[biorxiv.org] Microbial communities living on and in rocks operate at the microscale, where interactions with minerals fundamentally shape community structure and function. Yet the relationship between micron scale mineralogical configurations and microbial distributions remains poorly understood.
We tested the hypothesis that microbial biomass spatially correlates with areas of heightened mineralogical heterogeneity by applying Raman microspectroscopy to rock samples from three geologically distinct substrates: authigenic carbonates from a marine methane seep, volcanic basalt from Iceland, and polymetallic nodules from the abyssal seafloor.
Using spectral decomposition and multiple complementary metrics of compositional heterogeneity, we evaluated intra-pixel and inter-pixel heterogeneity patterns in relation to biomass distribution.
Our analyses reveal three patterns across all sample types. 1) When spectra are deconstructed into their constituent components, biomass zones are disproportionately dominated by the biomass spectral component compared with primary mineral components in zones of different minerals. 2) Biomass spectra have more homogeneous compositional profiles than mineral spectra. 3) Biomass is surrounded by more heterogeneous microhabitats than mineral pixels.
These findings demonstrate that biomass exerts a distinctive and consistent influence on Raman spectral signatures, both within and between pixels, in ways that mineral components do not. Our results establish generalizable principles linking microscale mineralogical properties to microbial biogeography; these properties could be used as a potential biosignature and may provide a standardized workflow applicable to diverse rock systems and astrobiological exploration strategies.
Pixel identity and decomposition as determined by Non-negative Matrix Factorization (NMF) analysis. A), C), E) Stacked bar plots showing the average component composition of pixels assigned to each identity class in the three samples: Seep Carbonate (SC), Polymetallic Nodule (PN), Volcanic Basalt (VB), respectively. Pixels are grouped according to their dominant component identity. In each stacked bar, the bottom segment represents the dominant (primary) component used to assign pixel identity, whereas the upper segments represent the average contributions of the remaining (minor) components. The y-axis indicates the mean component weight across all pixels within each identity class, and the x-axis lists the identified components for each sample. B), D), F) Representative FOVs for samples SC, PN and VB, respectively. Each pixel is colored according to its assigned identity, defined as the component with the highest NMF weight, using the same color scheme as the corresponding stacked bar plots. — biorxiv.org
Compositional Heterogeneity Structures Microbial Microhabitats across Distinct Mineral Substrates, biorxiv.org
Astrobiology,
