Take home message

（1）利用相关性：耐旱细菌与干旱富集的植物化学物质（如茉莉酸和哌啶酸）呈显著正相关。

（2）比较基因组：Maximum-likelihood phylogenetic tree shows the phylogenetic distributions of carbon (C), nitrogen (N), phosphorus (P) and sulfur (S) cycling-related genes in bacterial genome from the rhizosphere bacteria, including both DTB (labelled as ‘sorting’) and non-DTB (labelled as ‘sequencing’) 最大似然系统发生树显示了根际细菌的细菌基因组中碳(C)、氮(N)、磷(P)和硫(S)循环相关基因的系统发生分布，包括DTB(标记为“分类”)和非DTB(标记为“测序”) 【应该是宏基因组拼装的基因组】

（3）功能菌株的全球模式：In parallel, to explore the global distribution of DTB across diverse soil ecosystems, we reanalysed a previously published dataset from the previously published data51, which includes 3,304 soil metagenomes from various global regions. The original study processed these metagenomes to recover 40,039 high-quality metagenome-assembled genomes (MAGs)51. In our analysis, we specifically examined this dataset to identify the previously characterized DTB. We also conducted a search for secondary metabolite biosynthetic gene clusters (BGCs) in the DTB using antiSMASH v6.1.0, focusing on contigs larger than 5 kb (ref. 52). 与此同时，为了探索DTB在不同土壤生态系统中的全球分布，我们重新分析了之前发布的数据集，该数据集来自之前发布的数据51，其中包括来自不同全球区域的3，304个土壤元基因组。最初的研究处理了这些宏基因组，回收了40，039个高质量的宏基因组组装基因组(MAGs)51。在我们的分析中，我们专门检查了这个数据集，以确定先前表征的DTB。我们还使用anti mash v 6 . 1 . 0对DTB中的次级代谢物生物合成基因簇(bgc)进行了搜索，重点是大于5 kb的重叠群(参考。52)

摘要：

（1）Drought stress impacts plant–microbe interactions, reshaping microbial community composition and biogeochemical cycling, thereby reducing crop productivity and threatening food security. 干旱胁迫影响植物与微生物的相互作用，通过重塑微生物群落结构和生物地球化学循环，从而降低作物生产力并威胁粮食安全。

（2）However, the specific microbial responses and roles of plant-derived metabolites remain underexplored. 然而，特定微生物的响应机制及植物源代谢物的作用尚未得到充分探索。

（3）Here we reveal that drought stress shifts the composition of wheat-associated microbiota across the phyllosphere, rhizosphere and root endosphere by favouring Actinobacteria and Ascomycota while depleting Proteobacteria and Basidiomycota. 本研究发现，干旱胁迫通过富集放线菌门和子囊菌门、同时消耗变形菌门和担子菌门，改变了小麦叶际、根际和根内三大生态位的微生物组成。

（4）Targeted single-cell sorting and sequencing identified 21 active drought-tolerant bacteria (DTB) enriched in genes related to plant fitness and nutrient cycling. 通过靶向单细胞分选与测序技术，我们鉴定出21种活跃的耐旱细菌（DTB），这些细菌富含与植物适应性和养分循环相关的功能基因。

（5）These DTB showed significant positive correlations with drought-enriched plant phytochemicals such as jasmonic acid and pipecolic acid. 耐旱细菌与干旱富集的植物化学物质（如茉莉酸和哌啶酸）呈显著正相关。

（6）Moreover, the inoculation of synthetic community including four identified drought-tolerant taxa significantly stimulates the wheat growth under drought stress. 进一步研究发现，由四种已鉴定的耐旱类群组成的合成群落可显著促进干旱胁迫下的小麦生长。

（7）A global exploration confirmed the widespread distribution of DTB, underscoring their promising potential to enhance crop resilience. This study provides new insights into drought-induced microbiome shifts and highlights microbial candidates for improving crop resilience in a changing climate.全球尺度分析证实了耐旱细菌的广泛分布，凸显其提升作物抗逆性的巨大潜力。该研究为解析干旱诱导的微生物组变化提供了新见解，并为提升气候变化背景下作物韧性指明了微生物候选资源。

（8）Result

（9）The impact of drought on wheat metabolome 干旱对小麦代谢组的影响

Fig. 1: Responses of wheat physiologies to drought. 小麦生理对干旱的响应

a, Image showing the experimental design and the impact of soil moisture on wheat growth. 图a：实验设计示意图及土壤湿度对小麦生长的影响。

b, PLS-DA with Pareto scaling illustrating the effects of drought stress on the overall metabolome patterns in leaf and root tissues. Statistical significance was assessed using permutational multivariate analysis of variance (PERMANOVA, two-sided, 200 permutations), and model robustness was evaluated with R2Y and Q2 values. The colour represents the drought treatment, and the shape indicates the wheat growth stage. 图b：采用偏最小二乘判别分析与帕累托标度法展示干旱胁迫对叶片和根组织整体代谢组模式的影响。统计显著性通过置换多元方差分析（双尾检验，200次置换）评估，模型稳健性通过R2Y和Q2值衡量。颜色代表干旱处理程度，形状表示小麦生长阶段。

c,d, KEGG enrichment analysis revealing the most significantly enriched KEGG pathways of upregulated metabolites in wheat leaf (c) and root (d) tissues under severe drought. Statistical significance was determined by one-way ANOVA (two-sided, P < 0.05), and post-hoc tests corrected using the Benjamini–Hochberg method. 图c-d：KEGG富集分析显示重度干旱下小麦叶片（c）和根组织（d）中上调代谢物的显著富集通路。统计显著性通过单因素方差分析（双尾检验，P＜0.05）判定，事后检验采用Benjamini-Hochberg法校正。

e, Classification of upregulated phytochemical compounds in leaf tissues under severe drought. The inset bar chart shows the impact of drought on jasmonic acid levels in leaf tissues. Data are presented as mean ± standard error (s.e.) across biological replicates (n = 8 samples). Statistical significance was determined by one-way ANOVA (two-sided), with multiple comparisons corrected using the Bonferroni method.图e：重度干旱条件下叶片组织中上调植物化学物质的分类。插图为干旱对叶片茉莉酸水平影响的柱状图。数据以生物重复样本均值±标准误表示（n=8），统计显著性经单因素方差分析（双尾检验）及Bonferroni多重比较校正确定。

 f, Classification of upregulated phytochemical compounds in root tissues under severe drought. The inset bar chart shows the impact of drought on pipecolic acid levels in root tissues, with data presented as mean ± s.e. across biological replicates (n = 8 samples). Statistical significance was determined by one-way ANOVA (two-sided), followed by Bonferroni correction for multiple comparisons.图f：重度干旱下根组织中上调植物化学物质的分类。插图为干旱对根组织哌啶酸水平影响的柱状图。数据以生物重复样本均值±标准误表示（n=8），统计显著性经单因素方差分析（双尾检验）及Bonferroni多重比较校正判定。

（10） Drought shifts the wheat microbial diversity and community compositions 干旱改变了小麦微生物多样性和群落组成

Fig. 2: The effect of drought on wheat microbiome. 图2:干旱对小麦微生物的影响。 

a,b, Stacked bar plots illustrating the composition of bacterial (a) and fungal (b) communities in the phyllosphere, rhizosphere and root endosphere of wheat under control, moderate and severe drought treatments at the tillering and heading stages. c–e, Heatmaps showing the Spearman correlations (two-sided tests) between upregulated phytochemicals and the abundant bacteria and fungi phyla in the phyllosphere (c), rhizosphere (d) and root endosphere (e). The leaf metabolome data were used for the phyllosphere, and the root metabolome data were used for both the rhizosphere and root endosphere. P values were adjusted for multiple comparisons using the false discovery rate method. Significance levels: *P < 0.05, **P < 0.01, ***P < 0.001. a、b、堆积条形图说明了在分蘖期和抽穗期处于对照、中度和重度干旱处理下的小麦叶层、根际和根内层中细菌(a)和真菌(b)群落的组成。c–e，显示上调的植物化学物质与叶层(c)、根际(d)和根内层(e)中丰富的细菌和真菌门之间的Spearman相关性(双边测试)的热图。叶代谢组数据用于叶层，根代谢组数据用于根际和根内层。使用假发现率方法对多重比较的p值进行了调整。显著性水平:*P < 0.05，**P < 0.01，***P < 0.001。 

（11） Raman-based phenotypic profiling of drought tolerance in the rhizosphere 基于拉曼光谱的根际耐旱表型分析

Raman-based phenotypic profiling of drought tolerance in the rhizosphere. 根际耐旱性的拉曼表型分析【科学家想找出那些在干旱中不仅活着，而且还在“努力干活”（代谢活跃）的细菌。】

a, Box plots showing the C–D ratio distributions in bacterial populations under control, moderate and severe drought at the tillering and heading stages. Boxes represent the interquartile range (IQR; 25th–75th percentiles), centre lines denote medians, whiskers extend to 1.5× IQR, and white circles indicate means. The red dashed line (1.264) indicates the detection threshold, defined as the mean + 3 s.d. of unlabelled cells. Top bar chart: percentage of DTB across treatments (mean ± s.e., n = 3 individual experiments). Bottom bar chart: drought-tolerance levels of bacterial populations (mean ± s.e., n > 10 cells). Significance determined by one-way ANOVA (two-sided), and multiple comparisons corrected using the Bonferroni method. b, Active DTB with normalized C–D ratios >1.6 (top 2%, red dots) were selected for sorting (left). Representative single-cell Raman spectra of highly active bacteria (red line) and less active or dormant bacteria (green line) (right). c, The composition and relative abundance of rhizosphere bacterial communities before sorting and the recovered taxa of highly active DTB after sorting at the heading stage. a，箱线图显示了在分蘖期和抽穗期处于控制、中度和重度干旱下的细菌种群中的C–D比率分布。方框代表四分位数范围(IQR；第25–75个百分点)，中线表示中间值，须延伸至1.5× IQR，白色圆圈表示平均值。红色虚线(1.264)表示检测阈值，定义为未标记细胞的平均值+ 3标准差。顶部条形图:各处理的DTB百分比(平均值±标准差，n = 3个单独实验)。底部条形图:细菌群体的耐旱水平(平均标准偏差，n > 10个细胞)。通过单向ANOVA(双侧)确定显著性，并使用Bonferroni方法校正多重比较。【三个箱子： 分别代表了在“正常”、“中度干旱”和“重度干旱”环境下，所有细菌的C-D比率（活跃度）分布。箱子越靠上，说明整体细菌团队越活跃。】

b，选择归一化C–D比值> 1.6的活性DTB(顶部2%，红点)进行分类(左)。高活性细菌(红线)和低活性或休眠细菌(绿线)的代表性单细胞拉曼光谱(右图)。【左图： 科学家设定了一个很高的录取分数线（C-D比率 > 1.6，相当于 top 2%），把所有成绩在这个线以上的“超级抗旱特种兵”（用红点表示）一个个地筛选出来。右图： 这张图展示了“超级特种兵”（红线）和“普通士兵或休眠士兵”（绿线）的 “体能特征图谱”（拉曼光谱）。可以看到，红线的多个峰值都更高，这就像是特种兵更强壮的心跳、更发达的肌肉信号，从科学上证明了它们确实极度活跃。】c、整理前根际细菌群落的组成和相对丰度，以及抽穗期整理后高活性DTB的恢复类群。 

Genomic functions of drought-tolerant bacteria in the rhizosphere. 根际耐旱细菌的基因组功能

a, Metagenome-assembled genomes showing the distribution of stress response gene on DTB. a，宏基因组-显示DTB【drought-tolerant bacteria】胁迫反应基因分布的组装基因组。【它展示了这些细菌的基因组里，富含哪些与“胁迫响应”相关的基因。】

b, Maximum-likelihood phylogenetic tree shows the phylogenetic distributions of carbon (C), nitrogen (N), phosphorus (P) and sulfur (S) cycling-related genes in bacterial genome from the rhizosphere bacteria, including both DTB (labelled as ‘sorting’) and non-DTB (labelled as ‘sequencing’). 

b，最大似然系统发育树显示了根际细菌的细菌基因组中碳(C)、氮(N)、磷(P)和硫(S)循环相关基因的系统发育分布，包括DTB(标记为“排序”)和非DTB(标记为“测序”)。【它比较了“抗旱特种兵”（排序组）和“普通细菌”（测序组）在碳、氮、磷、硫等关键营养元素循环方面的基因能力。】

c, Pearson correlations between the abundances of upregulated phytochemicals in leaf and root tissues for severe drought tolerance and the proportions of drought-tolerant cells in rhizosphere samples. 严重耐旱的叶和根组织中上调的植物化学物质的丰度与根际样品中耐旱细胞的比例之间的Pearson相关性。

Fig. 5: Effects of DTB on wheat growth under drought stress. 图5:干旱胁迫下DTB对小麦生长的影响。

a, Bar charts illustrating the effects of individual bacterial strains (Pantoea endophytica, Pantoea vagans, Brevibacterium sediminis and Acinetobacter seifertii) and their SynCom on wheat plant height, maximum root length, maximum leaf length and maximum leaf widths under 30% PEG stress. Data are presented as mean ± s.e. from biological replicates (n = 5 wheat plants). Statistical analyses were performed using a Kruskal–Wallis test with post-hoc Bonferroni adjustment for multiple comparisons. b, Images showing the influence of DTB SynCom on wheat fitness under severe drought in soil. c, Bar chart depicting the effect of DTB SynCom on aboveground fresh biomass of wheat under severe drought in soil. SynCom, wheat inoculated with the DTB synthetic community; control, wheat without DTB inoculation. Each treatment was replicated four pots, with each replicate consisting of three wheat plants. Data are presented as mean ± s.e. from biological replicates (n = 4 independent experiments). Statistical analyses were performed using a Kruskal–Wallis test with post-hoc Bonferroni adjustment for multiple comparisons. a，柱状图说明了单个细菌菌株(内生泛菌、迷走神经泛菌、沉降短杆菌和塞弗氏不动杆菌)及其SynCom在30% PEG胁迫下对小麦植株高度、最大根长、最大叶长和最大叶宽的影响。数据以生物重复(n = 5株小麦植株)的平均值±标准差表示。使用Kruskal–Wallis检验进行统计分析，并对多重比较进行事后Bonferroni调整。图片显示了在土壤严重干旱条件下，DTB SynCom对小麦适应性的影响。c，柱状图描述了土壤严重干旱条件下DTB SynCom对小麦地上鲜生物量的影响。接种了DTB合成群落的小麦；对照，没有DTB接种的小麦。每个处理重复四个盆，每个重复由三个小麦植株组成。数据以生物重复(n = 4次独立实验)的平均值±标准差表示。使用Kruskal–Wallis检验进行统计分析，并对多重比较进行事后Bonferroni调整。 

（12） Global patterns of DTB DTB的全球模式

a, The map of sorted DTB at a global scale.球范围内的分类DTB地图。【它展示了从世界不同地区的农田和自然环境中，科学家们都成功地找到了这些耐旱细菌（DTB）】

 b, The relationships between the abundance and Shannon index of DTB and standardized precipitation evapotranspiration index on a global scale. Solid black lines represent OLS linear regressions, with shaded areas indicating 95% confidence intervals. Two-sided significance tests were applied. b . DTB的丰度和香农指数与全球范围内标准化降水蒸散指数的关系。黑色实线代表OLS线性回归，阴影区域表示95%的置信区间。应用双侧显著性检验。【它分析了这些细菌的数量（丰度）和多样性（香农指数）与一个国际通用的“干旱指数”（SPEI）之间的关系。】

c, Bubble plot showing the number of stress response genes in DTB. c，显示DTB压力反应基因数量的气泡图。

d, Secondary metabolite BGCs in DTB as predicted by antiSMASH. The sequence similarities between query and reference BGCs in the MIBiG database are shown in brackets. Basemap in a from Natural Earth 根据antiSMASH的预测，DTB的次级代谢物BGCs。在MIBiG数据库中查询和参考bgc之间的序列相似性显示在括号中。来自自然地球的底图 【次级代谢产物基因簇】

Proposed model illustrating a potential bacterial solution for enhancing drought resilience in wheat. 提出了一个模型，说明了一种增强小麦抗旱能力的潜在细菌解决方案

Drought reshapes the plant microbiome, creating distinct metabolic profiles and increasing the abundance of drought-tolerant bacteria that enhance key functions such as carbon, nitrogen, phosphorus and sulfur cycling. These bacteria carry functional genes involved in osmolyte production and plant growth hormone synthesis, thereby improving plant fitness. 干旱重塑了植物微生物群，创造了独特的代谢模式，增加了耐旱细菌的数量，增强了碳、氮、磷和硫循环等关键功能。这些细菌携带与渗透液生产和植物生长激素合成有关的功能基因，从而提高植物的适应性

 resilience 弹性