Materials and Methods

Animals—The postpartum periods of 9 lactating mares (8 Shetland ponies and 1 Haflinger), between 4 and 18 years of age, were studied after uncomplicated delivery. Two mares foaled for the first time, and the others were pluriparous. Mares were kept in single loose boxes on straw and were fed hay 3 times daily. Mineral supplements and water were always available. All mares were kept in the same facilities, and nutrition did not differ among them. The study was approved by the Committee for Animal Experimentation of the Austrian Ministry for Education and Science (license No. BMBWK-68.205/0016-BrGT/2007).

Experimental procedures—To verify the onset of cyclic ovarian activity, 10 mL of blood from a jugular vein was obtained daily in EDTA tubes from postpartum days 1 to 20 for determination of plasma progesterone concentrations. The last day with a plasma progesterone concentration < 1 ng/mL was defined as the day of ovulation.^22–24 Samples were centrifuged at 20°C and 900 × g for 10 minutes, and plasma was frozen at −20°C until analysis. Progesterone was analyzed as described.²⁵ Mares were checked also for estrous behavior with a pony stallion once daily, starting on postpartum day 4.

The mares were examined within 24 hours (day 1) and on postpartum days 9 and 16 via rectal and vaginal examination, including transrectal ultrasonography. The uterus was evaluated for dimension and symmetry and the ovaries for dimension and the presence of follicles and corpora lutea. On days 1, 9, and 16, tissue was obtained transcervically from randomly selected locations of the uterine horns, by use of routine techniques.²⁶ One half of the biopsy specimen was frozen at −80°C for qPCR analysis, and the other half was prepared for histologic examination. In addition, on days 1, 9, and 16, material for bacteriologic culture was obtained with a guarded uterine culture swab^a from the uterine lumen. Uterine swabs were placed in transport medium, stored at 5°C, and used for culture within 24 hours, as described.²⁷

Histologic and immunohistochemical evaluation—Specimens for histologic examination were fixed in 4% formalin and embedded in paraffin. Sections were cut at a thickness of 5 μm and stained with H&E.²⁸ Luminal epithelium, microcaruncles, glands, and stroma of the stratum compactum and spongiosum were evaluated in each sample. The amounts of neutrophils, lymphocytes, macrophages, and siderophages were classified as absent (–), mild (+, < 25% of cells), moderate (++, 25% to 50% of cells), or strong (+++, > 50% of cells). Fibrosis, hypertrophy, and cystic distension of uterine glands and their openings were recorded.

Lectin histochemical analysis was performed as described.^17,26 Paraffin sections were cut at a thickness of 3 μm and rehydrated, and endogenous peroxidase activity was blocked by incubation in 0.6% H₂O₂ in methanol for 15 minutes at 20°C. To minimize nonspecific lectin binding, preincubation in 1% bovine serum albumin^b in PBS solution at 20°C was performed for 30 minutes. Subsequently, slides were incubated for 1 hour with the respective biotinylated lectin (GSL,^c RCA,^d UEA,^e WGA,^f or HPA^g) at a concentration of 10 μg/mL. After incubation, sections were washed in PBS solution and incubated with avidin-biotin-peroxidase complex^h according to the manufacturer's instructions, washed, and developed in 3,3′diaminobenzidine-tetrahydrochloride substrateⁱ for 10 minutes. Slides were washed with distilled water, counterstained with hemalum, dehydrated, and mounted by use of xylene-soluble medium.^j The intensity of staining of the glycocalyx, classified as absent (–), mild (+), moderate (++), or strong (+++), was evaluated in luminal epithelium, microcaruncles, and uterine glands as described.²⁹ Staining of the Golgi region and cytoplasm of the glandular and luminal epithelium as well as the glandular contents, lamina propria, and blood vessels was described.

The expressions of Ki-67, lysozyme, and caspase 3 were determined via immunohistochemical analysis. Sections for immunohistochemical analysis were cut at a thickness of 3 μm. Sections were rehydrated, and endogenous peroxidase activity was blocked by incubation in 0.6% H₂O₂ in methanol for 15 minutes at 20°C. Goat serum^k (75 μL/5mL of PBS solution) was used to block nonspecific binding of proteins. Sections were incubated with the primary antibody (polyclonal anti-rabbit lysozyme^l: dilution, 1:150; anti-mouse Ki-67/MM1^m: dilution, 1:200; polyclonal anti-rabbit caspase 3ⁿ: dilution, 1:4,000) overnight at 4°C. Antigen retrieval by heating slides in 0.01M citrate buffer (pH, 6.0) in a microwave 4 times for 5 minutes was necessary for caspase 3 and Ki-67 immunostaining. After rinsing with PBS solution, the slides were incubated with the secondary antibody^o for 30 minutes. Subsequently, sections were washed in PBS solution and developed in diaminobenzidine substrate for 10 minutes. Slides were washed with distilled water, counterstained with hemalum, dehydrated, and mounted with a xylene-soluble mounting medium.^j

Caspase 3- and lysozyme-positive cells were counted at 400× magnification in 4 fields (1 field = 340 × 260 μm)/slide, comprising the stratum compactum (including microcaruncles on day 1). Ki-67–positive cells were also counted at 400× magnification in 4 fields. Two included the luminal epithelium and the stratum compactum, and the other 2 fields corresponded to the deep uterine glands and the stratum spongiosum.

qPCR assay—Endometrial biopsy specimens stored at −80°C were homogenized on ice in RNA-DNA-protein isolation reagent^p by use of a glass homogenizator previously heated for 2 hours at 250°C. Samples were centrifuged at 12,000 × g for 10 minutes at 4°C to remove insoluble material. The transferred supernatant was allowed to remain for 5 minutes at 20°C. Then, 0.2 mL of chloroform/mL of RNA-DNA-protein isolation reagent was added and the mixture was shaken vigorously for 15 seconds, allowed to stand for 5 minutes at 20°C, and subsequently centrifuged at 12,000 × g for 15 minutes at 4°C. By this centrifugation step, the mixture was separated into 3 phases. The RNA-containing, colorless upper aqueous phase was transferred to a fresh tube and frozen at −80°C for further analysis. Total RNA was extracted by use of a viral RNA kit^q by use of the protocol recommended by the manufacturer but starting with an ethanol treatment. To exclude contamination with DNA, PCR runs of each gene of the prepared endometrial samples were carried out. No DNA amplification could be detected in any of the samples used.

For preparation of the master mix (12.5 μL of reaction mix, 0.5 μL of 5-carboxy-X-rhodamine reference dye, 0.5 μL of enzyme mix, 0.5 μL of each primer, 0.5 μL of probe, and 7.5 μL of diethyl pyrocarbonate), a ready-to-use reaction mix for 1-step qPCR^r was employed. To 22.5 μL of the master mix, 2.5 μL of template was added. All primers and probes were used in 10μM concentrations. Equine gene-specific primers and internal oligonucleotide probes for IL-1β, -6, and -8 and TNF-α were applied as described.³⁰ Equine specific primers and probes of β-actin, GAPDH, COX-2, and PGS were used as described.³¹ Primers and probes for estrogen receptor and progesterone receptor were designed by use of a specific software program,^s and primers and probes were then synthesized for the study (Table 1).^t,u Probes were labeled at the 5′ end with the reporter dye 6-carboxyfluoresceine and at the 3′ end with the quencher dye 6-carboxytetramethyl-rhodamine. Correlation coefficients were determined by generating standard curves for primer and probes by use of total RNA from endometrial biopsy specimens.

Amplification and detection were performed by use of a qPCR system^v with a 96-well optical reaction plate, closed by optical cap stripes. The procedure was started via reverse transcription at 50°C for 15 minutes, followed by 2 minutes of denaturation of the reverse transcriptase and Taq DNA polymerase activation at 95°C. Forty amplification cycles were done (at 95°C for 15 seconds and annealing-elongation at 60°C for 30 seconds). Each sample was assayed in triplicate, and the arithmetic mean value was calculated and used for further analyses. Results were analyzed by use of qPCR calculation software.^w Relative expression of the genes of interest was calculated in comparison with β-actin and GAPDH. In a previous study³² on equine endometrial gene expression, these had been determined to be the most stable housekeeping genes.

Statistical analysis—All statistical comparisons were made with computer software.^x For all comparisons, nonparametric tests were used. Comparisons between times were made by use of the Friedman test, taking into account the sequential nature of data. Individual pairs were further analyzed by use of the Wilcoxon test. Comparisons between mares that had ovulated by the respective day (day 9 or 16) and those that had not ovulated were made by use of the Mann-Whitney test. For all comparisons, a P value < 0.05 was considered significant. Data given are mean ± SD.

Results

Clinical findings—Five of the 9 mares ovulated within the experimental period (mean postpartum day 13.8 ± 1.7), and 2 mares ovulated between postpartum days 16 and 20. One mare ovulated on day 32, and the remaining mare had not ovulated by day 20 and was no longer available after that day. Bacteria considered as specific or potential genital pathogens were not found in uterine swabs taken from mares on postpartum day 1. Mild growth of bacteria was found in 3 mares on day 9 (Escherichia coli, β-hemolytic streptococci, and normal vaginal flora in 1 mare each) and in 3 other mares on day 16 (E coli, β-hemolytic streptococci, and normal vaginal flora in 1 mare each). On postpartum days 9 and 16, none of the mares had lochial fluid in the uterus or had clinical signs of endometritis.

Histologic findings—On postpartum day 1, microcaruncles were evident in the stratum compactum. The luminal epithelium was tall columnar, and in 1 mare, it was cubic. High numbers of neutrophils were found in the stratum compactum under the luminal epithelium, around and in the microcaruncles (Figure 1). Lymphocytes were also present in microcaruncles and connective tissue. Accumulations of erythrocytes were apparent, and apoptotic bodies were visible in the luminal epithelium and the microcaruncles. In the stratum spongiosum, neutrophils, lymphocytes, and erythrocytes were only sporadically present. On day 9, microcaruncles were no longer identified. The luminal epithelium was tall columnar, cylindric to pseudostratified, or cubic to tall columnar. Neutrophils were scant, and lymphocytes, erythrocytes, and macrophages were present. In the stratum spongiosum, the predominant cells were lymphocytes. The numbers of siderophages and macrophages in connective tissue were scant to high. In 1 mare, a major infiltration of erythrocytes was found. In 2 mares, no siderophages were evident, whereas in the others, scant to high numbers were detectable. On day 16, the luminal epithelium had the same appearance as on day 9. In the stratum compactum and spongiosum, lymphocytes and macrophages were visible (Figure 2). Siderophages existed as on day 9.

View Full Size

Photomicrographic view of a section of endometrial tissue obtained from a mare on postpartum day 1 (A; notice a microcaruncle [MC] and accumulation of neutrophils under the luminal epithelium) and postpartum day 9 (B; notice siderophages around deep uterine glands in the stratum spongiosum). H&E stain; bar = 100 μm.

Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1525

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Mean ± SD cellular content scores (range, 0 to 3) in stratum compactum (A) and stratum spongiosum (B) of H&E stained endometrial biopsy specimens taken on the day of foaling (left columns) and on postpartum days 9 (middle columns) and 16 (right columns). ^a,bValues with different superscript letters are significantly (P < 0.05) different.

Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1525

• Download Figure
• Download figure as PowerPoint slide

Uterine glands were markedly dilated on day 1 (score, 2.1 ± 0.3) but not on days 9 and 16 (scores, 0.1 ± 0.3 and 0.0 ± 0.0, respectively [P < 0.05 vs day 1]). Uterine glands were coiled on day 1 and elongated on day 9. On day 16, uterine glands were elongated in mares that had ovulated and coiled in the others. No glandular contents were detectable.

Lectin binding patterns—On day 1, RCA I binding to galactose residues strongly labeled the glycocalyx and Golgi region of cells of the luminal epithelium and glandular epithelium near to the uterine lumen. Within the microcaruncles, the glycocalyx and cytoplasm of epithelial cells also had RCA binding. On day 9, glycocalyx and Golgi apparatus of luminal epithelium and luminal gland openings were still intensely stained. Macrophages were moderately stained, and siderophages were strongly stained with this lectin. On day 16, RCA binding of the deep uterine glands was reduced, compared with binding on day 9.

The glycocalyx of the luminal epithelium, including that of microcaruncles, was strongly stained for HPA, indicating a high number of surface N-acetylgalactosamine residues. The deep uterine glands had a noncontinuous but strongly labeled glycocalyx. On days 9 and 16, the HPA binding to the surface epithelium was still strong.

On day 1, the glycocalyx of the luminal epithelium reacted strongly with GSL I, whereas the epithelium of the microcaruncles yielded negative results. The glandular epithelium near to the uterine lumen had intense GSL binding (Figure 3). The binding pattern of GSL I, indicating α-galactose residues, had changed significantly on day 9. The luminal epithelium remained nearly unstained, and a positively stained glycocalyx was seen in deep uterine glands. On day 16, α-galactose residues on the luminal epithelium reappeared.

View Full Size

Photomicrographic views of a section of endometrial tissue obtained from a mare. A—Notice lectin binding pattern with GSL in luminal gland (LG) openings next to a microcaruncle on postpartum day 1. B—Notice luminal epithelium and luminal gland openings remain nearly unstained with GSL on postpartum day 9. Macrophages did not react with this lectin. C—Notice UEA binding patterns to the luminal epithelium, microcaruncles, and luminal gland openings on postpartum day 1. The luminal epithelium remains nearly unstained, whereas luminal gland openings react strongly with this lectin. D—Notice strongly UEA-colored luminal epithelium and luminal gland openings in uterine tissue on postpartum day 16. Lectinhistochemical stains; bar = 100 μm. See Figure 1 for remainder of key.

Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1525

• Download Figure
• Download figure as PowerPoint slide

On day 1, no binding of UEA I lectin was seen at the luminal surface epithelial cells (Figure 3). Ulex europacus agglutin I binding of luminal gland openings, glycocalyx, and cytoplasm in the microcaruncles was variable, and deep uterine glands were sparsely labeled. On day 9, the luminal epithelium had only minor fucose residues and the glandular openings were clearly marked by UEA I. On day 16, UEA I binding had significantly changed; the luminal epithelium was now intensively stained.

The lectin WGA, indicating N-acetyl-glucosamine residues, reacted strongly with the glycocalyx and Golgi apparatus of the luminal epithelium and epithelium of glandular openings on day 1. The glycocalyx of the microcaruncles and apoptotic bodies were also strongly stained. On day 9, the glycocalyx of the luminal epithelium and the glandular epithelium close to the uterine lumen remained WGA reactive. The glycocalyx of the deep uterine glands had intense binding of WGA, whereas the cytoplasm and Golgi apparatus had negative results. The macrophages were intensely stained with this lectin. An overview on lectin binding patterns to the cellular glycocalyx was composed (Table 2).

Immunohistochemical analysis—Caspase 3 staining, indicating apoptotic cells, was high on day 1, especially in the microcaruncles (Figure 4). It decreased significantly by day 9 and remained stable by day 16. On days 9 and 16, caspase 3–positive cells were mainly located in the stratum compactum. Also, macrophages and siderophages were mildly to strongly positive but were not included in the cell count. On postpartum day 1, only few cells stained for the proliferation marker Ki-67 in the microcaruncles and even fewer in the uterine glands and luminal epithelium. By day 9, a significant increase in the number of marked cells in the luminal epithelium and uterine glands had occurred versus day 1. On day 16, a slight but nonsignificant decrease in the number of marked cells was found (Figure 5). Lysozymepositive cells in great quantities were mainly located in the microcaruncles on day 1. Frequently, they were also detectable in blood vessels. A significant decrease was found on day 9, and the number of marked cells, mainly located in the stratum compactum, remained stable thereafter. Neutrophils were positive on day 1, whereas macrophages were not stained on days 9 and 16. For caspase 3, Ki-67, or lysozyme, no differences were found between mares that had already ovulated and those that had not yet ovulated on the respective days.

View Full Size

Photomicrographic views of a section of endometrial tissue obtained from a mare. A—Notice apoptotic cells stained for caspase 3 in a microcaruncle. In the luminal epithelium and in luminal gland openings, only sporadic cells are stained. B—Notice that on postpartum day 9 the number of caspase 3-stained cells (arrows) has decreased severely. C—Notice immunohistochemical staining of Ki-67 cells (arrow) in luminal gland openings, luminal epithelium, and microcaruncles on postpartum day 1. D—Notice proliferative activity in luminal epithelium and luminal gland openings marked by Ki-67 on postpartum day 16. Immunohistochemical stains; bar = 100 μm. See Figures 1 and 2 for remainder of key.

Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1525

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Mean ± SD numbers of cells per microscopic field of view immunohistochemically stained for Ki-67, caspase 3, and lysozyme in endometrial biopsy specimens obtained on the day of foaling (left columns) and on postpartum days 9 (middle columns) and 16 (right columns). See Figure 2 for remainder of key.

Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1525

• Download Figure
• Download figure as PowerPoint slide

qPCR assay—No significant changes in mRNA expressions existed among days 1, 9, and 16 for proinflammatory cytokines TNF-α, PGS, and COX-2, irrespective of the housekeeping gene used. Relative to β-actin, IL-6 mRNA expression was significantly (P = 0.01) higher on postpartum days 1 and 9, compared with expression on day 16. Relative to GAPDH, IL-8 mRNA expression was significantly higher on postpartum day 9 than on days 1 and 16, and expression of mRNA for PGS was higher (P = 0.01) on days 9 and 16 versus day 1 (Table 2)

On postpartum day 1, expressions of mRNA for progesterone and estrogen receptors were minimal versus days 9 and 16 (P = 0.01). By day 9, expressions of progesterone and estrogen receptors relative to β-actin and GAPDH housekeeping genes had markedly increased (P = 0.01 vs day 1; Figure 6) and remained largely stable thereafter.

View Full Size

Mean ± SD values for expressions of mRNA for progesterone (PR) and estrogen receptors (ER) in the endometrium of 9 mares in relation to β-actin on postpartum days 1 (d1), 9 (d9), and 16 (d16). See Figure 2 for remainder of key.

Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1525

• Download Figure
• Download figure as PowerPoint slide

Discussion

Seven of 9 mares ovulated by postpartum day 20. The resumption of ovulatory cycles thus occurred within a short and physiologic time frame.^1–3 Intrauterine fluid was gone quickly after parturition, which was in agreement with published data.¹³

The rapid removal of microcaruncles within the first postpartum days is in agreement with previous reports. Resorption of microcaruncles is complete by postpartum day 7,^4,33 and we did not find microcaruncles in biopsy specimens taken on day 9. Uterine glands were numerous, dilated, and coiled on day 1 but less dilated and straight instead of coiled on days 9 and 16, which was in agreement with previous reports.^4,14,33,34 On the day of foa1ing, large numbers of neutrophils were found in the stratum compactum with some lymphocytes present in microcaruncles and connective tissue. By day 9, neutrophils had largely disappeared and lymphocytes and macrophages had taken their place, which was in agreement with other studies.^5,6 The high number of neutrophils and the intense staining for lysozyme, marking neutrophil activity, on day 1 but no longer on days 9 and 16, indicated an acute process that might aid to clear the endometrium from cell debris and bacteria ascending into the uterus during foaling.

From these data, it appears that microcaruncles are removed in a programmed process. A high number of cells staining for the intracellular protease caspase 3 on day 1 indicated that cells undergo apoptosis. In addition, apoptotic bodies, resulting from shrinkage and fragmentation of cells, were found to a large extent in the luminal epithelium and microcaruncles on the day of foaling. Apoptotic bodies were then phagocytized by macrophages. The marked decrease in the number of caspase 3-positive cells by postpartum day 9 indicated that apoptotic processes were largely completed by then. Overall, the histologic changes occurred in all mares before foal heat ovulation.

It appears that, in parallel with the removal of tissue in the microcaruncles, regeneration of the endometrium occurs. This was indicated by a strong increase in cells that stained for the cell proliferation marker Ki-67 on postpartum days 1 and 9. The number of Ki-67–positive cells was still high on day 16, indicating ongoing proliferative activity. Endometrial cell proliferation in nonpregnant, cyclic mares is stimulated by high estrogen concentrations, as occur during estrus.¹⁶ Ki-67 staining in the luminal epithelium, endometrial gland openings, and superficial stroma thus is high in estrus and low in diestrus. An increased number of luminal epithelial cells in anestrous mares with small follicles in their ovaries indicates that in nonpregnant mares even low estrogen concentrations stimulate cell proliferation.¹⁶ Although, in our study, plasma estrogen concentrations were not determined, it can be assumed that the rapid resumption of ovulatory cycles is associated with an increase in plasma estrogen concentrations. This may stimulate endometrial proliferation after foaling.

Lectins bind specifically to glycoconjugates on the glycocalyx and within the cytoplasm. They thus provide information on secretory functions of the endometrial epithelium and have also been suggested to mark binding sites possibly involved in cell-to-cell recognition between the embryo and endometrium.¹⁷ In a previous study,²⁶ we were able to detect altered glycoconjugate patterns in mares with degenerative changes of the endometrium. Because uterine secretions should generate a microenvironment favorable for the early embryo, the endometrial glycoconjugate pattern was analyzed by means of lectin histochemistry. Binding of lectins to the glycocalyx of the endometrial epithelium changed over time. Compared with UEA binding in the nonpostpartum period, cyclic mares,²⁶ UEA binding at all locations was lower on days 1 and 9 but was similar on day 16. Also, WGA binding on postpartum day 16 was comparable with values for nonpostpartum mares, whereas binding was higher on days 1 and 9. The HPA and RCA binding in the luminal epithelium and luminal gland openings was similar in postpartum and nonpostpartum mares, whereas binding of these lectins to the deep glandular epithelium was higher in postpartum mares. Thus, despite rapid regeneration of the endometrium, uterine secretions on postpartum day 16 were, in part, different from the situation in cyclic, nonpostpartum mares. Although final conclusions on fertility require data on embryo survival, our results suggest that functional changes of the postpartum endometrium may take longer than histologic regeneration. Whether breeding at foal heat is associated with lower conception rates or higher fetal loss rates than breeding at second or later postpartum heats is discussed controversially. Although some authors do not find a generally reduced fertility in foal heat,^35,36 other authors report opposite findings.^10,12 Fertility in foal heat is clearly affected by uterine involution³⁶ but also by farm-related management factors.¹²

Interleukin-1β, -6, and -8 and TNF-α modulate acute inflammatory responses,^18,19 and IL-6 also regulates the transition from a neutrophil-dominated early inflammation to a macrophage-dominated process.²¹ Relative mRNA expressions for cytokines and key enzymes of prostaglandin synthesis were not markedly increased on day 1 (ie, in association with shedding of the microcaruncles). The increase in mRNA expression for IL-1β only by day 9 and the transient increases in IL-8 mRNA expression seem to indicate that inflammatory processes do not occur to a major extent immediately after foaling but may develop later during the puerperium. Our data, based on mRNA expression for proinflammatory cytokines, are in agreement with previous histologic findings, suggesting that removal of the microcaruncles in postpartum mares occurs without major inflammation.⁴

Expressions of progesterone and estrogen receptors were minimal on postpartum day 1, indicating that, at this time, the endometrium may be unresponsive to these steroids. By day 9, relative mRNA expressions for both receptors had markedly increased. Steroid receptor expression is regulated mainly at the level of mRNA transcription and mRNA stability.²¹ Although we did not analyze the abundance of these steroid receptors at the protein level, our data support the hypothesis that during foal heat increased expression of estrogen receptors allows the endometrium to respond to estrogens with a cycle-dependent cell proliferation. During subsequent diestrus, the endometrium should be able to respond to progesterone. The release of reproductive hormones, follicular development, and ovulation are not inhibited during foal heat.³⁷ Results of the present study indicated that uterine progesterone and estrogen receptors were present, and thus, it is unlikely that responsiveness of the endometrium to these hormones at first postpartum estrus is a factor limiting fertility.

Expressions of mRNA for progesterone and estrogen receptors in the endometrium were minimal on the day of foaling in postpartum mares. This indicated that the endometrium in foaling mares may be unresponsive to ovarian steroids. Within a few days, mRNA expressions for these steroid receptors increased markedly, suggesting that, at the time of foal heat and the subsequent luteal phase, the endometrium had fully regained its ability to respond to gonadal steroids. Removal of the endometrial epithelium occurred rapidly by apoptosis without a major inflammatory reaction. Thereafter, proliferative processes predominated, leading to histologic regeneration, but uterine secretions may still be altered by postpartum day 16.