The ECM fungi Suillus bovinus, Suillus luteus, and Scleroderma citrinum; the dark septate endophytes (DSE) Phialocephala fortinii, and the saprophytic fungus Lycoperdon perlatum were provided by the Microbiology Laboratory at the Institute for Forest Resources & Environment of Guizhou, Guizhou University. These fungi were subcultured on modified Melin Norkran’s (MMN) medium (Marx 1969) at 25°C in the dark. The MMN medium composition included: 25 mg/L NaCl, 250 mg/L (NH₄)₂HPO₄, 500 mg/L KH₂PO₄, 5 mg/L FeCl₃, 50 mg/L CaCl_2, 150 mg/L MgSO₄·7H₂O, 100 mg/L VB1, 10 g/L glucose, 1.00 g/L casamino acids, 5.00 g/L malt, and 10 g/L agar. Fusarium oxysporum is a type pathogenic fungus that caused root rot in trees, was obtained from the China General Microbiological Culture Collection Center and maintained on a medium containing glucose, peptone, yeast extract, and agar at 28°C. All fungi were cultured until suitable for inoculation.

Pinus massoniana seeds were collected from the high-quality breeding base of P. massoniana in Maanshan Forest Farm, Duyun City, Guizhou Province, China. Seeds were surface sterilized following Feng et al. (2022) and grown in a climate chamber for 28 days. Wildtype Arabidopsis thaliana ecotype Columbia-1 (col1) seeds were provided by Prof. Fuhua Fan, and the T-DNA insertion mutant lines, including yuc1, aux1, pin1, pin3, arf19, and tir1, were purchased from the AraShare Technology Service Center (Table 1). A. thaliana seeds were sterilized with 75% alcohol and stratified at 4°C for 3 days before use.

Data were analyzed using SPSS 25.0. Significant differences were determined by Student’s t-test or ANOVA with Duncan’s test (P < 0.05). Graphs were produced using Origin 2021. Venn diagrams of fungal VOCs were generated using Venny 2.1.0. (https://bioinfogp.cnb.csic.es/tools/venny/index.html).

To investigate the impact of VOCs produced by ECM fungi on plant growth, we examined the effects of S. bovinus-emitted VOCs on root growth and plant hormone levels in P. massoniana. After 28 days of exposure to S. bovinus VOCs, the P. massoniana lateral root number, root length and biomass were 46.97% (Fig. 1a), 29.04% (Fig. 1b), and 35.48% greater (Fig. 1c), respectively, than those of P. massoniana grown in the absence of S. bovinus VOCs. IAA accumulation in P. massoniana exposed to S. bovinus VOCs was 58.11% greater than that of P. massoniana grown in the absence of S. bovinus VOCs but had no significant effect on the accumulation of other phytohormones (Fig. 2). In the “Y-tube” experiment, roots grew toward the pipe containing the S. bovinus culture, with minimal or no growth toward the uninoculated medium (Suppl. material 1: fig. S1b).

Figure 2. 

Effects of Suillus bovinus (Sb) volatile organic compounds (VOCs) on phytohormone accumulation in Pinus massoniana. Seedlings grown in the absence of S. bovinus VOCs acted as controls (NM). Phytohormones: a indole-3-acetic acid (IAA); b abscisic acid (ABA); c gibberellic acid (GA3); d salicylic acid (SA); e trans-zeatin (tZA); f cis-zeatin (cZA); g 1-aminocyclopropane-1-carboxylate (ACC). Data shown are mean values ± the SE. n = 3. ** P < 0.01.

To assess the specificity of the effects of S. bovinus VOCs on plant growth, we examined the influence of VOCs emitted by other symbiotic, pathogenic, and saprophytic fungi on the root growth of P. massoniana. P. massoniana exposed to VOCs emitted by three other symbiotic fungi (i.e., S. luteus, S. citrinum, or P. fortinii) also showed significantly enhanced growth, including LRF and root elongation (Fig. 3a, b). However, the root growth of P. massoniana exposed to saprophytic fungi L. perlatum VOCs was not significantly different to that of the control (Fig. 3c, d). Conversely, the LRF of P. massoniana exposed to VOCs emitted by pathogenic fungi F. oxysporum was inhibited (Fig. 3e, f).

Figure 3. 

Effects of volatile organic compounds (VOCs) emitted by three different types of fungi on the root growth of Pinus massoniana. a–f represent the number of root branches and root length of Pinus massoniana exposed to VOCs emitted by symbiotic fungi (Sl, Suillus luteus; Sc, Scleroderma citrinum; Pf, Phialocephala fortinii), a saprophytic fungus (Lp, Lycoperdon perlatum) and a pathogenic fungus (Fo, Fusarium oxysporum), respectively. Due to the fast growth rate of F. oxysporum, only 14 days of P. massoniana root growth data were collected, while the rest of the fungi collected 28 days of data. n = 10. NM, P. massoniana grown in the absence of fungal VOCs. Different letters above bars indicate significant differences between treatments (P < 0.05).

To identify the signaling compounds in VOCs that promote plant root growth, we compared the volatile profiles of four symbiotic fungi: S. bovinus, S. luteus, S. citrinum, and P. fortinii. The VOC spectrum emitted by four symbiotic fungi, including terpenoids, alcohols, esters, hydrocarbons, ketones, and heterocyclic compounds, etc (Suppl. material 1: table S1). Terpenoids were the most abundant, with SQTs being particularly prevalent. Among the four symbiotic fungi, the S. bovinus VOC profile had the largest proportion of terpenoids (63.83%) and the P. fortinii VOC profile had the smallest proportion of terpenoids (20.00%) (Suppl. material 1: fig. S2b). The VOCs produced by DSE P. fortinii promote P. massoniana LRF more effectively than those produced by ECM fungi; however, there was no significant difference in the VOC effects among the ECM fungi (Fig. 3a). Only twenty-six VOCs were shared by all three ECM fungi, nine of which were terpenoids, most of which were SQTs (7) (Table 2; Suppl. material 1: fig. S2a). β-Cedrene and α-humulene were among the shared SQTs produced by the three ECM fungi but were not present in the P. fortinii VOC profile. Previous researches described that these two SQTs analogues, such as β-caryophyllene (Minerdi et al. 2011; Yamagiwa et al. 2011) and cedrene (Li et al. 2022) can promote plant growth, therefore, it is speculated that the terpenoids β-cedrene and α-humulene play an important role in the presymbiotic stage between P. massoniana and S. bovinus.

To investigate the growth-promoting effects of SQTs, we examined the impact of two SQTs, β-cedrene and α-humulene, common to the three ECM fungi, on root growth in the host P. massoniana and non-host A. thaliana. β-Cedrene and α-humulene both significantly promoted P. massoniana root development. After 35 days, the number of root branches of seedlings exposed to 1000 ppb β-cedrene was significantly greater than those exposed to 0, 1 or 10 ppb β-cedrene (Fig. 4a ,e). However, root length was not significantly affected by exposure to β-cedrene (Fig. 4b). By contrast, root development was promoted by 100 ppb α-humulene (Fig. 4e). Compared with seedlings in the control group (0 ppb), seedlings exposed to 100 ppb α-humulene exhibited a 37.92% increase in the number of root branches (Fig. 4c) and a 29.65% increase in root length (Fig. 4d). When the biosynthesis of SQTs by S. bovinus was blocked by lovastatin, there was no significant effect on P. massoniana root branching or root length. However, when 1000 ppb β-cedrene or 100 ppb α-humulene were also added to the fungal medium, P. massoniana root branching (Fig. 4f) and root length (Fig. 4g) were not significantly different to that of seedlings grown in the presence of S. bovinus without lovastatin.

Figure 4. 

Effects of β-cedrene and α-humulene, sesquiterpenes (SQTs) emitted by Suillus bovinus (Sb), on the growth of Pinus massoniana. Effects of different concentrations of β-cedrene on a root branching and b root length. Effects of different concentrations of α-humulene on c root branching and d root length. e Images of P. massoniana seedlings after 35 days of 1000 ppb β-cedrene or 100 ppb α-humulene treatment, CK, control seedlings; Scale bars: 2 cm. Effects of adding of β-cedrene or α-humulene to culture medium containing lovastatin (Lov) inhibits (which inhibits the synthesis of S. bovinus SQTs) on f root branching and g root length. n = 15. Error bars indicate ± the SE. Different letters between treatment after the same number of days of VOC exposure in (a, c), and different letters above the columns in (b, d, f, g) indicate significant differences between treatments (P < 0.05).

β-Cedrene and α-humulene also promote the LRF in the non-host A. thaliana. Concentrations of 10–100 ppb β-cedrene and 10 ppb α-humulene significantly increased the number of root branches (Fig. 5a). Furthermore, β-cedrene (10 ppb) and α-humulene (10 ppb) restored LRF in A. thaliana when lovastatin blocked SQT synthesis in S. bovinus (Lov + Sb + β-cedrene; Lov + Sb + α-humulene) (Fig. 5b).

Figure 5. 

Effects of β-cedrene and α-humulene on Arabidopsis thaliana root growth. a Number of root branches produced by seedlings exposed to different concentrations of the sesquiterpenes (SQTs) β-cedrene or α-humulene; b effects of adding β-cedrene (10 ppb) or α-humulene (10 ppb) to culture medium containing lovastatin (Lov) (which inhibits the synthesis of Suillus bovinus (Sb) SQTs) on root branching, n = 15. Data shown are means ± the SE. Different letters above bars indicate significant differences between treatments (P < 0.05).

To investigate the involvement of the auxin signaling pathway in root growth induced by S. bovinus VOCs, we used an auxin inhibitor and A. thaliana auxin-related mutants to determine whether the root growth promotion effect was retained under different treatments. The promoting effect of S. bovinus, β-cedrene and α-humulene on P. massoniana root growth was repressed by the auxin synthesis inhibitor yucasin (Fig. 6a,b). At yucasin concentrations of 50 μM and 100 μM, P. massoniana lateral roots exposed to β-cedrene or α-humulene did not develop or even wilted (Fig. 6a). By contrast, the auxin transport inhibitor NPA had little effect on P. massoniana lateral root growth induced by S. bovinus or β-cedrene but weakened the promotion effect of α-humulene on root branching (Fig. 6c, d). At an NPA concentration of 5 μM, the number of P. massoniana lateral roots was significantly reduced by 30.83% (Fig. 6c) compared with those of control seedlings, which were not subject to NPA.

Figure 6. 

Effects of the auxin synthesis inhibitor yucasin and the auxin transport inhibitor 1-N-naphthylphthalamic acid (NPA) on Pinus massoniana root development induced by Suillus bovinus volatile organic compounds (VOCs). Effects of adding different concentrations of the indole-3-acetic acid (IAA) synthesis inhibitor yucasin to medium inoculated with S. bovinus, or containing α-humulene or β-cedrene on a the number of P. massoniana root branches and b root length. The absence of some data for seedlings subjected to 50 M or 100 M yucasin is because P. massoniana wilted under these treatments. Effects of adding different concentrations of the IAA transport inhibitor NPA to medium inoculated with S. bovinus or containing α-humulene or β-cedrene on c the number of P. massoniana root branches and d root length. n = 15. Different letters above bars indicate significant differences between treatments (P < 0.05).

Similarly, both yucasin and NPA restricted the promoting of S. bovinus, β-cedrene, and α-humulene on the lateral root development of the nonhost A. thaliana (Fig. 7a, b). However, NPA treatment did not significant impact β-cedrene-induced LRF (Fig. 7b).

Figure 7. 

Effects of the auxin synthesis inhibitor yucasin and the auxin transport inhibitor 1‐N‐naphthylphthalamic acid (NPA) on Arabidopsis thaliana lateral root branch development induced by Suillus bovinus volatile organic compounds (VOCs). a Effects of adding different concentrations of yucasin to medium inoculated with S. bovinus or containing β-cedrene or α-humulene; b effects of adding different concentrations of NPA to medium inoculated with S. bovinus or containing β-cedrene or α-humulene. n = 15. Different letters above bars indicate significant differences between treatments (P < 0.05).

Similar to the wildtype A. thaliana experiment, exposure to S. bovinus VOCs induced more root branching in most A. thaliana mutants compared with those that were not exposed to S. bovinus VOCs (NM treatment) (Fig. 8). Interestingly, exposure to S. bovinus VOCs did not increase the number of lateral roots in pin1 and tir1 lines (Fig. 8a) but did promote primary root elongation of the pin1 line (Fig. 8b). The promotion of LRF by β-cedrene and α-humulene was not observed in the arf19 line (Fig. 9a, c). Furthermore, α-humulene did not induce branching in the tir1 line (Fig. 9b, c). Additionally, while S. bovinus VOCs seemed to stimulate shoot growth (Fig. 8b, data not collected), neither β-cedrene nor α-humulene had this effect (Fig. 9c).

Figure 8. 

Effects of auxin-related gene function on Suillus bovinus-induced Arabidopsis thaliana root development. a Number of root branches in A. thaliana col1 wildtype and in the T-DNA insertion mutant lines yuc1, aux1, pin1, pin3, arf19, and tir1 grown in the presence of S. bovinus (Sb) volatile organic compounds (VOCs) and in the absence of S. bovinus VOCs (NM) (n = 15). Error bars indicate ± the SE. *, P < 0.05; **, P < 0.01; ns, no significant difference between treatments. b Photographs showing the influence of S. bovinus VOCs on root architecture. Scale bar: 2 cm.

Figure 9. 

Effects of auxin-related gene function on β-cedrene- and α-humulene-induced Arabidopsis thaliana root development. Number of root branches in A. thaliana col1 wildtype and in the T-DNA insertion mutant lines yuc1, aux1, pin1, pin3, arf19, and tir1 (a) grown in the presence of β-cedrene (shown in red) or in the absence of β-cedrene (shown in black), and (b) grown in the presence of α-humulene (shown in red) or in the absence of α-humulene (shown in black) (n = 15). Error bars indicate ± the SE. “*, P < 0.05; **, P < 0.01; “ns, no significant difference between treatments. c Photographs showing the influence of β-cedrene and α-humulene on root architecture. Scale bar: 2 cm.

To explore the role of SQTs in establishing ECM symbiosis, we observed the formation of ECM symbionts between P. massoniana and S. bovinus treated with β-cedrene and α-humulene. Overall, there was no significant difference in P. massoniana growth across NM, M, M+β-cedrene, and M+α-humulene treatments; however, the root development of seedlings inoculated with S. bovinus (M) was significantly greater than that of uninoculated seedlings (NM) (Suppl. material 1: table S2). Compared with the NM treatment, the M, M+β-cedrene and M+α-humulene treatments significantly increased the root/shoot ratio of P. massoniana by 47.06%, 47.06% and 41.18%, respectively (Suppl. material 1: table S2). Furthermore, the presence of β-cedrene and α-humulene increased the proportion of P. massoniana roots infected by S. bovinus, particularly β-cedrene (Suppl. material 1: fig. S3). Colonization of P. massoniana roots by S. bovinus led to the development of dichotomous branches mycorrhizae with well-developed mantle and Hartig net (Fig. 10). By contrast, under the NM treatment, there were no hyphae present between cells or on the root surface (Fig. 10d, h). The presence of β-cedrene or α-humulene did not significantly affect the depth of Hartig nets and the thickness of the mantle in the ECM symbiotic between P. massoniana and S. bovinus (Fig. 10i, j).

Figure 10. 

Effects of β-cedrene and α-humulene on the formation of an ectomycorrhizal (ECM) symbiosis between Pinus massoniana and Suillus bovinus. Morphology of dichotomous root branches of seedlings subjected to: a S. bovinus inoculation (M), b M+β-cedrene; c M+α-humulene treatment; or d no inoculation (NM). Cross-sections showing the microstructure of the P. massoniana–S. bovinus ECM association of seedlings subjected to: e M; f M+β-cedrene; g M+α-humulene; or h NM; the dye used was 0.03% trypan blue. i Depth of Hartig nets and j thickness of mantle under different treatments. Error bars indicate ± the SE, n = 5. Different letters above bars indicate significant differences between treatments. Red arrows indicate extraradical mycelium, white arrows indicate the mantle, and yellow arrows indicate Hartig nets.

Traditionally, the beneficial functions of rhizosphere microbes in plant growth and development were thought to rely on direct microbe–root contact (Bais et al. 2004; Beauregard et al. 2013). However, VOCs released by microorganisms have been shown to enhance plant growth from a distance (Ditengou et al. 2015; Li et al. 2022; He et al. 2023; Wang et al. 2023). In this study, S. bovinus significantly promoted LRF and roots elongation in P. massoniana via VOCs, without direct root contact. Additionally, it induced root growth toward the source of the VOCs. LRF was considered to be a typical morphology in the early stage of mycorrhizal colonization because it was enhancing mycelium–root contact opportunities (Barker et al. 1998). These findings suggested that S. bovinus VOCs facilitates the communication and recognition by promoting P. massoniana LRF and directional root growth during the presymbiotic stage. Significantly, the effect of S. bovinus VOCs in promoting LRF may not be host-specific (Feng et al. 2022).

In addition to VOCs emitted by S. bovinus, we observed that VOCs emitted by other symbiotic fungi, including the ECM fungi S. luteus and S. citrinum, as well as the DSE P. fortinii, also stimulated P. massoniana LRF. Previous studies have similarly demonstrated that VOCs from other ECM fungi (Ditengou et al. 2015) and non-ECM fungi (Li et al. 2022; He et al. 2023; Liu et al. 2024) exhibit comparable functions. These findings suggest that plant root responses to VOCs from beneficial symbiotic fungi may not be specific to particular fungal species. Interestingly, pathogens also interact with plants via VOCs before physical contact (Venturi and Fuqua 2013). In our study, we observed that VOCs released by F. oxysporum inhibited the root growth of P. massoniana. In contrast, Moisan et al. (2019) and Hernández-Calderón et al. (2018) reported that VOCs emitted by pathogens can promoted plant growth, similar to the effects of VOCs from symbiotic fungi. Conversely, other studies have shown that VOCs released by pathogens may not affect plant growth (Tahir et al. 2017). Variations in plant responses to pathogen-emitted VOCs may result from differences in VOCs composition and concentration. Additionally, Peng et al. (2021) found that saprophytic fungi can act as mycorrhizal partners to promote host plant growth, suggesting that their VOCs may also play a similar role. However, in this study, VOCs emitted by the saprophytic fungus L. perlatum had no significant effect on P. massoniana growth. This lack of effect may be attributed to differences in fungal species, or it may suggest that saprophytic fungi L. perlatum necessitate direct contact to facilitate plant growth, which deserves further study. In this study, we propose that P. massoniana’s response to VOCs depends on the relationship between the fungi and the plant: VOCs from mutualistic fungi tend to promote root growth, those from pathogens inhibit growth, and those from saprophytic fungi have no effect.

Previous studies have shown that the ECM fungus Cenococcum geophilum, which does not release SQTs, fails to induce LRF before physical contact, indicating the importance of SQTs in this process (Agger et al. 2009; Ditengou et al. 2015). In this study, the emission of SQTs by S. bovinus was crucial for promoting lateral roots’ emergence. This conclusion was further supported by experiments using lovastatin, an inhibitor of SQT biosynthesis, which significantly reduced the positive effect of S. bovinus VOCs on P. massoniana LRF, with similar effects observed in the non-host A. thaliana. These findings are consistent with previous reports (Ditengou et al. 2015; Li et al. 2022) and underscore that the effects of S. bovinus VOCs are not specific to host plants.

Furthermore, we identified two SQTs, β-cedrene and α-humulene, which were exclusively present in the VOC profiles of the three ECM fungi tested, indicating that these two SQTs may be specific to ECM fungi. These SQTs were sufficient to stimulate LRF in both the host P. massoniana and the non-host A. thaliana in the absence of S. bovinus, and even when SQT biosynthesis was inhibited by lovastatin. The roles of these two SQTs in plant–microbe interactions have not been previously demonstrated; however, their analogs (such as β-caryophyllene and cedrene) have been shown to induce plant growth (Minerdi et al. 2011; Yamagiwa et al. 2011; Li et al. 2022). Our study demonstrates their function as signaling compounds in the interaction between P. massoniana roots and S. bovinus at the presymbiotic stage, which is a novel finding.

Further analysis of their roles during the symbiotic formation of the P. massoniana–S. bovinus symbiosis revealed that β-cedrene and α-humulene do not significantly influence ECM symbiosis development, including the formation of the mantle and Hartig net. This is consistent with findings by Plett et al. (2024), who also observed that the addition of terpenes did not affect Hartig net development. Given that SQTs enhance the proportion of P. massoniana roots infected by S. bovinus, we hypothesize that VOCs signaling molecules may increase the likelihood of contact between roots and ECM fungal hyphae, thereby promoting favorable conditions for ECM formation. Meanwhile, other signaling molecules, such as phytohormones like ABA, as reported by Hill et al. (2022), may play crucial roles in initiating symbiosis. However, it remains unclear whether other VOC profiles contain signaling molecules besides β-cedrene and α-humulene that are involved in communication during the ECM presymbiotic stage. Future research should focus on large-scale screening for additional volatile signaling compounds and further analysis of their molecular mechanisms to determine their role in triggering LRF.

Auxin, a crucial plant hormone, is widely recognized for its role in regulating LRF (Fukaki and Tasaka 2009). Previous studies have shown that microbial VOCs stimulate root branching by activating the host auxin signaling pathway (Felten et al. 2009; He et al. 2023; Liu et al. 2024). In this study, we demonstrated that S. bovinus VOCs significantly enhanced P. massoniana root branching and increased IAA accumulation in P. massoniana. These findings highlight the pivotal role of auxin in root development triggered by S. bovinus VOCs. Experiments using auxin biosynthesis and transport inhibitors further supported this conclusion. Yucasin, an auxin biosynthesis inhibitor, attenuated the effects of S. bovinus β-cedrene and α-humulene on LRF in both host and non-host plants, suggesting that YUC –mediated auxin biosynthesis is crucial in S. bovinus VOCs-induced LRF. This corroborates findings by Li et al. (2022), Zhang et al. (2022) and Wang et al. (2021). However, Ditengou et al. (2015) reported that SQTs released by L. bicolor could stimulate LRF independently of the auxin pathway. The discrepancy between these findings and those of other studies may stem from differences in the VOC profiles of different fungal species. Moreover, studies have shown that disrupting auxin polar transport alters auxin accumulation in roots and affects root growth (Zamioudis et al. 2013; Li et al. 2022; He et al. 2023). For instance, the ECM host Populus and non-host A. thaliana exposed to L. bicolor VOCs showed restricted LRF after NPA treatment (Felten et al. 2009), and a recently reported study had similar results (Qin et al. 2024). In our study, only the non-host A. thaliana was affected by NPA treatment, whereas S. bovinus VOCs continued to promote LRF in the host plant P. massoniana. This suggests that S. bovinus VOCs may contain auxin-independent signaling molecules that stimulate P. massoniana LRF. Importantly, S. bovinus VOC effects are not host specific, and the underlying mechanism may vary, potentially operating differently in different plant species. Furthermore, we observed that α-humulene induced LRF in both P. massoniana and A. thaliana via auxin polar transport, whereas β-cedrene did not, indicating that different VOCs enhance LRF through distinct mechanisms.

Interestingly, S. bovinus VOCs, β-cedrene and α-humulene, promoted LRF in A. thaliana yuc1 mutants, contrasting with the yucasin results, which inhibits auxin biosynthesis. Since 11 members of the YUCCA gene family have been detected in A. thaliana (Mashiguchi et al. 2011), other YUCCA genes may be involved in this process. The AUX1/LAX and PIN protein families have distinct roles in auxin influx and efflux, respectively (Pacifici et al. 2015). The effect of S. bovinus VOCs on A. thaliana lateral root growth was abolished in pin1 mutants but remained unchanged in aux1 mutants, indicating that PIN1 is involved in the development of lateral roots stimulated by S. bovinus VOCs. This further supports our hypothesis that S. bovinus VOCs stimulate A. thaliana LRF through auxin polar transport, requiring a functional auxin efflux system, as also suggested by previous studies (Sun et al. 2020; He et al. 2023). Treatment with NPA attenuated the ability of α-humulene to stimulate LRF in both P. massoniana and non-host A. thaliana, whereas β-cedrene was unaffected. Interestingly, experiments with A. thaliana mutants indicated that the promotion effects of α-humulene may not rely on auxin polar transport, given that it induced LRF in aux1, pin1, and pin3 mutants. These results contradict the outcomes of the NPA inhibitor trail, suggesting that α-humulene might stimulate LRF through other auxin influx and efflux carriers, which warrants further investigation.

Auxin signal transduction is crucial for regulating plant LRF in response to microbial VOCs (Tyagi et al. 2018). Studies by Li et al. (2022) and He et al. (2023) suggest that microbial VOCs promote LRF through complete auxin signal transduction, a finding supported by our study. In plants, the primary auxin signaling pathway involves the auxin receptor TIR1/AFB, AUX/IAA signal response factors, and ARF transcription factors (Hayashi 2012). While S. bovinus VOCs continued to promote root branching in A. thaliana arf19 mutants, neither β-cedrene nor α-humulene alone could induce LRF in these mutants, indicating that the regulatory mechanism of S. bovinus VOCs on root branching likely involves multiple signaling molecules rather than a single pathway.

Additionally, S. bovinus VOCs appeared to stimulate shoot growth in A. thaliana. This could be due to enhanced lateral root formation, which improves nutrient absorption and promotes shoot growth (Zhang et al. 2022), or the VOCs may contain unidentified plant growth regulators or signaling molecules, as neither β-cedrene nor α-humulene had this effect. Further research is needed to explore these hypotheses.

The authors have declared that no competing interests exist.

No ethical statement was reported.

All the fungal strains used in this study have been legally obtained, respecting the Convention on Biological Diversity (Rio Convention).

This research was supported by the Guizhou Provincial Science and Technology Projects (QKHZHYD[2024]044), the National Natural Science Foundation of China (32360372 and 31971572) and the Cultivation Project of Guizhou University [2020]47.

WYF was involved in the formal analysis, investigation, methodology, writing – original draft, review and editing. GYY was involved in the investigation. XGS was involved in conceptualization, funding acquisition, methodology, supervision, writing – review and editing. GJD was involved in supervision, funding acquisition, writing – review and editing.

Wanyan Feng https://orcid.org/0000-0001-5645-6742

Xueguang Sun https://orcid.org/0000-0002-7895-6702

Guiyun Yuan https://orcid.org/0009-0005-5345-4238

Guijie Ding https://orcid.org/0009-0008-5760-0425

The original data presented in this study are included in the article/Suppl. material 1. Further inquiries can be directed to the corresponding author.