The use of herbicides for weed control is very common, but some of them represent a threat to human health, are environmentally detrimental, and stimulate herbicide resistance. Therefore, using microorganisms as natural herbicides appears as a promising alternative. The mycoflorae colonizing different species of symptomatic and asymptomatic weeds were compared to characterize the possible mycoherbicidal candidates associated with symptomatic weeds. A collection of 475 symptomatic and asymptomatic plants belonging to 23 weed species was established. A metabarcoding approach based on amplification of the internal transcribed spacer (ITS) region combined with high-throughput amplicon sequencing revealed the diversity of fungal communities hosted by these weeds: 542 fungal genera were identified. The variability of the composition of fungal communities revealed a dispersed distribution of taxa governed neither by geographical location nor by the botanical species, suggesting a common core displaying nonspecific interactions with host plants. Beyond this core, specific taxa were more particularly associated with symptomatic plants. Some of these, such as Alternaria, Blumeria, Cercospora, Puccinia, are known pathogens, while others such as Sphaerellopsis, Vishniacozyma, and Filobasidium are not, at least on crops, and constitute new tracks to be followed in the search for mycoherbicidal candidates.
IMPORTANCE This approach is original because the diversity of weed-colonizing fungi has rarely been studied before. Furthermore, targeting both the ITS1 and ITS2 regions to characterize the fungal communities (i) highlighted the complementarity of these two regions, (ii) revealed a great diversity of weed-colonizing fungi, and (iii) allowed for the identification of potential mycoherbicides, among which were unexpected genera.


The negative impact of weeds on agricultural production and on many an ecosystem service no longer needs to be demonstrated (1, 2). According to an international survey carried out over 3 years, weeds are responsible for an estimated loss of yield ranging between 32 and 34% in a dozen crops depending on crop and region, and this impact is observed worldwide (37). Weeds are also putative reservoirs for pests or pathogens that alter harvests (8). Furthermore, invasive weeds disrupt existing balances as well as the natural diversity of preexisting vegetation and of the associated fauna, while this fauna includes insects that are valuable auxiliaries for agriculture and more generally for the environment (9). Moreover, weeds such as ragweed are particularly allergenic and cause health problems for humans (10, 11). Weed management requires implementing numerous chemical and mechanical means, whose impact on the environment is far from neutral. Synthetic herbicides, whose residues persist in the environment, affect biological resources other than the initial target; they can also lead to resistant weed populations (1214). Mechanical weeding has a globally negative carbon balance because it causes too rapid organic matter mineralization and increases the carbon footprint (15, 16). Conversely, weeds participate in ecosystem services in agricultural landscapes, particularly by attracting and feeding pollinating insects (17, 18). Yet, they globally have a negative image that has restricted interest in weed-microorganism interactions compared to the numerous studies dealing with interactions between so-called noble plants (crops) and microorganisms. However, weeds, like cultivated plants, harbor both epiphyte, endophyte, and rhizosphere microflorae including fungi and fungus-like organisms, among which are pathogenic fungi that appear as potential mycoherbicides (1921). The use of bioherbicides is indeed one alternative method to the use of synthetic phytopharmaceutical products for weed management in addition to allelopathy, as well as granivory and agroecological cropping systems (18, 22, 23). The development of bioherbicides starts with the isolation of microorganisms associated with symptoms observed on weeds posing a real management problem locally (for a given crop or a given region) (2427). A recent literature review showed that this approach resulted in an increasing number of publications proposing fungal candidates that, in theory at least, perform well on one or more weed species. Unfortunately, too few of these candidates are already available on the market, but the process must be encouraged and pursued (28). Although very pragmatic, the publication of scientific data about these candidates only gives too partial a view of the microbial potential hosted by weeds. High-throughput sequencing (HTS) provides access to the diversity of microorganisms present in complex environments and, in particular, to the diversity of weed endophytic fungi (2931). Actually, it would be more correct to speak of microorganisms colonizing plants than endophytes stricto sensu when their study is based on molecular methods that do not specifically target plant internal tissues or when microorganisms have not been isolated from internal tissues. Indeed, the presence of DNA of epiphytic microorganisms on the surface of plant organs cannot be excluded despite surface disinfection with ethanol, NaOCl, and/or H2O2. Such a targeted approach has been successful in identifying microorganisms associated with dieback occurrence in the invasive leguminous tree Parkinsonia aculeata (parkinsonia) (32, 33). More and more publications highlight the interest of focusing on endophytes (3436). Indeed, plant-colonizing microorganisms, including endophytes, ensure a continuum between mutualism and pathogenicity. They provide services to the host weed, such as pest protection or resistance to abiotic stresses, but they can also be pathogenic to the host weed, with limited symptomatic infections in natural populations (37, 38). Such is the case for fungi belonging to the genus Epichloë when they interact with grasses of the Poaceae family (3941). Although the evolutionary origin of endophytic fungi is not proven, it cannot be excluded that single-gene mutations could switch an endophyte to a pathogen and vice versa (42). Weed-colonizing fungi have been poorly studied, unlike those of cultivated plants such as wheat, soybean, maize, common bean, and grapevine, which are more interesting from an economic viewpoint (4349). Analyzing the taxonomic composition of weed-colonizing fungal communities should provide a source of information for weed management. In addition to acquiring knowledge about the diversity of weed-colonizing fungi of various origins, HTS may provide indications on the occurrence of pathogens acting as potential mycoherbicides (35).
Weeds obviously appear as an unexplored reservoir of fungi among which some are possible mycoherbicides. Using an HTS-based metabarcoding approach targeting the ribosomal internal transcribed spacer 1 (ITS1) and ITS2 regions, this study was aimed at (i) deciphering the diversity of fungi colonizing symptomatic and asymptomatic weeds in four French areas, and (ii) identifying fungi associated with symptomatic weeds likely to be potential mycoherbicides.


Global analysis, diversity, and richness.

Fungal communities colonizing symptomatic (S) and asymptomatic (AS) weeds belonging to 23 weed species were characterized (see Table S1 in the supplemental material). S weeds exhibited various symptoms including necrosis, chlorosis, discoloration, rust-like symptoms, or white mold symptoms (Fig. 1). This global analysis included all samples from 23 weed species originating from four French areas. The ITS1 region was successfully amplified for 466 samples out of the 475 initial samples. A total of 13,973,652 high-quality sequences were recovered after quality filtering and assigned to 2,510 operational taxonomic units (OTUs). Among these sequences, 1,207,541 of them represented by 60 OTUs were plant sequences. Their relative abundance was 3.7% in S weeds and 14.6% in AS weeds. The ITS1 data set also contained 0.06% of sequences identified as Oomycota. The ITS2 region was successfully amplified for 451 samples out of the 475 initial samples. A total of 13,175,803 high-quality sequences were recovered and assigned to 2,908 OTUs. The proportion of plant sequences was much higher than in the ITS1 data set, with 3,571,569 sequences represented by 1,030 OTUs, and relative abundances of 24.5% and 29.9% in S weeds and AS weeds, respectively. Overall, the ITS2 data set (S and AS weeds) contained 27.1% of plant sequences, with a variable number of plant sequences/weed species, while the ITS1 data set contained 8.6% of plant sequences.
FIG 1 Different symptoms observed on the sampled symptomatic weed species. Symptoms were identified in different groups as follows: black necrosis (blN), brown felting (brF), brown necrosis (brN), chlorosis (C), orange pustule (oP), pink necrosis (pN), powdery mildew (pM), rust (R), and white pustule (wP).
The total number of detected fungal OTUs was 2,450 in the ITS1 data set and 1,876 in the ITS2 data set (see Table S2 in the supplemental material). Rarefied data represented 435 and 353 weed samples in the ITS1 and ITS2 data sets, respectively. The ITS1 data set included 246 and 189 S and AS weed samples, respectively. The ITS2 data set included 180 and 173 S and AS weed samples, respectively. Symptomatic weeds harbored lower fungal richness than asymptomatic weeds. Thus, the median was 107 OTUs per sample in S weeds and 119 OTUs per sample in AS weeds in the ITS1 data set and 122 OTUs per sample in S weeds and 135 OTUs per sample in AS weeds in the ITS2 data set. Furthermore, the number of OTUs per modality was significantly higher in AS weeds than in S weeds in both the ITS1 and ITS2 data sets (Wilcoxon test, P value = 1.05e−03 and P value = 9.53e−03, respectively).
The evenness index was significantly lower in S samples than in AS samples in both ITS data sets. In the ITS1 data set, the median value was 0.47 for S samples versus 0.56 for AS samples (Wilcoxon test, P value = 2.51e−18). In the ITS2 data set, it was 0.43 versus 0.47 (Wilcoxon test, P value = 1.26e−04).

Fungal community composition.

In the ITS1 data set, members of the Ascomycota phylum accounted for 69.2% and 55.9% of the total number of fungal sequences in S and AS weeds, respectively. Basidiomycota was the second most abundant phylum, with relative abundances of 30.6% and 43.9% in S and AS weeds, respectively. A total of 1,665 OTUs were assigned to Ascomycota and 671 to Basidiomycota. The proportions of all other phyla (Chytridiomycota, Mortierellomycota, Mucoromycota, and Rozellomycota) were 0.03% and 0.05% in S and AS weeds, respectively. Finally, 0.2% of the fungal sequences were not identified at the phylum level. The proportions of fungal sequences affiliated to the different phyla were significantly different in S and AS weeds (X squared = 240,900; df = 3; P value < 2.2e−16).
Ascomycota sequences in the ITS1 data set were identified as members of the orders Capnodiales (13.5% of the total fungal sequences), Erysiphales (3.8%), Helotiales (2%), Hypocreales (4.4%), Pleosporales (35%), and Xylariales (2.3%) (see Fig. S1A in the supplemental material). The other Ascomycota orders represented less than 2% of the sequences. Within Basidiomycota, Filobasidiales, Sporidiobolales, and Tremellales were the most abundant orders, with relative abundances of 3.5%, 8.5%, and 11.3% of the total fungal sequences, respectively.
Regarding Ascomycota, similar relative abundances of S and AS weeds were found for the orders Capnodiales (13% and 14.2% in S and AS weeds, respectively) and Helotiales (1.8% and 2.3%). Pleosporales were more abundant in S weeds (40%) than in AS weeds (28%) (Fig. S1A). Alternaria and Cladosporium were the most abundant genera, with relative abundances of 10.2% to 18% in S and AS weeds, respectively, for Alternaria, and 9.3 to 12.5% in S and AS weeds, respectively, for Cladosporium (Fig. 2A). The Blumeria genus was detected more in S weeds (2.8%) than in AS weeds (0.06%). Similarly, Neoascochyta and Sphaerellopsis were more abundant in S weeds than in AS weeds. Epicoccum and Fusarium were detected with similar abundances of about 2.6% and 2.4% in S and AS weeds, respectively (Fig. 2A).
FIG 2 Relative abundances of different fungal genera detected in the ITS1 (A) and ITS2 (B) data sets, expressed as percentages of total sequences. S, symptomatic weeds; AS, asymptomatic weeds.
Regarding the Basidiomycota phylum, the dominant order Tremellales was more abundant in AS weed samples (13.3%) than in S weed samples (9.8%). Overall, all orders belonging to Basidiomycota were more abundant in AS weed samples than in S weed samples. For example, the relative abundance of Filobasidiales was 4.6% in AS weeds and 2.8% in S weeds (Fig. S1A). Similarly, Basidiomycota genera were more abundant in AS weeds than in S weeds. Thus, genera such as Bullera, Filobasidium, Sporobolomyces, and Vishniacozyma had relative abundances of 4.8%, 4.6%, 10.3%, and 7% in AS weeds and 2.7%, 2.7%, 6.8%, and 5.8% in S weeds, respectively (Fig. 2A).
In the ITS2 data set, members of the phylum Ascomycota accounted for 67.5% and 62.5% of the total number of fungal sequences in the S and AS samples, respectively. The Basidiomycota phylum represented 32.4% and 37.3% of relative abundance in the S and AS weed samples, respectively. A total of 1,202 OTUs were assigned to Ascomycota and 601 to Basidiomycota. The proportions of Chytridiomycota, Mortierellomycota, Mucoromycota, and Rozellomycota were 0.03% and 0.02% in S and AS weeds, respectively. The unidentified fungi at the phylum level accounted for relative abundances of 0.07% and 0.2% in S and AS weeds, respectively. As for the ITS1 analysis, the proportions of fungal sequences in the different phyla were significantly different in S and AS weeds (X squared = 28,428; df = 3; P value < 2.2e−16).
Ascomycota sequences were mainly identified as belonging to Capnodiales (19% of the total fungal sequences), Erysiphales (5%), and Pleosporales (29.5%) (Fig. S1B). Helotiales (2.9%), Hypocreales (3.1%), and Xylariales (2.5%) represented smaller groups. Six other orders were represented by less than 2% of the sequences. Within Basidiomycota, the orders Pucciniales, Sporidiobolales, and Tremellales were dominant, with relative abundances of 10.4%, 5.5%, and 11%, respectively (Fig. S1B). It is worth noting that Pucciniales was one of the most abundant orders among Basidiomycota in the ITS2 data set, while it was represented by only 0.02% of the sequences in the ITS1 data set.
Regarding Ascomycota, Capnodiales were more abundant in the ITS2 data set than in the ITS1 data set (Fig. S1B). The members of this order showed similar abundance in S weeds (18.7%) as in AS weeds (19.5%). As in the ITS1 data set, Erysiphales were more abundant in S weeds (6.5%) than in AS weeds (3.4%). Helotiales were more represented in S weeds (3.8%) than in AS weeds (1.8%), whereas this order was more abundant in AS weeds in the ITS1 data set. Pleosporales were equally abundant in AS and S weeds (approximatively 29%), whereas this order was more abundant in S weeds in the ITS1 data set. Greater diversity was observed for fungi of the Pleosporales order in the ITS2 data set as in the ITS1 data set. Alternaria (12.5 and 9.5%) and Cladosporium (15.7 and 11.7%) were also dominant genera in S and AS weeds (Fig. 2B). Erysiphe was a relatively common genus in S weeds (3%), while it was detected at only 0.1% relative abundance in AS weeds. The Sphaerellopsis genus was relatively abundant in S weeds (2.9%) and present at 1.5% in AS weeds. Neoascochyta was a dominant genus in S and AS weeds, with relative abundances of 5.4% and 6.8%, respectively. Neosetophoma was not detected in the ITS1 data set but was found in the ITS2 data set with a relative abundance of 0.6%. It was more abundant in S weeds than in AS weeds. Phaeosphaeria was slightly more abundant in AS weeds than in S weeds. Septoria had a relative abundance of 0.3% in the ITS2 data set, while it was not detected in the ITS1 data set.
Regarding Basidiomycota, Pucciniales was the predominant order in S weeds (12.6% of the total fungal sequences), while Sporidiobolales were more represented in AS weeds (6.8%) than in S weeds (4.3%), and Tremellales were also dominant in AS weeds (13.1%) (Fig. S1B). Apart from the fact that the genus Puccinia was more abundant in S weeds than in AS weeds, and poorly detected by ITS1 analysis, all of the Basidiomycota genera were more abundant in AS weeds than in S weeds, as already observed in the ITS1 data set. Among these genera, Bullera, Filobasidium, Sporobolomyces, and Vishniacozyma occurred in 4.4%, 2.8%, 6.7%, and 5.4% of AS weeds, respectively (Fig. 2B).

Comparison of genus diversity in the ITS1 and ITS2 data sets.

ITS1-based and ITS2-based metabarcoding strategies revealed that quite different but complementary levels of diversity were hosted by each of these two ITS regions. The fungal community compositions obtained from ITS1 and ITS2 analyses were compared at the genus level. A total number of 542 genera was identified using both data sets (Fig. 3A). The cloud on the top right of the graph shows that many genera were identified in both data sets. Conversely, the points aligned in the column on the left of the graph and at the bottom of the graph clearly indicate that many genera were only identified in one ITS data set. Indeed, only 44% of the genera were detected in both data sets (Fig. 3B), hence the need to compile the two analyses. For example, only the primers targeting the ITS1 region detected the genera Ascochyta, Beauveria, Laccaria, Neurospora, and Rhizoctonia. Conversely, only the primers targeting the ITS2 region detected the genera Bannoa, Neosetophoma, Septoria, Sphaerellopsis, Stegonsporium, and Tilletiopsis. Some of the abundant genera were detected in similar relative abundance by the primers targeting the ITS1 and ITS2 regions, e.g., Alternaria (13% and 12% relative abundance in the ITS1 and ITS2 data sets, respectively), Blumeria (1.3% and 1.4%), and Fusarium (2% and 1.8%). Conversely, the genus Puccinia was found to be minor in the ITS1 data set (0.03%), while it was one of the most abundant ones in the ITS2 data set (10.3%). Consequently, the two ITS1 and ITS2 regions did not reveal the same picture of the taxonomic diversity of weed-colonizing fungi and proved complementary for analyzing the whole sample set.
FIG 3 (A) Relative abundances of genera in the ITS1 and ITS2 data sets. (B) Venn diagram indicating the numbers of shared and unique genera between the ITS1 and ITS2 data sets.

Fungal communities in relation with the sampling area and the weed species.

In the ITS1 data set, significant differences between fungal communities from different geographical areas were revealed (permutational multivariate analysis of variance [PERMANOVA], P = 0.001). On the nonmetric multidimensional scaling (NMDS) graph (see Fig. S2A in the supplemental material), the Beauce area differed from the other areas. In the ITS2 data set, PERMANOVA indicated a less significant difference between the fungal communities of the four areas (PERMANOVA, P = 0.015) (Fig. S2B). The fungal communities colonizing the S and AS weeds appeared to be different from one to another. However, their distribution—at the genus level—across weeds belonging to a same or different species and across a same or different plots suggested that the geographic area had no influence on the composition of those colonizing fungal communities. Similarly, the NMDS analysis according to the weed species indicated significant differences between fungal communities associated with different weed species (PERMANOVA, P = 0.001) in the two data sets (see Fig. S3 in the supplemental material). However, this global analysis also showed that several weed species shared a common core fungal diversity. Therefore, a targeted statistical approach per plot and weed species was performed to compare S and AS weeds in order to identify candidate OTUs/species with a positive or a negative effect on the occurrence of pathogenic symptoms.

Fungi associated with symptomatic and asymptomatic weeds.

The fungal OTUs differentially abundant between S and AS weeds were identified for each weed species in each plot (Table 1). The analysis was performed per symptom for the six plots where two different symptoms were observed. Many OTUs assigned to the genus Alternaria were found differentially abundant between S and AS weeds in several situations (weed species and plots) in the two ITS1 and ITS2 data sets. Alternaria-assigned OTUs were found differentially abundant in 21 comparisons between S and AS weeds, including those depicted in both data sets. In addition, several OTUs identified as Alternaria were found in different weed species, in the same or different plots, in either S or AS weeds (Table 1). This is also true for other genera such as Epicoccum and Puccinia. Alternaria was associated with symptoms of well-defined black necrosis type. Thus, in S weeds of the ITS1 data set, one Alternaria OTU detected on Sonchus asper in a southern Burgundy plot stood out with a relative abundance of 87%. Interestingly, the same Alternaria OTU was also associated with S samples of the same weed species but in a plot in Beauce where it accounted for 75% of the sequences. Among Senecio vulgaris colonizing communities, another Alternaria-assigned OTU was dominant and significantly more abundant in S weeds than in AS weeds in plots T and Z in the Beauce area, where it accounted for 53% (plot T) and 76% (plot Z) of the total sequences in S weeds and 0% of the total sequences in AS weeds in these two plots. Other Alternaria-assigned OTUs were found to be more abundant in S samples than in AS samples, in Papaver rhoeas (80%), Amaranthus retroflexus (28%), and Ambrosia artemisiifolia (25%). A few other Alternaria-assigned OTUs were found to be more abundant in AS weeds than in S weeds, e.g., in Alopecurus myosuroides (10.4%), Bromus sterilis (14.2%), and Chenopodium album (19%).
TABLE 1 Genera associated with differentially abundant operational taxonomic units in symptomatic and asymptomatic weeds as revealed by ITS1 and ITS2 analysisa
WeedLocationPlotCode of the symptom observedITS1 analysisITS2 analysis
Elytrigia repensBeauceABblNVishniacozyma (17; 16)NoneNoneNone
brN + CNeoascochyta (19; 8)NoneNoneNone
Southern BurgundyAGblNRhodotorula (12; 65)NoneNoneNone
Northern BurgundyMblN + brN + CMicrodochium (34; 19)
Neoascochyta (15; 8)
Alopecurus myosuroidesSouthern BurgundyQbrN + CNoneNonePuccinia (89; 12)Alternaria (10; 10)
Cladosporium (13; 1)
Claviceps (22; 70)
Neoascochyta (7; 14)
Sporobolomyces (7; 4)
Northern BurgundyLpMBlumeria (60; 40)NoneCladosporium (8; 1)Neoascochyta (40; 8)
Phaeosphaeria (13; 17)
Epicoccum (5; 20)
brFBlumeria (90; 31)NoneAlternaria (12; 10)
Claviceps (11; 70)
Amaranthus retroflexusBeauceVwPNoneNoneNoneCladosporium (17; 1)
Southern BurgundyAEbrNAlternaria (28; 17)
Epicoccum (35; 11)
ObrN + CPhoma (14; 43)Fusarium (14; 78)NoneFusarium (14; 93)
Ambrosia artemisiifoliaSouthern BurgundyAJbrN + pMAlternaria (25; 17)NoneNoneEpicoccum (8; 20)
Pithomyces (9; 23)
Avena fatuaSouthern BurgundyCRPuccinia (6; 252)NoneNoneSphaerellopsis (40; 11)
EblNNoneNonePuccinia (30; 6)Zymoseptoria (35; 16)
Microdochium (6; 15)
Parastagonospora (13; 6,443)
Northern BurgundyKblN + RNoneNoneZymoseptoria (32; 16)Neoascochyta (26; 8)
Bromus sterilisSouthern BurgundyQoPNoneNonePuccinia (93; 24)Alternaria (14; 10)
Pyrenophora (6; 192)
JuraJblN + brN + CNoneParastagonospora (6; 23)NoneAlternaria (5; 10)
Cladosporium (9; 1)
Chenopodium albumBeauceVbrN + CFusarium (14; 58)NoneNoneNone
Southern BurgundyEbrFNoneNoneBlumeria (62; 42)Blumeria (81; 27)
OpNNoneClaviceps (8; 39)NoneAlternaria (19; 10)
Claviceps (6; 70)
Cirsium arvenseSouthern BurgundyBblN + CAlternaria (6; 6)Sporobolomyces (24; 1)
Bullera (8; 5)
Dioszegia (6; 34)
Convolvulus arvensisBeauceADblN + CNoneNoneNoneAlternaria (10; 10)
Cladosporium (25; 1)
UbrNNoneNoneVishniacozyma (8; 3,261)None
Southern BurgundyPblNNoneNoneColletotrichum (6; 105)None
SbrN + CNonePithomyces (30; 26)Bullera (14; 7)
Papiliotrema (11; 92)
Sporobolomyces (16; 4)
Symmetrospora (7; 28)
Cladosporium (21; 1)
Pithomyces (36; 23)
Echinochloa crus-galliSouthern BurgundySblNAcremonium (8; 64)NoneNoneNone
Elytrigia repensBeauceABblNVishniacozyma (17; 16)NoneNoneNone
  brN + CNeoascochyta (19; 8)NoneNoneNone
Southern BurgundyAGblNRhodotorula (12; 65)NoneNoneNone
Northern BurgundyMblN + brN + CMicrodochium (34; 19)
Neoascochyta (15; 8)
Lolium rigidumBeauceABRNoneClaviceps (33; 39)Alternaria (50; 5)None
Southern BurgundyEblNEpicoccum (6; 11)NoneNoneNone
FbrNMicrodochium (40; 19)Parastagonospora (33; 27)Phaeosphaeria (5; 121)Bannoa (42; 97)
RblNRamularia (7; 107)NoneNoneNone
JuraIblN + brN + CSphaerellopsis (49; 14)
Bensingtonia (5; 15)
NonePuccinia (44; 6)Phaeosphaeria (13; 17)
Mercurialis annuaSouthern BurgundyAFbrNCercospora (64; 28)Filobasidium (11; 9)NoneCladosporium (16; 1)
Papaver rhoeasSouthern BurgundyEblNAlternaria (80; 46)Alternaria (7; 6)NonePuccinia (74; 6)
Polygonum aviculareBeauceYRNoneNoneNoneCladosporium (17; 1)
Southern BurgundyAHblNNoneNoneNoneEpicoccum (15; 20)
Alternaria (6; 4,102)
pMNoneNoneErysiphe (41; 173)None
Senecio vulgarisBeauceTCAlternaria (53; 3,365)Cladosporium (6; 2)NoneNone
ZbrN + pMNoneSporobolomyces (12; 1)NoneNone
blNAlternaria (76; 3,365)Sporobolomyces (12; 1)NoneNone
Sinapis arvensisSouthern BurgundyAHpMHoltermanniella (6; 18)Alternaria (5; 6)NoneNone
C + blNFilobasidium (25; 13)Alternaria (5; 6)Filobasidium (5; 83,044)None
Solanum nigrumBeauceXC + brNCladosporium (27; 2)NoneNoneEpicoccum (9; 20)
Vishniacozyma (8; 9)
Sonchus asperBeauceACC + blNAlternaria (75; 7)Alternaria (22; 6)NoneNone
Southern BurgundyRblNAlternaria (87; 7)NoneNoneNone
Taraxacum officinaleSouthern BurgundyAGblNFusarium (13; 20)
Sphaerellopsis (60; 14)
NoneNoneTilletiopsis (5; 50)
Pairwise comparisons were carried out based on S weeds versus AS weeds for each weed species, each plot, and each type of symptom, in each ITS data set. Differentially abundant OTUs with a relative abundance higher than 5% on average in the modality are included. Numbers in parentheses indicate the relative abundance of the OTU followed by the OTU number. Symptoms are shown in Fig. 1. Symptoms were identified in different groups as follows: black necrosis (blN), brown felting (brF), brown necrosis (brN), orange pustule (oP), pink necrosis (pN), powdery mildew (pM), rust (R), and white pustule (wP).
The genus Blumeria was generally poorly detected in S and AS weeds. However, some highly abundant Blumeria-assigned OTUs stood out in four comparisons: in one plot in northern Burgundy, up to 90% of the fungal sequences detected in S Alopecurus myosuroides were associated with a Blumeria OTU. Two different symptoms were observed in this plot, and each symptom was associated with a different OTU of Blumeria. The genus Blumeria includes species responsible for powdery mildew, and the observed symptoms (Fig. 1) strongly resembled those of white molds characteristic of this disease. The other two comparisons for which a Blumeria-assigned OTU stood out were found in Chenopodium album: one Blumeria OTU had a relative abundance of 62% in the fungi community of the S weeds, while another had a relative abundance of 82% in the AS weeds.
Cladosporium OTUs were found differentially abundant in 10 comparisons between S and AS weeds. The most abundant one was associated with symptomatic Solanum nigrum (27% of relative abundance). Another Cladosporium-assigned OTU stood out in S weeds when associated with Alopecurus myosuroides, and in AS weeds in eight comparisons between AS and S weeds. As in the case of Alternaria, Cladosporium-assigned OTUs were present in weeds belonging to different species, including Solanum nigrum, Mercurialis annua, Polygonum aviculare, Alopecurus myosuroides, and Amaranthus retroflexus, which suggests an absence of specificity in this weed-fungus interaction.
The genus Cercospora, which includes weed-pathogenic species, was associated with OTUs whose relative abundance was 64% of all fungal sequences in the ITS1 data set of symptomatic Mercurialis annua, while it was lowly detected or undetected in AS weeds. The observed symptoms (Fig. 1) were of the cercosporiosis type.
In several plots and on several monocotyledonous weeds, Neoascochyta OTUs stood out in both S and AS weeds in both ITS1 and ITS2 data sets. Indeed, the same OTU assigned to Neoascochyta was abundant in S samples of Elytrigia repens in Beauce and northern Burgundy. However, it was also abundant at 40% in AS Alopecurus myosuroides in northern Burgundy. Its presence in AS weeds may correspond to a situation of latency prior to a possible outburst of pathogenic activity.
Analyzing ITS data sets allowed us to detect obligate pathogens such as Puccinia. This was particularly true when analyzing the ITS2 data set. Puccinia OTUs stood out in nine comparisons, eight of which were related to S weeds and one to AS weeds. This genus was also associated with rust-type symptoms (Fig. 1) observed on several weeds in different plots. Puccinia-assigned OTUs significantly stood out in S weeds, with high relative abundances up to 93% in Bromus sterilis in southern Burgundy, 89% in Alopecurus myosuroides in the same plot, and 44% in Lolium rigidum in the Jura area. The relative abundance of one Sphaerellopsis-assigned OTU was 49% in S Lolium rigidum in the Jura area in the ITS1 data set, while another Sphaerellopsis-assigned OTU significantly stood out in AS weeds in southern Burgundy, with a relative abundance of 40% in Avena fatua in the ITS2 data set.
The relative abundance of one Erysiphe OTU was 41% in the ITS2 data set of S Polygonum aviculare in southern Burgundy. The identified symptom (Fig. 1) was powdery mildew type, which is consistent with the nature of the symptoms usually caused by Erysiphe spp. on different weeds.
Statistical analysis revealed differences in the composition of the fungal communities colonizing S and AS weeds. Among the OTU-assigned genera found at a high relative abundance, some were known pathogens likely to be present in AS weeds at more than 10% relative abundance but without causing apparent symptoms at the time the weeds were collected. Alternaria, Blumeria, Microdochium, Neoascochyta, and Zymoseptoria were part of these genera. Those pathogenic fungi were simultaneously present in S and AS weeds: they had already expressed their infectious activity in S weeds and had not (yet) expressed it in AS weeds. This pointed to the fact that statistical analysis had difficulty in distinguishing the fungal genera really associated with the symptoms from those truly associated with asymptomatic weeds. A more detailed analysis of Table 1 also revealed slightly less known fungal genera, such as Dioszegia, Leucosporidium, Holtermanniella, Sphaerellopsis, and Vishniacozyma, associated with OTUs found to be more abundant in S weeds than in AS weeds, i.e., possible mycoherbicidal candidates.


The diversity of weed-colonizing fungi is poorly documented and therefore poorly known. Mukhtar et al. (50) used a microbiological approach to describe the diversity of epiphytic and endophytic microorganisms in some weeds, such as Chenopodium album, Euphorbia helioscopia, Parthenium hysterophorus, and Convolvulus arvensis, commonly found in Pakistan. This study underlined complex interactions between the hosts and their epiphytic and endophytic microflora. It especially showed that endophytic fungal communities were as specific as epiphytic communities, and genera such as Aspergillus, Drechslera, Alternaria, Penicillium and Cladosporium were common to both microflorae. In the present study, a metabarcoding approach targeting the ITS1 and ITS2 regions was used to reveal a maximum level of diversity among fungi associated with the most detrimental weeds in France. The OTU richness observed per weed sample was slightly lower in the ITS1 data set (median values of 107 and 119 OTUs for the S and AS samples, respectively) than in the ITS2 data set (122 and 135 OTUs). Higher richness among soil fungal communities was previously found using ITS86F/ITS4 targeting the ITS2 region compared to using ITS1-F/ITS2 primers targeting the ITS1 region (51). Other comparisons of the ITS1 and ITS2 barcodes also suggested targeting ITS2 rather than ITS1 in fungal metabarcoding analyses (5254).
Although similar total numbers of genera (385 and 389 in the ITS1 and ITS2 data sets, respectively) were detected, only 44% of them were detected in both data sets. Furthermore, differences in the abundance of classes, orders, families, and genera were observed between the two data sets. For example, the primer pair ITS1-F/ITS2 targeting the ITS1 region did not detect the Pucciniales order, while the primer pair ITS86F/ITS4 targeting the ITS2 region did not detect Cantharellales. Thus, the two ITS regions provided different, complementary information. Our results confirm the finding that different pictures of fungal communities may be obtained depending on the ITS region targeted and the primers used (5155). These discrepancies may be due to amplicon length differences, mismatches between primers and targets, or the presence of an intron at the 3′ end of the nuclear small subunit (51, 53, 56, 57). ITS1 and ITS2 barcodes are currently analyzed separately since the read length is a limiting factor when using amplicon sequencing based on Illumina technology.
At the phylum level, Ascomycota were found dominant in the fungal communities whatever the ITS marker (63.6% and 65.2% in the ITS1 and ITS2 data sets, respectively). However, Basidiomycota also represented a significant part of the fungal communities associated with weeds (36.2% and 34.7% in the ITS1 and ITS2 data sets, respectively). By contrast, using the same primers as in our study, Op de Beeck et al. (51) found that only 15% to 17% of the OTUs detected in soil belonged to Basidiomycota. All other fungal phyla accounted for less than 0.1% of the sequences in our study. This very low relative abundance may not be a methodological bias since both primer pairs ITS1-F/ITS2 and ITS86F/ITS4 revealed 3% to 4% of Chytridiomycota and Zygomycota in soil fungal communities (51). Thus, we may conclude that the fungal communities colonizing weeds were mainly composed of Ascomycota and Basidiomycota in our conditions. In our study, the two primer pairs also amplified a substantial amount of plant DNA, especially the primer pair ITS86F/ITS4 that yielded 27.1% of plant sequences. Thus, the primer pair ITS86F/ITS4 originally designed for soil samples lacks specificity when working with plant material. Similar fungal microbiome compositions at the phylum level have been found in different plant species, with dominant Ascomycota followed by still abundant Basidiomycota (52, 58). Conversely, using a conventional cultural method, 97% to 99% of the fungal communities were found to belong to Ascomycota in the toxic weed Stellera chamaejasme (59), in Hevea brasiliensis leaves (60), and in Glycine max leaves (61). The almost total absence of Basidiomycota detected in these studies is likely due to the culture method that favored the selection of fast-growing isolates like Ascomycota. Another relevant factor could be that some of these basidiomycetes have a yeast-like growth form more difficult to detect than those having a mycelial growth, hence a possible disregard.
The use of both ITS1 and ITS2 regions evidenced that a wide variety of fungal taxa inhabit weeds, with 542 genera. Some of these fungi are known to be resident on crops and others had never been reported on weeds before. The presence of the genus Cystofilobasidium had never been reported on weeds, while that of the genus Puccinia had been reported (6264). The genus Cystofilobasidium was detected on Sinapis arvensis, Elytrigia repens, Amaranthus retroflexus, and Convolvulus arvensis. Alpha diversity analysis showed that the composition, abundance, and diversity of fungal communities varied significantly between S and AS weeds. Conversely, other metabarcoding-based studies report similar diversity in fungal communities colonizing symptomatic and asymptomatic plants (65, 66). In our study, AS weeds harbored more diverse fungal communities than S weeds. It is possible that the growth of a pathogenic fungus in weed tissues significantly increases its relative abundance, which masks that of other fungal taxa without their absolute abundance being affected. Another hypothesis would be that the pathogenic fungus grows abundantly in the weed tissues and outcompetes other endophytes, either by exploiting nutrient and spatial resources or by producing fungitoxic secondary metabolites. Consequently, its relative abundance increases at the expense of other colonizing species, which are lowly detected or not detected at all. However, the large diversity or specific richness observed in AS weeds may be only an intermediate phase that can still lead to the death of the plant because a pathogen develops or because a balance between colonizing populations is broken. In the case of pine needles for instance, endophytic fungi may accumulate with time, and once their density exceeds a certain tissue-specific threshold, normally just before the onset of natural senescence, they resume their growth and kill the needle. However, under some adverse conditions, the threshold value may be reached earlier to prematurely eliminate weak needles to preserve needle mass. (67). Thus, the symptom (dead or necrotic needle) is indeed the result of a positive interaction between the plant and the endophytic fungal community and not that of a pathogenic activity. A recent analysis of healthy woody weeds and dieback-affected woody weeds (symptomatic weeds) also reported that colonizing fungal communities strongly differed according to the disease status (32). Thus, although the ITS1 and ITS2 data sets sometimes showed different but complementary aspects of the diversity of colonizing fungi in weed tissues, they also revealed that S and AS weeds harbored different fungal communities. This result, though not really surprising, is quite original.
The fungal community associated with weeds was characterized at two scales, i.e., (i) a global scale to determine the most abundant genera hosted by weeds by targeting both the ITS1 and ITS2 regions and (ii) a specific scale (per weed species and per plot) to determine the differentially abundant genera in S and AS weeds likely to be potential mycoherbicides. At both the global and specific scales, Cladosporium was one of the most abundant genera in S and AS weeds in the two ITS data sets. This genus is known to be very common and to include endophytic species (6871). Some species of Cladosporium are weed-pathogenic, biotrophic, often host-specific fungi that cause leaf spots and other lesions (discoloration, necrosis, etc.) (72). Specific OTUs of Cladosporium were rarely more abundant in S weeds than in AS weeds. However, Cladosporium was abundantly present in Solanum nigrum and may have been responsible for the observed symptoms. Solanum nigrum could be a host weed for this fungus since it belongs to the Solanaceae family like tomato, on which Cladosporium fulvum causes leaf mold disease (73). Some Cladosporium spp. are also known to have weed-protective effects (74). Cladosporium stood out in asymptomatic leaves of several weed species, such as Alopecurus myosuroides and Bromus sterilis. In both cases, the Puccinia genus was associated with symptomatic leaves. According to Torres et al. (75), Cladosporium species (Cladosporium sphaerospermum, Cladosporium uredinicola, and two newly described species Cladosporium cladosporioides and Cladosporium pseudocladosporioides) are fungal antagonists of Puccinia horiana, the causal agent of chrysanthemum white rust. Thus, Cladosporium could be antagonistic to grass rust. Cladosporium was also found abundant in asymptomatic Mercurialis annua leaves, whereas Cercospora was abundant in symptomatic leaves. Therefore, Cladosporium could be an antagonist of Cercospora. Furthermore, Cladosporium cladosporioides appears to be very effective in biological control of the apple scab fungus, Venturia inaequalis (71). Cladosporium spp. could serve as a protector at the beginning of plant (weed and crop) life, and other fungi could multiply over time and gradually overtake it.
Alternaria was one of the most abundant genera in S and AS weeds in the two ITS data sets. This genus stood out in several comparisons between S and AS weeds for some weed species. It has been reported as a ubiquitous fungal genus including saprophytic, endophytic, and pathogenic species (38). Endophytic Alternaria have been isolated from many plant species such as Quercus ilex (76), Quercus emoryi (77), or Taxus globosa (78). This endophytic fungus has also been isolated from Parthenium hysterophorus, an invasive weed that colonizes urban and agricultural zones in Australia, India, and Madagascar (79). The presence of Alternaria spp. over other fungal genera was somewhat expected considering its high frequency of isolation from environmental samples using conventional culture methods. Alternaria belongs to the order Helotiales, which includes many species shifting between endophytic and pathogenic growth phases (80). Abundant taxa, such as Alternaria, Cladosporium, Epicoccum, and Fusarium, associated with weeds have also been found on other plants (strawberry, grapevine, and in forests) but did not cause any symptoms (49, 65, 76). Although the existence of weed-colonizing fungi has been partially neglected up to now, these fungi are fully part of the plant holobiont. The present study suggests that commonly found fungi are present in several weed species as a common core whose interactions with the host weed are not as specific as weed-pathogen interactions can be.
Neoascochyta was an abundant genus at the global scale in S and AS weeds in the two ITS data sets. Like Alternaria, Neoascochyta was differentially detected in S and AS weeds for some weed species. Neoascochyta belongs to the Didymellaceae family that includes a wide range of plant-pathogenic fungal species. The phylogeny of Didymellaceae was revised recently, and Neoascochyta was reported to be the correct genus for some species previously identified as Ascochyta, including plant pathogens (81). Some Neoascochyta strains were recently found to be pathogenic on wheat (82). However, in our study, Neoascochyta was specifically associated with several asymptomatic weeds. To our knowledge, this genus had not been identified as a plant endophytic fungus or a fungal antagonist until now.
Neosetophoma was only detected by the primers targeting the ITS2 region. This genus stood out in one situation on symptomatic Lolium rigidum leaves and was associated with several other fungal genera that were more abundant in S weeds than in AS weeds. These genera could belong to a fungal complex responsible for the observed symptoms. Within the newly revisited taxonomy of Phaeosphaeriaceae, Neosetophoma is described as a genus belonging to Pleosporales, the largest order in Dothideomycetes (83). Neosetophoma is typified by Neosetophoma samararum, a species previously identified as Phoma samararum and isolated from nettle (Urtica dioica) (84). It is described in articles on fungal systematics as being a saprophyte or an endophyte but also a pathogen causing leaf spots on various host plants whose species names are unfortunately not provided (8587).
Sphaerellopsis includes species that are cosmopolitan mycoparasites attacking rust-causing fungi (88). Sphaerellopsis was detected in the two ITS data sets and was an abundant genus among Ascomycota in the ITS1 data set. It was also dominant in the ITS1 data set on symptomatic Lolium rigidum leaves, while Puccinia was dominant in the ITS2 data set. Thus, these two genera were found on the same symptomatic weed where the mycoparasitic fungus Sphaerellopsis could attack the rust-causing fungus Puccinia. Sphaerellopsis filum is mainly known as a mycoparasite of Puccinia graminis subsp. graminicola, a rust fungus on perennial ryegrass (89). The potential biocontrol activity of Sphaerellopsis isolates has been assessed, but unfortunately, the rust reduction achieved was too low to keep on with this strategy (88, 90).
Among Basidiomycota, the Sporobolomyces and Vishniacozyma yeast genera were the most abundant ones in the two data sets. At the specific scale, Sporobolomyces stood out on several S or AS weed species. Little information is available on this genus, except that it can be an endophyte. Endophytic yeasts, among them Sporobolomyces, have been isolated from leaves, flowers, and fruits of healthy apple trees in Brazil (91). Sporobolomyces has also been found in soybean leaves (92). In this genus, the species Sporobolomyces roseus is known to be a biological control agent of aerial contaminations of flowering spikes by Fusarium graminearum (93). Similarly, the Vishniacozyma genus has been differentially detected on several S and AS weed species. It was identified as an endophytic yeast in sugarcane leaves (94). To our knowledge, the Vishniacozyma endophytic genus has no known pathogenic or antagonistic activity or any other plant-protective activity. However, the epiphytic Vishniacozyma genus could have an antagonistic activity against blue mold and gray mold decay caused by Penicillium expansum and Botrytis cinerea, respectively, which are the most detrimental postharvest pear diseases (95). Therefore, both endophytic Sporobolomyces and Vishniacozyma could play a role in protecting their host plants and nullifying any mycoherbicidal biocontrol activity (93). More generally, fungal endophytes are potential weed disease biocontrol agents because they can colonize the same ecological niche as many invading pathogens (96) and produce metabolites through microbe-microbe interactions. However, this beneficial situation for weed control could turn into disadvantages when one or several colonizing fungi play against the biocontrol products. The fungal communities of colonizers revealed by the present analysis highlight the importance of understanding how these fungi synergize or oppose one another and of understanding the interactions of microbial communities associated with weeds. Thus, the development and exploitation of such colonizing fungi, including more specifically endophytes as biocontrol agents, will have to overcome numerous challenges (97).
To conclude, analyzing the huge diversity of colonizing fungi by metabarcoding provides new and original information on the fungal microflora hosted by weeds. This analysis allowed us to (i) detect genera associated with symptoms observed on weeds and (ii) understand that the colonizing community certainly plays a role in the expression of pathogenicity. This, along with other parameters, such as environmental conditions, could explain the current difficulty in implementing effective biocontrol solutions.
The differences between colonizing fungal communities observed in this study depended on (i) the suitability of the substrate (weed species, weed phenological state, etc.), (ii) competition associated with antagonism or any systemic plant defense reactions elicited by colonizing fungi, and (iii) pedoclimatic conditions. Weeds obviously do not host only pathogens but also antagonists and yeasts. Fungi like Alternaria can be avirulent strains able to become pathogenic (34). These strains are capable of pathogenicity if the host conditions change (nutrient status, hydric, or biotic stress), or they can have long latent periods (34). The interactions between the different components of the fungal communities (closely associated epiphytes, endophytes, pathogens, and saprotrophs) are complex and misunderstood. However, closely associated epiphytic, endophytic, and saprophytic fungal taxa are quite close and may swap roles (98, 99).
In addition to the acquisition of previously unexplored knowledge, our study highlights the existence of weed-pathogenic fungi that could be used for biological weed management, but it also points out that the implementation of biocontrol may come up against antagonistic colonizing fungi and the elicitation of weed defense reactions by other plant-associated fungi. The protector fungal taxa can impede the application of biocontrol products. Therefore, understanding the role of each of the microorganisms is important to find appropriate fungal taxa to control weeds.
Analyzing diversity using a molecular approach revealed some interesting tracks to follow to better understand weed-colonizing fungi interactions and propose strategies for targeting taxa that may host robust and sustainable candidates for use in biocontrol. However, the present analysis is based on relative abundances of DNA fragments associated or not with the presence of different symptoms. The third-generation sequencing Pacific Biosciences (PacBio) technology is a promising technology for targeting the whole ITS region in metabarcoding analyses of fungi and provides species names rather than genus names (100). More concretely, a pathology study implies testing the virulence and the aggressiveness of putative pathogenic fungal strains on a spectrum of hosts, including targeted weeds and untargeted cultivated plants that have to be preserved. That is why a classical microbiological approach should be carried out in a complementary way to definitively confirm the hypotheses and finalize the proposals, especially with a view to managing weeds in a sustainable manner.


Weed sampling and DNA extraction.

Sample collection took place in 38 arable plots in four areas in France from July to September 2016. These areas were located in the north of central France (Beauce) or eastern France (northern Burgundy, southern Burgundy, and Jura). Weed sampling was performed using a preestablished list of the 23 most harmful weed species in arable crops in these regions (see Table S1 in the supplemental material). Sampling was performed in plots harboring both symptomatic (S) and asymptomatic (AS) weeds. S weeds exhibited various symptoms including necrosis, chlorosis, rust-like symptoms, or white mold symptoms (Fig. 1). Whenever possible, four S weeds and four AS weeds at the same phenological stage were collected in each plot. Each set of four S or four AS weeds is referred to as one weed modality in the rest of the text. However, in rare situations, it was not possible to collect four specimens of each weed, and the modality included less than four weeds. When two different symptoms were observed in the same plot, four S weeds (i.e., four leaves, one per plant) per symptom and four AS weeds were collected. The final collection included 475 weeds corresponding to 65 S modalities and 55 AS modalities.
To carry out a comparative analysis of the weed endophytic microflora, one leaf per weed was taken from each S or AS weed. Leaves were surface-sterilized by immersion in 70% ethanol for 20 s and then rinsed in two successive sterile water baths and dried on sterile absorbent paper. Then, they were transferred to cryotubes, immediately frozen at −20°C, freeze-dried (Beta 1-8 LD, Christ, Osterode am Harz, Germany) for 48 h, and stored at −20°C. For each sample, 10 to 30 mg of lyophilized weed tissues were transferred to 2-mL screw-cap tubes containing three silica beads (diameter, 4 mm). Samples were ground 2 × 45 s at 4 m s−1 with a FastPrep-24 (MP Biomedicals, Solon, OH, USA). DNA extractions were performed with NucleoSpin Weed II kits (Macherey-Nagel, Düren, Germany). DNA concentrations were measured by fluorometry using a QuantiFluor double-stranded DNA (dsDNA) dye system kit (Promega, Madison, WI, USA) and a plate reader (Tecan, Männedorf, Switzerland), and DNA were stored at −20°C.

DNA amplification, amplicon library preparation, and amplicon sequencing.

A comparative metabarcoding analysis of the fungal communities was conducted by targeting the ITS1 and ITS2 regions. The ITS1 region was amplified using the specific fungal primers ITS1-F (5′-CTTGGTCATTTAGAGGAAGTAA-3′) and ITS2 (5′-GCTGCGTTCTTCATCGATGC-3′) (101, 102). Each PCR was conducted in a total volume of 50 μL containing 20 ng of DNA extracted from weed, 1.5 mM MgCl2, 200 μM deoxynucleoside triphosphates (dNTPs), 0.2 μM each primer, 0.56 mg mL−1 of bovine serum albumin (BSA) (Merck, Darmstadt, Germany), 2 U of Taq DNA polymerase (Expand high fidelity PCR system; Roche, Merck, Darmstadt, Germany), and 1× PCR buffer. Amplification conditions were as follows: initial denaturation at 94°C for 3 min, followed by 35 cycles of 30 s at 94°C, 30 s at 53°C, 45 s at 72°C, and a final extension step at 72°C for 10 min. The ITS2 region was amplified using the specific fungal primers ITS86F (5′-GTGAATCATCGAATCTTTGAA-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) (51, 101). Each PCR was conducted in a total volume of 50 μL containing 20 ng of weed DNA, 1.8 mM MgCl2, 200 μM dNTPs, 0.4 μM each primer, 2.5 U of Taq DNA polymerase, and 1× PCR buffer. Reaction conditions were as follows: initial denaturation at 95°C for 2 min, followed by 40 cycles of 30 s at 95°C, 30 s at 55°C, 1 min at 72°C, and a final extension step at 72°C for 10 min. Primers were linked with Illumina adaptors at their 5′ end (CTTTCCCTACACGACGCTCTTCCGATCT for the forward primers ITS1-F and ITS86F, and GGAGTTCAGACGTGTGCTCTTCCGATCT for the reverse primers ITS2 and ITS4) as recommended by the sequencing platform (GeT-PlaGe; Genotoul, Toulouse, France).
Amplifications were performed in a Mastercycler (Eppendorf, Hamburg, Germany). Amplicon yield and quality were evaluated by agarose gel electrophoresis at 2%. The following steps were performed by the GeT-PlaGe platform. (i) Amplicons were purified and indexed with multiplex identifier (MID) to identify each sample through a short PCR with primers corresponding to the Illumina adaptor and a unique MID for each sample. (ii) Amplicons were quantified and pooled in equimolar concentrations to establish libraries. (iii) The libraries were sequenced using the HTS Illumina MiSeq technology using the reagent kit v3 (2 × 250 bp).

Bioinformatic analysis.

Sequence data were analyzed using the homemade Jupyter Notebooks software program (INRAE, Dijon, France) piping together different bioinformatics tools, and according to Kluyver et al. (103), R1 and R2 sequences were assembled using PEAR (paired-end read merger) (104) with default settings, except for the sequence quality criteria set to 29 instead of 19 to obtain a better quality in sequence assembly. Further quality checks were conducted using the QIIME pipeline (105), and sequences shorter than 200 bp were removed. Reference-based and de novo chimera detection, as well as clustering in operational taxonomic units (OTUs), were performed using VSEARCH (106) and the UNITE reference database v.7-08/2016 (107). OTUs were generated after these two clustering steps (reference-based and de novo) were carried out at 97% similarity. OTUs represented by less than five sequences in the global data sets were removed. OTU taxonomic assignment was performed using BLAST (108) and the UNITE reference database. Taxonomic assignment of the most abundant OTU was verified by manual examination of the BLAST hits. Only the sequences associated with a publication were retained.

Statistical analyses.

Analyses were performed with R (v3.5.1) (R Core Team, 2018) using different statistical tests in the vegan package (v2.5-3), phyloseq package (v1.24.2), DEseq2 package (v1.20.0), and ggplot2 (v3.0.0) for the figures. Phyloseq and DEseq2 are language packages in R program available through the Bioconductor platform (https://www.bioconductor.org/biocLite.R). Nonfungal sequences were removed. Diversity metrics, i.e., richness (numbers of OTU) and evenness (Simpson’s reciprocal index), were calculated based on rarefied data (10,000 reads per sample) obtained with the function “rarefy” in the vegan package. Diversity metrics were compared between fungal communities originating from S and AS weeds using Wilcoxon tests. The OTU tables obtained from the ITS1 and ITS2 data sets were used to perform a nonmetric multidimensional scaling (NMDS) analysis targeting the fungal communities present in S and AS weeds according to the sampling area or to the weed species sampled. A permutational analysis of variance (PERMANOVA) was applied using the stress value given by the Bray-Curtis measure of pairwise dissimilarity to check for community differences between S and AS weeds and between areas.
The DEseq2 package was used to identify the differentially abundant OTUs in the S and AS weed samples in the ITS1 and ITS2 data sets. Using first an important step of normalization, DEseq2 then estimated the geometric mean in HTS data and tested for differential expression based on a model using the negative binomial distribution (109). Then DEseq2 removed the OTU absent in at least one sample. We considered DEseq2 results when more than 70% of the total fungal sequences were included in the analysis. If this criterion was not fulfilled, a Kruskal-Wallis test (with n = 4 per modality S and AS weeds) was performed. The OTUs were defined as differentially present when padj is <0.05 (here, padj refers to the raw P value adjusted for multiple testing using the Benjamini-Hochberg method). R functions allowed for automating the analyses (R scripts are provided in the supplemental material).

Data availability.

Data are accessible under BioProject number PRJNA622886.


This work, through the involvement of technical facilities of the GenoSol platform of the infrastructure ANAEE-Services, received a grant from the French state through the National Agency for Research under the program “Investments for the Future” (reference ANR-11-INBS-0001), as well as a grant from the Regional Council of Bourgogne-Franche-Comté. The BRC GenoSol is a part of BRC4Env, the pillar “Environmental Resources” of the Research Infrastructure AgroBRC-RARe.
We thank Arnaud Mounier and Aimé Spor for their assistance in bioinformatics analyses and GeT-PlaGe platform (Genotoul, Toulouse, France) for library preparation and Illumina sequencing.
PhD funding from the National Association of Technical Research (ANRT) (CIFRE no 2015-1280) financially supported Marion Triolet.
We declare no conflict of interest.

Supplemental Material

File (aem.02177-21-s0001.pdf)
File (aem.02177-21-s0002.xlsx)
ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.


Neve P, Barney JN, Buckley Y, Cousens RD, Graham S, Jordan NR, Lawton-Rauh A, Liebman M, Mesgaran MB, Schut M, Shaw J, Storkey J, Baraibar B, Baucom RS, Chalak M, Childs DZ, Christensen S, Eizenberg H, Fernandez-Quintanilla C, French K, Harsch M, Heijting S, Harrison L, Loddo D, Macel M, Maczey N, Merotto A, Jr, Mortensen D, Necajeva J, Peltzer DA, Recasens J, Renton M, Riemens M, Sonderskov M, Williams M. 2018. Reviewing research priorities in weed ecology, evolution and management: a horizon scan. Weed Res 58:250–258.
Storkey J, Neve P. 2018. What good is weed diversity? Weed Res 58:239–243.
Oerke EC, Dehne HW. 2004. Safeguarding production—losses in major crops and the role of crop protection. Crop Protection 23:275–285.
Pimentel D, Zuniga R, Morrison D. 2005. Update on the environmental and economic costs associated with alien-invasive species in the United States. Ecol Econ 52:273–288.
Oerke EC. 2006. Crop losses to pests. J Agric Sci 144:31–43.
Shabi T, Islam AKM, Hasan A, Juraimi AS, Anwar MP. 2018. Differential weed suppression ability in selected wheat varieties of Bangladesh. Acta Sci Malays 2:1–7.
Gharde Y, Singh PK, Dubey RP, Gupta PK. 2018. Assessment of yield and economic losses in agriculture due to weeds in India. Crop Protection 107:12–18.
Mézière D, Petit S, Granger S, Biju-Duval L, Colbach N. 2015. Developing a set of simulation-based indicators to assess harmfulness and contribution to biodiversity of weed communities in cropping systems. Ecol Indicators 48:157–170.
Panetta F, Gooden B. 2017. Managing for biodiversity: impact and action thresholds for invasive plants in natural ecosystems. NeoBiota 34:53–66.
Chen KW, Marusciac L, Tamas PT, Valenta R, Panaitescu C. 2018. Ragweed pollen allergy: burden, characteristics, and management of an imported allergen source in Europe. Int Arch Allergy Immunol 176:163–180.
Mouttet R, Augustinus B, Bonini M, Chauvel B, Desneux N, Gachet E, Le Bourgeois T, Müller-Schärer H, Thibaudon M, Schaffner U. 2018. Estimating economic benefits of biological control of Ambrosia artemisiifolia by Ophraella communa in southeastern France. Basic Appl Ecol 33:14–24.
Keshtkar E, Mathiassen SK, Moss SR, Kudsk P. 2015. Resistance profile of herbicide-resistant Alopecurus myosuroides (black-grass) populations in Denmark. Crop Protection 69:83–89.
Helander M, Saloniemi I, Omacini M, Druille M, Salminen J-P, Saikkonen K. 2018. Glyphosate decreases mycorrhizal colonization and affects plant-soil feedback. Sci Total Environ 642:285–291.
Sarangi D, Jhala AJ. 2018. A statewide survey of stakeholders to assess the problem weeds and weed management practices in Nebraska. Weed Technol 32:642–655.
Gilbert PA, Vanasse A, Angers DA. 2009. Harrowing for weed control: impacts on mineral nitrogen dynamics, soil aggregation and wheat production. Soil Tillage Res 103:373–380.
Aggestam V, Buick J. 2017. A comparative analysis of vehicle-related greenhouse gas emissions between organic and conventional dairy production. J Dairy Res 84:360–369.
Marshall EJP, Brown VK, Boatman ND, Lutman PJW, Squire GR, Ward LK. 2003. The role of weeds in supporting biological diversity within crop fields. Weed Res 43:77–89.
Cordeau S, Dessaint F, Denieul C, Bonin L, Vuillemin F, Delattre M, Rodriguez A, Guillemin J-P, Chauvel B. 2016. La nuisibilité directe des adventices en grandes cultures: quelles réponses nous apportent les essais désherbage? AFPP – 23e Conf Columa Journées Internationales Sur la Lutte Contre Les Mauvaises Herbes Dijon.
Bailey KL. 2014. The bioherbicide approach to weed control using plant pathogens, p 245–266. In Abrol DP (ed), Integrated pest management current concepts and ecological perspective. Academic Press, San Diego, CA.
Harding DP, Raizada MN. 2015. Controlling weeds with fungi, bacteria and viruses: a review. Front Plant Sci 6:659.
Hershenhorn J, Casella F, Vurro M. 2016. Weed biocontrol with fungi: past, present and future. Biocontrol Sci Technol 26:1313–1328.
Hernández-Aro M, Hernández-Pérez R, Guillén-Sánchez D, Torres-Garcia S. 2016. Allelopathic influence of residues from Sphagneticola trilobata on weeds and crops. Planta Daninha 34:81–90.
Petit S, Cordeau S, Chauvel B, Bohan D, Guillemin J-P, Steinberg C. 2018. Biodiversity-based options for arable weed management: a review. Agron Sustain Dev 38:48.
Charudattan R, Dinoor A. 2000. Biological control of weeds using plant pathogens: accomplishments and limitations. Crop Protection 19:691–695.
Ray P, Vijayachandran LS. 2013. Evaluation of indigenous fungal pathogens from horse purslane (Trianthema portulacastrum) for their relative virulence and host range assessments to select a potential mycoherbicidal agent. Weed Sci 61:580–585.
Tehranchian P, Adair RJ, Lawrie AC. 2014. Potential for biological control of the weed angled onion (Allium triquetrum) by the fungus Stromatinia cepivora in Australia. Australas Plant Pathol 43:381–392.
Macedo DM, Pereira OL, Hora Júnior BT, Weir BS, Barreto RW. 2016. Mycobiota of the weed Tradescantia fluminensis in its native range in Brazil with particular reference to classical biological control. Australas Plant Pathol 45:45–56.
Triolet M, Guillemin J-P, André O, Steinberg C. 2020. Fungal-based bioherbicides for weed control: a myth or a reality? Weed Res 60:60–77.
Kao-Kniffin J, Carver SM, Ditommaso A. 2013. Advancing weed management strategies using metagenomic techniques. Weed Sci 61:171–184.
Marsberg A, Slippers B, Wingfield MJ, Gryzenhout M. 2014. Endophyte isolations from Syzygium cordatum and a Eucalyptus clone (Myrtaceae) reveal new host and geographical reports for the Mycosphaerellaceae and Teratosphaeriaceae. Australas Plant Pathol 43:503–512.
Daru BH, Bowman EA, Pfister DH, Arnold AE. 2019. A novel proof of concept for capturing the diversity of endophytic fungi preserved in herbarium specimens. Philos Trans R Soc B 374:20170395.
Steinrucken TV, Bissett A, Powell JR, Raghavendra AKH, van Klinken RD. 2016. Endophyte community composition is associated with dieback occurrence in an invasive tree. Plant Soil 405:311–323.
Steinrucken TV, Raghavendra AKH, Powell JR, Bissett A, van Klinken RD. 2017. Triggering dieback in an invasive plant: endophyte diversity and pathogenicity. Australas Plant Pathol 46:157–170.
Arnold AE. 2007. Understanding the diversity of foliar endophytic fungi: progress, challenges, and frontiers. Fungal Biol Rev 21:51–66.
Brader G, Compant S, Vescio K, Mitter B, Trognitz F, Ma L-J, Sessitsch A. 2017. Ecology and genomic insights into plant-pathogenic and plant-nonpathogenic endophytes. Annu Rev Phytopathol 55:61–83.
De Silva NI, Brooks S, Lumyong S, Hyde KD. 2019. Use of endophytes as biocontrol agents. Fungal Biol Rev 33:133–148.
Bálint M, Tiffin P, Hallström B, O'Hara RB, Olson MS, Fankhauser JD, Piepenbring M, Schmitt I. 2013. Host genotype shapes the foliar fungal microbiome of balsam poplar (Populus balsamifera). PLoS One 8:e53987.
Eram D, Arthikala M-K, Melappa G, Santoyo G. 2018. Alternaria species: endophytic fungi as alternative sources of bioactive compounds. Italian J Mycol 47:40–54.
Saikkonen K, Wäli P, Helander M, Faeth SH. 2004. Evolution of endophyte–plant symbioses. Trends Plant Sci 9:275–280.
Saikkonen K, Young CA, Helander M, Schardl CL. 2016. Endophytic Epichloë species and their grass hosts: from evolution to applications. Plant Mol Biol 90:665–675.
Schardl CL, Leuchtmann A, Spiering MJ. 2004. Symbioses of grasses with seedborne fungal endophytes. Annu Rev Plant Biol 55:315–340.
Tejesvi MV, Pirttilä AM. 2018. Endophytic fungi, occurrence, and metabolites, p 213–230. In Anke T, Schüffler A (ed), Physiology and genetics: selected basic and applied aspects. Springer, Cham, Switzerland.
Pancher M, Ceol M, Corneo PE, Longa CMO, Yousaf S, Pertot I, Campisano A. 2012. Fungal endophytic communities in grapevines (Vitis vinifera L.) respond to crop management. Appl Environ Microbiol 78:4308–4317.
Impullitti AE, Malvick DK. 2013. Fungal endophyte diversity in soybean. J Appl Microbiol 114:1500–1506.
Gonzaga LL, Costa LEO, Santos TT, Araújo EF, Queiroz MV. 2015. Endophytic fungi from the genus Colletotrichum are abundant in the Phaseolus vulgaris and have high genetic diversity. J Appl Microbiol 118:485–496.
Nettles R, Watkins J, Ricks K, Boyer M, Licht M, Atwood LW, Peoples M, Smith RG, Mortensen DA, Koide RT. 2016. Influence of pesticide seed treatments on rhizosphere fungal and bacterial communities and leaf fungal endophyte communities in maize and soybean. Appl Soil Ecol 102:61–69.
Gdanetz K, Trail F. 2017. The wheat microbiome under four management strategies, and potential for endophytes in disease protection. Phytobiomes J 1:158–168.
Dissanayake AJ, Purahong W, Wubet T, Hyde KD, Zhang W, Xu H, Zhang G, Fu C, Liu M, Xing Q, Li X, Yan J. 2018. Direct comparison of culture-dependent and culture-independent molecular approaches reveal the diversity of fungal endophytic communities in stems of grapevine (Vitis vinifera). Fungal Divers 90:85–107.
Kraus C, Voegele RT, Fischer M. 2019. Temporal development of the culturable, endophytic fungal community in healthy grapevine branches and occurrence of GTD-associated fungi. Microb Ecol 77:866–876.
Mukhtar I, Khokhar I, Mushtaq S, Ali A. 2010. Diversity of epiphytic and endophytic microorganisms in some dominant weeds. Pakistan J Weed Science Res 16:287–297.
Op De Beeck M, Lievens B, Busschaert P, Declerck S, Vangronsveld J, Colpaert JV. 2014. Comparison and validation of some ITS primer pairs useful for fungal metabarcoding studies. PLoS One 9:e97629.
Bazzicalupo AL, Bálint M, Schmitt I. 2013. Comparison of ITS1 and ITS2 rDNA in 454 sequencing of hyperdiverse fungal communities. Fungal Ecol 6:102–109.
Tedersoo L, Anslan S, Bahram M, Põlme S, Riit T, Liiv I, Kõljalg U, Kisand V, Nilsson RH, Hildebrand F, Bork P, Abarenkov K. 2015. Shotgun metagenomes and multiple primer pair-barcode combinations of amplicons reveal biases in metabarcoding analyses of fungi. MycoKeys 10:1–43.
Yang R-H, Su J-H, Shang J-J, Wu Y-Y, Li Y, Bao D-P, Yao Y-J. 2018. Evaluation of the ribosomal DNA internal transcribed spacer (ITS), specifically ITS1 and ITS2, for the analysis of fungal diversity by deep sequencing. PLoS One 13:e0206428.
Monard C, Gantner S, Stenlid J. 2013. Utilizing ITS1 and ITS2 to study environmental fungal diversity using pyrosequencing. FEMS Microbiol Ecol 84:165–175.
Tedersoo L, Anslan S. 2019. Towards PacBio-based pan-eukaryote metabarcoding using full-length ITS sequences. Environ Microbiol Rep 11:659–668.
Taylor D, Walters W, Lennon N, Bochicchio J, Krohn A, Caporaso J, Pennanen T. 2016. Accurate estimation of fungal diversity and abundance through improved lineage-specific primers optimized for Illumina amplicon sequencing. Appl Environ Microbiol 82:7217–7226.
Scibetta S, Schena L, Abdelfattah A, Pangallo S, Cacciola SO. 2018. Selection and experimental evaluation of universal primers to study the fungal microbiome of higher plants. Phytobiomes J 2:225–236.
Jin H, Yan Z, Liu Q, Yang X, Chen J, Qin B. 2013. Diversity and dynamics of fungal endophytes in leaves, stems and roots of Stellera chamaejasme L. in northwestern China. Antonie Van Leeuwenhoek 104:949–963.
Gazis R, Chaverri P. 2010. Diversity of fungal endophytes in leaves and stems of wild rubber trees (Hevea brasiliensis) in Peru. Fungal Ecol 3:240–254.
Fernandes EG, Pereira OL, Da Silva CC, Bento CBP, De Queiroz MV. 2015. Diversity of endophytic fungi in Glycine max. Microbiol Res 181:84–92.
Paul N, Ayres P. 1987. Effects of rust infection of Senecio vulgaris on competition with lettuce. Weed Res 27:431–441.
Barreto RW, Evans HC. 1995. Mycobiota of the weed Cyperus rotundus in the state of Rio de Janeiro, with an elucidation of its associated Puccinia complex. Mycol Res 99:407–419.
Talhinhas P, Ramos AP, Tavares D. 2015. Genome size estimates for six rust (Pucciniales) species. Revista de Ciências Agrárias 38:176–183.
Abdelfattah A, Wisniewski M, Li Destri Nicosia MG, Cacciola SO, Schena L. 2016. Metagenomic analysis of fungal diversity on strawberry plants and the effect of management practices on the fungal community structure of aerial organs. PLoS One 11:e0160470.
Schlegel M, Queloz V, Sieber TN. 2018. The endophytic mycobiome of European ash and Sycamore maple leaves - geographic patterns, host specificity and influence of ash dieback. Front Microbiol 9:2345.
Sieber TN. 2007. Endophytic fungi in forest trees: are they mutualists? Fungal Biol Rev 21:75–89.
Wang FW, Jiao RH, Cheng AB, Tan SH, Song YC. 2007. Antimicrobial potentials of endophytic fungi residing in Quercus variabilis and brefeldin A obtained from Cladosporium sp. World J Microbiol Biotechnol 23:79–83.
Zhang P, Zhou PP, Yu LJ. 2009. An endophytic taxol-producing fungus from taxus media, Cladosporium cladosporioides MD2. Curr Microbiol 59:227.
Thakur A, Singh V, Kaur A, Kaur S. 2013. Insecticidal potential of an endophytic fungus, Cladosporium uredinicola, against Spodoptera litura. Phytoparasitica 41:373–382.
Wang X, Radwan MM, Taráwneh AH, Gao J, Wedge DE, Rosa LH, Cutler HG, Cutler SJ. 2013. Antifungal activity against plant pathogens of metabolites from the endophytic fungus Cladosporium cladosporioides. J Agric Food Chem 61:4551–4555.
Bensch K, Braun U, Groenewald JZ, Crous PW. 2012. The genus Cladosporium. Stud Mycol 72:1–401.
Rivas S, Thomas CM. 2005. Molecular interactions between tomato and the leaf mold pathogen Cladosporium fulvum. Annu Rev Phytopathol 43:395–436.
Shah S, Shrestha R, Maharjan S, Selosse M-A, Pant B. 2018. Isolation and characterization of plant growth-promoting endophytic fungi from the roots of Dendrobium moniliforme. Plants 8:5.
Torres DE, Rojas-Martínez RI, Zavaleta-Mejía E, Guevara-Fefer P, Márquez-Guzmán GJ, Pérez-Martínez C. 2017. Cladosporium cladosporioides and Cladosporium pseudocladosporioides as potential new fungal antagonists of Puccinia horiana Henn., the causal agent of chrysanthemum white rust. PLoS One 12:e0170782.
Fisher P, Petrini O, Petrini L, Sutton B. 1994. Fungal endophytes from the leaves and twigs of Quercus ilex L. from England, Majorca and Switzerland. New Phytol 127:133–137.
Bashyal BP, Wellensiek BP, Ramakrishnan R, Faeth SH, Ahmad N, Gunatilaka AAL. 2014. Altertoxins with potent anti-HIV activity from Alternaria tenuissima QUE1Se, a fungal endophyte of Quercus emoryi. Bioorg Med Chem 22:6112–6116.
Rivera-Orduña FN, Suarez-Sanchez RA, Flores-Bustamante ZR, Gracida-Rodriguez JN, Flores-Cotera LB. 2011. Diversity of endophytic fungi of Taxus globosa (Mexican yew). Fungal Divers 47:65–74.
Romero A, Carrion G, Rico-Gray V. 2001. Fungal latent pathogens and endophytes. Fungal Divers 7:81–87.
Cross H, Sønstebø JH, Nagy NE, Timmermann V, Solheim H, Børja I, Kauserud H, Carlsen T, Rzepka B, Wasak K, Vivian‐Smith A, Hietala AM. 2017. Fungal diversity and seasonal succession in ash leaves infected by the invasive ascomycete Hymenoscyphus fraxineus. New Phytol 213:1405–1417.
Chen Q, Hou LW, Duan WJ, Crous PW, Cai L. 2017. Didymellaceae revisited. Stud Mycol 87:105–159.
Golzar H, Thomas G, Jayasena KW, Wright D, Wang C, Kehoe M. 2019. Neoascochyta species cause leaf scorch on wheat in Australia. Australas Plant Dis Notes 14:1.
Karunarathna A, Papizadeh M, Senanayake I, Jeewon R, Phookamsak R, Goonasekara ID, Wanasinghe DN, Wijayawardene NN, Amoozegar MA, Shahzadeh FS, Camporesi E, Hyde KD, Weerahewa HLD, Lumyong S, McKenzie E. 2017. Novel fungal species of Phaeosphaeriaceae with an asexual/sexual morph connection. Mycosphere 8:1818–1834.
De Gruyter J, Woudenberg JHC, Aveskamp MM, Verkley GJM, Groenewald JZ, Crous PW. 2010. Systematic reappraisal of species in Phoma section Paraphoma, Pyrenochaeta and Pleurophoma. Mycologia 102:1066–1081.
Hyde KD, Jones EBG, Liu J-K, Ariyawansa H, Boehm E, Boonmee S, Braun U, Chomnunti P, Crous PW, Dai D-Q, Diederich P, Dissanayake A, Doilom M, Doveri F, Hongsanan S, Jayawardena R, Lawrey JD, Li Y-M, Liu Y-X, Lücking R, Monkai J, Muggia L, Nelsen MP, Pang K-L, Phookamsak R, Senanayake IC, Shearer CA, Suetrong S, Tanaka K, Thambugala KM, Wijayawardene NN, Wikee S, Wu H-X, Zhang Y, Aguirre-Hudson B, Alias SA, Aptroot A, Bahkali AH, Bezerra JL, Bhat DJ, Camporesi E, Chukeatirote E, Gueidan C, Hawksworth DL, Hirayama K, De Hoog S, Kang J-C, Knudsen K, Li W-J, Li X-H, et al. 2013. Families of Dothideomycetes. Fungal Divers 63:1–313.
Phookamsak R, Liu JK, Mckenzie EHC, Manamgoda DS, Ariyawansa H, Thambugala KM, Dai D-Q, Camporesi E, Chukeatirote E, Wijayawardene NN, Bahkali AH, Mortimer PE, Xu J-C, Hyde KD. 2014. Revision of Phaeosphaeriaceae. Fungal Divers 68:159–238.
Tibpromma S, Hyde KD, Jeewon R, Maharachchikumbura SSN, Liu J-K, Bhat DJ, Jones EBG, McKenzie EHC, Camporesi E, Bulgakov TS, Doilom M, de Azevedo Santiago ALCM, Das K, Manimohan P, Gibertoni TB, Lim YW, Ekanayaka AH, Thongbai B, Lee HB, Yang J-B, Kirk PM, Sysouphanthong P, Singh SK, Boonmee S, Dong W, Raj KNA, Latha KPD, Phookamsak R, Phukhamsakda C, Konta S, Jayasiri SC, Norphanphoun C, Tennakoon DS, Li J, Dayarathne MC, Perera RH, Xiao Y, Wanasinghe DN, Senanayake IC, Goonasekara ID, de Silva NI, Mapook A, Jayawardena RS, Dissanayake AJ, Manawasinghe IS, Chethana KWT, Luo Z-L, Hapuarachchi KK, Baghela A, Soares AM, et al. 2017. Fungal diversity notes 491-602: taxonomic and phylogenetic contributions to fungal taxa. Fungal Divers 83:1–261.
Trakunyingcharoen T, Lombard L, Groenewald J, Cheewangkoon R, To-Anun C, Alfenas A, Crous P. 2014. Mycoparasitic species of Sphaerellopsis, and allied lichenicolous and other genera. IMA Fungus 5:391–414.
Gordon TC, Pfender WF. 2012. Effects of the mycoparasite Sphaerellopsis filum on overwintering survival of stem rust in perennial Ryegrass. Plant Dis 96:1471–1481.
Freeman T. 1981. Use of conidial fungi in biological control, p 143–165. In Cole GT, Kendrick B (ed), Biology of conidial fungi, vol 2. Academic Press, San Diego, CA.
Camatti-Sartori V, Da Silva-Ribeiro RT, Valdebenito-Sanhueza RM, Pagnocca FC, Echeverrigaray S, Azevedo JL. 2005. Endophytic yeasts and filamentous fungi associated with southern Brazilian apple (Malus domestica) orchards subjected to conventional, integrated or organic cultivation. J Basic Microbiol 45:397–402.
de Souza Leite T, Cnossen-Fassoni A, Pereira O, Mizubuti E, de Araújo E, de Queiroz M. 2013. Novel and highly diverse fungal endophytes in soybean revealed by the consortium of two different techniques. J Microbiol 51:56–69.
Bergstrom GC, Corio Da Luz W. 2005. Biocontrol for plants with Bacillus subtilis, Pseudomonas putida, and Sporobolomyces roseus. US Patent 6,896,883 B2.
Khunnamwong P, Jindamorakot S, Limtong S. 2018. Endophytic yeast diversity in leaf tissue of rice, corn and sugarcane cultivated in Thailand assessed by a culture-dependent approach. Fungal Biol 122:785–799.
Gramisci BR, Lutz MC, Lopes CA, Sangorrín MP. 2018. Enhancing the efficacy of yeast biocontrol agents against postharvest pathogens through nutrient profiling and the use of other additives. Biol Control 121:151–158.
Ownley B, Pereira R, Klingeman W, Quigley N, Leckie B. 2004. Beauveria bassiana, a dual-purpose biological control with activity against insect pests and plant pathogens, p 255–269. In Lartey RT, Caesar AJ (ed), Emeging concepts in plant health management. Research Signpost, Kerala, India.
Rabiey M, Hailey LE, Roy SR, Grenz K, Al-Zadjali MAS, Barrett GA, Jackson RW. 2019. Endophytes vs tree pathogens and pests: can they be used as biological control agents to improve tree health? Eur J Plant Pathol 155:711–729.
Fröhlich J, Hyde KD. 1999. Biodiversity of palm fungi in the tropics: are global fungal diversity estimates realistic? Biodivers Conserv 8:977–1004.
Promputtha I, Lumyong S, Dhanasekaran V, Mckenzie EHC, Hyde KD, Jeewon R. 2007. A phylogenetic evaluation of whether endophytes become saprotrophs at host senescence. Microb Ecol 53:579–590.
Tedersoo L, Tooming-Klunderud A, Anslan S. 2018. PacBio metabarcoding of fungi and other eukaryotes: errors, biases and perspectives. New Phytol 217:1370–1385.
White TJ, Bruns T, Lee S, Taylor J. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics, p 315–322. In Innis MA, Gelfand DH, Sninsky JJ, White TJ (ed), PCR protocols: a guide to methods and applications. Academic Press, San Diego, CA.
Gardes M, Bruns D. 1993. ITS primers with enhanced specificity for basidiomycetes - application to the identification of mycorrhizae and rusts. Mol Ecol 2:113–118.
Kluyver T, Ragan-Kelley B, Pérez F, Granger B, Bussonnier M, Frederic J, Kelley K, Hamrick J, Grout J, Corlay S, Ivanov P, Avila D, Abdalla S, Willing C, Jupyter Development Team. 2016. Jupyter Notebooks—a publishing format for reproducible computational workflows, p 87–90. In Loizides F, Scmidt B (eds), Positioning and power in academic publishing: players, agents and agendas. IOS Press, Amsterdam, Netherlands.
Zhang J, Kobert K, Flouri T, Stamatakis A. 2014. PEAR: a fast and accurate Illumina paired-end reAd mergeR. Bioinformatics 30:614–620.
Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Peña AG, Goodrich JK, Gordon JI, Huttley GA, Kelley ST, Knights D, Koenig JE, Ley RE, Lozupone CA, McDonald D, Muegge BD, Pirrung M, Reeder J, Sevinsky JR, Turnbaugh PJ, Walters WA, Widmann J, Yatsunenko T, Zaneveld J, Knight R. 2010. QIIME allows analysis of high-throughput community sequencing data. Nat Methods 7:335–336.
Rognes T, Flouri T, Nichols B, Quince C, Mahé F. 2016. VSEARCH: a versatile open source tool for metagenomics. PeerJ 4:e2584.
Abarenkov K, Nilsson RH, Larsson K-H, Alexander IJ, Eberhardt U, Erland S, Høiland K, Kjøller R, Larsson E, Pennanen T, Sen R, Taylor AFS, Tedersoo L, Ursing BM, Vrålstad T, Liimatainen K, Peintner U, Kõljalg U. 2010. The UNITE database for molecular identification of fungi—recent updates and future perspectives. New Phytol 186:281–285.
Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local alignment search tool. J Mol Biol 215:403–410.
Anders S, Huber W. 2010. Differential expression analysis for sequence count data. Genome Biol 11:R106.

Information & Contributors


Published In

cover image Applied and Environmental Microbiology
Applied and Environmental Microbiology
Volume 88Number 622 March 2022
eLocator: e02177-21
Editor: Irina S. Druzhinina, Nanjing Agricultural University
PubMed: 35080907


Received: 3 November 2021
Accepted: 20 December 2021
Accepted manuscript posted online: 26 January 2022
Published online: 22 March 2022


Request permissions for this article.


  1. fungal community
  2. metabarcoding
  3. high-throughput sequencing
  4. ITS1
  5. ITS2
  6. weed biocontrol



Agroécologie, INRAE, Institut Agro, Université Bourgogne Franche-Comté, Dijon, France
De Sangosse—Bonnel, CS 10005, Pont du Casse, France
Agroécologie, INRAE, Institut Agro, Université Bourgogne Franche-Comté, Dijon, France
Agroécologie, INRAE, Institut Agro, Université Bourgogne Franche-Comté, Dijon, France
Agroécologie, INRAE, Institut Agro, Université Bourgogne Franche-Comté, Dijon, France
Carole Reibel
Agroécologie, INRAE, Institut Agro, Université Bourgogne Franche-Comté, Dijon, France
De Sangosse—Bonnel, CS 10005, Pont du Casse, France
Agroécologie, INRAE, Institut Agro, Université Bourgogne Franche-Comté, Dijon, France
Agroécologie, INRAE, Institut Agro, Université Bourgogne Franche-Comté, Dijon, France


Irina S. Druzhinina
Nanjing Agricultural University


The authors declare no conflict of interest.

Metrics & Citations



  • For recently published articles, the TOTAL download count will appear as zero until a new month starts.
  • There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.
  • Citation counts come from the Crossref Cited by service.


If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.

View Options

Figures and Media






Share the article link

Share with email

Email a colleague

Share on social media

American Society for Microbiology ("ASM") is committed to maintaining your confidence and trust with respect to the information we collect from you on websites owned and operated by ASM ("ASM Web Sites") and other sources. This Privacy Policy sets forth the information we collect about you, how we use this information and the choices you have about how we use such information.
FIND OUT MORE about the privacy policy