The prevalence of allergic skin diseases in horses, including atopic dermatitis and insect-bite hypersensitivity (IBH), ranges from 3% to 71% depending on the geographical region.1,2 Pruritus can considerably affect quality of life in affected horses and leads to self-trauma.1 Available treatments currently rely on glucocorticoids, allergen-specific immunotherapy, and fly control.3–7 Skin inflammation and self-trauma lead to secondary infections and repeated courses of antibiotics, promoting antibiotic resistance.8
Several mediators of pruritus have been described in humans and animals.9–11 In horses, knowledge of the mediators of pruritus is limited. Histamine does not appear to be an important mediator of pruritus in horses as urticaria is frequently nonpruritic in horses and antihistamines are frequently ineffective in reducing itch.3–5
In mice, humans, and dogs, the role of IL-31 in allergic inflammation and pruritus is well established, and IL-31 receptor α (IL-31RA) has been identified on sensory neurons, keratinocytes, and immune cells in these species.11–14 In humans and mice, binding of IL-31 to its receptor leads to phosphorylation of Janus kinase 1 and 2 (JAK1/2), which, in turn, triggers phosphorylation of signal transducer and activator of transcription 3 (STAT3) and, to a lesser extent, STAT1 and STAT5.12 Treatments targeting IL-31 are used in dogs and humans.14–18 Transcription of both IL-31 and IL-31RA has been identified in equine leukocytes; however, information on IL-31 in horses is limited.19 Oclacitinib, a selective JAK1/2 inhibitor used in dogs, has been tried in horses with variable success.20 Increased IL-31 transcription was identified in lesional skin of IBH horses compared with nonlesional skin from the same horse or normal skin of healthy controls.21 A significant upregulation of IL-31 transcription after allergen stimulation with Culicoides nubeculosus extract in peripheral blood mononuclear cells (PBMCs) was also reported in 19 IBH horses but not in PBMCs from 3 healthy horses.21 This information led to the development of anti–IL-31 vaccination, which was shown to decrease chronic pruritus of unknown origin in horses.22
We aimed to further characterize the role of IL-31 as a mediator of pruritus in horses. We produced 2 recombinant proteins based on the confirmed mRNA sequence for equine IL-31 (eIL-31) and demonstrated their pruritogenic effects in vivo after ID injection in normal horses. We also demonstrated that activation of the IL-31 receptor on equine adherent monocytes by recombinant eIL-31 resulted in downstream phosphorylation of STAT3 in vitro.
Methods
Animals
The horses used were either privately owned and used with the informed written consent of their owners or were part of the equine herd of the University of Florida. All procedures were approved by the University's IACUC (#201910660 for ID injections; #201709803, #202009803, and #202300000007 for blood samples).
The normal horses were 6 to 10 years old and had no history or clinical signs of skin disease. The allergic horses were actively pruritic and had been diagnosed with a combination of environmental allergies and IBH by a board-certified veterinary dermatologist based on suggestive history, compatible clinical signs, and exclusion of other pruritic skin diseases. For many but not all allergic horses, ID skin tests had been done confirming that they had a positive ID response to various pollens as well as Culicoides allergen. None of the horses had received any medication in the previous month.
Rapid amplification of cDNA ends sequencing of eIL-31 mRNA
Two eIL-31 mRNA and translated protein sequences were available on the US National Center for Biotechnology Information (NCBI) and Ensembl databases (NCBI reference sequence XM_023648068.1; Ensembl EquCab3.0 horse genome transcript ENSECAT00000018354.3).23,24 Both sequences were predicted from the equine genome, had not been previously confirmed, and were not the same. We therefore confirmed the full-length eIL-31 mRNA sequence using rapid amplification of cDNA ends (RACE) sequencing.
To obtain IL-31 mRNA for sequencing, 5 mL blood from an allergic American Quarter Horse mare was collected, the buffy coat was separated by centrifugation (1,000 X g, 15 minutes), and the decanted buffy coat was mixed for 5 to 10 minutes in 10 mL whole-blood lysis buffer (NH4Cl, 155 mM; KHCO3, 10 mM; EDTA-Na2, 0.1 mM, pH 7.1) to lyse erythrocytes, washed twice in Dulbecco PBS (DPBS), and then cultured overnight in 4 mL lymphocyte culture medium (RPMI 1640 medium containing 10% heat-inactivated fetal bovine serum [FBS], 10,000 U/mL penicillin, 10,000 µg/mL streptomycin, and 25 µg/mL amphotericin B) at 37 °C in 5% CO2. The cells were cultured in 2 aliquots of 2 mL, and 1 was also stimulated with phorbol 12-myristate 13-acetate 32.5 nM (20 ng/mL as is commonly used for nonspecific lymphocyte stimulation). Ribonucleic acid was extracted from both adherent and nonadherent cells using a column RNA extraction kit for mammalian cells (RNeasy Mini Kit; Qiagen) with 5 mM tris(2-carboxyethyl)phosphine added to the lysis buffer and freeze-thawing of the lysate at −80 °C for 5 minutes, and the resulting RNA was combined into 1 pooled sample.
For 5′ sequencing, intron-spanning nested reverse primers were designed targeting conserved regions of the IL-31 mRNA sequences predicted for Equus species as described previously.19,25 For 3′ sequencing, a forward primer was similarly designed. Primer sequences are listed in Table 1. Complementary DNA was generated from 100 ng of total RNA using a reverse-transcription kit and random hexamers according to the manufacturer's instructions. A RACE kit (Firstchoice RLM-RACE kit; Invitrogen) and high-fidelity taq polymerase (Platinum SuperFi; Invitrogen) were used to amplify both the 3′ and 5′ ends of the eIL-31 cDNA. Two products were produced by the 5′ RACE, which were separated by agarose gel electrophoresis, purified using a DNA gel purification kit, and sequenced separately by Sanger sequencing. The resulting sequences were submitted to GenBank (accession numbers PP565359 and PP565360).
Rapid amplification of cDNA ends and PCR primer sequences.
| Primer | Sequence |
|---|---|
| IL-31 5′ outside reverse | CTCTTTGTTCAGGTAGTCATCCAA |
| IL-31 5′ inside reverse | GTTGATGTTGCCTGGTGCTTG |
| IL-31 3′ forward | TCCACTAGGTTTGCCCTGTTC |
| IL-31RA forward | GGGAATCAATGGCACCAGGA |
| IL-31RA reverse | AGGTGCTTCAATTTGTTGGGT |
IL-31RA = IL-31 receptor α.
For 5′ sequencing, intron-spanning nested reverse primers were designed targeting conserved regions of the IL-31 mRNA sequences predicted for Equus caballus, E asinus, and E przewalskii. For 3′ sequencing, a single forward primer was similarly designed.
Recombinant eIL-31 protein production
A recombinant eIL-31 protein (r1eIL-31) was first produced by GenScript USA Inc according to the horse IL-31 sequence identified by RACE sequencing. This protein was produced using a mammalian Expi293F cell expression system and expression vector pcDNA3.4, without the use of a tag, and purified using Q Sepharose anion exchange chromatography as previously done to produce recombinant canine IL-31.26 The resulting protein was diluted in sterile PBS, pH 7.2, and stored in aliquots at −80 °C. The protein concentration was determined by a Micro-Bradford protein assay using bovine serum albumin standards, and purity was estimated by SDS-PAGE densitometry analysis using nonreducing conditions and Coomassie Blue stain. Endotoxin contamination was determined using a Limulus Amebocyte Lysate Endotoxin Assay Kit (Xiamen Bioendo Technology Co Ltd).
A second recombinant eIL-31 protein (r2eIL-31) was later produced by KanPro Research using a T7-based protein expression system in Escherichia coli as previously done to produce biologically functional human and cynomolgus monkey recombinant IL-31.27 The resulting E coli IL-31 inclusion bodies were dissolved in 6 M urea and refolded by dialysis, removing the denaturant used. Endotoxin was removed using an endotoxin removal kit (Proteus Endotoxin Removal Midi Kit; Bio-Rad) according to the manufacturer's protocol, and the final product was formulated in PBS containing 88 mM mannitol, 0.2 µm filtered, and frozen at −80 °C in 1 mL aliquots of 330 µg/mL (10X concentration compared with the previous mammalian protein r1eIL-31). The protein concentration was determined by bicinchoninic acid assay, and purity was estimated by SDS-PAGE densitometry analysis. Mannitol was added by KanPro as a cryoprotectant to prevent both pH changes and protein denaturation during freezing, and protein stability was validated after 3 freeze-thaw cycles.
Injection of recombinant eIL-31 in normal horses
Mammalian r1eIL-31 studies
From November 20 through November 27, 2019, 4 normal American Quarter Horse research horses, 1 gelding and 3 mares, were used to examine the pruritogenic effects of r1eIL-31 (normal horses 1 through 4). Symmetrical patches were clipped on either side of the neck, and each horse was randomized to be injected ID first with either r1eIL-31 diluted in sterile saline solution (0.9% sodium chloride for injection) or an equivalent volume of sterile saline. After 48 hours, the horses received the other injection on the opposite side of the neck. Doses of r1eIL-31 were 3 µg (n = 1; normal horse 1), 5 µg (n = 1; normal horse 2), and 10 µg (n = 2; normal horses 3 and 4). Doses were selected based on published studies11,27 in dogs and cynomolgus monkeys.
The horses were watched continuously by 2 unblinded observers for 1 hour before and 4 hours after the injection. The cumulative number of seconds spent rubbing or biting at the site of injection was recorded for each 20-minute block.
Bacterial r2eIL-31 studies
From February 17 through February 25, 2021, 8 normal research horses, 1 Thoroughbred mare, and 7 American Quarter Horses (1 gelding and 6 mares) were used to examine the pruritogenic effects of r2eIL-31 (normal horses 5 through 12). The horses were injected ID with 5 µg r2eIL-31 diluted to 0.3 mL in sterile saline and 0.3 mL sterile saline alone, on opposite sides of the neck, at the same time. The horses were watched continuously by 2 unblinded observers for 15 minutes before (baseline) and 270 minutes (4.5 hours) after injection. Cameras were also set at 2 sides of the stall, and video recordings were later evaluated by a blinded observer. The cumulative number of seconds spent exhibiting pruritus-associated behavior (rubbing/biting, shaking, and/or skin fasciculations) were recorded. The video recording for normal horse 10 was poor quality, so this horse was injected a second time, 3 weeks later.
Generation of equine adherent monocytes transcribing IL-31RA
Adherent equine monocytes were produced to examine cellular responses to recombinant eIL-31 proteins as IL-31RA is known to be expressed on macrophages and monocytes in other species.14,28 Blood (10 mL) was collected from 2 Andalusian mares, 1 allergic and 1 normal, and centrifuged for 10 minutes at 400 X g. The buffy coat was removed and resuspended in 1 mL autologous plasma, then added to 10.5 mL warm lymphocyte culture medium. Cells were cultured in a cell culture–treated 6-well cell culture plate (2 mL/well) and a cell culture–treated 4-well chamber slide (0.5 mL/chamber) for 24 hours at 37 °C in 5% CO2. The adherent cells were then rinsed with warm DPBS to remove nonadherent cells and cultured overnight at 37 °C in 5% CO2. RNA was extracted from the 6-well plates as above, and transcription of IL-31RA was confirmed by PCR as described previously.19 The chamber slides were stained with H&E to characterize the morphology of the adherent cells.
Transcription of IL-31 mRNA was confirmed in the adherent cells, which comprised eosinophils and monocytes, from both the normal and the allergic horses. Ficoll gradient separation was added to the protocol to isolate PBMCs and remove eosinophils prior to monocyte adhesion and culture. In our finalized protocol, 2 mL heparinized blood was diluted in 2 mL DPBS, layered onto 3 mL Ficoll (Ficoll-Paque PLUS; Cytiva) according to the manufacturer's recommendations, and then centrifuged at room temperature for 30 minutes at 700 X g without braking. The plasma and PBMCs were decanted separately. PBMCs were washed twice in lymphocyte culture medium, resuspended in lymphocyte culture medium with 30% added autologous plasma, seeded at a density of 6.0 X 105 cells/well in a cell culture–treated 6-well plate, and incubated for 24 hours at 37 °C, 5% CO2. After incubation and washing with DPBS, the adherent cells were incubated in starving medium (RPMI 1640 medium containing 10,000 U/mL penicillin, 10,000 µg/mL streptomycin, and 25 µg/mL amphotericin B without FBS) for 1.5 hours before use in stimulation experiments.
Phosphorylation of STAT3 in equine adherent monocytes in response to recombinant eIL-31
Following the results of the in vivo studies, we focused on r2eIL-31 for the in vitro experiments because it was produced to a higher purity and concentration. To confirm the biological activity of this recombinant IL-31 protein, we used western blotting to measure the phosphorylation of STAT3 in adherent monocytes generated from allergic horses and stimulated with r2eIL-31. Allergic horses were chosen for this work because IL-31 is a key mediator of pruritus in allergic dogs and humans; thus, the response in allergic horses was of particular relevance. Phosphorylation of STAT3 has been used as a marker of IL-31 signaling in canine and human cells stimulated with recombinant IL-31 protein.26,29
Adherent monocytes were generated from 3 allergic horses (1 Andalusian mare and 2 American Quarter Horses, 1 mare and 1 stallion [allergic horses 1, 2, and 3]), then stimulated with 0.1 µg/mL, 1 µg/mL, 10 µg/mL, or 100 µg/mL r2eIL-31 in starving medium or starving medium alone (unstimulated, 0 µg/mL) for 3 hours at 37 °C in 5% CO2. Adherent monocytes from 2 of the horses were also stimulated with an equivalent volume of PBS/mannitol protein diluent. The medium was then removed, and lysates were made using radioimmunoprecipitation assay buffer supplemented with a protease and phosphatase inhibitor cocktail. The concentration of denatured protein lysates was determined by using a bicinchoninic acid assay kit, and they were frozen in single-use aliquots at −80 °C. The levels of total STAT3 (tSTAT3), phosphorylated STAT3 (pSTAT3), and β-actin in each lysate were measured by western blot. Aliquots of 18 to 20 µg of denatured protein lysate were separated on a 4% to 20% tris-glycine polyacrylamide gel in tris-glycine SDS running buffer and transferred to nitrocellulose membranes by electrophoresis in tris-glycine transfer buffer. The membranes were blocked using blocking buffer (5% bovine serum albumin in tris-buffered saline, pH 7.4, with 0.05% Tween 20) for 2 hours, then incubated overnight at 4 °C with the primary antibodies against tSTAT3 (rabbit IgG monoclonal anti-STAT3, clone D3Z2G; Cell Signaling Technology) or pSTAT3 (rabbit IgG monoclonal anti–phospho-STAT3 Tyr705, clone D3A7; Cell Signaling Technology), diluted 1:1,000 in blocking buffer. Membranes were washed in tris-buffered saline, pH 7.4, with 0.05% Tween 20, then incubated with goat anti-rabbit IgG horseradish peroxidase (HRP)-conjugated secondary polyclonal antibody (4030-05; Southern Biotechnology) diluted 1:1,000 in blocking buffer for 1 hour. The membranes were washed, then imaged using chemiluminescent substrate (Crescendo Western HRP Substrate; EMD Millipore). β-actin was used as an internal loading control using the same western blotting procedure with modifications: mouse IgG1 anti–β-actin monoclonal antibody (clone AC-15; Santa Cruz Biotechnology) was used, diluted 1:20,000 for 1 hour, followed by goat anti-mouse IgG1 HRP-conjugated secondary polyclonal antibody (1071-05; Southern Biotechnology). Bands on the resulting images were analyzed by densitometry. Each set of stimulated and unstimulated adherent monocyte protein lysates from the same horse was run on the same western blot. The set of protein lysates generated from the first horse (allergic horse 1) was run on triplicate pSTAT3 western blots to establish the repeatability of technical replicates. The second and third sets of lysates from the remaining 2 horses were run as biological replicates, without technical replicates of each.
In addition to western blots, analysis of STAT3 phosphorylation in adherent monocytes stimulated with r2eIL-31 was also measured using AlphaLISA assays at a later date using an updated cell stimulation protocol. A second blood sample was collected from allergic horse 1, and adherent monocytes were generated as before with modifications: Dulbecco Modified Eagle Medium was used instead of RPMI 1640, and the cells were starved for 1 hour before stimulation with 0.0001 µg/mL, 0.001 µg/mL, 0.01 µg/mL, 0.1 µg/mL, 1.0 µg/mL, or 10 µg/mL bacterial r2eIL-31 in Dulbecco Modified Eagle Medium supplemented with 10% FBS or medium alone (unstimulated, 0 µg/mL). Cells were also stimulated with PBS/mannitol protein diluent at equivalent dilutions. After stimulation, the cells were lysed using AlphaLISA Surefire Ultra Lysis Buffer (Revvity Health Sciences Inc) supplemented with the same protease and phosphatase inhibitor cocktail used in preparing lysates for western blotting. The AlphaLISA Surefire Ultra Human and Mouse Phospho-STAT3 (Tyr705) Detection Kit and AlphaLISA Surefire Ultra Total STAT3 Detection Kit (both Revvity Health Sciences Inc) were used to compare tSTAT3 and pSTAT3 levels in the lysates, in duplicate, following the manufacturer's protocol.
Statistical analysis
Statistical analysis of the pilot data obtained during the r1eIL-31 in vivo study was not appropriate as only 1 of the dosages was used in duplicate. The cumulative total number of seconds spent exhibiting pruritus-associated behavior following injection of r2eIL-31 was normalized to 15 minutes to account for the difference in observation time before and after injection. A 1-tailed Wilcoxon matched-pairs signed rank test (Prism, version 10.3.1; Graphpad Software) was used to identify significant increases in itching after injection. The level of phosphorylation of STAT3 was calculated by normalizing the intensity of each pSTAT3 and tSTAT3 band to its respective β-actin control, expressing the normalized pSTAT3 as a percentage of the normalized tSTAT3 for each lysate, then calculating the fold change from the unstimulated lysate for each horse. For allergic horse 1, the mean of the 3 pSTAT3 technical replicates for each lysate was used. Normality testing of the data was followed by a Friedman test and post hoc Dunn multiple comparisons tests comparing to unstimulated cells. The AlphaLISA data were similarly processed, but statistical analysis was not possible as there was only 1 set of lysates. The cutoff for statistical significance was 0.05.
Results
Equine IL-31 mRNA RACE sequencing and recombinant protein production
Rapid amplification of cDNA ends sequencing identified 2 IL-31 mRNA sequences, which differed in the length of the 5′ untranslated region. Both sequences encoded the same protein when translated using the “translate nucleotide” translation tool available on the Expert Protein Analysis System.30 The sequence differed from the predicted Equus caballus IL-31 protein sequences listed in the NCBI GenBank (NCBI reference sequence: XP_023503836.1) and UniProt (accession number: F7AHG9) databases: the first 6 amino acids at the 5′ end of the translated protein matched the UniProt sequence, and the remainder matched the 3′ end of the NCBI sequence (Figure 1).
Alignment of the predicted horse IL-31 protein sequences from Uniprot (F7AHG9) and the National Center for Biotechnology Information (XP_023503836.1), with the IL-31 mRNA sequence produced in this study (equine IL-31 [eIL-31]). Green text indicates an amino acid match between our eIL-31 sequence and 1 or both of the predicted sequences. *All residues or nucleotides in the above column are identical. :Conserved substitutions have been observed in the above column. Semiconserved substitutions have been observed in the above column. Blank space means there is no match between all 3 sequences in the above column.
Citation: American Journal of Veterinary Research 85, 12; 10.2460/ajvr.24.05.0144
The r1eIL-31 from GenScript was 70% pure at a concentration of 33 µg/mL, with an endotoxin level of < 6.0 EU/mg, which meets recommended limits for injection (5 EU/kg/h).31 The r2eIL-31 from KanPro was calculated to be > 95% pure, with a level of endotoxin < 1 EU/mg (based on previous KanPro data from similarly produced proteins), and was diluted to a concentration of 330 µg/mL.
Intradermal injection of recombinant eIL-31 is pruritogenic in horses
During the first in vivo study, no pruritus-associated behavior (itching) was observed in the horses during the 1 hour before or 4 hours after saline injection, except in 1 horse, which itched for 2 seconds during the postsaline injection period. The cumulative number of seconds spent itching by each horse during the 1-hour pre–r1eIL-31 injection period and 4-hour post–r1eIL-31 injection period is summarized in Supplementary Table S1. For comparison with the preinjection period, the postinjection total in Supplementary Table S1 is also expressed normalized to 1 hour. However, the itching was not evenly distributed over the observation period and was concentrated in bursts (Figure 2). After the lowest dose of r1eIL-31 (3 µg, normal horse 1), 6 seconds of itching were recorded between 160 to 180 minutes after injection. After 5 µg of r1eIL-31 (normal horse 2), a cumulative 317 seconds of itching were recorded in 2 peaks between 20 to 40 and 140 to 200 minutes postinjection. After 10 µg of mammalian recombinant r1eIL-31 (n = 2), 1 horse (normal horse 4) itched for a cumulative 79 seconds, peaking between 100 to 120 and 140 to 160 minutes postinjection. The other horse administered a 10-µg dose of r1eIL-31 (normal horse 3) only exhibited a cumulative 15 seconds of itch-related behavior, which occurred in the periods between 40 to 60 and 240 to 260 minutes postinjection. However, this horse spent a much larger proportion of the time appearing irritated (expressed as skin fasciculations, which were noted but not recorded as itching behavior according to the predefined criteria used at the time, and a tense facial expression) without rubbing or scratching at the injection site. Skin fasciculations were therefore included in the later study to capture this behavior.
Pruritus-associated behavior following ID injection of mammalian cells produced recombinant equine IL-31 protein (r1eIL-31). During November 2019, 4 normal horses were injected ID: 1 each with 0.2 mL sterile saline containing 3 µg (normal horse 1) or 5 µg (normal horse 2) r1eIL-31 and 2 with 0.4 mL containing 10 µg r1eIL-31 (normal horses 3 and 4). Observation started 1 hour prior to the injection of the recombinant and continued for 4 hours after. On the y-axis, the mean cumulative seconds spent rubbing or biting at the site of r1eIL-31 injection during each 20-minute block; on the x-axis, time relative to injection.
Citation: American Journal of Veterinary Research 85, 12; 10.2460/ajvr.24.05.0144
The cumulative number of seconds that each horse spent itching during the 15-minute pre–r2eIL-31 injection period and the 4.5-hour post–r2eIL-31 injection period in the second in vivo study is summarized in Supplementary Table S2. For comparison with the preinjection period, the postinjection total in Supplementary Table S2 is also expressed normalized to 15 minutes. The itching was again not evenly distributed over the observation period and was concentrated in bursts (Figure 3). The itch response was variable, with 3 horses itching for less than 100 seconds in total (normal horses 7, 9, and 12) and 4 for between 115 and 542 seconds (normal horses 5, 6, 8, and 11). Normal horse 10 responded most strongly to the injection of r2eIL-31 and was also the only horse to direct pruritus-associated behavior towards the site of the saline injection. The video recording of normal horse 10 was poor quality, so injection of this horse was repeated 3 weeks later. Following the first injection, normal horse 10 spent a cumulative 1,053 seconds rubbing the r2eIL-31 injection site and 137 seconds rubbing the saline injection site over the 4 hours postinjection. Following her second injection 3 weeks later, normal horse 10 spent a cumulative 2,225 seconds rubbing the r2eIL-31 injection site and 573 seconds rubbing the saline injection site. None of the other horses directed itch responses against the site of the saline injection. A video clip example of the pruritus resulting from injection of r2eIL-31 is supplied in Supplementary Video S1.
Pruritus-associated behavior following ID injection of bacteria-produced recombinant IL-31 protein (r2eIL-31). During February 2021, 8 normal horses (normal horses 5 through 12) were injected ID with 0.3 mL sterile saline containing a 5-µg dose of r2eIL-31 on 1 side of the neck and 0.3 mL sterile saline alone on the other side. Observation started 15 minutes prior to the injection of the recombinant and continued for 4.5 hours after. The y-axis illustrates the cumulative seconds in each minute spent exhibiting pruritus (skin fasciculations, scratching, shaking, and/or rubbing at the site of r2eIL-31 injection). The x-axis illustrates the times in minutes relative to injection of the r2eIL-31. A—Normal horses 5, 6, and 7. B—Normal horses 8, 9, and 10. C—Normal horses 11 and 12 and the second r2eIL-31 challenge of normal horse 10 three weeks after the first.
Citation: American Journal of Veterinary Research 85, 12; 10.2460/ajvr.24.05.0144
The distribution of the itching was variable. Four of the horses expressed the majority of their itching later, after the first 2 hours post r1eIL-31 injection (normal horses 5, 8, 10, and 11). One horse itched most within the first 2 hours (horse 6), and the remaining 3 horses, which all exhibited less than 100 seconds of itching in total, expressed itch in short bursts spread more evenly through the 4.5-hour observation period (normal horses 7, 9, and 12).
Unexpectedly, the difference in the number of seconds spent itching before and after injection of r2eIL-31 did not reach the threshold for statistical significance (P = .0742). This may be due to the small number of horses used in the study and the fact that the majority of the horses recorded 0 seconds of itch during the 15-minute baseline observation period. The Wilcoxon signed rank test assumes continuous data, and in this case “zero inflation” may have compromised its ability to detect a significant difference. This could be improved in further studies by increasing the duration of the preinjection observation period, which would be more likely to record small, baseline amounts of itching behavior.
Bacterial r2eIL-31 stimulates STAT3 phosphorylation in equine adherent monocytes
Transcription of IL-31 mRNA in adherent equine leukocytes (comprising mixed eosinophils and monocytes) was confirmed in cells from both a normal and an allergic horse and was 10-fold higher in the cells from the normal horse.
Stimulation of adherent monocytes from allergic horses with r2eIL-31 resulted in a significant increase in the percentage of pSTAT3, measured by western blot (Figure 4), when unstimulated cells were compared with cells stimulated with 10 µg/mL r2eIL-31 (P = .018). No other significant differences were detected. Data on STAT3 phosphorylation in response to PBS/mannitol diluent were only obtained from 1 horse (allergic horse 2) due to technical errors and an insufficient sample for this to be repeated.
Phosphorylation of signal transducer and activator of transcription 3 (STAT3) in equine adherent monocytes following stimulation with bacterial recombinant IL-31 protein r2eIL-31. Adherent monocytes were generated from a single blood sample from each of 3 pruritic, allergic horses. These cells were then stimulated in medium containing either 0 µg/mL (unstimulated), 0.1 µg/mL, 1 µg/mL, 10 µg/mL, or 100 µg/mL r2eIL-31. The amount of total and phosphorylated STAT3 (pSTAT3) and β-actin in each resulting lysate was measured by western blot. The lysates generated from allergic horse 1 were run on triplicate pSTAT3 western blots (shown separately) to assess repeatability of the assay (the mean for each lysate was used for analysis). The level of pSTAT3 in each lysate was calculated as a percentage of the total STAT3, after normalization of both to β-actin, and expressed as a fold change from 0 µg/mL. Normality testing of the data was followed by a Friedman test and post hoc Dunn multiple comparisons tests comparing to 0 µg/mL r2eIL-31. *Significant difference from 0 µg/mL; P = .018.
Citation: American Journal of Veterinary Research 85, 12; 10.2460/ajvr.24.05.0144
After stimulation of adherent monocytes from a single horse with escalating concentrations of r2eIL-3, phosphorylation of STAT3 measured by AlphaLISA appeared to be increased in a dose-dependent manner (Figure 5). Statistical analysis of this data was not possible as there was only 1 sample of each lysate. Stimulation with an equivalent dilution of PBS/mannitol protein diluent also appeared to provoke STAT3 phosphorylation but to a lesser extent.
Phosphorylation of STAT3 in equine adherent monocytes following stimulation with bacterial recombinant IL-31 protein r2eIL-31 or an equivalent dilution of PBS/mannitol. Adherent monocytes were generated from a single blood sample from 1 allergic horse and stimulated with 0 µg/mL (unstimulated), 0.0001 µg/mL, 0.001 µg/mL, 0.01 µg/mL, 0.1 µg/mL, 1 µg/mL, or 10 µg/mL r2eIL-31 or an equivalent dilution of PBS/mannitol. The levels of total and pSTAT3 were measured by AlphaLISA in triplicate. Both the pSTAT3 and total STAT3 in each stimulated lysate were normalized to the unstimulated sample. The normalized pSTAT3 was then divided by the normalized total STAT3. No statistical analysis was done as only 1 horse was used.
Citation: American Journal of Veterinary Research 85, 12; 10.2460/ajvr.24.05.0144
Discussion
In this proof-of-concept study, RACE sequencing allowed the identification of discrepancies between our translated IL-31 protein sequence and the available horse IL-31 sequences in the GenBank and Uniprot databases. We demonstrated that recombinant IL-31 protein can elicit a pruritic response in normal horses and that stimulation of equine adherent monocytes with r2eIL-31 can produce an increase in STAT3 phosphorylation. Both of these findings are consistent with studies11,27 of IL-31 in other species.
The pruritogenic response after ID injection of recombinant IL-31 protein in our horses was variable, and was not immediate, similar to what has been reported in other species.11,27,32 In 6 cynomolgus monkeys, the response after ID injection of 19.5 μg recombinant cynomolgus monkey IL-31 protein peaked in the second hour postinjection.27 In 10 normal laboratory beagle dogs, variable levels of itching were observed following ID injection of 1.75 μg/kg recombinant canine IL-31 protein.32 No pruritus was observed in the first hour, and delayed onset pruritus peaked at 270 minutes.32 In mouse models, ID injection of recombinant IL-31 protein caused a gradual increase in long-lasting generalized pruritus and scratching about 3 hours after administration.33 The speed of response appears to be linked to the route of administration. When cynomolgus monkeys were injected IV, a generalized pruritus response was evident within minutes.27 Interestingly, a second, delayed pruritus response at 3 hours was also noted with IV administration.27 In this study, we wanted to minimize the chance of a generalized pruritus response; thus, only the ID route was used.
The response to recombinant IL-31 protein injection is dose dependent. In the cynomolgus monkey study,27 the dose was calculated per body weight, with a maximum dose of 3 µg/kg, and the authors concluded that, to induce a reproducible response, a dose of 19.5 µg/ID injection was needed. Pearson et al32 injected normal laboratory beagle dogs with recombinant canine IL-31 ID based on weight (1.75 µg/kg). For our study, a fixed dose per injection was used rather than a dose calculated according to body weight. If we had used the protocol used in dogs, an ID dose of 787 µg would have been needed, which was not feasible. Our horses received doses ranging from 3 to 10 μg/injection (approx 0.007 to 0.022 μg/kg), but a larger and more consistent itch response may have been observed using a higher dose. Thus, the pruritogenic response observed using a lower and possibly more physiologically relevant dose is meaningful. Humans with pruritus and allergic disease, such as chronic urticaria, have circulating serum levels of IL-31 protein in the range of 10,000 to 20,000 pg/mL (0.01 to 0.02 μg/mL).34
We observed a stronger response when we injected normal horse 10 the second time. This exaggerated response after a second injection has also been described in dogs: when normal laboratory beagles received 2 IV injections of recombinant IL-31 protein, with a 2-week interval in between, the response was stronger after the second injection.35 The exact reason for this response is not known, but it is proposed that IL-31 could upregulate IL-31 receptors in dorsal root ganglia as described in mice receiving multiple IL-31 injections.36
The observation that IL-31RA transcription in adherent leukocytes was higher in the cells from the normal horse is similar to our previous findings that some normal dogs exhibited raised IL-31RA transcription in skin biopsies compared with atopic dogs.37 Whether the IL-31RA PCR assay used in this study identifies only full-length, functional IL-31RA mRNA or includes unidentified splice variants is currently unknown. Interleukin 31 receptor α mRNA splice variants, encoding protein isoforms that may have varying functions, have been identified in dogs and humans but not yet in horses.37,38
For the in vitro experiments, we focused on allergic horses because any novel therapeutics would be used in allergic horses. Although we only tested adherent monocytes from 3 allergic horses, phosphorylation of STAT3 was consistent. The exaggerated STAT3 phosphorylation observed in 1 of the 3 technical replicates of the pSTAT3 western blot from allergic horse 1 is likely due to a technical error during imaging of the blot as frozen aliquots of the same lysate were used for all 3 replicates. The largest response was seen with 10 µg/mL of r2eIL-31 rather than the highest concentration (100 µg/mL). This was possibly due to induction of negative feedback mechanisms (as has been identified via SOCS3 in mice) or direct or indirect (via release of other cytokines) cytotoxicity inhibiting IL-31 signaling at the higher concentration.39
This work was limited by the small numbers of horses available and errors in the study design. In a pilot study (data not shown), we found that ID injection of saline and PBS/mannitol protein diluent did not elicit any irritation/reaction; thus, we used saline as our negative control for the in vivo experiments. The r2eIL-31 in vivo challenge would have been improved by the use of an equivalent dilution of the PBS/mannitol protein diluent as the control, randomized administration of this control or r2eIL-31, and a washout period to allow any response to subside. Blinding of the observers and a longer preinjection observation period would also have improved the robustness of the study design. No adverse effects were reported by study staff, but video monitoring over the following 24 hours would have allowed us to identify any additional delayed-onset pruritus.
This demonstration that IL-31 can elicit pruritus and activate STAT3 in horses, as in other species, supports this cytokine as a suitable target for novel therapies for equine pruritus. More work is needed to understand the events triggered by IL-31 receptor binding in various tissues, the subsequent signaling cascades and inflammatory responses, and the modulation of the receptor expression and signal transduction.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org.
Acknowledgments
The authors wish to thank the horse owners, the staff at the University of Florida's equine facilities and Interdisciplinary Center for Biotechnology Research Monoclonal Antibody Core (research resource identifier: SCR_019147), and Dr. Philip Gao.
Disclosures
The authors have nothing to disclose. No AI-assisted technologies were used in the generation of this manuscript.
Funding
This study was funded by University of Florida College of Veterinary Medicine Research Grants and the American College of Veterinary Dermatology.
ORCID
N. M. Craig https://orcid.org/0000-0001-7370-4437
N. S. Munguia https://orcid.org/0000-0003-4827-5188
R. Marsella https://orcid.org/0000-0003-2329-0453
References
- 1.↑
Lomas HR, Robinson PA. A pilot qualitative investigation of stakeholders’ experiences and opinions of equine insect bite hypersensitivity in England. Vet Sci. 2018;5(1):3. doi:10.3390/vetsci5010003
- 2.↑
van Grevenhof EM, Ducro B, Heuven HC, Bijma P. Identification of environmental factors affecting the prevalence of insect bite hypersensitivity in Shetland Ponies and Friesian Horses in the Netherlands. Equine Vet J. 2007;39(1):69–73. doi:10.2746/042516407X153020
- 3.↑
Marsella R, White S, Fadok VA, et al. Equine allergic skin diseases: clinical consensus guidelines of the world association for veterinary dermatology. Vet Dermatol. 2023;34(3):175–208. doi:10.1111/vde.13168
- 4.
Loeffler A, Herrick D, Allen S, Littlewood JD. Long-term management of horses with atopic dermatitis in southeastern England: a retrospective questionnaire study of owners’ perceptions. Vet Dermatol. 2018;29(6):526–e176. doi:10.1111/vde.12685
- 5.↑
Olsén L, Bondesson U, Broström H, et al. Pharmacokinetics and effects of cetirizine in horses with insect bite hypersensitivity. Vet J. 2011;187(3):347–351. doi:10.1016/j.tvjl.2009.12.030
- 6.
Friberg C, Logas D. Treatment of Culicoides hypersensitive horses with high dose n-3 fatty acids: a double-blinded cross-over study. Vet Dermatol. 1999;10(2):117–122.
- 7.↑
Ginel PJ, Hernández E, Lucena R, Blanco B, Novales M, Mozos E. Allergen-specific immunotherapy in horses with insect bite hypersensitivity: a double-blind, randomized, placebo-controlled study. Vet Dermatol. 2014;25(1):29–e10. doi:10.1111/vde.12092
- 8.↑
Marshall K, Marsella R. Evolution of the prevalence of antibiotic resistance to Staphylococcus spp. isolated from horses in Florida over a 10-year period. Vet Sci. 2023;10(2):71. doi:10.3390/vetsci10020071
- 9.↑
Sutaria N, Adawi W, Goldberg R, Roh YS, Choi J, Kwatra SG. Itch: pathogenesis and treatment. J Am Acad Dermatol. 2022;86(1):17–34. doi:10.1016/j.jaad.2021.07.078
- 10.
Hassan I, Haji ML. Understanding itch: an update on mediators and mechanisms of pruritus. Indian J Dermatol Venereol Leprol. 2014;80(2):106–114. doi:10.4103/0378-6323.129377
- 11.↑
Gonzales AJ, Fleck TJ, Humphrey WR, et al. IL-31-induced pruritus in dogs: a novel experimental model to evaluate anti-pruritic effects of canine therapeutics. Vet Dermatol. 2016;27(1):34–e10. doi:10.1111/vde.12280
- 12.↑
Furue M, Yamamura K, Kido-Nakahara M, Nakahara T, Fukui Y. Emerging role of interleukin-31 and interleukin-31 receptor in pruritus in atopic dermatitis. Allergy. 2018;73(1):29–36. doi:10.1111/all.13239
- 13.
Datsi A, Steinhoff M, Ahmad F, Alam M, Buddenkotte J. Interleukin-31: the "itchy" cytokine in inflammation and therapy. Allergy. 2021;76(10):2982–2997. doi:10.1111/all.14791
- 14.↑
Nemmer JM, Kuchner M, Datsi A, et al. Interleukin-31 signaling bridges the gap between immune cells, the nervous system and epithelial tissues. Front Med (Lausanne). 2021;8:639097. doi:10.3389/fmed.2021.639097
- 15.
Van Brussel L, Moyaert H, Escalada M, Mahabir SP, Stegemann MR. A masked, randomised clinical trial evaluating the efficacy and safety of lokivetmab compared to saline control in client-owned dogs with allergic dermatitis. Vet Dermatol. 2021;32(5):477–e131. doi:10.1111/vde.12984
- 16.
Marsella R, Doerr K, Gonzales A, Rosenkrantz W, Schissler J, White A. Oclacitinib 10 years later: lessons learned and directions for the future. J Am Vet Med Assoc. 2023;261(S1):S36–S47. doi:10.2460/javma.22.12.0570
- 17.
Serra-Baldrich E, Santamaría-Babí LF, Francisco Silvestre J. Nemolizumab: an innovative biologic treatment to control interleukin 31, a key mediator in atopic dermatitis and prurigo nodularis. Actas Dermosifiliogr. 2022;113(7):674–684. doi:10.1016/j.ad.2021.12.014
- 18.↑
Chovatiya R, Paller AS. JAK inhibitors in the treatment of atopic dermatitis. J Allergy Clin Immunol. 2021;148(4):927–940. doi:10.1016/j.jaci.2021.08.009
- 19.↑
Craig NM, Munguia NS, Trujillo AD, Wilkes R, Dorr M, Marsella R. Transcription of interleukin 31 and its receptor by leukocytes after Culicoides sp stimulation is dose dependent but is not exaggerated in allergic horses or correlated with pruritus. J Am Vet Med Assoc. 2023;261(S1):S75–S85. doi:10.2460/javma.22.12.0588
- 20.↑
Visser M, Cleaver D, Cundiff B, King V, Sture G. Oclacitinib maleate (Apoquel) dose determination in horses with naturally occurring allergic dermatitis. 2020 ACVIM forum on demand research abstract program. J Vet Int Med. 2020;34:2977–2978.
- 21.↑
Olomski F, Fettelschoss V, Jonsdottir S, et al. Interleukin 31 in insect bite hypersensitivity-alleviating clinical symptoms by active vaccination against itch. Allergy. 2020;75(4):862–871. doi:10.1111/all.14145
- 22.↑
Fettelschoss V, Olomski F, Birkmann K, Kündig TM, Bergvall K, Fettelschoss-Gabriel A. Interleukin 31 and targeted vaccination in a case series of six horses with chronic pruritus. Equine Vet Educ. 2021;33(12):e490–e500. doi:10.1111/eve.13408
- 23.↑
Sayers EW, Bolton EE, Brister JR, et al. Database resources of the national center for biotechnology information. Nucleic Acids Res. 2022;50(D1):D20–D26. doi:10.1093/nar/gkab1112
- 24.↑
Cunningham F, Allen JE, Allen J, et al. Ensembl 2022. Nucleic Acids Res. 2022;50(D1):D988–D995. doi:10.1093/nar/gkab1049
- 25.↑
Ye J, Coulouris G, Zaretskaya I, Cutcutache I, Rozen S, Madden TL. Primer-BLAST: a tool to design target-specific primers for polymerase chain reaction. BMC Bioinformatics. 2012;13:134. doi:10.1186/1471-2105-13-134
- 26.↑
Gonzales AJ, Humphrey WR, Messamore JE, et al. Interleukin-31: its role in canine pruritus and naturally occurring canine atopic dermatitis. Vet Dermatol. 2013;24(1):48–53.e11-2. doi:10.1111/j.1365-3164.2012.01098.x
- 27.↑
Lewis KE, Holdren MS, Maurer MF, et al. Interleukin (IL) 31 induces in cynomolgus monkeys a rapid and intense itch response that can be inhibited by an IL-31 neutralizing antibody. J Eur Acad Dermatol Venereol. 2017;31(1):142–150. doi:10.1111/jdv.13794
- 28.↑
Ghilardi N, Li J, Hongo JA, Yi S, Gurney A, de Sauvage FJ. A novel type I cytokine receptor is expressed on monocytes, signals proliferation, and activates STAT-3 and STAT-5. J Biol Chem. 2002;277:16831–16836. doi:10.1074/jbc.M201140200
- 29.↑
Yaseen B, Lopez H, Taki Z, et al. Interleukin-31 promotes pathogenic mechanisms underlying skin and lung fibrosis in scleroderma. Rheumatology. 2020;59(9):2625–2636. doi:10.1093/rheumatology/keaa195
- 30.↑
Gasteiger E, Gattiker A, Hoogland C, Ivanyi I, Appel RD, Bairoch A. ExPASy: the proteomics server for in-depth protein knowledge and analysis. Nucleic Acids Res. 2003;31(13):3784–3788. doi:10.1093/nar/gkg563
- 32.↑
Pearson J, Leon R, Starr H, Kim SJ, Fogle JE, Banovic F. Establishment of an intradermal canine IL-31-induced pruritus model to evaluate therapeutic candidates in atopic dermatitis. Vet Sci. 2023;10(5):329. doi:10.3390/vetsci10050329
- 33.↑
Arai I, Tsuji M, Takeda H, Akiyama N, Saito S. A single dose of interleukin-31 (IL-31) causes continuous itch-associated scratching behaviour in mice. Exp Dermatol. 2013;22(10):669–671.
- 34.↑
Raap U, Wieczorek D, Gehring M, et al. Levels of serum IL-31 in chronic spontaneous urticaria. Exp Dermatol. 2010;19(5):464–466.
- 35.↑
Pearson J, Denley T, Blubaugh A, et al. Characterisation of the pruritus responses and pruritic behaviours in an interleukin 31-induced canine model of pruritus. Vet Dermatol. 2024;35(3):296–304. doi:10.1111/vde.13231
- 36.↑
Arai I, Tsuji M, Miyagawa K, Takeda H, Akiyama N, Saito S. Repeated administration of IL-31 upregulates IL-31 receptor a (IL-31RA) in dorsal root ganglia and causes severe itch-associated scratching behaviour in mice. Exp Dermatol. 2015;24:75–78. doi:10.1111/exd.12587
- 37.↑
Craig N, Ahrens K, Wilkes R, Marsella R. Reduced IL-31 receptor alpha splice variant mRNA following allergen challenge in a canine model of atopic dermatitis. Allergy. 2021;76(10):3206–3209. doi:10.1111/all.15005
- 38.↑
Cevikbas F, Wang X, Akiyama T, et al. A sensory neuron–expressed IL-31 receptor mediates t helper cell–dependent itch: involvement of TRPV1 and TRPA1. J Allergy Clin Immunol. 2014;133(2):448–460.
- 39.↑
Maier E, Mittermeir M, Ess S, et al. Prerequisites for functional interleukin 31 signaling and its feedback regulation by suppressor of cytokine signaling 3 (SOCS3). J Biol Chem. 2015;290(41):24747–24759. doi:10.1074/jbc.M115.661306
