IlA.. lliL4.. andiliaD, encode for the permease. nitrite reductase. andnitrate reductase. respectively.

\Vithin the nitrate assimilation cluster. niL4. and iliaD aredivergently transcribed and share some but not all cisacting factors. For instance. both genes are under thecontrol of NirA. the positiveI;' acting specific pathwayactivator (Burger et aZ., 1991), but the four NirA bindingsites are not equally important for niiA and niaD eX1Jres­sion (Punt et aZ., 1995). Scattered throughout this regionare also several AreA sites; only those which are centrallylocated appear to have physiological importance (Punt etaZ., 1995). Regulation of nitrate assimilation in N. crassa isex-pected to be similar. Chiang and Marzluf (1995) proposetllat a Nit2lNit4 interaction (AreA and NirA homologs.respectively) might be significant for ex-pression of lIit-3(iliaD homolog).

Although homologs of A. llidllZans nitrate assimilationand metabolism genes have been identified in several fungi(Diolez et aZ., 1993; Unkles et aZ., 1992) and the oomycetePhytophthora infestalls (Pieterse et oZ., 199.5). lliL4. andiliaD linkage is not always conserved. Still. the fact thatlliiA and iliaD clusters have been reported in the yeastHallsellllZa poZY17lorpha (Brito et al., 1996) is evidence thata nutrient utilization pathway cluster can be ,\idespread inthe fungal kingdom.

II. NATURAL PRODUCT PATHWAYS

lla. Penicillin

Penicillins and cephalosporins are j?>-lactam containingantibiotics that are produced by members of several fungalgenera and cephalosporin in a variety of prokaryotes.Penicillin biosynthesis begins with the fonnation of thetripeptide, o-(L-a-aminoadipyl)-L-cysteinyl-D- valine (ACV),and is then followed by the cyclization of ACV to yieldisopenicillin N (IPN). IPN is the branch point for thevarious penicillin and cephalosporin pathways. Penicillin

Copyright" I 997 by Academic PressAll nghts of reproduction in any form reserved.

22

pathwa~' genes are clustered in Pellicillill11l chrysogellll11l(Diez et al., 1990) and A. nidllialls (.MacCabe et al., 1990).while cephalosporin pathway genes are clustered in Acre­17l0nilll7l chrysogellll11l (s}n. Cephalosp0rilll7l acrenwnium,Mathison et a!., 1993). In P chrysogenum and A nidulallsthe ACV synthetase (pcbAB), IPN s}nthetase (pclJC), andacvltransferase (penDEl genes are contained within are~on of approximately 20 kb in the same order andorientation (Martin and Gutierrez, 1995). In A chnJsoge­IWIIl, the pcbAB and pcbC genes are present on a fragmentof about 15 to 18 kb and in the same orientation as thecorresponding genes in P chrysogenllm and A nidllians.Cephalosporin C pathway-specific genes have also beenisolated from A chrysogenll11l (Mathison et al., 1993). ThecefEF (expandase/hydrm.:ylase) and cefC (acetyltransfer­ase) gene products catalyze the last two steps in thepathway and are closely linked to each other but are notlinked to the pcbAB and pcbC genes. In A chrysogenum,the pcbAB and pcbC genes reside on chromosome VI,while the cefEF and cefC genes are located on chromo­some II (Mathison et al., 1993).

A pathway-specific regulator has not been identified forthe penicillin pathway, although recently a protein, PENRl,has been implicated in such a role (Bergh et al., 1996).J\'lore thoroughly characterized is the role of PacC inregulating penicillin synthesis. It has been shO\m thatPacC, a C2H2-type zinc finger transcription factor involvedin activation of alkaline-ex-pressed genes, is required toactivate at least one gene, ipllA, in the A nidlilanspenicillin pathway (Espeso et al., 1993; Tilburn et al.,1995). There are also weak CreA and CreC effects onpenicillin cluster regulation in A nidulans (Espeso et al.,1995). Interestingly, carbon catabolite repression has stron­ger effects than PacC regulation on penicillin bios}nthesisin P. clmJsogenll11l (Suarez and Penalva, 1996).

Comparisons between the penicillin/cephalosporin path­ways in the fungi and prokaryotes have led to the hypoth­esis that portions of these pathways were acquired throughhorizontal transfer from a prokaryote (Penalva et al., 1990;Weigel et al., 1988). Fungal penicillin and cephalosporinpathway gene products are closely related; the IPN s}nthe­tases from A clmJsogenllm, P. chrysogenum, and Anidulans display about 80% identity. Surprisingly, compari­sons between fungal and prokaryotic IPN synthetasesrevealed approximately 60% identity (Penalva et al., 1990).The high degree of relatedness between fungal andprokaryotic IPN genes has been cited as evidence insupport of the horizontal transfer of pathway genes (Lan­dan et a!., 1990; Penalva et a!., 1990; Weigel et a!., 1988).

Copyright 0 1997 by Academic PressAll rights of reproduction in any form reserved.

Kl'vil'\\"

However. this analysis has been challenged based on therelativel}' small data set employed and because it is basedsolely on assumptions about rates of change in genesequences (Smith et a!., 1992).

lIb. Tricl1otl1ecenes

Trichothecenes are sesquiterpenoid m}'cotoxins pro­duced by several genera of fUamentous fungi. In additionto their vertebrate toxicity, trichothecenes have beenimplicated as virulence factors in some plant diseases(Desjardins et al., 1993). Production of trichothecenes byF graminearum was recently shO\\n to be required for highlevels of virulence in Fusarium wheat head scab (Desjar­dins et al., 1996; Proctor et al., 1995a). Analysis of thetrichothecene pathway gene cluster in F sporotricllioideshas so far resulted in tile identification of nine genes withina 25-kb region (Hohn et al., 1995b). The functions of eightpatllway gene products have been determined and includesix biosynthetic enZJwes, a transcription factor. and atransport protein. Of the six knO\m biosynthetic enZJ11les.two are cytochrome P450s that function eitller in tileaddition of a hydroxyl group at C-15, TRIll (Hohn et a!.,unpublished), or in the ox')rgenation of trichodiene (TRI4)to yield a product of unknO\m structure (Holm et al.,1995a). Both pathway CJtochrome P450s represent new'gene families within the P450 superfamily and are mostclosely related to other fungal P450s. The remainingenZJTI1e encoding genes in tile cluster include trichodienesynthase (Hohn and Beremand, 1989), the branch pointstep in the pathway catalyzing tile cyclization of famesyldi­phosphate, and three acetyltransferases that are involvedin the acetylation of the trichothecene hydroxyl groups(McCormick et a!., 1996). The cluster gene, Tri6, encodesa transcription factor (Proctor et al., 1995b) and is requiredfor pathway gene ex-pression. TRI6 contains sequencemotifs similar to the C2H2-type zinc fingers commonlyfound in many eukaryotic DNA binding proteins. Finally, agene (Tri12) encoding an apparent transport protein hasbeen identified (Hohn et a!., unpublished).

Macrocvclic trichothecenes are structurelv more com-./ ~

plex and frequently more cytotoxic than Fusariumtrichothecenes (Jarvis, 1991). Macrocyclic trichothecenesare characterized by the presence of a macrocycle ofvariable structure esterified through the trichotheceneC-15 and C-4 hydroxyl groups. Macrocyclic trichothecenepathway genes have been studied in the fUamentousascomycete, Myrothecium roridum, and sho\\TI to beorganized in a gene cluster. Apparent homologs of theFusarium pathway genes Tri4, Tri5, and Tri6 have been

identified \\ithin a 40-kb region in M. roridlllll (Holm etal., unpublished). One of tilese, ''MrTli4, has also beensho\\11 to be functionally identical to its suspected FlI­sarill1ll homolog through the complementation of a Tri4­mutant in F. sprorotrichioides. The macrocyclictrichothecene cluster differs significantly from the FlI­sarilllll trichothecene cluster \\ith respect to both thedistances between genes and pathway gene orientations,While the Tri4 and TriS genes are separated by about 8 kbof DNA in F. sporotlichioides, the MrTri4 and MrTriSgenes are almost 40 kb apart in AI. r01idll11!. In addition,transcription of the Tri4 and Tli6 genes occurs in adivergent fashion in F. ,sporotrichioides, but the MrTri4and MrTli6 genes are arranged so that transcription occursconvergently. Comparisons between the correspondinggene products in the two pathways indicate sequenceidentities ranging between 4.5 and 7.5%.

Ile. Aflatoxins and Sterigmatocystins

Aflatoxins and sterigmatocystins are a group ofpolyketidemycotoxins derived from the same biochemical pathway(Bennett and Papa, 1988; Brown et al., 1996b). Thesecompounds are produced by several genera, most notablythe genus Aspergilllls. Although long thought to be gener­ated only by a polyketide synthase (PKS), it has beenrecently established that aflatoxin and sterigmatocystinrequire both a PKS and a fatty acid synthase (FAS) to formthe first stable intermediate of the pathway (Brown et al.,1996a; Mallanti et al., 1995; Watanabe et al., 1996). TheFAS, composed of an a and a 13 subunit. likely produces aunique si:'\-carbon fatty acid which serves as the starter unitfor the PKS.

In the 1980s, studies of A. parasiticus and A. jlavlIsemplo;ing pathway mutants had established the likelyclustering of at least some of the genes in the aflatoxinpathway (for a review see Bennett and Papa, 1988). Thishas subsequently been confirmed by molecular geneticstudies. The complexity of this cluster is most thoroughlyillustrated by the characterization of the A. nidulanssterigmatocystin gene cluster (Fig. 1; Brown et al., 1996b).A sequencing and mRNA mapping investigation of this60-kb region revealed 25 coordinately regulated transcriptswhich accumulate speCifically when grO\\th conditionsfavor sterigmatocystin production. Most of the clustergenes appear to encode enzymes with the necessaryfunctions predicted for aflatoxin and sterigmatocystin bio­svntllesis. These functions include the PKS and FASmentioned earlier, five monooll:ygenases, several reducta­ses, dehydrogenases, a methyltransferase, and an esterase

23

(Brm\11 et al.. 1996b!. Site-directed mutations in nine ofthese genes have unambiguousI;' placed their position inthe gene cluster (Brm\11 et a!.. 1996a: :Kelkar ct al., 1996.1997; :Keller et al.. 1994, 1995b: Yu and Leonard. 19951.Other genes which have been dismpted include std, stcX.stcQ, stci: and stcW (Kelkar, Adams, and :Keller. unpub­lished), and their function is currentl;· being determined.Several genes in the A. parasiticlls cluster have also beencharacterized (e.g., Chang et al., 1995: Mahanti et al.,1995: Silva et al., 1996: Skor" et a!.. 1993: Woloshuk ct al..1994) and show functional and stmctural similarities tohomologs in A. nidulans. A comparison of the sterigmato­cystin and aflatoxin cluster genes shows that gene orderand direction of transcription have not been perfectl;·conserved (Fig. 1: Brown et al., 1996b: Trail et al., 1995: Yuet al., 1995).

All three clusters contain a cop;' of a gene, aflR,encoding a binuclear zinc cluster transcription factor(Chang et al., 1995; Woloshuk et al., 1994; Yu et al., 1996).Several lines of evidence support the hypothesis that AflRfunctions as a sequence-specific DNA binding protein thatis required for pathway gene ell:pression in both sterigmato­cystin and aflatoxin pathways (Chang et al., 1995: PaYlle etal., 1993; Yu et al., 1996; Fernandes, Keller, and Adams,unpublished). Sterigmatocystin and aflatoxin pathwa;' geneex-pression is also pH regulated (Keller et al., 1997).GrO\\th medium pH differentially affects penicillin andaflatoxinlsterigmatocystin production such that acidic pHfavors mycotoxin production (Keller et al., 1997) andalkaline pH favors penicillin production (Espeso et al.,1993; Tilburn et al., 1995). Unlike penicillin, a formal rolefor pace regulation of the aflatoxinlsterigmatocystin geneclusters has not been established, although several putativePacC consensus motifs are present in these clusters,

Ild, Melanin

Melanin is a high-molecular-weight, dark brown or blackpigment produced by numerous fungi. The biosynthesis ofsome fungal melanins begins with the synthesis of thepolyketide intermediate, 1,3,6,8-tetrahydroxynaphthalene,which is converted to 1,8-dihydroxynaphthalene (DHN)after two cycles of dehydration and reduction. DHN maybe then oxidized and polymerized to form melanin (Belland \Vheeler, 1986). Melanin contributes to the survival offungal propagules (Bell and Wheeler, 1986) and the abilityof certain fungal plant pathogens to penetrate host plantcells (Kubo and Fumsawa, 1991).

Genes involved in melanin smthesis are clustered insome fungi and not in others. A melanin pathway gene

Copyright l) 1997 by Academic PressAll rights of reproduction in any form reserved.

24

duster has been identified in Alterllaria altemata thatcontains at least three pathwa:-' bios:-llthetic genes withina 30-kb region (Kimura and Tsuge, 1993). Characteriza­tion of the 30-kb region by complementation and genedisruption analysis lead to the localization of genes encod­ing the pol:-'ketide s:-uthase, 1,3.6.8-tetrallydrox:uaphtha­lene S)llthase (All1l-), the sc:-talone dehydratase (Bn1l1-).

and the L3,8-trihydrox:uaphthalene reductase (Bnn2-).

Sequence data have not been reported for the A. aZtematagenes; however. melanin pathway gene sequences havebeen reported for the 1,3.6.8-tetrallydrox:uaphthalenes)uthase (PKS1, Takano et aZ., 1995), the L3,8-trihy­drox:uaphthalene reductase (THR1, Perpetuaet aZ., 1996).and the sc)talone dehydratase (SCD1; Kubo et aZ., 1996a)from Colletotriclllll1l Zagellariwll. In contrast to A. alter­nata it appears that these genes are not closely linked in C.lagellarilll1l (Kubo et aZ., 1996b). Individual cosmids fromC. ZagelJariwll containing either the PKSl or THR or SCDlgenes do not overlap. The 1,3,6,8-tetrallydrox)uaphthalenelL3,8-trihydrox:uaphthalene reductase gene has also beenisolated from Magnaporthe grisea (Vidal-Cros et aZ., 1994)but classical genetic analysis with melanin mutants indi­cates none of the pathway genes are closely linked (Chum­ley and Valent, 1990). Although pathway genes have notbeen isolated for CochZioboZlls miyabealJus and C. heter­ostrophus, classical genetic analysis indicates that in bothorganisms the 1,3,6,8-tetrahydroX)uaphthalene synthaseand the 1,3,8-trihydrox')uaphthalene reductase genes arelinked but not the scytalone dehydratase (Kubo et aZ.,1996b).

Ill. GENE CLUSTERS AND THEREGULATION OF PATH\\TAY GENEEXPRESSION

One possible explanation for the existence of dispens­able pathway gene clusters is that the close linkage ofpathway genes is somehow involved in the regulation ofpathway gene expression. It is clear that gene clustering isnot necessary for dispensable metabolic pathway functionbecause there are several examples where some if not allgenes for a specific pathway are unlinked (Sections Id andIId). However, the linkage of pathway genes could beimagined to result in chromatin structures that somehowinfluence pathway gene expression. There are presently noexamples of these types of regulatory mechanisms in fungiand scant data available to indicate whether regulation ofmetabolic pathway gene expression is influenced by the

Copynght" 1997 by AcademiC PressAll rights of reproduction in any form reserved.

localization of genes \\ithin a gene cluster. What littlee\idence there is suggests that when some cluster genesintegrate ectopically they are expressed in a manner similarto that of genes residing \\ithin the cluster (e.g.. penicillincluster genes, Brakhage et aZ., 1994). In contrast. geneslocated within the sporulation-specific gene duster of A.nidllZans (SpoCI cluster. Miller et aT.. 1987) are notregulated appropriately following ectopic integration. How­ever, comparisons between metabolic pathway gene dus­ters and the SpoCI cluster are difficult to interpret becausethe function of SpoCI cluster genes has not been deter­mined. If some aspect of cluster chromatin structure isinvolved in gene regulation then .this regulatory mecha­nism is largely independent of the gene rearrangementsobserved for some natural product pathways (e.g..trichothecenes pathways).

IV. PATHWAY EVOLUTION

Attempts to explain fungal pathwa)' gene clusteringinclude the proposal that some pathwa)'s were acquiredfrom prokaryotes via horizontal gene transfers. This hypoth­esis would account for the clustering of pathway genessince prokaryotic metabolic pathways are generally clus­tered. The strongest case for such an event involves thepenicillin and cephalosporin path\vays because the relatedpathways can still be found in prokaryotes. Penicillin andcephalosporin pathway genes also contain features charac­teristic of the corresponding prokarotic pathway genessuch as the absence of introns and high GC content(Weigel et aZ., 1988). Interestingly, there is e\idence that atleast a portion of the cephalosporin C pathway of A.clmjSogelJlll1l evolved in fungi (Mathison et aT., 1993). ThecefC gene is located next to the cefEF gene and encodes anacetyltransferase catalyzing the last step in cephalosporinC biosynthesis. As pointed out by Mathison et aZ. (1993),CefG appears to have no prokaryotic counterpart incephalosporin biosynthesis. Further, in contrast to othercephalosporin pathway genes, cefC contains interveningsequences raising the possibility that it evolved as part of afungal pathway. This observation could be interpreted asfurther support for the fungal origin of this pathway orevidence that the evolution of this pathway, for reasons notyet understood, requires gene clustering even if portions ofthe pathway were acquired from prokaryotes. The fact thatvery few fungal natural product pathways appear to haveprokaryotic counterparts argues that pathway gene trans­fers across these Kingdoms are the exception.

It is important to point out that not all dispensablemetabolic pathwa~.. genes are clustered. Both the cephalo­sporin aIltI melanin pathwa~'s prO\ide examples of dispens­able pathwa~'s where the dustering ofpathway genes variesbetween different fungi. One interpretation of these obser­\'ations is that the loss of pathwa~'gene clustering is due tothe breakup of the OIiginal gene clusters. Gene dusterdissolution ma~' in tum signal the first step in the eventualloss of the pathwa~' or may simply reflect the lack ofselection pressure required to maintain the cluster. Oncepathwa~' genes are dispersed they may be more availablefor recruitment by other dispensable pathways or may actto seed the development of new pathways. lvlany of thegenes in dispensable pathways show significant sequencerelatedness to genes with related functions in other meta­bolic pathways. One possible example of dispensablepathwa~' gene recyding involves the 1,3,8-trihydrox;maph­thalene reductase in the melanin path\vay and the versicol­orin A reductase \stcU, formerly verA, and cer-l) in thesterigmatocystin and aflatoxin pathways (Keller et al.,1994: Perpetua et al.. 1996: Skory et al., 1993: Vidal-Cros et

al., 1994). Both enz:mes catalyze the dehydroxylation of apolyphenol substrate. and their amino acid sequencesshare 54% identi~·. Other examples in natural productpathways include cytochrome P450s, polyketide synthases,and peptide synthetases. There are examples of dispens­able pathway metabolic enZ:1nes that appear to be unique(McConnick et al., 1996). but it is possible that as moremetabolic pathways are characterized, their relationshipsto existing genes \\ill become apparent.

Perhaps the most interesting question raised by fungaldispensable pathway gene clusters is the question of whythese clusters occur. What possible advantages are therefor the clustering of pathway genes? Certainly it wouldseem that clustering creates opportunities for pathways tobe transferred in a horizontal fashion: this point haspreviously been emphasized in discussions of fungal peni­cillin pathway evolution. In fact. heterologous expressionof a fungal penicillin pathway has been demonstrated usinga cosmid clone from P chnjSogenlllJl carrying the pcbAB,pcbC, and penDE genes (Smith et al., 1992). Transforma­tion of N. crassa and A niger, neither of which possesses apenicillin pathway, resulted in the identification of transfor­mants able to produce penicillin V However, despite thisdemonstration that dispensable pathways can be trans­ferred between fungi. it remains to be determined if suchtransfers have occurred.

From an evolutionarv' standpoint. one advantage ofpath\vay gene clustering is that it could increase the

probabili~' that patlnnn' genes are transferred as a unitduring sexual reproduction or parasexual recombination. Amajor difference in these two ~}1es of genetic exchange isthe presence of meiotic crossing over in sexual rec:ombina­tion. although the relative effects of this process onmaintenance of gene clusters has not been assessed. \lan~'

fungi that contain dispensable patln\'a~' gene clustersappear to lack a sexual c~'cle. and parasexual processes arethought to play important roles in the genetics of thesefungi (Glass and Kuldau. 1992: Leslie. 19931. Under theseconditions the frequency of parasexual events could deter­mine if gene clustering is important for the evolution ofdispensable pathways. Ifgene clusters function to facilitatedispensable pathway evolution then the breakup of geneclusters would be predicted to occur when the conditionsdriving cluster evolution no longer exist. Evidence that thedissolution of the melanin pathway gene cluster hasoccurred to different extents in different fungal species(Kubo et al., 1996bl is now available but the reasons forthese changes are unclear.

V. CONCLUSIONS AND FUTURERESEARCH

It is important to emphasize that dispensable pathwayscan be distinguished from general metabolic pathwaysbased on the different roles they play in cellular metabo­lism. Differences between dispensable and general meta­bolic pathway functions may also reflect fundamentallydifferent mechanisms for pathway regulation and evolu­tion. Currently, it appears tllat most clustered pathwaygenes do not differ significantly from other fungal meta­bolic pathway genes and there is a lack of data to support aregulatory role for gene clustering. Gene clustering may beimportant for patlnvay evolution but many questionsremain concerning evolution-based eX}1lanations for theexistence of gene clusters. Future research on geneclusters will continue to focus on the regulation of pathwaygene ex-pression and the question of whether this regula­tion is related to gene clustering. Regulation experimentsinvolving the manipulation of pathway promoters andpathway-specific transcription factors are now possible inseveral systems. Studies employing these research toolsshould help to further characterize the effects of thecluster emironment on gene regulation. At the same timesequence infonnation continues to accumulate on pathwaygenes and pathway gene organization for specific dispens­able pathways present in different fungi. The application of

Copyright \) 1997 by Academic PressAll rights of reproduction in any form reserved.

26

molecular phylogenetic approaches to these data mayanswer questions concerning the· evolution of individualpatl1\vays and the relationships between pathways that existas clusters in some fungi but not in others. Finally,questions of whether reproduction strategies or ecologicalniches affect gene clustering should be ex-plored to fullyunderstand the phenomenon of fungal gene clusters.

ACKl'JO\VLEDGMENTS

We thank D. Bhatnagar. 1". Kubo, J. Linz, and C Scazzocchio forprmiding unpublished information.

REFERENCES

Arst. H. 1\ .. Jr.. and Cove. D. J. 1973. Nitrogen metabolite repression inAspergillus nidulans. Mol. Gen. Genet. 126: 111-141.

Arst. H. 1\ .. Jr.. Jones. S. A.. and Bailey, C R. 1981. A method for theselection of deletion mutations in the L-proline catabolism gene clusterin Aspergillus nidulans. Genet. Res. 38: 171-195.

Arst. H.1\ .. Jr.. and Scazzocchio. C 1985. Formal genetics and molecularbiology of the control of gene expression in Aspergillus nidulans. InGene Manipulation in Filamentous Fungi (J. W. Bennett and L. L.

Lasure. Eds.), pp. 309-343. Academic Press, London.BelL A. A., and \Vheeler. M. H. 1986. Biosynthesis and functions of fungal

melanins. All1ll1. Rer:. Phytopathol. 24: 411-451.Bennett. J. W., and Papa. K. E. 1988. The aflatoxigenic ..'1spergillus

species. Adr:. Plant Patlwl. 6: 263-280.Bergh. K. T. Litzka. 0 .. and Brakhage. A. A. 1996. Identification of a

major cis-acting DNA element controlling the bidirectionally tran­scribed penicillin biosynthesis genes acr:A (pcb.4B) and ipnA (pcbCl ofAspergillus nidulans.]. Bacterio!. 178: 3908-3916.

Brakhage, A. A., Browne, P., and Turner, G. 1994. Analysis of theregulation of penicillin biosynthesis in Aspergillus nidulans by targeteddisruption of the acr:A gene. ,Hoi. Gen. Genet. 242: 57-64.

Brandriss, M. C. and 1\hgasanik, B. 1979. Genetics and physiology ofproline utilization in Saccharomyces cerevisiae: enzyme induction byproline.]. Bacteriol. 140: 498-503.

Brito. N., Avila, J.. Perez. M. D.. Gonzalez, C., and Siverio, J. M. 1996.The genes YNll and YNRl. encoding nitrite reductase and nitratereductase respectively in the yeast Hansenula polymorpha. are clus·tered and coordinately regulated. Biochem.]. 31i: 89-95.

Brown, D. \Y. Adams, T H., and Keller, N. P. 1996a. Aspergillus hasdistinct fatty acid synthases for primary and secondary metabolism.Proc. Nat!. Acad. Sci. USA 93: 14873-14877.

Brown, D. w.. Yu, J.-H.. Kelkar, H. S., Fernandes, M.. Nesbitt. T C.Keller, N. E, Adams, T. H.. and Leonard. T J. 1996b. Twenty-fiveco-regulated transcripts define a sterigmatocystin gene cluster inAspergillus nidulans. Froc. Natl. Acad. Sci. USA 93: 1418-1422.

Burger, G., Tuburn, J., and Scazzocchio, C. 1991. Molecular cloning and

Copyright (> 1997 by Academic PressAll rights of reproduction in any form reserved.

HI'vic\\"

functional characterization of the pathwav-specifk regnlator\" gellt'nirA. which controls nitrate assimilation in Aspergillus nidlllallS. M"I.Cell. Bioi. 11: 79.')...802.

Chang. P.-K.. Ehrlich. K. C. I'u. J.. Bhatnagar. D.. and Cle\·eland. T E.1995. Increased expression of Asper!!,illus parasitictls aflR. encoding asequence-specific DNA-binding protein. relieves nitrate inhibition ofaflatoxin biosynthesis. Appl. Environ. Microbiol. 61: 2372-2377.

Charles. 1. G.. Keyte, J. W.. Brammar. \Y. J.. Smith. 1\1.. and Hawkins.A. R. 1986. The isolation and nucleotide sequence of the complex aro11l

locus ofAspergillus nidulans. Nucleic Acids Res. 14: 2201-2213.Chiang. T Y.. and MarzluL G. A. 199.5. Binding affinity and functional

significance of nit2 and nit4 binding sites in the promoter of the hig:hh­regulated NIT-3 gene. which encodes nitrate reductase in Scurospora

crassa. ]. Bacterial. 177: 6093-6099.

Chumley. F. G.. and Valent. B. 1990. Genetic analysis of melanin­deficient, non-pathogenic mutants of Magnaporthe grisea. Mol. Plant­

Microbe Interact. 3: 135-143.Clutterbuck, A. J. 1992. Sexual and parasexual genetics of Aspergillus

species. In Aspergillus: Biology and lndustrial ApplicatiollS (j. \\'.Bennett and M. A. Klich. Eds.), pp. 3-18. Butterworth-Heinemann.Boston.

Davis. R. H. 1986. Compartmental and regulatory mechanisms in thearginine pathways of Neurospora crassa and Saccharomyces cerer:isiae.

Microbiol. Rev. 50: 280-313.

DeSjardins. A. E.. and Hohn. T M. 1997. Mycotoxins in plant pathogen­esis. Mol. Plant-Microbe Interact. in press.

DeSjardins. A. E., Hohn. T M.. and McConnick. S. P. 1993. Trichotheeenebiosynthesis in Fusarium species: Chemistry. geneties. and signifi­cance. Microbial. Ret;. 57: 595-604.

Desjardins. A. E., Proctor. R. H.. Bai, G., MeCormick. S. P.. Shaner. G..Buechley, G.. and Hohn. T M. 1996. Reduced virulence oftrichothecene-nonproducing mutants of Gibberella ;;cae in wheat fieldtests. Mol. Plant-Alicrobe Interact. 9: 775-781.

Diez. B.. Gutierrez. S.. Barredo, J. L., yan Solinger. P.. van der Voort.L. H. M.. and Martin. J. F. 1990. The cluster of penicillin biosynthetiegenes. Identification and characterization of the pcbAB gene encodingthe a-aminoadipyl-cysteinyl-valine synthetase and linkage to the pcbC

andpcbDE genes.]. BioI. Chelll. 265: 16358-16365.

Diolez. A., Langin. T, Gerlinger. C.. Brygoo. Y.. and Daboussi. M. J. 1993.The nia gene ofFusarium oxysponlln: isolation. sequence ,md develop­ment of a homologous transforn1ation system. Gene 131: 61-67.

Dowzer. C. E .. and Kelly, J. M. 1991. Analysis of the ereA gene. aregulator of carbon catabolite repression in Aspergillus nie/ulans. Mol.

Cell. Bioi. 11: 5701~5709.

Espeso. E. A., Fernandez-Canon. J. M.. and Penalva, M. A. 1995. Carbonregul~tion of penicillin biosynthesis in A~pergillus nidulans: a minoreffect of mutations in creB and creGo FEMS Microbiology Lett. 126:

63-68.Espeso. E. A., Tuburn, J., Arst, H. N.. and Penalva, M. A. 199.3. pH

regulation is a major determinant in expression of a fungal penicillinbiosynthetic gene. EMBO]. 12: 3947-3956.

Felenbok, B. 1991. The ethanol utilization regulon of Aspergillus

nidulans: the alcA-alcR system as a tool for the expression of recombi­nant proteins.]. Biotechnol. 17: Il-I8.

Fillinger. S., and Felenbok, B. 1996. A newly identified gene cluster inAspergillus nidulans comprises five novel genes localized in the alc

Review

region th,lt are controlled both I", the spedfic transactivator AleR and

the general carbon-catabolite repressOl' Cre:'.. Mol. Microviol. 20:

475-1SS.

Geever. R. F.. Baum. J. A.. Case. M. E.. and Giles. N. H.1987. Regulation

of the Q:\ gene cluster of Neurospora cmssa. Antonie r.;an LeelHEel/­

hock 53: 34:1-348.

Giles. N. H.. Gee\·er. R. F.. Ascll. D. 1\:.. Avalos. J.. and Case. M. E. 1989.

Organization and regulation of the qa (quinit' acid) genes in Neuros­

pora cmssa and other fungi.]. Heredity 82: 1-7.

Ghss. N. L.. and Kuldan. G. A. 1992. Mating t;pe and vegetative

incompatibilitv in filamentous ascomvcetes. AllIlII. Re[. Phytopatlwl.

30: 2UI-224.

Griffln. D. H. 1994. Fungal Physiology. pp. 215-300. Wiley-Liss. Ne\\'

York.

Haas. H .. and Marzluf. G. A. 1995. NRE. the major nitrogen regulatory

protein of Penicilliu11l chrysogelllllll. binds specifically to elements in

the intergenic promoter regions of nitrate assimilation and penicillin

biosynthetic gene clusters. Curr. Genet. 28: 177-173.

Ha\\'kins. A. R.. Lamb. H. K.. Moore. J. D .. and Roberts. C. F. 1993.

Genesis of eukaryotic transcriptional activator and repressor proteins

lw splitting a multidomain anabolic en~,ne. Gene 136: 49-54.

Hawkins. A. R.. Lamb. H. 1\:.. Radford. A.. and Moore. J. D. 1994.

Evolution of transcription-regulating proteins b~, en~'Tlle recruitment:

Molecular models for nitrogen metabolite repression and ethanol

utilisation in eukarvotes. Gene 146: 145-158.

Holm. T. M.. and Beremand. P. D. 1989. Isolation and nucleotide

sequence of a sesquiterpene cyclase gene from the trichothecene­

producing fungus Fusariu11l spomtriehioides. Gene 79: 1.31-138.

Hohn. T. M.. Desjardins. A. E.. and McConnick. S. P. 1995a. The Tri4

gene of Fusariu11l sporotrichioides encodes a cytochrome P450 mo­

noox-ygenase involved in trichothecene biosynthesis. Mol. Gen. Genet.

248: 95-102.

Holm. T. M.. Desjardins. A. E .. McConnick. S. P.. and Proctor. R. H.

1995b. Bios~nthesis of trichothecenes. genetic and molecular aspects:

Issues invohing toxic microorganisms. In Molecular Approaches to

Food Safety (M. Eklund. J. L. Richard. and M. Katsutoshi. Eds.). pp.

2:39-247. Alaken. Fort Collins. CO.

HulL E. P.. Green. P. M.. Arst. H. N.. Jr.. and Scazzocchio. C. 1989.

Cloning and physical characterization of the L-proline catabolism gene

cluster of Aspergillus nidulans. Mol. Microbiol. 3: 553-.5.59.

Janis. B. B. 1991. Macrocyclic Trichothecenes. In l'tfycotoxillS and

PhytoalexillS (R. P. Shanna and D. K. Salunklle. Ecis.). pp. 361-121.

CRC Press. Boca Raton. FL. Griffin. D. H. 1994. Fungal Physiology.

Wile~·-Liss. New York.

Johnstone. 1. L.. McCabe. P. C .. Greaves. P.. Gurr. S. J.. Cole. G. E .. Brow.

M. A. D.. Unkles. S. E.. Clutterbuck. A. J.. Kinghorn. J. R.. and Innis.

~L A. 1990. Isolation and characterisation of the cmA-niiA-niaD gene

cluster for nitrate assimilation in Aspergillus nidulans. Gene 90:

181-192.

Jones. S. A.. Arst. H. N., Jr.. and MacDonald. D. W. 1981. Gene roles in

the pm cluster of Aspergillus nidulans. Curr. Genet. 3: 49-.56.

Kale. S. P.. Cary. J. W.. Bhatnagar. D.. and Bennett. J. W. 1996.

Characterization of experimentally induced. nonaflatoxigenic variant

strains of Aspergillus pamsiticus. Appl. Enr.;iron. Microbiol. 62: 3.399­

.3404.

Kelkar. H. S.. Keller. N. P.. and Adams. T. H. 1996. A.spergillus nidulam

27

stcP encodes an 0 -methvltransfenLse that is required for sterigmHto('vs­tin bios~llthesis. Appl. EIlr.;irOlI. Microbiol. 62: 4296-42%.

Kelkar. H. S.. Skloss. T. \\'.. Haw. J. F.. Keller. N. P.. and Adams. T. H.

1997. Aspergillus nidulans stcL encodes a putative P --I5U monoox'ygen­

ase required for bisfuran desaturation during aflatoxilvsterigmato('vstin

biosmthesis.]. Bioi. Chem. 272: 1.559-1594.

Keller. N. P.. Kantz. N. .I .. and Adams. T. A. 1994. Aspergillus lIirlulan'

r;erA is required for production of the mvcotoxin sterigmatoevstin.

".l,ppl. Environ. Microbiol. 60: 144-1-1450.

Keller. N. P.. Bro\\~. D .. Butchko, R. A. E .. Femandes. M.. Kelkar. H..Nesbitt. C .. Segner. S., Bhatnagar. D .. Cleveland. T. E.. and Adams.

T. H. 1995. A consen'ed polyketide mycotoxin gene cluster in ..l,spergil­

Ius nidulans. In Molecular Approaches to Food Safety Is.mcs lllwlr;ill!!,

Toxic Microorganisms (R. J. L.. Ed.l. pp. 26:1-277. Alaken. Fort

Collins. CO.

Keller. N. P.. Nesbitt. C.. Sarr. B.. Phillips. T. D.. and Buro\\'. G. B. 1997.

pH regulation ofsterigmatocystin and aflatoxin biosmthesis in Aspcl'gil­

Ius spp. Phytopathology. accepted pending re\ision.

Keller. N. P.. Segner. S.. Bhatnagar. D .. and Adams. T. H. 1995. stcS. a

putative P--I50 monoox-ygenase. is reqnired for the conversion of

versicolorin A to sterignlatocvstin in Aspergillus nidulmls. Appl.

Enr.;iron. Microbial. 61: 3628-3632.

Kimura. N., and Tsuge. T. 1993. Gene duster involved in melanin

biosynthesis of the filamentous fungus Alten/aria altemata. J. Bacte­

rial. 1i5: 4425-143.5.

Kubo. Y.. and Furusawa. 1. 1991. Melanin bios\11thesis: Prerequisite for

the successful invasion of the plant host hv appressoria of Colletotri­

chum and PyriCtllaria. In The Fungal Spore alld Disease Initiatioll in

Plants and Animals (G. T. Cole and H. C. Hocll. Ecis.). pp. 205-218.

Plenum. New York.

Kubo. Y.. Takano. Y.. Endo. N.. Yasuda. N.. Tajima. S.. and FurusH\\'a. 1.1996a. Cloning and structural analysis of the melanin bios~~thesis gene

SCD] encoding sc~talone dehydratase in Colletotriclllllll lagellarilllli.

Appl. Environ. Microbiol.. in press.

Kubo. 1".. Takano. Y.. and Furusawa. 1. 1996b. Molecular gpnetic analysis

of melanin biosynthetic genes essential for apprpssorium function in

Colletotrichtlln lagenarill11l. In MoleClllar As]wcts of Pathogenicity:

Reqllirements for Signal Transduction (D. Mills. H. Kunoh. N. Keen.

and S. Mayama. Ecis.). pp. 73-82. APS Press. St. PauL MN.

Kuclla. B.. Caddick. M. X.. Langdon. T.. Marinez-Rossi. N. M.. Bennet.

C. F.. Sibley. S.. Davies. R. W.. and Arst. H. N., Jr. 1990. The regulatory

gene areA mediating nitrogen metabolitp repession in Aspergillus

nidulam. Mutations affeCting specificity of gene activation alter a loop

residue of a putativp zinc finger. EMBO ]. 9: 1355-1364.

Kulmburg. P.. Judp\\icz. N.. Mathieu. M .. Lenouvel. F.. Sequeval. D .. and

Felenbok. B. 1992. Specific binding sites for the activator protein.

AlcR. in the alcA promoter of the ethanol regulon of Aspergillus

nirhtlans.]. Bioi. Chelll. 267: 21146-21153.

Kulmburg. P.. Mathieu. M., Dowzer. C.. Kelly. J.. and Felenbok. B. 1993.

Specific binding sites in the alcR and alcC promoters of the ethanol

regulon of the CreA repressor mediating carbon catabolite repression

in Aspergillus nidulans. l'vlol. Microbiol. i: 847-857.

Lamb. H. K.. Hawkins. A. R.• Smith. M.. Han·ey. 1. J.. Brown, J.. Turner,

G.. and Roberts. C. F. 1990. Spatial and biological characterisation of

the complete quinic acid utilisation gene duster in A\pergillus nidu­

lans. Mol. Gen. Genet. 223: 17-23.

Copynght C I 997 by Academic PressAll rights of reproduction in any form reserved.

28

Lamb. H. 1\... "'e\\ton. G. H.. uYett. L. J.. Caims. E.. Rob,.,rts. C. F.. andHawkins. A. R. 1886. Th,., QUT.-\ acm'utor and QUTR repressorproteins of Aspcrgilllls lIidlllalls interact to regulate transcription of the

quinate utilization rathwa~' genes. Microbiology 142: 1477-1490.Landan. G.. Cohen. F.. AharonO\\itz. Y.. Shuali. Y., Graur. D.. and

Shiffman. D. 1890. Eyolution of isopenicillin N synthase genes ma\,have in\'olved horizontal gene transfer. Mol. Bioi. Eml. 7: 399-406.

Leslie. J. F. 1993. Fungal \'egetati\'e incompatibility. All/HI. Ret:. Phyto­

]Jalhol. 31: 127-150.!\lacCabe. A. P.. Riach. M. B. R.. Unkles. S. E.. and lUnghom, J. R. 1990.

The Aspcrgilhts lIidlllalls IlpeA locus consists of three contiguous genes

requirpd for ppnicillin bios~l1thesis. EMBO J. 9: 279-287.

Malmnti. N., Bhatnagar. D.. and Linz. J. E. 1995. Structure and function

ofras-lA. a genp encoding a putati\'e fatty acid synthetase directly

in\'ol\'ed in aflatoxin biosmthesis in Aspergilllls parsiticus. Appl.

Ellt:inllt. Microbiol. 62: 191-195.

Martin. J. F.. and Gutierrez. S. 1995. Genes for beta-Iactam antibiotic

biosmthesis. Alliollie Leeur.cellllOek Inl. j. Gen. M. 67: 181-200.

Mathieu. M.. and Felenbok. B. 1994. The Aspergillus nidulans CREAprotein mediates glucose repression of the ethanol regulon at various

levels through compptition \\ith the ALCR-specifil' transacti\'ator.EMBO j. 13: 4022-1027.

Mathison. L.. Soliday. C.. Stepan. T.. Aldrich, T.. and Rambosek. J. 1993.

Cloning. characterization. and use in strain impro\'ernent of theCephalosp0riulll acrelllOllillln gene cefG encoding acetyl transferase.C,m,: Genel. 23: 33-11.

McConniek. S. P.. Hohn. T. !vi .. and Desjardins. A. E. 1996. lsolation andcharacterization of Tri3, a gene encoding 15-0 -acetyltransferase from

Fusariulll sporotrichioides. Appl. Environ. Microbiol. 62: 353-359.Miller. B. L.. 1\liller. I\.. Y.. Roberti. K. A.. and Timberlake. W. E. 1987.

Position-dependent and -independent mechanisms regulate cell­specific pXl'ression of the SpoC1 gene cluster of .4spergillus nidulans.

Mol. Cell. Bioi. 7: 427-134.

Pame. G. A.. Nystrom. G. J.. Bhatnagar. D.. Cleveland. T. E.. andWoloshuk. C. P. 1993. Cloning of the afl-2 gene involved in aflatoxin

bios\1lthesis from l\spergillus flavus. Appl. Environ. Microbiol. 59:

156-162.

Penalva. M. A.. Moya. A.. Dorazo. J.. and Ramon. D. 1990. Sequences ofisopenicillin N ~l1thetase genes sug.,gest horizontal gene transfer from

prokaryotes to eukary·otes. Proc. R. Soc. London 241: 164-169.Perpetua. N. S.. Kubo. Y.. Takano. Y.. and Furusawa. 1. 1996. Cloning and

characterization of a melanin bios;l1thetic THRI reductase geneessential for appressorial penetration of Colletotrichulll lagellarilllli.

Mol. Plant-Microbe Illteract. 9: 323-329.

Pieterse. C. M.. van't Klooster, J.. van den Berg.Velthuis. G. c.. and

Govers. F. 1995. NiaA. the structural nitrate reductase gene ofPhytophthara infestails: Isolation. characterization and eXl'ression

analvsis in A~pergillus nidulans. CUrT. Genet. 27: 359-366.Proctor. R. H.. Hohn. T. M.. and McConnick. S. P. 1995a. Reduced

\irulence of Gibberella ::eae caused by disruption of a trichothecenetoxin biosynthetic gene. Mol. Plant-Microbe Interact. 8: 593-601.

Proctor, R. H.. Holm. T. M.. McConnick. S. P.. and Desjardins. A. E.1995b. Tri6 encodes an unusual zinc finger protein involved in the

regulation of trichothecene bio~Tlthesis in Fusariulll sporotrichioides.

Appl. Environ. Microbiol. 61: 1923-1930.

Punt. P. J.. Strauss. J.. Smit. R.. lUnghom. J. R.. Van den Hondel.

Copyright <D 1997 by AcademiC PressAll rights of reproduction in any form reserved.

C. A. M. J. J.. and Scazzocchio. C. 1995. Thp int"rgpnic re!!ion hptwP'>llth" di\,prgentl~·transcribed lliiA and iliaD !?:enes ofAspergilllls ,tidlliall'

contains multiplp NirA binding sites which ad bidirpctionalh-. Mol.

Cell. Bioi. 15: 5688-5699.

Scazzocchio. C. 1992. Control of gpnp eX11ression in thp catabolit­

pathways of Aspergilllls nidllialls: A personal and bhLs"d account. InAspergilllls: Biology alld Il1dllsfrial Applications (J. \\'. Benn"tt and

M. A. Klich. Eels.), pp. 43-68. Butterworth-Heinemann. Boston.

Sharma. K. K.. and Arst. H. N.. Jr. 1985. Th" product of the regulaton'

gene of the proline catabolism !?:ene cluster of Aspcrgilllls nidllians is apositive·acting protein. CIIrT. Genct. 9: 299-304.

Silva. J. c.. Minto. R. E.. Barry. C. E.. Holland. 1\:. A.. ami TO\\11send.

C. A. 1996. Isolation and chamcterization of the wrsieolorin B S\11thase

gene from Aspergilllls parasitictIS--eXl',l1Jsion of the aflatoxin BI

bios;Tlthetic gene cluster. J. Bioi. Chelll. 271: 13600-13608.

Skory. C. D., Chang. P. K.. and E.. L. J. 1993. Regulated eXl'ression of thenor-l and ver·l genes associated \\ith aflatoxin biots\l1thesis. Appl.

Ellviron. Microbial. 59: 1642-1646.Smith. M. w.. Feng. D.. and Doolittle. R. F. 1992. Evolution b\·

acquisition: The case for horizontal gene trausfers. Trellds Biochelll.

Sci. 17: 489-193.Suarez. T.. and Penalva. M. A. 1996. Characterization of a Pl'nicillilwi

e/lnjsogenlllll gene encoding a PacC transcription factor and its bindingsites in the divergent pcbAB-pcbC promoter of the ppnicillin bios;11the­tic cluster. Mol. Microbial. 20: 529-540.

Takano. Y.. Kubo. Y.. Shimizu. K.. Mise. 1\... Ok'llno. T.. and FllnIs;\wa. 1.1995. Structural analysis of PKSI. a polyketide s;l1thase gene involvedin melanin bios;l1thesis in CalletatrielwlIl lagellari/wi. Mol. Gen.Genet. 248: 162-167.

Tilburn, J.. Sarkar. S.. Widdick, D. A.. Espeso. E. A.. Orejas. M..Mungroo, J.. Penalva. M. A.. and Arst. H. N.. Jr. 1995. The A~pcrgilllls

PacC zinc finger transcription factor mediatcs regulation of both acid·and alkaline-exl'ressed genes by ambient pH. EMBO j. 14: 779-790.

Trail. F.. Mallanti, N.. Rarick. M.. Mehigh. R.. Liang. S. H.. Zholl. R.. andLinz, J. E. 1995. Physical and transcriptional map of an aflatoxin gene

cluster in A~pergilllls pamsitictlS and functional disruption of a genein\'olved early in thp aflatoxin pathway. Appl. Enviroll. Microbiol. 61:

2665-2673.

Unkles. S. E.. Campbell. E. I.. Punt. P. J.. Hawker. K. L.. Contreras. R..

Hawkins. A. R.. Van den Hondel. C. A.. and lUnghorn. J. R. 1992. Th"

Aspergilllls niger niaD gene encoding nitmtp reductas": Upstreamnucleotide and amino acid sequenl'" comparisons. Gelle 111: 149-155.

Vidal-Cros. A.. Vi\iani. F.. Labesse. G.. Boccam. M.. and Gauchy. M.

1994. Polyhydrox)l1aphthalenp reductase involved in melanin bio~'TI­

thesis in Magnaportlte grisca. Ellr. j. Bioe/wlIl. 219: 985-992.

Walton. J. D. 1996. Host-selective toxins: Agents of compatibility. Plant

Cell 8: 1723-1733.

Watanabe. C. M. H.. Wilson. D.. Linz. J. E.. and TO\\l1Send. C. A. 1996.

Demonstration of the catalytiC roles and e\idence for the phYSical

association of type I fatty acid ~thases and a polyketide ~thase inthe bios;l1thesis of aflatoxin BI . CltCIll. Bioi. 3: 463-169.

\Veigel. B. J.. Burgett. S. G.. Chen, V J., Skatrud. P. L.. Frolik. C. A.,Queener. S. w.. and Ingolia. T. D. 1988. Cloning and expression inEscherichia coli of isopenicillin N s:mthetase genes from Streptalllyces

liplllallii and Aspergilllls nidllians. j. Bacterial. 170: 3817-3826.

Woloshuk, C. P.. Fount. K. R.. Brewer. J. F., Bhatnagar. Do, Cleveland.

Review

T. E.. and Pame. G. A. W\)4. \Ioleenlar eharaeterization of af/R. a

regulatory lac-us for aflatoxlII bios\11th<csis. Appl. EIll"inJll . •HicrolJiol.

60: 24()~2414.

Yu. J.. Chang. P.. Cary. J.. \\"right. \1.. Bhatnagar. D .. Cleveland. T. E ..

Pa\1le. G. :L and Linz. J. E. 1\)9.5a. Comparati\"E> mapping of aflatoxin

path,,",l\' gene dusters in Aspecgilllls parasiticlls and Aspergilllls.f/aws.

Appl. EIlr:iroll .•\Jicrobiol. 61: 236,5-2371.

29

Yu. J.-H .. Butehko. R. A. E .. Femandes. \1.. b:eller. !\. P.. Leonard. T. .1 ..

and Adams. T. H. 19\)6. ConsE'rvation of stmC'turl' and runetion of tilt'

aflatoxin re~latof\' gt'nt' af/R from Aspcrgilllls llidlllll1lS amI A. .f/IIW.I.

CI/n: Gcnct. 29: 549-.555.

Yu. J.-H.. and Leonard. T. J. 1995h. Sterigmato("\'stin hios\11tlU;>sis in

Aspergilll/s I/idllialls requin:>s a nO\'el t\lw 1 pokketid'> S\11th'L">. J.Baetenol. Iii: 4792-!SOO.

Copynght '0 1997 by Academic PressAll rights of reproduction in any fonn reserved.