Antibiotic Resistome Biomarkers associated to the Pelagic Sediments of the Gulfs of Kathiawar Peninsula and Arabian Sea

Abstract

Antibiotic resistance has been one of the most persistent global issue. Specifically, marine microbiomes have served as complex reservoirs of antibiotic resistant genes. Molecular advancements have enabled exploration of the uncultured microbial portion from hitherto difficult to sample niches such as deeper oceans. The Gulfs of Kathiawar Peninsula have been known for their unique properties like extreme tidal variations, different sediment textures and physicochemical variations. Pelagic sediment cores across four coordinates each of the Gulf of Kutch, Gulf of Khambhat and an open Arabian Sea were collected, processed for metagenomic sequencing and assessed for antibiotic and metal resistome. The dominant genes were mostly resistant to macrolides, glycopeptides and tetracycline drugs. Studied samples divided into three clusters based on their resistome with carA, macB, bcrA, taeA, srmB, tetA, oleC and sav1866 among the abundant genes. Samples from creek of Gulf of Kutch and mouth of Gulf of Khambhat were most diverse in resistance gene profile. Biomarkers observed include gyrA mutation conferring resistance gene in the Arabian Sea; Proteobacteria species in Gulf of Kutch and Arabian sea; while Aquificae, Acidobacteria and Firmicutes species in the Gulf of Khambhat. Region-wise differentially abundant 23 genes and 3 taxonomic biomarkers were proposed for antibiotic resistance monitoring.

Figure 1

Bacteriocide and metal resistant gene difference as computed by STAMP using the Welch’s t-test representing a minimum variation at a significant level [p (corrected) <0.05] between (a) Khambhat and other regions, (b) Kutch and other regions, (c) TMS and other samples (d) AFM and other samples.

Figure 2

Thermal dendogram of metal resistant genes with samples grouped by similarity (UMPGA clustering) and significant genes (p < 0.05 between the three clusters as obtained based on drug class) ranked by their mean abundance.

Figure 3

(a) Proportion (%) of the antibiotic resistant mechanism of observed ARGs in studied sites (b) Relative abundance of drug classes among the studied samples.

Figure 4

Distribution of top 63 ARGs across the sites (the depicted ARGs were those that were having at least an abundance of >0.5% in one of the nine samples) (% Abundance in Y-axis versus ARG names in X-axis).

Figure 5

(a) Phylum-level difference as obtained based on drug class output as computed by STAMP using the Welch’s t-test representing a minimum variation at a significant level [p (corrected) <0.05] between: Upper - the cluster 1 against rest Middle - cluster 2 against rest, Lower -cluster 3 against rest, (b) Thermal dendogram of genus level abundance of microbial community with samples grouped by similarity (UMPGA clustering) and significant genera (p < 0.05 between the three clusters as obtained based on drug class) were ranked by their mean abundance.

Figure 6

(a) Region specific biomarker ARGs (BM-ARG) as analyzed by LefSe as differentially abundant ARGs by the Kruskal Wallis test (p < 0.05) and LDA score >2.0, (b) Region specific biomarker bacterial communities (BM-BC) Left - Differentially abundant microbial taxa at different levels of classification as analyzed by LefSe using the Kruskal Wallis test (p < 0.1; p < 0.05 marked with *) and LDA score >2.0; Right - Cladogram of the differentially abundant microbial taxa, the root of the cladogram denotes domain, all the taxonomic levels depicted (up to genus) are abbreviated, different colours indicate the region with most abundance of the biomarker and the size of each node represents their relative abundance.