Low Concentration of Anti-Müllerian Hormone in Mares with Delayed Uterine Clearance

Luteal dysfunction has been observed in mares with defective uterine clearance. Association of low ovarian reserve with luteal dysfunction and abnormal endometrial thickness has been reported in bovine. Anti-Müllerian hormone (AMH) has been indicated as a marker for ovarian reserve in bovine and originates primarily from the ovary in equine. The present study evaluated serum AMH concentration in mares with delayed uterine clearance versus that in mares without delayed uterine clearance. Of 49 mares assigned to the study, 12 and 37 mares were diagnosed with and without delayed uterine clearance, respectively. Delayed uterine clearance was determined based on history and the observation of intrauterine fluid in ultrasonographic examination 24 hours after natural breeding. Serum AMH concentration was measured during estrus. Concentration of AMH was lower in mares with delayed uterine clearance (0.4 ± 0.1 ng/mL) than in those without delayed uterine clearance (1.1 ± 0.1 ng/mL). In conclusion, the present study indicated possible associations between ovary-lined mechanisms and uterine clearance failure in mares.     

犬正常发情期与患子宫蓄脓的子宫和卵巢组织结构及乳铁蛋白表达的对比

本研究旨在探明犬子宫和卵巢在正常发情期与患子宫蓄脓时，其组织结构及乳铁蛋白（lactoferrin，LF）表达的变化特征。应用Masson’s、VVG、PAS组织化学染色方法观察乏情期、发情期、患子宫蓄脓时犬子宫及卵巢的组织结构特点，用免疫组织化学SP法观察LF的分布特征。结果显示，正常发情期犬：乏情期、发情期子宫内膜肌层厚度比分别为0.762 0、0.924 3；乏情期子宫固有层中胶原纤维含量大于发情期，发情期卵巢中胶原纤维含量大于乏情期；子宫血管层及卵巢血管内弹性膜清晰完整；子宫腺在乏情期时浅层管腔小，深层较大、腺管上皮为单层柱状上皮、上皮细胞及管腔内PAS阳性反应较强，发情期子宫腺管腔变大、腺管上皮为单层立方上皮、上皮细胞及管腔内有PAS阳性反应；子宫黏膜上皮在乏情期和发情期均为单层柱状上皮，但胞核位置不同，乏情期胞核位于中央，发情期胞核位于顶部。患子宫蓄脓犬：子宫内膜肌层厚度比为1.615 0；子宫固有层和卵巢中胶原纤维含量少于正常发情期；子宫腺管腔大，且形状不规则，管腔内有炎性细胞浸润，腺管上皮为单层立方上皮，有淋巴细胞位于基膜，上皮细胞及管腔内PAS阳性反应较弱；子宫血管层及卵巢血管内弹性膜较正常发情期变薄，且有断裂现象；子宫黏膜上皮为单层柱状上皮，胞核位于基底。LF在乏情期子宫腺上皮和卵巢中的表达水平高于发情期，而在子宫黏膜上皮中发情期的表达水平高于乏情期。患子宫蓄脓时犬子宫和卵巢中LF的表达水平均较低。综上表明，犬正常发情期与患子宫蓄脓时的子宫和卵巢组织结构特点显著不同，LF的表达水平也存在差异。     

Degenerative Endometrial Changes do not Change the Functional Capacity of Immigrating Uterine Neutrophils in Mares

An endometritis model was used to investigate the influence of degenerative endometrial changes (endometrosis) on functional parameters of uterine neutrophils in the horse. Six hours after intrauterine application of recombinant human interleukin‐8 (rhIL‐8), the uteri of 15 mares were flushed with phosphate‐buffered saline. Quantitative and qualitative flow cytometric assays were then made to determine the absolute numbers, viability, phenotype, generation of reactive oxygen species (ROS), and phagocytic activity of immigrated polymorphonuclear neutrophilic granulocytes (PMN). Recombinant hIL‐8 attracted similarly high numbers of similarly viable PMN into the uteri of mares with or without degenerative endometrial changes. Compared with blood PMN, immigrated uterine neutrophils displayed significantly upregulated expression of CD11a/CD18 (LFA‐1) on uterine PMN whereas major histocompatibility complex class I molecules were expressed at lower densities. The ability to phagocytose opsonized streptococci did not differ between uterine and blood PMN. However, uterine PMN displayed a higher capacity to generate ROS. On average, uterine PMN of mares with degenerative endometrial changes showed phenotypical and functional characteristics similar to those of mares with a histologically healthy endometrium. Therefore, degenerative endometrial changes per se did not reduce the functional capacity of equine uterine neutrophils in mares.     

绵羊发情周期子宫内膜微血管密度与VEGF的表达关系

以绵羊发情周期的子宫内膜为研究对象,采用免疫组化方法定量检测绵羊发情周期子宫内膜的微血管密度（MVD）和血管内皮生长因子（VEGF）表达量。结果表明：MVD的标记物CD34和VEGF在绵羊发情周期的子宫内膜中呈现相同的表达特征,即表达位点均在子宫内膜上皮固有层及肌层;两者均在发情后0d开始表达,5d最高。5d开始到15d表达量缓慢下降。子宫内膜VEGF表达量和MVD相关系数r=0.669,P=0,表明子宫角中VEGF表达量和MVD显著正相关。     

Physiological and pathological expression of intermediate filaments in the equine endometrium

The aim of this study was to investigate the expression of the intermediate filaments cytokeratin, vimentin and desmin in the equine endometrium by immunohistological techniques. For this purpose, endometrial biopsies of 151 mares were examined to determine physiological cycle patterns and changes resulting from endometriosis. During the physiological cycle epithelial cells and mesenchymal cells express cytokeratin and vimentin, respectively, whilst desmin and vimentin were coexpressed by the smooth muscle cells. Epithelial coexpression of cytokeratin and vimentin was seen in numerous fibrotic glands and in the uterine glands of three mares with pathologically inactive endometria. Three different staining patterns (basal, perinuclear, diffuse) of vimentin were associated with typical morphological alterations of the affected epithelia. In addition, in 14 cases a stromal coexpression of vimentin and desmin was found, indicating an atypical stromal differentiation in inactive endometria of older mares, barren for several years.     

Histological Changes in the Uterus during Postpartum in the Mare

An histological study of the postpartum period in 29 mares was carried out. Uterine biopsies were taken daily during the first 10 days postpartum in a total of 87 samples. At day 0, equine endometrium was characterized in the surface by the presence of regularly ordered microcaruncles; the stratum spongiosum was oedematous and contained distended and scarce glands. Degenerative changes in microcaruncles and endometrial glands were present on day 1 postpartum. The epithelium of the microcaruncles from 2 to 5 days postpartum showed cytoplasmic vacuolization, karyorrhexis and an inflammatory reaction with neutrophils and phagocitic cells. On day 7 postpartum, the histology of the endometrium was similar to the normal proestrus with cuboidal luminal epithelium and an oedematous stromal tissue. The changes are usually completed within days 9 and 10 postpartum with the histologically typical appearance of estrus in mares.     

Uterine clearance mechanisms during the early postovulatory period in mares

Uterine response to inoculation with Streptococcus zooepidemicus organisms, 51Cr-labeled 15-microns microspheres, and charcoal was evaluated in 9 mares (4 resistant and 5 susceptible to endometritis) to determine mechanical and cellular clearance rates during the early postovulatory period. Mares were inoculated at estrus prior to ovulation during estrous cycles 1, 3, and 5. Uterine swab specimens for aerobic and anaerobic bacteriologic culture and serum for progesterone determination were obtained on postovulation day 3 during estrous cycle 1, on the day of ovulation during estrous cycle 3, and on postovulation day 5 during estrous cycle 5. Immediately thereafter, the uterus was irrigated with 50 ml of sterile physiologic saline solution containing tracer amounts of 125I-labeled human serum albumin. Streptococcus zooepidemicus was isolated from 10 of 15 (67%) uterine specimens collected from susceptible mares and incubated aerobically. Escherichia coli also was isolated from 2 of the 10 specimens incubated aerobically. Anaerobic bacteriologic culture of specimens from all mares yielded no growth. Chromium-labeled microspheres were recovered twice from 2 susceptible mares, on day 0 and day 5. Charcoal was retained in 5 specimens collected from 3 susceptible mares. Bacteriologic culture of specimens from resistant mares did not yield growth. On day 0, chromium-labeled microspheres and charcoal were recovered once from 1 resistant mare. Mares susceptible to endometritis accumulated more fluid within the uterine lumen after ovulation than did resistant mares (mean +/- SEM, 52.73 +/- 15.22 ml and 7.41 +/- 1.96 ml, respectively; P less than 0.01). From this study, it appeared that uterine cellular and bactericidal mechanisms are dysfunctional during the early postovulatory period. However, there appeared to be no disruption of the mechanisms responsible for mechanical clearance of materials inoculated in the uterus.     

Microvascular development and growth of uterine tissue during the estrous cycle in mares

OBJECTIVE: To document uterine growth and microvascular development in the endometrium of uteri with differing degrees of fibrosis as well as uterine growth throughout the estrous cycle of mares. ANIMALS: 30 mares. PROCEDURE: Uterine tissue was obtained during the breeding season from a slaughter facility. Stage of estrous cycle of the mares was assessed on the basis of ovarian structures and plasma progesterone concentrations. Endometrium was characterized by use of light microscopy, and blood vessel walls were marked by histochemical techniques. Microvascular development was evaluated by a computerized image analysis system. Growth of uterine tissue was based on cellular content of DNA and RNA, RNA:DNA, and protein:DNA. RESULTS: Significant differences in vascular density were not observed in the endometrium of uteri obtained from mares euthanatized during the follicular or luteal phase of the estrous cycle, regardless of whether endometrial classification of degree of fibrosis was considered. There was a 3-fold increase in amount of DNA and RNA of endometrial cells in the follicular phase when compared to myometrium. Hypertrophy of endometrial tissue during the luteal phase was reflected by a significant increase in cell protein content and protein:DNA. CONCLUSIONS AND CLINICAL RELEVANCE: Endometrial growth of vascular tissues during the estrous cycle may be coordinated with development of nonvascular tissue. Estrogen and progesterone may play a role in regulation of uterine growth and angiogenesis.     

Expression of the uterine Mx protein in cyclic and pregnant cows,gilts, and mares

Pregnancy and interferon-tau (IFN tau) upregulate uterine Mx gene expression in ewes; however, the only known role for Mx is in the immune response to viral infection. We hypothesize that Mx functions as a conceptus-induced component of the anti-luteolytic mechanism and/or regulator of endometrial secretion or uterine remodeling during early pregnancy. This study was conducted to determine the effects of early pregnancy on uterine Mx expression in domestic farm species with varied mechanisms of pregnancy recognition. Endometrium from cows, gilts, and mares was collected during the first 20 d of the estrous cycle or pregnancy, and total messenger RNA (mRNA) and protein were analyzed for steady-state levels of Mx mRNA and protein. Northern blot analysis of Mx mRNA detected an approximately 2.5 Kb of mRNA in endometrium from each species. In pregnant cows, steady-state levels of Mx mRNA increased 10-fold (P < 0.05) above levels observed in cyclic cows by d 15 to 18. In cyclic gilts, slot blot analysis indicated that endometrial Mx mRNA levels did not change between d 5 and 18 of the cycle. However, in pregnant gilts, Mx levels tended (P = 0.06) to be elevated two-fold on d 16 only, and in situ hybridization indicated that this increase occurred in the stroma. In mares, Mx mRNA was low, but detectable, and did not change between ovulation (d 0) and d 20, regardless of reproductive status. Western blot analysis revealed multiple immunoreactive Mx protein bands in each species. One band was specific to pregnancy in cows. As in ewes, in situ hybridization analysis indicated that Mx mRNA was strongly expressed in the luminal epithelium, stroma, and myometrium by d 18 in cows. However, on d 14 in gilts, Mx was expressed primarily in the stroma, and on d 14 in mares, low levels of Mx expression were confined largely to the luminal epithelium. The uteruses of cows, gilts, and mares express Mx, and expression is upregulated during pregnancy in cows and gilts--animals whose conceptuses secrete interferons during early pregnancy, but that possess different mechanisms for pregnancy recognition.     

SCINTIGRAPHIC MEASUREMENT OF UTERINE CLEARANCE IN MARES

Uterine clearance of technetium 99m-albumin colloid (^99mTc-μAA) was qualitatively and quantitatively measured in 5 reproductively normal mares and 5 mares susceptible to endometritis (infertile). The percentage of 370 MBq ^99mTc-μAA cleared from the uterine lumen within 2 hr of intrauterine infusion was measured in 10 mares on day 3 of estrus and 48 hr after ovulation. The procedure was repeated 3 times on day 3 of estrus in 6 mares to determine repeatability. Six mares were infused with 1110 MBq ^99mTc-μAA on day 3 of estrus to evaluate the effect of increasing the dose to reduce the imaging time. There was no statistically significant difference in the mean percentage of radiocolloid cleared from the uterus during day 3 of estrus or 48 hr after ovulation or in the percent cleared when the studies were repeated in individual mares. There was no statistically significant difference in uterine clearance between the 370 and 1110 MBq dose studies in each mare from 15 to 120 min. Reproductively normal mares cleared approximately 50% of the radiocolloid from the uterus by 120 min while susceptible mares cleared less than 15%.     

Investigations on the Endometrial Response to Intrauterine Administration of N‐Acetylcysteine in Oestrous Mares

In mares, mating‐induced persistent endometritis contributes to low fertility. The condition is in part related to delayed clearance of mucus accumulated within the uterine lumen. The objective of this study was to investigate the endometrial response of healthy mares to intrauterine (i.u.) treatment with N‐acetylcysteine (NAC). Oestrous mares (n = 12) were randomly assigned to a treatment (TM) or control (C) group and received an i.u. infusion of 5% NAC and saline (total volume 140 ml), respectively. Endometrial biopsies were collected in five of the mares 24 h after treatment, in the remaining seven mares 72 h after treatment. Endometrial biopsies were evaluated for integrity of the luminal epithelium, number of polymorphonuclear neutrophils (PMN), staining for cyclooxygenase 2 (COX2), staining with Kiel 67 antigen (Ki‐67), lectins and periodic acid‐Schiff (PAS). The integrity of endometrial epithelial cells was not affected by treatment (no statistical differences between groups or times). At 24 h after treatment, the mean number of PMN in endometrial biopsies from NAC‐ and C‐mares did not differ, but at 72 h after treatment, number of PMN was significantly higher (p < 0.05) in C (3.9 ± 0.6 PMN/field) compared with NAC‐treated mares (2.3 ± 0.2 PMN/field). At 72 h after treatment, the intensity of staining for COX2 was significantly higher after saline than after NAC treatment (p < 0.05). In the epithelium, no differences in staining for the proliferation marker Ki‐67 were seen with respect to time and treatment. Score for the lectin wheat germ agglutinin (WGA) was slightly higher in NAC‐treated mares than in C‐mares 72 h after treatment (p < 0.05). Score for PAS staining of mucus in deep uterine glands differed significantly between groups at 24 h after treatment (p < 0.05). The present study demonstrates that NAC does not adversely affect the endometrial function. Moreover, an anti‐inflammatory effect on the equine endometrium was observed.     

Lectin binding patterns of uterine glands in mares with chronic endometrial degeneration

OBJECTIVE: To evaluate changes of glycoconjugate in uterine glands of endometrial tissues obtained from mares. ANIMALS: adult mares. PROCEDURE: Uterine biopsy samples were collected during the breeding season and analyzed histologically for signs of chronic endometrial degeneration. Stage of the estrous cycle was established, using clinical examination and determination of hormonal status. Uterine tissue samples were analyzed, using lectin histochemical and immunohistochemical techniques (estrogen and progesterone receptors). Connective tissues were stained to determine alterations of ground substance in periglandular fibrosis. RESULTS: Of 50 mares, 30 (60%) were classified as normal or having modest alterations, and 20 (40%) were classified as having moderate or severe endometrial degeneration. In normal equine endometrium, several lectins (Helix pomatia agglutinin, Lotus tetragonolobus agglutinin, Ricinus communis I agglutinin, Ulex europaeus agglutinin, and wheat germ agglutinin) bound to glycoconjugates of the luminal epithelium and openings of uterine glands. Lectin binding patterns of cystic dilated glands or fibrotic glands in endometrial samples were remarkably strong, whereas normal surrounding cells remained unstained. Lotus tetragonolobus lectin was not suitable for detecting endometrial alterations. Connective tissues stained with Alcian blue and results of Hale colloidal-iron binding revealed acidic ground substance in periglandular fibrosis. Estrogen and progesterone receptors were evenly distributed in healthy and affected endometrial samples. CONCLUSIONS AND CLINICAL RELEVANCE: Glycoconjugate patterns of uterine glands were altered in mares with chronic endometrial degeneration. Therefore, uterine secretions are likely to be altered. These changes are not induced by changes in content of estrogen and progesterone receptors in endometrial tissues.     

Treatment of Persistent Mating‐Induced Endometritis in Mares with the Non‐Steroid Anti‐Inflammatory Drug Vedaprofen

Recently, successful treatment of mares with a history of persistent mating‐induced endometritis (PMIE) with dexamethasone has been reported. As systemic treatment of horses with glucocorticoids should be handled with caution, we tested the hypothesis that treatment with the non‐steroid anti‐inflammatory drug (NSAID) vedaprofen, an inhibitor of cyclooxygenase‐2, may have comparative, positive effects on fertility. Barren mares with a history of repeated PMIE were treated with vedaprofen (n = 8; initially 2 mg/kg bodyweight followed by 1 mg/kg orally twice daily) from 1 day before the first insemination to 1 day after ovulation or left untreated (n = 9). All mares received oxytocin (20 I.E. s.c.) thrice daily. Uterine swabs were collected for bacteriology and cytology. The day after ovulation, fluid accumulation was detected in three control mares and four treated mares (n.s.). The percentage of neutrophils in uterine cytology was significantly increased in comparison to the day before ovulation irrespective of treatment. Pregnancy was confirmed in two of nine mares in the control group and seven of eight mares in the treatment group (p < 0.05). NSAIDs may positively affect fertility in mares with a history of PMIE.     

Influence of age and parity on the distribution of cells expressing major histocompatibility complex class II, CD4, or CD8 molecules in the endometrium of mares during estrus

OBJECTIVE: To evaluate effect of age and parity on distribution and number of cells expressing major histocompatibility complex (MHC) class II, CD4, or CD8 molecules in the endometrium of mares during estrus. ANIMALS: 32 gynecologically healthy mares, categorized as young (3 to 8 years; n = 17) or old (9 to 16 years; 15) and nulliparous (n = 6), nulliparous embryo donors (16), or parous (10). PROCEDURES: Endometrial specimens collected from the uterine body and horns during estrus were stained by use of the avidin-biotin-peroxidase method, using monoclonal antibodies against equine MHC class II, CD4, and CD8 molecules. Labeled cells in the stratum compactum within 5 randomly selected fields at 400x magnification (total area = 0.31 mm2) were counted, and numbers were compared among groups and between locations. RESULTS: Age did not affect cell numbers within the 3 cell subsets examined. Numbers in each subset were higher in the uterine body than in the horns, although the difference was not significant for cells expressing MHC class II. Significantly more cells expressing MHC class II molecules were detected in the uterine body of nulliparous and parous mares than in embryo donors, whereas in the horns, these cells were significantly higher in number only in parous mares. Parity did not affect number of CD4+ or CD8+ cells. CONCLUSIONS AND CLINICAL RELEVANCE: The increased likelihood for endometritis to develop in mares as they age cannot be explained by a decrease in number of cells expressing MHC class II, CD4, or CD8 molecules within the endometrium. However, greater number of cells within these 3 subsets detected in the uterine body, compared with the horns, during estrus suggests a local readiness to act against microorganisms or semen introduced during mating or insemination.     

Endometrial nitric oxide synthase activity in mares susceptible or resistant to persistent breeding‐induced endometritis and the effect of a specific iNOS inhibitor in vitro

Emerging research suggests that the nitric oxide system may play a role in persistent breeding‐induced endometritis (PBIE) in the mare. Differences in uterine nitric oxide (NO) levels between mares susceptible or resistant to PBIE and a dose‐dependent inhibitory effect of NO on uterine contractility have been demonstrated. The objectives of this study were to investigate the difference in total nitric oxide synthase (NOS) activity of the endometrium between susceptible and resistant mares and the effect of a specific inducible nitric oxide synthase (iNOS) inhibitor on the endometrial NOS activity in vitro. Six susceptible and six resistant mares were selected based on preset criteria and the results of an intrauterine challenge with killed spermatozoa during oestrus. Endometrial biopsy samples were collected 24 hr post‐challenge and cultured at 37°C for 24 hr in L‐arginine supplemented minimum essential medium with or without a specific iNOS inhibitor (1,400 W dihydrochloride, 1 mM). The medium and the cultured endometrial tissue were collected after 24 hr of culture and assayed for NO and total protein, respectively. Total NO content of the medium, normalized to endometrial tissue wet weight or total protein, was used as a measure of endometrial NOS activity. Non‐parametric tests were applied for statistical analysis. Susceptible mares had significantly greater endometrial NOS activity than resistant mares. The iNOS inhibitor treatment significantly reduced NOS activity in endometrial samples derived from susceptible and resistant mares. These findings provide a basis for in vivo testing of specific iNOS inhibitors as preventative or therapeutic options for PBIE in mares.     

Making sense of equine uterine infections: the many faces of physical clearance

Equine uterine infections inflict major losses on the equine industry. Persistent inflammation of the oviduct and uterus leads to loss of the conceptus and mares susceptible to infection have weakened uterine defences partly due to retention of inflammatory exudate. Bacteria may trigger inflammation, resist phagocytosis, or adhere to the endometrium and types of infection range from genital commensals in susceptible mares to reproductive pathogens in normal mares. Uterine infections are diagnosed by history, detection of uterine inflammation, and isolation of typical organisms and susceptible mares may be identified by detection of intrauterine fluid during oestrus, or at 6-48 h post-breeding. Therapy includes oxytocin, uterine lavage, antibiotics, and prostaglandin analogues and clinical studies indicate additive benefits of oxytocin and antibiotics. Improved conception rates have been associated with autologous, intrauterine plasma, despite controversy about its bactericidal efficacy. Because of the potential for endometrial damage, intrauterine antiseptics require caution.     

Opsonins of Streptococcus in uterine flushings of mares susceptible and resistant to endometritis: control of secretion and partial characterization

The release of opsonins into the uterine lumen of mares susceptible or resistant to endometritis was examined after intrauterine inoculation of a filtrate of Streptococcus culture fluid or vehicle. Uterine flushings were collected at 0.5 hour before and 2, 4, 6, 8, and 24 hours after inoculation on day 2 or 3 of estrus and on day 7 or 8 after ovulation. Amounts of opsonins in flushings were quantified as the H2O2 produced by leukocytes incubated with flushings-opsonized bacteria, compared with H2O2 produced by leukocytes incubated with nonopsonized bacteria. Opsonin values in flushings increased (P less than 0.025) in all mares after inoculation of filtrate or vehicle. For mares resistant to endometritis, opsonin values were greater at diestrus than at estrus. The opposite was true for mares susceptible to endometritis, resulting in a status (susceptible vs resistant) X stage of cycle interaction (P less than 0.025). Overall, opsonins were higher (P less than 0.05) in flushings of mares susceptible to endometritis than in flushings of mares resistant to endometritis, but this difference was only apparent at estrus. Preliminary characterization of opsonins in uterine secretions by ammonium sulfate fractionation and gel filtration indicated that opsonins were mainly associated with an ammonium sulfate-soluble fraction of high molecular weight (greater than 4 X 10(6] and an ammonium sulfate-precipitable fraction that was associated with immunoglobulin G.     

Cellular and humoral defence mechanisms in mares susceptible and resistant to persistent endometritis

Both random and directional migration of blood neutrophils from 9 mares susceptible to persistent endometritis were significantly less (p less than 0.05) than neutrophils from 8 resistant mares. Serum from susceptible mares had significantly more (p less than 0.01) chemotactic activity than serum from resistant mares. Although phagocytosis of yeast blastospores by blood neutrophils from 4 resistant and 3 susceptible mares was similar, uterine neutrophils from susceptible mares were significantly worse (p less than 0.01) at phagocytosis than uterine neutrophils from resistant mares. Uterine washings from 17 susceptible mares were significantly better at opsonising yeast blastospores than washings from 14 resistant mares; however, washings from both groups had a similar ability to promote killing of S. zooepidemicus by neutrophils. When an immunologically non-specific endometritis was induced, washings from 3 susceptible mares were significantly worse at promoting bactericidal activity by 144 h than washings from 4 resistant mares (p less than 0.01). Haemolytic complement activity was significantly greater (p less than 0.001) in washings from 17 susceptible mares than from 14 resistant mares. Induction of acute endometritis resulted in high levels of haemolytic complement activity in 2 of 3 susceptible mares at 24 and 144 h, but only in small increases in 4 resistant mares. Thus, some abnormalities in neutrophil function were detected and a possible defect in promotion of neutrophil bactericidal activity by uterine secretions from susceptible mares but there was no evidence for any deficiency in haemolytic complement activity.