Biology and Fertility of Soils (2022) 58:917–935 https://doi.org/10.1007/s00374-022-01678-1 ORIGINAL PAPER Occurrence and diversity of arbuscular mycorrhizal fungi colonising off‑season and in‑season weeds and their relationship with maize yield under conservation agriculture Blessing Mhlanga1 · Laura Ercoli1 · Gaia Piazza1,2 · Christian Thierfelder2 · Elisa Pellegrino1 Received: 21 April 2022 / Revised: 24 September 2022 / Accepted: 16 October 2022 / Published online: 26 October 2022 © The Author(s) 2022 Abstract Weeds are responsible for major crop losses worldwide but can provide beneficial agroecosystem services. This study aimed to elucidate how arbuscular mycorrhizal fungi (AMF) in weeds respond to host identity and conservation agricultural practices. The study was carried out at two locations in Southern Africa during off-season and in-season maize cultivation. Off-season AMF root colonisation, diversity indices and community composition significantly differed among weed species at both locations. Glomus sp. VTX00280 explains most of the AMF community differences. In-season, implementation of conventional tillage with mulching alone (CT + M) or together with crop rotation (CT + M + R) resulted in a 20% increase in AMF colonisation of the constantly occurring weed species, Bidens pilosa (BIDPI) and Richardia scabra (RCHSC), com- pared with conventional tillage plus rotations (CT + R). The diversity of AMF was highest under no-tillage plus mulching (NT + M). Off-season and in-season AMF structures of both BIDPI and RCHSC were not related, but 39% of the taxa were shared. Structural equation modelling showed a significant effect of the cropping system on weed AMF diversity parameters and weed and maize root colonisation, but no significant influence of weed root AMF traits and maize colonisation was detected on maize yield. This may be explained by the improvement in weed competitive ability, which may have offset the AMF-mediated benefits on yield. Our findings highlight that implementing M and CR to CT and NT positively affected weed AMF colonisation and diversity. The similarity between the off-season and in-season AMF composition of weeds supports the fact that weeds functionally host AMF during the non-crop period. Keywords Agroecosystem services · No-tillage · Mulching · Crop rotation · Root colonisation · Host specificity Introduction Among these practices, the management of weeds is one of the major challenges worldwide in large- and small-scale The increasing demand for food for the continuously grow- agriculture systems. This is especially true in southern ing world population calls for more sustainable management Africa under smallholder cropping systems, where weeds practices that promote yield whilst reducing the impact on account for 10 to 100% of yield losses in cereals, depend- the environment and biodiversity (MacLaren et al. 2020). ing on the involved weed species and the level of manage- ment (Heyl 2022). In this area, the high losses are mainly Blessing Mhlanga and Elisa Pellegrino contributed equally to this work. due to inadequate weed management practices, such as late weeding caused by a lack of manpower and inadequate * Blessing Mhlanga associated practices, such as rotation, intercropping and blessing.mhlangah@gmail.com mulching (Silva et al. 2019). On the other hand, successful * Elisa Pellegrino weed management requires a deep knowledge of the weed e.pellegrino@santannapisa.it communities occurring under different cropping systems, 1 especially under multi-component systems, such as conser- Crop Science Research Center, Scuola Superiore Sant’Anna, Pisa, Italy vation agriculture (CA), where management practices can 2 be implemented in different combinations (Derrouch et al. International Maize & Wheat Improvement Centre (CIMMYT), Southern Africa Regional Office (SARO), 2021). Conservation agriculture is based on three main prin- Harare, Zimbabwe ciples: minimum soil disturbance, permanent soil cover and Vol.:(012 3456789) 9 18 Biology and Fertility of Soils (2022) 58:917–935 crop diversification (FAO 2019). Modification of cropping et al. 2016; El Omari and El Ghachtouli 2021). Moreover, systems, especially through changing of cropping sequences weeds can also become alternative AMF hosts during the and including mulching, altered weed species composition off-season (non-crop growing dry periods) (Massenssini (Koocheki et al. 2009; Zhang et al. 2021). et al. 2014). This is particularly important in southern Africa, Despite their negative effects on crop productivity due which is characterised by short crop growing seasons and long to competition for water, radiation and nutrients, weeds can winter dry periods within the year. Furthermore, arbuscular provide beneficial agroecosystem services (MacLaren et al. mycorrhizas (AM) formed with native weed species can 2020; El Omari and El Ghachtouli 2021). Thus, several increase their competitive ability against invasive species and options, different from full weed eradication, have been hence prevent their dominance (Zhang et al. 2018; El Omari suggested with the goal of achieving a trade-off between and El Ghachtouli 2021). Despite abundant reports on the AM negative impacts and positive effects resulting from the fungal host preference/specificity in field trials, supporting conservation of weed diversity and functionality. Indeed, a specialisation between plant and associated AMF community more diverse weed community has recently been shown to (e.g. Gollotte et al. 2004; Helgason et al. 2007; Martínez- be less competitive with many crops, as well as to promote García and Pugnaire 2011; Li et al. 2019), no information crop health and beneficial bees (e.g. Storkey and Neve is available on the effect of host identity on AM fungal 2018; Ferrero et al. 2017; Bretagnolle and Gaba 2015). assemblages in off-season and in-season weeds. This effect can Regarding winter cereals, Adeux et al. (2019) have recently also be modulated by agronomical management practices that demonstrated that the reduction in yield loss was better might affect the weed outcome, ranging from suppression to explained by the increase in weed diversity rather than the promotion (Bever 2002; Zhang et al. 2010). This is particularly decrease in weed density. In addition, ecosystem services important in the CA systems of southern Africa where farmers can be rendered by some weed species, such as black-jack implement the components in different combinations, resulting (Bidens pilosa L.), barnyard grass (Echinochloa crus-galli in different weed communities potentially hosting diversified (L) P. Beauv) and black nightshade (Solanum nigrum L.), AM fungal assemblages having differential functions. forming associations (called mycorrhizas) with arbuscular Thus, in the present study, we aimed to verify if mycorrhizal mycorrhizal fungi (AMF) (Veiga et al. 2011; Massenssini weeds occurring under distinct cropping systems could act as et al. 2014) that are taxonomically classified either as a hosts of AMF during the off-season and in-season, and if AM phylum, Glomeromycota (Schüβler et al. 2001; Hibbett fungal assemblages would be affected by host identity and by the et al. 2007; Tedersoo et al. 2018), or as the sub-phylum CA components (cropping system). We hypothesised that (i) the Glomeromycotina, which together with Mortierellomycotina implementation of all three CA components leads to a promotion and Mucoromycotina, make up the phylum Mucoromycota of AMF through the increase of AMF colonisation and diversity (Spatafora et al. 2016; James et al. 2020; Li et al. 2021). within weed roots and that these traits are shaped by the identity As obligate mutualistic symbionts, AMF acquire nutrients of the hosting weed species; (ii) the identity of the hosting weed (e.g. phosphorus (P), nitrogen (N), sulphur (S)), through species would also affect the AM fungal colonisation and the extraradical mycelium, which acts as an extension of community composition and diversity of the weeds off-season; the host root system and transfer them to the host plant in (iii) off-season and in-season AM fungal composition of weeds exchange for photosynthetically assimilated carbon (4% to are related. Moreover, we aimed to dissect the potential effects 20% of total fixed C) (Smith and Read 2008; Gavito et al. of AM fungal diversity and root colonisation of maize and 2019). Thus, it is expected that the relationship that AMF weeds on maize grain yield. The elucidation of these topics is form through mycelial fungal networks with host plants, necessary to set up optimal weed control strategies with the goal such as the mycorrhizal weeds, will increase their growth of looking for an equilibrium between the control of damage and thus help them to proliferate through the acquisition caused by weeds and the conservation of biodiversity, ecosystem of nutrients and water (Wilson and Hartnett 1998; van der functioning and soil quality. Heyde et al. 2017). However, the AMF-weed interaction might not be of the mutualistic type, and this is the case of some ruderal plants, including several agricultural Material and methods suppressive weeds that respond negatively to AM fungal colonisation (Vatovec et al. 2005; Veiga et al. 2011). Some Experimental field locations non-mycorrhizal and mycorrhizal weeds exhibited growth suppression induced by AMF (i.e. reduced biomass, growth The experiment was done at two sites namely the Dom- rate and survival) through direct antagonistic effects, such boshawa Training Centre (DTC) and the University of Zim- as fungal parasitism and defence response of plants, and babwe (UZ). The geographical location and climate of the indirect effects in the triple interaction of AMF-weed-crop two locations are given in Table 1. Soil sampling was carried species by benefiting the associated mycorrhizal crop (Qiao out in November 2018 before maize sowing. Soil properties 1 3 Biology and Fertility of Soils (2022) 58:917–935 919 Table 1 Geographical location, Geographic location, soil characteristics and Location soil characteristics, and climate climate at the Domboshawa Training Domboshawa Training Centre University of Zimbabwe Centre (sandy location) and the University of Zimbabwe (clay Latitude 17.62° S 17.73° S location) Longitude 31.17° E 31.02° E Altitude (m asl) 1560 1503 Clay (g kg−1) 220 400 Sand (g kg−1) 730 390 Organic carbon (C) (g k g−1)a 7.3 16.8 Soil pH (0.01 M C aCl2)b 4.5 4.9 Soil total nitrogen (N) (g k g−1)c 0.6 2.3 Soil type Sandy clay loam Clay Soil classificationd Arenosols Rhodic Lixisols Average annual temperature (°C) 18.8 18.6 a Organic C was determined using the Walkley–Black wet combustion method (Nelson and Sommers 1982) b Soil pH was determined using the calcium chloride method (McLean 1982) c Total N was determined using the macro Kjeldahl digestion procedure (Bremner and Mulvaney 1982) d IUSS Working Group WRB (2015) and the corresponding analytical methods are given in iv. Conventional tillage plus mulch and rotation Table 1. The experiments at the two locations started in the (CT + M + R)—land preparation was done through summer crop growing season of 2013, and in this study, we digging with a hand hoe to simulate ploughing, maize report data collected in the 2019 growing seasons only. The was rotated with cowpea in 1-year rotations and crop climate of the two sites is classified as warm temperate with residues were retained on the soil surface at a rate of dry winters and hot summers (Kottek et al. 2006). Between 2.5 t ha−1 at the beginning of the season. the two sites, DTC had the highest average daily temperature v. No-tillage (NT)—no soil inversion was done, and of 29.0 °C and rainfall of 630 mm, whilst UZ had an average maize was sown as a monocrop in either riplines cre- daily temperature of 27.5 °C and rainfall of 383 mm during ated using a Magoye (DTC) or basins (UZ) created the study period (Fig. S1). afterwards. Crop residues were removed from the field after harvesting. vi. No-tillage plus mulch (NT + M)—no soil inver- Experimental set‑up and crop management sion was done, and maize was sown as a monocrop in either riplines created using a Magoye (DTC) or The experiments consisted of eight treatments (referred to as the basins (UZ) created afterwards. Crop residues were cropping system hereafter), which were arranged in a randomised retained on the soil surface at a rate of 2.5 t h a−1 at the complete block design (RCBD), and these were as follows: beginning of the season. vii. No-tillage plus rotation (NT + R)—land preparation i. Conventional tillage (CT)—land preparation was done and crop residue management were done as in treat- through digging with a hand hoe to simulate plough- ment (v), maize was sown in rotation with cowpea and ing. Maize was sown as a monocrop in either riplines crop residues were removed at harvest. created using a Magoye (DTC) or basins (UZ) created viii. No-tillage plus mulch and rotation (NT + M + R); afterwards. Crop residues were removed from the field referred to as CA herein—land preparation was done after harvesting. as in treatment (v), maize was rotated with cowpea in ii. Conventional tillage plus mulch (CT + M)—land 1-year rotations and crop residues were retained on the preparation and crop sowing as in treatment (i). Crop soil surface at a rate of 2.5 t ha−1 at the beginning of residues were retained on the soil surface at a rate of the season. 2.5 t ha−1 at the beginning of the season. iii. Conventional tillage plus rotation (CT + R)—land All treatments were replicated four times. For treatments preparation was done as in treatment (i), but maize involving rotation, plots were split into half and maize was rotated with cowpea (Vigna unguiculata (L.) was sown in a 1-year rotation with cowpea, with phases Walp.) in 1-year rotations. of the rotation present in each year. The in-plots measured 12 m × 6 m (72 m2). The maize rows were spaced at 90 cm 1 3 9 20 Biology and Fertility of Soils (2022) 58:917–935 and the maize plants at 25 cm. The cowpea rows were spaced all cropping systems at UZ, whereas at DTC, no common at 45 cm whilst the cowpea plants were spaced at 25 cm. The species were found across cropping systems. For both the maize population was 44,444 plants h a−1 whilst the cowpea pre-season and anthesis samplings, we randomly collected population was 88,888 plants ha−1. Both maize and cow- the roots of four plants (replicates) for each species at both pea received a basal fertiliser in the form of compound D locations (Fig. S2). At anthesis, the four plants were col- (7:14:7 NPK) at the rate of 11.6 kg N  ha−1, 10.1 kg P h a−1 lected from each plot (four replicates per cropping system), and 9.6 kg potassium (K) ha−1 at sowing, and maize further and then mixed to form one composite sample per plot. For received a top-dressing fertiliser in the form of ammonium maize, we assessed if cropping systems influenced the per- nitrate (NH4NO3) at the rate of 46 kg N  ha−1, split applied 4 centage of AM fungal root colonisation by sampling four and 7 weeks after sowing. Weeds were controlled by spray- plants from each plot at the anthesis stage, i.e. the pollina- ing glyphosate (N-(phosphono-methyl) glycine) at the rate tion stage (R1). The percentage of AM fungal root coloni- of 1.025 L active ingredient h a−1 at the beginning of the sation and root length containing arbuscules and vesicles season. Weeds were then manually controlled using hand were assessed using the magnified intersections method of hoes whenever weeds were 10 cm tall or 10 cm in diameter McGonigle et al. (1990) after clearing and staining (Phillips for stoloniferous weeds. and Hayman 1970). Details are given in Section 1 of the Supplementary Materials and Methods. Maize yield assessment Molecular analysis For each plot, maize plants were harvested from four rows that were 5 m long (i.e. an area of 18 m2). Maize cobs Plant DNA was extracted from 0.02 g of fine roots of the and stover were separated and a fresh weight of 10 cobs weeds, collected at pre-season and anthesis (pre-season: a was recorded. These cobs were air-dried for 4 weeks and total of 20 samples, four replicates per five plant species; reweighed for dry weight. The moisture content of the grain anthesis: 48 samples, three replicates per two plant species was determined, and the yield was expressed at 12.5% mois- per eight cropping systems only at the UZ location), using ture content. Maize stover was dried, and the stover was the DNeasy® Plant Mini Kit (QIAGEN, Hilden, Germany), determined on a dry weight basis. following the manufacturer’s instructions. Taking into con- sideration the patchy distribution of AMF within roots, AM fungal root colonisation of weeds and maize weed root fragments, i.e. 20 mg used for DNA extraction, and intraradical AMF diversity and community were chosen by randomly sampling from fine roots and then composition selecting those having good AM fungal colonisation. Root pieces (2–3 cm long) were mounted on slides in water and Assessment of AM fungal root colonisation of maize observed under a Zeiss Jenamed2 microscope with tungsten and weeds and UV lamps. Filter combinations used for fluorescence microscopy were BPF510 Excitation BPF475 (× 3)/Barrier To assess if weeds can act as alternative hosts of AMF dur- G247, G245 (Ames et al. 1982; Merryweather and Fitter ing the off-season (pre-season; which was at 30 days before 1991). Root samples showing a detectable level of autofluo- crop sowing) and during the season (at 60 days after crop rescence were selected for DNA extraction. The extracted sowing, which corresponds to the silking stage (R1); hereaf- genomic DNA was quantified by a Nanodrop spectropho- ter indicated as anthesis) and to also assess the effect of host tometer (Thermo Fisher Scientific, Vantaa, Finland) and identity (weed species) and cropping system, we determined then stored at − 20 °C for further analyses. The DNA was AM fungal root colonisation and molecularly characterised amplified using an amplicon-specific polymerase chain the diversity and community compositions within the roots reaction (PCR). A two-step nested PCR approach was used of weeds. For the pre-season, we identified five previously with two primer pairs to amplify the small subunit riboso- reported mycorrhizal weed species that were growing at the mal RNA (SSU) fragments. In the first step, the forward edges of the experimental field in the winter dry period. primer AML1 (5′-ATC AAC TTT CGA TGG TAG GAT These weed species were as follows: Bidens pilosa L. (Aster- AGA-3′) and reverse primer AML2 (5′-GAA CCC AAA aceae) (BIDPI), Cynodon nlemfuensis Vanderyst (Poaceae) CAC TTT GGT TTCC-3′) (Lee et al. 2008) were used, and (CYNNL), Erigeron sumatrensis Retz. (ERISU), Melinis in the second step, the forward primer WANDA-ill (5′-TCG repens (Willd.) Zizka (Poaceae) (RHYRE) and Richardia TCG GCA GCG TCA GAT GTG TAT AAG AGA CAG scabra L. (Rubiaceae) (RCHSC). For the anthesis assess- ANN NHN NNW NNN HGC AGC CGC GGT AAT TCC ment, we identified the species from the list of weeds col- AGCT-3′) (Dumbrell et al. 2011) and reverse primer AML2- lected in the pre-season present in all experimental plots. ill (5′-GTC TCG TGG GCT CGG AGA TGT GTA TAA Two species (i.e. BIDPI and RCHSC) were common to GAG ACA GGA ACC CAA ACA CTT TGG TTT CC-3′) 1 3 Biology and Fertility of Soils (2022) 58:917–935 921 (Lee et al. 2008) were used (the adaptors for the Illumina using the USEARCH ‘cluster_otus’ command. During reaction are in bold type). Both PCR reactions were per- the process, chimeric sequences and singletons were also formed in a 25-µl volume using 1 µl of the genomic DNA removed. The resulting OTUs were assigned to virtual taxa template (undiluted DNA at 13.5 ± 1.2 ng µl−1), 1.25 U µl−1 (VT) using the MaarjAM database (https://m aarja m.b otany. of GoTaq® Hot Start Polymerase (Promega Corporation, ut.e e). All representative sequences (143 in total) were sub- WA, USA), 0.2 µM of each primer, 0.2 mM of dNTPs, 2 mM mitted to the NCBI sequence read database (submission of MgCl2 and 1 × reaction buffer. The PCR cycle for both number SUB10794739), and these correspond to the acces- steps involved an initial denaturation at 95 °C for 2 min fol- sion numbers OM049043–OM049185. The representative lowed by 25 cycles of 94 °C for 30 s, 59 °C for 45 s, 72 °C sequences were aligned using the MAFFT online service for 1 min 30 s and 72 °C for 10 min. All PCR reactions (Katoh et al. 2019), and a Neighbour-Joining tree was built were carried out using an S1000 Thermal Cycler™ (Bio- using MEGA11 (Tamura et al. 2021), following the boot- Rad, Hercules, CA, USA). The quality of the PCR products strap test of phylogeny with 1000 bootstraps. The substitu- was checked by gel electrophoresis using 2% agarose gel tion model used was the Kimura 2-parameter with uniform in 1 × TBE buffer and then purified with magnetic beads rates among sites, pairwise deletion and 7 threads. (Agencourt AMPure® XP, Beckham Coulter, USA) and freshly prepared 80% ethanol. The concentration of DNA was then quantified using fluorimetry (Invitrogen™ Qubit™ Calculations and statistical analyses 4 fluorometer) by the Qubit4™ 1 × dsDNA High-Sensitivity Assay Kit (Invitrogen, Thermo Fisher Scientific, CA, USA), Weed community diversity and AMF diversity in weed roots following the manufacturer’s instructions. The cleaned and quantified amplicons of each library were then adjusted to Data analyses were done separately for the clay and sand an equimolar ratio of 10 ng µl−1 for the addition of dual- locations. Community diversity of AMF within weed roots index barcodes using the Nextera® XT DNA library prepa- was also computed using Shannon’s diversity (H′), Pielou’s ration kit (Illumina Inc., CA, USA). For more information evenness (J') and richness (S) based on VT counts. Shannon- on the dual-indexing procedure, please refer to Section 2 Weiner index (Shannon and Weaver 1949) was calculated as: of the Supplementary Materials and Methods. The gener- S ated metabarcoding libraries were run on an Illumina MiSeq ∑ ( ) � H − P InP sequencer at the University of York (UK), loading a 12-pM i i i=1 final library concentration with 20% PhiX library spike-in (Illumina) and an Illumina MiSeq V3 600 cycle sequencing where H′ is the Shannon-Weiner diversity index, Pi is the kit. proportion of individuals belonging to the ith VT and S is the total number of VT. Bioinformatics analyses Pielou’s evenness index (Pielou 1969) was calculated as the ratio of observed diversity to maximum diversity as Raw sequence data were processed and analysed using the follows: QIIME2 (2018.11) pipeline and plugins (Bolyen et al. 2019). � J = � H ∕ � H = H ∕lnS Demultiplexed forward and reverse paired-end reads were max joined using the ‘-fastq_mergepairs’ of the USEARCH where H′ is the Shannon’s diversity, and Hmax or lnS is the plugin (Edgar 2010). Out of the 1,867,193 reads exposed to maximum Shannon diversity in which all present VT appear merging, 89% (1,662,165 reads) were successfully merged in equal abundances for a community, and S is the number and 47% (874,286 reads) were aligned with zero differences. of observed VT. Evenness values range between 0 and 1, Primer sequences were trimmed off from the sequences representing absolute dominance and equal VT abundance, using the cutadapt plugin 1.18 with Python 3.5.5, and respectively. 1,659,726 valid sequences were obtained after optimisation. AM fungal data were assessed for normality and where The average read length was approximately 250 base pairs necessary, data were fourth-root transformed before further (bp) based on the maximum expected error (MaxEE). The analyses. The effect of host weed species identity, cropping command USEARCH ‘fastq_eestats2’ was used to check systems and their interaction (treated as fixed factors) on H′, sequence quality and, based on the percentage MaxEE, reads J′ and S were assessed using linear mixed models using the were truncated at the drop-off point of 250 bp using the ‘lme4’ package (Bates et al. 2015) in the R environment (R USEARCH ‘fastq_filter’ command. Quality-filtered reads Core Team 2021). Replicates were included in the analyses were dereplicated using the USEARCH ‘fastx_uniques’ as a random factor. The means and standard errors of back- command, and operational taxonomic units (OTUs) were transformed data were reported herein. F-tests were used to generated by clustering reads at a 97% similarity threshold test the significance of fixed effects, and where means were 1 3 9 22 Biology and Fertility of Soils (2022) 58:917–935 significantly different, the mean comparison procedure was (called the RELATE test). All the multivariate analyses were used to contrast them based on the Tukey’s tests (P < 0.05) performed using Primer 7 with PERMANOVA + software using the ‘emmeans’ package (Lenth 2019) in R. (Anderson 2001; Clarke et al. 2014). Finally, Venn diagrams were constructed to show the shared and unique AM fungal Weed and maize root colonisation by AMF taxa within the roots of BIDPI and RCHSC weed species collected at UZ at pre-season and anthesis (BIDPI pre- For AM fungal root colonisation percentage of weeds and season vs. BIDPI anthesis; RCHSC pre-season vs. RCHSC maize, data were checked for normality and fourth-root anthesis) and between BIDPI and RCHSC at anthesis. The transformed before analysis. For weed data, mixed models diagrams were built using Venny 2.1 (Oliveros 2015). were used to assess the effects of host species identity, crop- ping systems and their interaction (fixed effects) on the per- Relationship between cropping systems, weed AMF centage of AM fungal colonisation, and root length contain- diversity, maize AM fungal colonisation, weed AM ing arbuscules and vesicles. For maize data, mixed models fungal colonisation and maize grain yield were used to assess the effects of cropping systems on the colonisation rate. In both cases, replicates were included To assess the effect pathway of weed AMF diversity param- as random factors. The significance of the fixed effects was eters (H′, J′, and S′, maize AM fungal colonisation, and weed tested using F-tests and where means were significantly AM fungal colonisation on maize grain yield, we used piece- different, they were contrasted using a mean comparison wise structural equation modelling (pSEM). The pSEM anal- procedure following Tukey’s tests (P < 0.05). However, the ysis was based on multiple regression and was done using back-transformed means and standard errors were reported. the ‘piecewiseSEM’ package in R (Lefcheck 2016). Models were fit using linear models, and variables were standard- Intraradical AM fungal community composition ised for the effects to be directly comparable, and for each pathway, a standardised coefficient (λ) was estimated. In the To analyse the effect of weed species identity and cropping models, we also calculated the covariance of weed AM fun- system on AMF community structure in weed roots, we gal H′ and J′; weed AM fungal H′ and S; and RCHSC and used type III permutational multivariate analysis of vari- BIDPI colonisation rate. Model fits were estimated by the ance (PERMANOVA). As a semiparametric multivariate Fisher’s C test. test, PERMANOVA generates pseudo-F ratios and P values using the Monte Carlo permutation P(MC) test by permu- tating the resemblance measures (Anderson 2001). In our Results analyses, 999 permutations were employed. AM fungal species relative abundances were fourth-root Effect of cropping systems and host identity on AM transformed, and a resemblance matrix was constructed fungal root colonisation of weeds and on AM fungal based on the Bray–Curtis dissimilarity index (Bray and Cur- diversity and community structure in weed roots tis 1957) before carrying out PERMANOVA. Where group differences in community composition were detected, simi- After blast matching of OTUs against the MaarjAM data- larity percentage analysis (SIMPER) was done to detect the base, 143 AM fungal virtual taxa (VT) were retrieved, and species responsible for 70% of the differences by calculating these belonged to seven families, namely Acaulosporaceae, the percentage contribution of the species to the total effects. Archaesporaceae, Claroideoglomeraceae, Diversispo- Further, we carried out a permutation test for homogeneity raceae, Gigasporaceae, Glomeraceae and Paraglomer- of multivariate dispersions (PERMDISP) on each significant aceae (Fig. S3). Of all the VT, 57% belonged to the Glomer- factor level. This test is used as a measure of multivariate aceae, 10% to Acaulosporaceae and 10% to Gigasporaceae beta diversity to check whether the significant group differ- (Fig. S3). ences observed in PERMANOVA were also not influenced At pre-season, the percentage of AM fungal root coloni- by differences in the dispersion of group objects from the sation and of root length containing arbuscules and vesicles group centroid (alpha diversity). significantly differed among weed species at DTC (sandy Principal coordinates analysis (PCoA) was then per- location) (Fig. 1a). CYNNL showed an AM fungal root formed to visualise relevant patterns in the data. Finally, we colonisation significantly lower than ERISU and RCHSC used the RELATE procedure based on Spearman rank corre- (12% vs 25%). Arbuscules and vesicles were higher within lation to test if there was a relationship between the AM fun- RHYRE as compared to other weed species, whilst RCHSC gal communities observed in the roots of BIDPI and RCHSC had the lowest percentage of arbuscules and CYNNL and collected during the pre-season and the anthesis periods at ERISU the lowest percentage of vesicles (Fig. 1a). At UZ UZ using the Primer 7 (with PERMANOVA +) software (clay location), CYNNL and RHYRE showed a significantly 1 3 Biology and Fertility of Soils (2022) 58:917–935 923 Fig. 1 Effect of weed species identity on the percentage of arbus- Organization coding are BIDPI, Bidens Pilosa; CYNNL, Cynodon cular mycorrhizal (AM) fungal root colonisation and of root length nlemfuensis; ERISU, Erigeron sumatrensis; RHYRE, Melinis repens; containing arbuscules and vesicles of five weed species collected RCHSC, Richardia scabra. Abbreviations of the cropping systems are from the experimental boundaries at the Domboshawa Training Cen- CT, conventional tillage; CT + M, CT plus mulch; CT + R, CT plus tre (DTC; sandy location) (a) and the University of Zimbabwe (UZ; crop rotation; CT + M + R, CT plus mulch and rotation; NT, no-till- clay location) (b) during the off-season (pre-season). AM fungal root age; NT + M, NT plus mulch; NT + R, NT plus rotation; NT + M + R, colonisation as affected by different cropping systems (c), weed spe- NT plus mulch and rotation. Values are means ± SE of four replicates cies (d), vesicles as affected by weed species (e) and arbuscules as for each cropping system per season. Bars with different letters are affected by the interaction of cropping system and weed species (f) significantly different from each other based on P values: **P < 0.01; at UZ during the in-season (anthesis) in 2019. Abbreviations of the *P < 0.05 (see Table 2) weed species based on European and Mediterranean Plant Protection lower AM fungal root colonisation (18%) compared with of crop rotation alone (CT + R) (30%) (Fig. 1c). Moreover, BIDPI and ERISU (30%) (Fig. 1b). Arbuscules differed RCHSC had a higher AM fungal root colonisation than among weed species, with the RCHSC having the highest BIDPI (45% vs. 36%) (Fig. 1d). Percentage of vesicles dif- percentage and RHYRE the lowest (Fig. 1b). At anthesis, fered among the two weed species with the RCHSC having we found only two plant species (BIDPI and RCHSC) over higher colonisation (11%) as compared to that of BIDPI’s the five pre-season mycorrhizal weeds that were common (8%) (Fig. 1e). The interaction of cropping system and weed to all cropping systems at UZ, whereas no common plant identity had a significant effect on the percentage of arbus- species were found at DTC. Indeed, at the UZ location, the cules, with RCHSC under CT + R and NT + R having the cropping system and weed identity had significant effects on highest rate (Fig. 1f). AM fungal root colonisation (Figs. 1c, d). The implementa- At pre-season, the AM fungal communities showed sig- tion of CT with mulching alone (CT + M) or together with nificantly different H′ and J′ among weed species at both crop rotation (CT + M + R) resulted in a higher AM fungal locations (Table 2). The weed species BIDPI consistently root colonisation (50%) with respect to the implementation exhibited the highest H′ and J′ (Figs. 2a–d). At anthesis, 1 3 9 24 Biology and Fertility of Soils (2022) 58:917–935 Table 2 Effect of weed identity on arbuscular mycorrhizal fungal (Figs. 3b, d, f). The SIMPER analyses revealed that AM fungal (AMF) species Margalef richness (S), Shannon’s diversity (H′), and taxa, such as Glomus sp. VTX00280, explained most of the AM Pielou evenness (J′) within roots of plants collected at the Dom- boshawa Training Centre (DTC; sandy location) and the University fungal community differences in most of the weed species col- of Zimbabwe (UZ; clay location) along the experiment borders (off- lected at pre-season in both locations (Figs. 4a, b). However, at season: called ‘pre-season’) and within the UZ plots at maize anthesis DTC, Glomus sp. VTX00092 and Glomus sp. VTX00264 also (in-season: called ‘anthesis’) showed high contributions to the AM fungal community dif- Location and period Source DF H′a J′ S ferences (up to 30% in the CYNLE) (Fig. 4a). As for the weeds collected during the anthesis period at the UZ, although the AM DTC at pre-season Weed identityb 4 10.7** 4.6* 2.3 fungal taxa, such as Acaulospora sp. VTX00028, Claroideoglo- UZ at pre-season Weed identityb 4 6.8** 3.4* 2.3 mus sp. VTX00193, Dentiscutata sp. VTX00255, Gigaspora sp. UZ at anthesis Weed identityc 1 1.8 0.2 1.9 VTX00039 and Glomus sp. VTX00112, strongly contributed to UZ at anthesis Systemd 7 2.5* 1.1 2.9* the community differences among cropping systems (Fig. 4c). UZ at anthesis Weed identity × sys- 7 – – – tem An analysis comparing the AM fungal communities retrieved from the roots of the same weed species (i.e. BIDPI Effect of cropping system on S, H′, and J′ within the roots of weed and RCHSC) collected at pre-season with those collected at sampled at maize anthesis in the UZ location. F-values and degrees anthesis revealed different AM fungal community compositions of freedom (DF) were derived from linear mixed-effect models (Rho = 0.268; P > 0.05) (Fig. S4a). However, some AM fungal a F-values in bold were significantly different: **P < 0.01, *P < 0.05 taxa were common to the weed species. For example, in terms b Five mycorrhizal weed species: Bidens pilosa (BIDPI), Cynodon nlemfuensis (CYNNL), Erigeron sumatrensis (ERISU), Melinis of AM fungal composition, BIDPI sampled at pre-season had repens (RHYRE) and Richardia scabra (RCHSC) 36% of the same AM fungal taxa as BIDPI sampled at anthesis c Two weed species: Bidens pilosa and Richardia scabra (plant spe- (Fig. S4b). Similarly, RCHSC sampled at pre-season and anthe- cies belonging to the list of weeds collected in the pre-season and sis showed 42% of common AM fungal taxa (Fig. S4c). Finally, present in all experimental plots at maize anthesis; this occurred only the comparison between the AM fungal community composi- at the UZ location) tion of the two weed species at anthesis showed that 53% of d One season (2019) and eight cropping systems the retrieved AM fungal taxa were shared, 47% were unique to BIDPI, and no taxa were unique to RCHSC (Fig. S4d). whilst no differences were observed in terms of AM fungal diversity indices between BIDPI and RCHSC, the cropping Effect of cropping system on maize productivity system had a significant effect on H′ and S, with the NT + M and relationship with weed AMF diversity, maize system showing higher values at both locations as compared AM fungal colonisation and weed AM fungal with CT, CT + R and NT + M + R (Figs. 2e, f). Interestingly, colonisation under the NT + M system, AMF had also a higher S than under NT + M + R (Figs. 2e, f). Data on maize productivity and AM fungal root colonisa- Based on the PERMANOVA results, weed species col- tion has already been reported by Mhlanga et al. (2022) and lected at pre-season hosted different AM fungal communi- hence will not be reported herein, but we refer the reader to ties at both locations, whereas at anthesis, only the cropping the aforementioned paper. In this current analysis, we will system significantly shaped the AM fungal communities use the same data to relate to weed AMF diversity, maize (Table 3). This is also supported by the PCoA plots explaining AM fungal colonisation and weed AM fungal colonisation. 63% and 50% of the total variance in pre-season at DTC and Structural equation modelling resulted in an overall signifi- UZ, respectively, where the group centroids of the AM fungal cant fit (Fig. 5) (Fisher’s C = 30.63; P-value = 0.242; DF = 26). communities retrieved at pre-season within the roots of the The cropping system had a significant and positive influence weed species were clearly separated on the ordination space on weed AMF diversity (J′), weed AMF richness (J′), maize (Figs. 3a, c). Similarly, the PERMANOVA results at anthe- AMF root colonisation and colonisation of BIDPI and RCHSC sis were supported by the PCoA plot, explaining 41% of the (Fig. 5) (Table 4). All the investigated factors did not have a sig- total variance, in which the group centroids of the AM fungal nificant influence on maize grain yield. The cropping system communities retrieved in the cropping systems were clearly NT + M resulted in the highest path coefficient estimates for H, separated in the ordination space (Fig. 3e). This occurred S, and RCHSC, whilst CT + M had the highest coefficient for despite the high percentage of AMF common to all cropping BIDPI AMF root colonisation (Table 4). The cropping system systems (core community: 24% among CT-based systems and NT + M + R had the highest maize AMF root colonisation and 26% among NT-based systems; 60% between CT-based and grain yield path coefficients, whilst the CT systems had the least NT-based systems) (Fig. S4). For both pre-season and anthe- grain yield coefficient. Weed AMF diversity showed a positive sis data, the distances of the group object to their centroid correlation with weed AMF evenness (λ = 0.74) and weed AMF did not significantly differ, supporting similar alpha diversity richness (λ = 0.93). 1 3 Biology and Fertility of Soils (2022) 58:917–935 925 Fig. 2 Effect of species identity on arbuscular mycor- rhizal (AM) fungal community Shannon diversity (H′) and Pielou (J') evenness during the off-season (pre-season) at the Domboshawa Training Centre (DTC; sandy location) (a and b) and at the University of Zimbabwe (UZ; clay loca- tion) (c and d), and effect of cropping system on AM fungal Shannon diversity index (H′) and taxonomic richness (S) at UZ in-season (anthesis) (e and f). Abbreviations of the weed species based on European and Mediterranean Plant Protec- tion Organization coding are BIDPI, Bidens pilosa; CYNNL, Cynodon nlemfuensis; ERISU, Erigeron sumatrensis; RHYRE, Melinis repens; RCHSC, Rich- ardia scabra. Abbreviations of the cropping systems are CT, conventional tillage; CT + M, CT plus mulch; CT + R, CT plus crop rotation; CT + M + R, CT plus mulch and rotation; NT, no-tillage; NT + M, NT plus mulch; NT + R, NT plus rotation; NT + M + R, NT plus mulch and rotation. Values are means ± SE of four replicates for each cropping system per season. Bars with different letters are significantly different from each other based on P values reported in Table 2 Discussion AM fungal diversity during the off-season (pre-season) in an area surrounding the crop fields, when crops are absent, Host species identity and cropping system or during the season (anthesis) along with the crops. This determine AM fungal colonisation, diversity can signify the abundance of AMF in the soil (Barceló et al. and community structure within the roots of weeds 2020). The five mycorrhizal weeds collected during the off- season differed in the percentage of AM fungal root coloni- Despite the potential negative effect of weeds on crop pro- sation and of root length containing arbuscules and vesicles. ductivity and production costs, weeds can offer agroecosys- Similarly, at anthesis, RCHSC showed a higher AM fungal tem services beneficial to crops (MacLaren et al. 2020; El root colonisation than BIDPI. These differences may be Omari and El Ghachtouli 2021). One of the services that attributed to the host specificity of AMF, the selectivity or could be provided by weeds is hosting AMF and supporting mycorrhizal dependency of the host weeds (Eom et al. 2000; 1 3 9 26 Biology and Fertility of Soils (2022) 58:917–935 Table 3 Permutational multivariate analysis of variance (PER- experiment borders (off-season: called ‘pre-season’) and within the MANOVA) results for the effect of weed identity on the arbuscular UZ plots at maize anthesis (in-season: called ‘anthesis’) and PER- mycorrhizal fungal (AMF) community within roots of weed plants MANOVA results for the effect of cropping system on the AMF root collected at the Domboshawa Training Centre (DTC; sandy loca- community of weeds at maize anthesis in the UZ location tion) and the University of Zimbabwe (UZ; clay location) along Location Source DF Pseudo-F P(MC)a Explained variation (%) DTC at pre-season Weed identityb 4 4.848 0.001 56.19 UZ at pre-season Weed identityb 4 3.713 0.001 47.49 UZ at anthesis Weed identityc 1 0.943 0.469 − 0.42 UZ at anthesis Systemd 7 2.382 0.001 31.54 a P values based on Monte-Carlo permutational test, P(MC) b Five weed species: Bidens pilosa (BIDPI), Melinis repens (RHYRE), Cynodon nlemfuensis (CYNNL), Erigeron sumatrensis (ERISU), and Richardia scabra (RCHSC) c Two weed species: Bidens pilosa and Richardia scabra (plant species belonging to the list of mycorrhizal weeds collected in the pre-season and present in all experimental plots at maize anthesis; this occurred only at the UZ location) d Eight cropping systems Yang et al. 2012; Ciccolini et al. 2016; Säle et al. 2022). and mulching improved soil hydrothermal conditions (Lal This specificity was also revealed by the different AM fun- 2000) and positively affected fungal diversity (Brito et al. gal community compositions and diversity indices observed 2012; Piazza et al. 2019; Pellegrino et al. 2020). Moreo- within the weed root systems at pre-season. However, the ver, at anthesis, in our study, crop rotation reduced the AM unexplained variance in the PCoA analyses highlights that fungal diversity within the roots of BIDPI and RCHSC. other factors, in addition to host specificity, play a role in Earlier studies have demonstrated that crop rotation shaping AM fungal community composition. Glomus sp. increased or did not modify AM fungal diversity (Oehl VTX00280, explaining most of the AM fungal community et al. 2003, Hijri et al. 2006). The positive effect on AMF differences, suggests for the first time that host specificity diversity was found with an extensive crop rotation includ- during dry periods is mainly driven by a unique VT. This ing a perennial grass-clover mixture (Oehl et al. 2009). indicates that under stress conditions (off-season), the differ- Thus, the decrease of AM fungal diversity we found in ential supply of host C to the most functional VT promotes the systems with crop rotation compared to maize mono- its prevalence in roots, improving plant tolerance against dry cropping could be linked to the cowpea selection for a few conditions (Kiers et al. 2011; Omirou et al. 2013). Accord- dominant AM fungal species compared to maize (John- ingly, it was previously demonstrated that some species of son et al. 2013; Alaux et al. 2021). Since different CA AMF are less sensitive to water stress than others (Baha- practices result in different weed communities (Mhlanga dur et al. 2019). However, in the absence of drought stress, et al. unpublished results), this gives different AM fungal the host specificity observed in BIDPI and RCHSC at pre- communities a higher chance of being promoted within a season was no longer detectable in terms of the AM fungal system. Indeed, tillage alters the seedbank and its vertical community composition during the cropping cycle. distribution, the germination, predation and viability and Between the two species (BIDPI and RCHSC) that were dispersal of weed seeds and the weed community com- sampled in the plots at anthesis, the highest AM fungal position and diversity (Nichols et al. 2015). Moreover, colonisation was observed in the CT-based systems, either crop residues can affect seed germination via physical with mulch alone or with mulch and rotation. Despite that and chemical changes in the seed environment, whilst intensive tillage was previously observed to reduce AM rotating crops change the selection pressures, precluding fungal root colonisation in different crops (Castillo et al. one weed from repeatedly establishing itself. Overall, the 2006; van der Heyde et al. 2017; Mhlanga et al. 2022), our implementation of mulching either in NT or CT systems data suggest that mulching preserves soil moisture and modified the AM fungal community composition as com- promotes the proliferation of AMF, leading to increased pared to the other systems, despite the high percentage root colonisation (Wilkes et al. 2021). Moreover, mulching of the core AM fungal taxa. Thus, mulching also plays a added to NT increased AM fungal diversity and richness. crucial role in shaping AM fungal assemblages, as well This is in agreement with Lu et al. (2018) observing that as in improving AM fungal colonisation and diversity. In NT with crop straw retention promoted soil AM fungal accordance, mulching has recently been highlighted as a diversity with respect to CT. As previously reported, NT major driver of improving the stability and resilience of 1 3 Biology and Fertility of Soils (2022) 58:917–935 927 Fig. 3 Principal coordinates analysis (PCoA) based on Bray– Curtis distance dissimilarity of fourth-root transformed AMF community relative abundances. Plots show the AM fungal dif- ferences among weed species and cropping systems at the Domboshawa Training Centre (DTC; sandy location) (a) and at the University of Zimbabwe (UZ; clay location) (c and e) (see Table 2). Permutational dis- persion (PERMDISP) tests on the same data matrices at DTC and UZ (b and d weed species; f cropping system) represented by the distances of the objects from the centroid and standard error (SE). Abbreviations of the weed species based on European and Mediterranean Plant Protec- tion Organization coding are BIDPI, Bidens pilosa; CYNNL, Cynodon nlemfuensis; ERISU, Erigeron sumatrensis; RHYRE, Melinis repens; RCHSC, Rich- ardia scabra. Abbreviations of the cropping systems are CT, conventional tillage; CT + M, CT plus mulch; CT + R, CT plus crop rotation; CT + M + R, CT plus mulch and rotation; NT, no-tillage; NT + M, NT plus mulch; NT + R, NT plus rotation; NT + M + R, NT plus mulch and rotation. Bars with different letters are significantly different based on the reported P-permutational values (Pperm) maize-based rainfed systems in southern Africa (Kodzwa large variability of response to tillage. As an example, et al. 2020; Mhlanga et al. 2021). Most of the AM fun- Glomus sp. VTX00112 was abundant under the NT-based gal taxa retrieved from weed roots during the crop cycle systems and was rare under CT systems, whereas Glomus belonged to Glomeraceae. Indeed, species of this family, VTX00132 showed the opposite behaviour. These results such as Funneliformis mosseae, have a short life cycle that support the high functional variability within the fam- may reduce their sensitivity to discontinuous plant pres- ily Glomeraceae, as previously reported in some studies ence and disruption of the extraradical mycelia by frequent (Avio et al. 2006; Munkvold et al. 2004). In contrast to tillage (Oehl et al. 2003; Pellegrino et al. 2020). However, previous studies that found Gigasporaceae propagating our data evidenced among the retrieved Glomus taxa a from intact mycelia to be abundant under NT systems but 1 3 9 28 Biology and Fertility of Soils (2022) 58:917–935 1 3 Biology and Fertility of Soils (2022) 58:917–935 929 ◂Fig. 4 Similarity percentages analysis (SIMPER) identifying the Since the off-season and in-season AM fungal compo- arbuscular mycorrhizal (AM) fungal taxa that were responsible for sition of the two weeds occurring in all cropping systems the AM fungal community differences among weed species at the Domboshawa Training Centre (DTC; sandy location) (a) and the Uni- were similar, this finding supports the fact that weeds versity of Zimbabwe (UZ; clay location) (b) and among cropping sys- functionally host AMF during the dry periods, playing tems at UZ (c). The listed species explain approximately 70% of the key roles in the proliferation of AMF during the cropping contribution. Abbreviations of the weed species based on European cycle. Thus, weeds could be crucial for the maintenance and Mediterranean Plant Protection Organization coding are BIDPI, Bidens Pilosa; CYNNL, Cynodon nlemfuensis; ERISU, Erigeron of an active pool of beneficial fungi able to colonise and sumatrensis; RHYRE, Melinis repens; RCHSC, Richardia scabra. connect plants in cropping systems, potentially stimulat- Abbreviations of the cropping systems are CT, conventional tillage; ing crop defence pathways (Nerva et al. 2022) under the CT + M, CT plus mulch; CT + R, CT plus crop rotation; CT + M + R, drought conditions characterising the study area. This CT plus mulch and rotation; NT, no-tillage; NT + M, NT plus mulch; NT + R, NT plus rotation; NT + M + R, NT plus mulch and rotation evidence reinforces the ecological role played by weeds in the agroecosystem. In addition, the similarity in com- position between off-season (sampled at the edge of the scarce under intensive tillage (Daniell et al. 2001; Pel- experimental field) and in-season weeds (sampled inside legrino et al. 2020), Gigaspora sp. VTX00039 largely the plots) supports the fact that the common mycorrhiza occurred under conventionally tilled systems. Our results network is able to connect plants and transfer nutrients can be supported by some studies (Schalamuk and Cabello and signals at a long distance (Barto et al. 2012; Bennett 2010; Hart and Reader 2004) stating that Gigasporaceae et al. 2016). is less sensitive to soil disturbance than Glomeraceae Since weeds inside plots were controlled by glyphosate because after disturbance, some hyphal fragments lose at the beginning of the season, the residual effect of glyphosate viability due to cytoplasmatic leakage, whereas spores may have had an effect on the weed community. Glyphosate are not greatly affected, and Gigasporaceae colonise roots in soil dissipates almost completely 30 days after application primary from spores. Thus, there is still conflicting evi- under high temperatures, which are normally recorded in our dence on the ability of Glomeromycota families to use study area (Bento et al. 2016). However, the main metabolite of propagules type and to reconnect once they are disrupted glyphosate, aminomethylphosphonic acid (AMPA), can persist by tillage (De La Providencia et al. 2005). in soil, as has been detected at 20% of the applied glyphosate Our study applied a nested PCR approach using the rate after 30 days of glyphosate application (Bento et al. 2016; primer pair AML1/AML2 and the primer pair WANDA- Guijarro et al. 2018). Following Guijarro et al. (2018), the ill/AML2-ill (Lee et  al. 2008). Recently, Suzuki et  al. glyphosate exposure history affected the rate of persistence as (2020) evaluated primer pairs’ suitability for AM fungal the herbicide was degraded rapidly with long-term exposure community assessment by comparing five approaches, and slowly when glyphosate was never applied to the soil. three targeting the 18S rRNA gene (one using the AM Thus, in our experiment, the persistence of glyphosate is fungal-specific primer pair AMV5.4NF/AMDGR (Sato likely to be negligible at 60 days after maize sowing, when et al. 2005); a nested PCR approach using the AM fungal- in-season weeds were sampled. In contrast, the metabolite specific primer pair AML1/AML2 (Lee et al. 2008) and AMPA can be detected in the soil after long-term glyphosate the N-AMV5.4NF/AMDGR primer set; a nested PCR application, although its concentration at in-season sampling approach using the AM fungal-specific primer pair AML1/ time can be hypothesised to be low (degradation time for 90% AML2 and the NS31/AML2 primer set (Simon et al. 1992; of the initial concentration (DT90) between 88 and 148 days), Lee et al. 2008)), one targeting the 28S rRNA gene (using according to Bento et al. (2016) and Guijarro et al. (2018), the AMF-specific primer pair Glo454/NDL22 (van Tuinen and not affected by tillage treatments, as reported by Okada et al. 1998)) and one targeting the ITS region (using the et al. (2019). Moreover, since AMF can sporulate already at fungal universal primer pair ITS1-F KYO1/ITS2-KYO1 the early plant growth stage by draining host C during the (Toju et al. 2012)). AM fungal detection rate ranged from plant development (Harinikumar and Bagyaraj 1989), the 98% with nested AMV5.4NF/AMDGR to 0.04% with ITS1- manual removal of weeds inside the plots at the vegetative F KYO1/ITS2-KYO1 (Suzuki et al. 2020). For the NS31/ phase that does not involve the complete elimination of all AML2 approach, similar to the one applied in this study, the mycorrhizal roots is not likely to affect weed-mediated AM AM fungal detection rate was high (87%) and gave a high fungal propagule abundance in soil. This is also confirmed number of unique sequences, great phylogenetic diversity by a study comparing the effect of different methods of weed and low evenness. Moreover, AMF community composition control, including manual weeding, on spore number and AMF detected by single AMV5.4NF/AMDGR and NS31/AML2 root colonisation of several weeds, including B. pilosa and was relatively similar at the genus level (Suzuki et al. 2020), maize (Ramos-Zapata et al. 2012). although nested PCR has been shown to affect AMF com- munity analysis (Yu et al. 2015). 1 3 9 30 Biology and Fertility of Soils (2022) 58:917–935 Fig. 5 Structural equation model (SEM) (path analysis) showing the effect of cropping systems on AMF diversity (Shannon diversity (H′)), AMF evenness (Pielou evenness (J′)) and AMF richness (Margalef richness (S) in weed roots and maize, Bidens pilosa (BIDPI) and Richardia scabra (RCHSC) AM fungal root colonisation on grain yield at the University of Zimbabwe (UZ; clay loca- tion). The black lines represent positive influence, whilst the red lines represent negative influ- ence. Solid lines and dashed lines represent significant (P < 0.05) and non-significant (P > 0.05) influences, respec- tively. Standardised path coef- ficients are reported for each effect pathway Table 4 Standardised path coefficients of cropping system effect on colonisation, Bidens pilosa AMF root colonisation, and maize grain weed AMF diversity (H′), weed AMF evenness (J′), weed AMF rich- yield at University of Zimbabwe (UZ; clay location) ness (S), maize AMF root colonisation, Richardia scabra AMF root Cropping system Standardised coefficientsa Weed AMF Weed AMF Weed AMF Maize AMF root Richardia scabra Bidens pilosa AMF Maize grain diversity (H′) evenness (J′) richness (S) colonisation (%) AMF root colonisa- root colonisation (%) yield (kg tion (%) ha−1) CT 2.88 bc 0.91 a 24.33 bc 41.98 b 30.81 c 36.02 ab 2314.85 c CT + M 3.08 abc 0.91 a 30.33 abc 43.16 b 49.60 ab 51.00 a 2643.66 b CT + R 2.75 c 0.90 a 21.33 c 51.61 ab 37.00 c 22.38 b 3329.46 ab CT + M + R 3.30 ab 0.93 a 34.67 ab 55.09 ab 54.36 a 44.61 ab 3211.39 ab NT 3.14 abc 0.94 a 28.33 abc 49.31 ab 42.54 b 31.31 b 2813.91 b NT + M 3.35 a 0.94 a 36.00 a 47.21 b 55.65 a 35.60 ab 2552.71 b NT + R 2.94 abc 0.93 a 24.00 c 68.13 a 40.08 b 28.68 b 3615.74 ab NT + M + R 2.71 c 0.90 a 20.33 c 56.16 a 45.34 ab 38.06 ab 3935.50 a P value P < 0.001 ns P < 0.001 P < 0.001 P < 0.001 P < 0.01 P < 0.01 a Standardised coefficient estimates with different letters are significantly different from each other based on Tukey’s post hoc tests Abbreviations of the cropping systems are: CT conventional tillage, CT + M CT plus mulch CT + R CT plus crop rotation, CT + M + R CT plus mulch and rotation, NT no-tillage, NT + M NT plus mulch, NT + R NT plus rotation, NT + M + R NT plus mulch and rotation Relationship between cropping systems, weed AMF diversity, evenness, richness and AM fungal colonisation) diversity, maize AM fungal colonisation weed AM and maize AM fungal root colonisation, contrary to our fungal colonisation and maize grain yield hypothesis, all these traits did not significantly influence maize grain yield. Although we expected that the diversity As expected, weed AMF diversity showed a positive cor- of AMF in weed roots would result in the improvement of relation with weed AMF evenness and weed AMF richness the yield of associated maize plants through the improve- (van der Heijden et al. 1998). However, although the crop- ment of AMF-mediated traits, this effect may have been ping system directly affected all AMF traits in weeds (i.e. masked by other factors. Mycorrhizal weeds also benefit 1 3 Biology and Fertility of Soils (2022) 58:917–935 931 from the mutual association with AMF, and these under- that colonised the weeds and that colonised the maize ground interactions may improve the invasiveness and com- crops. Overall, our findings suggest that drought-resistant petitiveness of weeds against maize plants; thus, this com- weeds, growing off-season along the field borders, can petition may have neutralised the AMF-mediated benefits act as AMF hosts during the dry season when there are on crops (Massenssini et al. 2014; Callaway et al. 2004). On no crops in the field, and part of this AMF community is the other hand, since we only identified two weed species carried over into the fields. These new insights support that were common to all cropping systems and used these the need to find an equilibrium between the control of to assess AMF community response, this may have limited weeds and the maintenance of their diversity to guaran- the resolution at which we dissected the relationship. This tee crop yield and AMF-mediated ecosystem services. would mean that it is necessary to molecularly characterise For example, farmers could consider adopting cropping the AM fungal community in maize roots and relate these to systems that result in less competitive and diverse weed communities in the roots of more weed species. communities instead of complete eradication of weeds to conserve biodiversity and improve ecosystem services. However, further research needs to focus on the assess- Conclusion ment of the effect of weeds on maintaining or increasing the AM fungal propagules in the soil during the non- Arbuscular mycorrhizal fungi are important in agricul- cropping period and on the AMF shared among in-season tural systems as they assist their host plants in taking and off-season weeds and crops. Finally, the AM fungal up nutrients through the extraradical mycelium whilst community composition should be studied by applying obtaining photosynthetic assimilates from the host plant. innovative sequencing methods (i.e. the third-generation Since AMF are obligate mutualistic symbionts, they long-read sequencing technology) that allow for improved require a host for their proliferation. In southern Africa, specificity and enhanced resolution compared to Illumina where short crop growing seasons are experienced, under- sequencing. This, together with the assessment of other standing how weed communities host AMF as alternative weed AMF-mediated services (i.e. soil structure and hosts is important in agroecosystem management since nutrient retention), would allow us to understand the link this symbiosis determines the promotion of biodiversity between AMF, weeds, crop growth and nutrient uptake. and hence ecosystem services. Here, for the first time, we assessed if mycorrhizal weeds surrounding the experi- Supplementary Information The online version contains supplemen- tary material available at https://d oi.o rg/1 0.1 007/s 00374-0 22-0 1678-1. mental fields and, among these weeds, those commonly found in all experimental plots would act as hosts of AMF Acknowledgements The PhD fellowship for BM was funded by the during the off-season and during the season, respectively, Scuola Superiore Sant’Anna. All the authors are grateful to the Inter- and if AM fungal assemblages would be affected by host national Maize and Wheat Improvement Centre for funding the setup and running of the experimental trials and to donors of the MAIZE identity and the different combinations of CA compo- CGIAR Research Program (www. maize. org) who supported the trials nents. In this work, we have also shown that weeds grow- until 2018. The authors are thankful to the DSCATT project for co- ing during the dry off-season can host AMF, and that, funding the CIMMYT step-trials since 2018. The DSCATT project (N° although the AM fungal community composition in the AF 1802-001, N° FT C002181) is funded by the Agropolis Founda- tion (through the ‘Programme Investissement d’Avenir’ Labex Agro, dry winter period was not predictive of the composition funding ANR-10-LABX-0001-01) and by the TOTAL Foundation, at anthesis, a large proportion of AM fungal taxa were as part of a sponsorship agreement. The financial support of Project shared between sampling stages. This is a novel finding PSR 2014-2020—Misura 16.2 PS-GO 2017—Progetto AGROCIR- indicating that weeds in off-season can exert a functional COLIVE—CUP ARTEA 831728 is gratefully acknowledged. Special thanks go to the technical teams at the two experimental locations, led role during the dry periods since they represent the pool by Sign Phiri and Herbert Chipara, who assisted in data collection. for later AM fungal colonisation of crops. Moreover, we demonstrated for the first time that host specificity Author contribution E.P. and B.M.: AM fungal experimental idea and set is modulated by drought conditions, usually occurring up of the AM fungal experiment. C.T.: CA experiment idea and set up of omission trials. C.T., B.M., L.E. and E.P.: coordination of data collection. at our site in the off-season period, inducing the plants B.M., E.P. and G.P.: formal data analysis. B.M. and EP: writing the original under severe stress to select the most functional AMF. manuscript draft. E.P., L.E., C.T. and G.P.: writing, review and editing. All Finally, the models describing the response of maize authors have read and agreed to the published version of the manuscript. yield to weed AMF traits and maize AM fungal coloni- Funding Open access funding provided by Scuola Superiore Sant'Anna sation showed no significant influence. This absence of within the CRUI-CARE Agreement. influence may reflect that the competitive ability of the weeds was improved, hence overshadowing the antici- Data availability Data used in this study are stored in a public pated AMF-mediated benefits to maize productivity. It data repository and can be made available upon reasonable request following data-sharing regulations. The R scripts used in data may also reflect the absence of a link between the AMF 1 3 9 32 Biology and Fertility of Soils (2022) 58:917–935 analyses are available from the corresponding author upon request. of arbuscular mycorrhiza in soils is linked to the total length of Sequences generated in this study were uploaded to the NCBI data- roots colonized at ecosystem level. PLoS ONE 15:e0237256. base (submission number SUB10794739) and accession numbers https:// doi. org/ 10. 1371/ journ al. pone. 02372 56 OM049043–OM049185. Barto EK, Weidenhamer JD, Cipollini D, Rillig MC (2012) Fungal superhighways: do common mycorrhizal networks enhance Declarations below ground communication? Trends Plant Sci 17:633–637. https:// doi. org/ 10. 1016/j. tplan ts. 2012. 06. 007 Ethics approval All ethics committees of the organisations with which Bennett JA, Cahill JF, van der Heijden M (2016) Fungal effects on the authors are affiliated have no objections to the publication of this plant-plant interactions contribute to grassland plant abundances: work. evidence from the field. J Ecol 104:755–764. https:// doi. org/ 10. 1111/ 1365- 2745. 12558 Consent to participate Not applicable. Bates D, Maechler M, Bolker B, Walker S (2015) Fitting linear mixed- effects models using lme4. J Stat Softw 67:1–48. https://d oi.o rg/ Consent for publication Not applicable. 10. 18637/ jss. v067. i01 Bento CPM, Yang X, Gort G, Xue S, van Dam R, Zomer P, Mol HGJ, Ritsema CJ, Geissen V (2016) Persistence of glyphosate and Conflict of interest The authors declare no conflict of interest. aminomethylphosphonic acid in loess soil under different com- binations of temperature, soil moisture and light/darkness. Sci Open Access This article is licensed under a Creative Commons Total Environ 572:301–311. https:// doi. org/ 10. 1016/j. scito tenv. Attribution 4.0 International License, which permits use, sharing, 2016. 07. 215 adaptation, distribution and reproduction in any medium or format, Bever JD (2002) Host-specificity of AM fungal population growth rates as long as you give appropriate credit to the original author(s) and the can generate feedback on plant growth. Plant Soil 244:281–290. source, provide a link to the Creative Commons licence, and indicate https:// doi. org/ 10. 1023/A: 10202 21609 080 if changes were made. The images or other third party material in this Bolyen E, Rideout JR, Dillon MR, Bokulich NA, Abnet CC, Al-Ghalith article are included in the article's Creative Commons licence, unless GA, Alexander H, Alm EJ, Arumugam M, Asnicar F, Bai Y, indicated otherwise in a credit line to the material. If material is not Bisanz JE, Bittinger K, Brejnrod A, Brislawn CJ, Brown CT, Cal- included in the article's Creative Commons licence and your intended lahan BJ, Caraballo-Rodríguez AM, Chase J, Cope EK, Da Silva use is not permitted by statutory regulation or exceeds the permitted R, Diener C, Dorrestein PC, Douglas GM, Durall DM, Duvallet use, you will need to obtain permission directly from the copyright C, Edwardson CF, Ernst M, Estaki M, Fouquier J, Gauglitz JM, holder. To view a copy of this licence, visit http:// creat iveco mmons. Gibbons SM, Gibson DL, Gonzalez A, Gorlick K, Guo J, Hill- org/ licen ses/ by/4. 0/. mann B, Holmes S, Holste H, Huttenhower C, Huttley GA, Jans- sen S, Jarmusch AK, Jiang L, Kaehler BD, Kang KB, Keefe CR, Keim P, Kelley ST, Knights D, Koester I, Kosciolek T, Kreps J, Langille MGI, Lee J, Ley R, Liu YX, Loftfield E, Lozupone References C, Maher M, Marotz C, Martin BD, McDonald D, McIver LJ, Melnik AV, Metcalf JL, Morgan SC, Morton JT, Naimey AT, Adeux G, Vieren E, Carlesi S, Bàrberi P, Munier-Jolain N, Cord- Navas-Molina JA, Nothias LF, Orchanian SB, Pearson T, Peoples eau S (2019) Mitigating crop yield losses through weed SL, Petras D, Preuss ML, Pruesse E, Rasmussen LB, Rivers A, diversity. Nat Sustain 2:1018–1026. https:// doi. org/ 10. 1038/ Robeson MS, Rosenthal P, Segata N, Shaffer M, Shiffer A, Sinha s41893- 019- 0415-y R, Song SJ, Spear JR, Swafford AD, Thompson LR, Torres PJ, Alaux PL, Mison C, Senés-Guerrero C, Moreau V, Manssens G, Fou- Trinh P, Tripathi A, Turnbaugh PJ, Ul-Hasan S, van der Hooft cart G, Cranenbrouck S, Declerck S (2021) Diversity and species JJJ, Vargas F, Vázquez-Baeza Y, Vogtmann E, von Hippel M, composition of arbuscular mycorrhizal fungi across maize fields Walters W, Wan Y, Wang M, Warren J, Weber KC, Williamson in the southern part of Belgium. Mycorrhiza 31:265–272. https:// CHD, Willis AD, Xu ZZ, Zaneveld JR, Zhang Y, Zhu Q, Knight doi. org/ 10. 1007/ s00572- 020- 01007-0 R, Caporaso JG (2019) Reproducible, interactive, scalable and Ames RN, Ingham ER, Reid CPP (1982) Ultra-violet induced autofluo- extensible microbiome data science using QIIME 2. Nat Bio- rescence of arbuscular mycorrhizal root infection: an alternative technol 37:852–857. https://d oi.o rg/1 0.1 038/s 41587-0 19-0 209-9 to clearing and staining methods for assessing infections. Can J Bray JR, Curtis JRT (1957) An ordination of the upland forest commu- Microbiol 28:351–355. https:// doi. org/ 10. 1139/ m82- 052 nities of southern Wisconsin. Ecol Monogr 27:325–349. https:// Anderson MJ (2001) A new method for non-parametric multivariate doi. org/ 10. 2307/ 19422 68 analysis of variance. Austral Ecol 26:32–46. https:// doi. org/ 10. Bremner JM, Mulvaney CS (1982) Nitrogen - total. In: Page AL, Miller 1046/j. 1442- 9993. 2001. 01070.x RH, Keeney DR (eds) Methods of soil analysis, part 2, chemical Avio L, Pellegrino E, Bonari E, Giovannetti M (2006) Functional diver- and microbiological properties agronomy monograph. American sity of arbuscular mycorrhizal fungal isolates in relation to ext- Society of Agronomy, Madison, pp 595–624 raradical mycelial networks. New Phytol 172:347–357. https:// Bretagnolle V, Gaba S (2015) Weeds for bees? A review. Agron Sustain doi. org/ 10. 1111/j. 1469- 8137. 2006. 01839.x Dev 35:891–909. https:// doi. org/ 10. 1007/ s13593- 015- 0302-5 Bahadur A, Batool A, Nasir F, Jiang S, Mingsen Q, Zhang Q, Pan J, Liu Brito I, Goss MJ, De Carvalho M (2012) Effect of tillage and crop on Y, Feng H (2019) Mechanistic insights into arbuscular mycor- arbuscular mycorrhiza colonization of winter wheat and triticale rhizal fungi-mediated drought stress tolerance in plants. Int J Mol under Mediterranean conditions. Soil Use Manag 28:202–208. Sci 20:1–18. https:// doi. org/ 10. 3390/ ijms2 01741 99 https:// doi. org/ 10. 1111/j. 1475- 2743. 2012. 00404.x Bàrberi P, Bocci G, Carlesi S, Armengot L, Blanco-Moreno JM, Sans Callaway RM, Thelen GC, Barth S, Ramsey PW, Gannon JE (2004) FX (2018) Linking species traits to agroecosystem services: a Soil fungi alter interactions between the invader Centaurea functional analysis of weed communities. Weed Res 58:76–88. maculosa and North American natives. Ecology 85:1062–1071. https:// doi. org/ 10. 1111/ wre. 12283 https:// doi. org/ 10. 1890/ 02- 0775 Barceló M, van Bodegom PM, Tedersoo L, den Haan N, Veen GF(, Castillo CG, Rubio R, Rouanet JL, Borie F (2006) Early effects of Ostonen I, Trimbos K, Soudzilovskaia NA (2020) The abundance tillage and crop rotation on arbuscular mycorrhizal fungal 1 3 Biology and Fertility of Soils (2022) 58:917–935 933 propagules in an Ultisol. Biol Fertil Soils 43:83–92. https:// doi. Heyl A (2022) Important weeds in maize. https:// www. up. ac. za/ org/ 10. 1007/ s00374- 005- 0067-0 sahri/ artic le/ 18103 72/ impor tant- weeds- in- maize. Accessed 8 Ciccolini V, Ercoli L, Davison J, Vasar M, Öpik M, Pellegrino E (2016) Mar 2022 Land-use intensity and host plant simultaneously shape the Harinikumar KM, Bagyaraj DJ (1989) Effect of cropping sequence, composition of arbuscular mycorrhizal fungal communities in a fertilizers and farmyard manure on vesicular-arbuscular mycor- Mediterranean drained peatland. FEMS Microbiol Ecol 92:1–13. rhizal fungi in different crops over three consecutive seasons. https:// doi. org/ 10. 1093/ femsec/ fiw186 Biol Fertil Soils 7:173–175. https:// doi. org/ 10. 1007/ BF002 Clarke KR, Gorley RN, Somerfield PJ, Warwick RM (2014) Change 92578 in marine communities: an approach to statistical analysis and Hibbett DS, Binder M, Bischoff JF, Blackwell M, Cannon PF, Eriksson interpretation, 3rd edn. PRIMER-E Ltd, Plymouth, UK: Plym- OE, Huhndorf S, James T, Kirk PM, Lücking R, Lumbsch HT, outh Marine Lab. pp 262 Lutzoni F, Matheny PB, McLaughlin DJ, Powell MJ, Redhead Daniell TJ, Husband R, Fitter AH, Young JPW (2001) Molecular diver- S, Schoch CL, Spatafora JW, Stalpers JA, Vilgalys R, Aime MC, sity of arbuscular mycorrhizal fungi colonising arable crops. Aptroot A, Bauer R, Begerow D, Benny GL, Castlebury LA, FEMS Microbiol Ecol 36:203–209. https:// doi. org/ 10. 1111/j. Crous PW, Dai YC, Gams W, Geiser DM, Griffith GW, Gueidan 1574- 6941. 2001. tb008 41.x C, Hawksworth DL, Hestmark G, Hosaka K, Humber RA, Hyde De La Providencia IE, De Souza FA, Fernández F, Delmas NS, KD, Ironside JE, Kõljalg U, Kurtzman CP, Larsson KH, Licht- Declerck S (2005) Arbuscular mycorrhizal fungi reveal dis- wardt R, Longcore J, Miadlikowska J, Miller A, Moncalvo JM, tinct patterns of anastomosis formation and hyphal healing Mozley-Standridge S, Oberwinkler F, Parmasto E, Reeb V, Rog- mechanisms between different phylogenic groups. New Phytol ers JD, Roux C, Ryvarden L, Sampaio JP, Schüssler A, Sugiyama 165:261–271. https://d oi.o rg/1 0.1 111/j.1 469-8 137.2 004.0 1236.x J, Thorn RG, Tibell L, Untereiner WA, Walker C, Wang Z, Weir Derrouch D, Chauvel B, Cordeau S, Dessaint F (2021) Functional A, Weiss M, White MM, Winka K, Yao YJ, Zhang N (2007) shifts in weed community composition following adoption of A higher-level phylogenetic classification of the Fungi. Mycol conservation agriculture. Weed Res 62:103–112. https://d oi.o rg/ Res 111:509–547. https://d oi.o rg/1 0.1 016/j.m ycres.2 007.0 3.0 04 10. 1111/ wre. 12517 Hijri I, Sýkorová Z, Oehl F, Ineichen K, Maeder P, Wiemken A, Dumbrell AJ, Ashton PD, Aziz N, Feng G, Nelson M, Dytham C, Redecker D (2006) Communities of arbuscular mycorrhizal Fitter AH, Helgason T (2011) Distinct seasonal assemblages fungi in arable soils are not necessarily low in diversity. Mol of arbuscular mycorrhizal fungi revealed by massively parallel Ecol 15:2277–2289. https:// doi. org/ 10. 1111/j. 1365- 294X. 2006. pyrosequencing. New Phytol 190:794–804. https:// doi. org/ 10. 02921.x 1111/j. 1469- 8137. 2010. 03636.x IUSS Working Group WRB (2015) World Reference Base on Soils Edgar RC (2010) Search and clustering orders of magnitude faster than (World Soil Resources Reports No. 106), International soil clas- BLAST. Bioinformatics 26:2460–2461. https://d oi.o rg/1 0.1 093/ sification system for naming soils and creating legends for soil bioin forma tics/ btq461 maps. FAO-ISRIC, Rome, Italy El Omari B, El Ghachtouli N (2021) Arbuscular mycorrhizal fungi- James TY, Stajich JE, Hittinger CT, Rokas A (2020) Toward a fully weeds interaction in cropping and unmanaged ecosystems: resolved fungal tree of life. Annu Rev Microbiol 74. https:// doi. a review. Symbiosis 83:279–292. https:// doi. org/ 10. 1007/ org/ 10. 1146/ annur ev- micro- 022020- 051835 s13199- 021- 00753-9 Johnson JM, Houngnandan P, Kane A, Sanon KB, Neyra M (2013) Eom AH, Hartnett DC, Wilson GWT (2000) Host plant species effects Diversity patterns of indigenous arbuscular mycorrhizal fungi on arbuscular mycorrhizal fungal communities in tallgrass prai- associated with rhizosphere of cowpea (Vigna unguiculata (L.) rie. Oecologia 122:435–444. https://d oi.o rg/1 0.1 007/s 00442 0050 Walp.) in Benin. West Africa Pedobiologia 56:121–128. https:// 050 doi. org/ 10. 1016/j. pedobi. 2013. 03. 003 FAO (2019) Conservation Agriculture. Available online at: http://w ww. Katoh K, Rozewicki J, Yamada KD (2019) MAFFT online service: fao.o rg/c onser vatio n-a gricu lture/o vervi ew/w hat-i s-c onser vatio n- multiple sequence alignment, interactive sequence choice and agric ulture/ en/ (accessed August 2019) visualization. Brief Bioinform 20:1160–1166. https://d oi.o rg/1 0. Ferrero R, Lima M, Davis AS, Gonzalez-Andujar JL (2017) Weed 1093/ bib/ bbx108 diversity affects soybean and maize yield in a long-term experi- Kiers ET, Duhamel M, Yugandhar B, Mensah AJ, Oscar F, Verbruggen ment in Michigan, USA. Front Plant Sci 8:236–236. https:// doi. E, Fellbaum CR, Kowalchuk GA, Hart MM, Bago A, Palmer org/ 10. 3389/ fpls. 2017. 00236 TM, West SA, Vandenkoornhuyse P, Jansa J, Bücking H (2011) Gavito ME, Jakobsen I, Mikkelsen TN, Mora F (2019) Direct evidence Reciprocal rewards stabilize cooperation in the mycorrhizal for modulation of photosynthesis by an arbuscular mycorrhiza- symbiosis. Science 333:880–882. https://d oi. org/ 10. 1126/ scien induced carbon sink strength. New Phytol 223:896–907. https:// ce. 12084 73 doi. org/ 10. 1111/ nph. 15806 Kodzwa JJ, Gotosa J, Nyamangara J (2020) Mulching is the most Gollotte A, van Tuinen D, Atkinson D (2004) Diversity of arbuscular important of the three conservation agriculture principles in mycorrhizal fungi colonising roots of the grass species Agrostis increasing crop yield in the short term, under sub humid tropical capillaris and Lolium perenne in a field experiment. Mycorrhiza conditions in Zimbabwe. Soil till Res 197:104515. https:// doi. 14:111–117. https:// doi. org/ 10. 1007/ s00572- 003- 0244-7 org/ 10. 1016/j. still. 2019. 104515 Guijarro KH, Aparicio V, De Gerónimo E, Castellote M, Figuerola Koocheki A, Nassiri M, Alimoradi L, Ghorbani R (2009) Effect of EL, Costa JL, Erijman L (2018) Soil microbial communities cropping systems and crop rotations on weeds. Agron Sustain and glyphosate decay in soils with different herbicide applica- Dev 29:401–408. https:// doi. org/ 10. 1051/ agro/ 20080 61 tion history. Sci Total Environ 634:974–982. https:// doi. org/ Kottek M, Grieser J, Beck C, Rudolf B, Rubel F (2006) World map of 10. 1016/j. scito tenv. 2018. 03. 393 the Köppen-Geiger climate classification updated. Meteorol Z Hart MM, Reader RJ (2004) Do arbuscular mycorrhizal fungi recover 15:259–263. https:// doi. org/ 10. 1127/ 0941- 2948/ 2006/ 0130 from soil disturbance differently? Trop Ecol 45:97–112 Lal R (2000) Mulching effects on soil physical quality of an alfisol in Helgason T, Merryweather JW, Young JPW, Fitter AH (2007) Speci- western Nigeria. Land Degrad Dev 11:383–392. https:// doi. org/ ficity and resilience in thearbuscular mycorrhizal fungi of a 10. 1002/ 1099- 145X(200007/ 08) 11:4% 3C383:: AID- LDR393% natural woodland community. J Ecol 95:623–630. https:// doi. 3E3.0. CO;2-6 org/ 10. 1111/j. 1365- 2745. 2007. 01239.x 1 3 9 34 Biology and Fertility of Soils (2022) 58:917–935 Lee J, Lee S, Young JP (2008) Improved PCR primers for the detection properties agronomy monograph. American Society of Agron- and identification of arbuscular mycorrhizal fungi. FEMS Micro- omy, Madison, pp 539–579 biol Ecol 65:339–349. https://d oi.o rg/1 0.1 111/j.1 574-6 941.2 008. Nerva L, Giudice G, Quiroga G, Belfiore N, Lovat L, Perria R, Volpe 00531.x MG, Moffa L, Sandrini M, Gaiotti F, Balestrini R, Chitarra W Lefcheck J (2016) piecewiseSEM: Piecewise structural equation mod- (2022) Mycorrhizal symbiosis balances rootstock-mediated eling in R for ecology, evolution, and systematics. Methods Ecol growth-defence tradeoffs. Biol Fert Soils 58:17–34 Evol 7:573–579. https:// doi. org/ 10. 1111/ 2041- 210X. 12512 Nichols V, Verhulst N, Cox R, Govaerts B (2015) Weed dynamics Lenth R (2019) emmeans: estimated marginal means, aka least-squares and conservation agriculture principles: a review. Field Crop Res means. R package version 1.3.4. https:// CRAN.R- proje ct. org/ 183:56–68. https:// doi. org/ 10. 1016/j. fcr. 2015. 07. 012 packa ge= emmea ns Oehl F, Sieverding E, Ineichen K, Mäder P, Boller T, Wiemken A Li M, Jordan NR, Koide RT, Yannarell AC, Davis AS (2019) Interspe- (2003) Impact of land use intensity on the species diversity cific variation in crop and weed responses to arbuscular myc- of arbuscular mycorrhizal fungi in agroecosystems of central orrhizal fungal community highlights opportunities for weed Europe. Appl Environ Microbiol 69:2816–2824. https://d oi.o rg/ biocontrol. Appl Soil Ecol 142:34–42. https://d oi.o rg/1 0.1 016/j. 10. 1128/ AEM. 69.5. 2816- 2824. 2003 apsoil. 2019. 05. 016 Oehl F, Sieverding E, Ineichen K, Mader P, Wiemken A, Boller T Li Y, Steenwyk JL, Chang Y, Wang Y, James TY, Stajich JE, Spata- (2009) Distinct sporulation dynamics of arbuscular mycorrhizal fora JW, Groenewald M, Dunn Casey W, Hittinger CT, Shen fungal communities from different agroecosystems in long-term XX, Rokas A (2021) A genome-scale phylogeny of the kingdom microcosms. Agr Ecosyst Environ 134:257–268. https://d oi.o rg/ fungi. Curr Biol 31:1653.e5-1665.e5. https:// doi. org/ 10. 1016/j. 10. 1016/j. agee. 2009. 07. 008 cub. 2021. 01. 074 Okada E, Costa JL, Bedmar F (2019) Glyphosate dissipation in dif- Lu X, Lu X, Liao Y (2018) Effect of tillage treatment on the diversity ferent soils under no-till and conventional tillage. Pedosphere of soil arbuscular mycorrhizal fungal and soil aggregate-associ- 29:773–783. https:// doi. org/ 10. 1016/ S1002- 0160(17) 60430-2 ated carbon content. Front Microbiol 9:1–10. https:// doi. org/ 10. Oliveros JC (2015) Venny. An interactive tool for comparing lists with 3389/ fmicb. 2018. 02986 Venn’s diagrams. https://b ioinf ogp.c nb.c sic.e s/t ools/v enny/i ndex. MacLaren C, Storkey J, Menegat A, Metcalfe H, Dehnen-Schmutz K html (2020) An ecological future for weed science to sustain crop Omirou M, Ioannides IM, Ehaliotis C (2013) Mycorrhizal inoculation production and the environment. A Review Agron Sustain Dev affects arbuscular mycorrhizaldiversity in watermelon roots, but 40:24. https:// doi. org/ 10. 1007/ s13593- 020- 00631-6 leads to improved colonization and plant response under water Merryweather JW, Fitter AH (1991) A modified method for elucidat- stress only. Appl Soil Ecol 63:112–119. https://d oi.o rg/1 0.1 016/j. ing the structure of the fungal partner in a vesicular-arbuscular apsoil. 2012. 09. 013 mycorrhiza. Mycol Res 95:1435–1437. https:// doi. org/ 10. 1016/ Pellegrino E, Gamper HA, Ciccolini V, Ercoli L (2020) Forage rota- S0953- 7562(09) 80399-7 tions conserve diversity of arbuscular mycorrhizal fungi and soil Martínez-García LB, Pugnaire FI (2011) Arbuscular mycorrhizal fungi fertility. Front Microbiol 10:2969. https://d oi.o rg/1 0.3 389/f micb. host preference and site effects in two plant species in a semi- 2019. 02969 arid environment. Appl Soil Ecol 48:313–317. https://d oi.o rg/1 0. Phillips JM, Hayman DS (1970) Improved procedures for clearing roots 1016/j. apsoil. 2011. 04. 003 and staining parasitic and vesicular-arbuscular mycorrhizal fungi Massenssini AM, Bonduki VHA, Tótola MR, Ferreira FA, Costa for rapid assessment of infection. Trans Br Mycol Soc 55:158– MD (2014) Arbuscular mycorrhizal associations and occur- 161. https:// doi. org/ 10. 1016/ S0007- 1536(70) 80110-3 rence of dark septate endophytes in the roots of Brazilian Piazza G, Ercoli L, Nuti M, Pellegrino E (2019) Interaction between weed plants. Mycorrhiza 24:153–159. https:// doi. org/ 10. 1007/ conservation tillage and nitrogen fertilization shapes prokaryotic s00572- 013- 0519-6 and fungal diversity at different soil depths: evidence from a McGonigle TP, Miller MH, Evans DG, Fairchild GL, Swan JA (1990) 23-year field experiment in the Mediterranean area. Front Micro- A new method which gives an objective measure of colonization biol 10:1–20. https:// doi. org/ 10. 3389/ fmicb. 2019. 02047 of roots by vesicular-arbuscular mycorrhizal fungi. New Phytol Pielou EC (1969) An introduction to mathematical ecology. Wiley- 115:495–501. https:// doi. org/ 10. 1111/j. 1469- 8137. 1990. tb004 Interscience Publications, New York. pp. 294. https:// doi. org/ 76.x 10. 4039/ Ent10 3288-2 McLean EO (1982) Soil pH and lime requirement. In: Page AL, Miller Qiao X, Bei S, Li H, Christie P, Zhang F, Zhang J (2016) Arbuscu- RH, Keeney DR (eds) Methods of soil analysis, part 2, chemi- lar mycorrhizal fungi contribute to overyielding by enhancing cal and microbiological properties agronomy monograph, vol 9, crop biomass while suppressing weed biomass in intercrop- 2nd edn. American Society of Agronomy, Madison, pp 199–224 ping systems. Plant Soil 406:173–185. https:// doi. org/ 10. 1007/ Mhlanga B, Ercoli L, Pellegrino E, Onofri A, Thierfelder C (2021) The s11104- 016- 2863-8 crucial role of mulch to enhance the stability and resilience of Ramos-Zapata JA, Marrufo-Zapata D, Guadarrama P, Carrillo-Sánchez cropping systems in southern Africa. Agron Sustain Dev 41:29. L, Hernández-Cuevas L, Caamal-Maldonado A (2012) Impact of https:// doi. org/ 10. 1007/ s13593- 021- 00687-y weed control on arbuscular mycorrhizal fungi in a tropical agro- Mhlanga B, Pellegrino E, Thierfelder C, Ercoli L (2022) Conserva- ecosystem: a long-term experiment. Mycorrhiza 22:653–661. tion agriculture practices drive maize yield by regulating soil https:// doi. org/ 10. 1007/ s00572- 012- 0443-1 nutrient availability, arbuscular mycorrhizas, and plant nutrient R Core Team (2021) R: A language and environment for statistical uptake. Field Crops Res 277:108403. https:// doi. org/ 10. 1016/j. computing. RFoundation for Statistical Computing, Vienna, Aus- fcr. 2021. 108403 tria. URL https:// www.R- proje ct. org/. Version 4.1.0 Munkvold L, Kjøller R, Vestberg M, Rosendahl S, Jakobsen I (2004) Säle V, Sieverding E, Oehl F (2022) Growth responses of three Euro- High functional diversity within species of arbuscular mycor- pean weeds on different AMF species during early development. rhizal fungi. New Phytol 164:357–364. https://d oi.o rg/1 0.1 111/j. Plants 11:2020. https:// doi. org/ 10. 3390/ plant s1115 2020 1469- 8137. 2004. 01169.x Sato K, Suyama Y, Saito M, Sugawara K (2005) A new primer for Nelson DW, Sommers LE (1982) Total carbon, organic carbon and discrimination of arbuscular mycorrhizal fungi with polymerase organic matter. In: Page AL, Miller RH, Keeney DR (eds) chain reaction-denature gradient gel electrophoresis. Grassl Sci Methods of soil analysis, part 2, chemical and microbiological 51:179–181. https://d oi.o rg/1 0.1 111/j.1 744-6 97X.2 005.0 0023.x 1 3 Biology and Fertility of Soils (2022) 58:917–935 935 Schalamuk S, Cabello M (2010) Arbuscular mycorrhizal fungal prop- van der Heyde M, Ohsowski B, Abbott LK, Hart M (2017) Arbus- agules from tillage and no-tillage systems: possible effects on cular mycorrhizal fungus responses to disturbance are context- Glomeromycota diversity. Mycologia 102:261–268. https:// doi. dependent. Mycorrhiza 27:431–440. https:// doi. org/ 10. 1007/ org/ 10. 3852/ 08- 118 s00572- 016- 0759-3 Schüβler A, Schwarzott D, Walker C (2001) A new fungal phylum, the van Tuinen D, Jacquot E, Zhao B, Gollotte A, Gianinazzi-Pearson V Glomeromycota: phylogeny and evolution. Mycol Res 105:1413– (1998) Characterization of root colonization profiles by a micro- 1421. https:// doi. org/ 10. 1017/ S0953 75620 10051 96 cosm community of arbuscular mycorrhizal fungi using 25S Shannon CE, Weaver W (1949) The mathematical theory of commu- rDNA-targeted nested PCR. Mol Ecol 7:879–887 nication. University of Illinois Press, Urbana. https://d oi.o rg/1 0. Vatovec C, Jordan N, Huerd S (2005) Responsiveness of certain agro- 1002/j. 1538- 7305. 1948. tb013 38.x nomic weed species to arbuscular mycorrhizal fungi. Renew Silva JV, Baudron F, Reidsma P, Giller KE (2019) Is labour a major Agric Food Syst 20:181–189. https:// doi. org/ 10. 1079/ RAF20 determinant of yield gaps in sub-Saharan Africa? A study of 05115 cereal-based production systems in Southern Ethiopia. Agric Veiga RSL, Jansa J, Frossard E, van der Heijden MGA (2011) Can Syst 174:39–51. https:// doi. org/ 10. 1016/j. agsy. 2019. 04. 009 arbuscular mycorrhizal fungi reduce the growth of agricultural Simon L, Lalonde M, Bruns TD (1992) Specific amplification of 18S weeds? PLoS ONE 6:27825. https:// doi. org/ 10. 1371/ journ al. fungal ribosomal genes from vesicular-arbuscular endomycorrhi- pone. 00278 25 zal fungi colonizing roots. Appl Environ Microbiol 58:291–295 Wilkes TI, Warner DJ, Edmonds-Brown V, Davies KG, Denholm I Smith SE, Read D (2008) 1 - The symbionts forming arbuscular mycor- (2021) Zero tillage systems conserve arbuscular mycorrhizal rhizas. In: Smith SE, Read D (eds) Mycorrhizal symbiosis (Third fungi, enhancing soil glomalin and water stable aggregates with Edition). Academic Press, London, pp 13–41 implications for soil stability. Soil Syst 5:1–13. https:// doi. org/ Storkey J, Neve P (2018) What good is weed diversity? Weed Res 10. 3390/ soils ystem s5010 004 58:239–243. https:// doi. org/ 10. 1111/ wre. 12310 Wilson GWT, Hartnett DC (1998) Interspecific variation in plant Spatafora JW, Chang Y, Benny GL, Lazarus K, Smith ME, Berbee responses to mycorrhizal colonization in tallgrass prairie. Am J ML, Bonito G, Nicolas C, Grigoriev I, Gryganskyi A, James Bot 85:1732–1738. https:// doi. org/ 10. 2307/ 24465 07 TY, O’Donnell K, Roberson RW, Taylor TN, Uehling J, Vilgalys Yang H, Zang Y, Yuan Y, Tang J, Chen X (2012) Selectivity by host R, White MM, Stajich JE (2016) A phylum-level phylogenetic plants affects the distribution of arbuscular mycorrhizal fungi: classification of zygomycete fungi based on genome-scale data. evidence from ITS rDNA sequence metadata. BMC Evol Biol Mycologia 108:1028–1046. https:// doi. org/ 10. 3852/ 16- 042 12:50. https:// doi. org/ 10. 1186/ 1471- 2148- 12- 50 Suzuki K, Takahashi K, Harada N (2020) Evaluation of primer pairs for Yu G, Fadrosh D, Goedert JJ, Ravel J, Goldstein AM (2015) Nested studying arbuscular mycorrhizal fungal community compositions PCR biases in interpreting microbial community structure in 16S using a MiSeq platform. Biol Fertil Soils 56:853–858. https://d oi. rRNA gene sequence datasets. PLoS ONE 10:e0132253 org/ 10. 1007/ s00374- 020- 01431-6 Zhang Q, Yang R, Tang J, Yang H, Hu S, Chen X (2010) Positive Tamura K, Stecher G, Kumar S (2021) MEGA11: Molecular Evolu- feedback between mycorrhizal fungi and plants influences plant tionary Genetics Analysis version 11. Mol Biol Evol 38:3022– invasion success and resistance to invasion. PLoS ONE 5:12380. 3027. https:// doi. org/ 10. 1093/ molbev/ msab1 20 https:// doi. org/ 10. 1371/ journ al. pone. 00123 80 Tedersoo L, Sánchez-Ramírez S, Kõljalg U, Bahram M, Döring M, Zhang F, Li Q, Yerger EH, Chen X, Shi Q, Wan F (2018) AM fungi Schigel D, May T, Ryberg M, Abarenkov K (2018) High-level facilitate the competitive growth of two invasive plant species, classification of the fungi and a tool for evolutionary ecological Ambrosia artemisiifolia and Bidens pilosa. Mycorrhiza 28:703– analyses. Fungal Divers 90:135–159. https:// doi. org/ 10. 1007/ 715. https:// doi. org/ 10. 1007/ s00572- 018- 0866-4 s13225- 018- 0401-0 Zhang H, Hobbie EA, Feng P, Niu L, Hu K (2021) Can conservation Toju H, Tanabe AS, Yamamoto S, Sato H (2012) High-coverage ITS agriculture mitigate climate change and reduce environmental primers for the DNA-based identification of ascomycetes and impacts for intensive cropping systems in North China Plain? basidiomycetes in environmental samples. PLoS ONE 7:e40863 Sci Total Environ 806:151194. https:// doi. org/ 10. 1016/j. scito van der Heijden MG, Klironomos JN, Ursic M, Moutoglis P, Streitwolf- tenv. 2021. 151194 Engel R, Boller T, Wiemken A, Sanders IR (1998) Mycorrhizal fungal diversity determines plant biodiversity, ecosystem vari- Publisher's note Springer Nature remains neutral with regard to ability and productivity. Nature 396:69–72. https:// doi. org/ 10. jurisdictional claims in published maps and institutional affiliations. 1038/ 23932 1 3