Do commercial dog extracts cross-react with Felis domesticus allergen 1

Abstract

Background:

Half of U.S. households own a dog despite 10% of individuals being sensitized to dog. Assessment and treatment options for dog allergy include the use of commercial dog extracts which have inconsistent performance, making diagnosing and managing dog allergy challenging. Contamination of dog extracts with other allergens has previously been reported.

Objective:

We sought to determine whether commercial dog extracts contain other aeroallergens.

Methods:

An extract purity and quantification study on acetone precipitated dog hair and dander extract (AP dog) was performed, 6 aeroallergens; Alternaria (Alt a 1), Ragweed (Amb a 1), German Cockroach (Bla g 2), Dust Mite (Der p t), Cat (Fel d 1), and Rye Grass (Lol p 1). Following, conventional dog hair and dander extract (CV dog) and the new ultrafiltered dog hair and dander extract (UF dog) were also assessed based on the initial results of AP dog. SDS-PAGE was performed on all three dog extracts to compare allergen content. Lastly, serology results and aeroallergen immunotherapy prescriptions were compared.

Results:

The ELISA trays with Alt a 1, Amb a 1, Bla g 2, Der p 1, and Lol p 1 antibodies did not capture AP dog, while the ELISA tray with Fel d 1 antibody captured AP dog, CV dog, and UF dog. SDS-PAGE results of the 3 dog extracts did not reveal a band at the molecular weight for Fel d 1.

Conculsion:

Contamination of commercial dog extracts was not found. However, our findings are supportive of commercial dog extracts containing a Fel d 1-like dog allergen that is cross-reactive to Fel d 1. Cross-reactivity between commercial dog extracts and Fel d 1 could be responsible for double positivity to cat and dog in serology. Additional studies are needed to better illustrate this Fel d 1-like dog allergen.

Per the 2023–2024 American Pet Products Association, the majority of American households have pets (86.9 million of 125.7 million households [70%]), 74.9% of which own a dog, which makes the dog the most popular pet in the United States.^1,2 Not only do man’s best friends serve as our domestic companions, dogs also fulfill important jobs by working closely alongside us. There are roughly 500,000 service dogs assisting U.S. citizens with disabilities.³ Across the U.S. federal government, >5500 canines are used as working dogs, 1800 of them within the Department of Defense (DOD).⁴ Approximately 10% of U.S. individuals are sensitized to dog and 99.9% of households demonstrate detectable levels of Canis familiaris 1 (Can f 1), one of the major dog allergens, despite the data that only half of U.S. households have dogs.^5,6

Treatment of dog allergy includes avoidance, medication management, and/or allergen immunotherapy. Often, complete avoidance or rehoming a dog is not an acceptable option for patients or their families. Much research, including the frequency of canine bathing, shortening of hair length, breeding of “hypoallergenic dogs,” and using high-efficiency particulate air-filtered vacuum cleaners and air filters, has demonstrated the ability to decrease Can f 1 allergen levels by varying amounts but did not eliminate them.⁷ In addition, because dog allergen is present in homes where they do not reside, complete avoidance is not possible.⁶ Medications such as nasal steroids and long-acting antihistamines may ameliorate symptoms but at times may not provide sufficient relief. Consequently, the allergist may recommend subcutaneous aeroallergen immunotherapy (AIT) as a treatment strategy.

Dog extracts are typically prepared by using dog hair and dander. These extracts are not standardized; the allergen content is variable.⁸ In the United States, there are three commonly used dog extracts, which differ with regard to their concentrations and extraction method: (1) conventional dog hair and dander extract (CV dog) has a 1:10 weight per volume (w/v) or 1:20 w/v and is extracted from the source material without extra processing^9,10; (2) acetone-precipitated dog hair and dander extract (AP dog) has a 1:100 w/v and is manufactured by adding acetone to an aqueous extract, which concentrates the proteins^9,10; (3) the new ultrafiltered dog hair and dander extract (UF dog) has a 1:650 w/v and is processed from an ultrafiltered aqueous extract, which results in a comparable Can f 1 dosage to AP dog.^9,11 UF dog eliminates the occurrence of precipitation in AP dog extracts, replacing the use of acetone with ultrafiltration and diafiltration.^10,11 AP dog and UF dog have a significantly higher content of Can f 1, which allows smaller volumes to be used for AIT.¹⁰ Can f 1 concentrations for CV dog (1:10 w/v) have been reported to be 2–10 µg/mL, whereas AP dog and UF dog concentrations are reported to be ∼171 µg/mL and 183 µg/mL, respectively.¹⁰ However, these concentrations are expected to vary among lots because they are not standardized.¹⁰ The recommended, most effective dose for dog immunotherapy is 15 µg of Can f 1.^12,13 Therefore, an adequate maintenance dose for dog immunotherapy is 0.5 mL of AP dog or UF dog diluted with a 1:5 ratio and 3 mL of CV dog (1:10 w/v) undiluted.¹⁰

The performance of commercial dog extracts has been inconsistent, which makes diagnosing and treating dog allergy challenging.⁷ This could be due to many reasons. First, there are several significant dog allergens (Can f 1–7) and there is not just one dominant major allergen as with domestic cat’s major allergen, Felis domesticus allergen 1 (Fel d 1).⁷ Cox et al.¹³ defines a major allergen as “an antigen that binds to the IgE [immunoglobulin E] sera from 50% or more of a clinically allergic group of patients.” In patients with dog allergy, Can f 1 sensitization has been found in 40–70%, whereas Can f 5 has been found in 70%.^7,14 Because commercial dog extracts are not standardized, potency can vary significantly among extracts, up to 1000-fold, and may lack important dog allergens.⁸ Lastly, contamination of dog extracts with other allergens has been reported. In 1996, Der p 1 and Der p 2 (house-dust mite allergens of Dermatophagoides pteronyssinus) were detected in all commercial extracts.¹⁵

OBJECTIVE

To our knowledge, assessment for contamination of dog extracts had not been recently reassessed. We sought to determine whether commercial dog extracts contain other aeroallergens besides dog allergens. This study was funded by the Clinical Investigation Program at Joint Base San Antonio, Texas.

METHODS

In assessing for contamination, we performed an extract purity and quantification study on AP dog, our facility’s standard extract for skin-prick testing and AIT. We tested AP dog for six other aeroallergens (Alternaria, ragweed, German cockroach, dust mite, cat, rye grass) by using enzyme-linked immunosorbent assay (ELISA) trays with those specific capture antibodies (Alt a 1, Amb a 1, Bla g 2, Der p 1, Fel d 1, Lol p 1, respectively). These aeroallergens were chosen to include a wide aeroallergen variety (mold, weed, insect, cat, grass) and commercial ELISA plates with their specific capture antibody were available from Indoor Biotechnologies, Inc. (Charlottesville, VA). The lower limit of detection (LLOD) of these ELISA plates with capture antibody was in ng/mL, 1000-fold more sensitive than µg/mL because we were assessing for contamination. In addition, each allergen extract was also tested against its major allergen as a positive control measure of binding the targeted allergen (e.g., Alternaria to Alt a 1, ragweed to Amb a 1, etc.). Standards, controls, and extract dilutions were prepared and plated in duplicate on appropriate ELISA trays that contained capture antibody. ELISA trays with capture antibody and extracts were incubated, then followed by the addition of an enzyme-conjugated detection antibody, a color development substrate, and, finally, a stop reagent. Parameters were entered into the plate reader (Opsys MRTM; Dynex Technologies, Chantilly, VA) for each different assay plate. Sample and control values were generated off the standard curve. After initial results were obtained, the same method was repeated on CV dog and UF dog but testing for only cat with an ELISA tray with Fel d 1 capture antibody. In addition, sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was performed on all three dog extracts (AP, CV, UF) to review potential allergen content.

Clinical data were assessed to compare the rates of sensitization to dog and/or cat and AIT prescription habits for dog and/or cat from December 2022 to December 2023. De-identified patient serum-specific immunoglobulin E (sIgE) serology results were reviewed from our facility, which services the entire DOD. In addition, de-identified data from U.S. Army Centralized Allergen Extract Laboratory, which mixes AIT prescriptions for the entire DOD, were examined. The percentages of dog and cat, dog only, and cat only from sIgE serology results and AIT prescription results were compared via 3 × 2 χ² test.

RESULTS

First, a Can f 1 concentration of 23.7 µg/mL was detected, well below the previously reported concentration of 171 µg/mL.¹⁰ Each ELISA tray with capture antibody appropriately seized its specific antigen (e.g., Alt a 1 antibody captured Alternaria extract, Amb a 1 antibody captured ragweed, etc.). ELISA trays with Alt a 1 (LLOD 0.1 ng/mL), Amb a 1 (LLOD 0.78 ng/mL), Bla g 2 (LLOD 0.39 ng/mL), Der p 1 (LLOD 0.78 ng/mL), and Lol p 1 (LLOD 7.81 ng/mL) antibodies did not capture AP dog. It does not seem that AP dog is contaminated with Alternaria, ragweed, roach, dust mite, and rye grass. However, the ELISA tray with Fel d 1 antibody did capture AP dog, detecting a concentration of 1.34 µg/mL (Table 1).

Table 1.

Extract purity and quantification results of AP dog, CV dog, and UF dog

AP dog = Acetone precipitated dog hair and dander extract; CV dog = conventional dog hair and dander extract; UF dog = ultrafiltered dog hair and dander extract; LLOD = lower limit of detection; w/v = weight per volume.

To determine if these results would be reproducible in other dog extract formulations, we tested CV dog and UF dog for cat (Table 1). A Can f 1 concentration of 1.6 µg/mL was detected in CV dog and 42.9 µg/mL was detected in UF dog. The UF dog concentration was also well below the previously reported concentration of 183 µg/mL.¹⁰ The ELISA tray with Fel d 1 antibody did capture CV dog and UF dog, which detected a concentration of 0.023 µg/mL and 7.08 µg/mL, respectively. The concentration of Fel d 1 was five times higher in UF dog than in AP dog.

Given that Fel d 1 capture antibody bound three different dog extracts, this raised the suspicion that there could be contamination or, possibly, cross-reactive allergens between them. SDS-PAGE results of AP dog, CV dog, and UF dog (Fig. 1) were accomplished to determine if a band between 35 and 38 kDa, the molecular weight of Fel d 1, was present. No band was present at this molecular weight, which suggests that the three dog extracts were not contaminated with but rather cross-reacted to Fel d 1. A large band was noted at ∼20 kDa.

Figure 1.

AP dog, CV dog, and UF dog SDS-PAGE. SDS-PAGE = sodium dodecyl-sulfate polyacrylamide gel electrophoresis; AP = acetone precipitated dog hair and dander extract; CV = conventional dog hair and dander extract; UP = ultrafiltered dog hair and dander extract; MW = molecular weight.

To examine further the correlation between dog and cat allergy, we reviewed de-identified patient sIgE serology results from our facility from December 2022 to December 2023. We discovered 2047 samples positive to dog and/or cat (sIgE value of ≥0.35 kU/L) (Fig. 2A), 1327 (65%) were sensitized to both dog and cat, whereas 438 (21%) were sensitized only to dog and 282 (14%) were sensitized only to cat. In seeking to examine how sensitization data compared with AIT prescription patterns, we pulled de-identified data from the U.S. Army Centralized Allergen Extract Laboratory. From December 2022 to December 2023, 9075 patients were prescribed dog and/or cat extract in their AIT prescription (Fig. 2B), 4908 (54%) were sensitized to both dog and cat, whereas 2045 (23%) were sensitized only to dog and 2122 (23%) were sensitized only to cat. A 3 × 2 χ² test that compared the percentages between the sIgE results and AIT results found a p value of < 0.00001, which is considered highly significant.

Figure 2.

(A) Positive sIgE (≥0.35 kU/L) to dog and/or cat and (B) AIT prescriptions with dog and/or cat. sIgE = Specific immunoglobulin E; AIT = aeroallergen immunotherapy.

DISCUSSION

On a review of the literature, many dog allergens are cross-reactive to various cat allergens due to similar protein structures within lipocalins and albumins.¹⁴ The known dog allergens and their descriptions, including potentially cross-reactive cat allergens, are reviewed in Table 2. The lipocalins, Can f 1, 2, and 6, have demonstrated cross-reactivity to specific cat lipocalins.^14,16–20 Demonstrated in sera IgE cross-reactivity studies from patients sensitized to 13 rFel d 7 and rCan f 1, Can f 1 and Fel d 7 have compatible structures and identical epitopes, which resulted in cross-reactivity between the two allergens.¹⁶ Can f 2 shares a <22% sequence homology with Fel d 4 but has demonstrated up to 58% cross-reactivity to Fel d 4 in sera from five patients sensitized to Can f 2.¹⁷ In addition, several studies have supported cross-reactivity between Can f 6 and Fel d 4.^18–20 The dog albumin, Can f 3, has been reported to be cross-reactive with cat albumin, Fel d 2.²¹ Lastly, Can f 7 was demonstrated to cross-react with Cat-NPC2 in inhibition IgE studies.²²

Table 2.

Significant dog allergens in patients with dog allergy and their reported cross-sensitization to cat^*

^*Adapted from Refs. 7, 8, 14.

^#Adapted from Ref. 26.

Fel d 1, which is the capture antibody used in our study that reacted to three different prepared dog allergen extracts, is not reported as being cross-reactive with the important dog allergens Can f 1–7.²³ Fel d 1 is a secretoglobulin found in domestic cat saliva and dander, with a molecular weight of 38 kDa.^24,25 Cross-reactivity of Fel d 1 is found within the feline family to include big (wild or nondomestic) cats.²⁴ However, in 2007, Reininger et al.²⁶ reported the first demonstration of a “Fel d 1–like allergen” present in commercial dog dander extracts that is cross-reactive with Fel d 1. Quantitative serum sIgE inhibition studies when using rFel d 1 on the sera from 36 patients with cat allergy were examined and demonstrated that 25% of the patients had >50% inhibition of IgE reactivity to dog allergens with the rFel d1.^23,26 In addition, 20% of patients sensitized to cat and/or dog produced sIgE to the Fel d 1–like dog allergen and none reported symptoms to dog.²⁶

We appreciated no SDS-PAGE bands between 35 and 38 kDa, which suggests that our results were not due to contamination with Fel d 1. These results could be explained by cross-reactivity to a Fel d 1–like dog allergen. Many of the major dog allergens are clustered around 20 kDa, where this Fel d 1–like dog allergen has previously been reported.²⁶ This suggests that AP dog, CV dog, and UF dog are not contaminated with Fel d 1 but contain an allergen cross-reactive with Fel d1. This could also explain the double positivity appreciated in our population.

Our sIgE analysis demonstrates a 3:2 ratio of unique dog versus unique cat sensitization, and two thirds of all cat and/or dog sensitizations are positive to both. Therefore, patients who are sensitized to cat are more likely to also be sensitized to both when compared with patients who are sensitized to dog. This could indicate the possibility of a cat-like allergen in dog but not dog-like allergens in cat. AIT prescription data demonstrate a 1:1 ratio of unique dog versus unique cat prescriptions, and just more than one half of all cat and/or dog prescriptions contain both. The differences in percentages between sIgE results and AIT results were highly significant. Based on these data, patients are more likely to have positive sIgE results to dog but are more frequently prescribed AIT for cat. This discordance could be due to variety of reasons: (1) skin-prick testing results to dog and/or cat may differ than in sIgE testing; (2) patients may not be reporting symptoms to dog, despite positive sIgE results; and (3) cat allergens may be more allergenic than dog allergens, etc.

Limitations

There were several limitations to our study. First, by using capture antibody to assess for extract purity and quantification, we were unable to specify with certainty whether Fel d 1 captured dog extracts due to contamination versus cross-reactivity. In addition, only one sample from each extract preparation was tested and there can be significant variability among nonstandardized extract vials. With regard to our clinical data, sIgE results and AIT prescriptions were obtained from two different clinical programs, which do not include the exact same patient population. The data were de-identified, therefore, chart reviews to assess clinical history were not obtained and we were unable to determine whether patients sensitized to dog or prescribed dog AIT were symptomatic to dogs. Patients sensitized but not symptomatic to dog may be more likely to be sensitized and/or symptomatic to Fel d 1.

CONCLUSION

We did not find any evidence of contamination of AP dog with Alternaria, ragweed, roach, dust mite, and rye grass. Additional studies are needed to assess whether the new commercial dog extract, UF dog, contains other aeroallergens besides dog allergens. Our findings are supportive of commercial dog extracts, to include AP dog, CV dog, and UF dog that contain a Fel d 1–like dog allergen that is cross-reactive to Fel d 1. Cross-reactivity between commercial dog extracts and Fel d 1 could be responsible for frequent double positivity found in cat and dog in serology and potentially with skin-prick testing. We recommend that the clinician obtain a thorough history because sensitization does not always equal allergy and dog AIT may not be recommended. Future research could include reevaluating immunoblot inhibition studies to confirm whether commercial dog extracts are contaminated with Fel d 1 or contain a Fel d 1–like dog allergen. Lastly, future research should correlate patient symptomatology with sIgE levels, skin-prick test results, and AIT prescription habits.