Cryogenic electron microscopy (cryoEM) has become an indispensable technique for determining the structures of isolated biological macromolecules and imaging biomolecular structures within cells. While thin films of frozen-hydrated macromolecule suspensions can be directly prepared and imaged, these molecules must be extracted from their cellular context using existing sample preparation methods. On the other hand, cellular specimens preserve the native context, but the region of interest must be sectioned or subjected to focused ion-beam milling at cryogenic temperatures to achieve the necessary thickness for transmission electron imaging. Currently, no method exists for targeted cytoplasmic extraction of a subcellular volume from individual cells for direct vitrification and cryoEM imaging. In this study, a method is presented that addresses this gap. A system has developed that utilizes a force-sensitive microfluidic cantilever pipette to aspirate and dispense sample volumes as small as 204 fL onto conventional cryoEM sample supports, maintained at the dew point. This is followed by automatic vitrification for cryoEM imaging. Coupled with a fluorescence microscope, this setup allows for the extraction of a targeted subcellular volume from an individual cell and subsequent dispensing of the aspirated content onto an electron microscopy grid. A proof-of-concept is demonstrated by dispensing femtolitre volumes of the standard cryoEM single-particle sample, tobacco mosaic virus, and performing a subcellular biopsy from a single HeLa cell. Additionally, the challenges of manipulating such small volumes for cryoEM sample preparation are discussed, highlighting the current limitations of this approach and potential solutions for overcoming them.

We introduce a two-channel microfluidic atomic force microscopy (AFM) cantilever that combines the nanomechanical sensing functionality of an AFM cantilever with the ability to manipulate fluids of picolitres or smaller volumes through nanoscale apertures near the cantilever tip. Each channel is connected to a separate fluid reservoir, which can be independently controlled by pressure. Various systematic experiments with fluorescent liquids were done by either injecting the liquids from the on-chip reservoir or aspirating directly through the nanoscale apertures at the tip. A flow rate analysis of volume dosing, aspiration and concentration dosing inside the liquid medium was performed. To understand the fluid behaviour, an analytical model based on the hydrodynamic resistance, as well as numerical flow simulations of single and multi-phase conditions were performed and compared. By applying pressures between -500 mbar and 500 mbar to the reservoirs of the probe with respect to the ambient pressure, flow rates ranging from 10 fl s-1 to 83 pl s-1 were obtained inside the channels of the cantilever as predicted by the analytical model. The smallest dosing flow rate through the apertures was 720 fl s-1, which was obtained with a 10 mbar pressure on one reservoir and ambient pressure on the other. The solute concentration in the outflow could be tuned to values between 0% and 100% by pure convection and to values between 17.5% and 90% in combination with diffusion. The results prove that this new probe enables handling multiple fluids with the scope to inject different concentrations of analytes inside a single living cell and also perform regular AFM functionalities.

Microfluidic atomic force microscopy (AFM) cantilever probes have all the functionalities of a standard AFM cantilever along with fluid pipetting. They have a channel inside the cantilever and an aperture at the tip. Such probes are useful for precise fluid manipulation at a desired location, for example near or inside cells. They are typically made by complex microfabrication process steps, resulting in expensive probes. Here, we used two different 3D additive manufacturing techniques, stereolithography and two-photon polymerization, to directly print ready-to-use microfluidic AFM cantilever probes. This approach has considerably reduced the fabrication time and increased the design freedom. One of the probes, 564 μm long, 30 μm wide, 30 μm high, with a 25 μm diameter channel and 2.5 μm wall thickness had a spring constant of 3.7 N m^-1 and the polymer fabrication material had an elastic modulus of 4.2 GPa. Using these 3D printed probes, AFM imaging of a surface, puncturing of the cell membrane, and aspiration at the single cell level have been demonstrated.