In our recent work, we developed an optofluidic platform that allows a direct link to be made between the phenotypes (functions) and the genotypes (genes) of microbial cells within natural communities. By combining stable isotope probing, optical tweezers, Raman microspectroscopy, and microfluidics, the platform performs automated Raman-based sorting of taxa from within a complex community in terms of their functional properties. In comparison with manual sorting approaches, our method provides high throughput (up to 500 cells per hour) and very high sorting accuracy (98.3 ± 1.7%), and significantly reduces the human labour required. The system provides an efficient manner to untangle the contributions of individual members within environmental and host-associated microbiomes. In this News and Thoughts, we provide an overview of our platform, describe potential applications, suggest ways in which the system could be improved, and discuss future directions in which Raman-based analysis of microbial populations might be developed.

The microbial environment is typically within a fluid and the key processes happen at the microscopic scale where viscosity dominates over inertial forces. Microfluidic tools are thus well suited to study microbial motility because they offer precise control of spatial structures and are ideal for the generation of laminar fluid flows with low Reynolds numbers at microbial lengthscales. These tools have been used in combination with microscopy platforms to visualise and study various microbial taxes. These include establishing concentration and temperature gradients to influence motility via chemotaxis and thermotaxis, or controlling the surrounding microenvironment to influence rheotaxis, magnetotaxis, and phototaxis. Improvements in microfluidic technology have allowed fine separation of cells based on subtle differences in motility traits and have applications in synthetic biology, directed evolution, and applied medical microbiology.

Maintaining the identity of chromatin states requires mechanisms that ensure their structural integrity through the concerted actions of histone modifiers, readers, and erasers. Histone H3K9me and H3K27me are hallmarks of repressed heterochromatin, whereas H3K4me and H3K36me are associated with actively transcribed euchromatin. Paradoxically, several studies have reported that loss of Set2, the methyltransferase responsible for H3K36me, causes de-repression of heterochromatin. Here we show that unconstrained activity of the acetyltransferase complex Mst2C, which antagonizes heterochromatin, is the main cause of the silencing defects observed in Set2-deficient cells. As previously shown, Mst2C is sequestered to actively transcribed chromatin via binding to H3K36me3 that is recognized by the PWWP domain protein Pdp3. We demonstrate that combining deletions of set2⁺ and pdp3⁺ results in an epistatic silencing phenotype. In contrast, deleting mst2⁺, or other members of Mst2C, fully restores silencing in Set2-deficient cells. Suppression of the silencing defect in set2∆ cells is specific for pericentromeres and subtelomeres, which are marked by H3K9me, but is not seen for loci that lack genuine heterochromatin. Mst2 is known to acetylate histone H3K14 redundantly with the HAT Gnc5. Further, it is involved in the acetylation of the non-histone substrate and E3 ubiquitin ligase Brl1, resulting in increased H2B-K119 ubiquitylation at euchromatin. However, we reveal that none of these mechanisms are responsible for the Set2-dependent silencing pathway, implying that Mst2 targets another, unknown substrate critical for heterochromatin silencing. Our findings demonstrate that maintenance of chromatin states requires spatial constraint of opposing chromatin activities.