Simultaneous allergic traits in dogs and their owners are associated with living environment, lifestyle and microbial exposures

The study investigates whether shared living environments, lifestyles, and microbial exposures similarly influence allergic traits in cohabiting dogs and their owners. Prior work shows higher allergy prevalence in urban environments for both species and links environmental microbial exposures to allergy risk. Dogs typically manifest allergies with skin and gut symptoms while humans often show airway symptoms. Protective factors such as contact with animals, larger family size, and non-urban environments are thought to increase exposure to beneficial microbes. Previous findings indicate that cohabiting dogs and owners tend to share health status regarding allergic traits and partially share skin microbiota. The authors hypothesize that living environment and lifestyle similarly affect allergy risk in dogs and humans and assess whether microbiota act as a common factor by studying 168 dog–owner pairs for skin and gut microbiota, environmental/lifestyle variables, and allergic traits.

Background studies indicate urban living predisposes to allergy in both humans and dogs, potentially due to reduced exposure to environmental microbes. Farming-related dust has been associated with protection against asthma. Allergic dogs show altered skin microbiota; humans with asthma, sensitization, or atopic dermatitis show altered skin and/or gut microbiota. Cohabiting humans and dogs can share components of their skin microbiota. Prior surveys suggested concurrent allergic traits in dog–owner pairs, but the role of microbiota as a shared factor had not been tested in detail.

Study population comprised 168 dog–owner pairs recruited via breeders of Finnish Lapphunds and Labrador Retrievers with litters born in 2012–2013. Participants included 39 mother dogs and owners, and 130 adult puppies with 128 owners. Sampling took place in autumn 2014 in southern Finland. Skin swabs were collected from dogs (inner foreleg at carpus) and owners (volar forearm) using saline/Tween-moistened swabs. Whole blood from dogs and owners was collected, processed to serum, and frozen. Owners completed questionnaires on lifestyle, environment, and allergic symptoms (validated questionnaires: dogs via a veterinary dermatology instrument; humans via ISAAC). Fecal samples were collected by owners in November 2014 and frozen at home, then transported on dry ice; gut samples were missing for 11 dogs and 6 owners. A subset of owners (n_dogs=96) tracked dog exercise for one week in spring/summer 2015 using passive trackers. DNA extraction and 16S rRNA gene sequencing targeted the V1–V3 region on Illumina MiSeq. Skin DNA extraction followed prior protocols; gut DNA used QIAamp DNA Stool Mini Kit without modifications. Humans were tested for sensitization to common aeroallergens (birch, timothy, dog, cat, Dermatophagoides pteronyssinus) using Phadiatop; sensitization defined as sum of specific IgE > 0.99 kU/L. Dog sera IgE were measured against multiple environmental allergens, but canine allergy status was based on validated owner-reported symptom severity due to poor correlation of canine IgE with symptoms. Sequence processing included trimming, OTU clustering/annotation, removal of non-bacterial sequences (chloroplasts, mitochondria, unknown domain) and very rare OTUs (<10 sequences). Contaminant OTUs were identified using PCR and DNA extraction kit controls and removed based on known contaminant taxa or patterns indicating disproportionate influence in low-depth samples. Samples with <8000 sequences (n=8) were removed. To control library size effects across species and body sites, cumulative sum scaling normalization (metagenomeSeq) was applied across all samples. Raw sequence data deposited under PRJNA434794 (dog skin), PRJNA476220 (dog gut), PRJNA668051 (human skin), PRJNA668266 (human gut). Metadata available on request. Environmental variables: Land cover in home and exercise environments was reduced via PCA; lifestyle and dog owner–reported dog symptoms were reduced via PCoA using Gower distance. First axes were used, with positive values indicating more rural land-use; lifestyle represented a gradient from urban (negative) to rural (positive). Variables were then categorized into rural vs urban by median split; combinations across pairs yielded 35% rural/rural, 10% rural/urban, 20% urban/rural, and 35% urban/urban for environment/lifestyle. Dog allergic status was dichotomized based on PCoA axis cut-off (0.5); humans were sensitized if summed aeroallergen IgE ≥ 1 kU/L. Associations with environment/lifestyle used logistic regression (glm, binomial). Due to collinearity between living environment and lifestyle (dogs r=0.48; humans r=0.49), lifestyle was excluded from adjusted analyses. Microbiota analyses used phyloseq in R. Alpha diversity (Shannon) and beta diversity (Bray–Curtis) were computed; ordinations included PCoA and db-RDA (capscale). PERMANOVA (adonis) tested effects of variables on dissimilarity matrices. Random Forest regression predicted land cover and lifestyle axes from microbiota to estimate explained variation (OOB error reduction). SourceTracker estimated source contributions among microbiotas. Differential abundance testing employed Zero-inflated Log-Normal mixture models (metagenomeSeq fitFeatureModel) for allergic vs healthy and rural vs urban strata; p<0.05 considered significant. Analyses used R 3.5.2. Ethics: Canine sampling approved by Animal Ethics Committee of the State Provincial Office of Southern Finland (ESAVI/6054/04.10.03/2012). Human sampling approved by Coordinating Ethics Committee, Helsinki and Uusimaa Hospital District (188/13/03/00/14). Informed consent obtained for both dog and owner.

- Prevalence and sensitization: 31% of owners were sensitized to aeroallergens (mean summed specific IgE 15.30 kU/L; range 1.02–78.94); ~10% sensitized to dog allergen. About 20% of dogs had owner-reported allergic symptoms. In humans, sensitization associated with asthma (p=0.04), eczema (p=0.02), atopic dermatitis (p=0.04), and wheeze (p=0.003); rhinitis borderline (p=0.07). In dogs, IgE sensitization did not correlate with symptom severity (r=0.06, p=0.46). - Environment/lifestyle and allergy: Urban living environment increased risk of owner-reported allergy in dogs (p=0.003); humans showed borderline association (p=0.054). Urban-type lifestyle increased allergy risk in dogs (p=0.007); humans borderline (p=0.093). Sensitivity analysis excluding owners with allergic skin symptoms still showed dog allergy associated with urban living (p=0.042). - Concurrent allergic traits: Dogs and owners tended to be allergic/healthy concurrently. An allergic dog was more likely to have a sensitized owner (p=0.002), and a sensitized owner was more likely to have an allergic dog (p=0.002). After adjustment for age, sex, and living environment, the association remained significant for humans (p=0.009). In dogs, after adjustments (breed, sex, mother dog allergy, living environment), having a sensitized owner showed a trend (p=0.11). Additional dog risk factors: Labrador Retriever vs Finnish Lapphund (p=0.042), allergic mother (p<0.001), urban living (p=0.013). - Pair-level environment/lifestyle: Allergic dog–owner pairs were more urban in living environment (ANOVA, p=0.010) and lifestyle (p=0.028) than healthy pairs; exercise environment not associated. - Microbiota overview: Identified 52,589 OTUs from 43 million 16S sequences. Microbiotas primarily clustered by body site, secondarily by species. Dog and human gut microbiotas were distinct; skin microbiotas clustered closer between species. Fusobacteria abundant in dog gut, rare in human gut; Proteobacteria more abundant on dog skin. Dog skin had highest richness and Shannon diversity (p<0.0001); inter-individual variation was higher among humans than dogs for both skin (p=1.53e-07) and gut (p<2e-16). - Sharing of microbiota: SourceTracker indicated partial sharing of skin microbiota between dogs and owners; gut microbiotas largely independent. Skin microbiota of cohabiting pairs was more similar than random dog–human pairs (p<0.001). More universal OTUs in skin (30 OTUs in 90% of samples) than gut (9 OTUs in >60% of samples); Propionibacterium present in all but one skin sample. - Environment/lifestyle effects on microbiota: Skin microbiota structure differed by species, living environment (rural vs urban), and lifestyle (PERMANOVA); gut differed mainly by species. Random Forest: living environment predicted from dog skin better than human skin (24.27% vs 11.57% OOB variation); minimal prediction from gut (dog 3.82%, human 2.22%). Lifestyle prediction similar from skin (dog 17.71%, human 17.33%); only dog gut predicted lifestyle (15.66%) vs human gut (1%). Urban environment/lifestyle homogenized skin microbiota in owners (p=0.002 and p=0.02, respectively); opposite trend for gut (p=0.01). - Sharing modulated by lifestyle: Urban lifestyle increased proportion of human skin bacteria in dog skin (p=0.044). Rural lifestyle increased cross-sharing in gut: dog bacteria in human gut (p=0.028) and human bacteria in dog gut (p=0.009). - Differential taxa and allergy: No genera differed directly between allergic vs healthy overall. Several genera differed between rural vs urban environment/lifestyle; some also differed between allergic vs healthy within species. In dogs: urban-associated Rhodopseudomonas enriched in allergic dogs (p=0.007); rural-lifestyle associated Actinoplanes enriched in healthy dogs (p=0.02); Arthrobacter more common in healthy dogs, borderline (p=0.069). Summed rural-associated genera were marginally lower in allergic than healthy dogs (p=0.059). In humans: Sporosarcina (rural environment/lifestyle-associated) more common in healthy than sensitized (p=0.034). SourceTracker trends: higher dog-skin-related bacteria on human skin tended to associate with lower human sensitization (p=0.104); allergic dogs tended to have higher human-skin-related bacteria on their skin (p=0.13). A small but significantly higher proportion of human skin-related microbes was observed in the human gut of allergic vs healthy humans (median 0.43% vs 0.36%, p=0.003).

Cohabiting dogs and owners were more likely to exhibit allergic traits concurrently, with urban living environments and urban-type lifestyles associated with increased risk in both species. Skin microbiota composition was influenced by environment and lifestyle, and skin communities were partially shared within dog–owner pairs, whereas gut microbiotas remained distinct. Despite similar environmental associations, specific bacterial taxa linked to allergy differed between species, likely reflecting fundamental differences in skin habitat and immune function between dogs and humans. Urban environments homogenized skin microbiota in both species, supporting a role for environmental microbial exposure in allergy risk, potentially through reduced diversity or altered composition of skin-associated microbes. The findings suggest that microbial exposures contribute to allergy risk, based on: (1) associations between skin microbiota and environment/lifestyle; (2) associations between allergic traits and environment/lifestyle; (3) urban-driven homogenization of skin microbiota; (4) parallel clustering of skin microbiota and increasing allergy risk along rural–urban gradients in both species. Dogs, with highly diverse skin microbiota, may carry environmental microbes indoors; enrichment of dog-related microbes on human skin of healthy individuals hints at potential protective roles, though environmental microbes transported by dogs may be more relevant than resident dog skin taxa. Allergic dogs showed more human-skin-related microbes on their skin, consistent with lower environmental exposure and more indoor/urban influence. While gut microbiota were not shared, their role in allergy remains important, and unmeasured factors (diet, medication) could confound gut analyses. Overall, shared environmental factors beyond microbes (e.g., air pollution) may also contribute to concurrent allergic traits.

This study demonstrates that cohabiting dogs and owners tend to share allergic status, with higher prevalence in urban environments and among urban-type lifestyles. Skin microbiota, partially shared between dogs and owners, is shaped by environment and lifestyle and shows urban-driven homogenization in both species. However, allergy-associated taxa differed between species, indicating that while shared risk factors exist, species-specific microbial responses or other environmental influences likely mediate allergy risk. The work supports using pet dogs as real-life models for environment-linked non-communicable diseases. Future research should: (1) identify causal microbial exposures and mechanisms across species; (2) disentangle microbial from non-microbial environmental factors (e.g., air pollution); (3) incorporate longitudinal designs with detailed dietary/medication data; and (4) expand sample sizes to improve power for detecting shared microbial signals linked to allergy.

Key limitations include a relatively small number of allergic individuals, reducing statistical power; reliance on owner-reported symptoms for canine allergy classification (although a validated instrument was used) and poor correspondence of canine IgE with symptoms; potential reporting bias by allergic owners (sensitivity analysis did not support this); possible participation bias (mitigated by a 51% participation rate); cross-sectional design; limited data on diet, medications, and other factors that shape gut microbiota; and potential unmeasured environmental confounders such as air pollution.