Pesticide Biochemistry and Physiology 64, 167–184 (1999) Article ID pest.1999.2424, available online at http://www.idealibrary.com on Mechanisms of Resistance to Fenpropimorph and Terbinafine, Two Sterol Biosynthesis Inhibitors, in Nectria haematococca, a Phytopathogenic Fungus Alexandrine Lasseron-De Falandre, Danièle Debieu,1 Jocelyne Bach, Christian Malosse, and Pierre Leroux INRA, Unité de Phytopharmacie et Médiateurs Chimiques, 78026 Versailles Cédex, France Received November 6, 1998; accepted May 21, 1999 The mechanisms of resistance to terbinafine, a squalene epoxidase inhibitor, and to fenpropimorph, a sterol D14-reductase and/or D8 → D7-isomerase inhibitor, were investigated in laboratory mutants of the phytopathogenic fungus Nectria haematococca. Neither modified fungicide uptake nor fungicide metabolism could explain resistance in the mutants studied. The terbinafine-resistant mutants contained at least 10 times more squalene than the wild-type strain, and when cultivated in presence of terbinafine, they required a 30times higher fungicide concentration to obtain a very high level of accumulated squalene similar to that of the wild-type strain. Thus, a reduced affinity of the squalene epoxidase toward both the substrate and the fungicide could explain terbinafine resistance in these mutants. In some fenpropimorph-resistant mutants, the effects of fenpropimorph on sterol biosynthesis were similar to those in the wild-type strain, suggesting that resistance could be due to tolerance to sterol D8,14-sterol accumulation and to D5,7-sterol decrease. In one of them, the major sterol in absence of fungicide was not ergosterol, but ergosta-5,7,22,24(241)-tetraenol, 1 1 indicating reduced D24(24 )-reductase activity. The potential role of a modified D24(24 ) reductase in fenpropimorph resistance has yet to be elucidated. In the remaining fenpropimorph-resistant mutants, much higher fenpropimorph concentrations were required to cause ergosterol decreases similar to that in the wild-type strain, suggesting a D14-reductase modification. Among them, some mutants accumulated D8-sterols as well as D8,14-sterols, indicating that the mechanism of resistance may be due to reduced affinity of the D14reductase toward fenpropimorph, whereas the other mutants accumulated only D8,14-sterols. Those last mutants contained much larger amounts of 4,4-dimethyl and 4a-methyl sterols than the wild-type strain, in the absence of fungicide. The most abundant methylated sterol was 4,4-dimethylfecosterol, the presumed product of the D14-reductase, suggesting that resistance may be due to overproduction of the D14-reductase in these mutants. q1999 Academic Press INTRODUCTION Sterol biosynthesis inhibitors (SBIs)2 are widely used in agriculture against phytopathogenic filamentous fungi (1). The largest group 1 To whom correspondence should be addressed at INRA, Unité de Phytopharmacie et Médiateurs Chimiques, INRA, 78 026 Versailles Cédex, France. Fax: 33 1 30 83 31 19. Email: debieu@versailles.inra.fr. 2 Abbreviations used: SBI, sterol biosynthesis inhibitor; DMI, 14a-demethylation inhibitor; sterol trivial and systematic names: all sterols referred to in the text have the 5aconfiguration; eburicol, 4,4,14a-trimethylergosta-8,24(241)dien-3b -ol; 4,4-dimethylfecosterol, 4,4-dimethylergosta8,24(241)-dien-3b -ol; 4a-methylfecosterol, 4a-methylergosta-8,24(241)-dien-3b -ol; ergosterol, ergosta-5,7,22Etrien-3b -ol; episterol, ergosta-7,24(241)-dien-3b -ol; fecosterol, ergosta-8,24(241)-dien-3b -ol; ignosterol, ergosta8,14-dien-3b -ol; lanosterol, 4,4,14a-trimethylcholesta- inhibits the sterol 14a-demethylase (DMIs)2. These include pyridines, pyrimidines, piperazines, and azoles (1, 2). The intensive use of DMIs has led to the development of resistance in the field (3, 4). Another group of SBIs, including morpholines (e.g., fenpropimorph, tridemorph) and piperidines (e.g., fenpropidin), which are used mainly against cereal diseases, inhibits the sterol D14-reductase and/or the D8 → D7-isomerase (1, 5). To date, the only reported cases of reduced sensitivity to morpholines and piperidines concern Erysiphe graminis (6, 7). Allylamines, such as terbinafine, are squalene epoxidase inhibitors (8) that were originally 8,24(25)-dien-3b -ol; 3b-ol. lichesterol, ergosta-5,8,22E-trien- 167 0048-3575/99 $30.00 Copyright q 1999 by Academic Press All rights of reproduction in any form reserved. 168 LASSERON-DE FALANDRE ET AL. developed for use against pathogenic fungi in medicine (8, 9) but their in vitro activity against phytopathogenic fungi indicates that they are also potential agricultural fungicides (10–12). We were interested in the two latter SBI groups with regard to their risk of resistance. This was evaluated by the UV induction of laboratory mutants resistant to fenpropimorph or terbinafine in Nectria haematococca Berk. and Br., a phytopathogenic model fungus which has been the subject of genetic, pathogenicity, and biochemical investigations (13). The inheritance of resistance, fitness characteristics, and pathogenicity of the mutants have been previously reported (13). It was shown that fenpropimorph resistance can result from mutations in several genes. Four genes could be clearly identified, respectively designated Fen1, Fen2, Fen3, and Fen4, whereas genetic identity could not be assigned to two mutants (13, 14). Except for the strain carrying a mutation in the Fen4 gene, which exhibited a low resistance level, all the other strains exhibited a fenpropimorph resistance level greater than 10. Terbinafine resistance resulted from mutations in only one gene, designated Ter1. Mutations in the Ter1 gene led to a very high terbinafine resistance level, greater than 100. In some strains carrying mutations at the Fen1 locus and most of the strains carrying mutations at the Ter1 locus, growth rate, sporulation, and pathogenicity were not affected, suggesting that field resistance development could occur (13). In the present paper, we report on fenpropimorph and terbinafine uptake and metabolism and sterol profiles in these UVinduced laboratory N. haematococca mutants. Possible mechanisms of fenpropimorph and terbinafine resistance are discussed. MATERIALS AND METHODS Fungal Strains Two strains of the heterothallous ascomycete N. haematococca MP VI, S1 and S2, were used as wild-type strains (Fusarium solani f. sp. pisi, anamorph). They were kindly provided by Professor H. D. Van Etten and are identical to the strains 6-36 and 6-94 (15). Fenpropimorph-resistant strains (F) and terbinafine-resistant strains (T) were obtained from S1 or S2, following UV irradiation of the conidia, except for two spontaneously resistant strains selected on fenpropimorph (13). Resistant strains were designated by two numbers: the first refers to the original strain and the second to the mutant number (e.g., F1-1 or F2-1 and T11 or T2-1). The spontaneous mutants were designated by sp (e.g., F1-sp or F2-sp). The strains designated R1F1-sp and R2F1-sp are progenies from crosses between F1-sp and S2 and R1F1-38 and R2F1-38 strains are progenies from crosses between F1-38 and S2. The sensitivities of the mutants toward fenpropimorph, fenpropidin, tridemorph, and terbinafine, when grown on potato–dextrose agar, and the genetic analysis of resistance have already been published (13) and are summarized in Table 1. For two strains, F2-2 and F2-sp, genetic analysis of fenpropimorph-resistance was not possible due to lack of fertility (13). The strains carrying mutations in the Fen1, Fen2, Fen3, Fen4, or Ter1 genes were designated as Fen1, Fen2, Fen3, Fen4, or Ter1 mutants. Fungicides Fenpropimorph and tridemorph were provided by BASF AG. (Germany), fenpropidin by Dr. Maag AG. (Switzerland), and terbinafine by Sandoz AG. (Switzerland). [14C]Fenpropimorph was provided by Dr. Maag AG. (Switzerland). Fungicides were added to cultures in ethanolic solutions. Uptake of Fungicides To obtain mycelial suspensions, liquid nutrient medium (containing per liter: KH2PO4, 3 g; MgSO4, 0.3 g; NaNO3, 3 g; glucose, 8 g; and yeast extract, 2 g; pH 5.5) inoculated with 105 conidia ml21 was incubated at 258C under constant shaking at 150 rpm for 18, 19, or 27 h, depending on the growth rate of the strain. The mycelium was harvested, in late exponential growth phase, by filtration on a 125-mm-mesh FENPROPIMORPH- AND TERBINAFINE-RESISTANT MUTANTS OF N. haematococca 169 TABLE 1 Fungicide Sensitivity of Wild-Type Strains, Terbinafine- and Fenpropimorph-Resistant Strains of Nectria haematococca Tridemorph Fenpropidin Allylamine sensitivityc Terbinafine 5.1 3.6 1.7 1.3 2.9 2.1 3.6 6.7 6.4 4.5 3.3 0.9 5.1 5.5 3.2d 1.1d 2.5 1.0 0.9 1.0 0.9 0.8 0.9 10.9 16.9 18.1 18.0 7.0 2.7 26.9 31.5 15.4d 40.6d .59 1.3 0.7 1.0 1.1 1.5 0.8 0.9 1.3 1.9 1.8 — — 1.5 1.9 — — 2.6 .100 .100 .100 .140 .140 .140 Morpholine and piperidine sensitivityc Strainsa Wild-type S1 S2 Resistant F1-1 F2-1 F2-8 F1-sp R1F1-sp R2F1-sp F2-2 F2-sp R1F1-38 R2F1-38 F1-38 T1-1 T1-2 T1-10 T2-1 T2-3 T2-9 Mutationsb None None Fen1 Fen1 Fen1 Fen1Fen4 Fen1 Fen4 ND ND Fen2 Fen3 Fen2Fen3 Ter1 Ter1 Ter1 Ter1 Ter1 Ter1 Fenpropimorph 0.2 0.1 19.5 36.0 40.0 37.0 10.0 3.0 49.0 50.0 16.5d 46.5d 126.0 1.7 1.1 1.3 0.9 1.4 0.9 a Fenpropimorph-resistant mutant (F) and terbinafine-resistant mutants (T) were derived from the two wild-type strains S1 and S2 and they were designated by two numbers: the first one refers to the original strain and the second to the mutant number. All the resistant mutants were UV-induced, except that two spontaneous mutants designated sp. R1F1-sp and R2F1-sp are progeny from the F1-sp 3 S2 cross and R1F1-38 and R2F1-38 are progeny from the F1-38 3 S2 cross. b Genetic identity of fungicide resistance; ND, not determined due to lack of fertility. c For S1 and S2, expressed as EC50 value (mg ml21) for radial growth rate on potato–dextrose agar plates; however, for the resistant strains, expressed as RL value which is the resistance level obtained from the ratio of the EC50 value for the resistant strain relative to that for the wild-type strain; —, not done. d Resistance level of R1F1-38 and R2F1-38 strains, with average of S1 and S2 EC50 values as reference. gauze disc, washed, and resuspended (2 g wet weight/100 ml buffer) in potassium phosphate buffer (25 mM, pH 5.5 or pH 7.5) containing calcium chloride (0.1 mM) and glucose (1%). In the case of fenpropimorph uptake, mycelial suspensions were preincubated for 30 min and experiments were started by adding [14C]fenpropimorph (specific radioactivity 592 Bq mg21) in ethanolic solution to a final concentration of 16 mM (corresponding to 5 mg?ml21 containing less than 1% ethanol). Samples of mycelial suspension (5 ml) were taken after varying time intervals, filtered on preweighed Whatman GF/ A glass filters, washed twice with 5 ml buffer in order to eliminate fungicide not tightly associated with mycelium, lyophilized, and weighed in order to determine mycelial dry weight. Radioactivity in rehydrated mycelium (with 0.2 ml H2O) was extracted with lumagel scintillant liquid for 1 day at 28C and then counted in a liquid scintillation counter. A 200-ml aliquot of the filtrate (incubation medium plus the 23 5ml washings) was similarly counted in order to control that the radioactivity in the filtrate corresponded to the remaining radioactivity not tightly associated with the mycelium, showing that there was no loss of radioactivity in the experiment. 170 LASSERON-DE FALANDRE ET AL. In the case of terbinafine uptake studies, because [14C]terbinafine was not available, fungicide uptake was quantified by gas chromatography (GC) analysis. Samples of mycelial suspension (100 ml) were filtered; mycelium and filtrate were frozen and lyophilized. The mycelium and the filtrate were resuspended in a mixture of methanol (10 ml) and HCl 1% (2 ml) and incubated at 858C for 1 h 30 s and at 48C for 24 h, respectively, in order to extract terbinafine. The methanolic fraction was washed three times with hexane to remove lipids, evapored, resuspended in 2% NaCl aqueous solution, and then extracted three times with dichloromethane. The combined dichloromethane fractions containing terbinafine were then analyzed using a GC fitted with a thermoionic detector and an OV 1701 capillary column (15 m 3 0.32 mm), with N2 (1 bar) as carrier gas. The oven, injector, and detector temperatures were 200, 220, and 2808C, respectively. Uptake studies were carried out with at least duplicates, except for F1-sp, F2-2, and F2-8 mutants, for which no replicate was carried out. Under our conditions, standard deviation represents less than 10% of the mean value. Squalene and Sterol Extraction and Analysis The mycelium, in late exponential growth phase, was produced as described in the previous section. Fungicides were added to the culture medium in ethanolic solution (final concentration of ethanol 5.0 ml liter21) at the same time as the conidia; untreated controls received ethanol only. The mycelium was harvested, washed, lyophilized, and saponified in methanolic KOH (6%) at 708C for 2 h. Mycelial unsaponifiable lipids were extracted three times into hexane and then purified on silica gel TLC plates. CH2Cl2 was used as the developing solvent (two runs) as described elsewhere (16). The purified 4,4-dimethyl-, 4a-methyl-, and 4-desmethylsterol fractions were acetylated at room temperature for 15 h using a mixture of pyridine and acetic anhydride (ratio of 1:2) and purified on TLC plates (silica gel; CH2Cl2; one run). The steryl acetates were then analyzed by GC and GC/MS. The GC was fitted with a flame ionization detector and an OV-1 capillary column (30 m 3 0.32 mm), N2 0.5 bar. The oven temperature was 3008C. Cholesterol was used as a standard for relative retention time determination and sterol quantification. GC/MS analyses were performed with a Ribermag R10-10-C spectrometer. The GC was fitted with CPSIL5CB column and the oven temperature was programmed to increase from 250 to 3208C at a rate of 58C min21. The ionization potential was 70 eV as previously described (16). Mass spectra of sterols already published (16) are not reported; the only ones listed in Table 2 are those not previously reported. RESULTS Uptake of Fungicides Fenpropimorph was quickly accumulated in the wild-type strain mycelium during the first 5 min (Fig. 1). The maximum fungicide level obtained at pH 7.5 was about 15 times higher than that at pH 5.5. Fenpropimorph uptake was then studied for all the strains at pH 7.5. The two wild-type strains S1 and S2 exhibited similar uptake kinetics, the maximum fungicide levels, obtained after 30 min, being 7.8 and 9.0 nmol mg21 dry weight, respectively (Table 3). Then, the levels slightly decreased to 6.4 and 7.2 nmol mg21 dry weight for S1 and S2, respectively, after 4 h of incubation. The fenpropimorph-resistant mutants exhibited fenpropimorph uptake patterns similar to those of the wild-type strain from which they are issued (Table 3). However, F2-8, one of the three tested strains carrying a mutation in the Fen1 gene, and F2-2 appeared to show slightly lower uptake than the corresponding wild-type strain, S2. The maximum fungicide levels were 6.5 and 6.3 nmol mg21 dry weight for F2-8 and F2-2, respectively, compared with 9.0 nmol mg21 dry weight for the parental wild-type strain (S2). Metabolism studies conducted on TLC plates for S1, F21 (Fen1 mutant), and F1-38 (Fen2Fen3 mutant) mycelial extracts after 4 h of incubation with fenpropimorph did not reveal any metabolism of fenpropimorph either in the wild-type strain 171 FENPROPIMORPH- AND TERBINAFINE-RESISTANT MUTANTS OF N. haematococca TABLE 2 Ionic Species in the Mass Spectra of Steryl Acetates in Nectria haematococca Fenpropimorph-Resistant Strains Cultivated in either the Absence or the Presence of Fenpropimorph 4-Desmethylsterols RRtb Fragmentationc [M]+ [M-Me]+ [M-43]+ [M-Ac]+ [M-Ac-Me]+ [M-84]+ [M-84-Me]+ [M-43-AC]+ [M-SC-H]+ [M-SC-2H]+ [M-84-Ac-Me]+ [M-SC-Ac]+ [M-SC-Ac-H]+ [M-SC-Ac-2H]+ [M-SC-Ac-Me-H]+ [M-SC-26-Ac]+ [M-SC-42-Ac]+ A-H B 4a-Methyl sterol 4,4-Dimethyl sterol 1a 2a 3a 4a 5a 6a 1.32 1.35 1.36 1.42 1.56 1.68 434(9)d 434(2) 436(2) 436(21) 454(2) 439(10) 468(–) 453(7) 374(74) 359(11) 374(50) 359(100) 376(22) 361(100) 376(94) 361(100) 379(5) 370(23) 355(6) 356(6) 393(7) 384(19) 369(7) 370(6) 311(3) 311(6) 253(22) 253(45) 327(100) 295(3) 269(16) 341(100) 309(6) 283(15) 251(18) 251(32) 267(16) 281(19) 211(28) 157(47) 143(13) 211(32) 157(29) 143(18) 241(15) 227(37) 255(13) 241(17) 310(8) 309(8) 251(100) 250(16) 249(61) 235(35) 209(65) 157(31) 143(17) 235(22) 209(19) a Acetate of : 1, ergostapentaenol; 2, ergosta-5,8,14,22,24(241)-pentaenol; 3, ergosta-5,8,22,24(241)-tetraenol; 4, ergosta5,7,22,24(241)-tetraenol; 5, 4a-methylepisterol; 6, 4,4-dimethylepisterol. b Retention time of steryl acetate relative to cholesterol. c Ac, acetate (60); Me, methyl (15); 26, C2H2 (loss of C-16 and C-17); 42, C3H6 (loss of C-15 to C-17); 43, C3H7 (loss of C-25 to C-27); 56, C4H8 (loss of C-15 to C-17 and C-32 in a 14a-methyl sterol); 84, C6H12 (loss of C-23 to C-28 in a 24-methylene sterol); SC, sterol side chain; A and B, fragment of m/z 157 and 143, respectively. d Figures in parentheses are intensities of ions relative to the base peak (100) with m/z above 209. For 3 and 4 fragments of m/z 143 and 157 were also taken into account. TABLE 3 Accumulation of [14C]Fenpropimorph by Nectria haematococca Wild-Type and Fenpropimorph-Resistant Strains Fenpropimorph-resistant strains Wild-type strain Times 5 15 30 60 240 a b c d b S1 S2 c 6.7 7.8 7.8 7.6 6.4 7.5 8.2 9.0 8.2 7.2 Fen2Fen3 Fen1Fen4 NDa F2-8 F1-38 F1-sp F2-2 6.3 6.5 6.4 6.5 6.9 6.7 6.7 6.5 5.8 6.1 6.4 6.9 7.1 6.9 7.3 5.4 6.3 5.6 5.4 5.6 Fen1 F1-1 7.5 8.3 8.5 8.1 5.8 F2-1 d — 8.5 8.7 8.7 7.5 ND, not determined due to lack of fertility. Incubation times expressed as min, in presence of [14C] fungicide (5 mg ml21; 16 mM). Quantity of [14C] fungicide expressed as nmol mg21 of dry weight. —, not done. 172 LASSERON-DE FALANDRE ET AL. FIG. 1. [14C]Fenpropimorph uptake by N. haematococca wild-type strains; S1, pH 5.5 (L); S1, pH 7.5 ( ); S2 pH 7.5 (l). or in the fenpropimorph-resistant mutants (data not shown). Terbinafine uptake studies revealed that, after 1 h of incubation, the S2 wild-type strain and the T2-1 mutant showed similar amounts of terbinafine associated with the mycelium. However, the value for the wild-type strain was slightly lower (6.9 nmol mg21 dry weight) than that of the terbinafine-resistant strain (10.0 nmol mg21 dry weight). Squalene and Sterol Contents of Untreated Fenpropimorph-Resistant Mutants Squalene content and total sterol content varied from 0.02 to 0.08 and 2.7 to 5.8 mg mg21 dry weight, respectively, depending on the strain (Table 4). The fenpropimorph-resistant strains differed by factors lower than 1.7 for squalene amount and lower than 1.5 for sterol amount, relative to their parental wild-type strain. Among the fenpropimorph-resistant mutants, three groups could be distinguished according to their sterol profile in the absence of fungicide (Table 4). The first group comprised the Fen1 mutants (F1-1, F2-1, and F2-8) and the Fen3 mutant (R2F1-38; progeny issued from cross between F138 and S2) and was characterized by a sterol profile similar to that of wild-type strains. 4Desmethylsterols dominated the sterol profile (92 to 95% of total sterols), ergosterol being the predominant sterol (85 to 90% of total sterols), and 4,4-dimethyl and 4a-methyl sterols were minor components (3 to 7 and 1 to 2% of total sterols, respectively). Small amounts of other 4desmethylsterols (lichesterol, episterol, fecosterol, ergosta-5,7-dien-3b -ol, and an ergostatetraenol (ergosta-5,7,9(11),22-tetraen-3b -ol, as proposed in Ref. 16)) were also found. The 4,4dimethylsterols lanosterol, eburicol and 4,4dimethylfecosterol were found, while the only 4a-methylsterol present in nonnegligible amount was 4a-methylfecosterol. The second group of mutant strains included the Fen2 mutant (R1F1-38) and the Fen2Fen3 mutant (F1-38). The percentage of 4,4-dimethyl-, 4a-methyl-, and 4-desmethylsterols TABLE 4 Squalene and Sterol Contents in Nectria haematococca Untreated Wild-Type and Fenpropimorph-Resistant Strains S1 b Squalene Sterolsb Sterol classesc 4,4-Dimethylsterols 4a-Methylsterols 4-Desmethylsterols Sterol profilec Lanosterol Eburicol 4,4-Dimethylfecosterol 4,4-Dimethylepisterol 4a-Methylfecosterol 4a-Methylepisterol Ergostatetraen-3b -ol Ergostapentaen-3b -ol Lichesterol Ergosterol Ergosta-5,7,22,24(241)-tetraen-3b -ol Fecosterol Ergosta-5,7-dien-3b -ol Ergosta-5,7,24(241)-trien-3b -ol Episterol Other sterols a b c d e f 0.03 3.2 F1-1 0.05 3.2 F2-1 0.05 4.5 F2-8 0.03 2.7 R1F1-sp d nd 3.5 NDa Fen2 Fen3 Fen2Fen3 Fen4 Fen1Fen4 R1F1-38 R2F1-38 F1-38 R2F1-sp F1-sp F2-2 F2-sp 0.02 4.2 0.02 4.5 0.03 2.8 0.05 3.2 0.08 4.0 0.05 3.3 d nd 5.8 2.0 0.9 97.1 6.8 1.3 90.9 3.3 1.4 95.3 3.4 1.3 95.4 4.6 1.0 94.3 2.0 0.9 97.2 2.5 0.8 96.6 5.2 1.5 93.3 2.8 0.9 96.3 22.2 13.6 64.2 23.0 14.9 62.1 30.0 15.4 54.6 21.3 18.3 60.4 0.3 1.0 0.7 —f 0.9 — 1.5 — 1.9 91.2 — 0.5 0.3 — 1.8 — 1.3 3.5 2.0 — 2.3 — 0.7 — 0.9 84.5 — 0.7 1.1 — 3.1 0.1 0.5 1.4 1.3 — 1.3 — 1.3 — 2.4 87.0 — 0.6 1.6 — 2.2 0.5 0.7 1.6 1.1 — 1.2 — 0.9 — 1.3 88.3 — 0.8 0.9 — 3.1 0.1 0.4 3.0 1.2 — 1.0 — 1.2 — 1.2 86.4 — 0.9 1.1 — 3.4 0.2 0.3 0.9 0.7 Tr g 0.9 Tr 0.4 — 1.3 90.8 — 0.6 0.5 0.3 2.3 0.1 0.4 1.1 0.9 — 0.9 — 0.2 1.9 — 7.1 79.2 — — 1.7 4.2 2.3 0.6 2.5 2.0 — 1.5 — 0.9 — 1.2 86.9 — 0.1 1.2 — 2.6 0.5 0.2 1.3 1.2 — 0.9 — 0.4 2.5 — 11.0 74.8 — — 1.4 3.5 2.9 0.1 0.5 21.5 Tr 9.0 4.2 0.5 — 2.4 60.4 — 0.1 0.3 — 0.4 0.6 0.1 0.5 22.1 0.2 8.3 6.5 0.9 — 3.3 56.5 — 0.1 0.4 — 0.4 0.9 0.1 0.6 28.8 0.4 6.6 8.7 1.2 — 2.7 50.1 — 0.1 0.2 — 0.2 0.4 NIe 0.2 20.7 0.4 7.9 10.5 1.0 — 2.6 56.5 — — 0.1 — 0.1 — ND, not determined due to lack of fertility. Expressed as mg mg21 dry weight. Expressed as percentage of total sterols. nd, not done. NI, not integrated. —, not detected. Tr, below 0.1% of total sterols. 173 g 0.05 4.4 S2 Fen1 FENPROPIMORPH- AND TERBINAFINE-RESISTANT MUTANTS OF N. haematococca Fenpropimorph-resistant strains Wild-type strain 174 LASSERON-DE FALANDRE ET AL. and the type of 4,4-dimethyl- and 4a-methylsterols were similar to those of wild-type strains, whereas the individual 4-desmethylsterols present differed. The major sterol was not ergosterol but ergosta-5,7,22,24(241)-tetraen-3b -ol as revealed by the mass spectrum of the steryl acetate (Table 2), ([M]+ at m/z 436; base peaks at m/z 376 [M-Ac]+ and 361 [M-Ac-Me]+ and intense peaks, characteristic of D5,7-sterols (Ref. 17), at m/z 157 and 143); NMR studies (data not shown) and UV spectrum (lmax 231, 271, 282, and 293 nm) according to Barton et al. (18) supported this identification. Ergosterol represented only 7 and 12% of total sterols, for Fen2 and Fen2Fen3 mutant, respectively, whereas ergosta5,7,22,24(241)-tetraen-3b -ol accounted for 81 and 73% of total sterols. Among the minor 4desmethylsterols, as for wild-type strains, episterol and an ergostatetraenol (exhibiting a mass spectrum similar to that of ergosta-5,7,9(11),22tetraen-3b -ol) were found, but not lichesterol, fecosterol, or ergosta-5,7-dien-3b -ol. On the other hand, ergosta-5,7,24(241)-trien-3b -ol was found as well as a sterol with a mass spectrum revealing a fragmentation similar to that of ergostatetraenol but with additional unsaturation in the side chain. This particular sterol was called ergostapentaen-3b -ol (Table 2), and could be a 1 D24(24 )-ergostapentaenol. The third group was composed of the Fen4 mutant (R1F1-sp), the Fen1Fen4 mutant (F1-sp), and mutants F2-2 and F2-sp; it was characterized by percentages of 4,4-dimethyl-, 4a-methyl- and 4-desmethylsterols that differed strongly from all the other strains. Depending on the strain, 4,4-dimethyl- and 4a-methylsterols varied from 21 to 30% and 14 to 18% of total sterols, respectively, whereas 4-desmethylsterols represented only 55 to 60% and ergosterol 50 to 57% of total sterols. 4,4-Dimethylfecosterol (21 to 29% of total sterols) was the only 4,4-dimethylsterol accumulated in nonnegligible amount, whereas 4,4-dimethylepisterol was detected in tiny amounts. Among the 4a-methylsterols, 4amethylepisterol was found in amounts comparable to 4a-methylfecosterol, the only 4a-methylsterol detected in other strains. Squalene and Sterol Contents of Treated Fenpropimorph-Resistant Mutants Squalene content varied from 0.01 to 0.05 mg mg21 dry weight, depending on the fenpropimorph concentration and the strain (Tables 5–7). Moreover, those amounts were not very different from those in the absence of fungicide (Table 4). In the presence of fenpropimorph, the percentage of the normal major sterol, ergosterol for most of the strains and ergosta-5,7,22,24(241)tetraenol for the Fen2 and Fen2Fen3 mutants, decreased with simultaneous accumulation of either D8,14-sterols or of both D8,14- and D8-sterols, depending on the strain (Tables 5–7). In comparison with the S1 wild-type strain, the Fen1 mutant (F1-1) exhibited similar ergosterol decrease and D8,14-sterol accumulation at the same fenpropimorph concentrations, whereas its growth inhibition was lower (Table 5). As the total sterol amount increased with increasing fenpropimorph concentrations to a greater extent in the wild-type strain than in the Fen1 mutant (i.e., 4.4 to 9.3 and 3.2 to 4.6 mg mg21 dry weight, respectively), the amount of ergosterol was higher in the wild-type strain than in the Fen1 mutant (1 and 0.6 mg mg21 dry weight, respectively) when cultivated in the presence of 0.6 mg ml21 fenpropimorph. The D8,14-sterols comprised ergosta-5,8,14,22-tetraenol, ergosta-8,14,24(241)-trienol, and ignosterol, which was the major D8,14-sterol for both strains. In both strains treated with 0.6 mg ml21 fenpropimorph, a slight D8-sterol accumulation occurred, which was lower in the Fen1 mutant. The accumulation of D8-sterols did not increase at higher fenpropimorph concentrations in either strain. The D8-sterols found were lichesterol, ergosta-8-enol, and fecosterol. In the case of Fen2, Fen3, and Fen2Fen3 mutants (Table 6), ergosta-5,7,22,24(241)-tetraenol or ergosterol decreased with fenpropimorph treatment and the main sterols which accumulated were either D8,14-sterols alone (Fen2 mutant) or both D8,14- and D8-sterols (Fen3 and Fen2Fen3 mutants). At fenpropimorph concentrations leading to about 50% growth inhibition, FENPROPIMORPH- AND TERBINAFINE-RESISTANT MUTANTS OF N. haematococca 175 TABLE 5 Squalene and Sterol Content of S1 Wild-Type and F1-1 Fenpropimorph-Resistant Strains of Nectria haematococca in the Presence of Fenpropimorph Wild-type S1a b Squalene Sterolsb Sterol classesc 4,4-Dimethylsterols 4a-Methylsterols 4-Desmethylsterols Sterol profilec Ergosterol D8,14-4-Desmethylsterolsd Ergosta-5,8,14,22-tetraen-3b -ol Ergosta-8,14,24(241)-trien-3b -ol Ignosterol D8-4-Desmethylsterolsd Lichesterol Ergosta-8-en-3b-olf Fecosterol Other sterols Dry weight inhibitioni Fen1 F1-1a 0 0.6 2.4 0 0.6 2.4 20 0.05 4.4 0.02 6.5 0.05 9.3 0.05 3.2 0.01 4.7 0.04 4.6 0.05 5.5 2.0 0.9 97.1 1.0 0.4 98.6 0.5 0.2 99.3 3.3 1.4 95.3 0.9 0.5 98.6 1.2 0.4 98.4 0.9 0.4 98.7 91.2 —e — — — 2.4 1.9 — 0.5 6.4 0 15.7 59.5 3.9 12.1 43.5 11.1 f 2.6 5.9 2.6 13.7 45 5.2 87.6 5.3 35.1 47.2 4.0 f 1.6 2.4 NSh 3.2 88 87.0 — — — — 3.0 2.4 — 0.6 10.0 0 13.6 73.1 13.9 21.2 38.1 5.4 f 2.3 3.1 NSh 7.8 27 5.4 88.0 14.0 26.1 47.9 2.0 f NSg 2.0 NSh 4.6 38 2.5 87.4 13.1 22.9 51.4 5.4f 0.9 2.6 1.9 4.7 .50 a Conidia were inoculated in presence of 0.6, 2.4, or 20 mg ml 21 fenpropimorph and grown at 258C and 150 rpm for 18 h. b Expressed as mg mg21 dry weight. c Expressed as % of total sterols. d Sum of the different D8,14- or D8-4-desmethylsterols, expressed as % of total sterols. e —; not detectable. f Slightly overestimated because ergosta-8-en-3b -ol in mixture with an unidentified sterol. g NS; not separately integrated because of the very low level of lichesterol relative to ergosta-5,8,14,22-tetraen-3b -ol with very close RRt. h NS; not separately integrated because of the very low level of fecosterol relative to ignosterol, integrated with ignosterol. i Expressed as % of untreated. D8,14-sterols were more abundant than D8-sterols. However, for the Fen3 mutant, a lower fenpropimorph concentration, causing less than 50% inhibition, was also tested and in that case D8-sterols were more abundant than D8,14-sterols. For the Fen3 mutant, the D8,14-sterols that accumulated were the same as for the wild-type strain. Fecosterol was the main D8-sterol which accumulated, followed by ergosta-8-enol and lichesterol. For Fen2Fen3 and Fen2 mutants, the D8- and/or D8,14-sterols that accumulated 1 were all D24 -sterols: ignosterol was not detected, whereas ergosta-8,14,24(241)-trienol constituted the major D8,14-sterol and ergosta5,8,14,22,24(241)-pentaenol was found as a minor sterol instead of ergosta-5,8,14,22-tetraenol. Lichesterol and ergosta-8-enol were not detected, whereas fecosterol was found. In the case of the Fen2Fen3 mutant an ergosta5,8,22,24(241)-trienol was also observed but to a small extent. In the Fen4 and Fen1Fen4 mutants and in strain F2-2, mainly D8,14-sterols, rather than D8sterols, accumulated at all fenpropimorph concentrations tested. The 4-desmethyl D8,14-sterols found (Table 7) were the same as in the wildtype strain (Table 5). D8,14-Sterols other than 4desmethyl D8,14-sterols were found in appreciable amounts, whereas they were only found at trace levels in the other strains (data not shown). 176 LASSERON-DE FALANDRE ET AL. TABLE 6 Squalene and Sterol Content of R1F1-38, R2F1-38, and F1-38 Fenpropimorph-Resistant Strains of Nectria haematococca in the Presence of Fenpropimorph Fen2 R1F1-38a 0 b Squalene Sterolsb Sterol classesd 4,4-Dimethylsterols 4a-Methylsterols 4-Desmethylsterols Sterol profiled Ergosterol Ergosta-5,7,22,24(241)-tetraen-3b -ol D8,14-4-Desmethylsterols f Ergosta-5,8,14,22-tetraen-3b -ol Ergosta-5,8,14,22,24(241)-pentaen-3b -ol Ergosta-8,14,24(241)-trien-3b -ol Ignosterol D8-4-Desmethylsterols f Lichesterol Ergosta-5,8,22,24(241)-tetraen-3b -ol Ergosta-8-en-3b -olh Fecosterol Other sterols Dry weight inhibitioni 0.05 3.2 Fen3 R2F1-38a 1 0.02 4.9 0 0.08 4.0 3 0.03 3.2 Fen2Fen3 F1-38a 6 c nd 3.4 0 10 0.05 3.3 0.05 3.9 2.5 0.8 96.6 1.3 0.6 98.1 5.2 1.5 93.3 7.3 2.5 90.2 5.5 1.8 92.7 2.8 0.9 96.3 3.1 1.5 95.4 7.1 79.2 — — — — — — — — — — 13.7 0 2.0 3.3 79.6 — 0.9 78.7 — 6.3 g — — — 6.3g 8.6 45 86.9 —e — — — — — 1.3 1.2 — — 0.1 11.8 0 23.6 — 24.5 1.4 — 9.2 13.9 36.6h 6.4 — 9.0 21.2 15.3 28 15.9 — 43.4 2.0 — 19.3 19.0 32.5h 4.8 — 6.8 20.9 8.2 40 11.0 74.8 — — — — — — — — — — 14.2 0 2.2 6.0 49.5 — 2.4 47.1 — 23.1g — 2.5 — 20.6g 19.2 50 a Conidia were inoculated in presence of 1, 3, 6, or 10 mg ml21 fenpropimorph and grown at 258C and 150 rpm for 18 h for R1F1-38 and R2F1-38, and 27 h for F1-38. b Expressed as mg mg21 dry weight. c nd; not done. d Expressed as % of total sterols. e —; not detectable. f Sum of the different D8,14- or D8-4-desmethylsterols, expressed as % of total sterols. g Overestimated because fecosterol in mixture with an unidentified sterol. h Slightly overestimated because ergosta-8-en-3b -ol in mixture with an unidentified sterol. i Expressed as % of untreated. Thus, 4,4-dimethylergosta-8,14,24(241)-trienol and 4a-methyl-ergosta-8,14,24(241)-trienol amounted to 18 and 2% of total sterols, respectively (Table 7). However, the amounts of 4,4dimethyl- and 4a-methylsterols decreased, whereas those of 4-desmethylsterols increased with increasing fenpropimorph concentrations, thus resembling other strains in their relative proportions of 4,4-dimethyl-, 4a-methyl-, and 4-desmethylsterols. Squalene and Sterol Contents of TerbinafineResistant Mutants All the terbinafine-resistant strains, Ter1 mutants, had sterol amounts and profiles similar to those of wild-type strains, with ergosterol as the major sterol (Table 8). However, squalene amounts were 11 to 31 times higher in the terbinafine-resistant mutants relative to the wildtype strain from which they are issued. In the TABLE 7 Squalene and Sterol Contents of R1F1-sp, R2F1sp, F1-sp, and F2-2 Fenpropimorph-Resistant Strains of Nectria haematococca in the Presence of Fenpropimorph 0 c Squalene Sterolsc Sterol classese 4,4-Dimethylsterols 4a-Methylsterols 4-Desmethylsterols Sterol profilee 4,4-Dimethylergosta-8,14,24(241)-trien-3b -ol 4,4-Dimethylfecosterol 4a-Methylergosta-8,14,24(241)-trien-3b -ol 4a-Methylfecosterol 4a-Methylepisterol Ergosterol D8,14-4-Desmethylsterols j Ergosta-5,8,14,22-tetraen-3b -ol Ergosta-8,14,24(241)-trien-3b -ol Ignosterol D8-4-Desmethylsterolsj Lichesterol Ergosta-8-en-3b -olk Fecosterol Other sterols Dry weight inhibition l a b c d e f g i j k l NDa F2-2b Fen1Fen4 F1-spb 2.4 0 0.6 2.4 0 0.6 2.4 20 0 0.6 2.4 20 d nd 3.5 nd 5.2 nd 5.8 nd 4.4 nd 6.6 0.02 4.2 nd 3.6 nd 5.4 nd 3.1 0.03 4.5 nd 3.0 nd 1.9 0.04 5.0 2.0 0.9 97.2 0.4 0.1 99.5 22.2 13.6 64.2 12.7 5.8 81.5 4.5 1.9 93.6 23.0 14.9 62.1 16.8 16.9 66.3 21.3 6.2 72.5 2.4 1.2 96.4 30.0 15.4 54.6 32.4 8.9 58.7 31.3 7.7 61.0 13.7 2.7 83.6 —f 0.7 — 0.9 Tr 90.8 — — — — 1.3 1.3 — — 6.3 0 0.2g Tr i Tr Tr — 4.5 91.0 5.9 30.0 55.1 2.1k 1.0 1.1 — 2.2 36 — 21.5 0.5 9.0 4.2 60.4 — — — — 2.5 2.4 — 0.1 1.9 0 5.4h 6.9 1.1 4.5 0.2 27.9 42.8 5.5 11.1 26.2 6.2k 3.6 1.7 0.9 5.0 5 2.6h 1.4 0.7 1.1 0.1 10.8 70.9 5.3 23.4 42.2 5.7k 2.6 1.9 1.2 6.7 49 — 22.1 0.1 8.3 6.5 56.5 — — — — 3.4 3.3 — 0.1 3.1 0 — 12.0 1.9 13.0 1.9 45.3 9.6 1.9 1.8 5.9 6.3k 5.4 0.9 — 10.0 0 12.2 5.9 1.4 4.7 0.1 8.8 57.2 10.3 11.5 35.4 2.2k — 2.2 — 7.5 13 1.1 0.3 0.2 1.0 0.1 2.7 91.8 14.2 28.3 49.3 1.4k — 1.4 — 1.4 64 — 28.8 — 6.6 8.7 50.1 — — — — 2.8 2.7 — 0.1 3.0 0 10.1 21.5 0.4 7.3 1.3 48.8 5.5 1.6 0.4 3.5 2.9k 2.8 0.1 — 2.2 0 17.8 12.9 1.7 5.6 0.4 23.5 32.9 6.1 3.8 23.0 3.0k 2.5 0.5 — 2.2 0 11.5 1.3 1.7 0.9 — 3.8 76.1 8.7 15.3 52.1 2.9k 1.5 1.2 0.3 1.8 42 ND, not determined due to lack of fertility. Conidia were inoculated in presence of 0, 0.6, 2.4, or 20 mg ml21 fenpropimorph and grown at 258C and 150 rpm for 19 h. Expressed as mg mg21 dry weight. nd, not done. Expressed as % of total sterols. —; not detectable. Overestimated because in mixture with eburicol which was major. Slightly overestimated because in mixture with eburicol which was minor. Tr, below 0.1% of total sterols. Sum of the different D8,14- or D8-4-desmethylsterols, expressed as % of total sterols. Slightly overestimated because ergosta-8-en-3b -ol in mixture with an unidentified sterol. Expressed as % of untreated. 177 h Fen4 R2F1-spb FENPROPIMORPH- AND TERBINAFINE-RESISTANT MUTANTS OF N. haematococca Fen1 R1F1-spb 178 LASSERON-DE FALANDRE ET AL. TABLE 8 Squalene and Sterol Contents in Untreated Wild-Type and Terbinafine-Resistant Strains of Nectria haematococca Wild-type strains a Squalene Sterolsa 4,4-Dimethylsterolsb 4a-Methylsterolsb 4-Desmethylsterolsb Ergosterolb a b Terbinafine-resistant strains S1 S2 T1-1 T1-2 T1-10 T2-1 T2-3 T2-9 0.05 4.4 4.9 0.7 94.4 86.9 0.03 3.2 6.8 1.3 90.9 85.8 0.55 4.6 3.0 1.1 95.9 89.8 0.73 3.8 2.9 0.6 96.5 92.8 0.82 4.3 5.3 0.9 93.8 88.7 0.90 5.7 2.8 0.8 96.4 91.3 0.59 3.8 2.7 0.7 96.6 91.9 0.93 4.9 4.8 0.9 94.3 89.2 Expressed as mg mg21 dry weight. Expressed as % of total sterols. presence of terbinafine, ergosterol biosynthesis was slightly inhibited (7 to 27% inhibition, depending on the strain), whereas a very marked squalene accumulation occurred at terbinafine concentrations of 1 and 30 mg ml21 for the wildtype strain and resistant strains, respectively (Table 9). These large accumulations of squalene (up to 11 to 15.5 mg mg21 dry weight) were associated with growth inhibition of about 50% in both the wild-type and the resistant strains. This corresponded to a squalene accumulation factor of 310 for the wild-type strain and between 12 and 20 for the Ter1 mutants. DISCUSSION Fenpropimorph uptake kinetics in N. haematococca, with no transient accumulation level, appeared similar to that of fenpropimorph and tridemorph in P. italicum (19). This type of kinetics is different from the biphasic DMI uptake as reported for N. haematococca (20) and other fungi (19, 21–23), where uptake seems to be a balance between two phenomena, passive influx and energy-dependent efflux (19–23). The reasons for this difference in accumulation of SBIs are unknown. De Waard and van Nistelrooy (19) proposed that the efflux, possibly mediated by the electrochemical proton gradient, could be dependent upon differences in SBI chemical characteristics such as protonation or electron density. Fenpropimorph and tridemorph, which are amines, exist in neutral and protonated forms, the ratio of each form being dependent on the pH of the medium. The pKa of fenpropimorph has been determined to be 7.5 (24). At pH 5.5, fenpropimorph is entirely in its protonated form and thus much less lipophilic, whereas at pH 7.5, approximately 50% of the fungicide exists as the free base. This could explain why fenpropimorph toxicity was about 10 times lower in liquid medium at pH 5.5 than at pH 7.5 (25). The fenpropimorph-resistant strains exhibited uptake kinetics similar to that of the wild-type strains. However, two mutants (F2-8, Fen1 mutant, and F2-2) appeared to show a slightly reduced uptake and then only by a factor lower than 1.5. This low factor is not sufficient to explain the resistance of these strains; furthermore, F1-1 and F2-1, which are like F2-8 Fen1 mutants, did not show reduced uptake. Fenpropimorph metabolism studies conducted with Fen1 and Fen2Fen3 mutants did not reveal any difference between the resistant mutants and the wildtype strain (data not shown). So, metabolism does not appear to be involved in fenpropimorph resistance in these mutants. Resistance to terbinafine in Ter1 mutants could not be explained by reduced fungicide uptake. Although modified uptake does not explain resistance to either fenpropimorph or terbinafine in these mutants, it has been implicated in resistance to tebuconazole in several other mutants (Teb1, Teb2, and Teb3) previously studied (20). For these tebuconazole-resistant strains, reduced fungicide uptake was probably due to a constitutive 179 FENPROPIMORPH- AND TERBINAFINE-RESISTANT MUTANTS OF N. haematococca TABLE 9 Squalene and Sterol Contents in Wild-Type and Terbinafine-Resistant Strains of Nectria haematococca in Presence of Terbinafine Wild-type straina Terbinafine-resistant strainsa S1 b Squalene Sterolsb Ergosterolb Sterol classes c 4,4-Dimethylsterols 4a-Methylsterols 4-Desmethylsterols Dry weight inhibitiond a b c d T1-1 T2-1 0 1 0 1 30 0 30 0.05 4.4 4.0 15.5 3.0 2.9 0.55 4.6 4.1 1.32 4.9 4.0 11.2 4.0 3.8 0.90 5.7 5.2 11.1 4.7 4.3 4.9 0.7 94.4 0 1.1 0.5 98.4 55 3.0 1.1 95.9 0 8.2 2.3 89.5 17 2.9 0.8 96.3 60 2.8 0.8 96.4 0 4.2 1.0 94.8 50 Conidia were inoculated in presence of 0, 1, or 30 mg ml21 terbinafine and grown at 258C and 150 rpm for 19 h. Expressed as mg mg21 dry weight. Expressed as % of total sterols. Expressed as % of untreated. energy-dependent efflux (20) as described for several laboratory DMI-resistant strains of different fungi (19, 21–23). Moreover, reduced uptake was found in a variety of DMI-resistant clinical isolates of human pathogenic fungi (1, 3, 9) and could also be involved in the field isolates of phytopathogenic fungi resistant to DMIs, as recently found in Septoria tritici (26). Sterol analyses revealed modifications in sterol biosynthesis in several fenpropimorph-resistant mutants in comparison with the wild-type strains, either in the absence or in the presence of the fungicide. This suggests that modified fungicide targets could be a possible mechanism of resistance in some mutants. In wild-type strains, the major sterols were ergosterol when grown in absence of fenpropimorph, as described for most fungi (27, 28), and D8,14sterols when was grown in presence of fenpropimorph, as previously reported for N. haematococca (16, 29). This accumulation of D8,14sterols suggests that the sterol D14-reductase constitutes the major target for fenpropimorph in N. haematococca. Each resistant strain carrying a mutation in a different gene will be discussed below, together with the mutants with unkown genetic identity. Fen1 mutant exhibited a sterol composition similar to that of the wild-type strain both in the absence and in the presence of fungicide. Although ergosterol biosynthesis was inhibited to a similar extent in both the mutant and the wild-type strain at the same fenpropimorph concentration, mycelial growth was less inhibited. Thus, a modified D14-reductase does not explain fenpropimorph resistance in this case. D8,14-Sterol accumulation or ergosterol decrease or both (1) had been postulated to be responsible for fungitoxicity. Greater tolerance to D8,14-sterol accumulation and ergosterol decrease, relative to the wild-type strain, may be implicated in the resistance of the Fen1 mutant to fenpropimorph. It had been shown (13) that Fen1 mutants were also resistant to fenpropidin. This is consistent with the fact that the main fenpropidin target in N. haematococca is also the D14-reductase (29). Resistance was also expressed toward tridemorph but to a lesser extent. As it has been shown that tridemorph may be a better inhibitor of D8 → D7-isomerase than of D14-reductase in N. haematococca (29), this putative tolerance mechanism may also apply to D8-sterol accumulation in Fen1 mutants but to a lesser extent than for D8,14-sterol accumulation. Sterol analyses of a Fen2 mutant in absence of fenpropimorph revealed a modified sterol profile, characterized by a low ergosterol content, whereas the probable immediate precursor 180 LASSERON-DE FALANDRE ET AL. of ergosterol, ergosta-5,7,22,24(241)-tetraenol (28), constituted the major sterol. The large amount of this sterol and the presence of other 1 D24(24 )-sterols not found in the wild-type strains, 1 such as ergosta-5,7,24(241)-trienol and D24(24 )ergostapentaenol, indicate a decreased activity 1 of sterol D24(24 )-reductase resulting from a mutation in the Fen2 gene. This mutation should induce pleiotropic effects, such as a reduced growth rate and the absence of sporodochia with macroconidia, as published earlier (13). However, whether or not there is a causal relationship between these effects and the presence of ergosta-5,7,22,24(241)-tetraenol in place of ergosterol remains questionable. Neurospora 1 crassa strains exhibiting decreasing D24(24 )reductase activity and accumulating ergosta5,7,22,24(241)-tetraenol were characterized by a reduced rate of growth and were female sterile (30). However, the lower levels of ergosta-5, 7,22,24(241)-tetraenol found in untreated Saccharomyces cerevisiae (10 to 20% of total sterols) and in S. cerevisiae treated with a low 1 concentration of 23-azacholesterol, a D24(24 )reductase inhibitor (below 36% of total sterols), were not associated with any growth effect (31, 32). Growth of the Fen2 mutant in presence of fenpropimorph led mainly to D8,14-sterol accumulation, as in the wild-type strain. At similar fenpropimorph concentrations, ergosterol biosynthesis was inhibited to a similar extent, whereas growth inhibition was lower than in wild-type strain. As observed with the Fen1 mutant, tolerance to D8,14-sterol accumulation and major decrease in D5,7-sterols seems to be implicated in the resistance mechanism induced by the Fen2 mutation. Moreover, as with the Fen1 mutant, resistance was also expressed toward fenpropidin and to a lesser extent to tridemorph. The presence of ergosta-8,14,24(241)trienol instead of ignosterol as the major D8,14sterol in the fenpropimorph-treated Fen2 mutant 1 is also consistent with a deficiency of D24(24 )reductase activity due to the mutation in the Fen2 gene. It remains to be determined whether the tolerance mechanism depends on the accu1 mulation of D8,14,24(24 ) -sterols that might be less fungitoxic than the corresponding sterols saturated at C-24. Mutations leading to a modified enzyme, other than the known target of the fungicide in ergosterol biosynthesis, have already been suggested to be involved in resistance to DMIs (defective sterol D5(6)-desaturase), leading to circumvention of toxic sterol formation (1, 3). Otherwise, it could be asked if the 1 D24(24 )-reductase might also be a target of fenpropimorph. This hypothesis is suggested, taking into account that the catalysis of this enzyme might also involve carbocationic high-energy intermediates as proposed for D14-reductase and D8 → D7-isomerase (24, 33, 34), and because S. cerevisiae treated with relatively high concentra1 tions of 23-azacholesterol (32), a D24(24 )-reductase inhibitor, accumulated D8,14-sterols. The 1 D24(24 )-reductase has not yet been described as a target of fenpropropimorph but, this enzyme occurring at the very last step of ergosterol biosynthesis, its inhibition could be masked because of the major D14-reductase target. The Fen3 mutant exhibited a sterol profile similar to that of the wild-type strain in the absence of fenpropimorph but was different when the fungicide was present. Not only D8,14sterol accumulated but also D8-sterols in appreciable amounts compared to the wild-type strain. The D8-sterol accumulation could be higher than that of D8,14-sterols, depending on the fenpropimorph concentration. D8-Sterol accumulation indicates that D8 → D7-isomerase is also inhibited by fenpropimorph in these mutants, as already published for fenpropimorph-tolerant Fusarium species (16). As D14-reduction precedes D8 → D7-isomerization in ergosterol biosynthesis (27, 28), high levels of inhibition of D14-reduction would mask potential D8 → D7isomerization inhibition. In fact, in the case of S. cerevisiae, cell-free enzymatic studies have shown that both D14-reductase and D8 → D7isomerase are inhibited by fenpropimorph (35, 36), although mainly D8,14-sterols were detected by the analysis of sterols from yeast grown in presence of fenpropimorph (31). In the case of Fen3 mutant, a reduced sensitivity of the D14reductase toward fenpropimorph could explain FENPROPIMORPH- AND TERBINAFINE-RESISTANT MUTANTS OF N. haematococca the detection of D8 → D7-isomerization inhibition by D8-sterol accumulation. In order to obtain the same effects on ergosterol biosynthesis and growth inhibition as the wild-type strain, the Fen3 mutant required 10 times higher fenpropimorph concentration, supporting the view that the D14-reductase has reduced sensitivity. The resistance level value of the Fen3 mutant to fenpropidin (Table 1), a good D14-reductase inhibitor for this fungus as already mentioned, is also compatible with this hypothesis. The Fen3 mutant was as sensitive to tridemorph as the wild-type strain, suggesting that there was no change in sensitivity of the D8 → D7-isomerase, the main tridemorph target. The Fen4 mutant (R2F1-sp) exhibited a sterol profile different from that of the wild-type strain in the absence and in the presence of fenpropimorph, as did the mutant F2-2 and the mutant F2-sp, at least in the absence of fungicide. In the absence of fenpropimorph, these three mutants (R2F1-sp, F2-2, and F2-sp) contained C4-dimethylated and -monomethylated ergosterol precursors in greater amounts than the wild-type strain, indicating modification in ergosterol biosynthesis. The percentage of mono- and dimethylated sterols decreased with increasing fenpropimorph concentrations. A partial deficiency in C4demethylation is thus unlikely responsible for the C4-methylated sterol accumulation in the absence of fungicide. Among the 4a-methylsterols which accumulated in the absence of fungicide, 4a-methylepisterol (not detected in the wild-type strain) was found to be as abundant as 4a-methylfecosterol. Thus, the D8 → D7-isomerase appears to have low substrate specificity, allowing D8 → D7-isomerization of a 4a-methylsterol instead of a 4-desmethylsterol, as in normal ergosterol biosynthesis of filamentous fungi (27, 28, 34). This could be due to the higher 4amethylfecosterol level compared to wild-type strain, possibly coupled with higher D8 → D7isomerase activity in these mutants. Preliminary results from cell-free extract enzymatic studies currently in progress with the F2-2 strain were in agreement with an increased activity of this enzyme (Debieu, Taton, and Rahier, unpublished results). As Fen4 mutant resistance was not 181 observed toward tridemorph (13, 14), the increased activity of the D8 → D7-isomerase of the mutant could be due to a higher affinity toward its substrate than in the wild-type enzyme. Fenpropimorph treatment of the Fen4 mutant and the F2-2 strain led to D8,14sterol accumulation among 4,4-dimethyl-, 4amethyl-, and 4-desmethylsterols, indicating that the D14-reductase was the main target, as in the wild-type strain. To obtain percentages of ergosterol and D8,14-sterols similar to those of the wild-type strain, the Fen4 mutant and F2-2 strain required higher fenpropimorph concentrations. A modification of the D14-reductase could be involved in these cases. In order to explain the observation, such a modification must lead to a lower sensitivity to fenpropimorph to account for the phenotype resistant toward fenpropimorph and fenpropidin but should also confer increased D14-reductase activity to explain C4 mono- and dimethylated sterol accumulation. It might be due to a target structural modification leading to a reduced affinity toward fenpropimorph and to an increased affinity toward its own substrate. Another possibility could be overexpression of the gene, resulting in enzyme overproduction. The two mutants F2-2 and F2sp, for which genetic analysis of resistance has not been completed because of a lack of fertility (13), exhibited characteristics similar to those of the Fen4 mutant, considering both some phenotypic aspects (weakly reduced growth rate and pigment excretion) (13, 14) and sterol profile with (only F2-2 tested) or without fenpropimorph. This could mean that the F2-2 and possibly the F2-sp strains carry mutations similar to that carried by the Fen4 mutant. However, the F2-2 and F2-sp strains were largely more resistant than the Fen4 mutant (13, 14) and the fenpropimorph concentration required to obtain a similar effect on ergosterol biosynthesis was higher for the F2-2 mutant than for the Fen4 mutant. This could be due to a higher degree of modification of the D14-reductase in the case of the F2-2 strain. Moreover, comparison of ergosterol biosynthesis inhibition and growth inhibition had revealed a better tolerance to ergosterol biosynthesis inhibition in the case of the F2-2 182 LASSERON-DE FALANDRE ET AL. strain as well as the double mutant Fen1Fen4 than for the single mutant Fen4. Thus, it could be hypothetised that the resistance in the F22 strain might be due to two mechanisms of resistance, one induced by a Fen4-type mutation and another induced by a mutation leading to tolerance to ergosterol decrease and D8,14-sterol accumulation, as proposed for the Fen1-type mutation. It is noteworthy, that the Fen4-type mutation induced mycelial growth rate only slightly affected (83–91% of that of wild-type strain), whereas mono- and dimethylated sterols in C4 were present in appreciable amounts (38– 45% of total sterols). Thus, the sterols monoor dimethylated sterol on C4, which have been proposed to be fungitoxic (29), according to Nes et al. (37), induce only slight effects when they represent less than 50% of total sterols with ergosterol remaining the major sterol. As reported earlier (29), terbinafine treatment of N. haematococca led only to squalene accumulation, indicating that squalene epoxidase is the only target of terbinafine in ergosterol biosynthesis, as has been found in other fungi (8,12). All the terbinafine-resistant strains, Ter1 mutants, exhibited a sterol profile similar to that of the wild-type strain in terms of sterol composition and amount. However, they all accumulated squalene, which could suggest a modified squalene epoxidase in these mutants. Resistance expressed at the mycelial growth level correlated with effects at squalene epoxidase inhibition; a terbinafine concentration 30 times higher than for the wild-type strain was required to obtain similar growth inhibition and very high squalene amount. Thus, a modified squalene epoxidase with a reduced affinity for terbinafine could explain resistance in these Ter1 mutants. Furthermore, a reduced affinity of the enzyme for its substrate would account for squalene accumulation in the absence of terbinafine. Such a mechanism was found by studying the effects of terbinafine on squalene epoxidase activity in a cell-free extract, prepared from a laboratoryinduced terbinafine-resistant isolate of Ustilago maydis (39), which also accumulated squalene in the absence of terbinafine. In the Ter1 mutants the squalene accumulation factor in the absence of fungicide varied from 11 to 31, depending on the strain. This could correspond to different alleles of the same Ter1 gene, inducing different degrees of alteration to the squalene epoxidase. The Ter1 mutants exhibiting mycelial growth rates similar to that of the wild-type strain in the absence of terbinafine (13), it could be concluded that a squalene amount at least 10 times higher than in the wild-type strain, corresponding to about 1 mg mg21 dry weight, did not affect mycelial growth. Otherwise, in presence of terbinafine, as shown in the wild-type strain of N. haematococca (29), squalene amounts reaching about 10 mg mg21 dry weight seem to be implicated in fungicidal activity in the Ter1 resistant strains. Ergosterol level being decreased by less than 30%, the squalene amount represented more than two-times the sterol amount. In the absence of fungicide, ergosterol levels in Ter1 mutants were similar to those in wild-type strains despite a modified squalene epoxidase. This would probably be linked with regulation phenomenoma. Little information is known about ergosterol biosynthesis regulation; however, it has been shown that in some fungi ergosterol regulates its own biosynthesis by feedback inhibition at the 3-hydroxy-3-methylglutaryl coenzyme A reductase level (39). In conclusion, although most of the selected laboratory mutants of N. haematococca resistant toward the DMI tebuconazole had been found to be affected in fungicide uptake (20), it was not the case in strains selected for resistance to fenpropimorph and terbinafine. Tolerance toward the accumulation of D8,14-sterols and decreases in major D5,7-sterols could explain the resistance of some fenpropimorph-resistant strains. Modified enzymatic targets offer a better explanation for the resistance of the terbinafineresistant strains and the remainder of the fenpropimorph-resistant strains. Because fitness appears to be unaffected in the case of the Fen1 and Ter1 mutants (13), it could be hypothesized that tolerance toward ergosterol biosynthesis inhibition and modified targets are potential mechanisms that could be implicated in resistance in practice. To date, in the few reported cases of isolates of Erysiphe graminis exhibiting FENPROPIMORPH- AND TERBINAFINE-RESISTANT MUTANTS OF N. haematococca a reduced sensitivity to morpholines, fitness seems to be slightly affected (6, 7). Biochemical mechanisms responsible for reduced sensitivity have been recently investigated in only two isolates of E. graminis f. sp. tritici, showing that an altered sterol composition cannot explain the reduced sensitivity to fenpropimorph (40). A modified target previously described in DMI resistance in laboratory mutants (1, 3, 41) has recently been found in DMI-resistant field isolates of the phytopathogenic fungus Uncinula necator (42) and clinical isolates of the human pathogenic fungus Candida albicans (43) by identification of mutations in the gene encoding the fungicide target 14a-demethylase. ACKNOWLEDGMENTS The authors thank Catherine Albertini for helpful discussion and English corrections and the agrochemical companies, BASF AG. (Germany), Dr. Maag AG. (Switzerland), and Sandoz AG. (Switzerland), which kindly provided fungicides. REFERENCES 1. W. Köller, Antifungal agents with target sites in sterol functions and biosynthesis, in “Target Sites of Fungicide Action” (W. Köller, Ed.), pp. 119–206, CRC Press, Boca Raton, FL, 1992. 2. K. H. Kuck, H. Scheinpflug, and R. Pontzen, DMI fungicides, in “Modern Selective Fungicides— Properties, Applications, Mechanisms of Action” (H. Lyr, Ed.), pp. 205–258, Gustav Fisher Verlag, Jena, 1995. 3. M. A. De Waard, Resistance to fungicides which inhibit sterol 14a-demethylation, an historical perspective, in “Fungicide Resistance—Monograph N8609 (S. Heaney, D. Slawson, D. W. Hollomon, M. Smith, P. E. Russell, and D. W. Parry, Eds.), pp. 3–10, BCPC, Surrey, UK, 1994. 4. H. Buchenauer, DMI-fungicides-side effects on the plant and problems of resistance: Resistance of fungi to sterol demethylation inhibitors, in “Modern Selective Fungicides—Properties, Applications, Mechanisms of Action” (H. Lyr, Ed.), pp. 280–290, Gustav Fisher Verlag, Jena, 1995. 5. A. Kerkenaar, Mechanism of action of cyclic amine fungicides: Morpholines and piperidines, in “Modern Selective Fungicides—Properties, Applications, Mechanisms of Action” (H. Lyr, Ed.), pp. 185–204, Gustav Fisher Verlag, Jena, 1995. 6. D. W. Hollomon, Do morpholine fungicides select for resistance?, in “Fungicide Resistance—Monograph N8609 (S. Heaney, D. Slawson, D. W. Hollomon, M. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 183 Smith, P. E. Russell, and D. W. Parry, Eds.), pp. 281– 289, BCPC, Surrey, UK, 1994. A. J. G. Engels, and M. A. DE Waard, Fitness of isolates of Erysiphe graminis f. sp. tritici with reduced sensitivity to fenpropimorph, Crop Prot. 15, 771 (1996). N. S. Ryder, Squalene epoxidase-enzymology and inhibition, in “Biochemistry of Cell Walls and Membranes in Fungi” (P. J. Kuhn, A. P. J. Trinci, M. J. Jung, M. W. Goosey, and L. G. Copping, Eds.), pp. 189–203, Springer-Verlag, Berlin, 1990. H. Vanden Bossche, Chemotherapy of human fungal infections, in “Modern Selective Fungicides: Properties, Applications, Mechanism of Action” (H. Lyr, ed.), pp. 431–84, Gustav Fisher Verlag, Jena, 1995. G. A. Carter, S. J. Kendall, R. S. Burden, C. S. James, and T. Clarck, The lipid compositions of two isolates of Cladosporium cucumerinum do not explain their differences in sensitivity to fungicides which inhibit sterol biosynthesis, Pestic. Sci. 26, 181 (1989). P. Leroux, M. Gredt, and P. Boeda, Resistance to inhibitors of sterol biosynthesis in field isolates or laboratory strains of the eyespot pathogen Pseudocercosporella herpotrichoides, Pestic. Sci. 23, 119 (1988). A. B. Orth, and H. D. Sisler, Mode of action of terbinafine in Ustilago maydis and characterization of resistants mutants, Pestic. Biochem. Physiol. 37, 53 (1990). A. Lasseron-De Falandre, M. J. Daboussi, and P. Leroux, Inheritance of resistance to fenpropimorph and terbinafine, two sterol biosynthesis inhibitors, in Nectria haematococca, Phytopathology 81, 1432 (1991). D. Debieu, A. Lasseron, C. Gall, M. Gredt, J. Bach, and P. Leroux, Mechanisms of selectivity and resistance to fenpropimorph and related fungicides in Fusarium genus, German Phytomed. Soc. Ser. 4, 339 (1993). H. D. Van Etten, Identification of additional habitats of Nectria haematococca mating population VI, Phytopathology 68, 1552 (1978). D. Debieu, C. Gall, M. Gredt, J. Bach, C. Malosse, and P. Leroux, Ergosterol biosynthesis and its inhibition by fenpropimorph in Fusarium species, Phytochemistry 31, 1223 (1992). A. Rahier, and P. Benveniste, Mass spectral identification of phytosterols in “Analysis of Sterols and Other Biologically Significant Steroids” (W. D. Nes and E. J. Parish, Eds.), pp. 223–250, Academic Press, San Diego, 1989. D. H. R. Barton, T. Shioiri, and D. A. Widdowson, Biosynthesis of terpenes and steroı̈ds. Part V. The synthesis of ergosta-5,7,22,24(241)-tetraen-3b -ol, a biosynthetic precursor of ergosterol. J. Chem. Soc. 1968 (1971). M. A De Waard, and J. G. M. Van Nistelrooy, Accumulation of SBI fungicides in wild-type and fenarimol-resistant isolates of Penicillium italicum, Pestic. Sci. 22, 371 (1988). R. Akallal, D. Debieu, C. Lanen, M. J. Daboussi, R. Fritz, C. Malosse, J. Bach, and P. Leroux, Inheritance 184 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. LASSERON-DE FALANDRE ET AL. and mechanisms of resistance to tebuconazole, a sterol C-14 demethylation inhibitor, in Nectria haematococca, Pestic. Biochem. Physiol. 60, 147 (1998). A. E. Kalamaris, M. A. De Waard, B. N. Ziogas, and S. G. Georgopoulos, Resistance to fenarimol in Nectria haematococca var. cucurbitae, Pestic. Biochem. Physiol. 40, 212 (1991). M. A. De Waard, and J. G. M Van Nistelrooy, An energy-dependent efflux fenarimol-resistant mutant of Aspergillus nidulans, Pestic. Biochem. Physiol. 13, 255 (1980). C. Stehmann, and M. A. De Waard, Accumulation of tebuconazole by isolates of Botrytis cinerea differing in sensitivity to sterol demethylation inhibiting fungicides, Pestic. Sci. 45, 311 (1995). A. Rahier, and M. Taton, Fungicides as tools in studying postsqualene sterol synthesis in plants, Pestic. Biochem. Physiol. 57, 1 (1997). A. Lasseron-Ferault De Falandre, “Etude de la Résistance à des Fongicides Inhibiteurs de la Biosynthèse des Stérols chez un Champignon Filamenteux: Nectria haematococca,” Ph.D dissertation, Paris XI Orsay University, France (1991). T. Joseph-Horne, D. Hollomon, N. Manning, and S. L. Kelly, Investigation of the sterol composition and azole resistance in field isolates of Septoria tritici, Appl. Envir. Microbiol. 62, 184 (1996). J. D. Weete, Sterols of the fungi: Distribution and biosynthesis, Phytochemistry 12, 1843 (1973). E. I. Mercer, The biosynthesis of ergosterol, Pestic. Sci. 15, 133 (1984). D. Debieu, J. Bach, A. Lasseron, C. Malosse, and P. Leroux, Effects of sterol biosynthesis inhibitor fungicides, in the phytopathogenic fungus, Nectria haematococca: Ergosterol depletion versus precursor or abnormal sterol accumulation as the mechanism of fungitoxicity, Pestic. Sci. 54, 157 (1998). M. Grindle, and R. Farrow, Sterol content and enzyme defects of nystatin-resistant mutants of Neurospora crassa, Mol. Gen. Genet. 165, 305 (1978). R. I. Baloch, E. I. Mercer, T. E. Wiggins, and B. C. Baldwin, Inhibition of ergosterol biosynthesis in Saccharomyces cerevisiae and Ustilago maydis by tridemorph, fenpropimorph and fenpropidin, Phytochemistry 23, 2219 (1984). H. D. Pierce, A. M. Pierce, R. Srinivasam, A. M. Unrau, and A. C. Oehlschlager, Azasterol inhibitors in yeast: Inhibition of the 24-methylene sterol D24(28)-reductase 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. and D24-sterol methyltransferase of Saccharomyces cerevisiae by 23-azacholesterol, Biochem. Biophys. Acta 529, 429 (1978). A. Rahier, M. Taton, and P. Benveniste, Inhibition of sterol biosynthesis inhibitor in higher plants by analogues of high energy carbocationic intermediate, in “Biochemistry of Cell Wall and Membranes in Fungi”, (P. J. Kuhn, A. P. J. Trinci, M. J. Jung, M. W. Goosey, and L. G. Copping, Eds.), pp. 205–222, Springer-Verlag, Berlin, 1990. E. I. Mercer, Sterol biosynthesis inhibitors: Their current status and modes of action, Lipids 26, 584 (1991). R. I. Baloch, and E. I. Mercer, Inhibition of sterol D8 → D7-isomerase and D14-reductase by fenpropimorph, tridemorph and fenpropidin in cell-free enzyme systems from Saccharomyces cerevisiae, Phytochemistry 26, 663 (1987). P. Masner, C. C. Steel, R. Furter, H. P. Isenring, G. Oesterhelt, H. Ziegler, and P. Zobrist, Novel inhibitors of sterol C-14 demethylase and D14-reductase/D8 → D7isomerase for cereal disease control, Pestic. Sci. 35, 339 (1992). W. D. Nes, G. G. Janssen, F. G. Crumley, M. Kalinowska, and T. Akihisa, The structural requirements of sterols for membrane function in Saccharomyces cerevisiae. Arch. Biochem. Biophys. 300, 724 (1993). A. B. Orth, M. J. Henry, and H. D. Sisler, Mechanism of resistance to terbinafine in two isolates of Ustilago maydis, Pestic. Biochem. Physiol. 37, 182 (1990). M. Bard, and J. F. Downing, Genetic and biochemical aspects of yeast sterol regulation involving 3-hydroxy3-methyl-glutaryl coenzyme A reductase, J. Gen. Microbiol. 125, 415 (1981). A. J. G. Engels, and M. A. De Waard, Sterol composition of isolates of Erysiphe graminis f. sp. tritici differing in sensitivity to fenpropimorph, Pestic. Sci. 52, 258 (1998). M. A. De Waard, Molecular genetics of resistance in fungi to azole fungicides, in “Molecular Genetics and Evolution of Pesticide Resistance” (T. M. Brown, Ed.), ACS Symp. Ser. 645, pp. 62–71, Washington, DC, 1996. ´ C. Delye, F. Laigret, and M. F. Corio-Costet, A mutation in the 14a-demethylase gene of Uncinula necator that correlates with resistance to a sterol biosynthesis inhibitor, Appl. Environ. Microbiol. 63, 2966 (1997). D. Sanglard, F. Isher, L. Koymans, and J. Bille, Amino acid substitutions in the cytochrome P-450 lanosterol 14a-demethylase (CYP51A1) from azole-resistant Candida albicans clinical isolates contribute to resistance to azole antifungal agents, Antimicrob. Agents Chemother. 42, 241 (1998).