兰科植物种子细微如尘、无胚乳。自然条件下兰科种子只有依赖真菌菌丝侵入为其提供营养才能萌发，因而兰科植物都有天然的菌根共生特性。手参Gymnadenia conopsea (L.) R. Br.是一种广泛分布于欧亚大陆的多年生地生兰科植物，其块茎在我国作为传统中药，亦有悠久的民间食用历史。然而由于长期过度采挖、栖息地破坏和自然种群更新缓慢等因素，手参野生资源正面临严重枯竭，现已被列为国家II级保护植物。手参目前还未实现人工栽培，在本课题组前期研究中，成功获得了促进手参种子萌发的真菌菌株Ceratobasidium sp. GS2（下文统称为GS2），在实验室及野外均成功实现了手参种子的萌发，由此表明通过共生萌发实现手参的资源再生是可行的。虽然手参种子在菌株GS2作用下可以萌发，但对于其真菌共生萌发过程与机制方面目前仍知之甚少。

本研究以手参种子及前期获得的促萌发真菌GS2为研究材料，围绕“种子的真菌共生萌发”这一关键过程，系统性开展了微生物组、萌发真菌选择特异性、共生碳源偏好性、亚细胞水平和时间序列转录组等多层次研究，全面解析了手参种子共生萌发过程中的核心微生物群落结构、菌根共生特性及能量代谢机制，旨在建立宏观–微观相结合的研究框架，为手参的人工繁育与濒危兰科药用植物种质保育提供理论支撑。本文主要研究结果如下：

手参种子在与真菌共生萌发过程中，会招募周围环境中的细菌类群，尤其以假单胞菌属（Pseudomonas）细菌为优势类群。体外共生萌发结果显示，假单胞菌虽然不能直接促进种子萌发，但与GS2共同接种具有协同作用，能显著促进手参原球茎的发育；另外，研究还发现地理隔离与块茎发育阶段共同塑造了手参块茎的内生真菌群落结构且内生真菌群落存在明显的功能分化。

手参种子表现出菌根真菌选择的宽泛性。多种来源的角担菌属Ceratobasidium菌株可以促进手参种子萌发，且这些菌株并不局限于系统发育树某一分支，同时手参种子的萌发效率呈现出显著地域差异，西南地区手参种子较北方地区的手参种子具有更高的菌根特异性。

多数碳源处理GS2抑制手参种子的萌发，仅海藻糖和可溶性淀粉具有显著的促进作用，且海藻糖效果尤为突出。转录组分析表明，海藻糖通过激活促萌发真菌的核糖体生物合成、蛋白质折叠等通路，同时抑制细胞自噬，对真菌代谢状态进行调控。推测海藻糖可能导致真菌优先支持共生过程而非营养生长，抑或能够直接被真菌吸收，直接作为碳源转运给种子。

手参种子的真菌共生萌发过程总体上可划分为三个阶段：分别为吸胀期（0-12d），真菌侵染期（12-15d）和原球茎发育期（15-30d）。亚细胞水平显示手参种子在吸胀期可以产生乙醛酸循环体代谢自身储备的脂类物质，同时细胞内所有脂质体逐渐融合为一个大型脂滴并在真菌侵染前完成融合。真菌侵染时，菌丝内出现大量脂质体，推测脂类物质从种子转运至真菌。在原球茎发育期，共生界面周围可以观察到大量物质运输相关的细胞器，说明活跃的物质转运发生在共生界面上；同时衰老菌丝内观察到脂滴和光面内质网，表明其具有活跃的脂类代谢活动。

时间序列转录组研究表明，手参种子在吸胀期乙醛酸循环及脂肪酸β氧化通路基因表达量较高，脂类分解代谢的活力较强，表明种子可以利用自身储备的脂类营养。真菌侵染期种子发生了显著的基因表达重编程，其中脂类分解代谢相关基因被显著抑制，同时碳水化合物如蔗糖、海藻糖的分解代谢及糖酵解相关通路上的基因被激活，表明种子由代谢自身储备的脂类营养转为代谢真菌供给的碳水化合物。同时，对比胞内及胞外菌丝基因表达情况，发现胞内菌丝中不饱和脂肪酸生物合成和脂肪酸代谢通路中的基因显著上调。

综上所述，本研究聚焦于手参种子的真菌共生萌发这一复杂过程，发现假单胞菌属细菌和角担菌科真菌可能构成手参种子萌发过程的核心微生物群落，所获得的假单胞菌菌株亦有开发为生物菌肥的潜力；其次发现，手参种子对角担菌属真菌需求的宽泛性可能与其广域分布存在密切的相关性；最后，本文基于手参种子与GS2的共生萌发体系，提出了一个兰科植物种子萌发过程中营养转运的模型，即共生建立后真菌向种子转运碳水化合物，同时种子向真菌转运脂类，而非之前普遍认为的，真菌无偿地供给种子所需的营养。本研究不仅丰富了兰科植物种子-真菌互作的理论体系，也为手参等濒危兰科植物的人工繁育与资源可持续利用提供了理论基础与技术支撑。

Orchid seeds are dust-like and lack endosperm. Under natural conditions, their germination depends entirely on fungal hyphal invasion for nutrient acquisition, making orchid species inherently mycorrhizal. Gymnadenia conopsea (L.) R. Br., a perennial terrestrial orchid widely distributed across the Eurasian continent, has long been used in traditional Chinese medicine due to the medicinal properties of its tubers and also has a history of consumption in folk practices. However, owing to factors such as overharvesting, habitat destruction, and slow natural population turnover, wild resources of G. conopsea are now severely depleted, and the species has been listed as a National Class II Protected Plant in China. Although artificial cultivation has not yet been achieved, our research group previously isolated a mycorrhizal fungal strain Ceratobasidium sp. GS2 (hereafter referred to as GS2) that effectively promotes seed germination of G. conopsea, both under laboratory and field conditions, demonstrating the feasibility of regenerating G. conopsea resources via symbiotic germination. However, the mechanisms underlying fungal-induced germination in G. conopsea remain largely unknown.

In this study, using G. conopsea seeds and the previously identified GS2 strain as experimental materials, we systematically investigated the key process of symbiotic seed germination through a multi-level approach, including microbiome analysis, fungal specificity tests, carbon source preference assays, subcellular observations, and time-series transcriptomics. We aimed to comprehensively elucidate the structure of the core microbial community, the characteristics of orchid mycorrhizal symbiosis, and the underlying metabolic mechanisms, thereby establishing an integrative macro–micro research framework to inform the artificial propagation and germplasm conservation of endangered medicinal orchids. The major findings of this study are as follows:

1. During the process of symbiotic germination, G. conopsea seeds actively recruit bacterial taxa from the surrounding environment, with Pseudomonas spp. being dominant. Although Pseudomonas strains alone do not directly promote seed germination, co-inoculation with GS2 significantly enhances protocorm development, indicating a synergistic effect. Additionally, we found that both geographic isolation and developmental stages of tubers jointly shape the structure of the endophytic fungal community, which exhibits clear functional differentiation.

2. G. conopsea seeds exhibit broad specificity toward mycorrhizal fungi. Multiple Ceratobasidium strains from diverse sources were able to induce seed germination, and these strains were not restricted to a specific clade within the phylogenetic tree. Notably, the germination efficiency varied significantly by seed provenance: seeds from southwestern China showed higher specificity toward mycorrhizal fungi compared to those from northern regions.

3. Most carbon sources inhibited seed germination in the presence of GS2, except for trehalose and soluble starch, which had significant promotive effects, with trehalose showing the most pronounced effect. Transcriptome analysis revealed that trehalose promotes fungal metabolic activity by upregulating pathways related to ribosome biogenesis and protein folding while suppressing autophagy. This suggests that trehalose may redirect fungal metabolism toward symbiotic support rather than vegetative growth, or may be directly absorbed and utilized by the fungus as a carbon source and transferred to the seed.

4. The symbiotic germination of G. conopsea seeds can be divided into three phases: imbibition (0–12 days), fungal colonization (12–15 days), and protocorm development (15–30 days). Subcellular observations revealed that seeds initiate the glyoxylate cycle during imbibition, utilizing stored lipids for energy. Lipid bodies within seed cells gradually fused into a large single droplet prior to fungal colonization. During colonization, hyphae exhibited abundant lipid bodies, suggesting a transfer of lipids from seed to fungus. In the protocorm development phase, various transport-related organelles were observed near the symbiotic interface, indicating active material exchange. Lipid droplets and smooth endoplasmic reticulum were also found in senescent hyphae, implying ongoing lipid metabolism.

5. Time-series transcriptome analysis revealed that genes involved in the glyoxylate cycle and fatty acid β-oxidation were highly expressed during the imbibition phase, indicating robust lipid catabolism and self-sustained nutrient utilization by seeds. Upon fungal colonization, substantial transcriptional reprogramming occurred, with genes associated with lipid degradation being significantly downregulated, while those involved in carbohydrate metabolism—including the breakdown of sucrose and trehalose—and glycolysis were activated. This shift reflects a metabolic transition from endogenous lipid utilization to fungal-supplied carbohydrates. Comparative transcriptomic analysis of intra- and extraradical hyphae showed significant upregulation of genes involved in unsaturated fatty acid biosynthesis and fatty acid metabolism in the intracellular hyphae.

In conclusion, this study focuses on the complex process of fungal symbiotic germination in G. conopsea seeds and identifies Pseudomonas bacteria and Ceratobasidiaceae fungi as core microbial components potentially involved in seed germination. The isolated Pseudomonas strains also show promise as candidate biofertilizers. Furthermore, the broad compatibility of G. conopsea seeds with Ceratobasidium fungi may underlie its wide geographical distribution. Finally, based on the GS2-induced germination system, we propose a novel nutrient transfer model in orchid seed germination, in which fungi deliver carbohydrates to the seeds while seeds, in turn, supply lipids to the fungi—challenging the traditional view that nutrients are unidirectionally provided by fungi. This study not only advances the theoretical understanding of orchid seed–fungus interactions but also provides a solid theoretical and technical foundation for the artificial propagation and sustainable utilization of endangered medicinal orchids such as G. conopsea.