Leaf-cutter ant fungus gardens and and organic polymer characterization. (A) Foraging leaves, (B) fungus garden chambers, (C) Colony expansion, (D) vertical layers comparison of polymers (E) Cellulose breakdown, and (F) lignin breakdown.
Overview []
The leaf-cutter ant, Atta colombica, is a leaf forager social insect capable of coexisting in colonies of millions of workers. Mature colonies forage hundreds of kilograms of leaves each year in order to creat massive fungus gardens, which are special chambers were mutualistic a fungus is farmed as the colony's primary source of food.
The primary function of the fungus gardens is to convert plant biomass into nutrients for the ants. Previously, biomass breakdown was thought to be exclusively mediated by fungi, but recent studies report that the garden fungus is uncapable of braking down cellulose on its own, and suggest that a bacteria biota is reponsable of breaking down complex carbohydrates such as cellulose and lignin.
Addressing this hypothesis Suen et al. (2010) conducted a metagenomic and sugar-reduction profile analyses showing that fungus garden microbiome is composed of a diverse community of bacteria with high plant biomass-degrading capacity similar to that of the bovine rumen. This suggests evolutionary convergence of completely unrelated bacteria performing similar carbohydrate-degrading tasks.
Methods[]
A total of 25 fungus gardens from 5 healthy colonies (5 gardens each) of the leaf-cutter ant were collected. Five independent samples from three gardens per colony were sample for both top and bottom of the gardens. This material was tested for crystalline cellulose and hemicellulose (matrix polysaccharide) content. Bioassays on pure cultures were performed for bacteria Klebsiella variicola At-22 and Pantoea sp. At-9b to determine their capacity to degrade cellulose.
Extracted DNA from upper and lower fungus gardens was amplified with full-length 16S rDNA universal primers. DNA was then sequenced using standard Sanger-based capillary sequencing. A total of 703 unique sequences from the upper layer of the garden and 2,794 unique sequences from the lower layer were generated. The same samples were then pyrsequenced by first PCR amplifying prokaryote-specific V6-V8 regions and then sequenced on a Roche 454 FLX GS Titanium pyrosequencer. Maximun likelihood method was used for all phylogenetic analyses.
Fungus garden (A) Top and (B) bottom layer comparison of 16S microbial phylogenetic analyses.
To identify how the microbial community associated with plant polymer degradation, the study performed a carbohydrate-active enzyme (CAZy) characterization of the garden community metagenome. The analysis identified 69 gene modules across 28 families of glycosyl hydrolases, carbohydrate esterases, and polysaccharide lyases, which accounts for 58% of he seuences predicted to code for enzymes. The CAZy proflie and whole proteome-predicted metagenome were compared with 13 other metagenomes from environments that exhibit biomass degradation including animal guts and soil.
Findings[]
Quantification of plant biomass polymer content from these layers revealed that crystalline cellulose and sugars representing various plant polysaccharides, such as hemicelluloses, decreased in content from garden top to bottom.
Full-length 16S rDNA libraries contained 132 phylotypes (97% sequence identity) from 9 phyla in garden tops, and 197 phylotypes from 8 phyla in garden bottoms. Bacterial diversity comparisons among colonies and vertical layers revealed a number of consistent phylotypes, the majority of which are y-proteobacteria; however 16S libraries also showed that specific phyla may play specialized roles within vertical layers of the garden. Phylogenetic binning by community metagenomics indicated that the fungus garden is dominated by y-proteobacteria (30% of total bacterial sequences), a-proteobacteria (16%), actinobacteria (9%), d-proteobacteria (7%), and B-proteobacteria (7%).
Comparative clustering of fungus garden microbiome metagenome to 13 other bacterial metagenomes from similar environments; (A) Carbohydrate-active enzymes (CAZy) profile clustering; (B) Whole genome orthologous groups profile clustering.
Clustering analysis of the CAZy profiles showed that the fungus garden metagenome groups closest to bovine rumen, both including enzymes capable of degrading amylose, galactan, mannan, maltose, pectin, and xylan. Although similar in their carbohydrate-degrading activity, most bacteria from in the bovine rumen come from the genera Prevotella, Fibrobacter and Ruminococcus, whereas leaf-cutter and fungus gardens primarily contain bacteria from Proteobacteria, suggesting evolutionary convergence.
The two most abundant genera found accross all gardens were Klebsiella and Pantoea, which are simultaneous symbionts of the Agaricales fungus as well as the leaf-cutter ant. The predicted proteomes of two specific species, Klebsiella variicola At-22 and Pantoea sp. At-9b, cotained a number of sequences predicted to code for enzymes known to be involved in plant polymer degradation as well as the fixation of nitrogen in the fungus gardens.
References
Suen, G. et al. 2010. An insect herbivore microbiome with high plant biomass-degrading capacity. PLoS genetics, 6(9), e1001129.