We found previously that neutrophil-depleted mice exhibited significant blockading of both the regular estrous cycle and cyclic changes of steroid hormone levels. In this study, we aimed at elucidation of the underlying mechanism. To examine the possibility that an increase in bacteria in the vaginal vault of neutrophil-depleted mice causes blockading of the estrous cycle, we treated neutrophil-depleted mice with antibiotics but failed to restore the estrous cycle. We then examined another possibility that neutrophils regulate the estrous cycle via opioid peptides, because opioid peptides regulate steroidogenesis in theca and granulosa cells in the ovaries, and because neutrophils contain opioid peptides. In support of this possibility, naloxone, an opioid antagonist, blocked the estrous cycle and a μ opioid receptor agonist restored the estrous cycle in neutrophil-depleted mice. Pro-opiomelanocortin was immunohistochemically detected in peripheral blood neutrophils but not in ones that had infiltrated into the ovaries. i.v. injection of anti–MIP-2 polyclonal Ab caused blockading of the estrous cycle, whereas MIP-2 was detected in the ovaries, suggesting a role of MIP-2 in the regulation of the estrous cycle. Moreover, i.v. injection of MIP-2 decreased the pro-opiomelanocortin signal in peripheral blood neutrophils and caused blockading of the estrous cycle. Together, these results suggest that neutrophils maintain the estrous cycle via opioid peptides.

On bacterial infection, neutrophils infiltrate into the affected site where they exert bactericidal activity. The infiltration is mainly mediated by MIP-2 and KC, two major CXC chemokines specific to neutrophils, in mouse (1). Even without any infection, however, neutrophils also infiltrate into the vaginal vault at metestrus, which is mediated by MIP-2 (2). We recently showed with an anti–Gr-1 mAb that neutrophil-depleted mice exhibited significant blockading of the regular estrous cycle, as well as the cyclic changes of steroid hormone levels, suggesting a connection between neutrophils and the estrous cycle (3). One would expect that such infiltrating neutrophils should control the number of commensal bacteria in the vaginal vault, and so there is the possibility that an increase in such bacteria may cause blockading of the estrous cycle.

Neutrophils contain opioid peptides and IL-8 triggers the release of these opioid peptides (4, 5). Although neutrophil-derived opioid peptides are known to counter-control pain in peripheral inflamed tissues (6), it is currently unknown whether neutrophil-derived opioid peptides are involved in the regulation of the estrous cycle. In contrast, opioid peptides are known to participate in the action of steroid hormones in the brain. For instance, progesterone stimulates the production of opioid peptides in the hypothalamus, and these opioid peptides then act on gonadotropin-releasing hormone-secreting neurons in the hypothalamus to suppress the gonadotropin-releasing hormone pulse frequency (7).

In the ovaries, however, theca cells express different types of opioid receptors (8, 9), and opioid peptides regulate steroidogenesis in theca and granulosa cells (9, 10). Therefore, there is another possibility that neutrophils regulate the estrous cycle via opioid peptide release in the ovaries.

In this study, we aimed at elucidation of the mechanism underlying neutrophil-dependent regulation of the estrous cycle by focusing on these two possibilities described earlier.

Specific pathogen-free female ICR mice (5–7 wk old) were purchased from Sankyo Lab Service (Tokyo, Japan). The mice were maintained under a 12:12 light/dark cycle (lights-on from 6 am to 6 pm), and vaginal smears were prepared daily at 10 am. In this study, we used mice that had completed at least three estrous cycles (12–15 d). To examine the effects of antibiotic and antimycotic agents, we treated the vaginal vault with an Antibiotic-Antimycotic Mixed Stock Solution (Nacalai Tesque, Kyoto, Japan) every day for 2 wk. The project was approved by the Animal Experiment Committee of Toho University.

The phase of the estrous cycle was determined by analysis of vaginal smears as previously described (3). To deplete neutrophils, we administered 200 μg rat anti-neutrophil mAb (anti–Gr-1 mAb), which had been prepared from a supernatant of RB6-8C5 cells (provided by Dr. Sendo, Yamagata University), i.p. 9–10 h after lights-on. As a control, an equal dose of an anti-HLA mAb, which was prepared from a supernatant of SFR8-B6 cells (obtained from American Type Culture Collection), was administered i.p. at the same time.

Immunohistochemical staining was performed as previously described (3). In brief, tissues embedded in paraffin were cut into 5-μm-thick sections, followed by Ag retrieval with citric acid buffer (pH 6.0) for 5 min at 95°C. After blocking, a goat anti–pro-opiomelanocortin (POMC) polyclonal Ab (pAb; Santa Cruz Biotechnology, Santa Cruz, CA) was used at 1:100 dilution, followed by incubation at 4°C overnight. A VECTASTAIN Elite ABC Kit (Vector Laboratories, Burlingame, CA) and a Diaminobenzidine Substrate Kit (Vector Laboratories) were used according to the manufacturer’s instructions. As a control, some of the sections were incubated with normal goat IgG.

Peripheral blood neutrophils (pNeu) were obtained as follows. Mice were anesthetized with diethylether, and blood samples were obtained by postcaval vein puncture and kept in heparin-coated tubes. The RBCs were removed by the dextran method. The cell pellets were suspended in RPMI 1640 containing 20% FCS, and neutrophils were purified with a cell sorter (EPICS Altra; Coulter) based on forward scatter (FSC) and side scatter (SSC). The purity of the neutrophils was 86 ± 3%, as assessed by H&E staining. The cells were fixed for 6 h in Zamboni’s solution and then were embedded in paraffin in the same way as tissue samples.

Vaginal smears were fixed with methanol and then fixed for 30 min in Zamboni’s solution. After washing of the sections with PBS, immunohistochemical staining was performed as previously described (3). Each experiment was repeated three or more times.

An ovary was homogenized in 300 μl PBS containing 0.05% Tween 20 (Nacalai Tesque, Kyoto, Japan), followed by centrifugation at 4°C for 10 min at 12,000 rpm. The supernatant was stored at −80°C until the ELISA assay.

The plasma 17β-estradiol and progesterone levels were determined with an Estradiol EIA Kit or a Progesterone EIA Kit (Cayman Chemical, Ann Arbor, MI). The detection limits for estradiol and progesterone were 8 and 10 pg/ml, respectively. The mouse MIP-2 and KC levels were determined using ELISA development kits (PeproTech, Rocky Hill, NJ).

Opioid peptide release was assayed according to the method in Rittner et al. (11) with slight modification. Bone marrow neutrophils were separated from bone marrow cells by using Percoll density centrifugation. The purity of the neutrophils was 52 ± 5%, as assessed by H&E staining. The bone marrow neutrophils at 2 × 107 cells/ml were preincubated with 5 μg/ml cytochalasin B (Sigma-Aldrich) for 10 min in HBSS containing the protease inhibitors 5 μg/ml bestatin (Sigma-Aldrich), 40 μg/ml aprotinin (Wako Pure Chemicals, Tokyo, Japan), and 100 μM dl-thiorphan (Wako Pure Chemicals). Subsequently, cells were stimulated with 1 μg/ml MIP-2 (PeproTech). Release was terminated after 30 min by rapid cooling, centrifugation, and harvesting of the supernatant. Samples were stored at −20°C until assay. The mouse β-endorphin levels in the samples were determined using an ELISA kit (Wuhan EIAab Science, Wuhan, China).

To evaluate the involvement of opioid peptides in the regulation of the estrous cycle, we administered naloxone (NAL; 30 μg/mouse; Sigma-Aldrich Japan, Tokyo, Japan) (12), [D-Ala2, N-MePhe4, Gly-ol5]-enkephalin (DAMGO, 50 μg/mouse; Sigma-Aldrich Japan) (13), (5α, 7α, 8β)-(−)-N-methyl-N-(7-[1-pyrrolidinyl]-1-oxaspiro[4,5]dec-8-yl)benzenacetamide (U-69593, 50 μg/mouse; Sigma-Aldrich Japan) (13), or [D-Pen2,5]-enkephalin (DPDPE, 30 μg/mouse; Sigma-Aldrich Japan) (13) i.v. 9–10 h after lights-on on the day of estrus or every day.

A recombinant protein (MIP-2 or KC; PeproTech, Rocky Hill, NJ) or Abs (anti–MIP-2 pAb or anti-KC pAb; prepared in our laboratory) (14, 15) were administered i.v. at the dose of 100 ng/mouse or 400 μg/mouse, respectively, 9–10 h after lights-on on the day of estrus.

Heparinized peripheral blood was centrifuged, lysed, and washed. The resultant cells (2.5 × 105) were pretreated with 0.5 μg Fc block (anti-FcγRIII/II mAb prepared from the supernatant of 2.4G2 hybridoma cells) and 0.5 μg mouse IgG1, prepared from the ascites fluid of myeloma MOPC-21 cells for 30 min on ice. Cells were then stained for 30 min at 4°C with FITC-labeled anti–Gr-1 mAb or FITC-labeled anti-HLA mAb as an isotype control. Alternatively, cells were stained with biotinylated anti-CD69 mAbs (BD Pharmingen, San Diego, CA), washed, and then incubated with PE-conjugated streptavidin, for 30 min at 4°C. After washing with PBS containing 0.5% BSA twice, the cells were analyzed by flow cytometry with a FACSCalibur (BD Biosciences) using CellQuest or FlowJo software.

Differences between experimental groups were analyzed by means of one-way factorial ANOVA (one-factor ANOVA) and the post hoc test (Scheffe’s F) using Statcel (OMS-Publishing, Saitama, Japan) or Student t test. When p < 0.05, the difference was considered statistically significant.

Ninety percent of the ICR mice used in this study exhibited a 4- or 5-d estrous cycle when untreated, whereas 10% exhibited a 6- or 7-d estrous cycle with a 1- or 2-d diestrus for an unknown reason. The cycle length in untreated ICR mice was 4.7 ± 0.8 d (n = 54).

We showed previously with an anti–Gr-1 mAb that depletion of neutrophils led to blockading of the estrous cycle at diestrus (3). We confirmed in this study that an anti–Gr-1 mAb causes blockading of the estrous cycle at metestrus and/or diestrus (Fig. 1A–D), and that a control mAb (anti-HLA mAb) does not cause blockading of the estrous cycle (Fig. 1E), the cycle lengths being 7.7 ± 2.3 (n = 67) and 5.3 ± 0.5 d (n = 8), respectively. The difference in the cycle length between anti–Gr-1 mAb- and anti-HLA mAb-treated mice was statistically significant (p < 0.00001). Because many bacteria were detected in neutrophils in vaginal lavage fluid (VLF) (Fig. 1G) and neutrophil depletion caused an increase in bacteria such as cocci in the vaginal vault (Fig. 1H), we examined whether an increase in bacteria causes blockading of the estrous cycle. When we treated the vaginal vault of mice with antibiotic and antimycotic agents, bacteria were cleared from the VLF (Fig. 1I), but the estrous cycle remained blocked (Fig. 1F), the cycle length being 8.1 ± 0.8 d (n = 8). The difference in the cycle length between neutrophil-depleted and neutrophil-depleted/antibiotic and antimycotic agent-treated mice was not statistically significant. Therefore, the possibility is rather remote that an increase in bacteria in the vaginal vault causes blockading of the estrous cycle in neutrophil-depleted mice.

FIGURE 1.

The estrous cycle in neutrophil-depleted mice and Gram staining of vaginal smears. A–F, Mice were treated on days 0 and 4 (or 3 in F) with either an anti–Gr-1 mAb (A–D, F, open circles) or an anti-HLA mAb (E, gray circles). On day 0, mice were at proestrus (Pro) (A, F), estrus (Est) (B, E), metestrus (Met) (C), or diestrus (Di) (D). The vaginal vaults of mice were treated with antibiotic and antimycotic agents every day between days −5 and 9 (F). These are representative results for mice that exhibited a 4- or 5-d estrous cycle when untreated. G–I, Gram staining of metestrus vaginal smears from an untreated mouse (G), a neutrophil-depleted mouse treated with PBS (H), and a neutrophil-depleted mouse treated with antibiotic and antimycotic agents (I). An increase in bacteria was noticed in H but not in I. Original magnification ×1000 (G–I). Scale bars, 10 μm.

FIGURE 1.

The estrous cycle in neutrophil-depleted mice and Gram staining of vaginal smears. A–F, Mice were treated on days 0 and 4 (or 3 in F) with either an anti–Gr-1 mAb (A–D, F, open circles) or an anti-HLA mAb (E, gray circles). On day 0, mice were at proestrus (Pro) (A, F), estrus (Est) (B, E), metestrus (Met) (C), or diestrus (Di) (D). The vaginal vaults of mice were treated with antibiotic and antimycotic agents every day between days −5 and 9 (F). These are representative results for mice that exhibited a 4- or 5-d estrous cycle when untreated. G–I, Gram staining of metestrus vaginal smears from an untreated mouse (G), a neutrophil-depleted mouse treated with PBS (H), and a neutrophil-depleted mouse treated with antibiotic and antimycotic agents (I). An increase in bacteria was noticed in H but not in I. Original magnification ×1000 (G–I). Scale bars, 10 μm.

Close modal

Neutrophils contain opioid peptides. Moreover, theca cells express different types of opioid receptors (8, 9), and opioid peptides regulate steroidogenesis in theca and granulosa cells (9, 10). We therefore hypothesized that neutrophils may regulate the estrous cycle via opioid peptides.

We administered NAL, an opioid receptor antagonist, to mice on day 0 (Fig. 2B) or every day from days 0–11 (Fig. 2C -1C-4). This treatment did not affect the number of neutrophils in vaginal smears at metestrus. Nevertheless, the estrous cycle was blocked at diestrus, the cycle length being 7.9 ± 2.0 and 8.8 ± 0.7 d for mice treated with NAL once (n = 7) and ones treated with NAL consecutively (n = 13). The difference in the cycle length between untreated mice and mice treated with NAL either once or consecutively was statistically significant (p = 0.002 for treatment with NAL once, p < 0.00001 for consecutive treatment with NAL, respectively).

FIGURE 2.

The estrous cycle and hormone levels in NAL-treated mice. A–C, Mice were treated every day from days 0–11 with PBS (A, gray circles) or NAL (C-1–C-4, open circles), or treated with NAL on day 0 (B, open circles). On day 0, mice were at proestrus (C-1), estrus (B, C-2), metestrus (C-3), or diestrus (A, C-4). D and E, Each sample was taken at 3 pm. From neutrophil-depleted mice and those treated once with NAL, samples were taken on the second day after the estrous cycle was blocked at diestrus. The estrogen (E2) and progesterone (P4) levels were determined by specific ELISAs. Data are expressed as the means ± SEs for 3–5 mice. *p < 0.05.

FIGURE 2.

The estrous cycle and hormone levels in NAL-treated mice. A–C, Mice were treated every day from days 0–11 with PBS (A, gray circles) or NAL (C-1–C-4, open circles), or treated with NAL on day 0 (B, open circles). On day 0, mice were at proestrus (C-1), estrus (B, C-2), metestrus (C-3), or diestrus (A, C-4). D and E, Each sample was taken at 3 pm. From neutrophil-depleted mice and those treated once with NAL, samples were taken on the second day after the estrous cycle was blocked at diestrus. The estrogen (E2) and progesterone (P4) levels were determined by specific ELISAs. Data are expressed as the means ± SEs for 3–5 mice. *p < 0.05.

Close modal

The estrous cycle was blocked by NAL in exactly the same way as it was in anti–Gr-1 mAb-treated mice (Fig. 1A–D). When NAL administration or anti–Gr-1 mAb administration was initiated at proestrus or estrus, the estrous cycle continued up to diestrus, but thereafter it was blocked (Fig. 2C,-1, 2C,-2 versus Fig. 1A, 1B). On the contrary, when NAL administration or anti–Gr-1 mAb administration was initiated at metestrus or diestrus, the estrous cycle was not blocked at diestrus, but continued up to proestrus and then proceeded to the next diestrus, and thereafter was blocked (Fig. 2C,-3, 2C,-4 versus Fig. 1C, 1D). In addition, the estrogen (E2) and progesterone (P4) levels in plasma were affected in the same way as in neutrophil-depleted mice (Fig. 2D, 2E). It is noteworthy that when NAL was administered only once at estrus, but not other phases, the estrous cycle continued up to diestrus and thereafter was blocked at diestrus (Fig. 2B, data not shown), confirming the previous finding with anti–Gr-1 mAb that the estrous phase is the most critical among the four phases for neutrophil-dependent regulation of the estrous cycle (3).

We then examined whether opioid receptor agonists can restore the estrous cycle in neutrophil-depleted mice. An anti–Gr-1 mAb was administered on days −4 and 0, and the estrous cycle was blocked at diestrus from days −1 to 4 (Fig. 3A). When DAMGO, a μ-opioid receptor selective agonist, U-69593, a κ-opioid receptor-selective agonist, or DPDPE, a δ-opioid receptor-selective agonist, was administered i.v. to such neutrophil-depleted mice on day 0 (open circles with vertical centerlines), DAMGO, but not the other agents, shortened the blocked period at diestrus (Fig. 3B–D), the cycle lengths being 6.0 ± 0.7 d for DAMGO (n = 5), 7.0 ± 1.4 d for U-69593 (n = 5), and 8.0 ± 1.9 d for DPDPE (n = 5). The difference in the cycle length between neutrophil-depleted and neutrophil-depleted/DAMGO-administered mice was statistically significant (p = 0.0008), whereas that between neutrophil-depleted and neutrophil-depleted/U-69593 or DPDPE-administered mice was not significant. These results suggest that a μ-type opioid receptor agonist is required for regulation of a stable estrous cycle, and that neutrophils are either directly or indirectly involved in the production of such an opioid peptide.

FIGURE 3.

Restoration of the estrous cycle by opioid receptor agonists. A–D, Mice were treated with an anti–Gr-1 mAb on days −4 and 0 (A–D; open, gray, or lined circles), and they were treated with an opioid agonist (B–D, lined circles) or PBS (A, gray circle) on day 0. The agonists used were DAMGO (B), DPDPE (C), and U69593 (D). On day −4, the mice were at estrus and remained neutropenic until day 4.

FIGURE 3.

Restoration of the estrous cycle by opioid receptor agonists. A–D, Mice were treated with an anti–Gr-1 mAb on days −4 and 0 (A–D; open, gray, or lined circles), and they were treated with an opioid agonist (B–D, lined circles) or PBS (A, gray circle) on day 0. The agonists used were DAMGO (B), DPDPE (C), and U69593 (D). On day −4, the mice were at estrus and remained neutropenic until day 4.

Close modal

If neutrophils secrete opioid peptides to regulate the estrous cycle, then vaginally infiltrating neutrophils (vNeu) may not contain opioid peptides. We therefore examined POMC in pNeu and vNeu immunohistochemically. A population rich in pNeu was prepared with a cell sorter, as described in 1Materials and Methods. The signal for anti-POMC pAb was detected in 35.4 ± 7.8% of pNeu (Fig. 4D) but no vNeu (Fig. 4B). In this article, neutrophils were identified and discriminated from eosinophils as to nuclear morphology and eosin staining, with neutrophils and eosinophils amounting to 86.1 ± 2.7 and 6.9 ± 1.0%, respectively.

FIGURE 4.

Detection of POMC in pNeu but not in vNeu. vNeu were stained with H&E (A) or anti-POMC pAb (B). There are no positive signals in B. Neutrophils isolated from peripheral blood were stained with H&E (C) or anti-POMC pAb (D). There are many positive signals (brown) in D. Original magnification ×100. Scale bars, 100 μm.

FIGURE 4.

Detection of POMC in pNeu but not in vNeu. vNeu were stained with H&E (A) or anti-POMC pAb (B). There are no positive signals in B. Neutrophils isolated from peripheral blood were stained with H&E (C) or anti-POMC pAb (D). There are many positive signals (brown) in D. Original magnification ×100. Scale bars, 100 μm.

Close modal

We then conducted experiments to elucidate the mechanism underlying the release of opioid peptides by neutrophils. It has been reported that IL-8 triggers opioid peptide release from neutrophils in vitro (4, 5). If neutrophils secrete opioid peptides in response to chemokines such as IL-8 in vivo, neutrophils infiltrating into the genital organs may also secrete opioid peptides in response to IL-8, thereby modulating the estrous cycle. We thus examined whether MIP-2 or KC, a mouse IL-8 homolog, is involved in the estrous cycle. When anti–MIP-2 pAb or anti-KC pAb was administered i.v. at estrus, both Abs blocked the estrous cycle at diestrus (Fig. 5A,-1, 5A -2), although anti-KC pAb was not as effective as anti–MIP-2 pAb. The cycle length in anti–MIP-2– and anti-KC–treated mice was 7.3 ± 2.1 (n = 10) and 6.2 ± 1.3 d (n = 6), respectively, both being statistically significant as compared with in untreated mice (p = 0.0006 for anti–MIP-2; p = 0.02 for anti-KC).

FIGURE 5.

The effect of anti–MIP-2, anti-KC, MIP-2, or KC on the estrous cycle, and that of MIP-2 or KC on POMC in neutrophils. A-1 to B-3, Mice were treated with an anti–MIP-2 pAb (A-1), an anti-KC pAb (A-2), rMIP-2 (B-1), rKC (B-2), or PBS (A-3, B-3) on day 0 (open circles). b-1 and b-2, Immunohistochemical analysis of pNeu of mice treated with rMIP-2 (b-1) or rKC (b-2) was performed. Each sample was stained with an anti-POMC pAb, as described in 1Materials and Methods. Positive signals are indicated by arrows. Original magnification ×100. Scale bars, 100 μm.

FIGURE 5.

The effect of anti–MIP-2, anti-KC, MIP-2, or KC on the estrous cycle, and that of MIP-2 or KC on POMC in neutrophils. A-1 to B-3, Mice were treated with an anti–MIP-2 pAb (A-1), an anti-KC pAb (A-2), rMIP-2 (B-1), rKC (B-2), or PBS (A-3, B-3) on day 0 (open circles). b-1 and b-2, Immunohistochemical analysis of pNeu of mice treated with rMIP-2 (b-1) or rKC (b-2) was performed. Each sample was stained with an anti-POMC pAb, as described in 1Materials and Methods. Positive signals are indicated by arrows. Original magnification ×100. Scale bars, 100 μm.

Close modal

We then examined whether i.v. administration of rMIP-2 or recombinant KC (rKC) may also affect the estrous cycle. These chemokines did not affect the numbers of neutrophils in vaginal smears. Despite this, when mice were administered rMIP-2 or rKC at estrus (Fig. 5B,-1, 5B,-2) but not other phases (data not shown), the estrous cycle was blocked at diestrus, as it was in anti–Gr-1 mAb-treated mice. The cycle length in rMIP-2– or rKC-administered mice was 7.1 ± 2.0 (n = 10) or 6.8 ± 2.1 d, respectively, both being statistically significant as compared with in untreated mice (p = 0.004 for rMIP-2; p = 0.03 for rKC). The POMC signal was hardly detected in pNeu of such treated mice, as indicated by black arrows in Fig. 5b,-1 (for rMIP-2) and 5b -2 (for rKC). These findings suggest that rMIP-2 or rKC induces the release of opioid peptides from neutrophils in peripheral blood, and that infiltrating neutrophils cannot regulate the estrous cycle because of the absence of opioid peptides.

CD69 is induced on activation in lymphocytes and neutrophils (15), and thus CD69 serves as an activation marker. To corroborate the earlier finding regarding opioid peptides in pNeu after rMIP-2 administration, we then examined the activation of pNeu that had been obtained from rMIP-2–administered mice by means of flow cytometric analysis of induction of CD69. When the cells were gated with the FSC/SSC profile, as depicted in Fig. 6A (left panel), they were found to be Gr-1high–positive, as shown in Fig. 6A (center panel), and propidium iodide-negative (viable), as shown in Fig. 6A (right panel). When neutrophils were identified by means of the FSC/SSC profile, CD69 was found to be induced in the neutrophils that had been obtained from mice at estrus shortly after rMIP-2 administration (Fig. 6C). But neither neutrophils from mice treated at other phases nor untreated ones showed such induction (Fig. 6B, 6C).

FIGURE 6.

Activation of pNeu by i.v. administered rMIP-2. Peripheral blood was taken from mice 30 min after rMIP-2 had been administered i.v. Activation of pNeu was then examined by means of CD69 expression as described in 1Materials and Methods. A, All cells in the lysed peripheral blood were displayed as a dot plot of FSC versus SSC. When cells were gated with the FSC versus SSC profile (left panel, dotted circle), they were found to be Gr-1high– (center panel) and propidium iodide-negative (right panel), viable neutrophils. The histograms of such gated cells stained with an anti-CD69 mAb are shown as gray shaded areas, whereas those stained with the control mAb are shown as black lines. B, Expression of CD69 in the neutrophils of untreated mice. C, Expression of CD69 in the neutrophils of rMIP-2–treated mice.

FIGURE 6.

Activation of pNeu by i.v. administered rMIP-2. Peripheral blood was taken from mice 30 min after rMIP-2 had been administered i.v. Activation of pNeu was then examined by means of CD69 expression as described in 1Materials and Methods. A, All cells in the lysed peripheral blood were displayed as a dot plot of FSC versus SSC. When cells were gated with the FSC versus SSC profile (left panel, dotted circle), they were found to be Gr-1high– (center panel) and propidium iodide-negative (right panel), viable neutrophils. The histograms of such gated cells stained with an anti-CD69 mAb are shown as gray shaded areas, whereas those stained with the control mAb are shown as black lines. B, Expression of CD69 in the neutrophils of untreated mice. C, Expression of CD69 in the neutrophils of rMIP-2–treated mice.

Close modal

The results in Figs. 5 and 6 raise the possibility that MIP-2 induces the release of opioid peptides from neutrophils in some genital organs, perhaps the ovaries, thereby controlling the estrous cycle. We then determined the MIP-2 level in the ovaries and examined POMC in the ovaries immunohistochemically.

The MIP-2 level in the ovaries peaked at estrus as compared with in other phases (Fig. 7A), although the KC level did not change significantly (Fig. 7B). In contrast, the MIP-2 level in VLF was significantly higher at metestrus as compared with at other phases (Fig. 7C), which is partially responsible for the neutrophil infiltration into the vaginal vault (2). We then examined the localization of neutrophils and MIP-2 in the ovaries immunohistochemically. Neutrophils were detected in the vessels close to the granulosa cells (Fig. 7D, arrows), whereas the MIP-2 signal was detected in the cells situated close to the granulosa cells (Fig. 7E, arrows).

FIGURE 7.

The MIP-2 and KC levels in the ovaries and VLF, and immunohistochemical detection of neutrophils and MIP-2 in the ovaries. Ovaries were obtained at 6 pm on the day of estrus. MIP-2 levels in the ovaries (A) and VLF (C), or KC levels in the ovaries (B) were measured by ELISAs. The data are expressed as the means ± SEs for three to five mice. *p < 0.05. Sections were stained with either an anti–Gr-1 mAb (D) or anti–MIP-2 pAb (E). Original magnification ×400. Scale bars, 25 μm.

FIGURE 7.

The MIP-2 and KC levels in the ovaries and VLF, and immunohistochemical detection of neutrophils and MIP-2 in the ovaries. Ovaries were obtained at 6 pm on the day of estrus. MIP-2 levels in the ovaries (A) and VLF (C), or KC levels in the ovaries (B) were measured by ELISAs. The data are expressed as the means ± SEs for three to five mice. *p < 0.05. Sections were stained with either an anti–Gr-1 mAb (D) or anti–MIP-2 pAb (E). Original magnification ×400. Scale bars, 25 μm.

Close modal

We then examined immunohistochemically whether neutrophils release opioid peptides on infiltration into ovaries that had been obtained at 6 pm on the day of estrus. The POMC signal was detected in some of the intravascular neutrophils (Fig. 8A,a, 8A,b, black arrows) but not others (Fig. 8A,b; white arrows), with 26 ± 7.2% of the intravascular neutrophils being positive. In this study, the neutrophils were identified with nuclear morphology. In contrast, no POMC signal was detected in neutrophils that had extravasated into the stroma (Fig. 8B, 8C).

FIGURE 8.

Immunohistochemical detection of neutrophils and POMC in the ovaries. Ovaries were obtained at 6 pm on the day of estrus. Sections were stained with either an anti-POMC pAb (A, Aa, Ab, C) or an anti–Gr-1 pAb (B). Black and white arrows indicate positive and negative signals, respectively. Original magnification ×400. Scale bars, 25 μm.

FIGURE 8.

Immunohistochemical detection of neutrophils and POMC in the ovaries. Ovaries were obtained at 6 pm on the day of estrus. Sections were stained with either an anti-POMC pAb (A, Aa, Ab, C) or an anti–Gr-1 pAb (B). Black and white arrows indicate positive and negative signals, respectively. Original magnification ×400. Scale bars, 25 μm.

Close modal

The release of β-endorphin from bone marrow neutrophils was also examined according to the method described in 1Materials and Methods. As a result, MIP-2 caused β-endorphin release (3.2 ± 4.5 pg/ml), being statistically significant as compared with control (1.5 ± 3.9 pg/ml; n = 15; p = 0.033).

In this study, we examined the mechanism underlying neutrophil-dependent regulation of the estrous cycle by focusing on the following two possibilities. The first possibility is that an increase in bacteria in the vaginal vault causes blockading of the estrous cycle in neutrophil-depleted mice, whereas the second one is that neutrophils regulate the estrous cycle through the release of opioid peptides in response to MIP-2 in the ovaries.

The first possibility is rather unlikely, because treatment of mice with antibiotic and antimycotic agents cleared bacteria from the VLF but failed to restore the estrous cycle. In contrast, the second one is more likely, based on several lines of pharmacological and immunohistochemical evidence, although at present we cannot exclude the possibility that neutrophils regulate the estrous cycle through the release of opioid peptides by the cells other than neutrophils. First, NAL caused blockading of the estrous cycle. Second, DAMGO, a μ-opioid receptor-selective agonist, restored the estrous cycle in neutrophil-depleted mice. Third, anti–MIP-2 pAb caused blockading of the estrous cycle, whereas MIP-2 was detected in the ovaries. Fourth, i.v. injection of rMIP-2 decreased the POMC signal in pNeu and caused blockading of the estrous cycle. Fifth, the POMC signal was detected in intravascular neutrophils but not neutrophils that had extravasated into stroma of the ovaries. Sixth, MIP-2 caused the release of β-endorphin from bone marrow neutrophils significantly as compared with control.

We used anti–Gr-1 mAb to detect neutrophils immunohistochemically and to deplete neutrophils. On flow cytometric analysis, neutrophils were found to be highly positive for Gr-1, whereas eosinophils were moderately positive for Gr-1 (16, 17). However, our previous study demonstrated that anti–Gr-1 mAb detected neutrophils, but not eosinophils, in genital organs immunohistochemically, and that anti–Gr-1 mAb depleted neutrophils, but not eosinophils, from the circulation, as well as in the genital organs (3). Although the physiological ligand and biological function have not been clearly defined, there is a possibility that anti–Gr-1 mAb also affects cellular function independently of neutrophil depletion. Indeed, it has been reported that an anti–Gr-1 mAb induces apoptosis in peripheral inflammatory neutrophils (18). However, the F(ab′)2 fragment of anti–Gr-1 mAb did not affect the number of circulating neutrophils (19) or the estrous cycle (S. Sasaki, unpublished observations), supporting that the effect of an anti–Gr-1 mAb on the estrous cycle is mediated by depletion of neutrophils.

The POMC signal was detected in 35% of pNeu, but not in neutrophils that had extravasated into stroma of the ovaries. This suggests that neutrophils secrete opioid peptides on infiltration into the ovaries. Because it has also been reported that ∼40% of rat circulating leukocytes were immunoreactive for β-endorphin and POMC (20), there are two populations of pNeu in terms of POMC expression, and it is possible that the population having no POMC signal infiltrates into the ovaries selectively, although CXCR2, a receptor for MIP-2, is homogeneously expressed on mature Gr-1high bone marrow neutrophils (21). Further study is needed to clarify this point.

POMC mRNA is also expressed in theca and granulosa cells (22). Therefore, not only neutrophils, but also theca and granulosa cells, may secrete opioid peptides and participate in the local regulation of steroidogenesis. However, the intensity of the POMC signal in the stroma of ovaries was weakest at estrus (23), and so opioid peptides derived from theca and granulose cells may not participate in neutrophil-dependent regulation of the estrous cycle.

As shown in our previous study, during the blockading of the estrous cycle because of neutrophil depletion, the steroid hormone levels gradually returned to normal, suggesting the presence of a backup mechanism independent of neutrophils (3). In this study, we observed that a second NAL administration at estrus (day 8 in Fig. 2C,-2 and day 10 in Fig. 2C,-4) caused blockading at diestrus for 2 d, which was shorter than the blocked period for the first NAL administration (4 d; Fig. 2). This suggests that during the blockading of the estrous cycle caused by NAL administration, a backup mechanism may also come into operation to normalize the estrous cycle, and that the backup mechanism may be less dependent on opioid peptides.

There have been several reports on the relation between cytokines, infertility, and endometriosis. The serum levels of IL-6 and IL-8 in infertile patients are significantly higher than in fertile ones and more pronounced in the cervical mucus (24). In addition, the serum level of IL-8 is higher in women with endometriosis, whereas those of IL-6, RANTES, and IL-1β are not significantly different (25). Furthermore, it has been reported that the opioid concentration in peripheral mononuclear cells was reduced in the luteal phase in endometriosis patients compared with in controls (25), presumably because of opioid peptide release from neutrophils caused by IL-8 in peripheral blood. Our findings may also shed light on unexplained associations with ovarian dysfunctions and reduced fertility occurring mostly during the active stage of human inflammatory bowel disease (2628), because in a rat model of severe colitis, a significant decrease in uterine neutrophils was reportedly associated with estrous cycle disturbance (26). Our findings therefore may provide a novel therapy for some forms of infertility involving normalization of the serum IL-8 level. There is another example showing the relation between cytokine and reproductive disorder. The female athlete triad (anorexia, amenorrhea, and osteoporosis) is a syndrome that occurs in physically active women of reproductive age. Although it has been reported that stress affects reproductive function, the underlying mechanism is poorly understood (29). It has been reported, however, that acute psychological stress is able to provoke an increase in local IL-8 secretion (30), suggesting the possibility that opioid peptides in neutrophils are consumed through an increase in IL-8 caused by excessive stress, resulting in an abnormal estrous cycle in female athletes.

In conclusion, our results support the possibility that neutrophils regulate the estrous cycle through the release of opioid peptides, presumably β-endorphin, in response to MIP-2 in the ovaries. However, it is currently unknown whether the amount of β-endorphin released from neutrophils or other cells in the ovaries is sufficient for steroid hormone production, and therefore much work is needed in the future.

Abbreviations used in this article:

DAMGO

[D-Ala2, N-MePhe4, Gly-ol5]-enkephalin

DPDPE

[D-Pen2,5]-enkephalin

FSC

forward (light) scatter

NAL

naloxone

pAb

polyclonal Ab

pNeu

peripheral blood neutrophil

POMC

pro-opiomelanocortin

rKC

recombinant KC

SSC

side scatter (of light)

U-69593

(5α, 7α, 8β)-(−)-N-methyl-N-(7-[1-pyrrolidinyl]-1-oxaspiro[4,5]dec-8-yl)benzenacetamide

VLF

vaginal lavage fluid

vNeu

vaginally infiltrating neutrophil.

1
Kielian
T.
,
Barry
B.
,
Hickey
W. F.
.
2001
.
CXC chemokine receptor-2 ligands are required for neutrophil-mediated host defense in experimental brain abscesses.
J. Immunol.
166
:
4634
4643
.
2
Sonoda
Y.
,
Mukaida
N.
,
Wang
J. B.
,
Shimada-Hiratsuka
M.
,
Naito
M.
,
Kasahara
T.
,
Harada
A.
,
Inoue
M.
,
Matsushima
K.
.
1998
.
Physiologic regulation of postovulatory neutrophil migration into vagina in mice by a C-X-C chemokine(s).
J. Immunol.
160
:
6159
6165
.
3
Sasaki
S.
,
Nagata
K.
,
Kobayashi
Y.
.
2009
.
Regulation of the estrous cycle by neutrophil infiltration into the vagina.
Biochem. Biophys. Res. Commun.
382
:
35
40
.
4
Brack
A.
,
Rittner
H. L.
,
Machelska
H.
,
Leder
K.
,
Mousa
S. A.
,
Schäfer
M.
,
Stein
C.
.
2004
.
Control of inflammatory pain by chemokine-mediated recruitment of opioid-containing polymorphonuclear cells.
Pain
112
:
229
238
.
5
Rittner
H. L.
,
Labuz
D.
,
Richter
J. F.
,
Brack
A.
,
Schäfer
M.
,
Stein
C.
,
Mousa
S. A.
.
2007
.
CXCR1/2 ligands induce p38 MAPK-dependent translocation and release of opioid peptides from primary granules in vitro and in vivo.
Brain Behav. Immun.
21
:
1021
1032
.
6
Machelska
H.
,
Stein
C.
.
2006
.
Leukocyte-derived opioid peptides and inhibition of pain.
J. Neuroimmune Pharmacol.
1
:
90
97
.
7
Goodman
R. L.
1996
.
Neural systems mediating the negative feedback actions of estradiol and progesterone in the ewe.
Acta Neurobiol. Exp. (Warsz.)
56
:
727
741
.
8
Hamada
H.
,
Kishioka
S.
,
Yamoto
M.
,
Nakano
R.
.
1995
.
[3H]naloxone binding sites in porcine ovarian follicles and corpora lutea during the ovarian cycle.
Eur. J. Endocrinol.
132
:
622
626
.
9
Kaminski
T.
,
Siawrys
G.
,
Bogacka
I.
,
Okrasa
S.
,
Przala
J.
.
2003
.
The regulation of steroidogenesis by opioid peptides in porcine theca cells.
Anim. Reprod. Sci.
78
:
71
84
.
10
Kaminski
T.
,
Siawrys
C.
,
Bogacka
I.
,
Okrasa
S.
,
Przala
J.
.
2004
.
The influence of opioid peptides on steroidogenesis in porcine granulosa cells.
Reprod. Domest. Anim.
39
:
25
32
.
11
Rittner
H. L.
,
Labuz
D.
,
Schaefer
M.
,
Mousa
S. A.
,
Schulz
S.
,
Schäfer
M.
,
Stein
C.
,
Brack
A.
.
2006
.
Pain control by CXCR2 ligands through Ca2+-regulated release of opioid peptides from polymorphonuclear cells.
FASEB J.
20
:
2627
2629
.
12
Sirinathsinghji
D. J. S.
,
Audsley
A. R.
.
1985
.
Endogenous opioid peptides participate in the modulation of prolactin release in response to cervicovaginal stimulation in the female rat.
Endocrinology
117
:
549
556
.
13
Spanagel
R.
,
Almeida
O. F.
,
Shippenberg
T. S.
.
1994
.
Evidence that nor-binaltorphimine can function as an antagonist at multiple opioid receptor subtypes.
Eur. J. Pharmacol.
264
:
157
162
.
14
Iyoda
T.
,
Kobayashi
Y.
.
2004
.
Involvement of MIP-2 and CXCR2 in neutrophil infiltration following injection of late apoptotic cells into the peritoneal cavity.
Apoptosis
9
:
485
493
.
15
Tanimoto
N.
,
Terasawa
M.
,
Nakamura
M.
,
Kegai
D.
,
Aoshima
N.
,
Kobayashi
Y.
,
Nagata
K.
.
2007
.
Involvement of KC, MIP-2, and MCP-1 in leukocyte infiltration following injection of necrotic cells into the peritoneal cavity.
Biochem. Biophys. Res. Commun.
361
:
533
536
.
16
Daley
J. M.
,
Thomay
A. A.
,
Connolly
M. D.
,
Reichner
J. S.
,
Albina
J. E.
.
2008
.
Use of Ly6G-specific monoclonal antibody to deplete neutrophils in mice.
J. Leukoc. Biol.
83
:
64
70
.
17
Hestdal
K.
,
Ruscetti
F. W.
,
Ihle
J. N.
,
Jacobsen
S. E.
,
Dubois
C. M.
,
Kopp
W. C.
,
Longo
D. L.
,
Keller
J. R.
.
1991
.
Characterization and regulation of RB6-8C5 antigen expression on murine bone marrow cells.
J. Immunol.
147
:
22
28
.
18
Ribechini
E.
,
Leenen
P. J. M.
,
Lutz
M. B.
.
2009
.
Gr-1 antibody induces STAT signaling, macrophage marker expression and abrogation of myeloid-derived suppressor cell activity in BM cells.
Eur. J. Immunol.
39
:
3538
3551
.
19
Abbitt
K. B.
,
Cotter
M. J.
,
Ridger
V. C.
,
Crossman
D. C.
,
Hellewell
P. G.
,
Norman
K. E.
.
2009
.
Antibody ligation of murine Ly-6G induces neutropenia, blood flow cessation, and death via complement-dependent and independent mechanisms.
J. Leukoc. Biol.
85
:
55
63
.
20
Mousa
S. A.
,
Shakibaei
M.
,
Sitte
N.
,
Schäfer
M.
,
Stein
C.
.
2004
.
Subcellular pathways of beta-endorphin synthesis, processing, and release from immunocytes in inflammatory pain.
Endocrinology
145
:
1331
1341
.
21
Nguyen-Jackson
H.
,
Panopoulos
A. D.
,
Zhang
H.
,
Li
H. S.
,
Watowich
S. S.
.
2010
.
STAT3 controls the neutrophil migratory response to CXCR2 ligands by direct activation of G-CSF-induced CXCR2 expression and via modulation of CXCR2 signal transduction.
Blood
115
:
3354
3363
.
22
Staszkiewicz
J.
,
Skowronski
M. T.
,
Kaminski
T.
,
Siawrys
G.
,
Krazinski
B. E.
,
Kusmider
M.
,
Przala
J.
,
Okrasa
S.
.
2007
.
Expression of proopiomelanocortin, proenkephalin and prodynorphin genes in porcine theca and granulosa cells.
Anim. Reprod. Sci.
101
:
97
112
.
23
Lolait
S. J.
,
Autelitano
D. J.
,
Lim
A. T.
,
Smith
A. I.
,
Toh
B. H.
,
Funder
J. W.
.
1985
.
Ovarian immunoreactive beta-endorphin and estrous cycle in the rat.
Endocrinology
117
:
161
168
.
24
Naz
R. K.
,
Butler
A.
.
1996
.
Interleukins-6 and -8 levels in sera and cervical mucus of fertile, idiopathic infertile, and immunoinfertile women: implication in infertility.
Am. J. Reprod. Immunol.
35
:
534
540
.
25
Kalu
E.
,
Sumar
N.
,
Giannopoulos
T.
,
Patel
P.
,
Croucher
C.
,
Sherriff
E.
,
Bansal
A.
.
2007
.
Cytokine profiles in serum and peritoneal fluid from infertile women with and without endometriosis.
J. Obstet. Gynaecol. Res.
33
:
490
495
.
26
Houdeau
E.
,
Larauche
M.
,
Monnerie
R.
,
Bueno
L.
,
Fioramonti
J.
.
2005
.
Uterine motor alterations and estrous cycle disturbances associated with colonic inflammation in the rat.
Am. J. Physiol. Regul. Integr. Comp. Physiol.
288
:
R630
R637
.
27
Weber
A. M.
,
Ziegler
C.
,
Belinson
J. L.
,
Mitchinson
A. R.
,
Widrich
T.
,
Fazio
V.
.
1995
.
Gynecologic history of women with inflammatory bowel disease.
Obstet. Gynecol.
86
:
843
847
.
28
Bradley
R. J.
,
Rosen
M. P.
.
2004
.
Subfertility and gastrointestinal disease: ‘unexplained’ is often undiagnosed.
Obstet. Gynecol. Surv.
59
:
108
117
.
29
Tilbrook
A. J.
,
Turner
A. I.
,
Clarke
I. J.
.
2002
.
Stress and reproduction: central mechanisms and sex differences in non-rodent species.
Stress
5
:
83
100
.
30
Weik
U.
,
Herforth
A.
,
Kolb-Bachofen
V.
,
Deinzer
R.
.
2008
.
Acute stress induces proinflammatory signaling at chronic inflammation sites.
Psychosom. Med.
70
:
906
912
.

The authors have no financial conflicts of interest.