<i>Stigmatodactylus sikokianus</i> (Orchidaceae) mainly acquires carbon from decaying litter through association with a specific clade of Serendipitaceae

Abstract

Mycorrhizas usually engage in mutualism, in which soil fungi contribute minerals to plants that in turn supply photosynthetically fixed carbon (C) to their fungal partners (Smith & Read, 2008). However, the functional status of such symbioses in orchids remains somewhat controversial. Although all orchids depend on symbiotic fungi for essential nutrients (including C) during their initial development (Dearnaley et al., 2016), most orchid species exhibit some level of autotrophy at the adult stage. As such, some researchers have considered that green orchids are released from their dependency on heterotrophy for C gain at the adult stage (e.g. Alexander & Hadley, 1985). Indeed, radiocarbon tracer experiments have demonstrated that the green orchid Goodyera repens engages in a mutualistic symbiosis with its fungal partner Ceratobasidium cornigerum (Cameron et al., 2006, 2008), thereby indicating that orchid mycorrhizas can function as typical mycorrhizas whose net C flow is from plant to fungus. However, many green orchids obtain C both from photosynthesis and from symbiotic fungi (mixotrophy or partial mycoheterotrophy) (Gebauer & Meyer, 2003; Bidartondo et al., 2004; Selosse et al., 2004). Stable isotope analysis has identified that both fully mycoheterotrophic orchids and several green orchids are highly enriched in 13C and 15N, reflecting the incorporation of fungi-derived organic C and nitrogen (N) from their fungal symbionts (Gebauer & Meyer, 2003; Bidartondo et al., 2004). Intriguingly, although all fully mycoheterotrophic orchids and many mixotrophic orchids depend on either ectomycorrhizal (ECM) fungi or saprotrophic nonrhizoctonia fungi (Merckx, 2013; Suetsugu et al., 2019, 2021a), most green orchids are associated with nonECM rhizoctonias, including Ceratobasidiaceae, Tulasnellaceae and Serendipitaceae (Dearnaley et al., 2013). Therefore, nonECM rhizoctonias might not support the C demands of adult orchids exhibiting a high degree of mycoheterotrophy. Although recent studies have also suggested that mixotrophy may occur in rhizoctonia-associated orchids (Selosse & Martos, 2014; Gebauer et al., 2016; Suetsugu et al., 2021a), a net flow of C from fungus to plant in rhizoctonia-associated orchids is still debated (Lallemand et al., 2017; Jacquemyn et al., 2021). Current evidence also suggests that rhizoctonias are less suited to meet the large C demand of fully mycoheterotrophic adult orchids than ECM fungi (Schweiger et al., 2018). Several studies have also reported that shifts in associated fungi, from rhizoctonias to ECM fungi, often precede the evolution of full mycoheterotrophy (Bidartondo et al., 2004; Selosse et al., 2004). However, given that the activities of saprotrophic fungi are correlated with temperature and moisture (Martos et al., 2009), and at least some nonECM rhizoctonia fungi are saprotrophic (Dearnaley et al., 2013; Selosse & Martos, 2014), warm and high moisture conditions might enable rhizoctonias to meet the C demands of adult orchids. We also note that rhizoctonias are not pure saprotrophs. The nutritional mode of rhizoctonia fungi is difficult to understand in detail because rhizoctonias usually do not produce conspicuous fruiting bodies that can be sampled for isotopic analysis (Selosse & Martos, 2014); furthermore, the transition from saprotrophy to endophytism has evolved multiple times in rhizoctonias (Veldre et al., 2013; Weiß et al., 2016). Consequently, it remains controversial whether rhizoctonias that are otherwise detected from the orchid roots are saprotrophic or endophytic (Selosse & Martos, 2014). In this respect, transit time (i.e. the time from initial photosynthesis to final C utilization by mycoheterotrophic plants, analogous to diet age in animals) can be estimated by measuring the concentration of 14C derived from the nuclear bomb tests of the 1950s and 1960s (Hyodo et al., 2006; Hatté et al., 2020; Suetsugu et al., 2020). The radiocarbon approach will help investigate whether nonECM rhizoctonias detected from orchids exhibiting high levels of mycoheterotrophy are litter-decaying or endophytic, given that the association with endophytic fungi receiving current-year photosynthates must result in younger C ages. In this study, we focused on the nutritional mode of Stigmatodactylus sikokianus distributed in warm temperate regions (Fig. 1). The genus Stigmatodactylus has sometimes been classified as a fully mycoheterotrophic plant (Furman & Trappe, 1971; Campbell, 2014), although it retains chlorophyll in its stems and small leaves. Based on in vitro isolation, the mycobionts of S. sikokianus have been identified as members of Serendipitaceae, a group of nonECM rhizoctonias (Yagame & Yamato, 2008). Because S. sikokianus often grows in decaying litter of the nonECM Cryptomeria japonica plantations without any adjacent ECM trees, S. sikokianus may depend on C from decaying litter through an association with Serendipitaceae. Accordingly, we investigated the physiological ecology of S. sikokianus to improve our understanding of mycoheterotrophic evolution in orchids. Specifically, the fungal partners of S. sikokianus were identified using high-throughput DNA sequencing, while stable isotope (13C and 15N) analysis was used to confirm the mixotrophy in S. sikokianus. In addition, radiocarbon (Δ14C) analysis was used to test whether rhizoctonias associated with S. sikokianus are mainly litter-decaying, and consequently, whether S. sikokianus acquires 14C-enriched bomb C from decaying litter. All samples were collected in Kochi Prefecture, Japan on 2 September 2017 (see Supporting Information Methods S1 for detailed methodology). Based on the metabarcoding technique, community profiling revealed that S. sikokianus was exclusively associated with six operational taxonomic units (OTUs) belonging to Serendipitaceae (Weiß et al., 2016). The six OTUs belonging to Serendipitaceae (56 841 sequencing reads in total) were detected in all S. sikokianus individuals. We note that the selected primers are primarily designed for Basidiomycota. Therefore, although they are suitable for studying fungal diversity in orchids (Taylor & McCormick, 2008; Waud et al., 2014), it is possible that some other fungi such as ascomycetous endophytes were not amplified due to primer bias. Because the mycobionts detected in the present study belonged to Serendipitaceae, phylogenetic analysis was conducted on representative Sebacinales sequences. Maximum-likelihood phylogenetic trees showed that the mycobionts of S. sikokianus formed a highly supported clade with the mycobionts previously detected in S. sikokianus (Yagame & Yamato, 2008). These mycobionts also formed a highly supported clade with mycobionts identified in several green orchids and the Serendipita vermifera species complex (Fig. 2; Deshmukh et al., 2006; Yagame & Yamato, 2008). The S. vermifera species complex is known to comprise mycobionts of green orchids, including Acianthus, Caladenia, Elythranthera, Eriochilus, Glossodia, Microtis, Pheladenia and Prasophyllum (Deshmukh et al., 2006). Along with the genus Stigmatodactylus, these orchid species belong to the tribe Diurideae, suggesting that S. sikokianus fungal associations have not been acquired secondarily and instead have been derived from the fungal partnerships of their common ancestors. It has been suggested that endophytic and saprotrophic members of Serendipitaceae cannot provide sufficient C resources to adult orchids with high levels of mycoheterotrophy, whereas ECM members of Sebacinaceae may regularly obtain large amounts of C from trees (Weiß et al., 2016; Yagame et al., 2016). By contrast, our results indicate that such fungal shifts are not the sole prerequisite for increasing heterotrophy (shown in the following two sections). Despite the photosynthetic capability of S. sikokianus, the present study revealed that the orchid was highly enriched in 13C. The δ13C values of S. sikokianus (−23.3 ± 0.7‰; n = 6) were significantly higher than those of not only autotrophic reference plants (−33.8 ± 0.8‰; n = 8; P < 0.001; Fig. 3; Table S1) but also decaying litter entangled with S. sikokianus rhizomes (−28.4 ± 0.9‰; n = 6; P < 0.001). Notably, the 13C enrichment factor (10.6 ± 1.4‰, n = 6) was slightly higher than the mean enrichment factor of mycoheterotrophic orchids that exploit not only ECM fungi (8.2 ± 1.3‰, n = 94) but also litter-decaying fungi (8.7 ± 1.1‰, n = 24) (Martos et al., 2009; Ogura-Tsujita et al., 2009; Hynson et al., 2013; Lee et al., 2015; Suetsugu et al., 2019, 2020), thereby corroborating the hypothesis that S. sikokianus mainly acquires C nutrition from its fungal associations. Such a strong mycoheterotrophic ability could compensate for the low production of photosynthetic C associated with reduced leaf size and inhabiting deeply shaded forest understoreys. In contrast to the high 13C enrichment, the 15N enrichment factor of S. sikokianus (2.5 ± 0.6‰, n = 5) was much lower than that of mycoheterotrophic orchids that exploit ECM fungi (11.6 ± 3.1‰, n = 94) (Hynson et al., 2013) and was instead similar to the mean enrichment factor of mycoheterotrophic orchids that exploit saprotrophic fungi (4.8 ± 1.5‰, n = 42) (Martos et al., 2009; Ogura-Tsujita et al., 2009; Lee et al., 2015; Suetsugu et al., 2019, 2020). The relatively low 15N enrichment probably reflects the saprotrophic status of the mycobionts, which are generally less 15N-enriched than ECM fungi (Mayor et al., 2009). Therefore, the 13C and 15N enrichment patterns suggest that S. sikokianus is a nearly fully mycoheterotrophic plant species that obtains nutrients from litter-decaying fungi, although some uncertainties have remained due to limitations of the stable isotopic approach, including differences in fungal identities and plant metabolism (e.g. Suetsugu et al., 2018; Jacquemyn et al., 2021). It is known that the 13C enrichment of some rhizoctonia-associated mixotrophic orchids is too low to be detected (Schweiger et al., 2018). Several recent studies have provided evidence that endophytism results in depleted 13C levels (Halbwachs et al., 2013); therefore, the endophytic status of rhizoctonia fungi might be responsible for the absent or relatively low 13C enrichment observed in rhizoctonia-associated orchids (Selosse & Martos, 2014). However, the 13C enrichment factors in both S. sikokianus and albino mutants of so-called rhizoctonia-associated orchids, including Goodyera velutina (8.8 ± 0.9‰; Suetsugu et al., 2019) and Cypripedium debile (9.8 ± 0.5‰), are similar to those of mycoheterotrophic orchids exploiting saprotrophic nonrhizoctonia fungi. These results suggested that mycobionts behave more like saprotrophic fungi than endophytic fungi. Litter-decaying rhizoctonias as well as litter-decaying nonrhizoctonia fungi can support their development up to the adult stage in (nearly) fully mycoheterotrophic orchids, at least under favorable hot and wet conditions. By contrast, endophytic rhizoctonias may provide insufficient C to support their growth. Since the Δ14C values of atmospheric CO2 were globally elevated by nuclear weapons testing during the early 1950s but declined after the atmospheric nuclear test ban treaty in 1963 (Burchuladze, 1989), the age of C in organic matter (such as in orchid tissues) can be estimated by measuring levels of Δ14C (Hatté et al., 2020; Suetsugu et al., 2020). Indeed, Suetsugu et al. (2020) reported that the transit times of mycoheterotrophic plants associated with wood-decaying fungi (10- to 40-yr-old C) were much longer than those of mycoheterotrophs exploiting ECM fungi (0- to 1.4-yr-old C) (Hatté et al., 2020; Suetsugu et al., 2020). Both S. sikokianus (23.8 ± 4.6‰; n = 6; P < 0.001) and decaying litter entangled with S. sikokianus rhizomes (27.0 ± 7.7‰; n = 6; P < 0.001; Table S1) exhibited significantly higher Δ14C values than the two autotrophic plant species Neolitsea sericea (10.0 ± 2.9‰; n = 4) and Pinellia ternata (10.9 ± 4.2‰; n = 4), indicating that they had assimilated more bomb-derived C. The Δ14C values between S. sikokianus and decaying litter did not differ significantly. Given that the transit time of decaying litter is usually less than 10 yr (Hobbie et al., 2002), S. sikokianus and surrounding decaying litter must have used organic matter produced after the bomb peak (c. 1964) (Burchuladze, 1989). Consequently, the transit times of S. sikokianus and decaying litter were estimated to be 4.0 ± 0.8 and 4.5 ± 1.5 yr before the sampling period, respectively. These transit times are consistent with other litter feeders/consumers, including litter-decaying fungi (Hobbie et al., 2002; Hyodo et al., 2006, 2015). Therefore, although the ecology of the Serendipita spp. was not directly assessed, owing to their microscopic habits (Selosse & Martos, 2014), the transit time of C from the atmosphere to S. sikokianus suggests strongly that the symbionts are mainly litter-decaying, rather than endophytic, as the latter would result in much younger C ages. It should be noted that radiocarbon analysis provides the mean age of C (Hatté et al., 2020). Owing to the presence of the reduced but green leaves, the incorporation of 14C-enriched bomb C from decaying litter via mycobionts was probably diluted with current-year photosynthesis in S. sikokianus. However, because there was no significant difference between the ∆14C values of S. sikokianus and those of the adjacent decaying litter, the quantity of C fixed by photosynthesis in S. sikokianus is likely to be negligible. Even though high levels of mycoheterotrophy in green orchids are often associated with replacing saprotrophic rhizoctonias with other fungi that probably provide greater and more consistent C supplies (Selosse & Roy, 2009), the present study documents an exciting exception. The approaches included in this study (i.e. molecular barcoding, 13C and 15N measurements, and radiocarbon analysis) provide evidence that S. sikokianus obtains most of its C from decaying litter via the exploitation of saprotrophic rhizoctonias. These findings might contradict previous notions that rhizoctonias are always less capable of meeting the large C demands of adult orchids than nonrhizoctonia fungi (Martos et al., 2009; Dearnaley et al., 2013). However, we note that several orchids exploiting saprotrophic fungi have acquired full mycoheterotrophy (Martos et al., 2009; Ogura-Tsujita et al., 2009; Lee et al., 2015; Suetsugu et al., 2020, 2021b). Our findings indicate that saprotrophic mycobionts, including saprotrophic rhizoctonias, can meet the C demands of adult orchids with high mycoheterotrophy levels, especially under favorable hot and wet conditions. Therefore, other tropical (nearly) fully mycoheterotrophic orchids might acquire most of their C from their fungal partners, even without shifts from rhizoctonias to other fungi. The present study contributes to a deeper understanding of the diversity of mycoheterotrophic interactions in orchids and illustrates the potential importance of organic C re-entry into plants via saprotrophic rhizoctonias. We thank Drs Erik Hobbie, Marc-André Selosse, Jun Matsubayashi and other anonymous reviewers for their constructive comments on the manuscript. We also thank Hisanori Takeuchi for help with the fieldwork and Makoto Taniguchi, Takenori Yamamoto, Takako Shizuka, Takuto Shitara and Michiko Ishida for their technical assistance. We would like to thank Editage (www.editage.com) for English language editing. This work was financially supported by the Mitsubishi Foundation (KS) and JSPS KAKENHI (grant nos. 17H05016 (KS) and 16H02524 (IT). The present study was also supported by the Joint Research Grant for the Environmental Isotope Study of Research Institute for Humanity and Nature. KS planned and designed the research, collected the materials and wrote the initial draft. KS, TFH and HO conducted the laboratory experiments and carried out the analyses. IT supervised stable isotope and radiocarbon experiments. TFH, HO and IT revised the manuscript and approved the final version of the manuscript. The sequence data were deposited in the Sequence Read Archive of the DNA Data Bank of Japan (accession no. DRA011572). Further information can be found online in the Supporting Information section. Table S1 The natural abundance of 13C, 15N and 14C as well as the C and N concentrations in each sample of Stigmatodactylus sikokianus and its neighboring decaying litter and autotrophic plants. Please note: Wiley Blackwell are not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts