Academia.eduAcademia.edu
481 Mycol. Res. 98 (5): 481-492 (1994) Printed in Great Britain Chemotaxonomy of Penicillium aurantiogriseum and related species FLEMMING L U N D A N D TENS C. FRISVAD Department of Biotechnology, Building 221, The Technical University of Denmark, DK-2800 Lyngby, Denmark Penicillium aurantiogriseum sensu hto is the most common cereal-borne Penicillium species of worldwide occurrence and it has a complex taxonomy. A total of 519 isolates, including most of the NRRL cultures used by Raper & Thom (1949) in their P. cyclopium, P. viridicatum and P. ochraceum series, was examined for expressions of differentiation, especially of secondary metabolites. The fungi were micromorphologically quite similar. Isolates previously allocated to P. cyclopium and P. viridicatum were separated into nine species based on biosynthetic families of secondary metabolites and from several less conspicuous morphological and physiological differences. The species accepted were P. aurantiogriseum, P. aurantiovirens, P. cyclopium, P. freii, P. melanoconidium, P. neoechinulaturn, P. polonicum, P. tricolor and P. viridicatum. These species produced characteristic combinations of the following mycotoxins: xanthomegnin, viomellein, vioxanthin, penicillic acid, terrestric acid, viridic acid, venucosidin, penitrem A and the other secondary metabolites: viridamine, aurantiamine, auranthine, venucofortine, pubemline, cyclopeptin, dehydrocyclopeptin, cyclopenin, cyclopenol, 3-methoxyviridicatin, viridicatol, brevianamide A & B, meleagrin and oxaline. A11 the known secondary metabolites could be allocated to biosynthetic families based on standards and liquid chromatography - diode array detection. Some isolates of P. aurantiogriseum and P. polonicum are known to produce nephrotoxic glycopeptides possibly associated with Balkan endemic nephropathy. It is recommended that isolates should be identified to the species level using a combination of morphological, physiological and secondary metabolite data. Penicillium aurantiogriseum Dierckx and related species are probably the most common fungal species on cereals and they are of worldwide occurrence (Raper & Thom, 1949; Mulinge & Chesters, 1970; Ciegler et al., 1973; Pitt, 1979; Abramson, 1991; Frisvad & Samson, 1991; Lund ef al., 1992). They have very often been identified under other names such as P. cyclopium Westling, P. puberulum Bainier, P. martensii Biourge, P. viridicafum Westling, P. ochraceum Bainier apud Thom, P. olivicolor Pitt, P. aurantiovirens Biourge, P. olivinoviride Biourge and P. auranfiocandidum Dierckx (Frisvad. 1989). Based on physiological, ecological and secondary metabolite data, Frisvad (1981) suggested that isolates of P. yclopium, P. viridicatum and related taxa in the series of Raper & Thorn (1949) should be combined into one taxon. This Frisvad provisionally named 'P. cyclopium p', the 'p' standing for penicillic acid. Puberulic acid, xanthomegnin and viomellein were later found in several isolates of these taxa (Wirth & Klosek, 1972; Stack & Mislivec, 1978; Frisvad & Filtenborg, 1983). We propose to emend the series Viridicafa Raper & Thom ex Pitt (Pitt, 1979) to include only the cereal associated polythetic group of terverticillate penicillia in which the taxa have different combinations of the following characters: bluegreen or grey-green or green or dark green colour, small (2.5-3.5 1.im)globose to subglobose (rarely ellipsoidal),smooth to finely roughened (rarely echinulate) conidia; finely to conspicuously roughened conidiophore stipes; yellow reverse on yeast extract sucrose (YES) agar; inhibited growth on creatine-sucrose agar; and the secondary metabolites listed in Table 1. Several other secondary metabolites have been reported from P. auranfiogriseum (Mantle, 1987) but the data were based on misidentified isolates (Frisvad, 1989). Based solely on morphological and physiological criteria, P. auranfiogriseum is difficult t o delimit against other common terverticillate Penicillium species, and therefore data on production of secondary metabolites are necessary for the species identification. Frisvad & Filtenborg (1989) suggested a subdivision of P. aurantiogriseum into the varieties aurantiogriseum, melanoconidium Frisvad, neoechinulaturn Frisvad, Filt. & Wicklow, polonicurn (W. Zalessky) Frisvad and viridicatum (Westling) Frisvad & Filt, because there were consistent patterns in the profiles of the secondary metabolites which were related to macromorphological features. Other isolates placed by Raper & Thom (1949) in their P. cyclopium or P. viridicafum series, or identified as belonging to those series by later researchers (Samson, Stolk & Hadlok, 1976) are now included in P. solifum Westling, P. commune Thom, P. crtlsfosum Thom and P. verrucosum Dierckx. These species could be distinguished from P. aumntiogriseum and related species by the ~roduction of different secondary metabolites, different ecological features and different physiological characters (Frisvad, 1981, 1985 b; Frisvad, 1989, Frisvad & Filtenborg, 1989; Bridge ef al., 1989; Pitt & Cruickshank Penicillium auranfiogriseum chemotaxonomy 482 Table 1. Reported production of mycotoxins' and other secondary metabolites by Penicillium aurantiogriseum and related species Representatives of metabolite families References Penicillic acid' 1 Orsellinic acid Xanthomegnin' Viomellein' Vioxanthina Rubrosulphinb Viopurpurinb Xanthoviridicatin D & Gb Vermcosidin" 2 + + Lindenfelser & Ciegler, 1977; Wirth & Klosek, 1972; Ciegler & Kurtzman, 1970; Northolt et al., 1979; Thorpe & Johnson, 1974; Albright et al., 1964; Birkinshaw, Oxford & Raistrick, 1936; Le Bars, 1979; Alsberg & Black, 1913; Wirth, Gilmore & Noval, 1956; Ciegler et al., 1972a & b, 1973; Kobayashi, Tsunoda & Tatsuno, 1971; Bentley & Kiel, 1962; Kurtzman & Ciegler, 1970 Bentley & Keil, 1961 + + 3 Normethylvermcosidinb Viridic acid' Penitrem A' Penitrem S F h Nephrotoxic glycopeptidesa,b Aurantiamine Viridamine 3-methoxyviridicatin Cyclopeptin Dehydrocyclopeptin Cyclopenin Cyclopenol Viridicatol Puberulic acidb Puberulonic acid Brevianamide A & B Terrestric acid Verrucofortine ( = Verrucosine = fructigenine B) Puberuline (fructigenine A) + + + + + + Rugulosuvine Leucyltryptophanyldiketopiperazine Oxaline Meleagrin Auranthine a Stack & Mislivec, 1978; Stack et al., 1977 Burka, Ganguli & Wilson, 1983; El-Banna & Leistner, 1987, 1989; Frisvad & Filtenborg, 1989; Wilson, Byerly & Burka, 1981 Hodge, Hanis & Harris, 1988 Holzapfel, Koekemoer & van Dyk, 1986 Frisvad & Filtenborg, 1989 De Jesus et al., 1983 MacGeorge & Mantle, 1990; Yeulet et al., 1988; Mantle et a!., 1991a Larsen, Frisvad & Jensen, 1992 Holzapfel & Marsh, 1977 Frisvad 1985a; Hodge, Harris & Harris, 1988; Frisvad & Filtenborg, 1989; Cutler et al., 1984 Birkinshaw & Raistrick, 1932; Oxford, Raistrick & Smith, 1942 Wilson, Yang & Harris, 1973 Frisvad & Filtenborg, 1983, 1989 Hodge et al., 1988; Arai et al., 1989 Kozlovsky, 1990 Solovyeva et al., 1992 Solovyeva et a/., 1989 Frisvad & Filtenborg, 1989; Nagel et al., 1976 Frisvad & Filtenborg, 1989; Kawai et al., 1984 Yeulet et al., 1986 Known mycotoxins. Standards not available for this study. (1990), Bridge, Kozakiewicz & Paterson (1992); Stolk ef a]., 1990) and will not be considered further in this treatment. Three different types (groups) of nephrotoxins may be produced by P. auranfiogriseum, P. viridicatum and P. verrucosum, the three most common filamentous fungi associated with cereals (Frisvad & Filtenborg, 1989). Glycopeptides and/or xanthomegnin are produced by the two former species and ochratoxin A, citrinin and oxalic acid by the latter. It is also of interest that Zwicker, Carlton & Tuite (1983) reported on carcinogenicity in mice caused by P. viridicafum (isolate Purdue 66-68-3) grown on rice. Fungi within P. auranfiogriseum sensu Frisvad & Filtenborg (1989) have been reported to cause nephropathy in swine and renal and hepatic lesions in mice. The fungi causing these experimental mycotoxicoses were P. viridicafum Purdue 6668-2 = NRRL 6430, and P. ochraceum NRRL 869 and NRRL 871. These cultures were unable to produce the important nephrotoxins ochratoxin A, citrinin and oxalic acid (Carlton, Tuite & Mislevic, 1968; Budiarso, Carlton & Tuite, 1971; Carlton, Tuite & Caldwell, 1972; Zimmermann, Carlton & Tuite, 1979). It was later shown that the mycotoxins involved were xanthomegnin and viomellein (Stack ef al., 1977; Robbers ef al., 1978; Stack & Mislivec, 1978; Ueno, 1991). Ciegler, Lee & Dunn (1981) and Stack & Mislivec (1978) found that several strains of P. viridicafum and one out of nine isolates of P. cyclopium produced xanthomegnin and viomellein. Viomellein was later found to occur naturally in Danish barley (Hald ef al., 1983) and xanthomegnin, viomellein and vioxanthin in British wheat, barley and rape (Scudamore, Atkin & Buckle, 1986). P. aumntiogriseum has been suggested as a possible factor in the etiology of Balkan endemic nephropathy (Barnes ef al., 1977). Later Yeulet ef al. (1988) and Macgeorge & Mantle (1990, 1991) found nephrotoxicity in rats using cultured F. Lund and J. C. Frisvad 483 Table 2. Production of mycotoxins and other secondary metabolites by Penicillium aurantiogriseum and related taxa specified as % frequency. Metabolite families 6 and 9 are based on the literature (Table I) Speciesa Metabolite family 9 I0 I1 12 I3 14 Penicillic acid Xanthomegnin Vermcosidin Viridic acidb Penitrem A Nephrotoxic glycopeptides Aurantiamine Viridamine 3-methoxyviridicatin Puberulic acid Brevianamide A Terreshic acid Vermcofortineb and/or Pubem1ineD Oxaline Auranthineb I I1 + III N V VI VII VIII IX 7 7 7 7 7 7 + 0 91 0 0 0 0 0 100 100 0 0 100 0 0 100 88 0 0 0 0 100 0 0 0 0 0 0 0 100 0 0 0 0 0 0 0 0 0 0 0 0 100 0 0 0 7 Number of isolates analyzed by TLC: by HPLC: " I. P. aurantiogriseum; 11, P. freii; 111, P. tricolor; IV, P. polonicum; V, P. aurantiovirens; VI, P. viridicatum; VII, P. cyclopium; VIlI, P. melanoconidium; IX, P. neoechinulatum. The metabolite was only detected by HPLC. Produced in trace amounts only detected by HPLC. A ' ' shows that at least some isolates produce the metabolites A '7' shows that no data are available. + mycelia of several isolates from Balkan countries, and showed that the toxins were glycopeptides (Mantle ef al., 1991b). They also reported that P. commune produced these glycopeptides. This work examines the morphology, mycotoxins and other secondary metabolite profiles of isolates in the Penicillium aurantiogriseum complex and their possible allocation to homogeneous taxa. University of Denmark, DK-2800 Lyngby, Denmark (Lund ef al., 1992). Media Substrates included Czapek yeast autolysate agar (CYA) (Pitt, 1979), yeast extract sucrose agar (YES) (Frisvad & Filtenborg, 1989) and malt extract agar (MEA) (Pitt, 1979). The YES was made from Difco yeast extract to which was added 0.5 g MATERIALS A N D METHODS MgS0,. 5H,O 1-I (Filtenborg, Frisvad & Thrane, 1990), and MEA was made up from Difco ingredients. The media Fungi contained an additional trace element solution (Frisvad & Isolates were obtained from the National Center for Filtenborg, 1983). The media were three-point inoculated Agricultural Utilization Research (NRRL and Q M numbers), using a conidium suspension made from cultures grown on Peoria, IL, U.S.A.; Division of Food Research (FRR numbers), CYA for 7 days. Representative isolates of each species were Ryde, New South Wales, Australia; Centraalbureau voor also inoculated on DG18 (Oxoid) (Hocking & Pitt, 1980) and Schimmelcultures (CBS),Baarn, the Netherlands; International Merck's malt extract agar (El-Banna & Leistner, 1989) as these Mycological Institute (IMI), Egham, Surrey, United Kingdom; media were optimal for production of verrucosidin, xanthoAll Union Collection of Microorganisms (VKM), Moscow, megnin and viomellein (El-Banna & Leistner, 1989; Lund and Russia; Council for Scientific and Industrial Research (CSIR), Frisvad, unpublished). Other media used were SYES made National Chemical Research Laboratory, Pretoria, Republic of from Sigma yeast extract (YES modified to include Sigma YSouth Africa; National Collection of Fungi (PREM), Pretoria, 4000 yeast extract instead of Difco yeast extract), oat meal Republic of South Afnca; and University of Alberta agar (OAT) (Samson & van Reenen-Hoekstra, 1988), and Microfungus Collection and Herbarium (UAMH), Edmonton, creatine-sucrose agar (CREA) (Frisvad, 1981). Alberta, Canada. In addition, we examined all the P. aurantiogriseum and P. viridicaturn isolates and representatives Growth conditions of all other species in series Viridicafa from the collection at the Department of Biotechnology (IBT), The Technical The fungi were incubated for 7-14 days at 25 OC without Penicillium auranfiogriseum chemotaxonomy 484 Table 3. Profiles of mycotoxins and other secondary metabolites produced by P. aurantiogriseum and related species Metabolite profile P. aurantiogriseum IBT numbers* + + + + Aurantiamine Vermcosidin Terrestric acid Penicillic acid Auranthine" + + + Aurantiamine + Venucosidin + Aurantiamine Venucosidin Terrestric acid Auranthinea Auranthinea Aurantiamine Penicillic acid P. freii + Vermcosidin+ + I0527 = FRR 343, 11282 Aurantiamine 3-methoxyviridicatin Xanthomepin Viomellein + 3464 = CBS 183.89, 3514, 3515, 3516, 3517, 3520, 3521, 3523, 3525, 3527, 3529. 3530. 3531, 3532 = FRR 1103, 3699, 3903, 4363, 4368, 4373, 4532, 4585. 4775. 4777. 5124. 5132, 5137 = FRR 2935 = IMI 285515 = CBS 476.84. 5143, 5144, 5146 = NRRL A-27011. 5153,6167, 6690'. 6694b, 10004,10013, 10056, 10107,10167, 10190,10204,11238,11242, 11244, 11247, 11249, 11250 = IMI 357297, 11255, 11256, 11257, 1125Sb, 11260, 11266b, 11271. 11273, 11274, 11276, 11280. 11281, 11284'. 11289 = IMI 357298, 11290, 11298, 11306, 11307, 11308, 11310, 11314', 11316'. 11318', 11326', 11327, 11329, 11357, 11358, 11364, 11365, 11366. 11369. 11374 = IMI 357296, 11386, 11392, 11400, 11514, 11517, 11662, 11996 = CSIR 1876, 12794, 12796, 12835 = NRRL 951 = FRR 951, 14258 = IMI 114931 = CBS 630.66 5147, 5168, 11285°, 11295'. 11521' + + Aurantiamine 3-methoxyviridicatin Xanthomepin Viomellein Penicillic acid Aurantiamine 3-methoxyviridicatin + + P. tricolor + + + + + Xanthomegnin Viomellein Terrestric acid Vermcofortinec P. polonicum + + Venucosidin 3-methoxyviridicatin Penicillic acid Venucofortinec f Pubemlinea + + + + + + + P. aurantiovirens + 12176' = IMI 321328, 14121" = IMI 183170, NRRL 2035* 3-methoxyviridicatin Penicillic acid Vermcofortinec Pubemlined + 11663e = IMI 357306, 1247IC",12493cd.12494cd= IMI 357307, 12957cd 3109, 3180, 3226, 3447" = PREM 47750, 3451Cd= NRRL A-23312, 3463a, 4571, 4681". 5131, 5157" = NRRL 5570, 6156c = NRRL 2027, 6285' = CBS 690.77, 11245", 11294, l133Sd = IMI 357305, 11378, 11381, 11382, 11383" 11388, 11410Cd= NRRL 3608, 12663". 12822cd= NRRL 6314, 12826C= NRRL 6316, 12828' = NRRL 6315 = IM1357303, 14320' = IMI 351304, l432lcd, UAMH 4009 3449'". 345OCd,3522, 4501, 6687, 6692, 11239 = IMI 357304, 11319, 11334, 12821C= NRRL 995, 12827"" = NRRL 2029, 14250"". IMI 280215" + Vermcosidin 3-methoxyviridicatin Vermcofortinec Pubemlined 3-methoxyviridicatin Vermcofortinec Pubemline" 3-methoxyviridicatin Venucofortinec Puberdined+ Penicillic acid + 3471, 3992a = IMI I80922a. 5134" = NRRL 3672, 4686 = FRR 343, 6215, 6689, 6691, 10023, 10047. 11252. 11261, 11264, 11265 = IMI 357289, 11277, 11304 = IMI 357290, 11309' = NRRL 3612, 11311, 11313, 11321" = NRRL 3564, 11325, 11359, 11361, 11377, 11379, 11384, 11390, 11393, 11402 = IMI 357291, 11516, I1624a, 116358, 11660". 11672%, 12480a, 12482". 12726' = NRRL 6318, 12836a = NRRL 6317, 12954%, 12958'. 13548" = NRRL 971, 14016 = CBS 324.89. 14118, 14264 = NRRL 953 4144CCF 1940, 5268 = CCF 1275, 6103, 10037, 11275, 11278, 11283, 11287, 11291, 11293, 11296, 11301 = IM192235, 11303, 11335, 11342, VKM F-1298" 3984" = IMI 180922, 4095, 12834a = NRRL 3747, 14136 = UAMH 4334 + 3544C= NRRL 2138 = IMI 34846, 4151d, 6275'" = CBS 475.84 = IMI 285514 = FRR 2934, 6329, 6464, 10086. 11241, 11246 = IMI 103744, 11259cd, 11270, 1130OCd,11320, 11322, 11330, 11336'" = IMI 357292, 11337, 11362, 11362, 11367, 11368, 11373. 11375'", 11376, 11385'" = IMI 357293, 11387, 11389'", 11391, 11394. 11395, 11401, 11413, 11414, 12793Cd,12839C,14123Cd= IMI 159109. 14258 = CBS 630.66, 1431SC= NRRL 952, VKM F-310' 3946, 6168'". 6274Cd,6465C,6693, 11240, 11243, 11245, 11254, 11292, 11297c= CBS 162.81, 11305, 11317, 11323. 11328Cd,1136OCd.11371Cd, 12165", 12841Cd= NRRL 956, 12843' = NRRL 881, 14124cd= IMI 204208. 14134 = CBS 434.73. VKM F-64aC F. Lund and J. C. Frisvad 485 Table 3 (cont.) Metabolite profile IBT numbers' - + 12820d = NRRL 1899, 12838d = NRRL 2137, CBS 742.74 = NRRL 2010, NRRL 884 Pubemlined Penicillic acid P. viridicatum + + + + + + + + + + + + + Xanthomegnin Viomellein Viridamine Brevianamide A Viridic add' + Xanthomegnin ViomeUein Viridamine Xanthomegnin Viomellein Brevianamide A Viridic acid' Xanthomegnin Viomellein Viridamine Brevianamide A Penicillic add Viridic acid' P. cyclopium + 3-methoxyviridicatin Xanthomegnin Viomellein Penicillic acid VemcofortineCt Puberulined + + + 3-methoxyviridicatin Xanthomegnin Viomellein Vermcofortinec Pubemlined Xanthomegnin Viomellein Penicillic acid + + + + + P. melanoconidium + + + Vermcosidin Oxaline Penitrem A Penicillic acid + Aurantiamine 3-methoxyviridicatin Penicillic acid + Aurantiamine viridicatin + 3-methoxy- 3-methoxyviridicatin Penicillic acid + 3110, 345F = NRRL 963, 4545e' = IMI 39758iiw, 5123 = NRRL A-17919, 5329 = NRRL A-27014, 5142, 5145e = NRRL A-26909, 5151 = NRRL 963, 5154, 5162, 5170, 5180' = IMI 39758 iiw, 5183, 5195 = NRRL 6507, 5273', 5274e = NRRL A-14307, 5278, 5279e = IM1357308, 5283e = NRRL 870 M, 5284e = NRRL A-15105, 5289 = NRRL 870, 5292e = NRRL 961, 5357, 5370, 5834, 10057, 12683, 12816e = NRRL 962, 12817' = NRRL 3586, 1 2 8 W = NRRL A-15402, 12823' = NRRL 959 = NRRL 2028, 12824e = NRRL 3600, 12825e = NRRL A-18563, 14246 = IMI 351305 5112 = NRRL A-19118, 5136, 5188, 5189, 12814'8 = NRRL 871, 12829'8 = NRRL 869 5121 = NRRL A-26932. 5404' = NRRL A-26932 5114, 5192. 5193e = NRRL 5569 = FRR 1636, 5310, 5364, 11636ef, 11664ef, 14245e = IMI 351306 3221, 3397, 3453C,3454', 3465, 3526, 4359"' = NRRL 2040, 4365, 4367, 4543, 4544, 4552, 4671, 5117, 5118, 5119, 5120, 5122, 5128, 5130' = NRRL 1888, 5138, 5140, 5141e = NRRL 2040, 5150" = NRRL 1889, 5152, 5155, 5156, 5158, 5159, 5160, 5161, 5163, 5165, 5167" = Q M 7314, 5169, 5171' = CBS 477.84 = IM1285516 = FRR 2935, 5172, 5173, 5175, 5176, 5177. 5178. 5179'. 5181, 5182, 5184, 5186'. 5187, 5190d = IMI 357294, 5194, 5196, 5197, 5266, 5267. 5269, 5270, 5277, 5280, 5282, 5285, 5287, 5288' = IMI 357295, 5290, 5295, 5296, 5297, 5298, 5311, 5359, 5360, 5360 = FRR 1347, 5363, 5365. 5368, 5369, 5371, 5373, 5374, 5375, 5377, 6302, 6695, 6696', 10009, 10055, 10085, 11272, 11339, 11415'. 14164 = CBS 154.66 3452' 5115, 5157, 5276, 5372, 6688, 11333, 11398 = CBS 118.64, 11403 = CBS 118.64. 12840' = NRRL 970 3442, 3443 = CBS 218.90, 3444 = IMI 321503, 3445, 3446 = IMI 321502, 3928 = IMI 286235, 3937, 4107 = IMI 357301, 4572, 5702 = CBS 219906282 = CBS 21990, 6671, 6672 = NRRL 958, 6681 = ATCC 64627, 6684 = NRRL 13628, 6794, 10005, 10012, 10015, 10017, 10031, 10034, 10036, 10041 = IMI 357300, 10087, 11406 = IMI 357299 4557 = NRRL A-26681, 5407 = NRRL A-26839, 5409 = NRRL A-27004, 5411 = NRRL A-26897, 5414 = A-27001, 5422 = NRRL A-27149, 5423 = A-27166, 5433 = NRRL A-27021 = IMI 321491, 5435 = NRRL A26837, 5578 = NRRL A-26848, 5583 = NRRL A-26679 = IMI 321490, 5585 = NRRL A-26840, 5587 = NRRL A-26838, 5589 = NRRL A-26678, 5590 = NRRL A-27151, 5591 = NRRL A-27147, 5596 = NRRL A-27230, 5600 = NRRL A-26680 = ATCC 64628 = IMI 357302, 5605 = NRRL A26886, 5603 = NRRL A-26677 4558 = NRRL A-26845, 5405 = NRRL A27005, 5406 = NRRL A-27160, 5408 = NRRL A-27006, 5421 = NRRL A-26876, 5424 = NRRL A-27003, 5432 = NRRL A-26925, 5582 = NRRL A-26859, 5586 = NRRL A-26847, 5588 = NRRL A-27180, 5594 = NRRL A-26844, 5595 = NRRL A-26842, 5597 = NRRL A-26860, 5598 = NRRL A-26846 5410 = NRRL A-27178 = CBS 16987 = IMI 296937 = NRRL 13486 Abbreviations IBT, NRRL, FRR, CBS, IMI, VKM, CSIR, PREM, UAMH, Q M see materials and methods. Metabolites could only be detected by HPLC and the actual producer are marked at the IBT number. IBT-numbers with no mark have not been examined for ability to produce these four metabolites. Growth on MEA and DG18 was required to detect xanthomegnin and viornellein. Viridamine-like metabolite. 8 Examined by HPLC; viridic acid could not be detected. Penicilliurn aurantiogriseum chemotaxonomy illumination in an upright position in 9 an triple vented plastic Petri dishes. Standards Auranthine, aurantiamine, brevianamide A, puberulic acid, penicillic acid, penitrem A, oxaline, venucosidin, verrucofortine, cyclopenin, cyclopeptin, cyclopenol, viridicatin, 3methoxyviridicatin, viornellein, xanthomegnin, vioxanthin, orsellinic acid, puberuline, venucofortine, rugulosuvine, leucyltryptophanyl-diketopiperazine, terrestric acid and viridamine were used as standards. Thin layer chromatography (TLC) CYA, YES and MEA, Merck's malt extract agar and DG18 for representative isolates were used for metabolite production. Cultures were analysed b y the TLC-agar plug method (Filtenborg & Frisvad, 1980; Filtenborg, Frisvad & Svendsen, 1983). Small agar plugs were cut from the fungal colony using a sharpened stainless steel metal tube (inner diam. 4 mm). The plugs were wetted by a drop of chloroform/methanol (2: 1, V/V) and immediately applied onto a TLC plate (Silicagel 60, Merck Art 5721), 2.5 cm from the bottom line. As an external standard, 5 pl of a griseofulvin solution was applied at the centre lane of the TLC plate. The eluents TEF (toluene/ethyl acetate/90% formic acid, 5:4: 1, v/v/v) and for representative isolates CAP (chloroform/acetone/2propanol, 17 :3 :4, v/v/v) (Filtenborg et al., 1983) were used. The TLC plates were inspected in daylight and uv-light (254 nm and 365 nm), before and after spraying with ANIS: 0.5 % (v/v) p-anisaldehyde in ethanol (absolute)/glacial acetic acidlconcentrated H,SO, (17/2/1 v/v/v) and heating at 130° for 8 min. High performance liquid chromatography (HPLC) The contents of three plates each of YES, SYES, CYA, MEA and O A T were extracted after 2 wk of growth to ensure that a broad spectrum of secondary metabolites from each isolate was detected. Identification of the metabolites was confirmed by agreement between retention data and uv spectra of standards and the secondary metaboIites were detected by HPLC with diode array detection (Frisvad & Thrane, 1987). RESULTS All the isolates were allocated to one of nine species based on the profile of expressions of differentiation, i.e. families of secondary metabolites (biosynthetic systems) (Tables 2 and 3) and morphology, supplemented with physiological characters (Table 4). All the metabolites except penicillic acid from the taxa related to P. aurantiogriseurn were detected as being produced by either 0 % or more than 8 8 % of the isolates in each species (Table 2). Each species in the series Viridicata is circumscribed by a unique combination of secondary metabolite families (Table 2). Several isolates lack one or two secondary metabolite 486 families of the full profile known from typical isolates of the species (Table 3), but even these isolates cbuld be allocated to the correct species from a comparison of the remaining secondary metabolites and the use of distinctive morphological features. For example, IBT 4778, 11237, 11370 and 11504 only produced the aurantiamine and 3-methoxyviridicatin families, but as they had smooth conidia, they were allocated to P. freii and not to P. neoechinulafum. Deteriorated strains, e.g. NRRL 1889, NRRL 884 and NRRL 2010 (Table 3), could be allocated to one of the nine species, in this case P. aurantiovirens. Assignment was based on unknown secondary metabolites detected by HPLC-DAD, macromorphological features, and the descriptions in Raper & Thom (1949). Certain metabolite families or individual compounds (auranthine, brevianamides, oxaline, penitrems, viridarnine and viridic acid) were only found in particular species and unambiguously determined the identification. For example, viridic acid has only been found in P. viridicatum and not in other species. Other secondary metabolites, such as the penitrems, are produced by many species (Frisvad & Filtenborg, 1989) but only by P. melanoconidium in the series Viridicafa. Other metabolites were shared by more than one species. For instance, terrestric acid is produced by two species, aurantiamine is produced by three species, verrucofortine and/or puberuline is produced by four species, and xanthomegnin, viomellein and vioxanthin or the 3methoxyviridicatin family is produced by five species. In each secondary metabolite family all known chemical members were usually detected by HPLC, e.g. cyclopeptin, dehydrocyclopeptin, cyclopenol, cyclopenin, viridicatol and 3-methoxyviridicatin in the biosynthetic family named after the latter metabolite. The two closely related indol diketopiperazines verrucofortine and puberuline were either both present or at least one of them was present (see Table 2). In addition, other members of these two closely related secondary metabolite families were often detected: rugulosuvine a precursor for puberuline ( = fructigenine A) and leucyltryptophanyldiketopiperazine, a precursor for verrucofortine ( = verrucosine) (Solovyeva et al., 1992). Detection of xanthomegnin was always followed by detection of viomellein and also often by vioxanthin. Occasionally, secondary metabolites with similar chromophores, probably rubrosulphin, viopurpurin and xanthoviridicatin D and G, were detected (Table I). Production of penitrem A was often followed by other penitrems; however, penitrem A was the major product in that biosynthetic family. For P. melanoconidium, oxaline production was often followed by production of meleagrin, but only trace amounts of the precursor (Mantle, 1987) roquefortine C were detected. Aurantiamine and viridamine, belonging to the same secondary metabolite family, did not co-occur, but other compounds with the same chromophore were occasionally found in good producers. The structures for these compounds have not been elucidated. Production of viridic acid was always followed by an unknown, possibly biosynthetically closely related, secondary metabolite with a slightly different uv spectrum. Brevianamide A and B co-occurred, but the amount produced of the former F. Lund and J. C. Frisvad Table 4. Morphological characters on MEA and physiological characters on CYA, MEA, CREA and YES of Penicillium aurantiogriseum and related species v I I1 111 IV Conidium colour, on CYA Reverse colour on CYA greygreenb dark orangebrown cum bluegreen creamish yellow to brown grey Conidium colour on MEA Exudate droplets on CYA Obverse mycelium colour on YES Sporulation on YES Growth on CREA Colony diameter (mm) 7 days, 25' on CYA on MEA on YES Tonidiophore stipes on MEA Conidia shape on MEA greygreenD blue-green grey-green greyishgreyishgreen blue-green blue-green yellowish pinkish orange to to to beige pinkishpinkish brown to light brown blue-green blue-green green few many many few many white to yellow variable white to yellow weak brown weak white to cream strong poor poor poor 22+5< 30k4 33k5 23+4 26f 5 31f5 Conidial ornamentation on MEA rough finely rough subsubglobose to globose ellipsoidal smooth smooth VII VIII 1X greyishgreen orange to pinkishbrown dark green blue-green cuny yellow orange brown blue-green green blue-green many many few many white to yellow weak white to yellow medium vividly yellow weak white to yellow strong vividly yellow weak quite good poor poor poor poor poor 25+2 25+2 34+3 31f4 33f5 43+5 24k4 26k4 33+4 25f5 31f 4 31 +6 23k5 29+4 33f 4 22+3 31+3 33 f 7 28+5 31+2 30f 7 tuberculate subglobose finely fmely rough rough rough subsubsubglobose to globose to globose ellipsoidal ellipsoidal finely smooth smooth rough hely rough subglobose finely rough rough subsubglobose to globose ellipsoidal smooth rough dark brown smooth VI smooth " I, P. aurantiogriseum; 11, P. freii; 111, P. tricolor; W , P. polonicum; V, P. aurantiovirens; VI, P. viridicatum; VII, P. cyclopium; VIII, P. melanoconidium; IX P. neoechinulatum. With a blue tint. Average f one standard deviation was more than ten times larger than the amount of the latter. Venucosidin production was followed by another metabolite with a quite similar uv spectrum and anisaldehyde colour reaction (probably normethylverrucosidin, Hodge ef al., 1988). Production of penicillic acid was not accompanied by any of its precursors, e.g. orsellinic acid (Bentley & Keil, 1961). Other biosynthetic end products in this family are not known. Terrestric acid and several metabolites with similar chromatographic characteristics were found in P. aurantiogriseum and P. tricolor. Terrestric acid, however, is not biochemically related to penicillic acid (Mantle, 1987). The nine species are very dificult to separate (Table 4) when based only on morphology and a few physiological characters. Certain species in series Viridicafa may be recognized quite easily, however, by traditional diagnostic characters, e.g. rough conidia in P. neoechinulafum;grey conidia on all substrates and very tuberalate stipes of P. tricolor; pure green conidia on all substrates in P. viridicafum; and dark green conidia on CYA in P. melanoconidium. The chemotaxonomical characteristics of the nine species in series Viridicafa are described below. A formal taxonomic treatment and species descriptions will be given in other papers (Frisvad ef al., 1994; Frisvad ef al., 1994). Penicillium aurantiogriseum Dierckx, Annls Soc. Scient. Brux. 25 : 88, 1901. P. aurantiogriseurn produces secondary metabolites in the five metabolite families auranthine, aurantiamine, penicillic acid, terrestric acid and verrucosidin (Table 2). In accordance with the name, P. auranfiogriseurn has grey-green conidia on CYA and MEA, but is less obviously characterized by rough conidiophore stipes and subglobose to ellipsoidal conidia (Table 4). The ellipsoidal form of the conidia in some strains of P. auranfiogriseum is also seen in a few strains of other species, notably P. polonicum and P. aurantiovirens. Those strains of P. aurantiogriseum with ellipsoidal greyish-green conidia were earlier identified as either P. commune (IMI 180922a) (Macgeorge & Mantle, 1990) or P. granulatum (FRR 343) (Pitt, 1979) on the basis of ellipsoidal conidia, rough stipes and conidiurn colour. Some isolates only produced a part of the secondary metabolites listed in Table 2, but the combination of metabolites actually found was sufficient to identify the isolate as this species (Table 3). Strain IMI 180922 and its subculture IMI 180922a, differ macroscopically, but both produce nephrotoxic glycopeptides (Barnes ef al., 1977; Yeulet et al., 1988). Strain IMI 180922 has apparently lost the ability to Penicillium aumnfiogriseum chemotaxonomy produce the secondary metabolites terrestric acid and penicillic acid. Similarly, subcultures of FRR 343 and NRRL 3612 have lost their ability to produce terrestric acid. Isolates of P. auranfiogriseum are more prevalent on cereals in subtropical than temperate climates. Penicillium freii Frisvad & Samson, Mycologia, 1994, in press. P. freii produces secondary metabolites in the independent biosynthetic families aurantiamine, xanthomegnin, and 3methoxyviridicatin. A few isolates produced penicillic acid (Table 2). The combination of aurantiamine and either xanthomegnin or 3-methoxyviridicatin along with formation of smooth blue-green conidia and poor sporulation on YES was sufficient to identify an isolate as P. freii. All the diagnostic metabolites are produced on CYA. A few cases however were encountered where some deteriorated isolates did not produce xanthomegnin and viomellein on CYA, but large amounts of these mycotoxins were later detected on DG18 or MEA (Table 3). Most isolates of P. freii have been isolated from cereals in temperate climates (the Nordic countries, Great Britain and Canada). Penicillium tricolor, Frisvad, Seifert, Samson & Mills, Can. J. Bot., 1994, in press. P. tricolor produces the secondary metabolite biosynthetic families xanthomegnin, terrestric acid and verrucofortine. Furthermore, a metabolite with a chromophore characteristic of asteltoxin (Frisvad & Thrane, 1987) (RI 988-989) was consistently produced. Macromorphologically, P. fricolor was characterized by tuberculate stipes, grey conidia on CYA and MEA, dark brown reverse and exudate droplets on CYA and brown obverse mycelium on YES. Five strains of P. tricolor were examined that are representative of a large number of isolates found in Canadian cereals (Mills ef al., 1992). Penicillium polonicum W. Zalessky, Bull. Int. Acad. Pol. Sci. Lett., SCr. B, 1927: 445, 1927. P. polonicum produces secondary metabolites in the independent biosynthetic families verrucosidins, penicillic acid, 3-methoxyviridicatins and verrucofortines (Tables 2 & 3). Other distinguishing features were a high growth rate, sporulation on all media tested, and conidia that are larger than those produced by the other species (Table 4). Two known producers of nephrotoxic glycopeptides (M3 = IBT 14321,6Gd = IBT 14250) obtained from P. Mantle were typical of P. polonicum. Two features are characteristic of species with relative preference for growth on meat namely large broadly ellipsoidal conidia and rather good growth on CREA. P. polonicum share these characters with P. commune and other species associated with meat and cheese. P. polonicum has been found worldwide in cereals, but is also the most common species of the series Viridicafa in meat and indoor air. P. polonicum is abundant in cereals in certain 488 localities of temperate regions and also common in subtropical and tropical cereal samples. Penicillium aurantiovirens Biourge, Cellule 33: 119, 1923. P. aurantiovirens produces secondary metabolites in the biosynthetic families represented by penicillic acid, puberulic acid, 3-methoxyviridicatin and verrucofortine (Table 2). Some older strains (e.g. NRRL 2137) did not produce penicillic acid (Table 3), but have formerly been reported to do so (Wirth & Klosek, 1972). The original strains (NRRL 2137, NRRL 2138 and NRRL 2010) allocated to P. auranfiovirens, as circumscribed by Raper and Thom (1949), produced puberulic and puberulonic acid (Table I). We have found puberulic acid in one further strain and indications of secondary metabolites with chromophores like puberulic acid in HPLC chromatograms of several strains of P. aurantiovirens; however, a better method for puberulic acid production and detection is needed for a definitive identification. As mentioned by Raper & Thom (1949), P. aurantiovirens sporulation is quite poor on MEA. We have found that the isolates produce much more marginal mycelium on MEA and no conidia at all on YES. Growth by the culture ex neotype of P. auranfiocandidum, NRRL 884, was strongly deteriorated, so we propose IMI34846= NRRL 2138 = IBT 3544 as neotype for P. auranfiovirens. The name P. auranfiocandidum may also be a synonym of P. cyclopium or P. polonicum and, as it has not been taken up by taxonomists other than Raper & Thom (1949), we prefer the better known taxon P. auranfiovirens. P. auranfiovirens differs from P. polonicum by its poor sporulation, poor growth on CREA, and production of yellow coloured metabolites including puberulonic acid. It differs from P. freii by its ability to produce vemcofortine. P. aumntiovirens is of worldwide occurrence and has mostly been found in cereals. Penicillum viridicatum Westling, Ark. Bot. 11: 88, 1911. P. viridicafum produces secondary metabolites in the xanthomegnin, penicillic acid, viridic acid and brevianamide A metabolic families and viridamine. Penicillic acid is not commonly produced by this species. The few isolates that did not produce brevianarnide A always produced viridamine and thus such isolates could easily be assigned to this species. Those that did not produce viridamine produced brevianamide A, viridic acid and others (Tables 2 & 3). P, viridicafum characteristically shows pure green conidia on all substrates. On YES agar, the visually detectable sporulation usually covered approximately half of the colony obverse. P. viridicafum is much more common in subtropical climates than temperate regions, and it is most often found on cereals. Penicillium cyclopium Westling, Ark. Bot. 11: 90, 1911. P. cyclopium produces secondary metabolites in the xanthomegnin, penicillic acid, 3-methoxyviridicatin and vermcofortine metabolic families. Other features are green to slightly greyish-green conidia on CYA, blue-green conidia on MEA and strongly yellow mycelium on YES agar. Only 4 out of 103 strains did not produce 3-methoxyviridicatin (Tables 2 F. Lund and J. C. Frisvad & 3). P. cyclopium is most closely related to P. freii and P. aurantiovirens, but differs primarily from the former by production of verrucofortine and greyish-green conidia on CYA and from the latter by production of xanthomegnin and viomellein. Like P. freii, P. cyclopium has most often been found in cereals in temperate climates. Penicillium melanoconidium (Frisvad) Frisvad & Samson, Mycologia, 1994, in press. P. melanoconidium produces secondary metabolites in the penitrem A, oxaline, penicillic acid, verrucosidin and xanthomegnin families (Table 2). Furthermore, it produces a secondary metabolite with a chromophore identical to auranthine, but at a different retention index (RI 791-794). All isolates examined in this taxon have produced the full profile of secondary metabolites (Tables 2 & 3). Macromorphologically it is best characterized by dark green conidia on CYA and heavy sporulation on YES agar (see Table 4). In contrast to other species in Viridicata, P. melanoconidium only produces trace amounts of xanthomegnin and viomellein on CYA, but quite large amounts on DG18. Isolates of P. melanoconidium have onIy been found in cereals in Denmark, Great Britain and Germany. Penicillium neoechinulatum (Frisvad, Filtenborg & Wicklow) Frisvad & Samson, Mycologia, 1994, in press. P. neoechinulatum produces secondary metabolites in the penicillic acid, aurantiamine and 3-methoxyviridicatin families (Table 2). Only one culture did not produce aurantiamine (Table 3). The species is also characterized by echinulate conidia (Frisvad, Filtenborg & Wicklow, 1987) (Table 4). P. neoechinulatum has only been found on seeds in burrows and in cheek pouch swabs of kangaroo rats in Arizona, U.S.A. DISCUSSION The nine taxa treated in this paper can be given taxonomic status according to the criteria of Rogers (1989) i.e. that 'variety status should be given to those taxa which have different biosynthetic pathways' and 'species status is justified if chemistry is correlated with morphological or proven physiological differences, or if more than one major biosynthetic system is involved'. The chemotype concept described by Hawksworth (1974)who states that 'chemotypes are chemically characterized parts of morphologically indistinguishable populations' and 'chemotype concepts have wide applications in taxa which.. . produce chemicals sporadically' is thus not applicable to these different taxa. Based on these statements, we have chosen to give the nine groups related to P. aurantiogriseum species status despite our earlier proposal to reduce them to varieties of P. aurantiogriseum (Frisvad & Filtenborg, 1989). The nine taxa appear to be as different as the other taxa generally accepted at the species level in subgenus Penicillium (for example P. solitum and P. eckinulatum). Our present classification of species in series Viridicata is based on combinations of secondary metabolite data and a series of traditional characters. Earlier classifications were 489 based mainly on a few groups of characters such as micromorphology (Samson et al., 1976), colony texture and odour (Raper & Thom, 1949), conidiurn colour, and colony diameters (Pitt, 1979). Even though many of those classifications were accepted, users of these taxonomic systems often encountered difficulties. We emphasize that we have used the concept of metabolic families (or biosynthetic systems), not individual secondary metabolites, to characterize the nine species. The profile of metabolic families and micromorphological characters are regarded by us as the fundamental basis of phenotypic characterization of filamentous fungi, even though other phenotypic expressions of differentiation (Frisvad, 1992) may also be used. In a few cases, however, individual members (biosynthetic endproducts?) of what appear to belong to the same biosynthetic family are produced consistently by different species, e.g. P. viridicatum produce viridamine, while P. aumnfiogriseum, P. neoechinulatum and P. freii produce the closely related aurantiamine. Production of auantiamine by P. viridicatum has not been observed and viridamine production has likewise not been observed in the two other species. Some isolates did not produce one or more metabolites usually associated with a given species in Viridicata. These were often old deteriorated cultures. For example, most of the cultures of P. freii that did not produce xanthomegnin or viomellein on CYA and YES, produced these metabolites when grown on DG18. Of the remaining four non-producing isolates (Table 3), one produced vioxanthin. Thus, these isolates may be naturally occurring mutants of P. freii or they may have silent genes for the production of the naphthoquinones. A large number of the isolates examined have been cultured on a series of different media to show that the 0% frequency in the other species are valid using other conditions than mentioned in the Materials and Methods. For example, 24 isolates of P. polonicum did not produce xanthomegnin and viomellein after growth on Merck's malt agar, DG18, CYA, MEA, and YES agars. We have shown that the Viridicata series, as represented by Penicillium aurantiogriseum and related species, produces distinct combinations of mycotoxins and other secondary metabolites. Because of this, we consider it very important to identify potential mycotoxigenic isolates in the Viridicata series to species level. Isolates in series Viridicata received as P. aurantiogriseum (or P. commune), and shown to produce ; & nephrotoxic glycopeptides (Mantle et al., 1 9 9 1 ~ Mantle McHugh, 1993), were subsequently reexamined and shown here to be P. aurantiogriseurn and P. polonicum, while isolates received as, and confirmed to be P. viridicatum, did not produce those glycopeptides (P. Mantle, personal communication). It is not clear yet whether the two former species are dominating the actively growing mycoflora of Balkan cereals even though a recent report indcates that this is the case (Mantle & McHugh, 1993). However, it cannot be excluded that other species in the emended Viridicata series proposed here produce similar nephrotoxic glycopeptides. Such information could perhaps help in explaining the etiology of Balkan endemic nephropathy. In future work on prevention of mycotoxin production the species of the Viridicata series will be important in ecological Penicillium aurantiogriseum chemotaxonorny studies. These future studies may show which taxa are present in different geographic regions and ecological situations. A comparison of the nine taxa using molecular biological methods may indicate phylogenetic relationships and give rise to new rapid methods for their detection. In that way, they may be detected early and mycotoxin production may be prevented. We thank John T. Mills, Ulf Thrane and Ole Filtenborg for critically reading the manuscript and Ellen Kirstine Lyhne for excellent technical assistance. We gratefully acknowledge Drs R. A. Samson, S. W. Peterson, D. T. Wicklow, Z. Kozakiewicz, J. T. Mills, K. A. Seifert, J. I. Pitt, P. G. Mantle and K. A. Scudamore for providing us with many cultures of P. aurantiogriseum. We thank Dr T. F. Solovyeva and Dr A. G. Kozlovsky for supplying standards of auranthine, puberuline, rugulosuvine and leu~yltryptophan~l-diket~~iperazine. REFERENCES Abramson, D. (1991). Development of molds, mycotoxins and odors in moist cereals during storage. In Cereal Grain. Mycotoxins, Fungi and Quality in Drying and Storage (ed. J. Chelkowski), pp. 119-147. Elsevier: Amsterdam. Albright, J. L., Aust, S. D., Byers, J. M., Fritz, T. E., Brodie, B. O., Olsen, R. E., Link. R. P., Simon, J., Rhoades, H. E. & Brewer, R. L. (1964). Moldy corn mycotoxicosis in cattle. Journal of the American Veterinary and Medical Association 144, 1013-1019. Alsberg, C. L. & Black, 0. F. (1913). Contributions to the study of maize deterioration. Biochemical and toxicological investigations of Penicillium puberulum and Penicillium stoloniferurn. United States Department of Agriculture Bureaux of Plant Industry Bulletin 270, 7-48. Arai. K.. Kimura, K., Mushiroda, T. & Yamamoto, Y. (1989). Structures of hctigenines A and B, new alkaloids isolated from Pencillium ffudigenum Takeuchi. Chemical and Pharmaceutical Bulletin 37, 293 7-2939. Barnes. J. M., Carter, R. L., Peristianis, G. C., Austwick, P.K. C., Flynn, F. V. & Aldridge, W. N. (1977). Balkan (endemic) nephropathy and a toxinproducing strain of Penicillium verrucosum var. cyclopium: an experimental model in rats. Tke Lancet i, 6 7 1 4 7 5 . Bentley, R. & Keil, J. G. (1961). The role of acetate and malonate in the biosynthesis of penicillic acid. Proceedings of the Chemical Society 1961, 111-112. Bentley, R. & Keil, J. G. (1962). Tetronic acid biosynthesis in molds. 11. Formation of penicillic acid in Penicillium cyclopium. Journal of Biological Chemistry 23 7, 867-873. Birkinshaw, J. H., Oxford, A. E. & Raistrick, H. (1936). Studies in the biochemistry of microorganisms. XLVIII. Penicillic acid, a metabolic product of Penicillium puberulum Bainier and P. cyclopium Westling. Biochemical Journal 30, 394-411. Birkinshaw, J. H. & Raistrick, H. (1932). Studies in the biochemistry of microorganisms. XXIII. Puberulic acid, C,H,O,, and an acid C,H,O,, new products of the metabolism of glucose by Penicillium puberulum Bainier and Penicillium aurantio-virens Biourge. With an appendix on certain dihydroxyben~enedicarbox~lic acids. Biochemical ]ournu1 26, 441-453. Bridge. P. D., Hawksworth, D. L., Kozakiewicz, Z., Onions, A. H. S.. Paterson, R. R. M.. Sackin, M. J. & Sneath, P. H. A. (1989). A reappraisal of the terverticillate penicillia using biochemical, physiological and morphological features. I. Numerical taxonomy. Journal of General Microbiology 135, 2941-2966. Bridge, P. D., Kozakiewicz, Z. & Paterson, R. R. M. (1992). PENIMAT: a computer assisted identification scheme for tewerticillate Penicillium isolates. Mycological Papers 165, 1-59. Budiarso. I. T., Carlton, W. W. & Tuite, 1. (1971). The influence of some cultural conditions o n toxigenicity of Penicillium viridicatum. Toxicology and Applied Pharmacology 20, 194-205. Burka, L. T., Ganguli, M. & Wilson, B. J. (1983). Vemcosidin, a tremorgen from Penicillium verrucosum var. cyclopium. lournal of the Chemical Society Chemical Communications 1983, 544-545. 490 Carlton. W. W. & Tuite, J. (1970). Nephropathy and edema syndrome induced in miniature swine by corn cultures of Penicillium viridicatum. Pathologia Veterinaria 7,68-80. Carlton. W.W., Tuite. J. & Caldwell, R. W. (1972). Mycotoxicosis introduced in mice by Penicillium ochraceum. Toxicology and Applied Pharmacology 21, 13CF142. Carlton, W. W.. Tuite, J. & Mislivec, P. B. (1968). Investigations of the toxic effect in mice of certain species of Penicillium. Toxicology and Applied Pharmacology 13,372-387. Ciegler, A., Fennell. D. I., Sansing, G. A,, Detroy, R. W. & Bennett, G. A. (1973).Mycotoxin-producing strains of Penicillium viridicatum: classification into subgroups. Applied Microbiology 26, 271-278. Ciegler, A. & Kurtzman, C. P. (1970): Penicillic acid production by blue-eye fungi on various agricultural commodities. Applied Microbiology 20, 761-764. Ciegler, A,, Lee, L. S. and Dunn, J. J. (1981). Production of napthoquinone mycotoxins and taxonomy of Penicillium viridicatum. Applied and Environmental Microbiology 42, 44&449. Ciegler, A., Mintzlaff, H.-J., Machnik, W. & Leistner, L. ( 1 9 7 2 ~ ) . Untersuchungen iiber das Toxinbild~ngsvermo~envon Rohwiirsten isolierter Schimmelpilze der gattung PeniciNium. Die Fleichwirtschaft 52, 1311-1318. Ciegler, A., Mintzlaff, H.-J., Weisleder, D. & Leistner, L. (1972b). Potential production and detoxification of penicillic acid in mold-fermented sausage (salami). Applied Microbiology 24, 114-119. Cutler, H. G., Crumley, F. G., Cox, R. H., Wells, J. M. and Cole, R. J. (1984). The biological properties of cyclopenin and cyclopenol. Plant and Cell Physiology 25, 257-263. DeJesus, A. E.. Steyn, P. S., Van Heerden, F. R., Vleggaar, R., Wessels, P. L. & Hull, W. E. (1983). Tremorgenic mycotoxins from Penicillum crwtosum: isolation of penitrem A-F and the structure elucidation and absolute configuration of penitrem A. Journal of the Chemical Society, Perkin Transactions 1, 1847-1856. El-Banna, A. A. & Leistner, L. (1987). Quantitative determination of vermcosidin produced by PeniciNium auranfiogriseum. Microbiologie Aliments Nutrition 5, 191-195. El-Banna, A. A. & Leistner, L. (1989). Production of vermcosidin by Penicillium species. Microbiologie Aliments Nutrition 7 , 161-164. Filtenborg, 0. & Frisvad, J. C. (1980). A simple screening method for toxigenic moulds in pure cultures. Lebensrnittel Wissenschaft und Technologic 13, 128-130. Filtenborg, 0.. Frisvad, J. C. & Svendsen, I. A. (1983). Simple screening method for molds producing intracellular mycotoxins in pure culture. Applied and Environmental Microbiology 45,5 81-585. Filtenborg, 0..Frisvad, 1. C. & Thrane, U. (1990). The significance of yeast extract composition on metabolite production in Penicilliurn. In Modern Concepts in Penicillium and Aspergillus Classijicafion (ed. R. A. Samson & J. I. Pitt), pp. 433-440. Plenum Press: New York. Frisvad, J. C. (1981). Physiological criteria and mycotoxin production as aids in identification of common asymmetric penicillia. Applied and Environmental Microbiology 41,56&5 79. Frisvad, J. C. (1985~).Profiles of primary and secondary metabolites of value in classification of Penicillium viridicatum and related species. In Advances in Penicillium and Aspergillus Systematics (ed. R. A. Samson & J. I. Pitt), pp. 311-325. Plenum Press: New York. Frisvad, J. C. (1985 b). Classification of asymmetric penicillia using expressions of differentiation. In Advances in Pe~cilliumand Aspergillus Systemtics (ed. R. A. Samson & J. I. Pitt), pp. 327-333. Plenum Press: New York. Frisvad, J. C. (1989). The connection between the peniclllia and aspergilli and mycotoxins with special emphasis on misidentified isolates. Archives of Environmental Contamination and Toxicology 18, 452-467. Frisvad, J. C. (1992). Chemometrics and c h e m o t a ~ o n o m ~ a :comparison of multivariate statistical methods for the evaluation of binary fungal secondary metabolite data. Chemometrics and Intelligent Laboratory Systems 14,253-269. Frisvad, J. C. & Filtenborg, 0.(1983). Classification of terverticillate penicillia based on profiles of mycotoxins and other secondary metabolites. Applied and Environmental Microbiology 46, 1301-1310. Frisvad, J. C. & Filtenborg, 0. (1989). Terverticillate Penicillia: Chemotaxonomy and mycotoxin production. Mycologia 81. 837-861. F. Lund and J. C. Frisvad Mantle, P. G., McHugh, K. M., Adatia. R., Gary, T. & Turner, D. R. ( 1 9 9 1 ~ ) . Frisvad, J. C., Filtenborg, 0 . & Wicklow, D. T. (1987).Terverticillate penicillia isolated from underground seed caches and cheek pouches of banner-tailed Persistent karyornegaly caused by Penicillium nephrotoxins in the rat. Proceedings of the Royal Society, London, Series B 246, 251-259. kangaroo rats (Dipodomys spectabilis). Canadian Journal of Botany 65, Mantle, P. G., McHugh, K. M., Adatia, R., Heaton, J. M., Gray, T. & Turner, 765-773. D. R. (1991b). Penicillium aurantiogriseum-induced persistent renal histoFrisvad, J. C. & Samson, R. A. (1991). Mycotoxins produced by species of pathological changes in rats; an experimental model for Ballcan Endemic Penicillium and Aspergillus occumng in cereals. In Cereal Grain. Mycotoxins, Nephropathy competitive with ochratoxin A. In Mycotoxins, Endemic Fungi and quality in dying and storage (ed. J. Chelkowski), pp. 441-476. Elsevier : Amsterdam. Nephropathy and Urinary Tract Tumors (ed. M. Castagnero, R. Plestina, G. Dirheirner, I. N. Chernozemsky & H. Bartsch), pp. 119-127. IARC Frisvad. J. C., Samson, R. A,, Svendsen, A,, Lund, F. & van der Horst, M. I. Scientific Publications No. 115. International Agency for Research on (1994). New toxigenic species related to P. aurantiogriseum. Mycologia, in press. Cancer: Lyon. Frisvad. J. C., Seifert. K. A., Samson, R. A. & Mills, I. T. (1994). PenciNium Mills. J. T., Seifert, K. A,, Frisvad, J. C., Abramson, D. & Louis-Seize, G. (1992). tricolor, a new species from canadian wheat producing several mycotoxins. Relative importance of Penicillium species producing nephrotoxins in Canadian Journal of Botany, in press. Canadian grain products. In International Symposium on Stored-Grain Frisvad, J. C. & Thrane, U. (1987). Standardized high-performance liquid Ecosystems (ed. D. S. Jayas, N. D. G. White, W. E. Muir & R. N. Sinha), chromatography of 182 mycotoxins and other fungal metabolites based on pp. 29-31, University of Manitoba: Winnipeg. alkylphenone retention indices and UV-VIS spectra (diode array detection). Mulinge, S. K. & Chesters, C. G. C. (1970). Ecology of fungi associated with Journal of Chromatography 404, 195-214. moist stored barley grain. Annals of Applied Biology 65, 277-284. Hald, B., Christensen, D. H. & Krogh, P. (1983). Natural occurrence of the Nagel, D. W., Pachler, K. G., Steyn, P. S., Vleggaar, R. & Wessels, P. L. (1976). mycotoxin viomellein in barley and the associated quinone-producing The chemistry and 13CNMR assignments of oxaline, a novel alkaloid from penicillia. Applied and Environmental Microbiology 46, 1311-1317. Penicillium oxalicum. Tetrahedron 22, 2625-2631. Hawksworth, D. L. (1974). Mycologist's handbook. Commonwealth MyNortholt, M.D., van Egmond, H. P. & Paulsch, W. E. (1979). Penicillic acid cological Institute, Kew. production by some fungal species in relation to water activity and Hocking, A. D. & Pitt, J. I. (1980). Dichloran-glycerolmedium for enumeration temperature. Journal of Food Protection 42, 476-484. of xerophilic fungi from low-moisture foods. Applied and Environmental Oxford, A. E., Raistrick, H. & Smith, G. (1942). Antibacterial substances from Microbiology 37, 488-492. moulds. VI. Pubemlic acid, C,H,O,, and puberulonic acid, C,H,O,, Hodge, R. P., Hams, C. M. & Hams, T. M. (1988). Venucofortine, a major metabolic products of a number of species of Penicillium. Chemistry and metabolite of Penicillium verrucosum var. cyclopium, the fungus that Indushy 61, 485-487. produces the mycotoxin venucosidin. Journal of Natural Products 51, Pitt, J.I. (1979). The genw Penicillium and its teleomorphic states Eupenicillium 6&73. and Talaromyces. Academic Press: London. Holzapfel, C. W., Koekemoer, J. M. & van Dyk, M. S. (1986). Isolation, Pitt, I. I. & Cmidcshank, R. H. 1990. Speciation and synonymy in Penicillium structure and synthesis of viridic acid, a new tetrapeptide mycotoxin of subgenus Penicillium - towards a definitive taxonomy. In Modem Concepts Penicillium viridicatum. South African Journal of Chemistry 39, 75-80. in Penicillium and Aspergillus Clnssification (ed. R. A. Samson & J. I. Pitt), Holzapfel, C. W. & Marsh, J. J. (1977). Isolation and structure of viridamine, pp. 103-119. Plenum Press: New York. a new nitrogeneous metabolite of Penicillium viridicatum Westling. South Raper, K. B. & Thorn, C. (1949). A Manual of the penicillia. Williams and African Journal of Chemistry 30. 197-204. Wilkins; Baltimore. Kawai, K., Nozawa, K., Nakajima, S. & litaka, Y. (1984). Shidies on hngal Robbers, J. E., Hong, S., Tuite, J. & Carlton, W. W. (1978). Production of products. VII. The structures of meleagrin and 9-0-p-bromobenzoylxanthomegnin and viomellein by species of Aspergillus correlated with meleagrin. Chemical and Pharmaceutical Bulletin 32, 94-98. mycotoxicosis produced in mice. Applied and Environmental Microbiology Kobayashi, H., Tsunoda, H. & Tatsuno, T. (1971). Recherches toxicologique 36, 819-823. sur les mycotoxines qui polluent de fourrage artificiel du porc. Chemicaland Rogers, R. W. (1989). Chemical variation and the species concept in lichenized Pharmaceutical Bulletin 19,829-842. ascomycetes. Botanical Journal of the Linnean Sociefy 101, 2 2 F 2 4 0 . Kozlovsky, A. G. (1990). Nitrogen-containing mycotoxins of the fungi Samson, R. A. & van Reenen-Hoekstra, E. S. (ed.). (1988). Introduction to food32, Penicillium. Proceedings of the Japanese Association of Mycoto~icolog~ borne fungi. Centraalbureau voor Schimmelcultures, Baam. 31-34. Samson, R. A., Stolk, A. C. & Hadlok, R. (1976). Revision of the subsection Kurtzman, C. P. & Ciegler, A. (1970). Mycotoxin from a blue-eye mold of Fasciculata of Penicillium and some allied species. Studies in Mycology corn. Applied Microbiology 20. 204-207. (Baam) 11, 1-47. Larsen, T. O., Frisvad, J. C. and Jensen, S. R. (1992). Aurantiamine, a Scudamore, K. A., Atkin, P. M. & Buckle, A. E. (1986). Natural occurrence of diketopiperazine from two varieties of Penicillium aurantiogriseum. Phytothe naphthoquinone mycotoxins, xanthomegnin, viomellein and vioxanthin chemistry 31, 1613-1615. in cereals and animal feedstuffs. Journal of Stored Products Research 22, Le Bars, J. (1979). Contamination alimentaire par le Penicillium cyclopium 81-84. Westling. Frhquence de souches productrices d'acide penicillique. Anwles Solovyeva, T. F., Baskunov, B. P., Kiselyeva, 0 . V. & Kozlovsky, A. G. de Recherches Vitirinaires 10, 119-124. (1992). Biosynthesis of alkaloids by the culture Penicillium puberulum. Lindenfelser, L. A. & Ciegler, A. (1977). Penicillic acid production in Microbiology 61, 395-403. (In Russian.) submerged culture. Applied and Environmental Microbiology 34, 553-556. Solovyeva, T. F., Baskunov, B. P., Nefedova, M. Y. & Kozlovsky, A. G. Lund. F., Thrane, U.. Frisvad, J. C. & Filtenborg, 0 . (1992). IBT List of fungal (1989). Biosynthesis of leucylhyptophanyldiketopiperazine by a Penicillium cultures and their secondary metabolites. Department of Biotechnology, The aurantio-virens culture and the characteristics of its production. Microbiology Technical University of Denmark, Lyngby, Denmark. 5 8 , 308-313. Macgeorge, K. M. & Mantle, P. G. (1990). Nephrotoxicity of Pen~cillium Stack, M. E., Eppley, R. M., Dreifuss. P. A. & Pohland, A. E. (1977). Isolation aurantiogriseum and P. commune from an endemic nephropathy area of and identification of xanthomegnin, viomellein, rubrosulphin and vioYugoslavia. Mycopathogia 112, 139-145. purpurin as metabolites of Penicillium uiridicahrm. Applied and Environmental Macgeorge, K. M. & Mantle, P. G. (1991).Nephrotoxic fungi in a Yugoslavian Microbiology 33, 351-355. community in which Ballcan nephropathy is hyperendemic. Mycological Stack. M. E. & Mislivec, P. B. (1978). Production of xanthomegnin and Research 95, 66-64, viomellein by isolates of Aspergillus ochracew, Penicillium cyclopium and Mantle, P. G. (1987). Secondary metabolites of Penicillium and Acremonium. In P. viridicatum. Applied and Environmental Microbiology 36, 552-554. Penicillium and Acremonium (ed. 1. F. Peberdy) Biotechnology Handbooks Stok, A. C., Samson, R. A., Frisvad. 1. C. & Filtenborg, 0. (1990). The 1,pp. 161-243. Plenum Press: New York. systematics of the terverticillate penicillia. In Modem Concepts in Penicillium Mantle, P. G. & McHugh. K. M. 1993. Nephrotoxic fungi in foods from and Aspergillus Classification (ed. R. A. Samson & J. I. Pitt), pp. 121-136. nephropathy households in Bulgaria. Mycological Research 97,205-212. Plenum Press: New York. Penicillium aurantiogriseum chemotaxonomy Ueno, Y. (1991). Biochemical mode of action of mycotoxins. In Mycotoxins and animal foods (ed. J.E. Smith & R. Henderson), pp. 438-453. CRC Press: Boca Raton, Florida. Thorpe, C. W. & Johnson, R. L. (1974).Analysis of penicillic acid by gas-liquid chromatography. Journal of the Association of Official Analytical Chemists 57, 861-865. Wilson, B. J., Byerly, C. S. & Burka, L. T. (1981). Neurologic disease of fungal origin in three herds of cattle. Journal of the Veterinary and Medical Association 179, 480-481. Wilson, B. J.,Yang, D. T. C. & Harris, T. M. (1973). Production, isolation and preliminary toxicity studies of brevianamide A from cultures of Penicillium uiridicatum. Applied Microbiology 26, 6 3 3 4 3 5 . Wirth, 1. C., Gilmore, T. E. & Noval, J. J. (1956). Penicillic acid the antibiotic responsible for the activity of a culture filtrate of a strain of PeniciNium martensii. Archives of Biochemistry and Biophysics 63, 452-453. (Accepted 27 September 1993) 492 Wirth, J. & Klosek, R. (1972). Fungi. Moniliales. Relationships in Penicillium aurantiouirens. Phytochemistry 11, 2615. Yeulet. S. E., Mantle, P. G., Bilton, J. N., Rzepa, H. S. & Sheppard, R. N. (1986). Auranthine a new benzodiazepinone metabolite of Penicillium. Journal of the Chemical Society Perkin Transactions. I , 1891-1894. Yeulet, S. E.,Mantle, P. G., Rudge, M. S. & Greig, J. B. (1988). Nephrotoxicity of Penicillium aurantiogriseum, a possible factor in the aetiology of Balkan endemic nephropathy. Mycopathologia 102, 21-30. Zimmermann, J. L., Carlton, W. W. & Tuite, J. (1979). Mycotoxicosis produced in swine by cultural products of an isolate of Aspergillus ochraceus. Veterinaria Pathologia 16, 583-592. Zwicker. G.M., Carlton, W. W. & Tuite, J. (1973). Prolonged administration of Peniciiiium uiridicafum to mice: preliminary report of carcinogenicity. Food and Cosmetics Toxicology 11, 989-994.