Full paper
Effects of different mycorrhizal types of dispersedly retained trees on the diversity of ectomycorrhizal fungi in neighboring Abies sachalinensis seedlings
2025 Volume 66 Issue 3 Pages 195-200
Details
2025 Volume 66 Issue 3 Pages 195-200
Retention forestry conserves biodiversity by retaining forest structures in logged areas. It has been demonstrated that dispersedly retained broad-leaved ectomycorrhizal (EcM) trees can mitigate the effect of logging on the diversity of EcM fungi in the surrounding Abies sachalinensis seedlings. However, it remains unclear how retained trees of different mycorrhizal types affect the diversity of EcM fungi in Abies seedlings. We investigated the neighborhood effect of different mycorrhizal types of retained trees on the diversity of EcM fungi symbiotic with surrounding Abies seedlings. At dispersed retention sites, the roots of Abies seedlings were collected near mature EcM trees (ET) or arbuscular mycorrhizal (AM) trees (AT), or in open areas where no retained trees existed within ten meters (NT). EcM fungi were identified based on ITS barcoding of the EcM roots. The diversity measures of EcM fungi under AT and NT were comparable and lower, respectively than those under ET. The community composition of the EcM fungi was similar between AT and NT, and both were significantly different from that of ET. These results indicate that AM trees do not have significant impact on EcM fungul community in the surrounding EcM seedlings.
Ectomycorrhizal (EcM) fungi are important soil biological components involved in nutrient cycling and carbon sequestration in boreal and temperate forests (Becquer et al., 2019). EcM symbiotic associations were commonly found in seedlings in sites disturbed by deforestation (Hawkins et al., 2015; Jones et al., 2003). The species richness and community composition of EcM fungi in seedlings regenerated in such environments is affected by a variety of abiotic factors (e.g., soil moisture and temperature); however, biotic factors, such as inoculum loss, are also known to have a strong influence (Jones et al., 2003). For example, it has been reported that the distances from neighboring mature EcM trees that were left uncut in logged areas affect the EcM fungal community composition of the surrounding EcM seedlings (Cline et al., 2005; Jones et al., 2003, 2008; Obase et al., 2024). In other words, the roots of neighboring mature trees function as subsequent inoculum sources for diverse EcM fungi and affect the diversity of EcM fungi symbiotic with surrounding EcM seedlings.
However, it is not well understood how such neighborhood effects vary depending on the mycorrhizal type of neighboring trees, as there are no studies on this topic in logged areas. Most relevant studies that have examined such neighborhood effects have been conducted in cases where the target trees and their neighboring trees have the same mycorrhizal type. Few studies have examined the effects of different mycorrhizal types; EcM and arbuscular mycorrhizal (AM). In volcanic deserts where mycorrhizal inoculum is almost absent, Nara and Hogetsu (2004) reported that few willow seedlings (the dual mycorrhizal type that can symbiose with both AM and EcM fungi) associated with EcM fungi when transplanted in bare ground and patches occupied with only AM plants (i.e., EcM plants were absent), even though seedlings that were transplanted in patches with EcM plants (Salix reinii) associated with several EcM fungi with high frequency. In a greenhouse microcosm experiment, Fernández et al. (2022) documented that the AM plant Hieracium caespitosum and its associated AM fungi alter the composition of EcM fungal communities in the rhizosphere soils of Betula pendula planted in the vicinity. Incidentally, in mixed forest stands of AM and EcM trees, several studies have reported that the interaction with roots of neighboring AM trees altered EcM fungal community compositions in EcM trees (Bermúdez-Contreras et al., 2022; Haskins & Gehring, 2004; Hubert & Gehring, 2008; McHugh & Gehring, 2006), although a recent study indicated the possibility that focal AM trees Juniper spp. can associate with EcM fungi (Bermúdez-Contreras et al., 2022). Thus, it remains unclear whether these studies have focused on the effects of differences in mycorrhizal type among tree species or the effects of differences in preference and/or host specificity for EcM fungi among tree species. These results suggest that neighboring AM trees can have different impacts on EcM communities in surrounding trees when compared with neighboring EcM trees in logged areas.
Dispersed retention is a management practice in retention forestry that conserves biodiversity by dispersing forest materials and structures (e.g., living trees) in logged areas (Franklin et al., 1997). In a demonstration experiment for retention forestry initiated in Hokkaido, Japan, naturally regenerated broadleaved mature trees that include both AM and EcM tree species were dispersedly retained in logged coniferous artificial forests (Yamaura et al., 2018). At the experimental sites, it was shown that retained EcM trees can mitigate the decline in species richness of EcM fungi symbiotic with the surrounding Abies sachalinensis seedlings and maintain unique community compositions that differ from those in clear-cut areas (Obase et al., 2022). However, no studies have been conducted on EcM fungal communities in Abies seedlings that regenerate around retained AM trees. The experimental sites are suitable fields to study the neighborhood effect of how trees with different mycorrhizal types affect the mycorrhizal symbiosis of surrounding trees, and the results obtained from this study will also contribute to understanding of some of the effects of retention forestry on how diversity of EcM fungi can be conserved by leaving what tree species in place.
This study aimed to deepen our understanding of the neighborhood effects of retained trees of different mycorrhizal types on the surrounding root-associated EcM fungal communities by comparing EcM fungal communities in seedlings at two different types of microhabitats in logged areas―surrounding retained AM mature trees and EcM mature trees. We also examined EcM fungal communities in seedlings in open areas far from any retained mature trees in order to determine the effect of the presence of retained mature trees in the vicinity.
This study was conducted at four sites established in a retention experiment for plantation forestry in Sorachi, Hokkaido, Japan (REFRESH) (Yamaura et al., 2018). Each study site was a nearly 50-y-old A. sachalinensis planted forest, with areas ranging from 5.76 ha to 7.72 ha. Clear-cutting and ground preparation were performed at site SS1 in 2014, site SM2 in 2015, and sites SS3 and SM3 in 2016 when broadleaved mature trees were retained in a distributed manner at different densities (SS1 and SS3; 10 trees/ha, SM2, and SM3; 50 trees/ha) to understand the effect of the tree retention on biodiversity and other public functions. Abies sachalinensis saplings were planted in a striated pattern the following year in the logged areas. The species composition of the retained trees varied among the study sites, consisting primarily of EcM trees such as Betula spp., Tilia spp., and Quercus crispula Blume var. crispula, and AM trees such as Kalopanax septemlobus (Thunb.) Koidz. and Fraxinus mandshurica Rupr. No mature trees of A. sachalinensis were retained at study sites because they were empirically known to be susceptible to wind and thus deemed unsuitable for dispersed retention. Detailed information about the study sites and project outlines has been previously reported (Akashi et al., 2017; Yamaura et al., 2018).
2.2. Field sampling of EcM rootsIn May and Jun 2024, naturally regenerating A. sachalinensis seedlings (5-30 cm tall, Supplementary Table S1) were collected from three types of microhabitats―surrounding the retained EcM trees (ET), retained AM trees (AT), and open areas far from any retained trees (NT) (Supplementary Fig. S1). Each microhabitat was covered by mainly herbaceous plants or shrubs which are presumed as AM. Eight retained EcM and AM trees and eight points in the open areas were selected at each study site, and one A. sachalinensis seedling was collected near each retained tree or point. Abies seedlings were targeted for sampling because they were often found at all locations. The area surrounding the retained tree was defined as the area within three meters of the trunks of a targeted retained tree, whereas the open area was defined as the area at least ten meters far from any of the retained trees because the neighborhood effects of retained trees would be achieved (< 6 m; Cline et al., 2005, a few meters; Obase et al., 2022) or eliminated (10 m; Jones et al., 2008, Obase et al., 2024) at the range of distances. The targeted retention trees in the ET and AT were selected to be isolated from any other retention trees by at least ten meters to exclude the influence of neighborhood effects from non-targeted retention trees. Seedlings were collected at least 10 m apart. The species and number of retained trees targeted at each study site are listed in Table 1. The mycorrhizal type of the targeted tree species was based on previous studies [Kalopanax; Aggangan & Moon (2013); other tree species, Wang & Qiu (2006)]. Due to the low number of seedlings available at site SS3, the number of seedlings collected was lower than at the other study sites. Unless otherwise noted, the data obtained from site SS3 were included in the analysis.
| Trees | Family | Microhabitat* | SS1 | SS3 | SM2 | SM3 |
| Betula ermanii | Betulaceae | ET | 1 | 5 | 5 | 3 |
| Betula maximowicziana | Betulaceae | ET | 5 | 3 | 2 | |
| Betula platyphylla var. japonica | Betulaceae | ET | 1 | |||
| Quercus crispula | Fagaceae | ET | 1 | 1 | 1 | |
| Tilia japonica | Malvaceae | ET | 1 | 2 | ||
| Kalopanax septemlobus | Araliaceae | AT | 7 | 2 | 1 | |
| Fraxinus mandshurica | Oleaceae | AT | 5 | 6 | ||
| Cornus controversa var. controversa | Cornaceae | AT | 1 | |||
| Juglans mandshurica var. sachalinensis | Juglandaceae | AT | 1 | |||
| - | - | NT | 8 | 6 | 8 | 8 |
* AT: around retained arbuscular mycorrhizal trees, ET: around retained ectomycorrhizal trees, NT: open areas far from any retained trees.
The EcM root tips of A. sachalinensis seedling were briefly morphotyped under a dissecting microscope (SSZ-B; Kyowa Optical Co., Ltd., Kanagawa, Japan), and then 1-2 root tips of each morphotype in each sample were individually stored in a 0.2 mL microtube (1-1599-03; AS ONE, Osaka, Japan) at −20 °C. The DNA extraction, polymerase chain reaction (PCR), purification, and sequencing methods were the same as those described by Obase et al. (2022, 2024). Briefly, total DNA was extracted from EcM root tips by Extract-N-AmpTM Plant PCR Kit (XNAP2-1KT, Sigma-Aldrich, MO, USA). Internal transcribed spacer (ITS) region of nuclear ribosomal DNA was amplified by PCR with the use of TaKaRa Ex Taq (Takara Bio Inc., Shiga, Japan) or Extract-N-AmpTM Plant PCR Kit, and a primer pair of ITS1F (Gardes & Bruns, 1993) and ITS4 (White et al., 1990). The reaction mixture was purified using exonuclease I and Antarctic phosphatase (New England Biolabs, Inc., MA, USA) or the EnzSAP PCR Clean-Up Reagent Kit (EdgeBio, CA, USA). Sequencing reactions were performed by Eurofin Genomics (Tokyo, Japan) using ITS1 primers (White et al., 1990). Representative sequences were deposited in the International Nucleotide Sequence Database (INSD) through a process initiated by the DNA Data Bank of Japan (DDBJ) under accession number LC860144-LC860191.
2.4. Grouping into OTUThe ITS sequences were grouped into phylogenetic lineages of EcM fungi (Tedersoo et al., 2010), which were deduced using a BLAST search (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Sequences of the same lineages obtained in this study and previous studies (Obase et al., 2022, 2024) were aligned using MAFFT v.7 (Katoh & Standley, 2013) and then subjected to MOTHUR v.1.39 (Schloss et al., 2009) to classify them into OTUs based on a 97% similarity threshold. OTUs that contained sequences from Obase et al. (2022, 2024) were assigned names given by the literatures. OTUs consisting only of the sequences obtained in this study were assigned new names.
2.5. Statistical analysisThe occurrence frequency data of OTUs (i.e., the data matrix of the number of samples in which each OTU was detected) for each type of microhabitat at each study site were used for the analysis. All analyses were performed using R v.4.4.1 (R Core Team, 2024). Sample size-based rarefaction and extrapolation curves (Chao et al., 2014) for microhabitats were drawn using iNEXT with 1,000 bootstrap replications and ggiNEXT functions in the iNEXT (Hsieh et al., 2022) and ggplot2 packages (Wickham, 2016) to compare diversity measures among microhabitats―species richness, exponential of Shannon entropy, and the inverse of Simpson concentration, which were interpreted as the effective numbers of all, common, and dominant species, respectively (Hsieh et al., 2016).
To compare the diversity measures among the study sites and microhabitats, generalized linear model analyses with a Poisson distribution for OTU richness and gamma distribution for other diversity measures (link function = log) were conducted using the glm function in the stats package. Diversity measures were set as the objective variable, and study sites, microhabitats, and number of samples (offset log-transformed) as explanatory variables. A one-way ANOVA (type II) was performed using the Anova function in the car package (Fox & Weisberg, 2019) to check if the study sites and microhabitats were significant variables for the model. Pairwise Tukey post hoc tests were performed using the emmeans function in the emmeans package (Lenth, 2024) to analyze differences in diversity measures between different microhabitats.
To visualize the dissimilarity of EcM fungal communities among microhabitats, non-metric multidimensional scaling (NMDS) was conducted using the Bray-Curtis dissimilarity index and metaMDS function in the vegan package (Oksanen et al., 2022). To determine whether species composition differed among microhabitats, a permutational multivariate analysis of variance (PERMANOVA) was conducted using the adonis2 function based on the Bray-Curtis dissimilarity index, followed by pairwise comparisons between microhabitats (AT, ET, and NT) using the pairwiseAdonis function with the Bray-Curtis dissimilarity index, 999 permutations, and p-adjustment using the Benjamini and Hochberg method in the pairwiseAdonis package (Arbizu, 2017).
Indicator species analysis (INSPAN) was performed to determine whether the given OTUs or lineages preferentially colonized specific microhabitats using the multipatt function with 9,999 permutations in the indicspecies package (De Cáceres & Legendre, 2009). For indicator OTUs and lineages, pairwise comparisons of the proportions of the numbers of associated seedlings among microhabitats were conducted by using Fisher’s exact test with Benjamini and Hochberg correction, using the fisher.multcomp function in the RVAideMemoire package (Herve, 2023).
A total of 71 OTUs were detected in 313 sequences from 84 Abies seedlings (Supplementary Table S2). The microhabitat with the highest number of OTUs was ET (41 OTUs in total), followed by NT (25 OTUs) and AT (22 OTUs) (Fig. 1). A similar trend was found for other diversity measures (exponential of Shannon entropy and inverse of Simpson concentration).
ANOVA indicated that microhabitat, and not study site, was a significant explanatory variable in the model for the diversity measures of EcM fungi (Supplementary Table S3). OTU richness was significantly different between AT and ET (Tukey’s post-hoc test, p = 0.030), but there were no significant differences between AT and NT (p = 0.870) or between ET and NT (p = 0.061) (Supplementary Table S4). A similar trend was found for the other diversity measures, except that there were significant differences between ET and NT (p < 0.05).
As a whole, tomentella-thelephora lineage (including Tomentella and Thelephora spp.) had the highest number of OTUs (21 OTUs), followed by sebacina lineage (sebacina spp.) (ten OTUs), russula-lactarius lineage (Russula and Lactarius spp.) (nine OTUs) and inocybe lineage (inocybe spp.) (seven OTUs) (Fig. 2A; Supplementary Table S5). OTU cenococcum_hozan1 (closely related to the Cenococcum geophilum species complex) was the most frequently occurring taxon (detected in one to five seedlings at each study site), followed by OTU tomentella_hozan1 (closely related to Thelephora ellisii) and OTU tomentella_hozan3 (closely related to Tomentella sublilacina) (Supplementary Table S6). The number of OTUs detected across multiple microhabitats was low (3-6 OTUs), and the number of OTUs detected in only one type of microhabitat [particularly in ET (n = 30) compared with AT (n = 10) and NT (n = 13)] was higher (Fig. 2B).
At the lineage level, INSPAN selected two lineages as indicator taxa for ET (lineage boletus, p = 0.0099; lineage sebacina, p = 0.046), one lineage for both AT and NT (lineage laccaria, p = 0.029), and one lineage for both ET and NT (lineage russula-lactarius, p = 0.008). At the OTU level, INSPAN selected one OTU as an indicator taxon for ET (OTU xerocomellus_hozan1; p = 0.006). Fisher’s exact test indicated significant differences in the proportions of the numbers of associated seedlings with the members of lineages boletus (p = 0.049) and russula-lactarius (p = 0.020) between AT and ET (Supplementary Table S7). For OTU xerocomellus_hozan1, significant differences were found between AT and ET (p = 0.049), and ET and NT (p = 0.049). No significant differences were found in the other pairwise comparisons. When AT and NT were combined (AT+NT) and compared with ET, significant differences were found in all lineages and OTU (p = 0.003, 0.043, 0.013, 0.004 and 0.002 for lineages boletus, laccaria, russula-lactarius and sebacina and OTU xerocomellus_hozan1). When ET and NT were combined (ET+NT) and compared with AT, significant differences were only found in the lineage russula-lactarius (p = 0.016).
In the NMDS ordination plot, the AT and NT of each study site were relatively close to each other, whereas those of ET were dispersed and somewhat separated from those of AT and NT (Fig. 3). Significant differences were detected in the EcM fungal communities among microhabitats (PERMANOVA, R2 = 0.292, p = 0.009). Pairwise comparisons indicated significant differences between AT and ET (R2 = 0.262, p = 0.043) and between ET and NT (R2 = 0.258, p = 0.043), but not between AT and NT (R2 = 0.147, p = 0.66).
This study demonstrated that retained trees of different mycorrhizal types have different effects on the diversity and community composition of EcM fungi in surrounding Abies seedlings. Retained AM trees did not have significant impact on EcM fungul community in the surrounding EcM seedlings.
The main mechanism by which neighboring trees can retain the species richness of EcM fungi and alter the EcM community composition of the surrounding EcM seedlings seems to be whether the retained trees serve as an inoculum source for EcM fungi (e.g., Moeller et al., 2015; Nara & Hogetsu, 2004). In fact, the preferential occurrence of OTU xerocomellus_hozan1 in ET in this study is likely due to the incorporation of the surrounding seedlings into the mycelial network of the EcM taxon maintained by the retained EcM trees (Obase et al., 2022), and because of the absence of such types of inoculum sources in the case that the retained trees were AM or retained trees were absent. The results of comparisons between microhabitats with different dominant mycorrhizal types (ET vs AT+NT) also suggest that colonization of the members of the lineages boletus, russula-lactarius and sebacina on the surrounding seedlings are also induced by the presence of retained EcM trees, possibly via mycorrhizal networks from the retained EcM trees.
On the other hands, in AT and NT, probably resistant propagules of EcM fungi such as spores and sclerotia of Cenococcum geophilum, or spores that have been introduced from the surrounding areas after logging serve as the main inoculum source for EcM formation in Abies seedlings. For example, inoculum source of OTU tomentella_hozan1 that was frequently occurring taxon in all types of microhabitats in the study site is likely spores because EcM roots of this taxon were infrequently found in forests (Obase et al., 2022); the probability that it was originally symbiotic with the roots of the retained EcM tree is quite low. Also, the members of the lineage laccaria include species with high spore germination rates and well adapted to associate with seedlings under disturbed sites by spore dispersal (Ishida et al., 2008).
The results of this study suggest that the retained trees may inhibit the mycorrhization of certain EcM taxa in neighboring EcM seedlings. One possible explanation is the priority effect (Bogar & Kennedy, 2013), that is, a root-associated fungal community that inhabits a previously established plants strongly controls the fungal community of a later-established individual in the vicinity. It is possible that several of the retained EcM trees targeted in this study were not, or were infrequently, symbiotic with members of the laccaria lineage from the time before logging, and that the mycelial networks that captured the surrounding seedlings may have inhibited the invasion of subsequent EcM fungi (e.g., the laccaria lineage) into the surrounding seedlings. Alternatively, the members of the lineage laccaria may be outcompeted in the competition for the bare roots of Abies seedlings, given the relatively high frequency of occurrence by spore- or sclerotia-derived OTUs tomentella_hozan1 and cenococcum_hozan1 on the seedlings.
Complex context-dependent factors, possibly involving differences in plant traits, such as plant species in the neighborhood (Hubert & Gehring, 2008) and soil conditions (Becklin et al., 2012, Chen et al., 2018), may be involved in structuring EcM fungal communities in neighboring seedlings. In addition, the possibility of a low sampling effort, that is, a low number of samplings and site replications, should also be considered. These issues need to be clarified in the future through a series of additional field surveys and operational experiments.
The conservation effects of retention forestry on the diversity of mycorrhizal fungi have often been examined in relation to the quantity of retention, such as the density of the retained trees in logged areas (e.g., Sterkenburg et al., 2019); however, to our knowledge, no studies have focused on the quality of what is retained. The results of this study highlight the need to understand the effects of retained tree traits, such as mycorrhizal type, on the conservation effects of retained trees on the diversity of EcM fungi. This is an important perspective to comprehensively target not only EcM fungi but also diverse soil fungal guilds related to other important ecosystem services in forest soils (Philpott et al., 2018). Future research is needed to clarify the relationship between the quality of retained trees and the soil microbial community retained to examine forest management practices that increase soil fungal diversity and enhance soil ecosystem functioning.
The authors declare no conflict of interest. All experiments undertaken in this study complied with the current laws of the country in which they were performed.
We thank Dr. Nobuhiro Akashi for providing information on vegetation at the study sites, and Dr. Kenichi Ozaki for providing maps for access to the study sites. We would like to thank Editage (www.editage.jp) for English language editing. This study was supported by Grant-in-Aid for Scientific Research (C) JP21K05676 from the Japan Society for the Promotion of Science to Obase.
