Evaluating aggressiveness and host range of Alternaria dauci in a controlled environment


  • C. Boedo,

    1. Université d’Angers, IFR Quasav 149, UMR PaVé A77, UFR Sciences, 2 boulevard Lavoisier
    2. Agrocampus Ouest Centre d’Angers – Institut National d’Horticulture et de Paysage, IFR Quasav 149, UMR GenHort 1259, 2 rue Le Nôtre
    Search for more papers by this author
    • Contributed equally to the present work.

  • S. Benichou,

    1. Université d’Angers, IFR Quasav 149, UMR PaVé A77, UFR Sciences, 2 boulevard Lavoisier
    Search for more papers by this author
    • Contributed equally to the present work.

  • R. Berruyer,

    1. Université d’Angers, IFR Quasav 149, UMR GenHort 1259, UFR Sciences, 2 boulevard Lavoisier, 49045 Angers Cedex 01, France
    Search for more papers by this author
  • S. Bersihand,

    1. Université d’Angers, IFR Quasav 149, UMR PaVé A77, UFR Sciences, 2 boulevard Lavoisier
    Search for more papers by this author
  • A. Dongo,

    1. Université d’Angers, IFR Quasav 149, UMR PaVé A77, UFR Sciences, 2 boulevard Lavoisier
    Search for more papers by this author
  • P. Simoneau,

    1. Université d’Angers, IFR Quasav 149, UMR PaVé A77, UFR Sciences, 2 boulevard Lavoisier
    Search for more papers by this author
  • M. Lecomte,

    1. Agrocampus Ouest Centre d’Angers – Institut National d’Horticulture et de Paysage, IFR Quasav 149, UMR GenHort 1259, 2 rue Le Nôtre
    Search for more papers by this author
  • M. Briard,

    1. Agrocampus Ouest Centre d’Angers – Institut National d’Horticulture et de Paysage, IFR Quasav 149, UMR GenHort 1259, 2 rue Le Nôtre
    Search for more papers by this author
  • V. Le Clerc,

    1. Agrocampus Ouest Centre d’Angers – Institut National d’Horticulture et de Paysage, IFR Quasav 149, UMR GenHort 1259, 2 rue Le Nôtre
    Search for more papers by this author
  • P. Poupard

    Corresponding author
    1. Université d’Angers, IFR Quasav 149, UMR PaVé A77, UFR Sciences, 2 boulevard Lavoisier
      E-mail: pascal.poupard@univ-angers.fr
    Search for more papers by this author

E-mail: pascal.poupard@univ-angers.fr


The aggressiveness of Alternaria dauci isolates was investigated in greenhouse conditions. Twenty-seven isolates were pre-selected from a large collection to represent high diversity according to geographic or host origins and intergenic spacer (IGS) polymorphism. IGS sequence analysis revealed that isolates were grouped within three different clusters. Eleven isolates were selected and inoculated on a susceptible carrot cultivar. Three criteria (mean lesion number, mean necrotic leaf area and mean disease index) were used to assess the aggressiveness of isolates. Continuous variation in aggressiveness was shown and no clear division into isolate classes was evident. For the host range study, two isolates were inoculated under greenhouse conditions onto nine cultivated Apiaceae species, two wild Daucus species and six cultivated non-Apiaceae species representing six botanical families. Lesions varying in severity were observed on all dicot species (Apiaceae and non-Apiaceae), but no symptoms developed on the two monocots studied (leek and sweetcorn). Plant species were also differentiated on the basis of expanding lesions (cultivated and wild carrot, dill and fennel) or non-expanding lesions (other dicot species). Typical A. dauci conidia were observed after in vitro incubation of leaves with symptoms. Fungal structures were isolated from lesions and A. dauci was confirmed on the basis of conidial morphology and specific conventional PCR results. Genotyping of individual isolates performed with microsatellite markers confirmed the presence of the inoculated isolate. The results clearly showed that, in controlled conditions, the host range of A. dauci is not restricted to carrot.


Alternaria leaf blight (ALB), caused by the necrotrophic fungus Alternaria dauci, is one of the most destructive foliar diseases of cultivated carrot, Daucus carota subsp. sativus. The disease is present in carrot production areas worldwide. Different potential sources of disease inoculum are known, including infected carrot seeds, volunteer carrot plants, decaying carrot debris, and wild carrot (Farrar et al., 2004). After plant infection, young greenish-brown lesions of different shapes and sizes form on leaves, usually along the leaflet margins, while elongated lesions develop on petioles. Older lesions become dark brown and can be surrounded by a chlorotic halo. Lesions spread gradually, giving the leaf a blighted appearance, and in the worst cases the entire leaf can be affected. Infection of inflorescences, seeds and developing seedlings also occurs (Joly, 1964; Farrar et al., 2004). Two main effects are associated with the disease: first, symptoms can considerably reduce leaf photosynthetic activity and carbohydrate production; secondly, necrotic foliage can separate from the taproot, significantly reducing harvesting efficiency. In severely diseased fields, both consequences contribute to a significant reduction in carrot yield. Although integrated control measures can be implemented to limit disease impact, severe losses caused by ALB have been reported for this crop (Ben-Noon et al., 2003). For example, a reduction up to 74% in commercial fields was reported in Israel (Ben-Noon et al., 2001).

The importance of studies aimed at gaining greater insight into the aggressiveness of plant pathogens was recently highlighted in a review by Pariaud et al. (2009). In several plant pathogen species, aggressiveness is a classical quantitative trait, influenced by genotype and environment, especially climatic parameters. As a quantitative trait, it is subject to selection, following a complex pattern in which several trade-offs are made, especially with qualitative virulence and quantitative survival ability. Differential interaction between isolates and cultivars or a link between aggressiveness levels and population structures can also indicate the quantitative adaptation of the pathogen to host cultivars.

Little is known about pathogenicity components of A. dauci, especially when considering aggressiveness. Recently, Rogers & Stevenson (2010) showed that the aggressiveness of A. dauci significantly varied among isolates originating from a relatively small area in the northern USA. Differences in aggressiveness among A. dauci isolates collected worldwide has, however, never been investigated. In previous papers focusing on different aspects of A. dauci– carrot interactions, such as infection processes, ALB resistance evaluation or genetic determinism (Dugdale et al., 2000; Pawelec et al., 2006; Boedo et al., 2008, 2010; Le Clerc et al., 2009), only one or two isolates were generally used. It would be very important to assess the aggressiveness of A. dauci isolates before using them in future studies.

The virulence of A. dauci has been considered in previous studies in which A. dauci was reported to be capable of infecting plant species other than carrot, including cultivated and wild Apiaceae species (Felix & Orieux, 1963; Joly, 1964; Neergaard, 1977; Soteros, 1979). Seed transmission of A. dauci has been reported in carrot and other cultivated Apiaceae species, such as parsley and celery (Strandberg, 1992). The fungus has also been isolated from seeds of Daucus maximus, a wild Apiaceae species (Netzer & Kenneth, 1969). The different observations described above suggest that the A. dauci host range may not be restricted to cultivated carrot, but at this point it is still unclear. Indeed, in Israel, field crops of parsley, celery and fennel situated next to carrot fields with ALB outbreaks were not infected by A. dauci as determined by regular inspections of the neighbouring fields for disease symptoms (Netzer & Kenneth, 1969).

The first objective of the present study was to compare the aggressiveness of a set of A. dauci isolates after inoculation of a susceptible carrot cultivar which had previously been used as a positive disease control in several studies (Ben-Noon et al., 2001; Pawelec et al., 2006; Boedo et al., 2008, 2010; Le Clerc et al., 2009). For that purpose, A. dauci isolates were chosen from a worldwide collection of this fungal species to represent high diversity according to geographic origin and polymorphism in the intergenic spacer (IGS) regions of nuclear rDNA. The second objective of this study was to examine the host range of A. dauci, using cultivated non-carrot species belonging to the Apiaceae and other plant families. Two wild Daucus species (D. carota and D. bicolor) were also investigated as they could potentially represent a source of disease inoculum.

Materials and methods

Plant and fungal material

For the study of A. dauci aggressiveness, the susceptible carrot cv. Presto (Vilmorin) was grown in boxes (60 × 40 × 23 cm) with five plants per box containing a peat moss (Traysubstrat Klasmann)/sand (67:33, v/v) mixture in a greenhouse with day/night temperatures of 20 ± 2°C, a 16-h photoperiod and 80–85% RH, as previously described (Boedo et al., 2008). Eleven A. dauci isolates were selected from a subcollection of 27 (Table 1) originating from a worldwide collection of 120 (UMR Pathologie Végétale, Angers, France).

Table 1.   Sources of Alternaria spp. isolates evaluated in this study for variability based on EF-1α and IGS sequence polymorphism and for some isolates, aggressiveness and host range
Isolate accessionOther collection codeHost of originGeographic originYear of isolationGenBank accession IGS, EF-1αIGS patterna
  1. AD: A. Dongo; BA: B. Andersen; BMP: B. M. Pryor; CBS: Centraal Bureau voor Schimmelcultures; EGS: E. G. Simmons; MJB, M. J. Barbetti; WAC: Western Australia Plant Pathogen Collection; n.a.: not applicable; NA: not available.

  2. aIGS patterns correspond to clusters described in Figure 3.

Alternaria dauci
 AUS00110631Carrot (leaf)AustraliaNAHM481422, HQ403585C3
 BRA002Vil1(93.19)3Carrot (leaf)Brazil2006HM481423, HQ403580C1
 FRA001A2Carrot (leaf)France (Maine-et-Loire)2000HM481424, HQ403577C1
 FRA005B4Carrot (leaf)France (Indre)2000HM481425, –C2
 FRA012C5Carrot (leaf)France (Gers)2000HM481426, –C2
 FRA016P1SCarrot (leaf)France (Gironde)2000HM481427, –C1
 FRA017P2Carrot (leaf)France (Gironde)2000HM481428, HQ403581C1
 FRA018P3Carrot (leaf)France (Gironde)2000HM481429, HQ403582n.a.
 FRA019R9228-4Cress (seed)FranceNAHM481430, HQ403584C3
 FRA026R6552-1Carrot (leaf)France (Loiret)2003HM481431, –C2
 FRA032R6557-1Carrot (leaf)France (Gers)2003HM481432, –C2
 FRA046Brion3Chicory (leaf)France (Maine-et-Loire)2003HM481433, –C3
 FRA047Brion4Carrot (leaf)France (Maine-et-Loire)2003HM481434, –C1
 GER001AllCarrot (leaf)Germany1989HM481435, HQ403578C2
 GER002AN180Carrot (leaf)Germany2005HM481436, –C3
 GER003AN181Carrot (leaf)Germany2005HM481437, –C3
 HUN002AN189Carrot (leaf)Hungary2005HM481438, –C1
 ITA002AdIt03Carrot (leaf)Italy2006HM481439, HQ403579C3
 JPN002AdJa02Carrot (leaf)Japan2007HM481440, –C3
 NED001CBS101592Carrot (seed)Netherlands1993HM481441, HQ403586C1
 USA002RSot104641Carrot (leaf)USA1997HM481442, HQ403583C1
 USA003Hw104A16Carrot (leaf)USA1997HM481443, –C1
 USA006BMP155Carrot (seed)USA1992HM481444, –C2
 USA012BMP161Carrot (seed)USA1994HM481445, –C2
 USA014BMP167Carrot (leaf)USA (Florida)NAHM481446, –C1
 USA016BMP171CarrotUSA (California)NAHM481447, –C1
 THA003C5356-3Tomato (seed)Thailand2008HM481448, HQ403576C1
Alternaria tomatophila
 BA1523EGS-44-024Tomato (leaf)Australia1996–, HQ403565n.a.
 BA1443EGS-44-074Tomato (leaf)USA1996–, HQ403566n.a.
Alternaria porri
 AD238 LeekFranceNA–, HQ403569n.a.
Alternaria solani
 CBS110.41 Solanum aviculare (leaf)Netherlands1941–, HQ403570n.a.
 MJB10245WAC10245PotatoAustralia2002–, HQ403572n.a.
Alternaria zinniae
 BA1441AD234Zinnia elegans (leaf)New Zealand1996–, HQ403571n.a.
 AD254 Zinnia elegans (leaf)Hungary1984–, HQ403575n.a.
Alternaria bataticola
 CBS531.63 Ipomea batatasJapan1963–, HQ403574n.a.

For the A. dauci host range study, Apiaceae and non-Apiaceae plant species (Table 2) were grown in 5-L pots (three pots per species, 10 plants per pot) in the greenhouse conditions indicated above. Carrot cv. Presto was used as a positive control. As some isolates in the A. dauci collection originated from non-carrot species, cress and tomato were included in this study. Leek, sweetcorn and corn salad, which are often introduced into carrot crop rotations, were tested as members of botanical families differing from the Apiaceae, Brassicaceae and Solanaceae. Radish was chosen on the basis of preliminary results (C. Boedo, Université d'Angers, unpublished data). A. dauci isolates FRA001 and FRA017 obtained from carrot leaves and, for some experiments, A. dauci isolate THA003 from tomato seeds (cv. Caporal) were used (Table 1). FRA001 and FRA017 were chosen as reference isolates because of their use in previous studies (Pawelec et al., 2006; Boedo et al., 2008, 2010; Le Clerc et al., 2009). For production of conidia, fungal isolates were grown in Petri dishes on V8 juice agar medium (200 mL V8 vegetable juice, 3 g CaCO3, 15 g agar L−1, pH 6·8), incubated in darkness at 22 ± 2°C for 7 days, and then exposed to 12 h near-ultraviolet light (Philips, TL-D 36 W) alternating with 12 h darkness at 22 ± 2°C for 10–15 days.

Table 2.   Characteristics of plant species used for the Alternaria dauci host range study comprising Apiaceae species (cultivated and wild Daucus species) and cultivated non-Apiaceae species
Common name or reference used in the studyScientific nameReference of seed sample and supplierBotanical family
  1. WD: Wild Daucus species.

  2. aDaucus sp. collection from INHP, Agrocampus Ouest, Angers, France.

  3. bDaucus sp. collection from HRI, Warwick University, UK.

CarrotDaucus carotacv. Presto – VilmorinApiaceae
CeleriacApium graveolens var. rapaceumcv. Neve – ClauseApiaceae
CeleryApium graveolenscv. Deacon – ClauseApiaceae
ChervilCerefolium sativumClauseApiaceae
CorianderCoriandrum sativumCLX 3610 – ClauseApiaceae
DillAnethum graveolensClauseApiaceae
FennelFoeniculum vulgarecv. Cormo – ClauseApiaceae
ParsleyPetroselinum crispumcv. Race Calito – ClauseApiaceae
ParsnipPastinaca sativaCLX 3651 – ClauseApiaceae
WD 1D. carota sg. carota ssp. carotaINH 650aApiaceae
WD 2D. carota sg. carotaINH 617Apiaceae
WD 3D. carota sg. carota ssp. carota var. mauritanicusINH 653Apiaceae
WD 4D. carota sg. gummifer ssp. gummifer var. fontanesiiINH 667Apiaceae
WD 5D. carota sg. gummiferHRI 9217bApiaceae
WD 6D. bicolorINH 610Apiaceae
Corn saladValerianella olitoracv. Gala – ClauseValerianaceae
CressNasturtium officinalecv. De Fontaine Claudia – ClauseBrassicaceae
LeekAllium porrumcv. Atal – ClauseLiliaceae
RadishRaphanus sativuscv. Java – ClauseBrassicaceae
SweetcornZea mayscv. Landmark – ClausePoaceae
TomatoSolanum lycopersicumcv. Supersweet – ClauseSolanaceae

Morphological and molecular identification of A. dauci isolates

For identification of A. dauci conidia, the surface of colonies was scraped in the presence of 150 μL distilled water and conidial suspensions were collected. Conidial morphology was observed under a microscope (Leica DM4500B; Leica Microsystems). Morphological identification was performed according to Simmons (2007). The identification was confirmed by molecular criteria based on the comparative analysis of elongation factor 1α (EF-1α) and ITS sequences within A. dauci isolates and eight additional isolates representing five species from the porri species group (Table 1). Extraction of DNA from fungal colonies was performed according to Goodwin & Lee (1993). Partial EF-1α and ITS sequences were obtained from PCR products using primer sets EF1-728F/EF1-986R and ITS-1/ITS-4 in the conditions described by Carbone & Kohn (1999) and White et al. (1990), respectively. Phylogenetic analysis was carried out using the Phylogeny.fr web service (Dereeper et al., 2008). Sequences were aligned with muscle software (Edgar, 2004) and a dendrogram was constructed using the neighbour-joining algorithm (Saitou & Nei, 1987). Bootstrap confidence values were calculated from 1000 randomly resampled datasets.

Intraspecific variability within A. dauci isolates based on IGS sequences

The complete IGS sequence of A. dauci reference isolate NED001 (CBS101592, GenBank acc. no. HM536199) was obtained from a cloned PCR fragment using the conditions described in Hong et al. (2005b) and the primer set 26S3111F and IGS27. Internal primers (IGSup1: 5′-GTCGTGTAAGTAGTCGAGTAG-3′ and IGSdo1: 5′-GCCCAATAGCCAACTATCCGC-3′) were designed from the IGS sequence of NED001 to amplify a large sequence portion from the IGS of A. dauci isolate FRA019. Intergenic spacer sequences of NED001 and FRA019 were aligned using ClustalW software and their structural features (sequence repeats) and nucleotide differences (SNPs, indels) were analysed. The 5′ variable domain from the IGS of the 27 A. dauci isolates included in the subcollection (Table 1) was amplified using IGSup1 and IGSdo2 (5′-CCTAACTAAATAGTAAGAGAG-3′) specific primers. The reaction mixture (50 μL) consisted of 2 μL of 10-fold diluted fungal DNA, 1 U GoTaq® polymerase (Promega), 1× reaction buffer, 1·4 mm MgCl2, 20 pmol of each primer set and 0·2 mm dNTPs. Amplification by PCR was performed with the following parameters: initial denaturation at 94°C for 3 min, 34 cycles of denaturation at 94°C for 30 s, annealing at 56°C for 50 s and extension at 72°C for 2 min, and a final extension at 72°C for 10 min. Amplification products were resolved on 1·2% agarose gels in 0·5× TAE buffer (20 mm Tris-acetate pH 8, 0·5 mm EDTA) followed by ethidium bromide staining. Phylogenetic analysis from sequenced amplified fragments (GenBank acc. nos HM481422HM481448, see Table 1) was carried out as described above, except that gap positions were removed before constructing the dendrogram. For these analyses, Alternaria tenuissima (IGS sequence of the CH1 isolate from Solanum tuberosum, GenBank acc. no. EU519465) was used as an outgroup to root the tree.

Assessment of A. dauci aggressiveness

Conidial suspensions were prepared as described in Pawelec et al. (2006). Specifically, fungal cultures were flooded with sterile water and conidia were gently dislodged with a glass plate. Mycelial and conidial suspensions were filtered through two layers of cheesecloth. Spore density was counted using a haemocytometer and adjusted to 1 × 103 conidia per mL. Tween 20 was added to the suspension at a final concentration of 0·05%. An inoculation chamber realized with a Petri dish was set up around the third true leaf (f3) without detaching it from the plant, as described in Boedo et al. (2010). For each fungal isolate studied, five leaves were inoculated by spraying 400 μL conidial suspension per leaf. Greenhouse temperature was set to 24 ± 2°C. Sterile distilled water (1 mL) was added to each Petri dish at 4, 8 and 12 days post-inoculation (d.p.i.) to maintain high moisture. The number of lesions was counted on each leaf at 8 and 13 d.p.i., while taking into account that large lesions were actually a merged combination of several smaller lesions (a large lesion was considered equivalent to 10 small lesions). Disease severity was rated at 13 d.p.i. by visual assessment based on a 0–5 scale adapted from Pawelec et al. (2006), according to the percentage of necrotic leaf area (0: no visible disease symptoms, 1: < 5% leaf area affected, 2: 5% ≤ leaf area affected < 20%, 3: 20% ≤ leaf area affected < 40%, 4: 40% ≤ leaf area affected < 60%, 5: ≥ 60% leaf area affected). Mean lesion number (LN) at 8 and 13 d.p.i., mean disease index (disease index for aggressiveness, DIa) and mean percentage of necrotic leaf area (NLA) at 13 d.p.i. were calculated. The whole experiment was repeated twice for each fungal isolate studied.

Disease ratings were statistically analysed using R-2·9·1 software (R Development Core Team, 2005), completed with the Agricolae package (version 1·0-3, F. de Mendiburu, Universidad Nacional de Ingenieria, Lima, Peru). The homoscedasticity of LN, log10 (LN), DIa and NLA were checked (Breusch–Pagan test, α = 1%). LN and NLA did not show homoscedasticity, whereas DIa and log10 (LN) did. A linear model from DIa and log10 (LN) was designed with repetition considered as a block treatment. Residuals were found to be normally distributed (normal Q–Q plot) in each case. Multiple comparisons were performed using a Waller–Duncan k-ratio test (Waller & Duncan, 1969) with k set at 100. Nonparametric statistics were used to analyse NLA data. Significance of isolate effects was analysed using a Kruskal–Wallis rank sum test (α = 1%) followed by multiple comparisons (Wilcoxon tests with a Bonferoni correction, α = 5%).

Alternaria dauci host range study

Six weeks after sowing, plants (three pots per species and per isolate) were inoculated with 25 mL per pot (i.e. before inoculum runoff) of conidial suspension prepared as described above. Control plants (three pots per species) were sprayed with 25 mL of a 0·05% Tween 20 solution in sterile distilled water. Greenhouse temperature was set to 24 ± 2°C. Plants were incubated under plastic covers. Boxes and plastic covers were sprayed twice a day for 2 days with sterile water in order to maintain high RH and then plastic covers were removed. Disease severity was rated at 4, 10, 15, 21 and 25 d.p.i. using the 0–9 scale described by Pawelec et al. (2006) (0: no visible disease symptoms; to 9: severe defoliation, only new leaves remaining). A single score was given to each inoculated pot and an average disease index (disease index for virulence or host range, DIv) was calculated for the three pots. The diameter of nine randomly selected necrotic lesions per plant species (three spots in one plant per pot) was measured at 4, 10, 15, 21 and 25 d.p.i. If a chlorotic halo was present around the necrosis, it was not considered in the diameter measurement.

For each plant species, three leaves showing necroses (one leaf per pot) were sampled between 15 and 35 d.p.i. for fungal sporulation. Each sampled leaf was individually incubated in a 90-mm-diameter Petri dish containing an 85-mm-diameter filter paper disc humidified with 2 mL distilled water. Incubation was performed for 2–5 days at 24°C in darkness. The formation of conidiophores and conidia was observed under a binocular magnifying glass (Leica MZ 9·5) and photographed using a Canon Power Shot S70 digital camera.

Fungal isolates were reisolated from inoculated plant species. At 15 and 21 d.p.i., surface-sterilized leaf explants (2 × 2 mm) were removed at the symptom periphery and put in Petri dishes containing V8 juice agar medium. Fungal growth, sporulation and identification of conidia were performed in the conditions described above. For molecular characterization at a species level, conventional PCR amplification using the A. dauci IGSup1 and IGSdo2 specific primer pair and agarose gel visualization were carried out as described above.

Amplifications for individual genotyping (isolates FRA001 and FRA017) were performed using the microsatellite markers defined in Benichou et al. (2009): F 5′-CACGACGTTGTAAAACGTACATAATGAAGGAGACAGCAACG-3′/R 5′-AGTCTCGTCACTGCATGAGCTG-3′ (locus Admic7, repeat motif (GAA)33) and F 5′-CACGACGTTGTAAAACGTTCCAACCCCTTCACGATACCC-3′/R 5′-TCAGTCGTGACCACTGCAGCTC-3′ (locus Admic8, repeat motif (GAA)15). The reaction mixtures (20 μL final volume) contained 1 μL of 10-fold diluted template DNA, 1 U GoTaq® DNA polymerase, 1× colourless GoTaq® Flexi Buffer, 1·5 mm MgCl2, dNTPs at 0·2 mm each, 0·025 μm IRD700-labelled universal primer and each specific primer at 0·125 μm. The sequence of the first 17 nucleotides of the specific forward primers (underlined part) was identical to that of the universal primer. Samples were processed through 25 cycles, consisting of 30 s at 94°C, 30 s at 54°C and 30 s at 72°C. Amplification products were separated on a LI-COR® 4300 genetic analyser using a 6·5% matrix gel (KBPLUS; LI-COR® Biosciences), 0·4 μL ladder (IRDye® 50–700 bp Size Standard; LI-COR® Biosciences) and 0·4 μL PCR product. Sample migration was carried out for 1 h 25 min at 20·3 mA (1350 V, 27·3 W) and 43°C. Amplification products were analysed using sagaGT software (LI-COR® Biosciences). The whole experiment from inoculation to individual genotyping of reisolated isolates was repeated twice.


Characterization of isolates and intraspecific variability within A. dauci based on IGS polymorphism

In order to further study the aggressiveness of A. dauci, a subcollection of 27 isolates (Table 1) was pre-selected from 120 isolates of A. dauci collected in America, Asia, Europe and Oceania, mostly from important commercial carrot production areas. The 27 isolates were chosen on the basis of the geographic and plant host origins: 24 isolates originated from carrot leaves or seeds, while three were isolated from chicory, cress or tomato. The selected isolates, including the A. dauci reference isolate NED001 from the Centraal Bureau voor Schimmelcultures (CBS 101592), were grown on V8 juice agar medium. All isolates showed colonies exhibiting typical characteristics of A. dauci by comparison with the NED001 reference colony. Conidial morphology of the three isolates originating from non-carrot hosts (chicory, cress or tomato) was specifically examined. For these three isolates, the conidia showed a typical terminal filamentous beak, highly resembling the conidia produced by the reference isolate NED001 (results not shown). In order to further confirm the identification of A. dauci isolates, a molecular approach was used to discriminate A. dauci from other Alternaria species. For this, a portion of c. 300 bp overlapping the large second intron of EF-1α was amplified and sequenced from 10 isolates chosen to represent diverse geographic origins among the A. dauci subcollection. For this analysis, eight additional isolates representing five closely related species (Alternaria tomatophila, A. porri, A. solani, A. zinniae and A. bataticola) were also included. They belong to the same porri species group as A. dauci, as defined by Hong et al. (2005a). Phylogenetic analysis of EF-1α sequences showed a tight clustering of all the A. dauci isolates, apart from the other tested Alternaria species (Fig. 1). Similar results were obtained by phylogenetic analysis of ITS sequences, although the discrimination between Alternaria species different from A. dauci was lower than with EF-1α analysis (results not shown). The morphological and molecular criteria described above provide evidence that the 27 selected isolates belong to the A. dauci species.

Figure 1.

 Phylogenetic tree reconstructed by the neighbour-joining method from the alignment of EF-1α partial gene sequences of 10 Alternaria dauci isolates and eight Alternaria spp. isolates belonging to five different species within the porri species group. Bootstrap confidence measures greater than 50% from 1000 bootstrap replicates are indicated. The EF-1α sequence of Alternaria alternata EGS 34-016 (GenBank acc. no. AY438647) was used to root the tree.

In order to evaluate isolate diversity in the subcollection, sequence polymorphism in the IGS region of the 27 isolates was investigated. First, the complete IGS sequence (1856 bp) of the reference isolate NED001 was determined. Two large direct repeats (32–64 nt in length) and several smaller ones (9–16 nt in length) were found to be mainly clustered in the central part of the IGS sequence (Fig. 2). Based on previously published data showing that the 5′ part of the IGS region was highly polymorphic at the interspecific level within the genus Alternaria (Hong et al., 2005b), internal IGSup1 and IGSdo1 primers were defined and used to amplify a large portion (c. 1300 bp, including the 5′ part) of the IGS region from isolate FRA019. Comparison of sequences from isolates NED001 and FRA019 revealed at least 40 SNPs and six indels. As the central portion of the IGS region proved difficult to amplify and sequence because of the presence of numerous nucleotide sequence repeats, the more informative 5′ variable domain of the 27 isolates was amplified using IGSup1 and a newly designed primer IGSdo2 (located at position 381 relative to the IGS start). Analysis of the aligned nucleotide sequences revealed that A. dauci isolates were grouped into three clusters, named C1–C3 (Fig. 3). Isolate FRA018 was distinct from all other isolates which fell into three clusters. There was no evidence of clustering according to geographic origin.

Figure 2.

 Schematic representation of the IGS region from Alternaria dauci. Fine black lines indicate flanking 26S and 18S rDNA sequences. Thick black lines correspond to the IGS sequence. Hatched rectangles indicate direct repeats. Arrows indicate the position of PCR primers (IGSup1, IGSdo1 and IGSdo2) used to amplify a portion of the variable domain. The positions of three unique restriction sites are shown: N = NheI, H = HindIII, X = XhoI.

Figure 3.

 (a) Dendrogram reconstructed by the neighbour-joining method from the alignment of the 5′ variable domain from IGS of 27 Alternaria dauci isolates. Bootstrap confidence measures greater than 50% from 1000 bootstrap replicates are indicated. Alternaria tenuissima was used as an outgroup to root the tree. Three main A. dauci clusters, named C1–C3, are shown. Asterisks correspond to the isolates used for the A. dauci aggressiveness study. (b) Schematic representation of the 5′ variable domain from IGS showing sequence deletions (white rectangles) in isolate FRA018, C1 and C3 patterns.

Assessment of A. dauci aggressiveness

For the A. dauci aggressiveness study, isolates were selected from each cluster (C1, C2 and C3) in order to represent the IGS sequence polymorphism. Another criterion for isolate selection was the capacity to produce high numbers of conidia on V8 juice agar medium for inoculum preparation. Eleven of the 27 isolates were retained for further analysis of aggressiveness (four isolates from cluster C1, three from cluster C2, three from cluster C3 and isolate FRA018, see Fig. 3). The aggressiveness of A. dauci was assessed in greenhouse conditions using the 11 isolates inoculated on the susceptible carrot cv. Presto. For each isolate studied, typical dark brown necrotic lesions showing a sharp or diffuse outline with a yellow halo or no halo were observed on leaves at 8 and 13 d.p.i. The mean lesion number (expressed as log10 (LN)), mean necrotic leaf area (NLA) and mean disease index (DIa) are shown in Table 3. For the same criterion used to assess aggressiveness, the results suggest continuous variation in the values and no clear division into isolate classes. However, using mean LN, some isolates could be significantly differentiated from others: (i) ITA002 and FRA018 were the most and the least aggressive isolates, respectively, and (ii) there were four mid-range isolates: FRA016, which was less aggressive than ITA002; and FRA001, GER001 and AUS001, which were more aggressive than FRA018. On the basis of mean NLA and mean DIa, FRA012 and AUS001 were the most and the least aggressive isolates, respectively. Mean NLA data (i) distinguished the moderately aggressive isolate FRA016 from isolates FRA012, AUS001 and FRA018, and (ii) identified isolate FRA018, together with AUS001, as the least aggressive isolates in the test. Mean DIa values for isolates JPN002, HUN002 and FRA005 were significantly different from those for FRA012 and AUS001. These three isolates were mid-range in aggressiveness.

Table 3.   Assessment of aggressiveness of 11 Alternaria dauci isolates on the susceptible carrot cv. Presto using different criteria based on lesion number or necrotic leaf area
A. dauci isolatelog10 (LN)aMean NLAbMean DIac
  1. Data are means calculated from five leaves. Numbers followed by the same letter are not significantly different.

  2. aMean lesion number per leaf (expressed as log10 (LN)) at 8 and 13 days post-inoculation.

  3. bMean necrotic leaf area at 13 days post-inoculation.

  4. cMean disease index based on a 0–5 scale (0: no visible damage, to 5: > 60% leaf area damaged) at 13 days post-inoculation.

ITA0022·48 a15 ab2·2 ab
FRA0122·38 ab21·6 a2·53 a
JPN0022·31 ab14·8 ab2 bc
HUN0022·29 ab10·2 bcd2 bc
FRA0162·14 bc10·7 bc1·7 bcde
FRA0051·97 cd8·7 bcd1·89 bcd
FRA0171·95 cd5·5 cd1·5 cde
FRA0011·77 d6·9 cd1·7 bcde
GER0011·76 d7·1 cd1·6 cde
AUS0011·74 d2·8 d1·2 e
FRA0181·47 e2·8 d1·4 de

Alternaria dauci host range

After inoculation of a panel of Apiaceae and non-Apiaceae species with A. dauci, symptoms were observed on all tested plant species (Fig. 4 and Table 4), except leek and sweetcorn. Brown necroses with or without a halo developed on all cultivated and wild Apiaceae species. Leaf blight symptoms were very distinctive on the fine foliage of dill and fennel. On non-Apiaceae species, brown necroses were observed on radish, tomato and corn salad, while dark lesions were also present on radish, tomato, corn salad and cress. At 21 d.p.i., a moderate to relatively high disease index (DIv) was recorded on wild Daucus species, carrot, chervil, coriander, dill and fennel inoculated with isolate FRA017 or FRA001: the range was 3–8·3, covering 75% of the index scale. By comparison, celeriac, celery, parsley and parsnip showed lower DIv values (0·7–2). On non-Apiaceae species, DIv values remained very low in tomato and corn salad hosts (DIv = 1) and were only slightly higher in radish or cress (DIv between 1 and 2·3). In general, the disease indexes obtained with isolates FRA017 and FRA001 were quite similar within hosts. Across hosts, lesion diameter increased between 4 and 25 d.p.i. in carrot, dill, fennel and wild Daucus species, while it remained constant in other studied cultivated Apiaceae and non-Apiaceae species. In summary, the plant species could be divided into four categories on the basis of disease index values and necrosis patterns: (i) species exhibiting a relatively high disease index (≥ 3) with expanding lesions: carrot, dill, fennel and wild Daucus species; (ii) species also showing a relatively high disease index but with non-expanding lesions: chervil and coriander; (iii) species exhibiting a relatively low disease index (≤ 2) with non-expanding lesions: celeriac, celery, parsley, parsnip, corn salad, cress, radish and tomato; and (iv) symptomless species: leek and sweetcorn.

Figure 4.

 Lesions (circles) observed after inoculation of different Apiaceae or non-Apiaceae plant species with A. dauci. Cultivated Apiaceae species observed at 25 days post-inoculation (d.p.i.) with isolate FRA001. (a) Carrot, (b) celeriac, (c) celery, (d) chervil, (e) coriander, (f) dill, (g) fennel, (h) parsley, (i) parsnip. Wild Daucus species observed at 25 d.p.i. with isolate FRA001. (j) Daucus carota sg. carota ssp. carota (WD1), (k) D. carota sg. carota (WD2), (l) D. carota sg. carota ssp. carota var. mauritanicus (WD3), (m) D. carota sg. gummifer ssp. gummifer var. fontanesii (WD4), (n) D. carota sg. gummifer (WD5), (o) D. bicolor (WD6). Non-Apiaceae species observed at 4 d.p.i. with the FRA017 isolate. (p) Corn salad, (q) cress, (r) radish, (s) tomato.

Table 4.   Effect of Alternaria dauci on the development of blight symptoms in a greenhouse on different plant species belonging to the Apiaceae or other families and fungal sporulation on leaves
Plant speciesIsolate FRA017Isolate FRA001
SymptomsDisease indexaChanges in necrotic lesion diameterbSporulationcSymptomsDisease indexChanges in necrotic lesion diameterSporulation
  1. aMean disease index (DIv) at 21 days post-inoculation rated on a 0–9 scale (0 = no symptoms; 9 = completely destroyed foliage).

  2. bLesion diameter increased (+) or remained constant (−) between 4 and 25 days post-inoculation.

  3. cPresence (+) or absence (−) of conidiophores and conidia on detached leaves.

CarrotBrown necrosis with yellow halo5++Brown necrosis with yellow halo7++
ChervilBrown necrosis with or without yellow halo; yellow or purple basal leaves4·7+Brown necrosis with or without yellow halo; yellow or purple basal leaves4·3+
CorianderBrown necrosis4+Brown necrosis3·7+
DillBrown necrosis with or without yellow halo; yellowing of basal leaves5++Brown necrosis with or without yellow halo; yellowing of basal leaves4++
FennelBrown necrosis; yellowing of basal leaves3++Brown necrosis; yellowing of basal leaves4++
ParsleyBrown necrosis1Brown necrosis1·3+
ParsnipBrown necrosis with yellow halo1+Brown necrosis with yellow halo2+
CeleriacBrown necrosis0·7+Brown necrosis1+
CeleryBrown necrosis1+Brown necrosis1·3+
WD1Brown necrosis with yellow or red halo; yellowing of leaves6·7++Brown necrosis with yellow halo; yellowing of leaves8·3++
WD2Brown necrosis with yellow or red halo6·3++Brown necrosis with yellow or red halo5·7++
WD3Brown necrosis with yellow or red halo7++Brown necrosis with yellow or red halo6·3++
WD4Brown necrosis with yellow or red halo; yellowing or reddening of leaves7++Brown necrosis with yellow halo; yellowing or reddening of leaves7·7++
WD5Brown necrosis with yellow or red halo; yellowing or reddening of leaves6·3++Brown necrosis with yellow or red halo; reddening of leaves5·7++
WD6Completely destroyed foliage8·3++Brown necrosis with yellow halo7·7++
Corn saladBrown necrosis with or without diffuse yellow halo or dark spots1+Brown necrosis with or without diffuse yellow halo or dark spots1+
CressDark spots1+Dark spots2+
RadishBrown necrosis with yellow halo or dark spots2+Brown necrosis with yellow halo or dark spots2·3+
TomatoBrown diffuse necrosis with yellow halo or dark spots1+Brown diffuse necrosis with yellow halo or dark spots1+
LeekNo symptoms0No symptoms0
SweetcornNo symptoms0No symptoms0

Large densities of conidiophores and conidia were produced on detached leaves showing necroses in all plant species (two examples are presented in Fig. 5), except on parsley inoculated with FRA017, leek and sweetcorn. The fungus was successfully reisolated from all Apiaceae and non-Apiaceae species, except from cress inoculated with FRA017, and leek and sweetcorn, where reisolation attempts were unsuccessful. In all cases, A. dauci was identified on the basis of conidial morphology and specific PCR amplification of IGS sequences (results not shown). Genotyping of reisolated isolates using two polymorphic microsatellite markers (loci Admic7 and Admic8, see Benichou et al., 2009) confirmed the presence of the inoculated isolate, i.e. FRA017 or FRA001. The results obtained using locus Admic7 are shown in Figure 6.

Figure 5.

 Production of Alternaria dauci conidiophores and conidia on the upper side of detached leaves in (a) carrot and (b) radish. The leaves were collected at 25 days post-inoculation with isolate FRA017 and incubated as described in the Materials and methods. Scale bar = 100 μm.

Figure 6.

 Individual isolate genotyping after reisolation of Alternaria dauci from symptoms developed on cultivated Apiaceae (CA), non-Apiaceae (NA) or wild Daucus species (W). Genotyping was performed to identify isolate FRA017 (lanes 1–21) or FRA001 (lanes 22–43) using a specific microsatellite marker (locus Admic7). Band sizes indicated on the left side of the gel correspond to the size marker. Lanes 1 and 22, Daucus carota sg. gummifer ssp. gummifer var. fontanesii (WD4); 2 and 23, D. bicolor (WD6); 3 and 24, D. carota sg. carota ssp. carota (WD1); 4 and 25, D. carota sg. carota (WD2); 5 and 26, D. carota sg. gummifer (WD5); 6 and 27, D. carota sg. carota ssp. carota var. mauritanicus (WD3); 7, 11, 20, 28, 33 and 42, carrot; 8 and 29, corn salad; 9 and 30, radish; 10 and 31, tomato; 32, cress; 12 and 34, parsley; 13 and 35, celery; 14 and 36, coriander; 15 and 37, chervil; 16 and 38, parsnip; 17 and 39, fennel; 18 and 40, celeriac; 19 and 41, dill. FRA017 and FRA001 controls correspond to DNA extracted from reference fungal cultures of the relevant isolates.

In a next step, the virulence of THA003 isolated from tomato seed was tested on carrot (cv. Presto) and tomato in greenhouse conditions. Isolate FRA017 was used as a control. Symptoms observed after inoculation with isolate THA003 were identical to those induced by FRA017: brown necroses with a yellow halo were observed in carrot, while brownish–grey necroses with a yellow halo were present on tomato leaves. For both isolates, the disease index was high in carrot (DIv = 7 and 7·7 with THA003 and FRA017, respectively) and low in tomato (DIv = 2 for both isolates). Typical A. dauci conidia were produced in each case, while reisolation of the fungus followed by species diagnosis and individual genotyping allowed confirmation that isolate FRA017 or THA003 isolate was responsible for the lesions observed.


To study variations in A. dauci aggressiveness, isolates were selected to represent maximum diversity with regard to their origins and IGS patterns. Several studies have already highlighted the relevance of IGS sequence analyses for studying intraspecific diversity in fungi (Risède & Simoneau, 2002; Pecchia et al., 2004). Within the genus Alternaria, Hong et al. (2005b) previously showed, using restriction mapping of the IGS region, that closely related species could be efficiently resolved. In agreement with that study, the present results confirmed that, based on nucleotide sequence analysis, the IGS region of Alternaria spp. could be divided into a conserved domain located at the 3′ end and a highly variable domain at the 5′ end. Comparison of sequences from this variable domain revealed intraspecific polymorphism within the species A. dauci and allowed three major groups of isolates to be defined. Therefore, the IGS 5′ domain may be used to structure intraspecific diversity: this was applied here in order to aid isolate selection for the aggressiveness study. The 11 A. dauci isolates selected in this study for pathogenicity assays were representative of the three clusters and also exhibit polymorphism at microsatellite markers (Benichou et al., 2009). In this study, a significant variation in aggressiveness was observed independently of the criterion used, but the lesion number index criterion represented a more objective disease scoring evaluation than criteria based on percentages of necrotic leaf area. No correlation between IGS clustering (Fig. 3) and the aggressiveness of isolates (Table 3) was obtained. In agreement with the present work, Rogers & Stevenson (2010) showed, using a foliar and a petiole disease index, that the aggressiveness of 22 A. dauci isolates from six carrot-growing regions in the northern USA varied significantly. Diversity within A. dauci in terms of aggressiveness can thus be found at very different spatial levels, i.e. worldwide and in more restricted areas. The characterization of A. dauci isolates exhibiting different aggressiveness levels will be of great interest for the selection of carrot genotypes showing different levels of resistance to ALB, e.g. when considering the choice of adequate inoculum for plant resistance screening. In order to find new sources of partial resistance, a moderately aggressive isolate may be used, while a very aggressive one would be appropriate for the final evaluation of hybrids carrying several resistance factors. It is also crucial to have substantial knowledge of aggressiveness levels when conducting some epidemiological studies: for example, it would be important to evaluate the capacity of several A. dauci isolates, varying in aggressiveness, to infect carrot umbels and colonize seeds. In future experiments, A. dauci aggressiveness should be assessed under field conditions, since differential responses in terms of pathogen aggressiveness may be less detectable under optimal environmental conditions (Pariaud et al., 2009).

The host range of A. dauci was studied in controlled conditions using 17 plant species belonging to six different botanical families inoculated with fungal isolates FRA001 or FRA017, previously shown to be moderately aggressive. Both isolates, which were used in a recent study to evaluate different carrot resistance characterization methods (Boedo et al., 2010), belong to the same IGS cluster (C1). A complementary study using three different A. dauci populations of field isolates from two locations in France (a total of 115 isolates) revealed that the IGS pattern of the C1 type was highly represented (c. 50%) in these fungal populations (S. Benichou, Université d'Angers, unpublished data).

In the present study, A. dauci was able to infect and sporulate on all dicot plant species (Apiaceae and non-Apiaceae), while the two monocots tested (leek and sweetcorn) were confirmed as non-host plants. Two main categories of plant species were differentiated on the basis of expanding lesions (cultivated and wild carrot, dill and fennel) or non-expanding lesions (other dicot species studied). Potentially, the first category could correspond to ‘main’ hosts for A. dauci and the second category to ‘alternative’ hosts. In future experiments, the A. dauci host range should be studied using more aggressive or less aggressive isolates than FRA001 or FRA017 to determine if host spectrum may vary with isolate aggressiveness. The results obtained in controlled conditions, which are very favourable for the development of the pathogen, should be evaluated under field conditions. Under natural conditions, non-carrot species showing A. dauci lesions should be considered as an inoculum reservoir if fungal sporulation is confirmed in the field.

In agreement with the present results, it was previously reported that A. dauci can infect coriander (Felix & Orieux, 1963), celery, celeriac, parsnip and parsley (Joly, 1964; Neergaard, 1977). Bulajic et al. (2009) recently demonstrated that A. dauci can cause reductions in carrot, parsley, parsnip and celery seed emergence. The present results suggest that A. dauci host range could, at least partly, depend on the phylogenetic proximity between the plant species within the Apiaceae: for example, a relatively high disease index was assessed on chervil, which belongs to the tribe Scandiceae like carrot, whereas a low index was observed on parsnip, which is in a distinct tribe (Downie et al., 2010). In other respects, Soteros (1979) indicated that A. dauci was present on wild Daucus and wild Pastinaca species collected near cultivated carrot plots. The fungus was isolated from tissue with symptoms on three wild Apiaceae species: Daucus maximus, Ridolfia segetum and Caucalis tenella (Netzer & Kenneth, 1969). The fact that the wild Daucus species studied here were very susceptible to A. dauci stresses the need to regularly inspect carrot field edges, as wild Daucus species could represent important inoculum sources. Another aspect of A. dauci control would consist of better management of carrot fields, including the choice of previous crops used in rotations. Apiaceae species as carrot precedents must be considered as a risk factor for A. dauci development: it was previously demonstrated that volunteer plants could indeed be found growing for at least 8 years after harvest and that A. dauci could thrive on decaying plant debris for several years (Carette & Jolin, 1992; Pryor et al., 2002).

In this study A. dauci lesions were observed in four non-Apiaceae species, belonging to the Brassicaceae, Solanaceae and Valerianaceae. Joly (1964) and Strandberg (1992) reported the artificial transmission of A. dauci to plantlets of species in different families, including the Asteraceae (lettuce), Solanaceae (tomato, tobacco, aubergine), Brassicaceae (radish, cabbage) and Caryophyllaceae (carnation). Soteros (1979) noted the presence of A. dauci on Fumaria muralis, a wild Papaveraceae species. Remarkably, isolate FRA046 used in the present work was isolated from chicory (leaf) grown in an experimental plot situated next to a carrot experimental field at the GEVES (Groupe d’Etude et de Contrôle des Variétés et des Semences, Brion, France) research station; FRA047, which was also included in this study, originated from this naturally infested carrot plot (V. Cadot, GEVES Beaucouzé, France, personal communication). These two isolates are genetically different, as they belong to different IGS clusters, i.e. C3 and C1, respectively.

FRA019 and THA003, also used in the present work, were isolated from cress and tomato seeds, respectively. No further information is available concerning isolate FRA019. Seven A. dauci isolates originating from tomato seeds produced in the same plot in Thailand, including THA003, are present in the worldwide A. dauci collection used in this study. Molecular characterization of the seven isolates revealed polymorphism in IGS sequences and at microsatellite loci (S. Benichou, unpublished data), indicating that more than a single fungal genotype was collected from tomato seeds. It is currently not known whether the presence of A. dauci on cress or tomato seeds is the result of postharvest contamination or if the fungus was originally transmitted from its host plant to seeds. Further work is needed to shed light on this point. In the present study, no difference in aggressiveness between THA003 and FRA017 isolated from carrot leaf was observed on carrot or tomato, suggesting an absence of isolate specialization on hosts. A. dauci is a carrot seedborne pathogen, which can also be potentially transmitted by seeds of other Apiaceae species such as parsley, celery (Strandberg, 1992) and perhaps non-Apiaceae species (this study). It is very likely that the international seed trade is at least partly responsible for the worldwide spread of A. dauci. The present IGS analysis did not reveal any clustering by geographic origins, possibly because of the long-distance spread of the fungus through contaminated seeds.

In conclusion, A. dauci is a necrotrophic fungal pathogen with some diversity in terms of aggressiveness which could reflect a relatively high genetic variability for this mitosporic haploid fungus. A significant level of DNA polymorphism has been previously described using RAPD, ISSR (Rogers & Stevenson, 2006) and SSR (Benichou et al., 2009) markers and in this study using IGS. The virulence data showed that the host range of A. dauci studied in controlled conditions is not restricted to cultivated carrot or wild Daucus species. The results strongly suggest that several non-carrot species could constitute alternative hosts. The lifestyle of A. dauci could be midway between necrotrophic fungal species showing a limited host spectrum (such as Cochliobolus and Alternaria spp. producing host-selective toxins) and those that are clearly described as being polyphagous with a broad host range, such as Botrytis cinerea or Sclerotinia sclerotiorum (van Kan, 2006). As aggressiveness and virulence may correspond to variable pathogenicity components in A. dauci, both factors should be considered in the future to improve disease control methods.


The authors would like to thank A. Suel, S. Huet and B. Hamon for their technical assistance, R. Gardet and J. Granger for their help in the greenhouse experiments, D. Manley for reviewing the English. Dr E. Geoffriau (Agrocampus Ouest Centre d’Angers) is also gratefully acknowledged for his help in the determination of wild Daucus species. This work was financed by the French General Board for Companies (‘Création Variétale Potagère’ FUI project, 2007–2010). C. Boedo and S. Benichou were granted doctoral fellowship by ANRT (CIFRE)/Clause Vegetable Seeds and by the Algerian Government, respectively.