The microbial diversity of moonmilk, a hydrated calcium carbonate speleothem, was evaluated from two Italian caves to provide context for the food web of highly-specialized troglobitic beetles, Cansiliella spp. (Leptodirinae), with distinctive carbon and nitrogen isotope values indicative of a novel food source. The moonmilk and associated percolating waters had low to no extractable chlorophyll, with an average organic C:N ratio of 9, indicating limited allochthonous input and a significant contribution from microbial biomass. The biomass from moonmilk was estimated to be ~104 micro- and meiofaunal individuals per m2 and ~107 microbial cells/ml. Betaproteobacteria dominated the 16S rRNA gene sequences retrieved from the moonmilk from both caves. The distribution of other proteobacterial classes and phyla in the moonmilk were statistically similar to each other, even though the two caves are geographically separated from each other. Comparing the moonmilk gene sequences to sequences from previously described environmental clones or cultured strains revealed the uniqueness of the moonmilk habitat, as ~15% of all of the moonmilk sequences were more closely related to each other than to sequences retrieved from any other habitat. However, comparative analyses confirmed that as much as ~34% of the clones sequences were also closely related to environmental clones and cultured strains derived from soil and freshwater habitats, which is likely due to the fact that the putative inoculation source for the moonmilk bacterial communities is from overlying soil and percolating fluids from the surface. Prior to our studies of Cansiliella spp., moonmilk has not been considered a food source for cave animals. Our findings provide unique insight into moonmilk microbial diversity that could reveal the underpinnings of the moonmilk carbon and nitrogen cycle that influences the isotopic composition and the morphological adaptations of the troglobitic beetles associated with the moonmilk.



Supplemental Fig 1 Engel et al.pdf (105 kB)
supplemental figure S1

Supplemental Fig 2 Engel et al.pdf (26 kB)
supplemental figure S2

Supplemental Fig 3 Engel et al.pdf (34 kB)
supplemental figure S3

Supplemental Fig 4 Engel et al.pdf (26 kB)
supplemental figure S4