Metagenomics and Composition of Prokaryotes on Lewis Glacier and its Foreland in Mount Kenya, Nyeri County

Abstract

The Lewis glacier in Mt. Kenya, Kenya is one of the best documented tropical glaciers globally, with about 80 years of frequent observations of the length, area and volume change. These studies have, however, focused on the measurements of the surface mass balance and the recession of Lewis Glacier. Currently the Lewis Glacier is less than 0.105 ± 0.001 × 106 m², indicating a massive loss of the glacier mass and its endogenous biodiversity. The main objective of this study was to isolate and characterize glacial prokaryotic populations and determine their community structures on Lewis glacier using the next generation sequencing techniques by targeting the 16S rDNA V3-V4 variable gene regions. Samples collected included cryoconite materials, foreland soil and foreland plant rhizosphere from 12-year-old, 23-year-old and 42-year-old foreland chronosequence. R2A and BG-11 media were used to inoculate melt water into a hundred folds. Inoculants were incubated at 4°C, 10°C, and 25°C. Total community DNA was extracted from samples using phenol-chloroform. The 16S rDNA gene variable region (V3 – V4) of the extracted DNA library construction was performed as per the Illumina sequencing protocol. Sequences were analysed using QIIME2 pipeline. The isolates were partially sequenced by Sanger and the sequences aligned to their closest relatives from the NCBI gene bank. The QIIME2 output were subsequently introduced into R programming language (3.5.1) for the analyses of species abundance, composition, diversity and associations between the bacterial community structure and the environmental constraints. Packages tidyverse, superheat, vegan, dplyr, indicspecies, indval were used for diversity, composition, ordinations indicator species analyses. Packages ggrepel, ggcorrplot, were used for the multivariate analyses. The concentration of dissolved organic matter and nutrients were measured as eluted supernatant of the 10% UWA medium using a Spectroquant test kit and a spectral photometer. All samples were filtered through a polycarbonate membrane. The nutrient eluted supernatant of the 10% UWA agar medium were collected after shaking (110 r.p.m.) for 24 h. Albedo reflectance was measured in five replicates using albedo-meter from the field sites. Phylogenetic analyses of the 16S rDNA of axenic isolates revealed three major phyla Firmicutes, Proteobacteria, and Actinobacteria to dominate the snowpack. Cyanobacteria was the most prominent phylum (35 % - 37 %) on the glacier. The distribution of these genera were significantly different between the upper and the lower glacier (r = 0.094, p = 0.027) and between the foreland soil and rhizosphere (r = 0.27, p = 0.001). Other phyla, Proteobacteria, Actinobacteria and Bacteroidetes were also prominent in the glacier and at the foreland chrono-sequence. The community structure and distribution of these phyla were significantly different in the sample types (r = 0.84, p = 0.001), age of the last glacier terminal (r = 0.707, p = 0.001), sample sites (r = 0.816, p = 0.001) and altitude (r = 0.63, p = 0.011). There were a total of 27,367 ASVs across the glacier and foreland ecosystem. 185 ASVs were shared across the glacier, rhizosphere and soil samples, 423 ASVs were shared across the 12-year-old, 23-year-old and 42-year-old foreland rhizosphere while 455 ASVs were shared across the 12-year-old, 23-year-old and 42-year-old foreland soil. Foreland plant rhizosphere had the highest microbial community structure that increased down the foreland age. Primary foreland had the highest number of ASVs while glacier sites had the lowest microbial community structures, but high ASV numbers. The 3 % distance analyses of the of the 16S rDNA revealed that the rarefaction sequencing depth is far from exhaustive sampling in a number of samples, including the largest samples. Ordination analyses showed samples to be clustered together according to the sample types and sampling sites. This was also supported by the correspondence analyses of proximities (CAP) between the environment and bacterial community matrices. The analyses of correlation between the environmental variables and bacterial community structure indicated cryoconite material and mineral to be strongly correlated (r = 1). The Euclidean distance and Bray-Curtis dissimilarity were maximally correlated and revealed altitude and Chlorophyll.a to strongly influence the bacterial community structure (r = 0.318). Mantel test further confirmed the relationship between the environmental factors to bacterial community structure (r = 0.1763, p = 0.004). Like any other glacier in the world, the Lewis Glacier is dominated with the phylum Cyanobacteria which is less distributed across the foreland chrono-sequence. Other phyla such as Proteobacteria, Actinobacteria and Bacteroidetes also inhabits the glacier ecosystem, but highly abundant in the foreland. The current results are important as reference database tropical cryophilic prokaryotes. The study has also led to the conservation of low temperature adapted microorganisms in the local laboratory for subsequent analyses and screening for biotechnological applications.