Date Published: February 12, 2016
Publisher: Public Library of Science
Author(s): Ehtesham Mofiz, Torsten Seemann, Melanie Bahlo, Deborah Holt, Bart J. Currie, Katja Fischer, Anthony T. Papenfuss, Joseph M. Vinetz. http://doi.org/10.1371/journal.pntd.0004384
Abstract: The scabies mite, Sarcoptes scabiei, is an obligate parasite of the skin that infects humans and other animal species, causing scabies, a contagious disease characterized by extreme itching. Scabies infections are a major health problem, particularly in remote Indigenous communities in Australia, where co-infection of epidermal scabies lesions by Group A Streptococci or Staphylococcus aureus is thought to be responsible for the high rate of rheumatic heart disease and chronic kidney disease. We collected and separately sequenced mite DNA from several pools of thousands of whole mites from a porcine model of scabies (S. scabiei var. suis) and two human patients (S. scabiei var. hominis) living in different regions of northern Australia. Our sequencing samples the mite and its metagenome, including the mite gut flora and the wound micro-environment. Here, we describe the mitochondrial genome of the scabies mite. We developed a new de novo assembly pipeline based on a bait-and-reassemble strategy, which produced a 14 kilobase mitochondrial genome sequence assembly. We also annotated 35 genes and have compared these to other Acari mites. We identified single nucleotide polymorphisms (SNPs) and used these to infer the presence of six haplogroups in our samples, Remarkably, these fall into two closely-related clades with one clade including both human and pig varieties. This supports earlier findings that only limited genetic differences may separate some human and animal varieties, and raises the possibility of cross-host infections. Finally, we used these mitochondrial haplotypes to show that the genetic diversity of individual infections is typically small with 1–3 distinct haplotypes per infestation.
Partial Text: The scabies mite is an ectoparasitic arachnid that causes an itchy skin infection, known as scabies. Each year around 300 million people worldwide are affected by scabies . Scabies is responsible for a significant disease burden in affected populations through its obligate parasitic lifecycle, which facilitates secondary infections by other pathogens. A severe, but more rare form of scabies, known as crusted scabies, is characterised by hyper-infestation. It generally occurs in immune-compromised individuals , although it can occur in patients with no overt immunological deficiency . Cases of crusted scabies can play a significant role in transmission . The mite also infects more than a hundred species of mammals, creating an animal welfare and economic burden in primary industry [5–7].
Sequencing of the two human and four pig samples (unwashed and 3 washed technical replicates) generated 46 (patient B) to 62 (pig washed 1) million read pairs per sample (S1 Table). As expected, the libraries also contained host and microbial DNA. We estimated the level of contamination using two approaches. PhymmBL was applied to the patient B mite mitochondrial assembly to classify contigs into taxa. This revealed that 44% of contigs were from bacteria or other non-arachnid species (S1A Fig). Species abundance was estimated by re-aligning the reads back to the contigs. This revealed that 6% of the reads were derived from contaminants (S1B Fig).
We de novo assembled the mitochondrial genome of the scabies mite using massively parallel sequencing data from thousands of pooled whole mites obtained from two clinical isolates from different parts of northern Australia and from a laboratory pig model. Our approach was to initially perform a full metagenomic de novo assembly. As expected, the samples were contaminated by bacterial reads presumably from the scabies mite gut and the scabies lesion micro-environment. We then iteratively selected Mt contigs and used these as bait to recruit and assemble genuine mite reads. Our bespoke de novo assembly approach has some similarity to an existing bait-and-assembly method called MITObim . MITObim first directly baits the short reads using a closely related mitochondrial genome, then iteratively maps reads and performs contig extension using MIRA. De novo assembly is also supported, but only for “well behaved” data. In contrast, our approach performs full de novo assembly of the metagenomic mixture, baiting of contigs using the house dust mite genome and filtering of host mitochondrial contigs, then 1–2 iterations of read alignment and de novo assembly using velvet. Our results substantially extend the limited case studies and simulations used to validate MITObim and demonstrate that the general approach of mitochondrial genome bait-and-assembly works on real examples that are highly complex metagenomic mixtures involving both genetic heterogeneity and host/bacterial contamination.