Microhabitat conditions determine the magnitude and speed of microbial processes but have been challenging to investigate. In this study we used microfluidic devices to determine the effect of the spatial distortion of a pore space on fungal and bacterial growth, interactions, and substrate degradation. The devices contained channels differing in bending angles and order. Sharper angles reduced fungal and bacterial biomass, especially when angles were repeated in the same direction. Substrate degradation was only decreased by angles when fungi and bacteria were grown together. Investigation at the cellular scale suggests that this was caused by fungal habitat modification, which branched in sharp and repeated turns, blocking the dispersal of bacteria and the substrate. Our results demonstrate how the geometry of microstructures can influence microbial activity. This can be transferable to soil pore spaces, where spatial occlusion and microbial feedback on microstructures is thought to explain organic matter stabilization.

Microscopy

Epifluorescence microscopy was used for visualization of P. putida, C. Cinerea, and AMC using a fully motorized Nikon Ti2-E inverted microscope with PFS4 hardware autofocus, full 25 mm field-of-view, CoolLED pE300-White MB illumination connected via a 3 mm liquid light guide (LLG), and a Nikon Qi2 camera with 1x F-mount adapter. The filters used were LED-DAPI-A-2360A Semrock Filter Cube (Ex: 380-405 nm, Em: 413-480 nm), GFP-4050B Semrock Filter Cube (Ex: 444-488 nm, Em: 498-553 nm), mCherry-C Semrock Filter Cube (Ex: 520-585 nm, Em: 600-680 nm). The entire chip images for overall fluorescence quantification were captured using a (MRH00041) CFI Plan Fluor 4X, N.A. 0.13, W.D. 17.1 mm objective, with an exposure time of 20 ms for GFP, 100 ms for DAPI, and 100 ms for mCherry. For high magnification pictures a (MRD31905) CFI Plan Apochromat DM Lambda 100X Oil N.A. 1.45, W.D. 0.13 mm and a (MRD30405) CFI Plan Apochromat DM Lambda 40X, N.A. 0.95, W.D. 0.21 mm objective were used. NIS-Elements software was used for coordination of the multipoint imaging. Pictures were taken for every chip for 14 days. The days selected for analysis were the ones of maximum biomass, namely day 2 for Pseudomonas putida biomass and its AMC consumption, and day 6 for Coprinopsis cinerea biomass, its AMC consumption, and the AMC consumption of the fungal+bacterial conditions.

Image Analysis

The fluorescence intensity was quantified using ImageJ 1.52n [49]. Background was subtracted using the ImageJ rolling ball algorithm [50] using 7 pixels as radius of rolling ball for images taken with 4X objective. The rolling ball radius was given based on the size of the biggest fluorescent object, which was the width of a channel. After the subtraction, the mean florescence intensity per pixel was quantified inside each channel using the ROI manager tool. The rectangular ROIs were of the same size and covered every individual channel of the experiment.

To attain a deeper understanding of the fluorescence distribution along the channels, fluorescent profiles were obtained for every type of channel. For this purpose, the segmented line tool and the measure tool were used to cover manually the entire length of the channels.

Each file corresponds to a chip in an specific timepoint. The chip and time point are indicated in the file name, for instance: the file with the name "3t5" corresponds to the chip 3 at the time point 5 (days). The file contains the three fluorescent channels used: GFP, dTomato, and Methylcoumarin.