Flukeprint eDNA Reveals Skin Bacteria of Free-Ranging Transient Killer Whales

Environmental DNA collected from the disturbed patch of water left behind when a killer whale dives — its flukeprint — can reveal bacteria associated with the animal’s skin, according to a new study of West Coast Transient (Bigg’s) killer whales published in Metabarcoding and Metagenomics. The work offers a non-invasive alternative to biopsy sampling for monitoring the skin health of free-ranging cetaceans, including endangered populations that are poor candidates for invasive sampling.

Skin condition is a major indicator of health in killer whales and other cetaceans. Vessel strikes, entanglement in fishing gear, and ocean pollution can damage skin directly, while stress from pollution, reduced prey availability, and disturbance can increase susceptibility to chronic bacterial, viral, and fungal infections. Skin lesions in killer whales have previously been found to contain unclassified poxviruses and papillomaviruses, among other unknown causal agents. Until now, however, characterizing the skin microbiomes of free-living cetaceans required either bacteriological swabs of captive or deceased animals or the collection of skin biopsies, a mildly invasive approach that samples only a small surface area of skin.

The research team, led by scientists studying the mammal-eating West Coast Transient killer whale ecotype, targeted two regions where these predators concentrate: the Monterey Bay region of central California, where the whales prey on migrating gray whale calves in spring and sea lions through summer and fall, and the southern Salish Sea of southern British Columbia, which the whales frequent year-round due to abundant harbor seals. Between January and November 2024, the researchers opportunistically collected flukeprint eDNA samples from individual whales encountered at sea, along with paired lateral photo-identification images to confirm identities, sexes, and ages using published catalogues.

Sample collection followed a strict protocol designed to capture DNA shed from the whale rather than the surrounding ocean. Upon encountering West Coast Transient killer whales, a target individual was approached from a parallel position in a motorized research vessel at an approximate distance of 100 meters until a terminal dive was observed. After the whale dived, the vessel maneuvered immediately beside the flukeprint — the disturbed patch of water produced as the whale dives — and collected seawater 30 seconds to one minute after the terminal dive, taking care not to cross in front of the animal’s last observed direction of travel. For each whale encounter, a seawater control sample was also collected to characterize the local baseline bacterial community. All samples were filtered within three hours of collection through 0.45-micrometer self-preserving filters, and only collected under sea conditions resembling less than 4 on the Beaufort Scale to ensure flukeprints were visible.

In the laboratory, DNA was extracted using a modified Qiagen DNeasy protocol in a dedicated decontaminated eDNA room, with extraction blanks processed alongside each batch to monitor contamination. The V4/V5 region of the 16S rRNA gene was amplified using dual-indexed fusion primers and sequenced on an Illumina NextSeq platform. Bioinformatic processing with the MetaWorks pipeline, including denoising with a minimum cluster size of three to reduce spurious sequences in low-concentration eDNA samples, yielded more than 15 million reads. After filtering out ESVs present in extraction blanks and rarefying to equal depth, 274,749 reads across 2,189 bacterial exact sequence variants (ESVs) were retained for analysis.

The results showed that flukeprint samples were more diverse than seawater controls, both overall and when matched samples per encounter were compared. Flukeprints produced 1,870 ESVs representing 417 bacterial genera, compared with 1,514 ESVs representing 359 genera in seawater controls. Using a paired approach, each flukeprint and its corresponding seawater control shared an average of 52 bacterial genera, while each contained an average of 28 unique genera. In total, 117 bacterial genera were unique to flukeprint samples, compared with 51 unique to seawater controls. Among the genera found exclusively in flukeprints, the most abundant included Chryseobacterium, previously isolated from marine mammal skin; Shewanella, reported from killer whale and bowhead whale skin; Carnobacterium, isolated from killer whale and bowhead whale skin; and Pseudoalteromonas, previously identified in humpback whale and captive killer whale skin microbiomes.

Source-tracking analysis using the FEAST package revealed that flukeprints contained contributions from both known seawater sources and unknown sources, likely originating from the sampled individual whale. For whales sampled in California, unknown sources accounted for 20 to 41 percent of the bacterial signal, while for samples collected in British Columbia, known seawater contributions ranged from 94 to 100 percent, with unknown sources accounting for 0 to 6 percent. Functional pathway prediction using PICRUSt2 inferred 3,729 enzymes and 520 metabolic pathways from bacterial ESVs unique to flukeprints. Oxidoreductases, transferases, and hydrolases — enzyme classes with established roles in oxidative stress protection, adhesion, and degradation of host lipids and proteins — were proportionally higher among pathways likely associated with skin, and 14 of the top 20 metabolic pathways were classified as likely skin-associated.

To link bacterial findings to skin condition, the researchers assessed lateral images of 15 whales with unique flukeprint bacteria, dividing each image into four body regions — dorsal fin, anterior, mid-flank, and peduncle — and measuring skin lesions and scarring using ImageJ software. Gray patches were the most common lesion type observed. Whales sampled in California exhibited marginally higher percentages of skin lesions than whales sampled in British Columbia, though the difference was not statistically significant. Linear models revealed that the number of unique bacterial ESVs detected in flukeprints was explained by a combination of whale age, sampling month, and the percentage of visible skin lesions, while unique bacterial genera were explained by geographic region, whale sex, and sampling month. Notably, age exhibited a significant negative relationship with skin bacterial diversity, with higher flukeprint bacterial ESV diversity observed in younger individuals.

Geographic location emerged as the strongest driver of bacterial diversity overall, explaining 47 percent of the variation in assemblage composition. This regional structuring is consistent with previous skin microbiome studies of killer whales, beluga whales, and humpback whales. The researchers also noted the absence of human-associated bacteria such as Enhydrobacter, Paracoccus, and Sphingomonas — taxa common on human skin and previously detected on captive cetaceans — supporting the lack of contamination during sample collection and processing. Similarly, no key killer whale respiratory bacteria such as Staphylococcus, Bacillus, or Vibrio were detected, suggesting the respiratory microbiome does not contribute substantially to flukeprint bacterial communities.

The study demonstrates that non-invasive flukeprint eDNA sampling can detect bacteria likely derived from the skin of free-ranging killer whales, with variability influenced by a combination of demographic and environmental factors. Because seawater serves as the sampling medium, a higher proportion of shared taxa between flukeprints and controls is expected, and the authors caution that functional predictions from PICRUSt2 represent phylogenetically inferred potential rather than empirically measured pathways. Nevertheless, the approach offers greater skin surface representation than a biopsy, which typically yields only a 5-to-8-millimeter epidermal sample. For endangered populations such as the Southern Resident killer whales, flukeprint eDNA sampling provides an opportunity to assess health status with limited disturbance, and the authors suggest that integrating flukeprint eDNA data with lateral and drone imagery may eventually facilitate machine-learning approaches to detect, characterize, and classify skin lesions in wild cetaceans.

Subject of Research: Killer whale skin microbiome monitoring via flukeprint eDNA sampling

Article Title: Applying flukeprint environmental DNA to detect skin-associated bacteria from free-ranging West Coast Transient killer whales Orcinus orca rectipinnus

Article References: Robinson, C. V., Visona-Kelly, B. C., Laqua, E. J., Sutton, G. J., Talty, C. M., & Judson, M. W. D. (2026). Applying flukeprint environmental DNA to detect skin-associated bacteria from free-ranging West Coast Transient killer whales Orcinus orca rectipinnus. Metabarcoding and Metagenomics, 10, Article e174578. https://doi.org/10.3897/mbmg.10.174578

Image Credits: AI Generated

DOI: 10.3897/mbmg.10.174578

Keywords: environmental DNA, killer whales, skin microbiome, cetacean health, non-invasive sampling, Metabarcoding and Metagenomics, Applying, flukeprint, environmental, detect, skin-associated, bacteria