Preserving the past, protecting the future: targeted DNA enrichment detects zoonotic pathogens in museum specimens

By Dr. Liji Thomas, MDReviewed by Lily Ramsey, LLMJun 28 2023

Global health is at risk from multiple zoonotic diseases. Pathogens harbored by small mammals cause more than 60 illnesses.

A recent paper in Emerging Infectious Diseases describes a novel method to search for zoonotic pathogens from animal tissues archived in museums.

Study: Prospecting for Zoonotic Pathogens by Using Targeted DNA Enrichment. Image Credit: ElizavetaGalitckaia/Shutterstock.com

Introduction

Zoonoses that have caused the most cases of illness and death worldwide include the Rabies virus, Yersinia pestis (Y. pestis, author of the bubonic plague that killed anywhere between a third to a fifth of Europeans), Ebola virus, and Tuberculosis, as well as the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) which caused the coronavirus disease 2019 (COVID-19) pandemic.

Animal specimens are stored in museums of natural history. Previously, dried skin or bone was archived, or formaldehyde-fixed animal tissue.

Nowadays, liquid nitrogen or mechanical freezers are used to preserve soft tissue samples, aiding in retrieving their DNA and RNA.

These can be used to go back and detect zoonotic pathogens, thus filling gaps in the natural history of many such killers. Such studies have shown, in one case, the presence of Sin Nombre virus, a hantavirus, in wild rodent tissue specimens collected from the US Southwest.

These specimens were archived almost two decades before the first human cases.

A team of scientists set out to develop a method to identify the presence of such pathogens using museum collections of tissues. The way they chose to focus on targeted sequence capture using probe enrichment.

Those methods are designed to enrich small amounts of DNA target from a background of contaminating DNA.”

In many cases, probes are designed to pick up a few pathogens out of many. Such panels may be commercially available or customized.

However, many of the available probe sets can target certain pathogens that may not infect other hosts.

The researchers produced a panel of biotinylated probes in the current study with approximately 40,000 RNA probes of 80 base pairs (bp) each. These were directed towards over 30 zoonoses caused by different pathogen groups, including bacteria, fungi, helminths, and protozoa.

The identified pathogens could be known or related to known microbes, exploiting the probe’s ability to lock in on sequences with 10% difference in genetic sequence from the exact match.

The scientists used a version of the ultraconserved element (UCE) targeted sequencing technique so that pathogen DNA could be preferentially enriched.

The biotinylated baits bind to conserved regions of the genome, that is, regions that remain the same between different types of pathogens. They are then exposed to a collection of DNA that could contain pathogen DNA.

This enables pathogen DNA fragments to be separated from the rest by binding to the probe. These fragments are enriched and then sequenced.

The final yield contains thousands of copies of the same genetic loci with point mutations that can help trace the phylogenetic pathway or the population affected by the pathogen over time.

Forty-nine loci were targeted for each pathogen group, though a few had 98 loci. 

The researchers generated samples of tissue controls in the laboratory. These consisted of host-pathogen mixtures, enriched or not enriched for pathogen DNA. These were used to test the ability of the probes to pick out target loci at higher frequencies from the pathogen DNA.

What did the study show?

The results of the probe study indicated its ability to enrich targeted loci from pathogen DNA by ~3,000- to 7,000-fold, depending on the pathogen. The greatest increase in mean coverage per locus was for Mycobacterium. The lowest was for Plasmodium loci.

In the control samples, almost 43% of all reads were from control pathogens (Mycobacterium, Plasmodium, and Schistosoma) in the 1% enriched control sample. In contrast, the 1% unenriched control samples yielded only 0.03% reads from target loci.

The scientists assembled 0-23 target loci for each pathogen group in the control samples, varying in size from 109 to 1,991 bp. This helped them produce a phylogenetic tree for each group where they could capture two or more loci.

The resulting loci were sister groups to the reference genomes.

When tested on museum-archived tissues, they generated an average of 17.1 million reads per sample, almost 650 million reads for all 38 samples. Over 4% of reads corresponded to loci in the database, belonging to 93 genera, but 78 were at less than 1,000 reads per sample. Many of these low-frequency hits may be attributable to non-significant signals.

Bartonella and Plasmodium were found in 36 of the 38 samples, but while 18 pieces had over a thousand reads per sample, an equal number had less than 12. Conversely, Plasmodium reads at a median of ~160 per sample vs. ~550 for Bartonella.

There were 15 genera with 1,000 or more reads, but 13 did not yield two or more target loci for a pathogen. There were 16 Bartonella reads and one sample containing Paraburkholderia and Ralstonia, respectively. The first had 3-20 loci, the second three, and the third eight loci.

Mitochondrial genomes were also compared to the reads from each sample, though they were relatively few. Despite this, models with 50 or more mitochondrial reads matched the museum identifications. For the other samples, over 85% of reads matched the correct genus.

What are the implications?

The bait panel developed and used in this study enriched pathogen DNA in the control sample from 0.03% to ~43%. The enriched control sample yielded information that helped trace the phylogenetic pathway for three species.

This finding indicates that the probe set is able to detect these pathogens from even a few thousand genome copies per sample.”

The targeted loci are substantially enriched even in the presence of large amounts of host DNA in the sample. Enrichment is an initial step toward screening for pathogens in archived tissues.

In museum-archived samples, the process generated phylogenies for samples containing three species, Bartonella, Ralstonia, and Paraburkholderia. Multiple clades of Bartonella were discovered, fitting real-world host-pathogen interactions. The other two species are less sure of their identification.

The bait panel is redundant, and the targeted loci cover a span of divergences. This makes it possible to capture genetic loci from taxa other than those used to design the probe set, such as multiple species of Bartonella.

While these results are promising, work remains to refine the method. Host DNA is still present in large amounts despite the enrichment of targeted loci. Secondly, it is relatively expensive.

Again, those missed species cannot be identified despite targeting pathogens over a large spectrum, making mock controls essential.

Finally, there is a tradeoff between sensitivity and efficiency of capture. This limits its ability to identify pathogens if there are 3,000 or fewer genome copies per sample.

Although further effort is required to resolve these issues, we believe that enrichment of pathogen DNA from museum tissue samples is a viable tool worth further development.”