Compound biodegradability tests with natural microbial communities form an important keystone in the ecological assessment of chemicals. However, biodegradability tests are frequently limited by a singular focus either on the chemical and potential transformation products or on the individual microbial species degrading the compound. Here, we investigated a methodology to simultaneously analyze community compositional changes and biomass growth on dosed test compound from flow cytometry (FCM) data coupled to machine-learned cell type recognition. We quantified the growth of freshwater microbial communities on a range of carbon dosages of three readily biodegradable reference compounds, phenol, 1-octanol, and benzoate, in comparison to three fragrances, methyl jasmonate, myrcene, and musk xylene (as a nonbiodegradable control). Compound mass balances with between 0.1 to 10 mg C · liter⁻¹ phenol or 1-octanol, inferred from cell numbers, parent compound analysis, and CO₂ evolution, as well as use of ¹⁴C-labeled compounds, showed between 6 and 25% mg C · mg C⁻¹ substrate incorporation into biomass within 2 to 4 days and 25 to 45% released as CO₂. In contrast, similar dosage of methyl jasmonate and myrcene supported slower (4 to 10 days) and less (2.6 to 6.6% mg C · mg C⁻¹ with 4.9 to 22% CO₂) community growth. Community compositions inferred from machine-learned cell type recognition and 16S rRNA amplicon sequencing showed substrate- and concentration-dependent changes, with visible enrichment of microbial subgroups already at 0.1 mg C · liter⁻¹ phenol and 1-octanol. In general, community compositions were similar at the start and after the stationary phase of the microbial growth, except at the highest used substrate concentrations of 100 to 1,000 mg C · liter⁻¹. Flow cytometry cell counting coupled to deconvolution of communities into subgroups is thus suitable to infer biodegradability of organic chemicals, permitting biomass balances and near-real-time assessment of relevant subgroup changes.

中文翻译：

---

使用流式细胞术评估微生物群落生长中化合物的生物降解性

天然微生物群落的复合生物降解性测试是化学品生态评估中的重要基石。但是，生物降解性测试经常受到单一关注的局限，要么只关注化学和潜在的转化产物，要么只关注降解该化合物的单个微生物。在这里，我们研究了一种方法，该方法可以同时从流式细胞仪（FCM）数据和机器学习的细胞类型识别中分析剂量测试化合物上的群落组成变化和生物量增长。与三种香料，茉莉酸甲酯，月桂烯和麝香二甲苯（作为不可生物降解的对照）相比，我们通过三种易生物降解的参考化合物苯酚，1-辛醇和苯甲酸酯的碳剂量范围来量化淡水微生物群落的生长。 。从细胞数，母体化合物分析和CO ₂演变以及使用^14种C标记的化合物推断出^-1酚或1-辛醇，表明6％至25％mg C·mg C ^-1底物掺入生物质在2至4天之内释放出25至45％的CO ₂。相反，相同剂量的茉莉酸甲酯和月桂烯支持较慢（4至10天）和较少（2.6至6.6％mg C·mg C ^-1，含4.9至22％CO ₂）的群落生长。从机器学习的细胞类型识别和16S rRNA扩增子测序推断出的群落组成显示出底物和浓度依赖性变化，微生物亚组的可见富集量已经达到0.1 mg C·升^-1酚和1-辛醇。通常，除了在使用的最高底物浓度为100至1,000 mg C·L ^-1之外，微生物生长的开始阶段和静止阶段之后的群落组成相似。流式细胞仪细胞计数与社区解卷积耦合成亚组，因此适合于推断有机化学物质的生物降解性，允许生物量平衡和相关亚组变化的近实时评估。