of 11/11
Vol. 40, No. 1 APPLIED AND ENVIRONMENTAL MICROBIOLOGY, July 1980, p. 145-155 0099-2240/80/07-0145/11$02.00/0 Degradation of Melanin by Aspergillus fumigatus J. P. LUTHER AND H. LIPKE* Biology Department, University of Massachusetts/Boston, Dorchester, Massachusetts 02125 A strain of Aspergillus fumigatus from composted coffee and garden wastes utilized natural deproteinized insect, banana, hair, octopus, and synthetic tyrosine and dopa melanins as sole sources of carbon. With a sucrose supplement, degra- dation was essentially complete after 50 days in Czapek medium, pH 6.5 at 30°C. The catabolic rate differed for each substrate pigment, as did the molecular weight distribution of products accumulating in the medium. After incubation with L-[U-'4C]melanin, over 50% was recovered in a dark fungal pigment, the remainder appearing as cell protein, chitin, lipid, C02, and polar metabolites. When grown on melanin, the normally pale mycelia darkened with the production of a fungal allomelanin, with infrared spectrum and alkali fusion products differing from those of the substrate pigment. Isotope distribution in amino acids for A. fumigatus grown on labeled melanin supplemented with sucrose suggested sepa- rate pools for synthesis of cell proteins and melanoproteins. Deposition of allo- melanin increased resistance of conidia, sterigma, and conidiophores to lytic carbohydrases as judged by scanning electron microscopy. Natural melanins from plants, microorga- nisms, and animnals consist of a polyaromatic backbone with proteins attached by covalent bonds (23, 24, 28). The complex is subject to slow degradation by soil flora, with carbon and nitrogen being released from the protein moiety at a rate slightly higher than from the benzenoid pigment component (12). Melanin of low protein content prepared by incubation of [U-14C]tyro- sine with polyphenol oxidase released only 9% of the isotope as labeled carbon dioxide in the course of 8 weeks of incubation with potting soil (2). Similarly, a natural melanin deproteinized by treatment of humic acid with 6 M HCI af- forded only 2% of the residual pigment as carbon dioxide after 30 weeks of exposure to a sandy loam (17). Although a portion of the nonprotein component of melanoproteins is clearly resistant to degradation, the more labile aromatic com- ponents of fungal melanins and plant lignins contribute significantly to the formnation of humic acids (7, 15, 16). Melanins from cultures of Eurotium echinulatum and Stachyobotrys chartarum release phenols and anthraquinones into the media which are subsequently recycled for synthesis of humic acid-like materials (14, 23). Together with wall material from heavily melanized fungi, these polymers constitute a significant portion of the stable components of soil humus (23). The origin of this fraction is not confined to lignified tissues and fungal debris, however. Melanins occur widely in both marine and terrestrial animals, with composition vary- ing in the ratio of indolic and phenolic substitu- ents (1, 21). The effect of this structural diversity on mineralization, catabolism, and the identity of the organisms contributing to the initial stages of polymer degradation has not been reported. This contribution addresses these issues by re- porting the fate of deproteinized melanins of animal and synthetic origin after exposure to soil and to cultures of Aspergillus fumigatus isolated from these soils. In addition to its role as a humin precursor, melanin influences the processing of soil organic matter by conferring resistance to enzymatic degradation of associated polysaccharides and proteins. Mixtures of f8-(1,3)-glucanase and chi- tinase hydrolyzed the melanin-free hyaline hy- phae of A. phoenicia but failed to act on mel- anized sclerotia (2). The resistance of A. nidu- lans to these carbohydrases was in proportion to the melanin/polysaccharide ratio, and addi- tion of purified melanin to casein extended the time required for proteolysis by subtilisin (14). The organism described in the present study, A. fumigatus, develops a black color in the course of melanin degradation, providing the opportu- nity to compare susceptibility to lysis in colored and uncolored variants of the same strain. We examined mycelia by scanning electron micros- copy after incubation with depolymerizing car- bohydrases. Extensive damage to the uncolored population was revealed but litte change in the melanized samples was evident, confirmiing chemical studies with pigmented and unpig- mented strains isolated from natural sources (2). MATERIALS AND METHODS Isolation of organisms. A. fumigatus NRRL 6463 was isolated from a compost heap in Watertown, Mass., consisting of garden wastes and a significant 145 on May 26, 2018 by guest http://aem.asm.org/ Downloaded from

Degradation of Melanin Aspergillus fumigatusaem.asm.org/content/40/1/145.full.pdf · melanin utilization, parallel replicates, onewith inoc-ulum but prepared without melanin, and

  • View
    220

  • Download
    3

Embed Size (px)

Text of Degradation of Melanin Aspergillus fumigatusaem.asm.org/content/40/1/145.full.pdf · melanin...

  • Vol. 40, No. 1APPLIED AND ENVIRONMENTAL MICROBIOLOGY, July 1980, p. 145-1550099-2240/80/07-0145/11$02.00/0

    Degradation of Melanin by Aspergillus fumigatusJ. P. LUTHER AND H. LIPKE*

    Biology Department, University ofMassachusetts/Boston, Dorchester, Massachusetts 02125

    A strain of Aspergillus fumigatus from composted coffee and garden wastesutilized natural deproteinized insect, banana, hair, octopus, and synthetic tyrosineand dopa melanins as sole sources of carbon. With a sucrose supplement, degra-dation was essentially complete after 50 days in Czapek medium, pH 6.5 at 30C.The catabolic rate differed for each substrate pigment, as did the molecularweight distribution of products accumulating in the medium. After incubationwith L-[U-'4C]melanin, over 50% was recovered in a dark fungal pigment, theremainder appearing as cell protein, chitin, lipid, C02, and polar metabolites.When grown on melanin, the normally pale mycelia darkened with the productionof a fungal allomelanin, with infrared spectrum and alkali fusion products differingfrom those of the substrate pigment. Isotope distribution in amino acids for A.fumigatus grown on labeled melanin supplemented with sucrose suggested sepa-rate pools for synthesis of cell proteins and melanoproteins. Deposition of allo-melanin increased resistance of conidia, sterigma, and conidiophores to lyticcarbohydrases as judged by scanning electron microscopy.

    Natural melanins from plants, microorga-nisms, and animnals consist of a polyaromaticbackbone with proteins attached by covalentbonds (23, 24, 28). The complex is subject toslow degradation by soil flora, with carbon andnitrogen being released from the protein moietyat a rate slightly higher than from the benzenoidpigment component (12). Melanin of low proteincontent prepared by incubation of [U-14C]tyro-sine with polyphenol oxidase released only 9% ofthe isotope as labeled carbon dioxide in thecourse of 8 weeks of incubation with potting soil(2). Similarly, a natural melanin deproteinizedby treatment of humic acid with 6 M HCI af-forded only 2% of the residual pigment as carbondioxide after 30 weeks of exposure to a sandyloam (17). Although a portion of the nonproteincomponent of melanoproteins is clearly resistantto degradation, the more labile aromatic com-ponents of fungal melanins and plant ligninscontribute significantly to the formnation ofhumic acids (7, 15, 16). Melanins from culturesof Eurotium echinulatum and Stachyobotryschartarum release phenols and anthraquinonesinto the media which are subsequently recycledfor synthesis of humic acid-like materials (14,23). Together with wall material from heavilymelanized fungi, these polymers constitute asignificant portion of the stable components ofsoil humus (23). The origin of this fraction is notconfined to lignified tissues and fungal debris,however. Melanins occur widely in both marineand terrestrial animals, with composition vary-ing in the ratio of indolic and phenolic substitu-ents (1, 21). The effect of this structural diversityon mineralization, catabolism, and the identity

    of the organisms contributing to the initial stagesof polymer degradation has not been reported.This contribution addresses these issues by re-porting the fate of deproteinized melanins ofanimal and synthetic origin after exposure tosoil and to cultures of Aspergillus fumigatusisolated from these soils.

    In addition to its role as a humin precursor,melanin influences the processing of soil organicmatter by conferring resistance to enzymaticdegradation of associated polysaccharides andproteins. Mixtures of f8-(1,3)-glucanase and chi-tinase hydrolyzed the melanin-free hyaline hy-phae of A. phoenicia but failed to act on mel-anized sclerotia (2). The resistance of A. nidu-lans to these carbohydrases was in proportionto the melanin/polysaccharide ratio, and addi-tion of purified melanin to casein extended thetime required for proteolysis by subtilisin (14).The organism described in the present study, A.fumigatus, develops a black color in the courseof melanin degradation, providing the opportu-nity to compare susceptibility to lysis in coloredand uncolored variants of the same strain. Weexamined mycelia by scanning electron micros-copy after incubation with depolymerizing car-bohydrases. Extensive damage to the uncoloredpopulation was revealed but litte change in themelanized samples was evident, confirmiingchemical studies with pigmented and unpig-mented strains isolated from natural sources (2).

    MATERIALS AND METHODSIsolation of organisms. A. fumigatus NRRL 6463

    was isolated from a compost heap in Watertown,Mass., consisting of garden wastes and a significant

    145

    on May 26, 2018 by guest

    http://aem.asm

    .org/D

    ownloaded from

    http://aem.asm.org/

  • 146 LUTHER AND LIPKE

    amount of spent coffee grounds. Identity was verifiedby Margaret Bigelow, Department of Botany, Univer-sity of Massachusetts/Amherst, and by the TaxonomyGroup of the U.S. Department of Agriculture Fermen-tation Laboratory, Peoria, Ill. This strain was isolatedin the course of a screening program in which soils andcomposts were added at a level of 2 g/150 ml to 0.8%nutrient broth fortified with 1.5% sucrose in 500-miflasks. The flasks were incubated at 23C for 48 hwithout agitation, after which inocula were transferredto Czapek broth supplemented with insoluble insectmelanin, 5 mg/2.5 ml. The broth contained, per literof distilled water: NH4Cl, 3 g; K2HPO4, 1 g; MgSO4,0.5 g; KCI, 0.5 g; FeSO4, 0.01 g; and yeast extract, 1 g,pH 6.5. The cultures were incubated without agitationfor 6 weeks at 23C. Melanin utilization wasmonitored by dark coloration of the medium, whichwas initially straw yellow, and by disappearance of theinsoluble melanin pellet. Samples yielding the highestrate of melanin utilization were streaked on Czapekagar, and individual colonies were again tested formelanin utilization by the qualitative method de-scribed above. The plating step was repeated withthose cultures capable of melanin catabolism. Isolatesfrom the second platings were free of contaminants asassessed by streaking on nutrient agar. In all tests formelanin utilization, parallel replicates, one with inoc-ulum but prepared without melanin, and one withmelanin but not inoculated, served as controls. Mela-nin remained insoluble in the former group, and pig-mented cells failed to appear in the latter cultures.

    Utilization and analysis of [U-'4CJmelanin.Degradation was followed for 5 weeks in a 500-mlGledhill shake flask containing 120 ml of Czapek me-dium fortified with 3% sucrose and 40 mg of melaninprepared from [U-'4C]tyrosine (3.6 x 10' cpm, totalradioactivity). An uninoculated control failed to re-lease CO2 or show evidence of microbial growth duringthe sampling period. The design of the flask permittedfree exchange of filtered air for the C02-free gasesabove the culture surface as well as sampling of boththe liquid medium and CO2 (9). Carbon dioxide wasrecovered at weekly intervals from the gas phase bywithdrawal and replenishment of the Hyamine trap-ping solution (Packard Instrument Co., Inc.) sus-pended in a vial above the culture by a syringeequipped with a no. 13 needle, 15 cm long. Additionalradioactive CO2 was recovered by distillation from 20-ml samples withdrawn from the medium, and thevalues for dissolved CO2 were added to those recoveredfrom the Hyamine trap. Recovery of CO2 was quanti-tative as determined with Na2-'4CO3. Total nonvolatileradioactive metabolites were assessed after evapora-tion. Nonvolatile metabolites in the liquid medium oflow molecular weight were isolated by molecular siev-ing in polyacrylamide gels. Concentrated supernatantsfrom the media clarified by sedimentation at 10,000x g were taken up in 0.5 M ammonium acetate andchromatographed on Bio-Gel P-4 (2 by 30 cm) in thesame buffer. Metabolites eluting between the voidvolume and the peak containing inorganic salts werepooled and lyophilized in preparation for chromatog-raphy on Bio-Gel P-2 (2 by 50 cm). The column wasstandardized with aromatic markers and peptidesranging from 80 to 13,000 daltons. Column develop-

    ment in 0.5 M ammonium acetate, pH 7.8, substan-tially eliminated sorption of benzenoid derivatives bythe gel.

    Distribution of radioactivity in mycelial compo-nents released by hydrolysis in 6 M HCl (0.01% phenol,wt/vol) under N2 in a sealed tube at 100C for 18 hwas monitored by chromatography on a sulfonic acidresin according to a program designed for amino acidanalysis (25). A stream splitter facilitated collection ofamino acids for counting. Chitin was quantified asglucosamine. Materials not retained by the column atpH 2 were classed as acidic metabolites. Additionalradioactivity eluted from the cation exchange afterarginine in 0.2 M NaOH was considered to be associ-ated with basic metabolites. Lipid was isolated inchloroform/methanol (2:1).

    Determination of melanin. Each variable wasassessed by including two controls with each treat-ment. The first tube consisted of the experimentalmedium (2.5 ml) plus 5 mg of insoluble pigment andthe fungal inoculum; the second contained a similaramount of pigment but no inoculum; and the third wasdevoid of pigment but was inoculated. Experimentswere replicated at least four times. Disappearance ofsubstrate was assessed by spectrophotometric assay ofinsoluble melanin remaining at the bottom of thecultures at the end of the incubation period. Fungalpigment was quantified separately. After incubation,the cultures were diluted with 10 ml of 1% NaCl andmixed by mechanical blending. When centrifuged at2,500 x g, all of the residual substrate melanin sedi-mented, whereas the finely dispersed mycelia wereconfined to the supernatant solution. The pelletedmelanin was washed twice by sedimentation in waterand 0.05 M acetic acid and solubilized for spectropho-tometry by heating for 2 h to 100C in 5 M NaOHcontaining 0.15 M sodium borohydride (5). Absorb-ance was measured at 310 nm, and values were com-pared with a standard pure melanin of the same deri-vation. Frequent checks of the spectrophotometricprocedure by gravimetry or nitrogen determinationshowed concordance of values. Recovery of melaninfrom uninoculated controls was quantitative, and ma-terial responding to the spectrophotometric assay formelanin was absent from the culture when melaninwas omitted.

    A. fumigatus is normally silver-yellow and devoidof dark pigment unless grown on melanins or melan-ogens. When pigmented mycelia were present, inter-ference with melanin determination was prevented bydestruction of allomelanin with NaOH-borohydride.When recovery of the fungal pigment was desired, thesaline supernatant with the blended cells was lyophi-lized and extracted first for 48 h in 15 ml of 2 M NaOHand then for 6 days in 15 ml of 6 M HCl (bothtreatments at 20C). The insoluble material harvestedby sedimentation at 1,000 x g was hydrolyzed foramino acid analysis at 110C for 3 h in 6 M HCl underN2 in a sealed tube. The dark residue was pelleted,washed, and dried for gravimetry, nitrogen determi-nation, alkali fusion, and infrared and ultraviolet spec-trophotometry.

    Preparation of melanins. Pigments were pre-pared from puparia of the fleshfly Sarcophaga bullata(8), aged banana skin (21), black human hair (24), and

    APPL. ENVIRON. MICROBIOL.

    on May 26, 2018 by guest

    http://aem.asm

    .org/D

    ownloaded from

    http://aem.asm.org/

  • MELANIN DEGRADATION BY A. FUMIGATUS 147

    inksac of Octopus bimaculoides (22), or were synthe-sized from L-dihydroxyphenylalanine (L-dopa) withalkaline copper sulfate (5). Melanin labeled with 14Cwas prepared from L-[U-'4C]tyrosine with tyrosinasepurified from Psalliotia arvense (20, 26). The reactionmedium contained: L-tyrosine, 0.5,mol; L-[U-'4C]ty-rosine, 85,lCi; and tyrosinase (28 U/mg), 10 mg in 10ml of phosphate buffer, pH 6.8. After incubation for90 h at 25C, the insoluble pigment was sedimented,washed, and extracted twice with 20 ml of 2 M HCl at80C, again washed with water, and dried. Specificactivity was 9 x 105 cpm/mg. L-[U-'4C]tyrosine (spe-cific activity, 75 mCi/mmol) was purchased from NewEngland Nuclear Corp., Boston, Mass. Natural andsynthetic melanins were subjected to elemental anal-ysis and examined in an infrared spectrophotometeras KBr disks. Excellent agreement with publishedvalues was observed (21).

    Analytical methods. Bound phenolic and indolicsubstituents were identified by alkali fusion (11) fol-lowed by thin-layer chromatography on Whatmancellulose 300 CM sheets irrigated with n-butanol-acetic acid-water (60:15:25, vol/vol). Identities wereverified by high-voltage electrophoresis in 0.2 M pyr-idine acetate, pH 4.5. Compounds were visualized byfluorescence quenching at 254 nm or by the Folin-Ciocalteu reagent for phenols. Phenols and indoleswere revealed with FeCl3 in ethyl alcohol and withdiazotized sulfanilic acid (6). Amino acids were de-tected in a Beckman model 119C automatic analyzer(25). Radioactivity was measured in a Packard radi-ochromogram scanner and a Tri-Carb scintillationspectrometer with 10 ml of Aquasol (New EnglandNuclear Corp.); the counting efficiency was 64%. Sam-ples were corrected for quenching by addition of aninternal standard. Fractions from molecular sieveswere scanned in the ultraviolet portion of the spectrumin a Cary recording split-beam spectrophotometer andin the infrared as KBr disks.Enzymatic hydrolysis of fungal cell walls. Pale

    or darkly colored mycelia (100 mg, dry weight) weretreated with 10 mg of a 1:1 mixture of commercialcellulase (Sigma Chemical Co., St. Louis, Mo.) andchitinase (Gallard-Schlesinger, Long Island, N.Y.) for30 min at pH 6.8 at 25C in 10 ml of 0.1 M ammoniumacetate. Samples were prepared for scanning electronmicroscopy by coating with silver conducting paint.After addition of a drop of antistatic formulation anddrying, the stubs were immersed in liquid nitrogen andexamined in an American Metals Research instru-ment, model AMB 1000.

    RESULTSIsolation and growth characteristics of

    A. fumigatus. Soil microflora from garden, for-est, and compost were first enriched on sucrose-nutrient broth and then examined for utilizationof melanin in Czapek medium devoid of sucrose.Organisms from a dark compost fortified withcoffee grounds consumed 10% of the insect mel-anin in 5 weeks, after which the rate increaseduntil 60% was consumed over the next 3 weeks.Lesser rates were maintained by organisms from

    an acid loam rich in pine needles; 15% of thepigment was utilized in 8 weeks. Streak platingof the culture derived from coffee compost af-forded a colony capable of relatively rapid deg-radation of melanin. This isolate was subse-quently identified as A. fumigatus. Microflora ofalternate soil samples were not investigated fur-ther, although significant melanolysis character-ized several other specimens. A significant in-crease in the rate of melanin utilization by A.fumigatus was recorded when sucrose-nutrientbroth was replaced with Czapek broth, whichfavors fungal development; over 80% of the sub-strate was rendered soluble in 6 weeks. Mycelialgrowth was generally restricted to the meniscus;the liquid broth remained transparent with grad-ual darkening. Rate and hue were characteristicfor each type of melanin. Cultures devoid ofmelanin remained clear buff. Microscopic ex-amination of the melanin particles at x400 mag-nification revealed a pale thin coat of myceliasurrounding each granule of substrate. Optimumconditions for growth of A. fumigatus on insectmelanin were assessed by culture in Czapek me-dium adjusted from pH 4.5 to 9.5. Cells at eachpH increment were grown at 23, 30, and 37C toestablish the temperature preference (Fig. 1).Utilization was greatest at 30C, pH 6.5; hence,these conditions were maintained in subsequentstudies. Substitution of molar equivalents of am-monium nitrate for ammonium chloride, or as-paragine for sucrose, did not enhance catabo-lism. Slow agitation with a glass bead improvedpigment breakdown, although the amount ofmixing was insufficient to displace the granulesfrom the bottom of the tube or the fungal matfrom the meniscus. Disappearance of the sub-strate was not hindered by accumulation of darkcleavage products in the medium; active mela-nolysis characterized the entire growth interval.Consumption of different melanins by A.

    pH

    FIG. 1. Growth of A. fumigatus on insect melaninin sucrose-Czapek broth at selectedpH and temper-atures. Time of incubation, 8 weeks.

    VOL. 40, 1980

    on May 26, 2018 by guest

    http://aem.asm

    .org/D

    ownloaded from

    http://aem.asm.org/

  • 148 LUTHER AND LIPKE

    fumigatus. The ability of diverse melanins tosustain growth was ascertained with five pig-ments ofwidely different origin and composition.In agreement with published values for percent-age of nitrogen, the range included: insect, 6.8%;hair, 4.0%; banana, 0.74%; octopus, 6.33%; and L-dopa (synthetic), 8.4%. Polymer individualitywas also confirmed by ultraviolet and infraredspectra and by the products of alkali fusion asrevealed by thin-layer chromatography. Purifi-cation included reflux in strong base or acidwhich effectively reduced amino acid titers to10% or less as verified by amino acid analysis.All melanins examined were degraded by thefungus when presented on the basis of equalweights per incubate (5 mg/2.5 ml) (Fig. 2). Withthe exception of banana black, rate was uniformfrom weeks 5 to 8; synthetic (L-dopa) melaninsustained the most and insect pigment sustainedthe least catabolism. The relative position of thecurves did not vary, although the rate of utili-zation for any one melanin fluctuated 10%between experiments. It is evident, furthermore,that reduced breakdown was a consequence ofomission of sucrose and that the decline wasgreater for L-dopa melanin than for blowfly pig-ment. The color of the liquid medium was notrelated directly to the solubilization rate; prod-ucts of L-dopa melanin degradation imparted ared-beige coloration, whereas insect and bananaafforded chocolate or black.On incubation with melanin generated from

    L-[ U-"C]tyrosine with mushroom tyrosinase,substantial quantities were converted to 14C02,probably at the expense of soluble metabolites(Fig. 3). Whereas insoluble radioactive pigmentwas converted to soluble materials at a fairlyconstant rate for the entire 5-week interval, C02production was enhanced fourfold from weeks 3to 5. When the uninoculated control was exam-

    ~8O

    J20-

    Time (weeks)

    FIG. 2. Degradation of various melanins by A. fu-migatus. Symbols: ( ) complete medium with su-crose added; (---) medium without sucrose; (-) in-sect; (A) dopa; (I) hair; (0) banana; (0) octopus.

    Q4_-S

    ;, 3

    ,2 -L -

    2 3 4 5Time (weeks)

    FIG. 3. Utilization of[U-'4C]melanin (tyrosine) byA. fumigatus. Total counts initially, 3.6 x 10i cpm.Symbols: (A) '4CO2; (5) nonvolatile soluble metabo-lites.

    ined for melanin at the end of the experiment,95% of the starting particulate material was re-covered, but the flask with the fungus containedonly 8%. In both cases, specific radioactivity ofpurified substrate melanin recovered after incu-bation was unchanged from that of the startingmaterial.Chemical studies on products of melanin

    degradation. Radioactive melanin synthesizedwith mushroom tyrosinase was converted to C02and soluble metabolites to the extent of 7 and24%, respectively. Of the remaining label, 56%could be accounted for in the mycelial mat.Fractionation of the radioactive fungus (Fig. 4)revealed 55% of the radioactivity in the cells tobe localized in the dark fungal pigment synthe-sized in the course of development. Fungal lipidretained 4% of the label, with insoluble chitinand soluble protein containing 5 and 6%, respec-tively. Compounds not retained by sulfonic acidcation-exchange columns and not reacting withninhydrin, including neutral sugars, are reportedas unidentified acidic components, and repre-sented about 10% of the label recovered fromthe fungus. Strongly basic ninhydrin-reactivecomponents were eluted with 0.2 M NaOH andcomprised an additional 10% of the label.

    Incorporation of isotope into fungal melaninand protein-derived amino acids is recorded inTable 1. The degree of incorporation into aminoacids could be divided into two groups, thosewith specific activity less than 100 and those of2,200 or greater. Specific activity of the depro-teinized cellular pigment exceeded that ofaminoacids from protein by an order of magnitude ormore. Although half the total radioactivity wasrecovered in the fungal pigment (Fig. 4), thebulk of the carbon in the depot derived fromsucrose as revealed by a 10-fold diminution of

    APPL. ENVIRON. MICROBIOL.

    on May 26, 2018 by guest

    http://aem.asm

    .org/D

    ownloaded from

    http://aem.asm.org/

  • MELANIN DEGRADATION BY A. FUMIGATUS 149

    20r

    FIG. 4. Distribution of radioactivity in A. fumiga-tus cultured for 5 weeks on enriched media with [U-14C]melanin (tyrosine). For preparation of fractions,see text. Total radioactivity in original melanin, 3.6x 107 cpm.

    TABLE 1. Amino acid composition and specificactivity ofA. fumigatus allomelanin-proteincomplex grown on [U- '4C]melanin (tyrosine)a

    Constituent Amt (mol/105 Sp act (cpm/daltons) pxnol)Substrate melanin (ty- 152,(((b

    rosine)

    A. fumigatus melano-protein

    Aspartic acid 110 4,920Threonine 62 4,200Serine 68 4,050Glutamic acid 110 60Proline 34 9,520Glycine 101 50Alanine 90 42Half-cystine 79 40Valine 42 40Methionine 63 5,710Isoleucine 19 40Leucine 20 30Tyrosine 18 4,880Phenylalanine 24 40,B-Alanine 27 3,660Histidine 22 6,250Lysine 68 2,210Arginine 30 5,300Glucosamine 9 2,700

    Allomelanin pigment 14,000ba Procedures for separating amino acids from pig-

    ment are described in the text.b Based on assumed residue weight of 150 daltons.

    specific activity. On the basis of low specificactivity, less than 100 cpm/,umol, the contribu-tion of sucrose to the carbon pool of the proteinfraction exhibited noteworthy selectivity; glu-

    tamic acid, glycine, alanine, valine, leucines, cys-tine, and phenylalanine originated almost en-tirely from exogenous carbohydrate. On theother hand, aspartic acid, serine, threonine, pro-line, methionine, tyrosine, ,B-alanine, histidine,lysine, and arginine were partly derived fromsubstrate melanin, since specific activity rangedfrom 2,210 to 9,520 cpm/,umol for these residues.Although tyrosine furnished carbon directly tosubstrate melanin via enzymatic synthesis, spe-cific activity did not exceed that of other labeledresidues in protein.Approximately 80% of the radioactive mela-

    nin, 29 x 106 cpm, was recovered as cellularconstituents (Fig. 4). An additional 10% waspresent in the medium in soluble form at theend of the incubation period. The average mo-lecular weight of the latter fraction was assessedas 250 by chromatography on Bio-Gel P-2 (Fig.5A). Thin-layer chromatography of peak I re-vealed two components of equal radioactivity atRf 0.35 and 0.63; the former was fluorescent, andthe latter responded to color tests indicative ofindoles. For the component exhibiting Rf 0.35,Amax was 261 nm and Xmim was 280 nm; for thecomponent exhibiting Rf 0.63, Amax was 280 nm

    Mo/ecul/or Weight ( Do/tons)

    q)

    '-3

    I

    1880

    6.0 ,

    4.0 '-3

    2.0 i

    40 80 120 160 200 240 280Eff/uent Vo/ume (m/)

    FIG. 5. Chromatography on Bio-Gel P-2 (2 by 50cm) of metabolites accumulating in media from cul-tures incubated on diverse melanins with 0.5M am-monium acetate buffer. (A) L -[U-'4C]tyrosine mela-nin; (B) insect or L-dopa melanin.

    VOL. 40, 1980

    on May 26, 2018 by guest

    http://aem.asm

    .org/D

    ownloaded from

    http://aem.asm.org/

  • 150 LUTHER AND LIPKE

    FIG. 6. Mycelia grown (A) without and (B) with insect melanin (x400); bar represents 50 jum. Photographedat equal sample thickness and development and exposure times.

    and .min was 330 nm. Both substances reactedpositively with reagents such as FeCl3 and diazo-tized sulfanilic acid, indicating the presence ofphenolic hydroxyl groups. Differences in fungalmetabolites were revealed by chromatographyof products accumulating in the medium aftergrowth on melanins of separate origin (Fig. 5A).Insect melanin afforded two fractions on chro-matography, one component issuing from thecolumn at the void volume and a smaller entityof about 170 daltons. On the other hand, verylittle high-molecular-weight material character-ized the products of L-dopa or tyrosine melanins,the elution patterns of which were similar andin keeping with the close chemical relation be-tween the two precursors (Fig. 5B).Properties ofmycelial pigments from cul-

    tures grown on melanins. Light microscopyrevealed that darkening of the fungal mat origi-nated from deposition of pigment granuleswithin the mycelial strands (Fig. 6). Transfer ofspores from pigmented cultures reared on mel-anin to Czapek broth or slants prepared withoutmelanin resulted in vigorous growth of pigment-free mycelia. The purified mold pigment differedfrom substrate in solubility in alkali and in alighter brown color. The pigment formed by A.fumigatus, from ["4C]melanin (tyrosine), al-though highly radioactive, differed significantlyin infrared spectral features from the startingmaterial, especially at 3.5, 6.5 and 7.2,im (Fig.

    7). It is also evident from infrared scans thatfungal pigment varied in composition with thesource of precursor material. When grown oninsect melanin, a deep inflection was evident at6.6 jim that was absent from pigment derivedfrom tyrosine melanin. Noteworthy differenceswere also recorded in the hydroxyl band at 3 jim.Thin-layer chromatography on cellulose sheetsirrigated with n-butanol-acetic acid-water (60:15:25, vol/vol) revealed 5,6-dihydroxyindole (Rf0.65) and 3,4-dihydroxybenzoic acid (R1 0.81) asproducts of alkali fusion of both substrate mel-anins, but not of the corresponding fungal pig-ments.Fungal susceptibility to enzymatic lysis.

    Scanning electron microscopy clearly showedthat melanized mycelia were more resistant to30-min digestion by f3-(1,3)-glucanase and chiti-nase than cultures grown in the absence of mel-anin. There was extensive pitting of the conidio-phore and loss of sculptured detail on sterigmataand conidia from pale specimens grown on mel-anin-free media (Fig. 8B). When enzyme treat-ment was prolonged to 90 min, many features ofthe darkened mycelia were still recognizable,whereas controls were reduced to thin amor-phous nodules. Mycelia developing on tyrosinemelanin and insect melanin could not be distin-guished on the basis of stability to lytic carbo-hydrases in spite of differences in chemical struc-tures.

    APPL. ENVIRON. MICROBIOL.

    on May 26, 2018 by guest

    http://aem.asm

    .org/D

    ownloaded from

    http://aem.asm.org/

  • MELANIN DEGRADATION BY A. FUMIGATUS 151

    m

    st.inm

    re

    o08ci

    sil

    ot

    w,

    tbw,

    n(bEti4ic

    bffotlhal

    Wavelength (nm) nous structural polysaccharides (4). High-molec-4 6 8 10 12 14 ular-weight tannins from pine bark retained

    A ^ more than half the polymer weight after 10 days.10- of exposure to a strain of Penicillium adametzi.201 / \ =capable of degrading the (+)-catechin precursor.40- W and procyanidin dimers and trimers essentially.6C _ to completion in 50 h (10). Substantial levels of

    L.0- humic acids persisted in cultures of Aspergillus,Arthrobacter, Pseudomonas, and PenicilliumB after 7 weeks (19). In several instances, melan-

    .10C Aogenic phenols were released, suggesting a ready.20 XJJ - exchange of carbon between the two polyanions.

    A. fumigatus developed optimally at 30C, pH.64r6.5, when cultured in Czapek medium (Fig. 1).Visual inspection ascertained that fungal growth

    r.0- was essentially complete in 2 weeks, probablyc coinciding with exhaustion of sucrose. That mel-

    .20- anin could be utilized as a carbon source was

    .40 V >indicated by disappearance of 40 to 50% of the

    .60- 7/substrate after 8 weeks in medium with sucrosel.0o omitted (Fig. 2). In general, a heavier mycelialmat was observed with sucrose added, suggest-

    D / ~ i ing that the increased melanin catabolism in the10 / \ ,presence of this auxiliary carbon source resulted

    .2c-0 / \ {\- / from a greater number of active organismsrather than a higher titer of melanoclastic en-

    .40 zymes per cell. Although a lag period of 5 or 6.60- weeks usually preceded the interval with great-IofC est utilization on minimal medium, a linear rate4000 2000 1200 900 700 was representative of cultures offered sucrose

    Wovenumber (cm') initially.Eumelanins show great diversity with respect

    FIG. 7. Infrared spectra of substrate and fungal t uce, molecular weit,and assoctelanins. (A) Substrate melanin (insect); (B) sub- to structure, molecular weight, and associationrate melanin (tyrosine); (C) fungal allomelanin from with protein when in the native state (24). Pig-tsect melanin; (D) fungal allomelanin from tyrosine ments from hair, insect, octopus, banana, andNelanin. the two synthetic polymers derived from tyro-

    sine and L-dopa were consumed by the fungus

    DISCUSSION after removal of 80 to 85% of the protein byprolonged extraction with concentrated HCI orAs pH 4.5 to 8.5 and 23 to 37C, A. fumigatus NaOH. Growth, therefore, was at the expense ofXadily degraded melanins of widely different the benzenoid and indolic polymer constituents.rigins and compositions over the course of 4 to The identity of the substrates was confirmed byweeks on Czapek medium (Fig. 1 and 2). The examination in the infrared, by elemental anal-ilture from mixed coffee and garden compost ysis, and by alkali fusion where the differencesas chosen for further study on the basis of were manifest, principally in the ratio of the twognificantly higher lytic activity compared with most stable cleavage products, 3,4-dihydroxben-ther sources; nevertheless, melanin degradation zoic acid and 5,6-dihydroxyindole (10). Regard-as measured in six different soil samples from less of the origin of the polymer, A. fumigatusie Boston area. The rate of substrate utilization subsisted vigorously on the pigment, convertingas comparable with that of other arylated phe- more than 80% to cellular components such as)lic polymers joined by carbon-carbon bonds protein, polysaccharide, fungal pigment, and sol-etween adjacent rings, the resonance stabiliza- uble by-products accumulating in the mediumon of the benzenoid substituents and free-rad- (Fig. 3 and 4). Whereas slow evolution of CO2al longevity protecting not only cyclized loci from tyrosine melanin added to potting soilut side chains as well (14, 24). For example, could be construed as decarboxylation of periph-rest soils required 700 h for oxidation of 5% of eral phenolic acids (2, 14), in the present studyie rings and side chains of hemlock lignin, conversion of 90% of the polymer toCO2, solublepproximately 1/io the rate sustained by endoge- metabolites, and other cellular constituents at-

    VOL. 40, 1980

    q)

    on May 26, 2018 by guest

    http://aem.asm

    .org/D

    ownloaded from

    http://aem.asm.org/

  • 152 LUTHER AND LIPKE

    FIG. 8. Effect of /3-glucanase and chitinase on exterior fine structure of A. fumigatus with normal andmelanizedpigmentation. Scanning electron microscopy (x1,600). (A) Czapek-sucrose grown; (B) same mediumplus insect melanin. For details, see text.

    tests to appreciable depolymerization and ringscission by A. fumigatus. Gel fitration estab-lished that the bulk of the radioactivity in themedium was in the molecular weight range of150 to 250 (Fig. 5). Ultraviolet-absorbing mate-rial from the major fraction, peak I, comprisedthe bulk of the counts applied to the column and

    also responded to the FeCl3 test for hydroxyin-doles. The ultraviolet spectrum, infrared scan,and mobility on thin-layer chromatograms weresimilar to those of authentic hydroxyindoles,although definitive characterization was impos-sible because of instability of both fungal prod-ucts and standards. Soluble metabolites could

    APPL. ENVIRON. MICROBIOL.

    on May 26, 2018 by guest

    http://aem.asm

    .org/D

    ownloaded from

    http://aem.asm.org/

  • MELANIN DEGRADATION BY A. FUMIGATUS 153

    FIG. 8B.

    originate directly from polymer scission or couldbe secondary products of fungal metabolism.With the exception of C02, the products accu-mulating in the medium probably did not rep-resent initial catabolites, the majority of whichmay have been highly reactive and prone tospontaneous oxidation and rearrangement underthe culture conditions.

    Tyrosine melanin uniformly labeled with "C

    contributed materially to cell composition (Fig.4). Protein plus chitin accounted for 11% of theradioactivity, lipid accounted for 4%, and uni-dentified basic and acidic components accountedfor 10% each. The highly charged acidic andbasic fractions included aromatics judged byabsorption maxima between 250 and 280 nm,probably consisting of phenolic acids, amino-phenols, and indoles. The mole ratios of the

    VOL. 40, 1980

    on May 26, 2018 by guest

    http://aem.asm

    .org/D

    ownloaded from

    http://aem.asm.org/

  • 154 LUTHER AND LIPKE

    amino acids in the fungal hydrolysate showed apreponderance of aspartic and glutamic acidsand little histidine, in many respects resemblinga similar fraction from E. echinulatum (23).After deproteinization, the bulk of the radioac-tivity (55%) was recovered as a dark fungal pig-ment, insoluble in HCI and resembling allo-melanins from A. niger and pigmented Verticil-lum and Penicillium in hydrolysis products (27).Production of fungal pigment exceeded exoge-nous melanin sixfold, establishing that de novosynthesis accompanied melanolysis. The readytransfer of radioactivity from substrate to fungalpigment, therefore, was not the consequence ofengulfment or occlusion, but did result frominduction of an enzyme system in the unpig-mented cells transforming the products of mel-anin catabolism to a new polymer. This systemwas moderate in specificity, producing dark fun-gal polymers with different infrared records de-pending on the melanin of origin (Fig. 7).The pathway of melanin catabolism in the

    transfer of carbon to allomelanin and cell proteinis obscure. In the presence of sucrose, significantrandomization of label was observed in asparticacid, serine, threonine, methionine, tyrosine, 86-alanine, histidine, lysine, arginine, and glucosa-mine as judged by a diminution of specific activ-ities compared with the substrate. The virtualabsence ofradioactivity in glutamic acid, glycine,alanine, and cystine indicates distinct compart-ments for enzyme systems responsible for syn-thesis of melanoproteins and cell proteins in thepresence of exogenous carbohydrate. Such a dis-tribution would obtain if both peptide and pig-ment moieties of allomelanin were synthesizedsolely from radioactive substrate melanin, aminoacids associated directly with the fungal pigmentconsisting of those listed above as characterizedby appreciable radioactivity. Synthesis of thoseresidues in the cell protein, on the other hand,could derive principally from sucrose and am-monium salts, affording amino acids of low spe-cific activity on hydrolysis.A. niger, A. nidulans, A. phoenicia, and other

    dark species release lesser amounts of cell wallcarbohydrate than do unpigmented strains afterexposure to chitinase and glucanase (2, 3, 14).These observations have been extended to A.fumigatus, using microscopy as a criterion. Evi-dence of extensive modification of unpigmentedmycelia can be deduced from Fig. 8; the damageincluded fusion of conidial protuberances, invag-ination of interconidial regions, and blistering ofthe sterigma and conidiophore. Digestion wasrapid and essentially complete in 90 min. Inhi-bition of hydrolases by exogenous melanin hasbeen ascribed to free radicals trapped by the

    pigment interacting with sulfhydryl or other sen-sitive groups on the enzymes (3, 14). Undernatural conditions, rates of melanin degradationwould be influenced not only by soil pH and theavailability of nutrients, but also by the presenceof other organisms utilizing intermediates re-leased from the melanoprotein complex by A.fumigatus.

    ACKNOWLEDGMENTSWe thank Margaret Bigelow, of the University of Massa-

    chusetts/Amherst, and the Taxonomy Group of the U.S.Department of Agriculture Fernentation Laboratory, Peoria,Ill., for taxonomic advice. Jerome Woodinsky, of BrandeisUniversity, generously provided inksacs from Octopus bima-culoides. Invaluable assistance with isolation of melanin ca-tabolites was provided by E. Zomer, P. Banda, and KeithStrout.

    This investigation was supported in part by National Sci-ence Foundation grant PCM 75-22123.

    LITERATURE CITED1. Blois, M. S. 1978. The melanins: their synthesis and

    structure. Photochem. Photobiol. Rev. 3:115-134.2. Bloomfield, B. J., and M. Alexander. 1967. Melanins

    and resistance of fungi to lysis. J. Bacteriol. 93:1276-1280.

    3. Bull, A. T. 1970. Inhibition of polysaccharases by melanin:enzyme inhibition in relation to mycolysis. Arch. Bio-chem. Biophys. 137:345-356.

    4. Crawford, R. L., D. L. Crawford, C. Olofsson, L.Wikstrom, and J. M. Wood. 1977. Biodegradation ofnatural and man-made recalcitrant compounds withparticular reference to lignin. J. Agric. Food Chem. 25:704-707.

    5. Das, K. C., M. B. Abramson, and R. Katzman. 1976. Anew chemical method for quantifying melanin. J. Neu-rochem. 26:695-699.

    6. Egger, K. 1969. Hydrophilic plant constituents and theirderivatives, p. 687-706. In E. Stahl (ed.), Thin layerchromatography. Springer-Verlag, New York.

    7. Faiziev, I. 1965. Enzymatic hydrolysis of a melanin-likesubstance produced by actinomycete. Dokl. Akad. NaukTadzh. SSR 8:27.

    8. Fogal, W., and G. Fraenkel. 1969. Melanin in the pu-parium and adult integument of the fleshfly, Sarco-phaga bullata. J. Insect. Physiol. 15:1437-1447.

    9. Gledhill, W. E. 1975. Screening test for assessment ofultimate biodegradability: linear alkylbenzene. Appl.Environ. Microbiol. 30:922-929.

    10. Grant, W. D. 1976. Microbial degradation of condensedtannins. Science 193:1137-1140.

    11. Hackman, R. H., and M. Goldberg. 1971. Microchemi-cal detection of melanins. Anal. Biochem. 41:279-285.

    12. Haider, K., and J. P. Martin. 1970. Humic acid-typephenolic polymers from Aspergillus sydowi culture me-dium, Stachybotrys spp. cells and autoxidized phenolmixtures. Soil Biol. Biochem. 2:145-156.

    13. Haider, K., J. P. Martin, and Z. Filip. 1975. Humusbiochemistry, p. 218-227. In E. A. Paul and A. D.McLaren (ed.), Biochemistry, vol. 4. Marcel Dekker,Inc., New York.

    14. Kuo, M. J., and M. Alexander. 1967. Inhibition of thelysis of fungi by melanins. J. Bacteriol. 94:624-629.

    15. Linhares, L. F., and J. P. Martin. 1978. Decompositionin soil of the humic acid-type polymers (melanins) ofEurotium echinulatum, Aspergillus glaucus sp. andother fungi. Soil Sci. Soc. Am. J. 42:738-743.

    16. Martin, J. P., and K. Haider. 1971. Microbial activity in

    APPL. ENVIRON. MICROBIOL.

    on May 26, 2018 by guest

    http://aem.asm

    .org/D

    ownloaded from

    http://aem.asm.org/

  • MELANIN DEGRADATION BY A. FUMIGATUS 155

    relation to soil humus formation. Soil Sci. 111:54-63.17. Martin, J. P., and K. Haider. 1979. Biodegradation of

    '4C-labeled model and cornstalk lignins, phenols, modelphenolase humic polymers, and fungal melanins as in-fluenced by a readily available carbon source and soil.Appl. Environ. Microbiol. 38:283-389.

    18. Martin, J. P., K. Haider, and D. Wolf. 1972. Synthesisof phenols and phenolic polymers by Hendersonulatoruloidea in relation to humic acid formation. Soil Sci.Soc. Am. Proc. 36:311-315.

    19. Mathur, S. P., and E. A. Paul. 1966. A microbiologicalapproach to the problem of soil humic acid structure.Nature (London) 212:646-649.

    20. Miller, W. H., and C. R. Dawson. 1942. Factors influ-encing the catecholase activity and activation of tyro-sinase. J. Am. Chem. Soc. 63:3368-3374.

    21. Nicolaus, R. A. 1968. Melanins. Hermann, Paris.22. Nicolaus, R. A., M. Piatelli, and E. Fattorusso. 1964.

    The structure of melanins and melanogenesis, part IV.Tetrahedron 20:1163-1172.

    23. Saiz-Jiminez, C., K. Haider, and J. P. Martin. 1975.Anthraquinones and phenols as intermediates in theformation of dark-colored humic acid-like pigmnents by

    Eurotium echinulatum. Soil Sci. Soc. Am. Proc. 39:649-653.

    24. Swan, G. A. 1974. Structure, chemistry and biosynthesisof the melanins Fortschr. Chem. Org. Naturst. 31:521-582.

    25. Thathachari, Y. T. 1971. Physical studies on melanins.J. Sci. Ind, Res. 30:529-537.

    26. Thomson, A. R., and B. J. Miles. 1964. Ion exchangechromatography of amino acids: improvements in thesingle column system. Nature (London) 203:483-484.

    27. VanWoert, M. H., K. N. Prasad, and D. C. Borg. 1967.Spectroscopic studies of substantia nigra pigment inhuman subjects. J. Neurochem. 14:707-716.

    28. Wheeler, M. H., W. J. Thomsoff, A. A. Bell, and H. H.Mollenhaur. 1978. Ultrastructural and chemical dis-tinction of rnelanins formed by Verticillium dahliaefrom (+)-scytalone 1,8-dihydroxynaphthalene, catecholand L-3,4-dihydroxyphenylalanine. Can. J. Microbiol.24:289-297.

    29. Whittaker, J, R. 1979. Absence of a direct melanin-protein relationship in chick embryo melanocytes.Biochim, Biophys. Acta 583:378-387.

    VOL. 40, 1980

    on May 26, 2018 by guest

    http://aem.asm

    .org/D

    ownloaded from

    http://aem.asm.org/