ORIGINAL RESEARCH published: 10 September 2020 doi: 10.3389/fvets.2020.00631 Collagen and Eosinophils in Jenny’s Endometrium: Do They Differ With Endometrial Classification? Jordi Miró 1*, Miguel Gutiérrez-Reinoso 1 , Joana Aguiar da Silva 2 , Carina Fernandes 2 , Maria Rosa Rebordão 2,3 , Graça Alexandre-Pires 2 , Jaime Catalán 1 and Graça Ferreira-Dias 2 1 Equine Reproduction Service, Department of Animal Medicine and Surgery, Faculty of Veterinary Medicine, Autonomous University of Barcelona, Cerdanyola del Vallès, Spain, 2 Faculdade de Medicina Veterinária, CIISA - Centro de Investigação Interdisciplinar em Sanidade Animal, Universidade de Lisboa, Lisbon, Portugal, 3 Coimbra College of Agriculture, Polytechnic Institute of Coimbra, Coimbra, Portugal Collagen fibers and inflammatory cells are the basis for jenny endometrium Kenney and Doig’s classification developed for the mare. The infiltration of a large number of eosinophils in the jenny endometrium is intriguing. Eosinophil and fibroblast produced IL33, which has been related to fibrosis development and chronicity. This work on the Edited by: endometrium consisted of (i) quantification of collagen type I (COL1A2), type III (COL3A1), Amy Katherine McLean, University of California, Davis, and IL33 transcripts; (ii) histological localization and quantification of COL1 and COL3 United States proteins; and (iii) eosinophil and neutrophil count and correlation with collagen area Reviewed by: and IL33 transcripts. Localization of COL protein in the jenny endometrium was also Marco Antonio Alvarenga, São Paulo State University, Brazil compared to the mare endometrium. As fibrosis increased, eosinophil and neutrophil Miguel Quaresma, count decreased (P < 0.05). A 5-fold increase in IL33 transcripts was noted from Universidade de Trás os Montes e Alto categories IIA to III. There was a tendency toward a positive correlation between Douro, Portugal eosinophil count and IL33 transcripts in category IIA endometrium (P = 0.055). Neither *Correspondence: Jordi Miró transcripts of COL1A2 nor COL3A1 nor the areas of COL1 or COL3 differed with jordi.miro@uab.cat endometrial categories. Unlike for the mare, and regardless of the jenny endometrium classification, COL3 was always found to different extents in the stratum compactum, Specialty section: This article was submitted to while COL1 was mainly present in deep stroma. As fibrosis progressed in the mare, Comparative and Clinical Medicine, an extensive increase in COL1 fibers was notorious under the surface epithelium. a section of the journal Correlations between neutrophil count and COL1 and COL3 areas were observed in the Frontiers in Veterinary Science jenny endometrium, although no correlation was found for eosinophil count. Neutrophil Received: 01 May 2020 Accepted: 03 August 2020 count positive correlation with the COL1 area and negative correlation with the COL3 area Published: 10 September 2020 in endometria with mild lesions suggest that neutrophils in the jenny endometrium may Citation: be involved in fibrogenesis. In addition, when eosinophilia subsides, the endometrium Miró J, Gutiérrez-Reinoso M, Silva JAd, Fernandes C, reacts with fibrosis establishment, which could be stimulated by the pro-fibrotic cytokine Rebordão MR, Alexandre-Pires G, IL33, whose release might then be ascribed to fibroblasts. Further studies are needed Catalán J and Ferreira-Dias G (2020) to analyze the effect of the presence of COL3 next to the surface epithelium in the Collagen and Eosinophils in Jenny’s Endometrium: Do They Differ With stratum compactum, or around the endometrial glands on jenny’s endometrial function Endometrial Classification? and fertility. Front. Vet. Sci. 7:631. doi: 10.3389/fvets.2020.00631 Keywords: donkey, endometrium, collagen, fibrosis, endometrosis, eosinophils, IL-33 Frontiers in Veterinary Science | www.frontiersin.org 1 September 2020 | Volume 7 | Article 631 Miró et al. Jenny Endometritis and Endometrosis INTRODUCTION for 75 days (April 1st/June 15th, 2017) with ten Catalan jackasses, aged 5–9 years and of proven fertility, for natural breeding. In jenny’s endometrium, the infiltration of eosinophils and Jennies were assigned to different groups according to selection the presence of fibrosis to different extents might influence criteria. All animals were included in the Catalonian donkey endometrial function and gestation in a dissimilar fashion to the Studbook and under a selection program. One month after mare. The frequent infiltration of a large number of eosinophils jackass removal, a pregnancy diagnosis was done by transrectal in jenny’s endometrium, unlike for the mare, appears to be ultrasonography using a MyLabTM Gamma (Esaote, Genova, a physiological feature of the jenny endometrium (1). The Italy) device. eosinophils appear to regulate tissue homeostasis and contribute All animals were owned by a commercial donkey farm, for inflammatory reactions in tissue remodeling and innate FUIVES (Berga, Barcelona, Spain). Donkeys were kept in immunity (2). In general, eosinophils are considered as the paddocks and fed grain forage, straw, hay, and water ad libitum. major effector cells in type two inflammatory diseases, including A total of 29 jennies did not get pregnant (63.75% fertility asthma, helminthic infection, and allergy (2). In fact, eosinophils rate), and 23 of them were included in the present study. result from the differentiation of CD117+ progenitor cells, Six jennies were discarded for different reasons: three of them directly stimulated by interleukin 33 (IL33) (3). In airway presented cervical stenosis, one died, one showed a severe skin inflammation, IL33 besides exacerbating eosinophil infiltration disease, and the other one had never foaled before. Endometrial in an autocrine and paracrine manner also attracts macrophages, biopsies were obtained during estrous from 23 non-pregnant lymphocytes, IL-13, and TGF-β1 (3). A relationship between jennies through the cervix, with an alligator jaw biopsy punch eosinophil infiltration and the expression of IL33 has also been (Kevorkian’s uterine biopsy forceps, Hauptner and Herberholz, referred in chronicity and development of fibrosis in liver, skin, Solingen, Germany), at estrus. All 23 jennies used in this study and intestine (4–6). had foaled at least once before. Each biopsy was divided into In the mare endometrium, the excessive deposition of two pieces. One portion was immediately immersed in buffered collagen in the extracellular matrix, predominantly around formaldehyde, for paraffin blocks, and further processed for the endometrial glands and stroma, which is associated histological examination. The other piece of the endometrium with a chronic degenerative condition, destruction of tissue was placed in cryotubes with RNAlater R (AM7020; Ambion, architecture, and impairment of endometrial function, is named Applied Biosystems, Foster, CA, USA) and kept at −80◦ C, for endometriosis (7, 8). In tissue repair, collagen type III (COL3) gene transcription studies, by real-time PCR. is the first one to be expressed, being replaced by collagen type I (COL1) in later stages of fibrogenesis (9). Nevertheless, in the Histopathological Analysis mare endometrium at the transcript level, age appeared not to For histopathological analysis, 5-µm-thick sections of influence COL1 or COL3 mRNA (10). In contrast, quantification formaldehyde-fixed endometrium were stained with of the areas in the endometrium occupied by COL1 and COL3, hematoxylin (05-06014E; Bio-Optica) and eosin (HT1103128; stained with histochemical picrosirius red staining and visualized Sigma-Aldrich). Every jenny endometrial biopsy was evaluated by polarized light microscopy (11, 12), showed an increase in by light microscopy (Leica DM500) regarding the histopathology COL1 in the tissues with severe endometriosis, in comparison to of endometrial glands, presence of inflammatory cells healthy endometria or with mild fibrotic changes, where COL3 (neutrophils, eosinophils), and fibrosis in the stroma. was the most predominant type (10). A previous morphometric Neutrophil and eosinophil count was performed at random analysis of periglandular fibrosis in equine endometria using in 10 fields/biopsy, at the magnification of × 400. Due to the picrosirius red also referred a predominance of COL1 in severe inexistence of a classification method for jenny’s endometrium, endometriosis (13). the Kenney and Doig (14) grading system for the mare The main purpose of this work was to contribute to the endometrium was used. Thus, a healthy endometrium was understanding of the inflammatory processes and fibrosis in considered as category I; an endometrium with mild fibrotic jenny’s endometrium, classified according to Kenney and Doig’s lesions and inflammation as category IIA; when these lesions mare grading system (14). Therefore, the main goals of the become moderate as category IIB; or when they are severe as present study consisted of (i) quantification of transcripts of category III (14). collagen type I (COL1A2), type III (COL3A1), and IL33 in the endometrium; (ii) localization and quantification of COL1 Real-Time PCR for COL1A2, COL3A1, and and COL3 protein in histological sections; and (iii) assessment IL33 of eosinophil and neutrophil count and their correlation with Quantification of COL1A2, COL3A1, and IL33 transcripts was collagen and IL33 transcripts. In addition, localization of the accomplished by reverse transcriptase, followed by real-time COL protein in jenny’s endometrium was compared to mare polymerase chain reaction (qPCR), as previously described endometrium histological sections. (15). Briefly, after endometrial total RNA extraction with TRI Reagent R (T9424; Sigma) according to the manufacturer’s MATERIALS AND METHODS instructions, a NanoDrop system (ND200C; Fisher Scientific, Hampton, PA, USA) was used for RNA quantification. Animals and Sample Collection Visualization of 28S and 18S rRNA bands on a 1.5% agarose gel A total of 80 Catalonian jennies, between 3 and 22 years of age, confirmed RNA quality. Reverse transcription was performed were divided in eight jenny groups, which were brought together using M-MLV Reverse Transcriptase (M170B, Promega R ) from Frontiers in Veterinary Science | www.frontiersin.org 2 September 2020 | Volume 7 | Article 631 Miró et al. Jenny Endometritis and Endometrosis TABLE 1 | Primer sequences used in real-time PCR analysis. Gene (Accession Sequence 5′ -3′ Amplicon number) (base pairs) COL1A2 Forward: CAAGGGCATTAGGGGACACA 196 (NM_001323780.1) Reverse: ACCCACACTTCCATCGCTTC COL3A1 Forward: CAAAGGAGAGCCAGGAGCAC 98 (XM_014852914.1) Reverse: CTCCAGGCGAACCATCTTTG GAPDH Forward: CACCCACTCTTCCACCTTCG 173 (XM_014834961.1) Reverse: CTTGCTGGGTGATTGGTGGT IL33 Forward: CCCCAGCAAAGATGAACAGC 179 (XM_014860749.1) Reverse: TGGTACATGCCTTCTGGTGGT COL1A2—Equus asinus collagen type I alpha 2 chain; COL3A1—Equus asinus collagen, type III, alpha 1; GAPDH—Equus asinus glyceraldehyde-3-phosphate dehydrogenase; IL33—Equus asinus interleukin 33. 1 µg of total RNA in a 20-µL reaction volume using an oligo (dT) primer (C1101, Promega R ). Real-time PCR was used for the assessment of mRNA transcription of COL1A2, COL3A1, and IL-33. Glyceraldehyde- 3-phosphate dehydrogenase (GAPDH) was chosen as the reference gene [<2-fold changes between different biological conditions; (16)] from four potential validated genes (ribosomal protein L32, succinate dehydrogenase A complex, subunit A, flavoprotein, and beta-2-microglobulin). Internet-based program Primer-3 (17) and Primer Premier softwares (Premier Biosoft Interpairs, Palo Alto, CA, USA) were used to design the primers listed in Table 1. Target and reference gene amplification was performed on StepOnePlusTM Real-Time PCR System (Applied Biosystems R , California, USA) using Power SYBR Green PCR Master Mix (Applied Biosystems R ), after primer concentration optimization. Polymerase chain reaction (PCR) products were FIGURE 1 | Eosinophil (A) and neutrophil (B) count in different endometrial run on a 2.5% agarose gel to confirm specificity. Results were categories (IIA, IIB, or III) of the jenny. Bars represent mean ± SEM. Asterisk evaluated using Real-time PCR Miner 4.0 software. For gene indicates significant differences between endometrial categories (*P < 0.05). quantification, the average efficiency of each gene (E) and the average cyclic threshold (CT) of each sample were used, in the formula r0 = [1/(1 + E) CT] (18). Relative expression values RGB), with a pixel size of 0.441 µm/pixel (12). Collagen were calculated by normalizing the expression level of the target type one fibers appeared in red, while COL3 fibers stained genes against that of the reference gene. green. A qualitative evaluation of the endometrium sections was performed to enable the identification of the areas where each Picrosirius Red Staining type of collagen was present. Moreover, a quantitative analysis In order to assess the extent and localization of COL1 and COL3 was performed with ImageJ program (Version 1.52a, National in jenny’s endometrium, a histological section, other than the Institutes of Health, USA). one stained with hematoxylin and eosin for histopathological In order to better understand if the localization of collagen evaluation, was stained with Picrosirius Red Stain (ab#150681, fibers, namely, COL1 and COL3, in jenny’s endometrium was Abcam plc, Cambridge, UK), following the manufacturer’s specific to this species and could be the possible explanation instructions. Thus, a picrosirius red solution was placed onto for their longevity in reproduction as compared to the mare, deparaffinized endometrium sections and incubated for 1 h. endometrial histological sections from mares, graded according Afterward, slides were rinsed with acetic acid solution and to Kenney and Doig (14), were also stained with PSR and absolute alcohol. The slides were observed with an upright observed for a qualitative assessment. widefield microscope (Olympus BX51, Olympus Corporation, Tokyo, Japan) at 200× magnification (20×/0.75NA objective) Statistical Analysis under polarized light, and ten random images from each stained Data regarding the area in the endometrial biopsy occupied endometrium section were acquired with a camera (TIS 2MP by COL1 or COL3, and jenny’s age in relation to endometrial Frontiers in Veterinary Science | www.frontiersin.org 3 September 2020 | Volume 7 | Article 631 Miró et al. Jenny Endometritis and Endometrosis FIGURE 2 | Relative IL33 mRNA levels in different endometrial categories (IIA, IIB, or III) of the jenny. Bars represent mean ± SEM. category were statistically analyzed by one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. In order to compare the number of eosinophils or neutrophils with IL33, COL1, or COL3 transcripts between endometrial categories, a Kruskal–Wallis analysis followed by Dunn’s multiple-comparison test was used. Spearman rank correlation analysis between eosinophil or neutrophil count and IL33 mRNA levels and area occupied by COL1 or COL3 within each endometrial category was also performed. The software GraphPad Prism, Version 6.0, San Diego, CA, USA, was used to analyze these data. Significance was defined with the value of P < 0.05. Unless otherwise specified, the presented results are expressed as mean ± SEM. RESULTS Histological Classification FIGURE 3 | Relative mRNA levels (A) and proportion (B) of collagen type I From the 23 endometrial biopsies obtained from the jennies, (COL1) and type III (COL3) in jenny endometrium. Histologic sections were six endometria were graded as category IIA (12.5 ± 3.9 years stained with picrosirius red. Bars represent mean ± SEM. old), 11 as category IIB (13.8 ± 4.2 y), and six as category III (13.6 ± 2.2 y), according to Kenney and Doig’s mare endometrial classification (14). No age differences were found among endometrial categories (P > 0.05). As fibrosis increased, (data not shown). Nevertheless, a 4.9-fold increase in IL33 from categories IIB to III, a significant 14-fold decrease in the mRNA levels was noted from category IIA to category III number of eosinophils was depicted (Figure 1A; P < 0.05). endometrium, even though not significant (Figure 2). When Regarding neutrophil count, there was a fall from the category all endometrial categories were considered, no correlation was IIA endometrium to category IIB, and from categories IIA to III. found between eosinophil count and IL33 transcripts (data No correlation was found between age and any of the studied not shown). However, there was a tendency toward a positive factors (PMN, eosinophils, COL1, COL3, COL1, COL3, and IL33 correlation between eosinophil count and IL33 transcripts in transcripts) (Figure 1B; P > 0.05). category IIA endometrium (ρ = 0.7714; P = 0.055), mainly due to the effect of a single sample (one out of six) that clearly tipped Real-Time PCR the line toward significance (data not shown). The transcripts of IL-33, a cytokine involved in eosinophilia Neither transcripts of COL1A2 nor COL3A1 differed between and in chronicity and development of fibrosis (6), showed jenny’s endometrial categories (Figure 3A). However, a 4- no difference, as endometrial fibrosis advanced, from jenny’s fold drop in COL1A2 and a 3-fold fall in COL3A1 in endometria with little collagen fibers, such as category IIA, to the mRNA levels between category IIA and category III category III endometria, where fibrosis is visibly established were noted. Frontiers in Veterinary Science | www.frontiersin.org 4 September 2020 | Volume 7 | Article 631 Miró et al. Jenny Endometritis and Endometrosis FIGURE 4 | Jenny endometrium sections stained with picrosirius red and observed under polarized light microscopy. Collagen type I (COL1) stains red (arrow head), while collagen type III (COL3) stains green (white arrow). (A) Category IIA endometrium—a larger extent of COL3 is present in the stratum compactum under the surface epithelium and periglandular. Some COL1 fibers are present in the deeper stroma. (B) Category IIB endometrium—COL3 fibers are depicted under the basement membrane in the stratum compactum and periglandular, but COL3 fibers are also intertwined with some COL1 fibers. (C) Category III endometrium—COL3 fibers are still found under the basement membrane and periglandular, but more COL1 is present in the deep stroma, surrounding endometrial glands. Picrosirius Red Staining TABLE 2 | Qualitative evaluation of the presence of collagen type I (COL1) and The quantitative assessment of the different types of COL collagen type III (COL3) in jenny endometrium, stained with picrosirius red. in the jenny endometrium has shown no differences in the II A II B III area occupied by either COL1 or COL3 (Figure 3B). When considering the qualitative evaluation of collagen fibers in jenny’s Under basement membrane COL3 COL3 COL3 category IIA endometrium, green-stained COL3 was mostly Periglandular COL3 COL3/COL1 COL3/COL1 found under the basement membrane and in the stroma, Stroma COL3 COL3/COL1 COL1 ↑ surrounding endometrial glands. However, COL1 fibers, which Deep stroma COL1 COL1 COL1 ↑ stained red, were sparse and mainly present in the deeper stroma Around lymphatic COL1 (Figure 4, Table 2). In category IIB donkey endometrium, the Around blood vessels COL1 location of COL3 fibers was also under the basement membrane in the stratum compactum and periglandular, but COL3 fibers were also intertwined with some COL1 fibers. In the stroma, both and for category IIB (ρ = 0.6770; P = 0.03; Figure 6B) but COL types were found. When fibrosis was strongly established, not significantly in category III (Figure 6C). Between neutrophil such as in category III endometrium, COL3 fibers were still found count and COL3 area, a negative correlation was detected when under the basement membrane and periglandular, but more all endometrial categories were considered (ρ = −0.4716; P = COL1 was seen in the deep stroma, surrounding endometrial 0.03; Figure 6D) and in category IIB (ρ = −0.6102; P = 0.03; glands, lymphatic lacunae, and blood vessels, in comparison to Figure 6E). A tendency toward a negative correlation between the other Kenney and Doig’s endometrial categories (Figure 4, neutrophil count and COL3 area was also noticed in category III Table 2). When these data were matched to mare’s endometrial endometrium (ρ = −0.7537; P = 0.07; Figure 6F). localization of COL fibers, visible differences were found between species. In mare endometria graded as category I (not found DISCUSSION in the jennies in the present study), COL3 fibers were seen under the basement membrane and in the stratum compactum. To the best of our knowledge, this is the first study that has As endometriosis progressed, an increase in COL1 fibers was evaluated IL33 expression in donkey endometrium and related notorious under the basement membrane, which was not seen in its transcript levels with eosinophil count and collagen transcripts the jenny. Actually, in some category III endometria, only COL1 and deposition. Since IL33 is a critical regulator of a number of fibers were present (Figure 5, Table 3). processes including inflammation, vascularization, and fibrosis The correlation between eosinophil or neutrophil count and (19, 20), its role on the progression of endometrial inflammation the area occupied by COL1 or COL3 was also evaluated. and fibrosis in the jenny is worth evaluating. In fact, IL33 Although no correlation was found for eosinophil count (data not can exacerbate inflammation in collagen-induced arthritis, and shown), correlations between neutrophil count and both COL1 in allergic processes (21, 22). Based on studies on peritoneal and COL3 areas were observed. Positive correlations were seen fluid, serum, or samples from endometriotic lesions in women, for the endometrial area occupied by COL1 when all endometrial it has been suggested that IL33 might play an important role categories were considered (ρ = 0.520; P = 0.01; Figure 6A) toward inflammation present in advanced-stage endometriosis Frontiers in Veterinary Science | www.frontiersin.org 5 September 2020 | Volume 7 | Article 631 Miró et al. Jenny Endometritis and Endometrosis FIGURE 5 | Mare endometrium sections stained with picrosirius red and observed under polarized light microscopy. Collagen type I (COL1) stains red (arrowhead), while collagen type III (COL3) stains green (white arrow). (A) Category I endometrium—a larger extent of COL3 is present in the stratum compactum under the surface epithelium and periglandular. (B) Category IIA endometrium—COL3 fibers are depicted under the basement membrane in the stratum compactum and periglandular, but COL3 fibers are also intertwined with some COL1 fibers. (C) Category IIB endometrium—COL3 fibers are sparsely found under the basement membrane, while COL1 is predominant and present in the deep stroma, surrounding endometrial glands. (D) Category III endometrium—COL1 is found almost exclusively in the stratum compactum, periglandular, and under the basement membrane. TABLE 3 | Qualitative evaluation of the presence of collagen type I (COL1) and jenny endometrial samples, regardless of their histopathological collagen type III (COL3) in mare endometrium, stained with picrosirius red. classification. Even though not reaching significance, IL33 mRNA II A II B III levels in category III jenny endometrium were 5-fold increased, with respect to category IIA, while eosinophil count was about Under basement membrane COL3 COL3/COL1 COL1 14-fold lower in severe fibrosis than in healthier tissue (category Periglandular COL3 COL3/COL1 COL3/COL1 IIA). The development of fibrosis and its chronicity in liver, skin, Stroma COL3 COL3/COL1 COL1 ↑ and intestine has been related to eosinophil infiltration and the Deep stroma COL1 COL1 COL1 ↑ expression of IL33 (4–6). Since a direct relationship between Around lymphatic vessels COL1 tissue eosinophil infiltration and the expression of IL33 has been Around blood vessels COL1 referred, the present data on jenny endometria appears somewhat unexpected and cumbersome. In other species and tissues, accumulated observations have improved the understanding patients and on the progression of the disease (23, 24). Even of eosinophil stimulation and ability to infiltrate tissues and though endometriosis in women, which consists of the abnormal participate in disease pathogenesis (30, 31). Even though rarely growth of endometrial tissue outside the uterus (24, 25), and described, endometritis eosinophilica in the mare is a separate jenny’s endometrial fibrosis are different conditions, at a certain endometrial disease responsible for infertility (31). Nevertheless, stage of their development they both undergo an inflammatory deciphering the role of eosinophils in jenny’s endometrium process that ultimately may lead to fibrosis, as an end-point is still a challenge. It is plausible that jenny’s endometrium (25–29). We have shown that IL33 transcripts were detected in all with mild to moderate fibrosis (categories IIA and IIB) can Frontiers in Veterinary Science | www.frontiersin.org 6 September 2020 | Volume 7 | Article 631 Miró et al. Jenny Endometritis and Endometrosis FIGURE 6 | Correlation between neutrophil count and relative proportion of collagen type I (COL1—A–C) and type III (COL3—D–F) in all (A,D) and in IIB (B,E), and III (C,F) jenny endometrial categories. ρ—Spearman rank correlation. Frontiers in Veterinary Science | www.frontiersin.org 7 September 2020 | Volume 7 | Article 631 Miró et al. Jenny Endometritis and Endometrosis mount an inflammatory reaction characterized by the presence (38). Instead, the arrangement of fibroblasts around endometrial of eosinophils and neutrophils. This eosinophil infiltration might glands, in one or more layers, which produce the extracellular be mediated by IL33. In fact, our results appear to indicate a trend matrix proteins such as collagen IV, laminin, and fibronectin, toward a positive correlation between eosinophil count and IL33 should be taken into consideration (38). transcripts in category IIA endometrium. Unlike other cytokines, The connection between inflammation and fibrosis is well- when epithelial or endothelial cell damage or cell necrosis occurs, recognized (31). One putative link between neutrophil and IL33 is passively released, suggesting that it may function as an fibrogenesis has been provided by neutrophil extracellular traps alarmin that stimulates the immune system [reviewed by Chan (NETs, released by neutrophils) triggering the differentiation et al. (22)]. However, in severe fibrosis (category III), as in other of myofibroblasts, cells responsible for COL deposition (17). tissues where fibrosis is linked to impairment of organ function In the mare, NET components induced an in vitro increase (32), the endometrial cells may no longer be capable of mounting in COL1 in the endometrium (39). Likewise, in mares with an inflammatory reaction. Alternatively, eosinophil infiltration closed cervix, a marked endometritis and intense neutrophil subsides, and the tissue reacts with fibrosis establishment, which infiltration, and permanent pathological endometrial changes, could be stimulated by the still present pro-fibrotic cytokine IL33 including fibrosis were observed (40). In a recent study on (19), or by other pro-fibrotic stimuli (33). In spite of a decline in jenny’s endometrial cytology, an increase in neutrophil count in eosinophil infiltration in category III endometrium, the release older and multiparous jennies was referred (41). In the present of IL33 might be ascribed to other cells rather than eosinophils, study, the neutrophil count positive correlations with the area such as fibroblasts [reviewed by Chan et al. (22)]. Nevertheless, occupied by COL1, and the negative correlation with the area this should be further investigated in jenny’s endometrium. of COL3 observed in endometria with mild lesions, suggest that In the present study, as fibrosis progressed from categories the presence of neutrophils in jenny’s uterus may be involved in IIA to III, a 4-fold decrease in COL1A2 and a 3-fold decrease endometrial fibrogenesis. in COL3A1 in the mRNA levels were noted, although not In conclusion, there are important differences in the collagen significant. In addition, the percentage of areas occupied by distribution between jenny and mare endometria. The presence COL1 and COL3 fibers in jenny endometria was similar, of COL3 next to the surface epithelium, in the stratum despite endometrial histopathological characteristics, in contrast compactum, and around endometrial glands might not have a to what we have previously shown for the mare endometrium. deleterious effect on implantation and endometrial function, as A preliminary work on Masson’s trichrome-stained histological it appears to occur in mares with category III endometrium. sections of jenny endometrium classified as I, IIA, or IIB was not Further studies are needed to correlate these findings with jenny’s conclusive about putative differences in the total area occupied fertility and age Therefore, donkey endometrium classification by COL fibers (34). In contrast, in the mare, COL3 was the should be reconsidered. In addition, further studies are needed most predominant type of collagen in healthy endometrium, in to unravel the unknown role of neutrophils and eosinophils comparison to severe endometriosis where COL1 was mostly present in jenny’s endometrium on the reproductive performance found (10). of this species. Interestingly, when evaluating the topographic histological distribution of COL1 and COL3 in jenny’s endometrium, a different pattern was consistent, despite the histopathological DATA AVAILABILITY STATEMENT classification and the progress of fibrosis. Even in category III The raw data supporting the conclusions of this article will be endometrium, there was an area occupied by COL3 fibers under made available by the authors, without undue reservation. the basement membrane adjacent to the luminal epithelium, in the stratum compactum. In contrast, in the mare endometrium, COL3 fibers under the luminal epithelium were replaced by ETHICS STATEMENT COL1, as fibrogenesis progressed. In fact, the arrangement of collagen fibers in the mare endometrium and their characteristics Written informed consent was obtained from the owners for the were related to the evolution of the process (35, 36). Mare foaling participation of their animals in this study. rates do not overcome 11% in category III endometrium (37). However, this correlation remains unknown in donkeys. The AUTHOR CONTRIBUTIONS analysis of the Catalonian donkey stud book showed that only 3.3% of foaled jennies are older than 15 years of age1 . Moreover, JM and GF-D conceived the study, participated in the sample the generation interval index (average age of one jenny and its acquisition and analysis, helped conduct the data analysis, selected offspring) of this donkey breed is 7.75 ± 0.25 years1 . contributed to the writing of the original draft, and reviewing However, based on these data, it is impossible to calculate jenny’s and editing the manuscript. MG-R and JC participated in reproductive longevity or overall fertility rate, let alone for each the jenny control, sample acquisition, and process and data endometrial category. In addition, it has been referred that in curation. JS, CF, and GA-P performed the sample formal mare endometriotic tissue, the incidence of periglandular fibrosis analysis, data curation, and analysis. MR participated in sample should not be based on the increased presence of collagen fibers formal analysis, data curation, and analysis and writing of the original draft. All authors contributed to the article and approved 1 Rac.uab.cat/enllacos/racenlla.htm the submitted version. Frontiers in Veterinary Science | www.frontiersin.org 8 September 2020 | Volume 7 | Article 631 Miró et al. Jenny Endometritis and Endometrosis FUNDING ACKNOWLEDGMENTS This work was supported by a competitive grant from We would like to thank Mrs. Maria do Rosário Luís, from the the Fundação para a Ciência e Tecnologia (PTDC-CVT- Laboratory of Pathology (FMV) and Dr. Telmo Nunes from the REP-4202-2014) and by Project UIDP/CVT/00276/2020 Laboratory of Microscopy, for the valuable assistance on biopsies (funded by FCT). histological preparation and observation, respectively. REFERENCES 17. Chrysanthopoulou A, Mitroulis I, Apostolidou E, Arelaki S, Mikroulis D, Konstantinidis T, et al. Neutrophil extracellular traps promote 1. Vilés K, Rabanal R, Rodríguez-Prado M, Miró J. Effect of ketoprofen treatment differentiation and function of fibroblasts. J Pathol. (2014) 233:294–307. on the uterine inflammatory response after AI of jennies with frozen semen. doi: 10.1002/path.4359 Theriogenology. (2013) 79:1019–26. doi: 10.1016/j.theriogenology.2013. 18. Zhao S, Fernald RD. Comprehensive algorithm for quantitative real- 01.006 time polymerase chain reaction. J Comput Biol. (2005) 12:1047–64. 2. Rothenberg ME, Hogan SP. The eosinophil. Annu Rev Immunol. (2006) doi: 10.1089/cmb.2005.12.1047 24:147–74. doi: 10.1146/annurev.immunol.24.021605.090720 19. Gao Q, Li Y, Li M. The potential role of IL-33/ST2 signaling in 3. Gervasi MG, Osycka-Salut C, Caballero J, Vazquez-Levin M, Pereyra E, Billi fibrotic diseases. J Leukoc Biol. (2015) 98:15–22. doi: 10.1189/jlb. S, et al. Anandamide capacitates bull spermatozoa through CB1 and TRPV1 3RU0115-012R activation. PLoS ONE. (2011) 6:e16993. doi: 10.1371/journal.pone.0016993 20. Millar NL, O’Donnell C, McInnes IB, Brint E. Wounds that heal 4. Mchedlidze T, Waldner M, Zopf S, Walker J, Rankin AL, Schuchmann M, et and wounds that don’t – the role of the IL-33/ST2 pathway in al. Interleukin-33-dependent innate lymphoid cells mediate hepatic fibrosis. tissue repair and tumorigenesis. Semin Cell Dev Biol. (2017) 61:41–50. Immunity. (2013) 39:357–71. doi: 10.1016/j.immuni.2013.07.018 doi: 10.1016/j.semcdb.2016.08.007 5. Rankin AL, Mumm JB, Murphy E, Turner S, Yu N, McClanahan TK, et 21. Liew FY. IL-33: a janus cytokine. Ann Rheum Dis. (2012) 71:101–4. al. IL-33 induces IL-13–dependent cutaneous fibrosis. J Immunol. (2010) doi: 10.1136/annrheumdis-2011-200589 184:1526–35. doi: 10.4049/jimmunol.0903306 22. Chan BCL, Lam CWK, Tam LS, Wong CK. IL33: roles in allergic 6. Masterson JC, Capocelli KE, Hosford L, Biette K, McNamee EN, De Zoeten inflammation and therapeutic perspectives. Front Immunol. (2019)10:364. EF, et al. Eosinophils and IL-33 perpetuate chronic inflammation and fibrosis doi: 10.3389/fimmu.2019.00364 in a pediatric population with stricturing crohn’s ileitis. Inflamm Bowel Dis. 23. Santulli P, Borghese B, Chouzenoux S, Vaiman D, Borderie D, Streuli I, et al. (2015) 21:2429–40. doi: 10.1097/MIB.0000000000000512 Serum and peritoneal interleukin-33 levels are elevated in deeply infiltrating 7. Kenney RM. The aetiology, diagnosis and classification of chronic endometriosis. Hum Reprod. (2012) 27:2001–9. doi: 10.1093/humrep/ degenerative endometritis (CDE) (endometriosis). In: Allen WR, editor. des154 Equine Endometritis Worship. Proceedings of the John P. Hughes International 24. Miller JE, Monsanto SP, Ahn SH, Khalaj K, Fazleabas AT, Young SL, et Workshop on Equine Endometritis; 1992 Aug 19-20. Newmarket (1993). al. Interleukin-33 modulates inflammation in endometriosis. Sci Rep. (2017) doi: 10.1111/j.2042-3306.1993.tb02940.x 7:1–11. doi: 10.1038/s41598-017-18224-x 8. Hoffmann C, Ellenberger C, Mattos RC, Aupperle H, Dhein S, Stief B, et al. 25. Bulun SE. Endometriosis. N Engl J Med. (2009) 360:268–79. The equine endometrosis: new insights into the pathogenesis. Anim Reprod doi: 10.1056/NEJMra0804690 Sci. (2009) 111:261–78. doi: 10.1016/j.anireprosci.2008.03.019 26. Itoga T, Matsumoto T, Takeuchi H, Yamasaki S, Sasahara N, Hoshi T, et al. 9. Bochsler PN, Slauson DO. Inflammation and repair of tissue. In: Slauson DO, Fibrosis and smooth muscle metaplasia in rectovaginal endometriosis. Pathol Cooper BJ, editors. Mechanisms of Disease: A Textbook of Comparative General Int. (2003) 53:371–5. doi: 10.1046/j.1440-1827.2003.01483.x Pathology. St. Louis, MO: Mosby (2002). p. 140–245. 27. Vicino M, Scioscia M, Resta L, Marzullo A, Ceci O, Selvaggi LE. 10. Pinto-Bravo P, Reborddo MR, Amoral A, Fernandes C, Cuello C, Parrilla I, Fibrotic tissue in the endometrioma capsule: surgical and physiopathologic et al. Is mare endometrosis linked to oviduct fibrosis? Pferdeheilkunde. (2018) considerations from histologic findings. Fertil Steril. (2009) 91:1326–8. 34:43–6. doi: 10.21836/PEM20180107 doi: 10.1016/j.fertnstert.2008.02.157 11. Junqueira LCU, Cossermelli W, Brentani R. Differential staining of collagens 28. Sokkar SM, Hamouda MA, El-Rahman SM. Endometritis in she donkeys type I, II and III by sirius red and polarization microscopy. Arch Histol Jap. in Egypt. J Vet Med B Infect Dis Vet Public Health. (2001) 48:529–36. (1978) 41:267–74. doi: 10.1679/aohc1950.41.267 doi: 10.1046/j.1439-0450.2001.00457.x 12. Podico G, Canisso IF, Roady PJ, Austin SM, Carossino M, Balasuriya U, 29. Canisso IF, Panzani D, Miró J, Ellerbrock RE. Key aspects of donkey and et al. Uterine responses and equine chorionic gonadotropin concentrations mule reproduction. Vet Clin North Am Equine Pract. (2019) 35:607–42. after two intrauterine infusions with kerosene post early fetal loss in mares. doi: 10.1016/j.cveq.2019.08.014 Theriogenology. (2020) 147:202–10. doi: 10.1016/j.theriogenology.2019.11.014 30. Akuthota P, Weller PF. Eosinophils and disease pathogenesis. Semin Hematol. 13. Porto CD. Caracterização histoquímica do colágeno e expressão de MMP-2, (2012) 49:113–9. doi: 10.1053/j.seminhematol.2012.01.005 MMP-9 e TIMP-1 nas endometrites crônicas das éguas (Dissertation/Master’s 31. Grimm AL, Schoon H-A, Schöniger S. Histopathological features of thesis). Botucatu: University Estadual Paulista “Júlio de Mesquita Filho” endometritis eosinophilica in mares. Histol Histopathol. (2017) 32:1161–73. (2006). doi: 10.14670/HH-11-872 14. Kenney R, Doig PA. Equine endometrial biopsy. In: Morrow DA, editor. 32. Zeisberg M, Kalluri R. Cellular mechanisms of tissue fibrosis. 1. Common Current Therapy in Theriogenology. Philadelphia, PA: WB Saunders (1986). and organ-specific mechanisms associated with tissue fibrosis. J Physiol Cell p. 723–9. Physiol. (2013) 304:216–25. doi: 10.1152/ajpcell.00328.2012 15. Rebordão MR, Amaral A, Lukasik K, Szóstek-Mioduchowska A, Pinto- 33. Sziksz E, Pap D, Lippai R, Béres NJ, Fekete A, Szabó AJ, et al. Fibrosis Bravo P, Galvão A, et al. Impairment of the antifibrotic prostaglandin E related inflammatory mediators: role of the IL-10 cytokine family. Mediators 2 pathway may influence neutrophil extracellular traps–induced fibrosis Inflamm. (2015) 2015:764641. doi: 10.1155/2015/764641 in the mare endometrium. Domest Anim Endocrinol. (2019) 67:1–10. 34. Pires MA, Ferreira-Dias G, Catarino J, Bastos-de-Carvalho C, Silvestre doi: 10.1016/j.domaniend.2018.10.004 M, Nóvoa M, et al. Correlation between morphological characterization 16. Dheda K, Huggett JF, Bustin SA, Johnson MA, Rook G, Zumla A. Validation of jenny (Equus asinus) endometrial biopsy and quantification of of housekeeping genes for normalizing RNA expression in real-time PCR. collagen deposition by image analysis. J Comp Pathol. (2019) 166:139. Biotechniques. (2004) 37:112–9. doi: 10.2144/04371RR03 doi: 10.1016/j.jcpa.2018.10.125 Frontiers in Veterinary Science | www.frontiersin.org 9 September 2020 | Volume 7 | Article 631 Miró et al. Jenny Endometritis and Endometrosis 35. Masseno, APB. Assessment of endometrial fibrosis and myofibroblasts in 40. Reilas T, Rivera del Alamo MM, Liepina E, Yeste M, Katila T. Effects on the endometriosis active and inactive mares (Dissertation/Ph.D. thesis). Botucatu: equine endometrium of cervical occlusion after insemination. Theriogenology. University Estadual Paulista “Júlio de Mesquita Filho” (2012). (2016) 85:617–24. doi: 10.1016/j.theriogenology.2015.09.053 36. Costa, LD. Histochemical and immunohistochemical characterization of fibrotic 41. Quartuccio M, Cristarella S, Medica P, Fazio E, Mazzullo G, Rifici C, et changes in mares endometrosis (Dissertation/Master’s thesis). Botucatu: al. Endometrial cytology during the different phases of the estrous cycle in University Estadual Paulista “Júlio de Mesquita Filho” (2015). jennies: new evidences. Animals. (2020) 10:1062. doi: 10.3390/ani10061062 37. Kenney RM. Cyclic and pathologic changes of the mare endometrium as detected by biopsy, with a note on early embryonic death. J Am Vet Med Assoc. Conflict of Interest: The authors declare that the research was conducted in the (1978) 172:241–62. absence of any commercial or financial relationships that could be construed as a 38. Walter I, Handler J, Reifinger M, Aurich, C. Association of endometriosis potential conflict of interest. in horses with differentiation of periglandular myofibroblasts and changes of extracellular matrix proteins. Reproduction. (2001) 121:581–6. Copyright © 2020 Miró, Gutiérrez-Reinoso, Silva, Fernandes, Rebordão, Alexandre- doi: 10.1530/reprod/121.4.581 Pires, Catalán and Ferreira-Dias. This is an open-access article distributed under the 39. Rebordão MR, Amaral A, Lukasik K, Szóstek-Mioduchowska A, Pinto- terms of the Creative Commons Attribution License (CC BY). The use, distribution Bravo P, Galvão A, et al. Constituents of neutrophil extracellular or reproduction in other forums is permitted, provided the original author(s) and traps induce in vitro collagen formation in mare endometrium. the copyright owner(s) are credited and that the original publication in this journal Theriogenology. (2018) 113:8–18. doi: 10.1016/j.theriogenology.2018. is cited, in accordance with accepted academic practice. No use, distribution or 02.001 reproduction is permitted which does not comply with these terms. Frontiers in Veterinary Science | www.frontiersin.org 10 September 2020 | Volume 7 | Article 631